6-Chloro-3-alkylamino-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxide derivatives potently and selectively activate ATP sensitive potassium channels of pancreatic β-cells

Article abstract of DOI:10.1021/jm0208121

6-Chloro-3-alkylamino-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxide derivatives were synthesized and characterized as activators of adenosine 5′-triphosphate (ATP) sensitive potassium (K_ATP) channels in the β-cells by measuring effects on membrane potential and insulin release in vitro. The effects on vascular tissue in vitro were measured on rat aorta and small mesenteric vessels. Selected compounds were characterized as competitive inhibitors of [³H]glibenclamide binding to membranes of HEK293 cells expressing human SUR1/Kir6.2 and as potent inhibitors of insulin release in isolated rat islets. 6-Chloro-3-(1-methylcyclobutyl)amino-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxide (54) was found to bind and activate the SUR1/Kir6.2 K_ATP channels in the low nanomolar range and to be at least 1000 times more potent than the reference compound diazoxide with respect to inhibition of insulin release from rat islets. Several compounds, e.g., 3-propylamino- (30), 3-isopropylamino- (34), 3-(S)-sec-butylamino- (37), and 3-(1-methylcyclopropyl)amino-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxide (53), which were found to be potent and β-cell selective activators of K_ATP channels in vitro, were found to inhibit insulin secretion in rats with minimal effects on blood pressure and to exhibit good oral pharmacokinetic properties.

Full text of DOI:10.1021/jm0208121

J . Med. Chem. 2002, 45, 4171-4187
4171
6
-Ch lor o-3-a lk yla m in o-4H-th ien o[3,2-e]-1,2,4-th ia d ia zin e 1,1-Dioxid e Der iva tives
P oten tly a n d Selectively Activa te ATP Sen sitive P ota ssiu m Ch a n n els of
P a n cr ea tic â-Cells
Flemming E. Nielsen, Thora B. Bodvarsdottir, Anne Worsaae, Peter MacKay, Carsten E. Stidsen,
Harrie C. M. Boonen, Lone Pridal, Per O. G. Arkhammar, Philip Wahl, Lars Ynddal, Finn J unager,
Nils Dragsted, Tina M. Tagmose, J ohn P. Mogensen, Anette Koch, Svend P. Treppendahl, and J . Bondo Hansen*
Novo Nordisk Research and Development, Novo Nordisk Park, DK 2760 M a˚ løv, Denmark
Received J anuary 3, 2002
6
-Chloro-3-alkylamino-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxide derivatives were synthe-
sized and characterized as activators of adenosine 5′-triphosphate (ATP) sensitive potassium
(KATP) channels in the â-cells by measuring effects on membrane potential and insulin release
in vitro. The effects on vascular tissue in vitro were measured on rat aorta and small mesenteric
3
vessels. Selected compounds were characterized as competitive inhibitors of [ H]glibenclamide
binding to membranes of HEK293 cells expressing human SUR1/Kir6.2 and as potent inhibitors
of insulin release in isolated rat islets. 6-Chloro-3-(1-methylcyclobutyl)amino-4H-thieno[3,2-
e]-1,2,4-thiadiazine 1,1-dioxide (54) was found to bind and activate the SUR1/Kir6.2 KATP
channels in the low nanomolar range and to be at least 1000 times more potent than the
reference compound diazoxide with respect to inhibition of insulin release from rat islets. Several
compounds, e.g., 3-propylamino- (30), 3-isopropylamino- (34), 3-(S)-sec-butylamino- (37), and
3
-(1-methylcyclopropyl)amino-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxide (53), which were
found to be potent and â-cell selective activators of KATP channels in vitro, were found to inhibit
insulin secretion in rats with minimal effects on blood pressure and to exhibit good oral
pharmacokinetic properties.
In tr od u ction
Ch a r t 1
Diabetes is becoming a global health problem with
an increasing number of people affected by either type
1
or type 2 diabetes. Whereas type 1 diabetes is
characterized by rapid, autoimmune depletion of the
insulin-producing pancreatic â-cells, type 2 diabetes is
a slowly progressing disease affecting glucose and lipid
homeostasis, frequently associated with obesity and
elevated risk for cardiovascular diseases. A pivotal
component of type 2 diabetes is dysfunctional insulin
release including reduced first phase insulin release,
hyperinsulinaemia, and impaired insulin pulsatility.
Type 2 diabetes ultimately leads to â-cell degeneration
resulting in insufficient insulin release and the need for
administration of exogenous insulin.
In pancreatic â-cells, adenosine 5′-triphosphate (ATP)
sensitive potassium channels (KATP channels) couple
changes in blood glucose concentrations to insulin
The genes for two closely related SURs, SUR1 and
SUR2 have been cloned. Two different splice variants
of SUR2, SUR2A, and SUR2B have been reported.
SUR1 combines with Kir6.2 to form the KATP channels
of pancreatic â-cells, whereas the cardiac and the
skeletal muscle type consist of SUR2A and Kir6.2 and
the smooth muscle type consists of SUR2B and Kir6.1
or Kir6.2.
1
secretion. KATP channels also have important functions
in various other tissues, and their structure and func-
tion have been described in detail.2
-4
The KATP channel exists as an octameric complex of
the sulfonylurea receptor (SUR) and the pore-forming
inwardly rectifying potassium channel (Kir) in a 4 + 4
stoichiometry. Intracellular nucleotides and different
drugs regulate its activity. Whereas ATP and certain
sulfonylurea and benzoic acid insulin secretagogues are
inhibitors (blockers), MgADP (adenosine 5′-diphosphate)
and potassium channel openers stimulate potassium
currents.
Potassium channel openers, which activate the KATP
channels of vascular smooth muscle, are well-described.5
These include cyanoguanidines such as pinacidil (1),
chromanes such as cromakalim (2), and 1,2,4-thiadiaz-
6-9
ines such as diazoxide (3) (Chart 1).
Whereas pina-
cidil and cromakalim have little or no effect on KATP
1
0-12
channels of â-cells,
diazoxide nonselectively acti-
*
To whom correspondence should be addressed. Tel: +45 4443
4
857. Fax: +45 4443 4547. E-mail: jbha@novonordisk.com.
vates both KATP channels of â-cells (SUR1/Kir6.2) and
1
0.1021/jm0208121 CCC: $22.00 © 2002 American Chemical Society
Published on Web 08/16/2002

4
172 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Nielsen et al.
those of smooth muscle (SUR2B/Kir6.2) and has weak
stimulatory effects on KATP channels of the heart
Sch em e 1. Synthesis of
6-Chloro-3-alkylamino-4H-thieno[3,2-e]-1,2,4-thiadiazine
1,1-Dioxide Derivatives (Routes A-C)
a
1
3
(SUR2A/Kir6.2).
The ability of diazoxide to activate the KATP channels
of pancreatic â-cells results in inhibition of glucose-
stimulated insulin release.14,15 Human studies have
shown that diazoxide reduces glucose-stimulated insulin
release in healthy individuals16 and ameliorates the
abnormal hyperinsulinaemia in patients suffering from
1
7
insulinoma and nesidioblastosis (PHHI, persistent
1
8
hyperinsulinaemia and hypoglycemia of infancy). Sev-
eral preclinical and clinical studies have indicated that
diazoxide can be used to reduce the hyperinsulinaemia
commonly observed in obese, prediabetic individuals to
reduce weight and to prevent the development of type
2
diabetes in animal models.19,20 Recently, it has been
suggested that administration of diazoxide in combina-
tion with intensive insulin treatment, to induce â-cell
rest, will prevent or delay the progression of type 1
diabetes in newly diagnosed patients.21 The use of
diazoxide in this study was hampered by side effects
such as severe oedema and headaches. Additional side
effects associated with diazoxide include excessive hair
2
2
growth, which probably is related to the vasodilatory
effect of the compound.23
Structural modifications of the diazoxide molecule
have been described previously. These include varia-
tions of the size of the 3-alkyl substituent, in the
substitution of position 724,25 and in the introduction of
3
-alkylamino side chains to obtain hypotensive and
2
6,27
antiinflammatory compounds.
By combining com-
ponents from the structures of pinacidil and diazoxide,
2
8,29
Pirotte et al.
e]-1,2,4-thiadiazine 1,1-dioxides, among which BPDZ 44
4) (Chart 1) was reported to be a more potent and
have prepared a series of pyrido[4,3-
(
selective opener of KATP channels of pancreatic â-cells
than diazoxide. These structures have subsequently
been developed to include other potent benzo-1,2,4-
thiadiazine 1,1 dioxide derivatives.30
a
Reagents: (a) HCl(g) in EtOH, reflux. (b) NH2NH2‚H2O, room
temperature. (c) 1 N HCl, NaNO2, H2O, 0 °C. (d) Toluene, reflux.
(e) P O , N,N-dimethylcyclohexylamine, RNH ‚HCl, 180 °C. (f)
2
5
2
POCl3, pyridine, 100 °C or POCl3, pyridine‚HCl, H3PO4, 100 °C.
i
(
g) RNH2, ∆ or RNH2, R′OH (R′ ) Pr, Et), ∆ or RNH2, K2CO3,
In analogy, pinacidil-like cyanoguanidine derivatives
have recently been characterized as potent activators
of â-cell KATP channels with little effects on vascular
i
PrOH, ∆. (h) CsF, DMSO, 150 °C or KF, hexadecyltrimethylam-
monium bromide, NMP, 130 °C. (i) RNH2‚HCl, TEA, EtOH, 50-
100 °C.
3
1
smooth muscle.
On the basis of the above findings, we have synthe-
sized a series of 3-alkylamino-4H-thieno[3,2-e]-1,2,4-
thiadiazine 1,1-dioxides in which we have varied the
in Schemes 1-3. The strategy for the synthesis of key
intermediate 6-chloro-2,3-dihydro-3-oxo-4H-thieno[3,2-
e]-1,2,4-thiadiazine 1,1-dioxide (9) (Scheme 1) was based
on a previously reported synthetic approach to the
imidazo analogue 4,5-dihydroimidazo[4,5-e]-1,2,4-thia-
3
-alkylamino side chain to evaluate the structure-
activity relationship based on activity of pancreatic
â-cells and vascular smooth muscle in vitro and in vivo.
It is expected that such compounds could be valuable
tools for exploring the potential of reducing abnormal
hypersecretion of insulin as a treatment of metabolic
diseases. One compound of the series, described here,
NNC 55-0118 (34), has previously been found to protect
â-cells of rat islets when subjected to cytotoxic concen-
35
diazin-3(2H)-one 1,1-dioxide. The starting material
2
-(tert-butylamino)sulfonyl-5-chlorothiophene-3-carbox-
ylic acid (5) was prepared in two steps from 2-chlo-
rothiophene as described by Unterhalt and Moghadd-
36
am. The carboxylic acid 5 was converted to the
corresponding ethyl ester 6 with simultaneous cleavage
of the tert-butyl group by heating in anhydrous ethanol
saturated with hydrogen chloride. Treatment of 6 with
excess hydrazine hydrate at room temperature gave the
hydrazide 7, which was converted to the carbonyl azide
3
2
tration of streptozotocin in vitro, to prevent â-cell
3
3
degeneration in the BB rat model of type 1 diabetes
and to prevent development of type 2 diabetes in the
OLETF rat model of type 2 diabetes.34
8
in almost quantitative yield by reaction with nitrous
acid. Because of the potential explosion hazard of acyl
Ch em istr y
3
7
azides, crude 8 was cautiously dried without further
purification and then subjected to a Curtius rearrange-
ment by heating in refluxing dry toluene. The interme-
diate isocyanate spontaneously cyclized to the desired
The different synthetic routes used to prepare the
3
-alkylamino-6-chloro-4H-thieno[3,2-e]-1,2,4-thiadiaz-
ine 1,1-dioxides described in this paper are illustrated

Potassium Channels of Pancreatic â-Cells
Sch em e 2. Synthesis of
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4173
at elevated temperatures and had longer reaction times
without improving the yields due to side reactions.
6
1
-Chloro-3-alkylamino-4H-thieno[3,2-e]-1,2,4-thiadiazine
,1-Dioxide Derivatives (Route D)
a
To increase the nucleophilic substitution rate, we
changed the leaving group from chloro to fluoro as
outlined in Scheme 1 (route C), where the labile 3-chloro
substituent of 10 was displaced in a reaction with
cesium fluoride or activated potassium fluoride40 to give
the more reactive fluoro derivative 11. Compound 11
was either purified prior to amination (53-56) or
aminated in situ (57) but in low yield despite the better
leaving group. However, amination of 11 with methyl-
cyclopropylamine actually doubled the yield of 53 to 54%
as compared to that obtained with 10 as the starting
material.
An alternative synthesis (route D) of 3-alkylamino-
6
-chloro-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxides
is shown in Scheme 2. The known N-(tert-butyl)-5-
3
6
a
chloro-2-thiophenesulfonamide (12) was converted
to 3-amino-N-(tert-butyl)-5-chloro-2-thiophenesulfon-
amide (13) by ortho metalation with n-butyllithium
followed by quenching with p-toluenesulfonyl azide and
finally reduction with sodium borohydride in the pres-
ence of hexadecyltributylphosphonium bromide.4
Cleavage of the tert-butyl group in sulfonamide 13 by
heating in fuming hydrochloric acid afforded the desired
key intermediate 14. Compound 14 was reacted with
the appropriate alkyl isothiocyanate at 0-25 °C to give
the sulfonyl isothiourea derivative 15, which was readily
cyclized to the corresponding thieno[3,2-e]-1,2,4-thiadi-
azine 1,1-dioxide by treatment with phosgene in toluene
at 0 °C in the presence of triethylamine.
Reagents: (a) (i) n-BuLi, THF; (ii) p-toluenesulfonyl azide; (iii)
hexadecyltributylphosphonium bromide, NaBH4. (b) 12.5 N HCl,
0 °C. (c) t-BuOK, RNCS, DMF. (d) TEA, 1.93 M COCl2 in toluene,
THF.
6
compound 9. Our first attempt to obtain the target
compounds by direct amination of compound 9 (route
1-43
A) under conditions previously described for 2H-1,2,4-
benzothiadiazin-3(4H)-one 1,1-dioxides38 only resulted
in low yields (21 and 11%, respectively) of the desired
products 29 and 34. In route B (Scheme 1), compound
was first chlorinated with phosphorus oxychloride in
9
the presence of pyridine or preferable with activated
phosphorus oxychloride39 in the presence of pyridine
hydrochloride and ortho-phosphoric acid, to afford 3,6-
dichloro-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxide
10). Nucleophilic aromatic substitution of 10 with the
As shown in Scheme 3, it was possible to avoid direct
handling of the potential hazardous azides used in the
preparation of the required intermediates of routes A
and D. The acid 5 was converted to the ethyl ester 16,
from which the hydrazide 17 was obtained by reaction
with hydrazine hydrate in NMP. Compound 17 was
reacted with isoamyl nitrite in an acid-catalyzed Curtius
reaction in refluxing methanol followed by hydrochloric
acid cleavage of the tert-butyl group to afford the methyl
carbamate 18a without isolation of the in situ-formed
carbonyl azide. Alternatively, compound 18 could be
made from 8 as exemplified by preparation of the ethyl
carbamate 18b. The key intermediate 18 of route E was
then reacted with alkyl isothiocyanates in refluxing
(
appropriate amines gave the desired products in varying
yields (9-94%) by heating in a sealed flask. The
reactions were performed either neat in a large excess
of the amine, but preferentially with a few equivalents
as an alcoholic solution, or with 1-2 equiv in 2-propanol
in the presence of potassium carbonate. The best yields
were obtained with aqueous ammonia (27, 94%) or with
simple unbranched amines as, for example, methyl (28,
8
0%), while bulky amines generally resulted in low
yields (e.g., 53, 9%) due to incomplete conversion of 10
at the applied reaction conditions. The displacement
reactions with bulky amine nucleophiles were slow even
Sch em e 3. Synthesis of 6-Chloro-3-isopropylamino-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-Dioxide (34) (Route E)a
a
Reagents: (a) CH(OEt)3, reflux. (b) NH2NH2‚H2O, NMP. (c) (i) (CH3)2CHCH2CH2ONO, MeOH, TFA, reflux; (ii) MeOH, 12.5 N HCl,
reflux. (d) EtOH, reflux. (e) K2CO3, acetone, RNCS, reflux. (f) TEA, 1.93 M COCl2 in toluene, THF. (g) (i) 2 N NaOH, reflux or 1 N NaOH,
H2O, EtOH at pH 9-10; (ii) 4 N HCl.

