Crystallographic studies on two bioisosteric analogues, N-acetyl-β-D-glucopyranosylamine and N-trifluoroacetyl-β-D- glucopyranosylamine, potent inhibitors of muscle glycogen phosphorylase

Article abstract of DOI:10.1016/j.bmc.2005.08.010

Structure-based inhibitor design has led to the discovery of a number of potent inhibitors of glycogen phosphorylase b (GPb), N-acyl derivatives of β-d-glucopyranosylamine, that bind at the catalytic site of the enzyme. The first good inhibitor in this class of compounds, N-acetyl-β-d- glucopyranosylamine (NAG) (K_i = 32 μM), has been previously characterized by biochemical, biological and crystallographic experiments at 2.3 A resolution. Bioisosteric replacement of the acetyl group by trifluoroacetyl group resulted in an inhibitor, N-trifluoroacetyl-β-d- glucopyranosylamine (NFAG), with a K_i = 75 μM. To elucidate the structural basis of its reduced potency, we determined the ligand structure in complex with GPb at 1.8 A resolution. To compare the binding mode of N-trifluoroacetyl derivative with that of the lead molecule, we also determined the structure of GPb-NAG complex at a higher resolution (1.9 A). NFAG can be accommodated in the catalytic site of T-state GPb at approximately the same position as that of NAG and stabilize the T-state conformation of the 280s loop by making several favourable contacts to Asn284 of this loop. The difference observed in the K_i values of the two analogues can be interpreted in terms of subtle conformational changes of protein residues and shifts of water molecules in the vicinity of the catalytic site, variations in van der Waals interaction, and desolvation effects.

Full text of DOI:10.1016/j.bmc.2005.08.010

Bioorganic & Medicinal Chemistry 14 (2006) 181–189
Crystallographic studies on two bioisosteric analogues,
N-acetyl-b-D-glucopyranosylamine and
N-trifluoroacetyl-b-D-glucopyranosylamine,
potent inhibitors of muscle glycogen phosphorylase
Eleni Anagnostou,a, Magda N. Kosmopoulou,a Evangelia D. Chrysina,a
Demetres D. Leonidas,a Theodoros Hadjiloi,a Costantinos Tiraidis,a
a
Spyros E. Zographos, Zoltan Gyo¨rgydeak, Laszlo Somsak, Tibor Docsa,
c
c
d
´
´
´
´
´
Pal Gergely, Fragiskos N. Kolisis and Nikos G. Oikonomakos
d
e
a,b,
*
´
aInstitute of Organic and Pharmaceutical Chemistry, The National Hellenic Research Foundation, 48,
Vas. Constantinou Ave. 116 35 Athens, Greece
bInstitute of Biological Research and Biotechnology, The National Hellenic Research Foundation, 48,
Vas. Constantinou Ave. 116 35 Athens, Greece
cDepartment of Organic Chemistry, Research Centre for Molecular Medicine, University of Debrecen, Hungary
dDepartment of Medical Chemistry, Research Centre for Molecular Medicine, University of Debrecen, Hungary
eBiotechnology Laboratory, School of Chemical Engineering, The National Technical University of Athens, Greece
Received 30 March 2005; accepted 2 August 2005
Available online 4 October 2005
Abstract—Structure-based inhibitor design has led to the discovery of a number of potent inhibitors of glycogen phosphorylase b
(GPb), N-acyl derivatives of b-D-glucopyranosylamine, that bind at the catalytic site of the enzyme. The first good inhibitor in this
class of compounds, N-acetyl-b-D-glucopyranosylamine (NAG) (Ki = 32 lM), has been previously characterized by biochemical,
˚
biological and crystallographic experiments at 2.3 A resolution. Bioisosteric replacement of the acetyl group by trifluoroacetyl group
resulted in an inhibitor, N-trifluoroacetyl-b-D-glucopyranosylamine (NFAG), with a Ki = 75 lM. To elucidate the structural basis of
˚
its reduced potency, we determined the ligand structure in complex with GPb at 1.8 A resolution. To compare the binding mode of
N-trifluoroacetyl derivative with that of the lead molecule, we also determined the structure of GPb–NAG complex at a higher res-
˚
olution (1.9 A). NFAG can be accommodated in the catalytic site of T-state GPb at approximately the same position as that of
NAG and stabilize the T-state conformation of the 280s loop by making several favourable contacts to Asn284 of this loop. The
difference observed in the Ki values of the two analogues can be interpreted in terms of subtle conformational changes of protein
residues and shifts of water molecules in the vicinity of the catalytic site, variations in van der Waals interaction, and desolvation
effects.
Ó 2005 Elsevier Ltd. All rights reserved.
1. Introduction
developing inhibitors of glycogen phosphorylase (GP),
an enzyme proposed to be exploited as a therapeutic
molecular target.1–5 Several regulatory binding sites of
GP have been identified and are currently under investi-
gation for the design and synthesis of potent inhibitors
of the enzyme.3,6–13 More specifically, the catalytic site
has been probed with glucose analogue inhibitors, de-
signed on the basis of information derived from the crys-
tal structure of the inactive T-state GPb–a-D-glucose
complex.14–26 A common feature of these compounds
is that upon binding at the catalytic site, they promote
the (less active) T-state conformation of the enzyme
Efforts towards improving glycaemic control in type 2
diabetes mellitus have been lately directed towards
Keywords: Type 2 diabetes; Glycogen phosphorylase; N-acetyl-b-D-
glucopyranosylamine; N-trifluoroacetyl-b-D-glucopyranosylamine;
Bioisosteric inhibition; X-ray crystallography.
