Benzoxazole benzenesulfonamides as allosteric inhibitors of fructose-1,6-bisphosphatase

Article abstract of DOI:10.1016/j.bmcl.2006.01.014

A series of novel benzoxazole benzenesulfonamides was synthesized as inhibitors of fructose-1,6-bisphosphatase (FBPase-1). Extensive SAR studies led to a potent inhibitor, 53, with an IC₅₀ of 0.57 μM. Compound 17 exhibited excellent bioavailabi

Full text of DOI:10.1016/j.bmcl.2006.01.014

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1807–1810
Benzoxazole benzenesulfonamides as allosteric inhibitors
of fructose-1,6-bisphosphatase
Chunqiu Lai,a,* Rebecca J. Gum,a Melissa Daly,a Elizabeth H. Fry,b Charles Hutchins,b
Celerino Abad-Zapaterob and Thomas W. von Gelderna,*
aMetabolic Disease Research, GPRD, Abbott Laboratories, 100 Abbott Park Road, Abbott Park, IL 60064, USA
bStructural Biology Departments, GPRD, Abbott Laboratories, 100 Abbott Park Road, Abbott Park, IL 60064, USA
Received 22 November 2005; revised 4 January 2006; accepted 5 January 2006
Available online 30 January 2006
Abstract—A series of novel benzoxazole benzenesulfonamides was synthesized as inhibitors of fructose-1,6-bisphosphatase
(FBPase-1). Extensive SAR studies led to a potent inhibitor, 53, with an IC50 of 0.57 lM. Compound 17 exhibited excellent bioavail-
ability and a good pharmacokinetic profile in rats.
Ó 2006 Elsevier Ltd. All rights reserved.
In the previous paper,1 we identified benzoxazole ben-
zenesulfonamide 1 as a novel inhibitor of human fruc-
tose-1,6-bisphosphatase (hFBPase-1) with an IC50 of
3.4 lM in a colorimetric malachite green assay. X-ray
crystallographic studies indicate that 1 has a unique
binding mode, partially occupying the AMP allosteric
regulatory site. In this communication, we will report
the synthesis and structure–activity profile of benzoxaz-
ole benzenesulfonamide analogs 2, leading to an
improvement in potency.
O
O
N
O
O
O
a
N
S
R1
+
S
NH2
R2
R1
N
H
Cl
R2
O
4
3
2
Scheme 1. Reagents and condition: (a) pyridine, CH2Cl2, microwave
130 °C.
A rapid systematic variation of substituents on the
phenyl ring was accomplished in a parallel synthesis
fashion through the key intermediate 5. Compound 6
was prepared by Suzuki coupling reaction;3 7, 8, and 9
were prepared via corresponding copper assisted nucle-
ophilic substitution reactions4 (Scheme 2).
Cl
O
O
O
O
N
S
Cl
S
N
O
R2
R1
N
H
NH
O
Cl
1
2
The introduction of substituents to the benzene ring was
performed as Scheme 3. Treating 10 with either cyano-
gen bromide5 or di(imidazole-1-yl)methanimine6 affor-
ded substituted benzoxazol-2-ylamine 11, which, in
turn, reacted with 2,5-dichlorobenzenesulfonyl chloride
to form intermediate 12. Coupling 12 with aryl boronic
acid via Suzuki reaction gave rise to the desired com-
pound 13.
Benzoxazole benzenesulfonamide analogs 2 were pre-
pared by treating substituted benzoxazol-2-ylamines 3
with sulfonyl chlorides 4 in the presence of pyridine as
shown in Scheme 1.2 Both 3 and 4 were either commer-
cially available or prepared according to the literature
procedures.
The inhibitory activity of synthesized molecules was
evaluated by a colorimetric malachite green hFBPase-
1 assay as described in our previous article.1 The effects
of replacement of the benzenesulfonamide on FBPase-1
inhibitory activity are summarized in Table 1. The aro-
matic character of this group appears to be crucial.
Keywords: Fructose-1,6-bisphosphatase; Enzyme inhibitor; Allosteric
regulator; Benzoxazole benzenesulfonamide; Structure–activity
studies; X-ray crystallography.
*
Corresponding authors. Tel.: +1 847 937 9171; fax: +1 847 938 1674
(T.v.G.); e-mail addresses: Chunqiu.lai@abbott.com; thomas.vongeldern
@abbott.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.01.014

