Selective α-glucosidase substrates and inhibitors containing short aromatic peptidyl moieties

Article abstract of DOI:10.1016/j.bmcl.2011.02.062

We constructed a library of sugar-dipeptide conjugate to find out the best complementary against hydrophobic pocket of α-glucosidase. The best substrate showed 150-fold improved K_m value relative p-acetaminophenyl-α-d-glucopyranoside for α-glucosidase from Bacillus stearothermophillus. Using information from the complementary, we synthesized sp-WY and β-Glc-sp-WY, which selectivity inhibited the cognate enzyme.

Full text of DOI:10.1016/j.bmcl.2011.02.062

Bioorganic & Medicinal Chemistry Letters 21 (2011) 2441–2444
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Selective
a-glucosidase substrates and inhibitors containing short aromatic
peptidyl moieties
⇑
Sangeun Park, Soonsil Hyun, Jaehoon Yu
Department of Chemistry & Education, Seoul National University, Seoul 151-742, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
We constructed a library of sugar-dipeptide conjugate to find out the best complementary against
Received 27 December 2010
Revised 11 February 2011
Accepted 15 February 2011
hydrophobic pocket of
p-acetaminophenyl-
a
-glucosidase. The best substrate showed 150-fold improved Km value relative
a
-
D
-glucopyranoside for -glucosidase from Bacillus stearothermophillus. Using infor-
a
mation from the complementary, we synthesized sp-WY and b-Glc-sp-WY, which selectivity inhibited
the cognate enzyme.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
a-Glucoside-dipeptidyl conjugate
Substrate library
Chromogenic assay
a-Glucosidase inhibitor
a
-Glucosidases comprise one of the most abundant families of
by properly introducing aromatic moieties into inhibitors that
can be recognized by residues outside of the carbohydrate binding
region, it might be possible to create selectivity in inhibitor binding
enzymes found in various living systems, ranging from microor-
ganism to more evolved animals and plants. Because these en-
zymes are essential for maintaining catabolic metabolism, they
are potentially important targets for treating pathogenic features
of organisms. Many types of inhibitors of members of this family
of enzymes exist, including even commercially available chemo-
therapeutic agents for the treatment of human metabolic disease.1
to different isozymes of
For the most part, typical carbohydrate containing inhibitors
have been designed to target carbohydrate binding sites of -glu-
a-glucosidases.
a
cosidases and, as a result, they bind competitively with substrates
of the enzymes. However, several substrates have been designed
that contain simple chromogenic groups covalently attached to
the glycosidic centers in carbohydrates.7 Typical chromogenic
groups incorporate aromatic residues, such as nitrophenol and
coumarin, which change color or fluorescence intensity when the
glycosidic bond is cleaved during the enzyme catalyzed hydrolysis
reactions. Thus, sufficient space must be available (+1 and +2 re-
gion) to accommodate these chromogenic moieties in the active
Some of the inhibitors of a-glucosidases contain several monosac-
charide rings and multiple hydroxyl functional groups,2 which lead
to strong nanomolar binding affinities at carbohydrate binding
sites. However, substances of this type generally lack specificity
in which they inhibit a range of isozymes.
Other inhibitors of a-glucosidases, which do not contain carbo-
hydrate structures and polyhydroxyl substitution, derive from nat-
ural sources.3 In the family of non-carbohydrate type inhibitors,
aromatic residues are frequently observed to be the key pharmaco-
phores. Consequently, substances of this type display noncompet-
itive or mixed type inhibition owing to the fact that they partially
share binding sites with carbohydrate substrates. In the early re-
sites of
a-glucosidases. This portion of the enzyme active sites,
however, has seldom been targeted in the design of inhibitors for
a
-glucosidases and their isozymes.8
We hypothesized that inhibitors, which incorporate appropriate
aromatic residues that interact with groups in the non-carbohydrate
recognizing areas of these enzymes, might display binding selec-
tively to the target enzymes. Consequently, we felt that a study
involving construction and evaluation of libraries of various aro-
matic ring containing substances that can bind to the non-carbohy-
ported crystal structures of endo-a-glucosidases, roomy binding
sites (+1 and +2 region) have been observed, which are composed
not only of hydrophilic amino acid side chains but also hydropho-
bic4 and aromatic5 groups of amino acids that interact with aro-
matic substituents of the inhibitors through
p–p
interactions.