4
174 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Nielsen et al.
Ta ble 1. pKa, LogP, LogDpH 7.4, and mLogP Values for
Compounds 30, 34, 36, 37, and 47
and the extracellular medium depending on the mem-
brane potential of the cells. Inside the cells, DiBAC
associates with proteins and thereby enhances its
fluorescence dramatically. After the cell is depolarized,
DiBAC accumulates inside the cells, increasing their
fluorescence. By repolarization, DiBAC escapes from the
cells and their fluorescence decreases.
compd
pKa
LogP
LogDpH 7.4
mLogP
3
3
3
3
4
0
4
6
7
7
8.23
8.35
8.54
8.42
8.29
2.49
1.84
2.57
2.42
3.17
2.42
1.63
2.52
2.38
3.14
2.01
2.01
2.31
2.31
2.60
The ability of the compounds to inhibit the release of
insulin from the mouse â-cell line, âTC6 incubated in
the presence of 22 mM glucose and 0.1 mM IBMX to
increase the insulin release, was determined. The
vasorelaxing effects of the compounds were initially
determined by their ability to relax phenylephrine-
contracted rat aorta rings as previously described.45
acetone followed by ring closure as described for 15
leading to the thieno[3,2-e]-1,2,4-thiadiazine carbamates
2
3-26 via the sulfonyl isothiourea derivatives 19-22,
respectively. Careful alkaline hydrolysis at pH 9-10
finally afforded the desired products 34, 62, 64, and 65
in modest overall yields.
In Table 2, the results from testing a series of
The physicochemical properties of selected compounds
have been examined (Table 1). The 3-alkylamino-6-
chloro 4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxides
were all weak acids with pKa values around 8.3. This
suggests that the compounds at physiological pH pri-
marily will be in a neutral form. The water/octanol
distribution coefficients LogDpH 7.4 were measured and
found to correlate to the calculated mLogP value. The
LogDpH 7.4 values were dependent on the hydrophobicity
of the 3-alkylamino-side chain ranging from 1.63 (34)
to 3.14 (47) all lying within an acceptable range for
6
-chloro-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxides
substituted in position 3 with linear alkylamino groups
are compared to the 3-amino derivative 27 and diazox-
ide. The data support that diazoxide is a nonselective
potassium channel opener, which hyperpolarizes mem-
branes of âTC3 cells, inhibits insulin release of âTC6
cells, and relaxes phenylephrine-contracted aortic rings
with approximately equal potency. This corresponds to
previous studies, which have shown that diazoxide
inhibits glucose-stimulated insulin release from rat
islets (IC50 ) 24.7 ( 3.4 µM) and relaxes KCl-induced
contraction of rat aortic rings (IC50 ) 22.0 ( 2.6 µM).4
In comparison, we found that pinacidil is a potent
vasodilator on rat aortic rings (EC50 ) 0.8 ( 0.2 µM,
data not shown) without exerting any effects on â-cell
membrane potential or glucose-stimulated insulin re-
lease at concentrations up to 100 µM.
4
4
potential drugs.
6,47
Resu lts a n d Discu ssion
In Vitr o Biology. On the basis of our aim to identify
potent and â-cell selective potassium channel openers,
we decided to initially screen our compounds in assays,
which predict activation of KATP channels of â-cells and
vascular smooth muscle (Tables 2-4).
The 6-chloro-3-alkylamino-4H-1,2,4-thieno[3,2-e]thia-
diazine 1,1-dioxides with primary alkylamino groups in
position 3 as presented in Table 2 were in general able
to activate KATP channels of â-cells as determined by
the ability of the compounds to hyperpolarize mem-
branes of âTC3 cells and to inhibit glucose-stimulated
insulin release of âTC6 cells. The size of the 3-substitu-
ent has importance for the inhibition of insulin release
from â-cells. The n-butyl (31) (IC50 ) 1.39 ( 0.63 µM),
the ethyl (29, IC50 ) 1.9 ( 0.52 µM), the propyl (30,
IC50 ) 2.4 ( 0.44 µM), and the octyl (33, IC50 ) 1.5 (
0.5 µM) derivatives are approximately equipotent but
considerably more potent than the unsubstituted ami-
noderivative (27, IC50 ) 21.1 ( 17.4 µM), the methyl
Glucose-induced pancreatic â-cell insulin release is
mainly a consequence of closure of ATP-regulated
potassium channels mediated by changes in the ATP/
ADP ratio. The closure leads to â-cell plasma membrane
depolarization, influx of Ca2 through voltage sensitive
+
2
+
calcium channels, increase in cytoplasmic Ca , and
activation of exocytosis of insulin from their pool of
secretory granules.
Qualitative changes in membrane potential caused
primarily by changes in potassium fluxes were assessed
in the glucose responsive insulin-producing murine cell
line âTC3, by measuring DiBAC4(3) fluorescence. The
probe is cell permeable and distributes between the cells
Ta ble 2. Structures and Biology Screening Data for Compounds 27-33 and Diazoxide
membranea
potential âTC3
insulin release
in vitro âTC6
relaxation rat
a
a
aorta rings
compd
R
EC50 (µM)
IC50 (µM)
Emax (%)
EC50 (µM)
2
2
2
3
3
3
3
7
8
9
0
1
2
3
H
CH3
CH2CH3
(CH2)2CH3
(CH2)3CH3
(CH2)5CH3
(CH2)7CH3
NTc
>100
21.1 ( 17.4
21.18 ( 9.61
1.9 ( 0.52
2.4 ( 0.44
1.39 ( 0.63
7.1 ( 1.6
27 ( 10.2
35 ( 7.11
74 ( 1.3
65 ( 3.0
66 ( 4.0
67 ( 3.7
76 ( 2.2
23 ( 3.9
262.0 ( 38.0
17.3 ( 8.8
135.5 ( 36.4
64.7 ( 9.2
13.4 ( 2.4
0.15 ( 0.15
176.0 ( 48
12.8 ( 2.5
9 ( 4
2.0 ( 0.8
0.6 ( 0.6
0.2 ( 0.2
0.5 ( 0.1
77.4 ( 4b
1.5 ( 0.5
diazoxide
22.4 ( 3.7
a
Shown are means ( SEM calculated from at least three independent experiments. b Data from Lebrun et al.30 NT, not tested.
c

Potassium Channels of Pancreatic â-Cells
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4175
Ta ble 3. Structures and Biology Screening Data for Compounds 34-48
membranea
potential âTC3
insulin release
in vitro âTC6
relaxation rat
aorta rings
a
a
compd
R
EC50 (µM)
IC50 (µM)
Emax (%)
EC50 (µM)
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
CH(CH3)2
0.6 ( 0.02
1.3 ( 1.0
0.3 ( 0.02
0.8 ( 0.1
0.4 ( 0.3
0.2 ( 0.1
41 ( 150
0.4 ( 0.05
2.0 ( 1.1
NT
0.23 ( 0.02
0.16 ( 0.03
0.24 ( 0.06
0.41 ( 0.15
0.17 ( 0.09
0.03 ( 0.01
0.22 ( 0.04
24.95 ( 13.2
4.89 ( 1.21
0.17 ( 0.01
0.56 ( 0.26
1.01 ( 0.53
0.24 ( 0.09
0.15 ( 0.03
0.75 ( 0.28
85 ( 1.1
74 ( 6.1
82 ( 2.3
79 ( 3.3
77 ( 5.0
82 ( 7.5
56 ( 3.84
36 ( 3.82
72 ( 2.31
79 ( 2.6
80 ( 3.17
62 ( 4.23
38 ( 1.36
48 ( 3.85
56 ( 5.3
266.0 ( 59.8
65.0 ( 10.9
70.0 (9.1
51.0 ( 15.1
169.0 ( 39.9
11.8 ( 3.4
83.0 ( 12.5
2.0 ( 0.9
25.2 ( 7.9
34.8 ( 3.5
>300
75.3 ( 12.5
8.0 ( 4.2
20.5 ( 6.8
2.0 ( 0.3
(RS)-CH(CH3)CH2CH3
(R)-CH(CH3)CH2CH3
(S)-CH(CH3)CH2CH3
C(CH3)3
C(CH3)2CH2CH3
(RS)-CH(CH3)(CH2)2CH3
(RS)-CH(CH3)(CH2)4CH3
(RS)-CH(CH3)(CH2)3CH(CH3)2
(RS)-CH(CH3)CH(CH3)2
(S)-CH(CH3)CH(CH3)2
(R)-CH(CH3)CH(CH3)2
CH2CH(CH2CH3)2
14 ( 33
0.3 ( 0.3
>100
(RS)-CH2CH(CH3)CH2CH3
(RS)-CH(CH3)C(CH3)3
2.0 ( 2.0
17 ( 30
a
Shown are means ( SEM calculated from at least three independent experiments; NT, not tested.
(
7
28, IC50 ) 21.18 ( 9.61 µM), and the hexyl (32, IC50 )
compounds could inhibit the insulin secretory process
by interaction with other targets than the KATP chan-
nels.
.1 ( 1.6 µM) derivatives.
In addition, the compounds have been tested for their
ability to potentiate the efflux of 86Rb from preloaded
RIN5F rat insulinoma cells. As expected for compounds
that activate KATP channels of â-cells, there is an overall
good correlation between the ability of the compounds
to increase the ion currents of RIN5F cells, to hyper-
polarize âTC3 cells, and to inhibit insulin release from
âTC6 cells.
In comparison, these compounds were in general
considerably less potent with respect to relaxation of
rat aorta rings. Potency again seems to be dependent
on the size of the alkyl group with the hexyl derivative
+
Compounds 34-58 exhibit major differences with
respect to vasodilatory activity. Among the compounds,
there are some vasorelaxants such as, for example, 41
(EC50 ) 2.0 ( 0.9 µM), but also compounds with little
or no effects on rat aorta like 34 (EC50 ) 266.0 ( 59.81
µM), 38 (EC50 ) 169.0 ( 39.9 µM), and 44 (EC50 >300
µM). Overall, there is a dependency on chain length for
activity on rat aorta rings. The most potent vasodilators
are those having a six carbon side chain as in 32 (EC50
) 0.15 ( 0.15 µM) and 41 (EC50 ) 2.0 ( 0.88 µM) with
only small effect of a single methyl substituent (for
example, 32 as compared to 41). Reducing the length
of the 3-alkylamino side chain to a 2-C or a 3-C entity
considerably reduces potency, giving rise to compounds
with considerable â-cell selectivity. This is in agreement
with a series of N-alkyl-N′-pyrido cyanoguanidine (pi-
3
2 being the most potent (EC50 ) 0.15 ( 0.15 µM) as
compared to the relatively inactive ethyl (29, EC50 )
35.5 ( 36.4 µM), octyl (33, EC50 ) 176.0 ( 48 µM),
1
and 3-amino (27, EC50 ) 262.0 ( 38.0 µM) derivatives.
Compound 32 is more potent than diazoxide (EC50 )
1
0
4
9
2.8 ( 2.5 µM) and as potent as pinacidil (EC50 ) 0.8 (
.2 µM).
In Table 3, a series of 6-chloro-4H-thieno[3,2-e]-1,2,4-
nacidil) derivatives and within a series of 3-alkyl-
amino-4H-pyrido[4,3-e]-1,2,4-thiadiazine 1,1-dioxides28
among which compounds with N-methyl and N-propyl
groups are weak vasodilators.
thiadiazine 1,1-dioxides substituted in position 3 with
branched alkylamino groups are presented. As found for
compounds having linear alkylamino substituents, size
seems to be important for â-cell activity. R-Methyl
substitution (for example, 34 as compared to 29 or 35
and 39 as compared to 30) increases potency. In
contrast, â-ethyl substitution (for example, 46 as com-
pared to 31) reduces potency. In this series of potassium
channel openers, chirality of the side chain bears little
importance for â-cell activity. This is in agreement with
data generated on a pair of pyrido-1,2,4-thiadiazine 1,1-
dioxides.48
The lipophilic compounds 40, 46, and 47 are in
general low efficacious inhibitors of glucose-stimulated
insulin release from âTC6 cells and stimulate mem-
brane hyperpolarization of âTC3 cells with low and
variable potency. Nonspecific interactions due to the
high lipophilicity or lack of aqueous solubility could
contribute to these inconsistencies. Alternatively, these
By contrast to the â-cell effects, the effects on vascular
tissue (rat aorta) seem to be dependent on chirality. The
(S) forms of the compounds (e.g., 37 as compared to 36
or 44 as compared to 45) are equal or less potent than
the corresponding (R) isomers. This is in accordance
with previous reports on pyrido-1,2,4-thiadiazine 1,1-
4
8
dioxide derivatives and on pinacidil derivatives acting
4
8,50
on vascular tissue.
6-Chloro-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-diox-
ides substituted in position 3 with cycloalkylamino
groups are presented in Table 4. The size of the
carbocycle is important for potency with respect to
hyperpolarization of â-cell membranes and inhibiton of
insulin release with the cyclobutyl (50) and the cyclo-
pentyl (51) being more potent than the cyclopropyl (49)
and the cyclohexyl (52) derivatives. Increasing the
branching at the R-atom as in compounds 53-55 and
57 leads to considerably increased potency. Among these

4
176 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Nielsen et al.
Ta ble 4. Structures and Biology Screening Data for Compounds 49-57
a
membrane
potential âTC3
insulin release
in vitro âTC6
relaxation rat
a
a
aorta rings
compd
R
n
EC50 (µM)
IC50 (µM)
Emax (%)
EC50 (µM)
4
5
5
5
5
5
5
5
5
9
0
1
2
3
4
5
6
7
H
H
H
H
CH3
CH3
CH3
CH3
CH2CH3
1
2
3
4
1
2
3
4
2
>30
7.38 ( 1.23
0.49 ( 0.21
1.15 ( 0.18
25.57 ( 19.35
0.15 ( 0.03
0.02 ( 0.004
0.07 ( 0.03
5.33 ( 3.84
0.02 ( 0.01
37 ( 4.5
78 ( 4.64
67 ( 3.0
40 ( 2.76
69 ( 5.0
91 ( 1.1
79 ( 5.7
62 ( 3.0
81 ( 4.5
28.2 ( 9.6
27 ( 10.8
>300
6.8 ( 1.9
119.2 ( 12.1
6.5 ( 1.7
6.2 ( 0.7
1.4 ( 0.8
3.3 ( 2.0
1.4 ( 1.8
0.4 ( 0.4
8 ( 6
0.6 ( 0.3
0.014 ( 0.6
0.08 ( 0.08
1.0 ( 1.0
0.1 ( 0.07
a
Shown are means ( SEM calculated from at least three independent experiments.
3
Ta ble 5. Inhibition of [ H]Glibenclamide Binding to HEK293 Cells Expressing SUR1/Kir6.2 and Glucose-Stimulated Insulin Release
from Rat Islets; Relaxation of Precontracted Rat Mesenteric Arteries
[3
binding IC50 (µM)
H]glibenclamide
a
insulin release
rat isletsb
vasodilation of
mesenteric arteries
c
+
ATP
-ATP
compd
H
L
IC50 (µM)
EC50 (µM)
3
3
3
3
3
3
5
5
5
5
0
4
6
7
8
9
0
1
3
4
2.27 ( 0.89
0.34 ( 0.16
0.64 ( 0.39
0.46 ( 0.31
0.23 ( 0.12
0.18 ( 0.10
0.12 ( 0.04
NT
584 ( 229
187 ( 42
224 ( 79
309 ( 109
393 ( 169
375 ( 93
146 ( 36
NT
924 ( 148
554 ( 64
487 ( 89
420 ( 48
1350 ( 885
899 ( 420
943 ( 536
NT
0.7 ( 0.17
0.18 ( 0.08
23.3 ( 5.4
6.2 ( 2.1
0.06 ( 0.03
25.2 ( 7.0
19.2 ( 8.4
60.6 ( 26.6
29.4 ( 9.2
38.6 ( 0.9
10.8 ( 4.3
14.3 ( 5.0
7.2 ( 1.7
0.07 ( 0.02 (2)
0.06 ( 0.05 (2)
0.02 ( 0.01 (2)
0.05 ( 0.05 (2)
0.44 ( 0.07 (2)
0.21 ( 0.08
0.32 ( 0.14
0.003 ( 0.001
165 ( 55
10.8 ( 5.0
198 ( 36
685 ( 110
859 ( 213
750 ( 120
0.01 ( 0 (2)
20.28 ( 11.7
diazoxide
20.8 ( 6.1
a
3
IC50 values of KATP channel modulators from [ H]glibenclamide competition studies in the absence or presence of 2 mM MgATP. In
3
the presence of MgATP, [ H]glibenclamide is displaced in a biphasic manner by KATP channel openers from a high (H) and low (L) affinity
state. Inhibition of glucose (22 mM)-stimulated insulin release from isolated Wistar rat islets. c Effects on Wistar rat mesenteric arteries
precontracted with phenylephrine. Shown are means ( SEM of three or more independent experiments unless indicated (n); NT, not
tested.
b
compounds, 54 with the methylcyclobutyl amino side
chain is more potent (IC50 ) 0.02 ( 0.004 µM) than the
methylcyclopropyl 53 (IC50 ) 0.15 ( 0.03 µM), the
methylcyclopentyl 55 (IC50 ) 0.07 ( 0.03 µM), or the
methylcyclohexyl derivative 56 (IC50 ) 5.33 ( 3.84 µM)
with respect to inhibition of insulin release. In compari-
son, the ethylcyclobutyl derivative 57 (IC50 ) 0.02 (
The present results and previous reports indicate that
the size and the shape of the alkylamino side chains of
KATP channel openers of the cyanoguanidine and the
thiadiazine 1,1-dioxide series are very important for the
vascular activity. These lipophilic groups could occupy
a pocket on the SUR2B subunit necessary for inducing
ion channel activation.
0
.01 µM) is more potent than the cyclobutylamino
On the basis of potency and â-cell selectivity, com-
pounds 30, 34, 36-39, 50, 51, 53, and 54 were selected
for extensive evaluation with respect to effects on
recombined human SUR1/Kir6.2 KATP channels, freshly
isolated rat islets and vascular tissue in vitro (Table 5),
and on blood pressure and insulin release in vivo (Table
6).
compound 50 (IC50 ) 0.49 ( 0.21 µM) and equipotent
with 54. The increased â-cell activity of the branched
cycloalkylamine derivatives is in accordance with the
high potency of the tert-butyl and tert-amyl derivatives
3
8 and 39 as compared to, for example, the isopropyl
derivative 34. In this series, there is an overall good
correlation between the effects on âTC3 cells and the
effects on âTC6 cells. With respect to smooth muscle
effects, the size of the cycloalkyl groups increases
potency exemplified by 56 (methylcyclohexyl) (EC50 )
To measure the direct interaction of the compounds
with recombined KATP channels of the â-cells, we
examined the displacement of radiolabeled glibencla-
mide from membranes of HEK293 cells expressing
human SUR1/Kir6.2 potassium channels. Glibenclamide
potently binds to the SUR1 part of the â-cell K
1
.4 ( 0.8 µM), which is more potent than 55 (methyl-
cyclopentyl) (EC50 ) 6.2 ( 0.7 µM) and 54 (methylcy-
clobutyl) (EC50 ) 6.5 ( 1.7 µM) and considerably more
potent than 53 (methylcyclopropyl) (EC50 ) 119.2 ( 12.1
µM). The ethylcyclopropyl derivative 57 (EC50 ) 3.3 (
ATP
5
1
52,53
channels, and it has previously been shown
that
3
competition by diazoxide of [ H]glibenclamide binding
to SUR1/Kir6.2 is highly dependent on ATP and Mg2
+
.
2
.0 µM) is slightly less potent than 56 and approxi-
This was found to be the case also for compounds 30,
34, 36-39, 50, 53, and 54 (Figure 1, Table 5), whereas
mately equipotent with 54.

