Synthesis and biological evaluation of novel C-indolylxylosides as sodium-dependent glucose co-transporter 2 inhibitors

Article abstract of DOI:10.1016/j.ejmech.2012.06.053

Sodium-dependent glucose co-transporter 2 (SGLT2) inhibitors are the current focus on the indication for the management of hyperglycemia in diabetes. Here, a novel series of C-linked indolylxyloside-based inhibitors of SGLT2 has been discovered. Structure-activity relationship studies revealed that substituents at the 7-position of the indole moiety and a p-cyclopropylphenyl group in the distal position were necessary for optimum inhibitory activity. The pharmacokinetic study demonstrates that the most potent compound 1i is metabolically stable with a low clearance in rats. In further efficacy study, 1i is found to significantly lower blood glucose levels of streptozotocin (STZ)-induced diabetic rats.

Full text of DOI:10.1016/j.ejmech.2012.06.053

European Journal of Medicinal Chemistry 55 (2012) 32e38
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Synthesis and biological evaluation of novel C-indolylxylosides
as sodium-dependent glucose co-transporter 2 inhibitors
Chun-Hsu Yao, Jen-Shin Song, Chiung-Tong Chen, Teng-Kuang Yeh, Tsung-Chih Hsieh,
Szu-Huei Wu, Chung-Yu Huang, Yu-Lin Huang, Min-Hsien Wang, Yu-Wei Liu,
*
Chi-Hui Tsai, Chidambaram Ramesh Kumar, Jinq-Chyi Lee
Institute of Biotechnology and Pharmaceutical Research, National Health Research Institutes, 35 Keyan Road, Zhunan Town, Miaoli County 35053, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history:
Sodium-dependent glucose co-transporter 2 (SGLT2) inhibitors are the current focus on the indication for
the management of hyperglycemia in diabetes. Here, a novel series of C-linked indolylxyloside-based
inhibitors of SGLT2 has been discovered. Structureeactivity relationship studies revealed that substitu-
ents at the 7-position of the indole moiety and a p-cyclopropylphenyl group in the distal position were
necessary for optimum inhibitory activity. The pharmacokinetic study demonstrates that the most potent
compound 1i is metabolically stable with a low clearance in rats. In further efficacy study, 1i is found to
significantly lower blood glucose levels of streptozotocin (STZ)-induced diabetic rats.
Received 13 March 2012
Received in revised form
25 June 2012
Accepted 27 June 2012
Available online 4 July 2012
Keywords:
C-indolylxylosides
Ó 2012 Elsevier Masson SAS. All rights reserved.
Sodium-dependent glucose co-transporter
Structureeactivity relationship
Type 2 diabetes mellitus
1. Introduction
not degraded by glucosidase enzymes present in the gastrointes-
tinal tract, and obviates any need to derivatize them as prodrugs,
The therapeutically beneficial effects of sodium-dependent
glucose co-transporter 2 (SGLT2) inhibitors for the treatment of
type 2 diabetes mellitus (T2DM) patients has attracted the atten-
tion of many researchers in recent years. SGLT2, located on the S1
segment of the proximal tubule in the kidney, mediates the reup-
take of the majority (w90%) of glucose filtered by the kidney
glomeruli [1,2]. Uptake of the remainder is accomplished by SGLT1,
which is mainly expressed in the small intestine but also presents
in the S2/S3 segment of the proximal tubule. Accordingly, inhibition
of SGLT is considered a feasible approach for the management of
hyperglycemia by increasing the urinary glucose excretion and
consequently lowering the blood glucose concentration. To avoid
the gastrointestinal side effects associated with suppression of
SGLT1, inhibitors selective for SGLT2 are most highly desired [3e5].
A number of SGLT2 inhibitors have been developed and
submitted to clinical trials [6]. The reported glycoside SGLT2
inhibitors at advanced stages of development are C-glucosides,
which are metabolically stable and combine higher oral bioavail-
ability and plasma exposure [7,8]. The inherent structural stability
of the C-linked glycosidic bond helps ensure these molecules are
which usually require a higher dosage if they are to reach the
therapy target.
In the course of our recent investigations into the replacement
of glucose with
D-xylose coupled with 3-substituted indoles, the
resulting N-indolylxylosides were found to behave as potent SGLT2
inhibitors (Fig. 1) [9]. A series of C-indolylglucosides has also been
disclosed as SGLT2 inhibitors with good inhibitory activities [10].
These results suggested to us that the use of C-linked indolylxylo-
sides 1 to inhibit SGLT2 would be feasible. Herein, we describe the
synthesis of a novel series of metabolically stable C-indolylxylo-
sides and disclose their inhibitory activities against SGLT2 and
SGLT1.
2. Results and discussion
2.1. Chemistry
An array of C-indolylxylosides 1ae1k was prepared using a five-
step synthetic sequence, starting with the coupling of fully pro-
tected xylonolactone 2 [11] with a variety of 3-bromo-1-tosyl-1H-
indoles 3, Scheme 1. The novel C-xylosides 1ae1k used in this work
were synthesized according to the well-established chemistry of
Kissei, with appropriate modifications [10]. Commercially available
* Corresponding author. Tel.: þ886 37 246166x35781; fax: þ886 37 586456.
E-mail address: jinqchyi@nhri.org.tw (J.-C. Lee).
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2012.06.053

C.-H. Yao et al. / European Journal of Medicinal Chemistry 55 (2012) 32e38
33
R
R1
THF to provide the desired
b-linked C-indolylxylosides 1ae1g.
Attempts to synthesize compound 1h using this strategy were
unsuccessful; the n-propyl product 1k was obtained after hydro-
genolysis instead. Accordingly, an alternative strategy was pursued
for compounds bearing a p-cyclopropyl group on the distal phenyl
ring. First, the benzyl ether groups of tri-benzyl-xylopyranosyl
indole were removed under hydrogenolysis to furnish
compounds 10he10j. Then, N-alkylation of 10he10j with p-cyclo-
propylbenzyl bromide 11 in the presence of cesium carbonate
(Cs2CO3) provided the desired products 1he1j.
Ar
R2
Y
N
Z
H
HO
O
X
O
OH
HO
OH
OH
OH
1
N-indolylxylosides: X= N, Y = C, Z = H
C-indolylglucosides: X= C, Y = N, Z = CH2OH
Fig. 1. Design of C-indolylxyloside SGLT2 inhibitors.
