3-(2-Carboxy-ethyl)-4,6-dichloro-1H-indole-2-carboxylic acid: An allosteric inhibitor of fructose-1,6-bisphosphatase at the AMP site

Article abstract of DOI:10.1016/S0960-894X(03)00310-X

3-(2-Carboxyethyl)-4,6-dichloro-1H-indole-2-carboxylic acid (MDL-29951), an antagonist of the glycine site of the NMDA receptor, has been found to be an allosteric inhibitor of the enzyme fructose 1,6-bisphosphatase. The compound binds at the AMP regulatory site by X-ray crystallography. This represents a new approach to inhibition of fructose 1,6-bisphosphatase and serves as a lead for further drug design.

Full text of DOI:10.1016/S0960-894X(03)00310-X

Bioorganic & Medicinal Chemistry Letters 13 (2003) 2055–2058
3-(2-Carboxy-ethyl)-4,6-dichloro-1H-indole-2-carboxylic Acid:
An Allosteric Inhibitor of Fructose-1,6-bisphosphatase at
the AMP Site
Stephen W. Wright,a,* Anthony A. Carlo,a Dennis E. Danley,a David L. Hageman,a
George A. Karam,a Mahmoud N. Mansour,a Lester D. McClure,a Jayvardhan Pandit,a
Gayle K. Schulte,a Judith L. Treadway,a Ing-Kae Wanga and Paul H. Bauerb
aPfizer Central Research, Eastern Point Road, Box 8220-3141, Groton, CT 06340, USA
bPfizer Discovery Technology Center, 620 Memorial Drive, Cambridge, MA 02139, USA
Received 17 October 2002; accepted 6 March 2003
Abstract—3-(2-Carboxyethyl)-4,6-dichloro-1H-indole-2-carboxylic acid (MDL-29951), an antagonist of the glycine site of the
NMDA receptor, has been found to be an allosteric inhibitor of the enzyme fructose 1,6-bisphosphatase. The compound binds at
the AMP regulatory site by X-ray crystallography. This represents a new approach to inhibition of fructose 1,6-bisphosphatase and
serves as a lead for further drug design.
# 2003 Elsevier Science Ltd. All rights reserved.
Fructose-1,6-bisphosphatase (F16BPase) is an enzyme
expressed predominantly in the liver and kidney, and is
one of the rate limiting enzymes of hepatic gluconeo-
genesis. Liver F16BPase activity is elevated in insulin-
deficient and insulin-resistant animal models of diabetes,
highlighting the importance of this enzyme in the con-
trol of blood glucose.1 The physiologically relevant
form of F16BPase is a homotetramer,2 which exists in
distict conformational states (T and R), depending on
the relative concentrations of active site and allosteric
site ligands. It is subject to competitive substrate inhib-
ition by fructose-2,6-bisphosphate3 and to allosteric
inhibition by adenosine-5-monophosphate (AMP),
which afford the inactive T-state conformation.4 The
kinetically active conformer (R state) is stabilized by
fructose 1,6-bisphosphate. The regulation and mole-
cular basis of F16BPase enzyme activity has been
deduced from crystallographic studies utilizing the
recombinant human fructose-1,6-bisphosphatase pro-
tein.5 A F16BPase inhibitor should reduce hepatic glu-
cose output and lower blood glucose by inhibiting the
elevated rate of gluconeogenesis present in diabetic
patients, and would thus represent a useful therapy for the
treatment of Type 2 diabetes. To this end, the naturally
occurring AMP analogue ZMP (AICA-Riboside
monophosphate)6 as well as a number of synthetic
purine and other heterocyclic phosphonic acids have
been described as potential antidiabetic agents,7 as have
a series of piperazinediones.8
We sought an allosteric, low molecular weight (<500)
inhibitor of F16BPase that was not a phosphonic acid
or phosphate ester. Recently, we reported9 that a series
of anilinoquinazolines, currently of interest as inhibitors
of tyrosine kinases, were identified as allosteric inhibi-
tors of F16Bpase by screening a library of compounds
known to be AMP and/or ATP competitive enzyme
inhibitors against purified recombinant human
F16BPase.10 As a result of subsequent file screening, we
identified 3-(2-carboxyethyl)-4,6-dichloro-1H-indole-2-
carboxylic acid (MDL-29951, 1), an antagonist of the
glycine site of the NMDA receptor,11 as an inhibitor of
F16Bpase (Fig. 1).
*Corresponding author. Tel.: +1-860-441-5831; fax: +1-860-715-
4483; e-mail: stephen_w_wright@groton.pfizer.com
Figure 1. Compound 1.
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00310-X