Potassium Channels of Pancreatic â-Cells
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4177
3
F igu r e 1. [ H]Glibenclamide (1 nM) binding to membrane preparations from HEK293 cells stably expressing human SUR1/
2
+
Kir6.2. Binding was displaced in the absence (open symbols) or presence (closed symbols) of 2 mM ATP and 2 mM Mg by (A)
3
compound 34, (B) compound 54, (C) diazoxide, and (D) glibenclamide. In the presence of MgATP, [ H]glibenclamide is displaced
in a biphasic manner by KATP channel openers 34 and 54 from a high (H) and low (L) affinity state. Values are mean ( SEM of
three or more independent experiments.
competition by glibenclamide is independent of ATP and
rat islets occurs at concentrations 2-3 times lower than
those inhibiting insulin release from the murine âTC6
cell line. The effect of diazoxide on rat islets (IC50 )
20.28 ( 11.7 µM) is similar to that reported previously
2
+
Mg (pIC50 ) 8.73 ( 0.06 µM (+ATP); pIC50 ) 8.66 (
.05 µM (-ATP)) (Figure 1). The competition curves
0
indicate that these compounds compete with glibencla-
mide in a biphasic manner, giving rise to two inhibition
constants: a high affinity component, which largely
correlates with the potency of the compounds in the
â-cell membrane potential and the insulin release
assays, and a low affinity component, which correlates
with the inhibition by the present compounds in the
4
7
(IC50 ) 24.8 ( 3.4 µM). The high potency of compound
54 in this assay (IC50 ) 0.01 µM) corresponds to its
potency on âTC6 cells (IC50 ) 0.002 µM) and its ability
3
to displace [ H]glibenclamide from human SUR1/Kir6.2
channels (IC50 ) 0.003 µM).
All of the selected compounds of this series induced
concentration-dependent maximal vasodilatation of small
mesenteric arteries that were precontracted with phen-
ylephrine (Table 5). The relative high potency with
respect to dilatation of mesenteric arteries of all of the
tested compounds is surprising in view of the observed
effects in isolated rat aorta. Recent studies, however,
have suggested that the KATP channels of some vascular
tissues consist of SUR2B and Kir6.1 instead of Kir6.2.5
This potentially could lead to the discrepancy between
the effects on aorta and mesenteric arteries. To further
examine the effects of these compounds on smooth
muscle, 30, 36, and 37 were added to rat bladder rings
that were precontracted with the muscarinic agonist
bethanechol. The compounds were only very weak
dilators of bladder tissue (30, IC50 ) 146.3 ( 46.9 µM;
36, IC50 ) 204.1 ( 70.2 µM; and 37, IC50 > 300 µM;
data not shown). Overall, the 6-chloro-4H-thieno[3,2-
e]-1,2,4-thiadiazine 1,1-dioxides presented in Table 5
exhibit a high degree of â-cell selectivity as measured
by their ability to inhibit insulin release from rat islets
as compared to their vasorelaxant potency. In contrast,
diazoxide affects rat islets and mesenteric vessels with
equal potency.
2
+
absence of ATP and Mg . In contrast, diazoxide exhibits
a monophasic competition curve (Figure 1) suggesting
that its mode of interaction with the KATP channel is
different from the new 6-chloro-4H-thieno[3,2-e]-1,2,4-
thiadiazine 1,1-dioxide derivatives. Selected compounds
(
[
i.e., 30 and 36) were found also to inhibit binding of
H]glibenclamide in an ATP-dependent manner to
3
5,56
membranes of HEK293 cells expressing SUR1 but not
Kir6.2 (data not shown). This suggests that these
compounds bind to the SUR part of the KATP channel
complex to induce channel activation.
When measured by patch clamp in the whole cell
configuration, the present compounds (e.g., 30, 34, and
5
3) potently increase the ion current of human SUR1/
Kir6.2, stably expressed in a HEK293 cell line and
tranciently expressed in Xenopus oocytes, but do not
activate ion currents of SUR2A/Kir6.2 or SUR2B/Kir6.2
channels (data not shown).54
When freshly isolated rat pancreatic islets are sub-
jected to a high glucose concentration, insulin secretion
is stimulated. Compounds 30, 34, 36-39, 50, 51, 53,
5
4, and diazoxide dose dependently inhibit the insulin
secretion with an order of potency similar to that found
in the â-cell screening assay discussed above (Table 5).
In general, the inhibitory effect on insulin release from
The in vitro results strongly indicate that the present
potassium channel openers, e.g., 30, 34, 36, 37, 53, and

4
178 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Nielsen et al.
Ta ble 6. Acute Effects on Plasma Insulin Levels and Blood Pressure,a and Pharmacokinetic Propertiesb in Rats
pharmacokinetics
max reduction in
plasma insulin (%)
reduction in plasma
insulin, 30 min (%)c
max change
in MAP (%)d
c
MAPAUC (%)d
fpo (%)b
t1/2 (min)b
compd
3
3
3
3
3
3
5
5
5
5
0
4
6
7
8
9
0
1
3
4
100 ( 0
96 ( 1
97 ( 2
96 ( 2
85 ( 8
87 ( 7
100 ( 0
99 ( 1
85 ( 14
82 ( 8
53 ( 8
16 ( 5
96 ( 1
13 ( 2
-19 ( 4
-17 ( 2
-7 ( 4
-36 ( 13
-34 ( 9
5 ( 4
ND
69
65
90
49
NT
47
72
164
163
58
NT
99
113
NT
250
NT
79
-1187 ( 314
-1133 ( 923
-362 ( 930
-7205 ( 2574
-7870 ( 2676
-1237 ( 572
3133 ( 997
-268 ( 240
-10950 ( 3277
-3008 ( 195
94 ( 2
94 ( 6
71 ( 12
77 ( 16
100 ( 0
22 ( 9
49
29 ( 5
NT
103
NT
154
77 ( 23
76 ( 13
53 ( 8
-4 ( 3
-48 ( 7
-26 ( 2
diazoxide
a
Effects of compounds (3 mg/kg, except for compound 54, 0.3 mg/kg) on plasma insulin levels and mean arterial blood pressure (MAP)
b
after iv administration to anaesthetized Sprague-Dawley rats. Values are mean ( SEM of 2-4 animals.c Oral bioavailability (fpo) and
half-life (t1/2) after oral administration to Sprague-Dawley rats. The values are means of two animals. Maximal reduction in plasma
insulin and reduction in plasma insulin 30 min postinjection (calculated as percentage reduction of the level prior to administration).
d
Maximal change in MAP and area under the MAP curve measured over 5 min; ND, not determined; NT, not tested.
5
4 through binding to the SUR1 receptor, are potent
both as percent maximal change from baseline and as
the area under the curve measured during the first 5
min after administration. Compounds 30, 37, 50, and
53 had minimal effects on blood pressure while 34 and
36 decreased blood pressure less potently than diazox-
ide. Compounds 38, 39, 51, and 54 markedly affected
blood pressure. The potent hypotensive effect of 54 at
0.3 mg/kg was retained at 0.1 mg/kg iv (data not shown)
indicating that the compound in vivo is a considerably
more potent hypotensive drug than expected from the
in vitro data. In analogy, the hypotensive effects of 38
and 39 and the hypertensive effect of 51 were not
predicted from the screening assays and highlight that
both in vitro and in vivo tests are needed to verify the
selectivity of the present compounds.
activators of KATP channels of pancreatic â-cells with
minimal effects on vascular KATP channels. On the basis
of studies on recombined ATP-activated potassium
channels (SUR2A/Kir6.2, data not shown), it is unlikely
that these compounds will exert potassium channel
opening effects on cardiac tissues. When subjected to a
PanLab in vitro pharma screen, 34 and 53 (10 µM) did
not show any significant effects on a series of receptors
and ion channels. In contrast to diazoxide,57 34 does not
potentiate the effect of glutamate on AMPA receptors
(data not shown).
Among the tested compounds, 54 stands out as being
the most potent activator of â-cell KATP channels ever
described. With respect to all parameters tested in vitro,
this compound is at least 1000 times more potent than
diazoxide. This is the first example of a compound that
is able to activate the SUR1/Kir6.2 KATP channels in the
low nanomolar range. Only KATP channel blockers of
SUR1/Kir6.2 channels and activators of SUR2A/Kir6.2
and SUR2B/Kir6.2 channels have previously been de-
scribed to act with similar potency.58
In Vivo Biology. Diazoxide has been shown to
inhibit insulin release in several species including
humans
be attributed to the high bioavailability and long half-
life.61 Diazoxide induces a significant hypotensive re-
sponse when given either intravenously (iv) or per os
Pharmacokinetic studies in Sprague-Dawley rats
using cassette dosing (Table 6) have shown that com-
pounds 30, 34, 36, 37, 39, 50, and 53 all have good oral
bioavailability (f , 47-103%) with a plasma t after
po
1/2
oral administration ranging from 58 (37) to 250 min
(53). In comparison, diazoxide was found to have an oral
bioavailability of 154% with a plasma t1/2 of 79 min after
po administration. This is consistent with a reported
6
2
plasma half-life of diazoxide of 1-2 h in rats. The high
oral bioavailability (fpo) above 100% most likely indicates
hepatic recirculation. Despite the good oral bioavail-
ability and the ability to potently inhibit insulin release
upon iv administration, 50 was not able to reduce
plasma insulin levels in obese Zucker rats after per os
administration (data not shown).
2
1,59,60
with a long lasting effect that can largely
5
9
(po). Using anaesthetized, postprandial adult Spra-
gue-Dawley rats as a screening model, we examined
the effects of selected 6-chloro-3-alkylamino-4H-thieno-
To determine the pharmacodynamic properties of the
compounds, 34 was given per os to rats (Figure 2).
Compound 34 suppressed the mean plasma insulin
levels both before (2.8 ( 2.8 vs 43.0 ( 3.5 pmol/L in
vehicle-treated rats, p < 0.01) and after glucose chal-
lenge (66.1 ( 37.1 vs 268 ( 39.2 pmol/L in vehicle-
treated rats, p < 0.05) when dosed 3 h before the latter
(Figure 2A). Although the insulin secretion at 3 h after
administration of 34 was suppressed by the potassium
channel opener, the â-cells were able to respond to the
glucose challenge. When dosed 6 h before glucose, no
suppression of resting plasma insulin was seen but
glucose-stimulated insulin was still reduced to 100.3 (
23.5 pmol/L. No effect on resting or glucose-stimulated
insulin was seen in rats dosed 12 h or more before
challenge.
[3,2-e]-1,2,4-thiadiazine 1,1-dioxide derivatives (30, 34,
3
6-39, 50, 51, 53, 54, and diazoxide) for effects on
insulin release and blood pressure upon iv administra-
tion of a fixed dose of 3 mg/kg for all compounds except
5
selected 6-chloro-3-alkylamino-4H-thieno[3,2-e]-1,2,4-
thiadiazine 1,1-dioxide derivatives were equally effica-
cious (>80%) in decreasing plasma insulin levels al-
though the effects of 30 and 51 were of shorter duration
4, for which 0.3 mg/kg was used (Table 6). All of the
(measured as percent decrease in plasma insulin at 30
min). In comparison, an identical dose of diazoxide (3
mg/kg iv) only gave a partial (53%) but sustained
reduction of the concentration of insulin in plasma.
In the present study, diazoxide (3 mg/kg iv) was able
to reduce mean arterial blood pressure when determined

Potassium Channels of Pancreatic â-Cells
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4179
significant effects on blood pressure, heart rate, and
body temperature at the doses tested.
Con clu sion
By using structural elements from the cyanoguani-
dine, pinacidil, and the thiadiazine 1,1-dioxide, diazox-
ide, a series of potent activators of ATP-regulated
potassium channels have been synthesized. Selected
3
-alkylamino-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-di-
oxides of the described series bind to the SUR1 compo-
nent of the SUR1/Kir6.2 KATP channel and increase
channel activity to inhibit glucose-stimulated insulin
release. The activation of the KATP channel seems to be
species-independent suggested by the fact that the
compounds bind to and activate recombined human
KATP channels and inhibit insulin release from rat islets
and from mouse â-cell lines with approximately the
same order of potency. Although previous reports have
described compounds that are able to suppress insulin
release through activation of ATP-regulated potassium
2
8,29-31
channels,
the present compounds, e.g., 30, 34, 36,
3
7, 50, and 53, seem superior by showing good phar-
macokinetic and pharmacodynamic properties. Com-
pounds such as 34 (NNC 55-0118) and 53 (NN414)
represent a new class of potassium channel openers
highly potent and selective for the SUR1/Kir6.2 KATP
channels both in vitro and in vivo. In addition, 34 is
able to reduce plasma insulin levels after per oral
administration to normal rats with minimal effect on
cardiovascular parameters and therefore could be ex-
pected to possess less side effects than diazoxide.
3
2
Compound 34 has been found to protect â-cells in vitro
and in vivo
2
3
2,33
and to prevent the development of type
diabetes in OLETF rats.34 The present data strongly
suggest that these effects could be mediated through
activation of SUR1/Kir6.2 potassium channels. Of par-
ticular interest is also compound 54, which is able to
activate the SUR1/Kir6.2 KATP channels in the nano-
molar range. Additional studies are needed for full
exploration of the pharmacology and therapeutic po-
tentials of these molecules.
F igu r e 2. Plasma insulin (A) and blood glucose (B) levels in
Sprague-Dawley rats pre- and postglucose challenge at dif-
ferent timepoints after per os administration of compound 34
(
30 mg/kg). The values at t ) 0 h indicate administration of
vehicle. *p < 0.01 as compared to untreated t ) 0, **p < 0.05
as compared to glucose challenged rats (t ) 0) (ANOVA).
Exp er im en ta l Section
Ch em istr y. Melting points were determined in open capil-
lary tubes on a B u¨ chi B-545 apparatus and are uncorrected.
H NMR spectra were recorded on a Bruker DPX 200 (200
MHz), a Bruker DRX 300 (300 MHz), or a Bruker AMX 400
Figure 2B shows mean blood glucose values in the
rats before and after glucose challenge. A slight but
statistically insignificant blood glucose increase was
seen before challenge at 3 and 6 h. In combination,
Figure 2A,B suggests that 34 (30 mg/kg), after per oral
administration to normal rats, reduces both resting and
glucose-stimulated plasma insulin for up to 6 h without
inducing resting hyperglycemia.
1
(
400 MHz) spectrometer with tetramethylsilane as an internal
standard. Chemical shifts are given in parts per million (δ),
and splitting patterns are designated as follows: s, singlet; d,
doublet; t, triplet; q, quartet; quint, quintet; sext, sextet; sept,
septet; m, multiplet; and br, broad. The 70 eV EI solid mass
spectra were recorded on a Finnigan MAT-TSQ 70 mass
spectrometer. Liquid chromatography-mass spectrometry
To measure effects on cardiovascular parameters in
conscious animals, 54 (30 mg/kg po), 30, 36, 37, and 53
(
LC-MS) analysis was performed on HP1100 MSD equipped
with Waters Xterra MS C-18 × 3 mm column. IR spectra were
recorded on a Perkin-Elmer 1600 Series Fourier transform
infrared (FT-IR) spectrophotometer. Reactions were followed
by thin-layer chromatography (TLC) performed on silica gel
60 F254 (Merck) or ALUGRAMSIL G/UV254 (MACHEREY-
NAGEL) TLC aluminum sheets. Elemental analyses (C, H,
N) were performed by Novo Nordisk, Microanalytical Labora-
tory, Denmark, and were within (0.4%.
(
100 mg/kg po), 34 (100 and 300 mg/kg po), and
diazoxide (100 and 300 mg/kg po) were given to female
Sprague-Dawley rats. Blood pressure, heart rate, and
body temperature were measured by telemetry for 4.5
(30, 36, 37, and 53), 24 (54), or 48 h (34 and diazoxide).
Diazoxide induced a minor but significant and long-
lasting increase in heart rate at 100 mg/kg and a
pronounced effect on heart rate and mean arterial blood
pressure at 300 mg/kg (Figure 3A). Compound 54
induced a potent depression of blood pressure and heart
rate. Neither 30, 36, 37, 53, nor 34 (Figure 3B) had any
Eth yl 5-Ch lor o-2-su lfam oylth ioph en e-3-car boxylate (6).
A solution of 53
2,36
(60.0 g; 0.201 mol) in 700 mL of absolute
EtOH saturated with HCl gas was heated at reflux with
stirring for 17 h. The cooled solution was evaporated to
dryness, and the residue was triturated with water and dried