1808
C. Lai et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1807–1810
Table 1. Inhibitory activity of selected FBPase-1 inhibitors
O
S
O
S
NH
O
O
Cl
Cl
N
O
N
O
NR1COR2
O
S
NH
Ar
N
H
O
Cl
R
N
O
a
6
7
b
O
S
O
Cl
N
O
Compound
R
FBPase-1 IC50 (lM)
Br
N
H
Cl
5
d
c
1
3.4
Cl
O
S
O
S
NH
O
O
Cl
Cl
N
N
OR3
N Het
N
H
14
15
16
17
18
7.7
13
O
O
9
8
S
Scheme 2. Reagents and conditions: (a) ArB(OH)2, PdCl2(PPh3)2,
Na2CO3, DMF, microwave 150 °C; (b) R2CONHR,1 CuI, 1,10-
phenanthroline, Cs2CO3, ODCB, microwave 200–220 °C; (c) R3OH,
CuI, 1,10-phenanthroline, Cs2CO3, tol, microwave 160–170 °C; (d) N-
H heteroarene, CuI, 1,10-phenanthroline, K2CO3, NMP, 190 °C.
N
9.4
2.5
3.4
N
N
O
NH2
OH
a
b
X
X
NH2
X = Br, I
X = Br, I
10
11
Cl
Cl
19
20
Me
>50
>50
O
O
O
O
n-Bu
S
N
O
c
S
N
O
Ar
NH
X
NH
Cl
Cl
X = Br, I
13
Table 2. Inhibitory activity of selected FBPase-1 inhibitors
12
6
5
O
S
Scheme 3. Reagents and conditions: (a) BrCN (3 M in CH2Cl2) or
di(imidazole-1-yl)methanimine, MeOH, rt; (b) 2,5-dichlorobenzenesulfonyl
chloride, pyridine, CH2Cl2, microwave 130 °C; (c) ArB(OH)2,
PdCl2(PPh3)2, Na2CO3, DMF, microwave 150 °C.
O
4
R
Cl
N
O
N
H
3
2
Compound
R
FBPase-1 IC50 (lM)
14
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
H
7.7
6.5
6.0
3.8
23
Heteroaromatic (15, 16) and fused aromatic rings (17,
18) are well tolerated. Replacement of the phenyl ring
with alkyl groups (19, 20) dramatically diminishes the
activity.
2-Br
2-CN
2-Ph
2-(Imidazol-1-yl)
3-Cl
3-NO2
1.3
1.8
9.8
>50
2.8
7.8
6.2
24
Further SAR studies focused on the effects of substitu-
ents of the phenyl ring (Table 2). Several substituents
provide some improvement in potency, for instance, 3-
Cl (25, IC50 = 1.3 lM) and 3-NO2 (26, IC50 = 1.8 lM),
represent a 2–3-fold boost in potency relative to initial
lead 1. In general, mono-substitution at the ortho-, meta-
or para-position is desirable, but bulky groups such as t-
Bu (32, IC50 = 24 lM) and 4-t-Bu-benzyloxy (28,
IC50 > 50 lM) lead to reduced activity; ortho-, ortho-
di-substitution (36, 37) is strongly disfavored.
3-Ph
3-(4-t-Bu-benzyloxy)
4-F
4-OCF3
4-Me
4-t-Bu
4-Ph
4.0
6.7
12
4-(3-Furanyl)
4-NHCOCH2CH2CH3
2,6-Di-Cl
>50
>50
2-Cl, 6-Me
Manipulation of substituents on the benzoxazole ring
also led to a series of new compounds (Table 3). The
4-position is the most restrictive site on the molecule;
even small substituents such as iodo and methyl lead
to a substantial loss of potency. This result is consistent
with the previously reported crystallographic binding
mode,1 which indicates that 4-substituents are directed
toward the interior of the FBPase tetramer. In contrast,
the 7-position is directed toward the outside of the
FPBase tetramer, allowing for a broader range of
substituents, preferably hydrophilic. Small substituents
are tolerated at the 5-position and are acceptable at
the 6-position. A combination of 5-MeO and various
7-substituents significantly boosts the potency. Of note
is 53, in which a combination of 5-MeO with 7-(3-ami-
nophenyl) gives
(IC50 = 0.57 lM) versus 1. Compound 53 also shows im-
a
6-fold boost in potency

