Design, synthesis and biological evaluation of 6-deoxy O-spiroketal C-arylglucosides as novel renal sodium-dependent glucose cotransporter 2 (SGLT2) inhibitors for the treatment of type 2 diabetes

Article abstract of DOI:10.1016/j.ejmech.2019.07.032

In this work, aiming at finding a novel, potent, and selective sodium-dependent glucose cotransporter 2 (SGLT2) inhibitor with good pharmacokinetic profiles for the treatment of diabetes, we focus on modifying the sugar moiety of SGLT2 inhibitors, which dominates the binding with glucose binding site of hSGLT, via removing the C-6 hydroxy group to adjust the physicochemical properties and target-recognition manners of SGLT2 inhibitors. In addition, tofogliflozin containing a special O-spiroketal C-arylglucoside scaffold, displayed good efficacy and bioavailability both in animals and in humans. Therefore, a series of 6-deoxy O-spiroketal C-arylglucosides as novel SGLT2 inhibitors were designed, synthesized, and evaluated in this work. The structure-activity relationship (SAR) research on this novel series and a comprehensive in vitro and in vivo biological evaluation afforded compound 39 with high in vitro hSGLT2 inhibitory activity (IC₅₀ = 4.5 nM), good pharmacokinetic profiles, and more remarkable efficacy in C57BL/6J mice and Sprague-Dawley rats than marketed drug tofogliflozin.

Full text of DOI:10.1016/j.ejmech.2019.07.032

Accepted Manuscript
Design, synthesis and biological evaluation of 6-deoxy O-spiroketal C-arylglucosides
as novel renal sodium-dependent glucose cotransporter 2 (SGLT2) inhibitors for the
treatment of type 2 diabetes
Yibing Wang, Yang Lou, Jiang Wang, Dan Li, Hui Chen, Tiannan Zheng, Chunmei
Xia, Xiaohan Song, Tiancheng Dong, Jingya Li, Jia Li, Hong Liu
PII:
S0223-5234(19)30653-1
DOI:
https://doi.org/10.1016/j.ejmech.2019.07.032
EJMECH 11529
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 12 May 2019
Revised Date: 9 July 2019
Accepted Date: 9 July 2019
Please cite this article as: Y. Wang, Y. Lou, J. Wang, D. Li, H. Chen, T. Zheng, C. Xia, X. Song, T. Dong,
J. Li, J. Li, H. Liu, Design, synthesis and biological evaluation of 6-deoxy O-spiroketal C-arylglucosides
as novel renal sodium-dependent glucose cotransporter 2 (SGLT2) inhibitors for the treatment
of type 2 diabetes, European Journal of Medicinal Chemistry (2019), doi: https://doi.org/10.1016/
j.ejmech.2019.07.032.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Title:
Design, Synthesis and Biological Evaluation of 6-Deoxy
O-spiroketal C-arylglucosides as Novel Renal
Sodium-Dependent Glucose Cotransporter 2 (SGLT2)
Inhibitors for the Treatment of Type 2 Diabetes

ACCEPTED MANUSCRIPT
Graphical Abstract

ACCEPTED MANUSCRIPT
Highlights:
ꢀ
ꢀ
ꢀ
Novel 6-deoxy O-spiroketal C-arylglucosides as SGLT2 inhibitors were designed,
synthesized, and evaluated.
The structure-activity relationship (SAR) of this novel series of 31 compounds
was demonstrated.
Compound 39 demonstrated excellent in vitro inhibitory activities against
hSGLT2 (IC = 4.5 nM), which was higher than positive control (dapagliflozin,
5
0
IC = 8.3 nM in this assay).
5
0
ꢀ
ꢀ
Compound 39 displayed more remarkable efficacy on promoting urinary glucose
excretion and improving oral glucose tolerance than marketed drug tofogliflozin
did in C57BL/6J mice.
The half-life of compound 39 in mice was longer than tofogliflozin after oral
administration (5.95 h vs 2.57 h).

ACCEPTED MANUSCRIPT
Design, Synthesis and Biological Evaluation of 6-Deoxy
O-spiroketal
C-arylglucosides
as
Novel
Renal
Sodium-Dependent Glucose Cotransporter 2 (SGLT2)
Inhibitors for the Treatment of Type 2 Diabetes
†
,‡,
║
†,
║
†,‡
†
†
†,‡
Yibing Wang,
Yang Lou, Jiang Wang, Dan Li, Hui Chen, Tiannan Zheng,
†
†,‡
†
†,‡
*,†,‡
Chunmei Xia, Xiaohan Song, Tiancheng Dong, Jingya Li, Jia Li,
and Hong
*,†,‡
Liu
†
State Key Laboratory of Drug Research and CAS Key Laboratory of Receptor
Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555
Zu Chong Zhi Road, Shanghai, 201203, China.
‡
University of Chinese Academy of Sciences, No.19A Yuquan Road, Beijing 100049,
China.
║These authors contributed equally to this study.
Corresponding Authors
*
H. Liu: Phone, +86-21-50807042; E-mail, hliu@simm.ac.cn.
J. Li: Phone, +86-21-50806027; E-mail, jli@simm.ac.cn
*

ACCEPTED MANUSCRIPT
ABSTRACT:
In this work, aiming at finding a novel, potent, and selective sodium-dependent
glucose cotransporter 2 (SGLT2) inhibitor with good pharmacokinetic profiles for the
treatment of diabetes, we focus on modifying the sugar moiety of SGLT2 inhibitors,
which dominates the binding with glucose binding site of hSGLT, via removing the
C-6 hydroxy group to adjust the physicochemical properties and target-recognition
manners of SGLT2 inhibitors. In addition, tofogliflozin containing a special
O-spiroketal C-arylglucoside scaffold, displayed good efficacy and bioavailability
both in animals and in humans. Therefore, a series of 6-deoxy O-spiroketal
C-arylglucosides as novel SGLT2 inhibitors were designed, synthesized, and
evaluated in this work. The structure-activity relationship (SAR) research on this
novel series and a comprehensive in vitro and in vivo biological evaluation afforded
compound 39 with high in vitro hSGLT2 inhibitory activity (IC50 = 4.5 nM), good
pharmacokinetic profiles, and more remarkable efficacy in C57BL/6J mice and
Sprague-Dawley rats than marketed drug tofogliflozin.
Keywords: SGLT2 inhibitors, Diabetes, Sugar modification, Structure-activity
relationship, Urinary glucose excretion, Oral glucose tolerance

