Evaluation of a series of anticonvulsant 1,2,3,4-tetrahydroisoquinolinyl-benzamides

Article abstract of DOI:10.1016/S0968-0896(00)00149-8

SAR studies around a series of N-(tetrahydroisoquinolinyl)-2-methoxybenzamides, identified by high-throughput screening at the novel SB-204269 binding site, have provided compounds such as 13, 29-33 with high affinity and excellent anticonvulsant activity in animal models. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00149-8

Bioorganic & Medicinal Chemistry 8 (2000) 2085±2094
Evaluation of a Series of Anticonvulsant 1,2,3,4-
Tetrahydroisoquinolinyl-benzamides
Wai N. Chan, Michael S. Hadley, John D. Harling, Hugh J. Herdon, Barry S. Orlek,
Graham J. Riley, Rachel E. A. Stead, Tania O. Stean, Mervyn Thompson,*
Neil Upton and Robert W. Ward
SmithKline Beecham Pharmaceuticals, New Frontiers Science Park, Third Avenue, Harlow, Essex, CM19 5AW, UK
Received 10 November 1999; accepted 18 May 2000
AbstractÐSAR studies around a series of N-(tetrahydroisoquinolinyl)-2-methoxybenzamides, identi®ed by high-throughput
screening at the novel SB-204269 binding site, have provided compounds such as 13, 29±33 with high anity and excellent anti-
convulsant activity in animal models. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
demonstrated4 6 that trans 4S benzamides of this type
bind enantioselectively at a novel unique site in various
regions (e.g., forebrain, hippocampus, cerebellum, cere-
bral cortex) of the brains of several animal species
(mouse, cat, dog, marmoset) and man with similar high
anity for the [3H] SB-204269 site. It was also shown
that there was a good correlation between in vitro
potency and the level of in vivo activity in antic-
onvulsant models (e.g., rodent MEST and MES (max-
imal electroshock seizure) tests).6
Epilepsy is one of the most serious neurological condi-
tions for which there is a clear major unmet medical need
for compounds with enhanced ecacy and safety.1,2 For
example, around 20±30% of su€erers do not have their
seizures fully controlled with currently available drugs
(many of which also have problems with side e€ects)
and the prognosis for some childhood epileptic syn-
dromes and partial seizures in adults is poor. It was
against this background that we embarked upon our
studies with novel benzopyrans in an attempt to identify
new improved anticonvulsant agents.
In an earlier publication,3 we reported that novel trans 4S
benzamido-3,4-dihydro-2H-benzopyrans showed good
anticonvulsant activity in the mouse maximal electro-
shock seizure threshold (MEST) model. Subsequent
exploration of structure±activity relationships (SAR) led
to the identi®cation of the novel 4-¯uorobenzamide 1 SB-
204269 with potent anticonvulsant activity in a range of
seizure models and an overall pro®le indicative of poten-
tial utility for the treatment of partial and generalised
tonic-clonic seizures.4
The compound is currently undergoing clinical evaluation
as a pioneer treatment for epilepsy disorders and has
progressed to Phase II of clinical development. It was
We wished to identify alternative structural classes to fully
exploit this novel CNS binding site and capitalise on the
superior anticonvulsant pro®le a€orded by its modula-
tion. High-throughput screening of the SB compound
bank in the [3H] SB-204269 assay in rat forebrain6
*Corresponding author. Fax: 01279 622790; e-mail: mervyn_thomp-
son@sbphrd.com
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00149-8

2086
W. N. Chan et al. / Bioorg. Med. Chem. 8 (2000) 2085±2094
identi®ed an interesting high anity 5-benzamido-tet-
rahydroisoquinoline (THIQ) (2) which was actually 3-
fold more potent than SB-204269 itself. Preliminary
SAR studies revealed that attachment of the benzamide
moiety at the 5- or 7-positions was much preferred over
attachment at the 6- or 8- positions of the THIQ
nucleus.7 Early molecular modelling overlap studies
using SYBYL were used to rationalise the equivalent
anities of the 5- and 7-substituted isomers (see Fig. 1).
To further exploit these THIQ leads and optimise the in
vivo pro®le, a full SAR analysis was carried out and the
results of this work (see Tables 1 and 2) are reported
Figure 1. Overlapped energy minimised conformations of 6 (green) and 28 (yellow) generated using the SYBYL package (O=red, N=blue).
Table 1. Biological data for 5-substituted compounds
Compound
R
X
[3H]SB-204269
bindinga pKi
Mouse MESTb % increase in
seizure threshold at
10 mg/kg po 1 h post-dose
1
2
3
4
5
6
7
8
SB-204269
2-OMe, 4-NH2 5-Cl
2-OMe
3-Cl
4-But
2-OMe, 4-But
2-OMe, 5-Cl
2-OMe, 3,5-diCl
2,4-diOMe
2,4-diOMe, 6-Me
2,4-diOMe, 5-Cl
2-OMe, 4-OEt, 5-Cl
2-OMe, 4-OPri, 5-Cl
2-OMe, 4-But, 5-Cl
2-OMe, 4-Pri, 5-CF3
2,4-diOMe, 5-Cl
2,4-diOMe, 5-Cl
2,4-diOMe, 5-Cl
2,4-diOMe, 5-Cl
2-OMe, 4-But
2-OMe, 4-But
2-OMe, 4-But
2-OMe, 4-But
Ð
Ð
NMe
NMe
NMe
NMe
NMe
NMe
NMe
NMe
NMe
NMe
NMe
NMe
NMe
NMe
NH
7.3
7.8
5.8
<5.0
6.0
7.7
7.0
5.0
6.5
5.0
7.6
8.0
8.3
6.8
7.7
7.6
102***
25*
Nd
Nd
Nd
140***
20*
Nd
Nd
Nd
63***
95***
120***
Nd
128***
34*
Nd
Nd
Nd
52**
Nd
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
NEt
6.7
NCH2Ph
NCO2But
NH
NCO2Me
NSO2Me
CH2
<6.0
<5.9
7.5
6.2
6.4
4ns
Nd
Nd
<5.8
6.1
Ð
aProcedures as detailed in refs 5 and 6; all determinations were carried out in triplicate, SEM <Æ0.05.
bProcedures as detailed in ref 12; *P<0.05, **P<0.01, ***P<0.001, compared to vehicle-treated controls according to analysis following Lich®eld
and Wilcoxon in ref 17; doses refer to free base. ns: not signi®cant Nd: not determined.

W. N. Chan et al. / Bioorg. Med. Chem. 8 (2000) 2085±2094
Table 2. Biological data for 7-substituted THIQ benzamides
2087
Compound
R1
R2
[3H]SB-204269
bindinga pKi
Mouse MESTb % increase in
seizure threshold at
10 mg/kg po 1 h post-dose
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
H
3-Cl
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
H
6.6
7.5
8.1
7.6
8.6
8.7
8.3
8.8
8.9
8.3
7.3
7.9
7.5
8.9
8.5
7.7
8ns
24***
37***
130***
128***
195***
192 ***
163***
Nd
62**
Nd
72***
Nd
Nd
4-But
2-OMe, 4-But
3-Cl, 4-OMe
3-Cl, 4-OEt
3-Cl, 4-OPri
3-Br, 4-OMe
3-CF3, 4-OMe
3-CN, 4-OMe
3-OMe, 4-Cl
2,4-diOMe, 5-Br
3,5-diCl, 4-OMe
3-CF3, 4-OMe
3-CF3, 4-OMe
3-CF3, 4-OMe
Et
Pri
Nd
Nd
aProcedures as detailed in refs 5 and 6; all determinations were carried out in triplicate, SEM <Æ0.05.
bProcedures as detailed in ref 12; *P<0.05, **P<0.01, ***P<0.001, compared to vehicle-treated controls according to analysis following Lich®eld
and Wilcoxon in ref 17; doses refer to free base. ns: not signi®cant Nd: not determined.
here. The SAR information provided by these latest
studies has led to a re®nement of the original published7
pharmacophore model and this is discussed herein.
isomer, the N-2 tri¯uoroacetyl amine (41a) was pre-
pared in a manner similar to that of Stokker8b and base
hydrolysis to the NH compound followed by methyla-
tion10 or reductive alkylation11 and subsequent hydro-
genation of the 7-nitro to a 7-amino a€orded the desired
N-2 alkyl templates (42b to 42d). The required highly
functionalised benzoic acids that were not commercially
available were prepared as previously described.9 Cou-
pling of these two precursors, using standard proce-
dures, gave the desired benzamides in good overall yields
from 42. Any necessary puri®cation was carried out by
¯ash chromatography through silica gel using dichloro-
methane: methanol: ammonia (95:4.5:0.5) as eluent.
