Synthetic, enzyme kinetic, and protein crystallographic studies of C-β-D-glucopyranosyl pyrroles and imidazoles reveal and explain low nanomolar inhibition of human liver glycogen phosphorylase

Article abstract of DOI:10.1016/j.ejmech.2016.06.049

C-β-D-Glucopyranosyl pyrrole derivatives were prepared in the reactions of pyrrole, 2-, and 3-aryl-pyrroles with O-peracetylated β-D-glucopyranosyl trichloroacetimidate, while 2-(β-D-glucopyranosyl) indole was obtained by a cross coupling of O-perbenzylated β-D-glucopyranosyl acetylene with N-tosyl-2-iodoaniline followed by spontaneous ring closure. An improved synthesis of O-perbenzoylated 2-(β-D-glucopyranosyl) imidazoles was achieved by reacting C-glucopyranosyl formimidates with α-aminoketones. The deprotected compounds were assayed with isoforms of glycogen phosphorylase (GP) to show no activity of the pyrroles against rabbit muscle GPb. The imidazoles proved to be the best known glucose derived inhibitors of not only the muscle enzymes (both a and b) but also of the pharmacologically relevant human liver GPa (K_i?=?156 and 26?nM for the 4(5)-phenyl and -(2-naphthyl) derivatives, respectively). An X-ray crystallographic study of the rmGPb-imidazole complexes revealed structural features of the strong binding, and also allowed to explain the absence of inhibition for the pyrrole derivatives.

Full text of DOI:10.1016/j.ejmech.2016.06.049

Accepted Manuscript
Synthetic, enzyme kinetic, and protein crystallographic studies of C-β-D-
glucopyranosyl pyrroles and imidazoles reveal and explain low nanomolar inhibition of
human liver glycogen phosphorylase
Anastassia L. Kantsadi, Éva Bokor, Sándor Kun, George A. Stravodimos, Demetra
S.M. Chatzileontiadou, Demetres D. Leonidas, Éva Juhász-Tóth, Andrea Szakács,
Gyula Batta, Tibor Docsa, Pál Gergely, László Somsák
PII:
S0223-5234(16)30539-6
10.1016/j.ejmech.2016.06.049
EJMECH 8710
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 27 May 2016
Revised Date: 19 June 2016
Accepted Date: 28 June 2016
Please cite this article as: A.L. Kantsadi, E. Bokor, S. Kun, G.A. Stravodimos, D.S.M. Chatzileontiadou,
D.D. Leonidas, E. Juhász-Tóth, A. Szakács, G. Batta, T. Docsa, P. Gergely, L. Somsák, Synthetic,
enzyme kinetic, and protein crystallographic studies of C-β-D-glucopyranosyl pyrroles and imidazoles
reveal and explain low nanomolar inhibition of human liver glycogen phosphorylase, European Journal
of Medicinal Chemistry (2016), doi: 10.1016/j.ejmech.2016.06.049.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Synthetic, enzyme kinetic, and protein crystallographic studies of C-β-D-glucopyranosyl pyrroles
and imidazoles reveal and explain low nanomolar inhibition of human liver glycogen
phosphorylase
A. L. Kantsadi, É. Bokor, S. Kun, G. A. Stravodimos, D. S.M. Chatzileontiadou, D. D. Leonidas, É.
Juhász-Tóth, A. Szakács, Gy. Batta, T. Docsa, P. Gergely, and L. Somsák

ACCEPTED MANUSCRIPT
Synthetic, enzyme kinetic, and protein
crystallographic studies of C-β- -glucopyranosyl
D
pyrroles and imidazoles reveal and explain low
nanomolar inhibition of human liver glycogen
phosphorylase
Anastassia L. Kantsadi,1# Éva Bokor,2# Sándor Kun,2# George A. Stravodimos,1 Demetra S.M.
Chatzileontiadou,1 Demetres D. Leonidas,1 Éva Juhász-Tóth,2 Andrea Szakács,2 Gyula Batta,2 Tibor
Docsa,3 Pál Gergely,3 and László Somsák2*
1Department of Biochemistry and Biotechnology, University of Thessaly, 26 Ploutonos Str. 41221 Larissa,
Greece
2Department of Organic Chemistry, University of Debrecen, POB 400, H-4002 Debrecen, Hungary
3Department of Medical Chemistry, Faculty of Medicine, University of Debrecen, Egyetem tér 1, H-4032
Debrecen, Hungary
#These authors have equally contributed to this work.
Corresponding authors: DDL ‒ tel. +302410565278, fax +302410565290, e-mail ddleonidas@bio.uth.gr;
LS ‒ tel. +3652512900 ext 22348, fax +3652512744, e-mail somsak.laszlo@science.unideb.hu.
1

ACCEPTED MANUSCRIPT
Abstract
C-β- -Glucopyranosyl pyrrole derivatives were prepared in the reactions of pyrrole, 2-, and 3-aryl-
pyrroles with O-peracetylated β- -glucopyranosyl trichloroacetimidate, while 2-(β- -glucopyranosyl)
indole was obtained by a cross coupling of O-perbenzylated β- -glucopyranosyl acetylene with N-tosyl-
2-iodoaniline followed by spontaneous ring closure. An improved synthesis of O-perbenzoylated 2-(β-
D
D
D
D
D-
glucopyranosyl) imidazoles was achieved by reacting C-glucopyranosyl formimidates with α-
aminoketones. The deprotected compounds were assayed with isoforms of glycogen phosphorylase
(GP) to show no activity of the pyrroles against rabbit muscle GPb. The imidazoles proved to be the best
known glucose derived inhibitors of not only the muscle enzymes (both a and b) but also of the
pharmacologically relevant human liver GPa (Ki = 156 and 26 nM for the 4(5)-phenyl and -(2-naphthyl)
derivatives, respectively). An X-ray crystallographic study of the rmGPb-imidazole complexes revealed
structural features of the strong binding, and also allowed to explain the absence of inhibition for the
pyrrole derivatives.
Keywords: Diabetes type 2; Glycogen phosphorylase inhibitor; C-Glucopyranosyl derivative; Pyrrole;
Indole; Imidazole.
2

