Imidazolines as efficacious glucose-dependent stimulators of insulin secretion

Article abstract of DOI:10.1016/S0223-5234(03)00041-2

Synthesis of a series of imidazolines with glucose dependent effects on insulin exocytosis from pancreatic β-cells is reported. Regioisomers and enantiomers were found to exhibit marked differences in exocytotic effects as well as different activities on the K_ATP-channel; the (R (+)) isomer of 2-[2-(4,5-dihydro-1H-imidazol-2-yl)-1-thiophene-2-ylethyl]pyridin e (4a) and the (+) isomer of 2-[2-(4,5-dihydro-1H-imidazol-2-yl)-1-thiophene-3-ylethyl]pyridin e (4d) were found to give a significant increase in insulin release - in contrast to findings for their enantiomers - without influence on the K_ATP-channel. The (+) isomer (4a) showed glucose dependent insulin release from β-cells at concentrations above 2.5 mM and a marked glucose lowering effect in ob/ob mice as well as in fed but not in fasted rats.

Full text of DOI:10.1016/S0223-5234(03)00041-2

European Journal of Medicinal Chemistry 38 (2003) 357ꢁ362
/
www.elsevier.com/locate/ejmech
Imidazolines as efficacious glucose-dependent stimulators of insulin
secretion
Palle Jakobsen *, Peter Madsen, HenrikSune Andersen
Medicinal Chemistry Research, Novo Nordisk A/S, DK-2760 M a˚ løv, Denmark
Received 27 September 2002; received in revised form 19 December 2002; accepted 26 January 2003
Abstract
Synthesis of a series of imidazolines with glucose dependent effects on insulin exocytosis from pancreatic b-cells is reported.
Regioisomers and enantiomers were found to exhibit marked differences in exocytotic effects as well as different activities on the
KATP-channel; the (R (ꢀ
-[2-(4,5-dihydro-1H-imidazol-2-yl)-1-thiophene-3-ylethyl]pyridine (4d) were found to give a significant increase in insulin release*
in contrast to findings for their enantiomers*without influence on the KATP-channel. The (ꢀ) isomer (4a) showed glucose
/
)) isomer of 2-[2-(4,5-dihydro-1H-imidazol-2-yl)-1-thiophene-2-ylethyl]pyridine (4a) and the (ꢀ) isomer of
/
2
/
/
/
dependent insulin release from b-cells at concentrations above 2.5 mM and a marked glucose lowering effect in ob/ob mice as well as
in fed but not in fasted rats.
´
#
2003 Editions scientifiques et m e´ dicales Elsevier SAS. All rights reserved.
Keywords: Glucose dependent insulin release; Imidazolines; KATP current
1
. Introduction
-Substituted imidazolines have for years been targets
discover imidazoline containing compounds that could
increase insulin release from the b-cell at elevated
glucose concentrations only, without or with very little
effect on the pancreatic KATP-channel as well as on
peripheral or central imidazoline- and a-receptors.
A Japanese patent [13] claimed racemic 4a to be able
to increase insulin release, but no details regarding the
mechanism of action were given. We have found that
2
for the search for compounds that can modulate e.g.
blood pressure, heart rate, and CNS diseases, and drugs
like clonidine and phentolamine have been marketed for
the treatment of hypertension [1ꢁ3]. Several imidazo-
/
lines have been reported to modulate insulin release
from pancreatic b-cells through activity on the adrener-
gic a - a -, imidazoline I -, or I -receptors and/or the
this compound as the racemic mixture*/in addition to
its insulin releasing property*also exhibited a pro-
/
1
,
2
1
2
nounced inhibitory activity at the KATP-channel, while
one of the enantiomers was without this effect. This
prompted us to investigate a series of close analogues in
more detail.
KATP channel [4,5]. This is, however, not an ideal
mechanism of action as severe cardiovascular side
effects as well as hypoglycaemic events can arise [6].
The actual target for the imidazolines in the b-cell is far
from fully explored but it seems to be accepted that
besides activity on the above mentioned receptors and
on the KATP-channel, there might be a third imidazoline
receptor, or other mechanisms involved in the increase
of insulin release for compounds which do not exhibit
their activity through a blocking of the KATP-channel
2. Chemistry
Several methods for the preparation of 2-substituted
imidazolines are described in the literature, mainly using
nitriles or esters as starting material via condensation
[
7ꢁ
/
12]. The aim of the present investigation was to
with ethylenediamine or salts thereof [14ꢁ16]. Also
/
methods utilising ketones and imidazoline derivatives
are described [17]. In order to prepare different 2-
* Corresponding author.