2.2. Biological evaluations
indoles 4 were first subjected to N-tosylation with p-toluene-
sulfonyl chloride (pTsCl) and sodium hydride (NaH) to afford N-
tosyl indoles 5, which were then treated with bromine in
dichloromethane at 0 ꢀC for 1 h to provide the coupling precursors
3-bromo-1-tosyl-1H-indoles 3. Lithium-halogen exchange fol-
Biological evaluation of these synthesized C-indolylxylosides
could now take place; the results are presented in Table 1. EC50
values were calculated by measuring inhibition of the sodium-
dependent uptake of [14C]-labeled
a-methyl-D-glucopyranoside
lowed by the addition of 2,3,4-tri-O-benzyl- -xylonolactone 2
D
(AMG) into Chinese hamster ovary (CHO) cells stably expressing
human SGLT2 (hSGLT2) or hSGLT1 [12,13]. Because differences in
protein expression levels of the cells will result in differences of
EC50 values, the SGLT2 inhibitor 12, dapagliflozin [14], was used as
reference compound in this in vitro activity evaluation system.
Our previous studies suggested that N-indolylxylosides bearing
a para-substituent at the distal phenyl ring would impart increased
inhibitory activity against hSGLT2 [9]. In this study, we began the
structureeactivity relationship (SAR) studies by investigating the
effect of the para-substituent at the distal phenyl ring by fixing the
yielded a mixture of lactols 6, which were reduced with trie-
thylsilane (Et3SiH) and boron trifluoride etherate (BF3$OEt2) to give
the C-linked
b-xylosides 7. Efficient indole N-detosylation using
potassium hydroxide (KOH) in a mixture of THF and EtOH at 60 ꢀC
gave the free indoles 8.
Two routes to the target C-indolylxylosides 1ae1k were devel-
oped. Alkylation of 8 with benzyl halides or naphthyl bromide gave
the corresponding diarylmethanes 9ae9g, which in turn under-
went hydrogenolysis of the benzyl ethers over 10% Pd/C in MeOH/
a
1
1
R
R
N
H
N
Ts
4
5
1
1
R
R
b
HO
O
Br
N
N
Ts
Ts
O
O
O
c
d
e
1
R
+
BnO
OBn
N
BnO
OBn
BnO
OBn
Ts
OBn
2
OBn
6
OBn
7
3
n
Pr
1
1
1
R
R
R
2
2
R
R
NH
N
N
N
O
O
O
O
f
g
BnO
OBn
BnO
OBn
HO
OH
HO
OH
OBn
8
OBn
9a-9g
OH
1a-1g
OH
1k
g
1
1
R
R
N
NH
O
O
h
HO
OH
HO
OH
OH
1h-1j
OH
10h-10j
Scheme 1. Reagents and conditions: (a) NaH, pTsCl, DMF, rt, 4 h, 67e98%; (b) Br2, CH2Cl2, 0 ꢀC, 1 h, 55e98%; (c) nBuLi, THF/toluene (2/1), ꢁ78 ꢀC / rt, 2.5 h; (d) Et3SiH, BF3$OEt2,
CH3CN, 0 ꢀC / rt, 1 h, 15e54% over two steps; (e) KOH, THF/EtOH (1/2), 60 ꢀC, 20 h, 63e88%; (f) NaH, RPhCH2Cl or 2-(bromomethyl)naphthalene, DMF, rt, 16 h, 88e100%; (g) 10% Pd/
C, H2(g), MeOH/THF (1/1), rt, 3 h, 28e77% (for 1ae1g, 1k); (h) Cs2CO3, 4-cyclopropylbenzyl bromide 11, DMF, rt, 24 h, 10e51% over 2 steps (for 1he1j from 8).

34
C.-H. Yao et al. / European Journal of Medicinal Chemistry 55 (2012) 32e38
Table 1
indole moiety. The unsubstituted phenyl compound 1a
(EC50 ¼ 4153 nM) was used as template, to which all structural
modifications were compared. When the electron-withdrawing
fluoro group was introduced into the para-position, the resulting
compound 1b showed a drastic loss of activity (EC50 ¼ 25505 nM).
Replacing the fluoro group with an electron-donating methoxy
group gave xyloside 1c (EC50 ¼ 736 nM), which was a significantly
more potent inhibitor than 1a. Regarding weakly electron-donating
substituents, the n-propyl analogue 1k showed slightly improved
potency (EC50 ¼ 588 nM) compared to 1c, and xyloside 1d bearing
a bulky t-butyl group (EC50 ¼ 3632 nM) was found to impart similar
activity to the unsubstituted phenyl analogue 1a. It is noteworthy
that when the p-cyclopropyl substituent was introduced, the
potency of the resulting molecule 1h against hSGLT2
(EC50 ¼ 87 nM) improved 48-fold compared with that of the
unsubstituted compound 1a. Replacement of the phenyl group
with a benzofused ring, naphthalene gave compound 1e with an
EC50 value of 722 nM; an inferior inhibitor of hSGLT2 than 1h, and
approximately as potent as 1c.
Next, we turned our attention to the effect of the indole moiety.
Previous studies suggested that 4-substituted N-indolyl glycosides
and C-indolylglucosides bearing a 7-substituted or 5,7-disubstituted
indole moiety would possess good hSGLT2 inhibition [9,10], and
therefore the effect of the substituents incorporated onto the 4- or 7-
position of indole was explored. Substitution of 1c with the fluoro
group gave compounds 1f (4-fluoro) and 1g (7-fluoro); the former
wasfoundtoimpartanEC50 of1110nM, about7times lesspotentthan
the latter (EC50 ¼ 151 nM), in which the distal position was occupied
with p-methoxyphenyl ring. Incorporating a p-cyclopropyl phenyl
group gave 1i (EC50 ¼ 47 nM), the inhibitory activity of which was
significantly enhanced relative to the parent compounds 1h and 1g.
Similar activity was observed when the electron-donating methyl
group (1j, EC50 ¼ 50 nM) was attached in place of the fluoro group.
