Glucose-6-phosphatase catalytic enzyme inhibitors: Synthesis and in vitro evaluation of novel 4,5,6,7-tetrahydrothieno[3,2-c]- and -[2,3-c]pyridines

Article abstract of DOI:10.1016/S0968-0896(00)00153-X

The discovery of the first class of potent glucose-6-phosphatase catalytic site inhibitors, substituted 4,5,6,7-tetrahydrothieno[3,2-c]- and -[2,3-c]pyridines, is described. Optimisation of this series involved solution phase combinatorial synthesis and very potent compounds were prepared with IC₅₀ values down to 140 nM. The structure-activity relationship (SAR) of these compounds indicates that: a tetrahydrothieno[3,2-c]pyridine core ring system and the isomeric [2,3-c] system are equipotent and much better than the corresponding benzo analogue, 1,2,3,4-tetrahydro-isoquinoline. The 4-substituent of the tetrahydrothieno[3,2-c]pyridine ring has to be a phenyl group, optionally substituted with a lipophilic 4-substituent, such as trifluoromethoxy or chloro. The 5-substituent of the tetrahydrothieno[3,2-c]pyridine ring has to be a substituted benzoyl; anisoyl and (E)-3-furan-3-ylacryloyl are the best of the investigated groups. Substitution in the benzoyl ortho position seems to be forbidden, whereas substitution in the meta position is tolerated only if a methoxy para substituent is present. These SAR findings were parallel to those obtained in the 4,5,6,7-tetrahydrothieno[2,3-c]pyridine system. Enantioselectivity in enzyme recognition was observed and the activity resided in all cases only in one of the enantiomers. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00153-X

Bioorganic & Medicinal Chemistry 8 (2000) 2277±2289
Glucose-6-Phosphatase Catalytic Enzyme Inhibitors: Synthesis
and In Vitro Evaluation of Novel 4,5,6,7-Tetrahydrothieno[3,2-c]-
and -[2,3-c]pyridines
Peter Madsen,a,* Jane M. Lundbeck,a Palle Jakobsen,a Annemarie R. Varmingb
and Niels Westergaardc
a
Ê
Medicinal Chemistry Research, Novo Nordisk A/S, Health Care Discovery, Novo Nordisk Park DK-2760 Malùv, Denmark
b
Ê
Pharmaceutical Chemistry, Novo Nordisk A/S, Health Care Discovery, Novo Nordisk Park DK-2760 Malùv, Denmark
Ê
Diabetes Biochemistry & Metabolism, Novo Nordisk A/S, Health Care Discovery, Novo Nordisk Park DK-2760 Malùv, Denmark
c
Received 17 December 1999; accepted 22 May 2000
AbstractÐThe discovery of the ®rst class of potent glucose-6-phosphatase catalytic site inhibitors, substituted 4,5,6,7-tetra-
hydrothieno[3,2-c]- and -[2,3-c]pyridines, is described. Optimisation of this series involved solution phase combinatorial synthesis
and very potent compounds were prepared with IC50 values down to 140 nM. The structure±activity relationship (SAR) of these
compounds indicates that: a tetrahydrothieno[3,2-c]pyridine core ring system and the isomeric [2,3-c] system are equipotent and
much better than the corresponding benzo analogue, 1,2,3,4-tetrahydro-isoquinoline. The 4-substituent of the tetrahydrothieno[3,2-
c]pyridine ring has to be a phenyl group, optionally substituted with a lipophilic 4-substituent, such as tri¯uoromethoxy or chloro.
The 5-substituent of the tetrahydrothieno[3,2-c]pyridine ring has to be a substituted benzoyl; anisoyl and (E)-3-furan-3-ylacryloyl
are the best of the investigated groups. Substitution in the benzoyl ortho position seems to be forbidden, whereas substitution in the meta
position is tolerated only if a methoxy para substituent is present. These SAR ®ndings were parallel to those obtained in the 4,5,6,7-
tetrahydrothieno[2,3-c]pyridine system. Enantioselectivity in enzyme recognition was observed and the activity resided in all cases only
in one of the enantiomers. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
inorganic phosphate and glucose, respectively.6,7 It has
been shown that the rate of hydrolysis of G-6-P and the
hepatic glucose output were increased under diabetic
conditions.8,9 The increased activity could mainly be
accounted for by increased G-6-Pase catalytic enzyme
protein.10,11 Recently, peroxyvanadium compounds have
been found to be potent inhibitors of the G-6-Pase cata-
lytic enzyme,12 which makes this site a potential target in
control of excess glucose production seen in diabetes.12
Regulation of hepatic glucose production is an attractive
way to reduce the increased blood glucose levels seen in
type 2 diabetic patients.1 4 Inhibition of di€erent liver
enzymes involved in either glycogenolysis or gluconeo-
genesis seem therefore to be potential targets for devel-
opment of new drugs as alternatives to the existing
treatments of type 2 diabetes.5 Glucose-6-phosphatase
(G-6-Pase), which is predominantly found in the liver,
catalyses the terminal step in both glycogenolysis and
gluconeogenesis by converting glucose-6-phosphate (G-
6-P) to glucose, and is, therefore, a key enzyme in blood
glucose homeostasis.1 The G-6-Pase is a multicomponent
system comprising the G-6-Pase catalytic enzyme with its
active site located at the luminal site of the endoplasmic
reticulum, a speci®c transporter T1 which mediates entry
of G-6-P into the luminal compartment, and transpor-
ters T2 and T3 which mediates export to the cytosol of
Very few compounds with activity on the G-6-Pase
enzyme complex are known. A series of chlorogenic acid
derivatives have been found to inhibit the T1 translocase,13
2-hydroxy-5-nitrobenzaldehyde is described14 as an
inhibitor of both T1 and T2, and a report on a carbohy-
drate (6R-C-methylglucose) as a speci®c inhibitor of the
G-6-Pase enzyme has been published.15 The Ki of the 6R-
C-methylglucose was reported to be in the mM range.
We have screened our compound library for compounds
which are able to inhibit the activity of the G-6-Pase
catalytic site, and in this publication we report a series
of 4,5,6,7-tetrahydrothienopyridines, the ®rst report on
*Corresponding author. Tel.: +45 44 43 48 94; fax: +45 44 66 34 50;
e-mail: pem@novo.dk
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00153-X

2278
P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
non-carbohydrate compounds with potent activity at
the catalytic site only.
the described compounds.16 19 Ohkubo et al.18 have
described a series of 4,5,6,7-tetrahydrothieno[3,2-c]pyr-
idines bearing hydrogen, methyl or ethyl on the 5-posi-
tion with anticonvulsant activities. Russell20 reported a
series of 5-aminoalkyl substituted tetrahydrothieno-
pyridines as antiarrythmic agents and Vecchietti et al.21
have investigated a series of 4-aminoalkyl-5-arylacetyl
derivatives and found that they possess potent activity
as kappa opioid analgesics.
Neither 4,5,6,7-tetrahydrothieno[3,2-c]pyridines carry-
ing an acyl substituent in the 5-position and a cyclic
substituent in the 4-position nor 4,5,6,7-tetrahydro-
thieno[2,3-c]pyridines carrying an acyl substituent in the
6-position and a cyclic substituent in the 7-position are
well described in the literature.
The preparation of 4-substituted 4,5,6,7-tetrahydro-
thieno[3,2-c]pyridines have been reported, but no 4-aryl-5-
acyl-4,5,6,7-tetrahydrothieno[3,2-c]pyridines were among
Chemistry
Some 4-substituted 4,5,6,7-tetrahydrothieno[3,2-c]thieno-
pyridines (4-chlorophenyl 3j, 4-methoxyphenyl, 3k, 4-tri-
¯uoromethylphenyl 3l as well as 4-tri¯uoromethoxy-
phenyl 3m) were commercially available, new compounds
were prepared via the imine and amide routes as depicted
in Schemes 1 and 2, respectively.
The imine route (Scheme 1 and Table 1, entries a±f, h,
n±y) gave good yields for benzaldehyde derivatives (70±
100%) while thiophene, cyclohexyl as well as 1-methyl-
piperidine-4-carboxaldehydes gave low yields both at
room temperature and at elevated temperatures. These
compounds were more conveniently prepared as shown
in Scheme 2 using the amide route. Commercially
available 2-(2-thienyl)ethylamine (1) or 2-(3-thienyl)
ethylamine (1a) were treated with an aldehyde under
Scheme 1. Preparation of 4,5,6,7-tetrahydrothienopyridines, imine
route. (i) R1CHO, EtOH, TEA; (ii) TFA, rt; (iii) a. R1CHO, PhH, ~,
Dean±Stark, b. TFA.
Scheme 2. Preparation of 4,5,6,7-tetrahydrothieno[3,2-c] pyridines, amide route. (i) a. R1COCl, TEA, DCM, or b. R1CO2H, HOBt, EDAC, DMF,
(ii) POCl3/P2O5, ~; (iii) NaBH4, MeOH.
Table 1. Physicochemical data for 2a,b,d and 3a±h, n±ye
Compounds
R1
Yield 2 (%)
Mp (ꢀC)
Yielda 3 (%)
Mp (ꢀC)
[3,2-c]:
a
b
c
d
e
f
g
4-Dimethylaminophenyl
4-Nitrophenyl
4-Pyridyl
2-Thienyl
5-Chloro-2-thienyl
Cyclohexyl
77
76
76.8±77
83.9±84.4
95
85
71
15
10
9 (81)b
95±98
128±129.5
81.8±83.8
96±97.3
Oil
Oil
Oil
83.8±84.2
36
NDe
b
1-Methyl-4-piperidinyl
Phenyl
69
24
h
[2,3-c]:
n
o
p
q
r
s
t
u
v
x
y
4-Tri¯uoromethoxyphenyl
4-Chlorophenyl
4-Methoxyphenyl
2,4-Dichlorophenyl
2,4-Dimethoxyphenyl
3-Methoxyphenyl
3-Fluoro-4-methoxyphenyl
3,4-Methylenedioxyphenyl
2-Methoxyphenyl
16
81
42
91
95
100
97
100
100
190±195c
93.6±93.8
Oil
d
Oil
Oil
Oil
Oil
Oil
Oil
Oil
Oil
d
d
d
d
d
d
3,5-Dichlorophenyl
3-Fluorophenyl
94
100
d
aYields refer to imine route overall yields.
bAmide route, overall yields.
cHydrogen oxalate.
dYields determined by HPLC (ELS-detection).
eND: Not determined.

