Parallel synthesis and biological activity of a new class of high affinity and selective δ-opioid ligand

Article abstract of DOI:10.1016/S0968-0896(01)00017-7

A considerable number of research papers describing the synthesis and testing of the delta opioid receptor (DOR) ligands, SNC-80 and TAN-67, and analogues of these two compounds, have been published in recent years. However, there have been few reports of the discovery of completely new structural classes of selective DOR ligand. By optimising a hit compound identified by high throughput screening, a new series of tetrahydroisoquinoline sulphonamide-based delta opioid ligands was discovered. The main challenge in this series was to simultaneously improve both affinity and physicochemical properties, notably aqueous solubility. The most active ligand had an affinity (IC₅₀) of 6 nM for the cloned human DOR, representing a 15-fold improvement relative to the original hit 1 (IC₅₀ 98 nM). Compounds from this new series show good selectivity for the DOR over μ and κ opioid receptors. However the most active and selective compounds had poor aqueous solubility. Improved aqueous solubility was obtained by replacing the phthalimide group in 1 by basic groups, allowing the synthesis of salt forms. A series of compounds with improved affinity and solubility relative to 1 was identified and these compounds showed activity in an in vivo model of antinociception, the formalin paw test. In the case of compound 19, this analgesic activity was shown to be mediated primarily via a DOR mechanism. The most active compound in vivo, 46, showed superior potency in this test compared to the reference DOR ligand, TAN-67 and similar potency to morphine (68% and 58% inhibition in Phases 1 and 2, respectively, at a dose of 10 mmol/kg i.v.).

Full text of DOI:10.1016/S0968-0896(01)00017-7

Bioorganic & Medicinal Chemistry 9 (2001) 2609–2624
Parallel Synthesis and Biological Activity of a New Class
of High Affinity and Selective ꢀ-Opioid Ligand
D.R. Barn,a W.L. Caulfield,a J. Cottney,a K. McGurk,b J.R. Morphy,a,* Z. Rankovica
and B. Robertsc
aOrganon Laboratories Ltd, Newhouse, ML1 5SH, Scotland, UK
bN.V. Organon, Molenstraat 110, 5430 BH, Oss, The Netherlands
cAstraZeneca, Alderley Park, Macclesfield, Cheshire, SK10 4TG, UK
Received 6 October 2000; revised 11 January 2001; accepted 15 January 2001
Abstract—A considerable number of research papers describing the synthesis and testing of the delta opioid receptor (DOR)
ligands, SNC-80 and TAN-67, and analogues of these two compounds, have been published in recent years. However, there have
been few reports of the discovery of completely new structural classes of selective DOR ligand. By optimising a hit compound
identified by high throughput screening, a new series of tetrahydroisoquinoline sulphonamide-based delta opioid ligands was dis-
covered. The main challenge in this series was to simultaneously improve both affinity and physicochemical properties, notably
aqueous solubility. The most active ligand had an affinity (IC50) of 6 nM for the cloned human DOR, representing a 15-fold
improvement relative to the original hit 1 (IC50 98 nM). Compounds from this new series show good selectivity for the DOR over m
and k opioid receptors. However the most active and selective compounds had poor aqueous solubility. Improved aqueous solu-
bility was obtained by replacing the phthalimide group in 1 by basic groups, allowing the synthesis of salt forms. A series of com-
pounds with improved affinity and solubility relative to 1 was identified and these compounds showed activity in an in vivo model
of antinociception, the formalin paw test. In the case of compound 19, this analgesic activity was shown to be mediated primarily
via a DOR mechanism. The most active compound in vivo, 46, showed superior potency in this test compared to the reference
DOR ligand, TAN-67 and similar potency to morphine (68% and 58% inhibition in Phases 1 and 2, respectively, at a dose of
10 mmol/kg i.v.). # 2001 Elsevier Science Ltd. All rights reserved.
Introduction
last few years, selective non-peptide agonists and
antagonists for the DOR have been discovered (Fig. 1).
The first non-peptide antagonist to possess reasonable
selectivity for the DOR, naltrindole (NTI), was dis-
covered by applying the message-address concept.3
Opioid receptors are classified into three major sub-
types: m, k and d. Most currently used m-opioid analgesic
drugs, such as fentanyl and its analogues, exhibit a
range of serious side effects including respiratory
depression, dependence and muscle rigidity. Extensive
studies over the past decade have shown that d-opioid
receptor (DOR) agonists produce antinociception in
animal models of pain whilst appearing to lack the
characteristic side-effect profile of m-opioid receptor
(MOR) agonists.1,2 For this reason the DOR is an
attractive target for the design of novel opioid analge-
sics. A number of peptide agonists that display high
affinity and selectivity for the DOR have been reported,
but their clinical use has been hampered by poor meta-
bolic stability and low systemic availability. Over the
More recently,
a highly selective peptidomimetic
antagonist, H-Dmt-Tic-OH, based upon the N-terminal
message domain of the endogenous enkephalins, was
discovered.4 One of the first non-peptide agonists to
show good affinity and selectivity at the DOR, TAN-67,
is structurally related to naltrindole.5 However, a sec-
ond agonist, SNC-80, is clearly unrelated to the mor-
phinoid skeleton.6 Despite their structural diversity, all
these selective delta opioid ligands have one feature in
common, a basic amine. Earlier reports of non basic
opioids are few, such as the discovery that casomor-
phin-derived hexapeptides, such as N-t-Boc-Tyr-Pro-
Gly-Phe-Leu-Thr-OH, are delta opioid selective
antagonists.7 Recently we published the structure of a
new class of delta opioid ligand 1, which lacks a proto-
natable nitrogen.8 Very recently the structures of cyclic
*Corresponding author: Fax: +44-01698-736187; e-mail: r.morphy@
organon.nhe.akzonobel.nl
0968-0896/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00017-7

2610
D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
regions. Method B was used to make changes in the
aryl-sulphonamide region.
Compound 1 was also synthesised using the solid phase
routes shown in Scheme 2 (Methods C and D). This
novel methodology by which a range of cyclic imides
can be formed by cyclative cleavage has recently been
published.8 Method D allowed the synthesis of a two-
dimensional library in which both the phthalimide and
tosyl groups could be varied simultaneously.
We attempted to synthesise a larger amount of 20 using
a more traditional solution phase method, opening of
the anhydride followed by thermolytic ring closure
(Scheme 3, Method E). However decarboxylation of
regioisomer 17 occurred at the intermediate carboxylic
acid stage, giving a mixture of the amide 19 and the
imide 20.11
A library of 312 amides was synthesised using a solution
phase route, utilising solid phase reagents and sca-
vengers (Scheme 4, Method F). The primary amine 15
was converted to 12 different amides 21 using carboxylic
acids and 1-(3-dimethylaminopropyl)-3-ethyl-carbodii-
mide hydrochloride (EDC). After a simple purification
by aqueous extraction, the Boc group was removed
using 25% trifluoroacetic acid (TFA) in dichloro-
methane (DCM). In the second diversity-generating
step, the free secondary amine was converted to various
sulphonamides using sulphonyl chlorides, in the pre-
sence of resin-bound piperidine. A slight excess of sul-
phonyl chloride (1.25 mol equiv) was used to drive the
reaction to completion, and this was scavenged using
aminomethyl-polystyrene resin. The 312 amides 22 were
obtained in an average yield of 29.3 mg (84%) and
81.4% purity, as measured by HPLC. Each of the 312
products was analysed by flow injection MS, and 60
products were selected at random and structures
Figure 1. Selective ligands for the delta opioid receptor.
peptides, which are also non basic ligands at the DOR,
have also been published.9 Compound 1, identified
through high throughput screening of the Organon
compound collection, was attractive for reasons of
novelty, high selectivity over the other opioid receptor
sub-types and synthetic accessibility. The main dis-
advantage, particularly from the perspective of an
intravenously-administered analgesic, was its poor
aqueous solubility (<0.1 mg/mL). This paper describes
our efforts to optimise both the affinity and physico-
chemical properties of 1.
1
confirmed by H NMR.
Structure–activity relationships. For the purposes of
simplicity, we divided the molecule into three fragments:
the aryl sulphonamide region (R1), the phthalimide side
chain region (R2) and the tetrahydroisoquinoline
(THIQ) region (R3) (Fig. 1). We explored the basic SAR
in each of these regions, by conducting deletion studies,
before embarking on the synthesis of hit optimisation
libraries.
Results and Discussion
Chemistry
Compound 1 was synthesised in three steps, starting
from
2-phenethylamine
2
and
N-(3-hydro-
Region R1. The 4-toluene sulphonamide group is extre-
mely important for affinity (Table 1). Complete removal
of the tosyl group or replacement of the aromatic ring
by a methyl group, compounds 27 and 26 respectively,
produce a complete loss of affinity. More conservative
changes such as replacement of the sulphonyl group by
a carbonyl, 24, or removal of the 4-methyl group, 23,
also cause all activity to be lost.
xypropyl)phthalimide 4 (Scheme 1, Method A). 2-phe-
nethyl amine was tosylated using tosyl chloride in
pyridine. N-(3-hydroxypropyl)phthalimide was oxidised
using pyridinium chlorochromate in dichloromethane.
Pictet Spengler cyclisation in neat TFA provided 1 in
89.8% overall yield. An alternative route to 1 started
from N-(2-carboxyethyl)phthalimide 6 and involved
conversion to the amide 7, followed by a Bischler-
Napieralski reaction (Scheme 1, Method B). Protona-
tion and hydrogenation produced the tetra-
hydroisoquinoline 1010 and finally tosylation gave 1 in
65% overall yield. Method A was used to make changes
in the tetrahydroisoquinoline (THIQ) and phthalimide
Region R2. The phthalimide-containing side chain is
also important for activity, but the SAR appears to be
somewhat less tight in this region compared to R1
(Table 2). Whereas substitution of the phthalimidoethyl
group by a methyl group 29 causes a 25-fold decrease in

D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
2611
activity, its replacement by an acetamidoethyl 28 gives
only a 3-fold drop in affinity suggesting that the aro-
matic ring is not an essential pharmacophoric group.
Extension of the side chain by one carbon 30 gives a
slight decrease in affinity, but a one carbon reduction 31
has a more profound effect, with a near 30-fold
attenuation.
Overall, the preliminary SAR study suggests that the
tosyl group is very important for activity. In particular,
a para-substituent and the SO2 group appear to be cru-
cial and these features were therefore retained in the
subsequent hit optimisation libraries. There appeared to
be more scope for varying the phthalimide group and it
was felt that this region offered the best opportunity for
manipulating the physicochemical properties of the
compound. Substituents on the THIQ ring appear to
offer no advantage in terms of affinity.
Region R3. Aromatic substituents on the THIQ ring
were not well tolerated (Table 3). The substituents were
limited to electron rich groups by the synthetic route,
since the Pictet-Spengler cyclization does not permit
electron-deficient aromatic precursors to be used. The
most active analogue was the 7-hydroxy compound 32,
but the affinity was 3-fold lower than the unsubstituted
parent. Ethers in the 5-, 6- or 7- positions are slightly
less active (33, 35 and 34 respectively), whereas an alkyl
group in the 7-position (37) or a dihydroxy-substituted
ring (36) gives analogues that are virtually inactive.
Removal of the THIQ aromatic ring 39, or its replace-
ment by an imidazole 40, led to compounds with little
or no affinity.
In the first library, we concentrated on the R2 region,
making modest changes to the structure of 1, by chan-
ging the phthalimide group for other cyclic imides,
whilst keeping the tosyl group constant (Table 4).
Compound 1 was remade as an internal standard within
the library and was found to exhibit similar affinity:
112 nM for 41{1} compared to 98 nM for 1 (Table 4).
The SAR apparent from testing the library at the DOR
suggests that an electron withdrawing substituent in the
3-position on the phthalimide ring is favoured, since
both the 3-nitro 41{3} and 3-fluoro 41{5} derivatives
Scheme 1.