ACCEPTED MANUSCRIPT
Introduction
A key player in glycogen metabolism is glycogen phosphorylase (GP; EC 2.4.1.1), an enzyme that
catalyses glycogen degradation to glucose-1-phosphate (Glc-1-P).1 Three isoforms of GP exist in the
brain, muscle and liver tissue that share ~80 % sequence homology. GP is controlled by allosteric
effectors and reversible phosphorylation. It exists in at least two inter-convertible forms, a low activity
and substrate specificity T state and a high activity and substrate specificity R state, which are in
equilibrium. Phosphorylation by glycogen phosphorylase kinase (PhK) or the binding of various allosteric
activators, such as AMP, shift the equilibrium towards the R state. In contrast, dephosphorylation by
glycogen phosphorylase phosphatase or the binding of inhibitors such as glucose promote the transition
from the R to T state. The active phosphorylated enzyme (GPa) is the main form of GP in the liver
whereas the unphosphorylated inactive form (GPb) is the prevalent form in the resting muscle.
Extensive conformational changes occur in the tertiary structure upon transition from the T to R state.2, 3
These mainly occur at the inter-subunit interface and result in the rearrangement of the flexible 280s
loop (residues 282 to 285) from a conformation that blocks access to the catalytic site of the enzyme (T
state) to one that allows substrate binding (R state). X-ray crystallographic studies have revealed that
apart from the catalytic site there are other six different ligand binding sites: the allosteric site, the new
allosteric site, the inhibitor site, the glycogen storage site, the benzimidazole site and the quercetin
binding site.4, 5 Although AMP binds to both the muscle and the liver enzyme, only the muscle isozyme is
activated cooperatively.6 There are no insertions or deletions in the sequence of the liver enzyme as
compared to the muscle and out of the 171 amino acid differences between the two enzymes, 49% (85
amino-acids) are conservative and most of the amino acid changes are functionally neutral. The active
site is conserved in both amino acid sequence and structural architecture.6
3

ACCEPTED MANUSCRIPT
The liver is capable of storing glucose as glycogen and, through a reverse process, to produce and
release glucose to the blood stream. Therefore, the liver is intimately linked to glycaemic control and
human liver GP (hlGP) is a validated pharmaceutical target for the discovery of novel antihyperglycaemic
agents.1, 4, 7 Glycogen phosphorylase inhibitors (GPIs) have been shown to have a pharmaceutical
potential to combat type 2 diabetes (the non-insulin dependent form of diabetes)8-10 while they were
also found to display cardioprotective effects11, 12 and anticancer activity13-16 as well as to reduce
cerebral ischemia.17, 18 The design of GPIs has mainly been focused on derivatives of glucose (the
physiological inhibitor of the enzyme) and during the last decade, the active site of GP has been probed
with more than 150 glucose derivatives.1, 4, 19-24 The main goal of the design was to maximize the number
of interactions with active site residues and thus increase inhibitory potency. In this context much effort
has been devoted to introduce C-1 substituents suitable to exploit interactions with a hydrophobic cleft
inside the active site, the so-called β-cavity. As a result of these studies, three compound classes
emerged as submicromolar inhibitors of GP:19, 23, 24 glucopyranosylidene-spiro-heterocycles, N-acyl-N’-β-
D-glucopyranosyl ureas, and C-β-D-glucopyranosyl heterocycles. A collection of the latter type of
compounds in Chart 1 allows one to conclude that the presence of a hydrogen-bond donor (HBD)
heterocycle is in general a better inhibitor than the non-HBD counterpart (compare the pairs 1-2,25, 26 3-
4,27-30 5-6,31 and 7-831). From these series C-β-
highest potency against rabbit muscle GPb (rmGPb). Similarly to the other two inhibitor classes, the C-β-
-glucopyranosyl compounds with a larger aromatic substituent (cf phenyl a vs naphthyl b derivatives in
Chart 1) proved better inhibitors.
D
-glucopyranosyl imidazoles31 8 have so far shown the
D
4

ACCEPTED MANUSCRIPT
Chart 1. C-β-D-Glucopyranosyl heterocycles as
inhibitors (Ki [μM]) of rabbit muscle glycogen
phosphorylase b (rmGPb)
1
2
22925
7626
1125
926
R
a
b
6427, 28
12 (2.4)28
3
4
729, 30
0.4129, 30
no inh. at 625
μM31
-
5

ACCEPTED MANUSCRIPT
5
6
40031
31031
0.2831
-
15831
7
8
0.03131
Synthetic target compounds of this study
In this paper we disclose our studies on the preparation of several C-β-D-glucopyranosyl pyrrole
and indole derivatives to answer the question if heterocycles with one HBD atom would make good
inhibitors of GP, and present an improved synthesis of 2-(β- -glucopyranosyl) imidazoles. In addition,
D
the latter compounds have been tested against human liver GPa (hlGPa), the pharmaceutical target, and
the crystal structure of rmGPb in complex with the most efficient compounds have been determined to
elucidate the structural basis of their inhibitory potency.
6