P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
2279
Preparation of compounds 3f±g proceeded in good yields
using the amide route depicted in Scheme 2. Amides 4
were formed either from reaction of 1 with the appro-
priate carboxylic acid chloride/triethylamine or from car-
boxylic acids using HOBt/EDAC as coupling reagents.
Cyclisation of amides 4 with POCl3 or mixtures of P2O5/
POCl3 gave yields of >80%, subsequent reduction with
NaBH4 gave the desired tetrahydrothieno[3,2-c]pyr-
idines in yields >69%.
The target substituted 4,5,6,7-tetrahydrothieno[3,2-c]-
and -[2,3-c]pyridines 7±44 and 47±57 were prepared from
the various substituted 4,5,6,7-tetrahydrothienopyr-
idines 3a±y by means of conventional amide formation
procedures (Scheme 3). In order to prepare a large num-
ber of compounds with a large diversity for exploring the
structural requirements of the inhibitors for activity, the
procedure was modi®ed to work in a solution phase
parallel fashion. This method was conveniently used to
prepare >200 analogues of this type. Selected com-
pounds were prepared and characterised fully while all
the library compounds were characterised by HPLC-
MS. Biological results obtained from the active library
compounds did not di€er signi®cantly from the corre-
sponding data obtained from pure, characterised com-
pounds (Table 7). (Test solutions were made by
weighing out the evaporation residues). Further, selec-
ted compounds were resolved into the pure enantiomers
using semi-preparative HPLC. Generally, HPLC-reso-
lution of an active library compound gave two optical
Scheme 3. Preparation of substituted 4,5,6,7-tetrahydrothienopyr-
idines 7±44 and 47±57. (i) R2COCl, Et3N, DMF, or R2CO2H, HOBt,
EDAC, DMF.
formation of an imine which was isolated either by ®l-
tration or by concentration in vacuo. The imines were
cyclised immediately without any puri®cation. The crude
imines were highly prone towards hydrolysis to amine
and aldehyde, as column chromatography resulted in
hydrolysis to the starting materials.
Ring closure of the imines were performed in CF3CO2H
(TFA) either at room temperature or at re¯ux to a€ord
the desired 4,5,6,7-tetrahydro[3,2-c]- and -[2,3-c]pyridines
(3a±f, h, n±y).
Table 2. 4,5,6,7-Tetrahydrothieno[3,2-c]pyridines. SAR of the 5-benzoyl group
HPLC-MSa
Compound
R1
R2
IC50 (mM)
Rt (min)
M+1 (m/z)
ELS purity (%)
7
8
9
Cl
OCF3
OMe
CF3
4-OMe
4-OMe
4-OMe
4-OMe
3-OMe
2-Cl
4-OEt
4-F
4-Cl
4-NH2
4-OH
4-CH2OH
4-CH2OH
4-Ph
4-tBu
0.62
0.81
0.80
0.51
>100
>78
1.1
15.90
16.43
14.73
16.15
16.15
15.10
17.03
16.03
16.83
11.65
14.10
13.73
12.10
18.00
18.48
15.97
16.78
17.68
15.98
15.72
9.68
384
434
380
418
418
384
448
372
388
369
370
384
380
480
460
354
468
472
448
443
444
100
93
100
97
98
100
88
100
100
100
59
100
100
96
96
99
94
92
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
CF3
OMe
OCF3
Cl
Cl
Cl
Cl
Cl
OMe
OCF3
OCF3
Cl
OCF3
OCF3
OCF3
OCF3
OCF3
1.8
3.3
1.4
1.7
0.51
>100
>100
>100
4.9
H
3-Cl-4-OMe
3,4-di-Cl
3,4-OCH2O
5-Indolylb
5-Benzimidazolylc
1.7
>100
2.5
2.7
94
94
93
4.4
aHPLC-MS conditions: See Experimental.
bWhole 5-substituent is 5-indolylcarbonyl.
cWhole 5-substituent is 5-benzimidazolylcarbonyl.

2280
P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
Table 3. 4,5,6,7-Tetrahydrothieno[3,2-c]pyridines. SAR of the 4-phenyl group
HPLC-MSa
Compound
R1
R2
IC50 (mM)
Rt (min)
14.83
M+1 (m/z)
ELS purity (%)
92
28
OMe
0.61
350
354
350
29
30
Cl
1.1
1.9
15.90
12.33
77
82
CH2OH
31
32
OMe
OMe
0.75
1.2
9.55
393
395
85
66
14.80
33
34
35
36
CH2OH
OMe
100
6.0
12.44
14.58
16.38
8.50
395
356
391
351
48
79
58
91
OMe
6.1
OMe
>100
37
38
OMe
OMe
>100
>100
8.75
371
356
100
16.18
>98
aHPLC-MS conditions: See Experimental.
isomers of which one was approximately 2-fold more
potent as compared to the library compound, whereas
the other isomer was inactive.
tetrahydrothieno[2,3-c]pyridines were prepared and tes-
ted, along with other substituents/substitution patterns
in the 7-aryl group as listed in Table 5. Lastly, selected
compounds were resolved into their optically pure
enantiomers as listed in Table 6.
Initially we investigated the SAR of the 5-benzoyl group
using the commercially available thieno[3,2-c]pyridines
3j±m as seen in Table 2 where selected compounds are
listed. Using the thieno[3,2-c]pyridines 3a±h the SAR of
the 4-phenyl group was explored further as seen in
Table 3. To investigate compounds with a spacer
between the benzene ring and the carbonyl group of the
benzamides a series of compounds were prepared, some
of which are listed in Table 4. The isomeric 4,5,6,7-
Biological results
Initial information on the ability of these compounds to
inhibit the glucose-6-phosphatase activity at the cataly-
tic site using TritonX-100 disrupted pig microsomes
were obtained using ®ve concentrations (0, 1, 25, 50, 100
mM) of test compounds. Compounds of interest were

P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
2281
Table 4. 4,5,6,7-Tetrahydrothieno[3,2-c]pyridines. Further SAR of the 5-benzoyl group
HPLC-MSa
M+1 (m/z)
Compound
R1
R2
IC50 (mM)
Rt (min)
16.25
ELS purity (%)
93
39
OCF3
>100
448
40
41
42
43
OCF3
CF3
2.3
>100
>100
1.8
16.64
16.41
16.54
14.50
460
446
444
366
88
59
89
93
CF3
OMe
44
OCF3
>100
16.03
440
96
aHPLC-MS conditions: See Experimental.
further evaluated by determination of full dose±response
curves to calculate the IC50 values. Moreover, in some
cases the inhibitory mode of action was investigated.
For these compounds, the mode of action was always
competitive (data not shown).
capable of being a hydrogen bond donor and/or accep-
tor, compare for example 23 with 24 (Table 2). Activity
is lost, however, compared to the 4-methoxy group.
Ring systems, as in 27, 26 and 25 are also allowed with
respect to activity.
Many other acyl groups than substituted benzoyl have
been tried and the general picture is that positioning of a
methoxy group or another hydrogen bond acceptor
within narrow limits is crucial for activity. Thus, 4-meth-
oxyphenylacetyl (39) and its homologue 3-(4-methoxy-
phenyl)propionyl (41) are inactive, whereas 4-methox-
ycinnamyl (40) is active (Table 4). Moving this methoxy
group to the meta position gave an inactive compound
(42). It was not possible with other aromatic sub-
stituents to regain activity for any of the three men-
tioned `non-benzoyl' systems. Another acyl group that
confers good activity is the (E)-3-furan-3-ylacryloyl
group (43). This group presents the oxygen atom of the
furan ring roughly in the same space, as does the 4-
methoxybenzoyl group. It is remarkable that a fully
saturated acyl group (4-methoxycyclohexylcarboxyl, 44,
Table 4) is devoid of any activity, indicating that inter-
actions with the p-electrons of the aromatic ring or
those of the 3-furan-3-ylacryloyl group are implicated in
the enzyme inhibition.
Discussion
When comparing 7, 8, 9 and 10 (Table 2), it is evident
that the p-substituent of the 4-phenyl group has no sig-
ni®cant impact on potency.
The substitution pattern of the 5-benzoyl group is much
more important for inhibitory activity, the 4-methoxy
group being by far the best. Moving this group to the 3-
position (11) results in loss of all activity. A 4-Cl-sub-
stituent is also tolerated, although about 5-fold less potent
(15). Similarly, moving this group to the 2-position (12)
all activity is lost. An unsubstituted 5-benzoyl group
(22) is roughly as potent as a 4-chlorobenzoyl group.
As can be seen in Table 2, many other groups are tol-
erated in this para-position, for example hydroxy (17),
hydroxymethyl (18), amino (16), and ¯uoro (14) groups
confer good activity. There are, however, steric con-
straints, as compounds with more bulky aliphatic and
aromatic para substituents are not active, for example
tert-butyl (21) and phenyl (20).
It is evident that the combination of `allowed' sub-
stituents on the 4-phenyl and 5-benzoyl groups is not
additive with respect to inhibitory activity, as exempli®ed
by 18 and 19. With the 4-methoxyphenyl building block
(3k) there were generally much fewer library hits than
Compounds with 3,4-disubstituted benzoyl groups can
be active, provided that the 4-substituent is a heteroatom