drate region, might uncover selective a-glucosidase substrates and
Furthermore, since these regions are distant from the conserved
catalytic core carboxylate residues, different amino acid residues
are frequently located at distant sites in various isozymes.6 Thus,
inhibitors. Below, we describe the results of an investigation in
which a library of peptidyl glucose substrates was prepared and
their activities against various a-glucosidases were carried out.
Substrates prepared in the initial library were designed to
contain tethered aromatic amino acid groups on opposite sides of
carbohydrate moieties. Since the plan was to have substrate
⇑
Corresponding author.
E-mail address: jhoonyu@snu.ac.kr (J. Yu).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.02.062

2442
S. Park et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2441–2444
OH
structures that are complementary with amino acid side chains in
the hydrophobic pockets of the enzymes, only three aromatic ami-
no acids (Trp, Phe, and Tyr) were used as building blocks. The nitro
group of p-nitrophenyl-a-D-glucopyranoside was reduced to
produce the corresponding aniline derivative, which enabled intro-
duction of a succinamide side chain required to make the carbohy-
drate suitable for solid phase peptide synthesis. Fmoc-protected
aromatic amino acids were used to make nine dimeric peptides
that were employed in reactions with 2,3,4,6-O-tetraacetyl-(p-suc-
O
HO
HO
O
R1
O
HO
H
N
H
N
O
N
H
NH2
O
O
R2
α-glc-sp-XX
: R1 = side chain of F,W, or Y
R2 =side chain of F,W, or Y
cinyl)anilino-
protected dipeptidyl-
moval followed by cleavage from the solid support afforded the
peptidyl substrates ( -Glc-sp-XX, Fig. 1) that were purified by
HPLC, giving 30–40% isolated yields for each substrate (Scheme 1).
-Glucosidases from Bacillus stearothermophillus, rice, and rat
a-
D-glucopyranoside to produce the corresponding
a-D
-glucopyranosides. Protecting group re-
a
HN
O
O
a
H
N
H
N
HO
intestine were used to assay the substrate activities of the peptidyl
substrate library. For initial screening, equimolar mixtures of
individual members of the substrate library along with the typical
N
H
NH2
O
O
sp-WY
non-peptidyl
substrate,
p-nitrophenyl-a-D-glucopyranoside
OH
(pNP-a
-D-Glc) were incubated with each enzyme while monitoring
the production of the chromogenic product, p-nitrophenol.9 These
assays led to interesting results. Firstly, the rates of formation of
p-nitrophenol were observed to be low for every sample relative
to that of the non-peptidyl substrate alone (Fig. 2). Secondly, the
peptidyl substrates caused a greater reduction in the activity of
HN
OH
O
HO
HO
O
O
H
N
H
N
O
HO
N
NH2
H
O
O
OH
the B. stearothermophillus
a-glucosidase as compared to those of
β-Glc-sp-WY
O
the rice and rat intestine enzymes as judged by the more greatly
reduced, peptidyl substrate induced rates of production of
p-nitrophenol (Fig. 2). These findings suggest that differences exist
in the chromogenic recognition regions of enzymes, which lead to
the differences in substrate susceptibilities. More importantly, the
peptidyl substrates that brought about the greatest reduction in
OH
HO
OH
HO
HO
O
O
HO
O
R1
O
HO
H
H
O
N
N
N
H
NH2
4
O
O
R2
the rates differed in all three of the
a-glucosidases. For example,
a
-Glc-sp-WY caused the largest decrease in p-nitrophenol produc-
α-Pnt-sp-XX
: R1 = side chain of F,W, or Y
tion for the
peptidyl substrates
effect on catalysis by the respective rice and rat intestine enzymes.