*
Corresponding author. Tel.: +30 210 7273761; fax: +30 210
7273831; e-mail: ngo@eie.gr
Present address: Institute of Marine and Coastal Science, RUT-
GERS, New Brunswick, NJ, USA.
0968-0896/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.08.010

182
E. Anagnostou et al. / Bioorg. Med. Chem. 14 (2006) 181–189
O6
OH
C6
C5
cal studies showed that NAG was indeed an effective
inhibitor of liver GPa in rat hepatocytes and that it
was able to enhance the dephosphorylation (inactiva-
tion) of GPa,27–29 indicating a potential hypoglycaemic
action in the treatment of type 2 diabetes. Information
available from the crystal structure of T-state GPb–
O5
O
C4
O4
N1
NH
C1
C8
CH3
HO
C7
HO
C2
O3
C3
OH
O2
O
˚
NAG–IMP complex at 2.3 A resolution has shown
O7
that NAG fits neatly into the so-called b-pocket, a side
channel from the catalytic site, lined by both polar and
nonpolar groups, but with no access to the bulk sol-
vent,14 and binding is stabilized through a strong
hydrogen bond formed between the amide nitrogen
and the main chain carbonyl O of His377.17 Following
crystallographic analysis of the NAG complex, model-
ling of other compounds was performed, torsion angles
were adjusted to allow favourable interactions with the
protein, and a large number of derivatives were synthe-
sized with the objective of identifying more potent
inhibitors.18 The compounds synthesized differed only
in their substituent atoms at the b-configuration at
the C1 position of the glucopyranose ring, but none
of the NAG analogues studied resulted in improved
Ki values compared with the lead molecule.18 In a re-
cent study,24 it has been demonstrated that a substitu-
ent with a longer chain and larger, more hydrophobic
group facilitates stronger binding (Ki = 3.5 lM for N-
(2-naphthylacryloyl)-b-D-glucopyranosylamine). A bet-
ter understanding of the mechanism of inhibition of
GP through identification of the structural determi-
nants contributing to inhibitor interactions at the b-
pocket should aid in the design of improved inhibitors
against GP.
NAG
O6
OH
C6
C5
F1
F
O5
O
F3
F
C4
O4
HO
N1
C8
NH
C1
C
C7
HO
C2 OH
O3
C3
F
O2
O
F2
O7
NFAG
Scheme 1. Chemical structures of N-acetyl-b-D-glucopyranosylamine
(NAG) and N-trifluoroacetyl-b-D-glucopyranosylamine (NFAG),
showing the numbering system used.
through stabilization of the closed position of 280s loop
(residues 282–287), which blocks access of the substrate
to the catalytic site.
One of the early successes was the design and synthesis
of N-acetyl-b-D-glucopyranosylamine (NAG, Scheme
1), which exhibited a potency to inhibit GP activity
(Ki = 32 lM), approximately 50 times better than that
of a-D-glucose (Ki = 1.7 mM).17 Extensive physiologi-
0,12
80
60
0,1
40
0,08
0,06
0,04
0,02
0
20
0
0
2
4
6
8 10 12 14 16 18 20
[Glucose 1-phosphate], mM
v
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
1/[Glucose 1-phosphate],mM-1
Figure 1. Kinetics of NFAG inhibition of GPb with respect to Glc-1-P (3, 4, 6, 10, 20 mM) in the direction of the glycogen synthesis (30 °C, pH 6.8).
Double reciprocal plots of initial reaction velocity versus [Glc-1-P] at constant concentrations of AMP (1 mM) and glycogen (1%) and various
concentrations of NFAG. Inhibitor concentrations (mM) were as follows: 0 (s), 0.044 (d), 0.089 (h) and 0.178 mM (j). Best-fit lines were computer
generated according to the equation for competitive inhibition43 by fitting all of the data at once. The kinetic parameters determined by this method
are as follows: Km = 1.8 0.1 mM, k = 87.9 0.9 U/mg and Ki = 0.075 0.004 mM. Inset (top left): Fit of the calculated lines to the initial rate
measurements.

E. Anagnostou et al. / Bioorg. Med. Chem. 14 (2006) 181–189
183
We report here on the kinetic and crystallographic
experiments of N-trifluoroacetyl-b-D-glucopyranosyl-
amine (NFAG) with GPb. The change from –CH3 in
NAG to a –CF3 in NFAG was found to result in a de-
crease in inhibitory action (Ki = 75 lM) towards GPb.
To obtain information on the mechanism of NFAG
inhibition of the enzyme, we determined the crystal
Glc-1-P, at constant concentrations of glycogen (1% w/
v) and AMP (1 mM). The inhibition constant of NFAG
for muscle GPb, assayed into the direction of glycogen
synthesis, is Ki = 75 4 lM, approximately 2.3 times
larger than NAG, which points to a difference in the
binding energy between NFAG and NAG of about
0.51 kcal/mol (in a previous report,20 a Ki value of
710 lM was erroneously given for the NFAG inhibition
of GPb). To elucidate the structural basis of this differ-
ence in affinity, we have determined the crystal struc-
tures of GPb in complex with both NAG and NFAG.
