Design and Synthesis of Labystegines, Hybrid Iminosugars from LAB and Calystegine, as Inhibitors of Intestinal α-Glucosidases: Binding Conformation and Interaction for ntSI

Article abstract of DOI:10.1021/acs.joc.5b00342

This paper identifies the required configuration and orientation of α-glucosidase inhibitors, miglitol, α-1-C-butyl-DNJ, and α-1-C-butyl-LAB for binding to ntSI (isomaltase). Molecular dynamics (MD) calculations suggested that the flexibility around the k

Full text of DOI:10.1021/acs.joc.5b00342

Article
pubs.acs.org/joc
Design and Synthesis of Labystegines, Hybrid Iminosugars from LAB
and Calystegine, as Inhibitors of Intestinal α‑Glucosidases: Binding
Conformation and Interaction for ntSI
Atsushi Kato,*,† Zhao-Lan Zhang,‡ Hong-Yao Wang,‡ Yue-Mei Jia,‡ Chu-Yi Yu,*,‡ Kyoko Kinami,†
Yuki Hirokami,† Yutaro Tsuji,† Isao Adachi,† Robert J. Nash,§ George W. J. Fleet,⊥,∥ Jun Koseki,#
Izumi Nakagome,# and Shuichi Hirono*,#
†Department of Hospital Pharmacy, University of Toyama, Toyama 930-0194, Japan
‡Beijing National Laboratory of Molecular Science (BNLMS), CAS Key Laboratory of Molecular Recognition and Function, Institute
of Chemistry, Chinese Academy of Sciences, Beijing 100190, China
§Institute of Biological, Environmental and Rural Sciences, Phytoquest Limited, Plas Gogerddan, Aberystwyth, Ceredigion SY23 3EB,
United Kingdom
⊥Chemistry Research Laboratory, 12 Mansfield Road, Oxford OX1 3TA, United Kingdom
∥National Engineering Research Center for Carbohydrate Synthesis, Jiangxi Normal University, Nanchang 330022, PR China
#School of Pharmaceutical Sciences, Kitasato University, Tokyo 108-8641, Japan
S
* Supporting Information
ABSTRACT: This paper identifies the required configuration
and orientation of α-glucosidase inhibitors, miglitol, α-1-C-
butyl-DNJ, and α-1-C-butyl-LAB for binding to ntSI
(isomaltase). Molecular dynamics (MD) calculations sug-
gested that the flexibility around the keyhole of ntSI is lower
than that of ctSI (sucrase). Furthermore, a molecular-docking
study revealed that a specific binding orientation with a CH−π
interaction (Trp370 and Phe648) is a requirement for
achieving a strong affinity with ntSI. On the basis of these
results, a new class of nortropane-type iminosugars, labyste-
gines, hybrid iminosugars of LAB and calystegine, have been
designed and synthesized efficiently from sugar-derived cyclic
nitrones with intramolecular 1,3-dipolar cycloaddition or
samarium iodide catalyzed reductive coupling reaction as the key step. Biological evaluation showed that our newly designed
3(S)-hydroxy labystegine (6a) inherited the selectivity against intestinal α-glucosidases from LAB, and its inhibition potency was
10 times better than that of miglitol. Labystegine, therefore, represents a promising new class of nortropane-type iminosugar for
improving postprandial hyperglycemia.
hyperglycaemia and macrovascular complications exists in
diabetic individuals.2,3 Therefore, efficient α-glucosidase in-
hibitors have considerable potential for preventing these
cardiovascular risks.
MGAM and SI, both possessing two homologous glycosyl
hydrolase family 31 (GH31) catalytic subunits, are anchored to
the small intestinal brush border epithelial cells: the C-terminal
subunit (ctMGAM and ctSI) is near the lumen, whereas the N-
terminal subunit (ntMGAM and ntSI) is near the proximal
membrane-bound end.4 ntMGAM and ctMGAM exhibit exo α-
glucosidase activities against α-1,4-linked maltose substrates but
display different specificities for malto-oligosaccharides with
different lengths. In contrast, the ntSI subunit, known as
INTRODUCTION
■
Glycosidases are involved in several important biological
processes, such as intestinal digestion, lysosomal catabolism
of glycoconjugates, and post-translational modification of
glycoproteins. Among these glycosidases, maltose-glucoamylase
(MGAM; EC 3.2.1.20 and 3.2.1.3) and sucrase-isomaltase (SI;
EC 3.2.148 and 3.2.10) are key enzymes responsible for
catalyzing the last glucose-cleaving step in starch digestion.
Inhibition of the α-glucosidase activity of MGAM and SI in the
catabolism of dietary sugars and starches would allow potential
control of blood glucose levels in individuals with type 2
diabetes. Furthermore, a recent meta-regression analysis
revealed that a close relationship exists between 2 h
postprandial glucose levels and cardiovascular risk below the
diabetic threshold,1 and there is growing epidemiological
evidence to show that an association between postprandial
Received: February 15, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 1. Chemical structures of piperidine-, pyrrolidine-, and nortropane-type iminosugars.
isomaltase, also causes hydrolysis of the α-1,6-linkages of starch
and the isomaltose branch points; in addition, the ctSI subunit,
classically referred to as sucrase, has glucosidase activity for the
α-1,2-linkage of sucrose.5 An ingenious synergism is proposed
to exist between these enzymes in order to allow for complete
digestion. These four enzymes have redundant α-1,4 activities,
whereas only ntSI has high activity for α-1,6-linkages of starch
and isomaltose. This difference in their activities and
distribution is closely related to the component ratio of α-1,4
linkages to α-1,6-linkages in human dietary starch, which is
19:1.
The ntMGAM and ntSI are ∼60% identical in amino acid
sequence and higher than with the N-and C-terminal domains
associated with the same enzymes (∼40% sequence iden-
tity).6−8 These sequence differences affect the inhibition
specificity. In comparison with binding potential against
MGAM and SI, acarbose has been shown to be an efficient
inhibitor of the ctMGAM and ctSI (sucrase) but a weaker
inhibitor of the ntSI (isomaltase).9−12 These results are in
agreement with previous reports on the structural of the
ntMGAM−acarbose complex. Acarbose appears to bind to the
ntMGAM-active site mainly through side-chain interactions
with its acarvosine unit; its glycone rings have almost no
interactions with amino acid residues.8 The systematic analysis
and comparison of the inhibition potencies and specificity of a
range of inhibitors is needed in order to obtain a good
understanding of the structural requirements for substrate
reorganization and tight binding in these intestinal α-
glucosidases.
Iminosugars are analogues of pyranoses and furanoses in
which the ring oxygen has been replaced by nitrogen (Figure
1).13−15 They inhibit glycosidases by mimicking the corre-
sponding substrate or the transition-state intermediates of
various glycosidases.16 The intestinal α-glucosidase inhibitor
miglitol was introduced on the market for the treatment of type
2 diabetes in 1996.17−19 Miglitol is designed from the naturally
occurring 1-deoxynojirimycin (DNJ), which closely mimics the
D-glucose structure. This structural similarity to D-glucose can
cause side effects such as diarrhea and impairment of liver
function because it shows nonspecific α-glucosidase inhibition
in various organs.20−22 For better drugs, it is therefore useful to
elucidate the structural requirements for the design of selective
α-glucosidase inhibitors lacking the piperidine structures that
can so closely resemble glucose. Following our recent work
concerning the structure−activity relationships of pyrrolidine
iminosugars,23−26 we have recently reported the synthesis and
biological evaluation of a series of α-1-C-alkylated 1,4-dideoxy-
1,4-imino-L-arabinitol (LAB) derivatives as a new class of
promising compounds that can be used to treat postprandial
hyperglycemia.27 Our results indicate that the structural
similarity to D-glucose is not always necessary for recognition
and tight-binding with ntMGAM. Considering these character-
istic ligand−ntMGAM interactions, we turned our attention to
the inhibition activities and binding interactions against ntSI
and ctSI. These enzymes are essential for the processing of
dietary carbohydrates in the same way as MGAM. A recent
study revealed that a close relationship exists between
expression of SI in colon mucosa and progression to
carcinoma28 and metastatic colon adenocarcinoma.29 The
reason for the observed up-regulation of SI in these colon
cancers is not apparent yet, but it is possibly related to its
metabolic activity.
The overall aim of this study was to generate iminosugars
with suitable structural moieties of iminosugar inhibitors for
active-site binding to SI and to provide general guidelines on
the effective design of nonpiperidine-based intestinal α-
glucosidase inhibitors. In the present study, we first constructed
the three-dimensional structures of the rat ntSI and ctSI by
performing homology modeling. Furthermore, on the basis of
these protein structures, we performed molecular docking
simulations to predict the docking configurations of miglitol, α-
1-C-butyldeoxynojirimycin (α-1-C-butyl-DNJ), and α-1-C-
butyl-1,4-dideoxy-1,4-imino-L-arabinitol (α-1-C-butyl-LAB)
into ntSI (isomaltase). Using this information, a new class of
nortropane-type iminosugars, labystegines, hybrid iminosugars
from LAB and calystegine, were designed as a novel class of
intestinal α-glucosidase inhibitors, which includes their
suppressive effects on blood glucose levels using an isomaltose
loading test and their binding conformations and stabilization
energies with ntSI with the intention of determining the extent
of differences to miglitol. Labystegines and their derivatives will
be synthesized via sugar-derived cyclic nitrone chemistry
developed by our research group as well as others.30−32
Employing D-xylose-derived cyclic nitrone 1 as starting material,
the nortropane scaffolds of labystegines can be constructed
following the protocol of intramolecular [3 + 2] 1,3-dipolar
B
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
cycloaddition developed by Goti30 and/or intramolecular
samarium iodide catalyzed reductive coupling reaction
developed by our research group.32
predicted rat ntSI. Figure 3 shows the binding conformations
(Figure 3A−C) and ligand/protein interactions (Figure 3D−F)
for rat ntSI. Miglitol, α-1-C-butyl-DNJ, and α-1-C-butyl-LAB
were observed to bind to the same hydrophobic pocket
consisting of Trp370, Ile434, Trp512, Trp612, and Phe648 of
rat ntSI. It is noteworthy that the binding orientation of miglitol
was slightly different from that of α-1-C-butyl-LAB and α-1-C-
butyl-DNJ, whereas the hydrogen bonds formed to Asp398,
Trp477, Asp514, and Asp615 in ntSI were maintained in all
models (Figure 3D−F). The butyl group of α-1-C-butyl-DNJ
and α-1-C-butyl-LAB has a favorable interaction with the
hydrophobic pocket (Ile276, Trp370, Trp477, and Phe648).
The hydroxyethyl group of miglitol heads for the same
hydrophobic pocket, whereas the terminal hydroxyl group
gravitates toward the hydrophilic part of the active site and
forms hydrogen bonds with Asp274 (Figure 3A,C). For the
clarification of the difference, we next analyzed the binding
stabilization energies (ΔEs) between each ligand and ntSI
(isomaltase).
RESULTS AND DISCUSSION
■
Protein Structures with Homology Modeling Method.
A comparison of the structures of ntSI (isomaltase) and ctSI
(sucrase) will provide improved insight into their inhibitory
activities on α-1,6- and α-1,2-linked substrates and should allow
the design of new potent and specific inhibitors. At first, we
created three-dimensional structures of the rat ntSI and ctSI
with a homology modeling method, based on human isomaltase
and human maltase, respectively. These predicted three-
dimensional structures were relaxed by a molecular dynamics
(MD) method. From these minimized structures as initial
coordinates, we performed MD calculations to sample the
keyhole structures. The 250 keyhole structures found were
saved and clustered. Parts A and B of Figure 2 show the
Table 1 shows that the negative logarithmic values of the
half-maximal inhibitory concentration (pIC50), calculated ΔE’s,
and relative energies (ΔΔE’s) of each ligand for rat ntSI
(isomaltase). The binding energies (ΔE) for miglitol, α-1-C-
butyl-DNJ, and α-1-C-butyl-LAB are −68.89, −77.11, and
−75.88 (kcal/mol), respectively. These results were consistent
with the experimental results (pIC50) showing that α-1-C-butyl-
DNJ has higher inhibitory activity against isomaltase than both
α-1-C-butyl-LAB and miglitol, with pIC50 for each ligand of
4.41, 6.35, and 5.33 [log (μM−1)], respectively. For further
confirmation, we also constructed the three-dimensional
models of the α-1-C-butyl-LAB−human ntSI complex. The
docking calculation was performed under the same conditions
as the docking calculations of rat ntMGAM using the X-ray
structures of human ntSI (PDB ID: 3LPP).
The binding energy (ΔE), as estimated from the three-
dimensional structures of the compound with human ntSI,
suggested that the α-1-C-butyl-LAB would be an inhibitor of
human ntSI as well as that from the rat (Table 2).
From the viewpoint of binding interactions, the main
difference in binding affinity could be explained by the
difference in van der Waals interactions. In particular, the
CH−π interaction is a very important factor for their affinity.
