Synthesis, Biological Activity, and Molecular Modeling Studies of 1H-1,2,3-Triazole Derivatives of Carbohydrates as a-Glucosidases Inhibitors

Article abstract of DOI:10.1021/jm901265h

A class of drugs in use for treating type II diabetes mellitus (T2D), typified by the pseudotetrasaccharide acarbose, act by inhibiting the a-glucosidase activity present in pancreatic secretions and in the brush border of the small intestine. Herein, we report the synthesis of a series of 4-substituted 1,2,3-triazoles conjugated with sugars, including D-xylose, D-galactose, D-allose, and D-ribose. Compounds were screened for a-glucosidase inhibitory activity using yeast maltase (MAL12) as a model enzyme. Methyl2,3-O-isopropylidene-β-D-ribofuranosides, such as the 4-(l-cyclohexenyl)-l,2,3-triazole derivative, were among the most active compounds, showing up to 25-fold higher inhibitory potency than the complex oligosaccharide acarbose. Docking studies on a MAL12 homology model disclosed a binding mode consistent with a transition-state-mimicking mechanism. Finally, the actual pharmacological potential of this triazole series was demonstrated by the reduction of postprandial blood glucose levels in normal rats. These compounds could represent new chemical scaffolds for developing novel drugs against T2D.

Full text of DOI:10.1021/jm901265h

2364 J. Med. Chem. 2010, 53, 2364–2375
DOI: 10.1021/jm901265h
Synthesis, Biological Activity, and Molecular Modeling Studies of 1H-1,2,3-Triazole Derivatives of
Carbohydrates as r-Glucosidases Inhibitors
Sabrina B. Ferreira, ,† Ana C. R. Sodero,‡ Mariana F. C. Cardoso, Emerson S. Lima,§ Carlos R. Kaiser,† Floriano P. Silva, Jr.,‡
and Vitor F. Ferreira*,
†
‡
Instituto de Quımica, Universidade Federal do Rio de Janeiro, LABRMN, Ilha do Fundao, 21949-900, Rio de Janeiro, RJ, Brazil, Laboratorio de
´
Bioquımica de Proteınas e Peptıdeos, Fundac-ao Oswaldo Cruz, Instituto Oswaldo Cruz, 21045-900, Rio de Janeiro, RJ, Brazil, Faculdade de
~
ꢀ
§
´
´
´
~
´
^
Cie^ncias da Sauꢀde, Universidade Federal do Amazonas, 69010-300, Manaus, AM, Brazil, and Departamento de Quımica Organica, Instituto de
´
ꢀ
Quımica, Universidade Federal Fluminense, CEG, Campus do Valonguinho, 24210-141, Niteroi, RJ, Brazil
Received August 24, 2009
A class of drugs in use for treating type II diabetes mellitus (T2D), typified by the pseudotetrasaccharide
acarbose, act by inhibiting the R-glucosidase activity present in pancreatic secretions and in the brush
border of the small intestine. Herein, we report the synthesis of a series of 4-substituted 1,2,3-triazoles
conjugated with sugars, including D-xylose, D-galactose, D-allose, and D-ribose. Compounds were
screened for R-glucosidase inhibitory activity using yeast maltase (MAL12) as a model enzyme. Methyl-
2,3-O-isopropylidene-β-D-ribofuranosides, such as the 4-(1-cyclohexenyl)-1,2,3-triazole derivative,
were among the most active compounds, showing up to 25-fold higher inhibitory potency than the
complex oligosaccharide acarbose. Docking studies on a MAL12 homology model disclosed a binding
mode consistent with a transition-state-mimicking mechanism. Finally, the actual pharmacological
potential of this triazole series was demonstrated by the reduction of postprandial blood glucose levels in
normal rats. These compounds could represent new chemical scaffolds for developing novel drugs
against T2D.
Introduction
of the low amino acid sequence similarity between the
evolutionarily distant members of this family, some highly
conserved residues are still present and structural studies
have demonstrated that all GH13 members share similar
folding.8
A glycosidic bond is a covalent chemical bond that joins
two simple sugars by means of an oxygen atom. Many
polysaccharides are formed by monosaccharides joined by
R or β glycosidic bonds. During the digestion process these
bonds are hydrolyzed by specific glycosidases liberating the
carbohydrate subunits as nutrients.1
In humans, four enzymes are responsible for the complete
degradation of starch into glucose, which is then able to enter
the bloodstream by a specific transport system.2,3 First,
salivary and pancreatic R-amylases (EC 3.2.1.1) cleave the
internal R-1,4bonds of starchintoshorter linearandbranched
dextrin chains. The dextrins are then further hydrolyzed into
glucose by two small-intestinal brush-border exohydrolases
(maltase-glucoamylase, EC 3.2.1.20 and 3.2.1.3, and sucrase-
isomaltase, EC 3.2.148 and 3.2.10).4
Human pancreatic R-amylase belongs to family 13 of the
glycoside hydrolases (GH13a), which gathers a large variety
of enzymes with varying substrate and product specificities
as reflected by the 26 different EC numbers found in this
family of enzymes: glycoside hydrolases (EC 3.2.1.X, the most
abundant), glycoside transferases (EC 2.4.1.X), and even
isomerases (EC 5.4.99.15 and EC 5.4.99.16).5-7 Regardless
The R-glucosidase from Saccharomyces cerevisiae (baker’s
yeast), coded by the MAL12 gene, is also a member of the
GH13 family. It can hydrolyze nonreducing 1,4-linked R-D-
glucose residues with release of R-D-glucose (EC 3.2.1.20)
and retention of configuration at the anomeric carbon, char-
acteristic of GH13enzymes.6,9 Because of itsready availability
and ease of handling, baker’s yeast R-glucosidase (MAL12)
has been widely used in enzymatic assays to discover new
R-glucosidase inhibitors with distinct biological activities.10-16
In addition, MAL12 has been targeted in several structure-
based drug design efforts where the lack of experimental
information at the protein structure level has been mainly
overcome by the use of theoretical models based on homology
with the GH13 family members for which structural informa-
tion is available.
Enzyme-catalyzed glycosidic bond cleavage of pyranosides
is usually mediated by two catalytic carboxylic acid residues
and typically proceeds via a positively charged oxocarbenium
ion transition state.17 The charge buildup is accompanied by
planarization of the anomeric carbon and concomitant dis-
tortion of the pyranose ring out of its chair conformation.
Many natural and synthetic glycosidase inhibitors mimic the
charge buildup and/or the conformational distortion of the
transition state.11 Among them, nojirimycin (1) and deoxy-
nojirimycin(2) are two classicalglycosidaseinhibitors isolated
*To whom correspondence should be addressed. Telephone: þ 55 21
26292345. Fax: þ 55 21 26292362. E-mail: cegvito@vm.uff.br.
aAbbreviations: GH13, family 13 of the glycoside hydrolases; GTase,
4-R-glucanotransferase; Ms, mesylate; Ts, tosylate; DMF, dimethylform-
amide; DMAP, dimethylaminopyridine; PDB, Protein Data Bank; LP,
lipophilic potential; T2D, type II diabetes mellitus; HRMS, high-resolution
mass spectrometry.
r
pubs.acs.org/jmc
Published on Web 02/19/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2365
Chemistry
The diversity of chemical structures of the 1,2,3-triazole
family and their useful biological activities made these com-
pounds attractive targets in synthetic organic chemistry. The
potential of this heterocycle led us to develop a synthetic route
for preparing glycoconjugated triazoles and examine their
activity. The methodology used for obtaining the triazoles
4a-n, 5a, 6a, and 7a-c was based on the Huisgen 1,3-dipolar
cycloaddition reaction.30 In order to obtain only the 1,4
regioisomer we used the reaction conditions known as “click
chemistry”, which uses Cu(I) as a catalyst. In addition to
increasing the reaction rate, the use of the copper catalyst
provides full control over reaction regioselectivity leading
only to the 1,4 regioisomer.31 As observed by Sharpless, speed
and reaction yields are much higher than with the traditional
method, with yields above 80% and 100% of regioselectivity.
Different terminal alkynes (8a-n) were used as substrates for
the reactions with azido derivatives of ribose (9), xylose (10),
galactose (11), and allose (12). Following preparation of
the carbohydrate acetonides, a better leaving group at carbon
C5, the target of the subsequent reactions, was obtained by
tosylation or mesylation. Then the azido carbohydrates were
prepared by a nucleophilic substitution reaction with sodium
azide. According to Scheme 1, to obtain the 1,2,3-triazoles,
a solution of an azido carbohydrate (D-ribose, D-xylose,
D-allose, or D-galactose), a terminal alkyne, and copper sulfate
pentahydrate (5 mol %) was treated with the slow addition
of sodium ascorbate (15 mol %).32 The reaction mixture
was stirred for 4-6 h depending on the initial substrate.
All the triazoles were obtained with good yields and fully
characterized by 1H NMR, 13C NMR, IR (infrared) spectro-
scopy, HRMS (high-resolution mass spectra), and elemental
analysis.
Figure 1. Natural glycosidase inhibitors.
from the cultured broth of a Streptomyces sp., acting on the
early stages of glycoprotein processing.
The search for R-glucosidase inhibitors led to the discovery
of acarbose (3), a pseudotetrasaccharide isolated from the
Actinoplanes strain SE 50 as a potent sucrase inhibitor.
Acarbose consists of a polyhydroxylated aminocyclohexene
derivative (valienamine) linked via its nitrogen atom to a
6-deoxyglucose, which is itself R-1,4-linked to a maltose moiety
(Figure1).18Thiscompound inhibitsporcineintestinalsucrase
with an IC50 of 0.5 μM.19 After many clinical investigations it
was introduced onthe market in1990 for the treatmentof type
2 diabetes mellitus,2 since inhibition of pancreatic R-amylase
and intestinal R-glucosidases has demonstrated great value in
the control of postprandial blood glucose levels by slowing the
starch digestion rate.20-24 The strong inhibition of human
amylasesby acarbose (micromolarKi) isattributed both to the
partial planarity of the valienamine ring and to the presence of
strong electrostatic interactions between the carboxyl groups
at the active site and the protonated nitrogen atom of the
inhibitor.2
The importance of triazolic compounds in medicinal chem-
istry is undeniable. Particularly, 1,2,3-triazoles have gained
increased attention in the drug-discovery field since the in-
troduction of the “click” chemistry concept by Sharpless.25,26
These types of compounds can actively participate in hydro-
gen bonding and dipole-dipole interactions because of their
strong dipole moments and are extremely stable to hydrolysis
and oxidative/reductive conditions. Moreover, contrary to
other azaheterocycles, the 1,2,3-triazole ring is not protonated
at physiological pH because of its poor basicity. Hence, it has
been proposed that the nonprotonated sp2-hybridized nitro-
gen atoms of 1,2,3-triazoles may better mimic the partial
positive charge at the anomeric carbon in the transition state
of the glucosidase catalyzed reaction than the corresponding
basic nitrogen atoms of iminosugars.27 By this reasoning,
sp2-azaheterocyclic glycomimetics may actually behave as
more specific glucosidase inhibitors.
r-Glucosidase Inhibition
All compounds were initially screened for R-glucosidase
inhibition at 500 μM, and the results are shown in Table 1. All
compounds inhibited yeast maltase to some degree, and most
of them presented an inhibitory profile higher than acarbose.