2612
D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
Scheme 2.
Scheme 3.

D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
2613
Scheme 4.
were 2-fold more active than 1. The presence of a sec-
ond fluorine atom, in the 6-position 41{6}, gave a fur-
ther increase in affinity to 16 nM, representing a 6-fold
improvement relative to 1. In contrast, the 3,6-dichloro
derivative 41{9} was much less active, possibly indicat-
ing that two electron withdrawing substituents are only
well tolerated if of small size. The pyridine derivative
41{2} was 4-fold more active than 1, whereas the pyra-
zine 41{12} was 2-fold less active. Substituents in the 4-
position are generally less well tolerated. For example,
the 4-nitro 41{4} and 4-fluoro derivatives 41{13} were
less active than 1, and much less active than the analo-
gues with the equivalent substituent in the 3-position.
Derivatives containing both 4- and 5-substituents
showed very poor activity, such as the dichloro-com-
pound 41{10} and the 3,4-naphthyl derivative 41{15}.
The tetrafluoro-analogue 41{10} gave only moderate
activity relative to 41{6}, again suggesting that 4- and 5-
substituents are not well tolerated. The only compound
containing a 3,4-disubstitution pattern, 41{14}, was
poorly active. The combination of a nitrogen in the 3-
position of the ring, with a 4-methyl substituent 41{16},
provided good affinity. In this case, the beneficial effect
of the ring nitrogen seems to counteract the generally
detrimental effect of introducing a 4-substituent.
provide higher affinity, although the SAR is generally
quite flat: cyclopropyl 41{25}>trans-cyclohexyl
41{20}>cis-cyclohexenyl 41{18}>bicycloheptene
41{22}>cis-cyclohexyl 41{19}>bicyclooctene 41{23}.
Overall, this first library produced eight compounds
with higher affinity than 1. Perhaps the most interesting
new compound was 41{16}, which was 3-fold more
active than 1, and due to the presence of a pyridine ring
offered the prospect of improved water solubility rela-
tive to the parent phthalimide.
Table 2. Biological Data for Ligands with variations in the R2 position
R2
IC50 d (nM) c
a
1
PhtCH2CH2
MeCONHCH2CH2ꢀ
98
293
2400
197
b
28
29
30
31
a
Meꢀ
a
PhtCH2CH2CH2ꢀ
a
PhtCH2ꢀ
2700
aCompounds were made using Method A (Scheme 1).
bCompounds were made from 11 (Scheme 1); Pht=Phthalimido.
cIC50 values were determined from the log concentration-inhibition
curves (at least six points) and are mean values of at least two
experiments.
The SAR for the succinimides within the library sug-
gests that small non-polar substituents are well toler-
ated. For a single substituent in the 3-position, the order
of activity is: Me 41{28}>Ph 41{27}>HCONH
41{24}>PhCH2OCONH 41{21}. Two methyl groups,
either on the same 41{26} or different 41{29} carbons,
are also tolerated. Where a second saturated ring is
fused to the succinimide ring, small rings appear to
Table 3. Biological data for ligands with variations in the R3 posi-
tiona
Table 1. Biological data for ligands with variations in the R1 positiona
R3
IC50 d (nM)b
1
H
7-OH
98
296
510
570
1400
6900
9313
568
32
33
34
35
36
37
38
39
40
R1
IC50 d (nM)b
5-OMe
7- OMe
6- OMe
6,7-(OH)2
7-Me
1
4-Me(C6H4)SO2ꢀ
C6H5SO2ꢀ
4-Me(C6H4)COꢀ
C6H5CH2ꢀ
MeSO2ꢀ
98
7044
>10,000
>10,000
>10,000
>10,000
23
24
25
26
27
7-(OCH2C6H4CN)
—
—
>10,000
8000
H
aCompounds were made using Method A (Scheme 1).
bIC50 values were determined from the log concentration-inhibition
curves (at least 6 points) and are mean values of at least 2 experiments.
aCompounds were made from 10 (Scheme 1).
bIC50 values were determined from the log concentration-inhibition
curves (at least 6 points) and are mean values of at least 2 experiments.

2614
D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
Table 4. Biological data for phthalimides, succinimides and maleimidesa
Phthalimides 41
Succinimides/Maleimides 41
Phthalimides:
41{1}
41{2}; W=N
41{3}; A=NO2
41{4}; B=NO2
41{5}; A=F
41{6}; A,D=F
41{7}; A,B,C,D=F
41{8}; B=Me
41{9}; A,D=Cl
41{10}; B,C=Cl
41{11}; B=Cl
IC50 d (nM)b
112
24
Succinimides:
IC50 d (nM)b
195
249
41{17};
41{18}; A-C=CH2CHCHCH2 (cis)
41{19}; A-C=CH2CH2CH2CH2 (cis)
41{20}; A-C=CH2CH2CH2CH2 (trans)
41{21}; A=PhCH2OCONH (S)
41{22}; A-C=bicyclo[2.2.1]hept-5-ene (endo)
41{23}; A-C=bicyclo[2.2.2]oct-5-ene (endo)
41{24}; A=HCONH (R)
41{25}; A-C=CH2
48
496
42
16
464
496
1659
6709
244
339
202
517
279
2152
627
86
165
41{26}; A,B=Me
41{27}; A=Ph
41{28}; A=Me
41{29}; A,C=Me
174
75
86
41{12}; W,Z=N
41{13}; B=F
164
173
41{14}; A-B=CHCHCHCH
41{15}; B-C=CHCHCHCH
41{16}; W=N, B=Me
2691
1302
35
Maleimide:
41{30}; A-C=CH2CH2CH2CH2
343
Unless stated otherwise A,B,C,D represent hydrogen and W,X,Y,Z represent carbon. A,B indicates both positions are substituted as shown. A,B
indicates a bridging group between the two positions. R=1-(2-ethyl)-2-(4-toluenesulphonyl)-1,2,3,4-tetrahydroisoquinoline.
aCompounds were made using Method C (Scheme 2).
bIC50 values were determined from the log concentration-inhibition curves (at least 6 points) and are mean values of at least 2 experiments.
Interestingly amide 19, a side product obtained during
the synthesis of 20, was also active at the DOR
(110 nM), with reasonable selectivity over m and k
opioid receptors, >100-fold and 30-fold respectively.
Significantly, whereas it was not possible to produce a
salt form of 20 with either hydrochloric acid or sul-
phuric acid in order to improve aqueous solubility, the
hydrosulphate salt of 19 could be obtained. Both the
molecular weight and cLogP are lower for 19 than for 1,
which suggested that the new compound may have
superior CNS penetration (Table 7).12
of these, the R2 region was held constant as the pyri-
dine-containing cyclic imide, found in 41{16}, and the
R1 region was varied (Table 5; Scheme 2). The main
driving force behind this library was an attempt to
improve further the affinity of 41{16} by introducing
alternative substituents on the aryl sulphonamide ring.
In the second of these libraries, we used a solution phase
route to make analogues of the amide 19 and varied
both the R1 and R2 regions combinatorially. The main
emphasis for this library was to explore a wider range of
other water-solubilising groups in the R2 region, since
we knew that this region was most likely to tolerate
basic groups (Table 6; Scheme 4).
Having made some progress towards our goal of
improving both the affinity and reducing the lipophili-
city of 1, we designed two follow-up libraries. In the first
Replacement of the tosyl group in 42{1} [=41{16}] wi th
a 4-(n-butylphenyl)-sulphonyl group, 42{2} gave a 9-
fold improvement in affinity. However, the 4-chloro
42{3}, diethylaminomethyl 42{6}, 3-methyl 42{7} and 4-
trifluoromethyl 42{8} analogues were much less active
than the parent. Two other analogues retained similar
affinity to the tosyl compound, the 4-(ethylphenyl)-sul-
phonamide 42{4} and the 4-(n-butoxyphenyl)-sulpho-
namide 42{5}. This SAR suggests that binding of the
butyl chain within a hydrophobic channel on the recep-
tor may be responsible for the improved affinity of
42{2}.
Table 5. Biological data for sulphonamide variationsa
A
B
IC50 d (nM)b
42{1}
42{2}
42{3}
42{4}
42{5}
42{6}
42{7}
42{8}
Me
nBu
Cl
Et
nBuO
Et2NCH2
H
H
H
H
H
H
H
Me
H
52
6
1100
53
The % inhibition of 3H-NTI binding for each of the 312
amides 22 in the DOR binding assay was determined at
two concentrations: 100nM and 1000 nM (Table 6).
Although single concentration inhibition data of this type
should be treated wi th cauti on, i t di d enable us to di scern
SAR trends which were consistent with those observed in
the imide library above. As in the imide library, replace-
ment of the tosyl group (sulphonamide No. 9) with a 4-(n-
butylphenyl)-sulphonyl group, (sulphonamide No. 23),
80
4350
608
800
CF3O
aCompounds were made using Method D (Scheme 2).
bIC50 values were determined from the log concentration-inhibition
curves (at least six points) and are mean values of at least two experiments.