The selectivity of hSGLT2 over hSGLT1 is the other important
consideration. Unfortunately, none of the C-indolylxyloside SGLT2
inhibitors tested showed significant specificity toward hSGLT1
(hSGLT1/hSGLT2 ¼ 0.8e10.2), as shown in Table 1. Although sup-
pressing the reuptake of hSGLT1 may be implicated in gastroin-
testinal side effects, our previous in vivo animal study showed that
Effect of C-indolylxylosides 1ae1k on human SGLT inhibitory activity and selectivity.
R1
R2
N
O
HO
OH
OH
1a-1k
Compd R1
R2
EC50 [nM]a,b
hSGLT2
EC50 [nM]a,c
hSGLT1
Selectivity
hSGLT1/hSGLT2
1a
1b
1c
1d
1e
H
H
H
H
H
4153 ꢂ 499
9523 ꢂ 1229
2.3
0.8
3.4
6.3
6.9
25,505 ꢂ 4726 20,937 ꢂ 9982
736 ꢂ 1636 2491 ꢂ 772
3632 ꢂ 7386 23,063 ꢂ 2662
F
OMe
t
Bu
722 ꢂ 166
4962 ꢂ 354
1f
4-F
7-F
H
1110 ꢂ 205
151 ꢂ 25
87 ꢂ 19
6427 ꢂ 2200
550 ꢂ 143
884 ꢂ 234
282 ꢂ 11
5.8
3.6
OMe
OMe
1g
1h
1i
mice treated with 4-chloro-3-(4-cyclopropylbenzyl)-1-(b-D-xylo-
10.2
6.0
pyranosyl)-1H-indole 13 (hSGLT1/hSGLT2 ¼ 1.8) were not observed
to have diarrhea within 8 h of oral administration of a single dose at
a range of 10, 50, and 200 mg/kg [9]. Therefore, compounds with
good potency against hSGLT2 need to be subjected to a diarrheo-
genic activity test, to determine whether any adverse effects are
associated with the lack of selectivity, and the results used to guide
their future development.
In comparison with our previously reported N-indolylxylosides
[9], C-indolylxylosides were found to be superior inhibitors against
hSGLT2. Of N-indolylxylosides, the most potent compound 13,
exhibiting an EC50 value of 205 nM, showed a 4.4-fold loss in
potency compared to C-indolylxyloside 1i (EC50 ¼ 47 nM). SARs
studies indicated that the p-cyclopropylphenyl group in the distal
position was necessary for optimum inhibitory activity against
hSGLT2 in these two series of compounds. And the 7-substituents
instead of the 4-substituents, the essential element for potent N-
xylosides, of the indole moiety are required for C-linked glycosides.
7-F
47 ꢂ 3
7-
Me
1j
50 ꢂ 11
55 ꢂ 5
1.1
5.5
n
1k
H
588 ꢂ 35
2.4 ꢂ 0.6
205 ꢂ 20
3208 ꢂ 949
Pr
12d
593 ꢂ 176 247.1
13d
364 ꢂ 152
1.8
Nevertheless, no matter the combination of D-xyloside with 1- or 3-
substituted indoles, the resulting compounds showed no signifi-
cant selectivity for hSGLT2 versus hSGLT1.
a
These data were obtained by at least of two independent experiments, each
experiment performed in triplicates.
b
Inhibition of uptake of [14C]-AMG in CHO-K1 cells stably transfected with
human SGLT2.
2.3. Pharmacokinetic profile of 1i
c
Inhibition of uptake of [14C]-AMG in CHO-K1 cells stably transfected with
human SGLT1.
d
Dapagliflozin (12); 4-chloro-3-(4-cyclopropylbenzyl)-1-(b-D-xylopyranosyl)-
Among the aforementioned C-indolylxylosides, the most potent
SGLT2 inhibitor 1i was chosen for further pharmacokinetic study.
1H-indole (13).

C.-H. Yao et al. / European Journal of Medicinal Chemistry 55 (2012) 32e38
35
Administration of a single 1 mg/kg intravenous dose to rats
revealed that 1i has the low total body clearance, suggesting that 1i
is metabolically stable (Table 2). After oral administration of a 1 mg/
kg dose of 1i to rats, the Cmax and Tmax of 1i were 161.7 ng/mL and
0.8 h, respectively. The elimination half-life (t1/2) of 1i was 5.2 h.
The results indicated that 1i has favorable pharmacokinetic prop-
erty in rats.
2.4. Antihyperglycemic effect of 1i
As shown in Fig. 2, the antihyperglycemic effect of 1i was
assessed in streptozotocin (STZ) induced (STZ at 65 mg/kg, ip)
diabetic SpragueeDawley (SD) rats (blood glucose of >400 mg/dL).
Blood samples were obtained from the tail vein at 0 (predose), 0.5,
1, 2, 3, 4, and 5 h after a single oral administration of 1i (20 mg/kg),
12 (20 mg/kg), or vehicle for blood glucose analysis. C-indolylxy-
loside 1i at 20 mg/kg was found to cause a 37% reduction in blood
glucose level compared with the control.
Fig. 2. Antihyperglycemic effect of single oral dosing (20 mg/kg) of 1i and dapagli-
flozin 12 in STZ-induced diabetic rats. Data were expressed as the mean ꢂ SEM (n ¼ 6/
group). Area under the curve of testing agents were significant different from vehicle:
*P < 0.05 vs vehicle.
3. Conclusion
In conclusion, a novel class of C-indolylxylosides found to have
promising biological activity against hSGLT2 has been designed and
synthesized. SAR studies indicated that 7-substituted indolylxlyo-
sides bearing a distal p-cyclopropylphenyl group, such as 1i and 1j,
exhibited the best SGLT2 inhibition. Further pharmacokinetic and
animal studies demonstrated that 1i was metabolically stable along
with significant efficacy on lowering blood glucose levels of
streptozotocin (STZ)-induced diabetic rats, and therefore hold
promise as candidates for the treatment of T2DM e provided that
the gastrointestinal side effects associated with the suppression of
SGLT1 are not observed in in vivo studies.