4
180 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Nielsen et al.
F igu r e 3. Mean values of mean arterial blood pressure, heart rate, and body temperature from vehicle-treated (n ) 4) (b),
diazoxide-treated (n ) 3), or 34-treated (n ) 3) telemetered rats (100 (4) or 300 mg/kg (O)). Arrows indicate time of dosing. There
was a decrease in mean blood pressure and an increase in heart rate at 100 and 300 mg/kg and a decrease in body temperature
after 300 mg/kg diazoxide dosing (left panels of A). Compound 34 had no effect on mean blood pressure, heart rate, or body
temperature (right panels of B).
1
to give 52.3 g (96%) of 6; mp 113-114 °C. H NMR (DMSO-
such a rate that the temperature did not exceed 10 °C. The
mixture was then heated at 95-100 °C for 16 h and cooled to
room temperature. A yellow byproduct was removed by filtra-
tion, and the filtrate was concentrated in vacuo. To the residue
was added 200 g of ice, and the resulting mixture was stirred
for 1 h and filtered. The isolated solid was dissolved in
d , 200 MHz): δ 1.31 (t, 3H), 4.32 (q, 2H), 7.55 (s, 1H), 7.77
6
+
(
br s, 2H). LC-MS m/z 270/272 (M + H) . Anal. (C
C, H, N.
-Ch lor o-3-(h yd r a zin oca r b on yl)t h iop h en e-2-su lfon -
7 8 4 2
H ClNO S )
5
a m id e (7). The ethyl ester 6 (50.0 g; 0.185 mol) was added in
one portion to 98% hydrazine hydrate (50 mL) with stirring
at ambient temperature. The reaction was slightly exothermic.
The solution was stirred for 90 min and concentrated in vacuo.
The residue was crystallized by trituration with 250 mL of
water, and the mixture was acidified with 12.5 N HCl and
stirred for 30 min at 0 °C. The product was isolated by
filtration, washed with water, and dried to give 42.4 g (89%)
, 200 MHz): δ 4.7
br s, 2H), 7.49 (s, 1H), 7.71 (br s, 2H), 9.94 (br s, 1H). LC-MS
H ClN O S ) C, H, N.
5 6 3 3 2
-Ch lor o-2-su lfa m oylth iop h en e-3-ca r bon yl Azid e (8).
A solution of sodium nitrite (3.0 g; 43.4 mmol) in 20 mL of
water was added dropwise with stirring at 0 °C to a solution
of 7 (10.0 g; 39.1 mmol) in 80 mL of 1 N HCl, and the mixture
was stirred for 15 min. The precipitate was isolated by
filtration, washed with water, and dried in vacuo over phos-
3
saturated aqueous NaHCO (100 mL) and filtered to remove
a small amount of insoluble material. After treatment with
decolorizing charcoal, the filtrate was acidified with 4 N HCl
and the precipitate that formed was filtered, washed with
water, and dried to give 5.55 g (48%) of 10; mp >240 °C (dec).
1
H NMR (DMSO-d , 300 MHz): δ 7.08 (s, 1H), no NH signal
6
was seen (hidden in a very broad H
258/260 (M ). Anal. (C
2
O peak). EI-MS m/z 256/
‚0.5H O) C, H, N.
1
+
of 7; mp 158-160 °C. H NMR (DMSO-d
6
5
H
2
Cl
2
N
2
O
2
S
2
2
(
Meth od B. A mixture of 9 (1.60 g, 6.7 mmol), phosphorus
oxychloride (7 mL), pyridine hydrochloride (2.55 g, 22 mmol),
and 85% ortho-phosphoric acid (0.47 mL, 4.1 mmol) was heated
at 100 °C for 15 h. The mixture was concentrated in vacuo at
50 °C, and the residual oil was carefully treated with ice water
(20 mL) with stirring at 0 °C. The crude product was isolated
by filtration, washed with water, and dissolved in saturated
+
m/z 256/258 (M + H) . Anal. (C
5
3
aqueous NaHCO (25 mL). After treatment with decolorizing
phorus pentoxide at room temperature to give 9.93 g (96%) of
charcoal, the filtrate was acidified with 12.5 N HCl with
stirring at 0 °C and the white precipitate that formed was
filtered, washed with water, and dried to give 1.26 g (73%) of
10.
1
8
2
. H NMR (DMSO-d
6
, 200 MHz): δ 7.55 (s, 1H), 7.97 (br s,
H). IR (KBr, cm -1): 3276 (NH); 2155, 1211 (N
).
-Ch lor o-2,3-d ih yd r o-3-oxo-4H-th ien o[3,2-e]-1,2,4-th ia -
); 1659 (Cd
3
O); 1349, 1173 (SO
2
6
6-Ch lor o-3-flu or o-4H-th ien o[3,2-e]-1,2,4-th iadiazin e 1,1-
Dioxid e (11). A mixture of compound 10 (2.57 g, 10 mmol)
and anhydrous cesium fluoride (6.09 g, 40 mmol) in dry DMSO
(10 mL) was stirred for 18 h at 150 °C in a sealed flask. The
cooled mixture was diluted with water (30 mL) and stirred
for 5 min before collection of the precipitate. The beige solid
was washed with water and dried to give 2.12 g (88%) of crude
d ia zin e 1,1-Dioxid e (9). Carbonyl azide 8 (7.0 g; 26.2 mmol)
was added in portions to 50 mL of refluxing dry toluene during
1
0 min. The mixture was stirred for 5 min and cooled on an
ice bath. The precipitate was isolated by filtration, washed
with toluene, and dried to give 5.90 g (94%) of 9; mp 258-260
1
°
C (decomposition with gas evolution). H NMR (DMSO-d
6
, 300
1
MHz): δ 6.93 (s, 1H), ca. 8 (very br s, 1H), 11.98 (s, 1H). LC-
11, which was used without further purification. H NMR
+
MS m/z 239/241 (M + H) . Anal. (C
5
H
3
ClN
2
O
3
S
2
) C, H, N.
6
(DMSO-d , 300 MHz): δ 7.09 (s, 1H), 7.34 (br s, 1H). EI-MS:
+
3
,6-Dich lor o-4H-th ien o[3,2-e]-1,2,4-th ia d ia zin e 1,1-Di-
m/z 240/242 (M ).
oxid e (10). Meth od A. A suspension of 9 (10.75 g, 0.045 mol)
in phosphorus oxychloride (100 mL) was cooled on an ice bath,
and dry pyridine (7.3 mL, 0.09 mol) was added dropwise at
3-Am in o-N-(ter t-b u t yl)-5-ch lor ot h iop h e n e -2-su lfon -
a m id e (13). A solution of 1236 (38.1 g, 0.15 mol) in dry THF
(300 mL) was cooled to -70 °C, and n-BuLi (1.6 N in hexane,

Potassium Channels of Pancreatic â-Cells
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4181
1
90 mL) was added, maintaining the temperature < -65 °C.
After this was added, the mixture was allowed to warm to -20
C and stirred at this temperature for 30 min. A solution of
washed with water, and dried to give 2.15 g (75%) of pure 18b;
1
mp 132-134 °C. H NMR (DMSO-d
6
, 200 MHz): δ 1.26 (t, 3H),
°
4.18 (q, 2H), 7.60 (s, 1H), 7.86 (br s, 2H), 8.85 (br s, 1H). Anal.
(C ClN ) C, H, N.
-Am in o-6-ch lor o-4H-th ien o[3,2-e]-1,2,4-th iadiazin e 1,1-
p-toluenesulfonyl azide (34 g, 0.17 mol) in dry THF (100 mL)
was added, maintaining the temperature at -20 °C, and the
cooling bath was removed. After the mixture had reached room
temperature, water (100 mL) was added. The organic phase
was isolated, and the aqueous phase was extracted with
toluene (2 × 50 mL). To the combined organic phases was
added hexadecyltributylphosphonium bromide (7.62 g, 15
mmol) followed by the dropwise addition of a solution of sodium
borohydride (6.58 g, 0.174 mol) in water (20 mL) with stirring
and cooling to room temperature. The mixture was stirred
overnight at room temperature, and water (100 mL) was
added. The organic phase was isolated, washed with water (2
7
H
9
2 4 2
O S
3
Dioxid e (27). A suspension of compound 10 (10.0 g, 38.9
mmol) in ammonia (35 mL of 25% solution in water) was
stirred for 16 h at 100 °C in a sealed flask. The mixture was
cooled, concentrated in vacuo, and filtered. The solid was
washed with water, dissolved in 2 N NaOH (30 mL), and
treated with decolorizing charcoal. The filtered solution was
stirred at 50 °C, 4 N HCl was slowly added until the mixture
was slightly acidic, and the resulting suspension was stirred
at 50 °C for 30 min and then at room temperature overnight.
The precipitate was isolated by filtration, washed with water,
and dried in vacuo at 50 °C overnight to give 9.39 g (94%) of
×
100 mL), dried, and evaporated to dryness. The oily residue
was dissolved in EtOAc (150 mL) and washed with 1 N NaOH
SO and
2
7 as a white solid (monohydrate). Anal. (C
5
H
4
ClN
3
O
2
S
2
2
‚1H O)
(6 × 100 mL). The organic phase was dried with Na
2
4
C, H, N. Recrystallization from EtOH followed by drying in
vacuo afforded the anhydrous product; mp >350 °C (dec). H
NMR (DMSO-d , 300 MHz): δ 7.02 (s, 1H), 7.10 (br s, 2H),
1
ClN
evaporated to afford a quantitative yield (40.6 g) of crude 13
1
as an oily substance, which was used without further purifica-
tion in the next step. H NMR (DMSO-d
6
1
, 300 MHz): δ 1.16
+
6
5 4
1.18 (s, 1H). LC-MS m/z 238/240 (M + H) . Anal. (C H -
(
s, 9H), 6.05 (br s, 2H), 6.59 (s, 1H), 7.40 (br s, 1H).
-Am in o-5-ch lor oth iop h en e-2-su lfon a m id e Hyd r och lo-
r id e (14). Compound 13 (40.4 g) was heated with stirring at
3
O
2
S
2
) C, H, N.
3
6
-Ch lor o-3-m eth yla m in o-4H-th ien o[3,2-e]-1,2,4-th ia d i-
a zin e 1,1-Dioxid e (28). A solution of compound 10 (514 mg,
5
0-60 °C in 12.5 N HCl (200 mL) for 2.5 h. The crude product
2
6
5
mmol) and methylamine (0.8 mL of 33% solution in EtOH,
.4 mmol) in absolute EtOH (4 mL) was stirred for 3 days at
5 °C in a sealed flask. The cooled mixture was concentrated
was isolated from the cooled mixture by filtration, dried, and
triturated with ether (60 mL) to give 17.8 g (48%) of the
hydrochloride salt 14; mp >165 °C (dec). H NMR (DMSO-d
3
2
1
6
,
in vacuo, and the residue was treated with water and acidified
to pH <2 with 1 N HCl. Stirring was continued until the
product crystallized, and the solid was isolated by filtration,
washed with water, and dried to give 406 mg (80%) of almost
pure 28. An analytical sample was prepared by crystallization
from AcOH/EtOH with decolorizing charcoal followed by
precipitation from an alkaline solution with 1 N HCl and
00 MHz): δ 6.64 (s, 1H), 6.84 (very br s, 5H). LC-MS m/z
+
13/215 (M + H) . Anal. (C
4
H
5
ClN
2
O
2
S
2
‚HCl 0.5H
2
O) C, H,
N.
Meth yl 5-Ch lor o-2-su lfam oylth ien -3-ylcar bam ate (18a).
3
6
A mixture of carboxylic acid 5 (100.0 g; 0.336 mol) in triethyl
orthoformate (700 mL, 4.2 mol) was refluxed overnight. The
cooled mixture was concentrated at reduced pressure to a
volume of about 150 mL. After it was stirred at 0 °C, the
resulting crystalline precipitate was isolated by filtration and
1
drying in vacuo at 40 °C; mp >325 °C (dec). H NMR (DMSO-
d , 300 MHz): δ 2.75 (d, 3H), 7.02 (s, 1H), 7.23 (br s, 1H), 11.2
6
+
6 6 3 2 2
(br s, 1H). EI-MS m/z 251/253 (M ). Anal. (C H ClN O S ) C,
dried to give 88.8 g (81%) of ethyl 2-(tert-butylamino)sulfonyl-
H, N.
6
1
5
-chloro-3-thiophenecarboxylate (16) as a white solid. H NMR
-Ch lor o-3-et h yla m in o-4H -t h ien o[3,2-e]-1,2,4-t h ia d i-
(
7
6
DMSO-d , 300 MHz): δ 1.20 (s, 9H), 1.30 (t, 3H), 4.32 (q, 2H),
a zin e 1,1-Dioxid e (29). Phosphorus pentoxide (2.84 g, 20
mmol), N,N-dimethylcyclohexylamine (3 mL, 20 mmol), and
ethylamine hydrochloride (1.63 g, 20 mmol) were carefully
mixed and heated with stirring on an oil bath at 180 °C for 20
min. To the homogeneous mass was added the oxo compound
9 (1.2 g, 5 mmol), and the mixture was stirred at 180 °C for 5
h. After it was cooled to about 100 °C, water (150 mL) was
carefully added and the mixture was stirred for 1 h at room
temperature. The resulting mixture was extracted with EtOAc
(3 × 100 mL), and the organic phase was washed with
.47 (br s, 1H), 7.53 (s, 1H).
A solution of crude 16 (10 g, 30 mmol) in NMP (7.5 mL)
was treated with hydrazine monohydrate (3 mL, 61.8 mmol)
at 20-30 °C under external cooling and stirred overnight at
room temperature. Methyl tert-butyl ether (25 mL) and water
30 mL) were added to the mixture with stirring, and the pH
was adjusted to about 2 with 6 N HCl. After it was stirred for
0 min, the organic phase was isolated and evaporated under
reduced pressure to give the crude N-(tert-butyl)-5-chloro-3-
hydrazinocarbonylthiophene-2-sulfonamide (17), which was
used in the next step without further purification.
A solution of crude 17 (10 g, ∼32 mmol) in MeOH (16 mL)
was added dropwise simultaneously with isoamyl nitrite (9
mL, 67 mmol) to a stirred refluxing mixture of MeOH (100
mL) and trifluoroacetic acid (0.4 mL). The mixture was
refluxed for an additional hour and then cooled and concen-
trated at reduced pressure. The resulting residue was hydro-
lyzed by heating at reflux in a mixture of MeOH (9 mL) and
(
3
3
saturated aqueous NaHCO , dried, and evaporated to dryness.
The residue was recrystallized from EtOAc/MeOH to give 282
1
mg (21%) of pure 29; mp 271-274 °C. H NMR (DMSO-d
6
,
2
00 MHz): δ 1.11 (t, 3H), 3.22 (quint, 2H), 7.04 (s, 1H), 7.3
br s, 1H), 11.1 (br s, 1H). EI-MS m/z 265/267 (M ). Anal. (C
+
(
7 8
H -
ClN ) C, H, N.
3
2 2
O S
6-Ch lor o-3-p r op yla m in o-4H-th ien o[3,2-e]-1,2,4-th ia d i-
a zin e 1,1-Dioxid e (30). A solution of 10 (15 g, 58.3 mmol)
and propylamine (40 mL, 0.49 mol) in propan-2-ol (50 mL) was
stirred for 20 h at 85 °C in a sealed flask. The cooled mixture
was concentrated in vacuo, the residue was treated with water
(150 mL) and acidified with 2 N HCl, and stirring was
continued until the product crystallized. The solid was isolated
by filtration, washed with water, and then dissolved in 1 N
NaOH (400 mL) by gentle heating. After treatment with
decolorizing charcoal and filtration, the filtrate was cooled and
acidified with 6 N HCl. The resulting precipitate was isolated
by filtration, washed with water, and recrystallized from EtOH
1
2.5 N HCl (9 mL) for about 6 h during which about 3 mL of
solvent was distilled off. After a further amount of MeOH was
added (6 mL), the hot mixture was treated with decolorizing
charcoal, filtered, and cooled. The resulting precipitate was
isolated by filtration, washed with a small amount of aqueous
EtOH, and recrystallized from MeOH to give 1.5 g (18%) of
the desired methyl carbamate 18a as pale yellow crystals; mp
1
1
7
78-180 °C. H NMR (DMSO-d
6
): δ: 3.73 (s, 3H), 7.58 (s, 1H),
ClN ) C, H, N.
Eth yl 5-Ch lor o-2-su lfa m oylth ien -3-ylca r ba m a te (18b).
.85 (s, 2H), 8.89 (s, 1H). Anal. (C
6
H
6
2 4 2
O S
and dried to give 10.72 g (66%) of pure 30 as a white solid;
1
mp 259-260 °C. H NMR (DMSO-d
6
, 300 MHz): δ 0.89 (t, 3H),
Solid carbonyl azide 8 (2.67 g; 10 mmol) was added in portions
to absolute EtOH (50 mL) with stirring and gentle reflux on
an oil bath. The resulting solution was refluxed for 4 h, cooled
to room temperature, and concentrated under reduced pres-
sure. The residue was triturated with water (100 mL), and
the resulting white precipitate was isolated by filtration,
1
.52 (sext, 2H), 3.15 (q, 2H), 7.05 (s, 1H), 7.29 (br s, 1H), 10.95
br s, 1H). EI-MS m/z 279/281 (M ). Anal. (C
+
(
8 3 2 2
H10ClN O S ) C,
H, N.
3-Bu t yla m in o-6-ch lor o-4H -t h ien o[3,2-e]-1,2,4-t h ia d i-
a zin e 1,1-Dioxid e (31). Potassium tert-butoxide (0.494 g, 4.4