As shown in Figure 3E,F, α-1-C-butyl-DNJ and α-1-C-butyl-
LAB were found to bind to isomaltase with the same
orientation. Both compounds showed hydrogen bonding
(Asp514 and Asp615) and the CH−π interaction (Trp370
and Phe648). However, the six-membered ring of α-1-C-butyl-
DNJ lies closer to Trp370 and Phe648 than the five-membered
ring of α-1-C-butyl-LAB (Figure 4). These differences in
CH−π interaction may explain the differences in potency of
inhibition against isomaltase.
Influence of α-1-C-Alkyl Chain Length against Binding
Energy in the Complex Structure. In order to clarify the
depth and space of the hydrophobic pocket of ntSI (isomaltase)
and to predict an acceptable alkyl chain length at the anomeric
position, we prepared α-1-C-butyl-LAB, α-1-C-heptyl-LAB, and
α-1-C-nonyl-LAB and constructed three-dimensional structures
of the ntSI (isomaltase) complexed with them. Table 3 shows
the negative logarithmic values of half-maximal inhibitory
concentration (pIC50) values, calculated binding energies
(ΔE’s), and relative energies (ΔΔE’s) of α-1-C-butyl-LAB, α-
1-C-heptyl-LAB, and α-1-C-nonyl-LAB for rat isomaltase. The
values of binding energies (ΔE’s) for each ligand are −75.88,
Figure 2. Clustering dendrogram for keyhole structures of predicted
(A) rat ntSI and (B) ctSI obtained from each molecular dynamics
simulation.
clustering dendrograms for the predicted rat ntSI and ctSI
obtained from these results. As shown in Figure 2, we found
that the number of representative structures for rat ctSI
(sucrase) was larger than for rat ntSI (isomaltase) when we
focused on the same criterion shown by the red dashed line.
This fact suggests that the flexibility around the keyhole of
sucrase is greater than that of isomaltase. Thus, to consider the
difference of keyhole flexibility between isomaltase and sucrase,
we have used the one and seven representative keyhole
structures selected above with the same criteria in each docking
calculation, respectively.
Docking Study of Miglitol, α-1-C-butyl-DNJ, and α-1-
C-butyl-LAB against ntSI. To understand the structural basis
of the interaction of miglitol, α-1-C-butyl-DNJ, and α-1-C-
butyl-LAB with isomaltase (ntSI), we first performed molecular
docking simulation of these inhibitors with one keyhole of
C
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 3. Binding conformations and views of interactions for rat ntSI (isomaltase). Panels A−C, binding of miglitol (A), α-1-C-butyl-DNJ (B), and
α-1-C-butyl-LAB (C) in the active site of ntSI. The molecular surface of ntSI was rendered by lipophilic potential. Panels D−F, schematic diagram of
hydrogen-bonding interactions and vdW interactions between ntSI and miglitol (D), α-1-C-butyl-DNJ (E), and α-1-C-butyl-LAB (F).
isomaltase than α-1-C-heptyl-LAB and α-1-C-nonyl-LAB, with
pIC50 for each ligand of 5.33, 4.80, and 3.77 [log (μM−1)],
respectively. As shown in Figure 5A, the pyrrolidine ring of α-1-
C-heptyl-LAB and α-1-C-nonyl-LAB bound to ntSI (isomal-
tase) has the same conformation as that of α-1-C-butyl-LAB. In
contrast, the binding orientation of these compounds was
different from that of α-1-C-butyl-LAB. These results would
indicate that ring orientation was changed due to accom-
modation of the long side chain in the hydrophobic pocket. As
a result, the distance between NH2+ in the pyrrolidine ring and
Phe648 was reduced from 4.32 (α-1-C-butyl-LAB) to 2.47 (α-
1-C-heptyl-LAB) or 2.31 Å (α-1-C-nonyl-LAB) (Figure 5B−
D), finally leading to loss of binding energies (ΔE’s). These
results strongly suggested that the longer α-1-C-alkyl chains
reduced the inhibition activities compared to α-1-C-butyl-LAB.
Design of Labystegines. In the present study, we
performed molecular dynamics (MD) calculations and found
that the flexibility around the keyhole of ntSI (isomaltase) is
lower than that of ctSI (sucrase) (Figure 2). Furthermore, our
Table 1. Negative Logarithmic Values of Half-Maximal
Inhibitory Concentration (pIC50), Calculated Stabilization
Energies (ΔE’s), and Relative Energies (ΔΔE’s) of Each
Ligand for ntSI (Isomaltase)
miglitol
α-1-C-butyl-DNJ
α-1-C-butyl-LAB
pIC50 (log μM−1
ΔE (kcal/mol)
ΔΔE (kcal/mol)
)
4.41
−68.89
0
6.35
−77.11
−8.22
5.33
−75.88
−6.69
Table 2. Binding Energy Estimated for α-1-C-Butyl-LAB
with Rat and Human ntSI (Isomaltase)
ΔE (kcal/mol)
rat isomaltase
−75.88
−75.44
human isomaltase
−68.64, and −63.76 kcal/mol, respectively. These results were
consistent with the experimental results (pIC50) showing that
of α-1-C-butyl-LAB has higher inhibitory activity against
D
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 4. Binding modes of miglitol (A), α-1-C-butyl-DNJ (B), and α-1-C-butyl-LAB (C) with rat ntSI (isomaltase). The residues related to van der
Waals interactions are drawn in space-filling representation. The hydrogen bonds are depicted by dashed lines. (D) Superposition of the binding
conformations of miglitol, α-1-C-butyl-DNJ, and α-1-C-butyl-LAB with Trp370 and Phe648 in rat ntSI.
class of iminosugar with the bridge head hydroxyl group
replaced by a hydroxymethyl group be named labystegine
(Figure 6). The C-2, C-3, and C-4 OH groups and ring
heteroatom in the six-membered ring of calystegine B2 lie in the
same region of space as the C-3, C-4 OH, and C-5 CH2OH
groups and ring heteroatom of isofagomine, respectively.38
Furthermore, previous studies revealed that replacement of the
C4 hydroxy group of 1,5-dideoxy-1,5-imino-D-xylitol (DIX) by
a hydroxymethyl substituent to give isofagomine bring a
dramatic increase in inhibitory potency against β-glucosidases.39
On a similar basis, noeurostegine (C-4 hydroxymethyl
calystegine B2) was designed and subsequently shown to be a
β-glucosidase inhibitor with potential as a valuable compound
against Gaucher’s disease.40,41 In sharp contrast, our newly
designed labystegines originate from pyrrolidine iminosugars
and can be considered as α-glucosidase-inhibiting LAB
derivatives with an extended bridge at the C-1 and C-4
positions. Furthermore, they differ from original calystegines in
that they have a hydroxymethyl substituent instead of a hydroxy
group in the bridgehead position. We envisaged that
labystegine would inherit the potency and selectivity against
intestinal α-glucosidases from LAB.
Table 3. Comparison of Negative Logarithmic Values of
Half-Maximal Inhibitory Concentration (pIC50), Calculated
Stabilization Energies (ΔE’s), and Relative Energies
(ΔΔE’s) for ntSI (Isomaltase) with the Length of Alkyl
Groups in LAB Derivatives
α-1-C-butyl-
α-1-C-heptyl-
α-1-C-nonyl-
LAB
LAB
LAB
pIC50 (log μM−1
ΔE (kcal/mol)
ΔΔE (kcal/mol)
)
5.33
−75.88
0.00
4.80
−68.64
7.24
3.77
−63.76
12.12
results indicated that the alkyl chain length at the anomeric
position of α-1-C-alkyl-LAB should be limited. These results are
supported by previous reports that the pseudotetrasaccharide
acarbose is an efficient inhibitor of the ctMGAM (maltase) and
ctSI (sucrase) but a weaker inhibitor of the ntSI (iso-
maltase).9−12 In the present study, we also revealed that the
binding orientation of iminosugars toward the keyhole of ntSI
(isomaltase) affected the inhibition potencies: a suitable
binding orientation with CH−π interaction (Trp370 and
Phe648) is a requirement for achieving a strong affinity for
isomaltase. On the basis of these remarkable results, we
envisaged that a further improvement in activity might be
gained in the case of pyrrolidine-based structures having a
bicyclic ring bridge with a suitable CH−π interaction with
Trp370 and Phe648. To test this hypothesis, we drew
inspiration from naturally occurring calystegines to prepare a
hybrid with the pyrrolidine iminosugar LAB. Calystegines are
characterized as nortropane alkaloids with a high degree of
hydroxylation and an unusual hemiaminal functionality at the
bridge head position.33−37 We propose that this new structural
Synthesis of Labystegines. The nortropane scaffolds of
labystegine 6 were prepared following the protocol of
intramolecular [3 + 2] cycloaddition developed by Goti30
(Scheme 1). The nitrone 1 prepared from D-xylose42 reacted
with allyl Grignard reagent to afford only a single diastereomer
2 (dr >95). The high trans-selectivity is in accord with previous
reports.43−46 Oxidation of hydroxylamine 2 with activated
47,48
MnO2
gave two separable regioisomers 3 and 4. The
cycloadduct 5 was obtained as the only product in high yield
E
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 5. Binding conformations of α-1-C-butyl-LAB, α-1-C-heptyl-LAB, and α-1-C-nonyl-LAB for rat ntSI (isomaltase): (A) superposition of
binding conformations of α-1-C-butyl-LAB, α-1-C-heptyl-LAB, and α-1-C-nonyl-LAB in the active site of ntSI. (B−D) Binding conformations of α-1-
C-butyl-LAB (B), α-1-C-heptyl-LAB (C), α-1-C-nonyl-LAB (D), and Phe648 of ntSI. The numbers represent the distance between a hydrogen
attached to the nitrogen of the five-membered ring and a carbon of Phe648.
Scheme 1. Synthesis of Nortropane with [3 + 2]
Cycloaddition
Figure 6. Design strategy of nortropane-type α-glucosidase inhibitors.
when 3 was heated in a dilute toluene solution at 120 °C for 14
h. The relative stereochemistry of the newly generated
stereogenic center was confirmed by 2D NMR (NOESY and
COSY) experiments. Catalytic hydrogenolysis (10% Pd/C) in
acidic methanol furnished the iminosugar 6 in high yield. NMR
data of compounds 2a−5a were completely identical to those
reported in ref 30.
N-Alkyl derivatives were obtained from 5a, the N−O bond of
which was then cleaved by Zn−Cu(OAc)2−AcOH (Scheme 2).
Subsequently, N-alkylation was achieved by reductive amina-
tion of aldehydes with secondary amine 7. After Pd/C-catalyzed
hydrogenolysis, N-alkylated iminosugars 9a,b were obtained in
high yields.
The synthesis of 16 commenced with a stereoselective
nucleophilic addition of Grignard reagent to nitrone 1 to afford
the hydroxylamine 11 in high yield (Scheme 3); 11 was
oxidized with activated MnO2 directly without isolation. After
conversion of the acetal 12 to aldehyde 14, the intramolecular
31,32,49,50
reductive coupling reaction promoted by SmI2
was
attempted; hydroxylamine 15 was obtained as the single
diastereomer. The configuration of 15 was assigned by careful
analyses of 1D and 2D NMR spectra. Catalytic hydrogenolysis
F
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
of 5.1 and 2.8 μM, whereas these calystegines exhibited less
than a 50% inhibition of rat intestinal α-glucosidases even at
concentrations as high as 1000 μM. In sharp contrast, our
designed 6a was a potent inhibitor of rat intestinal maltase,
isomaltase, and sucrase with IC50 value of 0.15, 2.8, and 0.18
μM, respectively. These inhibitory potencies were 10-times
better than those of miglitol. Furthermore, miglitol showed
lower selectivity with inhibition of β-glucocerebrosidase, with
an IC50 value of 219 μM, whereas 6a had no effect on this
enzyme. Thus, we next focused on the introduction of an N-
alkyl substituent into 6a and whether extending the alkyl chain
at this position could influence the inhibition activities of 6a
(Table 4). We found that the inhibitory potency of N-methyl
(9a: IC50 = 254 μM) and N-butyl derivatives (9b: IC50 = 66
μM) against isomaltase were both much weaker than the parent
6a. Furthermore, the branched C-methyl-compound (6b) also
had reduced inhibitory potency against this enzyme (IC50 = 104
μM). It is noteworthy that compound 16 kept its inhibitory
potency, with IC50 values of 4.0 μM, whereas compound 24
showed reduced inhibition (IC50 = 93 μM) compared to
compound 16.
Docking Study of 3(S)-Hydroxylabystegine (6a)
against ntSI. To understand the structural basis of the
interaction of 3(S)-hydroxylabystegine (6a) with isomaltase, we
first determined the mode of inhibition and the inhibition
constant (Ki) from Lineweaver−Burk plots. We found 6a
inhibited isomaltase (ntSI) in a competitive manner, with Ki
value of 0.48 μM. This result suggests that 6a occupies the
active-site of this enzyme in much the same manner as miglitol.