E-mail address: paj@novonordisk.com (P. Jakobsen).
´
0
223-5234/03/$ - see front matter # 2003 Editions scientifiques et m e´ dicales Elsevier SAS. All rights reserved.
doi:10.1016/S0223-5234(03)00041-2

358
P. Jakobsen et al. / European Journal of Medicinal Chemistry 38 (2003) 357ꢁ362
/
Fig. 1. Synthetic route for the preparation of compounds 4a
/
e.
ꢁ
(
pyridyl-thienylethyl)imidazolines we have used three
approaches as depicted in Figs. 1 and 2.
2- or 3-thienyl lithium in Et O or THF at ꢂ
/
78 8C. Acid
2
hydrolysis during work-up afforded ketones (1a f) in
fair to good yields. Subsequent reaction with triethyl
phosphonoacetate or diethyl cyanomethylphosphonate
/
ꢁ
Pyridyl-thienyl ketone starting materials (1a-f) were
conveniently prepared from pyridine carbonitriles and
Fig. 2. Synthetic route for the preparation of compound 4f.

P. Jakobsen et al. / European Journal of Medicinal Chemistry 38 (2003) 357ꢁ
/362
359
gave the unsaturated esters or nitrile 2a
/
e and 5a,
3. Results and discussion
ꢁ
respectively, in good yields. These unsaturated inter-
mediates were hydrogenated at atmospheric pressure or
up to 5.5 bar using Pd/C as catalyst. This reduction gave
low to good yields and often required long reaction
times and renewal of the catalyst. The saturated esters
Compounds 4a
/
e were tested in a patch clamp
ꢁ
exocytosis assay (mouse pancreatic b-cells) measuring
their ability to release insulin. Imidazolines that in-
creased the exocytotic rate were further tested for their
effect on the KATP-channel as well as in assays for
insulin release, activity on a-, and imidazoline-receptors.
Table 1 shows the in vitro data for selected compounds.
The in vitro results clearly indicates that there is a
difference in the degree to which the compounds could
increase the rate of the insulin exocytosis, both for
stereoisomers like in 4a, 4d and 4e and between
regioisomers, whereas compound 4c gave no increase
in the exocytotic rate.
3
a
/
e were conveniently reacted with ethylenediamine in
ꢁ
monotosylate affording imidazoline 4a in good yield
Fig. 1.
the presence of AlMe [5] to give 4a
3
/e, and the cyano
ꢁ
compound 6a was melted with dry ethylenediamineꢁ
/
The 3-thienyl-3-pyridyl compound 4f was not pre-
pared by these routes since the reduction of the cyano or
the ester-intermediates 2f or 5f gave very low yields.
Compound 4f was therefore prepared using the syn-
thetic pathway described in Fig. 2. Ketone 1f was
reacted with the Li-salt of Boc-protected 2-diethoxypho-
sphorylmethylimidazoline 8. Work up with HCl gave in
good yield the unsaturated compound 7f as a Z,E-
mixture. This was reduced to 4f without further
purification as attempts to isolate the pure diastereo-
mers resulted in decomposition. The yield was good
Also activity on the KATP-channel differed consider-
ably between the stereoisomers as well as between the
regioisomers. (R)-(ꢀ) 4a was further investigated for
/
insulin release from isolated mouse b-cells as well as in
in vivo experiments in rats and ob/ob mice. The results
from these investigations showed that (R)-(ꢀ) 4a
/
increased insulin release dose dependently only at blood
glucose concentrations above 2.5 mM (Fig. 3). In rats,
the plasma insulin level was increased significantly after
2 min in fed but not in fasted animals (20 mg/kg i.v.),
and the blood glucose levels were significantly decreased
in the fed compared to the fasted animals after 30 min.
(
65%) in contrast to previous findings [17].
Imidazolines that showed promising effects on insulin
exocytosis as racemic mixtures were separated into pure
enantiomers by HPLC using chiral OD columns.
Resolution by fractional crystallisation in the presence
of chiral acids was unsuccessful. The exact configuration
of the enantiomers of compound 4a was determined by
X-ray crystallography on the corresponding hydrochlo-
In ob/ob mice (R)-(ꢀ
blood glucose level significantly, (data not shown).