4.1.1. General procedure for the synthesis of compound 5
NaH (1.1 equiv) was added to a stirred solution of indole (1.0
equiv) in DMF at 0 ꢀC. After being stirring at 0 ꢀC for 10 min, pTsCl
(1.1 equiv) was added, and the reaction was allowed to stir at room
temperature for 4 h. The reaction was quenched by the addition of
H2O at 0 ꢀC and extracted with CH2Cl2. The organic layer was dried
over MgSO4, filtered and concentrated under reduced pressure. The
residue was purified by column chromatography to afford the
desired product 5.
4.1.2. General procedure for the synthesis of compound 3
A solution of bromine (1.2 equiv) in CH2Cl2 was added to a stir-
red solution of compound 5 (1.0 equiv) in CH2Cl2 at 0 ꢀC under
nitrogen. After being stirred at the same temperature for 1 h, the
solvent was removed under reduced pressure. The residue was
purified by column chromatography to provide the desired
bromide 3.
4. Experimental section
4.1. Chemistry
All chemicals were purchased as reagent grade and used
without further purification unless otherwise stated. Column
chromatography was performed with silica gel (Merck Kieselgel 60,
230e400 mesh). All reactions were monitored with thin-layer
chromatography (TLC) using Merck 60 F254 silica gel glass-
backed plates. 1H NMR and 13C NMR spectra were recorded on
a Varian Mercury-300 or Mercury-400 spectrometers. Chemical
4.1.3. General procedure for the synthesis of compound 6
A 1.6 M solution of nBuLi in hexanes (1.4 equiv) was added to
a stirred solution of 3 (1.4 equiv) in THF/toluene (2/1) at ꢁ78 ꢀC
under nitrogen. After 0.5 h, a solution of fully protected xylono-
lactone 2 in THF/toluene (2/1) was added to the reaction; the
mixture was then warmed to room temperature gradually and
stirred for another 2 h. The reaction was poured into saturated
aqueous ammonium chloride and the mixture was extracted with
CH2Cl2. The organic layer was dried over MgSO4, filtered, and
concentrated under reduced pressure. The residue was purified by
column chromatography to give 6.
shifts are reported in parts per million (ppm, d) relative to the
internal standard signal of CD3OD. LC/MS data was obtained on an
Agilent MSD-1100 ESI-MS/MS system. Purity of the final
compounds was determined on a Hitachi 2000 series HPLC system
with a reverse phase C18 column (Agilent ZORBAX Eclipse XDB-C18
5
m
m, 4.6 mm ꢃ 150 mm, 0.5 mL/min flow rate). Mobile phase A:
acetonitrile; mobile phase B: 10 mM NH4OAc aqueous solution
containing 0.1% formic acid. The gradient system started from A:B
(10%:90%) at 0 min to A:B (90%:10%) at 45 min.
4.1.4. General procedure for the synthesis of compound 7
Et3SiH (5.0 equiv) and then BF3$OEt2 (0.5 equiv) were added to
a stirred solution of 6 in CH3CN at 0 ꢀC under argon and then the
reaction was warmed to room temperature gradually. After 1 h, the
reaction was quenched by the addition of saturated aqueous
sodium bicarbonate at 0 ꢀC and extracted with CH2Cl2. The organic
layer was dried over MgSO4, filtered and concentrated under
reduced pressure. The residue was purified by column chroma-
tography to give 7.
Table 2
Pharmacokinetic profile of C-xyloside 1i in rats.
Parameter
Unit
IV
PO
1.0
5.2
Dose
t1/2
mg/kg
h
1.0
5.9
Clearance
Vss
mL/min/kg
L/kg
16.8
5.0
Cmax
Tmax
AUC(0-inf.)
ng/mL
h
ng/mL*h
161.7
0.8
764
4.1.5. General procedure for the synthesis of compound 8
KOH (20 equiv) was added to a stirred solution of 7 (1.0 equiv) in
THF/EtOH (1/2) at room temperature. The reaction was warmed to
930

36
C.-H. Yao et al. / European Journal of Medicinal Chemistry 55 (2012) 32e38
60 ꢀC and further stirred for 20 h. The reaction was cooled to 0 ꢀC
and neutralized by the addition of 1 N HCl(aq). The mixture was
extracted with EtOAc and the organic layer was dried over MgSO4,
filtered and concentrated under reduced pressure. The residue was
purified by column chromatography to afford the desired product 8.
4.43 (d, J ¼ 9.3 Hz, 1H), 3.98 (dd, J ¼ 11.1, 5.4 Hz, 1H), 3.76 (t,
J ¼ 9.3 Hz, 1H), 3.68 (ddd, J ¼ 10.5, 9.3, 5.4 Hz, 1H), 3.48e3.35 (m,
2H); 13C NMR (100 MHz, CD3OD):
d 138.5, 137.0, 135.0, 134.4, 129.6,
129.2, 129.0, 128.81, 128.80, 127.4, 127.1, 126.8, 126.3, 123.0, 121.2,
120.5, 114.3, 111.3, 80.2, 79.0, 75.1, 71.8, 71.4, 51.1; ESMS m/z: 390
(MHþ), 412 (MNaþ); HPLC purity 100%.
4.1.6. General procedure for the synthesis of compounds 9ae9g
NaH (1.2 equiv) and aryl bromides or aryl chlorides (1.2 equiv) were
sequentially added to a stirred solution of 8 (1.0 equiv) in DMF at 0 ꢀC.
The reaction was slowly warmed to room temperature and stirred for
16 h. The reaction was quenched by the addition of H2O at 0 ꢀC and
extracted with CH2Cl2. The organic layer was dried over MgSO4, filtered
and concentrated under reduced pressure. The residue was purified by
column chromatography to afford the desired products 9ae9g.
4.1.7.6. 4-Fluoro-1-(4-methoxybenzyl)-3-(
b
-
D
-xylopyranosyl)-1H-
indole (1f). 1H NMR (300 MHz, CD3OD):
d
7.34 (s, 1H), 7.16e7.01 (m,
4H), 6.85e6.80 (m, 2H), 6.75e6.69 (m, 1H), 5.27 (s, 2H), 4.51 (dd,
J ¼ 9.6, 1.2 Hz, 1H), 3.93 (dd, J ¼ 10.8, 5.1 Hz, 1H), 3.79e3.73 (m, 1H),
3.73 (s, 3H), 3.64 (ddd, J ¼ 10.8, 9.0, 5.1 Hz, 1H), 3.43 (t, J ¼ 9.0 Hz,
1H), 3.36 (t, J ¼ 10.8 Hz, 1H); 13C NMR (75 MHz, CD3OD):
d 160.8,
159.7, 156.5, 141.1, 140.9, 130.8, 129.7, 123.4, 123.3, 117.6, 117.3, 115.2,
112.61, 112.57, 107.72, 107.67, 106.0, 105.7, 80.3, 78.5, 75.2, 75.1, 71.7,
71.2, 55.8, 50.8; ESMS m/z: 388 (MHþ), 410 (MNaþ); HPLC purity
98.7%.