Thus, we next performed in silico molecular docking of this
compound to rat ntSI for the purpose of understanding the
increased inhibitory activity of 6a. Figure 8 shows the binding
conformation of 6a (Figure 8A) and its interactions (Figure
8B,C) with rat ntSI. These results suggested that the binding
mode of 6a is almost same as that of α-1-C-butyl-LAB and
miglitol, whereas a small difference was observed for protein−
ligand interaction. The hydrogen bonds formed to Asp398,
Trp477, and Asp514 of ntSI were maintained in 6a (Figure
8B). In contrast, the bridgehead hydroxymethyl group of 6a
Scheme 2. Synthesis of N-Alkyl Derivatives
(10% Pd/C) in acidic methanol furnished the iminosugar 16 in
quantitative yield.
A similar strategy (Scheme 4) was applied to synthesis of 24
which has a [2.2.1] skeleton; however, the ester bromide 17
was used as the nucleophile to afford the adduct 18. Reduction
of the lactone 18 with LiAlH4, followed by oxidation by
activated MnO2, gave the key nitrone 21. After oxidation of the
alcohol 21 to the aldehyde 22, the hydroxylamine 23 was
obtained as the single diastereomer by an intramolecular
reductive coupling reaction promoted by SmI2. Catalytic
hydrogenolysis (10% Pd/C) in acidic methanol afforded the
iminosugar 24 in quantitative yield.
Structure−Activity Relationships of Labystegine De-
rivatives against Isomaltase and β-Glucocerebrosidase.
We first investigated the inhibition potency of labystegine
derivatives (Figure 7) against rat intestinal maltase, isomaltase,
and sucrase. Human lysosomal β-glucocerebrosidase was also
tested in order to extend our understanding of the selectivity
against the intestinal α-glucosidases. The IC50 of labystegines,
calystegines, and miglitol are shown in Table 4. Classical
nortropane-type iminosugar calystegine B2 and C1 showed
potent inhibition against β-glucocerebrosidase, with IC50 values
Scheme 3. Synthesis of Nortropane by an Intramolecular Reductive Coupling Reaction
G
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 4. Synthesis of 24 with a [2.2.1] Skeleton
Table 4. IC50 Values (μM) for Labystegine Derivatives
against Intestinal α-Glucosidases and Human Lysosome β-
Glucocerebrosidase, Compared with Calystegine C1 and
Miglitol
IC50 (μM)
rat intestinal
human lysosome
β-
compd
maltase
isomaltase
sucrase glucocerebrosidase
a
b
6a
9a
9b
6b
16
24
0.15
33
2.8
254
66
0.18
6.6
NI (4.6%)
NI (6.1%)
NI (6.6%)
NI (3.6%)
NI (8.7%)
NI (3.6%)
5.1
4.8
5.1
1.8
66
4.1
104
4.0
93
11
0.36
18
calystegine B2
calystegine C1 NI (30.2%) NI (38.8%)
miglitol 1.3 39
NI (25.1%) NI (36.8%)
385
319
1.0
2.8
219
a
b
NI: less than 50% inhibition at 1000 μM. Numbers in parentheses
Figure 7. Chemical structures of labystegines, calystegines, and
miglitol.
indicate % inhibition at 1000 μM.
interacted with Asp436 and Trp477 instead of the interaction
with Asp615, which was observed in α-1-C-butyl-LAB. It is
noteworthy that 6a was located closer to Trp370 than miglitol
due to the van der Waals interaction between the bridge part of
6a and the Trp370 and Phe648 of ntSI (Figure 8C, D). The
hydrogen bond and van der Waals interactions may result in
the increase of inhibitory activity of 6a.
In Vivo Isomaltose-Loading Test. To confirm that 6a had
significant effects on blood glucose levels in vivo, an isomaltose
loading test was conducted using miglitol as a positive control
(Figure 9). The animal experimental protocols in this study
were approved by the Animal Experiments Committee of the
University of Toyama. Male C57BL/6J mice (17−20 g) after
an overnight fast were used for acute isomaltose loading tests.
The blood glucose levels were measured by the StatStrip Xpress
kit (Nova Biochemical Co. Ltd.). A control group was loaded
with injection of saline only. Administration of isomaltose (2.5
g/kg body weight p.o.) to fasted mice resulted in a rapid
increase in blood glucose concentrations from 74.8 2.4 to a
maximum of 157.0
4.0 mg/dL after 60 min. A significant
suppressive effect of the blood glucose level was achieved with
2.5 mg/kg body weight of 3(S)-hydroxylabystegine (6a) after
30, 60, and 120 min. A 2.5 mg/kg body weight miglitol also
significantly decreased the blood glucose concentrations at 30
min after isomaltose loading. However, the suppression effects
were clearly weaker than 6a. This effect is dose-dependent, and
5.0 mg/kg body weight of 6a showed a nearly perfect
suppression effect against postprandial hyperglycemia. These
results suggested that 6a is worthy of study as a potential
therapeutic agent for the treatment of diabetes.
H
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 8. Binding conformations and interactions of 6a with rat ntSI (isomaltase): (A) binding conformation of 6a in the active site of the molecular
surface rendered by lipophilic potentia; (B) schematic diagrams of hydrogen bond and van der Waals interactions of 6a; (C) binding mode of 6a.
The residues related to van der Waals interactions are drawn in space-filling representation. The hydrogen bonds are depicted by the dashed line.
(D) Superposition of the binding conformations of 6a and miglitol with Trp370 and Phe648 in rat ntSI.
of ntSI (isomaltase) is lower than that of ctSI (sucrase).
Molecular docking simulations of α-glucosidase inhibitors,
miglitol, α-1-C-butyl-DNJ, and α-1-C-butyl-LAB into ntSI
revealed that the suitable binding orientation with CH−π
interaction (Trp370 and Phe648) is a requirement for
achieving a strong affinity. The design of labystegines not
only inherited suitable hydrogen bonds formed to Asp398,
Trp477, and Asp514 of ntSI from LAB but also gained new
CH−π interactions with Trp370 and Phe648. It was widely
accepted that nortropane iminosugars are ideally suited for the
design of β-glucosidase inhibitors. However, labystegine opens
up new possibilities for α-glucosidase inhibition by bridge-type
bicyclic iminosugars while enhancing a designing flexibility and
providing a new strategy based on inhibition moieties of the
five-membered ring fragment. In particular, 3(S)-hydroxylabys-
tegine (6a) showed selective intestinal maltase, isomaltase, and
sucrase inhibition, and its potency was 10 times, an order of
magnitude, better than that of miglitol. Furthermore, 6a
Figure 9. Effects of miglitol and 6a on blood glucose levels. Blood
strongly suppressed postprandial hyperglycemia, similar to
miglitol in vivo. Compound 6a therefore represents a promising
new class of a nortropane-type iminosugar that has the
potential for treating postprandial hyperglycemia. Finally, the
synthetic method for the preparation of labystegines is general
and efficient for the synthesis of labystegine-type iminosugars
and therefore is one of the very important tools for our future
study on labystegines.
glucose concentrations of male C57BL/6J mice after an oral load with
isomaltose, 2.5 g/kg body weight, with (a) 2.5 (red) mg/kg body
weight miglitol, (b) 2.5 (orange), and (c) 5.0 (green) mg/kg body
weight 6a. The control group was loaded with saline (blue). Each
value represents the mean SEM (n = 4). **: significant difference (p
< 0.01). *: significant difference (p < 0.05) compared with the control.
CONCLUSIONS
■
In summary, we elucidated the best molecular size and
structural moieties of iminosugars for binding to the active
site of ntSI (isomaltase) by molecular dynamics (MD)
calculations and found that the flexibility around the keyhole
EXPERIMENTAL SECTION
■
General Experimental Procedures. All reagents were used as
received from commercial sources without further purification or
I
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
20
prepared as described in the literature. Tetrahydrofuran was distilled
from sodium and benzophenone immediately before use. Reactions
were stirred using Teflon-coated magnetic stirring bars. Analytical
TLC was performed with 0.20 mm silica gel 60F plates with 254 nm
fluorescent indicator. TLC plates were visualized by ultraviolet light or
by treatment with a spray of Pancaldi reagent [(NH4)6MoO4,
Ce(SO4)2, H2SO4, H2O] or a solution 0.5% ninhydrin in acetone.
Chromatographic purification of products was carried out by flash
column chromatography on silica gel (230−400 mesh). Acidic ion-
exchange chromatography was performed on Amberlite IR-120 (H+)
or Dowex 50WX8-400, H+ form. Melting points were determined
using an electrothermal melting point apparatus. Both melting points
and boiling points are uncorrected. Infrared spectra were recorded on
an FT/IR-480 plus Fourier transform spectrometer. NMR spectra
were measured in CDCl3 (with TMS as internal standard) or D2O on
a 300 M (1H at 300 MHz, 13C at 75 MHz), 400 M (1H at 400 MHz,
13C at 100 MHz), or 600 M (1H at 600 MHz, 13C at 150 MHz) NMR
spectrometer. Chemical shifts (δ) are reported in ppm, and coupling
constants (J) are reported in Hz. The following abbreviations were
used to explain the multiplicities: s = singlet, d = doublet, t = triplet, q
= quartet, m = multiplet. High-resolution mass spectra (HRMS) were
recorded on a LTQ/FT mass spectrometer or a GCT mass
spectrometer. Polarimetry was carried out using an AA-10R polar-
imeter, and the measurements were made at the sodium D line with a
0.5 dm path length cell. Concentrations (c) are given in grams per 100
mL.
Synthesis of 3(S)-Hydroxylabstegine (6a). (2S,3S,4S,5S)-1-
Hydroxy-2-allyl-3,4-bis(benzyloxy)-5-(benzyloxymethyl)pyrrolidine
(2a). A solution of allylmagnesium bromide in Et2O (1 M, 9.6 mL)
was slowly added by syringe under argon to a cooled (0 °C) solution
of nitrone 1 (1 g, 2.4 mmol) in THF; the resulting reaction mixture
was stirred for 30 min, quenched with aq NH4Cl, and allowed to warm
to room temperature. The aqueous layer was extracted with ethyl
acetate, the combined extracts were dried, and the solvents were
removed in vacuo. The residue was purified by flash chromatography
(EtOAc/petroleum ether, 1:2) to give the title compound 2a as a
colorless oil (947 mg, 86%): Rf = 0.73 (EA/PE = 1:2); [α]D20 +3.4 (c
0.48 CH2Cl2); IR (KBr, film) ν 3243, 3067, 3032, 2867, 1639, 1490,
1452, 1363, 1305, 1208, 1100, 914 cm−1; 1H NMR (300 MHz CDCl3)
δ 7.29−7.16 (m, 15H, Ph), 7.06 (br, 1H, OH), 5.83−5.78 (m, 1H, H-
8), 5.09−5.02 (m, 2H, H-9), 4.53−4.41 (m, 6H, CH2Ph), 3.95 (dd, J =
2.4, 3.9 Hz, 1H, H-4), 3.83−3.75 (m, 2H, H-3, H-6), 3.62 (dd, J = 6.6,
9.0 Hz, 1H, H-6), 3.51−3.46 (m, 1H, H-5), 3.33−3.27 (m, 1H, H-2),
2.74−2.65 (m, 1H, H-7), 2.33−2.23 (m, 1H, H-7); 13C NMR (75
MHz CDCl3) δ 138.31, 138.25 (Ph), 135.6 (C-8), 128.39, 128.37,
127.94, 127.90, 127.8, 127.7 (Ph), 117.0 (C-9), 85.5 (C-3), 84.3 (C-
4), 73.4, 71.7 (CH2Ph), 69.9 (C-5), 69.4 (C-2), 68.4 (C-6), 32.7 (C-
7); DEPT-135 (75 MHz CDCl3) δ positive: 135.6, 128.4, 128.0,
127.9, 127.8, 127. 7, 127.6, 85.5, 84.3, 69.9, 69.4, negative: 117.0, 73.4,
71.7, 32.7; HRMS (EI) m/z calcd for C29H33NO4[M]+ 459.2410•
found 459.2413.
for [C29H31NO4H+] 458.2326, found 458.2326. 4a: [α]D +29.1 (c
3.30, CH2Cl2); IR (cm−1) 3033, 2866, 1721, 1543, 1452, 1363, 1212,
1094; 1H NMR (300 MHz, CDCl3) δ 7.31−7.18 (m, 15H, Ph), 6.62−
6.51 (m, 2H, H-7, H-8), 4.84 (s, 1H, H-4), 4.64−4.45 (m, 6H, 3 ×
CH2Ph), 4.27 (s, 1H, H-5), 4.09 (s, 1H, H-3), 4.01 (dd, J = 6.0, 9.9
Hz, 1H, H-6), 3.88 (dd, J = 2.7, 9.6 Hz, 1H, H-6), 1.87 (d, J = 5.4 Hz,
3H, H-9); 13C NMR (75 MHz CDCl3) δ 147.8 (C-2), 137.7, 137.3,
137.2 (Ph), 137.2 (C-7), 128.6, 128.5, 128.4, 127.1, 128.1, 128.0,
127.9, 127.8 (Ph), 118.9 (C-8), 83.5 (C-4), 78.2 (C-3, C-5), 73.6, 71.5,
70.4 (CH2Ph), 67.4 (C-6), 19.5 (b C-9); HRMS (EI) m/z calcd for
+
[C29H31NO4 ] 457.2253, found 457.2257.