Results and Discussion
Chemistry
The compounds in Tables 1 and 2 were synthesised as
outlined in Scheme 1. The appropriate THIQ amine
templates were prepared according to literature proce-
dures8,9 and consequently only brief details are given in
the Experimental. For the 5-substituted isomer, N-2
alkyl templates were made according to the method of
Mathison et al.,8a (i.e., as for 42e); N-2 H and N-2 acyl
compounds were available using standard methods via
the N-2 Boct amine (42f). In the case of the 7-substituted
Biological Evaluation
Compounds were evaluated in the [3H] SB-204269
binding assay and those which showed a reasonable
level of potency (pKi>7.5) were then examined in vivo
in the mouse MEST test at a standard dose of 10 mg/kg
po 1 h post-dose. Compounds of interest were further
evaluated in the rat model and those which showed
good anticonvulsant activity were then examined to
determine their duration of action (see Table 3). As the
SAR's developed, this led to a re®nement of the screen-
ing protocol and later compounds (see Table 2) were
examined in the rat as the only species of choice.
A large number of compounds was prepared in order to
probe the SAR of THIQ benzamides and a selection has
been chosen to illustrate key ®ndings (see Tables 1 and 2).
Initially, variation of the substitution pattern of the
benzamide moiety was investigated, primarily in the 5-
substituted isomer series.
Scheme 1. Reagents and conditions: (i) 10% Pd/C, H2, ethanol, 4 h,
>90% yield; or SnCl2, EtOH, 3 h, 50 ꢀC, 40±60% yield; (ii) ArCOCl,
Et3N, CH2Cl2, 6 h, 25 ꢀC, 45±85% yield; (iii) ArCO2H, 1-(3-Dimethy-
laminopropyl)-3-ethylcarbodiimide, HOBT, DMF, 18 h, 25 ꢀC, 50±
85% yield.

2088
W. N. Chan et al. / Bioorg. Med. Chem. 8 (2000) 2085±2094
Table 3. Rat MEST data for selected compounds
Compound
pKi
Dose mg/kg po
% Increase in seizure thresholda
time post-dose (h)
0.25
50
190
240
380
50
Nd
180
105
70
120
70
0.5
90
1
180
400
810
760
300
470
690
380
225
250
230
2
420
580
1300
1100
520
340
740
620
310
450
850
4
570
330
1400
1100
395
110
700
830
410
650
1200
6
390
Nd
SB-204269
6
15
13
7.3
7.7
7.6
8.3
10
10
10
10
2
10
5
5
250
420
640
150
Nd
400
180
130
260
220
Nd
700
255
Nd
400
700
190
260
1100
28
29
30
31
32
33
7.6
8.6
8.7
8.3
8.8
8.9
2
5
2
aProcedure as detailed in refs 12 and 16; doses refer to free base; all ®gures P<0.05; Nd: Not determined.
From our earlier published initial SAR for mono-
substituted benzamides in the 5-series, 2-methoxy 3 and
4-t-butyl 5 showed modest anity (pKi 5.8 and 6.0.
respectively) but 3-chloro 4 was inactive (pKi<5). The
combination of a 2-methoxy with either a 4-t-butyl 6
(pKi 7.7) or 5-chloro 7 (pKi 7.0) resulted in markedly
enhanced potency.7 Additional analogues have since
been prepared to further probe the SAR of this class of
THIQ benzamides. In general, for disubstitution in the 5-
series the combination of a 2-methoxy with a lipophilic
4-substituent results in optimum potency, although the
2,4-dimethoxy 9 showed a relatively modest level of
potency (pKi 6.5).
binding site. Secondly, the 2-methoxy substituent on the
benzamide is no longer essential for good binding a-
nity (e.g. compare 3-chloro 4 (pKi<5) and 26 (pKi 7.5))
and is in fact detrimental (up to 10-fold) to optimum
potency (i.e. compare 27/28 (pKi 8.1, 7.6) and 32/36
(pKi 8.8, 7.9)). This has been rationalised by molecular
modelling (see later). The monosubstituted benzamides
(26 pKi 7.5, 27 pKi 8.1) showed potency levels over 100-
fold higher than in the 5-substituted series. The dis-
ubstituted benzamides (29, 30 and 32±34) containing a
3-electron withdrawing group combined with 4-meth-
oxy or 4-ethoxy were of low nanomolar potency. Inter-
estingly, the reversed combination was 20-fold less
potent (compare 35 (pKi 7.3) and 29 (pKi 8.6)). Intro-
duction of a second chloro group into the 5-position was
detrimental (37, pKi 7.5). Another di€erence from the 5-
series THIQs was that increasing the lipophilic bulk of
the 4-alkoxy group was slightly detrimental (OMe
29=OEt 30>OPri 31). However, in agreement with the
5-series, any increase in the size of the N-2 alkyl on the
THIQ was detrimental (compare 33 with 39 and 40)
with NH and NMe being equipotent (33, 38 pKi 8.9).
Introduction of a 5-chloro into 9 resulted in a trisub-
stituted compound 11 with a 10-fold increase in potency.
Increasing the lipophilic bulk of the 4-alkoxy group (12±
13) was bene®cial, the iso-propoxy compound 13 (pKi
8.3) being the most potent 5-series THIQ identi®ed. In
contrast, introduction of a 5-chloro into the 2-methoxy
4-t-butyl 6 gave compound 14 (pKi 6.8), which was 10-
fold less potent. Interestingly, the 3,5-dichloro 2-meth-
oxy 8 and 2-methoxy 6-methyl 10 were virtually inactive
(pKi 5.0).
The most potent compounds from both isomeric series
were examined in vivo. Initially, compounds with a
pKi>7.5 were examined in mouse MEST at 10 mg/kg
po for comparison with SB-204269 at the same dose.
Three 5-series THIQs (6, 13 and 15) and ®ve 7-series
THIQs (29±32) showed a good level of activity (>100%
increase in seizure threshold) and were further evaluated
in the rat MEST model along with the 4-methoxy-3-tri-
¯uoromethyl compound 33 (see Table 3).12 All these
compounds showed a rapid onset of activity (good
activity at 15 min post-dose) with the peak level of e€ect
being at around 4 h post-dose. The duration of e€ect
was also encouraging with activity maintained at the 6 h
time point, especially for (33) which was the most
potent compound and showed an excellent in vivo pro-
®le at the low dose of 2 mg/kg. Unfortunately, the 7-
THIQ compound 28 did not exhibit a robust response
in the rat in spite of showing good activity in the mouse
model. Compound 33, which showed the best pro®le in
the time-course studies, was further examined in the
more stringent MES test and had an ED50 of 0.67 mg/
kg po. This was some 10-fold lower than the ®gure for
SB-204269 which had an ED50 of 6.3 mg/kg po.