Although compound (R)-(ꢀ) 4a showed significant in
/) 4a (100 mg/kg p.o.) decreased the
/
ride salts, showing the configuration of the (ꢀ
/
) isomer
vivo effect on blood glucose levels in different animal
models of type 2 diabetes it also exhibited strong
to be R and the (ꢂ) isomer to be S.
/
Table 1
In vitro data for selected imidazolines
a
b
Insulin exocytosis (fF/s)
Compound
K
ATP channel (ratio)
IC50 (mM)
c
d
e
a
2a
a
2b
a
2ns
I
2c
I
2p
4
4
4
4
4
4
4
4
4
4
4
a ꢀ
a ꢀ
a ꢀ
b ꢀ,ꢂ
c ꢀ,ꢂ
d ꢀ,ꢂ
d ꢀ
d ꢂ
e ꢀ,ꢂ
e ꢀ
e ꢂ
/
,ꢂ
/
0.20
nd
12.3
4.0
1.98
6.94
0.76
1.94
7.9
0.66
3.06
0.6
0.43
3.15
0.5
3.51
10
0.9
/
ꢀ
/
/
1.03
0.67
nd
11.5
9.1
1.42
/
/*
/
/
4.8
/
/
nd
9.2
/
0.93
0.56
0.59
nd
9.3
5.8
8.03
5.02
7.34
3.6
2.4
4.8
3.97
0.92
1.89
3.59
0.89
0.72
/
2.56
3.14
/
/
10.9
5.9
5.85
/
/
0.23
8.9
The data from the a- and I-receptors were obtained from a standard Panlab-screen. Test concentration for the patch clamp measurements was 100
mMa .
The ratio of whole-cell KATP current with or without test compound is given.
Change in cell capacitance by application of test compound is given as a measure of exocytosis.
b
c
2
Non-selective a -receptor.
d
e
I
2
-central.
I
2
-peripheral.
*
The (ꢀ
/
) and the (Ã
/
) -isomers showed B30% inhibition of the KATP-channel in INS-1 cells.
/

360
P. Jakobsen et al. / European Journal of Medicinal Chemistry 38 (2003) 357ꢁ362
/
added to pH 1, followed by stirring for 30 min. KOH
was then added to pH 10. Ethyl acetate (50 mL) and
water (10 mL) were added and the organic phase was
separated, dried with MgSO and evaporated to dryness
4
resulting in yellow oil.
The crude product was treated with HCl (6 N, 50 mL)
for 1 h and the pH was then adjusted to 10 by addition
of solid KOH. Extraction, drying, and evaporation as
described above resulted in yellow crystals 2.4 g (85%) of
pyridine-3-yl-thiophene-2-yl-methanone 1c, Mp: 90-
ꢀ
1
9
9
7
1.5 8C. EI/MS: m/z 189 (M) . H-NMR (CDCl ): d
3
.10 (d, 1H); 8.80 (dd, 1H); 8.18 (dd, 1H); 7.80 (d, 1H);
1
.65 (dd, 1H); 7.48 (dd, 1H); 7.20 (t, 1H). C NMR
3
Fig. 3. Insulin release from mouse islets (fmol insulin/min/islet) as a
function of glucose concentration (mM) for compound (R)-(ꢀ) 4a at
00 mM.
(
1
CDCl ): d 186.6; 153.2; 150.3; 143.4; 136.9; 135.6;
3
/
35.5; 134.1; 128.7; 123.9. Anal. C H NOS (C,H,N).
10 7
1
The following intermediates were prepared from the
appropriate starting materials using the same procedure
as described for the preparation of 1c.
affinities for both the a - and the I -receptors thus not
2
2
being ideal for further exploration. Compound (ꢀ)-4d,
/
however, showed a more promising profile. In conclu-
sion, we have shown that it was possible to find
imidazolines that could increase insulin release without
having activity at the KATP-channel and with low
affinities for the a- and imidazoline-receptors. The
exocytotic effect was found to be strongly dependent
on the structure with great variance between the
individual enantiomers.
4
.3.2. Pyridine-2-ylthiophene-2-ylmethanone (1a)
1
Oil, H-NMR (CDCl ): d 8.75 (d, 1H); 8.37 (d, 1H);
3
8
7
.12 (d, 1H); 7.85 (t, 1H); 7.75 (d, 1H); 7.45 (dd, 1H);
.14 (dd, 1H).
