C-Aryl 5a-carba-β-d-glucopyranosides as novel sodium glucose cotransporter 2 (SGLT2) inhibitors for the treatment of type 2 diabetes

Article abstract of DOI:10.1016/j.bmc.2012.04.053

C-Aryl 5a-carba-β-d-glucopyranose derivatives were synthesized and evaluated for inhibition activity against hSGLT1 and hSGLT2. Modifications to the substituents on the two benzene rings resulted in enhanced hSGLT2 inhibition activity and extremely high h

Full text of DOI:10.1016/j.bmc.2012.04.053

Bioorganic & Medicinal Chemistry 20 (2012) 4117–4127
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
C-Aryl 5a-carba-b-D-glucopyranosides as novel sodium glucose cotransporter
2 (SGLT2) inhibitors for the treatment of type 2 diabetes
⇑
Yoshihito Ohtake , Tsutomu Sato, Hiroharu Matsuoka, Takamitsu Kobayashi, Masahiro Nishimoto,
Naoki Taka, Koji Takano, Keisuke Yamamoto, Masayuki Ohmori, Takashi Higuchi, Masatoshi Murakata,
Kazumi Morikawa, Nobuo Shimma, Masayuki Suzuki, Hitoshi Hagita, Kazuharu Ozawa, Koji Yamaguchi,
Motohiro Kato, Sachiya Ikeda
Research Division, Fuji Gotemba Research Laboratory, Chugai Pharmaceutical Co. Ltd, 1–135 Komakado, Gotemba, Shizuoka 412–8513, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
C-Aryl 5a-carba-b-D-glucopyranose derivatives were synthesized and evaluated for inhibition activity
Received 28 March 2012
Revised 26 April 2012
Accepted 27 April 2012
Available online 4 May 2012
against hSGLT1 and hSGLT2. Modifications to the substituents on the two benzene rings resulted in
enhanced hSGLT2 inhibition activity and extremely high hSGLT2 selectivity versus SGLT1. Using the cre-
ated superimposed model, the reason for the high hSGLT2 selectivity was speculated to be that additional
substituents occupied a new space, in a different way than known inhibitors. Among the tested com-
pounds, the ethoxy compound 5h with high hSGLT2 selectivity exhibited more potent and longer hypo-
glycemic action in db/db mice than our O-carbasugar compound (1) and sergliflozin (2), which could be
explained by its improved PK profiles relative to those of the two compounds. These results indicated that
5h might be a promising drug candidate for the treatment of type 2 diabetes.
Keywords:
Type 2 diabetes
SGLT2 inhibitor
Carbasugar
Ó 2012 Elsevier Ltd. All rights reserved.
5a-Carba-b-D-glucopyranose
1. Introduction
benzene ring of the aglycon.4 We also demonstrated that one of
them (1, Fig. 1) showed similar levels with longer duration
Type 2 diabetes (T2D) mellitus is a progressive metabolic dis-
ease characterized by decreased insulin secretion and increased
insulin resistance, and the prevalence has been increasing yearly.
However, it is generally accepted that current medication does
not adequately control the blood glucose in patients with diabetes.
In addition, hypoglycemia is reported as one of the side effects
when insulin or sulfonylureas is administered. Therefore, more
effective and safer new drugs for the treatment of T2D are strongly
desired.
In the last decade, sodium glucose cotransporter 2 (SGLT2) has
been attracting attention because it is uniquely responsible for
reabsorbing glucose from the renal filtrate.1 SGLT2 is primarily ex-
pressed at the proximal convoluted tubules, whereas SGLT1 is ex-
pressed in the small intestine, heart, brain, and renal tubules.
Furthermore, SGLT1 mutation is known to cause glucose–galactose
malabsorption,2 so SGLT1 inhibition is reported to have possible
gastrointestinal side effects.3 From these findings, SGLT2-selective
inhibitors are considered to be safer than non-selective inhibitors.
in vivo efficacy than sergliflozin5 (2, Fig. 1). The O-linked carbasug-
ar derivatives have good characteristics as selective SGLT2
OMe
O
OMe
O
OH
OH
OH
OH
O
HO
O
R
OH
OH
2 sergliflozin (R=CO2Et)
2-active (2-A) (R=H)
(O-glucopyranoside)
1
Cl
S
O
OH
O
OH
OH
O
Recently, we reported metabolically stable O-aryl 5a-carba-b-D-
glucopyranosides which are carbasugar analogs as novel SGLT2
inhibitors, revealing their SAR for the substituent on the distal
HO
HO
OH
F
OH
OH
4 canagliflozin
(C-glucopyranoside)
3 dapagliflozin
(C-glucopyranoside)
⇑
Corresponding author. Tel.: +81 550 87 8644; fax: +81 550 87 5326.
E-mail address: ohtakeysh@chugai-pharm.co.jp (Y. Ohtake).
Figure 1. Structures of some known SGLT inhibitors.
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.04.053

4118
Y. Ohtake et al. / Bioorg. Med. Chem. 20 (2012) 4117–4127
R1
(9) followed by removal of the hydroxyl group at benzylic position
gave the aglycons, 10a–10d, 10f, 10h, 10j and 10l.
Ar
Ar
C-Aryl 5a-carba-b-D-glucopyranoside derivatives were synthe-
O
R2
X
R
3OH
sized according to the route outlined in Scheme 3. After lithiation
of aglycon 10, it was coupled with the intermediate 8 at À78 °C
to give two diastereomers at position 1 on cyclohexane ring. After
separation of the epimers by silicagel column chromatography, the
more polar epimers were deprotected and deoxygenated to afford
tetraols (5a, 5b, 5f, 5h, and 5j) under catalytic hydrogenation con-
dition using Pd(OH)2/C and hydrogen gas. Tetraol 5d was obtained
by stepwise conversion (method c). On the other hand, when the
less polar epimers were subjected to the same hydrogenation con-
dition, deoxygenation at the benzylic positions did not occur and
pentaols (5e, 5g, 5i, 5k, and 5m) were obtained. The less polar
compounds were successfully deoxygenated using Et3SiH with
BF3ÁOEt2 at À40 °C, followed by debenzylation, to give the tetraol
5l with the same configuration as that in the more polar com-
pounds by method b. The configurations at position 1 on cyclohex-
ane ring of tetraol and pentaol compounds were determined by
NMR analysis.9
Cyclopropyl derivative 5c was prepared by a different route due
to the difficulty of the hydrogenation at the final step (Scheme 4).
Pentahydroxy compound 12, which was obtained from 8 via pen-
tabenzyloxy 11, was protected with two acetonide groups to give
13. Trifluoromethanesulfonylation of 13 with N-(2-pyridyl)-bis(tri-
fluoro-methanesulfonimide) (PyNTf2), followed by Suzuki coupling
reaction, led to 15. Finally, acetonide deprotection under acidic
condition afforded 5c successfully.
OH
OH
OH
OH
O
+
HO
HO
HO
OH
OH
OH
OH
O
-aryl carbasugar
C
-aryl glucoside
like 3 and 4
5a-5m
-aryl carbasugar
like 1
C
Figure 2. Design of C-aryl carbasugar type of SGLT2 inhibitors.
inhibitors (max. hSGLT2 IC50 = 11 nM and 510-fold selectivity for
hSGLT2 versus hSGLT1); however, further improvement of potency
and selectivity was thought to be desirable for clinically effective
and safe drugs for the treatment of T2D.
Originally, many clinical trials using SGLT2 inhibitors began
with the O-glucopyranoside type. However, the C-glucopyranoside
type of SGLT2 inhibitors, as represented by dapagliflozin6 (3) and
canagliflozin7 (4), have currently superseded the O-types due to
their high potency and metabolic stability.8 According to a report
by Bristol–Myers Squibb, 3 actually appears to be 8-fold more po-
tent and have greater hSGLT2 selectivity than 2-A in their in vitro
assay.6a
Taking this situation into account, we applied the carbasugar
strategy into the C-glucopyranoside type as shown in Figure 2 in
an effort to enhance hSGLT2 potency and metabolic stability. Here
we wish to report a novel C-aryl carbasugar class with potent and
selective SGLT2 inhibition, describing the SAR of compounds and
providing some in vivo data.
2.2. In vitro data and SAR
All the synthesized compounds 5a–5m were evaluated for their
in vitro inhibition activity against human SGLT1 (hSGLT1) and
SGLT2 (hSGLT2). The results are shown in Table 1. Of the three
compounds 5a–5c with a substituent (X) at para position on the
distal benzene ring of the aglycon, 5c (X = cPr) showed the most
potent hSGLT2 inhibition and the highest hSGLT2 selectivity,
denoting the same tendency of O-linked carbasugar 1 which we re-
ported previously. The cyclopropyl derivative 5c, however, showed
high CYP3A4 time-dependent inhibition (TDI).10 Therefore, we set
the ethyl group as the para substituent for further examination.
Next, we investigated the substituents (R1 and R2) on the central
benzene ring (5d, 5f, 5h, 5j, and 5l). As a result, introducing the R1
2. Results and discussion
2.1. Chemistry
First, a key intermediate 8 was prepared from the correspond-
ing cyclohexanol 74 by oxidation using Dess–Martin periodinane
(Scheme 1).
Next, aglycons 10 were synthesized according to the general
method shown in Scheme 2. Addition of the corresponding
Grignard reagents or lithiated reagents, generated by nBuLi and
aryl bromides, to substituted-3-bromobenzaldehyde derivatives
OH
O
Several steps
OBn
OBn
OBn
OBn
O
HO
BnO
BnO
OH
OH
OBn
7
OBn
6 (D-glucal)
Scheme 1. Reagents and conditions: (a) Dess–Martin periodinane, CH2Cl2, rt, 3 h.
R1
Br
R1
Br
10a; X=OMe, R =H, R2=H
1
CHO
a, b
10b; X=Et, R1=H, R2=H
10c; X=OBn, R1=H, R2=H
10d; X=Et, R1=OMe, R2=H
10f; X=Et, R1=H, R2=OMe
10h; X=Et, R1=H, R2=OEt
10j; X=Et, R1=H, R2=Me
R2
X
R2
9
10l; X=Et, R1=OMe, R2=OMe
Scheme 2. Reagents and conditions: (a) ArMgBr or ArBr/nBuLi, THF, 0 °C; (b) Et3SiH, BF3ÁOEt2, CH2Cl2, 0 °C to rt.