D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
2615
Table 6. The upper and lower figures are the % inhibition of 3H-Naltrindole binding to the DOR at 1000 nM and 100 nM respectively. If deter-
mined, the figure in bold is the IC50 (nM). Compounds made using Method F
1
2
3
4
5
6
7
8
9
10
11
12
1
2
41
9
49
25
46
22
9
25
>10000
17
9
12
44
29
22
35
21
32
17
7
26
16
20
30
16
46
15
52
24
46
23
23
27
13
18
20
25
29
26
31
13
35
25
13
33
21
20
37
25
>10000
3
50
17
48
19
59
30
37
ꢀ4
27
18
42
21
29
28
32
22
34
28
31
28
23
24
46
27
3470
4
55
18
52
16
60
25
34
15
23
14
46
28
30
25
36
23
30
22
35
36
22
17
41
18
6975
5
48
21
50
29
50
17
16
13
15
18
26
22
34
19
33
10
33
15
12
36
13
15
39
22
>10000
6
46
18
51
22
48
23
13
23
9
22
27
23
29
2
33
11
32
15
11
34
12
18
41
23
>10000
7
70
39
67
33
81
56
66
15
53
23
75
14
39
31
56
41
63
43
69
48
26
20
70
33
225
676
8
66
27
65
31
65
33
52
16
57
26
55
21
36
25
40
38
60
46
47
28
31
19
71
31
292
9
72
42
65
35
84
51
67
36
55
42
77
46
36
33
53
41
64
46
68
42
30
22
67
32
155
283
10
11
12
13
14
15
16
48
13
52
19
50
21
14
22
>10000
14
38
15
31
35
7
36
18
28
15
0
11
20
21
43
22
52
21
56
22
57
22
36
31
5963
36
31
41
39
37
8
39
30
38
35
16
13
21
21
48
25
73
34
71
32
86
50
166
66
47
252
62
15
78
31
40
39
66
45
66
49
58
35
42
18
69
33
78
33
77
43
84
56
185
61
57
175
72
18
76
20
51
41
3950
74
50
435
75
55
443
65
35
54
27
77
41
59
21
63
30
61
20
37
54
355
58
35
42
37
4326
35
45
35
55
43
31
26
42
21
64
34
72
30
68
43
77
41
61
37
291
55
43
64
41
35
27
58
41
64
49
78
44
40
21
73
51
1191
49
23
51
31
57
23
39
23
22
32
40
28
36
18
30
21
30
19
24
24
20
15
52
24
7940
(continued)

2616
D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
Table 6 (continued)
1
2
3
4
5
6
7
8
9
10
11
12
17
18
19
20
21
22
23
45
14
50
19
47
19
22
32
7150
15
40
25
37
38
15
31
17
28
12
21
20
181
16
492
28
49
15
50
18
53
27
24
20
6450
16
25
33
22
34
3
34
10
28
9
17
21
20
18
43
20
49
15
53
26
53
26
17
7
2555
27
20
24
22
39
17
37
22
32
31
18
23
26
17
52
22
54
22
59
22
53
28
36
12
2252
41
25
28
25
38
17
48
44
56
40
30
17
37
26
69
31
44
24
53
19
55
34
23
13
1328
19
20
52
35
35
31
37
37
40
36
54
29
43
26
59
22
58
18
51
23
61
35
39
20
2850
22
27
53
31
33
16
31
23
32
29
20
22
32
23
52
21
83
51
122
88
62
127
88
63
44
77
73
22
86
54
318
80
61
103
66
55
718
86
66
240
86
74
103
72
47
257
74
38
992
90
59
225
24
25
79
45
74
44
85
60
233
74
65
213
71
49
80
39
52
40
68
52
430
73
55
529
73
43
58
28
78
42
48
24
49
25
47
23
18
40
16
44
23
36
38
20
32
20
30
10
9
22
20
12
50
48
7600
26
52
25
46
26
50
27
24
43
8
43
26
22
34
14
33
12
28
5
23
32
25
14
52
47
2200
gave derivatives with much improved activity (Fig. 2).
Generally sulphonamides bearing 4-alkyl substituents
gave the highest activity, at both 100 and 1000 nM, in
the order: 4-nBu>3-Cl, 4-Me>4-nPr>4-iPrꢁ4-Meꢁ4-
Et. With the exception of 4-OMe (sulphonamide No. 7),
which gave reasonable inhibition, particularly at 1000
nM, more polar substituents in the 4-position were not
well tolerated. Likewise, multiply-substituted aromatics
produced low activity, e.g., 2,3,4,5,6-pentafluoro-, 2,4,6-
trimethyl and 2,3,4,5,6-pentamethyl-phenylsulphonyl.
The SAR for the 12 amine variants appeared to be flat-
ter than that for the sulphonamides, particularly at
100 nM (Fig. 3). At both 100 nM and 1000 nM, amine
No. 3 appeared to be the most active.
IC50s were also determined for all derivatives containing
either amine No. 4 or sulphonamide No. 23. The SAR
trend observed from the crude data was essentially con-
firmed for the 26 different sulphonamides in combina-
tion with amine No. 4 (Fig. 2). Again the most active
compound 22{23,4} contained the 4-(n-butylphenyl)-
sulphonyl group (22 nM), and the order of activity was
4-nBu>4-nPrꢁ3-Cl, 4-Meꢁ4-iPrꢁ4-Meꢁ4-Et. In com-
bination with sulphonamide No. 23, the amine variant
which gave rise to the most active compound was the
original (4-methyl)-3-pyridyl group, amine No. 4 (Fig.
3). The next most active contained a (3-pyridyl)methyl
group (98 nM). Confirming a relatively flat SAR in this
region, all 12 amines, in combination with sulphona-
mide No. 23, provided IC50s of less than 1 mM.
All compounds which produced >50% inhibition at
100 nM were selected for IC50 determination (Table 6;
bold text). In order to obtain more SAR information,
Four compounds from the amide library were selected
for resupply as hydrochloride salts for in vivo testing:

D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
2617
Figure 2. Trends in activity for the 26 sulphonamide variants in 22. The total % inhibition data (100 nM and 1000 nM; left hand linear scale)
represent the sum of the % inhibitions for the 12 different amines at these concentrations (shown in Table 6). The IC50 data (right hand logarithmic
scale) were determined for amine No. 4 only.
Table 7. Opioid binding and in vivo data for reference m and d opioids, compound 1 and analoguesa
IC50
(nM)a
d
IC50 m
(nM)
IC50 k
(nM)
mouse FPT: Phase 1bc
(dose in mmol/kg i.v.)
mouse FPT: Phase 2 b,c
(dose in mmol/kg i.v.)
M.Wt.
449.64
clogP
4.9
SNC-80
2.2d
9600
8200
48 (3)
6 (3 )
61 (20)
58 (10)
97 (0.1)
nte
55 (10)
62 (20)
36 (20)
57 (10)
53 (20)
68 (10)
39 (10)
79 (3)
37 (3)
52 (20)
69 (10)
79 (0.1)
nte
47 (10)
73 (20)
22 (20)
47 (10)
39 (20)
58 (10)
10 (10)
TAN-67
Morphine
Fentanyl
1
1.4d
257
254
98
1197
3.5
4.1
>10,000
>10,000
1160
105
485
>10,000
3300
344.46
285.34
336.48
460.55
449.57
3.8
0.6
3.6
5.5
4.9
19
110
43
44
45
46
256 (127)
37 (22)
172 (98)
265 (185)
>10,000
>10,000
>10,000
>10,000
1010
1285
1680
2005
457.63
491.65
491.65
480.62
5.3
6.4
5.1
4.9
aMean value of at least two experiments; in brackets IC50 for unpurified library compound for comparison.
b% Inhibition of paw licking behaviour.
cIn italics, % inhibition in the presence of Naltrindole (10 mg/kg ip) administered 35 min before and immediately prior to compound.
dAffinity data consistent with published data.15,16
ent=not tested due to low aqueous solubility of compound.
43, 44, 45 and 46, equivalent to 22{23,2}, 22{23,4},
22{23,9}, 22{23,12} respectively (Table 7). All com-
pounds appeared to have similar affinity to the equiva-
lent unpurified library compounds, albeit slightly lower.
Selectivity over the MOR was very good for all
compounds, but KOR selectivity was somewhat lower.
For 44, m selectivity was >270-fold and k selectivity was
35-fold. It would appear that the replacement of the
tosyl group with a 4-(n-butylphenyl)-sulphonyl group,
increases both k and d affinity.
higher activity being observed at a dose of 20 mmol/
kg i.v.: 65% in Phase 1 and 71% in Phase 2.
Unsurprisingly, considering its much lower affinity at
In vivo pharmacology. Analgesic activity of test com-
pounds was assessed in mice in the formalin paw test in
mice (FPT), a well-validated model of inflammatory
pain.13,14 Whilst compound 1 was too insoluble to be
tested, the more soluble compound 19 was found to
possess analgesic activity in the FPT (Table 7). At a
dose of 10 mmol/kg i.v., 19 inhibited paw licking beha-
viour by 58% in Phase 1 of the test and by 42% in
Phase 2. The inhibitory effect is dose dependent, with
Figure 3. Trends in activity for the 12 amine variants in 22. The total
% inhibition data (100 nM and 1000 nM; left hand linear scale) repre-
sent the sum of the % inhibitions for the 26 sulphonamides at these
concentrations (shown in Table 6). The IC50 data (right hand log.
scale) were determined for sulphonamide No. 23 only.

