Structure-activity relationship for aryl and heteroaryl boronic acid inhibitors of hormone-sensitive lipase

Article abstract of DOI:10.1016/j.bmc.2004.12.042

A range of aryl and heteroaryl boronic acids were tested for their in vitro hormone-sensitive lipase inhibitory properties. (2-Benzyloxy-5-fluorophenyl) boronic acid, (2-benzyloxy-5-chlorophenyl)boronic acid and 5-bromothiophene-2- boronic acid were found to be the most potent HSL inhibitors with IC ₅₀ values of 140, 17 and 350 nM, respectively.

Full text of DOI:10.1016/j.bmc.2004.12.042

Bioorganic & Medicinal Chemistry 13 (2005) 2305–2312
Structure–activity relationship for aryl and heteroaryl boronic
acid inhibitors of hormone-sensitive lipase
*
Søren Ebdrup, Poul Jacobsen, Anupma Dhanda Farrington and Per Vedsø
Novo Nordisk A/S, Novo Nordisk Park, 2760 M a˚ løv, Denmark
Received 9 July 2004; accepted 22 December 2004
Available online 21 January 2005
Abstract—A range of aryl and heteroaryl boronic acids were tested for their in vitro hormone-sensitive lipase inhibitory properties.
(
2-Benzyloxy-5-fluorophenyl)boronic acid, (2-benzyloxy-5-chlorophenyl)boronic acid and 5-bromothiophene-2-boronic acid were
found to be the most potent HSL inhibitors with IC50 values of 140, 17 and 350 nm, respectively.
2004 Elsevier Ltd. All rights reserved.
ꢀ
1
. Introduction
Due to the dysregulated metabolism of type 2 diabetic
patients and the elevated level of FA the interest in tar-
geting HSL with the aim of reducing insulin resistance
and dyslipidemia in obese, prediabetic and diabetic indi-
viduals has recently gained considerable interest.
Type 2 diabetes including the metabolic syndrome is a
chronic multifactorial metabolic disease. One of the
characteristics of the disease is elevated levels of fatty
acids (FAs) in the circulation, which is widely consid-
ered a major pathogenetic factor. It has been shown that
elevated FA levels are associated with peripheral insulin
resistance, decreased glucose uptake by muscle tissue
and increased hepatic triglyceride (TG) release to the cir-
8
Thus carbamoyl triazoles (1), carbamoyl isoxazole-5-
12
9
10,11
ones (2), oxadiazoles
(3), cyclipostines (4), carbaz-
1
3
14
ates (5) and pyrrolopyrazinediones (6) have been
identified as HSL inhibitors. Most of the HSL inhibitors
described are expected to be pseudosubstrate inhibi-
1
–3
culation.
8
tors (inhibition mechanism for pyrrolopyrazinediones
unknown). A few carbamoyltriazoles (1, R1 = Cl and
Hormone-sensitive lipase (HSL) is a multifunctional tis-
sue lipase that plays a central role in the overall energy
homeostasis as it is the rate limiting enzyme in the
8
R2 = CF CH or 1, R1 = OCF and R2 = CH ) and a
3
3
2
3
carbamoyl isoxazole-5-one (2, R1 and R2 forms a S-3-
methyl substituted piperidine ring and R3 = CH-
4
release of FA from adipocytes. HSL catalyzes the first
8
and second step in the breakdown of triglycerides,
releasing two molecules of FA and monoacylglycerol,
which is further hydrolyzed by monoacylglycerol
(CH ) ) are the only selective HSL inhibitors described
2
3
(Fig. 1).
5
lipase.
Reversible hydrolase inhibitors such as boronic acids
have been extensively used in the development of en-
zyme inhibitors of peptidases/proteases and lipases and
HSL has a broad substrate specificity as it catalyzes the
4
15
hydrolysis of both cholesteryl esters and acylglycerides.
The activity of HSL is regulated via phosphorylation/
dephosphorylation primarily controlled by catechol-
employed as pharmaceutical agents. In this paper the
HSL structure–activity relationship for a range of aryl
and heteroaryl boronic acids will be presented together
with counter screening data for selected hydrolases.
6
amines and insulin.
The enzyme belongs to a class of hydrolases that adapt
the a/b hydrolase fold and contains a catalytic triad of
serine, histidine and aspartic acid and an oxyanion hole.