ACCEPTED MANUSCRIPT
Results and Discussion
Syntheses
For the preparation of C-glycosyl pyrroles a generally used method is the reaction of an electron
rich pyrrole derivative with an electrophilic glycosyl donor. Typically, such reactions of 1H-pyrroles give
2-glycosylated products.32-34 For the synthesis of the target compounds 2- (11)35 and 3-arylpyrroles
(12a,36 12b37) were prepared according to previously reported methods. C-glycosylation of 10 and 11
with O-peracetylated β-D
-glucopyranosyl-trichloroacetimidate38 9 in the presence of BF3·OEt2 furnished
the β-configured (3J1’,2’ > 9 Hz) products 1333 and 15, respectively, in moderate to high yields as expected
(Scheme 1). Surprisingly, glycosylation of 3-arylpyrroles 12 took place in the sterically more crowded 2-
position to give 17. The site of C-glucosylation in 15b and 17b was established by 1H-1H NOE experiments
(Figure 1). In the NMR spectra of 15b the characteristic resonances of H-3 (6.53 ppm, J ≈ 3 Hz), H-4 (6.24
ppm, J ≈ 3 Hz) and C-3 (106.6 ppm), C-4 (110.2 ppm) indicate the 2,5-disubstituted pyrrole structure,39
and this is also proved by the observed NOEs. In case of 17b the 2,3-disubstituted pyrrole could not be
differentiated from a 2,4-disubstituted pyrrole by a simple analysis of 1H (pyrrole-H resonances at 6.43
and 6.86 ppm J ≈ 2.6 Hz) and 13C (pyrrole CH signals: 109.3, 119.2 ppm) spectra. However, NOE between
H-4 and H-5 as well as the proximity of the sugar and naphthyl moieties (H-1’‒naphthyl-H NOEs) indicate
the 2,3-disubstitution of the pyrrole ring. O-Acetyl protecting groups were removed by the Zemplén
protocol to obtain the test compounds 14, 16, and 18.
7

ACCEPTED MANUSCRIPT
Scheme 1. Reaction conditions: i) BF3·Et2O, 4 Å molecular sieves, dry CH2Cl2, ‒50 °C to rt; ii) cat. NaOMe
in MeOH, rt.
Figure 1. Nuclear Overhauser effects observed in C-glucopyranosylated (2-naphthyl)-pyrroles.
8

ACCEPTED MANUSCRIPT
Synthesis of 2-glycosyl indoles was studied in detail in the course of the preparation of C-
mannosyl tryptophan and analogous compounds.40 2-(β-
D-Glucopyranosyl) indole (23) was prepared by
adaptation and slight modification of the reported method. O-Perbenzylated glucosyl acetylene41 19 was
cross-coupled with N-tosyl-2-iodoaniline (20) and cyclized to indole 2140 in one step. Subsequent
detosylation with Bu4NF and debenzylation by catalytic hydrogenolysis yielded the target 2-β-D-
glucopyranosyl indole 23.
Scheme 2. Reaction conditions: i) Pd(PPh3)4, CuI, PPh3, dry Et3N, 75 °C, under Ar; ii) Bu4NF, THF, reflux; iii)
H2 (10 bar), Pd(C), dry EtOAc, rt.
9

ACCEPTED MANUSCRIPT
In a recent communication we presented the first preparation of 4(5)-aryl-2-(β-D-glucopyranosyl)
imidazoles 8 (Chart 1) based on the cyclisation of O-perbenzoylated C-β-D-glucopyranosyl formamidine
with α-bromoketones.31 However, this transformation furnished the protected imidazoles 26 only in low
yields (10-29 %). In order to find a more efficient procedure for the formation of 26 another ring-closing
reaction was investigated: ethyl C-β-D
-glucopyranosyl formimidate42 (24) was treated with α-amino
ketones 25 (obtained from the corresponding α-azido ketones as described in the supporting
information) in anhydrous pyridine providing the desired imidazoles in satisfactory yields (Scheme 3).
OBz
O
NH
C
BzO
BzO
Ar
OEt
OBz
24
O
.
NH2 HCl
25
i
a
Ar
OR
HN
O
Ar
RO
RO
N
OR
26b (R = Bz)
26a (R = Bz)
42 %
45 %
ii
ii
b
8b (R = H)
8a (R = H)
77 %
74 %
Scheme 3. Reaction conditions: i) dry pyridine, rt; ii) cat. NaOMe in MeOH, rt.
10

ACCEPTED MANUSCRIPT
Kinetic Studies
The test compounds were evaluated against GPs of various origin and activation state and the
data are collected in Table 1. The pyrrole derivatives 14, 16, 18, and indole 23 were tested as inhibitors
of rmGPb, the prototype of GPs,43 by using an assay described previously.44 The pyrroles 14, 16, and 18
proved inactive in the concentration range studied (the highest applied concentration was 625 μM)
while indole 23 exhibited very weak inhibition, therefore, these compounds were not investigated
further.
Since imidazoles 8 were shown to be the strongest glucose derived inhibitors of rmGPb known to
date,31 the inhibitory potency of 8a and 8b has been evaluated against hlGPa. For comparison purposes
we also determined their potency with rmGPa and measured the inhibitory potency of α-D-glucose for
hlGPa. Both compounds 8 are competitive inhibitors with respect to substrate Glc-1-P as revealed by the
Lineweaver-Burk plots that intersect at the same point on the y-axis for hlGPa and rmGPa. The Ki values
derived from the Dixon plots (1/v vs [I]) were for 8a 0.156 ꢀM and 0.226 ꢀM for hlGPa and rmGPa,
respectively. For 8b they were found to be 0.026 μM and 0.065 μM for hlGPa and rmGPa, respectively.
These values are in agreement with their respective inhibition constants for rmGPb31 and indicate that
8a and 8b are approximately 27400 and 164600 times, respectively, more potent than α-D-glucose. They
also indicate that the compound with the 2-naphthyl group is more efficient than its phenyl counterpart.
Furthermore, it seems that the two compounds do not display any preference for the three isoenzymes
(hlGPa, rmGPa and rmGPb) since they inhibit all three with similar potency.
11