ACCEPTED MANUSCRIPT
INTRODUCTION
Diabetes Mellitus (DM), as a chronical, progressive metabolic disorder causing
many devastating short and long-term effects, has represented one of the primary
threats to health and human development these decades. It mainly consists of three
types: type 1 diabetes (T1DM), type 2 diabetes (T2DM), and gestational diabetes
1
(
GDM), and T2DM accounts for around 90% of all cases. The common medications
for T2DM treatment are insulin and its analogs, as well as metformin, sulphonylureas,
1
GLP-1 analogs, and DPP4 inhibitors. For T1DM patients, current small molecular
drugs usually fail in reducing blood glucose levels, and long-term injection of insulin
2
is indispensable because the lack of basal insulin in T1DM patients. However,
various adverse effects of those medications such as hypoglycemia, cardiovascular
3
risk and weight gain cannot be ignored. Recently, the newly developed
sodium-dependent glucose cotransporter 2 (SGLT2) inhibitors have become
4
, 5
increasingly popular and widely accepted for T2DM therapy, and the clinical trials
of T1DM treatment using SGLT2 inhibitors combined with insulin are well
6
-11
underway
.
SGLT2, mostly distributed in proximal renal tubules, is a major transporter
responsible for over 90% of renal glucose reabsorption that is known as one of the
12
major pathways to maintain glucose homeostasis.
In diabetes, continuous
hyperglycemia leads to upregulation of SGLT2, resulting in stronger absorption of
1
3-15
renal glucose and consequently higher blood glucose levels.
Based on the
revelation that mutations of the SGLT2 gene caused familial renal glycosuria without
affecting daily activities, researchers have presumed that inhibiting the hyperactive
SGLT2 in kidneys of diabetic patients, thus increasing glucose excretion, should be a
1
6, 17
promising approach to reduce blood glucose levels.
Moreover, this approach
causes no hypoglycemia and weight gain because the remarkable anti-diabetic effects
1
2, 18, 19
is independent with insulin secretion or insulin action.
Most excitingly, the
EMPA-REG OUTCOME trial revealed that using SGLT2 inhibitor treatment with
empagliflozin displayed significant reduction of heart failure hospitalization and

ACCEPTED MANUSCRIPT
2
0, 21
cardiovascular deaths in T2DM patients
, and similar results of other gliflozins
4
, 22, 23
were also reported
anti-diabetic medications
benefits have been proposed, including the amelioration of ventricular preload and
, which has never been achieved by other classes of
3
, 24-28
. Several putative mechanisms of the cardiovascular
+
+
afterload, improvement in cardiac metabolism, inhibition of myocardial Na /H
2
9
exchange, reduction of necrosis and cardiac fibrosis, and so on. Besides, it has been
reported that T1DM patients treated with SGLT2 inhibitors need less dosage of insulin
injection to improve glycemic control, which may be associated with a lower rate of
6
-11, 30
hypoglycemia events.
The first reported SGLT2 inhibitor was phlorizin (1, Figure 1), a natural product
that was proved to cause glucosuria and hypoglycemic effect via nonselective SGLT
inhibition. Nonetheless, phlorizin was not suitable for clinical use mainly because of
its poor metabolic stability to β-glucosidases, moderate potency, and severe
gastrointestinal side effects resulting from its low selectivity of hSGLT2. Therefore,
large amounts of efforts have been promoted to overcome these drawbacks and plenty
3
1, 32
of promising SGLT2 inhibitors have sprung up
. Generally, SGLT2 inhibitors are
divided into three categories, O-glucosides, C-glycosides, and O-spiroketal
3
3
C-arylglucosides. The metabolic susceptibility to β-glucosidases limited the
development of O-glucosides, while the latter two structural categories, which are
proved highly stable against glucosidases, have been extensively and rapidly
investigated, yielding up to seven drugs launched for the treatment of T2DM (Figure
3
4
35
1
). Among the first few drugs on the market, dapagliflozin (2), canagliflozin (3),
36
37
38
empagliflozin (4), and ipragliflozin (5) are C-glucosides; tofogliflozin (6) is of
39
O-spiroketal C-arylglucoside class; luseogliflozin (7) and the latest approved
4
0
ertugliflozin (8) are special C-glycosides: the former one contains a C-phenyl
1
-thio-D-glucitol scaffold, and the latter one incorporates a unique bridged ketal ring.
Although up to seven gliflozins are already on the market, the hypoglycemic effects
as well as cardiovascular benefits of them still need improvement, and it is worthy of
efforts on developing new drugs to avert adverse effects such as ketoacidosis.

ACCEPTED MANUSCRIPT
Figure 1. Phlorizin and the marketed SGLT2 inhibitors.
It was noticed that all the currently marketed SGLT2 inhibitors share the same
pattern of four hydroxy groups in the glucose moiety. As some results of binding
41
36, 42, 43
studies of phlorizin and marketed gliflozins
with hSGLTs had been revealed,
the critical role of the sugar moiety in target-recognition attracted our attention to
searching for a novel and potent SGLT2 inhibitor with more satisfactory
pharmacokinetics profiles via sugar modifications. Herein, we report the design,
synthesis, and biological evaluation of a novel series of 6-deoxy O-spiroketal
C-arylglucosides 9-39, among which compound 39 was selected as a highly potent
SGLT2 inhibitor.
RESULTS AND DISCUSSION
Design of 6-deoxy O-spiroketal C-arylglucosides
Recent decades have witnessed large amounts of global efforts on the SAR
3
2, 33, 44-49
investigation of selective SGLT2 inhibitors
. Generally, three main elements
are proved critical to the maintenance of SGLT2 inhibitory activity: a glucose moiety
or its analogues with the same chair configuration, a proximal aromatic ring, and a
distal ring where a lipophilic substituent is favorable.
Although the interactions between SGLT2 inhibitors and the target have not been
achieved in detail because the crystal structure of SGLT2 is unavailable yet, several
studies have disclosed that it was the ectodomains of SGLT that recognized the