Y. Ohtake et al. / Bioorg. Med. Chem. 20 (2012) 4117–4127
4119
R1
R1
R1
O
R2
X
OBn
OBn
X
R2
10
X
Br
R2
OH
OBn
OH
OBn
1
1
BnO
+
a
BnO
BnO
OBn
OBn
OBn
OBn
(moOreBnpolar)
8
(less polar)
b or c
b
d
5a; X=OMe, R =R2=H
1
5b; X=Et, R =R2=H
R1
1
R1
5d, 5e; X=Et, R =OMe, R =H
1
2
5f, 5g; X=Et, R =H, R =OMe
R2
X
OH
OH
1
2
R2
X
1
1
OH
OH
5h, 5i; X=Et, R =H, R =OEt
1
2
HO
HO
OH
5j, 5k; X=Et, R =H, R =Me
1
2
OH
(TeOtrHaol)
5l, 5m; X=Et, R =R2=OMe
1
(Pentaol)
5e, 5g, 5i, 5k, 5m
5a, 5b, 5d, 5f, 5h, 5j, 5l
Scheme 3. Reagents and conditions: (a) nBuLi, THF or Et2O, À78 °C then separation by silica gel; (b) Pd(OH)2/C, H2, MeOH/THF, rt; (c) Et3SiH, TFA, CH2Cl2, À10 °C to rt then
Pd(OH)2/C, H2, MeOH/THF, rt; (d) Et3SiH, BF3ÁEt2O, CH2Cl2, À40 °C to À10 °C then Pd(OH)2/C, H2, MeOH/THF, rt.
OH
OH
OBn
c
b
a
HO
OH
OH
O
O
OBn
OBn
8
HO
BnO
O
O
OH
OBn
(more polar)
13
11
12
OTf
e
f
d
O
O
O
OH
OH
O
HO
O
O
O
O
OH
5c
15
14
Scheme 4. Reagents and conditions: (a) 10c, nBuLi, THF, À78 °C, 0.5 h (62%); (b) Pd(OH)2, H2, MeOH/THF, rt, 40 h (63%); (c) Me2C(OMe)2, DMF, p-TsOH, 0 °C to rt, 0.5 h (59%);
(d) PyNTf2, CH2Cl2, DMAP, 0 °C to rt, 0.5 h (83%); (e) cPrB(OH)2, Pd(PPh3)4, NaBr, K3PO4, toluene–H2O, 100 °C, 6 h (33%); (f) 2 N HCl, 1,4-dioxane, 3 h (51%).
and/or R2 substituents showed significant effects on hSGLT2 inhibi-
tion and selectivity toward SGLT2 versus SGLT1 in most cases. Of
the tested C-aryl carbasugar compounds, 5d (R1 = OMe) exhibited
the most potent hSGLT2 inhibition activity (IC50 = 8.3 nM); how-
ever, it also inhibited hSGLT1 most strongly (IC50 = 510 nM), lead-
ing to a drastic decrease of the selectivity for hSGLT2 (61-fold). On
the other hand, introducing alkoxy groups as the R2 substituent
turned out to be a very effective way of both enhancing hSGLT2
inhibition and lowering hSGLT1 inhibition activity. Methoxy and
ethoxy compounds (5f and 5h) showed improved hSGLT2 potency
and very low hSGLT1 inhibition compared to non-substituted 5b.
Double methoxy introduction to R1 and R2 (5l) also supported
the idea that the R2-methoxy substitution effectively diminishes
the hSGLT1 inhibition (versus 5d). Meanwhile, methyl introduction
(5j) had little effect on both hSGLT1 and hSGLT2 inhibition (com-
pared with 5b).
that of 5d, whereas the hSGLT2 inhibition was 10-fold lower. This
result suggested that the introducing hydrogen-bonding or a
hydrophilic group in R3 would be unfavorable for hSGLT1 and
hSGLT2 inhibition activity. A similar tendency was observed in
5k. In contrast, three compounds with an alkoxy group at R2 posi-
tion (5g, 5i and 5m) showed maintained hSGLT2 inhibition and
moderate to high hSGLT2 selectivity in comparison with the corre-
sponding non-hydroxy compounds (5f, 5h and 5l). This difference
implies the formation of an intramolecular hydrogen bond be-
tween the oxygen atom of the alkoxy group and the hydrogen atom
of the benzylic hydroxyl group, suggesting that the hydrogen-
bonding donating effect has disappeared.
Because comparing 3D structures of known inhibitors with our
compounds would be very helpful for us to understand the SAR pro-
files, we constructed a superimposed model (Fig. 3). Several known
SGLT2 inhibitors (phlorizin,11 2-A, 3, 4, ipragliflozin,12 luseogliflo-
zin13) and our compounds (1, 5h and 5i) were subjected to superpo-
sition with the Flexible Alignment module in Molecular Operating
Environment (MOE).14 This superimposed model showed that all
the inhibitors totally overlapped each other, indicating that they
Another interesting finding was the effect of the hydroxyl group
at benzylic position (R3 = OH). On the whole, introducing of a hy-
droxyl group at R3 position effectively reduced of hSGLT1 inhibi-
tion activity. The hSGLT1 inhibition of 5e was 60-fold lower than

4120
Y. Ohtake et al. / Bioorg. Med. Chem. 20 (2012) 4117–4127
Table 1
In vitro data for hSGLT inhibition activity and selectivity
R1
R2
X
R
3 OH
HO
OH
OH
5a-5m
Compounds
X
R1
R2
R3
SGLT IC50 (nM)a
hSGLT2 hSGLT1
Selectivityb
IC50 of CYP3A4 inhibition (lM)
Ratiod
TDIe
c
c
IC50 (À)
IC50 (+)
Phlorizin
2
2-A
3
16
190
22,000
2300
800
1900
48,000
33,000
27,000
510
31,000
73,000
>100,000
>100,000
>100,000
25000
>100,000
4500
12
85
480
620
280
660
670
2,500
61
ND
ND
ND
ND
ND
>50
>50
76
ND
ND
>50
ND
36.6
ND
ND
ND
25.7
29.0
ND
ND
ND
ND
ND
>50
>50
33
ND
ND
>50
ND
>50
ND
ND
ND
40.3
>50
ND
ND
ND
ND
ND
n.c.
n.c.
2.3
ND
ND
n.c.
ND
n.c.
ND
ND
ND
0.6
n.c.
ND
ND
ND
ND
ND
n.j.
n.j.
(+)
ND
ND
n.j.
ND
n.j.
ND
ND
ND
(À)
n.j.
260
4.8
1.3
6.7
73
49
11
8.3
100
12
4
5a
5b
5c
5d
5e
5f
5g
5h
5i
OMe
Et
cPr
Et
Et
Et
Et
Et
Et
Et
H
H
H
OMe
OMe
H
H
H
H
H
H
OMe
OMe
H
H
H
H
H
H
H
H
OH
H
OH
H
OH
H
OH
H
H
310
OMe
OMe
OEt
OEt
Me
Me
OMe
OMe
6,100
>7,100
>5,300
>10,000
480
>590
530
860
14
19
9.9
52
170
8.5
10
5j
5k
5l
Et
Et
Et
5m
OH
8600
ND = no data; n.c. = not calculated; n.j. = not judged
a
Phlorizin was always included in the assays as reference standards.
The selectivity values were calculated by IC50 hSGLT1/IC50 hSGLT2.
IC50 (À) and IC50 (+) represent the IC50 values in which the inhibitor was added to the reaction mixture simultaneously with BFC (coincubation assay) and before the
b
c
addition of BFC (preincubation assay), respectively.
d
The ratio were calculated by IC50 (À)/IC50 (+).
e
TDIs were judged from the literature.10 (+) = high risk (P2); (À) = no risk (<1.3)
construct a common pharmacophore. Also, it is supposed that the
central benzene ring of aglycon is aligned in a perpendicular direc-
tion to the sugar moiety and each central benzene ring is less over-
lapped, whereas each distal benzene ring is relatively well-
overlapped. From this model, it was found that only the ethoxy
group (R2) of our compounds was located in a region that could
not be occupied by other inhibitors. This finding suggests that some
sort of negative interactions between the ethoxy moiety and the
hSGLT1 protein may occur in this region. In addition, the oxygen
atom of the alkoxy group (R2) is so close to the hydrogen atom of
the hydroxyl group (R3) as to form an effective intramolecular
hydrogen bond, supporting our consideration mentioned above.
SGLT2 inhibitor 5i showed a little shorter and weaker in vivo activ-
ity than the less potent 5h is not clear,15 but may be the metabolic
liability attributable to the benzylic alcohol moiety.
Obtaining a positive result in in vivo pharmacological study, we
conducted a pharmacokinetic (PK) study of 5h in db/db mice at
oral single dosage of 10 mg/kg and compared the result with those
of 1 and 2 (Table 2). Outstandingly, the C-aryl carbasugar com-
pound 5h showed an apparent lower clearance (CL/F) and higher
AUCinf than 1 and 2. These results suggested that 5h may become
metabolically more stable than 1 and 2 by being converted from
the O-linked type to the C-linked type, being interpreted as mean-
ing that 5h has more potent and longer in vivo efficacy action than
1 and 2.
2.3. In vivo and PK data
3. Conclusion
We selected paired ethoxy compounds (5h and 5i) with high
hSGLT2 selectivity and evaluated their blood glucose-lowering ef-
fect in db/db mice at oral administration of 30 mg/kg to compare
with 2 (Fig. 4A). For 4 h after dosing, all the three compounds low-
ered blood glucose almost equivalently, but this changed after 4 h.
That is, while the glucose-lowering effect of 2 diminished rapidly
and that of 5i gradually, that of 5h was maintained for almost
8 h. In addition, compared with the reduction rate of blood glucose
AUC0–8 h of 1 at dosing of 100 mg/kg (À41.6 3.5%), those of 5h
and 5i at dosing of 30 mg/kg were almost the same (À42.6 1.3%
and À39.1 1.3%, respectively) (Fig. 4B). These results indicate that
the blood glucose-lowering effects of 5h and 5i were threefold
stronger than that of 1. The reason why the more potent in vitro
In the course of our research on the carbasugar class of SGLT2
inhibitors, we successfully synthesized C-aryl derivatives. Several
C-linked compounds showed more potent in vitro hSGLT2 inhibition
and higher hSGLT2 selectivity than those of O-linked compound 1.
Using the created superimposition model of some known SGLT2
inhibitors, the reason for the high hSGLT2 selectivity, especially
the diminished hSGLT1 inhibition, was speculated to be that addi-
tional substituents occupied a new space, in a different way than
known inhibitors. Of the tested compounds, highly SGLT2-selective
5h was found to have a more potent and longer lasting blood glu-
cose-lowering effect in db/db mice than 1 and 2, which might be
attributed to its stronger SGLT2 inhibition activity and the better

Y. Ohtake et al. / Bioorg. Med. Chem. 20 (2012) 4117–4127
4121
Figure 3. Superimposed model of SGLT2 inhibitors. Compound 5i is depicted in a
thicker stick model (carbon atoms in blue). Hydrogen atoms are not displayed for
clarity. See Experimental section for further explanation.
PK profiles than those of 1 and 2. These results demonstrate that 5h
might be a promising drug candidate for the treatment of T2D.