The observations indicate that the peptidyl substrates are compet-
itive inhibitors of all of the glucosidases owing to the fact that they
a
-glucosidase from B. stearothermophillus, whereas the
-Glc-sp-YF and -Glc-sp-FF had the largest
a
a
R2 = side chain of F,W, or Y
Figure 1. Structures of inhibitors and substrate libraries. Synthesises of these
compounds are described in Supplementary data.
OH
OAc
1.H2, Pd/C,MeOH, rt, 14 h
Ac2O, DMAP
Pyr, rt, 14 h
NO2
O
O
HO
HO
OAc
AcO
2.Succinic anhydride,
DIPEA, DMF, rt, 3 h
NO2
HO
AcO
O
O
1
2
R1
O
H
N
resin
H2N
N
OAc
O
H
OAc
O
R2
O
AcO
AcO
AcO
AcO
O
R1
O
HCTU, HOBt,
DIPEA, 2 h
AcO
O
H
N
H
N
resin
AcO
H
N
O
O
N
H
N
H
OH
O
O
R2
O
4
OH
O
HO
HO
O
R1
O
HO
H
N
H
N
1. 7N NH3 in MeOH, 1 h
O
N
H
NH2
2. Cleavage, 1 h
O
O
R2
α-Glc-sp-XX
: R1 = side chain of F,W, or Y
R2 = side chain of F,W, or Y
Scheme 1. Synthesis of peptidyl glucose library.

S. Park et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2441–2444
2443
Figure 2. Initial screening of the substrate library (a-Glc-sp-XX) using p-nitrophenyl-a-D-glucose (pNP-a-D-Glc) as a competitive substrate.
contain structural features that are complementary to then on-
carbohydrate recognizing regions of these enzymes. Owing to the
significantly large effects observed on the rates of p-nitrophenol
results showed that the
was completely inhibited by b-Glc-sp-WY (Fig. 1), which displayed
a high affinity for the enzyme (Ki = 110 M, Fig. 3). Importantly, b-
a-glucosidase from B. stearothermophillus
l
formation and the selectivity among substrates, the
from B. stearothermophillus was chosen for further study.
a
-glucosidase
Glc-sp-WY was observed to display almost no inhibition against
the activities of the other isozymes (Table 2).9 The data suggest
that the role played by the carbohydrate group of the inhibitor is
merely to anchor the inhibitor to the active site and, moreover, that
it does not govern binding selectivity of the inhibitors.
Two possible reasons exist for the reduced production of
p-nitrophenol. The first is that the selected peptidyl substrates
are competitive inhibitors of the glucosidase catalyzed reaction
of the chromogenic substrate. Another possibility is that the prod-
ucts of hydrolysis of the peptidyl substrates might inhibit the
activity of these enzymes. Both explanations suggest that the com-
plementary structures of each the glucosidases are recognized by
peptidyl residues in the substrates. Among the first generation
Owing to the fact that they are categorized as members of the
a
-glucosidase family10 and are important targets for dietary and
medical purposes, salivary and pancreatic
plored next. Since they are more accessible compared to the corre-
sponding pyranosides, p-nitrophenyl- -pentaosides were used
to synthesize a 9 peptidyl substrates ( -Pnt-sp-XX, Fig. 1) library
to probe the effects on
-amylase activities.9 An equimolar mixture
of each peptidyl substrate and p-nitrophenyl- -pentaoside was
a-amylases were ex-
a-D
substrates explored,
a-Glc-sp-WY displayed the most profound
a
inhibitory effect. The Km value of this substrate was determined
and compared with those with p-nitrophenyl- and p-acetamin-
a
a-D
ophenyl-
-Glc-sp-WY was about 90-fold and 150-fold less than that of
the two non-peptidyl substrates (Table 1). This finding demon-
strates that the peptidyl part of -Glc-sp-WY is well-recognized
a-D-glucopyranoside. As expected, the Km value of
a
a
by the complementary region of the cognate enzyme.