4
4
4
4
.3.3. Pyridine-4-ylthiophene-2-ylmethanone (1b)
ꢀ
M.p.: 101 8C, EI/MS: m/z 189 (M) .
.3.4. Pyridine-2-ylthiophene-3-ylmethanone (1d)
ꢀ
Oil. LC/MS: m/z 190 (Mꢀ1) .
/
4
. Experimental
.3.5. Pyridine-4-ylthiophene-3-ylmethanone (1e)
ꢀ
4
.1. General procedures
M.p.: 105ꢁ
/
106 8C, EI/MS: m/z 189 (M) .
The general synthetic methods described previously
.3.6. Pyridine-3-ylthiophene-3-ylmethanone (1f)
ꢀ
[
of the theoretical value.
18] were used. Combustion analyses were within ꢀ
/
0.4%
M.p.: 71.5ꢁ
/
72 8C, EI/MS: m/z 189 (M) .
4
(
.4. General procedure for the preparation of compounds
4a
4
.2. Pharmacological methods
/
e)
ꢁ
The assays used have been described previously
4
2
.4.1. [2-(4,5-Dihydro-1H-imidazol-2-yl)-1-thiophene-
-ylethyl]pyridine (4c)
Sodium (0.12 g, 5 mmol) was dissolved in dry ethanol
[
19,20].
4
(
.3. General procedure for the preparation of compounds
(
10 mL), the solution was cooled to ꢂ5 8C and triethyl
/
1aꢁf)
/
phosphonoacetate (1.13 g, 5 mmol) in ethanol (1 mL)
was added and the mixture stirred for 30 min. Pyridine-
3-ylthiophene-2-ylmethanone 1c (0.76 g, 4 mmol) dis-
solved in ethanol (1 mL) was added and the mixture was
heated at 55 8C for 3 days. The reaction mixture was
4
.3.1. Pyridine-3-yl-thiophene-2-yl-methanone (1c)
n-Butyl lithium (10.0 mL, 1.6 M in hexane, 16 mmol)
was added to dry diethyl ether (30 mL) under a nitrogen
atmosphere below ꢂ78 8C. 2-Bromothiophene (2.44 g,
5 mmol) in dry diethyl ether (25 mL) was added
dropwise at a temperature below ꢂ72 8C. The mixture
was subsequently stirred at ꢂ30 8C for 20 min. Further
cooling to ꢂ78 8C followed by addition of 3-cyanopyr-
idine (1.56 g, 15 mmol) in dry diethyl ether (30 mL) over
0 min. The mixture was kept at ꢂ78 8C for 20 min
followed by heating to 0 8C for 10 min. HCl (6 N) was
/
then evaporated to dryness to give crude Z ꢁE mixture
/
1
of 3-pyridine-3-yl-3-thiophene-2-ylacrylic acid ethyl es-
ters 2c as an oil (1.2 g). The crude product was dissolved
in dry ethanol (40 mL) and hydrogenated over Pd/C
(10%, 0.25 g) at 20 psi overnight. The reaction mixture
was filtered through celite and evaporated to dryness.
The crude product was purified on silica gel using
dichloromethane/MeOH (19:1) as eluent, which af-
/
/
/
1
/

P. Jakobsen et al. / European Journal of Medicinal Chemistry 38 (2003) 357ꢁ
/362
361
forded 265 mg of 3-pyridin-3-yl-3-thiophen-2-yl-propio-
4.4.2.2. Resolution of (4a). Racemic 4a was separated
into pure enantiomers using a Chiralcel OD column
ꢀ
nic acid ethyl ester 3c. LC/MS: m/z 262 (Mꢀ
To trimethyl aluminium (2 M in toluene, 0.75 mL, 1.5
mmol) cooled to ꢂ10 8C was slowly added ethylenedia-
mine (0.1 mL, 1.5 mmol) and the mixture was stirred for
0 min. 3c (260 mg, 1 mmol) in dry toluene (10 mL) was
/1) .
(250ꢁ/20 mm, Daicel, Japan) eluted with a mixture of n-
/
heptane/2-propanol/diethylamine (90:10:0.1). The two
2
D
2
0
enantiomers eluted at Rt. 46ꢁ
(cꢄ0.52, MeOH) and Rt. 74ꢁ
(cꢄ
/
64 min [a] ꢄ
/
ꢂ
/
67.78
0
3
/
/
98 min [a] ꢄ
/
ꢀ
/
65.68
D
added dropwise over 30 min. The mixture was heated at
reflux for 1 h and cooled to 0 8C. Water (1 mL), MeOH
/
0.51, MeOH). The configuration of the first eluting
isomer was found to be (S) by X-ray crystallography.