ACCEPTED MANUSCRIPT
17, 50-52
inhibitors
. The glucose-binding site of hSGLT dominates the interaction with
the sugar moiety of SGLT2 inhibitors, while the binding of aglucone moiety influence
41, 42, 52
the affinity of the entire inhibitor
. Furthermore, inhibition kinetics and binding
4
1
42
43
studies of phlorizin , dapagliflozin , and luseogliflozin suggested that modification
of the sugar moiety may profoundly influence the association and dissociation
manners of SGLT2 inhibitors with hSGLT2. For example, luseogliflozin, containing a
distinctive 1-thio-D-glucitol sugar ring, displayed a much longer dissociation half-life
4
3
36
(7 hours)
than empagliflozin (about 60 min) , which might interpret its long
5
3
duration of action on urinary glucose excretion .
At present, most SGLT2 inhibitors have been designed focusing on the
3
3
modification of aglucone , while the efforts in sugar modification were relatively
3
9, 40, 45-47
rare
. Besides, the same pattern of four hydroxy groups in the sugar moiety
was retained in all the marketed drugs (Figure 1). In the course of our efforts to the
development of novel, potent and selective SGLT2 inhibitors with good
pharmacokinetic profiles, we envisioned that the role of the sugar moiety in
target-recognition might need more in-depth investigation, and removing the C-6
hydroxy group from the glucose moiety might not only alter the physicochemical
properties of the molecule, but also influence the association and dissociation manners
of SGLT2 inhibitors with the target. On the other hand, the interesting scaffold of
3
8
O-spiroketal C-arylglucoside in tofogliflozin attracted our attention because of its
3
8, 54
high potency and selectivity
.
Therefore, we designed and synthesized a novel series of 6-deoxy O-spiroketal
C-arylglucosides 9-39 as SGLT2 inhibitors (Figure 2).
Figure 2. Design of 6-deoxy O-spiroketal C-arylglucosides as SGLT2 inhibitors.
Chemistry

ACCEPTED MANUSCRIPT
The synthetic route of 6-deoxy O-spiroketal C-arylglucosides 9-24 is outlined in
Scheme 1. Methyl glucopyranose 44 was converted to 45 in three consecutive steps to
selectively leave the 6-OH group free. Direct bromination of 45 with
triphenylphosphine and carbon tetrabromide, afforded quantitatively the bromide 46.
Efficient reduction of 46 with the tributyltin hydride furnished 47. Acid hydrolysis of
4
7 afforded hemiacetal 48, which was finally oxidized by DMSO to provide the key
intermediate 49. On the other hand, reduction of the two carboxyl groups of
commercially available 2-bromoterephtalic acid 50, followed by protection of the
resulting hydroxy groups with 2-methoxypropene, afforded intermediate 52. Addition
of lactone 49 to intermediate 52 gave the desired O-spiroketal C-arylglucoside 53 by
the treatment with p-toluenesulfonic acid. Oxidation of 53 provided aldehyde 54,
which was further reacted with Ar-X (X = Br or H) to afford 55a-55p. Alcohols
5
5a-55p were converted to 56a-56p by using triethylsilane and borontrifluoride
diethyletherate. Finally, debenzylation with boron trichloride or catalytic
hydrogenation with palladium hydroxide under a hydrogen atmosphere afforded the
target compounds 9-24.
a
Scheme 1. Synthetic routes of 6-deoxy O-spiroketal C-arylglucosides 9-24.

ACCEPTED MANUSCRIPT
a
Reagents and conditions: (a) Imidazole, TIPSCl, DMF, 0 °C to rt, 24 h; (b) NaH,
BnBr, DMF, 0 °C to rt, 12 h; (c) TBAF, THF, rt, 12 h; (d) Ph P, CBr , THF, 0 °C, 1 h;
3
4
(
e) Bu SnH, AIBN, toluene, 80 °C, 4 h; (f) 3 M H SO aq, AcOH, 85 °C, 2.5 h; (g)
3
2
4
.
Ac O, DMSO, rt, overnight; (h) BH SMe , THF, 0 °C to 70 °C, 4 h; (i) pyridinium
2
3
2
p-toluenesulfonate, 2-methoxypropene, THF, 0 °C, 2 h; (j) n-BuLi, THF, −78 °C, 3 h;
(
(
(
k) p-toluenesulfonic acid, THF-MeOH, rt, 15 h; (l) PCC, CH Cl , rt, 4 h; (m) Ar-X
2 2
X = H or Br), n-BuLi, THF, −78 °C, 3 h; (n) Et SiH, BF ·OEt , CH Cl , −40 °C, 1 h;
3
3
2
2
2
o) H , Pd(OH) /C, MeOH-AcOEt, 2 M HCl aq, rt, 5 h, or BCl , Me -benzene,
2
2
3
5
CH Cl , −78 °C, overnight.
2
2
Synthetic routes of compounds 25-28 and 33-37 are shown in Scheme 2. Diol 59 was
obtained from commercially available 57 via a three-step procedure consisting of
dibromination, nucleophilic substitution reaction, and hydrolysis. Compound 59 was
transformed into compounds 25-28 and 33-37 by a similar method to that in Scheme
1
.

ACCEPTED MANUSCRIPT
a
Scheme 2. Synthetic routes of compounds 25-28 and 33-37.
a
Reagents and conditions: (a) NBS, AIBN, EtOAc, reflux, 0.5 h; (b) NaOAc, DMF,
8
0 °C, 3 h; (c) KOH, THF-EtOH-H O, 80 °C, 4.5 h; (d) pyridinium
2
p-toluenesulfonate, 2-methoxypropene, THF, 0 °C, 2 h; (e) n-BuLi, THF, −78 °C, 3 h;
(
f) p-toluenesulfonic acid, THF-MeOH, rt, 15 h; (g) PCC, CH Cl , rt, 4 h; (h) Ar-X (X
2 2
=
H or Br), n-BuLi, THF, −78 °C, 3 h; (i) Et SiH, BF ·OEt , CH Cl , −40 °C, 1 h; (j)
3 3 2 2 2
BCl , Me -benzene, CH Cl , −78 °C, overnight.
3
5
2
2
Synthetic routes to compounds 29-32 are shown in Scheme 3. Aldehyde 62 was
coupled with 2-bromothiophene followed by a reduction reaction to give intermediate
6
6. A formyl group was introduced to compound 66 by the use of
N,N-dimethylformamide (DMF), giving aldehyde 67. Debenzylation of 67 afforded
compound 29, further reduction of the formyl group of 35 furnished the
corresponding alcohol 30. On the other hand, the oxidation of aldehyde 67 using
iodine under basic conditions afforded ester 68, which was further subjected to
debenzylation to furnish compound 31. Aldehyde 67 was treated with chloroform and
5
5
KOH to give intermediate 69, which was subsequently transformed into aryl acetic