2. Results and discussion
7
Boronic acids are strong Lewis acids due to the incom-
plete boron electron shell. Depending on the phenyl sub-
stitution, most phenylboronic acids have a pK in the
*
Corresponding author. Tel.: +45 4443 4830; fax: +45 4466 3939;
e-mail: sebd@novonordisk.com
a
0
968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.12.042

2306
S. Ebdrup et al. / Bioorg. Med. Chem. 13 (2005) 2305–2312
(
7a–e) are capable of inhibiting HSL with IC50 values
O
between 0.85 and 5.9 lM. The calculated pK values
O
a
R3
O
N
(
Table 1) of the most active meta substituted phenylbo-
ronic acids are about 7.5 whereas the inactive analogues
7h,7f–j) all have higher pK values (7.8–8.6) suggesting
N
R2
N
N
N
(
a
S
R1
N
O
that they are not capable of forming a strong interaction
with the active site serine as compared to 7a–e. Despite
R1
R2
2
1
the high pK value of 8.0, 7g is capable of inhibiting
a
HSL with an IC50 value of 4.6 lM. As a reference, the
O
R1
R1
O
experimental determined pK value of phenylboronic
16
a
N
O
N
acid is 8.8. and the HSL IC50 value of phenylboronic
acid is above 10 lM.
O
O
O
P
H
R2
R1
O
O
O
4
3
ortho Substituted phenylboronic acids were also tested
(
O
O
Table 2) but only 8f and 8g were active with IC50 values
R2
R1
O
O
of 3.4 and 1.1 lM, respectively. Due to the relatively
R2
O
N
R3
high calculated pK value of 8.6 it is expected that lipo-
N
N
a
R3
R1
N
O
philic interactions of the butyl and benzyl ether groups
are responsible for the surprisingly low IC50 values.
Compound 8a–e, which have lower pK values than 8f
5
O
6
a
and 8g were not active.
Figure 1. Structure of HSL inhibitors.
Only one para substituted phenylboronic acid (9c, Table
3
range 4.5–9.0 and suitable substitution of the aryl ring
can provide boronic acids with the correct properties for
ready conversion from a neutral and trigonal planar
) has an IC50 value in the low micromole range
(
1.3 lM) in contrast to its relatively high pK value of
a
8
.4. Surprisingly 9d,9e and 9g were not active despite
2
3
sp boron to an anionic tetrahedral sp boron under phy-
1
5
2
3
siological conditions. This sp –sp transition state con-
version is similar to that seen for the amide and ester
carbon atom during cleavage of natural substrates.
Table 2. Activity and selectivity of ortho substituted phenylboronic
acids
OH
Due to the described characteristics of aryl and hetero-
aryl boronic acids as reversible hydrolase inhibitors we
decided to explore if it was possible to find aryl or het-
eroaryl boronic acids, which were capable of inhibiting
HSL. Different substituted phenyl and thiophene-
boronic acids were tested and selected compounds were
also counter screened for butyrylcholin esterase (BChE)
and acetylcholine esterase (AChE) inhibitory activity.
B
OH
R
No.
R
HSL IC50, lM BChE IC50, lM Calculated pK
a
8a
8b
Br
Cl
F
>50
>50
>50
>50
>50
3.4
na
na
8.3
8.2
8.3
8.1
7.8
8.6
8.6
a
8c
>10
11
a
8
8
8
8
d
CF
3
a
e
NO
2
>10
>50
21
As can be seen from Table 1 phenylboronic acids bear-
ing an electron withdrawing group in the meta position
a
b
c
f
OBu
OBn
a
g
1.1
a
b
c
AChE IC50 > 50 lM.
OBu: Butyl ether.
OBn: Benzylether.
Table 1. Activity and selectivity of meta substituted phenylboronic
acids
OH
R
B
OH
Table 3. Activity and selectivity of para substituted phenylboronic
acids
No.
R
HSL IC50
,
BChE IC50
,
Calculated
pK
OH
lM
lM
a
B
OH
a
7
7
7
7
7
7
7
7
7
7
a
Br
Cl
F
5.9 ± 0.55
4.4
5.9
>10
>50
>50
>50
>50
na
7.6
7.6
7.5
7.4
7.6
7.8
8.0
8.1
8.1
8.6
a
b
R
a
c
a
No.