˚
structure of GPb complexed with NFAG at 1.8 A reso-
lution. The crystallographic results show that on binding
of NFAG to the enzyme there were subtle changes in the
vicinity of the catalytic site that can provide a rationale
for the kinetic properties of the inhibitor. A new crystal-
lographic data set of GPb–NAG complex was also
A summary of the crystallographic data processing and
refinement statistics is given in Table 1. A schematic
representation of the native T-state structure of GPb
with the distinct catalytic, allosteric effector and the
new allosteric inhibitor binding sites is shown in Figure
2. Electron density maps defined the position of each
inhibitor within the catalytic site (Fig. 3A and B), con-
sistent with the kinetic results. Furthermore, there was
clear additional density contributed by the fluorine
atoms. We describe briefly the GPb/NAG interactions
and in more detail the GPb/NFAG interactions at
the catalytic site.
˚
collected at 1.9 A resolution for a more accurate com-
parison with the GPb–NFAG complex.
2. Results
We have shown previously that NAG is an effective
competitive inhibitor of GPb, with a Ki value of
32 ( 1) lM.30 The inhibitory efficiency of NFAG was
also tested with GPb (Fig. 1). The compound displayed
competitive inhibition with respect to the substrate
Table 1. Diffraction data and refinement statistics for GPb–NAG and GPb–NFAG complexes
GPb–NAG
GPb–NFAG
Experiment
Space group
10 mM NAG
P43212
74 (0.7)a
a = b = 128.6, c = 116.1
100 mM NFAG
P43212
90 (0.5)a
a = b = 128.8, c = 116.2
No. of images (°)
Unit cell dimensions (A)
˚
˚
Resolution range (A)
No. of observations
29.36–1.90
507267
28.07–1.80
568131
No. of unique reflections
hI/r(I)i (outermost shell)b
Completeness (outermost shell) (%)
Rm (outermost shell)c
68661
6.6 (2.4)
89632
11.9 (1.9)
89.2 (89.5)
0.097 (0.445)
99.0 (97.5)
0.041 (0.420)
˚
Outermost shell (A)
Multiplicity (outermost shell)
1.93–1.90
4.1 (3.6)
1.83–1.80
3.7 (3.6)
˚
Refinement (resolution) (A)
29.36–1.90
65135 (3495)
28.07–1.80
85072 (4511)
No. of reflections used (free)
Residues included
(12–251), (261–341), (324–837)
6585
(12–252), (261–314), (324–837)
6596
No. of protein atoms
No. of water molecules
No. of ligand atoms
356
15 (PLP), 15 (NAG)
402
15 (PLP), 18 (NFAG)
d
Final R (Rfree
R (Rfree) (outermost shell) (%)
)
(%)
18.8 (21.1)
25.6 (27.3)
19.0 (21.5)
27.6 (29.9)
˚
Outermost shell in refinement (A)
˚
r.m.s.d. in bond lengths (A)
2.02–1.90
0.006
1.91–1.80
0.006
r.m.s.d. in bond angles (°)
r.m.s.d. in dihedrals (°)
r.m.s.d. in impropers (°)
1.3
22.1
1.3
22.0
0.93
(12–251), (261–314), (324–837)
0.77
(12–252), (261–314), (324–837)
2
˚
Average B (A ) for residues
Overall
32.4
30.3
32.2
30.1
CA,C,N,O
Side chain
Average B (A ) for ligands
34.4
19.9 (PLP), 20.2 (NAG)
41.6
34.3
19.8 (PLP), 21.8 (NFAG)
43.0
b
˚
b
˚
Average B (A ) for water molecules
a 0.5/0.7 is the rotation range per image.
b r(I) is the standard deviation of I.
P P
P P
c Rmerge
=
|hIhi ꢀ Iih |/
Iih, where hIhi and Iih are the mean and ith measurement of intensity for reflection h, respectively.
i
h
i
h
P
P
d Crystallographic R =
corresponding R value for a randomly chosen 5% of the reflections that were not included in the refinement.
| |Fo| ꢀ|Fc | |/ |Fo|, where |Fo| and |Fc| are the observed and calculated structure factor amplitudes, respectively. Rfree is the

184
E. Anagnostou et al. / Bioorg. Med. Chem. 14 (2006) 181–189
Figure 2. A schematic diagram of the T-state GPb dimeric molecule, viewed down the molecular dyad, showing the positions for the catalytic,
allosteric, inhibitor and new allosteric binding sites. The catalytic site is buried at the centre of the subunit accessible to the bulk solvent through a
˚
15 A long channel. NAG (shown in red) binds at this site and promotes the less active T state through stabilization of the closed position of the 280s
˚
loop (shown in white). The allosteric site, which binds the activator AMP (shown in magenta), is situated at the subunit-subunit interface some 30 A
from the catalytic site. The inhibitor site, which binds purine compounds, such as caffeine, and flavopiridol (shown in yellow) is situated at the
entrance to the catalytic site tunnel, formed by two hydrophobic residues of Phe285 and Tyr613. The new allosteric or indole binding site, located
inside the central cavity formed on association of the two subunits, binds the CP320626 molecule (shown in orange).
2.1. NAG binding
interacts directly with Gly135 N and Asp283 OD1 and
indirectly with Arg569 N, PLP O2P, Tyr573 OH and
Lys574 NZ, through an extensive water-mediated
hydrogen bonding network. The contacts that NAG
makes with Asn284 (1 hydrogen bond and 8 van der
Waals contacts) are maintained in both the 1.9 and
2.3 A resolution structures and promote a closed geom-
etry of the 280s loop (residues 282–287) occurring in the
T-state GPb.
The mode of binding and the interactions that NAG
makes with GPb are similar to those that have been de-
scribed previously in the 2.3 A resolution structure of
˚
the GPb–NAG–IMP complex.17 The glucopyranose
moiety makes one less hydrogen bond with the protein
than in the a-D-glucose complex due to the absence of
–OH at 1 position since NAG is a b-D-glucose analogue.