ACCEPTED MANUSCRIPT
Table 1. The chemical structures and the inhibition constants of the compounds studied. The
crystallographic numbering scheme for the inhibitor atoms in
D-glucose, 8a, and 8b is also shown.
Compound
number
Chemical Structure
Ki (μM)
hlGPa
rmGPa
200045
rmGPb
170046
α-D-Glc
4280 ± 410
8a
0.156 ± 0.008 0.226 ± 0.098
0.026 ± 0.006 0.065 ± 0.003
0.28031
8b
14
0.03131
no inh. at
ND
ND
ND
ND
625 µM
no inh. at
16a
625 µM
no inh. at
16b
ND
ND
625 µM
12

ACCEPTED MANUSCRIPT
no inh. at
18a
18b
ND
ND
625 µM
no inh. at
ND
ND
ND
ND
625 µM
23
IC50 625 μM
ND: not determined
Structural Studies
The crystal structures of rmGPb in complex with 8a and 8b have been determined at high
resolution (1.85 Å). Both inhibitors were found bound at the catalytic site with all atoms well defined
within the electron density map (Fig. 1). The binding of 8a and 8b does not trigger any significant
conformational change to rmGPb and the r.m.s.d. between the complex structures and the ligand-free
structure over well-defined residues gave a r.m.s.d. of 0.15 and 0.14 Å for Ca atoms for the 8a and 8b
complexes, respectively. Upon binding both inhibitors displace four water molecules from the active site
and they bind in similar structural mode (Fig. 2). The most characteristic feature is the anchoring of their
β-D-glucopyranose moiety at the rmGPb glucose binding site and its interactions with protein residues
(Table 2). These are essentially the same with those formed by glucose with rmGPb.47 The imidazole
linker engages in a hydrogen bond interaction with the main chain oxygen of His377 via N(H)2 and in
13

ACCEPTED MANUSCRIPT
water mediated interactions with Asp283 side chain via N5. The interaction with the carbonyl oxygen of
His377 has been previously observed for the amide nitrogen of N-acetyl-β-D-glucopyranosylamine (Ki =
32 ꢀM),48, 49 of glucopyranosylidene-spirohydantoin (Ki = 3.1 ꢀM),50, 51 and -thiohydantoin (Ki = 5.1 ꢀM),52
-glucopyranosyl formamide analogs (Ki = 310 – 1800 ꢀM),53 and C-(β-
D-glucopyranosyl)benzimidazole
β-D
(Ki = 9 ꢀM).26 The increased inhibitory potency of these compounds was attributed to the interaction
with the main chain oxygen of His377 and one of the design goals of 8 was to exploit this interaction. In
addition, participation of heterocyclic nitrogen atoms in water mediated interactions were observed for
28
C-β-D
-glucopyranosyl benzimidazole26 (2), 5-methyl-1,3,4-oxadiazole,26,
and 5-(2-naphthyl)-1,2,4-
oxadiazole21 (3b) derivatives. In these compounds the nitrogen atoms are in the same position as N5 in
the imidazoles. A comparison of the inhibitory efficiency of benzimidazole 2 (Ki = 9 μM) and indole 23
(IC50 ~ 625 μM) shows the practical loss of activity for the latter that may be explained by the absence of
the second nitrogen in 23 (assuming that the heterocycle is oriented towards the formation of the H-
bond to His377). The absence of this nitrogen in the pyrrole ring may also explain the loss of activity for
pyrroles 16 and 18 in which the orientation of NH towards His377 places the aromatic residues into
highly unfavourable positions. The opposite orientation allowing the participation of the ring N in water
mediated interactions would result in the loss of the H-bonding ability to His377 and this, even if the
position of the aromatic moieties would be optimal in 16, seems detrimental for the binding.
14

ACCEPTED MANUSCRIPT
Figure 1. Diagram showing the electron density map of 8a (A) and 8b (B) before incorporating the
atomic models of the inhibitors (presented as ball-and-stick-models).
Table 2. Hydrogen bond interactions of 8a, and 8b in the crystal. Numbers shown are distances in Å.
Inhibitor atom
O2’
Protein atom
Asn284 ND2
Tyr573 OH
Glu672 OE1
Water
8a
8b
2.9
3.1
3.1
2.9
3.0
2.7
3.3
3.1
3.2
2.9
2.9
2.8
Water
O3’
Glu672 OE1
15