We next designed selective glucosidase inhibitors using the
information gained from the above studies. The first question we
probed was if the non-carbohydrate part of the substrate might it-
self be a glucosidase inhibitor. Indeed, studies showed that sp-WY
(Fig. 1) was a low millimolar (Ki = 1.4 mM) inhibitor of the Bacillus
a-glucosidase and that it did not inhibit the other
a-glucosidases.
A second generation inhibitor, in which the peptide side chain of
a
-Glc-sp-WY was directly bonded to the carbohydrate backbone
Figure 3. Dixon plot for inhibition by b-Glc-sp-WY of the
Bacillus stearothermophillus. Each data point was triplicated and averaged.
a-glucosidase from
via non-hydrolysable b-glycosidic linkage, was studied.9 The assay
Table 1
Table 2
Enzymatic constants of
a-glucosidase from Bacillus stearothermophillus with various
substratesa
Comparison of inhibitory effects between the selected peptides and known inhibitors
using various
a
-glucosidasesa
pNP-a-
D-Glc
p-Acetaminophenyl-
-glucopyranosideb
a-Glc-sp-WY
sp-WY (
lM)
b-Glc-sp-WY
M)
Acarbose (lM)
a-D
(
l
Km (mM)
1.2
0.019
9.5
1.9
0.0015
0.76
0.013
0.00025
0.13
mmax
(l
M sÀ1
)
)
Bacillus stearothermophillus
Rice
Rat intestine
1400
NCb
NC
110
NC
NC
0.34
4.2
3.5
kcat (sÀ1
kcat/Km(mMÀ1 sÀ1
)
7.6
0.40
10
a
a
Ki value was calculated by Dixon plot. Measurement of each value was tripli-
Km value was calculated by Michaelis–Menten kinetics. Measurement of each
cated and averaged.
value was triplicated and averaged.
b
b
NC, the constants could not be calculated because of barely detectable product
The structure and synthesis of this compound were described in Supplementary
trace or little difference from those in the absence of the inhibitor.
data.

2444
S. Park et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2441–2444
Figure 4. Initial screening of the substrate library (a-Pnt-sp-XX) using p-nitrophenyl-a-D-pentaoside (pNP-a-D-Pnt) as a competitive substrate and various c-amylases. Each
data point was triplicated and averaged.
incubated with each enzyme.9 Reduced rates of formation of p-
nitrophenol were observed in each case in comparison with that
of the non-peptidyl chromogenic substrate alone. However, the
rate reductions were not as large as those seen in studies of the
and that it might be generally applied to the design of inhibitors
of any glucosidase.
Acknowledgments
a
-glucosidases (Fig. 4). Certain peptidyl substrates and their corre-
sponding hydrolysis products showed the same selectivity for both
-amylases. Even though the peptidyl substrates have a reduced
impact on the retardation of both -amylases compared to those
for -glucosidases, the observation show that strategy we have de-
vised might be general. The lower inhibition seen with the -amy-
This work was supported by a fund from Korea Research Foun-
dation (20100002030).
a
a
a
Supplementary data
a
lase substrates could be a consequence of the fact that peptidyl
moieties in these substances are small relative to the large pentao-
side unit. If this is the case, increasing the size of the non-carbohy-
drate component and/or reducing the size of the carbohydrate
group11 could lead to more affective and specific substrates for
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.02.062.
References and notes
1. Van de Laar, F. A.; Lucassen, P. L.; Akkermans, R. P.; Van de Lisdonk, E. H.;
Rutten, G. E.; Van Weel, C. Diabetes Care 2005, 28, 154.
2. (a) Jacob, G. S. Curr, Opin. Struct. Biol. 1995, 5, 605; (b) Scott, L. J.; Spencer, C. M.
Drugs 2000, 59, 521.
the a-amylases.