R
HSL IC50
,
AChE/BChE
Calculated
pK
d
OF
3
C
0.85 ± 0.45
1.0 ± 0.20
>50
a
lM
IC50, lM
a
e
f
CF
CO
3
2
Et
9a
9b
9c
9d
9e
9f
OCH
F
2
Ph
>50
>50
1.3
na
na
8.7
8.7
8.4
7.8
8.3
9.0
7.2
a
g
h
SMe
OEt
OBu
Me
4.6
>50
>50
na
Cl
>50
>50
>50
>50
na
a
b
i
j
21
>50
>50
na
CF
OCF
OMe
SO Me
3
>10
>50
>50
>50
3
a
AChE IC50 > 50 lM.
OBu: Butyl ether.
b
9g
2

S. Ebdrup et al. / Bioorg. Med. Chem. 13 (2005) 2305–2312
2307
lower calculated pK values. Steric hindrance might be
a
responsible for this lack of inhibitory activity.
low micromole range. 10c (3,5-difluoro), 10e (3-chloro-
4-fluoro) and 10f (3,4-difluoro) had IC50 values of 2.2,
2.6 and 6 lM, respectively. Surprisingly 10g (3,4-di-
As most of the potent phenylboronic acids have electron
withdrawing groups in the meta and para positions we
tested different disubstituted phenylboronic acids pos-
sessing two electron withdrawing groups 10a–g (Table
chloro) was not active even though both mono substi-
tuted analogues 7b and 9c were active.
The ortho ether structure element of 8f and 8g and the
meta and para electron withdrawing structure element
4
). Three dihalo-phenylboronic acids were active in the
Table 4. Activity and selectivity of disubstituted phenylboronic acids
No.
Structure
HSL IC50, lM
AChE/BChE IC50, lM
Calculated pK
a
F
F
OH
B
F
OH
10a
>50
na
6.6
F
F
F
OH
Cl
B
OH
1
0b
>10
2.2
na
6.6
6.5
Cl
OH
B
F
OH
1
0c
>50
F
OH
B
Br
OH
1
0d
>50
2.6
na
6.6
7.6
Br
OH
B
Cl
1
0e
OH
>50
F
F
OH
B
1
0f
OH
6
>50
na
7.6
7.4
F
OH
B
Cl
1
0g
0h
OH
OH
>50
Cl
Cl
OH
B
1
28
>50
na
7.6
7.5
O
OH
B
F
1
0i
0j
OH
>10
O
OH
B
OH
1
>50
>50
>50
>50
7.6
O
Br
OH
B
OH
10k
7.6
O
Br
(continued on next page)

2308
S. Ebdrup et al. / Bioorg. Med. Chem. 13 (2005) 2305–2312
Table 4 (continued)
No.
Structure
HSL IC50, lM
AChE/BChE IC50, lM
Calculated pK
a
O
OH
B
a
OH
1
0l
0.14 ± 0.028
>50
>10
7.5
F
O
OH
B
OH
1
0m
0.017 ± 0.070
7.6
Cl
O
OH
B
1
0n
0.76
>50
>10
na
8.7
7.6
OH
F
OH
B
F
OH
1
0o
O
OH
B
OH
10p
>50
na
7.5
F
a
Hepatic lipase (HL) IC50 > 10 lM; lipoprotein lipase (LPL) IC50 > 10 lM; and pancreatic lipase (PL) IC50: 7lM.
of 7a–e and 9c were combined in the search for more
potent inhibitors. The 2-methoxy-5-chloro analogue
IC50 values of 140 nm and 17 nm, respectively. The 4-
fluoro regioisomer 10n was also active with an IC50
1
0h was a weak inhibitor with an IC50 value of 28 lM.
value of 0.76 lM. The pK values of 10l and 10n were
a
The ortho propoxy and butoxy ethers 10j and 10k were
inactive but surprisingly 10l and 10m bearing a benzyl-
oxy group in the 2-position and a fluoro or a chloro
group in the 5-position were potent HSL inhibitors with
in the same range as the pK values of 7b,7c and 9b.
a
Compound 10o, the 4-benzyloxy isomer of 10l, was not
capable of inhibiting HSL which is in agreement with
data for 9a.
Table 5. Activity and selectivity of thiopheneboronic acids
No.