2282
P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
Table 5. 4,5,6,7-Tetrahydrothieno[2,3-c]pyridines. SAR of the 7-aryl
group
Table 6. Optically resolved compounds
Compound
ee
([a]2D0
IC50
(mM)
Column Rt Flow
(min) (mL/min)
)
7A
100%
98.0% ( 170.0)
>99.8%
>100
Wb,g 44±50b
12
1a
12
1a
10
1a
10
1a
10
1a
10
1a
10
1a
10
1a
10
1a
10
1a
10
1a
10
1a
15.5a
62±72
20.8a
24±30
12.0a
41±47
15.8a
27±30
9.3a
39±45
12.3a
40±43
12.6a
55±59
16.4a
28±32
9.9a
51±58
15.5a
24±29
8.9a
42±50
12.5a
16±19
8.4a
7B
0.31
0.31
>100
0.23
>100
0.18
>100
0.63
>100
1.0
>100
>100
8.70
>100
0.91
1.6
Wb
Wb
Wb
We
We
Wb
Wb
Wb
Wb
Wb
Wb
Cc,h
Cc
HPLC-MSa
8A
Compound
R1
IC50
(mM)
Rt
(min)
M+1
(m/z)
ELS
purity
8B
99.1%
>99.9%
98.8%
9A
a
47
0.35
16.50
Ðb
434
Ðb
Ðb
417
409
380
397
394
380
418
368
93
9B
10A
10B
15A
15B
17A
17B
27A
27B
43A
43B
46A
46B
47A
47B
49A
49B
>99.0%
99.0%
48
49
50
51
52
53
54
55
56
57
0.26
0.80
Ðb
100%
99.5%
Ðb
Ðb
100%
99.4%
>100
6.71
5.82
5.39
5.97
5.87
5.38
6.97
5.38
50%
77%
73%
64%
73%
75%
80%
67%
100%
6
0.6a
6
>99.0%
99.8%
27±35
14.7a
20±24
12.9a
30±36
19.8a
27±32
9.2a
0.6a
6
Cd
>100
0.55a
6
99.8%
Cd
0.55a
10
1a
100% (+175.4)
Wb
Wb
Wf
Wf
Wf
Wf
4.9
>99.5% ( 170.6) >100
62±74
17.0a
32±36
10.5a
45±51
14.5a
26±30
9.1a
10
1a
10
1a
10
1a
10
1a
>99.8%
>98.0%
100%
0.16
>100
0.24
>100
4.6
99.2%
38±43
11.9a
10
1a
0.78
aRetention times and ¯ow rates in italics refers to data from analytic
runs.
bEluted isocratically with a mixture of heptane: 2-propanol (1:1).
cEluted isocratically with a mixture of heptane:ethanol:diethylamine
(70:30:0.1).
>100
dEluted isocratically with a mixture of heptane:ethanol (3:2).
eEluted isocratically with a mixture of heptane:ethanol (1:1).
fEluted isocratically with a mixture of heptane:ethanol (4:1).
gW: Column: (R,R)-Whelk-O (Regis). 21.1Â250 mm for preparative
runs, 4.6 250 mm for analytical runs.
>100
25
hC: Column: Chiralpak AS. 20Â250 mm for preparative runs, 4.6Â
250 mm for analytical runs.
system (Table 3) originating from the intermediates 3a±i.
An unsubstituted phenyl group is as potent as any of
the 4-Cl-, 4-CF3-, and 4-OCF3-phenyl systems (compare
for example 28 with 8). Even with less favourable sub-
stituents on the 5-benzoyl group, such as 4-chloro- (29)
and hydroxymethyl (30), potency is maintained.
aHPLC-MS metod B used for compound 50±57. See Experimental.
bSee Table 7 for data.
obtained with 3j, 3l and 3m, and only compounds with
anisoyl and (E)-3-furan-3-ylacryloyl as 5-substituents
consistently showed up as potent library hits.
Likewise, a 4-dimethylamino substituent on the 4-
phenyl group (31) is about as good as the unsubstituted
or 4-Cl-, 4-CF3-, and 4-OCF3-phenyl systems. In con-
trast, neither a 4-pyridyl group (36), nor a 1-methyl-4-
We then turned to investigate the SAR of the 4-sub-
stituent of the 4,5,6,7-tetrahydrothieno[3,2-c]pyridine

P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
2283
Table 7. Physicochemical and biological data for fully characterised 4,5,6,7-tetrahydrothieno[3,2-c]- and-[2,3-c]pyridines
Compound
R1
R2
Mp
(ꢀC)
Yield
(%)
Anal.
IC50
(mM)
8
9
4-Tri¯uoromethoxyphenyl
4-Methoxyphenyl
4-Tri¯uoromethylphenyl
4-Chlorophenyl
4-Tri¯uoromethoxyphenyl
4-Methoxyphenyl
4-Chlorophenyl
4-Methoxyphenyl
4-Methoxyphenyl
4-Methoxyphenyl
4-Hydroxyphenyl
5-Benzimidazolyl
2-(3-Furanyl)ethenyl
4-Methoxyphenyl
4-Methoxyphenyl
68±70
126.5±127.3
86.1±88.8
207±212
187±>200
75±77
98
84
88
99
87
20
91
81
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
0.27
0.80
0.34
1.7
4.4
1.2
10
17
27
43
48
49
129.5±130.0
Oil
0.26
0.80
4-Methoxyphenyl
piperidyl group (37) are tolerated at all, indicating
unfavourable contacts with the basic nitrogens in these
systems. In the case of 37, the inactivity can also be
attributed to the absence of aromatic p-electrons (com-
pare with 38). Similarly as seen with 4-(4-methoxy-
phenyl) group, a 4-(4-nitrophenyl) group is only active
when paired with the 4-methoxybenzoyl group in the
5-position confers activity (32), and other `allowed'
substituents on the 5-benzoyl group are not active at all
(e.g. 4-hydroxymethyl 33).
The substituents of the 7-phenyl group of the [2,3-c]
series have a large impact on the inhibitory activity
(Table 5). The 4-tri¯uoromethoxy (47) and 4-chloro (48)
groups are equipotent and marginally better than the 4-
methoxy group (49). An ortho substituent is not toler-
ated at all (50, 51 and 55) while meta substituted phenyl
groups are active, although less active than the para
substituted ones. Compare for example 49 with 52
which is about 6-fold less active. A 3,5-dichlorophenyl
group (56) is not tolerated at all while a 3-¯uorophenyl
group is almost inactive (57). A 3,4-disubstituted phenyl
group is also tolerated. When comparing the 3-¯uoro-4-
methoxyphenyl group (53) with the 4-methoxyphenyl
Saturated systems replacing the 4-phenyl group are not
tolerated at all, since the cyclohexyl system only give
inactive compounds (e.g., 38), indicating that also this
aromatic ring system is involved in recognition and
inhibition of the enzyme.
A 2-thienyl (34) group as well as a 5-chloro-2-thienyl (35)
group as the 4-substituent of the 4,5,6,7-tetrahydro-
thieno[3,2-c]pyridine system is clearly less tolerated (by
about 10-fold) as compared to the parent phenyl or 4-
(4-chlorophenyl) groups.
Further, we have prepared a series of analogues wherein
the central tetrahydrothieno[3,2-c]pyridine ring was
replaced by a 1,2,3,4-tetrahydro-isoquinoline system, as
seen in 45. These analogues were all virtually inactive,
indicating that it is also in the core ring system that ben-
zene is not a good thiophene bioisoster for this target.
We then investigated the in¯uence of changing the
4,5,6,7-tetrahydro[3,2-c]pyridine core ring system to the
isomeric [2,3-c] ring system. The SAR of the 5-acyl
group of the [3,2-c] series was nicely paralleled with that
of the 6-acyl group of the [2,3-c] series, the 4-methoxy-
benzoyl group being the best followed by 2-(3-fur-
anyl)acryloyl>5-chlorothiophen-2-ylcarbonyl>4-chloro-
benzoyl (data not shown).
Figure 1. Dose±response curves for inhibition of pig G-6-Pase catalytic
enzyme activity by 10A, 10B, 47A and 47B. Dose-response curves for
inhibition of pig glucose-6-phosphatase activities using disrupted
microsomes by 10a(~) and 10b (&) (A) and 47a (~) and 47b (&) (B).
Glucose-6-phosphatase activities were measured using 0.5 mM glu-
cose-6-phosphatase as the substrate. Data are from 3±4 individual
experiments with SEM indicated by vertical bars if they extend beyond
the symbols. IC50 values (nM) were calculated to be 183Æ36 and
159Æ26 for compound 10a and 47a respectively.