Figure 4. Blood glucose-lowering effect of single oral administration in db/db mice;
(A) change in blood glucose levels of 2, 5h, and 5i; (B) Reduction rate of blood
4. Experimental
glucose AUC0–8 in db/db mice (1, 2, 5h, and 5i) versus vehicle control. Data are
h
expressed as means SE (n = 6). ⁄Based on our previous data.4b
4.1. Chemistry: instruments
Silica gel 60F254 precoated plates on glass from Merck KgaA
were used for thin layer chromatography (TLC) and preparative
TLC (PTLC). Column chromatography was carried out on Merck Sil-
ica Gel 60 (230–400 mesh) or Shoko Scientific Purif-Pack SI
Table 2
Pharmacokinetic parameters of 1, 2, and 5h in db/db mice
Compounds
1
2
5h
Dose (mg/kg; po)
T1/2 (h)
Tmax (h)
10
10
10
(60 lm), if not otherwise specified. Melting points (mp) were
2.68
0.25
1.64
3.07
12,586
3257
0.82
0.25
3.36
2.97
3315
2818
1.57
0.25
3.68
6.24
3629
1602
determined with a Yanagimoto micro melting point apparatus.
1H and 13C NMR spectra were recorded on JEOL EX-270
(270 MHz), Varian Mercury300 (300 MHz) or JEOL JNM-ECP-400
(400 MHz), and chemical shifts were reported in parts per million
(d) downfield from tetramethylsilane as an internal standard. The
peak patterns are shown as the following abbreviations: br, broad;
s, singlet; d, doublet; t, triplet; q, quartet; and m, multiplet. Mass
spectra (MS) were measured by a Thermo Electron LCQ Classic or
a Micromass ZQ of Waters (ESI). High resolution mass spectra
(HRMS) were recorded by a Micromass Q-Tof Ultima API mass
spectrometer. All reagents and the solvent were commercially
available unless otherwise indicated.
Cmax (lg/mL)
AUCinf (lgÁh/mL)
Vz/Fa (mL/kg)
CL/F (mL/hr/kg)
a
The apparent volume of distribution during terminal phase at oral
administration.
J = 14.3, 3.9 Hz), 2.64 (1H, dd, J = 14.3, 13.7 Hz), 3.39 (1H, dd,
J = 9.0, 2.5 Hz), 3.71 (1H, dd, J = 9.6, 9.3 Hz), 3.77 (1H, dd, J = 9.2,
3.5 Hz), 3.90 (1H, dd, J = 10.8, 9.0 Hz), 4.13 (1H, d, J = 9.3 Hz), 4.43
(2H, s), 4.55 (2H, dd, J = 11.7, 2.9 Hz), 4.78 (1H, d, J = 10.8 Hz),
4.91–4.97 (3H, m), 7.18–7.21 (2H, m), 7.26–7.40 (18H, m).
4.1.1. (2R,3S,4R,5R)-2,3,4-Tris(benzyloxy)-5-[(benzyloxy)
methyl]cyclohexanone (8)
4.1.2. 1-Bromo-3-(4-methoxylbenzyl)benzene (10a)
To a solution of (1S,2S,3S,4R,5R)-2,3,4-tris(benzyloxy)-5-[(ben-
4.1.2.1. (3-Bromophenyl)(4-methoxyphenyl)methanol.
In a
zyloxy)methyl]cyclohexanol
7
(236 mg, 0.44 mmol) in CH2Cl2
nitrogen stream, to a solution of 3-bromo-benzaldehyde (423 mg,
2.29 mmol) in THF (10 mL) was added 0.5 M methoxylphenylmag-
nesium bromide in THF solution (6.86 mL, 3.43 mmol) dropwise at
0 °C. The reaction mixture was stirred at the same temperature for
10 min and warmed up to rt. After being stirred for 30 min at rt, the
mixture was stopped by addition of satd. NH4Cl aq solution. The
obtained mixture was extracted with AcOEt, and the extracts were
(1.2 mL) was added Dess–Martin periodinane (280 mg, 0.66 mmol)
portionwise at room temperature (rt) and the mixture solution was
stirred for 3 h. The solvent was concentrated and the obtained resi-
due was purified by silica gel chromatography (AcOEt/hexane = 1:4)
to obtain the title compound (201 mg, 85%) as a white solid. mp
89–91 °C; 1H NMR (CDCl3) d: 1.84–1.94 (1H, m), 2.41 (1H, dd,

4122
Y. Ohtake et al. / Bioorg. Med. Chem. 20 (2012) 4117–4127
washed with water and then dried over MgSO4. The organic sol-
vent was concentrated, and the residue was purified by silica gel
column chromatography (AcOEt/hexane = 1:9 to 1:3) to obtain
the title compound (905 mg, quant) as a colorless oil. 1H NMR
(CDCl3) d: 2.75 (1H, br s), 3.80 (3H, s), 5.76 (1H, s), 6.86–6.90
(2H, m), 7.17–7.21 (1H, m), 7.24–7.29 (3H, m), 7.37–7.40 (1H,
m), 7.55–7.56 (1H, m).
yield was 86%. Colorless oil. 1H NMR (CDCl3) d: 1.22 (3H, t,
J = 7.5 Hz), 2.61 (2H, q, J = 7.5 Hz), 3.85 (5H, s), 6.79 (1H, d,
J = 8.4 Hz), 7.05–7.24 (5H, m), 7.35–7.37 (1H, m).
4.1.7. 2-Bromo-1-ethoxy-4-(4-ethylbenzyl)benzene (10h)
This compound was prepared from 3-bromo-4-ethoxybenzal-
dehyde and 0.5 M 4-ethylphenylmagnesium bromide in THF solu-
tion in the same manner as described in Section 4.1.2. The total
yield was 88%. Colorless oil. 1H NMR (CDCl3) d: 1.22 (3H, t,
J = 7.6 Hz), 1.45 (3H, t, J = 7.0 Hz), 2.62 (2H, q, J = 7.6 Hz), 3.85
(2H, s), 4.06 (2H, q, J = 7.0 Hz), 6.79 (1H, d, J = 8.4 Hz), 7.00–7.15
(5H, m), 7.36 (1H, d, J = 2.1 Hz).
4.1.2.2. 1-Bromo-3-(4-methoxylbenzyl)benzene.
To a solu-
tion of obtained (3-bromophenyl)(4-methoxyphenyl)methanol
(670 mg, 2.29 mmol) in CH2Cl2 (10 mL) were added Et3SiH
(0.73 mL, 4.57 mmol) and BF3ÁOEt2 (0.32 mL, 2.51 mmol) at 0 °C
and the mixture was stirred for 30 min. Water (4 mL) was added
to the reaction mixture and the resulting mixture was extracted
with AcOEt. The extracted organic layer was washed with brine,
and then dried over MgSO4. The organic solvent was concentrated,
and the residue was purified by silica gel column chromatography
(AcOEt/hexane = 1:9) to obtain the title compound (586 mg, 92%)
as a colorless oil. 1H NMR (CDCl3) d: 3.78 (3H, s), 3.88 (2H, s),
6.83–6.86 (2H, m), 7.07–7.16 (4H, m), 7.31–7.33 (2H, m).
4.1.8. 2-Bromo-1-methyl-4-(4-ethylbenzyl)benzene (10j)
This compound was prepared from 5-bromo-2,4-dimethoxy-
benzaldehyde and 0.5 M 4-ethylphenyl-magnesium bromide in
THF solution in the same manner as described in Section 4.1.2.
The total yield was 54%. Colorless oil. 1H NMR (CDCl3) d: 1.22
(3H, t, J = 7.6 Hz), 2.35 (3H, s), 2.61 (2H, q, J = 7.6 Hz), 3.87 (2H,
s), 7.01 (1H, dd, J = 8.0, 1.7 Hz), 7.06–7.14 (5H, m), 7.36 (1H, d,
J = 1.7 Hz).
4.1.3. 1-Bromo-3-(4-ethylbenzyl)benzene (10b)
This compound was prepared from 3-bromobenzaldehyde and
0.5 M 4-ethylphenylmagnesium bromide in THF solution in the
same manner as described in Section 4.1.2. The total yield was
63%. Colorless oil. 1H NMR (CDCl3) d: 1.22 (3H, t, J = 7.6 Hz), 2.64
(2H, q, J = 7.6 Hz), 3.90 (2H, s), 7.07–7.16 (6H, m), 7.31–7.33 (2H, m).
4.1.9. 1-Bromo-5-(4-ethylbenzyl)-2,4-dimethoxybenzene (10l)
This compound was prepared from 5-bromo-2,4-dimethoxy-
benzaldehyde and 0.5 M 4-ethylphenyl-magnesium bromide in
THF solution in the same manner as described in Section 4.1.2.
The total yield was 68%. White solid. mp 48–49 °C; 1H NMR (CDCl3)
d: 1.22 (3H, t, J = 7.6 Hz), 2.61 (2H, q, J = 7.6 Hz), 3.82 (3H, s), 3.83
(2H, s), 3.89 (3H, s), 6.47 (1H, s), 7.10 (4H, s), 7.20 (1H, s).
4.1.4. 1-Bromo-3-(4-benzyloxybenzyl)benzene (10c)
In a nitrogen stream, to a solution of 1-benzyloxy-4-bromoben-
zene (2.0 g, 7.6 mmol) in THF (70 mL) was added 2.7 M nBuLi in
hexane solution (3.1 mL, 8.36 mmol) dropwise at À78 °C and the
mixture solution was stirred for 30 min. Then, a solution of 3-bro-
mobenzaldehyde (1.17 g, 6.31 mmol) in THF (30 mL) was added
dropwise thereto and the reaction mixture was stirred at À78 °C
for 1 h. The mixture was stopped by addition of satd. NH4Cl aq solu-
tion. The mixture was extracted with AcOEt, and the organic layer
was washed with brine, and then dried over anhydrous MgSO4.
The solvent was concentrated, and the residue was purified by silica
gel chromatography (AcOEt/hexane = 1:8) to obtain (3-bromophe-
nyl)(4-benzyloxyphenyl)methanol (2.26 g, 97%) as a colorless oil.
To a solution of obtained (3-bromophenyl)(4-benzyloxyphe-
nyl)methanol (1.0 g, 2.70 mmol) in CH2Cl2 (3 mL) and MeCN
(3 mL) were added Et3SiH (0.52 mL, 3.25 mmol) and BF3ÁOEt2
(0.36 mL, 2.83 mmol) at À40 °C and the mixture was stirred for
30 min. Water (4 mL) was added to the reaction mixture and the
resulting mixture was extracted with AcOEt. The extracted organic
layer was washed with brine, and then dried over anhydrous
MgSO4. The organic solvent was concentrated, and the residue
was purified by silica gel column chromatography (AcOEt/ hex-
ane = 1:20) to obtain the title compound (863 mg, 90%) as a color-
less oil. 1H NMR (CDCl3) d: 3.87 (2H, s), 5.03 (2H, s), 6.88–6.92 (2H,
m), 7.05–7.15 (5H, m), 7.29–7.43 (6H, m).