2056
S. W. Wright et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2055–2058
F16BPase activity was assayed by measuring the in-
organic phosphate hydrolyzed from fructose-1,6-bis-
phosphate by the enzyme. The phosphate released was
quantified spectrophotometrically as a complex with
ammonium molybdate and malachite green.12 The assay
was run under saturating concentrations of substrate
(500 mM) due to the low Km of F16BPase for its sub-
strate and the sensitivity of the phosphate detection
method. Under these conditions, the assay was linear
with time and enzyme concentration, and it was able to
detect inhibition of F16BPase by AMP (IC50=0.8 mM).
MDL-29951 (1) was found to inhibit the human
F16Bpase under these conditions (IC50=2.5 mM).
with Thr31 as N1 of AMP. The pocket occupied by the
inhibitor is deeper than that occupied by AMP. This
appears to occur because Met177 adopts a rotamer dif-
ferent from that found in the AMP bound structure,
which allows the 6-chloro substituent on the indole to
be completely buried.
The indole 2-position carboxyl group makes a hydrogen
bond with a backbone amide nitrogen of Gly28, and
also makes a hydrogen bond to a water molecule dis-
posed between the carboxyl group and the side chain of
Tyr113. The indole 2-position carboxylic acid extends in
the direction of, but does not occupy, the AMP phos-
phate binding region (Fig. 3). The carboxylic acid resi-
due extending from the indole 3-position is in an
extended conformation, pointing out to the solvent
channel normally occupied by the ribose ring of AMP in
the T-state enzyme, and making a hydrogen bond to the
phenolic hydroxyl group of Tyr113.
It was considered unlikely that 1 binds at the F16BPase
active site in as much as enzyme inhibition is observed
at substrate concentrations well above saturation. In
order to better understand the binding mode of our
compounds, we co-crystallized recombinant porcine
kidney F16BPase with 1 in the presence of manganous
chloride, and fructose-6-phosphate.13 The space group
of these crystals at 2.15 angstron resolution was found
to be C2 wth unit cell dimensions a=118.87, b=73.45,
c=78.02 A, b=106.2ꢀ with two monomers of FBPase in
the crystallographic asymmetric unit. The structure was
solved by molecular replacement using the program
AmoRe,14 as implemented in the CCP4 suite of soft-
ware.15 The starting model for molecular replacement
was the structure of T-state F16BP complexed with
AMP and F6P (pdb reference 1FBP).16 Examination of
difference Fourier maps, after rigid body and individual
atom positional refinement of the protein atoms showed
clear density for inhibitor and F6P at the allosteric and
substrate binding sites respectively. Two Mn2+ ions
could also be placed in the electron density. The struc-
ture was refined using REFMAC17 to a final R-factor of
0.208 and free R-factor of 0.283.
Figure 3.
Some derivatives were prepared to examine the effects
of modification at the two carboxylic acid containing
residues. MDL-29951 (1) and analogues were prepared
from 3,5-dichloroaniline by Japp–Klingmann indole
synthesis and subsequent standard functional group
manipulation of the differentially protected intermediate
10 as reported in the literature (Scheme 1).18 All pro-
Our structure reveals that 1 binds at the same site as the
allosteric inhibitor AMP (Fig. 2), and presumably
achieves inhibition by the same mechanism as AMP.
The indole ring of the inhibitor occupies the same shal-
low hydrophobic pocket as the purine ring of AMP, and
the indole nitrogen N1 makes the same hydrogen bond
1
ducts were characterized by H NMR and APCI MS.
Scheme 1. (a) (i) NaNO2, HCl, H2O, 0 ꢀC; (ii) NaOAc, ethyl 2-
oxocyclopentanecarboxylate, 15ꢀC; (iii) Et3N , HO, reflux; (iv) p-
TsOH, toluene, reflux.
2
The carboxypropyl side chain at the indole 3-position
was found to be relatively tolerant of functional groups
that retained a polar hydrogen bonding array at this
position (Table 1, entries 1–6). For example, the amides
2 and 3, the hydroxamic acid 4, and the alcohol 6 all
were similar in potency to 1. In contrast, the carboxyl
group at the indole 2-position was found to be very
restrictive (Table 1, entries 7–11). Only the carboxylic
acid group was tolerated at this position.
Figure 2. X-Ray crystal structure of 1 in porcine kidney F16BPase. 1
is colored by atom type.