ACCEPTED MANUSCRIPT
acid 70. Final debenzylation of intermediate 70 followed by the after-treatment with
methanol yielded compound 32.
a
Scheme 3. Synthetic routes of compounds 29-32.
a
Reagents and conditions: (a) 2-bromothiophene, n-BuLi, THF, −78 °C, 3 h; (b)
Et SiH, BF ·OEt , CH Cl , −40 °C, 1 h; (c) DMF, n-BuLi, THF, −70 °C, overnight; (d)
3
3
2
2
2
BCl , Me -benzene, CH Cl , −78 °C, overnight; (e) NaBH , MeOH, rt, 2 h; (f) I ,
3
5
2
2
4
2
KOH, MeOH, 0 °C, 3 h; (g) CHCl , KOH, MeOH, DMF , −9 °C, 2 h; (h) NaOH
3
(
newly grinded powder), NaBH , t-BuOH, 35 °C, 48 h.
4
Compounds 38-39 were synthesized as shown in Scheme 4. Reduction of
commercially available 71 under sodium borohydride condition generated
(2-bromo-5-fluorophenyl)methanol 72, which was further subjected to formylation
and reduction and gave compound 73. Diol 73 was transformed into compound 38 by

ACCEPTED MANUSCRIPT
a similar synthetic method to that in Scheme 1. Compound 39 was prepared starting
from commercially available 79. Diol 82 was obtained from 79 via chloromethylation,
substitution reaction, acetylation and hydrolysis. Compound 82 was then transformed
into compound 39 by a similar synthetic route to that in Scheme 1.
a
Scheme 4. Synthetic routes of compounds 38-39.
a
Reagents and conditions: (a) NaBH , MeOH, 0 °C to rt, 2.5 h; (b) HTMP, n-BuLi,
4
DMF, THF, -78 °C to rt, 3.5 h; (c) pyridinium p-toluenesulfonate, 2-methoxypropene,
THF, 0 °C, 2 h; (d) ZnCl , paraformaldehyde, HCl (gas), 60 °C, 24 h; (e) NaOAc,
2
DMF, 80 °C, 3 h; (f) KOH, THF-EtOH-H O, 80 °C, 4.5 h; (g) n-BuLi, THF, −78 °C,
2
3
2
h; (h) p-toluenesulfonic acid, THF-MeOH, rt, 15 h; (i) PCC, CH Cl , rt, 4 h; (j)
2 2
-ethylthiophene, n-BuLi, THF, −78 °C, 3 h; (k) Et SiH, BF ·OEt , CH Cl , −40 °C, 1
3
3
2
2
2
h; (l) BCl , Me -benzene, CH Cl , −78 °C, overnight.
3
5
2
2
Structure−Activity Relationships of 6-deoxy O-spiroketal C-arylglucosides
The in vitro cell-based SGLT2 AMG (methyl-α-D-glucopyranoside) assay was

ACCEPTED MANUSCRIPT
performed to evaluate the inhibitory effects of all synthesized compounds 9-39 on
56
hSGLT2 and hSGLT1 inhibitory activities. In the SAR exploration, our first
attempts were focused on the series of compounds 9-24 with R as a hydrogen atom as
in the case of tofogliflozin. The distal aromatic ring, ranging from alkylphenyl,
alkoxyphenyl, alkylthiophenyl to arylthiophenyl, naphthyl and even heteroaromatic
fused rings, were investigated (Table 1). Unfortunately, most of these compounds
showed weak activities, and even the best one of them, namely, compound 10 with an
ethylphenyl ring, just displayed an IC of 63.5 nM against hSGLT2, while the IC
5
0
50
values of the others were all above 100 nM (Table 1). We assumed that the R
substitution of this novel scaffold of 6-deoxy O-spiroketal C-arylglucoside might be
quite different from the case of tofogliflozin, although they share similar spiroketal
moieties. Then, compound 25 was designed via the alteration of R group to a chloride
atom while the ethylphenyl group was reserved as the distal Ar moiety. To our
excitement, compound 25 was about twice more potent than the hydrogen-substituted
counterpart (10, Table 1). When the ethyl group on the distal phenyl ring was changed
into a para-ethyoxyl group, the activity of the corresponding compound was
significantly enhanced, with an IC50 of 8.8 nM against hSGLT2 (compound 26 in
Table 1). Next, several thiophenyl-substituted analogues were investigated.
Compound 27, containing a 5-methylthiophenyl substitution, was as potent as
compound 26 whose Ar group was a para-ethyoxylphenyl ring. Compound 28 with
the distal ring of 5-chlorothiophenyl group instead of alkylthiophenyl groups, showed
good inhibitory activity with an hSGLT2 IC of 15.9 nM. On the other hand, some
5
0
electron-deficient-group substituted thiophenyl rings were not quite favored in
promoting hSGLT2 inhibition, as well as in the cases of hydrophobic ester derivatives,
and the corresponding compounds possessed moderate IC50 values around 50 nM
against hSGLT2 (compounds 29, 30, 31 and 32 in Table 1). In addition, compounds
bearing hydrophobic chains that contain different alkoxy groups exhibited moderate
potencies, and the extension of the length of alkoxy groups might slightly improve the
potency and selectivity (compounds 33, 34, and 35 in Table 1). Compounds with
4
-phenyl substituted thiophenyl and benzothiophenyl groups as the Ar moieties were

ACCEPTED MANUSCRIPT
also investigated. Results showed that the compound with the latter substitution (37)
displayed much higher inhibitory activity (IC = 24.0 nM) than the compound with
5
0
the former substitution (36, IC50 > 100 nM). The above results indicated that some
extension of the size of R substituent from the smallest hydrogen atom might be
critical to the augmentation of hSGLT2 inhibitory activities. Therefore, other
substituents such as fluorine atom and methyl group as R group were explored. The
results showed that only the methyl-substituted compound 39 displayed satisfactory
activity against hSGLT2 (IC = 4.5 nM), while in the other case, the IC value was
5
0
50
around 26.2 nM (Table 1). Consequently, it was inferred that the size of R group as
large as a chlorine atom or a methyl group just matched this novel scaffold; smaller R
groups were not preferred.
To conclude, the SAR studies showed that, i) an R group as large as a chlorine
atom or a methyl group was critical to improve the hSGLT2 inhibitory activities of
compounds derived from the novel scaffold of 6-deoxy O-spiroketal C-arylglucoside;
ii) the electron-withdrawing group substituted aromatic rings, arylthiophenyl and
aromatic fusing rings as the distal Ar moiety, were not quite favored; iii) alkyl or
alkoxy substituted aromatic rings were beneficial to improving hSGLT2 activities
when R was a chlorine atom or a methyl group.
Table 1. Structures and in vitro inhibitory activity against hSGLT2 and hSGLT1.
IC ± SD
5
0
a
Compd.
9
R
Ar
hSGLT2
nM)
hSGLT1 Selectivity
(
(µM)
b
H
H
> 100
ND
ND
351
10
63.5 ± 7.1 22.3 ± 4.3