4.1.10. [1S,2R,3R,4R,6S]-4-Hydroxymethyl-6-[3-(4-methoxy-
benzyl)phenyl]cyclohexane-1,2,3-triol (5a)
4.1.10.1. [1R,2R,3S,4R,5R]- and [1S,2R,3S,4R,5R]-2,3,4-trisbenzyl-
oxy-5-benzyloxymethyl-1-[3-(4-methoxybenzyl)phenyl] cyclo-
hexanol.
To a solution of 3-(4-methoxybenzyl)-1-bromo-
benzene 10a (77 mg, 0.28 mmol) in THF (0.5 mL) was added 2.5 M
nBuLi in hexane solution (0.10 mL, 0.27 mmol) dropwise at À78 °C
and the mixture solution was stirred for 15 min. Then, a solution
of 8 (100 mg, 0.186 mmol) in THF (0.50 mL) was added dropwise
thereto and the reaction mixture was stirred for 1 h. To the reaction
mixture was added satd. NH4Cl aq solution and the mixture was
extracted with AcOEt, and the organic layer was washed with brine,
and then dried over anhydrous MgSO4. The solvent was concen-
trated, and the obtained residue was purified by silica gel chroma-
tography (AcOEt/hexane = 2:5) to obtain a less polar compound
(53 mg, 39%) and a more polar compound (68 mg, 50%).
(Less polar compound): Rf 0.35 (AcOEt/hexane = 1:4); 1H NMR
(CDCl3) d: 1.88 (1H, dd, J = 14.5, 3.9 Hz), 1.97–2.02 (1H, m), 2.26–
2.34 (1H, m), 2.98 (1H, br s), 3.39–3.43 (1H, m), 3.70–3.96 (10H,
m), 4.42–4.46 (3H, m), 4.62 (1H, d, J = 10.4 Hz), 4.83 (1H, d,
J = 10.8 Hz), 4.91 (1H, d, J = 10.7 Hz), 4.93 (1H, d, J = 10.7 Hz),
6.74–6.77 (3H, m), 7.04–7.21 (7H, m), 7.21–7.37 (18H, m).
(More polar compound): Rf 0.30 (AcOEt/hexane = 1:4); 1H NMR
(CDCl3) d: 1.40–1.54 (2H, m), 1.84–1.91 (1H, m), 2.42 (1H, dd,
J = 14.1, 3.7 Hz), 2.54 (1H, s), 3.37 (1H, dd, J = 9.0, 2.5 Hz), 3.58
(1H, dd, J = 9.0, 4.7 Hz), 3.64–3.69 (4H, m), 3.74 (1H, d,
J = 10.4 Hz), 3.90–3.97 (3H, m), 4.42 (2H, s), 4.51 (1H, d,
J = 10.8 Hz), 4.73 (2H, d, J = 11.0 Hz), 4.81 (1H, d, J = 11.0 Hz), 4.86
(1H, d, J = 10.8 Hz), 5.02 (1H, d, J = 11.7 Hz), 6.73–6.77 (2H, m),
7.06–7.09 (3H, m), 7.16–7.20 (2H, m), 7.22–7.35 (18H, m), 7.55–
7.58 (1H, m), 7.65–7.67 (1H, m).
4.1.5. 1-Bromo-4-methoxy-3-(4-ethylbenzyl)benzene (10d)
This compound was prepared from 5-bromo-2-methoxybenzal-
dehyde and 0.5 M 4-ethylphenylmagnesium bromide in THF solu-
tion in the same manner as described in Section 4.1.2. The total
yield was 99%. Colorless oil. 1H NMR (CDCl3) d: 1.22 (3H, t,
J = 7.8 Hz), 2.61 (2H, q, J = 7.8 Hz), 3.79 (3H, s), 3.88 (2H, s), 6.71
(1H, d, J = 8.7 Hz), 7.10–7.15 (5H, m), 7.26 (1H, dd, J = 8.7, 2.3 Hz).
4.1.6. 2-Bromo-1-methoxy-4-(4-ethylbenzyl)benzene (10f)
This compound was prepared from 3-bromo-4-methoxybenzal-
dehyde and 0.5 M 4-ethylphenylmagnesium bromide in THF solu-
tion in the same manner as described in Section 4.1.2. The total
4.1.10.2. [1S,2R,3R,4R,6S]-4-Hydroxymethyl-6-[3-(4-methoxy-
benzyl)phenyl]cyclohexane-1,2,3-triol.
To a solution of the
obtained more polar compound (50 mg, 0.068 mmol) in MeOH
(5.0 mL)-THF (1.0 mL) was added 20% Pd(OH)2 (27 mg) and the

Y. Ohtake et al. / Bioorg. Med. Chem. 20 (2012) 4117–4127
4123
mixture solution was stirred under hydrogen atmosphere for 3 h.
The reaction mixture was filtered and the filtrate was concen-
trated. The obtained residue was purified by PTLC (MeOH/
CH2Cl2 = 1:9) to obtain the title compound (11 mg, 45%) as a color-
less amorphous. 1H NMR (CD3OD) d: 1.44–1.50 (1H, m), 1.65–1.75
(1H, m), 1.85 (1H, dt, J = 3.7, 13.7 Hz), 2.58–2.64 (1H, m), 3.34–3.40
(2H, m), 3.51–3.57 (1H, m), 3.62 (1H, dd, J = 10.6, 6.5 Hz), 3.79 (3H,
s), 3.80 (1H, dd, J = 10.4, 4.1 Hz), 3.92 (2H, s), 6.83–6.87 (2H, m),
7.04–7.06 (1H, m), 7.10–7.16 (4H, m), 7.22–7.27 (1H, m); 13C
NMR (CD3OD) d: 159.4, 144.5, 143.2, 134.8, 130.8, 129.5, 129.4,
128.0, 126.5, 114.8, 81.2, 77.4, 75.1, 64.4, 55.6, 50.0, 45.4, 42.0,
34.5; MS (ESI) m/z: 376 ([M+H2O]+); HRMS calcd for C21H25O5
([MÀH]À) 357.1697. Found 357.1699.
4.1.12.3. 4-{3-[(3aS,4S,5aR,9aR,9bR)-2,2,8,8-Tetramethyl-hexa-
hydro[1,3]dioxolo[4’,5’:3,4]benzo-[1,2-d][1,3]dioxin-4-
yl]benzyl}phenol (13).
To a solution of the above obtained
compound (47 mg, 0.136 mmol) in DMF (1 mL) were added 2,2-
dimethoxypropane (142 mg, 1.36 mmol) and p-toluenesulfonic
acid hydrate (2 mg) at 0 °C. The mixture was stirred at rt for
30 min, and the reaction was stopped by addition of satd. NaHCO3
aq solution. The mixture was extracted with AcOEt, and the organic
layer was washed with water and dried over anhydrous MgSO4.
The solvent was concentrated, and the obtained residue was puri-
fied by PTLC (AcOEt/hexane = 1:4) to obtain the title compound
(34 mg, 59%). 1H NMR (CDCl3) d: 1.11–1.24 (1H, m), 1.42 (6H, s),
1.46 (3H, s), 1.53 (3H, s), 1.73 (1H, dt, J = 13.7, 3.9 Hz), 1.80–1.93
(1H, m), 2.94 (1H, dt, J = 11.0, 4.0 Hz), 3.59–3.92 (7H, m), 4.85
(1H, s), 6.71–6.76 (2H, m), 7.00–7.08 (5H, m), 7.19–7.23 (1H, m) .
4.1.11. (1R,2R,3S,4S,6R)-4-[3-(4-Ethylbenzyl)-phenyl]-6-(hydro-
xymethyl)cyclohexane-1,2,3-triol (5b)
This compound was prepared from 8 and 10b in the same
manner as described in Section 4.1.10. The total yield was 44%.
Colorless amorphous. 1H NMR (CD3OD) d: 1.24 (3H, t, J = 7.5 Hz),
1.43–1.53 (1H, m), 1.61–1.78 (1H, m), 1.85 (1H, dt, J = 3.7,
13.4 Hz), 2.56–2.68 (3H, m), 3.29–3.40 (2H, m), 3.50–3.66 (2H,
m), 3.80 (1H, dd, J = 10.9, 3.9 Hz), 3.94 (2H, s), 7.04–7.27 (8H, m);
13C NMR (CD3OD) d: 144.5, 143.0, 142.9, 139.9, 129.9, 129.6,
129.4, 128.8, 128.0, 126.5, 81.2, 77.4, 75.1, 64.4, 50.0, 45.4, 42.5,
34.5, 29.4, 16.3; MS (ESI) m/z: 357 ([M+H]+); HRMS calcd for
4.1.12.4. 4-{3-[(3aS,4S,5aR,9aR,9bR)-2,2,8,8-Tetramethylhexahy-
dro-3aH-[1,3]dioxolo[4’,5’:5,6]benzo [1,2-d][1,3]dioxin-4-yl]ben-
zyl}phenyl trifluoromethane sulfonate (14).
To a solution of
the obtained compound (34 mg, 0.08 mmol) in CH2Cl2 (0.8 mL)
were added pyridine (15 mg, 0.19 mmol), 2-[N,N-bis(trifluorome-
thane-sulfonyl) amino]pyridine (34 mg, 0.096 mmol), and DMAP
(1 mg) at 0 °C. The reaction mixture was stirred at rt for 30 min,
and the reaction was stopped by addition of water. The mixture
was extracted with CH2Cl2 and the organic layer was washed with
water and dried over anhydrous MgSO4. The solvent was concen-
trated, and the obtained residue was purified by PTLC (AcOEt/hex-
ane = 1:4) to obtain the title compound (37 mg, 83%). 1H NMR
(CDCl3) d: 1.11–1.24 (1H, m), 1.42 (6H, s), 1.43 (3H, s), 1.53 (3H,
s), 1.74 (1H, dt, J = 13.7, 3.9 Hz), 1.80–1.93 (1H, m), 2.96 (1H, dt,
J = 11.1, 3.6 Hz), 3.59–3.92 (5H, m), 3.98 (2H, s), 7.00–7.03 (2H,
m), 7.10–7.35 (6H, m).
C
22H27O4 ([MÀH]À) 355.1904. Found 355.1908.
4.1.12. (1R,2R,3S,4S,6R)-4-[3-(4-Cyclopropyl-benzyl)phenyl-6-
(hydroxymethyl)cyclohexane-1,2,3-triol (5c)
4.1.12.1. (1R,2R,3S,4R,5R)- and (1S,2R,3S,4R,5R)-2,3,4-Tris(ben-
zyloxy)-1-[3-(4-(benzyloxy)benzyl)phenyl-5-(benzyl-
oxy)methyl] cyclohexanol (11).
solution of 1-bromo-3-(4-benzyloxybenzyl)benzene
In a nitrogen stream, to a
(10c)
(753 mg, 2.1 mmol) in THF (10 mL) was added 2.5 M nBuLi in hex-
ane solution (0.8 mL, 2.0 mmol) dropwise at À78 °C and the reac-
tion mixture was stirred at the same temperature for 10 min. To
this solution was added a solution of 8 (715 mg, 1.33 mmol) in
THF (2 mL) dropwise at À78 °C and the resulting solution was stir-
red for 30 min. The reaction was stopped by addition of satd. NH4Cl
aq solution. The obtained solution was extracted with AcOEt, and
the organic layer was washed with water and then dried over
anhydrous MgSO4. The solvent was concentrated, and the obtained
residue was purified by silica gel column chromatography (AcOEt/
hexane = 1:5 to 1:1) to obtain a less polar compound (295 mg, 27%)
and a more polar compound (670 mg, 62%).