On completion of these initial optimisation studies, the
e€ects of changing the nature of the THIQ nucleus were
examined, using either the 2-methoxy 4-t-butyl or the
2,4-dimethoxy 5-chloro benzamide substitution patterns
(16±23). Comparison of N-2 alkyl substituents revealed
that there is steric intolerance with a sharp loss in a-
nity for substituents larger than methyl (NH=NMe>
NEt>NBenzyl). The carbamates 21 (pKi 6.2) and 19
(pKi<5.9) (Again, the size restriction around N-2 was
apparent with lower anity seen for the more bulky
t-butyl carbamate) and sulfonamide 22 (pKi 6.4) was also
less potent and tetralin 23 (pKi<5.8) was inactive, indi-
cating the need for a basic nitrogen for optimum potency.
The 30-fold lower potency of the aromatic quinoline 24
(pKi 6.1) in comparison with its THIQ counterpart 11
(pKi 7.6) probably re¯ects its much lower basicity.
Examination of the data for 7-substituted THIQ benz-
amides, revealed striking di€erences compared to the 5-
substituted series. Firstly, the compounds are generally
more potent showing very high anity at the SB-204269

W. N. Chan et al. / Bioorg. Med. Chem. 8 (2000) 2085±2094
2089
Figure 2. View of the alternative overlapped energy minimised 6 (green) and 28 (yellow) generated using the SYBYL package (O=red, N=blue).
Molecular modelling overlap studies using SYBYL13
(see Fig. 1) were used to rationalise the equivalent a-
nities of the 5- and 7-substituted isomers 6 and 28. These
were carried out on the charged molecules, ¯exibly ®tting
the 5-isomer onto an energy minimised 7-conformation.
Equivalent spring constants were applied to the basic N-
2 nitrogens, amide linkages and aromatic carbon atoms
bearing the lipophilic tert-butyl substituents; resulting in
good overlap of the important pharmacophoric elements.
If this overlap is correct, the directionality of the THIQ
nitrogen lone pair appears to be unimportant for optimum
activity. Intramolecular hydrogen bonding between the 2-
methoxy and the amide linker gives a virtual 6-membered
ring which stabilises a low energy conformation of the
compounds. However, this model does not explain why
the 2-methoxy group is detrimental in the 7-series.
potential therapy for epilepsy with partial seizures (with
or without secondary generalisation).14 In vitro selectiv-
ity cross-screening revealed that compounds 6, 13, and
33 were selective (>30-fold) over a range of other recep-
tors (e.g., amino acid receptors) and ion channels (e.g.
sodium, potassium and chloride), which are thought to
modulate neurotransmission. Further studies around
THIQ benzamides and related compounds will be repor-
ted in future publications.
Experimental
Unless otherwise noted, materials were obtained from
commercial suppliers and were used without further pur-
i®cation. Carboxylic acids were prepared as published in
ref 9. All extracts were washed with water, brine and
dried over anhydrous magnesium sulfate; all solvents
were removed with a Buchi rotary ®lm evaporator.
Chromatography refers to ¯ash chromatography which
was performed on either Merck Kieselgel 60 (0.040±0.063
mm) or Waters Sep-Pak1 Vac silica cartridges (5 or 10g).
1H NMR spectra were obtained using either a Bruker AC
250 or a Jeol GX 270 spectrometer. High-resolution
electron impact mass spectra were determined on a Jeol
JMS DX 303/DA 5000 system operating at 70eV.
Atmosperic pressure ionisation mass spectra [m/z (API)]
were obtained using a VG Micromass OpenLynx Plat-
form LCMS system. Melting points (mp) were determined
using a Ko¯er hot stage apparatus and are uncorrected.
Biological data for hydrochloride salts refer to the equva-
lent amount of free base in order to provide a direct
comparison.
Further molecular modelling studies suggested an alter-
native binding mode for 5- and 7-isomers where the
essential 2-OMe oxygen of the former overlapped with
the carbonyl oxygen of the latter (see Fig. 2). Such an
overlap provides a possible explanation of the key SAR
®nding that the 2-OMe is detrimental in the 7-isomer.
Its position in this mode could result in an unfavourable
steric interaction at the binding site. The basic THIQ N-
20s overlap well in this alternative model of binding as do
the lipophilic 4-substituents on the benzamide moiety,
although this extends further in the 7-series. This is
consistent with smaller lipophilic groups in the 7-series
having greater anity than would be predicted from the
corresponding 5-series SAR as they are able to extend
deeper into the presumed lipophilic pocket.
The full details for the synthesis of all the required
THIQ amino templates have been published9 and the
following procedures have been given to illustrate the
general principles involved.
Conclusion
Using [3H] SB-204269 as a ligand, a new chemical series
has been discovered with anities up to 50-fold higher
than that of SB-204269 and for a number of these
potent compounds, this is re¯ected in greatly enhanced
activity in in vivo anticonvulsant models. The overall
pharmacological pro®le of these compounds suggests a
7-Nitro-1,2,3,4-tetrahydro-2-tri¯uoroacetylisoquinoline
(41a). The N-2-(4-nitrophenyl)ethyl tri¯uoroacetamide

2090
W. N. Chan et al. / Bioorg. Med. Chem. 8 (2000) 2085±2094
(2.26g, 9.15mmol) and paraformaldehyde (0.45 g, 14.4
mmol) in acetic acid (10mL) and concd H2SO4 (15 mL)
were stirred at 25 ꢀC for 20 h and work up according to
the procedure of G.E. Stokker, a€orded the title com-
pound as a white solid (2.17 g, 86%). 1H NMR (CDCl3)
d 3.10 (2H, m), 3.92 (2H, m), 4.85+4.92 (2H, 2xs), 7.38
(1H, t), 8.10 (2H, m); m/z (EI): 274 (M+ 11%, C11H9
F3N2O3 requires M+ 274).
into dichloromethane. Work up and chromatography
on Kieselgel 60 in 10% methanol:dichloromethane gave
the title compound as a viscous yellow oil (0.26 g; 41%).
1H NMR (CDCl3) d 1.12 (6H, d, J=7 Hz), 2.76 (4H,
m), 2.84 (1H, m), 3.63 (2H, s), 6.37 (1H, d, J=7 Hz),
6.48 (1H, dd, J=7, 1 Hz), 6.86 (1H, d, J=7 Hz); m/z
(API+): 191 (MH+; 80%, C12H18N2 requires M+ 190).
7-Amino-2-ethyl-1,2,3,4-tetrahydroisoquinoline (42d). The
title compound was prepared in 14% overall yield from
acetaldehyde using a method similar to that described
2-Methyl-7-nitro-1,2,3,4-tetrahydroisoquinoline (41b).
The tri¯uoroacetamide 41a (17.22 g; 63mmol) was hydro-
lysed at 25 ꢀC using a solution of potassium carbonate
(46.6 g) in 10% aqueous methanol (660 mL). After 18 h,
extraction with dichloromethane gave the N-2 amine
(11 g). A portion of this foregoing N-2 amine (2.08 g,
11.7 mmol) was treated with 88% formic acid (3.45 mL)
and 37% aqueous formaldehyde (5.88 mL) at 80 ꢀC for
2 h according to the procedure of G.M. Carrera and
D.S. Garvey.10 Basi®cation with 10% sodium hydroxide
followed by extraction with ethyl acetate a€orded an
orange gum (2.3 g). Chromatography on Kiesegel 60 in
0 to 3% methanol:ethyl acetate gave the title compound
as an orange solid (1.7 g). 1H NMR (CDCl3) d 2.52 (3H,
s), 2.75 (2H, t, J=6 Hz), 3.03 (2H, t, J=6 Hz), 3.67 (2H,
s), 7.28 (1H, d, J=8 Hz), 7.94 (1H, d, J=2 Hz), 8.01
(1H, dd, J=8, 2 Hz); m/z (API+): 193 (MH+; 100%,
C10H12N2O2 requires M+ 192).