Notably, most of the β-D-ribosyl derivatives 4a, 4b, 4d, 4e, 4f,
and 4g as well as the R-D-galactosyl derivative 5a and the
R-D-xylosyl derivative 7a produced the highest inhibitory
values at 500 μM (Table 1).
Next, the determination of the IC50 for some of the most
active β-D-ribosyl derivatives revealed that all of the tested
triazoles presented higher inhibitory potencies against
R-glucosidase than acarbose, with values ranging from 4 to
25 μM (Table 2). Compounds 4b, 4e, and 4g, carrying,
respectively, the 1-cyclohexenyl, phenoxymethyl, and 1-cyclo-
hexanol moieties at C4 of the triazole ring, were the most active.
Rossi and Basu28 have reported the synthesis of 1-glycosyl-4-
phenyltriazoles linked to the anomeric carbon atom of the
sugar unit, with one galactosyl derivative showing Ki for
Escherichia coli β-galactosidase of about 300 μM. In another
Molecular Modeling
work, Perion et al.29 reported the synthesis of 6-deoxyglucosyl-
ꢀ
triazoles (linked via C4 of the sugar unit) with a 2-fold higher
inhibitory activity against yeast R-glucosidase (Ki = 73 μM)
than acarbose.
In this paper, we report the synthesis of novel 1H-1,2,3-
triazoles attached to carbohydrates, their in vitro evalua-
tion as R-glucosidases inhibitors and demonstrate their
hypoglycemic activity in normal rats for two of the most
potent inhibitors. A molecular modeling study was also
undertaken to provide a structural rationale for the observed
inhibitory potencies over yeast maltase.
Structure and Validation of Modeled Yeast Maltase
(MAL12). The R-glucosidase MAL12 model consists of a
single polypeptide chain of 576 residues, similar to Bacillus
cereus and Geobacillus sp. glucosidases, which are also
members of the GH13 family. Figure 2 shows the structural
alignment between MAL12 and the templates used to con-
struct the homology model.
As depicted in Figure 3, the model is composed of three
domains: the N-terminal (residues 1-104 and 176-500), the
subdomain (residues 105-175), and the C-terminal domain

2366 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Ferreira et al.
Scheme 1. Synthesis of 1H-1,2,3-Triazole Derivatives of Carbohydrates
glucosidases of the GH13 family.11 The substrate binding
cleft is aligned between the N-terminal domain and the
subdomain. No disulfide bond was found in the structure.
The stereochemical quality of the model was evaluated
using the ProCheck program33 with 99% of the residues
found with allowed backbone conformations, according to
the Ramachandran plot. Only four residues (0.8%) lay in
generously allowed regions, and just one residue (0.2%) had
a disallowed conformation (Phe302). Other stereochemical
parameters for the MAL12 model were inside the expected
values. The compatibility between the amino acid sequence
and the environment of the amino acid side chains in the
model was further checked with the Verify-3D program,
which uses a 3D score for each residue type in each environ-
mental class to pinpoint residues with unusually low scores
that may indicate an incorrect structure.34 In the MAL12
model, around 86% of the residues had an averaged 3D-1D
score of >0.2, indicating a possible correct folding of the
protein.
Identification of the Substrate Binding Subsites on MAL12.
On the basis of the fact that the catalytic residues (Asp206,
Glu268, and Asp341 in MAL12) and the residues important
for substrate binding (Tyr63, His103, and His340 in MAL12)
are highly conserved in the GH13 family,11,35 we used the
conformation of modified acarbose in the complex with
T. maritima 4-R-glucanotransferase (PDB code 1LWJ)9 to
define the binding site for docking acarbose to MAL12. The
resulting bound conformation of acarbose in MAL12 was
consistent with the binding mode of this pseudosaccharide or
its analogues to other GH13 glucosidases. The analysis of
this modeled complex enabled the identification of four
subsites involved in substrate binding to MAL12: -1, þ1,
þ2, and þ3 (following the nomenclature of Davies et al.36).
The -1 and þ1 subsites are involved in the recognition of
the acarviosine unit in acarbose. In addition to the catalytic
residues, Asp60, Tyr63, His103, Phe169, Gln173, and Arg431
Table 1. Inhibitory Activities of 1H-1,2,3-Triazole Derivatives at 500
μM over Yeast R-Glucosidasea
compd
inhibition (%)
acarbose
4a
4b
48.1 ( 1.9
94.4 ( 7.5
98.1 ( 0.3
46.5 ( 16.4
98.0 ( 1.3
98.4 ( 3.7
99.5 ( 0.1
98.5 ( 0.3
60.2 ( 13.1
44.9 ( 3.1
41.1 ( 18
nd
4c
4d
4e
4f
4g
4h
4i
4j
4l
4m
4n
79.5 ( 2.5
nd
5a
6a
99.4 ( 0.4
73.1 ( 1.4
87.9 ( 13.5
33.3 ( 46.1
68.8 ( 20
7a
7b
7c
a nd: not determined.
Table 2. IC50 Values of β-D-Ribosyltriazoles 4a, 4b, 4d, 4e, 4g, 4m, and
Acarbose over Yeast Maltase
compd
IC50 (μmol/L)
acarbose
108.8 ( 12.3
14.9 ( 4.2
3.8 ( 0.5
15.1 ( 0.9
5.7 ( 0.3
4a
4b
4d
4e
4g
4m
5.2 ( 0.9
24.7 ( 11.8
(residues 500-576). The catalytic residues (Asp206, Glu268,
and Asp341) are found in the N-terminal domain, as with all

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2367
Figure 2. Sequence alignment of MAL12, oligo-1,6-glucosidase from Bacillus cereus (PDB code 1UOK), and GH13 R-glucosidase from
Geobacillus sp (GSJ, PDB code 2ZE0). The alignment was shaded according to the Blosum62 matrix at the 100% conservation level.
Figure 3. Graphic representation of the homology modeling structure of MAL12. The N-terminal, subterminal, and C-terminal domains are
shown in blue, green, and orange, respectively. The catalytic residues are highlighted.

2368 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Ferreira et al.
surround the valienamine moiety while residues Thr207 and
His340 are closer to the 6-deoxyglucose subunit. The þ2 subsite
is formed by residues Phe149, His231, Asn233, and Ala270,
while the þ3 subsite is formed by residues His271, Glu296,
Thr299, and Phe303. In contradiction to the observations of
Tomich et al.,38 we found that the MAL12 structure could not
accommodate a -2 subsite (occupied by the 6-deoxyglucose
moiety in the nonreducing terminal of modified acarbose
modeled by these authors). In our model of MAL12, the
residues Asp60, Tyr63, Phe150, Asn173, and Arg431 form
the bottom of a substrate binding cleft that effectively blocks
the binding of an oligosaccharide extending beyond the -1
residue. This observation is consistent with the known specifi-
city of MAL12, which removes an R-linked glucose unit from
the nonreducing end of a carbohydrate chain.
Molecular Docking of 1,2,3-Glucotriazoles. All of the
compounds used in the docking study consist of a triazole
moiety coupled via a β-glycosidic linkage to D-ribose mono-
acetonide, and the chemical differences lie in their distinct R2
substituents (see Table 2). The MAL12-ligand complexes
were generated by docking the triazoles into the enzyme
active site, assuming a competitive inhibition mechanism,
similar to acarbose2 (Figure 4).
A visual inspection of the highest-scoring docking con-
formations for the triazole compounds revealed a recurrent
configuration where the R2 substituent was located at the -1
subsite, corresponding to the position of the valienamine
moiety in acarbose. Moreover, in this position the triazole
ring is found occupying the þ1 subsite, where it makes a
hydrogen bond with Glu268, while the D-ribose monoaceto-
nide moiety binds to the region encompassing the þ2 subsite
and part of the þ3 subsite (Figure 4). This binding mode is
consistent with a transition-state-mimicking mechanism, as
proposed for other triazoles,32 and acknowledges the im-
portance of the interaction with Glu268 as a significant
feature in the inhibition of yeast R-glucosidase and other
GH13 enzymes, as was previously proposed.15 Another
point considered to identify the most likely binding mode
for the ligands was the placement of the R2 group in the -1
subsite, since compounds with bulk R2 substituents (except
for 4a) were among the most potent of the series under study,
which indicates that this molecular region is likely to be
completely buried inside the enzyme.
Figure 4. Superposition of the docked conformations of acarbose
(red) and compound 4b (cyan) in the MAL12 binding site. The
protein surface was colored by the lipophilic potential (LP) calcu-
lated using the MOLCAD module of the Sybyl molecular modeling
package (Tripos Associates, St. Louis, MO). The color ramp ranges
from brown (most lipophilic) to blue (most hydrophilic). In the
detail (inset), the molecular surface regions corresponding to the
protein segments Gln58-Asp64, Glu397-Ser416, Lys147-Ser153,
Tyr211-His231, Arg167-Arg172, Ser300-Asn306 were omitted to
render a clearer view of the floor of the substrate binding cleft in
MAL12.
Figure 5. Binding mode of compound 4b in the MAL12 active site: (a) important residues for 4b binding in the MAL12 active site, according to
visual inspection, considering the amino acid residues in a radius of 6 A around the ligand; (b) Ligplot40 cartoon illustrating compound 4b
binding to MAL12.
˚

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2369
Figure 6. Comparison of the atomic contacts established between MAL12 and the ligands 4b, 4m stereoisomers R and S (4mR and 4mS), and
acarbose.
Figure 5 depicts the binding mode of compound 4b,
one of the most potent compounds of the series under
study. It is noted that the important residues for substrate
binding and/or catalysis39 are present in the binding site
(Tyr63, His103, Asp206, Glu268, His340, and Asp341).