However, there is a hydrogen bond between N1 of NAG
and carbonyl O of His377, an interaction that is con-
served in all b-D-glucopyranosylamine and spirohydan-
toin analogues of b-D-glucopyranose studied so
far.6,18,23,25,26 There are, in total, 15 hydrogen bonds
and 61 van der Waals interactions (6 nonpolar/nonpo-
lar, 10 polar/polar and 45 nonpolar/polar) (Table 2) in
the GPb–NAG complex. The hydrogen bonds formed
between the ligand and the protein are shown in Figure
4A. Three water molecules, not identified in the previous
structural study, were found to be located in the vicinity
of the catalytic site, Wat106, Wat110 and Wat314.
Wat106 is hydrogen bonded to Glu88 OE1, Asn282 O,
and makes a water-mediated link to Asn284 N and
ligand O7 atom (through Wat355). Wat110 is hydrogen
bonded to His341 NE2 of the b-pocket, Asp339 OD1
and Ala383 of O through water Wat109. Wat314 is
hydrogen bonded to Asp283 OD2, Lys574 NZ, Tyr573
OH and another water molecule (Wat104) which itself
˚
2.2. NFAG binding
The ligand binds at the catalytic site of the enzyme in the
same mode as the lead molecule, and exploits a total of
15 hydrogen bonds and 75 van der Waals interactions (4
nonpolar-nonpolar, 19 polar-polar and 52 nonpolar/po-
lar) (Table 3). The hydrogen bond of the amide nitrogen
(N1) with His377 O is retained. There are two additional
polar contacts in comparison to the GPb–NAG com-
plex; atoms F2 and F3 make rather short contacts
˚
(3.2 A) to Wat315. The water-mediated hydrogen bonds
between NFAG and GPb are also maintained, with the
exception of the hydrogen bonds between O2 of NFAG
and Lys574 NZ, Asn284 ND2, Tyr573 OH, Asp383
OD2 and Wat99 through Wat319, and F2, F3, O7 polar
contacts to Wat314 through Wat315. The fluorine atoms
of NFAG are involved in 21 van der Waals interactions
with the GPb molecule. However, the carbon C8 of the

E. Anagnostou et al. / Bioorg. Med. Chem. 14 (2006) 181–189
185
Figure 3. 2FoꢀFc electron density maps of the refined NAG (A) and NFAG (B) structures bound at the catalytic site of GPb. The maps are
contoured at the 1.0 r level.
inhibitor NFAG forms four less van der Waals interac-
tions with the protein; there is one interaction with
Asn284 OD1 and one with a water molecule (Wat315),
while in the GPb–NAG complex carbon C8 interacts
additionally to another water molecule and Asn284
CG, ND2 as well as with Asp339 OD1. Details of the
contacts made at the catalytic site for GPb–NFAG com-
plex are shown in Figure 4B.
the new allosteric, and the inhibitor sites or the tower/
tower subunit interface.
A structural comparison between the GPb–NAG com-
plex and the GPb–NFAG complex shows that the two
proteins also have very similar overall structures
(Fig. 5); the structures superimpose quite well and they
resemble each other in the vicinity of the catalytic site;
the positions of the Ca, main-chain and side-chain
atoms for residues 18–249, 262–312 and 326–829 deviate
2.3. Comparison between NAG and NFAG binding
˚
from their mean positions by 0.072, 0.094 and 0.465 A,
NFAG can be accommodated at the catalytic site, with
essentially no disturbance of the structure. The superpo-
sition of the GPb–NFAG complex with the native T-
state GPb over well-defined residues (18–249, 262–312
and 326–829) gave an r.m.s. deviation of 0.179 A for
Ca atoms, indicating that the two structures are overall
similar. There are no changes at the allosteric effector,
respectively, indicating overall negligible changes be-
tween the complex structures. However, comparison re-
veals an interesting conformational shift for residue
Asp339. In the GPb–NFAG complex (as compared with
the GPb–NAG complex), the CA, CB and CG of
˚
˚
Asp339 are shifted by 0.3, 0.45 and 0.7 A, respectively.
The dihedral angle v2 [CA-CB-CG-OD1] is rotated by

186
E. Anagnostou et al. / Bioorg. Med. Chem. 14 (2006) 181–189
Table 2. Hydrogen bond interactions between NAG, NFAG and
residues at the catalytic site of GPb
nated analogue. The result is reminiscent of that of the
binding of D-Ala (Kd ꢂ 3.0 0.8 mM) and D-F3-Ala
(Kd = 6.2 0.6 mM) to Rhodotorula gracilis D-amino
acid oxidase.38 Comparison of the NAG and NFAG
complexes revealed that the most significant differences
between the two structures are shifts of Asp339, and
two water molecules on binding NFAG, which are ab-
sent in the interaction with NAG. In fact, being larger
than hydrogen, the fluorine atom can cause unfavour-
able steric interactions with Asp339 and water structure
in the catalytic site; the terminal –CF3 does come close
to the side chain of Asp339, but the carboxylate group
rotates 90° about the CB-CG bond to avoid contact
with the inhibitor.