2618
D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
the DOR, 19 was less active than the literature d opioid,
SNC-80. SNC-80 produced 48% inhibition in Phase 1
and 79% inhibition in Phase 2 at a dose of 3 mmol/kg i.v.
The analgesic activity of both 19 and SNC-80 was
reduced significantly by the d-selective antagonist, nal-
trindole (NTI). Encouragingly, this confirms that com-
pound 19, like SNC-80, produces analgesia which is
mediated primarily via a DOR mechanism.
0.1% formic acid to 90% acetonitrile/0.1% formic acid)
at a flow rate of 4 mL/min. Analytical GC–MS (EI) ana-
lysis was carried out on a Hewlett Packard 5890 Series II
gas chromatogram (temperature gradient 100 ꢃC–310 ꢃC,
25 ꢃC/min; RTX1 column: 30 mꢂ0.25 mm, DF 0.25 mm)
and Hewlett Packard 5971 mass selective detector.
Analytical GLC was recorded on Shimadzu GC-14A
gas chromatograph (240 ꢃC; RTX1 column: 30
mꢂ0.25 mm; DF 0.25 mm). Preparative reversed-phase
HPLC purification was carried out on a Gilson system
using a LUNA C18 column (5 mm; 60ꢂ21.2 mm) under
gradient conditions (10 min; 90 % water/0.1 % TFA to
90% acetonitrile/0.1% TFA) at a flow rate of 5 mL/min.
Lipophilicity (clog P) was calculated using the
CLOGP3, PCModels version 4.34 (Daylight Chemical
Information Systems Inc. Irvine, CA, USA).
Compounds 43, 44, 45 and 46 were also tested in the
FPT in mice and all showed some analgesic activity. The
highest affinity ligand 44 showed weaker analgesic
activity than 43 or 45 in both phases of the test. Only
the imidazole-containing 46 was less active. It is tempt-
ing to speculate that the high lipophilicity of 44 (cLogP
6.4; Table 7) is a contributory factor, perhaps causing
extensive plasma protein binding which is reducing CNS
penetration.12
1-[2-(1,3-dihydro-1,3-dioxo-2H-isoindol-2-yl)ethyl]-1,2,3,4-
tetrahydro-2-[(4-methylphenyl)sulfonyl]-isoquinoline
(1)
(Method A). A mixture of 3 (9.12 g, 33.1 mmol) and 5
(7.40 g, 36.4 mmol) in trifluoroacetic acid (70 mL) was
stirred at reflux for 90 min and then allowed to cool to
room temperature. The reaction mixture was added to
200 mL of ice and water and the product was extracted
into CH2Cl2 (2ꢂ100 mL), and combined extracts were
washed with saturated aqueous KHCO3 and then dried
(Na2SO4) and concentrated to give a gum which crys-
tallised. The crude product was recrystallised from
CH2Cl2/EtOH to give a powder which was further pur-
ified by column chromatography on silica gel using
heptane/EtOAc (4:1) followed by CH2Cl2/CH3OH
(95:5) as eluent. The product was obtained as a crystal-
line solid (13.83 g, 90%). 1H NMR (CDCl3): d 1.95–2.29
(m, 2H), 2.31 (s, 3H), 2.48–2.60 (m, 2H), 3.53–3.73 (m,
1H), 3.85–4.06 (m, 3H), 5.06–5.18 (m, 1H), 6.84 (d,
1H),7.00-7.18 (m, 5H),7.60 (d, 2H), 7.64–7.76 (m, 2H),
7.76–7.89 (m, 2H).
Conclusion
The tetrahydroisoquinoline sulphonamide-based ligands
described in this paper represent a completely new
structural class of d opioid ligand. Whilst the com-
pounds were attractive leads in terms of affinity,
selectivity and synthetic accessibility, the main challenge
in this series was to improve physicochemical proper-
ties, notably aqueous solubility. The most active analo-
gue had an affinity of 6 nM for the cloned human DOR,
representing a 15-fold improvement relative to the ori-
ginal hit 1. However this analogue also had poor aqu-
eous solubility. It was difficult to combine in a single
molecule structural features which provided both
improved affinity and solubility. Improved aqueous
solubility was obtained by replacing the phthalimide
group in the original hit by basic groups, allowing the
synthesis of salt forms. The best compromise between
affinity and solubility was found in compound 44 which
had 3-fold better affinity than the original hit (37 nM)
and a >10-fold improvement in aqueous solubility.
Several analogues showed activity in an in vivo model
of antinociception, the formalin paw test. In the case of
compound 19, this analgesic activity was shown to be
mediated primarily via a d opioid mechanism. The most
active compound in vivo, 45, showed similar potency in
this test to morphine and the previously reported delta
opioid ligands, TAN-67 and SNC-80.
1-[2-(1,3-dihydro-1,3-dioxo-2H-isoindol-2-yl)ethyl]-1,2,3,4-
tetrahydro-2-[(4-methylphenyl)sulfonyl]-isoquinoline (1)
(Method B). To a solution of 1-[N-(2-phthalimido-
ethyl)]-1,2,3,4-tetrahydroisoquinoline
9.8 mmol) and diisopropylethylamine
(10)
(3.0 g,
(1.29 g,
10.0 mmol) in CH2Cl2 (30 mL) was added p-toluene-
sulfonyl chloride (1.91g, 10.0 mmol). Solution was stirred
at room temperature for 2 h then washed with water,
dried (Na2SO4) and solvent removed under reduced
pressure. The residue was purified by flash chromato-
graphy on silica gel using CH2Cl2/CH3OH 95:5 as
eluent to afford (1) as a crystalline solid (4.50 g, 99%).
Experimental
4-methyl-N-(2-phenylethyl)-benzenesulfonamide (3). To a
solution of 2-phenethylamine 2 (12.12 g, 0.1 mol) in dry
pyridine (100 mL) strirring under N2 in a cooling bath
at ꢀ20 ꢃC was added p-toluenesulfonyl chloride (25.0 g,
0.13 mol) over 10 min, keeping internal temperature
below ꢀ20 ꢃC during the addition. The resultant red
solution was stirred for 18 h at room temperature, then
quenched with water (400 mL). The mixture was
extracted into CH2Cl2 (3ꢂ150 mL), and the extracts
were combined and washed with 2 N HCl (3ꢂ200 mL),
brine, and then dried (Na2SO4). Solvent was evaporated
to obtain the product (27.23 g, 99%). 1H NMR
General chemical procedures
All reagents were either purchased from common com-
mercial sources or synthesised according to literature
references using commercial sources. Proton NMR (1H
NMR) were obtained on a Bruker DPX 400 spectro-
meter and are referenced to internal TMS. Mass spectra
were recorded on a Shimadzu LC-8A (HPLC) PE Sciex
API 150EX LC–MS. Analytical reversed-phase LC–MS
analysis was carried out on LUNA C18 column (5 mm;
30ꢂ4.6 mm) under gradient conditions (90% water/

D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
2619
(CDCl3): d 2.43 (s, 3H), 2.76 (t, 2H), 3.21 (t, 2H), 4.33
(br t, 1H), 7.06 (d, 2H), 7.20–7.35 (m, 5H), 7.70 (d, 2H).
carbonyl]-3-pyridinecarboxylic acid (18). To a stirring
suspension of 6-methyl-2,3-pyridinedicarboxylic acid
(90 mg, 0.5 mmol) in dry THF (10 mL) was added trie-
thylamine (102 mg, 1.0 mmol), forming a solution. Ethyl
chloroformate (108 mg, 1.0 mmol) was added and the
resultant suspension was stirred at room temperature
for 30 min. A solution of 11 (165 mg, 0.5 mmol) in THF
(1 mL) was added at a rapid dropwise rate and the
reaction mixture was stirred at reflux for 2 h. Solvent
was removed under reduced pressure and residue was
partitioned between EtOAc (10 mL) and water (10 mL).
Organic layer was washed with water (2ꢂ5 mL), dried
(Na2SO4) and concentrated, then purified by flash
chromatography on silica gel using CH2Cl2/CH3OH
(95:5) as eluent, to give a mixture of the amides 17 and
18 (150 mg). 1H NMR confirmed the product was a
mixture. The product was used in the next step without
any further purification.
1,3-dihydro-1,3-dioxo-2H-isoindole-2-propanal, (5). To a
stirred solution of N-(3-hydroxypropyl)phthalimide 4
(15.0 g, 73.1 mmol) in dry CH2Cl2 under N2 at 0 ꢃC was
added pyridinium dichromate (17.4 g, 80.7 mmol) over 2
min. Cooling bath was removed and mixture was stirred
at room temperature for 18 h. Reaction mixture was
filtered through a dicalite pad and filtrate was collected.
Solvent was evaporated under reduced pressure and the
resulting residue was purified by column chromato-
graphy on silica gel using CH2Cl2/CH3OH (99:1) as
eluent to afford the desired product as a crystalline solid
(8.88 g, 59%). 1H NMR (CDCl3): d 2.89 (t, 2H), 4.04 (t,
2H), 7.65–7.78 (m, 2H), 7.78–7.93 (m, 2H), 9.83 (s, 1H).
1,2,3,4-tetrahydro-2-[(4-methylphenyl)sulfonyl]-1-isoquino-
lineethanamine (11). To a stirring suspension of 1
(5.20 g, 11.3 mmol) in EtOH was added hydrazine
monohydrate (10.1 g, 214 mmol) in one portion, and the
resultant mixture was stirred at reflux for 18 h. The
reaction was cooled, quenched with water (300 mL) and
product was extracted into EtOAc (3ꢂ100 mL).
Extracts were combined, washed with water (50 mL)
and dried (Na2SO4). Solvent was evaporated to obtain
the product (3.62 g, 97%). 1H NMR (DMSO-d6): d
1.57–1.98 (m, 2H), 2.27 (s, 3H), 2.33–2.70 (br m, 4H),
3.34–3.54 (m, 1H), 3.65–3.82 (m, 1H), 4.95–5.10 (m, 1H),
6.89 (d, 1H), 6.98–7.17 (m, 3H), 7.22 (d, 2H), 7.60 (d, 2H).
1-[2-[[(6-methyl-3-pyridinyl)carbonyl]amino]ethyl]-1,2,3,4-
tetrahydro-2-[(4-methylphenyl)sulfonyl]-isoquinoline (19)
and
1-[2-(5,7-dihydro-2-methyl-5,7-dioxo-6H-pyrrolo
[3,4-b]pyridin-6-yl)ethyl]-1,2,3,4-tetrahydro-2-[(4-methyl-
phenyl)sulfonyl]-isoquinoline (20). The crude mixture of
17 and 18 (150 mg) was heated in a Pyrex test tube to
200 ꢃC for 45 min. The resultant glassy solid was pur-
ified by flash chromatography on silica gel using
CH2Cl2/CH3OH (99:1) as eluent, to give the amide 19
1
(38 mg, 17 %). H NMR (CDCl3): d 1.90–2.02 (m, 2H),
2.30 (s, 3H), 2.32–2.50 (m, 2H), 2.63 (s, 3H), 3.34–3.52
(m, 2H) 3.95–4.05 (m, 1H), 4.05–4.15 (m, 1H), 5.05–5.12
(m, 1H), 6.81 (d, 1H), 6.92–7.13 (m, 5H), 7.26 (d, 1H),
7.56 (d, 2H), 7.66 (br t, 1H), 8.12 (d, 1H), 9.10 (s, 1H),
MS 450.4 (MH+). The imide 20 (58 mg, 16%) was also
1-(2-aminoethyl)-3,4-dihydro-2(1H)-isoquinolinecarboxylic
acid-1,1-dimethylethyl ester (15). A mixture of 1010
(35.68 g, 0.116 mol), di-tert-butyl dicarbonate (25.23 g,
0.116 mol), and sodium hydrogen carbonate (28.81 g,
0.343 mol) in CH3OH (600 mL) was placed in an ultra-
sonic bath for 1 h. Insoluble material was filtered off,
and filtrate was evaporated under reduced pressure to
afford the crude intermediate. This was purified by flash
chromatography on silica gel using CH2Cl2/CH3OH
(98:2) followed by CH2Cl2/CH3OH (97:3) as eluent
1
isolated from the chromatography. H NMR (CDCl3):
d 2.10–2.28 (m, 2H), 2.31 (s, 3H), 2.47–2.58 (m, 2H),
2.76 (s, 3H), 3.50–3.68 (m, 1H), 3.85–4.05 (m, 3H),
5.05–5.18 (m, 1H), 6.86 (d, 1H), 7.00–7.30 (m, 6H), 7.44
(d, 1H), 7.59 (d, 2H), 8.03 (d, 1H), MS 475.9 (MH+).
General procedure for the synthesis of amides (21). A
solution of carboxylic acid (4.0 mmol), 1-(3-dimethyl-
aminopropyl)-3-ethyl-carbodiimide hydrochloride (0.92 g,
4.80 mmol) and triethylamine (0.49 g, 4.80 mmol) was
stirred at room temperature for 15 min (for acids 7, 8,
and 11, a further 0.41 g, 4.0 mmol of triethylamine was
added because the carboxylic acid starting material
contained an amine hydrochloride). To this solution
was added 1-hydroxybenzotriazole (0.54 g, 4.0 mmol)
and 15 (1.11 g, 4.0 mmol) and stirring at room tempera-
ture was continued for 18 h. Reaction mixture was
washed with water (2ꢂ10 mL) and brine (10 mL) then
dried (Na2SO4) and concentrated. Purification by flash
chromatography on silica gel using CH2Cl2/CH3OH
(95:5) or CH2Cl2/CH3OH/NH4OH (90:10:0.5) as eluent
afforded the desired amides which were characterised by
1H NMR.
1
(37.2 g, 78%). H NMR (CDCl3): d 1.48 (s, 9H), 2.05–
2.30 (br m, 2H), 2.64–3.12 (br m, 2H), 3.21–3.45 (br m,
1H), 3.64–4.36 (br m, 3H), 5.49–5.41 (m, 1H), 7.15 (m,
4H), 7.70 (m, 2H), 7.82 (m, 2H). The phthalimide
(37.2 g, 0.091 mmol) was then dissolved in EtOH
(400 mL) and hydrazine monohydrate (68.7 g, 1.37 mol)
was added rapidly. The mixture was stirred at room
temperature for 18 h. The resultant thick white pre-
cipitate was quenched with water (2 L) and product was
extracted into EtOAc (3ꢂ500 mL). Extracts were com-
bined, washed with brine (2ꢂ200 mL) and dried
(Na2SO4). Solvent was evaporated to afford 15 (24.7 g,
1
98%). H NMR (CDCl3): d 1.50 (s, 9H), 2.05–2.30 (br
m, 2H), 2.67–2.77 (br m, 1H), 2.85–3.05 (br m, 1H),
3.22–3.42 (br m, 1H), 3.67–4.30 (br m, 3H), 5.13–5.35
(br m, 1H), 7.08–7.22 (br m, 4H), 7.68–7.73 (br m, 2H),
7.80–7.86 (br m, 2H).
General procedure for BOC-deprotection of amides. The
amides 21 were dissolved in CH2Cl2 (20 mL), tri-
fluoroacetic acid (5 mL) was added and reaction was
stirred at room temperature for 1 h. Reaction was
evaporated to dryness and saturated K2CO3 solution
6-methyl-3-[2-[1,2,3,4-tetrahydro-2-[(4-methylphenyl)sul-
fonyl]-1-isoquinolinyl]ethyl]amino]carbonyl]-2-pyridinecar-
boxylic acid (17) and 6-methyl-2-[2-[1,2,3,4-tetrahydro-2
-[(4-methylphenyl)sulfonyl]-1-isoquinolinyl]ethyl]amino]-