4.1.12.5. (3aS,4S,5aR,9aR,9bR)-4-[3-(4-Cyclopropylbenzyl)phe-
nyl]-2,2,8,8-tetramethylhexahydro-3aH-[1,3]dioxolo[4’,5’:5,6]
benzo [1,2-d][1,3]dioxine (15).
In a nitrogen stream, to a
solution of the obtained compound (38 mg, 0.068 mmol) in toluene
(0.5 mL)-water (0.017 mL) were added K3PO4 (65 mg, 0.31 mmol),
NaBr (7 mg, 0.068 mmol), cyclopropylboronic acid (9 mg,
0.10 mmol) and Pd(PPh3)4 (8.0 mg, 0.007 mmol). The reaction mix-
ture was heated at 100 °C for 6 h. The reaction mixture was cooled
to 0 °C and water was added thereto. The resulting solution was
extracted with AcOEt, and the organic layer was washed with brine
and dried over anhydrous MgSO4. The solvent was concentrated,
and the obtained residue was purified by PTLC (AcOEt/hex-
ane = 1:4) to obtain the title compound (10 mg, 33%). 1H NMR
(CDCl3) d: 0.62–0.68 (2H, m), 0.89–0.96 (2H, m), 1.10–1.24 (1H,
m), 1.42 (3H, s), 1.43 (3H, s), 1.46 (3H, s), 1.54 (3H, s), 1.74 (1H,
dt, J = 13.7, 3.9 Hz), 1.80–1.93 (2H, m), 2.94 (1H, dt, J = 10.9,
3.6 Hz), 3.59–3.96 (8H, m), 6.96–7.07 (7H, m), 7.16–7.25 (1H, m).
(More polar): Rf 0.28 (AcOEt/hexane = 1:4); 1H NMR (CDCl3) d:
1.40–1.60 (2H, m), 1.82–1.92 (1H, m), 2.39 (1H, dd, J = 13.8,
3.6 Hz), 2.56 (1H, s), 3.36 (1H, dd, J = 8.9, 2.5 Hz), 3.54–3.76 (3H,
m), 3.90 (2H, s), 3.92–3.98 (1H, m), 4.41 (2H, s), 4.49 (1H, d,
J = 13.4 Hz), 4.70–4.90 (6H, m), 5.02 (1H, d, J = 11.7 Hz), 6.81–
6.86 (2H, m), 7.06 (2H, d, J = 8.6 Hz), 7.15–7.38 (27H, m), 7.54–
7.66 (2H, m).
4.1.12.6. (1R,2R,3S,4S,6R)-4-[3-(4-Cyclopropylbenzyl) phenyl]-6-
(hydroxymethyl) -cyclohexane-1,2,3-triol (5c).
To a solution
4.1.12.2. (1R,2R,3S,4S,6R)-4-[3-(4-Hydroxybenzyl)phenyl-6-(hyd-
of the obtained compound (10 mg, 0.022 mmol) in 1,4-dioxane
(0.2 mL) was added 2 N HCl aq solution (0.2 mL) at 0 °C dropwise.
The reaction mixture was stirred for 3 h, and then sat. NaHCO3 aq
solution was added thereto. The resulting solution was extracted
with CH2Cl2 and the organic layer was washed with brine and
dried over anhydrous MgSO4. The solvent was concentrated, and
the obtained residue was purified by PTLC (MeOH/CH2Cl2 = 1:10)
to obtain the title compound (4.5 mg, 51%) as a colorless amor-
phous. 1H NMR (CD3OD) d: 0.63–0.68 (2H, m), 0.91–0.98 (2H, m),
1.42–1.52 (1H, m), 1.62–1.78 (1H, m), 1.80–1.94 (2H, m), 2.54–
2.66 (1H, m), 3.33–3.40 (2H, m), 3.50–3.68 (2H, m), 3.79 (1H, dd,
J = 10.9, 3.9 Hz), 3.93 (2H, s), 6.85–7.16 (7H, m), 7.21–7.26 (1H,
roxymethyl)cyclohexane-1,2,3-triol (12).
To a solution of
the obtained more polar compound (659 mg, 0.81 mmol) in THF
(2 mL) and MeOH (10 mL) was added 20% Pd(OH)2 (200 mg) and
the mixture was stirred under hydrogen atmosphere for 64 h. After
filtering the catalyst, the filtrate was concentrated under reduced
pressure. The obtained residue was purified by silica gel column
chromatography (MeOH/CH2Cl2 = 1:10) to obtain the title com-
pound (176 mg, 63%) as a colorless oil. 1H NMR (CD3OD) d: 1.47
(1H, t, J = 13.2 Hz), 1.61–1.78 (1H, m), 1.85 (1H, dt, J = 13.5,
3.7 Hz), 2.52–2.68 (1H, m), 3.32–3.40 (2H, m), 3.50–3.66 (2H, m),
3.76–3.84 (1H, m), 3.88 (2H, s), 6.68–6.74 (2H, m), 7.01- 7.32
(6H, m).

4124
Y. Ohtake et al. / Bioorg. Med. Chem. 20 (2012) 4117–4127
m); MS (ESI) m/z: 386 ([M+H2O]+); HRMS calcd for C23H27O4
([MÀH]À) 367.1904. Found 367.1907.
3.77 (1H, d, J = 4.0 Hz), 3.82 (3H, s), 3.92 (2H, s), 6.91 (1H, d,
J = 8.2 Hz), 7.02–7.14 (6H, m); MS (ESI) m/z: 404 ([M+H2O]+); HRMS
calcd for C23H29O5 ([MÀH]À) 385.2010. Found 385.2011.
4.1.13. (1R,2R,3S,4S,6R)-4-[3-(4-Ethylbenzyl)-4-methoxyphenyl]
-6-(hydroxymethyl)cyclohexane-1,2,3-triol (5d)
4.1.13.4. (1R,2R,3S,4R,5R)-1-[3-(4-Ethylbenzyl)-4-methoxyphenyl]
4.1.13.1. (1R,2R,3S,4R,5R)- and (1S,2R,3S,4R,5R)-2,3,4-Tris(ben-
zyloxy)-5-[(benzyloxy)methyl]-1-[3-(4-ethylbenzyl)-4-methoxy-
-5-(hydroxymethyl)cyclohexane-1,2,3,4-tetraol (5e).
To a
solution of the less polar compound which was obtained at 1st step
as described in Section 4.1.13 (50 mg, 0.067 mmol) in THF (0.5 mL)
and MeOH (2 mL) was added 20% Pd(OH)2 (10 mg) and the mixture
was stirred under hydrogen atmosphere at room temperature for
15 h. After filtering the catalyst, the filtrate was concentrated un-
der reduced pressure. The obtained residue was purified by PTLC
(MeOH/CH2Cl2 = 1:10) to obtain the title compound (20 mg, 74%)
as a colorless amorphous. 1H NMR (CD3OD) d: 1.22 (3H, t,
J = 7.6 Hz), 1.65–1.84 (2H, m), 2.01–2.14 (1H, m), 2.60 (2H, q,
J = 7.6 Hz), 3.39–3.46 (1H, m), 3.62–3.75 (4H, m), 3.82 (3H, s),
3.94 (2H, s), 6.94 (1H, d, J = 8.4 Hz), 7.06 (2H, d, J = 8.2 Hz), 7.13
(2H, d, J = 8.2 Hz), 7.30–7.38 (2H, m); 13C NMR (CD3OD) d: 157.5,
142.7, 139.9, 139.5, 130.5, 129.8, 128.6, 128.6, 125.4, 111.2, 78.5,
77.8, 76.9, 75.0, 64.2, 55.9, 41.0, 40.2, 36.7, 29.4, 16.3; MS (ESI+):
420 [M+H2O]+; HRMS calcd for C23H29O6 ([M-H]À) 401.1959. Found
401.1956.
phenyl]cyclohexanol.
bromo-2-(4-ethylbenzyl)-1-methoxybenzene
In a nitrogen stream, to a solution of 4-
(10d) (908 mg,
2.98 mmol) in THF (10 mL) was added 2.70 M nBuLi in hexane
solution (1.03 mL) dropwise at -78 °C and the reaction mixture
was stirred for 10 min. To this solution was added a solution of 8
(1.00 g, 1.86 mmol) in THF (2 mL) at -78 °C dropwise and the
resulting solution was stirred for 30 min. The reaction was stopped
by addition of satd. NH4Cl aq solution. The obtained solution was
extracted with AcOEt, and the organic layer was washed with
water and then dried over anhydrous MgSO4. The solvent was con-
centrated, and the obtained residue was purified by silica gel col-
umn chromatography (AcOEt/hexane = 1:5 to 1:1) to obtain a
less polar compound (320 mg, 23%) and a more polar compound
(890 mg, 64%).
(Less polar): Rf 0.38 (AcOEt/hexane = 1:4); 1H NMR (CDCl3) d:
1.17 (3H, t, J = 7.6 Hz), 1.80–2.00 (2H, m), 2.20–2.36 (1H, m), 2.55
(2H, d, J = 7.6 Hz), 2.91 (1H, d, J = 2.0 Hz), 3.35–3.42 (1H, m),
3.65–4.05 (10H, m), 4.41–4.45 (3H, m), 4.59 (1H, d, J = 10.7 Hz),
4.81 (1H, d, J = 10.7 Hz), 4.87 (1H, d, J = 10.7 Hz), 4.90 (1H, d,
J = 10.7 Hz), 6.78–6.84 (3H, m), 6.99–7.20 (6H, m), 7.20–7.40
(18H, m).