4
182 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Nielsen et al.
mmol) was added to a solution of compound 14 (0.5 g, 2.0
mmol) in dry dimethyl formamide (DMF, 5 mL) with stirring
on an ice bath. The resulting mixture was stirred at 0 °C for
Rou te B. A solution of 10 (3.0 g, 11.67 mmol) in isopropyl-
amine (20 mL) was stirred in a Teflon-lined stainless steel
pressure bomb at 100 °C for 30 h. Excess amine was evapo-
rated under reduced pressure, and the residue was stirred in
water (50 mL) followed by the addition of 4 N HCl to pH 2.
The mixture was stirred on an ice bath for 1 h and the
precipitate was filtered off, washed with water, and recrystal-
lized from EtOAc/MeOH to give 2.42 g (74%) of pure 34 as a
white crystalline product.
5
min before the dropwise addition of butyl isothiocyanate (0.29
mL, 2.4 mmol). The mixture was stirred at 0 °C for an
additional 0.5 h, the cooling bath was removed, and stirring
was continued at room temperature for 30 min. The mixture
was concentrated in vacuo at <50 °C, and the residue was
taken up in water (25 mL), treated with decolorizing charcoal,
and filtered. The filtrate was acidified with AcOH to pH 3-4
and stirred at 0 °C until the product solidified. The solid was
isolated by filtration to give 0.26 g (39%) of N-(3-amino-5-
Rou te D. The title compound 34 was prepared via inter-
mediate 15f (R ) CH(CH ) as described for compound 31
3 2
using 14 (0.5 g, 2 mmol) and isopropyl isothiocyanate (0.26
chlorothien-2-yl)sulfonyl-N′-butylthiourea 15a (R ) (CH
2
)
3
-
mL, 2.4 mmol); yield 0.41 g (74%).
1
CH
3
); mp 115-117 °C (dec). H NMR (DMSO-d
6
, 200 MHz):
Rou te E. K CO (0.67 g, 4.9 mmol) was added to a solution
2
3
δ 0.85 (t, 3H), 1.22 (sext, 2H), 1.48 (quint, 2H), 3.47 (q, 2H),
of the ethyl carbamate 18b (1.14 g, 4.0 mmol) in dry acetone
(10 mL) with stirring at room temperature. Then, isopropyl
isothiocyanate (0.51 mL, 4.8 mmol) was added dropwise and
the resulting mixture was stirred at gentle reflux for 5 h. The
cooled mixture was filtered, and the filtrate was evaporated
to dryness under reduced pressure. The residue was dissolved
in water (25 mL), acidified with 4 N HCl to pH 2, and stirred
at 0 °C until the product solidified. The solid was isolated by
filtration, washed with water, and dried in vacuo to give 1.35
6
.48 (br s, 2H), 6.64 (s, 1H), 8.28 (br t 1H), 11.25 (br s, 1H).
Phosgene (0.44 mL of a 20% solution in toluene, 0.85 mmol)
was added dropwise to a solution of 15a (252 mg, 0.77 mmol)
and dry triethylamine (0.214 mL, 1.54 mmol) in dry THF (4
mL) with stirring at 0 °C. The mixture was stirred for 75 min
at 0 °C and concentrated in vacuo. The residue was triturated
with water, and the precipitate was isolated by filtration,
washed with water, and crystallized from EtOAc to give 162
1
g (87%) of N-{5-chloro-3-[(ethoxycarbonyl)amino]thien-2-yl}-
mg (71%) of 31; mp 218-219 °C. H NMR (DMSO-d
6
, 200
1
sulfonyl-N′-isopropylthiourea 20 (R ) CH(CH
DMSO-d
3H), 7.67 (s, 1H), 8.49 (br d, 1H), 9.0 (s, 1H).
3
)
2
). H NMR
MHz): δ 0.90 (t, 3H), 1.35 (sext, 2H), 1.50 (quint, 2H), 3.18
q, 2H), 7.02 (s, 1H), 7.25 (br s, 1H), 10.95 (br s, 1H). EI-MS
(
6
, 200 MHz): δ 1.12 (d, 6H), 1.25 (t, 3H), 4.2 (q + m,
(
+
m/z 293/295 (M ). Anal. (C
9 3 2 2
H12ClN O S ) C, H, N.
Phosgene (0.57 mL of a 20% solution in toluene, 1.1 mmol)
was added dropwise to a solution of 20 (386 mg, 1 mmol) and
dry triethylamine (0.28 mL, 2 mmol) in dry THF (5 mL) with
stirring at 0 °C. The mixture was stirred at this temperature
for 40 min and concentrated in vacuo. The residue was
triturated with water (5 mL), and the resulting sticky pre-
cipitate was isolated by decantation, washed with water, and
6
-Ch lor o-3-h exyla m in o-4H -t h ien o[3,2-e]-1,2,4-t h ia d i-
a zin e 1,1-Dioxid e (32). Potassium tert-butoxide (0.99 g, 8.8
mmol) was added to a solution of compound 14 (1.0 g, 4.0
mmol) in dry DMF (10 mL) with stirring on an ice bath. The
resulting mixture was stirred at 0 °C for 10 min before the
dropwise addition of hexyl isothiocyanate (0.69 mL, 4.8 mmol).
The mixture was stirred at 0 °C for an additional 1 h, the
cooling bath was removed, and stirring was continued at room
temperature for 30 min. The mixture was concentrated in
vacuo, and the residue was taken up in water (50 mL), treated
with decolorizing charcoal, and filtered. The filtrate was
acidified with AcOH to pH 6 and stirred at 0 °C until the
product solidified. The solid was isolated by filtration and dried
in vacuo at 40 °C to give 1.02 g (72%) of N-(3-amino-5-
crystallized from EtOH/Et
10%) of pure ethyl 6-chloro-3-isopropylamino-4H-thieno[3,2-
e]-1,2,4-thiadiazine-4-carboxylate 1,1-dioxide 24 (R ) CH-
CH , 200 MHz): δ
.20 (d, 6H), 1.30 (t, 3H), 3.98 (m, 1H), 4.33 (q, 2H), 7.57 (s,
H), 8.90 (br d, 1H). An additional crop of 24 (275 mg, 78%)
2
O/petroleum ether to give 35 mg
(
(
1
1
3
)
2
); mp 130-131 °C. 1H NMR (DMSO-d
6
was obtained as a sticky solid by concentration of the organic
mother liquor and was used without further purification.
Crude 24 (275 mg, 0.78 mmol) was refluxed in 2 N NaOH
2 5
chlorothien-2-yl)sulfonyl-N′-hexylthiourea 15b (R ) (CH ) -
1
CH
3 6
). H NMR (DMSO-d , 200 MHz): δ 0.86 (distorted t, 3H),
(2 mL, 4 mmol) for 17 h. The mixture was cooled to room
1
(
.22 (m, 6H), 1.49 (m, 2H), 3.45 (q, 2H), 6.50 (br s, 1H), 6.65
temperature, and 4 N HCl was added dropwise to pH 2 (gas
evolution). The precipitate was isolated by filtration, washed
with water, and crystallized from EtOAc/MeOH to give 116
mg (53%) of pure 34.
s, 1H), 8.29 (br t, 1H) 11.3 (br s, 1H).
Phosgene (1.5 mL of a 20% solution in toluene, 3.09 mmol)
was added dropwise to a solution of 15b (1.0 g, 2.81 mmol)
and dry triethylamine (0.78 mL, 5.6 mmol) in dry THF (8 mL)
with stirring at 0 °C. The mixture was stirred for 1 h at 0 °C
and concentrated in vacuo. The residue was triturated with
water (50 mL), and the resulting sticky precipitate was isolated
by decantation, washed with water, and crystallized from
(
RS)-3-sec-Bu tyla m in o-6-ch lor o-4H-th ien o[3,2-e]-1,2,4-
th ia d ia zin e 1,1-Dioxid e (35). A mixture of 10 (0.6 g, 2.33
mmol), sec-butylamine (0.25 mL, 2.43 mmol), and K CO (0.5
g, 3.62 mmol) in propan-2-ol (2 mL) was stirred at 100 °C for
6 h in a sealed flask. The cooled mixture was concentrated
2
3
1
1
EtOAc to give 302 mg (33%) of 32; mp 207-209 °C. H NMR
in vacuo, the residue was treated with water (20 mL) and
acidified with 1 N HCl, and stirring was continued until the
product crystallized. The solid was isolated by filtration,
washed with water, and then dissolved in 1 N NaOH (50 mL)
by gentle heating. After treatment with decolorizing charcoal
and filtration, the filtrate was cooled and acidified with 6 N
HCl. The resulting precipitate was isolated by filtration,
(
1
(
6
DMSO-d , 200 MHz): δ 0.87 (distorted t, 3H), 1.28 (m, 6H),
.51 (m, 2H), 3.17 (q, 2H), 7.05 (s, 1H), 7.30 (br s, 1H), 11.01
+
s, 1H). EI-MS m/z 321/323 (M ). Anal. (C11
H, N.
-Ch lor o-3-oct yla m in o-4H -t h ien o[3,2-e]-1,2,4-t h ia d i-
a zin e 1,1-Dioxid e (33). The title compound was prepared via
intermediate 15c (R ) (CH CH ) as described for compound
1 using 14 (1.0 g, 4.0 mmol) and octyl isothiocyanate (0.94
3 2 2
H16ClN O S ) C,
6
2
)
7
3
washed with water, and dried to give 0.42 g (61%) of 35 as a
3
1
white solid; mp 224-226 °C (EtOH). H NMR (DMSO-d
6
, 200
1
mL, 4.8 mmol); yield 0.55 g (39%); mp 202-205 °C. H NMR
DMSO-d , 200 MHz): δ 0.85 (t, 3H), 1.28 (m, 10H), 1.50 (m,
H), 3.18 (q, 2H), 7.03 (s, 1H), 7.25 (br s, 1H), 10.2 (br s, 1H).
MHz): δ 0.88 (t, 3H), 1.12 (d, 3H), 1.48 (m, 2H), 3.69 (m, 1H),
.07 (s, 1H), 7.14 (br s, 1H), 10.70 (br s, 1H). EI-MS m/z 293/
(
2
6
7
2
+
95 (M ). Anal. (C
9 3 2 2
H12ClN O S ) C, H, N.
+
EI-MS m/z 349/351 (M ). Anal. (C13
3 2 2
H20ClN O S ) C, H, N.
(
R)-3-sec-Bu t yla m in o-6-ch lor o-4H -t h ien o[3,2-e]-1,2,4-
6
-Ch lor o-3-isop r op yla m in o-4H-th ien o[3,2-e]-1,2,4-th ia -
th ia d ia zin e 1,1-Dioxid e (36). The title compound was pre-
pared as described for compound 35 using 10 (10 g, 38.9 mmol)
and (R)-(-)-sec-butylamine (5 mL, 48.8 mmol) by heating at
d ia zin e 1,1-Dioxid e (34). Rou te A. The title compound 34
was prepared as described for compound 29 using the oxo
compound 9 (1.2 g, 5 mmol) and isopropylamine hydrochloride
85 °C overnight; yield 7.22 g (63%); mp 224-225 °C (EtOH).
1
(
1.91 g, 20 mmol); yield 150 mg (11%); mp 281-283 °C. 1
H
6
H NMR (DMSO-d , 200 MHz): δ 0.88 (t, 3H), 1.11 (d, 3H),
NMR (DMSO-d
6
, 200 MHz): δ 1.15 (d, 6H), 3.88 (m, 1H), 7.08
1.48 (m, 2H), 3.69 (m, 1H), 7.07 (s, 1H), 7.14 (br s, 1H), 10.75
(br s, 1H). EI-MS m/z 293/295 (M ). Anal. (C H12ClN O S ) C,
9 3 2 2
+
(
(
s, 1H), 7.20 (br d, 1H), 10.74 (br d, 1H). EI-MS m/z 279/281
+
M ). Anal. (C
8
H
10ClN
3
O
2
S
2
) C, H, N.
H, N.