1
above. H NMR (CDCl3) d 1.19 (3H, t, J=7Hz), 2.57
(2H, q, J=7Hz), 2.74 (4H, m), 3.55 (2H, s), 6.37 (1H, d,
J=7Hz), 6.50 (1H, dd, J=7, 1 Hz), 6.87 (1H, d, J=7Hz);
m/z (API+): 177 (MH+; 100%, C11H16N2 requires M+
176).
5-Amino-2-methyl-1,2,3,4-tetrahydroisoquinoline (42e).
5-Aminoisoquinoline (14.4 g, 100 mmol) in acetone
(300mL) was treated with iodomethane (14.4 mL) and
then allowed to stand at 25ꢀC for 2 h. The yellow pre-
cipitate was then ®ltered, washed with acetone and dried
to a€ord 5-amino-2-methylisoquinolinium iodide (18.8 g,
65 mmol). This was dissolved in methanol (1.5 L) and
water (60 mL) with cooling to 0 ꢀC and sodium borohy-
dride (17.8 g, 0.47 mol) was added portionwise over 2 h.
The mixture was then allowed to stir at room tempera-
ture for 18 h. Concentration in vacuo followed by
extraction of the residue into dichloromethane gave the
title compound as a beige solid (8.87 g, 54%), mp 51±
7-Amino-2-methyl-1,2,3,4-tetrahydroisoquinoline (42b).
The 7-nitro compound 41b (0.25 g; 1.3 mmol) in metha-
nol (40 mL) was hydrogenated over 10% palladium on
carbon (100 mg) at atmospheric pressure overnight. The
catalyst was removed by ®ltration through a pad of Kie-
selguhr and evaporation in vacuo gave the title com-
pound as a cream solid (213 mg; 99%), mp 82±84 ꢀC
53 ꢀC (MeOH). H NMR (CDCl3) d 2.44 (3H, s), 2.60
1
(2H, d, J=7 Hz), 2.73 (2H, d, J=7 Hz), 3.53 (2H, s),
3.56 (2H, brs), 6.46±6.57 (2H, m), 6.96 (1H, t, J=8 Hz).
5-Amino-2-(tert-butoxycarbonyl)-1,2,3,4-tetrahydroiso-
quinoline (42f). A solution of 5-aminoisoquinoline
(10 g, 69 mmol) in glacial acetic acid (150 mL) and con-
centrated sulfuric acid (1 mL) was hydrogenated over
platinum oxide (1 g) at 55 psi for 20 h. Evaporation in
vacuo followed by basi®cation of the residue and
extraction with dichloromethane gave 5-amino-1,2,3,4-
tetrahydroisoquinoline (6.45 g, 44 mmol). This was dis-
solved in 1,4-dioxane (250 ml) containing 3 M sodium
hydroxide (14.7 mL, 44 mmol) and di-tert-butyl-dicar-
bonate (9.57 mL, 44 mmol). The solution was stirred at
25 ꢀC for 18 h. The reaction mixture was then poured
into water (400 mL) and extracted with ether. The
organic phases were dried and concentrated in vacuo to
a€ord a brown oil which solidi®ed on standing. Recrys-
tallisation from ethanol/petrol gave the title compound
1
(MeOH). H NMR (CDCl3) d 2.43 (3H, s), 2.64 (2H, d,
J=7 Hz), 2.79 (2H, d, J=7 Hz), 3.48 (2H, s), 3.51 (2H,
brs), 6.36 (1H, d, J=2 Hz), 6.50 (1H, dd, J=8, 2 Hz),
6.89 (1H, d, J=8 Hz); m/z (API+): 163 (MH+; 90%,
C10H14N2 requires M+ 162).
7-Amino-1,2,3,4-tetrahydro-2-tri¯uoroacetylisoquinoline
(42a). Was prepared from 41a in a similar manner to the
procedure used for (42b). 1H NMR (CDCl3) d 2.84 (2H,
m), 3.23 (2H, brs), 3.82 (2H, m), 4.66 (2H, 2s, rotamers),
6.47 (1H, m), 6.57 (1H, m), 6.96 (1H, m); m/z (API+):
245 (MH+; 100%, C11H11F3N2O requires M+ 244).
7-Amino-2-iso-propyl-1,2,3,4-tetrahydroisoquinoline (42c).
Nitro compound NH ex 41a/b (1.55 g, 8.7 mmol) and
acetone (2.52 g, 43.5mmol) in 1,2-dichloroethane (50 mL)
were treated with sodium triacetoxyborohydride (0.28 g,
13.1 mmol) and glacial acetic acid (0.6 mL, 9.0 mmol).11
The mixture was stirred at 25 ꢀC over 72 h and then
diluted with dichloromethane (50 mL). The mixture was
washed with saturated NaHCO3, dried (Na2SO4) and
evaporated in vacuo. Chromatography on Kieselgel 60
in ethyl acetate gave the N-iso-propyl compound
(0.73 g). This material (0.73 g, 3.32 mmol) in ethanol
(100 mL) was heated to 50 ꢀC and treated with a solu-
tion of tin (II) chloride (2.52 g, 13.27 mmol) in concd
HCl (10 mL) and stirring continued for 3 h. The mixture
was basi®ed with 40% NaOH and the product extracted
1
as an o€-white crystalline solid (5.1 g, 30%). H NMR
(CDCl3) d 1.48 (9H, s), 2.55 (2H, t, J=7 Hz), 3.60 (2H,
brs), 3.70 (2H, t, J=7 Hz), 4.54 (2H, s), 6.56 (2H, d,
J=8 Hz), 7.01 (1H, t, J=8 Hz).
The acylation procedure given for 9 was used for the pre-
paration of compounds 2, 3, 4, 5, 7, 25, 26, 27, and the
remaining benzamides were prepared in a manner similar
to that used for 6. Final compounds were isolated as
foams following chromatographic puri®cation and 6, 9,
13, 14, 15, 17, 18, 20, 24, 33 and 36 were converted into
monohydrochlorides using 1 M HCl in diethyl ether and
methanol mixtures.

W. N. Chan et al. / Bioorg. Med. Chem. 8 (2000) 2085±2094
2091
5-Series THIQ compounds in Table 1.
6 Hz), 7.40 (1H, d, J=6 Hz), 7.67 (1H, d, J=2 Hz),
7.85 (1H, d, J=2 Hz), 9.87 (1H, s); m/z (API+): 367,
365 (MH+; 65, 100%, C18H18Cl2N2O2 requires M+
365).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-2-meth-
1
oxybenzamide, hydrochloride. (3). H NMR (DMSO-d6)
d 2.90 (3H, s), 3.08 (2H, m), 3.38 (1H, m), 3.75 (1H, m),
4.00 (3H, s), 4.30 (1H, m), 4.50 (1H, m), 7.05 (1H, d,
J=8 Hz), 7.11 (1H, t, J=8 Hz), 7.23 (1H, d, J=8 Hz),
7.31 (1H, t, J=8 Hz), 7.55 (1H, t, J=8 Hz), 7.85 (2H, m),
9.85 (1H, s), 10.95 (1H, br, s); m/z (API+)+ 297.