Structure
HSL IC50, lM
BChE IC50, lM
Calculated pK
a
OH
1
1a
1b
B
S
>50
na
8.4
OH
OH
b
1
Cl
Br
B
0.71 ± 0.38
>50
>50
8.3
8.2
S
S
OH
OH
a
1c
1
B
0.35 ± 0.053
OH
OH
b
1
1d
B
>50
>50
>50
na
8.7
7.5
S
OH
OH
1
1e
B
S
O
OH
OH
b
1
1f
>10
>50
8.5
B
S
OH

S. Ebdrup et al. / Bioorg. Med. Chem. 13 (2005) 2305–2312
2309
Table 5 (continued)
No.
Structure
HSL IC50, lM
BChE IC50, lM
Calculated pK
a
O
1
1g
1h
>50
na
na
8.1
OH
B
S
S
OH
Br
1
OH
>50
8.1
B
OH
O
O
O
c
1
1i
1j
B
>50
>50
na
na
7.9
S
O
O
c
7.9
1
N
O
S
B
O
S
O
S
O
b
Cl
B
O
c
8.3
11k
1.28
1.09
1.1
>10
>10
>10
O
b
c
8.3
11l
Cl
Cl
B
S
S
O
O
b
c
8.3
11m
B
N
O
a
b
c
Hepatic lipase (HL) IC50 > 10 lM; lipoprotein lipase (LPL) IC50 > 10 lM; and pancreatic (PL) IC50: >10 lM.
AChE IC50 > 50 lM.
Calculated pK value of corresponding boronic acid.
a
Table 6. Activity and selectivity of 3-sulfonamide substituted phenylboronic acids
No.
Structure
HSL IC50, lM
BChE IC50, lM
Calculated pK
a
O
S
OH
B
a
a
N
N
1
1
1
1
3a
>50
>10
7.4
OH
O
O
S
OH
B
3b
>50
>50
>50
>10
>10
>50
7.4
7.4
7.4
OH
O
H
N
O
OH
B
a
S
3c
OH
O
H
N
O
OH
a
3d
S
B
OH
O
OH
O
a
1
3e
N
N
S
O
B
>50
>50
7.5
OH
OH
OH
B
O
a
1
1
3f
S
7.5
>10
>50
7.4
7.5
O
OH
O
a
N
S
B
3g
>50
OH
O
a
AChE IC50 > 50 lM.

2310
S. Ebdrup et al. / Bioorg. Med. Chem. 13 (2005) 2305–2312
The two HSL inhibitors 10l and 11c were also tested for
N
B
N
B
hepatic lipase (HL), lipoprotein lipase (LPL) and pan-
creatic lipase (PL) activity. None of the compounds
inhibited HL and LPL but unfortunately 10l inhibited
PL with an IC50 value of 7 lM. Further investigations
will be conducted with the aim of identifying selective
HSL inhibitors and to evaluate the biological impor-
tance of targeting HSL. Inhibitors of HSL might have
a future for the treatment of type 2 diabetes, the meta-
bolic syndrome and impaired glucose tolerance.
O
O
O
O
1
. NCS, CH Cl
2
2
B(OH)2
1
. BuLi, -78 °C
2. HNR R
1 2
2
. SO (g)
3. Dowex H
+
2
97%
15-69%
Br
SO Li
SO NR R
2 1 2
2
1
2a
13a-g
Scheme 1. Synthesis of 3-sulfonamide substituted phenylboronic acids.
A range of thiopheneboronic acids were also tested
(Table 5) and surprisingly the chloro and bromo substi-
tuted analogues 11b and 11c were potent HSL inhibitors
4. Experimental
with IC50 values of 710 nm and 350 nm, respectively.
The two compounds both have calculated pK values
a
4.1. General methods and materials
of about 8.3, which is considerably higher than the cor-
responding meta substituted phenylboronic acids (pK ꢀ
a
All reactions involving air-sensitive reagents were per-
formed under nitrogen using syringe-septum cap tech-
niques. The glassware was flame dried prior to use.
MgSO4 were used to dry solutions. Solvents were
removed in vacuo by rotary evaporation. Melting points
were recorded on a B u¨ chi 535 and are uncorrected.
NMR spectra were recorded on a Bruker AMX 400 or
Bruker DRX 300 instrument with tetramethylsilane
(TMS) as internal standard. Liquid chromatography-
mass spectrometry (LC–MS) analysis was performed
7.6). Surprisingly, the acetyl analogue 11e was inactive
despite a lower pK value (7.5). The extended analogues
a
1
1i and 11j were also inactive indicating that only small
substituents are accepted in the 5-position of the thio-
phene ring. The boronic ester analogues 11k–m of 11b
were all equipotent with the boronic acid 11b suggesting
that the boronic esters are hydrolyzed to 11b in the
aqueous assay media.