2620
D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
(20 mL) was added followed by brine (20 mL). Product
was extracted into EtOAc (4ꢂ20 mL) then combined
extracts were dried (Na2SO4) and solvent was evapo-
rated under reduced pressure to give the secondary
amine. All amines had structures confirmed by 1H
NMR and were used for sulfonamide formation with-
out further purification.
centrated under reduced pressure. Residue was purified
using flash chromatography on silica gel, using CH2Cl2/
CH3OH 98:2 as eluent to afford 24 (60 mg, 43%). H
NMR (CDCl3): d 2.03–2.35 (m, 2H), 2.41 (s, 3H), 2.62–
2.77 (br m, 1H), 2.80–3.05 (m, 1H), 3.40–4.18 (m, 4H),
5.85–6.00 (m, 1H), 6.98–7.45 (m, 8H), 7.64–7.90 (m, 4H).
MS 424.4 (MH+).
1
General procedure for formation of sulfonamides (22).
To solutions of the BOC-deprotected tetra-
hydroisoquinolines (0.07 mmol) in CH2Cl2 (3 mL) were
added the sulfonyl chlorides (0.0875 mmol). PS resin-
bound piperidine (2 mmol gꢀ1 substitution, 100 mg,
0.28 mmol) was added to each reaction, then reactions
were allowed to stand at room temperature (no agita-
tion was required) for 48 h. Aminomethyl PS resin
(1.39 mmol gꢀ1 substitution, 50 mg, 0.07 mmol) was
then added to each reaction, then reactions were
allowed to stand at room temperature for a further 48 h.
Resin was removed by filtration through a fritted tube,
washing resin with CH2Cl2 (2ꢂ2 mL). Filtrates were
concentrated to give the products.
1-[2-(1,3-dihydro-1,3-dioxo-2H-isoindol-2-yl)ethyl]-1,2,3,
4-tetrahydro-2-(phenylmethyl)-isoquinoline (25). A mix-
ture of 1010 (100 mg, 0.33 mmol), benzaldehyde (35 mg,
0.33 mmol), sodium acetate (44 mg, 0.53 mmol) and
glacial acetic acid (0.26 g, 4.30 mmol) was stirred in eth-
anol (2 mL) and cooled in an ice-water bath. Sodium
borohydride (62 mg, 1.63 mmol) was added over 90 min,
then cooling bath was removed and resultant suspen-
sion was stirred at room temperature for 30 min. Reac-
tion mixture was made basic with 5% Na2CO3, and the
product was extracted into CH2Cl2 (2ꢂ5 mL), dried
(Na2SO4) and solvent removed under reduced pressure
to afford the crude 25. Purification using flash chroma-
tography on silica gel, using toluene/EtOAc (9:1) as
eluent afforded the desired product (10 mg, 8%). 1H
NMR (CDCl3): d 2.00–2.09 (m, 1H), 2.14–2.25 (m, 1H),
2.54–2.62 (m, 1H), 2.78–2.85 (m, 1H), 2.86–2.96 (m,
1H), 3.22–3.31 (m, 1H), 3.66–3.87 (m, 4H), 3.89–3.97
(m, 1H), 7.03–7.15 (m, 4H), 7.21–7.35 (m, 3H), 7.38–
7.44 (m, 2H), 7.65–7.70 (m, 2H), 7.77–7.82 (m, 2H). MS
397.2 (MH+).
Modified procedure for formation of N,N-diethybenzyla-
mine sulfonamides (Table 6, sulfonamide No. 26). To a
solution of each of the BOC-deprotected tetra-
hydroisoquinolines (0.07 mmol) in CH2Cl2 (3 mL) was
added 4-bromomethylbenzenesulfonyl chloride (0.0875
mmol). PS resin-bound piperidine (2 mmol gꢀ1 substi-
tution, 100 mg, 0.28 mmol) was added to each reaction,
then the reactions were allowed to stand at room tem-
perature (no agitation was required) for 48 h. Diethyl-
amine (0.35 g, 4.83 mmol) was added to each reaction
vessel, then the reactions were allowed to stand at room
temperature for 48 h. Resin was removed by filtration
through a fritted tube, washing resin with CH2Cl2
(2ꢂ2 mL). Filtrates were concentrated to give the crude
products, which were purified by SPE (500 mg silica,
CH2Cl2/CH3OH/NH4OH (90:10:0.5) as eluent).
1-[2-(1,3-dihydro-1,3-dioxo-2H-isoindol-2-yl)ethyl]-1,2,3,
4-tetrahydro-2-(methylsulfonyl)-isoquinoline (26). Pre-
pared from 1010 (50 mg, 0.16 mmol) and methanesulfo-
nyl chloride (37 mg, 0.32 mmol) using the procedure
1
described for 23 (22 mg, 36%). H NMR (CDCl3): d
2.08–2.26 (m, 2H), 2.76 (s, 3H), 2.77–2.84 (m, 1H),
3.02–3.13 (m, 1H), 3.59–3.69 (m, 1H), 3.87–3.96 (m,
2H), 4.02–4.10 (m, 1H), 4.89–4.95 (m, 1H), 7.10–7.21
(m, 4H), 7.66–7.72 (m, 2H), 7.80–7.86 (m, 2H). GLC
tR=1.06 min (98.5%). MS 385.0 (MH+).
1-[2-(1,3-dihydro-1,3-dioxo-2H-isoindol-2-yl)ethyl]-1,2,3,
4-tetrahydro-2-(phenylsulfonyl)-isoquinoline (23). To a
solution of 1010 (90 mg, 0.29 mmol) in CH2Cl2 (10 mL)
was added benzenesulfonyl chloride (78 mg, 0.44 mmol)
followed by diisopropylethylamine (93 mg, 0.72 mmol).
Solution was stirred at room temperature for 2 h then
concentrated under reduced pressure. Residue was pur-
ified using flash chromatography on silica gel, using
CH2Cl2/CH3OH (98:2) as eluent to afford 23 (85 mg,
1-[2-(acetamido)ethyl]-1,2,3,4-tetrahydro-2-[(4-methyl-
phenyl)sulfonyl]-isoquinoline (28). To 11 (20 mg,
0.06 mmol) was added acetic anhydride (1 mL). The
mixture was stirred at 20 ꢃC for 5 h then evaporated to
dryness, with the aid of a toluene azeotrope. The residue
was taken up in DCM and evaporated once more. The
product was pure according to TLC and NMR. 1H
NMR: (CDCl3; 400 MHz) d 7.57 (d, J=8.3 Hz, 2H),
7.00–7.12 (m, 5H), 6.83 (d, J=7.5 Hz, 1H), 6.70 (m,
1H), 5.05 (dd, 1H), 3.95 (m, 1H), 3.85 (m, 1H), 3.40 (m,
1H), 3.15 (m, 1H), 2.35–2.50 (m, 2H), 2.31 (s, 3H), 2.07
(s, 3H), 2.00–2.15 (m, 1H), 1.85 (m, 1H).
1
66%). H NMR (CDCl3): d 2.03–2.26 (m, 2H), 2.50–
2.61 (m, 2H), 2.59–2.68 (m, 1H), 3.85–4.04 (m, 3H),
5.10–5.17 (m, 1H), 6.81–6.87 (d, 1H), 6.98–7.15 (m, 3H),
7.25–7.34 (m, 2H), 7.38–7.44 (m, 2H), 7.68–7.76 (m,
4H), 7.80–7.86 (m, 2H). MS 447.2 (MH+).
1-Methyl-1,2,3,4-tetrahydro-2-[(4-methylphenyl)sulfonyl]-
isoquinoline (29). Prepared by Method A from 4-
methyl-N-(2-phenethyl)benzenesulfonamide and acetal-
dehyde. Obtained as a colourless gum in 76% yield.
1H NMR (CDCl3) d 1.48 (d, 3H), 2.36 (s, 3H), 2.65
(m, 2H), 3.38 (m, 1H), 3.85 (m, 1H), 5.12 (q, 1H), 6.97
(d, 1H), 7.04 (d, 1H), 7.11 (m, 2H), 7.17 (d, 2H) and
7.67 (d, 2H). GLC tR=6.27 min (98.3%). GC–MS 301
(M+).
1-[2-(1,3-dihydro-1,3-dioxo-2H-isoindol-2-yl)ethyl]-1,2,3,
4-tetrahydro-2-(4-methylbenzoyl)-isoquinoline (24). To a
solution of 1010 (100 mg, 0.33 mmol) in CH2Cl2 (5 mL)
was added triethylamine (41 mg, 0.41 mmol) and the
solution was stirred at room temperature for 1min. p-
toluoyl chloride (51 mg, 0.33 mmol) was added and the
reaction was stirred for 1 h at room temperature then
washed with 5% Na2CO3, dried (Na2SO4) and con-