(1R*,3S*,4S*,5S*,6S*)-3-(Benzyloxymethyl)-4,5-bis(benzyloxy)-7-
aza-8-oxytricyclo[4.2.1.03,7]nonane (5a). A solution of nitrone 3a
(195 mg, 0.43 mmol) was dissolved in anhydrous toluene and heated
at reflux under an argon atmosphere. After the reaction was completed,
the solvent was evaporated and the resulting crude product was
purified by flash chromatography to give the title compound 5a as
20
yellow light oil (170 mg, 87%): [α]D +27.6 (c 3.48, CH2Cl2); IR
(cm−1) 3062, 3029, 2981, 2947, 2862, 1495, 1453, 1361, 1300, 1254,
1209, 1105, 1026; 1H NMR (300 MHz CDCl3) δ 7.40−7.22 (m, 15H,
PhH), 4.74−4.70 (m, 2H, H-1, CH2Ph), 4.59−4.43 (m, 5H, CH2Ph),
4.34 (d, J = 1.5 Hz, 1H, H-5), 3.90 (d, J= 2.4 Hz, 1H, H-4), 3.63 (dd, J
= 2.1, 10.2 Hz, 1H, H-6), 3.52 (d, J = 10.2 Hz, 1H, H-9), 3.35 (d, J =
10.2 Hz, 1H, H-9), 2.22−2.12 (m, 1H, H-7), 1.98 (d, J = 12.3 Hz, 1H,
H-2), 1.44−1.37 (m, 2H, H-2, H-7); 13C NMR (75 MHz, CDCl3) δ
138.4, 138.2, 137.9 (Ph), 128.44, 128.35, 128.33, 127.9, 127.9, 127.8,
127.7, 127.5 (Ph), 86.9 (C-4), 83.1 (C-5), 79.0 (C-1), 76.8 (C-3), 72.3
(C-9), 73.6, 72.9, 71.4 (CH2Ph), 67.3 (C-6), 39.7 (C-7), 37.1(C-2);
+
HRMS (EI) m/z calcd for [C29H31NO4 ] 457.2256, found 457.2253.
(1S,3R,5S,6S,7S)-1-(Hydroxymethyl)-8-azabicyclo[3.2.1]octane-
3,6,7-triol (6a). 10% Pd/C (30 mg) and concd HCl (1 mL) were
added to a solution of cycloadduct 5a (90 mg, 0.20 mmol) in MeOH.
The mixture was stirred under an atmosphere of H2 at room
temperature for 16 h when the cycloadduct was judged to diminish by
TLC, and then Ar was bubbled through the reaction mixture and the
Pd/C was filtered off. The solvent was removed under reduced
pressure, and NH3·H2O (25%, 10 mL) was added. After being stirred
for 10 min, the solvent was removed under reduced pressure and the
residue was purified by acid ion-exchange resin column (Dowex
20
50WX8−400, H+ form) to give 6a as yellow oil (33 mg, 89%): [α]D
+0 (c 1.66, MeOH); IR (cm−1) 3343, 2935, 1091, 1052; H NMR
1
(300 MHz, D2O) δ 4.04−3.93 (m, 2H, H-4, H-7), 3.86 (s, 1H, H-6),
3.48 (s, 2H, H-8), 3.12 (s, 1H, H-5), 2.11−2.07 (m, 1H, H-4), 1.98 (q,
J = 6.0 Hz, 1H, H-2), 1.50−1.41 (m, 1H, H-4), 1.31 (t, J = 12.3 Hz,
1H, H-2); 13C NMR (75 MHz, D2O) δ 81.8 (C-6), 81.1 (C-7), 66.1
(C-1), 63.8 (C-8), 63.0 (C-3), 60.6 (C-5), 36.6 (C-4), 35.3 (C-2); m/z
calcd for [C8H15NO4H+] 190.1074, found 190.1073.
Synthesis of 3(S)-3-Hydroxy-3-methyllabystegine (6b).
(2S,3S,4S,5S)-1-Hydroxy-2-(benzyloxymethyl)-3,4-bis(benzyloxy)-5-
(2-methylallyl)pyrrolidine (2b). A solution of (2-methylallyl)-
magnesium bromide in THF (1 M, 9.6 mL) was slowly added by
syringe under argon to a cooled (0 °C) solution of nitrone 1 (1 g, 2.4
mmol) in THF; the resulting reaction mixture was stirred for 30 min,
quenched with aq NH4Cl, and allowed to warm to room temperature.
The aqueous layer was extracted with ethyl acetate, the combined
extracts were dried, and the solvents were removed in vacuo. The
residue was purified by flash chromatography (EtOAc/petroleum
ether, 1:2) to give the title compound 2b as a colorless oil (1.16g,
91%): [α]D20 −27 (c 1.05, CHCl3); IR (KBr, film) ν 3239, 3065, 3030,
2915, 2862, 1646, 1495, 1453, 1362, 1307, 1206, 1177, 1098, 1027,
(3S,4S,5S)-2-(Benzyloxymethyl)-3,4-bis(benzyloxy)-5-allyl-1-pyr-
roline N-Oxide (3a) and (3S,4S,5S)-2-((E)-Prop-1-en-1-yl)-3,4-bis-
(benzyloxy)-5-(benzyloxymethyl)-1-pyrroline N-Oxide (4a). Acti-
vated manganese dioxide (0.38 g, 4.36 mmol) was added portionwise
to a cooled (0 °C) solution of hydroxylamine 2a (1 g, 2.18 mmol) in
CH2Cl2. The suspension was stirred at room temperature for 12 h.
The resultant mixture was filtered with Celite and concentrated in
vacuo. The crude product was purified by flash chromatography to
give the title compound 3a as a colorless oil (600 mg, 60%) and 4a
20
also as a colorless oil (270 mg, 27%): 3a [α]D +53.3 (c 3.15,
892, 818, 736, 696 cm−1; H NMR (300 MHz CDCl3) δ 7.31−7.24
1
CH2Cl2); IR (cm−1) 3033, 2916, 2865, 1722, 1590, 1494, 1451, 1361,
1248, 1210, 1009, 920; 1H NMR (300 MHz, CDCl3) δ 7.31−7.26 (m,
15H, Ph), 5.76−5.68 (m, 1H, H-8), 5.14−5.09 (m, 2H, H-9), 4.70−
4.51 (m, 6H, H-3, H-6, CH2Ph), 4.42−4.33 (m, 3H, H-6, CH2Ph),
3.98 (d, J = 6.9 Hz, 1H, H-5), 3.87 (s, 1H, H-4), 2.94−2.89 (m, 1H, H-
7), 2.51−2.44 (m, 1H, H-7); 13C NMR (75 MHz CDCl3) δ 142.6 (C-
2), 137.5, 137.1 (Ph), 132.5 (C-8), 128.6, 128.5, 128.1, 128.03, 127.97,
127.8 (Ph), 119.2 (C-9), 82.9 (C-3), 79.9 (C-4), 78.5 (C-5), 73.6,
72.5, 71.4 (CH2Ph), 62.7 (C-6), 35.4 (C-7); HRMS (ESI) m/z calcd
(m, 15H, PhH), 6.19 (br, 1H, OH), 4.79 (d, J = 8.7 Hz, 2H, H-9),
4.57−4.37 (m, 6H, 6 × CH2Ph), 3.93−3.92 (m, 1H, H-4), 3.84−3.83
(m, 1H, H-3), 3.80 (dd, J = 5.1, 9.6 Hz, 1H, H-6), 3.63 (dd, J = 6.6, 9.0
Hz, 1H, H-6), 3.53−3.42 (m, 2H, H-2, H-5), 2.69 (dd, J = 5.4, 13.8
Hz, 1H, H-7), 2.27 (dd, J = 9.3, 14.1 Hz, 1H, H-7), 1.75 (s, 3H, H-10);
13C NMR (75 MHz CDCl3) δ 143.2 (C-8), 138.2, 138.12, 138.08,
128.37, 128.36, 128.32, 128.0, 127.9, 127.71, 127.68, 127.65, 127.61,
112.7 (C-9), 85.9 (C-4), 84.3 (C-3), 73.3, 71.6, 71.6, 70.0 (C-2), 68.6
J
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(C-6), 68.0 (C-5), 35.9 (C-7), 22.5 (C-10); HRMS (ESI) m/z calcd
for [C35H35NO4H+] 534.2639, found 534.2643.
(Na2SO4) and concentrated at reduced pressure. The residue was
purified by flash column chromatography (silica gel, petroleum ether/
AcOEt 1/1) to afford amine 7 as a colorless solid (2.64 g, 88%): mp
(3S,4S,5S)-2-(Benzyloxymethyl)-3,4-bis(benzyloxy)-5-(2-methyl-
allyl)-1-pyrroline N-Oxide (3b) and (3S,4S,5S)-2-(2-Methylallyl)-3,4-
bis(benzyloxy)-5-(benzyloxymethyl)-1-pyrroline N-Oxide (4b). The
reaction of activated manganese dioxide (184 mg, 2.12 mmol) and
hydroxylamine 2b (500 mg, 1.06 mmol) gave 3b as colorless oil (290
20
123−124 °C; [α]D + 4.0 (c 1.0, CHCl3); IR (cm−1) 3235, 3026,
2944, 2885, 1496, 1452, 1391, 1361, 1214, 1126, 1092, 1072, 885, 755,
738, 696; 1H NMR (300 MHz, CDCl3) δ 7.32−7.22 (m, 15H), 4.55−
4.32 (m, 6H), 4.11−4.00 (m, 1H), 3.97 (s, 1H), 3.76−3.75 (d, J = 1.5
Hz, 1H), 3.41−3.30 (m, 3H), 2.62 (br, 2H), 2.06−1.96 (m, 2H),
1.61−1.52 (m, 1H), 1.27 (t, J = 11.2 Hz, 1H); 13C NMR (75 MHz,
CDCl3) δ138.4, 138.2, 138.0, 128.5, 128.41, 128.37, 128.0, 127.8,
127.3, 88.5, 88.4, 73.3, 72.9, 72.7, 71.2, 64.7, 64.1, 57.9, 39.4, 38.6;
HRMS (ESI) m/ z calcd for [C29H33NO4H +] 460.2488, found
460.2483.
20
mg, 58%) and 4b also as a colorless oil (115 mg, 23%). 3b: [α]D
+70.0 (c 1.0, CH2Cl2); IR (cm−1) 3063, 3031, 2917, 2863, 1649, 1587,
1496, 1453, 1358, 1246, 1208, 1092, 1069, 1028, 901, 739, 698; H
1
NMR (300 MHz, CDCl3) δ 7.32−7.21 (m, 15H, PhH), 4.84 (s, 1H,
H-9), 4.73−4.52 (m, 7H, H-6, H-9, H-3, 4 × CH2Ph), 4.39−4.32 (m,
3H, H-6, 2 × CH2Ph), 4.08 (d, J = 9.0 Hz, 1H, H-5), 3.88 (s, 1H, H-
4), 3.04 (dd, J = 2.7, 14.1 Hz, 1H, H-7), 2.34 (dd, J = 12, 14.1 Hz, 1H,
H-7), 1.76 (s, 3H, H-10); 13C NMR (75 MHz, CDCl3) δ142.3 (C-2),
141.0 (C-8), 137.52, 137.48, 137.2, 128.6, 128.5, 128.5, 128.2, 128.10,
128.07, 128.04, 127.99, 127.8, 114.1 (C-9), 82.6 (C-3), 80.0 (C-4),
78.1 (C-5), 73.6, 72.5, 71.3, 62.96 (C-6), 39.7 (C-7), 22.4 (C-10);
HRMS (ESI) m/z calcd for [C30H33NO4Na+] 494.2302, found
Synthesis of 3(S)-Hydroxy-N-methyllabystegine (9a).