ꢂ
Electron withdrawing groups in the meta position of the
phenylboronic acid seem in general to be the most
favourable pattern. Thus a series of 3-sulfonamide
substituted phenylboronic acids 13a–g were synthesized
on HP1100 MSD equipped with a Waters Xterra MS
C-18 5 lm 3.0 · 50 mm column. All solvents and re-
agents were obtained from commercial sources and used
without further purification. Butyl lithium was titrated
prior to use.
1
7
employing our recently described protocol in which
the 3-bromo substituted N-methyldiethanolamine pro-
tected phenylboronic acid undergoes selective bro-
mine–lithium exchange at À78 ꢁC upon treatment with
n-BuLi (Scheme 1). The lithio intermediate was trapped
with sulfur dioxide providing the corresponding lithium
sulfinate 12a in 97% yield. Oxidation with N-chlorosuc-
cinimide gives the sulfonyl chloride, which upon reac-
tion with a range of alkyl amines produce the
corresponding 3-sulfonamide substituted arylboronic
acids 13a and 13g (Table 6). However, all compounds
except 13f (IC50 of 7.5 lM) were inactive.
4.2. 3-[5-(5,5-Dimethyl-[1,3,2]dioxaborinan-2-yl)-thio-
phen-2-yl]-acrylic acid ethyl ester (11i)
A solution of 2,2,6,6-tetramethylpiperidine (1.53 mL,
mmol) in anhydrous THF (50 mL) was stirred under
9
nitrogen. The mixture was cooled to À5 ꢁC (internal
temperature). n-Butyllithium (1.6 M in hexanes,
5
.5 mL, 8.8 mmol) was added over 6 min. The mixture
was stirred for an additional 30 min at À5 ꢁC before
being cooled to À105 ꢁC N (l)/Et O. 3-Thiophen-2-yl-
2
2
acrylic acid ethyl ester (1.46 g, 8 mmol) dissolved in
THF (30 mL) was added over 5 min and the solution
was stirred for another 30 min. Then triisopropylborate
3. Conclusion
(
2.3 mL, 10 mmol) was added and the reaction mixture
was allowed to warm up to room temperature over
h. Glacial acetic acid (0.52 mL, 9 mmol) and neopent-
In conclusion, a range of substituted phenyl and
thiopheneboronic acids have been identified as HSL
inhibitors with IC50 values in the low to high nm
range. The phenylboronic acids 10l and 10m and the
thiopheneboronic acid 11c were the most potent HSL
inhibitors with IC50 values of 140, 17 and 350 nm,
respectively.
1
ylglycol (1.04 g, 10 mmol) was added followed by stir-
ring for 30 min. Addition of aqueous NH Cl followed
by extraction with CH Cl , drying of the combined
2
organic phases (MgSO ), filtration and evaporation pro-
4
duced a crude product, which was crystallized from hep-
tane–EtOAc to give 0.96 g (41%) of the title compound
4
2
It seems as if large substituents are only accepted in the
ortho and meta position of phenylboronic acids. In gen-
eral, it seems like steric hindrance and lipophilic interac-
tions are of greater importance for the inhibitory
1
as yellow crystals. H NMR (CDCl ): d 7.77 (d, 1H),
3
7
3
2
.48 (d, 1H), 7.27 (d, 1H), 6.27 (d, 1H), 4.26 (q, 2H),
.77 (s, 4H), 1.35 (t, 3H), 1.04 (s, 6H). LC–MS: m/z
27 (M-pinacol+H), 249 (M-pinacol+Na).
activity than the actual pK value of the boronic acid.
a
As 8g and 11l–n are more potent than what would have
been expected from their calculated pK values it is ex-
4.3. 5-(5,5-Dimethyl-[1,3,2]dioxaborinan-2-yl)-thiophene-
2-sulfonic acid tert-butylamide (11j)
a
pected that lipophilic interaction between the ortho ben-
zyloxy group of the inhibitors and HSL is important for
the inhibitory properties.
A solution of diisopropylamine (420 mg, 4.16 mmol) in
anhydrous THF (4 mL) was stirred under nitrogen.