4.1.7. General procedure for the synthesis of compounds 1ae1g
A mixture of 9ae9gand 10%Pd/C inMeOH/THF(1/1)wasstirred at
room temperature under 1 atm H2(g) for 1w3 h. The reaction was
filtered through a pad of Celite and the filtrate was concentrated
under reduced pressure. The residue was then purified by column
chromatography (MeOH/CH2Cl2) to providethe final products 1ae1g.
4.1.7.7. 7-Fluoro-1-(4-methoxybenzyl)-3-(
b
-
d
D-xylopyranosyl)-1H-
indole (1g). 1H NMR (300 MHz, CD3OD):
7.46 (d, J ¼ 7.8 Hz, 1H),
7.30 (s, 1H), 7.11 (d, J ¼ 8.7 Hz, 2H), 6.95 (td, J ¼ 7.8, 4.5 Hz, 1H),
6.85e6.78 (m, 3H), 5.39 (s, 2H), 4.37 (d, J ¼ 9.9 Hz, 1H), 3.96 (dd,
J ¼ 11.1, 5.1 Hz, 1H), 3.73 (s, 3H), 3.73e3.61 (m, 2H), 3.45e3.32 (m,
4.1.7.1. 1-Benzyl-3-(
(400 MHz, CD3OD):
b
-
D
-xylopyranosyl)-1H-indole (1a). 1H NMR
7.71e7.69 (m, 1H), 7.34 (s, 1H), 7.30e7.17 (m,
d
2H); 13C NMR (75 MHz, CD3OD):
d 160.7, 153.2, 150.0, 133.0, 132.9,
6H), 7.11 (ddd, J ¼ 8.0, 7.2, 1.2 Hz, 1H), 7.04 (ddd, J ¼ 8.0, 7.2, 1.2 Hz,
1H), 5.36 (s, 2H), 4.42 (d, J ¼ 9.2 Hz, 1H), 3.98 (dd, J ¼ 11.2, 5.2 Hz,
1H), 3.75 (t, J ¼ 9.2 Hz, 1H), 3.68 (ddd, J ¼ 10.4, 9.2, 5.2 Hz, 1H), 3.45
(t, J ¼ 9.2 Hz,1H), 3.38 (dd, J ¼ 11.2,10.4 Hz,1H); 13C NMR (100 MHz,
132.0, 130.5, 129.5, 125.9, 125.7, 120.8, 120.7, 117.3, 117.2, 115.4, 115.3,
115.1, 108.5, 108.2, 80.1, 78.6, 75.1, 71.7, 71.4, 55.8, 52.7, 52.6; ESMS
m/z: 388 (MHþ), 410 (MNaþ); HPLC purity 99.0%.
CD3OD):
d
139.5, 138.4, 129.8, 129.1, 128.8, 128.6, 128.2, 123.0, 121.1,
4.1.8. General procedure for the synthesis of compounds 10he10j
A mixture of 8 and 10% Pd/C in MeOH/THF (1/1) was stirred at
room temperature under 1 atm H2(g) for 3 h. The reactionwas filtered
through a pad of Celite and the filtrate was concentrated under
reduced pressure. The residue was then purified by column chro-
matography (MeOH/CH2Cl2) to provide the final products 10he10j.
120.5, 114.2, 111.2, 80.2, 79.0, 75.1, 71.8, 71.4, 51.0; ESMS m/z: 340
(MHþ), 362 (MNaþ); HPLC purity 89.5%.
4.1.7.2. 1-(4-Fluorobenzyl)-3-(
b
-
D
-xylopyranosyl)-1H-indole (1b).
1H NMR (400 MHz, CD3OD):
d
7.69 (d, J ¼ 8.0 Hz,1H), 7.33 (s,1H), 7.28
(d, J ¼ 8.0 Hz,1H), 7.21e7.17 (m, 2H), 7.12e7.08 (m,1H), 7.05e6.97 (m,
3H), 5.34 (s, 2H), 4.41 (d, J ¼ 9.2 Hz,1H), 3.97 (dd, J ¼ 10.8, 5.2 Hz, 1H),
3.74 (t, J ¼ 9.2 Hz, 1H), 3.67 (ddd, J ¼ 10.8, 9.2, 5.2 Hz, 1H), 3.44 (t,
J ¼ 9.2 Hz, 1H), 3.37 (t, J ¼ 10.8 Hz, 1H); 13C NMR (75 MHz, CD3OD):
4.1.9. General procedure for the synthesis of compounds 1he1j
Cs2CO3 (5 equiv) was added to a stirred solution of 10he10j (1.0
equiv) and 4-cyclopropylbenzyl bromide (1.2 equiv) in DMF at room
temperature. After 24 h, the reaction was quenched by the addition
of H2O, and the resulting mixture was extracted with EtOAc. The
organic layer was dried over MgSO4, filtered and concentrated
under reduced pressure. The residue was purified by column
chromatography to afford the desired products 1he1j.
d
165.3,162.1,138.3,135.6,135.5,130.2,130.1,129.0,128.8,123.0,121.2,
120.5,116.6,116.3,114.4,111.2, 80.2, 79.0, 75.1, 71.8, 71.4, 50.2;ESMSm/
z: 358 (MHþ), 380 (MNaþ); HPLC purity 91.5%.
4.1.7.3. 1-(4-Methoxybenzyl)-3-(
b
-
D
-xylopyranosyl)-1H-indole (1c).