4.1.14. (1R,2R,3S,4S,6R)-4-[5-(4-Ethylbenzyl)-2-methoxyphenyl]
-6-(hydroxymethyl)cyclohexane-1,2,3-triol (5f)
This compound was prepared from 8 and 10f in the same man-
ner as described in Section 4.1.13. The total yield was 3%. Colorless
amorphous. 1H NMR (CD3OD) d: 1.18 (3H, t, J = 7.6 Hz), 1.24–1.50
(1H, m), 1.54–1.85 (2H, m), 2.57 (2H, q, J = 7.6 Hz), 2.90–3.18
(1H, m), 3.24–3.34 (2H, m), 3.54 (1H, dd, J = 10.7, 5.9 Hz), 3.60–
3.78 (2H, m), 3.76 (3H, s), 3.83 (2H, s), 6.82 (1H, d, J = 8.4 Hz),
6.96 (1H, dd, J = 8.4, 2.0 Hz), 7.01–7.11 (5H, m); 13C NMR (CD3OD)
d: 157.4, 142.9, 140.4, 134.9, 131.9, 129.8, 128.8, 128.8, 128.5,
111.9, 81.4, 76.0, 75.2, 64.5, 56.0, 45.5, 41.8, 41.8, 33.2, 29.4,
16.3; MS (ESI) m/z: 404 ([M+H2O]+); HRMS calcd for C23H29O5
([MÀH]À) 385.2010. Found 385.2011.
(More polar): Rf 0.31 (AcOEt/hexane = 1:4); 1H NMR (CDCl3) d:
1.12 (3H, t, J = 7.6 Hz), 1.40–1.54 (2H, m), 1.78–1.88 (1H, m), 2.35
(1H, dd, J = 13.7, 3.3 Hz), 2.43 (1H, s), 2.47 (2H, q, J = 7.6 Hz), 3.34
(1H, dd, J = 8.9, 0.9 Hz), 3.53–3.72 (4H, m), 3.80 (3H, s), 3.86–3.93
(2H, m), 4.41 (2H, s), 4.49 (1H, d, J=10.7 Hz), 4.67–4.79 (2H, m),
4.83 (1H, d, J = 10.7 Hz), 5.02 (1H, d, J = 11.7 Hz), 6.78 (1H, d,
J = 8.4 Hz), 7.00 (2H, d, J = 7.9 Hz), 7.10 (2H, d, J = 7.9 Hz), 7.14–
7.20 (2H, m), 7.21–7.40 (18H, m), 7.56–7.63 (2H, m).
4.1.13.2. ((((1S,2R,3R,4R,6S)-4-[(Benzyloxy)methyl]-6-[3-(4-eth-
ylbenzyl)-4-methoxyphenyl]cyclohexane-1,2,3-triyl)tris(oxy))
4.1.15. (1R,2R,3S,4R,5R)-1-[5-(4-Ethylbenzyl)-2-methoxyphenyl]
-5-(hydroxymethyl)cyclohexane-1,2,3,4-tetraol (5g)
tris(methylene))tribenzene.
To a solution of the above more
This compound was prepared from 8 and 10f in the same man-
ner as described in Section 4.1.14. The total yield was 47%. Color-
less amorphous. 1H NMR (CD3OD) d: 1.13 (3H, t, J = 7.6 Hz), 1.56
(1H, dd, J = 14.1, 3.9 Hz), 1.94–2.04 (1H, m), 2.24 (1H, dd, J = 13.9,
13.2 Hz), 2.51 (2H, q, J = 7.5 Hz), 3.24–3.37 (1H, m), 3.57–3.63
(3H, m), 3.74 (3H, s), 3.80–3.90 (2H, m), 4.22 (1H, d, J = 9.1 Hz),
6.80 (1H, d, J = 8.4 Hz), 6.95–7.01 (5H, m), 7.41 (1H, d, J = 1.2 Hz);
13C NMR (CD3OD) d: 156.0, 142.9, 140.5, 134.9, 134.3, 129.8,
129.5, 129.3, 128.8, 112.5, 78.0, 77.5, 75.7, 75.3, 64.5, 55.7, 42.0,
40.9, 36.2, 29.5, 16.3; MS (ESI) m/z: 425 ([M+Na]+); HRMS calcd
for C23H29O6 ([MÀH]À) 401.1959. Found 401.1958.
polar compound (124 mg, 0.16 mmol) in CH2Cl2 (2 mL) were added
Et3SiH (0.39 mL, 2.44 mmol) and trifluoroacetic acid (0.12 mL,
1.6 mmol) at À5 °C. The reaction mixture was stirred for 1 h, and
then satd. NaHCO3 aq solution was added thereto. The resulting
mixture was extracted with CH2Cl2 and the organic layer was
washed with brine and then dried over anhydrous MgSO4. The sol-
vent was concentrated, and the obtained residue was purified by sil-
ica gel column chromatography (AcOEt/hexane = 1:9) to obtain the
title compound (more polar, 10 mg, 8%) and a diastereoisomer (less
polar, 93 mg, 77%). Rf 0.68 (AcOEt/hexane = 1:4); 1H NMR (CDCl3) d:
1.17 (3H, t, J = 7.6 Hz), 1.58–1.96 (3H, m), 2.56 (2H, q, J = 7.6 Hz),
2.55–2.70 (1H, m), 3.43–3.62 (5H, m), 3.78–4.00 (6H, m), 4.43 (2H,
s), 4.44 (1H, d, J = 9.9 Hz), 4.56 (1H, d, J = 10.8 Hz), 4.82–4.93 (3H,
m), 6.75–6.82 (3H, m), 6.99–7.18 (9H, m), 7.20–7.38 (15H, m).
4.1.16. (1R,2R,3S,4S,6R)-4-[2-Ethoxy-5-(4-ethylbenzyl)phenyl]-
6-(hydroxymethyl)cyclohexane-1,2,3-triol (5h)
4.1.16.1. (1R,2R,3S,4R,5R)- and (1S,2R,3S,4R,5R)-2,3,4-Trisben-
zyloxy-5-benzyloxy-methyl-1-[2-ethoxy-5-(4-ethylben-
4.1.13.3. (1R,2R,3S,4S,6R)-4-[3-(4-Ethylbenzyl)-4-methoxyphenyl]-
zyl)phenyl]-cyclohexanol.
In a nitrogen stream, to a solution
6-(hydroxymethyl)cyclohexane-1,2,3-triol.
To a solution of
of 2-bromo-1-ethoxy-4-(4-ethylbenzyl) benzene (10h) (1.18 g,
3.70 mmol) in THF (9 mL) was added 1.59 M nBuLi in hexane solu-
tion (2.30 mL, 3.66 mmol) dropwise at -78 °C and the reaction mix-
ture was stirred for 1 h. Then, to this solution was added a solution
of 8 (1.50 g, 2.80 mmol) in THF (4.5 mL) dropwise at -78 °C. After
being stirred for 10 min, the reaction was stopped by addition of
satd. NH4Cl aq solution. The resulting mixture was extracted with
AcOEt, and the organic layer was washed with brine and then dried
over anhydrous MgSO4. The solvent was concentrated, and the ob-
tained residue was purified by silica gel column chromatography
the obtained compound (10 mg, 0.013 mmol) in THF (0.2 mL) and
MeOH (1 mL) was added 20% Pd(OH)2 (10 mg) and the mixture
was stirred under hydrogen atmosphere at rt for 13 h. After filtering
the catalyst, the filtrate was concentrated under reduced pressure.
The obtained residue was purified by PTLC (MeOH/CH2Cl2 = 1:10)
to obtain the title compound (3.9 mg, 75%) as a colorless amor-
phous. 1H NMR (CD3OD) d: 1.23 (3H, t, J = 7.4 Hz), 1.34–1.48 (1H,
m), 1.60–1.76 (1H, m), 1.83 (1H, dt, J = 13.5, 3.5 Hz), 2.48–2.66
(3H, m), 3.30–3.38 (2H, m), 3.42–3.52 (1H, m), 3.58–3.65 (1H, m),

Y. Ohtake et al. / Bioorg. Med. Chem. 20 (2012) 4117–4127
4125
(AcOEt/hexane = 1:4) to obtain a less polar compound (884 mg,
41%) and a more polar compound (740 mg, 34%).
less amorphous. 1H NMR (CD3OD) d: 1.19 (3H, t, J = 7.6 Hz), 1.24–
1.37 (1H, m), 1.59–1.81 (2H, m), 2.30 (3H, s), 2.57 (2H, q,
J = 7.6 Hz), 2.93 (1H, m), 3.34 (2H, m), 3.50–3.63 (2H, m), 3.75
(1H, dd, J = 10.7, 4.2 Hz), 3.86 (2H, s), 6.87 (1H, d, J = 7.6 Hz), 7.03
(1H, dd, J = 7.6, 2.1 Hz), 7.07 (4H, m), 7.11 (1H, d, J = 2.1 Hz); 13C
NMR (CD3OD) d: 142.9, 142.3, 140.5, 140.2, 135.3, 131.4, 129.8,
128.8, 127.5, 127.2, 81.3, 77.3, 75.1, 64.4, 45.6, 44.4, 42.3, 34.0,
29.4, 19.5, 16.3; MS (ESI) m/z: 369 ([MÀH]À); HRMS calcd for
(Less polar): Rf 0.39 (AcOEt/hexane = 1:4); 1H NMR (CDCl3) d:
1.23 (3H, t, J = 7.6 Hz), 1.37 (3H, t, J = 7.0 Hz), 1.70 (1H, dd,
J = 14.3, 4.1 Hz), 2.20–2.35 (1H, m), 2.62 (2H, q, J = 7.6 Hz), 2.67–
2.82 (1H, m), 3.00–3.30 (1H, br), 3.43 (1H, dd, J = 8.8, 2.0 Hz),
3.71 (1H, dd, J = 9.8, 9.8 Hz), 3.83 (1H, dd, J = 8.9, 4.0 Hz), 3.86–
4.10 (4H, m), 3.91 (2H, s), 4.44 (2H, s), 4.48 (1H, d, J = 10.3 Hz),
4.58 (1H, br), 4.63 (1H, d, J = 10.3 Hz), 4.85 (1H, d, J = 11.0 Hz),
4.88 (1H, d, J = 11.0 Hz), 4.95 (1H, d, J = 10.8 Hz), 6.75 (1H, d,
J = 8.4 Hz), 6.79 (2H, d, J = 7.0 Hz), 7.00–7.40 (23H, m), 7.53 (1H,
d, J = 2.0 Hz).
C
23H29O4 ([MÀH]À) 369.2060. Found 369.2058.
4.1.19. (1R,2R,3S,4R,5R)-1-[5-(4-Ethylbenzyl)-2-methylphenyl]-
5-(hydroxymethyl)cyclohexane-1,2,3,4-tetraol (5k)
(More polar): Rf 0.32 (AcOEt/hexane = 1:4); 1H NMR (CDCl3) d:
1.12 (3H, t, J = 7.6 Hz), 1.30 (3H, t, J = 7.2 Hz), 1.40–1.55 (1H, m),
1.76 (1H, dd, J = 13.3, 13.3 Hz), 2.50 (2H, q, J = 7.6 Hz), 2.74 (1H,
dd, J = 13.7, 3.2 Hz), 3.38 (1H, dd, J = 9.0, 2.8 Hz), 3.54 (1H, dd,
J = 9.0, 5.3 Hz), 3.64 (1H, dd, J = 10.8, 8.7 Hz), 3.86 (1H, dd, J = 9.5,
9.5 Hz), 3.89 (2H, s), 4.00–4.10 (2H, m), 4.39 (2H, s), 4.53 (1H, d,
J = 11.0 Hz), 4.64–4.72 (3H, m), 4.77 (1H, d, J = 11.7 Hz), 4.79 (1H,
d, J = 10.5 Hz), 4.87 (1H, d, J = 10.8 Hz), 4.98 (1H, d, J = 11.7 Hz),
6.98–7.42 (26H, m), 7.67 (1H, d, J = 1.8 Hz).