ACCEPTED MANUSCRIPT
Ala673 N
Ser674 N
Gly675 N
Gly675 N
Water
3.3
3.1
3.2
2.9
2.7
2.7
2.8
2.9
2.8
15
3.3
3.1
3.1
2.9
2.6
2.7
2.8
2.9
2.8
15
O4’
O6’
His377 ND1
Asn484 ND2
His377 O
Water
N2
N5
Total
The phenyl and the naphthyl groups of 8a and 8b, respectively, bind at the β-cavity engaging in
numerous van der Waals interactions. Both phenyl and naphthyl are involved in 18 van der Waals
interactions with Glu88 (1), Asn282 (2), Asn284 (8), Phe285 (2), and His341 (5). In addition, the naphthyl
group is involved in 11 van der Waals interactions with Asn282 (1), Phe285 (3), Phe286 (1), Arg292 (2),
His341 (3), and Ala383 (1). The structural basis of the inhibitory potency of the two compounds seems to
lie on their interactions with the side chains of Asn282, Asp283, and Asp284 which hold the 280s loop in
its closed T-state conformation, therefore inhibiting access of the substrate to the catalytic site. The
additional van der Waals interactions of 8b seem to form the structural basis for the significant
difference in the inhibitory potencies of these two inhibitors.
16

ACCEPTED MANUSCRIPT
Figure 2. Stereo diagram showing the network of interactions formed between 8a (a) and 8b (b) with
residues at active site of rmGPb. Hydrogen bond interactions are represented as dotted lines and water
molecules as blue spheres.
17

ACCEPTED MANUSCRIPT
Conclusion
Syntheses of a series of C-glucopyranosyl pyrroles, indole, and an improved preparation of C-
glucopyranosyl imidazoles allowed us to study and compare their inhibitory efficiency against isoforms
of GP. While pyrroles had no effect with rmGPb, imidazoles proved to be the best known glucose
analogue inhibitors of rmGPb, rmGPa, and also of hlGPa, the physiologically relevant enzyme. Structural
investigations of the rmGPb-imidazole complexes by X-ray crystallography revealed important
interactions between the imidazole ring and the protein, such as the H-bridge from N(H)2 to His377 and
involvement of N5 in water mediated H-bond networks. The absence of the latter interactions in the
pyrroles may explain why these compounds show no inhibitory activity.
Experimental
Syntheses
General methods
Melting points were measured on a Kofler hot-stage and are uncorrected. Optical rotations were
determined with a Perkin-Elmer 241 polarimeter at rt. NMR spectra were recorded with Bruker 360
(360/90 MHz for 1H/13C) or Bruker Avance II 500 (500.13/125.76 for 1H/13C) spectrometers. Chemical
shifts are referenced to Me4Si (1H), or to the residual solvent signals (13C). Mass spectra were recorded
with a Thermo LTQ XL mass spectrometer (Thermo Electron Corp., San Jose, CA, USA) operated in a full
scan positive ion ESI mode. Microanalyses were performed on an Elementar Vario Micro cube
18

ACCEPTED MANUSCRIPT
instrument. TLC was performed on DC-Alurolle Kieselgel 60 F254 (Merck) plates, visualized under UV light
and by gentle heating. For column chromatography Kieselgel 60 (Merck, particle size 0.063-0.200 mm)
was used. EtOAc (P4O10), CH2Cl2 (P4O10) and Et3N (CaH2) were distilled from the given drying agents and
stored over 4 Å molecular sieves. MeOH was purified by distillation after refluxing for a couple of hours
with magnesium turnings and iodine. Pyridine (VWR), pyrrole, tetrakis(triphenylphosphine)palladium(0),
TOSMIC, 2-vinylnaphthalene (Sigma-Aldrich) were purchased from the indicated suppliers. 2,3,4,6-Tetra-
O-acetyl-β-
pyrroles (12a,36 12b37), 2,3,4,6-tetra-O-benzyl-β-
(20), ethyl C-(2,3,4,6-tetra-O-benzoyl-β-
-glucopyranosyl)formimidate42 (24) and the 2-azido-1-
arylethanones55 were synthesized according to published procedures.
D
-glucopyranosyl-trichloroacetimidate38 (9), the 2-aryl-1H-pyrroles35 (11a,b), the 3-aryl-1H-
D
-glucopyranosyl-ethyne41 (19), N-tosyl-2-iodoaniline54
D
General procedures
General procedure I for the synthesis of 2-(2’,3’,4’,6’-tetra-O-acetyl-β-D-glucopyranosyl)-pyrroles (13,
15, 17)
2,3,4,6-Tetra-O-acetyl-β-D-glucopyranosyl-trichloroacetimidate (9, 100 mg, 0.2 mmol) and pyrrole (10, 1
mmol) or the corresponding aryl-pyrrole (11a,b or 12a,b, 0.2 mmol) were dissolved in anhydrous CH2Cl2
(3 ml) and activated 4 Å molecular sieves (powder, 200 mg) were added. The mixture was cooled to ‒50
°C and BF3·Et2O (12 μl, 0.1 mmol) was added dropwise with continuous stirring. The reaction was stirred
for 1h at ‒50 °C, then slowly allowed to reach room temperature and stirred for the given time. After
total consumption of the trichloroacetimidate (judged by TLC, 1:1 hexane-EtOAc), molecular sieves were
removed by filtration and the filtrate was purified by column chromatography.
19