Additionally, a strong hydrogen bond is established bet-
ween the nitrogen atom of the triazole moiety and the
The distorted half-chair conformation of the cyclohexenyl
ring is an important part of the interaction of compound 4b
with MAL12. This substituent is stabilized by several hydro-
phobic contacts with Tyr63, His103, and Phe169, being
actually sandwiched by Tyr63 and Phe169, similar to the
valienamine moiety of modified acarbose in the complex
with T. maritima 4-R-glucanotransferase.37 The isopropyli-
dene group of 4b also forms hydrophobic interactions with
Phe149, Ala270, and His271.
˚
Thr207 (O-N distance of 3.0 A) and Glu268 (O-N dis-
tance of 3.2 A) residues.
˚
Figure 6 provides a detailed comparison of the atomic
contacts established between MAL12 and compounds 4b
and 4m (stereoisomers R and S of the latter) to delineate
features in the binding modes of the most and the least potent
triazole derivatives acting as R-glucosidase inhibitors, res-
pectively. Acarbose was also included as a reference drug.
Triazole 4d showed almost 3-fold lower potency than 4b
(IC50 = 15.1 ( 0.9 μol/L, Table 2) and is the only compound
presenting an acyclic and significantly more polar R2 sub-
stituent. The analysis of the 4d-MAL12 complex shows
that the triazole moiety of compound 4d makes hydrogen
˚
bonds with Thr207 (O-N distance of 3.9 A) and Glu268

2370 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Ferreira et al.
˚
(O-N distance of 2.6 and 3.6 A for OE2 and OE1 atoms,
respectively). However, the ligand does not seem to produce
hydrophobic contacts with Tyr63, His103, and Phe169,
probably because of the polar nature of the shorter R2
substituent (Figure 7). Moreover, the amide substituent
of 4d shows a destabilizing contact with the Tyr63 side
chain.
Compound 4m was the least potent of the series, although
it has a bulky R2 substituent. The key to understanding this
apparent inconsistency is that this compound was tested as a
racemic mixture. Consequently, the docking procedure was
performed with both enantiomers (R and S) aiming to
recognize how they individually contribute to the inhibition
of the enzyme.
The S configuration of compound 4m (4mS) interacts
through hydrogen bonds with Arg304 (4mS, O1-N distance
Figure 7. Docked conformation of compound 4d on the MAL12
model (gray). The slab shown depicts residues important to inhi-
bitor binding. The triazole moiety makes hydrogen bonds with
Thr207 and Glu268, but the ligand does not produce hydrophobic
contacts with Tyr63 and His103.
˚
˚
of 2.9 A), Thr207 (triazole moiety, O-N distance of 3.2 A),
˚
and Glu268 (4mS, N2-O distance of 3.2 A) (Figure 6).
Similar to the cyclohexyl moiety in compound 4b, the tetra-
hydropyranyl substituent of compound 4mS also forms
hydrophobic contacts with residues Tyr63, His103, Phe169,
and His340, and the isopropylidene group interacts with
Phe149, Ala270, and His271 (Figure 6).
With the R configuration of compound 4m (4mR), the
oxymethylene moiety displaces the triazole group in order to
maintain the favorable hydrophobic interactions of the
tetrahydropyranyl group. However, the molecule loses the
hydrogen bond with Glu268 (Figure 8). The absence of such
interaction for the R enantiomer may be a reason why 4m is
the least potent compound of the series, since it is described
as an important interaction to the R-glucosidase inhibition.15
One limiting factor in the docking approach used here
is the lack of a solvation term in the scoring function,
since solvation effects have been evoked previously to
explain poor structure-activity relationships with non-
carbohydrate glucosidase inhibitors.41 In fact, we could not
find any reasonable correlation between the docking energies
and inhibitory potencies of the examined compounds.
On another issue, compounds 4a, 4b, 4d, 4e, 4g, and 4m
showed higher inhibitory potencies than acarbose, and
the docking studies presented here were unable to offer an
Figure 8. Superposition of docked conformations of 4mS (CPK co-
lored with carbon atoms in yellow) and 4mR (CPK colored with carbon
atoms in green) on the MAL12 model. The figure was drawn in slab
mode to show residues important to inhibitor binding. As can be seen,
molecule 4mR does not make a hydrogen bond with Glu268.
Figure 9. Effect of triazoles and acarbose after a single oral administration of 3 g/kg maltose in normal rats. Five hundred milliliters of
50 mg/kg triazoles or acarbose was dosed. After 5 min, 1 mL of a 3 g/kg maltose solution was administered to each rat. Control group was
administered with the same volume of maltose solution without inhibitor. At 0, 30, 60, and 120 min, 20 μL blood samples were collected and
immediately subjected to blood glucose level assays by a disposable glucose sensor. Data are the mean (mg/dL of plasma) ( SD. Significant
differences between sample and control groups were examined with Dunnett’s t-test (n = 5, (/) P < 0.05).

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2371
unambiguous structural explanation for this, as the acarbose-
MAL12 complex shows extensive favorable contacts between
this pseudotetrasaccharide and the enzyme (Figure 6). One
possible origin for this difficulty is in the entropic contribution
to binding, which is usually not taken into account when a
docking study is performed. Indeed, acarbose is the most
flexible of the R-glucosidase ligands examined in this work,
having six free-rotating bonds connecting its four rings (three
more than compound 4b). Rigid ligands are likely to face less
entropic penalties upon binding than a more flexible one and
thus should have a more favorable binding energy.42 Given the
comparison with the bound conformations of the triazoles
(Figure 4), it is likely that the interactions made by the fourth
ring of acarbose cannot compensate for the entropic cost of
correctly positioning this additional ring in the MAL12 bind-
ing cleft. Consistent with this view, the pseudotrisaccharide,
acarviosine-glucose, has been shown to be a 400-fold more
potent inhibitor of yeast R-glucosidase than acarbose.43
important hydrophobic contacts with residues Tyr63,
His103, and Phe169 at the -1 subsite. The computational
study also predicted that the R enantiomer within the racemic
compound 4m may have little or no activity because of the
lack of the crucial interaction with Glu268 in the active site.
Although successful in offering a possible structural rationale
for the inhibition mechanism of these triazoles, the molecular
modeling study could not correlate interaction energies calcu-
lated from the docked conformations of the ligands with
experimental binding affinities. This difficulty may have
arisen from the oversimplistic physical model used in the
docking program, where neither entropic contributions nor
solvation terms were included in the binding-energy computa-
tion. Moreover, it is possible that some of these inhibitors may
not work as purely competitive inhibitors, presenting mixed
(noncompetitive) behavior instead. It is noteworthy that the
compounds were tested with all or most of the sugar hydroxyl
groups protected by an isopropylidene moiety. The presence
of these protecting groups may have a huge effect on
the inhibitory activities of these compounds because of the
increased molecular volume and loss of donor hydrogen-
bonding groups. Work is underway in our group to clarify
all of these issues. Indeed, the high efficiency of the proposed
synthesis, in combination with the plethora of readily avail-
able azide and acetylene reagents, makes the approach parti-
cularly well suited for combinatorial library schemes, which
should facilitate hypothesis testing and optimization of inhi-
bitor potencies.
Hypoglycemic Effect in Normal Rats
In order to investigate the pharmacological potential of the
most active R-glucosidase inhibitors in the triazole series as
leads for developing new drugs for T2D, we assayed com-
pounds 4a and 4e for hypoglycemic activity in normal rats.
The selected triazoles were resynthesized on a 2 g reaction
scale, and no significant changes in the reaction yields were
noted for this increased scale. Upon oral administration at a
50 mg/kg dose, compounds 4a and 4e were able to signifi-
cantly reduce glucose uptake by normal rats fed with maltose
(Figure 9). This hypoglycemic effect was more evident 30 and
60 min after maltose ingestion, when blood glucose levels were
peaking, similar to the effect displayed by the reference drug
acarbose. The latter clearly performed better than compound
4e in reducing glucose levels in blood samples collected at 30
and 60 min but was undistinguishable from compound 4a for
the 60 min sample.
Experimental Section
Chemistry. General Remarks. Reagents were used as pur-
chased without further purification. Reagents were purchased
from Aldrich or Acros Chemical Co. Dichloromethane and
DMF were distilled from calcium hydride. Pyridine was distilled
from potassium hydroxide. Acetone was distilled from phos-
phorus pentoxide and methanol from iodine and magnesium.
Column chromatography was performed with silica gel 60
(Merck 70-230 mesh). Analytical thin-layer chromatography
was performed with silica gel plates (Merck, TLC silica gel 60
F254), and the plots were visualized under UV light or devel-
oped by immersion in an ethanolic solution of vanillin. Yields
refer to chromatographically and spectroscopically homoge-
neous materials. Melting points were obtained on a Fischer-
Johns apparatus and are uncorrected. Infrared spectra data
were recorded from KBr pellets on a Perkin-Elmer model
1420 FT-IR spectrophotometer calibrated relative to the
1601.8 cm-1 absorbance of polystyrene. The optical rotations
([R]D20) were measured with a Perkin-Elmer 243B polarimeter
using 10 cm cells and a wavelength of 589 nm. NMR spectra
were recorded on a Varian Unity Plus VXR (300 MHz) system in
CDCl3 solutions using tetramethylsilane as the internal refer-
ence standard (0.0 ppm). Coupling constants (J) are reported in
hertz and refer to apparent peak multiplicities. HRMS were
recorded on a MICROMASS Q-TOF MICRO mass spectro-
meter using ESI-TOF (electrospray ionization time-of-flight).
Elemental analysis was used to ascertained purity of g95% for
all compounds for which biological data were determined.
General Method for the Preparation of 1H-1,2,3-Triazoles
Derivatives 4a-n, 5a, 6a, and 7a-c. To a solution containing
0.70 mmol of the appropriate alkyne, 0.83 mmol of azidocarbo-
hydrates in dichloromethane (0.7 mL), and water (0.7 mL) were
Conclusions
We have described the synthesis of several 1H-1,2,3-triazole
derivatives of carbohydrates and have demonstrated their
action as R-glucosidase inhibitors. The R-glucosidase inhibi-
tion screening showed that all compounds were active at
500 μM, although the triazoles conjugated with D-ribose
(4a-n) were in general more effective than the other triazoles
containing D-xylose, D-galactose, and D-allose as the carbohy-
drate portion (5a, 6a, 7a-c). The 1-O-methyl-2,3-O-isopro-
pylidene-β-D-ribofuranose derivatives were conjugated
through the C5 atom of the sugar and the N1 atom of the
1H-1,2,3-triazole ring. Some of these compounds, e.g., 4b, 4e,
and 4g, showed 20-fold higher potency in yeast maltase
inhibition than acarbose, as measured by the ratios of their
IC50 values. This observation was quite surprising considering
the specificity of this enzyme for substrates containing at least
one and up to four glucose residues linked by R-1,4 bonds.