(1S,3R,5S,6S,7S)-6,7-Bis(benzyloxy)-1-((benzyloxy)methyl)-8-meth-
yl-8-azabicyclo[3.2.1]octan-3-ol (8a). To a solution of 7 (138 mg, 0.3
mmol) in CH3OH (10 mL) were added 40% HCHO (67.5 mg, 0.9
mmol) and 10% Pd/C (30 mg), the mixture was stirred under an
atmosphere of H2 at room temperature for 16 h, the flask was bubbled
with Ar, and the Pd/C was filtered off. The solvent was removed under
reduced pressure, and the residue was purified by flash column
chromatography (silica gel, petroleum ether/AcOEt 3/1) to afford 8a
20
494.2298 (−0.81 ppm). 4b [α]D +68 (c 1.0, CH2Cl2); IR (cm−1)
3062, 3031, 2911, 2865, 1650, 1590, 1523, 1496, 1453, 1364, 1256,
1
1209, 1092, 1073, 1027, 901, 818, 738, 698; H NMR (300 MHz,
20
CDCl3) δ 7.33−7.22 (m, 15H, PhH), 4.82 (s, 1H, H-9), 4.74 (s, 1H,
H-9), 4.60−4.45 (m, 7H, 6 × CH2Ph, H-3), 4.23−4.22 (m, 1H, H-4),
4.10−4.09 (m, 1H, H-5), 4.02 (dd, J = 6.0, 10.2 Hz, 1H, H-6), 3.85
(dd, J = 3.0, 9.9 Hz, 1H, H-6), 3.46 (d, J = 15.0 Hz, 1H, H-7), 3.06 (d,
J = 15.0 Hz, 1H, H-7), 1.74 (s, 3H, H-10); 13C NMR (75 MHz,
CDCl3) δ 144.3 (C-8), 139.3 (C-2), 137.8, 137.4, 137.2, 128.6, 128.52,
128.46, 128.40, 128.1, 128.04, 127.98, 127.9, 127. 8, 113.4 (C-9), 84.5
(C-3), 78.2 (C-5), 77.7 (C-4), 73.5, 71.8, 71.7, 67.0 (C-6), 32.9 (C-7),
23.0 (C-10); HRMS (ESI) m/z calcd for [C30H33NO4Na+] 494.2302,
found 494.2298 (−0.77 ppm).
as a colorless oil (121 mg, 85%): [α]D + 12.0 (c 1.9, CHCl3); IR
(cm−1) 3378, 3030, 2933, 2860, 1662, 1496, 1454, 1308, 1073, 1039,
1
737, 700; H NMR (300 MHz, CDCl3) δ 7.34−7.18 (m, 15H), 4.62
(d, J = 12.3 Hz, 1H), 4.52−4.35 (m, 5H), 3.95−3.91 (m, 1H), 3.85 (s,
1H), 3.57 (s, 1H), 3.32 (s, 3H), 2.40 (s, 3H), 1.82 (br, 1H), 1.77−1.65
(m, 2H), 1.56−1.45 (m, 2H); 13C NMR (75 MHz, CDCl3) δ 138.4,
138.3, 138.0, 128.2, 128.1, 127.7, 127.6, 127.50, 127.46, 127.42, 87.7,
85.5, 77.4, 73.2, 72.2, 72.1, 71.2, 65.3, 64.0, 63.2, 31.2, 30.0, 29.2;
HRMS (ESI) m/z calcd for [C30H35NO4H+] 474.2639, found
474.2638.
(1R*,3S*,4S*,5S*,6S*)-1-Methyl-3-(benzyloxymethyl)-4,5-bis-
(benzyloxy)-7-aza-8-oxytricyclo[4.2.1.03,7]nonane (5b). Compound
5b was obtained as a colorless oil (146 mg, 73%): [α]D20 +29.1 (c 0.55,
CH2Cl2); IR (cm−1) 3062, 3030, 2929, 2868, 1496, 1453, 1384, 1361,
1208, 1097, 1028, 909, 832, 737, 698; 1H NMR (300 MHz, CDCl3) δ
7.32−7.23 (m, 15H, PhH), 4.75−4.43 (m, 6H, 6 × CH2Ph), 4.32 (s,
1H, H-5), 3.90 (d, J = 2.1 Hz, 1H, H-4), 3.77 (d, J = 10.2 Hz, 1H, H-
6), 3.56 (d, J = 10.2 Hz, 1H, H-10), 3.38 (d, J = 10.5 Hz, 1H, H-10),
2.12 (J = 12.3 Hz, 1H, H-2), 2.05 (td, J = 3.3, 11.1 Hz, 1H, H-9),
1.53−1.50 (m, 4H, H-9, H-11), 1.26 (dd, 1H, H-2); 13C NMR (75
MHz, CDCl3) δ138.5, 138.3, 137.9, 128.4, 128.31, 128.29, 127.9,
127.8, 127.73, 127.67, 127.64, 127.5, 87.6 (C-1), 87.0 (C-4), 83.6 (C-
5), 78.3 (C-3), 73.5, 73.1, 72.2, 71.3 (C-10), 69.1 (C-6), 45.0 (C-9),
42.6 (C-2), 17.1 (C-11); HRMS (ESI) m/z calcd for
[C30H33NO4Na+] 494.2302, found 494.2296 (−1.21 ppm).
(1S,3R,5S,6S,7S)-1-(Hydroxymethyl)-8-methyl-8-azabicyclo-
[3.2.1]octane-3,6,7-triol (9a). Compound 9a was obtained as a yellow
oil (49 mg, 95%): [α]D20 −4.0 (c 2.0, MeOH); IR (cm−1) 3348, 2938,
1470, 1301, 1173, 1130, 1095, 1041, 727; 1H NMR (400 MHz, D2O)
δ 3.96−3.91 (m, 1H), 3.88 (d, J = 3.0 Hz, 2H), 3.55 (d, J = 12.5 Hz,
1H), 3.51 (d, J = 12.6 Hz, 1H), 3.12 (s, 1H), 2.35 (s, 3H), 1.79−1.76
(m, 2H), 1.68 (dd, J = 6.3, 13.5 Hz, 1H), 1.48 (t, J = 11.1 Hz, 3H); 13C
NMR (100 MHz, D2O) δ 80.7, 79.1, 67.2, 66.7, 62.9, 62.4, 30.3, 28.7,
27.4; HRMS (ESI) m/z calcd for [C9H17NO4H+] 204.1230, found
204.1232 (1.05 ppm).
Synthesis of N-Butyl-3(S)-hydroxylabystegine (9b).
(1S,3R,5S,6S,7S)-6,7-Bis(benzyloxy)-1-((benzyloxy)methyl)-8-butyl-
8-azabicyclo[3.2.1]octan-3-ol (8b). To a solution of 7 (138 mg, 0.3
mmol) in CH3OH (10 mL) were added butyraldehyde (65 mg, 0.9
mmol) and 10% Pd/C (30 mg), the mixture was stirred under an
atmosphere of H2 at room temperature for 16 h, the flask was bubbled
with Ar, and the Pd/C was filtered off. The solvent was removed under
reduced pressure, and the residue was purified by flash column
chromatography (silica gel, petroleum ether/AcOEt 5/1) to afford 8b
(1S,3R,5S,6S,7S)-1-(Hydroxymethyl)-3-methyl-8-azabicyclo-
[3.2.1]octane-3,6,7-triol (6b). Compound 6b was obtained as a
20
colorless oil (70 mg, 91%): [α]D +44.4 (c 1.26, MeOH); IR (cm−1)
3355, 2938, 1217, 1061; 1H NMR (300 MHz, D2O) δ 4.09 (d, J = 3.3
Hz, 1H, H-7), 4.05 (d, J = 3.0 Hz, 1H, H-6), 3.47−3.35 (m, 3H, H-5, 2
× H-8), 2.06 (dd, J = 3.3, 12.0 Hz, 1H, H-4), 1.96 (d, J = 12.9 Hz, 1H,
H-2), 1.70 (dd, J = 2.7, 12.3 Hz, 1H, H-4), 1.43 (s, 3H, H-9), 1.27 (dd,
J = 3.3, 12.9 Hz, 1H, H-2); 13C NMR (75 MHz, D2O) δ 89.2 (C-3),
80.7 (C-6), 79.0 (C-1), 76.5 (C-7), 71.8 (C-5), 63.1 (C-8), 44.7 (C-4),
40.2 (C-2), 15.9 (C-9); HRMS (ESI) m/z calcd for [C9H17NO4H+]
204.1230, found 204.1233 (1.05 ppm).
20
as a colorless oil (128 mg, 83%): [α]D + 22.0 (c 1.6, CHCl3); IR
(cm−1) 3403, 3028, 2930, 2858, 1496, 1454, 1362, 1100, 1072, 735,
697; 1H NMR (300 MHz, CDCl3) δ 7.35−7.18 (m, 15H), 4.61 (d, J =
12.2 Hz, 1H), 4.50−4.39 (m, 5H), 4.01−3.90 (m, 1H), 3.80 (s, 1H),
3.58 (d, J = 2.1 Hz, 1H), 3.41 (s, 1H), 3.37 (d, J = 9.9 Hz, 1H), 3.32
(d, J = 9.9 Hz, 1H), 2.63 (t, J = 7.0 Hz, 2H), 1.74−1.25 (m, 9H), 0.91
(t, J = 7.2 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 138.43, 138.38,
138.2, 128.3, 128.22, 128.15, 128.0, 127.7, 127.6, 127.4, 87.6, 85.4,
77.4, 73.1, 72.7, 72.1, 71.2, 65.7, 64.2, 58.6, 42.5, 30.2, 30.0, 29.6, 20.8,
14.1; HRMS (ESI) m/z calcd for [C33H41NO4H+] 516.3108, found
516.3104.
Synthesis of N-Alkyl-3(S)-hydroxylabystegines.
(1S,3R,5S,6S,7S)-6,7-Bis(benzyloxy)-1-((benzyloxy)methyl)-8-
azabicyclo[3.2.1]octan-3-ol (7). Copper(II) acetate (132 mg, 0.66
mmol) was added to a suspension of Zn powder (4.29 g, 66 mmol) in
acetic acid (20 mL), and the mixture was stirred at room temperature
for 2 h after which the color turned to brown. A solution of 5a (3 g,
6.6 mmol) in AcOH (20 mL) was added, the reaction mixture was
stirred at room temperature for 3 h, and acetic acid was removed
under reduced pressure. The residue was dissolved in ethyl acetate, the
pH adjusted to 8 by addition of aqueous solution of NaHCO3, and the
mixture was filtered with Celite. The filtrate was extracted with ethyl
acetate (3 × 30 mL), and the combined organic layers were dried
(1S,3R,5S,6S,7S)-8-Butyl-1-(hydroxymethyl)-8-azabicyclo[3.2.1]-
octane-3,6,7-triol. Compound 9b was obtained as a yellow oil (55 mg,
20
91%): [α]D +30.0 (c 1.8, MeOH); IR (cm−1) 3359, 2934, 1456,
1
1377, 1300, 1160, 1096, 1022, 723; H NMR (400 MHz, D2O) δ
3.98−3.90 (m, 1H), 3.87 (s, 2H), 3.58 (d, J = 12.2 Hz, 1H), 3.52 (d, J
= 12.3 Hz, 1H),3.30 (s, 1H), 2.75−2.70 (m, 1H), 2.55−2.50 (s, 1H),
1.79−1.66 (m, 3H), 1.39−1.33 (m, 3H), 0.91 (t, J = 7.2 Hz, 3H); 13C
NMR (100 MHz, D2O) δ 80.9, 79.0, 67.3, 63.1, 62.8, 62.0, 42.2, 28.4,
K
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
28.1, 27.7, 20.3, 13.3; HRMS (ESI) m/z calcd for [C12H23NO4H+]
246.1700, found 246.1699.
H-6, 2 × CH2Ph), 3.89 (t, J = 5.7 Hz, 1H, H-5), 3.71 (s, 1H, H-4),
2.54 (t, J = 7.2 Hz, 2H, H-8), 2.16−2.05 (m, 2H, H-7); 13C NMR (75
MHz, CDCl3) δ 200.6 (C-9), 143.0 (C-2), 137.4, 137.3, 136.9, 128.63,
128.58, 128.5, 128.21, 128.18, 128.1, 128.0, 127.8, 82.5 (C-3), 81.0 (C-
4), 77.8 (C-5), 73.6, 72.7, 71.5, 62.7 (C-6), 39.6 (C-8), 23.6 (C-7);
HRMS (ESI) m/z calcd for [C29H31NO5H+] 474.2275, found
474.2262.