D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
2621
1-[3-(1,3-Dihydro-1,3-dioxo-2H-isoindol-2yl)propyl]-1,2,3,
4-tetrahydro-2-[(4-methylphenyl) sulfonyl]-isoquinoline (30).
Prepared by Method A from 4-methyl-N-(2-phenethyl)
benzenesulfonamide and 4-phthalimidobutyraldehyde.
quinoline (34). Prepared by Method A from 4-methyl-
N-[2-(4-methoxyphenyl)ethyl]benzenesulfonamide and
3-phthalimidopropionaldehyde. Obtained as a colour-
less gel in 83% yield. 1H NMR (CDCl3) d 2.08 (m, 2H),
2.28 (s, 3H), 2.44 (m, 2H), 3.48 (m, 1H), 3.69 (s+m,
6H), 4.98 (t, 1H), 6.59 (dd, 1H), 6.81 (m, 2H), 7.17 (d,
2H), 7.56 (d, 2H) and 7.84 (m, 4H). GLC tR=7.59 min
(98.2%). MS 491.4 (MH+).
1
Obtained as a colourless gum in 61% yield. H NMR
(CDCl3) d 1.73 (m, 1H) and 1.87 (m, 3H), 2.27 (s, 3H),
2.5 (m, 2H), 3.44 (m, 1H), 3.74 (m, 1H), 3.85 (m, 2H),
5.07 (m, 1H), 6.84 (d, 1H), 7.06 (m, 5H), 7.58 (m, 2H),
7.73 (m, 2H) and 7.86 (m, 2H). GLC tR=3.86 min
(99.7%). MS 475.2 (MH+).
1-[2-(1,3-Dihydro-1,3-dioxo-2H-isoindol-2yl)ethyl]-6-meth-
oxy-1,2,3,4-tetrahydro-2-[(4-methylphenyl)sulfonyl]-iso-
quinoline (35). Prepared by Method A from 4-methyl-
N-[2-(3-methoxyphenyl)ethyl]benzenesulfonamide3 and
3-phthalimidopropionaldehyde. Obtained as a colour-
less gel in 81% yield. 1H NMR (CDCl3) d 2.06 (m, 1H),
2.17 (m, 1H), 2.31 (s, 3H), 2.52 (m, 2H), 3.57 (m, 1H),
3.69 (s, 3H), 3.88 (m, 3H), 5.07 (dd, 1H), 6.36 (d, 1H),
6.65 (dd, 1H), 7.02 (d, 1H), 7.09 (d, 2H), 7.58 (d, 2H),
7.69 (m, 2H) and 7.81 (m, 2H). GLC tR=12.41 min
(94.1%). MS 491.4 (MH+).
1-[(1,3-Dihydro-1,3-dioxo-2H-isoindol-2yl)methyl]-1,2,3,
4-tetrahydro-2-[(4-methylphenyl) sulfonyl]-isoquinoline (31).
Prepared by Method A from 4-methyl-N-(2-phenethyl)
benzenesulfonamide and 2-phthalimidoacetaldehyde.
1
Obtained as a colourless solid in 81% yield. H NMR
(CDCl3) d 2.19 (s, 3H), 2.68 (m, 2H), 3.76 (m, 2H), 3.96
(m, 1H), 4.07 (m, 1H), 5.28 (dd, 1H), 6.93 (d, 2H), 7.03
(d, 1H), 7.18 (m, 2H), 7.27 (d, 2H), 7.71 (m, 2H), and
7.84 (m, 2H). GLC tR=6.21 min (98.8%). MS 447.4
(MH+).
1-[2-(1,3-Dihydro-1,3-dioxo-2H-isoindol-2yl)ethyl]-6,7-di-
hydroxy-1,2,3,4-tetrahydro-2-[(4-methylphenyl)sulfonyl]-
isoquinoline (36). An intermediate in synthesis of 27,
1,2,3,4-tetrahydro-6,7-dimethoxy-2[(4-methylphenyl)-
sulfonyl]-1-(2-phthalimidoethyl)isoquinoline, was pre-
1-[2-(1,3-Dihydro-1,3-dioxo-2H-isoindol-2yl)ethyl]-7-hy-
droxy-1,2,3,4-tetrahydro-2-[(4-methylphenyl)sulfonyl]-
isoquinoline (32). 4-Methyl benzensulphonyl chloride
(2.77 g, 14.53 mmol) was added into a solution of tyr-
amine (2.0 g, 14.58 mmol) in pyridine (40 mL) and the
resulting mixture was stirred over 16 h at room tem-
perature. The reaction mixture was diluted with water
(50 mL) and the product extracted with DCM
(2ꢂ50 mL). The organic layer was washed with 2 N HCl
(150 mL), water (100 mL), saturated aqueous NaHCO3
(100 mL) and brine (100 mL). The organic phase was
dried with MgSO4 and concentrated under reduced
pressure to obtain crude 4-methyl-N-[2-(4-hydro-
xyphenyl)ethyl]benzenesulfonamide which was purified
by recrystallization from hot ethanol (20 mL). Yield:
pared by Method
A
from 4-methyl-N-[2-(3,4-
dimethoxyphenyl)ethyl]benzenesulfonamide and 3-
phthalimidopropionaldehyde. Obtained as a colourless
gel in 52% yield. 1H NMR (CDCl3) d 2.05 (m, 1H), 2.17
(m, 1H), 2.32 (s, 3H), 2.41 (m, 2H), 3.57 (m, 1H), 3.74
(s, 3H), 3.86 (s, 3H), 3.91 (m, 3H), 5.03 (dd, 1H), 6.31 (s,
1H), 6.62 (s, 1H), 7.09 (d, 2H), 7.59 (d, 2H), 7.71 (m, 2H)
and 7.82 (m, 2H). MS 521.2 (MH+). A solution of 1,2,3,4-
tetrahydro-6,7-dimethoxy-2-[(4-methylphenyl)sulfonyl]-1-
(2-phthalimidoethyl) isoquinoline (128 mg, 0.245 mmol)
in DCM (5 mL) was added dropwise into a solution of
BBr3 (0.185 mL, 1.96 mmol) in DCM (10 mL) under
nitrogen at ꢀ5 ꢃC. After 2 h the reaction mixture was
allowed to warm up and stirred for another 16 h at
room temperature. The reaction was quenched with
water (20 mL) and extracted with ether (3ꢂ150 mL).
The combined organic layers were concentrated under
reduced pressure and the crude product was purified by
flash chromatography on silica gel eluting with EtOAc/
heptane (1:1), to obtain 37 as a colourless gum. Yield:
1
3.98 g (94%). H NMR (CDCl3) d 2.42 (s, 3H), 2.62 (t,
2H), 3.09 (t+bs, 3H), 6.74 (d, 2H), 6.92 (d, 2H), 7.31 (d,
2H) and 7.68 (d, 2H). MS 291 (MH+). Compound 32
was prepared by Method A from 4-methyl-N-[2-(4-
hydroxyphenyl)ethyl]benzenesulfonamide and 3-phtha-
limidopropionaldehyde. Obtained as a colourless gel in
1
61% yield. H NMR (CDCl3) d 2.08 (m, 1H), 2.16 (m,
1H), 2.31 (s, 3H), 2.44 (m, 2H), 3.56 (m, 1H), 3.89 (m,
3H), 5.02 (dd, 1H), 5.36 (s, 1H), 6.55 (dd, 1H), 6.65 (d,
1H), 6.71 (d, 1H), 7.09 (d, 2H), 7.58 (d, 2H), 7.68 (m, 2
H), and 7.81 (m, 2H). LC–MS purity, 93.4%. MS 477.2
(MH+).
1
14 mg (12%). H NMR (CDCl3) d 2.05 (m, 1H), 2.12
(m, 1H), 2.32 (s, 3H), 2.37 (m, 2H), 3.49 (m, 1H), 3.88
(m, 3H), 4.94 (dd, 1H), 5.58 (bs, 1H), 6.02 (bs, 1H), 6.37
(s, 1H), 6.64 (s, 1H), 7.1 (d, 2H), 7.58 (d, 2H), 7.68 (m,
2H) and 7.83 (m, 2H). MS 493.4 (MH+).
1-[2-(1,3-Dihydro-1,3-dioxo-2H-isoindol-2yl)ethyl]-5-meth-
oxy-1,2,3,4-tetrahydro-2-[(4-methylphenyl)sulfonyl]-iso-
quinoline (33). Prepared by Method A from 4-methyl-
N-[2-(2-methoxyphenyl)ethyl]benzenesulfonamide and
3-phthalimidopropionaldehyde. Obtained as a colour-
less gel in 82% yield.1H NMR (CDCl3) d 2.12 (m, 2H),
2.29 (s+m, 5H), 3.55 (m, 1H), 3.68 (s, 3H), 3.88 (m, 2H),
4.02 (m, 1H), 4.98 (m, 1H), 6.59 (dd, 1H), 6.74 (dd, 1H),
7.08 (m, 3H), 7.59 (d, 2H), 7.69 (m, 2H) and 7.79 (m,
2H). GLC tR=16.43min (88.9%). MS 491.4 (MH+).
1-[2-(1,3-Dihydro-1,3-dioxo-2H-isoindol-2yl)ethyl]-7-
methyl-1,2,3,4-tetrahydro-2-[(4-methylphenyl)sulfonyl]-
isoquinoline (37). Prepared by Method A from 4-
methyl-N-[2-(4-methylphenyl)ethyl]benzenesulfonamide
and 3-phthalimidopropionaldehyde. Obtained as a
1
colourless solid in 85% yield. H NMR (CDCl3) d 2.13
(m, 2H), 2.24 (s, 3H) 2.31 (s, 3H), 2.49 (m, 2H), 3.57 (m,
1H), 3.94 (m, 3H), 5.06 (dd, 1H), 6.76 (d, 1H), 6.86 (d,
1H), 6.93 (s, 1H), 7.09 (d, 2H), 7.61 (d, 2H), 7.69 (m,
2H) and 7.81 (m, 2H). GLC tR=9.15min (98.4%). MS
475.2 (MH+).
1-[2-(1,3-Dihydro-1,3-dioxo-2H-isoindol-2yl)ethyl]-7-meth-
oxy-1,2,3,4-tetrahydro-2-[(4-methylphenyl)sulfonyl]-iso-