Moreover, compounds 4a and 4e, reduced postprandial
glucose levels when administered orally to normal rats, high-
lighting the pharmacological potential of this synthetic series.
While assuming a reversible competitive inhibition modality
for these compounds, our molecular modeling study sup-
ported a binding mode compatible with a transition-state
mimicking mechanism, where the most potent inhibitors have
the N3 atom in the triazole ring bound to one of the catalytic
carboxylates, Glu268, and the R2 substituent engaging
added CuSO4 5H2O (9.3 mg, 0.04 mmol) and sodium ascorbate
3
(22 mg, 0.11 mmol). The resulting suspension was stirred at
room temperature for 4-6 h. After this time, the mixture was
diluted with 5 mL of dichloromethane and 5 mL of water. The
organic phases were separated, dried with sodium sulfate, and
concentrated at reduced pressure, furnishing the 1,2,3-triazoles

2372 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Ferreira et al.
derivatives, which were analyzed by 1H and 13C NMR spectro-
scopy and infrared.
3.36 (s, 3H, OCH3), 4.41-4.46 (m, 1H, H-40), 4.54-4.61 (m, 2H,
H-50), 4.66 (d, 1H, J=5.8Hz, H-30),4.75(d, 1H, J=5.8Hz,H-20),
5.00 (s, 1H, H-10), 5.22 (s, 2H, H-1a), 6.97-7.00 (m, 3H, H-4a and
1-O-Methyl-2,3-O-isopropylidene-5-(4-phenyl-1H-1,2,3-triazol-
1-yl)-β-D-ribofuranose (4a). Compound 4a was obtained as a white
powder in 98% yield. Mp 130-133 °C. [R]D20 -46 (c 1.0, CHCl3).
H-5a), 7.26-7.31 (m, 2H, H-2a and H-3a), 7.70 (s, 1H, H-5). 13
C
NMR (CDCl3, 75MHz) δ25.1 (CH3), 26.6 (CH3), 53.4 (C-50), 55.8
(OCH3), 62.2 (C-1a), 82.0 (C-40), 85.2 (C-30), 85.4 (C-20), 110.3
(C-10), 113.2 (C-60), 115.0 (C-3a), 121.5 (C-5a), 123.9 (C-5), C-4
not observed, 159.6 (CdO). HRMS (EI) calcd for C18H23N3O5:
361.1638. Found: 362.4054 (M þ H)þ. Anal. Calcd for C18H23-
N3O5: C, 59.82; H, 6.41; N, 11.63. Found: C, 59.85; H, 6.40; N,
11.65.
1
IR (film, CHCl3) ν (cm-1): 3075 and 1612. H NMR (CDCl3,
300 MHz) δ 1.31 (s, 3H, CH3), 1.47 (s, 3H, CH3), 3.41 (s, 3H,
OCH3), 4.43 (d, 1H, J = 5.8 Hz, H-30), 4.62-4.70 (m, 3H, H-50
and H-40), 4.79 (d, 1H, J = 5.8 Hz, H-20), 5.04 (s, 1H, H-10),
7.31-7.45 (m, 3H, ArH), 7.82-7.87 (m, 3H, ArH and H-5). 13C
NMR (CDCl3, 75 MHz) δ 24.7 (CH3), 26.2 (CH3), 55.4 (OCH3),
53.0 (C-50), 81.6 (C-20), 84.8 (C-30), 85.0 (C-40), 109.9 (C-10), 112.7
(C-60), 119.7 (C-5), 125.5 (C-Ar-2a or 3a or 4a), 128.0 (C-Ar-2a
or 3a or 4a), 128.6 (C-Ar-2a or 3a or 4a), 130.3 (C-Ar-1a), 147.8
(C-4). HRMS (EI) calcd for C17H21N3O4: 331.1532. Found:
332.3803 (M þ H)þ. Anal. Calcd for C17H21N3O4: C, 61.62; H,
6.39; N, 12.68. Found: C, 61.63; H, 6.40; N, 12.70.
1-O-Methyl-2,3-O-isopropylidene-5-[4-(3-chloropropyl)-1H-
1,2,3-triazol-1-yl]-β-D-ribofuranose (4f). Compound 4f was
obtained as a white powder in 90% yield. Mp 60-63 °C.
[R]D20 -56 (c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 2987
1
and 1632. H NMR (CDCl3, 300 MHz) δ 1.29 (s, 3H, CH3),
1.44 (s, 3H, CH3), 2.17 (q, 2H, J = 6.6 and 7.3 Hz, H-2a), 2.89
(t, 2H, J = 7.6 Hz, H-1a), 3.38 (s, 3H, OCH3), 3.57 (t, 2H, J =
6.6 Hz, H-3a), 4.37-4.43 (m, 1H, H-40), 4.50-4.59 (m, 2H,
H-50), 4.65 (d, 1H, J = 5.8 Hz, H-30), 4.74 (d, 1H, J = 5.8 Hz,
H-20), 5.00 (s, 1H, H-10), 7.43 (s, 1H, H-5). 13C NMR (CDCl3,
75 MHz) δ 22.4 (C-1a), 24.7 (CH3), 26.1 (CH3), 31.5 (C-2a),
43.9 (C-3a), 52.84 (C-50), 55.3 (OCH3), 81.6 (C-30), 84.8 (C-40),
85.0 (C-20), 109.6 (C-5), 109.8 (C-10), 112.7 (C-60), 112.5(C-4).
HRMS (EI) calcd for C14H22ClN3O4: 331.1299. Found: 332.8078
(M þ H)þ. Anal. Calcd for C14H22ClN3O4: C, 50.68; H, 6.68; N,
12.66. Found: C, 50.70; H, 6.70; N, 12.67.
1-O-Methyl-2,3-O-isopropylidene-5-(4-cyclohexene-1H-1,2,3-
triazol-1-yl)-β-D-ribofuranose (4b). Compound 4b was obtained
as a light-yellowish powder in 95% yield. Mp 105-108 °C.
[R]D20 -45 (c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 1374 and
1
1670. H NMR (CDCl3, 300 MHz) δ 1.29 (s, 3H, CH3), 1.45
(s, 3H, CH3), 1.62-1.70 (m, 2H, H-4a), 1.72-1.80 (m, 2H,
H-3a), 2.16-2.23 (m, 2H, H-2a), 2.36-2.41 (m, 2H, H-5a),
3.38 (s, 3H, OCH3), 4.40 (dd, 1H, J = 5.9 and 9.9 Hz, H-30),
4.50-4.59 (m, 2H, H-50), 4.66 (d, 1H, J = 5.9 Hz, H-40), 4.74
(d, 1H, J = 5.8 Hz, H-20), 5.00 (s, 1H, H-10), 6.51-6.53 (m, 1H,
H-6a), 7.48 (s, 1H, H-5 triazole). 13C NMR (CDCl3, 75 MHz)
δ 22.0 (C-4a), 22.2 (C-2a and C-1a), 24.7 (CH3), 25.0 (C-5a), 26.1
(CH3), 52.7 (C-50), 55.3 (OCH3), 81.6 (C-30), 84.8 (C-40), 85.0
(C-20), 109.9 (C-10), 112.6 (C-60), 124.9 (C-6a and C-5), 127.0
(C-Ar-1a), 151.1 (C-4). HRMS (EI) calcd for C17H25N3O4:
335.1845. Found: 336.4104 (M þ H)þ. Anal. Calcd for C17H25-
N3O4: C, 60.88; H, 7.51; N, 12.53. Found: C, 60.87; H, 7.49;
N, 12.51.
1-O-Methyl-2,3-O-isopropylidene-5-[4-(1-hydroxycyclohexil)-
1H-1,2,3-triazol-1-yl]-β-D-ribofuranose (4g). Compound 4g was
obtained as a white powder in 88% yield. Mp 113-115 °C.
[R]D20 -48 (c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 3421 and
1
2935. H NMR (CDCl3, 300 MHz) δ 1.29 (s, 3H, CH3), 1.45
(s, 3H, CH3), 1.57-2.00 (m, 4H, H-2a, 3a, 4a, and 5a), 2.39 (s,
1H, O-H), 3.38 (s, 3H, OCH3), 4.35-4.38 (m, 1H, H-40),
4.44-4.59 (m, 2H, H-50), 4.65 (d, 1H, J = 5.8 Hz, H-30), 4.75
(d, 1H, J = 5.8 Hz, H-20), 5.00 (s, 1H, H-10), 7.57 (s, 1H, H-5).
13C NMR (CDCl3, 75 MHz) δ 21.7 (C-3a), 24.7 (CH3), 25.2
(C-4a), 26.2 (CH3), 37.9 (C-2a), 53.0 (C-50), 55.4 (OCH3), 81.7
(C-30), 84.8 (C-40), 85.0 (C-20), 109.9 (C-10), 112.7 (C-60), 115.6
(C-5), C-4, not observed. HRMS (EI) calcd for C17H27N354:
353.1951. Found: 354.4267 (M þ H)þ. Anal. Calcd for C17H27-
N3O5: C, 57.77; H, 7.70; N, 11.89. Found: C, 57.79; H, 7.71;
N, 11.84.
1-O-Methyl-2,3-O-isopropylidene-5-[4-(1-hydroxy-1-methyl-
ethyl)-1H-1,2,3-triazol-1-yl]-β-D-ribofuranose (4h). Compound
4h was obtained as a white powder in 79% yield. Mp 104-
106 °C. [R]D20 -27 (c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1):
3461. 1H NMR (CDCl3, 300 MHz) δ 1.30 (s, 3H, CH3), 1.45 (s,
3H, CH3), 1.64 (s, 6H, 2 CH3), 2.47 (s, 1H, O-H), 3.39 (s, 3H,
OCH3), 4.38-4.44 (m, 1H, H-40), 4.51-4.60 (m, 2H, H-50), 4.66
(d, 1H, J = 5.8 Hz, H-30), 4.74 (d, 1H, J = 5.8 Hz, H-20), 5.00 (s,
1H, H-10), 7.54 (s, 1H, H-5). 13C NMR (CDCl3, 75 MHz) δ 24.7
(CH3), 26.2 (CH3), 30.2 (CH3), 30.3 (CH3), 52.0 (C-50), 55.4
(OCH3), 68.3 (C-1a), 81.6 (C-30), 84.2 (C-40), 85.0 (C-20), 109.9
(C-10), 112.2 (C-60), 119.3 (C-5), 161.3 (C-4). HRMS (EI) calcd
C14H23N3O5: 313.1638. Found: 314.3611 (M þ H)þ. Anal.
