O-Spiro C-aryl glucosides as novel sodium-dependent glucose co-transporter 2 (SGLT2) inhibitors

Article abstract of DOI:10.1016/j.bmcl.2009.08.030

Two series of O-spiro C-aryl glucosides were synthesized and tested for inhibition of hSGLT1 and hSGLT2. 6′-O-Spiro C-aryl glucosides exhibited potent in vitro hSGLT2 inhibitory activity but 2′-O-spiro C-aryl glucosides showed no in vitro hSGLT2 inhibitory activity at a screening concentration of 1 μM.

Full text of DOI:10.1016/j.bmcl.2009.08.030

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5632–5635
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
O-Spiro C-aryl glucosides as novel sodium-dependent glucose
co-transporter 2 (SGLT2) inhibitors
Baihua Xu a,b,c, , Binhua Lv a,b,c, Yan Feng b, Ge Xu b, Jiyan Du b, Ajith Welihinda d, Zelin Sheng b, Brian Seed d,
*
Yuanwei Chen a,b,
*
a Chengdu Institute of Organic Chemistry, Chinese Academy of Sciences, Chengdu, Sichuan 610041, PR China
b Egret Pharma (Shanghai) Limited, Shanghai 201203, PR China
c Graduate School of Chinese Academy of Sciences, Beijing 100049, PR China
d Theracos, Inc., Sunnyvale, CA 94085, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Two series of O-spiro C-aryl glucosides were synthesized and tested for inhibition of hSGLT1 and hSGLT2.
60-O-Spiro C-aryl glucosides exhibited potent in vitro hSGLT2 inhibitory activity but 20-O-spiro C-aryl glu-
cosides showed no in vitro hSGLT2 inhibitory activity at a screening concentration of 1 lM.
Received 11 May 2009
Revised 28 July 2009
Accepted 7 August 2009
Available online 12 August 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
SGLT
Spiro
Glucosides
According to the World Health Organization, more than 180
million people worldwide have diabetes mellitus. This number is
likely to more than double by 2030. Diabetes is characterized by
failing to produce enough insulin from pancreatic b-cells (type 1
diabetes),1 or failing to respond properly to the insulin produced
by the pancreas (type 2 diabetes).2 The essential characteristic of
diabetes is hyperglycemia, which is considered to be the major
pathogenic factor for the development of serious complications
including retinopathy,3 cardiovascular disease,4 nephropathy,5
neuropathy,6 ulcers7 and heart disease.8
dial blood glucose level in hyperglycemic streptozotocin (STZ)-in-
duced diabetic rats.12 BMS’ biological results with dapagliflozin
prompted us to design two novel types of O-spiro C-aryl gluco-
sides, 60-O and 20-O-spiro C-aryl glucosides (Fig. 1). Chugai Seiyaku
Kabushiki Kaisha disclosed a patent for application of 60-O-spiro C-
aryl glucosides without the 4-chloro substituent as SGLT inhibitors
prior to us.13 Herein, we report our initial design, synthesis, and
biological evaluation of the O-spiro C-aryl glucoside SGLT2
inhibitors.
Cellular glucose transport is conducted by two classes of glu-
cose transporters, the facilitative glucose transporters (GLUTs)
and sodium-dependent glucose co-transporters (SGLTs).9 Two
important isoforms of SGLT, SGLT1 and SGLT2, were identified,
the former is found predominantly in the intestinal brush border,
while the latter is localized in the renal proximal tubule and is
responsible for the majority of glucose re-absorption by the kid-
ney.10 Recent studies suggest that inhibition of SGLT at the kidney
may be a useful approach to decrease glucose absorption and this
could result in the increase of the amount of glucose excreted in
the urine.11
R2
1
6' 4'
OH
R
O
HO
O
HO OH
2 R1 = H, R2 = OEt
Cl
2'
4'
3 R1 = Cl, R2 = Et
6'
R2
O
HO
OH
OH
1 Dapagliflozin
1
2'
4'
R
OEt
Dapagliflozin 1, a C-aryl glucoside SGLT2 inhibitor developed by
Bristol-Myers Squibb Company, exhibits powerful ability of urinary
glucose excretion and can dramatically lower fasting and postpran-
O
HO
O
HO
OH
HO OH
4 R1 = H, R2 = Et
* Corresponding authors. Tel.: +86 28 85232730; fax: +86 28 85259387.
E-mail addresses: xubaihua2002@163.com (B. Xu), chenyw@cioc.ac.cn (Y. Chen).