ACCEPTED MANUSCRIPT
General procedure II for the preparation of 2-amino-1-arylethanone hydrochlorides (25)
To a degassed solution of a 2-azido-1-arylethanone (18.6 mmol) in methanol (100 mL) a solution of
tin(II) chloride dihydrate (12.65 g, 55.8 mmol) in methanol (160 mL) was added and the mixture was
stirred at rt under N2 atmosphere. When the TLC (3:1 hexane-EtOAc) showed total consumption of the
starting material (5-24 h) dry HCl gas was bubbled through the reaction mixture over 12 h. The solution
was then concentrated under reduced pressure and the residue was put into the freezer for overnight.
The resulting precipitates were filtered off and washed with cold methanol. The products were used in
the next step without further purification.
General procedure III for the synthesis of 4(5)-aryl-2-(2’,3’,4’,6’-tetra-O-benzoyl-β-D-glucopyranosyl)-
imidazoles (26)
Ethyl C-(2,3,4,6-tetra-O-benzoyl-β-D-glucopyranosyl)formimidate (24, 0.15 mmol) and the corresponding
2-amino-1-arylethanone hydrochloride (25, 0.31 mmol, 2 equiv.) were dissolved in anhydrous pyridine
(4 mL) and the reaction mixture was stirred at rt. After completion of the reaction (1-2 d) monitored by
TLC (3:2 hexane-EtOAc) the solvent was removed under diminished pressure and the residue was
purified by column chromatography.
General procedure IV for removal of the O-acetyl protecting groups (for the preparation of test
compounds 14, 16, 18)
An O-acetyl protected compound (100 mg) was dissolved in anh. MeOH (5 mL), a few drops of ~1M
solution of NaOMe in MeOH was added and the mixture was left to stand at rt. After complete
conversion (judged by TLC, 7:3 CHCl3-MeOH) the reaction mixture was neutralized by the careful
addition of AcOH and evaporated in vacuo. The crude product was purified by column chromatography.
20

ACCEPTED MANUSCRIPT
Protein production and purification
Rabbit skeletal muscle glycogen phosphorylase b (rmGPb) was isolated, and purified as described
previously.56 RmGPa was prepared from rmGPb by phosphorylation with a truncated form of the γ
(catalytic) subunit of rabbit skeletal muscle phosphorylase kinase produced as described previously.57
Human liver glycogen phosphorylase a (hlGPa) was expressed from the pET160DEST vector (Invitrogen)
with a His tag (6x) followed by the Lumio tag and a TEV protease prescission site in E.coli strain BL21
Gold (DE3) (Agilent Technologies). The strain was inoculated in LB medium (plus 50 μg/mL ampicillin,
100 mg/L pyridoxine, and 600 mg/L MnCl2) grown at 37 º C to a cell density of OD600= 0.7 and induced for
15 hr with 0.6 mM isopropyl-1-thio-β-D-galactoside (IPTG) prior to harvest. The E. coli cells were
resuspended in lysis buffer (50 mM Na2HPO4/NaH2PO4 [pH 8.0], containing 500 mM NaCl, 20 mM
imidazole, a Protease inhibitor cocktail tablet (EDTA-free, Roche) and 50 U Benzonase) and sonicated at
4º C. The lysate was clarified by centrifugation at 16,000 × g for 1 hr. hlGPb in the soluble fraction of the
lysate was purified using a Ni2+ Sepharose chromatography. His-tag tail was removed by cleaving with
TEV protease and Ni2+ Sepharose chromatography. A final step using anion-exchange chromatography
led to a pure hlGPb sample as judged by the appearance of a single band in SDS page.Phosphorylation to
hlGPa was performed using a truncated form of the γ (catalytic) subunit of rabbit skeletal muscle
phosphorylase kinase.
Kinetic studies
Kinetic studies with hlGPa were performed at 30° C in the direction of glycogen synthesis by measuring
the inorganic phosphate released in the reaction.56 Briefly, the enzyme (5 µg/ml rmGPb or rmGPa, and 1
21

ACCEPTED MANUSCRIPT
μg/mL hlGPa) was assayed in a 30 mM imidazole/HCl buffer (pH 6.8) containing 60 mM KCl, 0.6 mM
EDTA, and 0.6 mM dithiothreitol using constant concentrations of glycogen (0.2% w/v), AMP (1 mM;
only for the rmGPb experiments), and various concentrations of Glc-1-P and inhibitors. For the
measurement of the Ki for α-D-glucose for hlGPa solutions of four different concentrations were used 2
min after dissolution while the overall assay time was 1 minute following a previously described
method.46 Initial velocities were calculated from the pseudo-first order rate constants. Kinetic data
presented in the form of Lineweaver-Burk, or Dixon plots were analysed for the calculation of the kinetic
parameters by the non-linear regression program GRAFIT.58
X-ray crystallography
Tetragonal T state rmGPb crystals were grown as described previously.59 These crystals belong to space
group P43212 with unit cell dimensions a=b=128.6 Å and c=116.3 Å. rmGPb – inhibitor complexes were
formed by diffusion of a 10 mM solution of the inhibitor in the crystallization media (supplemented with
15% (v/v) DMSO) in preformed rmGPb crystals at room temperature for 6 hours prior to data collection.
X-ray diffraction data were collected using synchrotron radiation on station ID911-2 at MAX-Lab
Synchrotron Radiation Source in Lund, Sweden. Crystal orientation, integration of reflections, inter-
frame scaling, partial reflection summation, and data reduction was performed by the program
Mosflm.60 Scaling and merging of intensities were performed by SCALA.61 Crystallographic refinement of
the complexes was performed by maximum-likelihood methods using REFMAC.61 The starting model
employed for the refinement of the complexes was the structure of the native T state rmGPb complex
determined at 1.9 Å resolution (Leonidas et al., unpublished results). Ligand molecule coordinates and
topologies were constructed using JLigand62 and they were fitted to the electron density maps after
adjustment of their torsion angles. Alternate cycles of manual rebuilding with the molecular graphic
22