Potassium Channels of Pancreatic â-Cells
S)-3-sec-Bu t yla m in o-6-ch lor o-4H -t h ien o[3,2-e]-1,2,4-
th ia d ia zin e 1,1-Dioxid e (37). The title compound was pre-
pared as described for compound 35 using 10 (10 g, 38.9 mmol)
and (S)-(+)-sec-butylamine (4.8 mL, 46.8 mmol) by heating at
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4183
(
1.75 (m, 1H), 3.65 (m, 1H), 7.11 (br s, 2H), 10.68 (s, 1H). EI-
+
MS m/z 307/309 (M ). Anal. (C10
H
14ClN
3
O
2
S
2
) C, H, N.
Isola tion of Op tica lly Active Com p ou n d s. The enanti-
omers of 43 (44 and 45) were separated by high-performance
liquid chromatography (HPLC) using Chiralpak AS with a
mobile phase consisting of heptane/EtOH (50:50) at a flow rate
of 0.7 mL/min at ambient temperature. A UV detector set to
225 and 254 nm was used, and the first enantiomer to elute
was 44. Chiral purity was determined to be 100% ee for 44
and 99.2% ee for 45. The absolute configuration of the
enantiomers was determined by chiral HPLC analysis of 43
as compared with an authentic sample of 45 prepared from
1
20 °C overnight; yield 5.37 g (47%); mp 223-224 °C (EtOH).
1
H NMR (DMSO-d
1
6
, 300 MHz): δ 0.88 (t, 3H), 1.12 (d, 3H),
.48 (m, 2H), 3.69 (m, 1H), 7.07 (s, 1H), 7.14 (br s, 1H), 10.75
+
(
br s, 1H). EI-MS m/z 293/295 (M ). Anal. (C
H, N.
-ter t-Bu tyla m in o-6-ch lor o-4H-th ien o[3,2-e]-1,2,4-th ia -
9 3 2 2
H12ClN O S ) C,
3
d ia zin e 1,1-Dioxid e (38). The title compound was prepared
as described for compound 28 using 10 (5.0 g, 19.4 mmol) and
tert-butylamine (20 mL, 0.19 mol) except that the mixture was
stirred for 20 h at 125 °C; yield 2.91 g (52%); mp >364 °C (dec)
(R)-1,2-dimethylpropylamine.
(S)-6-Ch lor o-3-(1,2-d im eth ylp r op yl)a m in o-4H-th ien o-
1
[3,2-e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (44). mp 212-214 °C.
(
(
(
MeOH). H NMR (DMSO-d
6
, 300 MHz): δ 1.37 (s, 9H), 6.79
1
br s, 1H), 7.11 (s, 1H), 10.55 (br s, 1H). EI-MS m/z 293/295
M ). Anal. (C
H NMR (DMSO-d
6
, 400 MHz): δ 0.90 (d, 6H), 1.08 (d, 3H),
+
1.75 (m, 1H), 3.65 (m, 1H), 7.11 (br s, 2H), 10.68 (s, 1H). Anal.
9 3 2 2
H12ClN O S ) C, H, N.
(
C
10
H
14ClN
3
O
2
S
2
) C, H, N.
R)-6-Ch lor o-3-(1,2-d im eth ylp r op yl)a m in o-4H-th ien o-
3,2-e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (45). mp 218-219 °C.
6
-Ch lor o-3-(1,1-d im eth ylp r op yla m in o)-4H-th ien o[3,2-
(
e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (39). The title compound
was prepared as described for compound 28 using 10 (5.0 g,
[
1
H NMR (DMSO-d
.75 (m, 1H), 3.65 (m, 1H), 7.11 (br s, 2H), 10.68 (s, 1H). Anal.
6
, 400 MHz): δ 0.90 (d, 6H), 1.08 (d, 3H),
1
9.4 mmol) and 1,1-dimethylpropylamine (10 mL, 85.7 mmol)
1
except that the mixture was stirred for 30 h at 125 °C; yield
1
(C10
H
14ClN
3
O
2
S
2
) C, H, N.
3
d
(
.38 g (56%); mp 359-360 °C (dec) (MeOH). H NMR (DMSO-
, 300 MHz): δ 0.82 (t, 3H), 1.31 (s, 6H), 1.73 (q, 2H), 6.67
6
-Ch lor o-3-(2-eth ylbu tyla m in o)-4H-th ien o[3,2-e]-1,2,4-
6
th ia d ia zin e 1,1-Dioxid e (46). The title compound was pre-
pared as described for compound 28 using 10 (514 mg, 2.0
mmol) and 2-ethylbutylamine (0.41 mL, 4.0 mmol) except that
br s, 1H), 7.12 (s, 1H), 10.57 (br s, 1H). EI-MS m/z 307/309
+
(M ). Anal. (C10
3 2 2
H14ClN O S ) C, H, N.
(
RS)-6-Ch lor o-3-(1-m eth ylbu tyl)a m in o-4H-th ien o[3,2-
the mixture was stirred at 120 °C; yield 0.24 g (37%); mp 258-
e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (40). The title compound
was prepared as described for compound 35 using 10 (3 g, 11.7
mmol) and 1-methylbutylamine (2.75 mL, 23.3 mmol) by
1
2
(
1
59 °C. H NMR (DMSO-d
6
, 300 MHz): δ 0.88 (t, 6H), 1.29
quint, 4H), 1.46 (sept, 1H), 3.11 (t, 2H), 7.08 (s, 1H), 7.17 (br,
+
H), 10.86 (br s, 1H). EI-MS m/z 321/323 (M ). Anal. (C11
ClN ) C, H, N.
RS)-6-Ch lor o-3-(2-m eth ylbu tyl)a m in o-4H-th ien o[3,2-
16
H -
heating at 120 °C overnight; yield 2.22 g (62%); mp 195-196
3
2 2
O S
1
°
(
C (EtOH). H NMR (DMSO-d
6
, 200 MHz): δ 0.90 (t, 3H), 1.13
(
d, 3H), 1.20-1.54 (m, 4H), 3.79 (m, 1H), 7.07 (s, 1H), 7.12 (br
+
e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (47). The title compound
was prepared as described for compound 28 using 10 (9.0 g,
s, 1H), 10.74 (br s, 1H). EI-MS m/z 307/309 (M ). Anal. (C10
ClN ) C, H, N.
RS)-6-Ch lor o-3-(1-m eth ylh exyl)a m in o-4H-th ien o[3,2-
14
H -
3
2 2
O S
3
5 mmol) and 2-methylbutylamine (8.5 mL, 72 mmol), except
that the mixture was stirred at 120 °C overnight; yield 7.15 g
66%); mp 240-241 °C (EtOH). 1H NMR (DMSO-d
, 300
MHz): δ 0.86 (d, 3H), 0.87 (t, 3H), 1.13 (m, 1H), 1.39 (m, 1H),
.62 (m, 1H), 3.08 (m, 2H), 7.08 (s, 1H), 7.25 (br, 1H), 10.90
(
e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (41). The title compound
was prepared as described for compound 31 using 14 (0.5 g,
(
6
2
.0 mmol) and 2-isothiocyanatoheptane (0.38 g, 2.4 mmol),
except that the intermediate N-(3-amino-5-chlorothien-2-yl)-
sulfonyl-N′-(1-methylhexyl)thiourea 15g (R (RS)-CH-
) was isolated by extraction with CH Cl
(3 ×
5 mL) before the cyclization to 41; yield 180 mg (27%); mp
1
+
3 2 2
(br s, 1H). EI-MS m/z 307/309 (M ). Anal. (C10H14ClN O S )
)
C, H, N.
(CH
3
)(CH
2
)
4
CH
3
2
2
(
R S )-6-Ch lor o-3-(1,2,2-t r im e t h ylp r op yl)a m in o-4H -
1
1
th ien o[3,2-e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (48). The title
compound was prepared as described for compound 28 using
1
79-181 °C. H NMR (DMSO-d , 200 MHz): δ 0.86 (distorted
6
t, 3H), 1.13 (d, 3H), 1.18-1.55 (m, 8H), 3.75 (m, 1H), 7.06 (s,
1
0 (200 mg, 0.78 mmol) and 1,2,2-trimethylpropylamine (0.50
+
1
H), 7.13 (br s, 1H), 10.74 (s, 1H). EI-MS m/z 335/337 (M ).
Anal. (C12 ) C, H, N.
RS)-6-Ch lor o-3-(1,5-d im eth ylh exyl)a m in o-4H-th ien o-
3,2-e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (42). The title com-
pound was prepared as described for compound 31 using 14
1.0 g, 4.0 mmol) and 2-isothiocyanato-6-methylheptane (0.75
mL, 3.78 mmol), except that the mixture was stirred at 110
H
18ClN
3
O
2
S
2
1
°
C for 17 h; yield 137 mg (55%); mp 294-295 °C (EtOAc). H
(
NMR (DMSO-d , 300 MHz): δ 0.90 (s, 9H), 1.06(d, 3H), 3.68
(m, 1H), 6.90 (br s, 1H), 7.12 (s, 1H), 10.65 (s, 1H). EI-MS m/z
6
[
+
321/323 (M ). Anal. (C H ClN O S ) C, H, N.
1
1
16
3
2
2
(
6
-Ch lor o-3-cyclop r op yla m in o-4H -t h ien o[3,2-e]-1,2,4-
g, 4.3 mmol), except that the intermediate N-(3-amino-5-
chlorothien-2-yl)sulfonyl-N′-(1,5-dimethylhexyl)thiourea 15i (R
th ia d ia zin e 1,1-Dioxid e (49). The title compound was pre-
pared as described for compound 35 by heating a mixture of
)
(RS)-CH(CH
extraction with Et
chromatography on silica gel (CH
final cyclization step to 42; yield 133 mg (10%); mp 197-199
3
)(CH
O (3 × 30 mL) and purified by flash
Cl /MeOH 20:1) before the
2 3 3 2
) CH(CH ) ) was isolated as an oil by
1
0 (3.0 g, 11.7 mmol), cyclopropylamine (1.62 mL, 23.3 mmol),
2
and K CO (3.2 g, 23.3 mmol) in propan-2-ol (20 mL) for 18 h
2
3
2
2
at 80 °C in a sealed flask to give 2.17 g (67%) of 49 as a beige
H NMR (DMSO-d , 300
1
solid; mp 272-273 °C (EtOAc/MeOH).
6
1
°
C (EtOAc). H NMR (DMSO-d
6
, 300 MHz): δ 0.85 (d, 6H),
MHz): δ 0.52-0.61 (m, 2H), 0.75-0.82 (m, 2H), 2.54-2.65 (m,
1H), 7.11 (s, 1H), 7.89 (br s, 1H), 10.95 (br s, 1H). EI-MS m/z
277/279 (M ). Anal. (C H ClN O S ) C, H, N.
1
.12 (d, 3H), 1.1-1.6 (m, 7H), 3.77 (m, 1H), 7.07 (s, 1H), 7.12
+
+
(
br s, 1H), 10.72 (br s, 1H). LC-MS m/z 350/352 (M ). Anal.
) C, H, N.
RS)-6-Ch lor o-3-(1,2-d im eth ylp r op yl)a m in o-4H-th ien o-
8
8
3
2
2
13 3 2 2
(C H20ClN O S
6
-Ch lor o-3-cyclobu tyla m in o-4H-th ien o[3,2-e]-1,2,4-th i-
(
a d ia zin e 1,1-Dioxid e (50). The title compound was prepared
as described for compound 28 using 10 (3.0 g, 11.7 mmol) and
cyclobutylamine (5 g, 70 mmol) except that the mixture was
[
3,2-e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (43). A solution of 10
(0.5 g, 1.94 mmol) in 1,2-dimethylpropylamine (5 mL) was
stirred for 48 h at 100 °C in a sealed flask. The cooled mixture
was concentrated under reduced pressure, and the residue was
stirred with water (20 mL) followed by adjustment to pH 2
with 4 N HCl. The initially formed sticky mass crystallized
by stirring the mixture for about 2 h at 0 °C. The precipitate
was isolated by filtration, washed with water, and recrystal-
lized from EtOAc/MeOH followed by drying in vacuo at 60 °C
stirred for 20 h at 100 °C; yield 2.79 g (82%); mp 313-314 °C
1
(dec) (EtOH/water). H NMR (DMSO-d
6
, 300 MHz): δ 1.58-
1.75 (m, 2H), 1.89-2.05 (m, 2H), 2.19-2.30 (m, 2H), 4.16 (m,
1H), 7.06 (s, 1H), 7.62 (br s, 1H), 10.83 (br s, 1H). EI-MS m/z
+
291/293 (M ). Anal. (C
9
H
10ClN
3
O
2
S
2
) C, H, N.
6-Ch lor o-3-cyclop en t yla m in o-4H -t h ien o[3,2-e]-1,2,4-
th ia d ia zin e 1,1-Dioxid e (51). The title compound was pre-
pared as described for compound 31 using 14 (2.0 g, 8.0 mmol)
and cyclopentyl isothiocyanate (1.22 g, 9.6 mmol); yield 1.35
overnight to give 0.43 g (72%) of pure 43; mp 216.5-218 °C.
1
H NMR (DMSO-d , 200 MHz): δ 0.90 (d, 6H), 1.08 (d, 3H),
6

4
184 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Nielsen et al.
g (57%); mp 299-300 °C (EtOH). 1H NMR (DMSO-d
, 200
MHz): δ 1.40-1.70 (m, 6H), 1.90 (m, 2H), 3.95 (sext, 1H), 7.05
filtered, the solution was acidified to pH <2 with 4 N HCl and
the precipitate was filtered off and purified by chromatography
(dichloromethane/MeOH (19:1)). Recrystallization from EtOH
6
+
(
s, 1H), 7.3 (br, 1H), 10.70 (br s, 1H). EI-MS m/z 305/307 (M ).
Anal. (C10 ) C, H, N.
-Ch lor o-3-cycloh exyla m in o-4H-th ien o[3,2-e]-1,2,4-th i-
1
H
12ClN
3
O
2
S
2
afforded 55 mg (8%) of pure 56; mp 218-219 °C. H NMR
6
6
(DMSO-d , 300 MHz): δ 1.18-1.54 (m, 11H), 1.97-2.12 (m,
a d ia zin e 1,1-Dioxid e (52). The title compound was prepared
as described for compound 31 using 14 (1.0 g, 4.0 mmol) and
2H), 6.55 (br s, 1H), 7.12 (s, 1H), 10.60 (br s, 1H). LC-MS m/z
+
3 2 2
334/336 (M + 1) . Anal. (C12H16ClN O S ) C, H, N.
cyclohexyl isothiocyanate (0.656 mL, 4.8 mmol); yield 0.65 g
6-Ch lor o-3-(1-eth ylcyclobu tyl)a m in o-4H-th ien o[3,2-e]-
1,2,4-th ia d ia zin e 1,1-Dioxid e (57). A mixture of 10 (1.02 g,
3.95 mmol), freshly activated KF (688 mg, 11.9 mmol), and
hexadecyltrimethylammonium bromide (43 mg, 0.12 mmol) in
dry NMP (4 mL) was stirred for 20 h at 120 °C under nitrogen
to form the fluoro compound 11. The mixture was allowed to
cool to room temperature and then reacted directly with
1-ethylcyclobutylamine hydrochloride (0.8 g, 5.93 mmol) (pre-
pared from 1-ethylcyclobutanecarboxylic acid68 analogously
to 1-methylcyclopropylamine hydrochloride described by
1
(
51%); mp 282-284 °C (EtOH/MeOH). H NMR (DMSO-d
6
, 300
MHz): δ 1.1-1.9 (m, 10H), 3.55 (m, 1H), 7.08 (s, 1H), 7.19
+
(
br, 1H), 10.73 (br s, 1H). EI-MS m/z 321/319 (M ). Anal.
) C, H, N.
-Ch lor o-3-(1-m eth ylcyclop r op yl)a m in o-4H-th ien o[3,2-
e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (53). Rou te B. A solution
(
11 3 2 2
C H14ClN O S
6
6
3
of 10 (386 mg, 1.5 mmol) in 1-methylcyclopropylamine (1.0
g, 14 mmol) was stirred for 24 h at 85 °C in a sealed flask.
The cooled solution was concentrated in vacuo, and the residue
was stirred with EtOAc (1-2 mL) and filtered. The white
precipitate was stirred in 4 N HCl (5 mL) for 2 h and then
filtered off and chromatographed on silica gel with ethyl
6
4
Kirmse ) and triethylamine (1.65 mL, 11.9 mmol) for 3 days
at 75 °C in the sealed flask. The cooled mixture was poured
into water, acidified to pH <2 with 1 N HCl, and extracted
acetate to give 112 mg (26%) of pure 53; mp 251-252 °C dec
2 4
with ethyl acetate. The organic phase was dried with Na SO ,
1
(
(
1
MeOH/water). H NMR (DMSO-d
6
, 300 MHz): δ 0.65-0.79
evaporated to dryness, and recrystallized from AcOH to give
1
m, 4H), 1.36 (s, 3H), 7.11 (s, 1H), 7.82 (br s, 1H), 10.78 (br s,
H). EI-MS m/z 291/293 (M ). Anal. (C
214 mg (17%) of pure 57; mp 244-246 °C. H NMR (DMSO-
+
H
10ClN
3
O
2
S
2
) C, H, N.
6
d , 300 MHz): δ 0.79 (t, 3H), 1.70-1.93 (m, 4H), 1.96-2.08
9
(
m, 2H), 2.13-2.25 (m, 2H), 7.09 (s, 1H), 7.24 (br s, 1H), 10.57
Rou te C. A solution of the fluoro compound 11 (145 mg,
+
.6 mmol), 1-methylcyclopropylamine hydrochloride64 (78 mg,
.72 mmol), and triethylamine (0.184 mL, 1.32 mmol) in
3 2 2
(br s, 1H). LC-MS m/z 320/322 (M+1) . Anal. (C11H14ClN O S )
C, H, N.
0
0
ethanol (0.5 mL) was stirred for 21 h at 60 °C in a sealed flask.
The cooled mixture was concentrated in vacuo, and the residue
was triturated with water (3 mL) followed by adjustment to
pH <2 with 4 N HCl. The crude sticky product was dissolved
in 1 N NaOH (4 mL), treated with decolorizing charcoal, and
filtered. The filtrate was acidified with 4 N HCl, and the
resulting precipitate was isolated by filtration, washed with
water, and recrystallized from AcOH to give 95 mg (54%) of
LogP -p K
a
Mea su r em en ts. Potentiometric titration ex-
instrument from Sirius
periments were conducted on a GLpK
a
Analytical Instruments Ltd. The determination of the ioniza-
tion constant of the compound was performed in the following
way: approximately 3 mg of the compound was weighed into
a small glass vial and placed in a temperature-controlled
sample holder. A 10 mL amount of cosolvent (80% MeOH) was
added, and the pH was adjusted to approximately 2. Three
titrations were performed in the range of pH 2 to 10.5 on the
same sample, adding 3.00 mL of 0.15 M KCl before the second
titration and 3.00 mL of 0.15 M KCl before the third. For each
5
3.
6
-Ch lor o-3-(1-m eth ylcyclobu tyl)a m in o-4H-th ien o[3,2-
e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (54). A mixture of the fluoro
compound 11 (1.3 g, 5.3 mmol), 1-methylcyclobutylamine
titration, a p
The pK value at 0% cosolvent was determined by using a
s a
K value was measured by refining of the data.
6
5
a
hydrochloride (1.0 g, 8.1 mmol), and triethylamine (2.5 mL,
8.1 mmol) in EtOH (10 mL) was stirred for 16 h at 50 °C and
s a
Yasuda-Shedlovsky extrapolation procedure on the p K
values determined for the three measured difference curves.
1
then for 5 h at 70 °C in a sealed flask. The cooled mixture
was concentrated in vacuo, and the residue was triturated with
water (25 mL) followed by adjustment to pH <2 with 1 N HCl.
The crude product was isolated by filtration, recrystallized
from acetic acid, and finally purified by chromatography (C18;
The determination of the octanol/water distribution coef-
ficient, logP, of the compound was performed in the following
way: approximately 3 mg of the compound was weighed into
a small glass vial and placed in the temperature-controlled
sample holder. A 10.0 mL amount of 0.15 M KCl and 0.1 mL
of octanol were added. The pH was adjusted to 10.5 and
titrated to pH 2.0. Three titrations were performed on the same
sample, adding 1.00 mL of octanol before the second titration
and 10.00 mL before the third. Prior to data refinement, the
2
5
0-60% acetonitrile + 0.01% TFA) to give 363 mg (22%) of
1
4; mp 294-296 °C. H NMR (DMSO-d , 300 MHz): δ 1.47
6
(
2
3
s, 3H), 1.75-1.88 (m, 2H), 1.94-2.05 (m, 2H), 2.18-2.31 (m,
H), 7.07 (s, 1H), 7.33 (br s, 1H), 10.67 (br s, 1H). LC-MS m/z
+
06/308 (M+1) . Anal. (C10
3 2 2
H12ClN O S ) C, H, N.
measured value for the aqueous pK
the GLpK software. For each titration, the shift in the pK
a
had to be entered into
6
-Ch lor o-3-(1-m eth ylcyclop en tyl)a m in o-4H-th ien o[3,2-
e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (55). A mixture of 11 (0.60
a
a
g, 2.5 mmol), 1-methylcyclopentylamine hydrochloride6
6,67
value was measured. The three refined data sets were com-
bined in a multiset, and the octanol/water distribution coef-
ficient was extracted according to the possible partitioning
equations. The Moriguchi LogP (mLogP) was calculated using
commercial available software.44
(0.5
g, 3.7 mmol), and triethylamine (1.03 mL, 7.4 mmol) in EtOH
2.5 mL) was stirred for 16 h at 50 °C and then for 24 h at 65
C in a sealed flask. The cooled mixture was concentrated in
(
°
vacuo, and the residue was triturated with water followed by
adjustment to pH <2 with 1 N HCl. The crude product was
isolated by filtration, dried, and recrystallized from AcOH to
Effects on Mem br a n e P oten tia l in âTC3 Cells. Mea-
4
surement of DiBAC (3) fluorescence was made using the
1
30
give 208 mg (26%) of 55; mp >300 °C (dec). H NMR (DMSO-
procedure described by Lebrun et al. For each compound and
concentration (0.01, 0.1, 1, 10, and 100 µMol/L), a total of four
individual wells was used to construct the data sets underlying
the values in Tables 2-4. Briefly, data were normalized to the
initial value for each well and expressed as fractional fluores-
cence change during the 45 min incubation period using Excel
(Microsoft, U.S.A.). A four parameter nonlinear curve fit was
then made to extract EC50 values and SEM using MacCurveFit
d , 300 MHz): δ 1.43 (s, 1H), 1.53-1.72 (m, 6H), 1.92-2.10
6
(
m, 2H), 6.91 (br s, 1H), 7.10 (s, 1H), 10.52 (br s, 1H). LC-MS
+
m/z 320/322 (M + 1) . Anal. (C11
3 2 2
H14ClN O S ) C, H, N.
6
-Ch lor o-3-(1-m eth ylcycloh exyl)a m in o-4H-th ien o[3,2-
e]-1,2,4-th ia d ia zin e 1,1-Dioxid e (56). A mixture of 11 (0.5
g, 2.08 mmol), 1-methylcyclohexylamine hydrochloride6
6,67
(373
mg, 2.49 mmol), and triethylamine (0.58 mL, 4.16 mmol) in
EtOH (3 mL) was stirred for 20 h at 50 °C and then for 22 h
at 100 °C in a sealed flask. The cooled mixture was concen-
trated in vacuo, and the residue was triturated with water
followed by adjustment to pH <2 with 4 N HCl. The crude
product was isolated by filtration and dissolved in 1 N NaOH
followed by treatment with decolorizing charcoal. After it was
1
.3 (Kevin Raner Software, Australia).
In h ibition of In su lin Relea se In Vitr o. (a ) In h ibition
of Glu cose-In d u ced In su lin Relea se in âTC6 Cells. âTC6
cells69 were cultured at 5 × 10 cells/microtiter well in DMEM
+ 10% fetal calf serum (FCS), 11 mM glucose, 1% Glutamax,
4
2
and 20 mM Hepes for 3 days (95% humidity, 37 °C, 5% CO ).