S. Ebdrup et al. / Bioorg. Med. Chem. 13 (2005) 2305–2312
2311
The mixture was cooled to À5 ꢁC (internal temperature).
n-Butyllithium (1.6 M in hexanes, 2.5 mL, 4 mmol) was
added over 3 min. The mixture was stirred an additional
period at À78 ꢁC. The mixture was stirred at À78ꢁC
for 15 min. Then gaseous sulfur dioxide (ca. 5 g) was
added causing an immediate precipitation and a 40 ꢁC
increase in the internal temperature. The mixture was
allowed to warm to room temperature and stirred
for 1 h. The precipitated lithium sulfinate was
3
0 min at À5 ꢁC before being cooled to À78 ꢁC. Thio-
phene-2-sulfonic acid tert-butylamide (438 mg, 2 mmol)
dissolved in THF (4 mL) was added over 2 min and the
suspension was stirred for another 45 min. Triisoprop-
ylborate (760 mg, 4 mmol) was added and the reaction
mixture was allowed to warm up to room temperature
over 30 min. Addition of glacial acetic acid (0.24 mL,
isolated by filtration under N (g), washed with THF
2
(50 mL) and dried in vacuo providing 2.81 g (97%) of
1
the title compound as a solid: H NMR (DMSO-d ): d
6
7.66 (s, 1H), 7.39–7.32 (m, 2H), 7.17 (t, 1H), 3.97-3.84
(m, 4H), 3.27–3.21 (m, 2H), 2.97–2.89 (m, 2H), 2.18 (s,
3H).
4
.2 mmol) and neopentylglycol (235 mg, 2.26 mmol) fol-
lowed by stirring for 2 h. Addition of aqueous NH Cl
4
followed by extraction with CH Cl , drying of the com-
2
2
bined organic phases (MgSO ), filtration and evapora-
4
4.8. Representative procedure for preparation of sulfon-
amide substituted phenylboronic acids: 3-benzylsulfamo-
ylbenzeneboronic acid (13a)
tion produced 684 mg slightly impure product. This
was recrystallized from heptane–EtOAc (5:1) to give
1
1
06 mg (16%) of the title compound as crystals.
H
NMR (CDCl ): d 7.60 (d, 1H), 7.41 (d, 1H), 4.52 (br
s, 1H), 3.77 (s, 4H), 1.30 (s, 9H), 1.04 (s, 6H). LC–
MS: m/z 286 (M-pinacol+Na).
Lithium; 3-(6-methyl-[1,3,6,2]dioxazaborocan-2-yl)-benz-
enesulfinate (275 mg, 1.0 mmol) was suspended in
3
CH Cl
1
(2 mL).
.10 mmol) was added and the mixture was stirred at
N-Chlorsuccinimide
(147 mg,
2
2
4
[
.4. 2-(5-Chlorothiophen-2-yl)-4,4,5,5-tetramethyl-
1,3,2]dioxaborolane (11k)
room temperature for 1 h. Benzylamine (0.23 mL,
2.1 mmol) was added and the reaction mixture was stir-
red for 1 h at room temperature and then Dowex
50WX2-400 cation exchange resin (ca. 1 g) was added
and the mixture stirred for a further 1 h. The resin was
removed by filtration and extracted with CH Cl –
5
(
2
-Chloro-2-thiopheneboronic acid (1 mmol) and pinacol
1 mmol) in toluene was stirred at room temperature for
h. The solution was washed with water and evaporated
2
2
to dryness to give the title compound (20%) as a volatile
MeOH (9:1). 1 N NaOH was added to the combined or-
ganic filtrates and the aqueous phase was washed with
CH Cl . The aqueous phase was acidified with 1 N
1
oil. H NMR (CDCl ): d 7.41 (d, 1H), 6.98 (d, 1H), 1.34
3
(
s, 12H). LC–MS: m/z 162 (M-pinacol+H).
2
2
HCl and the resulting crystals were isolated by filtration
to give 107 mg (37%) of the title compound: H NMR
1
4.5. 2-(5-Chlorothiophen-2-yl)-5,5-dimethyl-[1,3,2]dioxa-
borinane (11l)
(DMSO-d + DCl): d 8.24 (s, 1H), 8.03 (d, 1H), 7.85
6
(dt, 1H), 7.56 (t, 1H), 7.31–7.21 (m, 5H), 3.96 (s, 2H).
LC–MS: m/z 314 (M+Na).