Synthesis of 2(R)-Hydroxylabystegine (16). (2S,3S,4S,5S)-1-
Hydroxy-2-(benzyloxymethyl)-3,4-bis(benzyloxy)-5-(3,3-
dimethoxypropyl)pyrrolidine (11). To a solution of cyclic nitrone 1
(20.0 g, 48.0 mmol) was added dropwise a solution of Grignard
reagent in THF (100 mL) [prepared by heating Mg turnings (3.5 g,
144 mmol) and bromide 10 (22 g, 120 mmol) under Barbier
conditions]. The reaction mixture was stirred for 30 min, quenched
with saturated NH4Cl (50 mL) solution, and extracted with AcOEt (3
× 100 mL). The combined extracts were dried with MgSO4, and the
solvents were removed in vacuo. The residue was recrystallized
(petroleum ether/AcOEt 2:1 15 mL) at 0 °C and filtered to give
hydroxylamine 11 as a white powder (17.4 g). The filtrate was
concentrated and purified by flash column chromatography (silica gel,
petroleum ether/AcOEt 3:1) to give hydroxylamine 11 as a white
(1R,2R,5S,6S,7S)-1-(Benzyloxymethyl)-6,7-bis(benzyloxy)-8-
azabicyclo[3.2.1]octane-2,8-diol (15). To a stirred and carefully
deoxygenated solution of aldehyde 14 (437 mg 0.92 mmol) in THF
(18 mL) was added deoxygenated water (1.32 mL, 73.6 mmol) under
an Ar atmosphere. The mixture was cooled to −78 °C to which was
added freshly prepared SmI2 in THF (23 mL, 1.84 mmol). The
reaction was completed within 10 min and then quenched by
successively adding aqueous solutions of Na2S2O3 (10 mL) and
NaHCO3 (10 mL); the aqueous layer was extracted with ethyl acetate
(3 × 10 mL), and the combined organic layers were washed with a
saturated aqueous solution of NaCl, dried over MgSO4. Filtration,
concentration in vacuo, and purification by chromatography (silica gel,
petroleum ether/AcOEt = 4/1) afforded the N-hydroxylamino alcohol
20
powder (5.6 g): yield 92%; mp 48−50 °C; [α]D + 7.6 (c 1.05,
CH2Cl2); IR (cm−1) 3416, 2931, 1649, 1454, 1122, 738, 698; 1H
NMR (300 MHz, CDCl3) δ 7.33−7.25, (m, 15H), 5.37 (s, 1H, OH),
4.59−4.41 (m, 7H), 3.96−3.94 (m, 1H), 3.81 (dd, J = 2.8, 3.5 Hz,
1H), 3.76−3.75 (m, 1H), 3.59−3.53 (m, 2H), 3.29 (s, 3H), 3.27 (s,
3H), 3.19−3.12 (m, 1H), 1.74−1.59 (m, 4H); 13C NMR (75 MHz,
CDCl3) δ 138.2, 138.2, 138.1, 128.4, 128.3, 128.0, 127.9, 127.7, 127.7,
127.6, 104.4, 86.7, 84.7, 73.3, 71.8, 71.6, 70.2, 69.6, 68.4, 52.9, 52.4,
29.5, 23.8; HRMS (ESI) m/z calcd for [C31H39NO6H+] 522.2850,
found 522.2847 (−0.43 ppm).
20
15 (284 mg, 65%) as a light yellow oil. 15: [α]D +2.04 (c 0.98,
CH2Cl2); IR (cm−1) 3505, 3330, 3062, 3030, 2934, 2865, 1495, 1453,
1417, 1362, 1306, 1207, 1188, 1075, 1013, 908, 845, 738, 698 cm−1;
1H NMR (300 MHz, CDCl3) δ 7.34−7.18 (m, 15H, PhH), 4.62−4.35
(m, 6H), 4.26−3.96 (m, 3H), 3.80−3.69 (m, 2H), 3.62−3.53 (m, 1H),
2.50−2.39 (m, 0.62H), 2.21−2.16 (m, 0.35H), 2.07−1.93 (m, 0.80H),
1.83−1.70 (m, 1.26H), 1.60−1.49 (m, 0.73H), 1.10−1.03 (m, 0.61H);
13C NMR (75 MHz, CDCl3) δ 138.2, 138.1, 138.0, 137.8, 137.4, 128.5,
(3S,4S,5S)-2-(Benzyloxymethyl)-3,4-bis(benzyloxy)-5-(3,3-
dimethoxypropyl)pyrroline N-Oxide (12) and (3S,4S,5S)-2-(3,3-
Dimethoxypropyl)-3,4-bis(benzyloxy)-5-(benzyloxymethyl)pyrroline
N-Oxide (13). Following the standard deprotection procedure,
reaction of activated manganese dioxide (2.78 g, 32 mmol) and
hydroxylamine 11 (8.3 g, 16 mmol) gave 12 as a brown oil (3.6 g,
43%) and 13 also as a brown oil (3.2 g, 39%). 12: [α]D20 +40.0 (c 0.65,
CH2Cl2); IR (cm−1) 3062, 3031, 2933, 2865, 1717, 1587, 1496, 1360,
128.3, 128.3, 128.1, 127.9, 127.8, 127.7, 127.64, 127.56, 127.5, 90.1,
88.3, 83.9, 82.6, 75.1, 73.9, 73.6, 73.1, 72.4, 71.8, 71.3, 71.1, 69.8, 69.2,
68.9, 67.4, 64.5, 26.7, 26.4, 25.3, 16.5; HRMS (ESI) m/z calcd for
[C29H33NO5H+] 476.2431, found 476.2445.
(1R,2R,5S,6S,7S)-1-(Benzyloxymethyl)-6,7-bis(benzyloxy)-8-
azabicyclo[3.2.1]octane-2-ol (16). Following the standard depro-
tection procedure, 16 was obtained as a yellow oil (34 mg, 89%):
1
1242, 1206, 1069, 1028, 739, 698; H NMR (300 MHz, CDCl3) δ
7.38−7.24 (m, 15H, PhH), 4.69−4.56 (m, 6H, H-6, H-3, 4 × CH2Ph),
4.24 (s, 2H, 2 × CH2Ph), 4.38−4.33 (m, 2H, H-6, H-9), 3.95 (d, J =
3.3 Hz, 1H, H-5), 3.83 (s, 1H, H-4), 3.30 (s, 3H, OCH3), 3.28 (s, 3H,
OCH3), 2.16−2.07 (m, 1H, H-7), 1.88−1.71 (m, 1H, H-7), 1.70−1.59
(m, 2H, H-8); 13C NMR (75 MHz, CDCl3) δ 142.4 (C-2), 137.5,
137.0, 128.5, 128.51, 128.45, 128.10, 128.06, 128.03, 127.99, 127.9,
127.7, 104.2 (C-9), 82.6 (C-3), 80.8 (C-4), 78.8 (C-5), 73.6, 72.5,
71.3, 62.7 (C-6), 53.2, 53.0 (OCH3), 28.5 (C-8), 26.3 (C-7); HRMS
(ESI) m/z calcd for [C31H37NO6Na+] 542.2513, found 542.2513. 13:
20
[α]D −16.0 (c 1.0, MeOH); IR (cm−1) 3363, 2935, 1128, 1074,
1
1048, 1011, 898, 773; H NMR (300 MHz, D2O) δ 4.05 (d, J = 2.4
Hz, 1H, H-7), 4.00 (s, 1H, H-6), 3.80 (s, 1H, H-2), 3.59−3.60 (m, 2H,
H-8), 3.12 (s, 1H, H-5), 1.90−1.82 (m, 2H, H-3, H-4), 1.77−1.58 (m,
2H, H-3, H-4); 13C NMR (75 MHz, D2O) δ 82.3 (C-6), 81.1 (C-7),
67.7 (C-1), 65.1 (C-2), 62.8 (C-8), 61.2 (C-5), 25.3, 24.3 (C-3, C-4);
HRMS (ESI) m/z calcd for [C8H15NO4H+] 190.1074, found
190.1071.
Synthesis of 24. (3aS,4S,5S,6S)-4,5-Bis(benzyloxy)-6-
((benzyloxy)methyl)tetrahydropyrrolo[1,2-b]isoxazol-2(3H)-one
(18). 1,2-Dibromoethane (0.45 g, 0.22 mL, 0.0024 mol) was added via
a syringe to a suspension of Zn powder (3.51 g, 0.054 mol) in
anhydrous THF (24 mL). After being heated to reflux for 10 min, the
mixture was cooled to room temperature; TMSCl (0.26 g, 0.31 mL,
0.0024 mol) was added, and the mixture was stirred for 15 min. Ethyl
bromoacetate 17 (4.01 g, 0.024 mol) and cyclic nitrone 1 (5.00 g,
0.012 mol) were added successively at such a rate that THF boiled
gently. After being stirred for 2 h, the reaction mixture was quenched
with 1 N hydrochloric acid (10 mL). The aqueous phase was extracted
with ethyl acetate (3 × 10 mL) and dried with MgSO4, and the solvent
was concentrated in vacuo. The lactone 18 thus obtained as a colorless
oil was used directly without purification to the next step: [α]D20 +80.4
(c 0.92, CH2Cl2); IR (cm−1) 3062 (w), 3030 (m), 2920 (m), 2865
(m), 1784 (s), 1494 (m), 1454 (m), 1415 (m), 1365 (m), 1208 (m),
20
[α]D +26.8 (c 0.82, CH2Cl2); IR (cm−1) 3062, 3031, 2933, 2866,
1715, 1599, 1496, 1453, 1362, 1250, 1206, 1122, 1071, 1028, 738, 698;
1H NMR (300 MHz, CDCl3) δ 7.25−7.16 (m, 15H, PhH), 4.59−4.35
(m, 7H, 6 × CH2Ph, H-3), 4.27 (t, J = 5.7 Hz, 1H, H-9), 4.15 (t, J =
2.1 Hz, 1H, H-4), 3.97 (s, 1H, H-2), 3.95 (dd, J = 5.4, 9.9 Hz, 1H, H-
6), 3.75 (dd, J = 2.7, 9.9 Hz, 1H, H-6), 3.20 (s, 6H, OCH3), 2.59−2.52
(m, 1H, H-7), 2.42−2.35 (m, 1H, H-7), 1.87−1.73 (m, 2H, H-8); 13C
NMR (75 MHz, CDCl3) δ 146.4 (C-2), 137.8, 137.3, 137.2, 128.6,
128.5, 128.4, 128.1, 128.04, 127.97, 127.9, 127.8, 104.1 (C-9), 84.9 (C-
3), 78.3 (C-4), 77.4 (C-5), 73.5, 71.7, 71.6, 66.9 (C-6), 53.2, 53.1
(OCH3), 27.4 (C-8), 20.6 (C-7); HRMS (ESI) m/z calcd for
[C31H37NO6H+] 520.2694, found 520.2696.
(3S,4S,5S)-2-(Benzyloxymethyl)-3,4-bis(benzyloxy)-5-(3-oxoprop-
yl)-1-pyrroline N-Oxide (14). To a well-stirred solution of nitrone 12
(0.68 g, 1.31 mmol) in acetone/H2O (50 mL, 4:1) was added catalytic
p-TsOH (25 mg, 0.13 mmol). The reaction mixture was stirred for 2 h
at reflux. The acetone was removed in vacuo. The aqueous layer was
extracted with ethyl acetate (3 × 10 mL). The combined the organic
layers were dried and filtered, and solvents were removed under
reduced pressure. The residue was purified by flash column
chromatography (silica gel, petroleum ether/AcOEt 1/2) to afford
the aldehyde 14 (437 mg, 79%) as a colorless oil: [α]D20 +42.0 (c 1.19,
CH2Cl2); IR (cm−1) 3062, 3031, 2925, 2864, 2727, 1723, 1588, 1496,
1389, 1358, 1247, 1208, 1093, 1071, 1028, 905, 818, 740, 699 cm−1;
1H NMR (300 MHz, CDCl3) δ 9.62 (s, 1H, H-9), 7.23−7.17 (m, 15H,
1
1186 (m), 1154 (m), 1099 (s), 1027 (m), 908 (m); H NMR (600
MHz, CDCl3) δ 7.38−7.26 (m, 15H, PhH), 4.61−4.47 (m, 6H,
CH2Ph), 4.19 (t, J = 3.9 Hz, 1H, H-4), 4.15 (td, J = 5.2, 9.8 Hz, 1H, H-
2), 4.09 (t, J = 3.9 Hz, 1H, H-3), 3.81 (q, J = 5.0 Hz, 1H, H-5), 3.67
(dd, J = 5.2, 9.9 Hz, 1H, H-6), 3.60 (dd, J = 6.3, 10.0 Hz, 1H, H-6),
2.94 (dd, J = 9.0, 17.8 Hz, 1H, H-7), 2.80 (dd, J = 6.0, 17.8 Hz, 1H, H-
7); 13C NMR (75 MHz, CDCl3) δ 174.7 (C-8), 137.8, 137.4, 137.1,
128.7, 128.53, 128.45, 128.2, 128.0, 127.90, 127.89, 127.8, 87.7 (C-3),
84.0 (C-4), 73.41, 72.42 (C-5), 72.38, 72.3, 68.6 (C-6), 68.0 (C-2),
34.4 (C-7); HRMS (ESI) m/z calcd for [C28H29NO5Na+] 482.1938,
found 482.1936 (−0.34 ppm).
PhH), 4.62−4.42 (m, 6H, H-3, H-6, 4 × CH2Ph), 4.36−4.25 (m, 3H,
L
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(2S,3S,4S,5S)-1-Hydroxy-2-(benzyloxymethyl)-3,4-bis(benzyloxy)-
5-(2-hydroxyethyl)pyrrolidine (19). LiAlH4 (0.91 g, 0.024 mol) was
added slowly to a solution of the lactone 18 in anhydrous THF (50
mL) at 0 °C (ice−water bath). After the addition was completed, the
reaction mixture was allowed to warm to room temperature and stirred
for an additional 1 h, and ethyl acetate was added to destroy excess
LiAlH4, followed by acidification with 1 N HCl. The aqueous phase
was extracted with ethyl acetate (3 × 15 mL), and the combined
organic layer was dried and concentrated. The crude product was used
directly for oxidation. 19: colorless oil; [α]D20 +2.9 (c 0.7, CH2Cl2); IR
(cm−1) 3369 (m), 3062 (w), 3030 (w), 2923 (m), 2869 (m), 1495
(m), 1453 (s), 1363 (s), 1310 (w), 1263 (m), 1209 (m), 1071 (s),
and NaHCO3 (5 mL), the aqueous layer was extracted with ethyl
acetate (3 × 10 mL), and the combined organic layers were washed
with saturated aqueous solution of NaCl and dried over MgSO4.