This compound was prepared from 8 and 10j in the same man-
ner as described in Section 4.1.14. The total yield was 20%. Color-
less amorphous. 1H NMR (CD3OD) d: 1.19 (3H, t, J = 7.6 Hz), 1.70
(1H, m), 1.94 (1H, m), 2.05 (1H, m), 2.53 (3H, s), 2.58 (2H, q,
J = 7.6 Hz), 3.40 (1H, d, J = 9.2 Hz), 3.67 (2H, d, J = 4.6 Hz), 3.73
(1H, d, J = 9.2 Hz), 3.87 (2H, s), 4.01 (1H, d, J = 9.2 Hz), 6.91 (1H,
dd, J = 7.6, 1.5 Hz), 7.01 (1H, d, J = 7.6 Hz), 7.07 (4H, m), 7.42 (1H,
d, J = 1.5 Hz); 13C NMR (CD3OD) d: 144.4, 142.9, 140.2, 140.0,
134.3, 134.0, 129.8, 128.8, 128.5, 128.3, 78.5, 78.0, 76.7, 74.8,
64.1, 42.3, 40.9, 37.8, 29.4, 22.5, 16.3; MS (ESI) m/z: 409
([M+Na]+); HRMS calcd for C23H29O5 ([MÀH]À) 385.2010. Found
385.2009.
4.1.16.2. (1R,2R,3S,4S,6R)-4-[2-Ethoxy-5-(4-ethylbenzyl)phenyl]
-6-(hydroxymethyl)cyclohexane-1,2,3-triol.
To a solution of
the obtained more polar compound (402 mg, 0.517 mmol) in THF
(4 ml) and methanol (2 mL) was added 20% Pd(OH)2 (78 mg) and
the mixture solution was stirred at rt under hydrogen atmosphere
for 19 h. After filtering the catalyst, the filtrate was concentrated
under reduced pressure. The obtained residue was purified by sil-
ica gel column chromatography (MeOH/CH2Cl2 = 1:10) to obtain
the title compound (42 mg, 20%) as a colorless amorphous. 1H
NMR (CD3OD) d: 1.18 (3H, t, J = 7.6 Hz), 1.25–1.50 (1H, m), 1.38
(3H, t, J = 7.0 Hz), 1.62 (1H, m), 1.78 (1H, m), 2.57 (2H, q,
J = 7.6 Hz), 2.95–3.15 (1H, br), 3.25–3.33 (2H, m), 3.54 (1H, dd,
J = 10.8, 6.1 Hz), 3.73 (1H, dd, J = 10.8, 3.8 Hz), 3.65–3.85 (1H, m),
3.83 (2H, s), 4.00 (2H, q, J = 7.0 Hz), 6.81 (1H, d, J = 8.3 Hz), 6.94
(1H, dd, J = 8.3, 2.1 Hz), 7.04 (1H, d, J = 2.1 Hz), 7.06 (4H, m); 13C
NMR (CD3OD) d: 156.7, 142.9, 140.4, 134.8, 132.0, 129.8, 129.8,
128.8, 128.5, 113.1, 81.4, 75.8, 75.3, 64.9, 64.6, 49.2, 45.5, 41.8,
33.0, 29.4, 16.3, 15.3; MS (ESI) m/z: 401 [M+H]; HRMS calcd for
4.1.20. (1R,2R,3S,4S,6R)-4-[5-(4-Ethylbenzyl)-2,4-dimethoxy-
phenyl]-6-(hydroxymethyl)cyclohexane-1,2,3-triol (5l)
To a solution of 10l (840 mg, 2.51 mmol) in Et2O (1 mL) was
added 1.58 M nBuLi in hexane solution (1.60 mL, 2.53 mmol) drop-
wise at À78 °C and the mixture solution was stirred for 15 min.
Then, to this solution was added a solution of 8 (1.03 g, 1.93 mmol)
in THF (3 mL) dropwise at À78 °C. The reaction mixture was stirred
at À78 °C for 2.5 h and then at À40 °C for 0.5 h. To the reaction
mixture was added satd. NH4Cl aq solution and the mixture was
extracted with AcOEt. The organic layer was washed with brine,
and then dried over anhydrous MgSO4. The solvent was
concentrated and the obtained residue was purified by silica
gel chromatography (AcOEt/hexane = 1:5 to 1:2) to obtain a less
polar compound (1100 mg, 72%) and a more polar compound
(108 mg, 7%).
To a solution of the above less polar compound (747 mg,
0.943 mmol) in CH2Cl2 (9.5 mL) were added Et3SiH (1.51 mL,
9.43 mmol) and BF3ÁOEt2 (1.19 mL, 9.43 mmol) at À40 °C. The reac-
tion mixture was stirred at À40 °C for 11 h and warmed up to
À10 °C for 1 h, and then satd. NaHCO3 aq solution was added
thereto. The resulting mixture was extracted with CH2Cl2 and the
organic layer was washed with brine and then dried over anhy-
drous MgSO4. The solvent was concentrated and the obtained res-
idue was purified by silica gel column chromatography (AcOEt/
hexane = 1:4) to obtain a more polar compound (164 mg, 22%)
and a less polar compound (475 mg, 65%) as a diastereoisomer.
To a solution of the above more polar compound (164 mg,
0.211 mmol) in THF (4 mL) and MeOH (2 mL) was added 20%
Pd(OH)2 (55 mg) and the mixture was stirred under hydrogen
atmosphere at rt for 40 min. After filtering the catalyst, the filtrate
was concentrated under reduced pressure. The obtained residue
was purified by PTLC (MeOH/CH2Cl2 = 1:10) to obtain the title
compound (81 mg, 92%) as a colorless amorphous. 1H NMR
(CD3OD) d: 1.17 (3H, t, J = 7.6 Hz), 1.24–1.48 (1H, m), 1.52–1.82
(2H, m), 2.56 (2H, d, J = 7.6 Hz), 2.86–3.08 (1H, m), 3.24–3.29
(1H, m), 3.48–3.67 (2H, m), 3.68–3.84 (2H, m), 3.79 (3H, s), 3.81
(3H, s), 6.56 (1H, s), 6.93 (1H, s), 7.01 (2H, d, J = 8.6 Hz), 7.05 (2H,
d, J = 8.6 Hz); 13C NMR (CD3OD) d: 158.5, 157.9, 142.5, 140.4,
130.6, 129.6, 128.5, 123.5, 122.7, 96.9, 81.5, 76.2, 75.3, 64.6, 56.2,
56.0, 45.5, 35.9, 34.9, 33.2, 29.4, 16.3; MS (ESI) m/z: 416 ([M]+);
HRMS calcd for C24H31O6 ([MÀH]À) 415.2115. Found 415.2118.
C
24H31O5 ([MÀH]À) 399.2166. Found 399.2163.
4.1.17. (1R,2R,3S,4R,5R)-1-[2-Ethoxyl-5-(4-ethylbenzyl)phenyl]-
5-(hydroxymethyl)cyclohexane-1,2,3,4-tetraol (5i)
To a solution of the less polar compound (600 mg, 0.772 mmol)
in THF (6 mL) and MeOH (3 mL) was added 20% Pd(OH)2 (60 mg)
and the mixture was stirred under hydrogen atmosphere for
2.5 h. After filtering the catalyst, the filtrate was concentrated un-
der reduced pressure. The obtained residue was purified by silica
gel column chromatography (MeOH/CH2Cl2 = 1:10) to obtain the
title compound (197 mg, 61%) as a colorless amorphous. 1H NMR
(CD3OD) d: 1.18 (3H, t, J = 7.6 Hz), 1.43 (3H, t, J = 7.0 Hz), 1.60
(1H, dd, J = 14.3, 4.1 Hz), 2.01 (1H, m), 2.43 (1H, dd, J = 14.0,
13.0 Hz), 2.57 (2H, q, J = 7.5 Hz), 3.39 (1H, dd, J = 10.5, 9.3 Hz),
3.60–3.68 (3H, m), 3.85 (2H, s), 4.03 (2H, m), 4.34 (1H, d,
J = 9.2 Hz), 6.82 (1H, d, J = 8.2 Hz), 6.99 (1H, dd, J = 8.4, 2.3 Hz),
7.04–7.14 (4H, m), 7.45 (1H, d, J = 2.3 Hz); 13C NMR (CD3OD) d:
155.2, 142.9, 140.5, 134.7, 134.3, 129.8, 129.4, 129.2, 128.8,
113.2, 77.9, 77.5, 75.7, 75.2, 64.9, 64.5, 42.0, 41.0, 36.2, 29.4,
16.3, 15.3; MS (ESI) m/z: 434 ([M+H2O]+); HRMS calcd for
C
24H31O6 ([MÀH]À) 415.2115. Found 415.2113.
4.1.18. (1R,2R,3S,4S,6R)-4-[5-(4-Ethylbenzyl)-2-methylphenyl]-
6-(hydroxymethyl)cyclohexane-1,2,3-triol (5j)
This compound was prepared from 8 and 10j in the same man-
ner as described in Section 4.1.13. The total yield was 25%. Color-

4126
Y. Ohtake et al. / Bioorg. Med. Chem. 20 (2012) 4117–4127
4.1.21. (1R,2R,3S,4R,5R)-1-[5-(4-Ethylbenzyl)-2,4-
dimethoxyphenyl]-5-(hydroxymethyl)cyclohexane-1,2,3,4-
tetraol (5m)
mals received a single oral dose of 2 (30 mg/kg), 5h and 5i (10 and
30 mg/kg) formulated as homogenous suspensions in 0.5% car-
boxymethyl-cellulose sodium salt via oral gavage. Blood samples
were collected just before and at 1, 2, 4, 6, and 8 h after dosing.
Blood glucose levels were measured by the hexokinase method
(Autosera S GLU, Daiichi Pure Chemicals, Japan).
This compound was prepared from 8 and 10l in the same man-
ner as described in Section 4.1.14. The total yield was 52%. Color-
less amorphous. 1H NMR (CD3OD) d: 1.17 (3H, t, J = 7.6 Hz), 1.61
(1H, dd, J = 13.9, 3.8 Hz), 1.90–2.10 (1H, m), 2.22 (1H, dd, J = 13.9,
13.2 Hz), 2.55 (2H, q, J = 7.6 Hz), 3.32–3.42 (1H, m), 3.58–3.70
(3H, m), 3.76–3.88 (2H, m), 3.80 (3H, s), 3.84 (3H, s), 4.19 (1H, d,
J = 9.1 Hz), 6.58 (1H, s), 7.00 (2H, d, J = 8.2 Hz), 7.06 (2H, d,
J = 8.2 Hz), 7.35 (1H, s); 13C NMR (CD3OD) d: 158.6, 156.9, 142.5,
140.4, 130.6, 129.7, 128.5, 126.0, 122.4, 97.3, 77.9, 77.2, 76.0,
75.3, 64.6, 56.1, 55.9, 40.8, 36.6, 36.0, 29.4, 16.3; MS (ESI) m/z:
455 ([M+Na]+); HRMS calcd for C24H31O7 ([MÀH]À) 431.2064.