2284
P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
group (49) it is evident, however, that only some activity
is lost. A 3,4-methylenedioxyphenyl group (54) is equi-
potent with the 4-methoxyphenyl group.
were obtained using a Perkin±Elmer Polarimeter Model
241. Microanalyses were performed by Novo Nordisk
Analytical Department.
As these compounds are so potent G-6-Pase catalytic
enzyme inhibitors as racemates, we wanted to study
the G-6-Pase inhibition properties of the individual
enantiomers. Generally the compounds were eciently
resolved with baseline separations for minutes as seen in
Table 6. The ®rst eluting enantiomers are denoted `A'
and the second eluting enantiomers are denoted `B'. In
all cases the activity of the racemate could be attributed
to the activity of one of the enantiomers, as the other
enantiomer was inactive. Also, as could be expected, the
activities of the active enantiomers were about twice as
much as the activity of the racemates. Representative
dose-response curves are given in Figure 1.
The HPLC-MS analyses were performed on a PE Sciex
API 100 LC/MS System using a WatersTM 3Â150 mm 3.5
m C-18 Symmetry column and positive ionspray with a
¯ow rate at 20 mL/min. The column was eluted with a
linear gradient of 5±90% A, 85±0% B and 10% C in 15
min at a ¯ow rate of 1 mL/min (solvent A=acetonitrile,
solvent B=water and solvent C=0.1% tri¯uoroacetic
acid in water).
The HPLC-MS method B used for analysing com-
pounds 50±57 is identical to that described above with
the following settings: Column: YMC ODS-A 120A s-
5m 3Â50 mm, gradient 5±90% acetonitrile in 0.05%
TFA linearly during 7.5 min at 1.5 mL/min.
Further, we have obtained information on absolute
con®guration (X-ray) of 46A and 10A. These com-
pounds were both found to be of (S) con®guration.
1-(4-Tri¯uoromethoxyphenyl)-1,2,3,4-tetrahydroisoquino-
line (45) was prepared using the `imine route' (similarly
as previously described22).
[4-(4-Methoxyphenyl)-4,5,6,7-tetrahydrothieno[3,2-c]-
pyridin-5-yl]-(4-tri¯uoromethylphenyl)-methanone (46)
is commercially available (Maybridge).
Conclusion
In summary, we have described the ®rst class of potent
non-carbohydrate glucose-6-phosphatase catalytic enzyme
inhibitors, 4-aryl-5-benzoyl-4,5,6,7-tetrahydrothieno[3, 2-
c]pyridines and 7 - aryl - 6 - benzoyl - 4,5,6,7 - tetrahydro-
thieno[2,3-c]pyridines.
Dimethyl-{4-[(2-thiophen-2-ylethylimino)methyl]phenyl}
amine, (2a). 2-(2-Thienyl)-ethylamine (5 g, 39.4 mmol),
4-dimethylaminobenzaldehyde (5.9 g, 94 mmol), triethy-
lamine (6 mL) and ethanol (150 mL) were mixed at room
temperature. The reaction mixture was stirred at room
temperature for 15 h. The reaction mixture was con-
centrated to 75 mL by evaporation in vacuo and the
solid formed was ®ltered and dried to a€ord 6.82 g
(67%) dimethyl-{4-[(2-thiophen-2-ylethylimino)methyl]-
phenyl}amine, (2a). Mp: 76.8±77 ꢀC. 1H NMR (200
MHz, CDCl3) d 3.0 (s, 6H), 3.15 (t, 2H), 3.8 (t, 2H),
6.65 (d, 2H), 6.8 (d, 1H), 6.95 (dd; 1H), 7.1 (dd; 1H), 7.6
(d, 2H), 8.1 (s, 1H), 13C NMR (CDCl3) d 32.3, 40.7,
63.3, 112.1, 124.0, 124.7, 125.6, 127.2, 130.1, 143.1,
152.5, 162.26. Calcd for C15H18N2S: C, 69.73; H, 7.02;
N, 10.84. Found: C, 69.84; H, 7.13; N, 10.84.
The structure±activity relationship of these compounds
indicates that: (1) Tetrahydrothieno[3,2-c]- and -[2,3-
c]pyridine core ring systems are much better than the cor-
responding benzo analogue, 1,2,3,4-tetrahydro-isoquino-
line. (2) The 4-substituent of the tetrahydrothieno[3,2-
c]pyridine ring system (or the 7-substituent of the [2,3-c]
system) has to be a phenyl group, optionally substituted
with a lipophilic 4-substituent, such as tri¯uoromethoxy
or chloro. A thienyl group is less favoured. Saturated
ring systems are not allowed. (3) The 5-substituent of
the tetrahydrothieno[3,2-c]pyridine ring system (or the
6-substituent of the [2,3-c] system) has to be a 4-sub-
stituted benzoyl such as anisoyl, although compounds
with a 5-(E)-3-furan-3-ylacryloyl group are roughly
equipotent. Substitution in the benzoyl meta position is
less tolerated, whereas substitution in the benzoyl ortho
position seems to be forbidden (at least in the [3,2-c]
series). Saturated ring systems are not allowed. (4) The
observed high enantioselectivity in enzyme recognition
and inhibition is in good accordance with the high
potency of the inhibitors.
Dimethyl-[4-(4,5,6,7-tetrahydrothieno[3,2-c]pyridin-4-yl)
phenyl]amine (3a). Dimethyl-{4-[(2-thiophen-2-yl-ethyl-
imino)methyl]phenyl}amine (1 g, 3.9 mmol) was added to
TFA (20 mL) at once (exothermic reaction). The reaction
mixture was stirred at room temperature for 72 h, and
then evaporated in vacuo. The crude oil was suspended
in dichloromethane (75 mL) and extracted with 1N
hydrochloric acid (50 mL). The aqueous phase was
added to 2N sodium hydroxide to pH 10, then extracted
with dichloromethane (3Â125 mL). The organic phase
was dried with MgSO4, ®ltered, evaporated in vacuo to
a€ord 0.96 g (95%) dimethyl-[4-(4,5,6,7-tetrahydro-
thieno[3,2-c]pyridin-4-yl)phenyl]amine (3a). Mp: 95±
98 ꢀC. 1H NMR (200 MHz, CDCl3) d 1.75 (broad s, 1H),
2.6±3.3 (m, 4H), 2.75 (s, 6H), 4.7 (s, 1H), 6.45 (d, 1H), 6.5
(d, 2H), 6.9 (d, 1H), 7.1 (d, 2H); 13C NMR (CDCl3) d
26.7, 41.2, 43.1, 60.0, 113.0, 121.9, 127.1, 129.6, 132.3,
135.2, 150.5. Calcd for C15H18N2S: C, 69.73; H, 7.02; N,
10.84. Found: C, 69.20; H, 7.06; N, 10.80.
Experimental
General
Melting points were determined on a Buchi 535 appa-
1
ratus and are uncorrected. H and 13C NMR spectra
were obtained on Bruker Avance DRX 300 and DPX
200 systems (300 MHz and 200 MHz, respectively)
operating at room temperature. Optical rotation data