Filtration, concentration in vacuo, and purification by chromatography
on silica gel afforded the N-hydroxyamino alcohol 23 (139 mg, 54%
20
for two steps) as a colorless oil: [α]D +21.2 (c 1.32, CH2Cl2); IR
(cm−1) 3541 (m), 3354 (m), 3062 (m), 3031 (m), 2920 (m), 2865
(s), 1496 (m), 1453 (m), 1419 (w), 1362 (m), 1309 (m), 1261 (m),
1
1208 (m), 1092 (s), 1027 (s); H NMR (600 MHz, CDCl3) δ 7.34−
7.23 (m, 15H, PhH), 4.62 (d, J = 12.0 Hz, 1H, CH2Ph), 4.52 (d, J =
11.8 Hz, 1H, CH2Ph), 4.51 (d J = 11.8 Hz, 1H, CH2Ph), 4.47 (dd, J =
11.8 Hz, 1H, CH2Ph), 4.45 (dd, J = 11.8 Hz, 1H, CH2Ph), 4.39 (d,
1H, J = 11.9 Hz, CH2Ph), 4.33−4.30 (m, 1H, H-6), 4.02 (d, J = 11.1
Hz, 1H, H-7), 4.00 (s, 1H, H-2), 3.88 (d, J = 11.1 Hz, 1H, H-7), 3.85
(d, J = 5.6 Hz, 1H, H-4), 3.18 (s, 1H, H-3), 2.67 (d, J = 10.6 Hz, 1H,
OH), 2.16−2.12 (m, 1H, H-5b), 2.00−1.97 (m, 1H, H-5a); 13C NMR
(75 MHz, CDCl3) δ138.2, 138.00, 137.95, 128.4, 128.33, 128.31,
127.8, 127.7, 127.6, 83.1 (C-3), 79.9 (C-2), 75.6 (C-1), 73.5, 72.2,
70.9, 72.1 (C-6), 69.0 (C-4), 64.7 (C-7), 37.7 (C-5); HRMS (ESI)
HRMS (ESI) m/z calcd for [C28H31NO5H+] 462.2275, found
462.2275 (0.09 ppm).
1
1027(s); H NMR (300 MHz, CDCl3) δ 7.32−7.24 (m, 15H, PhH),
4.59−4.43 (m, 6H, CH2Ph), 4.01−3.99 (m, 1H, H-3), 3.95 (dd, J =
3.0, 6.3 Hz, 1H, H-4), 3.78−3.69 (m, 3H, 2 × H-8, H-6), 3.61−3.54
(m, 2H, H-6, H-2), 3.33 (t, J = 6.6 Hz, 1H, H-5), 2.07−1.98 (m, 1H,
H-7), 1.86−1.76 (m, 1H, H-7); 13C NMR (75 MHz, CDCl3) δ138.1,
137.9, 137.7, 128.5, 128.44, 128.40, 128.3, 128.1, 128.0, 127.9, 127.8,
127.7, 87.0 (C-4), 84.5 (C-3), 73.3, 72.1, 71.7, 70.3 (C-2), 69.3 (C-5),
67.9 (C-6), 61.1 (C-8), 32.3 (C-7); HRMS (ESI) m/z calcd for
[C28H33NO5H+] 464.2431, found 464.2432 (−0.17 ppm).
(1R,2S,3S,4S,5R)-1-(Hydroxymethyl)-7-azabicyclo[2.2.1]heptane-
2,3,6-triol (24). Copper(II) acetate (1.6 mg, 0.008 mmol) was added
to a suspension of Fe powder (4.5 mg, 0.8 mmol) in acetic acid (1
mL), and the mixture was stirred at room temperature for 2 h after
which the color turned to brown. A solution of 23 (40 mg, 0.087
mmol) in AcOH (1 mL) was added, and after the reaction mixture was
stirred at room temperature for 3 h, acetic acid was removed under
reduced pressure. The residue was dissolved in ethyl acetate, and the
pH adjusted to 8 by adding aqueous solution of NaHCO3 and filtered
with Celite. The filtrate was extracted with ethyl acetate (3 × 2 mL),
and the combined organic layers were dried (Na2SO4) and
concentrated at reduced pressure to afford the amine as a brown oil.
10% Pd/C (10 mg) and concd HCl (2 mL) were added to a solution
of thus obtained amine in MeOH (10 mL). After the resulting
suspension had been stirred under an atmosphere of H2 at room
temperature for 24 h, the reaction was judged to be complete by TLC.
The flask was bubbled with Ar, and the Pd/C was filtered off. After the
solution was concentrated in vacuo, the residue was dissolved in
MeOH and neutralized with aqueous ammonium solution and
concentrated in vacuo. The above procedure was repeated three
times to ensure complete neutralization. The residue was purified with
an acid resin column (DOWEX 50W × 8, 100−200 mesh), eluting
with distilled water (50 mL) and then 1 N NH4OH (50 mL), affording
(3S,4S,5S)-2-(2-Hydroxyethyl)-3,4-bis(benzyloxy)-5-(benzyloxy-
methyl)-1-pyrroline N-Oxide (20) and (3S,4S,5S)-2-(Benzyloxymeth-
yl)-3,4-bis(benzyloxy)-5-(2-hydroxyethyl)-1-pyrroline N-Oxide (21).
Following the standard deprotection procedure, reaction of activated
manganese dioxide (2.08 g, 0.024 mol) and the crude hydroxylamine
19 gave 20 as a yellow oil (1 g, 18% for 3 steps) and 21 also as a
yellow oil (3 g, 54% for 3 steps). 20: [α]D20 +9.5 (c 1.05, CH2Cl2); IR
(cm−1) 3335 (m), 3052 (m), 3032 (m), 2918 (m), 2867 (m), 1601
(m), 1494 (m), 1454 (m), 1362 (m), 1266 (s), 1210 (m), 1075 (s),
1
1092 (s), 1027 (m); H NMR (300 MHz, CDCl3) δ 7.31−7.22 (m,
15H, PhH), 4.67−4.63 (m, 1H, H-4), 4.60−4.29 (m, 6H, 6 × CH2Ph),
4.26−4.24 (m, 1H, H-5), 4.08 (m, 1H, H-4), 4.04 (q, J = 5.1 Hz, 1H,
H-6), 3.89−3.59 (m, 3H, H-6, 2H-8), 2.81−2.65 (m, 1H, H-7); 13C
NMR (75 MHz, CDCl3) δ146.4 (C-2), 137.7, 137.1, 137.0, 128.6,
128.5,128.3, 128.04, 128.02, 127.84, 127.80, 85.4 (C-3), 78.2 (C-4),
77.6 (C-5), 73.5, 71.9, 71.8, 66.7 (C-6), 59.2 (C-8), 29.8 (C-7);
HRMS (ESI) m/z calcd for [C282H0 31NO5Na+] 484.2094, found
484.2090 (−0.96 ppm). 21: [α]D +89 (c 1.55, CH2Cl2); IR
(cm−1) 3363 (m), 3062 (m), 3030 (m), 2921 (m), 2865 (m), 1598
(m), 1495 (m), 1453 (m), 1394 (m), 1357 (m), 1232 (m), 1208 (m),
1
1070 (s), 1027 (m); H NMR (300 MHz, CDCl3) δ 7.35−7.24 (m,
15H, PhH), 4.69−4.56 (m, 6H, H-3, H-6, 4 × CH2Ph), 4.44−4.34 (m,
3H, H-6, 2 × CH2Ph), 4.07−4.05 (m, 1H, H-5), 3.86 (s, 1H, H-4),
3.80−3.67 (m, 2H, H-8), 2.18−2.06 (m, 1H, H-7), 2.01−1.95 (m, 1H,
H-7); 13C NMR (75 MHz, CDCl3) δ144.9 (C-2), 137.3, 137.2, 136.9,
128.7, 128.6, 128.5, 128.2, 128.11, 128.08, 128.05, 128.0, 82.5 (C-3),
81.9 (C-4), 78.7 (C-5), 73.7, 72.8, 71.6, 62.9 (C-6), 59.9 (C-8), 35.7
(C-7); HRMS (ESI) m/z calcd for [C28H31NO5Na+] 484.2094, found
484.2091 (−0.82 ppm).
20
24 (14 mg, 92% for two steps) as a yellow oil: [α]D −3.1 (c 0.65,
1
MeOH); IR (cm−1) 3343, 2939, 1086, 1037; H NMR (300 MHz,
D2O) δ 4.30 (dd, J = 2.7, 7.2 Hz, 1H, H-6), 3.87 (m, 3H, H-3, 2 × H-
7), 3.59 (d, J = 1.5 Hz, 1H, H-2), 3.44 (d, J = 4.2 Hz, 1H, H-4), 2.12−
2.04 (m, 1H, H-5), 1.63−1.55 (m, 1H, H-5); 13C NMR (75 MHz,
D2O) δ 80.0 (C-2), 78.6 (C-3), 72.9 (C-1), 67.0 (C-6), 62.0 (C-4),
58.2 (C-7), 37.5 (C-5); HRMS (ESI) m/z calcd for [C7H13NO4H+]
176.0917, found 176.0918.
(3S,4S,5S)-2-(Benzyloxymethyl)-3,4-bis(benzyloxy)-5-(2-oxoeth-
yl)-1-pyrroline N-Oxide (22). To a solution of oxalyl dichloride (0.17
mL, 1.43 mmol) in CH2Cl2 (10 mL) was added dropwise DMSO
(0.14 mL, 2.6 mmol) at −70 °C, and the resulting solution was stirred
for 20 min. A solution of alcohol 21 (0.6 g, 1.3 mmol) in CH2Cl2 (10
mL) was added dropwise at −70 °C, and the resulting solution was
stirred for 1 h. Then triethylamine (0.9 mL, 6.5 mmol) was added, and
the resulting solution was stirred at −70 °C for 15 min and at room
temperature for another 1 h. The resulting mixture was diluted with
CH2Cl2 (50 mL) and washed with H2O (3 × 20 mL). The organic
phase was dried (MgSO4) and concentrated under reduced pressure to
afford aldehyde 15, which was used in the next step without isolation.
(1R,2S,3S,4S,6R)-1-(Benzyloxymethyl)-2,3-bis(benzyloxy)-7-
azabicyclo[2.2.1]heptane-6,7-diol (23). To a stirred and carefully
deoxygenated solution of nitrone 22 (directly obtained from the last
step, 0.56 mmol) in THF (26 mL) was added deoxygenated water
(0.81 mL, 0.045 mol) under an Ar atmosphere. The mixture was
cooled to −78 °C to which was added freshly prepared SmI2 in THF
(14 mL, 1.12 mmol). The reaction was completed within 10 min and
quenched by successively added aqueous solutions of NaS2SO3 (5 mL)
Preparation of Protein Structures. To compare with the
corresponding experimental values for rat isomaltase or sucrase, we
wanted to use the rat proteins in docking simulations. However, their
structures did not exist in protein data banks. Thus, three-dimensional
structures of the rat isomaltase and sucrase were constructed by the
homology modeling method, based on human isomaltase (PDB ID:
3LPP;5 identity: 73%) and human maltase (PDB ID: 2QMJ;8 identity:
38%), respectively. Next, to relax these predicted 3D-structures, we
performed the annealing simulations with a molecular dynamic (MD)
method. After that, to consider the structural flexibility around the
keyhole of isomaltase and sucrase in our docking calculations, keyhole
samplings were performed with MD calculations at 300 K from above
annealing structures as initial coordinates. In the 200000 MD steps
(Δt = 6 [fs]), the 250 keyhole structures were saved. The keyholes
were then selected from the 250 sampled structures by clustering.
These homology modeling and molecular dynamics samplings were
performed with Prime package and Desmond package in Schrodinger
̈
Suite 2008,51 respectively.
M
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Molecular Docking Calculations and Binding Energy
Estimation. We have performed the docking simulations of miglitol,
α-1-C-butyl-DNJ, and α-1-C-butyl-LAB for each keyhole of isomaltase
(3) Ceriello, A. Diabetes 2005, 54, 1−7.
(4) Van Beers, E. H.; Buller, H. A.; Grand, R. J.; Einerhand, A. W.;
Dekker, J. Crit. Rev. Biochem. Mol. Biol. 1995, 30, 197−262.
(5) Sim, L.; Willemsma, C.; Mohan, S.; Naim, H. Y.; Pinto, B. M.;
Rose, D. R. J. Biol. Chem. 2010, 285, 17763−17770.