Calcd for C14H23N3O5: C, 53.66; H, 7.40; N, 13.41. Found:
C, 53.65; H, 7.40; N, 13.43.
1-O-Methyl-2,3-O-isopropylidene-5-(4-hydroxymethyl-1H-
1,2,3-triazol-1-yl)-β-D-ribofuranose (4c). Compound 4c was
obtained as a light-yellowish powder in 80% yield. Mp 93-
95 °C. [R]D20 -16 (c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1):
1
3452 and 1654. H NMR (CDCl3, 300 MHz) δ 1.30 (s, 3H,
CH3), 1.45 (s, 3H, CH3), 1.70 (s, 1H, OH), 3.39 (s, 3H, OCH3),
4.39-4.46 (m, 1H, H-30), 4.51-4.61 (m, 2H, H-50), 4.66 (d, 1H,
J = 5.8 Hz, H-30), 4.74 (d, 1H, J = 5.8 Hz, H-20), 4.79 -4.80
(m, 2H, H-1a), 5.01 (s, 1H, H-10), 7.64 (s, 1H, H-5). 13C NMR
(CDCl3, 75 MHz) δ 25.1 (CH3), 26.5 (CH3), 53.3 (C-50), 55.8
(OCH3), 56.4 (C-1a), 82.0 (C-20), 85.2 (C-30), 85.2 (C-30), 85.4
(C-40), 110.2 (C-10), 113.1 (C-60), 122.4 (C-5), 148.3 (C-4).
HRMS (EI) calcd for C12H19N3O5: 285.1325. Found:
286.3085 (M þ H)þ. Anal. Calcd for C12H19N3O5: C, 50.52;
H, 6.21; N, 14.73. Found: C, 50.53; H, 6.23; N, 14.70.
1-O-Methyl-2,3-O-isopropylidene-5-(4-carboxylate ethyl-1H-
1,2,3-triazol-1-yl)-β-D-ribofuranose (4d). Compound 4d was
obtained as a white powder in 87% yield. Mp 88-90 °C.
[R]D20 -44 (c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 2931 and
1
1730. H NMR (CDCl3, 300 MHz) δ 1.30 (s, 3H, CH3), 1.41
(t, 3H, J = 7.0 Hz, C-3a-CH3), 1.45 (s, 3H, CH3), 3.37 (s, 3H,
OCH3), 4.39 - 4.63 (m, 5H, H-40, H-50, H-2a), 4.66 (d, 1H, J =
5.8 Hz, H-30), 4.73 (d, 1H, J = 5.8 Hz, H-20), 5.00 (s, 1H, H-10),
8.23 (s, 1H, H-5). 13C NMR (CDCl3, 75 MHz) δ 14.1 (C-3a),
24.7 (CH3), 26.1 (CH3), 53.2 (C-50), 55.5 (OCH3), 61.1 (C-2a),
81.4 (C-20), 84.7 (C-30), 84.9 (C-40), 109.0 (C-10), 112.2 (C-60),
112.2 (C-4), 109.6 (C-5), 159.6 (CdO). HRMS (EI) calcd for
C14H21N3O6: 327.1430. Found: 328.3453 (M þ H)þ. Anal.
Calcd for C14H21N3O6: C, 51.37; H, 6.47; N, 12.84. Found: C,
51.31; H, 6.45; N, 12.89.
1-O-Methyl-2,3-O-isopropylidene-5-(4-propyl-1H-1,2,3-triazol-
1-yl)-β-D-ribofuranose (4i). Compound 4i was obtained as a white
powder in 85% yield. Mp 106-107 °C. [R]D20 -24 (c 1.0, CHCl3).
1
IR (film, CHCl3) ν (cm-1): 2939 and 2986. H NMR (CDCl3,
300 MHz) δ 1.30 (s, 3H, CH3), 1.40 (t, 3H, J = 7.5 Hz, CH3), 1.45
(s, 3H, CH3), 3.38 (s, 3H, OCH3), 4.39-4.63 (m, 7H, H-40, H-50
and H-2a and 1a), 4.66 (d, 1H, J = 5.8 Hz, H-30), 4.73 (d, 1H, J =
5.8 Hz, H-20), 5.01 (s, 1H, H-10), 8.19 (s, 1H, H-5). 13C NMR
(CDCl3, 75 MHz) δ 14.13 (C-3a), 19.2 (C-2a), 24.7 (CH3), 26.1
(CH3), 26.0 (C-1a), 53.2 (C-50), 55.5 (OCH3), 81.4 (C-30), 84.8
(C-40), 84.7 (C-20), 110.0 (C-10), 112.0 (C-60), 127.6 (C-5), 140.3
(C-4). HRMS (EI) calcd C14H23N3O4: 297.1689. Found: 298.2611
1-O-Methyl-2,3-O-isopropylidene-5-(4-phenoxymethyl-1H-1,2,3-
triazol-1-yl)-β-D-ribofuranose (4e). Compound 4e was obtained as
a light-yellowish powder in 80% yield. Mp 80-83 °C. [R]D20 -28
1
(c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 3065 and 1595. H
NMR (CDCl3, 300 MHz) δ 1.30 (s, 3H, CH3), 1.45 (s, 3H, CH3),

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2373
(M þ H)þ. Anal. Calcd for C14H23N3O4: C, 56.55; H, 7.80; N,
14.13. Found: C, 56.57; H, 7.82; N, 14.15.
1.57 (s, 3H, CH3), 4.21-4.26 (m, 2H, H-60), 4.34 (q, 1H, J =
4.8 Hz, H-20), 4.45-4.53 (m, 1H, H-30), 4.63-4.71 (m, 2H,
H-50and H-40), 5.54 (d, 1H, J = 4.8 Hz, H-10), 7.29-7.34 (m,
1H, H-3a), 7.39-7.44 (m, 2H, H-2a), 7.82-7.85 (m, 2H, H-1a),
7.95 (s, 1H, H-5). 13C NMR (CDCl3, 75 MHz) δ 24.6 (CH3), 25.1
(CH3), 26.2 (CH3), 26.2 (CH3), 50.8 (C-60), 67.5 (C-50), 70.6 (C-20),
71.0 (C-30), 71.4 (C-40), 96.5 (C-10), 109.3 (C-70), 110.1 (C-80), 121.1
(C-5), 126.0 (C-1a), 128.2 (C-3a), 129.0 (C-2a), 131.0 (C-ar), 147.7
(C-4). HRMS (EI) calcd C20H25N3O5: 387.1794. Found: 388.4431
(M þ H)þ. Anal. Calcd for C20H25N3O5: C, 62.00; H, 6.50; N,
10.85. Found: C, 62.03; H, 6.53; N, 10.87.
1-O-Methyl-2,3-O-isopropylidene-5-[4-(1-hydroxy-1-methyl-
propyl)-1H-1,2,3-triazol-1-yl]-β-D-ribofuranose (4j). Compound 4j
was obtained as a white powder in 84% yield. Mp 111-113 °C.
[R]D20 -41 (c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 3355, 2970,
and 2936. 1H NMR (CDCl3, 300 MHz) δ 0.85 (t, 3H, J = 7.3 Hz,
C-3a), 1.30 (s, 3H, CH3), 1.45 (s, 3H, CH3), 1.57 (s, 3H, C-4a), 2.49
(s, 1H, O-H), 3.38 (s, 3H, OCH3), 4.35-4.44 (m, 1H, H-40),
4.51-4.58 (m, 2H, H-50), 4.64 (d, 1H, J = 5.8 Hz, H-30), 4.75 (d,
1H, J = 5.8 Hz, H-20), 5.01 (s, 1H, H-10), 7.52 (s, 1H, H-5). 13C
NMR (CDCl3, 75MHz) δ8.0(C-3a), 24.7(CH3), 26.2(CH3), 27.7
(C-4a), 35.7 (C-2a), 52.9 (C-50), 55.3 (OCH3), 71.0 (C-1a), 81.6
(C-30), 84.8 (C-40), 85.0 (C-20), 109.9 (C-10), 112.2 (C-60), 120.0
(C-5), 154.6 (C-4). HRMS (EI) calcd C15H25N3O5: 327.1794.
Found: 328.3889 (M þ H)þ. Anal. Calcd for C15H25N3O5: C,
55.03; H, 7.70; N, 12.84. Found: C, 55.06; H, 7.73; N, 12.86.
1-O-Methyl-2,3-O-isopropylidene-5-(4-methylacetate-1H-1,2,3-
triazole-1-yl)-β-D-ribofuranose (4l). Compound 4l was obtained as
light-yellowish powder in 85% yield. Mp 73-74 °C. [R]D20 -23
(c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 2987, 2941, and 1740.
1H NMR (CDCl3, 300 MHz) δ 1.27 (s, 3H, CH3), 1.42 (s, 3H, s,
CH3), 2.04 (s, 3H, C-3a), 3.35 (s, 3H, OCH3), 4.34-4.43 (m, 1H,
H-40), 4.53-4.58 (m, 2H, H-50), 4.62 (d, 1H, J = 5.8 Hz, H-30),
4.73 (d, 1H, J = 5.8 Hz, H-20), 4.98 (s, 1H, H-10), 5.18 (s, 1H,
C-1a), 7.69 (s, 1H, H-5). 13C NMR (CDCl3, 75 MHz) δ 20.6
(C-3a), 24.7 (CH3), 26.1 (CH3), 52.9 (C-50), 55.4 (OCH3), 57.3
(C-1a), 81.5 (C-30), 84.7 (C-40), 84.9 (C-20), 109.9 (C-10), 112.7
(C-60), 123.9 (C-5), 142.8 (C-4), 170.6 (CdO). HRMS (EI) calcd
C14H21N3O6: 327.1430. Found: 328.3453 (M þ H)þ. Anal. Calcd
for C14H21N3O6: C, 51.37; H, 6.47; N, 12.84. Found: C, 51.38;
H, 6.49; N, 12.86.