1H NMR (400 MHz, CD3OD):
d
7.67 (d, J ¼ 8.0 Hz, 1H), 7.32e7.30 (m,
2H), 7.14e7.08 (m, 3H), 7.04e7.00 (m, 1H), 6.84e6.81 (m, 2H), 5.27
(s, 2H), 4.40 (d, J ¼ 9.2 Hz, 1H), 3.96 (dd, J ¼ 10.8, 5.2 Hz, 1H),
3.75e3.71 (m, 1H), 3.73 (s, 3H), 3.67 (ddd, J ¼ 10.8, 9.2, 5.2 Hz, 1H),
3.44 (t, J ¼ 9.2 Hz, 1H), 3.37 (t, J ¼ 10.8 Hz, 1H); 13C NMR (75 MHz,
4.1.9.1. 1-(4-Cyclopropylbenzyl)-3-(
b
-
D
-xylopyranosyl)-1H-indole
(1h). 1H NMR (400 MHz, CD3OD):
d
7.67 (d, J ¼ 8.0 Hz, 1H),
7.30e7.28 (m, 2H), 7.11e6.96 (m, 6H), 5.28 (s, 2H), 4.40 (d,
J ¼ 9.6 Hz,1H), 3.96 (dd, J ¼ 10.8, 5.2 Hz,1H), 3.73 (dd, J ¼ 9.6, 8.8 Hz,
1H), 3.67 (ddd, J ¼ 10.8, 8.8, 5.2 Hz, 1H), 3.43 (t, J ¼ 8.8 Hz, 1H), 3.37
(t, J ¼ 10.8 Hz, 1H), 1.86e1.82 (m, 1H), 0.93e0.88 (m, 2H), 0.62e0.58
CD3OD):
d 160.8, 138.4, 131.4, 129.7, 128.9, 128.8, 122.9, 121.1, 120.4,
115.1, 114.0, 111.3, 80.2, 79.0, 75.1, 71.8, 71.4, 55.8, 50.5; ESMS m/z:
370 (MHþ), 392 (MNaþ); HPLC purity 99.0%.
(m, 2H); 13C NMR (75 MHz, CD3OD):
d 144.9, 138.4, 136.4, 129.0,
128.8, 128.3, 126.9, 122.9, 121.1, 120.4, 114.1, 111.3, 80.2, 79.0, 75.1,
71.8, 71.4, 50.7, 16.0, 9.7; ESMS m/z: 380 (MHþ), 402 (MNaþ); HPLC
purity 96.6%.
4.1.7.4. 1-(4-tert-Butylbenzyl)-3-(
b-D-xylopyranosyl)-1H-indole (1d).
1H NMR (400 MHz, CD3OD):
d
7.68 (d, J ¼ 8.0 Hz, 1H), 7.32e7.30 (m,
4H), 7.12e7.08 (m, 3H), 7.04e7.01 (m, 1H), 5.31 (s, 2H), 4.41 (d,
J ¼ 9.6 Hz, 1H), 3.96 (dd, J ¼ 10.8, 5.2 Hz, 1H), 3.74 (dd, J ¼ 9.6, 8.8 Hz,
1H), 3.67 (ddd, J ¼ 10.8, 8.8, 5.2 Hz, 1H), 3.44 (t, J ¼ 8.8 Hz,1H), 3.37 (t,
4.1.9.2. 1-(4-Cyclopropylbenzyl)-7-fluoro-3-(
b
-
D
-xylopyranosyl)-1H-
indole (1i). 1H NMR (400 MHz, CD3OD):
d
7.46 (d, J ¼ 8.0 Hz, 1H),
J ¼ 10.8 Hz, 1H), 1.26 (s, 9H); 13C NMR (75 MHz, CD3OD):
d
151.7,
7.30 (s, 1H), 7.05e6.92 (m, 5H), 6.83e6.78 (m, 1H), 5.41 (s, 2H), 4.37
(d, J ¼ 9.6 Hz, 1H), 3.96 (dd, J ¼ 10.8, 5.2 Hz, 1H), 3.69e3.62 (m, 2H),
3.42 (t, J ¼ 8.8 Hz, 1H), 3.36 (t, J ¼ 10.8 Hz, 1H), 1.90e1.80 (m, 1H),
0.94e0.88 (m, 2H), 0.62e0.58 (m, 2H); 13C NMR (100 MHz, CD3OD):
138.4, 136.5, 129.0, 128.7, 128.1, 126.7, 122.9, 121.1, 120.4, 114.1, 111.3,
80.2, 79.0, 75.1, 71.8, 71.4, 50.6, 35.5, 31.9; ESMS m/z: 396 (MHþ), 418
(MNaþ); HPLC purity 95.4%.
d
152.9, 150.5, 144.9, 137.0, 133.0, 132.9, 130.6, 128.1, 126.9, 120.82,
4.1.7.5. 1-(2-Naphthylmethyl)-3-(
1H NMR (300 MHz, CD3OD):
7.45e7.39 (m, 3H), 7.35e7.30 (m, 2H), 7.12e7.01 (m, 2H), 5.51 (s, 2H),
b
-
D
-xylopyranosyl)-1H-indole (1e).
120.75, 117.3, 117.2, 115.4, 108.5, 108.3, 80.2, 78.7, 75.1, 71.8, 71.4,
53.0, 52.9, 16.0, 9.8; ESMS m/z: 398 (MHþ), 420 (MNaþ); HPLC
purity 100%.