P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
2285
(4-Nitrobenzylidene)-(2-thiophen-2-ylethyl)amine (2b). 2-
(2-Thienyl)ethylamine (5 g, 39.4 mmol), 4-nitro-
benzaldehyde (6 g, 39.7 mmol) and triethylamine (10
mL) were dissolved in ethanol (100 mL). The reaction
mixture was stirred at room temperature for 48 h. The
solid formed was ®ltered and dried to a€ord 7.75 g
(76%) (4-nitrobenzylidene)-(2-thiophen-2-ylethyl)amine
6.85 (d, 1H), 6.9 (dd; 1H), 7.05 (dd, 1H), 7.1 (d, 1H),
7.25 (d, 1H), 7.45 (d, 1H), 8.25 (s, 1H); 13C NMR
(CDCl3) d: 31.8, 63.0, 124.1, 125.7, 127.2, 127.8, 129.4,
131.0, 142.6, 142.8, 155.6.
4-(Thiophen-2-yl)-4,5,6,7-tetrahydrothieno[3,2-c]pyridine
(3d). The imine 2d (1.26 g, 5.7 mmol) was added to tri-
¯uoroacetic acid (25 mL) and the reaction mixture was
stirred at 60 ꢀC for 12 h. The reaction mixture was eva-
porated in vacuo. The residue was dissolved in di-
chloromethane (30 mL) and washed with 2N sodium
hydroxide (30 mL). The aqueous phase was extracted
with dichloromethane (3Â25 mL). The combined
organic phases were dried with MgSO4, ®ltered and
evaporated in vacuo. The residue (1.15 g) was puri®ed
by column chromatography on silica gel eluting with a
mixture of dichloromethane:methanol (19:1) to a€ord
0.18 g (5%) 4-(thiophen-2-yl)-4,5,6,7-tetrahydrothieno
[3,2-c]pyridine (3d). Mp: 96±97.3 ꢀC. 1H NMR (200
MHz, CDCl3) d 2.0 (s, 1H), 2.85 (dt; 2H), 3.1±3.35 (m,
2H), 5.3 (s, 1H), 6.15 (d, 1H), 6.9 (s, 1H), 6.95 (d, 1H),
7.05 (d, 1H), 7.2 (dd; 1H); 13C NMR (CDCl3) d: 26.3,
42.1, 55.1, 122.3, 125.3, 125.7, 126.7, 126.8, 135.3, 136.7,
148.5.
(2b). Mp: 83.9±84.4 ꢀC. H NMR (200 MHz, CDCl3) d
1
3.2 (t, 2H), 3.95 (t, 2H), 6.3 (dd, 1H), 6.4 (dd, 1H), 7.4
(d, 2H), 8.3 (d, 2H), 8.25 (s, 1H); 13C NMR (CDCl3) d
31.6, 63.3, 124.2, 124.3, 125.8, 126.2, 127.2, 129.2, 142.0,
142.3, 160.0.
4-(4-Nitrophenyl)-4,5,6,7-tetrahydrothieno[3,2-c]pyridine
tri¯uoroacetate (3b). (4-Nitrobenzylidene)-(2-thiophen-2-
ylethyl)amine (1 g, 3.8 mmol) was added to tri¯uoroacetic
acid (100 mL) at once (exothermic reaction). The reac-
tion mixture was stirred at room temperature for 72 h,
then evaporated in vacuo. Crystallization from a mix-
ture of diethyl ether and dichloromethane a€orded 1.2 g
(85%) 4-(4-nitrophenyl)-4,5,6,7-tetrahydrothieno[3,2-
c]pyridine tri¯uoroacetate (3b). Mp: 128±129.5 ꢀC. H
1
NMR (200 MHz, CDCl3) d 3.3 (m, 2H), 3.6 (m, 2H), 4.6
(broad s, 1H), 5.3 (d, 1H), 7.3 (d, 1H), 7.6 (d, 2H), 8.3
(d, 2H); 13C NMR (CDCl3) d 26.1, 45.1, 62.0, 128.5,
129.5, 129.8, 133.7, 135.1, 138.0, 146.2. Calcd for C15
H13N2SF3O4: C, 48.13; H, 3.48; N, 7.49. Found C,
47.96; H, 3.57; N, 7.32.
4-(5-Chlorothiophen-2-yl)-4,5,6,7-tetrahydrothieno[3,2-c]
pyridine (3e). 2-(2-Thienyl)-ethylamine (1 g, 7.9 mmol)
and 5-chlorothiohene-2-carboxaldehyde (1.15 g, 7.9
mmol) were mixed (exothermic reaction). The separated
water was decanted and the remaining oil was heated
until re¯ux. The mixture was allowed to cool to room
temperature. Then TFA (10 mL) was added under stir-
ring, and then re¯uxed for 16 h. The reaction mixture
was evaporated in vacuo, dissolved in dichloromethane
(50 mL) washed with 1N NaOH (50 mL). The aqueous
phase was extracted with dichloromethane (50 mL). The
combined organic phases were dried with MgSO4, ®l-
tered, and evaporated in vacuo. The residue (1.2 g) was
puri®ed on a silica gel column eluting ®rst with di-
chloromethane and then with a mixture of dichlor-
omethane:methanol (19:1) giving 0.2 g (9.9%) 4 - (5 -
chlorothiophen-2-yl)-4,5,6,7-tetrahydrothieno[3,2-c]pyr-
4-Pyridin-4-yl-4,5,6,7-tetrahydrothieno[3,2-c]pyridine (3c).
2-(2-Thienyl)ethylamine (2 g, 15.7 mmol), pyridine-4-
carboxaldehyde (1.68 g, 15.7 mmol), triethylamine (1
mL) and ethanol (15 mL) were mixed and the reaction
mixture was stirred at room temperature for 15 h, then
evaporated in vacuo. The crude oil was added to tri-
¯uoroacetic acid (75 m) at once (exothermic reaction).
The reaction mixture was stirred at room temperature
for 0.5 h, then evaporated in vacuo. The residue was
dissolved in dichloromethane (150 mL) and washed
with 2N sodium hydroxide (100 mL). The aqueous
phase was extracted with dichloromethane (3Â50 mL).
The combined organic phases were dried with MgSO4,
®ltered and evaporated in vacuo to give an oil (3.21 g)
which was crystallised from a mixture of dichloro-
methane and hexane to a€ord 2.4 g (71%) 4-pyridin-4-
yl-4,5,6,7-tetrahydrothieno[3,2-c]pyridine (3c). Mp:
81.8±83.8 ꢀC. 1H NMR (200 MHz, CDCl3) d 2.15
(broad s, 2H), 2.7±3.4 (m, 4H), 5.0 (s, 1H), 6.45 (d, 1H),
7.15 (d, 1H), 7.25 (dd; 2H), 8.05 (dd; 2H); 13C NMR
(CDCl3) d 26.3, 42.6, 59.3, 122.8, 123.8, 126.2, 135.3,
1
idine (3e) as an oil. H NMR (200 MHz, CDCl3) d 2.15
(broad s, 1H), 2.7 (m, 2H), 3.0±3.3 (m, 2H), 5.2 (s, 1H),
6.65 (m, 2.5H), 7.0 (d, 1.5H); 13C NMR (CDCl3) d 41.9,
55.3, 122.6, 124.8, 125.8, 126.4, 129.7, 135.3, 135.6,
147.6. Calcd for C11H10NS2Cl: C, 51.65; H, 3.94; N,
5.48. Found C, 51.75; H, 3.80; N, 5.44.
4-Cyclohexyl-4,5,6,7-tetrahydrothieno[3,2-c]pyridine (3f,
imine route). 2-(2-Thienyl)ethylamine (2 g, 15.7 mmol)
and cyclohexylcarboxaldehyde (1.76 g, 15.7 mmol) were
mixed in ethanol (15 mL). The reaction mixture was stir-
red for 5 days. The solvent was removed in vacuo. The
residual brown oil was added to tri¯uoroacetic acid (40
mL) and stirred for 72 h at room temperature. The reac-
tion mixture was evaporated in vacuo. The remaining oil
was added to dichloromethane (50 mL) washed with 1N
NaOH (100 mL). The aqueous phase was extracted with
dichloromethane (30 mL). The combined dichloro-
methane phases were added to activated carbon, dried
with MgSO4, ®ltered and evaporated in vacuo. The resi-
due (2.91 g) was puri®ed by column chromatography on
.
136.0, 150.4, 152.8. Calcd for C12H12N2S 1H2O: C,
61.51; H, 6.02; N, 11.96. Found C, 61.64; H, 5.98; N,
11.79.
(2-Thiophen-2-ylethyl)thiophen-2-ylmethyleneamine (2d).
2-(2-Thienyl)ethylamine (2 g, 15.7 mmol), 2-thienyl-car-
boxaldehyde (1.76 g, 15.7 mmol), triethylamine (1 mL)
and ethanol (15 mL) were mixed, and the reaction mix-
ture was stirred at room temperature for 96 h. The
mixture was evaporated in vacuo, and the residue was
crystallised from hexane, to a€ord 1.26 g (36%) (2-
thiophen-2-ylethyl)thiophen-2-ylmethyleneamine (2d).
1H NMR (200 MHz, CDCl3) d 3.2 (t, 2H), 3.8 (t, 2H),