5
-Chloro-2-thiopheneboronic acid (1 mmol) and
neopentylglycol (1 mmol) in toluene was stirred at
room temperature for 2 h. The solution was washed
with water and evaporated to dryness to give the
4.9. 3-Cyclohexylmethylsulfamoylbenzeneboronic acid
(13b)
1
title compound (54%) as crystals. Mp 42–48 ꢁC;
H
NMR (CDCl ): d 7.33 (d, 1H), 6.95 (d, 1H), 3.74 (s,
3-Cyclohexyl-methylamine was used instead of benzyl-
3
1
4H), 1.03 (s, 6H). LC–MS: m/z 162 (M-neopentyl-
glycol+H).
amine. Yield 21%. H NMR (DMSO-d + DCl): d 8.21
6
(s, 1H), 8.03 (d, 1H), 7.83 (dt, 1H), 7.58 (t, 1H), 2.55
(d, 2H), 1.68–1.65 (m, 4H), 1.40–1.25 (m, 1H), 1.20–
1.04 (m, 3H), 0.87–0.72 (m, 2H). LC–MS: m/z 298
(M+H), 320 (M+Na).
4
(
.6. 2-(5-Chlorothiophen-2-yl)-[1,3,6,2]dioxazaborocane
11m)
5
-Chloro-2-thiopheneboronic acid (1 mmol) and dietha-
4.10. 3-(3,3-Dimethyl-butylsulfamoyl)benzene-boronic
acid (13c)
nolamine (1 mmol) in toluene was stirred at room
temperature for 2 h. The solution was evaporated to
dryness and recrystallized from heptane–iso-propanol
3,3-Dimethyl-butylamine was used instead of benzyl-
1
(
4:1) to give the title compound (28%) as crystals. Mp
amine. Yield 24%. H NMR (DMSO-d + DCl): d 8.21
6
1
1
6
3
88–190 ꢁC; H NMR (DMSO-d ): d 7.10 (br s, 1H),
(s, 1H), 8.04 (d, 1H), 7.84 (dt, 1H), 7.58 (t, 1H), 2.77–
2.70 (m, 2H), 1.31–1.23 (m, 2H), 0.80 (s, 9H). LC–MS:
m/z 286 (M+H), 308 (M+Na).
6
.94 (d, 1H), 6.82 (d, 1H), 3.88–3.73 (m, 4H), 3.15–
.03 (m, 2H), 2.90–2.81 (m, 2H). LC–MS: m/z 162 (M-
diethanolamine+H).
4.11. 3-(3-Methyl-butylsulfamoyl)benzeneboronic acid
(13d)
4.7. Lithium; 3-(6-methyl-[1,3,6,2]dioxazaborocan-2-yl)-
benzenesulfinate (12a)
3
-Methyl-butylamine was used instead of benzylamine.
1
To a stirred solution of 3-bromobenzeneboronic acid N-
methyldiethanolamine cyclic ester (3.31 g, 11.7 mmol) in
dry THF (100 mL) was added dropwise 1.43 M solution
in hexanes n-BuLi (7.4 mL, 10.5 mmol) over a 3 min
Yield 15%. H NMR (DMSO-d + DCl): d 8.23
6
(s, 1H), 8.04 (d, 1H), 7.84 (dt, 1H), 7.58 (t, 1H), 2.74
(t, 2H), 1.54 (apparent sep, 1H), 0.78 (d, 6H). LC–MS:
m/z 272 (M+H), 294 (M+Na).

2312
S. Ebdrup et al. / Bioorg. Med. Chem. 13 (2005) 2305–2312
4
(
.12. 3-(Methyl-phenethylsulfamoyl)benzeneboronic acid
13e)
4.16. Calculation of pK values
a
The computed values were calculated using the pK soft-
a
N-Methyl-phenylethylamine was used instead of benzyl-
ware from ACDLabs (ACDLabs, Toronto, Canada,
version 6.0).
1
amine. Yield 69%. H NMR (DMSO-d + DCl): d 8.17
6
(
7
s, 1H), 8.08 (d, 1H), 7.78 (d, 1H), 7.60 (t, 1H), 7.33–
.19 (m, 5H), 3.18 (t, 2H), 2.77 (t, 2H), 2.69 (s, 3H).
LC–MS: m/z 320 (M+H), 342 (M+Na).
Acknowledgements
Inge Thøger Christensen is gratefully acknowledged for
calculation of the pK values. Lotte Gottlieb Sørensen,
4
.13. 3-(Phenethylsulfamoyl)benzeneboronic acid (13f)
a
Alice Ravn Madsen and Johnny Madsen are gratefully
acknowledged for analysis and synthetic support.