(MH+; 90%, C18H20N2O2 requires M+ 296).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-2,4-di-
methoxybenzamide, hydrochloride (9). 2,4-Dimethox-
ybenzoyl chloride (263 mg 1.54 mmol) was added to a
solution of 5-amino 42e (250 mg, 1.54 mmol) in CH2Cl2
(5 mL) containing triethylamine (0.4 mL) and the mix-
ture kept at 25 ꢀC for 18 h. The mixture was partitioned
between chloroform (50 mL) and water (50 mL) and
work up as above for 6 gave the title compound as a
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-3-chloro-
1
benzamide (4). H NMR (CDCl3) d 2.90±3.50 (7H, m),
white solid (330 mg, 60%), mp>200 ꢀC (dec). H NMR
1
4.40 (2H, m), 7.15 (1H, m), 7.36 (2H, m), 7.58 (1H, t,
J=7Hz), 7.70 (1H, m), 7.97 (1H, d, J=7Hz), 8.05 (1H, s),
10.22 (1H, s), 10.85 (1H, brs); m/z (API+): 303, 301
(MH+; 35, 100%, C17H17ClN2O requires M+ 300).
(DMSO-d6) d 2.90 (3H, s), 3.05 (2H, br. s), 3.35 (2H,s),
3.85 (3H, s), 4.03 (3H, s), 4.25±4.55 (2H, m), 6.70 (1H,
dd, J=1, 6 Hz), 6.75 (1H, m), 7.00 (1H, d, J=6 Hz),
7.30 (1H, t, J=6 Hz), 7.95 (2H, m), 9.75 (1H, s), 11.25
(1H, brs); m/z (API+): 327 (MH+; 100%, C19H22N2O3
requires M+ 326).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-4-tert-
butylbenzamide (5). Whitish solid, mp 226 ꢀC (ethyl ace-
1
tate). H NMR (270 MHz, DMSO-d6) d 1.31 (9H, s),
2.90 (3H, m), 3.42 (3H, m), 3.62 (1H, m), 4.32 (1H, m),
4.52 (1H, m), 7.11 (1H, m,), 7.31 (2H, m), 7.53 (2H, d,
J= 6 Hz), 7.95 (2H, d, J=6 Hz), 10.98 (1H, brs); m/z
(API+): 323 (MH+; 100%, C21H26N2O requires M+
322).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-2,4-di-
1
methoxy-6-methylbenzamide (10). H NMR (DMSO-d6)
d 2.28 (3H, s), 2.31 (3H, s), 2.60 (2H, m), 2.78 (2H, m),
3.48 (2H,s), 3.76 (6H, s), 6.46 (2H, d, J=7 Hz), 6.92
(1H, d, J=7 Hz), 7.23 (1H, t, J=7 Hz), 7.28 (1H, d, J=
7 Hz), 9.46 (1H, s); m/z (API+): 341 (MH+; 100%,
C20H24N2O3 requires M+ 340).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-4-tert-
butyl-2-methoxybenzamide hydrochloride (6). 4-t-Butyl-
2-methoxybenzoic acid (208 mg, 1.0 mmol) in DMF
(8 mL) was stirred with 1 equiv of 1-(3-dimethyl amino-
propyl)-3-ethylcarbodiimide hydrochloride (192 mg)
and 1-hydroxybenzotriazole (135 mg) at 25ꢀC for
30 min. A solution of 5-amino 42e (163 mg, 1.0 mmol) in
CH2Cl2 (2 mL) was added and the mixture kept at 25 ꢀC
for 18 h. The mixture was partitioned between chloro-
form (50 mL) and water (50 mL) and the material in the
organic phase was puri®ed by ¯ash chromatography.
Combination of appropriate fractions gave a solid
(253 mg, 75%) which was converted into a hydrochlo-
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-5-chloro-
2,4-dimethoxybenzamide (11). White crystals, mp 187±
189 ꢀC (from ethyl acetate). 1H NMR (DMSO-d6) d 2.34
(3H, s), 2.67 (2H, m), 2.75 (2H, m), 3.49 (2H, s), 3.99
(3H, s), 4.10 (3H, s), 6.87 (1H, d, J=5 Hz), 6.95 (1H, s)
7.15 (1H, t, J=5 Hz), 7.89 (1H, d, J=5 Hz), 7.94 (1H,
s), 9.70 (1H, s); m/z (API+)+ 361. (MH+; 70%,
C19H21ClN2O3 requires M+ 360).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-5-chloro-
1
4-ethoxy-2-methoxybenzamide (12). H NMR (DMSO-
ride salt, mp 218±220 ꢀC. H NMR (270 MHz, DMSO-
d6) d 1.42 (3H, t, J=6Hz), 2.34 (3H, s), 2.70 (4H, m), 3.50
(2H, s), 4.08 (3H, s), 4.28 (2H, m), 6.87 (1H, d, J=6Hz),
6.93 (1H, s), 7.15 (1H, t J=6Hz), 7.86 (1H, d, J=6Hz),
7.94 ( 1H, s), 9.70 (1H, s); m/z (API+): 375. (MH+; 100%,
C20H23ClN2O3 requires M+ 374).
1
d6) d 1.32 (9H, s), 2.91 (3H, s), 3.06 (2H, m), 3.35 (1H,
m), 3.70 (1H, m), 4.02 (3H, s), 4.45 (2H, m), 7.04 (1H, d,
J=10 Hz), 7.15 (2H, m), 7.31 (1H, m), 7.82 (1H, d,
J=12 Hz), 7.90 (1H, d, J=12 Hz), 9.83 (1H, s), 10.82
(1H, s); m/z (API+): 353 (MH+, 80%), Found: M+
352.21549, calcd for C22H28N2O2 352.21666.
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-5-chloro-
2-methoxy-4-iso-propoxybeꢀnzamide, hydrochloride (13).
1
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-5-chloro-
2-methoxy-benzamide (7). White crystals, mp 176±
178 ꢀC (from ethyl acetate). 1H NMR (DMSO-d6) d 2.36
(3H, s), 2.67 (2H, m), 2.73 (2H, m), 3.51 (2H, s), 3.98
(3H, s), 6.92 (1H, d, J=6 Hz), 7.16 (1H, t, J=7 Hz),
7.28 (1H, d, J=8 Hz), 7.61 (1H, dd, J=3, 10 Hz), 7.71
(1H, d, J=8 Hz), 7.83 (1H, d, J=3 Hz), 9.79 (1H, s); m/z
(API+): 331. (MH+; 100%, C18H19ClN2O2 requires
M+ 330).
White crystals, mp 190±193 C. H NMR (DMSO d6) d
1.37 (6H, d, J=6 Hz), 2.92 (3H, s), 3.05 (2H, m), 3.37
(1H, m), 3.73 (1H, m), 4.08 (3H, s), 4.32 (1H, m), 4.51
(1H, m), 4.96 (1H, m), 6.95 (1H, s), 7.06 (1H, d, J=6 Hz),
7.32 (1H, t, J=6 Hz), 7.89 (2H, m), 9.27 (1H, s), 11.20
(1H, br s); m/z (API+): 389 (MH+, 80%, C21H25
ClN2O3 requires M+ 388).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-4-tert-
butyl-2-methoxy-5-chlorobenzamide, hydrochloride (14).