1-O-Methyl-2,3-O-isopropylidene-5-[4-(tetrahydro-2H-pyran)-
methyl)-1H-1,2,3-triazole-1-yl]-β-D-ribofuranose (4m). Compound
4m was obtained as a white powder in 95% yield. Mp 63-66 °C.
[R]D20 -25 (c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 2942 and
2873. 1HNMR(CDCl3, 300 MHz) δ1.29(s, 3H, CH3), 1.44 (s, 3H,
CH3), 1.50-1.84 (m, 6H, C-3a, C-4a, C-5a), 3.38 (s, 3H, OCH3),
3.56-3.58 (m, 1H, C-6a), 3.86-3.92 (m, 1H, C-6a), 4.36 - 4.42 (m,
1H, H-40), 4.52-4.67 (m, 4H, H-50 and C-2a), 4.73-4.76 (m, 1H,
H-30), 4.73 (d, 1H, J = 5.8 Hz, H-20), 5.01(s, 1H, H-10), 7.64 (s, 1H,
H-5). 13C NMR (CDCl3, 75 MHz) δ 19.1 (C-4a), 24.7 (CH3), 25.1
(C-5a), 26.1 (CH3), 30.2 (C-3a), 52.8 (C-50), 55.3 (OCH3), 60.3
(C-1a), 62.1 (C-6a), 81.6 (C-30), 84.8 (C-40), 85.0 (C-20), 98.0
(C-2a), 109.8 (C-10), 112.6 (C-60), 122.6 (C-5), 145.3 (C-4). HRMS
(EI) calcd C17H27N3O6: 369.1408. Found: 370.4258 (M þ H)þ.
Anal. Calcd for C17H27N3O6: C, 55.27; H, 7.37; N, 11.37. Found:
C, 55.25; H, 7.35; N, 11.35.
1,2:5,6-Di-O-isopropylidene-3-(4-phenyl-1H-1,2,3-triazole-1-yl)-
r-D-alofuranose (6a). Compound 6a was obtained as a white
powder in 89% yield. Mp 155-157 °C. [R]D20 -87 (c 1.0, CHCl3).
1
IR (film, CHCl3) ν (cm-1): 2987 and 2935. H NMR (CDCl3,
300 MHz) δ1.32(s, 3H, CH3), 1.36(s, 3H, CH3), 1.43(s, 3H, CH3),
1,51 (3H, s, CH3), 3.90-4.00 (m, 2H, H40 and H-50), 4.20-4.28 (m,
2H, H-60), 4.33-4.38(m, 1H, H-30), 5.03 (d, 1H, J = 3.6 Hz, H-20),
6.27 (1H, d, J = 3,4 Hz, H-10), 7.29-7.34 (m, 1H, H-3a),
7.32-7.37 (m, 1H, H-3a), 7.40-7.46 (m, 2H, H-2a), 7.82-7.85
(m, 2H, H-1a), 7.91 (s, 1H, H-5). 13C NMR (CDCl3, 75 MHz) δ
25.0 (CH3), 26.0 (CH3), 26.5 (CH3), 26.7 (CH3), 66.2 (C-30), 67.5
(C-60), 72.9 (C-40), 80.4 (C-50), 83.3 (C-20), 104.9 (C-10), 106.2
(C-70), 109.4 (C-80), 112.2 (C-5), 125.6 (C-1a), 128.7 (C-2a), 128.1
(C-ar), 130.2 (C-4). HRMS (EI) calcd C20H25N3O5: 387.1794.
Found: 388.4431 (M þ H)þ. Anal. Calcd for C20H25N3O5: C,
62.00; H, 6.50; N, 10.85. Found: C, 62.01; H, 6.52; N, 10.86.
5-Deoxy-1,2-O-isopropylidene-5-(4-phenyl-1H-1,2,3-triazole-
1-yl)-r-D-xylofuranose (7a). Compound 7a was obtained as a
light-yellowish powder. Yield: 97%. Mp 50-53 °C. [R]D20 -17
1
(c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 3419 and 2861. H
NMR (CDCl3, 300 MHz) δ 1.32 (s, 3H, CH3), 1.50 (s, 3H, CH3),
2.00 (s, 1H, OH), 4.33 (dd, 1H, J = 4.9 Hz, H-30), 4.51-4.65(m,
3H, H-40 and H-50), 4.84 (dd, 1H, J = 5.1 Hz, H-20), 6.01 (d, 1H,
J = 3.6 Hz, H-10), 7.33-7.46 (m, 3H, H-2a and 4a), 7.75-7.80
(m, 3H, H-3a and H-5). 13C NMR (CDCl3, 75 MHz) δ 26.4
(CH3), 26.9 (CH3), 49.4 (C-50), 75.4 (C-20), 78.6 (C-30), 85.6
(C-40), 105.0 (C-10), 112.2 (C-60), 120.2 (C-5), 124.0 (C-1a), 128.1
(C-2a), 130.3 (C-3a and 4a), 128.7 (C-2a), 128.1 (C-ar), 156.4
(C-4). HRMS (EI) calcd C16H17N3O5: 317.1376. Found:
318.3526 (M þ H)þ. Anal. Calcd for C16H19N3O4: C, 60.56;
H, 6.03; N, 13.24. Found: C, 60.57; H, 6.05; N, 13.25.
5-Deoxy-1,2-O-isopropylidene-5-[4-(1-cyclohexene)-1H-1,2,3-
triazole-1-yl]-r-D-xylofuranose (7b). Compound 7b was obtained
as a light-yellowish powder in 98% yield. Mp 140-142 °C.
[R]D20 -15 (c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 3270 and
2931. 1H NMR (CDCl3, 300 MHz) δ 1.30 (s, 3H, CH3), 1.44 (s,
3H, CH3), 1.63-1.70 (m, 2H, H-4a), 1.72-1.79 (m, 2H, H-6a),
2.17-2.23 (m, 2H, H-5a), 2.35-2.40 (m, 2H, H-3a), 3.80-3.82
(m, 1H, H-30), 4.18-4.21 (m, 1H, OH), 4.39-4.53 (m, 2H, H-50),
4.58 (d, 1H, J = 3.4 Hz, H-20), 4.74-4.81 (m, 1H, H-40), 5.96 (d,
1H, J = 3.4 Hz, H-10), 6.46-6.50 (m, 1H, H-2a), 7.51 (s, 1H, H-
5). 13C NMR (CDCl3, 75 MHz) δ 22.3 (C-4a and C-5a), 22.6 (C-
3a), 25.5 (C-6a), 26.4 (CH3), 26.5 (C-3a), 27.0 (CH3), 49.2 (C-50),
74.4 (C-30), 79.7 (C-20), 85.6 (C-40), 105.4 (C-10), 112.1 (C-60),
120.2 (C-1a), 125.8 (C-2a), 127.2 (C-5), 149.7 (C-4). HRMS (EI)
calcd C16H23N3O4: 321.1689. Found: 322.3853 (M þ H)þ. Anal.
Calcd for C16H23N3O4: C, 59.80; H, 7.21; N, 13.08. Found: C,
59.83; H, 7.23; N, 13.09.
1-O-Methyl-2,3-O-isopropylidene-5-[4-(1-hydroxy-1,3-dimethyl-
butyl)-1H-1,2,3-triazole-1-yl]-β-D-ribofuranose (4n). Compound 4n
was obtained as a white powder in 79% yield. Mp 79-81 °C.
[R]D20 -32 (c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 3421, 2951,
and 2873. 1H NMR (CDCl3, 300 MHz) δ 0.83 (d, 3H, J = 6,5 Hz,
C-4a), 0.86 (d, 3H, J = 6.5 Hz, C-5a), 1.30 (s, 3H, CH3), 1.45 (s,
3H, CH3), 1.61 (s, 3H, C-6a), 1.66-1.86 (m, 3H, C-2a and C-3a),
2.33 (s, 1H, OH), 3.38 (s, 3H, OCH3), 4.37 - 4.45 (m, 1H, H-40),
4.51-4.60 (m, 2H, H-50), 4.66 (d, 1H, J = 5.8 Hz, H-30), 4.74 (dd,
1H, J = 2.4 and 5.8 Hz, H-20), 5.01 (s, 1H, H-10), 7.53 (s, 1H, H-5).
13C NMR (CDCl3, 75 MHz) δ 24.1 (C-4a), 24.1 (C-5a), 24.2
(C-6a), 24.7 (CH3), 26.1 (CH3), 29.1 (C-3a), 51.4 (C-50), 52.9
(C-2a), 55.3 (OCH3), 71.2 (C-1a), 81.6 (C-30), 84.1 (C-40), 84.9
(C-20), 109.9 (C-10), 112.2 (C-60), 119.9 (C-5), 135.7 (C-4). HRMS
(EI) calcd C17H29N3O5: 355.2107. Found: 356.4420 (M þ H)þ.
Anal. Calcd for C17H29N3O5: C, 57.45; H, 8.22; N, 11.82. Found:
C, 57.47; H, 8.25; N, 11.85.
5-Deoxy-1,2-O-isopropylidene-5-(4-hydroxymethyl-1H-1,2,3-
triazole-1-yl)-r-D-xylofuranose (7c). Compound 7c was ob-
tained as a white powder in 75% yield. Mp 50-51 °C.
[R]D20 -35 (c 1.0, CHCl3). IR (film, CHCl3) ν (cm-1): 3448,
1
2986, and 2936. H NMR (CDCl3, 300 MHz) δ 1.32 (s, 3H,
CH3), 1.50 (s, 3H, CH3), 2.88 (s, 1H, H-1a), 2.96 (s, 1H, H-1a),
3.55-3.67 (m, 2H, H-30 and 40), 4.24-4.31 (m, 2H, H-50), 4.53
(d, 1H, J = 3.6 Hz, H-20), 5.94 (d, 1H, J = 3.6 Hz, H-10), 8.00 (s,
1H, H-5). 13C NMR (CDCl3, 75 MHz) δ 26.9 (CH3), 26.5 (CH3),
48.5 (C-1a), 49.0 (1H, C-50), 74.8 (C-30), 78.73 (C-20), 85.1 (C-40),
105.4 (C-10), 104.6 (C-60), 127.7 (C-5), 129.9 (C-4). HRMS (EI)
1,2:3,4-Di-O-isopropylidene-6-(4-phenyl-1H-1,2,3-triazole-1-yl)-
r-D-galactopyranose. (5a). Compound 5a was obtained as a white
powder in 90% yield. Mp 152-154 °C. [R]D20 -98 (c 1.0, CHCl3).