2286
P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
silica gel eluting with a mixture of dichloromethane:
methanol (9:1) to a€ord 0.33g (9%) 4-cyclohexyl-
4,5,6,7-tetrahydrothieno[3,2-c]pyridine (3f). 1H NMR
(200 MHz, CDCl3) d 1.0±1.9 (m, 10H), 2.6 (m, 1H), 2.75
(m, 1H), 2.9 (m, 1.5H), 3.25±3.6 (m, 1.5H), 3.9 (dd; 1H),
6.75 (d, 1H), 7.0 (d, 1H).
2.05 (dt, 2H), 2.25 (s, 3H), 1.6 (h, 1H), 2.7 (t, 2H), 2.95
(broad d, 2H), 3.75 (t, 2H), 7.05 (d, 1H), 7.15 (d, 1H);
13C NMR (CDCl3) d 22.6, 30.3, 42.8, 46.9, 48.1, 56.4,
122.2, 124.0, 128.5, 129.4, 131.2, 143.7, 166.4.
4-(1-Methylpiperidin-4-yl)-4,5,6,7-tetrahydrothieno[3,2-
c]pyridine (3g). 4-(1-Methylpiperidin-4-yl)-6,7-dihydro-
thieno[3,2-c]pyridine (5g) (2.95 g, 12.6 mmol) was dis-
solved in methanol (75 mL) and 0.71 g (0.0189 mol)
sodium borohydride was added in portions. Stirring was
continued for 2 h at room temperature. Evaporation in
vacuo gave an oil, which was dissolved in dichloro-
methane (100 mL) and washed with H2O (100 mL). The
aqueous phase was extracted with dichloromethane
(2Â100 mL). The combined organic phases were dried
with MgSO4, ®ltered and evaporated in vacuo to give
2.98g (100%) 4-(1-methylpiperidin-4-yl)-4,5,6,7-tetra-
hydrothieno[3,2-c]pyridine (3g). 1H NMR (200 MHz,
CDCl3) d 1.5 (dd; 2H), 1.6±2.0 (m, 6H), 2.2 (t, 3H), 2.75
(m, 2H), 2.8±3.0 (m, 3H), 3.35 (dt; 1H), 3.8 (broad s,
1H), 6.8 (d, 1H), 7.05 (d, 1H); 13C NMR (CDCl3) d
26.6, 27.1, 29.3, 30.0, 40.8, 42.7, 46.8, 56.6, 56.7, 59.1,
Cyclohexanecarboxylic acid (2-thiophen-2-ylethyl)amide
(4f). 2-(2-Thienyl)ethylamine (3 g, 23.6 mmol) was added
to a solution of cyclohexanecarboxylic acid (3.02 g, 23.6
mmol) and HOBt (3.61 g, 23.6 mmol) in DMF (150
mL), EDAC (6.78 g, 35.4 mmol) was added and the
resulting mixture was stirred for 15 h at room temperature.
The mixture was evaporated in vacuo. The remaining
oil was added to ethyl acetate (200 mL) and washed
with brine (100 mL), 0.5 N HCl (100 mL) and 1N
NaOH (100 mL). The organic phase was dried with
MgSO4, ®ltered and evaporated in vacuo giving 5.51g
(99%) cyclohexanecarboxylic acid (2-thiophen-2-yle-
thyl)amide (4f) as white crystals. Mp 86.7±87.2 ꢀC. H
1
NMR (200 MHz, CDCl3) d 1.1±1.5 (m, 5H), 1.6±1.9 (m,
5H), 2.05 (tt, 1H), 3.0 (t, 2H), 3.5 (dt, 2H), 5.6 (broad s,
1H), 6.8 (dd, 1H), 6.95 (dd, 1H), 7.1 (dd, 1H); 13C NMR
(CDCl3) d: 26.1, 30.1, 30.4, 41.0, 45.9, 45.95, 124.3,
125.7, 127.5, 141.9, 176,5.
.
121.8, 125.5, 135.4, 137.2. Calcd for C13H20N2S 1.75
H2O: C, 58.28; H, 8.84; N, 10.46. Found C, 58.52; H,
8.62; N, 10.62.
[1-Methylpiperidin-4-carboxylic acid]-(2-thiophen-2-yl)
ethyl amide (4g). 2-Thiophen-2-ylethylamine (3 g, 23.6
mmol) in DMF (150 mL) was added to a premixed (2
min.) solution containing HOBt (3.61 g (23.6 mmol),
diisopropylethylamine (8.23 mL, 47.2 mmol) and N-
methyl-piperidin-4-yl-carboxylic acid, hydrochloride in
DMF (50 mL), then 6.78g (0.0354 mol) EDAC was
added. The reaction mixture was stirred at room tem-
perature for 48 h. The solvent was evaporated in vacuo.
The remaining oil was added to H2O (100 mL), 1N NaOH
(50 mL) and dichloromethane (100 mL). The aqueous
phase was extracted with dichloromethane (3Â75 mL).
The combined organic phases were dried with MgSO4,
®ltered and evaporated in vacuo giving 7.2 g. Recrystal-
lisation from a mixture of ethyl acetate and pentane
a€orded 4.1g (69%) [1-methylpiperidin-4-carboxylic
acid]-(2-thiophen-2-yl)ethyl amide (4g) as white crystals.
Mp: 114.2±114.7 ꢀC. 1H NMR (200 MHz, CDCl3) d
1.6±2.1 (broad multiplet, 6H), 2.25 (s, 3H), 2.9 (d, 2,
5H), 3.1 (t, 2H), 3.15 (p, 0.5H), 3.5 (q, 2H), 5.95 (broad
s, 1H), 6.8 (d, 1H), 7.9 (dd, 1H), 7.13 (d, 1H); 13C NMR
(CDCl3) d 29.3, 30.2, 41.1, 43.1, 46.7, 55.6, 124.3, 125.7,
127.4, 141.7, 175.
4-Cyclohexyl-6,7-dihydrothieno[3,2-c]pyridine (5f). Com-
pound 4f (5.5g, 23.2 mmol) was added to a solution of
xylene (75 mL) P2O5 (9.9 g, 69.9 mmol) and POCl3 (6.6
mL, 3 equiv) at 85 ꢀC. The reaction mixture was heated
at 85 ꢀC for 2 h. After cooling, the mixture was added
1N NaOH (100 mL) and solid NaOH until pH 10. The
mixture was extracted with toluene (2Â50 mL). The
combined organic phases were dried with MgSO4, ®l-
tered and evaporated in vacuo to give 4.96g (98%) 4-
cyclohexyl-6,7-dihydro-thieno[3,2-c]pyridine (5f) as a
1
yellow oil. H NMR (200 MHz, CDCl3) d 1.15±1.5 (m,
5H), 1.65 (m, 5H), 2.55±2.7 (m, 1H), 2.75 (dt, 2H), 3.75
(dt, 2H), 7.0 (d, 1H), 7.05 (d, 1H); 13C NMR (CDCl3) d
22.7, 26.6, 26.9, 31.2, 44.8, 48.1, 122.1, 124.2, 131.7,
143.6, 168.0. Calcd for C13H17NS: C, 71.18; H, 7.81; N,
6.39. Found C, 69.60; H, 7.64; N, 6.54.
4-Cyclohexyl-4,5,6,7-tetrahydrothieno[3,2-c]pyridine (3f,
amide route). Compound 5f (4.25g, 19.4 mmol) was dis-
solved in methanol (75 mL) and NaBH4 (1.1 g, 29
mmol) was added in portions. The mixture was stirred
for 1 h at room temperature. The mixture was then
evaporated in vacuo and the remaining oil was dissolved
in H2O (50 mL) and extracted with dichloromethane
(3Â75 mL). The organic phase was dried with MgSO4,
®ltered and evaporated in vacuo to a€ord 4.25g (98%) 4
-cyclohexyl-4,5,6,7-tetrahydrothieno[3,2-c]pyridine (3f)
4-(1-Methylpiperidin-4-yl)-6,7-dihydrothieno[3,2-c]pyri-
dine (5g). [1-Methylpiperidin-4-carboxylic acid]-(2-thio-
phen-2-yl)ethyl amide (1 g, 40 mmol) was dissolved in
toluene (25 mL) and added to POCl3 (1.1 mL, 12
mmol). The resulting mixture was heated at 80 ꢀC for 19
h. Toluene was decanted from a yellow precipitate. The
yellow crystals were dissolved in dichloromethane (150
mL), ice (50 mL) and 2N NaOH (50 mL). The aqueous
phase was extracted with dichloromethane (2Â75 mL)
the combined organic phase and was dried with MgSO4,
®ltered, and evaporated in vacuo to give 0.77 g (82%) of
4-(1-methylpiperidin-4-yl)-6,7-dihydro-thieno[3,2-c]pyr-
1
as a yellow oil. H NMR (200 MHz, CDCl3) d 1.0±1.9
(m, 11H), 2.7 (m, 2H), 2.9 (m, 1H), 3.25 (m, 1H), 3.75
(dd, 1H), 6.35 (d, 1H), 7.0 (d, 1H); 13C NMR (CDCl3) d
26.7, 27.0, 27.1, 27.2, 27.4, 31.0, 43.2, 43.5, 60.3, 121.8,
125.4, 135.2, 137.7. Calcd for C13H19NS: C, 70.45; H,
8.65; N, 6.39. Found C, 70.35; H, 8.75; N, 6.66.
4-Phenyl-4,5,6,7-tetrahydrothieno[3,2-c]pyridine (3h). 2-
(2-Thienyl)ethylamine (2 g, 15.7 mmol) and benzalde-
hyde (1.67 g, 15.7 mmol) were dissolved in toluene (50
1
idine (5g) as an oil. H NMR (CDCl3) d 1.8 (m, 4H),