ACCEPTED MANUSCRIPT
1
1
2
H
H
> 100
ND
ND
ND
1
> 100
ND
1
3
4
H
H
> 100
> 100
ND
ND
ND
ND
1
15
16
17
H
H
H
> 100
> 100
> 100
ND
ND
ND
ND
ND
ND
1
1
8
9
H
H
> 100
> 100
ND
ND
ND
ND
2
0
1
H
H
> 100
> 100
ND
ND
ND
ND
2
22
23
24
25
H
H
> 100
> 100
> 100
ND
ND
ND
ND
ND
ND
13
H
Cl
29.7 ± 18.4 0.4 ± 0.1
26
Cl
8.8 ± 1.6
0.5 ± 0.1
60
2
7
8
Cl
Cl
8.5 ± 0.8
0.4 ± 0.1
0.5 ± 0.1
46
30
2
15.9 ± 2.6
29
30
31
32
Cl
Cl
Cl
Cl
51.3 ± 14.2
47.2 ± 8.3
70
47
3
.6 ± 0.3
.2 ± 0.2
2
51.5 ± 11.7 3.0 ± 0.2
42.4 ± 11.9 6.4 ± 0.8
58
151

ACCEPTED MANUSCRIPT
33
34
35
Cl
Cl
Cl
69.4 ± 14.4 4.7 ± 0.3
68
71
67.5 ± 14.3 4.8 ± 0.4
42.3 ± 9.0
5.0 ± 0.5
117
36
Cl
> 100
ND
ND
37
38
39
Cl
F
24.0 ± 2.4
2.0 ± 0.1
81
26.2 ± 2.0 25.9 ± 4.3
988
216
Me
4.5 ± 0.8
1.0 ± 0.1
Dapagliflozin
-
-
-
-
8.3 ± 0.7
2.9
1
2.3 ± 1.1
8.44
1483
c
Tofogliflozin
2912
a
b
The selectivity values were calculated by IC hSGLT1/IC hSGLT2; ND refers to
5
0
50
c
Not Determined, similarly hereinafter; See ref 34.
Urinary Glucose Excretion in C57BL/6J Mice and Sprague-Dawley Rats
In the in vitro biological evaluation, compound 39 showed excellent in vitro
inhibitory activities against hSGLT2 (IC = 4.5 nM), and good selectivities against
5
0
hSGLT1 (selective ratio: 216-fold), which were comparable to the marketed drugs
dapagliflozin and tofogliflozin (Table 1). Therefore, compound 39 was selected for in
vivo biological evaluation.
We explored and compared the effects of 39 and tofogliflozin on urine volume,
urine glucose concentration and urinary glucose excretion (UGE) in C57BL/6J mice
after single oral administration at a dose of 1 mg/kg (Figure 3A-C and Table 2). The
amounts of 24-hour urinary glucose excretion caused by 39 and tofogliflozin were
110.53 mg/100 g and 48.85 mg/100 g, respectively (Figure 3C and Table 2),
indicating that compound 39 displayed higher SGLT2 inhibitory effect than
tofogliflozin. Similarly, 39 did not influence the urinary volume in in Sprague-Dawley
(
SD) rats (Figure 3D) but induced more excretion of urinary glucose (134.23 ± 46.72
mg/100 g) than tofogliflozin (73.68 ± 29.52 mg/100 g) on the same dose and duration
Figure 3F and Table 3).
(

ACCEPTED MANUSCRIPT
Figure 3. The effect of 39 on urinary glucose excretion during 24 h after oral
administration. The urinary volume (A, n=9), glucose concentration (B, n=9) and
glucose excretion (C, n=9) of 39 and tofogliflozin in C57BL/6J mice and the urinary
volume (D, n=6), glucose concentration (E, n=6) and glucose excretion (F, n=6) of 39
and tofogliflozin in SD rats were determined by the glucose-oxidase method. All
results are expressed as mean ± SEM. *, p value<0.05; **, p value<0.01; ***, p value
<
0.001 compared with the control vehicle group by two-tails t-test.
Table 2. Effects of single oral administration of 39 and tofogliflozin on urinary
glucose excretion in C57BL/6J mice during 24 h.

ACCEPTED MANUSCRIPT
Urine glucose
concentration
Urine
volume
(mL)
Urine glucose
excretion
Dose
Groups
Vehicle
(
mg/kg)
(mmol/L)
(mg/100 g)
0.5% MC
0.74±0.14
92.99±72.53
56.27±51.46
1.24±0.45
1.21±0.28
1.30±0.49
0.84±0.40
*
*
*
**
3
9
1
1
110.53±93.81
*
***
Tofogliflozin
48.85±24.34
All results are expressed as mean ± SD (n = 9). **, p value<0.01, ***, p value < 0.001;
compared with the vehicle group by two-tails t-test.
Table 3. Effects of single oral administration of 39 and tofogliflozin on urinary
glucose excretion in SD rats during 24 h.
Urine glucose
concentration
Urine
volume
Urine glucose
excretion
Dose
Groups
Vehicle
(
mg/kg)
(mmol/L)
(mL)
(mg/100 g)
0.51±0.40
0.5% MC
0.28±0.16
93.80±44.64
43.67±21.66
24.17±7.11
24.67±8.91
26.50±4.81
*
**
**
***
3
9
1
1
134.23±46.72
*
***
Tofogliflozin
73.68±29.52
All results are expressed as mean ± SD (n = 6). ***, p value < 0.001; compared with
the vehicle group by two-tails t-test.
Oral Glucose Tolerance Tests (OGTT) in C57BL/6J Mice
Based on the excellent effect of 39 on UGE in both mice and rats, it was
supposed that 39 should have good performance of improving glucose control.
Further investigation of compound 39 on its hypoglycemic effect on C57BL/6J mice
showed that single oral dose of compound 39 (3 mg/kg) significantly improved oral
glucose tolerance in mice in a dose-dependent manner after an oral glucose overload
(Figure 4A-B and Table 4). It is worth noting that compound 39 displayed more
remarkable improvement on oral glucose tolerance than tofogliflozin did in C57BL/6J
mice at the same dose (Figure 4A-B and Table 4), indicating its great potential of
treating diabetes.