Potassium Channels of Pancreatic â-Cells
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4185
Cells were washed twice with NN buffer (all in mM: NaCl,
48 000g for 10 min. The pellet was homogenized again in buffer
and centrifuged at 48 000g for 10 min. The pellet was
suspended in buffer with 250 mM sucrose by homogenization.
Total protein was determined using a BioRad kit. The receptor
preparation was kept in small aliquots (5 mg protein/mL) at
-80 °C until use.
Receptor preparation (100 µg/mL): [ H]glibenclamide (1 nM,
New England Nuclear, NET1024), test compound, and assay
2 2
buffer (30 mM HEPES, 2 mM MgCl , 2 mM Na -ATP, pH 7.4)
1
2
14; KCl, 47; KH
2 4 4 3
PO , 1.21; MgSO , 1.16; NaHCO , 25.5; CaCl,
H
2
O, 2.5; HEPES, 10) supplemented with 0.1% bovine serum
albumin (BSA) and incubated for 60 min in this buffer. All
wells were aspirated, and the cells were incubated for 3 h with
NN buffer, 22 mM glucose, 0.1 mM IBMX, and serial dilutions
of the compounds. A reference compound served as positive
control. A test for responsiveness toward a series of glucose
concentrations was included in every assay to ensure func-
tionality. The supernatant from each well was harvested, and
insulin content was measured by an in-house enzyme-linked
immunosorbent assay (ELISA) using guinea pig antiinsulin
antibodies and a rat insulin as standard. The results were
analyzed in Prism (Graphpad Software) and expressed as IC50
and Emax (the maximum inhibition obtained at 10 µM of the
compound). SEM was calculated for all compounds.
3
to a final volume of 250 µL was mixed. All test compounds
were dissolved in DMSO and diluted in assay buffer. Glipizide
(10 µM) was used to determine nonspecific binding. The assay
was incubated for 1 h at 37 °C. Incubation was terminated by
dilution with 4 mL of ice-cold 0.9% NaCl solution, followed by
rapid filtration through Whatman GF/B filters. Filters were
washed with 2 × 4 mL ice-cold 0.9% NaCl solution. The
radioactivity retained on the filters was dissolved with scintil-
lation cocktail and counted using a Packard scintillation
counter. IC50 values were calculated by nonlinear regression
analysis (sigmoid two site or one site) of binding data using
Prism for Windows 3.02, GraphPad Software, San Diego,
California, U.S.A.
(
b) In h ibition of Glu cose-In d u ced In su lin Relea se in
Wista r Ra t Islets. Islets were isolated by Collagenase and
gradient centrifugation in Ficoll gradient (40-13%). Isolated
islets were incubated in bulk overnight in RPMI, 10% FCS,
and 11 mM glucose. The islets were handpicked and placed
at 10 islets/microtiter well and cultured overnight in DMEM,
1
0% FCS, and 3 mM glucose. Essentially, the islets were tested
â-Cell Selectivity Stu d ies a fter In tr a ven ou s Ad m in -
istr a tion to Ra ts. Postprandial, adult male Sprague-Dawley
rats, weighing 280-320 g (M&B Breeding Center, Ll. Skensved,
Denmark), were anaesthetized with a mixture of Hypnorm/
as described for the âTC6 but with no addition of IBMX. The
insulin content was measured in the same ELISA as used for
the âTC6 cells.
7
2
Rela xa tion of P r econ tr a cted Sm ooth Mu scle In Vitr o.
Dormicum receiving supplemental anesthetic every 20 min.
(
a ) Aor ta a n d Bla d d er . Female Wistar rats weighing ap-
Body temperature was maintained at 37 °C by a homeothermic
blanket system (Harvard Homeothermic Blanket System,
Harvard Apparatus Ltd, U.K.). A polythene cannula (PE50)
coupled to a pressure transducer (Statham P23XC) and a
pressure coupler (HSE 561, Hugo Sachs Elektronik Kg,
Germany) was introduced into the carotid artery to measure
blood pressure (MAP), which was recorded continuously and
fed to a plotter (Linearcorder Mark VII, WR 3310, Hugo Sachs
Elektronik Kg, Germany) and a computer and analyzed offline.
A polythene cannula (PE50) was inserted into the jugular vein
for drug infusion and blood sampling. After a stabilization
period, the test substances were injected as a bolus lasting
for 20 s. Blood samples were drawn at 5-10 min intervals,
from 15 min preinjection until 30 min postinjection. Com-
pounds were dissolved in NaOH (0.15%) and saline and were
administered in a fixed dose of 3 mg/kg rat. Plasma insulin
levels were measured with in-house ELISA using guinea pig
antiinsulin antibodies and rat insulin as standard.
P h a r m a cok in etic P r op er ties in Ra t. The pharmacoki-
netic rat experiments were performed on Sprague-Dawley,
body weight about 180 g, rats using the cassette approach with
two compounds per cassette (n ) 2). Blood sampling was done
by heart puncture and each single data point originates from
one rat. The doses administered were selected in order to
obtain linear pharmacokinetics. For intravenous administra-
tion, rats were given 0.5 mg/kg of 34, 39, 50, or diazoxide and
0.89 mg/kg of 30, 36, 38, or 54. For oral administration, 1.89
mg/kg of 30, 36, 37, or 53 and 2.0 mg/kg of 34, 39, 50, or
diazoxide were used. Dosage vehicle for the compounds was
potassium phosphate buffer 0.1 M, pH 8.0, except for 34 for
which vehicle was PEG 300, 20-30% in water and for
diazoxide the vehicle was PEG 300, 40%, and Cremophor, 10%,
in water. The plasma concentrations of dosed compounds were
determined by an LC/MS/MS method. The concentration-time
data sets were analyzed by noncompartmental pharmacoki-
netic methods based on statistical moments theory.
proximately 150-200 g were killed by cervical dislocation, and
the thoracic aorta or bladder body was removed. Aorta and
bladder were cut into rings of approximately 5 mm wide. Ring
preparations were mounted in 5 mL (Danish Myo Technology,
Aarhus, Denmark) or 10 mL organ baths (Schuler Organ Bath
8
09, Hugo Sachs Elektronik, Germany) with a resting tension
of 2 or 0.5 g (for aorta and bladder, respectively) and bathed
in Krebs Ringer solution with the following composition
(
2
mM): NaCl, 118.5; NaHCO
.4; and glucose, 11.1 in double-distilled water. The ringer
solution was continuously aerated with 95% O /5% CO at 37
C. Cumulative concentration response curves (0.1-300 µM)
3 2 2
, 25.0; KCl, 4.7; CaCl , 6.8; MgCl ,
2
2
°
were constructed for all test compounds on top of a precon-
traction induced either by 0.3 µM phenylephrine or 10 µM
bethanechol, in aorta and bladder, respectively. All test
compounds were freshly dissolved in DMSO. Potency of a
compound (EC50), with regard to producing smooth muscle
relaxation, was defined, as the concentration required obtain-
ing a half-maximal dilating effect. EC50 was estimated for
individual concentration response curves by four parameter
nonlinear, logistic regression using Myodata software (Danish
Myo Technology, Aarhus, Denmark). Data generated in blad-
der smooth muscle were corrected for time decay of the
bethanechol-induced contractile response.
(
b) Mesen ter ic Resista n ce Ar ter ies. Adult male Wistar
rats (aged 12-16 weeks) were killed by a sharp blow on the
head and cervical dislocation. The mesentery was removed and
immersed in cold (4 °C) Krebs Ringer solution. Third to fourth
order side branches were dissected from the superior mesen-
teric artery. Arteries were chemically sympathectomized by
incubation in 6-hydroxydopamine. Vessels were then mounted
in a wire myograph at 37 °C, in Krebs Ringer solution (CaCl
2
2
7
0
2
,
,
2 4 4
.5; KCl, 4.7; KH PO , 1.2; MgSO , 1.2; NaCl, 118.5; NaHCO
3
5.0; and glucose, 5.5, all in mM) and continuously aerated
7
1
2 2
with 95% O /5% CO . Arteries were normalized to find their
individual optimal lumen diameter for isometric force develop-
ment. For all test compounds, cumulative concentration
response curves were constructed in vessels precontracted with
In Vivo Effect of Or a lly Ad m in ister ed 34 on P a n cr e-
a tic F u n ction in Glu cose-Ch a llen ged Ra ts. Groups of male
Sprague-Dawley rats, body weight about 360 g, were fasted
overnight and dosed orally with 34 (30 mg/kg, n ) 3/group)
3-12 h prior to experimentation or with vehicle at 3 h before
(n ) 6). Vehicle consisted of 1:1:4:4 (v/v) glycerol/water/0.5%
gelatine solution/10% methylhydroxypropyl-cellulose E3, and
a dose volume of 2 mL/kg was used. Before administration of
a glucose challenge, rats were anaesthetized with Hypnorm
and Dormicum.72 A 200 µL blood sample was taken from the
retro-orbital plexus into chilled, heparinized tubes for deter-
1
0 µM phenylephrine. Data were analyzed as described above.
3
[
H]Gliben cla m id e Bin d in g. HEK293 cells stably ex-
pressing SUR1/Kir6.2 were grown in DMEM with 1% penicil-
lin/streptomycin, 0.6 mg/mL G418, and 10% FCS. Cells were
washed with phosphate-buffered saline (PBS) and harvested
in PBS. The cells were centrifuged at 48 000g for 10 min. The
cell pellet was homogenized in buffer (30 mM Tris, pH 7.4)
using an Ultra Turrax homogenizer and then centrifuged at

4
186 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Nielsen et al.
mination of plasma glucose (EBIO Plus autoanalyzer; Eppen-
dorff, Germany) and plasma insulin by ELISA as above.
Glucose was injected at four different sc sites at a load of 2
g/6 mL/kg, and a second blood sample was collected 15 min
later. Statistical analysis was done using one way analysis of
variance (ANOVA) with Bonferroni’s post hoc test.
(15) Lebrun, P.; Devreux, V.; Hermann, M.; Herchuelz, A. Similarities
between the Effects of Pinacidil and Diazoxide on Ionic and
Secretory Events in Rat Pancreatic Islets. J . Pharmacol. Exp.
Ther. 1989, 250, 1011-1018.
(
16) Seltzer, H. S.; Allen, E. W. Hyperglycemia and Inhibition of
Insulin Secretion during Administration of Diazoxide and
Trichlormethiazide in Man. Diabetes 1969, 18, 19-28.
17) Grill, G. V.; Rauff, O.; MacFarlane, I. A. Diazoxide Treatment
for Insulinoma: a National UK Survey. Postgrad. Med. J . 1997,
7, 640-641.
In Vivo Mea su r em en t of Ca r d iova scu la r P a r a m eter s
in Telem eter ed Sp r a gu e-Da w ley Ra ts. Female 12-18
month Mol:SPRD rats (M&B Ltd., Ll.) weighing 250-350 g
were used. For insertion of the telemetry probe, which was
done at least 2 months before the start of the experiments,
each rat was anaesthetized with Isofluran. The abdomen of
the rat was opened with a 3-5 cm long incision. A catheter
was inserted in the aorta (1.5 cm), and the catheter was sealed
to the aorta by the use of surgical glue. Hereafter, the
telemetry probe (TL11M2-C50-PXT, Data Sciences Interna-
tional, St. Paul, MN) was placed in the abdomen. Between and
during the study, the rats were pair-housed (only one rat with
sensor) in type III cages with aspen bedding. The rats were
fed Altromin 1324 (Brogaarden, Gentofte, Denmark) and water
ad libitum. Lights were on from 6:00 am to 6:00 pm with no
twilight periods, and room air was changed 8-15 times per
hour.
(
(18) Meissner, T.; Brune, W.; Mayatepek, E. Persistent Hyperinsuli-
naemic Hypoglycaemia of Infancy: Therapy, Clinical Outcome
and Mutational Analysis. Eur. J . Pediatr. 1997, 156, 754-757.
(
19) Alemzadeh, R.; Slonim, A. E.; Zdanowicz, M. M.; Maturo, J .
Modification of Insulin Resistance by Diazoxide in Obese Zucker
Rats. Endocrinology 1993, 133, 705-712.
(
20) Aizawa, T.; Taguchi, N.; Sato, Y.; Nakabayashi, T.; Kobuchi, H.;
Hidaka, H.; Nagasawa, T.; Ishihara, F.; Itoh, N.; Hashizume,
K. Prophylaxis of Genetically Determined Diabetes by Diazox-
ide: a Study in a Rat Model of Naturally Occurring Obese
Diabetes. J . Pharmacol. Exp. Ther. 1995, 275, 194-199.
21) Bjork, E.; Berne, C.; Kampe, O.; Wibell, L.; Oskarsson, P.;
Karlsson, F. A. Diazoxide Treatment at Onset Preserves Re-
sidual Insulin Secretion in Adults With Autoimmune Diabetes.
Diabetes 1996, 45, 1427-1430.
(
(
(
22) Uno, H.; Kemnitz, J . W.; Cappas, A.; Adachi, K.; Sakuma, A.;
Kamoda, H. The Effects of Topical Diazoxide on Hair Follicular
Growth and Physiology of the Stumptailed Macaque. J . Derma-
tol. Sci. 1990, 1, 183-194.
23) Buhl, A. E.; Waldon, D. J .; Conrad, S. J .; Mulholland, M. J .;
Shull, K. L.; Kubicek, M. F.; J ohnson, G. A.; Brunden, M. N.;
Stefanski, K. J .; Stehle, R. G. Potassium Channel Conduc-
tance: a Mechanism Affecting Hair Growth Both in Vitro and
in Vivo. J . Invest. Dermatol. 1992, 98, 315-319.
Data acquisition was performed by using PONEMAH and
cardiovascular parameters, and body temperature was fol-
lowed continuously throughout the experiment. Mean values
(every 30 min) were calculated.
Su p p or tin g In for m a tion Ava ila ble: Assay method and
8
6
+
table of Rb efflux data for compounds 27-33 and diazoxide.
Synthetic procedures and table of biological data for additional
compounds 58-67. This material is available free of charge
via the Internet at http://pubs.acs.org.
(
24) Topliss, J . G.; Sherlock, M. H.; Reimann, H.; Konzelman, L. M.;
Shapiro, E. P.; Pettersen, B. W.; Schneider, H.; Sperber, N.
Antihypertensive Agents. I. Non-Diuretic 2H-1, 2, 4-Benzothia-
diazine 1,1-Dioxides. J . Med. Chem. 1963, 6, 122-127.
25) Topliss, J . G.; Konzelman, L. M.; Shapiro, E. P.; Sperber, N.;
Roth, F. E. Antihypertensive Agents. II. 3-Substituted 2H-1,2,4-
Benzothiadiazine 1,1-Dioxides. J . Med. Chem. 1964, 7, 269-273.
26) Parenti, C.; Costantino, L.; Di Bella, M. Heteroarylalkanoic Acids
with Possible Antiinflammatory Activities. Part 8: The Scaveng-
ing of the Oxygen-Free Radicals. Die Pharmazie 1990, 45, 680-
Refer en ces
(
(
(
(
(
(
(
1) Ashcroft, F. M.; Rorsman, P. ATP-Sensitive K+ Channels: a Link
between B-Cell Metabolism and Insulin Secretion. Biochem.
Soc. Trans. 1990, 18, 109-111.
2) Aguilar-Bryan, L.; Bryan, J . Molecular Biology of Adenosine
Triphosphate-Sensitive Potassium Channels. Endocr. Rev. 1999,
2
0, 101-135.
6
81.
3) Seino, S. ATP-Sensitive Potassium Channels: a Model of Het-
eromultimeric Potassium Channel/Receptor Assemblies. Annu.
Rev. Physiol. 1999, 61, 337-362.
(
27) Petersen, H. J . Synthesis of 3-Amino- and 3-Substituted Amino-
2H-1,2,4-Benzothiadiazine 1,1-Dioxides. Acta Chem. Scand.
1973, 27, 2655-2660.
4) Ashcroft, F. M.; Gribble, F. M. ATP-Sensitive K+ Channels and
Insulin Secretion: Their Role in Health and Disease. Diabeto-
logia 1999, 42, 903-919.
(28) de Tullio, P.; Pirotte, B.; Lebrun, P.; Fontaine, J .; Dupont, L.;
Antoine, M. H.; Ouedraogo, R.; Khelili, S.; Maggetto, C.; Ma-
sereel, B.; Diouf, O.; Podona, T.; Delarge, J . 3-and 4-Substituted
4H-Pyrido[4,3-e]-1,2,4-Thiadiazine 1,1-Dioxides as Potassium
Channel Openers: Synthesis, Pharmacological Evaluation, and
Structure-Activity Relationships. J . Med. Chem. 1996, 39, 937-
5) Coghlan, M. J .; Carroll, W. A.; Gopalakrishnan, M. Recent
Developments in the Biology and Medicinal Chemistry of Potas-
sium Channel Modulators: Update from a Decade of Progress.
J . Med. Chem. 2001, 44, 1627-1653.
6) Robertson, D. W.; Steinberg, M. I. Potassium Channel Modula-
tors: Scientific Applications and Therapeutic Promise. J . Med.
Chem. 1990, 33, 1529-1541.
7) Primeau, J .; Butera, J . Potassium Channel Activating Agents;
Emerging Trends. Curr. Pharm. Des. 1995, 1, 391-406.
8) Pirotte, B.; Fontaine, J .; Lebrun, P. Recent Advances in the
Chemistry of Potassium Channel Openers. Curr. Med. Chem.
(
9
48.
(
29) Pirotte, B.; de Tullio, P.; Lebrun, P.; Antoine, M. H.; Fontaine,
J .; Masereel, B.; Schynts, M.; Dupont, L.; Herchuelz, A.; Delarge,
J . 3-(Alkylamino)-4H-Pyrido[4,3-e]-1,2,4-Thiadiazine 1,1-Diox-
ides as Powerful Inhibitors of Insulin Release From Rat Pan-
creatic â-Cells: a New Class of Potassium Channel Openers? J .
Med. Chem. 1993, 36, 3211-3213.
(
(
1
995, 2, 573-582.
(
30) Lebrun, P.; Arkhammar, P.; Antoine, M. H.; Nguyen, Q. A.;
Hansen, J . B.; Pirotte, B. A Potent Diazoxide Analogue Activat-
ing ATP-Sensitive K+ Channels and Inhibiting Insulin Release.
Diabetologia 2000, 43, 723-732.
(
9) Atwal, K. S. Advances in the Structure-Activity Relationships,
Mechanisms of Action, and Therapeutic Utilities of ATP-Sensi-
tive Potassium Channel Openers. Drug Dev. Res. 1994, 33, 250-
2
62.
(
10) Shindo, T.; Yamada, M.; Isomoto, S.; Horio, Y.; Kurachi, Y. SUR2
Subtype (A and B)-Dependent Differential Activation of the
Cloned ATP-Sensitive K+ Channels by Pinacidil and Nicorandil.
Br. J . Pharmacol. 1998, 124, 985-991.
(31) Tagmose, T. M.; Mogensen, J . P.; Agerholm, P. C.; Arkhammar,
P. O. G.; Wahl, P.; Worsaae, A.; Hansen, J . B. Phenylcyan-
oguanidines As Inhibitors of Glucose-Induced Insulin Secretion
From â-Cells. Bioorg. Med. Chem. Lett. 2001, 11, 1749-1752.
(32) Kullin, M.; Li, Z.; Hansen, J . B.; Bjork, E.; Sandler, S.; Karlsson,
F. A. KATP Channel Openers Protect Rat Islets Against the Toxic
Effect of Streptozotocin. Diabetes 2000, 49, 1131-1136.
(
11) Moreau, C.; J acquet, H.; Prost, A. L.; D’hahan, N.; Vivaudou,
M. The Molecular Basis of the Specificity of Action of KATP
Channel Openers. EMBO J . 2000, 19, 6644-6651.
(
12) Reimann, F.; Gribble, F. M.; Ashcroft, F. M. Differential
Response of KATP Channels Containing SUR2A or SUR2B
Subunits to Nucleotides and Pinacidil. Mol. Pharmacol. 2000,
(
33) Rasmussen, S. B.; Sorensen, T. S.; Hansen, J . B.; Mandrup-
Poulsen, T.; Hornum, L.; Markholst, H. Functional Rest Through
Intensive Treatment With Insulin and Potassium Channel
Openers Preserves Residual Beta-Cell Function and Mass in
Acutely Diabetic BB Rats. Horm. Metab. Res. 2000, 32, 294-
5
8, 1318-1325.
(
13) D’hahan, N.; Moreau, C.; Prost, A. L.; J acquet, H.; Alekseev, A.
E.; Terzic, A.; Vivaudou, M. Pharmacological Plasticity of
Cardiac ATP-Sensitive Potassium Channels Toward Diazoxide
Revealed by ADP. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 12162-
3
00.
(
34) Aizawa, T.; Hansen, J . B.; Carr, R. D.; Brand, C. L.; Bodvars-
dottir, T. B.; Godnasson, B.; Komatsu, M.; Kaneko, T.; Nish-
izawa, T.; Itoh, N.; Yamauchi, K.; Hashizume, K. Diabetes
Prevention by NNC 55-0118, a Pancreo-Selective ATP-Sensitive
Potassium Channel Activator: Studies in Rat Models of Geneti-
cally Determined Obese Diabetes. IDF Abstr. 2000.
1
2167.
(
14) Trube, G.; Rorsman, P.; Ohno-Shosaku, T. Opposite Effects of
Tolbutamide and Diazoxide on the ATP-Dependent K+ Channel
in Mouse Pancreatic Beta-Cells. Pflugers Arch. 1986, 407, 493-
4
99.