2622
D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
1-[2-(1,3-Dihydro-1,3-dioxo-2H-isoindol-2yl)ethyl]-7-[(4-
benzonitrile)oxymethyl]-1,2,3,4-tetrahydro-2-[(4-methyl-
phenyl)sulfonyl]-isoquinoline (38). 4-Bromomethyl ben-
zonitrile (106 mg, 0.54 mmol) and K2CO3 (76 mg,
0.55 mmol) were added into a solution of phenol 32
(258 mg, 0.54 mmol) in DMF (20 mL) and the resulting
mixture heated at 80 ꢃC for 1h. The reaction mixture
was cooled to room temperature, diluted with toluene
and the resulting solution concetrated under reduced
pressure. The crude product was purified by flash chro-
matography on silica gel eluting with EtOAc/heptane
(1:1), to obtain compound 38 as a colourless gum.
Yield: 150 mg (47%). 1H NMR (CDCl3) d 2.06 (m, 1H),
2.15 (m, 1H), 2.32 (s, 3H), 2.47 (m, 2H), 3.54 (m, 1H),
3.88 (m, 3H), 5.08 (s+m, 3H), 6.68 (dd, 1H), 6.76 (d,
1H), 6.79 (d, 1H), 7.07 (d, 2H), 7.53 (d, 2H), 7.59 (d,
2H), 7.72 (m, 4H) and 7.84 (m, 2H). GLC tR=9.01 min
(98.9%). MS 592.4 (MH+).
ethanol (1 mL) and toluene (9 mL). After heating at
reflux for 3 h, the reaction mixture was stirred at 20 ꢃC
for 18 h, then evaporated to dryness under reduced
pressure. After purification by SiO2 flash chromato-
graphy (CH3OH/DCM; (1:4)), the tetrahydroisoquino-
line was obtained in 36% yield (130 mg). To a
suspension of the tetrahydroisoquinoline (120 mg,
0.405 mmol) in anhydrous pyridine (1 mL), was added
p-toluenesulphonyl chloride (77 mg, 0.405 mmol) in
anhydrous pyridine (0.5 mL). The reaction mixture was
stirred at 20 ꢃC for 3 h, then at 80 ꢃC for 18 h. Volatiles
were removed under reduced pressure and the residue
was chromatographed on flash SiO2, eluting with 5%
CH3OH in DCM. The product 40 was obtained as a
clear gum in 36% yield (59 mg). 1H NMR: (CDCl3;
400 MHz) d 7.90 (m, 2H), 7.75 (m, 2H), 7.40–7.60 (m,
3H), 7.13 (d, J=7.4Hz, 2H), 4.95 (m, 1H), 4.15 (m, 1H),
3.85 (m, 2H), 3.45 (m, 1H), 2.40 (m, 2H), 2.34 (s, 3H),
2.15 (m, 2H). MS (ES+) m/z 452 (M + H)+.
1-[(4-methylphenyl)sulfonyl]-2-(2-phthalimidoethyl)piperi-
dine
(39).
4-Methyl
sulfonylchloride
(4.42 g,
2-(dimethylamino)-N-[[[2-[1,2,3,4-tetrahydro-2-[(4-butyl-
phenyl)sulfonyl] - 1- isoquinolinyl]ethyl]amino]carbonyl] -
acetamide (43). The crude library product 22{23,2}
(25 mg) was purified by SPE (1 g silica cartridge,
CH2Cl2/CH3OH 98:2 followed by CH2Cl2/CH3OH 95:5
23.21 mmol) was added in small portions into a solution
of 2-piperidineethanol (1 g, 7.73 mmol) and pyridine
(2.5 mL, 30.9 mmol) in chloroform (40 mL) at 0 ꢃC. The
resulting solution was allowed to warm up to room
temperature and stirred for 16 h. The reaction mixture
was concentrated under reduced pressure and the
residue reconstituted in diethyl ether (100 mL). The
organic phase was washed with aqueous 2 N HCl
(50 mL), 5% NaHCO3 (50 mL) and water (50 mL). The
organic phase was dried with MgSO4 and the solvent
removed under reduced pressure to obtain a brown oil,
which after purification by flash chromatography on
silica gel eluting with 3:2 Et2O/heptane, afforded N,O-
bis(4-methyl benzenesulfonyl)-2-piperidineethanol as a
1
as eluent) to give (43) (20 mg, 0.044 mmol). H NMR
(CDCl3): d 0.87 (t, 3H), 1.20 (m, 2H), 1.47 (m, 2H), 1.95
(m, 1H), 2.07 (m, 1H), 2.36 (s, 6H), 2.45 (m, 2H), 2.55
(t, 2H), 3.01 (s, 2H), 3.30 (m, 1H), 3.45 (m, 1H), 3.72
(m, 1H), 3.95 (m, 1H), 5.00 (m, 1H), 6.80 (m, 1H), 7.00
(m, 2H), 7.09 (m, 3H), 7.59 (d, 2H), 7.93 (br t, 1H).
HPLC tR=3.78 min (95.0%). MS 458.0 (MH+).
1-[2-[[(6-methyl-3-pyridinyl)carbonyl]amino]ethyl]-1,2,3,4-
tetrahydro-2-[(4-butylphenyl)sulfonyl]-isoquinoline (44).
The crude library product 22 {23,4} (23 mg) was pur-
ified by SPE (1g silica cartridge, CH2Cl2/CH3OH (98:2)
followed by CH2Cl2/CH3OH (95:5) as eluent) to give 44
1
colourless solid. Yield: 1.64 g (48%). H NMR (CDCl3)
d 1.13, 1.41 and 1.74 (m, 6H), 1.78 (m, 1H), 2.07 (m,
1H), 2.42 (s, 3H), 2.46 (s, 3H), 2.92 (t, 1H), 3.78 (dd,
1H), 4.05 (m, 3H), 7.26 (d, 2H), 7.35 (d, 2H), 7.67 (d,
2H) and 7.80 (d, 2H). GLC tR=8.35 min (85%).
Potassium phthalimide (1.74 g, 9.4 mmol) was added
into a solution of N,O-bis(4-toluenesulfonyl)-2-piper-
idineethanol (3.16 g, 7.23 mmol) in anhydrous DMF
(75 mL) under vigorous stirring, at room temperature.
The reaction mixture was heated at 80 ꢃC for 3 h, cooled
down to room temperature and stirred over 16 h. The
reaction was quenched with water (100 mL) and the
product extracted with DCM (2ꢂ50 mL). The combined
DCM extract was washed with brine (50 mL), 0.2N
NaOH (50 mL) and brine again (50 mL). The organic
phase was dried over NaSO4 and concentrated under
reduced pressure to obtain crude 39 which was purified
by recrystallization from DCM/Et2O (1:1). Yield: 2.6 g
(87%). 1H NMR (CDCl3) d 1.21 and 1.51 (m, 6H), 1.68
(m, 1H), 2.04 (m, 1H), 2.41 (s, 3H), 3.09 (t, 1H), 3.67 (m,
2H), 3.85 (dd, 1H), 4.12 (m, 1H), 7.26 (d, 2H), 7.72 (m,
4H) and 7.83 (m, 2H). GLC tR=8.25 min (99.7%). MS
413.2 (MH+). GC–MS 238 (M+-199 (TsOCH2CH2)).
1
(20 mg, 0.041 mmol). H NMR (CDCl3): d 0.87 (t, 3H),
1.20 (m, 2H), 1.49 (m, 2H), 1.95 (m, 1H), 2.20 (m, 1H),
2.33–2.47 (m, 2H), 2.54 (m, 2H), 2.63 (s, 3H), 3.42 (m,
2H), 4.03 (m, 1H), 4.12 (m, 1H), 5.08 (m, 1H), 6.80 (m,
1H), 6.94–7.10 (m, 5H), 7.27 (m, 1H), 7.57 (m, 2H), 7.67
(br t, 1H), 8.14 (m, 1H), 9.10 (m, 1H). HPLC tR=3.96
min (93.6%). MS 492.5 (MH+).
N-[[[2-[1,2,3,4-tetrahydro-2-[(4-butylphenyl)sulfonyl]-1-
isoquinolinyl]ethyl]amino]carbonyl] - 3 - pyridineacetamide
(45). The crude library product 22{23,9} (19 mg) was
purified by SPE (1g silica cartridge, CH2Cl2/CH3OH
(98:2) followed by CH2Cl2/CH3OH (95:5) as eluent) to
1
give 45 (14 mg, 0.028 mmol). H NMR (CDCl3): d 0.87
(t, 3H), 1.20 (m, 2H), 1.48 (m, 2H), 1.85 (m, 1H), 1.98
(m, 1H), 2.28–2.45 (m, 2H), 2.55 (m, 2H), 3.20 (m, 1H),
3.33 (m, 1H), 3.63 (s, 2H), 3.83 (m, 1H), 3.93 (m, 1H), 4.79
(m, 1H), 6.75–6.90 (m, 3H), 6.98–7.12 (m, 4H), 7.32 (m,
1H), 7.49 (m, 2H), 7.80 (m, 1H), 8.55 (m, 1H), 8.66 (m,
1H). HPLC tR=3.83 min (94.0%). MS 492.5 (MH+).
4-[2-(1,3-Dihydro-1,3-dioxo-2H-isoindol-2-yl)ethyl]-6-[(4-
methylphenyl) sulphonyl] -4,5,6,7-tetrahydro-1H-imi-
dazo[4,5-c]pyridine (40). Histamine (111 mg, 1 mmol)
and 5 (244 mg, 1.2 mmol) were dissolved in a mixture of
N-[[[2-[1,2,3,4-tetrahydro-2-[(4-butylphenyl)sulfonyl]-1-
isoquinolinyl]ethyl]amino]carbonyl]-1H-imidazole-1-acet-
amide (46). The crude library product 22{23,12} (21
mg) was purified by SPE (1g silica cartridge, CH2Cl2/