1
IR (film, CHCl3) ν (cm-1): 2987 and 2932. H NMR (CDCl3,
300 MHz) δ1.29 (s, 3H, CH3), 1.37(s, 3H, CH3), 1.40(s, 3H, CH3),

2374 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Ferreira et al.
calcd C11H17N3O5: 271.1168. Found: 238.0905 (M þ H)þ. Anal.
Calcd for C11H17N3O5: C, 48.70; H, 6.32; N, 15.49. Found: C,
48.72; H, 6.33; N, 15.51.
The hydrogen atoms were added to the protein and ligand struc-
tures, and 1000 steps of energy minimization were performed by the
€
steepest descent method, using Gasteiger-Huckel charges and a
dielectric constant of 80, in the Tripos force field.
Biochemistry. The R-glucosidase inhibitory activity was ass-
ayed using the yeast maltase enzyme (Sigma-Aldrich, St. Louis,
MO) according to the procedure described by Matsui and
collaborators.44 The R-glucosidase activity was determined
by monitoring the p-nitrophenol released from p-nitrophenyl-
R-D-glycopyranoside at 405 nm. The compounds in the series
under study were first screened for R-glucosidase inhibition at a
fixed concentration of 500 μM, and activities were expressed as
percentage inhibitions relative to a control assay with no
inhibitor added. Compounds showing the highest inhibitory
activities were further characterized by determining the concen-
tration required to inhibit 50% of the R-glucosidase activity
under the assay conditions (defined as the IC50 value). Acarbose
(Glucobay, Bayer) was used in the assays for comparison.
Pharmacology. Theanimals(maleWistarrats, 2-3 months old,
200-250 g) were divided into four groups of five rats each. Group 1
consisted of nondiabetic control rats that received a physiological
NaCl solution containing DMSO at 1% (as vehicle). Group 2
received acarbose (50 mg/kg), group 3 received triazole 4a
(50 mg/kg), and group 4 received triazole 4e (50 mg/kg) in the
same vehicle. Following the general procedure of Kanthi and
collaborators,45 each triazole or acarbose was suspended in
0.5 mL of physiological NaCl solution containing DMSO at 1%
and administered orally by a cannula to 8 h fasted rats 5 min before
the oral administration of 1 mL of maltose (3 g/kg). The control
group was given the same volume of vehicle before the maltose
solution. Blood samples were obtained from the tail vein 5 min
before the oral administration of the drugs or the vehicle (T0) and
at 30, 60, and 120 min thereafter. Glucose concentration was
measured in total blood with an Accu-check compact machine
(Roche). Data were expressed as the mean ( SD. Significant
differences between sample and control groups were examined
with Dunnett’s t-test (n = 5, (/) P < 0.05).
Molecular Docking of the 1,2,3-Glucotriazoles. The molecular
models of the 1,2,3-glucotriazoles were built using the Sybyl
molecular modeling package, version 8.0 (Tripos Associates,
St. Louis, MO). The hydrogen atoms were added to the struc-
tures, and 1000 steps of energy minimization were performed by
€
the steepest descent method, using Gasteiger-Huckel charges
and a dielectric constant of 80, in the Tripos force field. The
structures were further optimized using the conjugated gradient
method. Ligand-protein docking was performed with the
molecular docking algorithm MolDock51 using the Molegro
Virtual Docker software, version 3.0.0 (Molegro ApS, Aarhus,
Denmark, http://www.molegro.com). MolDock uses a heuristic
search algorithm (termed guided differential evolution) which is
a combination of differential evolution and a cavity-prediction
algorithm. The docking scoring function is an extension of the
piecewise linear potential (PLP).51 After the ligands and the
protein coordinates were imported, all structural parameters
including bond type, hybridization, explicit hydrogen, charges,
and flexible torsions were assigned using the automatic prepara-
tion function in Molegro Virtual Docker software, version
3.0.0.51 For each compound, 50 docking runs were performed
with the initial population of 300 individuals. The software is
also able to work with side chain conformational changes by
softening the potentials (steric, hydrogen bonds, and electro-
static force) used during a docking simulation. The residues
considered flexible were those that were close enough to the
˚
reference ligand to interact with it (6 A around the acarbose,
69 residues). After each ligand had been docked, the side chains
chosen were energy-minimized by the Molegro Virtual Docker
software, version 3.0.0,51 with respect to the conformation
found, using the standard nonsoftened potentials. Only torsion
angles in the side chains were modified during the minimization;
all other properties (including bond lengths and backbone atom
positions) were held fixed.51 In order to identify which of the
ligand configurations most likely represented the correct bind-
ing mode of the triazole compounds, all complexes were ana-
lyzed considering evidence from sequence conservation, visual
inspection, and automated assignment of interatomic contacts
using the Ligplot program and the CSU/LPC server.52
Molecular Modeling. MAL12 Model Building and Validation.
Homology modeling of MAL12 was carried out using the
Modeller software, version 9v2,46 based on the crystal structures
˚
of B. cereus oligo-R-1,6-glucosidase (PDB code 1UOK, 2.0 A
resolution, 39% identity among 581 aligned residues)47 and of
˚
Geobacillus sp. R-glucosidase GSJ (PDB code 2ZE0, 2.0 A
resolution, 39% identity among 591 aligned residues).36 The
target sequence was found at the UniprotKB/Swiss-Prot data-
base under access number P53341, and the sequence alignment
was performed using the T-Coffee server.48 A set of 50 modeled
structures was generated and ranked by the Modeller objective
function and the top-scoring models were checked for their
stereochemical and overall structural quality, using the Pro-
Check,34 Verify-3D,35 and Whatcheck49 computer programs.
A single model was selected for further analysis and as a starting
structure for docking simulations.
Acknowledgment. This research was supported by grants
from CNPq (National Council of Research of Brazil),
CAPES, FAPERJ (PRONEX E-26/171.512/2006), UFRJ,
UFAM, Fiocruz, and UFF for contribution to the funding
of this program.
Supporting Information Available: Experimental procedures
and characterization data for azido derivatives and the precur-
sors derivatives; results of the analysis of all 1H-1,2,3-triazoles
synthesized and evaluated in this study (1H NMR, 13C NMR,
and mass spectra). This material is available free of charge via
the Internet at http://pubs.acs.org.
Defining the Binding Site on MAL12. According to the DALI
server (DALI, version 3),50 the 3D structure of T. maritima
4-R-glucanotransferase in complex with modified acarbose (PDB
code 1LWJ, 2.50 A resolution)35 shows significant folding similar-
˚
˚
ity, although low sequence identity, with 1UOK (1.45 A rmsd over
˚
317 CR atoms and 27% of identity) and 2ZE0 (1.52 A rmsd over
305 CR atoms and 29% of identity). Additionally, the catalytic
residues and some of the residues important for substrate binding
are highly conserved in all members of the GH13 family. Thus, the
definition of the putative binding site for acarbose in MAL12 was
based on the structural similarity with the homologous residues in
1LWJ involved in the binding of modified acarbose as found in the
complex. The same binding site was used to perform the molecular
docking of the triazoles. The MAL12/acarbose complex was
effectively generated by superposing the MAL12 modeled struc-
ture onto 1LWJ and then transferring the acarbose atomic
coordinates to the MAL12 model using the Sybyl molecular
modeling package, version 8.0 (Tripos Associates, St. Louis, MO).
References
(1) Dwek, R. A. Glycobiology: toward understanding the function of
sugars. Chem. Rev. 1996, 96, 683–720.
(2) Tattersall, R. Alpha-glucosidase inhibition as an adjunct to the
treatment of type-1 diabetes. Diabetic Med. 1993, 10, 688–693.
(3) Sim, L.; Quezada-Calvillo, R.; Sterchi, E. E.; Nichols, B. L.; Rose,
D. R. Human intestinal maltase-glucoamylase: crystal structure
of the N-terminal catalytic subunit and basis of inhibition and
substrate specificity. J. Mol. Biol. 2008, 375, 782–792.
(4) Henrissat, B. A Classification of glycosyl hydrolases based on
amino-acid-sequence similarities. Biochem. J. 1991, 280, 309–316.
(5) Henrissat, B.; Bairoch, A. Updating the sequence-based classifica-
tion of glycosyl hydrolases. Biochem. J. 1996, 316, 695–696.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2375
(6) Henrissat, B.; Callebaut, I.; Fabrega, S.; Lehn, P.; Mornon, J-P;
Davies, G. Conserved catalytic machinery and the prediction of a
common fold for several families of glycosyl hydrolases. Proc. Natl.
Acad. Sci. U.S.A. 1995, 92, 7090–7094.
(7) Stam, M. R.; Danchin, E. G. J.; Rancurel, C.; Coutinho, P. M.;
Henrissat, B. Dividing the large glycoside hydrolase family 13
into subfamilies: towards improved functional annotations of
R-amylase-related proteins. Protein Eng., Des. Sel. 2006, 19, 555–562.
(8) Henrissat, B.; Davies, G. Structural and sequence-based classification
of glycoside hydrolases. Curr. Opin. Struct. Biol. 1997, 7, 637–644.
(9) Ly, H. D.; Withers, S. G. Mutagenesis of glycosidases. Annu. Rev.
Biochem. 1999, 68, 487–522.
(10) Liu, Y.; Ma, L.; Chen, W. H.; Wang, B.; Xu, Z. L. Synthesis of
xanthone derivatives with extended pi-systems as alpha-glucosidase
inhibitors: insight into the probable binding mode. Bioorg. Med.
Chem. 2007, 15, 2810–2814.
(11) Melo, E. B.; Gomes, A. D.; Carvalho, I. Alpha- and beta-glucosidase
inhibitors: chemical structure and biological activity. Tetrahedron 2006,
62, 10277–10302.
(27) Garcia-Moreno, M. I.; Rodriguez-Lucena, D.; Mellet, C. O.;
Fernandez, J. M. G. Pseudoamide-type pyrrolidine and pyrrolizi-
dine glycomimetics and their inhibitory activities against glycosi-
dases. J. Org. Chem. 2004, 69, 3578–3581.
(28) Rossi, L. L; Basu, A. Glycosidase inhibition by 1-glycosyl-4-phenyl
triazoles. Bioorg. Med. Chem. Lett. 2005, 15, 3596–3599.
(29) Castro, S.; Cherney, E. C.; Snyder, N. L.; Peczuh, M. W. Synthesis
of substituted septanosyl-1,2,3-triazoles. Carbohydr. Res. 2007,
342, 1366–1372.