Potassium Channels of Pancreatic â-Cells
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4187
(
35) Ofitserov, V. I.; Mokrushin, Z. S.; Pushkareva, Z. V.; Nikiforova,
N. V. Synthesis of Analogues of 5(4)-Aminoimidazole-4(5)-
Carboxamide and Purinees VI. 5(4)-Aminoimidazole-4(5)-Sul-
fonamide Derivatives. Chem. Heterocycl. Compd. (Engl. Transl.)
Sulfonylurea Receptors. EMBO J . 1998, 17, 5529-5535.
(54) Wahl, P.; Dabrowski, M.; Boonen, H.; Olsen, H. L.; Kofod, H.;
Hansen, J . B. A Novel Diazoxide Analogue Selectively Activates
SUR1/Kir6.2 Channels. Abstr. 1468-P, ADA2001 2001.
(55) Yamada, M.; Isomoto, S.; Matsumoto, S.; Kondo, C.; Shindo, T.;
Horio, Y.; Kurachi, Y. Sulphonylurea Receptor 2B and Kir6.1
Form a Sulphonylurea Sensitive but ATP Insensitive K+ Chan-
nel. J . Physiol. (London) 1997, 499, 715-720.
(56) Suzuki, M.; Li, R. A.; Miki, T.; Uemura, H.; Sakamoto, N.;
Ohmoto-Sekine, Y.; Tamagawa, M.; Ogura, T.; Seino, S.; Marban,
E.; Nakaya, H. Functional Roles of Cardiac and Vascular ATP-
Sensitive Potassium Channels Clarified by Kir6.2-Knockout
Mice. Circ. Res. 2001, 88, 570-577.
(57) Randle, J . C.; Biton, C.; Lepagnol, J . M. Allosteric Potentiation
by Diazoxide of AMPA Receptor Currents and Synaptic Poten-
tials. Eur. J . Pharmacol. 1993, 247, 257-265.
(58) Manley, P. W.; Loffler-Walz, C.; Russ, U.; Hambrock, A.;
Moenius, T.; Quast, U. Synthesis and Characterization of a Novel
Tritiated K-ATP Channel Opener With a Benzopyran Structure.
Br. J . Pharmacol. 2001, 133, 275-285.
(59) Seltzer, H. S.; Allen, E. W. Hyperglycemia and Inhibition of
Insulin Secretion During Administration of Diazoxide and
Trichlormethiazide in Man. Diabetes 1969, 18, 19-28.
(60) Bjork, E.; Berne, C.; Karlsson, F. A. Induction of Beta-Cell Rest
in Type 1 Diabetes. Studies on the Effects of Octreotide and
Diazoxide. Diabetes Care 1998, 21, 427-430.
(61) Pruitt, A. W.; Faraj, B. A.; Dayton, P. G. Metabolism of Diazoxide
in Man and Experimental Animals. J . Pharmacol. Exp. Ther.
1974, 188, 248-256.
(62) Dayton, P. G.; Pruitt, A. W.; Faraj, B. A.; Israili, Z. H.
Metabolism and Disposition of Diazoxide. A Mini-Review. Drug.
Metab. Dispos. 1975, 3, 226-229.
(63) Vaidyanathan, G.; Wilson, J . W. Reaction of Cyclopropanamines
With Hypochlorite. J . Org. Chem. 1989, 54, 1815-1820.
(64) Kirmse, W.; Hellwig, G.; Van, C. P. Deamination Reactions 42.
Addition of Diazocyclopropanes to Carbonyl Compounds. Chem.
Ber. 1986, 119, 1511-1524.
(65) Cox, E. F.; Caserio, M. C.; Silver, M. S.; Roberts, J . D. Small-
Ring Compounds. XXXIV. Carbonium Ion Reactions of 1-Meth-
ylcyclobutyl, (1-Methylcyclopropyl)-Carbinyl and (Beta-Methyl-
allyl)-Carbinyl Derivatives. J . Am. Chem. Soc. 1961, 83, 2719-
2724.
1
976, 924-927.
(36) Unterhalt, B.; Moghaddam, S. 2,3-Dihydro-3-Oxo-Thienoisothia-
zole 1,1-Dioxides and Their 3- Thioxo Compounds. Pharmazie
1
994, 49, 115-117.
(
37) If water was present during the Curtius rearrangement, a
substantial amount of the symmetric urea derivative 3,3′-
ureylenedi-5-chlorothiophene-2-sulfonamide was formed as a
byproduct.
(
38) J ensen, K. G.; Pedersen, E. B. Phosphorus Pentoxide in Organic
Synthesis 4. Synthesis of 4H-1,2,4-Benzothiadiazin-3-Amine 1,1-
Dioxides. Chem. Scr. 1982, 20, 248-250.
(
39) Andersen, K.; Begtrup, M. Activation of Phosphorus Oxychloride
for Hydroxy to Chlorine Displacement Reactions: Preparation
of 3-Chloro- and 3-Ethylthio-4,5,6,7-Tetrahydroisoxazolo(4,5-c)-
pyridine. Acta Chem. Scand. 1992, 46, 1130-1132.
(
40) Smyth, T. P.; Carey, A.; Hodnett, B. K. Inexpensive, Active KF
for Nucleophilic Aromatic Displacement Reactions. Tetrahedron
1
995, 51, 6363-6376.
(
41) Melikian, A.; Boigegrain, R.; Kan, J . P.; Soubrie, P. Regioselective
Synthesis of 4-Chloro-3-Hydroxyanthranilic Acid, a Potent in
Vitro Inhibitor of 3-Hydroxyanthranilic Acid Oxygenase Activity
From Rat Brain. Eur. J . Med. Chem. 1990, 25, 267-270.
42) Reed, J . N.; Snieckus, V. Ortho-Amination of Lithiated Tertiary
Benzamides. Short Route to Polysubstituted Anthranilamides.
Tetrahedron Lett. 1983, 24, 3795-3798.
(
(
(
43) Spagnolo, P.; Zanirato, P.; Gronowitz, S. General Route for the
Facile Transformation of Ortho-Substituted Lithiobithienyls into
Amino Derivatives. J . Org. Chem. 1982, 47, 3177-3180.
44) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J .
Experimental and Computational Approaches to Estimate Solu-
bility and Permeability in Drug Discovery and Development
Settings. Adv. Drug Delivery Rev. 1997, 23, 3-25.
(
45) Videbaek, L. M.; Aalkjaer, C.; Mulvany, M. J . Effect of Pinacidil
on Norepinephrine- and Potassium-Induced Contractions and
Membrane Potential in Rat and Human Resistance Vessels and
in Rat Aorta. J . Cardiovasc. Pharmacol. 1988, 12 (Suppl. 2),
S23-S29.
(
46) Pirotte, B.; Fontaine, J .; Lebrun, P. New Activators of Potassium
Channels: Design, Pharmacology and Therapeutic Perspectives
J . Pharm. Belg. 1995, 50, 167-178.
(66) Hamlin, K. E.; Freifelder, M. The Synthesis of Some Tertiary
Carbinamines As Vasopressor Agents. J . Am. Chem. Soc. 1953,
75, 369-373.
(
47) Lebrun, P.; Antoine, M. H.; Ouedraogo, R.; Herchuelz, A.; de
Tullio, P.; Delarge, J .; Pirotte, B. Pyridothiadiazines As Potent
Inhibitors of Glucose-Induced Insulin Release. Adv. Exp. Med.
Biol. 1997, 426, 145-148.
48) Khelili, S.; de Tullio, P.; Lebrun, P.; Fillet, M.; Antoine, M. H.;
Ouedraogo, R.; Dupont, L.; Fontaine, J .; Felekidis, A.; Leclerc,
G.; Delarge, J .; Pirotte, B. Preparation and Pharmacological
Evaluation of the R- and S-Enantiomers of 3-(2-Butylamino)-
(67) Koziara, A.; Osowska-Pacewicka, K.; Zawadzki, S.; Zwierzak, A.
A Simple, One-Pot Transformation of t-Alkyl Chlorides into (t-
Alkyl)Amines. Synthesis 1987, 487-489.
(
(68) Ohuchida, S.; Tani, K. Omega-Cycloalkyl-Prostaglandin E1
Derivatives. European Patent Application EP-A1 0974580, J an
26, 2000.
(69) Efrat, S.; Linde, S.; Kofod, H.; Spector, D.; Delannoy, M.; Grant,
S.; Hanahan, D.; Baekkeskov, S. Beta-Cell Lines Derived From
Transgenic Mice Expressing a Hybrid Insulin Gene-Oncogene.
Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 9037-9041.
(70) Le Brigand, L.; Virsolvy, A.; Manechez, D.; Godfroid, J . J .;
Guardiola-Lemaitre, B.; Gribble, F. M.; Ashcroft, F. M.; Bataille,
D. In Vitro Mechanism of Action on Insulin Release of S-22068,
a New Putative Antidiabetic Compound. Br. J . Pharmacol. 1999,
128, 1021-1026.
(71) Aprigliano, O.; Hermsmeyer, K. Trophic Influence of the Sym-
pathetic Nervous System on the Rat Portal Vein. Circ. Res. 1977,
41, 198-206.
(72) Mulvany, M. J .; Halpern, W. Contractile Properties of Small
Arterial Resistance Vessels in Spontaneously Hypertensive and
Normotensive Rats. Circ. Res. 1977, 41, 19-26.
4
H- and 3-(3-Methyl-2-Butylamino)-4H-Pyrido[4,3-e]-1,2,4-Thia-
diazine 1,1-Dioxide, Two Tissue Selective ATP-Sensitive Po-
tassium Channel Openers. Bioorg. Med. Chem. 1999, 7, 1513-
1
520.
(
49) Petersen, H. J .; Nielsen, C. K.; Arrigoni-Martelli, E. Synthesis
and Hypotensive Activity of N-Alkyl-N′-Cyano-N′′-Pyridyl
Guanidines. J . Med. Chem. 1978, 21, 773-781.
50) Manley, P. W.; Quast, U. Structure-Activity Studies of Potas-
sium Channel Opening in Pinacidil Type Cyanoguanidines,
Nitroethenediamines, Thioureas, and Ureas. J . Med. Chem.
(
1
992, 35, 2327-2340.
(
51) Gribble, F. M.; Tucker, S. J .; Seino, S.; Ashcroft, F. M. Tissue
Specificity of Sulfonylureas: Studies on Cloned Cardiac and
Beta-Cell KATP Channels. Diabetes 1998, 47, 1412-1418.
(
52) Schwanstecher, M.; Sieverding, C.; Fischer, A.; Dorschner, H.;
Bassen, D.; Schwanstecher, C.; Gonzalez, G.; Clement, J . P.;
Aguilarbryan, L.; Bryan, J . Potassium Channel Openers Act
Through Sulfonylurea Receptors. Biophys. J . 1998, 74, A35.
53) Schwanstecher, M.; Sieverding, C.; Dorschner, H.; Gross, I.;
Aguilar-Bryan, L.; Schwanstecher, C.; Bryan, J . Potassium
Channel Openers Require ATP to Bind to and Act Through
(73) Brand, C. L.; Rolin, B.; J orgensen, P. N.; Svendsen, I.; Kris-
tensen, J . S.; Holst, J . J . Immunoneutralization of Endogenous
Glucagon With Monoclonal Glucagon Antibody Normalizes Hy-
perglycemia in Moderately Streptozotocin-Diabetic Rats. Dia-
betologia 1994, 37, 985-993.
(
J M0208121