S. W. Wright et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2055–2058
2057
Table 1. Effect of carboxyl modification on F16BPase inhibition Table 2. Effect of linker length modification on F16BPase inhibition
Compd
n
F16BPase
IC50, mMa
Compd
R1
R2
F16BPase
IC50, mMa
13
1
14
15
1
2
3
4
ꢁ10
2.5
1.0
1
2
3
4
5
6
7
8
CO2H
CONH2
CONMe2
CONHOH
CO2Me
CH2OH
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CONH2
CONMe2
CONHOH
CO2Et
2.5
2.2
1.7
1.5
7.0
2.7
1.2
aValues are means of three experiments. An IC50 of ꢁ10 indicates that
no curve was noted in the dose response up to 10 mM.
ꢁ10
ꢁ10
ꢁ10
ꢁ10
ꢁ10
9
10
11
Acknowledgements
CH2OH
aValues are means of three experiments. An IC50 of ꢁ10 indicates that
The authors wish to thank Dr. Walter C. Soeller for
expression of the recombinant human, rabbit, and rat
F16Bpase enzymes. The authors wish to thank Pro-
fessor Evan R. Kantrowitz (Boston College) for a sup-
ply of porcine kidney F16BPase. The authors wish to
thank Professor M. Raafat El-Maghrabi (SUNY Stony
Brook) for kind provision of the human and rat
F16BPase cDNAs.
no curve was noted in the dose response up to 10 mM.
Methylation of the indole nitrogen to furnish 12 also
eliminated F16BPase inhibitory activity (IC50ꢁ10 mM)
(Fig. 4).
References and Notes
1. (a) Insulin resistant model: Sugiyama, Y.; Shimura, Y.;
Ikeda, H. Endocrinologia Japonica 1989, 36, 65. (b) Insulin
deficient model: Wimhurst, J. M.; Manchester, K. L. Biochem.
J. 1970, 120, 95.
2. (a) For a review, see: Veneziale, C. M. In The Regulation of
Carbohydrate Formation and Utilization in Mammals; Vene-
ziale, C. M., Ed.; University Park Press: Baltimore, 1981 (b)
Pilkis, S. J.; Granner, D. K. Ann. Rev. Physiol. 1992, 54, 885.
3. (a) Van Schaftingen, E.; Hers, H. G. Proc. Natl. Acad. Sci.
U.S.A. 1981, 78, 2862. (b) Pilkis, S. J.; El-Maghrabi, M. R.;
Pilkis, J.; Claus, T. H. J. Biol. Chem. 1981, 256, 3619. (c)
Pilkis, S. J.; El-Maghrabi, M. R.; Cluas, T. H. Annu. Rev.
Biochem. 1988, 57, 755.
4. (a) Taketa, K.; Pogell, B. M. J. Biol. Chem. 1965, 240, 651.
(b) Ke, H.; Liang, J. Y.; Zhang, Y.; Lipscomb, W. N. Bio-
chemistry 1991, 30, 4412. (c) Zhang, Y.; Liang, J.; Huang, S.;
Lipscomb, W. N. J. Mol. Biol. 1994, 244, 609.
5. Gidh-Jain, M.; Zhang, Y.; van Poelje, P. D.; Liang, J.;
Huang, S.; Kim, J.; Elliott, J. T.; Erion, M. D.; Pilkis, S. J.; El-
Maghrabi, M. R.; Lipscomb, W. N. J. Biol. Chem. 1994, 269,
27732.
Figure 4. Compound 12.
The length of the alkyl chain extending from the indole
3-position was modified to afford both chain-shortened
and chain-elongated analogues of 1 (Table 2). One
methylene unit (13) was not sufficient to retain activity,
whereas longer chains (14 and 15) had inhibitory prop-
erties similar to those of 1.
In general, the sequences of F16BPase are conserved
between mammalian species (e.g., identity of human
and rabbit=91%). A comparison of the enzyme inhibi-
tory potency of 1 against some F16BPase isozymes
revealed that 1 inhibited the human liver (IC50=2.5
mM), porcine kidney19 (IC50=1.0 mM), and rabbit
liver20 (IC50=0.21 mM) isoforms of the enzyme, but was
significantly less potent against the rat liver21 isoform
(IC50=11 mM).
6. Gruber, H. E.; Tuttle, R. R.; Browne, C. E.; Ugarkar, B.
G.; Reich, J. W.; Metzner, E. K.; Marangos, P. J. U.S. Patent
5,658,889, 1997. Chem. Abstr. CAN114:214418 (1998).
7. (a) Dang, Q.; Erion, M. D.; Reddy, M. R.; Scarlato, G. R.;
Kasibhatla, S. R.; Reddy, K. R. WO 98/39343 A1, 1998.
Chem. Abstr. CAN129:245147 (1998). (b) Dang, Q.; Erion,
M. D.; Reddy, M. R.; Robinson, E. D.; Kasibhatla, S. R.;
Reddy, K. R. WO 98/39344 A1, 1998. Chem. Abstr. CAN
131:243287 (1998). (c) Dang, Q.; Erion, M. D.; Reddy, M. R.;
Scarlato, G. R.; Kasibhatla, S. R.; Reddy, K. R. WO 98/39345
A1, 1998. Chem. Abstr. CAN129:245033 (1998).
3-(2-Carboxyethyl)-4,6-dichloro-1H-indole-2-carboxylic
acid (1) represents the first low molecular weight (<500)
AMP site allosteric inhibitor of fructose 1,6-bisphos-
phatase that is not a phosphorus-based AMP mimic.
Given that 1 is a small molecule with known drug-like
properties, it represents a new approach to inhibition of
F16BPase. Enzyme inhibition is achieved by binding the
AMP binding site, with a similar hydrogen bond net-
work but deeper hydrophobic pocket than found in the
AMP bound structure. These compounds represent
potential leads for improving potency and selectivity for
fructose 1,6-bisphosphatase inhibition.
8. Mjalli, A. M. M.; Mason, J. C.; Arienti, K. L.; Short, K.
M.; Kimmich, R. D. A.; Jones, T. K. WO 99/47549 A1, 1999.
Chem. Abstr. CAN131:243287 (1999).