d
7.80e7.70 (m, 4H), 7.61 (s, 1H),

C.-H. Yao et al. / European Journal of Medicinal Chemistry 55 (2012) 32e38
37
4.1.9.3. 1-(4-Cyclopropylbenzyl)-7-methyl-3-(
1H-indole (1j). 1H NMR (400 MHz, CD3OD):
b
-
D
-xylopyranosyl)-
were purchased from Sigma (Deisenhofen, Germany). Briefly,
hSGLT2/CHO-K1 cells or hSGLT1/CHO-K1 cells were seeded into
a white-walled 96-well culture plate (Corning, NY, USA) at a density
of 30,000 cells/well (hSGLT2) or 20,000 cells/well (hSGLT1) and
incubated for 48 h at 37 ꢀC in a 5% CO2 atmosphere in growth
medium. After 48 h, the culture medium in the wells was removed
d
7.54 (d, J ¼ 8.0 Hz,1H),
7.23 (s, 1H), 6.96e6.89 (m, 3H), 6.81e6.78 (m, 3H), 5.54 (s, 2H), 4.41
(d, J ¼ 9.6 Hz, 1H), 3.97 (dd, J ¼ 10.8, 5.2 Hz, 1H), 3.75 (dd, J ¼ 9.6,
8.8 Hz, 1H), 3.67 (ddd, J ¼ 10.8, 8.8, 5.2 Hz, 1H), 3.44 (t, J ¼ 8.8 Hz,
1H), 3.38 (t, J ¼ 10.8 Hz, 1H), 2.45 (s, 3H), 1.86e1.79 (m, 1H),
0.93e0.85 (m, 2H), 0.61e0.57 (m, 2H); 13C NMR (100 MHz, CD3OD):
and wells were washed three times with 280 mL of KRH solution
d
144.5, 138.6, 137.1, 131.1, 129.9, 127.0, 126.6, 125.8, 122.6, 120.8,
and then incubated in KRH solution containing 3 uM [14C]-AMG in
the absence or presence of inhibitors for up to 120 min at 37 ꢀC. At
the end of the uptake period, the KRH solution was removed and
the uptake of [14C]-AMG was stopped by adding ice-cold KRH
solution (stop solution). The wells were rinsed three times with
119.1, 113.9, 80.3, 78.8, 75.0, 71.8, 71.4, 53.0, 19.8, 15.9, 9.7; ESMS m/
z: 394 (MHþ), 416 (MNaþ); HPLC purity 91.9%.
4.1.10. 1-(4-propylbenzyl)-3-(b-D-xylopyranosyl)-1H-indole (1k)
NaH (1.2 equiv) and 4-cyclopropylbenzyl bromide (1.2 equiv)
were sequentially added to a stirred solution of 8 (1.0 equiv) in DMF
at 0 ꢀC. The reaction was slowly warmed to room temperature and
stirred for 16 h. The reaction was quenched by the addition of H2O
at 0 ꢀC and extracted with CH2Cl2. The organic layer was dried over
MgSO4, filtered and concentrated under reduced pressure. The
residue was purified by column chromatography to afford the
alkylated intermediate.
150 mL stop buffer using the microplate washer (TEcan, Männedorf,
Switzerland). After the third rinse, the stop solution was completely
removed from the wells and the cells were solubilized by adding
0.1% sodium dodecyl sulfate (Sigma). After 24 h, the microtiter plate
was taken for scintillation counting of radioactive [14C]-AMG using
a TopCount (Perkin Elmer). The percent of inhibition of inhibitors
was calculated by comparing count per minute (CPM) in inhibitor-
containing well with CPM in wells containing only DMSO vehicle.
A mixture of the alkylated intermediate and 10% Pd/C in MeOH/
THF (1/1) was stirred at room temperature under 1 atm H2(g) for 3 h.
The reaction was filtered through a pad of Celite and the filtrate was
concentrated under reduced pressure. The residue was then puri-
fied by column chromatography (MeOH/CH2Cl2) to provide the
Dapagliflozin was evaluated in parallel in every assay. A
doseeresponse curve was fitted to a sigmoidal doseeresponse
model using GraphPad software to determine the inhibitor
concentration at half-maximal response (EC50).
final product 1k. 1H NMR (400 MHz, CD3OD):
d 7.69e7.67 (m, 1H),
7.31e7.29 (m, 2H), 7.12e7.00 (m, 6H), 5.31 (s, 2H), 4.41 (d, J ¼ 9.6 Hz,
1H), 3.96 (dd, J ¼ 10.8, 5.2 Hz, 1H), 3.74 (dd, J ¼ 9.6, 9.2 Hz, 1H), 3.67
(ddd, J ¼ 10.8, 9.2, 5.2 Hz, 1H), 3.43 (t, J ¼ 9.2 Hz, 1H), 3.37 (t,
J ¼ 10.8 Hz, 1H), 2.52 (t, J ¼ 7.6 Hz, 2H), 1.58 (sext, J ¼ 7.6 Hz, 2H),
4.3. In vivo pharmacokinetics evaluation of 1i in rats [9]
The animal study was approved by Institutional Animal Care
and Use Committee of National Health Research Institutes. A
solution of test compound (1 mg/mL) was prepared by dissolving
appropriate amount compound in a mixture of PEG 400/DMA
(80:20, v/v). Male SpragueeDawley rats, weighing 250e350 g each
(8e10 weeks old), were obtained from BioLASCO, Ilan, Taiwan. 1i
(powder form obtained from column purification) was adminis-
tered to three male rats each intravenously by a bolus injection to
the jugular vein or orally at 1 mg/kg dose. At 0 (prior to dosing), 2, 5
(IV only), 15, and 30 min and at 1, 2, 4, 6, 8 and 24 h after dosing,
0.89 (t, J ¼ 7.6 Hz, 3H); 13C NMR (100 MHz, CD3OD):
d 143.3, 138.4,
136.7, 129.9, 129.0, 128.8, 128.3, 122.9, 121.1, 120.4, 114.1, 111.3, 80.2,
79.0, 75.1, 71.8, 71.4, 50.8, 38.8, 25.9, 14.2; ESMS m/z: 382 (MHþ),
404 (MNaþ); HPLC purity 90.6%.
4.1.11. p-Cyclopropylbenzyl bromide (11)
To a stirred solution of p-cyclopropylbenzaldehyde [15] (1.14 g,
7.81 mmol) in a mixture of solvents (MeOH/THF ¼ 1/1, 30 mL) was
added sodium borohydride (450 mg, 7.81 mmol) at room temper-
ature. After 0.5 h, the solution was concentrated in vacuo and the
a blood sample (w150 mL) was collected from each animal via the
jugular-vein cannula and stored in ice (0e4 ꢀC). Plasma was sepa-
rated from the blood by centrifugation (14,000 ꢃ g for 15 min at
residue redissolved in CH2Cl2, and washed with H2O and NaCl(aq)
.