In summary, by carrying out assays, we were able to explore
substrates that have peptidyl structural components that are
3. (a) Lee, D. S.; Lee, S. H. FEBS Lett. 2001, 501, 84; (b) Yoshikawa, M.; Murakami,
T.; Shimada, H.; Matsuda, H.; Yamahara, J.; Tanabe, G.; Muraoka, O. Tetrahedron
Lett. 1997, 38, 8367; (c) de Melo, E. B.; Gomes, A. S.; Carvalho, I. Tetrahedron
2006, 62, 10277; (d) Gao, H.; Kawabata, J. Bioorg. Med. Chem. Lett. 2008, 18, 812.
4. (a) Ramasubbu, N.; Paloth, V.; Luo, Y.; Brayer, G. D.; Levine, M. J. Acta Crystallogr.
Sect. D 1996, 52, 435; (b) Ramasubbu, N.; Ragunath, C.; Mishra, P. J. J. Mol. Biol.
2003, 325, 1061; (c) Mishra, P. J.; Ragunath, C.; Ramasubbu, N. Biochem. Biophys.
Res. Commun. 2002, 292, 468; (d) Qian, M.; Haser, R.; Buisson, G.; Duee, E.;
Payan, F. Biochemistry 1994, 24, 6284; (e) Kadziola, A.; Sogaard, M.; Svensson,
B.; Haser, R. J. J. Mol. Biol. 1998, 278, 205.
complementary to the hydrophobic pocket of
best substrate, -glucose-spacer-Trp-Tyr ( -Glc-sp-WY) showed
a 150-fold lesser Km value (13 M) relative to that of p-acetamin-
ophenyl- -glucopyranoside (1.9 mM) with the -glucosidase
from B. stearothermophillus. A substrate lacking the carbohydrate
component, sp-WY, was observed to be only a moderate inhibitor
(1.4 mM) of the B. stearothermophillus glucosidase but it possessed
high selectivity for this enzyme in comparison to other isozymes.
With the goal of making a more potent inhibitor, this non-carbohy-
drate structure was tethered to a glucose moiety via a b-glycoside
linkage. The resulting substance, b-Glc-sp-WY was an exception-
a-glucosidases. The
a
a
l
a-
D
a
5. Watanabe, K.; Hata, Y.; Kizaki, H.; Katsube, Y.; Suzuki, Y. J. Mol. Biol. 1997, 269,
142.
6. (a) Kitamura, M.; Okuyama, M.; Tanzawa, F.; Mori, H.; Kitago, Y.; Watanabe, N.;
Kimura, A.; Tanaka, I.; Yao, M. J. Biol. Chem. 2008, 283, 36328; (b) Brumshtein,
B.; Greenblatt, H. M.; Butters, T. D.; Shaaltiel, Y.; Aviezer, D.; Silman, I.;
Futerman, A. H.; Sussman, J. L. J. Biol. Chem. 2007, 282, 29052.
7. Chapdelaine, P.; Tremblay, R. R.; Dube, J. Y. Clin. Chem. 1978, 24, 208.
8. Kandra, L.; Zajácz, A.; Remenyik, J.; Gyémánt, G. Biochem. Biophys. Res. Commun.
2005, 334, 824.
ally moderate but selective inhibitor (Ki = 110
lM) of the Bacillus
glucosidase. Moreover, -glucosidases from other origins were
a
not inhibited by this substance, a result which suggests that the
peptidyl moiety is essential for selectivity. This substance is much
more selective against the cognate enzyme than commercially
available inhibitor, acarbose. This effort has demonstrated that a
strategy employing aromatic peptides as an additive recognition
tool is viable for the generation of specific glucosidase inhibitors
9. See Supplementary data.
10. EC number of
a
-amylase from Human saliva Type XIII-A is 3.2.1.1 and EC
-amylase from Porcine pancreas Type VI-B is 3.2.1.1. And EC
-glucosidase from Bacillus stearothremophillus is 3.2.1.20.
number of
number of
a
a
11. Maurus, R.; Begum, A.; Kuo, H. H.; Racaza, A.; Numao, S.; Andersen, C.; Tams, J.
W.; Vind, J.; Overall, C. M.; Withers, S. G.; Brayer, G. D. Protein Sci. 2005, 14, 743.