D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
2623
CH3OH (98:2) followed by CH2Cl2/CH3OH/NH4OH
(90:10:0.5) as eluent) to give 46 (17mg, 0.035 mmol).
1H NMR (CDCl3): d 0.87 (t, 3H), 1.20 (m, 2H), 1.48
(m, 2H), 1.73–2.02 (m, 2H), 2.30–2.45 (m, 2H), 2.55
(m, 2H), 3.20–3.40 (m, 2H), 3.82 (m, 1H), 3.92 (m, 1H),
4.71 (s, 2H), 4.78 (m, 1H), 6.80 (m, 2H), 6.91 (m, 1H),
7.02 (m, 1H), 7.10 (m, 4H), 7.20 (s, 1H), 7.52 (m, 2H),
7.64 (m, 1H). HPLC tR=3.74 min (95.0%). MS 481.5
(MH+).
paration and incubated at 23 ꢃC for 3 h. Assays were
terminated by rapid filtration through Whatman GF/B,
which had been pre-soaked in ice-cold 0.5%(v/v) poly-
ethyleneimine for 1 h, using a Brandel M24R cell har-
vester (Brandel Gaithersburg, MD). Filters were
washed three times with ice-cold Milli-Q water, dried
and placed in 24-well plates with 1mL of Ultima Gold
scintillation fluid and counted for tritium using a
Microbeta (Wallac).
ꢀ Opioid binding assays. Briefly, crude homogenates
from 0.3ꢂ106 Chinese Hamster Ovary cells (CHO) cells
stably expressing the human delta opioid receptor
cDNA were incubated with 0.15 nM [3H] Naltrindole
(50, 70–3H], 50 mM Tris–HCl buffer (pH 7.4), 5 mM
MgCl2 in a final volume of 1.5 mL. Non-specific binding
was determined in the presence of 10 mM naltrexone.
Assays were initiated by the addition of the mem-
brane preparation and incubated at 25 ꢃC for 3 h.
Assays were terminated by rapid filtration through
Whatman GF/B Packard 96-well plates which had
been pre-soaked in ice-cold 0.5%(v/v) polyethyl-
eneimine for 1 h, using a Packard cell harvester.
Filters were washed three times with ice-cold 50 mM
Tris–HCl (pH 7.4), dried. Following the addition of
50 ml of MicroScint to each well, the plates were sealed
and counted for tritium using a Packard TopCount
Scintillation Counter.
Receptor binding assays
Membrane preparation. Rat brains with cerebellum
removed, from male Wistar rats were homogenised in 5
vol (w/v) of ice-cold 50 mM Tris–HCl buffer (pH 7.4)
using a Potter-Elvehejm homogeniser (six passes at
1000 rpm). The homogenate was centrifuged at 45,000 g
for 15 min at 4 ꢃC. The pellet was resuspended by
homogenisation in a further 5 vol of ice-cold buffer. To
remove endogenous opioids that may interfere with the
radioligand binding assay, the membranes were incu-
bated at 25 ꢃC for 30 min in a shaking water bath, and
re-centrifuged as described above. The membranes were
resuspended in 10 vol of ice-cold buffer (based on the
original weight of tissue), snap frozen and stored at
ꢀ80 ꢃC. For the preparation of guinea pig brain mem-
branes, six guinea pig cortex were homogenised in 5 vol
(w/v) in ice-cold 50mM Tris–HCl, pH 7.4. The homo-
genate was allowed to settle for 30 min on ice and
the larger particles re-homogenised in 10 mL Tris–HCl.
The combined homogenates were centrifuged at 45,000
g for 15 min at 4 ꢃC. The supernatant was discarded
and the pellet resuspended by homogenisation in 10 vol
(w/v) Tris–HCl, pH 7.4. The homogenate was incu-
bated at 25 ꢃC for 30 min in a shaking waterbath
and re-centrifuged, as above. The pellet was resus-
pended in a final volume of 130 mL ice-cold Tris–HCl,
snap frozen in liquid nitrogen and stored as aliquots at
ꢀ80 ꢃC.
Data analysis. IC50 values from competitive binding
data were determined by computer analysis using Prism,
whereby a non-linear repression analysis using four-
parameter logistic equation was utilised, assuming
a single set of binding sites. A curve was fitted to a
graph of binding (% total binding) values plotted
against the log values of the corresponding drug
concentrations.
In vivo assay
ꢁ Opioid binding assay. Briefly, 300 mg rat brain mem-
brane protein was incubated with 1.5 nM [3H] DAMGO
[Tyrosyl-3,5-3H(N)], 50 mM Tris–HCl buffer (pH 7.4),
0.1mg mLꢀ1 bacitracin in a final vol of 1 mL. Non-
specific binding was determined in the presence of
10 mM naltrexone. Assays were initiated by the addition
of the membrane preparation and incubated at 23 ꢃC
for 5 h. Assays were terminated by rapid filtration
through Whatman GF/B, which had been pre-soaked in
ice-cold 0.5%(v/v) polyethyleneimine for 1 h, using a
Brandel M24R cell harvester (Brandel Gaithersburg,
MD). Filters were washed three times with ice-cold
Milli-Q water, dried and placed in 24-well plates with
1 mL of Ultima Gold scintillation fluid and counted
for tritium using a Microbeta Scintillation Counter
(Wallac).
Formalin paw test (FPT). In this test, the injection of a
dilute solution of formalin into one hind paw of the
mouse induces an inflammatory response in the paw,
which may provide a more valid model for clinical pain
than acute threshold tests using a phasic, high intensity
stimulus e.g., hot plate test. The response to sub-plantar
formalin injection shows two distinct phases of noci-
ceptive behaviour in several species.13 The early phase
(Phase 1) begins immediately after formalin injection
and lasts for 4–5 min. This phase is followed by a period
of 10–15 min of quiescent behaviour, after which a sec-
ond, late phase (Phase 2) occurs. This phase starts ꢁ20
min post-formalin injection and continues for a further
20–30 min. In mice, recording manually the time spent
licking or biting the injected paw is the most common
method of behavioural assessment.14 Agonists at m, k
and d opioid receptors inhibit both phases of the
formalin response.
ꢂ Opioid binding assay. Briefly, 250 mg guinea pig brain
membrane protein was incubated with 1.5 nM [3H]
U69593 [Phenyl-3,4-3H], 50 mM Tris–HCl buffer (pH
7.4) in a final vol of 1 mL. Non-specific binding was
determined in the presence of 10 mM naltrexone. Assays
were initiated by the addition of the membrane pre-
Male ICR mice (22–30 g) were habituated to their test
environment by placing them, singly, into clear perspex
observation boxes for ꢁ2 h on the day prior to the

2624
D. R. Barn et al. / Bioorg. Med. Chem. 9 (2001) 2609–2624
experiment and for 1 h prior to drug administration on
the day of the experiment. Formalin solution, 0.3% in
sterile saline, was prepared as a fresh solution daily.
Due to the limited aqueous solubility of some of the
compounds, test compounds or vehicle were dissolved
in 10% cyclodextrin in water and administered intrave-
nously (iv), 10 mL kgꢀ1, 15 min prior to the subplantar
injection into one hind paw of 20 ml of formalin
solution. The total time in seconds that each animal
spent licking its injected paw was recorded in 5-min
epochs for 40 min, beginning immediately after formalin
injection. Mean and SEM values were calculated at each
time point for each treatment group. Area under the
curve was calculated in seconds between 0 and 5 min
after formalin injection for Phase 1 (early phase) and
between 15 and 40 min after formalin injection for
Phase 2 (late phase). Test compound-induced inhibition
of licking was calculated as follows:
Acknowledgements
The authors thank the analytical department at
Organon Newhouse for providing NMR, IR, MS and
HPLC data in support of this work.
References
1. Rapaka, R. S.; Porreca, F. Pharmacol. Res. 1991, 8, 1.
2. Dondio, G.; Clarke, G.D.; Giardina, G.; Petrillo, P.; Pet-
rone, G.; Ronzoni, S.; Visentin, L.; Vecchietti, V. Analgesia,
1995, 1, 394
3. Portoghese, P. S.; Sultana, M.; Takemori, A. E. J. Med.
Chem. 1990, 43, 1714.
4. Salvadori, S.; Balboni, G.; Guervini, R.; Tomatis, R.;
Bianchi, C.; Bryant, S. D.; Cooper, P. S.; Lazarus, L. H. Med.
Chem. 1997, 40, 3100.
5. Suzuki, T.; Tsuji, M.; Mori, T.; Misawa, M.; Endoh, T.;
Nagase, H. Life Sci. 1995, 57, 155.
6. Calderon, S. N.; Rothman, R. B.; Porreca, F.; Flippen-
Anderson, J. L.; McNutt, R. W.; Xu, H.; Smith, L. E.;
Bilsky, E. J.; Davis, P.; Rice, K. C. J. Med. Chem. 1994, 37,
2125.
7. Ronai, A. Z.; Botyanszki, J.; Hepp, J.; Medzihradszky, K.
Life Sciences 1992, 50, 1371.
8. Barn, D. R.; Morphy, J. R. J. Comb. Chem. 1999, 1, 151.
9. Schiller, P. W.; Berezowska, I.; Nguyen, T. M.-D.;
Schmidt, R.; Lemieux, C.; Chung, N. N.; Falcone-Hindley, M.
L.; Yao, W.; Liu, J.; Iwama, S.; Smith, A. B.; Hirschmann, R.
J. Med. Chem. 2000, 43, 551.
% Inhibition of Phase 1 ð0À5 min after formalinÞ
¼ 100%
Mean time ðsÞ spent licking in test compound-treated group
À
Mean time ðsÞ spent licking in control group
ꢀ 100%
% Inhibition of Phase 2 ð15À40 min after formalinÞ
¼ 100%
10. Hall, I. H.; Wong, O. T.; Futch, G.; Scovill, J. P. Biomed.
Biochim. Acta 1990, 49, 103.
11. Barnes, R. A.; Godfrey, J. C. J. Org. Chem. 1957, 22, 10431.
12. The Lipinski rules for drug-like activity, which were pub-
lished during the course of this work, suggest that good per-
meability is less likely if molecular weight is over 500 and that
good solubility and permeability is less likely if LogP is over 5:
Lipinski, C. A. Adv. Drug Deliv. Rev. 1997, 23, 3
13. Dubuisson, D.; Dennis, G. Pain 1977, 4, 161.
14. Hunskaar, S.; Fasmer, O. B.; Hole, K. J. Neurosci. Meth
1985, 14, 69.
15. Knapp, R. J.; Santoro, G.; De leon, I. A.; Lee, K. B.;
Edsall, S. A.; Waite, S.; Malatynska, E.; Calderon, S. N.; Rice,
K. C.; Rothman, R. B.Porreca F; Roeske W. R.; Yamamura
H. I. J. Pharm. Expt. Ther. 1996, 277, 1284.
Mean time ðsÞ spent licking in test compound-treated group
À
Mean time ðsÞ spent licking in control group
ꢀ 100%
For tests in which Naltrindole was adminstered in
addition to the test compound, Naltrindole (10 mg kgꢀ1
i.p.) was given 35 min before administration of the test
compound and again immediately prior to the test
compound. Inhibition of licking behaviour induced by
the mu agonist fentanyl (3 nmol kg-1; i.v.) in the FPT
was shown not to be significantly antagonised by this
dose of Naltrindole.
16. Nagase, H.; Wakita, H.; Kawai, K.; Endoh, T.; Matsura,
H.; Tanaka, C.; Takezawa, Y. Jpn. J. Pharmacol. 1994,
64 (Suppl 1), 35.