ACCEPTED MANUSCRIPT
Figure 4. The effect of compound 39 on improving oral glucose tolerance after
single-dose administration in C57BL/6J mice. (A) Blood glucose levels were
determined 2 h before (-120 minutes) and at 0, 15, 30, 60, 90, and 120 minutes after
oral glucose challenge in C57BL/6J mice (n=8). (B) Areas under curves (AUC0-2 h) in
C57BL/6J mice were calculated by the trapezoidal rule. All results are expressed as
mean ± SEM (n = 8). *, # and +, p value<0.05; **, ## and ++, p value<0.01; ***, ###
and +++, p value < 0.001 compared with the control vehicle group one-way ANOVA
with Dunnett’s correction for OGTT or by two-tails t-test for AUC0-2 h. The
significance of 3 mg/kg 39 and 1 mg/kg and 3 mg/kg Tofo compared with vehicle
group are marked as *, # and + respectively. Tofo, tofogliflozin.
Table 4. Areas under the curve (AUC0-2 h) of blood glucose after single oral
administration of 39 and tofogliflozin in C57BL/6J mice with oral glucose challenge.
AUC0-2 h
Dose
mg/kg)
Groups
Vehicle
N
(
(mmol/L·h)
0.5% MC
8
8
8
8
8
23.95±2.33
27.63±4.90
1
3
1
3
39
***
16.73±1.04
22.02±1.81
Tofogliflozin
+
+
20.52±2.04
All results are expressed as mean ± SD (n = 8). ***, p value < 0.001 (3 mg/kg 39); ++,

ACCEPTED MANUSCRIPT
p value < 0.01 (3 mg/kg tofogliflozin) compared with the control vehicle group by
two-tails t-test.
Pharmacokinetic Profiles
Administration of a single intravenous dose of 2 mg/kg and a single oral dose of
1
0 mg/kg to mice revealed that compound 39 possessed the following specialties:
acceptable oral bioavailability (53.4%), longer half-life than tofogliflozin (5.95 h vs
.57 h, p.o.), and moderate exposure in plasma but extremely large steady-state
volume of distribution (Vss_obs > 10 L/kg). Clearly, the Vss_obs value was much greater
2
34
than the total blood volume in mice , indicating remarkable enrichment of 39 in
3
4
39
extravascular space (e.g., the target organ kidney ), which may explain its
advantage over tofogliflozin on promoting urinary glucose excretion and improving
glucose tolerance.
Table 5. Pharmacokinetic parameters of 39 after i.v. and p.o. administration in mice.
Admin.
i.v.
2
p.o.
10
Dose (mg/kg)
T1/2 (h)
10.8 ± 10.7
-
5.95 ± 2.32
0.33 ± 0.14
740 ± 294
3009 ± 408
5576 ± 1535
-
Tmax (h)
Cmax (ng/mL)
AUClast (h*ng/mL)
AUCINF_obs (h*ng/mL)
CL_obs (mL/min/kg)
MRTINF_obs (h)
Vss_obs (mL/kg)
F (%)
-
1128 ± 309
2322 ± 1087
19.9 ± 11.0
13.1 ± 13.0
10107 ± 3155
8.9 ± 3.4
-
53.4
CONCLUSION
In this work, we modifed the sugar moiety of SGLT2 inhibitors via removing the
C-6 hydroxy group of glucose, aiming to find a novel and potent SGLT2 inhibitor
with good pharmacokinetic profiles. A novel series of 6-deoxy O-spiroketal

ACCEPTED MANUSCRIPT
C-arylglucosides as SGLT2 inhibitors were designed, synthesized, and evaluated, and
the structure-activity relationship (SAR) of this novel series was demonstrated.
Among these newly designed and synthesized SGLT2 inhibitors, compound 39 was a
promising anti-diabetic agent with high inhibitory activity against hSGLT2, favorable
pharmacokinetic profiles, and excellent effects on facilitating the glucose outflow in
C57BL/6J mice and SD rats and improving glucose tolerance in C57BL/6J mice,
which were better than the positive control tofogliflozin.
ꢁ
EXPERIMENTAL SECTION
Chemistry
All reagents (chemicals) were commercially available and used without further
purification. Analytical thin-layer chromatography (TLC) was performed on HSGF
2
2
54 (0.1 mm thickness). Column chromatography was performed on silica gel
00–300 mesh to purify the compounds. NMR spectra were recorded on
Varian-MERCURY Plus-400 and AVANCE III 500 in CDCl and Methanol-d .
3
4
Chemical shifts were reported in parts per million (ppm, δ) downfield from
tetramethylsilane. Proton coupling patterns were described as s = singlet, d = doublet,
dd = doublet of doublet, t = triplet, q = quartet, m = multiplet. Low- and
high-resolution mass spectra (LRMS and HRMS) were given with electrospray
ionization (ESI).
All target compounds 9-39 were confirmed with over 95% purity which were
determined by Agilent 1260 HPLC with binary pump, photodiode array detector
(
DAD), using Agilent XDB-C18 (4.6 × 250 mm, 5 ꢀm), MeOH/H O = 80/20 (v/v) at
2
1
.0 mL/min, or Agilent XDB-C18 (4.6 × 250 mm, 5 ꢀm), MeOH/H O = 75/25 (v/v) at
2
1
.0 mL/min, or Agilent Extend-C18 (4.6 × 250 mm, 5 ꢀm), MeOH/H O = 75/25 (v/v)
2
at 1.0 mL/min and calculated the peak areas at 254 nm.
(
1S,3'R,4'S,5'S,6'R)-6'-methyl-6-(4-methylbenzyl)-3',4',5',6'-tetrahydro-3H-spiro[
isobenzofuran-1,2'-pyran]-3',4',5'-triol (9). To solution of
-bromo-4-methylbenzene (1.00 g, 5.85 mmol) in anhydrous THF (30 mL), 2.4 M
a
1