(6) Nichols, B. L.; Eldering, J.; Avery, S.; Hahn, D.; Quaroni, A.;
Sterchi, E. J. Biol. Chem. 1998, 273, 3076−3081.
with Glide in Schrodinger Suite 2008.52 To consider the ionization
̈
states around pH 6.8, corresponding to the experimental conditions, in
our docking calculations, we used the Ionizer and Epik.53 Then, many
conformations of each ligand state were prepared with confgen
Advanced. The poses generated from docking calculations were ranked
by Glide Score, and the top 10 ranked poses were selected for each
ligand. These poses were optimized with OPLS2005 force field54 (ε =
4r), and the interaction energies were calculated. A pose with the
lowest value of the interaction energy was selected as the binding
conformation for each ligand. After these binding structures had been
optimized with GB/SA continuum solvation model, we calculated the
(7) Nichols, B. L.; Avery, S.; Sen, P.; Swallow, D. M.; Hahn, D.;
Sterchi, E. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 1432−1437.
(8) Sim, L.; Quezada-Calvillo, R.; Sterchi, E. E.; Nichols, B. L.; Rose,
D. R. J. Mol. Biol. 2008, 375, 782−792.
(9) Rossi, E. J.; Sim, L.; Kuntz, D. A.; Hahn, D.; Johnston, B. D.;
Ghavami, A.; Szczepina, M. G.; Kumar, N. S.; Sterchi, E. E.; Nichols, B.
L.; Pinto, B. M.; Rose, D. R. FEBS J. 2006, 273, 2673−2683.
(10) Quezada-Calvillo, R.; Sim, L.; Ao, Z.; Hamaker, B. R.; Quaroni,
A.; Brayer, G. D.; Sterchi, E. E.; Robayo-Torres, C. C.; Rose, D. R.;
Nichols, B. L. J. Nutr. 2008, 138, 685−692.
binding energies (ΔEs) by the below equation, where Ecomp, Eprotein
,
and Eligand are a total energy of protein−ligand complex, protein, and
ligand, respectively.
(11) Ren, L.; Cao, X.; Geng, P.; Bai, F.; Bai, G. Carbohydr. Res. 2011,
346, 2688−2692.
ΔE = Ecomp − (Eprotein + Eligand
)
(12) Ren, L.; Qin, X.; Cao, X.; Wang, L.; Bai, F.; Bai, G.; Shen, Y.
Protein Cell 2011, 2, 827−836.
Enzyme Assay. Brush border membranes were prepared from the
rat small intestine according to the method of Kessler et al.55 and were
assayed at pH 6.8 for rat intestinal isomaltase using the appropriate
disaccharide as substrate. The reaction mixture contained 25 mM
maltose and the appropriate amount of enzyme, and the incubations
were performed for 10−30 min at 37 °C. The reaction was stopped by
heating at 100 °C for 3 min. After centrifugation (600 g; 10 min), 0.05
mL of the resulting reaction mixture was added to 3 mL of the Glucose
CII-test Wako. The absorbance at 505 nm was measured to determine
the amount of the released D-glucose. Kinetic parameters were
determined by the double-reciprocal-plot method of Lineweaver−Burk
at increasing concentrations of the appropriate maltose.
(13) Winchester, B.; Fleet, G. W. J. Glycobiology 1992, 2, 199−210.
(14) Watson, A. A.; Fleet, G. W. J.; Asano, N.; Molyneux, R. J.; Nash,
R. J. Phytochemistry 2001, 56, 265−295.
(15) Asano, N. Glycobiology 2003, 13, 93R−104R.
(16) Nash, R. J.; Kato, A.; Yu, C-. Y.; Fleet, G. W. J. Future Med.
Chem. 2011, 3, 1513−1521.
(17) Joubert, P. H.; Bam, W. J.; Manyane, N. Eur. J. Clin. Pharmacol.
1986, 30, 253−255.
(18) Schnack, C.; Roggla, G.; Luger, A.; Schernthaner, G. Eur. J. Clin.
Pharmacol. 1986, 30, 417−419.
(19) Kennedy, F. P.; Gerich, J. E. Clin. Pharmacol. Ther. 1987, 42,
455−458.
ASSOCIATED CONTENT
■
(20) Wisselaar, H. A.; van Dongen, J. M.; Reuser, A. J. Clin. Chim.
Acta 1989, 182, 41−52.
S
* Supporting Information
1H and 13C NMR spectra of obtained compounds. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(21) Misago, M.; Tsukada, J.; Fukuda, M. N.; Eto, S. Biochem.
Biophys. Res. Commun. 2000, 269, 219−225.
(22) Belmatoug, N.; Burlina, A.; Giraldo, P.; Hendriksz, C. J.; Kuter,
D. J.; Mengel, E.; Pastores, G. M. J. Inherit. Metab. Dis. 2011, 34, 991−
1001.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: kato@med.u-toyama.ac.jp.
*E-mail: yucy@iccas.ac.cn.
■
(23) Best, D.; Jenkinson, S. F.; Saville, A. W.; Alonzi, D. S.; Wormald,
M. R.; Butters, T. D.; Norez, C.; Becq, F.; Bleriot, Y.; Adachi, I.; Kato,
́
A.; Fleet, G. W. J. Tetrahedron Lett. 2010, 51, 4170−4174.
(24) Best, D.; Wang, C.; Weymouth-Wilson, A. C.; Clarkson, R. A.;
Wilson, F. X.; Nash, R. J.; Miyauchi, S.; Kato, A.; Fleet, G. W. J.
Tetrahedron: Asymmetry 2010, 21, 311−319.
(25) da Cruz, F. P.; Newberry, S.; Jenkinson, S. F.; Wormald, M. R.;
Butters, T. D.; Alonzi, D. S.; Nakagawa, S.; Becq, F.; Nash, R. J.; Kato,
A.; Fleet, G. W. J. Tetrahedron Lett. 2011, 52, 219−223.
(26) Ayers, B. J.; Ngo, N.; Jenkinson, S. F.; Martínez, R. F.; Shimada,
Y.; Adachi, I.; Weymouth-Wilson, A. C.; Kato, A.; Fleet, G. W. J. J. Org.
Chem. 2012, 77, 7777−7792.
(27) Kato, A.; Hayashi, E.; Miyauchi, S.; Adachi, I.; Imahori, T.;
Natori, Y.; Yoshimura, Y.; Nash, R. J.; Shimaoka, H.; Nakagome, I.;
Koseki, J.; Hirono, S.; Takahata, H. J. Med. Chem. 2012, 55, 10347−
10362.
*E-mail: hironos@pharm.kitasato-u.ac.jp.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported in part by a Grant-in-Aid for Scientific
Research (C) (No: 26460143) (A.K.) from the Japanese
Society for the Promotion of Science (JSPS) and the
Leverhulme Trust (G.W.J.F.). Financial support from the
National Basic Research Program of China (Nos.
2011CB808603, 2012CB822101), the National Natural Science
Foundation of China (No. 21272240), the National Science
and Technology Major Projects for “Major New Drugs
Innovation and Development” (2013ZX09508104), the Devel-
opment Fund for Collaborative Innovation Center of
Glycoscience of Shandong University, and National Engineer-
ing Research Center for Carbohydrate Synthesis of Jiangxi
Normal University is gratefully acknowledged.
(28) Wiltz, O.; O’Hara, C. J.; Steele, G. D.; Mercurio, A. M. Surgery
1990, 108, 269−276.
(29) Wiltz, O.; O’Hara, C. J.; Steele, G. D.; Mercurio, A. M.
Gastroenterology 1991, 100, 1266−1278.
(30) Delso, I.; Tejero, T.; Goti, A.; Merino, P. J. Org. Chem. 2011, 76,
4139−4143.
(31) Chavarot-Kerlidou, M.; Rivard, M.; Chamiot, B.; Hahn, F.;
Rose-Munch, F.; Rose, E.; Py, S.; Herson, P. Eur. J. Org. Chem. 2010,
75, 944−958.
REFERENCES
(32) Zhang, Z.-L.; Nakagawa, S.; Kato, A.; Jia, Y.-M.; Yu, C.-Y. Org.
Biomol. Chem. 2011, 9, 7713−7719.
(33) Ducrot, P. H.; Lallemand, J. Y. Tetrahedron Lett. 1990, 31,
3879−3882.
■
(1) Coutinho, M.; Gerstein, H. C.; Wang, Y.; Yusuf, S. Diabetes Care
1999, 22, 233−240.
(2) Bonora, E.; Muggeo, M. Diabetologia 2001, 44, 2107−2114.
N
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(34) Goldmann, A.; Milat, M. L.; Ducrot, P. H.; Lallemand, J. Y.;
Maille, M.; Lepingle, A.; Charpin, I.; Tepfer, D. Phytochemistry 1990,
29, 2125−2127.
(35) Molyneux, R. J.; Pan, Y. T.; Goldmann, A.; Tepfer, D. A.; Elbein,
A. D. Arch. Biochem. Biophys. 1993, 304, 81−88.
(36) Asano, N.; Kato, A.; Oseki, K.; Kizu, H.; Matsui, K. Eur. J.
Biochem. 1995, 229, 369−376.
(37) Asano, N.; Kato, A.; Miyauchi, M.; Kizu, H.; Tomimori, T.;
Matsui, K.; Nash, R. J.; Molyneux, R. J. Eur. J. Biochem. 1997, 248,
296−303.
(38) Kato, A.; Nakagome, I.; Nakagawa, S.; Koike, Y.; Nash, R. J.;
Adachi, I.; Hirono, S. Bioorg. Med. Chem. 2014, 22, 2435−2441.
(39) Cheng, H. H.; Asano, N.; Ishii, S.; Ichikawa, Y.; Fan, J.-Q. FEBS
J. 2006, 273, 4082−4092.
(40) Rasmussen, T. S.; Jensen, H. H. Org. Biol. Chem. 2010, 8, 433−
441.
(41) Rasmussen, T. S.; Allman, S.; Twigg, G.; Butters, T. D.; Jensen,
H. H. Bioorg. Med. Chem. Lett. 2011, 21, 1519−1522.
(42) Wang, W.-B.; Huang, M.-H; Li, Y.-X.; Rui, P.-X.; Hu, X.-G.;
Zhang, W.; Su, J.-K.; Zhang, Z.-L.; Zhu, J.-S.; Xu, W.-H.; Xie, X.-Q.; Jia,
Y.-M.; Yu, C.-Y. Synlett 2010, 3, 488−492.
(43) Yu, C.-Y.; Huang, M.-H. Org. Lett. 2006, 8, 3021−3024.
(44) Merino, P.; Delso, I.; Tejero, T.; Cardona, F.; Marradi, M.;
Faggi, E.; Parmeggiani, C.; Goti, A. Eur. J. Org. Chem. 2008, 73, 2929−
2947.
(45) Hu, X.-G.; Bartholomew, B.; Nash, R. J.; Wilson, F. X.; Fleet, G.
W. J.; Nakagawa, S.; Kato, A.; Jia, Y.-M.; Well, R. V.; Yu, C.-Y. Org.
Lett. 2010, 12, 2562−2565.
(46) Hu, X.-G.; Jia, Y.-M.; Xiang, J.; Yu, C.-Y. Synlett 2010, 982−986.
(47) Cicchi, S.; Goti, A.; Brandi, A. J. Org. Chem. 1995, 60, 4743−
4748.
(48) Cicchi, S.; Marradi, M.; Goti, A.; Brandi, A. Tetrahedron Lett.
2001, 42, 6503−6505.
(49) Masson, G.; Py, S.; Vallee
1772−1775.
(50) Masson, G.; Philouze, C.; Py, S. Org. Biomol. Chem. 2005, 3,
2067−2069.
́
, Y. Angew. Chem., Int. Ed. 2002, 41,
(51) Bowers, K. J.; Chow, E.; Xu, H.; Dror, R. O.; Eastwood, M. P.;
Gregersen, B. A.; Klepeis, J. L.; Kolossvary, I.; Moraes, M. A.;
Sacerdoti, F. D.; Salmon, J. K.; Shan, Y.; Shaw, D. E. Proc. ACM/IEEE
Conf. Supercomputing (SC06), Tampa, FL, Nov 11−17, 2006.
(52) Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.;
Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.;
Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. J. Med. Chem. 2004,
47, 1739−1749.
(53) Shelley, J. C.; Cholleti, A.; Frye, L. L.; Greenwood, J. R.; Timlin,
M. R.; Uchimaya, M. J. Comput. Aided Mol. Des. 2007, 21, 681−91.
(54) Jorgensen, W. L.; Maxwell, D. S.; Tirado-R, J. J. Am. Chem. Soc.
1996, 118, 11225−11236.
(55) Kessler, M.; Acuto, O.; Strelli, C.; Murer, H.; Semenza, G. A. A.
Biochem. Biophys. Acta 1978, 506, 136−154.
O
DOI: 10.1021/acs.joc.5b00342
J. Org. Chem. XXXX, XXX, XXX−XXX