1H NMR (DMSO-d6) d 1.49 (9H, s), 2.91 (3H, s), 3.05
(2H, m), 3.18 (2H, s), 3.35 (1H, m), 4.03 (3H, s), 4.42 (1H,
m), 7.08 (1H, d, J=8 Hz), 7.20 (1H, s), 7.33 (1H, m), 7.79
(2H, m), 9.82 (1H, s), 10.66 (1H, br s); m/z (API+): 387
(MH+, 90%, C22H27ClN2O2 requires M+ 386).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-3,5-di-
chloro-2-methoxybenzamide (8). Whitish solid, mp
138 ꢀC (dichloromethane:hexanes). H NMR (DMSO-
1
d6) d 2.33 (3H, s), 2.58 (2H, m), 2.78 (2H, m), 3.49 (2H,
s), 3.88 (3H, s), 6.95 (1H, d, J=6 Hz), 7.16 (1H, t, J=

2092
W. N. Chan et al. / Bioorg. Med. Chem. 8 (2000) 2085±2094
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-4-iso-pro-
pyl-5-tri¯uoromethyl-2-methoxybenzamide, hydrochloride
(15). White crystals, mp 188±191 ꢀC. 1H NMR (DMSO-
d6) d 1.30 (6H, d), 2.90 (3H, s), 3.05 (2H, m), 3.30 (1H,
m), 3.50 (2H, m), 4.07 (3H, s), 4.39 (2H, m), 7.06 (1H, d,
J=9 Hz), 7.31 (2H, m), 7.74 (1H, d, J=10 Hz), 8.02
(1H, s), 9.86 (1H, s), 11.00 (1H, br s). m/z (API+): 407
(MH+; 90%, C22H25F3N2O2 requires M+ 406).
(1H, brs), 7.04 (1H, d, J=2 Hz), 7.18 (1H, dd, J=8,
2 Hz), 7.25 (1H, m), 8.10 (1H, brs), 8.25 (1H, d,
J=8 Hz), 9.70 (1H, brs); m/z (API+): 397 (MH+;
100%, C23H28N2O4 requires M+ 396).
N-(2-Methanesulfonyl-1,2,3,4-tetrahydroisoquinolin-5-
yl)-4-tert-butyl-2-methoxybenzamide (22). 1H NMR
(250MHz, CDCl3) d 1.45 (9H, s), 2.88 (3H, s), 2.92 (2H, t,
J=7Hz), 3.64 (2H, t, J=7Hz), 4.09 (3H, s), 4.48 (2H, s),
6.94 (1H, d, J=8 Hz), 7.0 (1H, d, J=2 Hz), 7.28 (1H,
dd, J=8, 2 Hz), 7.29 (1H, m), 8.05 (1H, d, J=8 Hz),
8.23 (1H, d, J=8 Hz), 9.70 (1H, brs); m/z (API+): 417
(MH+; 100%, C22H28N2O4S requires M+ 416).
N-(1,2,3,4-Tetrahydroisoquinolin-5-yl)-5-chloro-2,4-dime-
thoxybenzamide (16). To a solution of 5-chloro-2,4-dime-
thoxy-N-[2-(tert-butoxycarbonyl)-1,2,3,4-tetrahydroiso-
quinolin-5-yl]benzamide (1 g) in dichloromethane (30
mL) at 0 ꢀC was added tri¯uoracetic acid (3 mL). The
mixture was then stirred at room temperature for 3 h
before pouring into saturated aqueous sodium bicarbo-
nate (100 mL) and extracting with dichloromethane.
The organic phase was dried and concentrated in vacuo
gave the title compound as an o€ white solid (700 mg).
1H NMR (DMSO-d6) d 2.62 (2H, m), 3.06 (2H, m), 3.85
(2H, s), 4.00 (3H, s), 4.10 (3H, s), 6.86 (1H, d, J=7 Hz),
6.96 (1H, s), 7.13 (1H, t, J=7 Hz), 7.86 (1H, d, J=
7 Hz), 7.94 (1H, s), 9.70 (1H, s); m/z (API+): 347.
(MH+; 100%, C18H19ClN2O3 requires M+ 346).
N-(5,6,7,8-Tetrahydronaphthalen-1-yl)-4-tert-butyl-2-
1
methoxybenzamide (23). H NMR (250MHz, CDCl3) d
1.39 (9H, s), 1.70±2.00 (4H, m), 2.18 (2H, t, J=7Hz), 2.80
(2H, t, J=7Hz), 3.98 (3H, s), 6.90 (1H, d, J= 8 Hz), 7.04
(1H, d, J=2Hz), 7.08±7.22 (2H, m), 8.10 (1H, d, J=8Hz),
8.23 (1H, d, J=8Hz), 9.75 (1H, brs); m/z (API+): 338
(MH+; 100%, C22H27NO2 requires M+ 337).
N-(Isoquinolin-5-yl)-5-chloro-2,4-dimethoxybenzamide
1
hydrochloride (24). H NMR (DMSO-d6) d 4.03 (3H,
s), 4.13 (3H, s), 7.00 (1H, s), 7.89 (1H, s), 7.96 (1H, t,
J=7 Hz), 8.15 (2H, m), 8.41 (1H, d, J=7 Hz), 8.71 (1H,
d, J=7 Hz), 9.78 (1H, s), 10.47 (1H, brs); m/z (API+):
343 (MH+; 100%, C18H15ClN2O3 requires M+ 342).
N-(2-Ethyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-5-chloro-
1
2,4-dimethoxybenzamide, hydrochloride (17). H NMR
(DMSO-d6) d 1.36 (3H, t, J=7 Hz), 3.09 (2H, m), 3.12±
3.85 (4H, m), 4.00 (3H, s), 4.10 (3H, s), 4.30 (1H, m), 4.55
(1H, m), 6.94 (1H, s), 7.06 (1H, d, J=6 Hz), 7.32 (1H, t,
J=6 Hz), 7.89 (2H, m), 9.76 (1H, s), 10.85 (1H, brs); m/z
(API+): 375 (MH+; 100%, C20H23ClN2O3 requires M+
374).
7-Series THIQ compounds in Table 2.
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)benzamide
1
(25). H NMR (250 MHz, CDCl3) d 2.46 (3H, s), 2.69
(2H, t), 2.91 (2H, t), 3.58 (2H, s), 7.10 (1H, d), 7.30 (1H,
dd), 7.40±7.60 (4H, overlapping m), 7.75 (1H, br s), 7.87
(2H, m); m/z (API+): 267 (MH+; 100%, C17H18ClN2O
requires M+ 266).
N-(2-Benzyl-1,2,3,4-tetrahydroisoquinolin-5-yl)-5-chloro-
2,4-dimethoxybenzamide hydrochloride (18). 1H NMR
(DMSO-d6) d 3.08 (2H, m), 3.78 (2H, m), 4.00 (3H, s),
4.09 (3H, s), 4.37 (2H, m), 4.49 (2H, m), 6.95 (1H, s),
7.05 (1H, d, J=7 Hz), 7.31 (1H, t, J=7 Hz), 7.50 (3H,
m), 7.65 (2H, m), 7.90 (2H, m), 9.75 (1H, s), 10.98 (1H,
brs); m/z (API+): 437 (MH+; 100%, C25H25ClN2O3
requires M+ 436).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-3-chloro-
benzamide (26). 1H NMR (250 MHz, CDCl3) d 2.46
(3H, s), 2.68 (2H, t), 2.90 (2H, t), 3.57 (2H, s), 7.10 (1H,
d), 7.29 (1H, dd, overlapping with CHCl3), 7.39 (1H, s),
7.42 (1H, d), 7.52 (1H, m), 7.73 (1H, m), 7.83 (2H, m).
N-(2-t-Butyloxycarbonyl-1,2,3,4-tetrahydroisoquinolin-5-
yl)-5-chloro-2,4-dimethoxybenzamide (19). 1H NMR
(250 MHz, CDCl3) d 1.51 (9H, s), 2.75 (2H, m), 3.72
(2H, m), 3.97 (3H, s), 4.09 (3H, s), 4.60 (2H, s), 6.54
(1H, s), 6.94 (1H, d, J=7 Hz), 7.24 (1H, t, J=7 Hz),
8.11 (1H, m), 8.30 (1H, s), 9.52 (1H, brs).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-4-t-butyl-
1
benzamide (27). H NMR (250MHz, CDCl3) d 1.34 (9H,
s), 2.44 (3H, s), 2.68 (2H, t), 2.89 (2H, t), 3.55 (2H, s), 7.07
(1H, d), 7.29 (1H, dd overlapping with CHCl3 signal),
7.38±7.53 (3H, m, overlapping signals), 7.75±7.90 (3H,
m, overlapping signals).