Inhibitor
atom
GPb–NAG
GPb–NFAG
Protein atom Distance Protein atom Distance
˚
(A)
˚
(A)
N1
O2
His377 O
Asn284 ND2
Tyr573 OH
Glu672 OE1
Wat258
2.9
3.1
3.3
3.3
2.7
(3.4)
2.7
3.2
3.1
2.9
2.5
2.6
2.8
2.5
3.0
His377 O 2.8
Asn284 ND2 3.0
Tyr573 OH
Glu672 OE1
Wat254
3.1
3.2
2.8
3.3
2.7
3.1
3.2
2.9
2.6
2.7
Wat314
Wat319
O3
Glu672 OE1
Ser674 N
Gly675 N
Gly675 N
Wat200
Glu672 OE1
Ser674 N
Gly675 N
Gly675 N
Wat196
O4
O6
O7
The decrease in binding energy associated with trifluoro-
substitution of the acetyl group of NAG is however con-
tradictory with respect to the increase in the number of
van der Waals contacts. It is possible that, in the GPb–
NFAG complex, the energy gain due to the additional
van der Waals interactions is outbalanced by the energy
loss due to the conformational change of Asp339 and
the bulkier fluorine atoms—since the fluorine atom has
His377 ND1
Asn484 OD1
Wat105
His377 ND1
Asn484 OD1 2.8
Wat100
Wat315
Wat315
Wat315
2.8
3.1
Wat355
(F2
(F3
3.2)
3.2)
Water-mediated interactions. NAG: Wat258 is hydrogen bonded to
Ala673 N, Thr671 O, and to Thr671 O, Val379 N and Wat256 through
Wat257; Wat200 is hydrogen bonded to Thr676 N, OG1 and PLP
O3P; Wat105 is hydrogen bonded to Leu136 N, Asp283 OD1, and to
Leu136 N, Gly137 N, Gly134 N and Glu88 OE2 through Wat199;
Wat355 is hydrogen bonded to Asn284 N and to Asn133 ND2, Glu88
OE1 and Asn282 O through Wat106. NFAG: Wat245 is hydrogen
bonded to Ala673 N, Thr671 O and to Thr671 O, Val379 N and
Wat252 through Wat253; Wat196 is hydrogen bonded to Thr676 N,
OG1 and PLP O3P; Wat100 is hydrogen bonded to Leu136 N, Asp283
OD1 and to Leu136 N, Gly137 N, Gly134 N and Glu88 OE2 through
Wat195; Wat315 is hydrogen bonded to Asn284 N, and to Asn133
ND2, Glu88 OE1 and Asn282 O, and Asn284 N, Phe285 N and
Wat103 through Wat101 and Wat314, respectively.
˚
a van der Waals radius of 1.35 A compared to hydrogen
˚
with a radius of 1.2 A, leading to a lower binding affinity
(Ki = 75 lM) compared to NAG (Ki = 32 lM). Howev-
er, it cannot be excluded that other factors are involved
in the reduced binding energy for NFAG, including in-
creased hydrophilicity. It has been reported that fluorine
can also form hydrogen bonds with water, which could
influence its partitioning between freely solvated and en-
zyme-bound states.39 On the other hand, hydrophilicity
of the trifluoromethyl group may differ according to the
carbon skeleton to which it is attached. For fluorinated
aliphatic carbonyl compounds, the apparent hydrophi-
licity depends even on the choice of partitioning sol-
vents.40,41 Unfortunately, we were unable to locate
relevant literature data for the hydrophilicities of acet-
amide- and trifluoroacetamide-type compounds, except
for one report on 4-nitrophenyl 2-acylamido-2-deoxy-
b-D-glucosaminides.42 In this paper, a considerably
higher hydrophilicity is invoked for the trifluoroaceta-
mide than the acetamide derivative, which can be similar
for NFAG and NAG, respectively.
91° to avoid close contacts with the fluorine atoms of
NFAG. In its new position, the carboxyl group of
Asp339 is hydrogen bonded to His377 N. Following
the conformational change of Asp339, Wat318 also
˚
shifts by ꢁ1.1 A to maintain its hydrogen bond with
Asp339 OD1. Furthermore, Wat315 shifts away
˚
(0.4 A) from the ligand to create more space for the
trifluoroacetyl group to be accommodated without
causing steric hindrance.
4. Materials and methods
3. Discussion
The synthesis of NFAG (Scheme 1) was described previ-
ously.20 NAG was kindly provided by Prof. G. W. J.
Fleet. Rabbit GPb was isolated, purified, recrystallised
and assayed as described.17 Kinetic experiments were
performed in the direction of glycogen synthesis in the
presence of constant concentrations of glycogen (1%
w/v), AMP (1 mM) and various concentrations of Glc-
1-P (1.5–20 mM) and inhibitor (44–178 lM). The inor-
ganic phosphate released in the reaction was determined
as previously.44
Bioisosters have been used to improve the properties of
a molecule and obtain new classes of compounds.31–33
Substitution of a hydrogen by fluorine is one of the most
common methods of bioisosteric changes employed by
medicinal chemists, possibly endowing enhanced phar-
macological activity,34,35 novel electrostatic interactions
to the modified analogue, such as hydrogen bonding,
and favourable van der Waals interactions with protein
residues at the catalytic site of an enzyme.36,37
This study provides high-resolution structures of GPb in
complex with two bioisosteric molecules, NAG and
NFAG. The NFAG analogue binds with a 2.3-fold de-
creased affinity, relative to the corresponding unfluori-
Native T-state GPb crystals, grown in the tetragonal lat-
tice,45 space group P43212, were soaked with 10 mM
NAG (for 1 h) or 100 mM NFAG (for 30 min) in a buf-
fered solution (10 mM Bes, 0.1 mM EDTA and 0.02%

E. Anagnostou et al. / Bioorg. Med. Chem. 14 (2006) 181–189
187
Figure 4. Stereo diagrams showing interactions between NAG (A) and NFAG (B) and protein in the vicinity of the catalytic site.