Figure 1.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.08.030

B. Xu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5632–5635
5633
Two possible kinds of spiro compounds at the 60-position and
the 20-position of the proximal phenyl ring can be constructed con-
necting at the C-1 position of the glucose ring. Accordingly, we syn-
thesized 60-O-spiro compounds 2 and 3, and 20-O-spiro compound
4 (Fig. 1). Beginning with the synthesis of 60-O-spiro glucoside 2, as
shown in Scheme 1, commercially available 3-bromo-4-methyl-
benzoic acid 5 was converted to acid chloride with oxalyl chloride,
which subsequently underwent Friedel–Crafts acylation reaction
with phenetole to afford diaryl ketone 6. Bromination of the latter
ketone with N-bromosuccinimide followed by displacement with
sodium acetate provided acetate 8. Selective reduction of the car-
bonyl group of 8 with triethylsilane and boron trifluoride etherate
gave diarylmethane 9. Hydrolysis of 9 followed by silylation of the
ensuing alcohol with trimethylsilyl chloride furnished silyl ether
10. Treatment of the silyl ether with n-BuLi at ꢀ78 °C, followed
a same group in the corresponding position of 60-O-spiro glucoside
2 to obtain 60-O-spiro glucoside 3, as shown in Scheme 2. Bromin-
ation of commercially available 3-amino-4-methylbenzoic acid 13
with N-bromosuccinimide afforded bromide 14, which subse-
quently underwent esterification to give ester 15. Chlorination of
the ester under standard Sandmeyer reaction conditions provided
chloride 16, which was oxidized by KMnO4 to generate benzoic
acid 17. Friedel–Crafts acylation of ethylbenzene with benzoyl
chloride, prepared from benzoic acid 17 using oxalyl chloride, gave
diaryl ketone 18. Selective reduction of the latter ketone with tri-
ethylsilane in the presence of catalytic amount of trifluorometh-
anesulfonic acid gave the corresponding diarylmethane 19, which
was reduced with sodium borohydride, followed by protection of
the resulting benzyl alcohol with chloromethoxymethane to afford
methoxymethyl ether 20. Treatment of this ether using a similar
approach that was described for synthesis of compound 2 provided
the desired 60-O-spiro glucoside 3.16
by the addition of 2,3,4,6-tetra-O-trimethylsilyl-D-gluconolactone
11, prepared from corresponding gluconolactone in the presence
of trimethylsilyl chloride and N-methylmorpholine,14 afforded cor-
responding coupling intermediate 12, which finally underwent
deprotection and cyclization in one-pot with methylsulfonic acid
to provide the desired 60-O-spiro glucoside 2 as a single anomer15
in moderate yield.16
Scheme 3 depicts the synthesis of 20-O-spiro glucoside 4. Bro-
mination of commercially available 3-bromo-2-methylbenzoic
acid 22 with N-bromosuccinimide gave benzylbromide 23. Our ini-
tial attempts to displace the bromide in 23 with sodium acetate
failed to yield the desired acetate, probably due to intramolecular
lactonization. Thus, benzylbromide 23 was converted to the acid
chloride with oxalyl chloride and this material was subsequently
In view of that the 40-position of the proximal phenyl ring of
dapagliflozin is substituted by a chloro group, we thus introduced
OEt
OEt
OEt
R
Br
AcO
Br
a
b
d
Br
CO2H
Br
O
O
7
: R = Br
5
6
9
c
8
: R = OAc
e, f
O
O
TMSO
HO
O
TMSO
OEt
h
OEt
TMSO
11
OTMS
OEt
O
TMSO
Br
OTMS
HO
O
TMSO
OH
OH
OTMS
g
TMSO OTMS
HO
2
12
10
Scheme 1. Reagents and conditions: (a) oxalyl chloride, DMF, AlCl3, phenetole, CH2Cl2 (75%); (b) NBS, AIBN, CCl4, reflux (50%); (c) AcONa, DMF, 68 °C (95%); (d) BF3ꢁEt2O,
Et3SiH, CH3CN/ClCH2CH2Cl (2:1), rt (61%); (e) LiOHꢁH2O, THF/MeOH/H2O (2:3:1), rt (92%); (f) TMSCl, N-methylmorpholine, THF (95%); (g) n-BuLi, 2,3,4,6-tetra-O-
trimethylsilyl-
D-gluconolactone 11, THF/toluene (1:2), ꢀ78 °C, then H2O; (h) MeSO3H, THF, ꢀ78 °C to rt (37% two steps).