N-(1,2,3,4-Tetrahydroisoquinolin-5-yl)-4-tert-butyl-2-
methoxybenzamide hydrochloride (20). Whitish crystals,
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-4-tert-
butyl-2-methoxybenzamide (28). 1H NMR (250MHz,
CDCl3) d 1.35 (9H, s), 2.46 (3H, s), 2.68 (2H, t), 2.89 (2H,
s), 3.59 (2H, s), 4.06 (3H, s), 7.01 (1H, d), 7.07 (1H, d), 7.15
(1H, dd), 7.31 (1H, dd), 7.50 (1H, d), 8.19 (1H, d), 9.74
(1H, br s); m/z (API+): 353 (MH+; 90%, C22H28
ClN2O2 requires M+ 352).
mp 230 ꢀC. H NMR (DMSO-d6) d .31 (9H, s), 2.97
1
(2H, m), 3.37 (2H, m), 3.44 (2H, m), 4.06 (3H, s), 4.31
(2H, m), 7.08 (1H, d, J=9 Hz), 7.16 (2H, m), 7.28 (1H,
m), 7.84 (2H, m), 9.55 (2H, br s), 9.83 (1H, s). m/z
(API+): 339 (MH+; 80%, C21H26N2O2 requires M+
338).
N-(2-Methoxycarbonyl-1,2,3,4-tetrahydroisoquinolin-5-
yl)-4-tert-butyl-2-methoxybenzamide (21). 1H NMR
(250 MHz, CDCl3) d 1.39 (9H, s), 2.80 (2H, brt), 3.72
(3H, s), 3.78 (2H, br), 4.10 (3H, s), 4.65 (2H, s), 6.95
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-3-chloro-
1
4-methoxybenzamide (29). H NMR (CDCl3) d 2.47 (3H,
s), 2.70 (2H, t, J=6Hz), 2.91 (2H, t, J=6Hz), 3.59 (2H, s),
3.97 (3H, s), 6.99 (1H, d, J=9Hz), 7.09 (1H, d, J=8Hz),

W. N. Chan et al. / Bioorg. Med. Chem. 8 (2000) 2085±2094
2093
7.32 (1H, dd, J=2 and 8 Hz), 7.40 (1H, s), 7.79 (2H, m),
7.90 (1H, d, J=2 Hz).
407, 405 (MH+; 80%, C19H21BrN2O3 requires M+
405).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-3-chloro-
1
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-3,5-di-
1
4-ethoxybenzamide (30). H NMR (CDCl3) d 1.51 (3H,
chloro-4-methoxybenzamide (37). H NMR (CDCl3) d
t, J=7 Hz), 2.49 (3H, s), 2.74 (2H, t, J=6 Hz), 2.92 (2H,
t, J=6 Hz), 3.62 (2H, s), 4.18 (2H, q, J=7 Hz), 6.96
(1H, d, J=9 Hz), 7.09 (1H, d, J=8 Hz), 7.31 (1H, dd,
J=2 and 8 Hz), 7.39 (1H, d, J=2 Hz), 7.76 (2H, m),
7.89 (1H, d, J=2 Hz); m/z (API+): 345 (MH+; 100%,
C19H21ClN2O2 requires M+ 344).
2.46 (3H, s), 2.69 (2H, t, J=6 Hz), 2.90 (2H, t, J=6 Hz),
3.57 (2H, s), 3.96 (3H, s), 7.09 (1H, d, J=8 Hz), 7.30
(1H, dd, J=2 and 8 Hz), 7.34 (1H, d, J=2 Hz), 7.81
(2H, s), 7.89 (1H, s); m/z (API+): 365 (MH+; 100%,
C18H18Cl2N2O2 requires M+ 365).
N-(1,2,3,4-Tetrahydroisoquinolin-7-yl)-4-methoxy-3-tri-
¯uoromethylbenzamide (38). 1H NMR (250 MHz, CDCl3)
d 2.65 (2H, t, J=6 Hz), 3.00 (2H, t, J=6 Hz), 3.85 (3H,
s), 3.89 (2H, s), 6.95 (2H, d), 7.17 (1H, dd, J=6, 2 Hz),
7.25 (1H, s), 7.57 (1H, s), 7.93 (2H, m); m/z (API+):
351.1 (MH+; 100%) C18H17F3N2O2 requires M+ 350).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-3-chloro-
1
4-iso-propoxybenzamide (31). H NMR (CDCl3) d 1.42
(6H, d, J=6 Hz), 2.49 (3H, s), 2.74 (2H, t, J=6Hz), 2.92
(2H, t, J=6 Hz), 3.63 (2H, s), 4.67 (1H, quintet, J=6Hz),
6.98 (1H, d, J=9Hz), 7.09 (1H, d, J=8Hz), 7.28 (1H, dd,
J=2 and 8 Hz), 7.40 (1H, d, J=2Hz), 7.67±7.81 (2H,
bm), 7.88 (1H, d, J=2 Hz); m/z (API+): 359 (MH+;
100%, C20H23ClN2O2 requires M+ 358).
N-(2-Ethyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-4-meth-
1
oxy-3-tri¯uoromethyl benzamide (39). H NMR (CDCl3)
d 1.21 (3H, t, J=7 Hz), 2.65 (2H, q, J=7Hz), 2.80 (2H,
d, J=6 Hz), 2.92 (2H, t, J=6 Hz), 3.67 (2H, s), 3.97
(3H, s), 7.07 (2H, m), 7.30 (1H, m), 7.41 (1H, d,
J=2 Hz), 7.89 (1H, brs), 8.06 (2H, m); m/z (API+): 379
(MH+; 100%, C20H21F3N2O2 requires M+ 378).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-3-bromo-
4-methoxybenzamide (32). White powder, mp 134±136 ꢀC
(diethyl ether). 1H NMR (CDCl3) d 2.47 (3H, s), 2.71 (2H,
t), 2.91 (2H, t), 3.60 (2H, s), 3.97 (3H, s), 6.96 (1H, d), 7.10
(1H, d), 7.29 (1H, m), 7.40 (1H, s), 7.67 (1H, s), 7.84 (1H,
dd), 8.02 (1H, s), 8.05 (1H, d); m/z (API+): 377, 375
(MH+; 30%, C18H19BrN2O2 requires M+ 375).
N-(2-iso-Propyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-4-
methoxy-3-tri¯uoromethyl benzamide (40). 1H NMR
(CDCl3) d 1.13 (6H, d, J=7Hz), 2.84 (5H, m), 3.72 (2H,
s), 3.97 (3H, s), 7.07 (2H, m), 7.26 (1H, m), 7.43 (1H, d,
J=2 Hz), 7.83 (1H, brs), 8.04 (2H, m); m/z (API+): 393
(MH+; 100%, C21H23F3N2O2 requires M+ 392).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-4-meth-
1
oxy-3-tri¯uoromethyl benzamide (33). H NMR (CDCl3)
d 2.48 (3H, s), 2.72 (2H, t, J=6Hz), 2.92 (2H, t, J=6Hz),
3.60 (2H, s), 3.98 (3H, s), 7.09 (2H, m), 7.32 (1H, dd, J=2
and 8 Hz), 7.41 (1H, d, J=2Hz), 7.83 (1H, s), 8.07 (2H,
m); m/z (API+): 365 (MH+; 100%, C19H19F3N2O2
requires M+ 364). Converted into a monohydrochloride
obtained as white crystals, mp 186±188 ꢀC.