sodium azide, pH 6.7), prior to data collection. Data for
the GPb–NFAG complex were collected from a single
crystal on Station 9.6 at Daresbury Laboratory to
Molecular Simulations, Inc., 200 Fifth Avenue Wal-
tham, MA 02154), where the hydrogen atoms of the
methyl group of C8 of NAG were replaced by fluorine
atoms. Alternate cycles of manual rebuilding with the
program ÔOÕ48 and refinement with CNS improved the
quality of the models. The stereochemistry of the pro-
tein residues was validated by PROCHECK.49,50
Hydrogen bonds and van der Waals interactions were
calculated with the program CONTACT, as imple-
mented in CCP450 applying a distance cut-off of 3.3
˚
˚
1.8 A resolution (k = 0.87 A). Data for the GPb–NAG
complex were collected at ESRF, Grenoble, by translat-
ing a single crystal and successively exposing small sec-
˚
tions to the beam (k = 0.933 A), at a resolution of
˚
1.9 A. The reflections were recorded on an ADSC Q4
CCD detector for both complexes. Data reduction and
integration, followed by scaling and merging of the
intensities obtained, were performed with Denzo and
Scalepack, respectively, as implemented in a HKL
suite.46
˚
and 4.0 A, respectively. The program calculates the an-
gle O. . .H. . .N (where the position of hydrogen is
unambiguous) and the angle source . . .oxygen-bonded
carbon atom. Suitable values are 120° and 90°. The
Luzatti plots30 suggest an average positional error for
Crystallographic refinement of both complexes was
performed with CNS version 1.147 using positional
and individual B-factor refinement with bulk-solvent
correction. The starting model employed for the refine-
ment of both complexes was the structure of GPb–
˚
both structures of approximately 0.21 A. The protein
structures were superimposed using LSQKAB.50 The
schematic representation of the crystal structures pre-
sented in all figures were prepared with the program
XOBJECTS (M. E. M. Noble, unpublished results).
The coordinates of the new structures have been depos-
ited with the RCSB Protein Data Bank (http://
www.rcsb.org/pdb), with codes 1WW2 (GPb–NAG
complex) and 1WW3 (GPb–NFAG complex).
17
˚
NAG–IMP complex determined at 2.3 A resolution.
For both data sets, there was sufficient (2FoꢀFc) and
(FoꢀFc) electron density to accommodate the ligands
at the catalytic site. The NFAG model was built using
the program QUANTA (QUANTA Version 3.3, 1992,

188
E. Anagnostou et al. / Bioorg. Med. Chem. 14 (2006) 181–189
Table 3. van der Waals interactions between compounds NAG, NFAG and residues at the catalytic site of GPb
Inhibitor atom
GPb–NAG
GPb–NFAG
Protein atom
His377 O
No. of contacts
Protein atom
His377 O
Asn284 ND2; His377 C, CB
No. of contacts
C1
N
1
4
1
3
Asn284 OD1, ND2; His377 C,
CB
C2
His377 O; Glu672 OE1; Ala673
CB; Wat258
4
His377 O; Glu672 OE1; Wat254
3
O2
C3
O3
Asn284 CG, OD1; His377 O
Glu672 OE1; Gly675 N; Wat200
Glu672 CG, CD; Ala673 N;
Ser674 C, CA; Gly675 CA
3
3
6
Asn284 CG, OD1; His377 O
Glu672 OE1; Gly675 N; Wat196
Glu672 C, CG, CD; Ala673 N,
C, CA, CB; Ser674 CA; Gly675
CA; Wat254
3
3
10
C4
O4
Gly675 N; Wat200
2
7
Gly675 N; Wat196
Asn484 OD1; Ser674 C, CB;
Gly675 C, CA, O
2
6
Asn484 OD1; Ser674 C, CB;
Gly675 C, CA, O; Thr676 CG2
Gly135 C; Leu136 N; Wat200
His377 CB, CG, O, ND1
Gly135 C, O; Leu139 CD2;
His377 ND1; Asn484 OD1
Leu139 CD2; His377 CG, CE1;
Val455 CG1; Asn484 CG
Asn284 CG, OD1, ND2;
Wat105; Wat355
C5
O5
C6
3
4
5
Gly135 C; Leu136 N; Wat196
His377 CB, O, ND1
3
3
5
Gly135 C, O; Leu136 CA; His377
ND1; Asn484 OD1
O6
C7
5
5
Leu139 CD2; His377 CG, CE1;
Val455 CG1, CG2; Asn484 CG
Asn284 ND2; His377 CB, O;
Wat315
6
4
O7
C8
Leu136 CB, CD2; Asn284 ND2
Asn284 CG, OD1, ND2; Asp339
OD1; Wat110; Wat355
3
6
Leu136 CB; Asn284 ND2
Asn284 OD1; Wat315
2
2
F1
F2
F3
Asp339 OD2; His377 C, CA, CB,
O; Thr378 N, CB, CG2
8
8
Asn284 N, CA, CB, CG, ND2,
OD1; Thr378 CG2, Wat314
Leu136 CD2; Wat314; Wat318
3
75
Total
61
Figure 5. Stereo diagram of the superimposed structures of NAG complex (white) onto the NFAG complex (green) in the vicinity of the catalytic
site.