(30) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. A
stepwise Huisgen cycloaddition process: copper (I)-catalyzed re-
gioselective “ligation” of azides and terminal alkynes. Angew.
Chem., Int. Ed. 2002, 41, 2596–2599.
(31) Lu, Y.; Gervay-Hague, J. Synthesis of c-4 and c-7 triazole analogs
of zanamivir as multivalent sialic acid containing scaffolds. Carbo-
hydr. Res. 2007, 342, 1636–1650.
(32) Lee, B. Y.; Park, S. R.; Jeon, H. B.; Kim, K. S. A new solvent
system for efficient synthesis of 1,2,3-triazoles. Tetrahedron Lett.
2006, 47, 5105–5109.
(33) Laskowski, R. A.; MacArthur, M. W.; Moss, D. S.; Thornton,
J. M. PROCHECK, a program to check the stereochemical quality
of protein structures. J. Appl. Crystallogr. 1993, 26, 283–291.
(34) Luthy, R.; Bowie, J. U.; Eisenberg, D. Assessment of protein
models with 3-dimensional profiles. Nature 1992, 356, 83–85.
(35) Shirai, T.; Hung, V. S.; Morinaka, K.; Kobayashi, T.; Ito, S.
Crystal structure of GH13 alpha-glucosidase GSJ from one of
the deepest sea bacteria. Proteins 2008, 73, 126–133.
(12) Park, H.; Hwang, K. Y.; Kim, Y. H.; Oh, K. H.; Lee, J. Y.; Kim, K.
Discovery and biological evaluation of novel alpha-glucosidase
inhibitors with in vivo antidiabetic effect. Bioorg. Med. Chem. Lett.
2008, 18, 3711–3715.
(13) Park, H.; Hwang, K. Y.; Oh, K. H.; Kim, Y. H.; Lee, J. Y.; Kim, K.
Discovery of novel alpha-glucosidase inhibitors based on the
virtual screening with the homology-modeled protein structure.
Bioorg. Med. Chem. 2008, 16, 284–292.
(36) Davies, G. J.; Wilson, K. S.; Henrissat, B. Nomenclature for sugar-
binding subsites in glucosyl hydrolases. Biochem. J. 1997,321,557–559.
(37) Tschamber, T.; Gessier, F.; Dubost, E.; Newsome, J.; Tarnus, C.;
Kohler, J.; Neuburger, M.; Streith, J. Carbohydrate transition state
mimics: synthesis of imidazolo-pyrrolidinoses as potential nectri-
sine surrogates. Bioorg. Med. Chem. 2003, 11, 3559–3568.
(38) Tomich, C. H.; da Silva, P.; Carvalho, I.; Taft, C. A. Homology
modeling and molecular interaction field studies of alpha-glucosi-
dases as a guide to structure-based design of novel proposed anti-
HIV inhibitors. J. Comput.-Aided Mol. Des. 2005, 19, 83–92.
(39) Vocadlo, D.; Davies, J. Mechanistic insights into glycosidase
chemistry. Curr. Opin. Chem. Biol. 2008, 12, 539–555.
(40) Wallace, A. C.; Laskowski, R. A.; Thornton, J. M. Ligplot: a
program to generate schematic diagrams of protein-ligand inter-
actions. Protein Eng. 1995, 8, 127–134.
(41) Gloster, T. M.; Meloncelli, P.; Stick, R. V.; Zechel, D.; Vasella, A.;
Davies, G. J. Glycosidase inhibition: an assessment of the binding
of 18 putative transition-state mimics. J. Am. Chem. Soc. 2007, 129,
2345–2354.
(14) Park, J. H.; Ko, S.; Park, H. Toward the virtual, screening of alpha-
glucosidase inhibitors with the homology-modeled protein struc-
ture. Bull. Korean Chem. Soc. 2008, 29, 921–927.
(15) Pluempanupat, W.; Adisakwattana, S.; Yibchok-Anun, S.; Chavasiri,
W. Synthesis of N-phenylphthalimide derivatives as alpha-glucosidase
inhibitors. Arch. Pharmacal Res. 2007, 30, 1501–1506.
(16) Zhang, R.; McCarter, J. D.; Braun, C.; Yeung, W.; Brayer, G. D.;
Withers, S. G. Synthesis and testing of 2-deoxy-2,2-dihaloglyco-
sides as mechanism-based inhibitors of alpha-glycosidases. J. Org.
Chem. 2008, 73, 3070–3077.
(17) MacGregor, E. A.; Janecek, S.; Svensson, B. Relationship of
sequence and structure to specificity in the R-amylase family of
enzymes. Biochim. Biophys. Acta 2001, 1546, 1–20.
€
(18) Schmidt, D. D.; Frommer, W.; Muller, L.; Junge, B.; Wingender,
W.; Truscheit, E. R-Glucosidase inhibitors, new complex oligosac-
charides of microbial origin. Naturwissenschaften 1977, 64, 535–
536.
(19) Hanozet, G.; Pircher, H. P.; Vanni, P.; Oesch, B.; Semenza, G. An
example of enzyme hysteresis. The slow and tight interaction of
some fully competitive inhibitors with small intestinal sucrase.
J. Biol. Chem. 1981, 256, 3703–3711.
(20) Chiasson, J.-L.;Josse, R. G.;Hunt,J. A.;Palmason,C.;Rodger,N. W.;
Ross, S. A.; Ryan, E. A.; Tan, M. H.; Wolever, T. M. S. The efficacy of
acarbose in the treatment of patients with non-insulin-dependent
diabetes-mellitus: a multicenter controlled clinical-trial. Ann. Intern.
Med. 1994, 121, 928–935.
(21) Holman, R. R.; Cull, C. A.; Turner, R. C. A randomized double-
blind trial of acarbose in type 2 diabetes shows improved glycemic
control over 3 years (UK Prospective Diabetes Study 44). Diabetes
Care 1999, 22, 960–964.
(22) Jenkins, D. J.; Taylor, R. H.; Goff, D. V.; Fielden, H.; Misiewicz,
J. J.; Sarson, D. L.; Bloom, S. R.; Alberti, K. G. Scope and
specificity of acarbose in slowing carbohydrate-absorption in
man. Diabetes 1981, 30, 951–954.
(42) Chang, C. E. A.; Chen, W.; Gilson, M. K. Ligand configurational
entropy and protein binding. Proc. Natl. Acad. Sci. U.S.A. 2007,
104, 1534–1539.
(43) Kim, M. J.; Lee, S. B.; Lee, H. S.; Lee, S. Y.; Baek, J. S.; Kim, D.;
Moon, T. W.; Robyt, J. F.; Park, K. H. Comparative study of
the inhibition of alpha-glucosidase, alpha-amylase, and cyclo-
maltodextrin glucanosyltransferase by acarbose, isoacarbose, and
acarviosine-glucose. Arch. Biochem. Biophys. 1999, 371, 277–283.
(44) Matsui, T.; Yoshimoto, C.; Osajima, K.; Oki, T.; Osajima, Y.
In vitro survey of alfa-glucosidase inhibitory food components.
Biosci., Biotechnol., Biochem. 1996, 60, 2019–2022.
(45) Kanthi, J. M.; Abesundara, T. M.; Kiyoshi, M. Glucosidase
inhibitory activity of some Sri Lanka plant extracts, one of which,
Cassia auriculata, exerts a strong antihyperglycemic effect in rats
comparable to the therapeutic drug acarbose. J. Agric. Food Chem.
2004, 52, 2541–2545.
(46) Sali, A.; Blundell, T. L. Comparative protein modeling by satisfac-
tion of spatial restraints. J. Mol. Biol. 1993, 234, 779–815.
(47) Watanabe, K.; Hata, Y.; Kizaki, H.; Katsube, Y.; Suzuki, Y. The
refined crystal structure of Bacillus cereus oligo-1,6-glucosidase at
2.0 angstrom resolution: structural characterization of proline-
substitution sites for protein thermostabilization. J. Mol. Biol.
1997, 269, 142–153.
(48) Notredame, C.; Higgins, D. G.; Heringa, J. T-Coffee: a novel
method for fast and accurate multiple sequence alignment.
J. Mol. Biol. 2000, 302, 205–217.
(23) Taylor, R. H.; Jenkins, D. J.; Barker, H. M.; Fielden, H.; Goff,
D. V.; Misiewicz, J. J.; Lee, D. A.; Allen, H. B.; MacDonald, G.;
Wallrabe, H. Effect of acarbose on the 24-h blood-glucose profile
and pattern of carbohydrate-absorption. Diabetes Care 1982, 5,
92–96.
ꢀ
(24) Brogard, J. M.; Willemin, B.; Blickle, J. F.; Lamalle, A. M.; Stahl,
A. Alpha-glucosidase inhibitors. A new concept for the treatment
of diabetes and reactive hypoglycemias. Rev. Med. Interne 1989, 10,
365–374.
(25) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Click chemistry: diverse
chemical function from a few good reactions. Angew. Chem., Int.
Ed. 2001, 40, 2004–2021.
(49) Hooft, R. W.; Vriend, G.; Sander, C.; Abola, E. E. Errors in protein
structures. Nature 1996, 381, 272–272.
(26) Recent reviews: (a) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. The
growing impact of click chemistry on drug discovery. Drug Dis-
covery Today 2003, 8, 1128–1137. (b) Wang, Q.; Chittaboina, S.;
Barnhill, H. N. Advances in 1,3-dipolar cycloaddition reaction of azides
and alkynes. A prototype of “click” chemistry. Lett. Org. Chem. 2005,
2, 293–301. (c) Bock, V. D.; Perciaccante, R.; Jansen, T. P.; Hiemstra,
H.; Maarseveen, J. H. Click chemistry as a route to cyclic tetrapeptide
analogues: synthesis of cyclo-[Pro-Val-(triazole)-Pro-Tyr]. Org. Lett.
2006, 8, 919–922.
€€ €
€
(50) Holm, L.; Kaariainen, S.; Rosenstrom, P.; Schenkel, A. Searching
protein structure databases with DaliLite v.3. Bioinformatics 2008,
24, 2780–2781.
(51) Thomsen, R.; Christensen, M. H. MolDock: a new technique for
high-accuracy molecular docking. J. Med. Chem. 2006, 49, 3315–3321.
(52) Sobolev, V.; Sorokine, A.; Prilusky, J.; Abola, E. E.; Edelman, M.
Automated analysis of interatomic contacts in proteins. Bioinfor-
matics 1999, 15, 327–332.