Cl
Cl
NH2
NH2
NH2
HO2C
a
c
b
d
CO2H
CO2H
CO2Me
CO2Me
CO2Me
Br
14
Br
15
Br
16
Br
17
13
e
g, h
f
Cl
MeO2C
Br
Cl
MeO2C
Br
Cl
MOMO
Br
20
19
O
18
i
MOMO
HO
Cl
OTMS
Cl
O
j
TMSO
O
HO
O
OH
TMSO OTMS
HO OH
3
21
Scheme 2. Reagents and conditions: (a) NBS, DMF, 5 °C (87%); (b) SOCl2, MeOH, reflux (99%); (c) CuCl, NaNO2, conc. HCl, 1,4-dioxane, H2O, 0 °C (93%); (d) KMnO4, 18-crown-
6, MgSO4, t-BuOH/H2O (1:2), reflux (32%); (e) oxalyl chloride, DMF, AlCl3, ethylbenzene, CH2Cl2 (87%); (f) Et3SiH, CF3SO3H, TFA, rt (100%); (g) NaBH4, MeOH, THF
(quantitative); (h) MOMCl, DIPEA, CH2Cl2, rt (90%); (i) n-BuLi, 2,3,4,6-tetra-O-trimethylsilyl-
ꢀ78 °C to rt (63% two steps).
D-gluconolactone 11, THF/toluene (1:2), ꢀ78 °C, then H2O; (j) MeSO3H, THF,

5634
B. Xu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5632–5635
Br
Br
R
AcO
Br
O
Br
CO2H
CO2H
Br
a
b
d
24: R = Br
22
23
26
e, f
c
25: R = OAc
MOMO
Br
MOMO
h
g
O
HO
O
O
HO
TMSO
OH
OH
4
OTMS
OTMS
28
27
HO
TMSO
Scheme 3. Reagents and conditions: (a) NBS, AIBN, CCl4, reflux (95%); (b) oxalyl chloride, DMF, AlCl3, ethylbenzene, CH2Cl2; (c) AcONa, DMF, 68 °C; (d) BF3ꢁEt2O, Et3SiH,
CH3CN/ClCH2CH2Cl (2:1), rt (36% three steps); (e) LiOHꢁH2O, THF/MeOH/H2O (2:3:1), rt; (f) MOMCl, DIPEA, CH2Cl2, rt (44% two steps); (g) n-BuLi, 2,3,4,6-tetra-O-
trimethylsilyl-
D-gluconolactone 11, THF/toluene (1:2), ꢀ78 °C, then H2O; (h) MeSO3H, THF, ꢀ78 °C to rt (41% two steps).
ments of all spectra and Dr. Jacques Y. Roberge for thoughtful re-
Table 1
view of this Letter.
In vitro data for hSGLT inhibitory activity and selectivity
Compds
hSGLT2 IC50 (nM)
hSGLT1 IC50 (nM)
Selectivity
(hSGLT1/hSGLT2)
References and notes
1
2
3
4
6.7 (4.8–9.0)a
71 (52–96)a
6.6 (3.4–12.8)a
0%b
885 (528–1480)a
10,000–100,000
620 (450–853)a
38%c
132
141–1410
94
—
1. Del Prato, S.; Matsuda, M.; Simonson, D. C.; Groop, L. C.; Sheehan, P.; Leonetti,
F.; Bonadonna, R. C.; DeFronzo, R. A. Diabetologia 1997, 40, 687.
2. Bonadonna, R. C. Rev. Endocr. Metab. Disord. 2004, 5, 89.
3. The Diabetes Control and Complications Trial Research Group. Diabetes 1995,
44, 968.
4. Haffner, S. J.; Cassells, H. Am. J. Med. 2003, 115, 6s–11s.
5. The Diabetes Control and Complications Trial Research Group. Kidney Int. 1995,
47, 1703.
a
b
c
Numbers in parentheses indicate 95% confidence intervals.
Inhibition at a screening concentration of 1 M.
Inhibition at a screening concentration of 100 M.
l
l
6. The Diabetes Control and Complications Trial Research Group. Ann. Int. Med.
1995, 122, 561.
subjected to Friedel–Crafts acylation reaction with ethylbenzene to
afford diaryl ketone 24. The bromide 24 was substituted with so-
dium acetate, followed by selective reduction of the carbonyl with
triethylsilane, to yield acetate 26. Saponification of acetate 26, and
protection of the resulting alcohol with chloromethoxymethane
provided methoxymethyl ether 27. Treatment of 27 using a similar
method used for synthesis of analogue 2 gave target 20-O-spiro glu-
coside 4 in moderate yield.16
7. (a) Boulton, A. J. M.. In Textbook of Diabetes; Pickup, J. C., Williams, G., Eds.;
Blackwell Science: Oxford UK, 1997; Vol. 2, p 58; (b) Hoeldtke, R. D.; Davis, K.