Acknowledgments
project between Greece and Hungary (2005–2006) (to
N.G.O and L.S.), SRS Daresbury Laboratory (contract
IHPP HPRI-CT-1999-00012) and ESRF (contract
HPRI-CT-1999-00022). We also wish to acknowledge
the assistance of Ms. Rozina Kardakaris and Dr. Vicky
Skamnaki for help in kinetics and data collection.
This work was supported by Greek GSRT through PE-
NED-204/2001 (MNK), and ENTER-EP6/2001 (EDC),
the Hungarian Scientific Research Fund (OTKA
T43550 to Z. Gy.) the Joint Research and Technology

E. Anagnostou et al. / Bioorg. Med. Chem. 14 (2006) 181–189
189
22. Oikonomakos, N. G.; Kosmopoulou, M.; Zographos, S.
References and notes
´
E.; Leonidas, D. D.; Chrysina, E. D.; Somsak, L.; Nagy,
V.; Praly, J.-P.; Docsa, T.; Toth, B.; Gergely, P. Eur. J.
´
´
1. Aiston, S.; Hampson, L.; Gomez-Foix, A. M.; Guinovart,
Biochem. 2002, 269, 1684.
23. Chrysina, E. D.; Oikonomakos, N. G.; Zographos, S. E.;
Kosmopoulou, M. N.; Bischler, N.; Leonidas, D. D.;
J. J.; Agius, L. J. Biol. Chem. 2001, 276, 23858.
2. McCormack, J. C.; Westergaard, N.; Kristiansen, M.;
Brand, C. L.; Lau, J. Curr. Pharm. Des. 2001, 7, 1451.
3. Treadway, J. L.; Mendys, P.; Hoover, D. J. Expert Opin.
Investig. Drugs 2001, 10, 439.
´
Kovacs, L.; Docsa, T.; Gergely, P.; Somsak, L. Biocatal.
Biotransform. 2003, 21, 233.
´
´
24. Gyo¨rgydeak, Z.; Hadady, Z.; Felfo¨ldi, N.; Krakomperger,
A.; Nagy, V.; Toth, M.; Brunyanszki, A.; Docsa, T.;
4. Latsis, T.; Andersen, B.; Agius, L. Biochem. J. 2002, 368,
309.
´
´
´
Gergely, P.; Somsak, L. Bioorg. Med. Chem. 2004, 12,
4861.
5. Aiston, S.; Coghlan, M. P.; Agius, L. Eur. J. Biochem.
2003, 270, 2773.
25. Chrysina, E. D.; Kosmopoulou, M. N.; Kardakaris, R.;
Bischler, N.; Leonidas, D. D.; Kannan, T.; Loganathan,
D.; Oikonomakos, N. G. Bioorg. Med. Chem. 2005, 13,
765.
6. Oikonomakos, N. G. Curr. Protein Pept. Sci. 2002, 3, 561.
7. Kurukulasuriya, R.; Link, J. T.; Madar, D. J.; Pei, Z.;
Richards, S. J.; Rohde, J. J.; Souers, A. J.; Szczepankie-
wicz, B. G. Curr. Med. Chem. 2003, 10, 123.
26. Chrysina, E. D.; Kosmopoulou, M. N.; Tiraidis, C.;
Kardakaris, R.; Bischler, N.; Leonidas, D. D.; Hadady,
8. Lu, Z.; Bohn, J.; Bergeron, R.; Deng, Q.; Ellsworth, K. P.;
Geissler, W. M.; Harris, G.; McCann, P. E.; McKeever,
B.; Myers, R. W.; Saperstein, R.; Willoughby, C. A.; Yao,
J.; Chapman, K. Bioorg. Med. Chem. Lett. 2003, 13, 4125.
9. Ogawa, A. K.; Willoughby, C. A.; Bergeron, R.; Ells-
worth, K. P.; Geissler, W. M.; Myers, R. W.; Yao, J.;
Harris, G.; Chapman, K. T. Bioorg. Med. Chem. Lett.
2003, 13, 3405.
´
Z.; Somsak, L.; Docsa, T.; Gergely, P.; Oikonomakos, N.
G. Protein Sci. 2005, 14, 273.
27. Board, M.; Hadwen, M.; Johnson, L. N. Eur. J. Biochem.
1995, 228, 753.
28. Board, M.; Bollen, M.; Stalmans, W.; Kim, Y.; Fleet, G.
W. J.; Johnson, L. N. Biochem. J. 1995, 311, 845.
29. Board, M.; Johnson, L. N. Diabetes Res. 1995, 28, 95.
30. Luzatti, V. Acta Crystallogr. 1952, 5, 802.
31. Burger, A. Prog. Drug Res. 1991, 37, 287.
32. Papani, G. A.; LaVoie, E. J. Chem. Rev. 1996, 96, 3147.
33. Olesen, P. H. Curr. Opin. Drug Discov. Devel. 2001, 4,
471.
´
10. Somsak, L.; Nagy, V.; Hadady, Z.; Docsa, T.; Gergely, P.
Curr. Pharm. Des. 2003, 9, 1177.
11. Kristiansen, M.; Andersen, B.; Iversen, L. F.; Westerg-
aard, N. J. Med. Chem. 2004, 47, 3537.
12. Wright, S. W.; Rath, V. L.; Genereux, P. E.; Hageman, D.
L.; Levy, C. B.; McLure, L. D.; McCoid, S. C.; McPh-
erson, R. K.; Schelhorn, T. M.; Wilder, D. E.; Zavadoski,
W. J.; Gibbs, E. M.; Treadway, J. L. Bioorg. Med. Chem.
Lett. 2005, 15, 459.