Found 431.2059.
4.3.4. Pharmacokinetics studies
The db/db mice were administered a drug, and subjected to
blood sampling under the same conditions as the animals used
for the in vivo blood glucose- lowering test. Animals received a sin-
gle oral administration of 2 and 5h at a dose of 10 mg/kg (n = 6).
Blood samples were collected at 0.25, 0.5, 1, 2, 4, 6, and 8 h after
administration. The drug concentration in plasma sample was
measured using an LC-MS/MS system after deproteinization. LC-
MS/MS analysis was performed using a Shimadzu LC-10AD pump
and an Applied Biosystems/MDS Sciex API-300 mass spectrometer
with a TurboIonSpray source. Chromatographic separation was
4.2. Molecular modeling
All molecular modeling procedures were performed using
Molecular Operating Environment (MOE) ver.2010.10.14 Six known
SGLT2 inhibitors and our three compounds were constructed and
minimized with use of MMFF94 force field.16 Superposed models
of selected inhibitors were derived by the FlexibleAlignment mod-
ule of MOE in which H-bond donor and acceptor, hydrophobe, and
aromaticity were assigned for similarity terms. Although several
energetically equivalent models were generated because of higher
flexibility of inhibitors, the differences among these models are
due to the relative configurations of sugar and distal aromatic moi-
eties bound to the central aromatic ring. One of them, which has a
similar configuration to the crystal structure of phlorizin (Cam-
bridge Structure Database ID: CEWWAC01), was chosen for further
discussion. Global minimum energy of each inhibitor was esti-
mated by the LowModeMD search method. The energy difference
of each compound aligned was within 5 kcal/mol from the global
minimum.
achieved using
(2.0 Â 150 mm, 5
a
Shiseido CAPCELL PAK C18 column
lm). The mobile phase consisted of acetoni-
trile/10 mM ammonium acetate solution (4:6, v/v) and the flow
rate was set at 0.2 mL/min. The injection cycle of each sample
was set at 10 min. The transitions for multiple reaction monitoring
were 394 to 180 (2-active), 418 to 259 (5h). Non-compartmental
pharmacokinetic parameters were calculated based on the aver-
aged plasma concentration-time data using WinNonlin Profes-
sional 5.0 (Pharsight, Mountain View, CA).
Acknowledgements
We are very thankful to Y. Itezono for NMR analysis; H. Suda for
mass spectrometry; A. Higashida and N. Sekiguchi for CYP inhibi-
tion assay; T. Kawai, Y. Hasebe, M. Takeda of Chugai Pharmaceuti-
cal Co., Ltd and many biologists of Chugai Research Institute for
Medical Science, Inc. for evaluating the pharmacological profiles
in the animal experiments, and Editing Services at Chugai Pharma-
ceutical Co., Ltd for assistance with English usage.
4.3. Biology
4.3.1. In vitro SGLT inhibition assay
References and notes
Chinese hamster ovary-K1(CHO) cells stably expressing human
SGLT2 (NM003041) and human SGLT1 (NM000343) were used for
1. You, G.; Lee, W. S.; Barros, E. J. G.; Kanai, Y.; Huo, T. L.; Khawaja, S.; Wells, R. G.;
Nigam, S. K.; Hediger, M. A. J. Biol. Chem. 1995, 270, 29365.
2. Wright, E. M.; Turk, E.; Martin, M. G. Cell Biochem. Biophys. 2002, 36, 115.
3. Nomura, S. Curr. Top. Med. Chem. 2010, 10, 411.
4. (a) Matsuoka, H.; Sato, T.; Nishimoto, M.; Shimma N. PCT Int. Appl.
WO2006011469, 2006; EP1783110.; (b) Ohtake, Y.; Sato, T.; Matsuoka, H.;
Nishimoto, M.; Taka, N.; Takano, K.; Yamamoto, K.; Ohmori, M.; Higuchi, T.;
Murakata, M.; Kobayashi, T.; Morikawa, K.; Shimma, N.; Suzuki, M.; Hagita, H.;
Ozawa, K.; Yamaguchi, K.; Kato, M.; Ikeda, S. Bioorg. Med. Chem. 2011, 19, 5334.
5. (a) Fushimi, N.; Ito, F.; Isaji, M. PCT Int. Appl. WO2003011880, 2004.; (b)
Katsuno, K.; Fujimori, Y.; Takemura, Y.; Hiratochi, M.; Itoh, F.; Komatsu, Y.;
Fujikura, H.; Isaji, M. J. Pharmacol. Exp. Ther. 2007, 320, 323.
the sodium-dependent methyl-
a- -glucopyranoside (AMG) uptake
D
inhibition assay. The cells were incubated in reaction buffer with
compound and 1 mM AMG containing [14C]AMG for 45 min. AMG
uptake activity was determined by counting the radioactivity of
the cell lysates. IC50 values were calculated by curve fitting using
a four-parameter logistic model (XLfit, ID Business Solutions Ltd.).
4.3.2. CYP3A4 inhibition assay10
6. (a) Meng, W.; Ellsworth, B. A.; Nirschl, A. A.; McCann, P. J.; Patel, M.; Girotra, R.
N.; 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, W. N.; Janovitz, E. B.; Flint, O. P.; Whaley, J. M.; Washburn, W. N. J.
Med. Chem. 2008, 51, 1145; (b) Washburn, W. N. J. Med. Chem. 2009, 52, 1785;
(c) Ellsworth, B. A.; Meng, W.; Patel, M.; Girotra, R. N.; Wu, G.; Sher, P. M.;
Hagan, D. L.; Obermeier, M. T.; Humphreys, W. G.; Robertson, J. G.; Wang, A.;
Han, S.; Waldron, T. L.; Morgan, N. N.; Whaley, W. N.; Washburn, W. N. Bioorg.
Med. Chem. Lett. 2008, 18, 4770; (d) Obermeier, M.; Yao, M.; Khanna, A.;
Koplowitz, B.; Zhu, M.; Li, W.; Komoroski, B.; Kasichayanula, S.; Discenza, L.;
Washburn, W.; Meng, W.; Ellsworth, B. A.; Whaley, J. M.; Humphreys, W. G.
Drug Metab. Dispos. 2010, 38, 405.
The evaluation of CYP3A inhibition was carried out on 96-well
microtiter plates using MultiPROBE™ IIEX Liquid Handling Robotics
(PerkinElmerJapan, Yokohama, Japan). Fluorescence detection of
the metabolite of 7-benzyloxy–4-(trifluoromethyl)-coumarin
(BFC) was performed using a Spectra Max Gemini microplate read-
er (Molecular Devices, Sunnyvale, CA, USA). Incubations and calcu-
lations of IC50 were conducted based on the method provided on
the BD Biosciences website.17 IC50 (–) and IC50 (+) represent the
IC50 values in which the inhibitor was added to the reaction mix-
ture simultaneously with BFC (coincubation assay) and before
the addition of BFC (preincubation assay), respectively. The criteria
to determine positive TDI were based on a twofold value of the
coefficient of the variation (CV) of the IC50 (–) of ketoconazole.
7. Nomura, S.; Sakamaki, S.; Hongu, M.; Kawanishi, E.; Koga, Y.; Sakamoto, T.;
Yamamoto, Y.; Ueta, K.; Kimata, H.; Nakayama, K.; Tsuda-Tsukimoto, M. J. Med.
Chem. 2010, 53, 6355.
8. Hardman, T. C.; Rutherford, P.; Dubrey, S. W.; Wierzbicki, A. S. Curr. Pharma.
Des. 2010, 16, 3830.
9. Tetraol 5j and pentaol 5k were used for the determination of the configuration
at position 1 on cyclohexane ring. The configuration of 5j was determined by
1H NMR spectra. The coupling constant of C(1)-H and C(2)-H was 9.5 Hz. This
large coupling constant shows the axial–axial configuration. The spectra of
4.3.3. In vivo blood glucose-lowering test in db/db mice
Male db/db mice were purchased from CLEA Japan (Tokyo,
Japan) and maintained on a regular diet (CE2, CLEA Japan). The ani-

Y. Ohtake et al. / Bioorg. Med. Chem. 20 (2012) 4117–4127
4127
tertiary alcohol 5k by ROESY in DMSO-d6 shows the correlation between C(1)-
OH and C(3)-Hax, C(5)-Hax. It means that the configuration of C(1)-OH of 5k is
also axial as described in our Figures.
Io, F.; Yamamoto, D.; Miyata, N.; Takahashi, T.; Uchida, S.; Yamamoto, K. J. Med.
Chem. 2010, 53, 3247.
14. Molecular Operating Environment v2010.10, Chemical Computing Group, Inc.:
Montreal, Quebec.
15. All the prepared compounds were also evaluated for their in vitro inhibition
activity against mouse SGLTs (mSGLT1 and 2). Compound 5i showed twice
more potent mSGLT2 inhibitory activity (IC50 = 20 nM) than 5h (IC50 = 48 nM).
16. Halgren, T. A. J. Comput. Chem. 1996, 17, 490.
17. BD Biosciences Website (San Jose, CA: BD Biosciences) (available at: http://
www.bdbiosciences.com/discovery_labware/gentest/products/HTS_KITS/HTS/
hts_summary.shtml) (2000).
10. Sekiguchi, N.; Higashida, A.; Kato, M.; Nabuchi, Y.; Mitsui, T.; Takanashi, K.;
Aso, Y.; Ishigai, M. Drug Metab. Pharmacokinet. 2009, 24(6), 500.
11. Ehrenkanz, J. R. L.; Lewis, N. G.; Kahn, C. R.; Roth, J. Diabates/Metab. Res. Rev.
2005, 21, 31.
12. Imamura, M.; Murakami, T.; Shiraki, R.; Ikegai, K.; Sugane, T.; Iwasaki, F.;
Kurosaki, E.; Tomiyama, H.; Noda, A.; Kitta, K.; Kobayashi, Y. PCT int. Appl.
WO2004080990, 2004; Chem. Abstr. 2004, 141, 296242.
13. Kakinuma, H.; Oi, T.; Hashimoto-Tsuchiya, Y.; Arai, M.; Kawakita, Y.; Fukasawa,
Y.; Iida, I.; Hagima, N.; Takeuchi, H.; Chino, Y.; Asami, J.; Okumura-Kitajima, L.;