(
2 mL) and dichloromethane (2 mL) were added and the
mixture heated to 40 8C for 20 min. The organic phase
was separated and the aqueous phase was extracted with
4.4.3. [2-(4,5-Dihydro-1H-imidazol-2-yl)-1-thiophene-
2-ylethyl]pyridine (4b)
dichloromethane (3ꢃ
/
3 mL). The combined organic
The title compound was prepared from pyridine-4-
ylthiophene-2-ylmethanone using the method described
for 4c. The resulting product was purified twice on a
silica gel column. In the second purification EtOAc/
phases were dried (Na SO ) and evaporated to dryness.
4
2
Purification on silica gel using dichloromethane/MeOH/
triethylamine (45:5:3) as eluent afforded 70 mg (27%) of
3
-[2-(4,5-dihydro-1H-imidazol-2-yl)-1-thiophene-2-
ylethyl]pyridine 4c as an oil. LC/MS: m/z 258 (Mꢀ
ELS purity: 99%.
MeOH/ NH (25%) (4:1:1) was used as the eluent. 4b
3
ꢀ
ꢀ
/
1) ,
was isolated as an oil. MS: m/z 257 (M) .
4
.4.3.1. Resolution of (4b). Racemic 4b was separated
on Chiralcel OD using a mixture of 2-propanol/heptane/
diethylamine (50:50:0.1) as eluent. (ꢀ)-4b. Rt.: 10.4 min.
28.78 (2-propanol/heptan/
)-4b. Rt.: 13.0 min, purity:
26.38 (2-propanol/heptane/diethy-
4
2
.4.2. [2-(4,5-Dihydro-1H-imidazol-2-yl)-1-thiophene-
-ylethyl]pyridine (4a)
/
2
D
0
Purity: ꢀ
/
98.9% ee. [a] “
/
ꢀ
/
The title compound was prepared from pyridine-2-
ylthiophene-2-ylmethanone 1a using the method de-
1
scribed for 4c. H-NMR (CDCl ): d 8.55 (d, 1H); 7.65
diethylamine (50:50:0.1)). (ꢂ
/
2
0
ꢀ
/
98.2% ee. [a] ꢄ
/
ꢂ
/
D
lamine) (50:50:0.1)).
3
(
(
1
dt, 1H); 7.41 (d, 1H); 7.21 (dd, 1H); 7.15 (dd, 1H); 6.99
t, 1H); 6.91 (dd, 1H); 6.2ꢁ5.7 (broad, 1H); 5.02 (dd,
H); 3.60 (s, 4H); 3.58 (dd, 1H); 3.19 (dd, 1H).
NMR (CDCl ): d 168.5; 161.2; 149.4; 145.3; 137.3;
/
4.4.4. [2-(4,5-Dihydro-1H-imidazol-2-yl)-1-thiophene-
3-ylethyl]pyridine (4d)
1
3
C
The compound was prepared using the methodology
ꢀ
3
127.1; 125.7; 125.1; 123.9; 122.6; 47.5; 45.7; 34.8. Anal.
C H N S (C,H,N).
described for compound 4c. LC/MS: m/z 258 (Mꢀ
/
1) ,
ELS purity 99%.
1
4
15
3
4
.4.4.1. Resolution of (4d). Separation as described for
4b. (ꢀ) Rt.: 10.2 min; purity: 80.1% ee. (ꢂ) Rt.: 14.4
min.; purity: 99.5% ee. The assignment to (ꢀ) and (ꢂ)
was arbitrarily set in analogy to the findings for
compounds 4b and 4e.
4
.4.2.1. [2-(4,5-Dihydro-1H-imidazol-2-yl)-1-
/
/
thiophene-2-ylethyl]pyridine (4a). To a cooled (10 8C)
mixture of diethyl cyanomethylphosphonate (19.5 g,
/
/
0
0
.11 mol) in THF (400 mL) was added t-BuOK (14 g,
.125 mol). After stirring for 5 min 1a (18.9 g, 0.1 mol)
was added and the mixture was stirred at RT for 4 h.