34. Resnati, G. Farmaco 1990, 45, 1043.
35. Bryskier, A.; Chantot, J. F. Drugs 1995, 49, 16.
36. Chong, Y.; Choo, H.; Chu, C. K. Bioorg. Med. Chem.
Lett. 2004, 14, 437.
´
13. Somsak, L.; Nagy, V.; Hadady, Z.; Felfo¨ldi, N.;
˘
37. Lindberg, J.; Pyring, D.; Lo¨wgren, S.; Rosenquist, A;
Docsa, T.; Gergely, P. In Frontiers in Medicinal
Chemistry; Reitz, A. B., Kordik, C. P., Choudhary,
M. I., Atta-ur-Rahman, Eds.; Bentham, 2005; Vol. 2,
pp 253–272.
Zuccarello, G.; Kvarnstro¨m, I.; Zhang, H.; Vrang, L.;
Classon, B.; Hallberg, A.; Samuelsson, B.; Unge, T. Eur.
J. Biochem. 2004, 271, 4594.
38. Umhau, S.; Pollegioni, L.; Molla, G.; Diederichs, K.;
Welte, W.; Pilone, M. S.; Ghisla, S. Proc. Natl. Acad. Sci.
U.S.A. 2000, 97, 12463.
39. Schwartz, B.; Drueckhammer, D. G.; Usher, K. C.;
Remington, S. J. Biochemistry 1995, 34, 15459.
40. Welch, J. Tetrahedron 1987, 43, 3123.
41. Organofluorine Chemistry. Principles and Commercial
Applications, Banks, R. E., Smart, B. E., Tatlow, J.
C., Eds.; Plenum Press, New York and London,
1994.
42. Yamamoto, K. J. Biochem. 1974, 76, 385.
43. Leatherbarrow, R. J. Trends Biochem. Sci. 1990, 15, 455.
44. Saheki, S.; Takeda, A.; Shimazu, T. Anal. Biochem. 1985,
148, 277–281.
14. Martin, J. L.; Veluraja, K.; Johnson, L. N.; Fleet, G. W.
J.; Ramsden, N. G.; Bruce, I.; Oikonomakos, N. G.;
Papageorgiou, A. C.; Leonidas, D. D.; Tsitoura, H. S.
Biochemistry 1991, 30, 10101.
15. Watson, K. A.; Mitchell, E. P.; Johnson, L. N.; Son, J. C.;
Bichard, C. J. F.; Orchard, M. G.; Fleet, G. W. J.;
Oikonomakos, N. G.; Leonidas, D. D.; Kontou, M.;
Papageorgiou, A. C. Biochemistry 1994, 33, 5745.
16. Bichard, C. J. F.; Mitchell, E. P.; Wormald, M. R.;
Watson, K. A.; Johnson, L. N.; Zographos, S. E.; Koutra,
D. D.; Oikonomakos, N. G.; Fleet, G. W. J. Tetrahedron
Lett. 1995, 36, 2145.
17. Oikonomakos, N. G.; Kontou, M.; Zographos, S. E.;
Watson, K. A.; Johnson, L. N.; Bichard, C. J. F.; Fleet, G.
W. J.; Acharya, K. R. Protein Sci. 1995, 4, 2469.
18. Watson, K. A.; Mitchell, E. P.; Johnson, L. N.; Cruciani,
G.; Son, J. C.; Bichard, C. J. F.; Fleet, G. W. J.;
Oikonomakos, N. G.; Kontou, M.; Zographos, S. E. Acta.
Crystallogr. D 1995, 51, 458.
45. Oikonomakos, N. G.; Melpidou, A. E.; Johnson, L. N.
Biochim. Biophys. Acta 1985, 832, 248.
46. Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276,
307.
47. Brunger, A. T.; Adams, P. D.; Clore, G. M.; DeLano, W.
¨
L.; Gros, P.; Grosse-Kunstleve, R. W.; Jiang, J.-S.;
Kuszewski, J.; Nilges, M.; Pannu, N. S.; Read, R. J.;
Rice, L. M.; Simonson, T.; Warren, G. L. Acta. Crystal-
logr. D 1998, 54, 905.
19. Gregoriou, M.; Noble, M. E. M.; Watson, K. A.; Garman,
E. F.; Krulle, T. M.; Fuente, C.; Fleet, G. W. J.;
Oikonomakos, N. G.; Johnson, L. N. Protein Sci. 1998,
7, 915.
¨
20. Somsak, L.; Kovacs, L.; Toth, M.; Osz, E.; Szilagyi, L.;
48. Jones, T. A.; Zou, J. Y.; Cowan, S. W.; Kjeldgaard, M.
Acta. Crystallogr. A 1991, 47, 110.
49. Laskowski, R. A.; MacArthur, M. W.; Moss, D. S.;
Thornton, J. M. J. Appl. Crystallogr. 1993, 26, 283.
50. Collaborative Computational Project No 4. Acta Crystal-
logr. D 1994, 50, 760.
´
Gyo¨rgydeak, Z.; Dinya, Z.; Docsa, T.; Toth, B.; Gergely,
´
´
´
´
P. J. Med. Chem. 2001, 44, 2843.
´
¨
21. Oikonomakos, N. G.; Skamnaki, V. T.; Osz, E.; Szilagyi,
´
L.; Somsak, L.; Docsa, T.; Toth, B.; Gergely, P. Bioorg.
´
Med. Chem. 2002, 10, 261.
´