C. Lai et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1807–1810
1809
Table 3. Inhibitory activity of selected FBPase-1 inhibitors
100
10
Cl
O
4
7
O
5
S
N
O
R
NH
Cl
6
1
Mean IV
Compound
R
FBPase-1 IC50 (lM)
Mean Oral
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
4-I
4-Me
>50
55
0.1
0.01
4-(4-Pyridyl)
5-Br
5-Me
>50
1.9
6.4
32
0
2
4
6
8
10
12
Hours after Dose
5-t-Bu
5-(3-Furanyl)
6-Cl
6-Me
4.0
8.1
10
Figure 2. Rat pharmacokinetics of 17.
regulatory site and the benzoxazole moiety protrudes
into the space between subunits of the FBPase homo-
tetramer. In addition to the AMP-site interactions, the
amino group of the C7-substituent is exposed and makes
two additional hydrogen-bonding interactions with the
C@O of Gly26 and Thr27. Other analogs with 7-substit-
uents capable of H-bonding (e.g., 48, 49) interact simi-
larly with the protein backbone (data not shown) and
demonstrate improved potency, suggesting that the lat-
ter may be the result of these H-bonding interactions.
6-MeO
5-MeO, 7-(4-MeO-3-Pyridyl) 3.5
8.0
5-MeO, 7-(3-HO-Ph)
5-MeO, 7-(4-HO-Ph)
5-MeO, 7-(3-NH2CH2-Ph)
5-MeO, 7-(4-NH2-Ph)
5-MeO, 7-(3-NH2-Ph)
1.7
1.8
2.6
1.3
0.57
Compound 17 was selected as a representative to evalu-
ate the pharmacokinetic properties of these benzoxazole
benzenesulfonamides in rats. At 5 mg/kg dose, com-
pound 17 was shown to have an excellent oral bioavail-
ability (>100%) with a long oral half-life (7.5 h) and a
low plasma clearance (0.04 L/h kg) (Fig. 2).
In conclusion, we have synthesized a series of novel
benzoxazole benzenesulfonamides as inhibitors of
FBPase-1. Optimization of screening hit 1 led to
compound 53 (IC50 = 0.57 lM) with a 6-fold boost in
potency. X-ray crystal structures revealed that the
appropriately selected 7-substituent of 53 forms two
additional hydrogen bonding interactions with the en-
zyme, leading to the enhanced affinity. These results sug-
gest opportunities to design even more potent FBPase-1
inhibitors. Compound 17 with an IC50 of 2.5 lM exhib-
ited an excellent pharmacokinetic profile in rats. Further
optimization studies as well as the pharmacological
characterization of leading compounds will be reported
in due course.
Figure 1. Binding mode of 53 in the FBPase tetramer. The view is
approximately along one of the tetramer dyads and illustrates the
binding mode of 53 in the periphery of the FBPase tetramer. Each
molecule of inhibitor interacts with the main chain C@O of residues
Gly26 and Thr27 across the dyad. Thick dotted lines (left) highlight the
hydrogen-bonding interactions of the anilino group of molecule A of
53 with the main chain carbonyl oxygens of a second molecule of
FBPase trans- across the twofold symmetry axis (white symbols and
gray atoms). Thin dotted lines indicate the corresponding interaction
for molecule B of 53 with the opposite monomer of FBPase (yellow
symbols and stick bonds). There are four molecules of 53 per tetramer;
References and notes
˚
H-bond distances range from ꢀ2.7 to 3.6 A. Aromatic stacking of the
1. von Geldern, T. W.; Lai, C.; Gum, R. J.; Daly, M.; Sun, C.;
Fry, E. H.; Abad-Zapatero, C. Bioorg. Med. Chem. Lett.
2006, in press, doi:10.1016/j.bmcl/2006.01.015.
aniline groups across the symmetry axis is apparent.
proved potency in a ‘coupled’ assay format,1 with an
IC50 = 0.73 lM and Ki = 0.22 lM.
2. Kurahashi, Y.; Kume, T.; Yanagi, A.; Honda, M.; Arubu-
rehito, M. JP Patent 02,001,483, 1990; Chem. Abstr. 1990,
113, 19447.
3. Suzuki, A. Pure. Appl. Chem. 1991, 63, 419.
The X-ray crystal structure of 537 confirms the previous-
ly observed binding mode for this series of inhibitors
and is consistent with the SAR concepts outlined
(Fig. 1). Similar to compound 1,1 the arylsulfonamide
portion of 53 interacts with a part of the AMP allosteric
4. (a) Freeman, H. S.; Butler, J. R.; Freedman, L. D. J. Org.
Chem. 1978, 43, 4975; (b) Bacon, R. G. R.; Rennison, S. G.
J. Chem. Soc. (c) 1969, 312; (c) Wu, Y.-J.; He, H.;
L’Heureux, A. Tetrahedron Lett. 2003, 44, 4217.
5. Plampin, J. N.; Cain, C. K. J. Med. Chem. 1963, 6, 247.

1810
C. Lai et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1807–1810
6. Wu, Y.-Q.; Limburg, D. C.; Wilkinson, D. E.; Hamilton,
G. S. J. Heterocycl. Chem. 2003, 40, 191.
53 were orthorhombic (P212121) with cell constants
˚
a = 84.4, b = 108.7, c = 196.4 A and diffracted weakly only
˚
to approximately 3.5 A, using synchrotron radiation
7. Co-crystals of human FBPase-1 with 48 (and 53) were
grown by protocols similar to the ones used in Ref. 1 but
displacing the seeding compound ZMP with the relevant
ligand. The crystals with 48 were orthorhombic (P212121)
(Advanced Photon Source, APS, IMCA-CAT, 17-ID).
The crystals contained one tetramer of FBPase and four
molecules of 53 in the asymmetric unit. The structures of
FBPase with 48 and 53 were solved by molecular replace-
ment using one FBPase tetramer and refined using standard
methods. Crystallographic coordinates of FBPase com-
plexed with 48 and 53 have been deposited at the Protein
Data Bank with accession codes 2fie, 2fix.
˚
with cell constants a = 67.4, b = 83.3, c = 277.5 A and
˚
diffracted to 2.8 A resolution, using synchrotron radiation
(Advanced Photon Source, APS, IMCA-CAT, 17-ID). The
crystals contained one tetramer of FBPase and four
molecules of 48 in the asymmetric unit. The crystals with