The mixture was then poured on ice (300 mL), toluene
4.4.5. [2-(4,5-Dihydro-1H-imidazol-2-yl)-1-thiophene-
3-ylethyl]pyridine (4e)
(
100 mL) was added, the organic layer separated, dried
with Na SO and concentrated to give crude 19.11 g
Compound 4e was prepared from pyridine-4-ylthio-
phene-3-ylmethanone 1e as described for 4c. Yield 30%.
2
4
ꢀ
13
(
90%) of 5a as a Z/E-mixture. 5a (41 g, 0.19 mol) was
hydrogenated over Pd/C (10%, 3 g) in EtOH (300 mL) at
0 8C and 3.5ꢁ5.5 bar for 1 day, the mixture was filtered
EI/MS: m/z 257 (M) .
C NMR (CDCl ): d 165.8,
3
152.7, 150.3, 143.0, 127.5, 126.8, 123.4, 121.7, 49.9, 44.0,
35.7.
5
/
and more Pd/C (10%, 3 g) was added and the hydro-
genation continued. This procedure was repeated 2
times more resulting in 95% conversion to the saturated
compound 6a. The crude product was purified by
silicagel chromatography using toluene as eluent giving
4.4.5.1. Resolution of (4e). Separation as described for
2
0
4b. Rt.: 13.9 min, purity: ꢀ
/
99.3% ee, [a] ꢄ
/
ꢀ42.08 (2-
/
D
propanol/heptan/diethylamine (50:50:0.1)). Rt.: 17.6
2
0
D
min, purity: ꢀ
/
93.0% ee, [a] ꢄ
/
ꢂ41.08 (2-propanol/
/
3
2.57 g (80%) of pure 6a. This was reacted neat with dry
heptan/di-ethylamine (50:50:0.1)).
ethylenediamine mono-tosylate (2 equiv.) at 160 8C for 8
h. Subsequent cooling and addition of CHCl (300 mL),
4.4.6. [2-(4,5-Dihydro-1H-imidazol-2-yl)-1-thiophene-
3-ylethyl]pyridine (4f)
Z/E-3-[2-(4,5-Dihydro-1H-imidazol-2-yl)-1-thiophen-
3-ylvinyl]pyridine 7f (0.52 g) was hydrogenated over Pd/
3
followed by extraction with KOH solution (5%), separa-
tion of the organic phase, drying with Na SO , and
evaporation to dryness gave 31.3 g (80%) of 4a.
2
4

362
P. Jakobsen et al. / European Journal of Medicinal Chemistry 38 (2003) 357ꢁ362
/
C (10%, 50 mg) in ethanol (10 mL) at 30 psi for 4 h. The
reaction mixture was filtered through celite and evapo-
rated to dryness to give 360 mg (65%) 4f. LC/MS: m/z:
dichloromethane/methanol as eluent. Yield: 23.5 g light
ꢀ
1
red oil. LC/MS: m/z 321 (Mꢀ1) ; H-NMR (CDCl ):
/
3
d 4.14 (dq, 4H); 3.75 (m, 4H); 3.50 (d, 2H); 1.48 (s, 9H);
1.30 (t, 6H).
ꢀ
13
2
1
1
58 (Mꢀ
/
1) , ELS purity: 97%. C NMR (CDCl ): d
3
67.4, 149.5, 148.4, 143.0, 138.8, 135.7, 127.6, 126.8,
24.0, 121.7, 57.7, 48.1, 41.8, 35.1.
Acknowledgements
4.5. Z/E-3-[2-(4,5-Dihydro-1H-imidazol-2-yl)-1-
thiophen-3-yl-vinyl]pyridine (7f)
The authors would like to thank Drs. Hervør Lykke
Olsen, Jesper Gromada and Krister Bokvist for carrying
out the in vitro testing, Dr. Peter MacKay for perform-
ing the in vivo experiments, and Sandra Dols Rodriguez,
Bo Ritsmar Pedersen, Dorthe Egholm Jensen, and
Anette Peulicke Sørensen for their technical assistance.
Preparation from 3-thienyl-3-pyridyl ketone 1f (5 g)
and 2-(diethoxyphosphorylmethyl)-4,5-dihydro-imida-
zole-1-carboxylic acid tert-butyl ester 8 (8.45 g) by
means of n-BuLi (20 mL, 1.6 N in hexane) in THF at
ꢂ
/
78 8C. Subsequent stirring at ꢂ
/
20 8C for 20 min and
then at RT overnight followed by quenching by addition
of saturated aqueous NH Cl (25 mL) and extraction
with EtOAc afforded 9.2 g (98%) of BOC-protected 7f.