Phenethylamine was used instead of benzylamine. Yield
2
1
5%. H NMR (DMSO-d + DCl): d 8.24 (s, 1H), 8.05
6
(
2
d, 1H), 7.84 (dt, 1H), 7.58 (t, 1H), 7.40–7.14 (m, 5H),
.96 (t, 2H), 2.68 (t, 2H). LC–MS: m/z 306 (M+H),
References and notes
3
28 (M+Na).
1. Unger, R. H. Diabetes 1995, 44, 863–870.
2
3
4
5
. Boden, G. Diabetes 1997, 46, 3–10.
. Frayn, K. N. Diabetologia 2002, 45, 1201–1210.
. Raclot, T. Prog. Lipid Res. 2003, 42, 257–288.
. Sztalryd, C.; Komaromy, M. C.; Kraemer, F. B. J. Clin.
Invest. 1995, 95, 2652–2661.
4
.14. 3-(Butylmethylsulfamoyl)benzeneboronic acid (13g)
N-Methyl-butylamine was used instead of benzylamine.
Yield 68%. H NMR (DMSO-d + DCl): d 8.16 (s, 1H),
1
6. Langin, D.; Holm, C.; Lafontan, M. Proc. Nutr. Soc.
996, 55, 93–109.
7
6
1
. Contreras, J. A.; Karlsson, M.; Østerlund, T.; Laurell, H.;
8
(
.08 (d, 1H), 7.79 (d, 1H), 7.61 (t, 1H), 2.92 (t, 2H), 2.63
3, 3H), 1.43 (p, 2H), 1.27 (sextet, 2H), 0,87 (t, 3H). LC–
MS: m/z 272 (M+H), 294 (M+Na).
Svensson, A.; Holm, C. J. Biol. Chem. 1996, 271, 31426–
3
1430.
8. Ebdrup, S.; Sørensen, L. G.; Olsen, O. H.; Jacobsen, P.
J. Med. Chem. 2004, 47, 400–410.
9. Lowe, D. B. et al. Bioorg. Med. Chem. Lett. 2004, 14,
4
.15. Enzyme assay
3
155–3159.
The inhibition of the different lipases (HSL, HL,
LPL, PL) was determined in phosphorlipid stabilized
emulsion assays. The assays were based on a fluoro-
chrome-labelled triacylglyceride octadec-9-enoic acid
1
1
1
0. Shoenafinger, K.; Petry, S.; Mueller, G.; Baringhous,
K.-H. PCT int. Appl. WO 0166531, 2001.
1. Petry S.; Schoenafinger, K.; Mueler, G.; Baringhaus,
K.-H. PCT Appl. WO 0117981, 2001.
2. V e´ rtesy, L.; Beck, B.; Br o¨ nstrup, M.; Ehrlich, K.; Kurz,
8
2
-[12-(7-nitro-benzo[1,2,5]oxadiazol-4-ylamino)-dodeca-
noyloxy]-1-octadec-9-enoyloxymethyl-ethyl ester. The
AChE and BChE assays were based on the sub-
strates acetylcholine and butyrylthiocholine, respec-
M.; Muller, G.; Schummer, D.; Seibert, G. J. Antibiot.
¨
2002, 55, 480–494.
13. de Jong, C. J.; Sørensen, L. G.; Tornqvist, H.; Jacobsen,
P. Bioorg. Med. Chem. Lett. 2004, 14, 1741–1744.
4. Slee, D. H.; Bhat, A. S.; Nguyem, T. N.; Kisk, M.;
Lundeen, K.; Newman, M. J.; McConell, S. J. J. Med.
Chem. 2003, 46, 1120–1122.
8
tively.
1
Maximum concentration of inhibitors for determination
of IC50 values was 100 lM with 5-fold dilution steps in
six concentrations. Each data point was determined as
the average of two measurements. IC50 values were cal-
culated using Prism ver. 4.0 by fitting the data to a sig-
moidal dose–response (variable slope) nonlinear
regression analyses.
1
1
5. Yang, W.; Gao, X.; Wang, B. Med. Res. Rev. 2003, 23,
3
46–368.
6. Edwards, J. O.; Sederstrom, R. J. J. Phys. Chem. 1961, 65,
62–863.
8
17. Vedsø, P.; Olesen, P. H.; Hoeg-Jensen, T. Synlett 2004,
892–894.