M.; Hshieh, P. B.; Gaspar, S. R.; dworkin, G. E. J. Diabetes Compl. 1994, 8, 117.
8. Klein, R. Diabetes Care 1995, 18, 258.
9. (a) Thorens, B. Am. J. Physiol. 1996, 270, G541–G553; (b) Wright, E. M. Am. J.
Physiol. Renal Physiol. 2001, 280, F10–F18.
10. (a) Lee, W. S.; Kanai, Y.; Wells, R. G.; Hediger, M. A. J. Biol. Chem. 1994, 269,
12032; (b) You, G.; Lee, W. S.; Barros, E. J. G.; Kanai, Y.; Huo, T. L.; Khawaja, S.;
Wells, R. G.; Nigam, S. K.; Heidiger, M. A. J. Biol. Chem. 1995, 270, 29365; (c)
Kanai, L.; Lee, W. S.; You, G.; Brown, D.; Hediger, M. A. J. Clin. Invest. 1994, 93,
397; (d) Yano, H.; Seino, Y.; Imura, H. Diabetes Front. 1990, 1, 417.
11. (a) Arakawa, K.; Ishihara, T.; Oku, A.; Nawano, M.; Ueta, K.; Kitamura, K.;
Matsumoto, M.; Saito, A. Br. J. Pharmacol. 2001, 132, 578; (b) Oku, A.; Ueta, K.;
Arakawa, K.; Ishihara, T.; Nawano, M.; Kuronuma, Y.; Matsumoto, M.; Saito, A.;
Tsujihara, K.; Anai, M.; Asano, T.; Kanai, Y.; Endou, H. Diabetes 1999, 48, 1794.
12. Meng, W.; Ellsworth, B. A.; Nirschl, A. A.; McCann, P. J.; Patel, M.; Wu, G.; Sher,
P. M.; Morrison, E. P.; Biller, S. A.; Zahler, R.; Deshpande, P. P.; Pullockaran, A.;
Hagan, D. L.; Morgan, N.; Taylor, J. R.; Obermeier, M. T.; Humphreys, W. G.;
Khanna, A.; Discenza, L.; Robertson, J. G.; Wang, A.; Han, S.; Wetterau, J. R.;
Janovitz, E. B.; Flint, O. P.; Whaley, J. M.; Washburn, W. N. J. Med. Chem. 2008,
51, 1145.
All compounds were screened in a cell-based SGLT functional
assay,17 and hSGLT inhibitory activity (IC50
) and selectivity
(hSGLT1/hSGLT2) are presented in Table 1. Using dapagliflozin 1
as the reference compound we identified 60-O-spiro C-aryl gluco-
side 2 as our lead compound because of its good inhibitory activity
toward hSGLT2 (IC50 = 71 nM). Introduction of a chloro group at
the 40-position of the proximal phenyl ring led to analogue 3 with
a 10-fold elevation of the hSGLT2 inhibitory activity with an IC50
value of 6.6 nM, which was similar to that of dapagliflozin 1
(6.7 nM) in the same assay. This result suggested that introducing
a substituent in the 40-position of the phenyl ring was very impor-
tant for improvement of the inhibitory activity.12 However, 60-O-
spiro C-aryl glucoside 3 was less selective for hSGLT2 versus
hSGLT1 than dapagliflozin. On the other hand, 20-O-spiro C-aryl
glucoside 4 showed no inhibitory activity toward hSGLT2 at a
13. (a) Kobayashi, T.; Sato, T.; Nishimoto, M. PCT Int. Appl. WO2006080421, 2006;
Chem. Abstr. 2006, 145, 189115.; (b) Isaji, M. Curr. Opin. Invest. Drugs 2007, 8,
285.
14. 2,3,4,6-tetra-O-trimethylsilyl-D-glucolactone was prepared according to the
following literature procedures: (a) Horton, D.; Priebe, W. Carbohydr. Res. 1981,
94, 27; b Deshpande, P. P.; Ellsworth, B. A.; Singh, J.; Denzel, T. W.; Lai, C.;
Crispino, G.; Randazzo, M. E.; Gougoutas, J. Z. PCT Int. Appl. WO2004063209,
2004; Chem. Abstr. 2004, 141, 89317.