4
References
7
HCl (4 mL) and EtOAc (4 mL) by stirring overnight.
f (2 g, 5.63 mmol) was de-protected in a mixture of 3N
[1] D.J. Reis, P. Bousquet, A. Parini (Eds.), The Imidazoline
Receptor, Ann. NY Acad. Sci. 763 (1995).
[2] M. Gothert, G.J. Molderings, D.J. Reis (Eds.), Imidazoline
Receptors and their Endogenous Ligands, Ann. NY Acad. Sci.
Evaporation of the solvent afforded 1.2 g (90%) of 7f.
ꢀ
LC/MS m/z: 256 (Mꢀ1) .
/
8
81 (1999).
[
3] G.J. Molderings, Drug Future 22 (7) (1997) 757ꢁ
/772.
4.6. (Diethoxyphosphorylmethyl)-4,5-dihydro-1H-
imidazole-1-carboxylic acid tert-butyl ester (8)
[4] S.V. Zaitsev, et al., Ann. NY Acad. Sci. 881 (1999) 241ꢁ252.
/
[
5] F. Rondu, et al., J. Med. Chem. 40 (1997) 3793ꢁ
6] E.F. Pfeiffer, et al., Horm. Metab. Res. 6 (Suppl.) (1976) 112ꢁ
26.
7] N.G. Morgan, Exp. Opin. Invest. Drugs 8 (5) (1999) 575ꢁ
8] S.V. Zaitsev, et al., Diabetes 45 (1996) 1610ꢁ1618.
/
3803.
[
/
1
Diethyl cyanomethylphosphonate (50 g) was con-
verted to the imidate hydrochloride by addition of
[
[
/584.
/
ethanol (22 mL) and Et O (300 mL), cooling to 0 8C
[9] U. Hari, U. Roettig, M. Paal, Tetrahedron Lett. 42 (2001) 5187ꢁ
189.
10] M.A. Efanov, S.V. Zaitsev, H.-J. Mest, A. Raap, I.B. Appelkog,
O. Larsson, P.-O. Berggren, S. Efendic, Diabetes 50 (2001) 797ꢁ
02.
11] G. Morgan, S.L.F. Chan, C.A. Brown, E. Tsoli, Ann. NY Acad.
Sci. 763 (1995) 361ꢁ373.
12] S.L.F. Chan, C.A. Brown, K.E. Scarpello, N.G. Morgan, Ann.
NY Acad. Sci. 763 (1995) 153ꢁ156.
13] S. One et al., JP 04069385.
14] P. Oxley, W.F. Short, J. Chem. Soc. (1947) 497ꢁ
15] N. Sperber, R. Fricano, J. Am. Chem. Soc. 75 (1953) 2986ꢁ
/
2
5
followed by saturation with HCl (g).
The mixture was kept in a closed bottle for 48 h, and
subsequently evaporated to dryness.
The resulting imidate hydrochloride was added por-
tionwise to a mixture of ethylene diamine (16.6 g) in
ethanol (140 mL) and heated to 40 8C for 1h. Ethanol
[
/
8
[
[
/
/
(
350 mL) was added and the mixture concentrated
[
[
[
yielding 90% of 2-imidazolinylmethyl diethyl phospho-
nate hydrochloride. This crude product was dissolved in
dichloromethane (150 mL) and triethylamine (140 mL)
and reacted with BOC-anhydride (65 g) added in 3
portions over 30 min. The mixture was stirred at room
temperature for 20 h. Water was added and the organic
phase was isolated and evaporated to dryness. The crude
product was purified on a silicagel column using 95:5
/
505.
/
2988.
[16] W. Kuzmerkiewicz, F. Saczewski, H. Foks, Arch. Pharm. 319
1986) 830ꢁ834.
17] F. Ishikawa, Chem. Pharm. Bull. 28 (5) (1980) 1394ꢁ
18] P. Jakobsen, B.R. Pedersen, E. Persson, Bioorg. Med. Chem. 8
2000) 2095ꢁ2103.
19] J. Gromada, et al., J. Physiol. 518 (3) (1999) 745ꢁ
(
/
[
[
/
1402.
(
/
[
/759.
[20] H.S. Andersen, P. Jakobsen, P. Madsen, WO 200238559.