P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
2287
mL) and the reaction mixture was heated at re¯ux until
20 mL of toluene and water was distilled of in a Dean±
Stark trap. The remaining mixture was evaporated in
vacuo to give the crude imine (3.44 g). The crude imine
was added to tri¯uoroacetic acid (50 mL) at once (exo-
thermic reaction). The reaction mixture was stirred at
room temperature for 72 h, and then evaporated in
vacuo. The residue was dissolved in dichloromethane
(50 mL) and washed with 2N sodium hydroxide (30
mL). The aqueous phase was extracted with dichloro-
methane (3Â30 mL). The combined organic phases were
dried with MgSO4, ®ltered and evaporated in vacuo.
The residue was puri®ed by column chromatography on
silica gel eluting with a mixture of dichloromethane and
methanol (19:1). This a€orded 0.823 g (24%) 4-phenyl-
4,5,6,7-tetrahydrothieno[3,2-c]pyridine (3h) as white
crystals. Mp: 79.8±80.7 ꢀC 1H NMR (200 MHz, CDCl3)
d 1.95 (broad s, 1H), 2.7±3.3 (m, 4H), 5.0 (s, 1H), 6.45
(d, 1H), 6.95 (d, 1H), 7.25 (m, 5H); 13C NMR (CDCl3)
d: 25.5, 43.0, 60.6, 122.2, 126.8, 128.0, 128.9, 135.4,
137.3, 144.2. Calcd for C13H13NS: C, 72.52; H, 6.09; N,
6.51. Found C, 72.18; H, 6.10; N, 6.33.
was separated, dried with MgSO4 and evaporated
resulting in an oil which was puri®ed on silica-gel using
dichloromethane:methanol (9:1) as eluent. Yield of (3o)
50%, mp 93.6±93.8 C. 1H NMR (CDCl3) d 7.33±7.1 (m,
4H), 7.13 (d, 1H), 6.82 (d, 1H), 5.12 (s, H), 3.32±2.95
(m, 2H), 2.9±2.6 (m, 2H), 1.95 (broad, 1H NH); 13C
NMR (CDCl3): d 142.7, 137.7, 135.7, 134.0, 129.9,
129.1, 127.7, 123.9, 59.4, 42.7, 27.0. Calcd for C13H12
ClNS: C, 62.52%; H, 4.87%; N, 5.61%; Found C,
62.72%; H, 4.87%; N, 5.59.
7-(4-Methoxyphenyl)-4,5,6,7-tetrahydrothieno[2,3-c]pyri-
dine (3p). This compound was prepared from 2-(3-thie-
nyl)ethylamine (1.21 g, 9.5 mmol) and 4-methoxybenz-
aldehyde (1.30 g, 9.5 mmol) in dry ethanol (8 mL), giv-
ing 100% yield of the crude imine which was reacted
further with tri¯uoroacetic acid without further pur-
i®cation as described for (3o) resulting in a yield of 42%
of (3p) isolated as an oil. 1H NMR (CDCl3) d 7.24±7.21
(m, 2H), 7.1 (d, 1H), 5.09 (s broad, 1H), 3.78 (s, 3H),
3.37±3.21 (m, 1H), 3.12±2.97 (m, 1H), 2.90±2.58 (m,
2H), 1.86 (broad, 1H NH); 13C NMR (CDCl3): d 159.6,
138.9, 136.6, 135.4, 129.6, 127.6, 123.6, 114.2, 59.6, 55.7,
43.0, 27.1. Calcd for C14H15NS: C, 68.54%; H, 6.16%;
N, 5.71%; Found C, 68.40%; H, 6.19%; N, 5.72.
4- (4-Tri¯uoromethoxyphenyl)-4,5,6,7-tetrahydrothieno
[2,3-c]pyridine (3n). A solution of 2-(3-thienyl)ethyla-
mine23 (14.7 g, 0.115 mol) and 4-tri¯uormethoxybenz-
aldehyde (15.0 g, 0.08 mol) in benzene (200 mL) was
re¯uxed (Dean±Stark trap, H2O removed) for 4 h. Tri-
¯uoroacetic acid (2 mL) was added and the mixture was
re¯uxed for 8 h. After cooling it was made alkaline with
NH4OH and washed with water. The organic phase was
dried (K2CO3) and evaporated in vacuo to give a residue,
which was puri®ed by chromatography on silica-gel (200
g). A by-product was removed by elution with benzene,
Rf 0.68 (SiO2; CHCl3:EtOH:NH4OH, 200:10:1), prob-
ably Schi€ base (it was decomposed by an attempt to
prepare hydrogen oxalate). Crude 3n (3.9 g) was
obtained by elution with chloroform. Rf 0.47 (SiO2;
CHCl3:EtOH:NH4OH, 200:10:1). The hydrogen oxalate
was prepared by neutralisation of the solution of the
above base in diethyl ether with a solution of oxalic acid
dihydrated in acetone. Hydrogen oxalate was con-
taminated with hydrogen oxalate of 2-(3-thienyl)ethyla-
mine. A suspension of the mixture was repeatedly boiled
with water and ®ltrated. This a€orded, after drying,
pure (3n), hydrogen oxalate 2.5 g (8%), mp 190±195 ꢀC.
1H NMR (250 MHz, DMSO-d6) d 9.22 (s, 3H), 7.59 (d,
J=8.6 Hz, 2H), 7.47 (d, J=4.9 Hz, 1H), 7.42 (d, J=8.6
Hz, 2H), 6.97 (d, J=4.9 Hz,1H), 5.73 (s, 1H), 3.35 (bm,
2H), 2.95 (bm, 2H). Calcd for C14H12F3NOS, C2H2O4,
1/4 H2O: C, 48.79; H, 3.71; N, 3.56; F, 14.47; S, 8.14;
Found C, 48.69; H, 3.60; N, 3.42; F, 14.83; S, 8.38.
7-Substituted 4,5,6,7-tetrahydrothieno[2,3-c]pyridines (3q±
3y). The 7-substituted 4,5,6,7-tetrahydrothieno[2,3-c]-
pyridines were prepared from equimolar amounts of
appropriate substituted benzaldehyde (0.0078 mol) and
2-(3-thienyl)ethylamine (0.0078 mol) in dry ethanol (8
mL) by shaking for 3 days at room temperature. The
mixture was subsequently evaporated to dryness and the
resulting oil treated with tri¯uoroacetic acid (20 mL) by
stirring for 24 h followed by addition of NaOH (2M, 10
mL). Extraction with dichloromethane (10 mL) fol-
lowed by evaporation a€orded the desired 7-substituted
4,5,6,7-tetrahydrothieno[2,3-c]pyridines as oils. Identity
and yield were estimated from the HPLC/MS spectra,
data are presented in Table 1.
Preparation of amides: general procedure (Table 7)
(4-Methoxyphenyl)-[4-(4-tri¯uoromethoxyphenyl)-4,5,6,
7-tetrahydrothieno[3,2-c]pyridin-5-yl]methanone (8). 4-
Methoxybenzoic acid (0.64 g, 4.2 mmol) was dissolved
in DMF (25 mL) and HOBt (0.71 g, 5.0 mmol) was
added followed by EDAC (0.96 g, 5.0 mmol). The
resulting mixture was stirred at room temperature for 30
min. 4-(4-Tri¯uoromethoxyphenyl)-4,5,6,7-tetrahydro-
thieno[3,2-c]pyridine (3l, 1.5 g, 5.0 mmol) and DIPEA
(1.4 mL, 8.4 mmol) were added and the mixture was
stirred at room temperature for 16 h. Water (100 mL)
was added and the mixture was extracted with Et2O
(3Â20 mL). the combined organic phases were washed
with satd NH4Cl (20 mL), dried (MgSO4) and con-
centrated in vacuo. The residue was puri®ed by column
chromatography on silica gel eluting with a mixture of
EtOAc:heptane (1:2). This a€orded, after crystallisation
from a mixture of heptane and methyl tert-butyl ether,
7-(4-Chlorophenyl)-4,5,6,7-tetrahydrothieno[2,3-c]pyridine
(3o). 2-(3-Thienyl)ethylamine (3.15 g, 24.8 mmol) and
4-chlorobenzaldehyde (3.5 g, 2.48 mmol) were mixed
without solvent resulting in dissolution of the crystals of
the amine followed by precipitation of slightly yellow
crystals. The mixture was left at room temperature for 4 h.
Tri¯uoroacetic acid (20 mL) was added and the mixture
stirred overnight at room temperature. The solvent was
evaporated and the resulting mixture extracted between
NaOH (4 M) and dichloromethane, the organic phase
2.13
methoxyphenyl)-4,5,6,7-tetrahydro-thieno[3,2-c] pyridin-
g (98%) of (4-methoxyphenyl)-[4-(4-tri¯uoro
5-yl]methanone (8), mp 68-70 ꢀC. H NMR (300 MHz,
1