15. Barrett, A. G. M.; Pena, M.; Willardsen, J. A. J. Chem. Soc., Chem. Commun. 1995,
11, 1147; Barrett, A. G. M.; Pena, M.; Willardsen, J. A. J. Org. Chem. 1996, 61,
1082.
screening concentration of 1 lM. All above results demonstrated
that 60-O-spiro was the preferred conformation for the binding site.
In summary, we have identified a novel and potent hSGLT2
inhibitor series 60-O-spiro C-aryl glucosides. Glucoside 3 had simi-
lar in vitro hSGLT2 inhibitory activity and a little less selectivity as
compared to dapagliflozin 1. Further modification of this series will
be reported in due course.
16. All compounds were provided satisfactory spectra data (1H NMR and/or ESI
MS). The detailed experimental procedures have been published in the
following patent application: Chen, Y.; Feng, Y.; Xu, B.; Lv, B.; Dong, J.; Seed,
B.; Hadd, M. J. PCT Int. Appl. WO2007140191, 2007; Chem. Abstr. 2007, 147,
542063.
17.
A plasmid bearing the human full-length SGLT1 coding sequence in the
pDream 2.1 mammalian expression vector was purchased from GenScript
Corporation. A full-length human SGLT2 cDNA (GenScript Corporation) was
cloned into the pEAK15 mammalian expression vector. Human SGLT1
expression plasmid DNA was transfected into COS-7 cells (American Type
Culture Collection) using Lipofectamine 2000 (Invitrogen Corporation).
Acknowledgments
Transfected cells were evaluated for SGLT1 activity in methyl-
a-D-
We gratefully acknowledge Ying Chen for activity testing, the
analytical group of Egret Pharma (Shanghai) Limited for measure-
[U-14C]glucopyranoside (AMG) uptake assay and cryopreserved until use.

B. Xu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5632–5635
5635
L of
Plasmid containing human SGLT2 was linearized and stably transfected into
(SGLT1 assay) or 1.5 h (SGLT2 assay). Cells were washed twice with 150
wash buffer (137 mM N-methylglucamine, 10 mM Tris/Hepes, pH 7.2) and
methyl-
-[U-14C]glucopyranoside uptake was quantitated using a TopCount
scintillation counter (PerkinElmer, Inc.). Inhibitors were assayed at
l
HEK293.ETN cells. SGLT2-expressing clones were selected based on resistance
to puromycin (Invitrogen Corporation) and activity in AMG uptake assay. Cells
expressing SGLT1 or SGLT2 were seeded on 96-well ScintiPlates (PerkinElmer,
a-D
8
Inc.) in DMEM containing 10% FBS (1 ꢂ 105 cells per well in 100
l
L medium)
concentrations in triplicates. Sodium-dependent glucopyranoside uptake was
calculated by subtracting the values obtained with sodium-free buffer from
those obtained using sodium buffer. In general, ratios of sodium-dependent to
sodium-independent AMG uptake in SGLT1 and SGLT2 expressing cells were
10–15 and 15–20, respectively. Results of AMG uptake were analyzed using
GraphPad Prism (Intuitive Software for Science). IC50 calculations were
incubated at 37 °C under 5% CO2 for 48 h prior to the assay. Cells were washed
twice with 150 lL of either sodium buffer (137 mM NaCl, 5.4 mM KCl, 2.8 mM
CaCl2, 1.2 mM MgCl2, 10 mM tris(hydroxymeth-yl)aminomethane/N-2-
hydroxyethylpiperazine-N0-ethanesulfonic acid [Tris/Hepes], pH 7.2) or
sodium-free buffer (137 mM N-methyl-glucamine, 5.4 mM KCl, 2.8 mM CaCl2,
1.2 mM MgCl2, 10 mM Tris/Hepes, pH 7.2). Test compound in 50
l
L each of
performed using nonlinear regression with variable slope. As a reference
standard, a derivative of dapagliflozin was routinely included in the assays. In
26 independent evaluations, the reference compound inhibited SGLT2 activity
sodium or sodium-free buffer containing 40
l
Ci/mL methyl-
a-
D-
[U-14C]glucopyranoside (Amersham Biosciences/GE Healthcare) was added
per well of a 96-well plate and incubated at 37 °C with shaking for either 2 h
by 69.7 9.6% and SGLT1 by 72.7 6.7% at 10 nM and 10 lM, respectively.