ACCEPTED MANUSCRIPT
program COOT61 and refinement with REFMAC63 improved the quality of the models. A summary of the
data processing and refinement statistics for the inhibitor complex structures is given in Table 2. The
stereochemistry of the protein residues was validated by MolProbity.64 Hydrogen bonds and van der
Waals interactions were calculated with the program CONTACT as implemented in CCP461 applying a
distance cut off 3.3 Å and 4.0 Å, respectively. Figures were prepared with PyMol.65 The coordinates of
the new structures have been deposited with the RCSB Protein Data Bank (http://www.rcsb.org/pdb)
with codes presented in Table 3.
Table 3. Summary of the diffraction data processing and refinement statistics for the rmGPb complexes.
Values in parentheses are for the outermost shell.
GPb complex
8a
8b
Data collection and processing statistics
Resolution (Å)
Outermost shell (Å)
Reflections measured
Unique reflections
Multiplicity
38.39 – 1.85
38.76 – 1.85
1.95 – 1.85
1.95 – 1.85
323721 (46061)
82598 (11847)
3.9 (3.9)
326554 (46626)
74394 (11353)
4.4 (4.1)
Rsymm
0.069 (0.428)
99.3 (98.7)
0.111 (0.450)
89.9 (94.8)
Completeness (%)
23

ACCEPTED MANUSCRIPT
<I/σI >
12.2 (3.3)
4.9 (2.6)
Refinement statistics
Rcryst
0.158 (0.240)
0.188 (0.260)
321
0.170 (0.240)
0.203 (0.286)
324
Rfree
No of solvent molecules
r.m.s. deviation from ideality
in bond lengths (Å)
in angles (°)
0.01
1.3
0.01
1.3
Average B factor (Å2)
Protein atoms
Solvent molecules
Inhibitor atoms
PDB entry
30.7
36.9
18.0
5JTT
29.8
36.1
17.4
5JTU
Acknowledgments
This project was implemented under the "ARISTEIA" Action of the "Operational Programme Education
and Lifelong Learning" and is co-funded by the European Social Fund (ESF) and National Resources. This
24

ACCEPTED MANUSCRIPT
work was supported in part by the Postgraduate Programmes ‘‘Biotechnology-Quality assessment in
Nutrition and the Environment”, ‘‘Application of Molecular Biology-Molecular Genetics-Molecular
Markers”, Department of Biochemistry and Biotechnology, University of Thessaly. Work at the
Synchrotron Radiation Sources, MAX-lab, Lund, Sweden and EMBL Hamburg Outstation, Germany, was
supported from the EU FP7 Programme (FP7/2007-2013) under BioStruct-X (grant agreement
N°283570). The synthetic work was supported by the Hungarian Scientific Research Fund (Grants OTKA
PD 105808, ANN 110821). The BAROSS REG_EA_09-1-2009-0028 (LCMS_TAN) project also contributed
to the implementation of the study.
Supporting information
Detailed description of synthetic procedures and compound characterization, copies of NMR spectra of
the new compounds.
References
1.
Oikonomakos, N. G. Glycogen phosphorylase as a molecular target for type 2 diabetes therapy.
Curr. Protein Pept. Sci. 2002, 3, 561-586.
25

ACCEPTED MANUSCRIPT
2.
Barford, D.; Johnson, L. N. The Allosteric Transition of Glycogen-Phosphorylase. Nature 1989,
340, 609-616.
3.
Barford, D.; Hu, S. H.; Johnson, L. N. Structural Mechanism for Glycogen-Phosphorylase Control
by Phosphorylation and AMP. J. Mol. Biol. 1991, 218, 233-260.
4.
Somsák, L.; Czifrák, K.; Tóth, M.; Bokor, É.; Chrysina, E. D.; Alexacou, K. M.; Hayes, J. M.; Tiraidis,
C.; Lazoura, E.; Leonidas, D. D.; Zographos, S. E.; Oikonomakos, N. G. New inhibitors of glycogen
phosphorylase as potential antidiabetic agents. Curr. Med. Chem. 2008, 15, 2933-2983.
5.
glycogen phosphorylase: potential treatment for type 2 diabetes. Phytochem. Rev. 2014, 13, 471-498.
6. Rath, V. L.; Ammirati, M.; Danley, D. E.; Ekstrom, J. L.; Gibbs, E. M.; Hynes, T. R.; Mathiowetz, A.
Hayes, J.; Kantsadi, A.; Leonidas, D. Natural products and their derivatives as inhibitors of
M.; McPherson, R. K.; Olson, T. V.; Treadway, J. L.; Hoover, D. J. Human liver glycogen phosphorylase
inhibitors bind at a new allosteric site. Chem. Biol. 2000, 7, 677-682.
7.
Phosphorylase Inhibitors. Mini-Rev. Med. Chem. 2010, 10, 1175-1187.
8. Kurukulasuriya, R.; Link, J. T.; Madar, D. J.; Pei, Z.; Richards, S. J.; Rohde, J. J.; Souers, A. J.;
Agius, L. Physiological Control of Liver Glycogen Metabolism: Lessons from Novel Glycogen
Szczepankiewicz, B. G. Potential drug targets and progress towards pharmacologic inhibition of hepatic
glucose production. Curr. Med. Chem. 2003, 10, 123-153.
9.
Agius, L. New hepatic targets for glycaemic control in diabetes. Best Pract. Res. Clin. Endocrin.
Metab. 2007, 21, 587-605.
10.
Treadway, J. L.; Mendys, P.; Hoover, D. J. Glycogen phosphorylase inhibitors for treatment of
type 2 diabetes mellitus. Expert Opin. Invest. Drugs 2001, 10, 439-454.
11.
Tracey, W. R.; Treadway, J. L.; Magee, W. P.; Sutt, J. C.; McPherson, R. K.; Levy, C. B.; Wilder, D.
E.; Yu, L. J.; Chen, Y.; Shanker, R. M.; Mutchler, A. K.; Smith, A. H.; Flynn, D. M.; Knight, D. R.
26