2288
P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
CDCl3) d 2.84 (1H, m), 3.05 (1H, m), 3.28 (1H, m), 3.84
(3H, s), 3.90 (1H, m), 6.75 (1H, d), 6.92 (3H, m), 7.17
(2H, d), 7.21 (1H, d), 7.37 (4H, m). Anal. (C22H18
F3NO3S.1/4H2O): CHN.
Reagent Concentrate (cat no 360-3C) was obtained
from Sigma Chemical Company, St. Louis, Mo (USA).
Methods
Glucose-6-phosphatase activity. Pig liver microsomes
were prepared essentially as described previously.24
Microsomes were kept at 80 ꢀC until use. Prior to
measurement of G-6-Pase activity, microsomes were
treated with 0.5% v/v Triton X-10025 (the ®nal con-
centration of Triton X-100 in the assay was 0.04%), and
these are referred to as disrupted microsomes. Glucose-
6-phosphatase activity in disrupted microsomes (0.05
mg protein) was measured by incubation for 6 min at
30 ꢀC with 0.5 mM glucose-6-phosphate, 30 mM MES
(2-(N-morpholino)ethanesulfonic acid) pH 6.5, and
various compounds dissolved in 100% DMSO in a total
volume of 325 mL. The ®nal concentration of DMSO in
the assay did not exceed 5%. The reaction was terminated
by addition of 100 mL Sigma phosphorus reagent. This
mixture was allowed to stand for 2 min and, subsequently,
the absorbance was measured at 340 nm. All values were
corrected for blank. In some cases, microsomal protein
contents were measured using the Bio-Rad DC Protein
Assay and human serum albumin as the standard. The
IC50 values were calculated using graphpad prism
(GraphPad software Inc. San Diego, USA).
Preparation of amides 7±45 and 47±57: general library
procedure
The carboxylic acid (0.15 mmol) was dissolved in DMF
(0.25 mL) and a solution of 1-hydroxybenzotriazole
(HOBt, 0.15 mmol) in DMF (0.25 mL) was added. A
suspension of N-(3-dimethylaminopropyl)-N0-ethylcar-
bodiimide hydrochloride (EDAC) (0.225 mmol) in
DMF (0.25 mL) was added followed by addition of the
substituted 4,5,6,7-tetrahydrothienopyridine derivative
(0.15 mmol) in DMF (0.25 mL). The reaction mixture
was shaken vigorously at room temperature for 24 h.
The reaction mixture was added a saturated solution of
sodium chloride (2 mL) and ethyl acetate (1 mL) and
shaken vigorously for 1 h. The organic phase was
removed by pipetting and evaporated in a Savant
vacuum centrifuge to a€ord the product. The resulting
oil or crystals were analysed by HPLC-MS.
Optical resolution of amides: general procedure
(+)- and ( )-[4-(4-Chlorophenyl)-4,5,6,7-tetrahydro-
thieno[3,2-c]pyridin-5-yl]-(4-methoxyphenyl)methanone,
(7A and 7B, respectively). Compound 7 (30 mg) was
dissolved in a 1:1 mixture of n-heptane and 2-propanol
(5 mL) and fractionated by HPLC (2 runs) using a
21.1Â250 mm (R,R)-Whelk-O column (Regis). The col-
umn was eluted isocratically with a 1:1 mixture of n-
heptane:2-propanol at a ¯ow rate of 12 mL/min and
fractions collected corresponding to 0.8 min/fraction.
The eluting enantiomers were detected spectroscopically
by measuring absorbance at a wavelength of 225 nm.
Two eluting peaks were detected, one corresponding to
TR 44±50 min and one corresponding to TR 62±72 min.
Fractions from the two runs corresponding to TR 44±50
min were separately pooled and evaporated to yield 12.1
mg of the corresponding (+)-isomer.
Acknowledgements
The technical assistance of Sanne Kold, Claus Freder-
iksen, Dorthe Egholm Jensen, Bo R. Pedersen, Mads
Ole Nymand Jensen, and Liselotte Wul€-Hùyer are
gratefully acknowledged. Also, we thank Dr Karla
Frydenvang, Royal Danish School of Pharmacy and
Dr. Inger Sùtofte, Danish Technical University for the
X-ray absolute con®guration determinations.
References and Notes
1. Nordlie, R. C.; Foster, J. D.; Lange, A. J. Annu. Rev. Nutr.
1999, 19, 379.
2. Herling, A. W.; Burger, H.-J.; Schwab, D.; Hemmerle, H.;
Below, P.; Schubert, G. Am. J. Physiol., Gastrointest. Liver
Physiol. 1998, 37, G1087.
3. Parker, J. C.; Vanvolkenburg, M. A.; Levy, C. B.; Martin,
W. H.; Burk, S. H.; Kwon, Y.; Giragossian, C.; Grant, T. G.;
Carpino, P. A.; Mcpherson, R. K.; Vestergaard, P.; Treadway,
J. L. Diabetes 1998, 47, 1630.
4. Madsen, P.; Brand, C. L.; Holst, J. J.; Knudsen, L. B. Curr.
Pharm. Design 1999, 5, 683.
100% ee (Determined by HPLC using a 4.6Â250 mm
(R,R)-Whelk-O column eluted with a 1:1 mixture of n-
heptane and 2-propanol, the ¯ow rate was 1 mL/min,
eluting sample was monitored spectroscopically at 225
and 245 nm, TR 15.5 min).
Fractions from the two runs above corresponding to TR
62±72 min were separately pooled and evaporated to
yield 12.5 mg of the corresponding ( )-isomer.
5. Turner, R.; Cull, C. Ann. Intern. Med. 1996, 124, 136.
6. Nordlie, R. C.; Bode, A. M.; Foster, J. D. Proc. Soc. Exp.
Biol. Med. 1993, 203, 274.
7. Sukalski, K.A.; Nordlie, R.C. In Advances in Enzymology
and Related Areas of Molecular Biology; Meister, 1989, Ed.;
John Wiley and Sons: New York pp 93±117.
8. Lyall, H.; Grant, A.; Scott, H. M.; Burchell, A. Biochem.
Soc. Trans. 1992, 20, 271S.
98% ee (Conditions as described above, TR 20.8 min).
[a]2D0 170.0ꢀ (c 0.25, ethyl acetate).
Enzyme assays
Materials
9. DeFronzo, R. A.; Bonadonna, R. C.; Ferrannini, E. Dia-
betes Care 1992, 15, 318.
Liver from pigs (24 h fasted) were obtained from the
animal quarters at Novo Nordisk A/S. Phosphorus
10. Argaud, D.; Zhang, Q.; Pan, W.; Maitra, S.; Pilkis, S. J.;
Lange, A. Diabetes 1996, 45, 1563.

P. Madsen et al. / Bioorg. Med. Chem. 8 (2000) 2277±2289
2289
11. Burchell, A.; Cain, D. I. Diabetologia 1985, 28, 856.
12. Westergaard, N.; Brand, C. L.; Lewinsky, R. H.; Ander-
sen, H. S.; Carr, R. D.; Burchell, A.; Lundgren, K. Arch. Bio-
chem. Biophys. 1999, 366, 55.
17. Bremner, J. B.; Browne, E. J.; Chohan, V.; Yates, B. F.
Aust. J. Chem. 1984, 37, 1043.
18. Ohkubo, M.; Kuno, A.; Katsua, K.; Ueda, Y.; Shirakawa,
K.; Nakanishi, H.; Kinoshita, T.; Takasugi, H. Chem. Pharm.
Bull. 1996, 44, 778.
13. (a) Hemmerle. H.; Burger, H.-J.; Below, P.; Schubert, G.;
Rippel, R.; Schindler, P. W.; Paulus, E.; Herling, A. W. J. Med.
Chem. 1997, 40, 137. (b) Herling, A. W., Burger, H-J., Schwab,
D.; Hemmerle, H.; Below, P.; Schubert, G. Am. J. Physiol. 1998,
274, 1087. (c) Arion, W. J.; Can®eld, W. K.; Ramos, F. C.; Su,
M. L.; Burger, H.-J.; Hemmerle, H.; Schubert, G.; Below, P.;
Herling, A. W. Arch. Biochem. Biophys. 1998, 351, 279.
14. Arion, W. J.; Can®eld, W. K.; Ramos, F. C.; Burger, H.-
J.; Hemmerle, H.; Schubert, G.; Below, P.; Herling, A. W.
Arch. Biochem. Biophys. 1997, 339, 315.
19. Bremner, J. B.; Browne, E. J.; Davies, P. E. Aust. J. Chem.
1980, 33, 1335.
20. Russell, R. K. US patent 5,294,621 1994.
21. Vecchietti, V.; Clarke, G. D.; Colle, R.; Giardina, G.;
Petrone, G.; Sbacchi, M. J. Med. Chem. 1991, 34, 2624.
22. Gray, N. M.; Cheng, B. K.; Mick, S. J.; Lair, C. M.;
Contreras, P. M. J. Med. Chem. 1989, 32, 1242.
23. Cardelli, M.; Claudi, F.; Di Stefano, A.; Giorgioni, G.;
Cantalamessa, F.; Cagnotto, A.; Skorupska, M. Eur. J. Med.
Chem. 1994, 29, 423.
24. Arion, J. M.; Lange, A. J.; Walls, H. E. J. Biol. Chem.
1980, 255, 10387.
25. Arion, W. J.; Carlson, P. W.; Wallin, B. K.; Lange, A. J.
J. Biol. Chem. 1972, 247, 2551.
15. Bleriot, Y.; Smelt, K. H.; Cadefau, J.; Bollen, M.; Stal-
mans, W.; Biggadike, K.; Johnson, L. N.; Oikonomakos, N.
G.; Lane, A. L.; Crook, S.; Watkin, D. J.; Fleet, G. W. J.
Tetrahedron Lett. 1996, 37, 7155.
16. Descamps, M.; Binon, F. Bull. Soc. Chim. Belg. 1962, 71, 579.