ACCEPTED MANUSCRIPT
n-BuLi in hexane solution (2.44 mL, 5.85 mmol) was added dropwise at −78 °C under
a nitrogen atmosphere and the resultant mixture was stirred under the same condition
for 1 h. A solution of compound 54 (322.0 mg, 0.58 mmol) in anhydrous THF (5 mL)
was then added dropwise to the resultant mixture. The reaction mixture was stirred for
2
h, and then water was added thereto. The resultant mixture was extracted with
EtOAc. The organic layer was washed with brine, dried over Na SO , and then
2
4
evaporated under reduced pressure to remove the solvent. The obtained residue was
purified by column chromatography on silica gel (petroleum ether/EtOAc = 4/1) to
give compound 55a (338.5 mg, 90%) as a colorless oil.
To a solution of compound 55a (338.0 mg, 0.53 mmol) in CH Cl (30 mL),
2
2
triethylsilane (419.9 ꢀL, 2.63 mmol) and BF ·OEt (71.4 ꢀL, 0.58 mmol) were added
3
2
at −40 °C and the mixture was stirred at the same temperature for 1 h. After addition
of water, the reaction mixture was extracted with CH Cl . The organic layer was
2
2
washed with brine, dried over Na SO , and then evaporated under reduced pressure to
2
4
remove the solvent. The obtained residue was purified by column chromatography on
silica gel (petroleum ether/EtOAc = 10/1) to give compound 56a (310.0 mg, 94%) as
a colorless oil.
To a solution of compound 56a (310.0 mg, 0.49 mmol) and pentamethylbenzene
(733.2 mg, 4.95 mmol) in CH Cl (30 mL) was added 1.0 M BCl in toluene solution
2 2 3
(4.95 mL, 4.95 mmol) at −78 °C under a nitrogen atmosphere, and the mixture was
stirred at the same temperature overnight. After addition of MeOH (10 mL), the
reaction mixture was warmed to rt and evaporated under reduced pressure to remove
the solvent. The obtained residue was purified by column chromatography on silica
gel (CH Cl /MeOH = 20/1) to give compound 9 (100 mg, 57%) as a white solid.
2
2
1
HPLC purity: 95.84%. m.p. 145–146 °C; H NMR (400 MHz, Methanol-d ) δ 7.24 –
4
7
2
.18 (m, 2H), 7.16 – 7.12 (m, 1H), 7.12 – 7.05 (m, 4H), 5.15 – 5.04 (m, 2H), 3.96 (s,
H), 3.91 – 3.81 (m, 1H), 3.77 – 3.66 (m, 2H), 3.18 – 3.10 (m, 1H), 2.29 (s, 3H), 1.20
1
3
(
d, J = 6.3 Hz, 3H). C NMR (125 MHz, Methanol-d ) δ 142.7, 140.3, 139.7, 139.3,
4
1
1
36.7, 131.1, 130.1, 129.9, 123.4, 121.8, 111.5, 77.4, 76.1, 75.2, 73.4, 71.4, 42.2, 21.1,
+
+
8.2. LRMS (ESI) m/z: 357 [M+H] ; HRMS (ESI) cacld for C H O Na [M+Na] :
2
1
24
5

ACCEPTED MANUSCRIPT
3
79.1516; found: 379.1510.
((2R,3R,4S,5R,6S)-3,4,5-tris(benzyloxy)-6-methoxytetrahydro-2H-pyran-2-yl)met
hanol (45). To a cooled solution (0 °C) of α-D-Methylglucoside 44 (20.00 g, 103.00
mmol) and imidazole (21.04 g, 308.99 mmol) in DMF (180 mL), TIPSCl (24.27 mL,
1
13.30 mmol) was added dropwise over a period of one hour. After 24 h at rt, the
reaction was diluted with water and extracted with CH Cl . The combined organic
2
2
layer was washed with brine, dried over Na SO , concentrated in vacuo. A solution of
2
4
this crude product and BnBr (61.17 mL, 514.98 mmol) in DMF (350 mL) was cooled
to 0 °C, and NaH (60% in mineral oil, 20.60 g, 514.98 mmol) was added. The
reaction mixture was allowed to warm to r.t. After 12 h, the reaction mixture was
carefully quenched by water. The reaction mixture was extracted with EtOAc. The
combined organic layer was washed with brine, dried over Na SO , and concentrated
2
4
in vacuo. The crude product was combined with TBAF (53.86 g, 205.99 mmol) and
THF (350 mL) was added. The reaction solution was stirred at rt for 12 h, diluted with
H O, and extracted with EtOAc. The combined organic layer was washed with brine,
2
dried over Na SO , concentrated in vacuo, and purified by column chromatography
2
4
on silica gel (petroleum ether/EtOAc = 2/1) to give compound 45 (37.50 g, 78%) (3
1
steps) as a colorless syrup. H NMR (400 MHz, Chloroform-d) δ 7.40 – 7.27 (m,
1
5H), 4.99 (d, J = 10.9 Hz, 1H), 4.92 – 4.77 (m, 3H), 4.70 – 4.61 (m, 2H), 4.56 (d, J
=
3.6 Hz, 1H), 4.01 (t, J = 9.3 Hz, 1H), 3.81 – 3.61 (m, 3H), 3.58 – 3.45 (m, 2H),
+
3
.37 (s, 3H). LRMS (ESI, m/z): 487 [M+Na] .
(
2S,3R,4S,5R,6R)-3,4,5-tris(benzyloxy)-2-methoxy-6-methyltetrahydro-2H-pyran
47). At 0 °C, Ph P (25.41 g, 96.87 mmol) and CBr (32.12 g, 96.87 mmol) were
(
3
4
added successively to a solution of 45 (30.00 g, 64.58 mmol) in THF (300ml). The
mixture was stirred for 1 h and then filtered. The filtrate was concentrated in vacuo
and purified by column chromatography on silica gel (petroleum ether/EtOAc = 10/1)
to give compound 46 (33.70 g, 99%) as a colorless syrup. LRMS (ESI, m/z): 549
+
[
M+Na] .