Pharmacological methods
Mouse MEST model. Compounds were evaluated for
oral anticonvulsant activity in groups of 12 naive mice
(male CD1-Charles River, 25±30 g) in the mouse MEST
test using an `up and down' method of shock titration
as described in Dixon and Mood15 and Upton et al.16
Compounds were administered orally by gavage as a
®ne suspension in 1% methylcellulose in water in a dose
volume of 1 mL/kg. Percentage increases for drug-treated
groups are devised from studies where standard errors
were less than 10% of the CC50 values and with p<0.05
compared to vehicle control animals; measured at 1 h
post-dose. In all experiments, the CC50 values for vehicle-
treated controls fell within the range of 12±14 mA. Statis-
tical comparisons between vehicle and drug-treated
groups were made using the method of Litch®eld and
Wilcoxon.17
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-3-cyano-
4-methoxybenzamide (34). 1H NMR (CDCl3) d 2.47
(3H, s), 2.70 (2H, t, J=6 Hz), 2.91 (2H, t, J=6 Hz), 3.59
(2H, s), 4.02 (3H, s), 7.09 (2H, t, J=8 Hz), 7.29 (1H, dd,
J=2 and 8 Hz), 7.39 (1H, d, J=2 Hz), 7.80 (1H, s), 8.10
(2H, m); m/z (API+): 322 (MH+; 100%, C19H19N3O2
requires M+ 321).
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-4-chloro-
1
3-methoxybenzamide (35). H NMR (250 MHz, CDCl3)
d 2.47 (3H, s), 2.70 (2H, t, J=6 Hz), 2.88 (2H, t,
J=6 Hz), 3.59 (2H, s), 3.98 (3H, s), 7.11 (1H, d), 7.21
(1H, d), 7.30 (2H, m), 7.40 (1H, d), 7.45 (1H, d), 7.75
(1H, s); m/z (API+): 331.1 (MH+; 100%, C18H19ClN2O2
requires M+ 330.8).
Rat MES(T) models. For the MEST test, the threshold
for maximal (tonic hindlimb extension) electroshock
seizures in male rats (Sprague±Dawley, 80±150 g, 6
weeks old) was determined by a Hugo Sachs Electronik
stimulator which delivered a constant current (0.3 s
duration; from 1 to 300 mA in steps of 5±20 mA). For
the MES test a ®xed high-intensity supramaximal cur-
rent of 120 mA (ca. 5Â basal threshold current; 0.3 s
duration) was applied. The procedures are similar to
that outlined above for mouse and full details are as
published by Upton et al.12,16
N-(2-Methyl-1,2,3,4-tetrahydroisoquinolin-7-yl)-5-bromo-
2,4 - dimethoxybenzamide, hydrochloride (36). O€-white
solid, mp 190 ꢀ (dec). 1H NMR (250 MHz, CDCl3+
CD3OD) d 2.89±3.19 (4H, m and overlapping s at 2.98
and HOD signal), 3.27 (1H, m), 3.46 (1H, m, over-
lapping with CHD2OD signal), 3.70 (1H, m), 4.00 (3H,
s), 4.11 (4H, overlapping s and d), 4.58 (1H, br d), 6.54
(1H, s), 7.18 (1H, d), 7.38 (1H, d), 7.61 (1H, br s), 8.36
(1H, s), 9.69 (partially exchanged 1H, br s); m/z (API+):

2094
W. N. Chan et al. / Bioorg. Med. Chem. 8 (2000) 2085±2094
Jerman, J. C.; Orlek, B. S.; Stean, T. O.; Thompson, M.;
Acknowledgements
Upton, N.; Ward, R. W. Bioorg. Med. Chem. Lett. 1998, 8,
2903.
8. (a) Mathison, I. M.; Fowler, K. C.; Morgan, P. H.; Tidwell,
R. R.; Peters, E. R. J. Med. Chem. 1973, 16, 332. (b) Stokker,
G. E. Tetrahedron Lett. 1996, 37, 5433.
9. (a) Harling, J. D.; Orlek, B. S.; Thompson, M. Patent
Cooperation Treaty Application WO97/48683, 1997 128,
10213 and references cited therein; Chem. Abstr. 1997. (b)
Thompson, M.; Harling, J. D.; Edwards, P. D. Patent Coop-
eration Treaty Application WO99/21836 and references cited
therein; Chem. Abstr. 1999, 130, 325093.
We thank Anna Wright and Julie Carlisle for technical
assistance in the preparation of compounds 3, 4, 10, 39
and 40, Vicky Holland and Di Hoult for in vitro
screening information and Kellie Darker for manuscript
preparation.
References and Notes
10. Carrera, G. M.; Garvey, D. S. J. Heterocycl. Chem. 1992,
29, 847.
1. Leach, J. P.; Brodie, M. J. Seizure 1995, 4, 5.
2. Upton, N. Trends Pharmacol. Sci. 1994, 15, 456.
11. (a) Abdel-Magid, F. A.; Carson, K. G.; Harris, B. D.;
Maryano€, C. A.; Shah, R. D. J. Org. Chem. 1996, 61, 3849.
(b) Abdel-Magid, F. A.; Carson, K. G.; Maryano€, C. A.
Tetrahedron Lett. 1990, 31, 5595.
12. Upton, N.; Blackburn, T. P.; Campbell, C. A.; Cooper,
D.; Evans, M. L.; Herdon, H. J.; King, P. D.; Ray, A. M.;
Stean, T. O.; Chan, W. N.; Evans, J. M.; Thompson, M. Brit.
J. Pharmacol. 1997, 121, 1679.
3. Blackburn, T. P.; Buckingham, R. E.; Chan, W. N.; Evans,
J. M.; Hadley, M. S.; Thompson, M.; Upton, N.; Stemp, G.;
Vong, A. K. Bioorg. Med. Chem. Lett. 1995, 5, 1163.
4. Chan, W. N.; Evans, J. M.; Hadley, M. S.; Herdon, H. J.;
Jerman, J. C.; Morgan, H. K. A.; Stean, T. O.; Thompson,
M.; Upton, N.; Vong, A. K. J. Med. Chem. 1996, 39, 4537 and
references cited within. Additional information can also be
found in refs 12 and 14 in this manuscript..
13. SYBYL, Tripos Associates, Inc., 1699 S. Hanley Rd, Suite
303, St. Louis, MO 63144, USA.
5. (a) Preliminary results are in: Herdon, H.; Jerman, J.;
Stean, T. O.; Chan, W.; Middlemiss, D. N.; Upton, N. Eur. J.
Pharmacol. 1996, 314, R7; (b) Herdon, H.; Jerman, J. C.;
Chan, W. Patent Cooperation Treaty Application. WO96/
18650, 1996; Chem. Abstr. 1996, 125, 105148.
14. Upton, N.; Thompson, M. In Progress in Medicinal
Chemistry; King, F. D., Oxford, A. W., Eds.; Elsevier Science:
Oxford, 2000; Vol. 37, pp 177±200.
15. Dixon, W. J.; Mood, A. M. J . Amer. Stat. Assn. 1948, 43,
109.
6. Herdon, H. J.; Jerman, J. C.; Stean, T. O.; Middlemiss, D.
N.; Chan, W. N.; Vong, A. K.; Evans, J. M.; Thompson, M.;
Upton, N. Brit. J. Pharmacol. 1997, 121, 1687.
16. Upton, N.; Stean, T.; Middlemiss, D.; Blackburn; Ken-
nett, G. Eur. J. Pharmacol. 1998, 359, 33.
17. Litch®eld, J. T.; Wilcoxon, F. J. Pharmacol. Exp Ther.
1949, 96, 99.
7. Chan, W. N.; Hadley, M. S.; Harling, J. D.; Herdon, H. J.;