2058
S. W. Wright et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2055–2058
9. Wright, S. W.; Hageman, D. L.; McClure, L. D.; Carlo,
A. A.; Treadway, J. L.; Mathiowetz, A. M.; Withka, J. M.;
Bauer, P. H. Bioorg. Med. Chem. Lett. 2001, 11, 17.
10. El-Maghrabi, M. R.; Gidh-Jain, M.; Austin, L. R.; Pilkis,
S. J. J. Biol. Chem. 1993, 268, 9466.
11. Salituro, F. G.; Harrison, B. L.; Baron, B. M.; Nyce, P. L.;
Stewart, K. T.; McDonald, I. A. J. Med. Chem. 1990, 33,
2946.
maximum size of 0.075Â0.075Â0.2 mm. This crystal structure
can be accessed on the web at: http://www.rcsb.org/pdb/cgi/
explore.cgi?pid=14071046187361&pdbId=1LEV.
14. Bailey, S. Acta Crystallogr. 1994, D50, 760.
15. Navaza, J. Acta Crystallogr. 1994, A50, 157.
16. Ke, H.; Zhang, Y.; Lipscomb, W. N. Proc. Natl. Acad.
Sci. U.S.A. 1990, 87, 5243.
17. Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Acta
Crystallogr. 1997, D53, 240.
12. Baykov, A. A.; Evtushenko, O. A.; Avaeva, S. M. Anal.
Biochem. 1988, 171, 266.
18. (a) Salituro, F. G.; Harrison, B. L.; Baron, B. M.; Nyce,
P. L.; Stewart, K. T.; Kehne, J. H.; White, H. S.; McDonald,
I. A. J. Med. Chem. 1992, 35, 1791. (b) Rodriguez, A.; Nomen,
M.; Spur, B. W. Tetrahedron Lett. 1998, 39, 8563.
19. Williams, M. K.; Kantrowitz, E. R. Proc. Natl. Acad. Sci.
U.S.A. 1992, 89, 3080.
20. (a) Commercially available (Sigma) rabbit liver F16Bpase
(<10% pure) was purified to 95+% purity, see: Pontremoli,
S.; Traniello, S.; Luppis, B.; Wood, W. A. J. Biol. Chem. 1965,
240, 3459. (b) Pontremoli, S.; Grazi, E.; Accorsi, A. Biochem-
istry 1968, 7, 3628.
13. Crystallization: Porcine kidney fructose-1,6-bis- phospha-
tase used for crystallization as is or in some cases was further
purified by cation exchange chromatography on a Mono S
column (Pharmacia-Amersham). The protein was con-
centrated in 8 mM potassium phosphate buffer, pH 7.2, 1 mM
EDTA, 1 mM ß-mercaptoethanol, 2.0 mM MnCl2, and 1 mM
1 to 15 mg/mL. Fructose6-phosphate was added to a final
concentration of 2 mM and then incubated for 60 min at room
temperature prior to crystallization trials (Hanging drop vapor
diffusion method). The drops consisted of 2 mL of protein
mixed with 2 mL of 100 mM HEPES, pH 7.0, 200 mM sodium
acetate, 13.5% PEG 4000 and suspended over 0.6 mL of the
same buffer. Crystals grew within one week and reached a
21. El-Maghrabi, M. R.; Pilkis, J.; Marker, A.; Colosia, A. D.;
D’Angelo, G.; Fraser, B. A.; Pilkis, S. J. Proc. Natl. Acad. Sci.
U.S.A. 1988, 85, 8430.