The organic phase was dried over MgSO4, filtered, and concentrated
in vacuo. The residue was purified by column chromatography
(EtOAc/Hex ¼ 1/3) to provide p-cyclopropylbenzyl alcohol (1.12 g,
97%). Phosphorus tribromide (1.0 mL, 11.3 mmol) was added to
a stirred solution of p-cyclopropylbenzyl alcohol (1.12 g, 7.56 mmol)
in CH2Cl2/Et2O (1/1, 25 mL) at 0 ꢀC under argon. The reaction was
warmed up to room temperature gradually and stirred for 1 h. The
reaction was quenched by the addition of H2O and extracted with
Et2O. The organic phase was dried over MgSO4, filtered, and
concentrated in vacuo. The residue was purified by column chro-
matography (EtOAc/Hex ¼ 1/15) to afford p-cyclopropylbenzyl
4
ꢀC in a Beckman Model AllegraTM 6R centrifuge) and stored in
a freezer (ꢁ80 ꢀC). All samples were analyzed for the test
compound by LC-MS/MS (ABI 4000Q Trap). Plasma concentration
data were analyzed with standard non-compartmental method.
4.4. Blood glucose lowering effect of 1i in STZ-Induced diabetic rats
[9,14a]
Adult male SpragueeDawley rats (BioLasco, Ilan, Taiwan)
received three intraperitoneal injections of streptozotocin from
Sigma (catalog no. S0130, St. Louis, MO) at 65 mg/kg freshly
prepared in 0.01 M citrate buffer at one injection every other day.
The animals were then monitored by using blood glucometer
(ACCU-CHEK from Roche, Basel, Switzerland), for the glucose levels
in blood collected via stabbing the tail vein with 25G needle once
a week. A streptozotocin-induced diabetes was confirmed when
the blood glucose levels were up more than 400 mg/dL. The dia-
betic rats were divided into experimental groups at six rats each
and orally administered a single dose of 1i, 12, or vehicle as control.
Blood samples were obtained from the tail vein at 0 (predose), 0.5,
1, 2, 3, 4, and 5 h after the oral administration and measured for
blood glucose levels with the glucometer. A P value of less than 0.05
was considered statistically significant.
bromide (1.19 g, 74%). 1H NMR (300 MHz, CDCl3):
d 7.27 (d,
J ¼ 8.0 Hz, 2H), 7.03 (d, J ¼ 8.0 Hz, 2H), 4.82 (s, 2H), 1.84e1.93 (m,
1H), 0.89e1.00 (m, 2H), 0.67e0.72 (m, 2H).
4.2. In vitro human SGLT inhibition assays [9,12,13]
Stably transfected CHO-K1 cells were used for transporter
studies. SGLT was determined by uptake of [14C]-
a-methyl-D-glu-
copyranoside ([14C]-AMG, specific radioactivity 310 mCi/mmol)
purchased from Perkin Elmer (Boston, USA). For the purpose of this
study, Krebs-Ringer-Henseleit (KRH) solution containing 120 mM
NaCl, 4.7 mM KCl, 1.2 mM MgCl2, 2.2 mM CaCl2, and 10 mM Hepes
(pH 7.4 with Tris) was used for SGLT uptake assays. All chemicals

38
C.-H. Yao et al. / European Journal of Medicinal Chemistry 55 (2012) 32e38
[9] C.H. Yao, J.S. Song, C.T. Chen, T.K. Yeh, M.S. Hung, C.C. Chang, Y.W. Liu,
Acknowledgements
M.C. Yuan, C.J. Hsieh, C.Y. Huang, M.H. Wang, C.H. Chiu, T.C. Hsieh, S.H. Wu,
W.C. Hsiao, K.F. Chu, C.H. Tsai, Y.S. Chao, J.C. Lee, J. Med. Chem. 54 (2011)
166e178.
We are grateful to the National Health Research Institutes and
National Science Council of Taiwan (NSC 99-2323-B-400-003) for
financial support.
[10] S. Yonekubo, N. Fushimi, CA 2588963, 2006.
[11] (a) R. Lopez, A. Fernandez-Mayoralas, J. Org. Chem. 59 (1994) 737e745;
(b) C.G. Lucero, K.A. Woerpel, J. Org. Chem. 71 (2006) 2641e2647.
[12] F. Castaneda, R.K.H. Kinne, Mol. Cell. Biochem. 280 (2005) 91e98.
[13] J.T. Lin, J. Kormanec, F. Wehner, S. Wielert-Badt, R.K.H. Kinne, Biochim. Bio-
phys. Acta 1373 (1998) 309e320.
References
[14] (a) W. Meng, B.A. Ellsworth, A.A. Nirschl, P.J. McCann, M. Patel, R.N. Girotra,
G. Wu, P.M. Sher, E.P. Morrison, S.A. Biller, R. Zahler, P.P. Deshpande,
A. Pullockaran, D.L. Hagan, N. Morgan, J.R. Taylor, M.T. Obermeier,
W.G. Humphreys, A. Khanna, L. Discenza, J.G. Robertson, A. Wang, S. Han,
J.R. Wetterau, E.B. Janovitz, O.P. Flint, J.M. Whaley, W.N. Washburn, J. Med.
Chem. 51 (2008) 1145e1149;
(b) S. Han, D.L. Hagan, J.R. Taylor, L. Xin, W. Meng, S.A. Biller, J.R. Wetterau,
W.N. Washburn, J.M. Whaley, Diabetes 57 (2008) 1723e1729.
[15] A. Conte-Mayweg, H. Kuehne, T. Luebbers, C. Maugeais, W. Mueller, P. Pflieger,
US 7259183, 2007.
[1] E.M. Wright, Am. J. Physiol. Ren. Physiol. 280 (2001) F10eF18.
[2] E.M. Wright, E. Turk, Pflugers Arch. 447 (2004) 510e518.
[3] E. Turk, B. Zabel, S. Mundlos, J. Dyer, E.M. Wright, Nature 350 (1991) 354e356.
[4] M.G. Martín, E. Turk, M.P. Lostao, C. Kerner, E.M. Wright, Nat. Genet. 12 (1996)
216e220.
[5] M. Kasahara, M. Maeda, S. Hayashi, Y. Mori, T. Abe, Biochim. Biophys. Acta
1536 (2001) 141e147.
[6] L.T. Ho, S.S. Kulkarni, J.C. Lee, Curr. Top. Med. Chem. 11 (2011) 1476e1512.
[7] W.N. Washburn, Expert Opin. Ther. Patents 19 (2009) 1485e1499.
[8] W.N. Washburn, J. Med. Chem. 52 (2009) 1785e1794.