ACCEPTED MANUSCRIPT
Cardioprotective effects of ingliforib, a novel glycogen phosphorylase inhibitor. Am. J. Physiol.-Heart
Circul. Physiol. 2004, 286, H1177-H1184.
12.
Treadway, J. L.; Magee, W. P.; Hoover, D. J.; McPherson, R. K.; Martin, W. H.; Zavadoski, W. J.;
Gibbs, E. M.; Tracey, W. R. Cardioprotective effect of the glycogen phosphorylase inhibitor CP-380867.
Diabetes 2000, 49, A127-A127.
13.
Favaro, E.; Bensaad, K.; Chong, M. G.; Tennant, D. A.; Ferguson, D. J. P.; Snell, C.; Steers, G.;
Turley, H.; Li, J.-L.; Günther, U. L.; Buffa, F. M.; McIntyre, A.; Harris, A. L. Glucose Utilization via Glycogen
Phosphorylase Sustains Proliferation and Prevents Premature Senescence in Cancer Cells. Cell Metab.
2012, 16, 751-764.
14.
Zois, C. E.; Favaro, E.; Harris, A. L. Glycogen metabolism in cancer. Biochem. Pharmacol. 2014,
92, 3-11.
15.
Lew, C. R.; Guin, S.; Theodorescu, D. Targeting glycogen metabolism in bladder cancer. Nat. Rev.
Urol. 2015, 12, 383-391.
16.
Zois, C. E.; Harris, A. L. Glycogen metabolism has a key role in the cancer microenvironment and
provides new targets for cancer therapy. J. Mol. Med. 2016, 94, 137-154.
17.
to cerebral ischemia. Mini Rev Med Chem 2010, 10, 1188-93.
18. Guan, T.; Qian, Y. S.; Tang, X. Z.; Huang, M. H.; Huang, L. F.; Li, Y. M.; Sun, H. B. Maslinic Acid, a
Xu, L.; Sun, H. Pharmacological manipulation of brain glycogenolysis as a therapeutic approach
Natural Inhibitor of Glycogen Phosphorylase, Reduces Cerebral Ischemic Injury in Hyperglycemic Rats by
GLT-1 Up-Regulation. J. Neurosci. Res. 2011, 89, 1829-1839.
19.
Somsák, L. Glucose derived inhibitors of glycogen phosphorylase. Compt. Rend. Chimie 2011, 14,
211-223.
27

ACCEPTED MANUSCRIPT
20.
Somsák, L.; Nagy, V.; Hadady, Z.; Docsa, T.; Gergely, P. Glucose analog inhibitors of glycogen
phosphorylases as potential antidiabetic agents: recent developments. Curr. Pharma. Design 2003, 9,
1177-1189.
21.
Curr. Opin. Invest. Drugs 2008, 9, 379-395.
22. Somsák, L.; Nagy, V.; Hadady, Z.; Felföldi, N.; Docsa, T.; Gergely, P. Recent developments in the
Oikonomakos, N. G.; Somsák, L. Recent advances in glycogen phosphorylase inhibitor design.
synthesis and evaluation of glucose analog inhibitors of glycogen phosphorylases as potential
antidiabetic agents. In Frontiers in Medicinal Chemistry, Reitz, A. B.; Kordik, C. P.; Choudhary, M. I.;
Rahman, A. u., Eds. Bentham: 2005; Vol. 2, pp 253-272.
23.
of glycogen phosphorylase: potential treatment for type 2 diabetes. Phytochem. Rev. 2014, 13, 471-498.
24. Praly, J. P.; Vidal, S. Inhibition of Glycogen Phosphorylase in the Context of Type 2 Diabetes, with
Focus on Recent Inhibitors Bound at the Active Site Mini-Rev. Med. Chem. 2010, 10, 1102-1126.
25. Hadady, Z.; Tóth, M.; Somsák, L. C-(β-D-glucopyranosyl) heterocycles as potential glycogen
phosphorylase inhibitors. Arkivoc 2004, (vii), 140-149.
26. Chrysina, E. D.; Kosmopolou, M. N.; Tiraidis, C.; Kardarakis, R.; Bischler, N.; Leonidas, D. D.;
Hayes, J. M.; Kantsadi, A. L.; Leonidas, D. D. Natural products and their derivatives as inhibitors
Hadady, Z.; Somsák, L.; Docsa, T.; Gergely, P.; Oikonomakos, N. G. Kinetic and crystallographic studies on
2-(β-D-glucopyranosyl)-5-methyl-1,3,4-oxadiazole, -benzothiazole, and -benzimidazole, inhibitors of
muscle glycogen phosphorylase b. Evidence for a new binding site. Protein Sci. 2005, 14, 873-888.
27.
Benltifa, M.; Vidal, S.; Fenet, B.; Msaddek, M.; Goekjian, P. G.; Praly, J.-P.; Brunyánszki, A.; Docsa,
T.; Gergely, P. In the Search of Glycogen Phosphorylase Inhibitors: 5-Substituted 3-C-Glucopyranosyl-
1,2,4-Oxadiazoles from β-D-Glucopyranosyl Cyanides upon Cyclization of O-Acyl-amidoxime
Intermediates. Eur. J. Org. Chem. 2006, 4242-4256.
28