ACCEPTED MANUSCRIPT
To a solution of compound 46 (31.89 g, 60.46 mmol) in dry toluene (250 mL) at 20
o
C was added Bu SnH (19.45 mL, 72.55 mmol) and AIBN (992.82 mg, 6.05 mmol)
3
successively. The mixture was stirred for 4 h at 80 °C. The solution was evaporated to
syrup which was purified by column chromatography on silica gel (petroleum
1
ether/EtOAc = 10/1) to give compound 47 (23.59 g, 87%) as a colorless syrup. H
NMR (400 MHz, Chloroform-d) δ 7.42 – 7.22 (m, 15H), 4.98 (d, J = 10.9 Hz, 1H),
4
.90 (d, J = 10.9 Hz, 1H), 4.86 – 4.76 (m, 2H), 4.72 – 4.60 (m, 2H), 4.53 (d, J = 3.6
Hz, 1H), 3.95 (t, J = 9.3 Hz, 1H), 3.79 – 3.66 (m, 1H), 3.52 (dd, J = 9.7, 3.6 Hz, 1H),
3
.37 (s, 3H), 3.13 (t, J = 9.3 Hz, 1H), 1.24 (d, J = 6.3 Hz, 3H). LRMS (ESI, m/z): 471
+
[
M+Na] .
(3R,4S,5R,6R)-3,4,5-tris(benzyloxy)-6-methyltetrahydro-2H-pyran-2-one (49). To
a solution of compound 47 (15.10 g, 33.66 mmol) in glacial acetic acid (300 mL) was
o
added 3.0 M sulfuric acid (33.66 mL, 100.99 mmol). After stirring at 85 C for 2.5 h,
the reaction mixture was cooled to room temperature. CH Cl was added to the
2
2
system and the mixture was added sat. aq. NaHCO until no gas generated. The
3
organic layer was separated, washed with brine, dried over Na SO . Concentrated in
2
4
vacuo, and purified by column chromatography on silica gel (petroleum ether/EtOAc
=
2/1) to give compound 48 (13.01 g, 89%) as a white solid. LRMS (ESI, m/z): 457
+
[
M+Na] .
To a solution of compound 48 (25.20 g, 57.99 mmol) in DMSO (200 mL) was added
acetic acid anhydride (50 mL). The mixture was stirred overnight before diluted with
water, then sat. aq. NaHCO was added into the mixture until no gas generated. The
3
mixture was extracted with EtOAc. The combined organic layer was washed with
brine, dried over Na SO , concentrated in vacuo, and purified by column
2
4
chromatography on silica gel (petroleum ether/EtOAc = 10/1) to give compound 49
1
(
24.70 g, 98%) as a white solid. H NMR (400 MHz, Chloroform-d) δ 7.48 – 7.19 (m,
1
5H), 4.95 (d, J = 11.5 Hz, 1H), 4.74 – 4.61 (m, 3H), 4.61 – 4.48 (m, 3H), 4.12 (d, J
=
5.0 Hz, 1H), 3.97 – 3.87 (m, 1H), 3.46 (dd, J = 8.8, 5.7 Hz, 1H), 1.41 (d, J = 6.4 Hz,
+
3
H). LRMS (ESI, m/z): 433 [M+H] .

ACCEPTED MANUSCRIPT
2
-bromo-1,4-bis(((2-methoxypropan-2-yl)oxy)methyl)benzene (52). To a solution
of 2-bromo-terephthalic acid 50 (15.00 g, 61.22 mmol) in anhydrous THF (200 mL)
.
was added 2.0 M BH SMe in THF (91.83 mL, 183.65 mmol) at 0 °C. The mixture
3
2
was stirred for 2 h at 70 °C. After the exothermic reaction was over, the reaction was
cooled by ice-water bath. The reaction was quenched by dropwise addition of water.
The resulting mixture was extracted with EtOAc. The organic layer was washed with
brine and concentrated under reduced pressure to give compound 51 (11.90 g, 89%)
as an off-white solid.
To a solution of 51 (11.90 g, 54.82 mmol) in anhydrous THF (200 mL) was added
2
-methoxypropene (51.61 mL, 548.24 mmol) and pyridinium p-toluenesulfonate
o
(
275.55 mg, 1.10 mmol) at 0 °C, and the resultant mixture was stirred at 0 C for 2 h.
Sat. aq. NaHCO was added thereto, and then the resultant mixture was extracted with
3
EtOAc (500 mL) containing Et N (1.5 mL). The organic layer was washed with brine,
3
dried over Na SO , concentrated in vacuo, and purified by column chromatography
2
4
on silica gel (petroleum ether / EtOAc = 20/1) to give compound 52 (13.62 g, 69%) as
1
a colorless oil. H NMR (400 MHz, Chloroform-d) δ 7.54 (d, J = 1.3 Hz, 1H), 7.50 (d,
J = 7.9 Hz, 1H), 7.28 (dd, J = 7.9, 1.6 Hz, 1H), 4.53 (s, 2H), 4.44 (s, 2H), 3.24 (s, 3H),
3
.23 (s, 3H), 1.45 (s, 6H), 1.42 (s, 6H).
(1S,3'R,4'S,5'R,6'R)-3',4',5'-tris(benzyloxy)-6'-methyl-3',4',5',6'-tetrahydro-3H-s
piro[isobenzofuran-1,2'-pyran]-6-carbaldehyde (54). To a solution of compound 52
(10.00 g, 27.68 mmol) in anhydrous THF (100 mL), 2.4 M n-BuLi in hexane solution
(12.69 mL, 30.45 mmol) was added dropwise at −78 °C under a nitrogen atmosphere
and the resultant mixture was stirred under the same condition for 1 h. A solution of
compound 49 (10.54 g, 24.36 mmol) in anhydrous THF (50 mL) was then added
dropwise to the resultant mixture. The reaction mixture was stirred for 2 h, and then
water was added thereto. The resultant mixture was extracted with EtOAc. The
resultant organic layer was washed with water and brine and then dried over Na SO .
2
4
The solvent was then removed by distillation under reduced pressure. The obtained

ACCEPTED MANUSCRIPT
residue was dissolved in a mixed solvent of THF (100 mL) and MeOH (50 mL), and
p-toluenesulfonic acid (5.24 g, 30.45 mmol) was added thereto. The mixture was
stirred at rt for 15 h. After most of the MeOH was evaporated, the solution was
extracted with EtOAc. The combined organic layer was washed with sat. aq. NaHCO3,
brine, dried over Na SO , concentrated in vacuo, and purified by column
2
4
chromatography on silica gel (petroleum ether/EtOAc = 4/1) to give compound 53
+
(
7.67 g, 57%) as a colorless oil. LRMS (ESI, m/z): 553 [M+H] .
To a solution of compound 53 (11.20 g, 20.27 mmol) in CH Cl (120 mL), pyridinium
2
2
chlorochromate (PCC) (6.55 g, 30.40 mmol) and silica gel (200-300 mesh, 15 g) were
added at rt and the mixture was stirred for 4 h. After completion, the reaction mixture
was evaporated under reduced pressure to remove the solvent. The obtained residue
was purified by column chromatography on silica gel (petroleum ether/EtOAc = 10/1)
1
to give compound 54 (9.00 g, 81%) as a white solid. H NMR (400 MHz,
Chloroform-d) δ 9.86 (s, 1H), 7.88 (d, J = 7.7 Hz, 1H), 7.50 (s, 1H), 7.41 (d, J = 7.8
Hz, 1H), 7.38 – 7.27 (m, 10H), 7.13 (t, J = 7.2 Hz, 1H), 7.07 (t, J = 7.3 Hz, 2H), 6.77
(d, J = 7.3 Hz, 2H), 5.24 (s, 2H), 4.97 (d, J = 11.1 Hz, 3H), 4.73 (d, J = 10.9 Hz, 1H),
4
4
3
.65 (d, J = 11.4 Hz, 1H), 4.25 (d, J = 11.2 Hz, 1H), 4.15 (t, J = 9.3 Hz, 1H), 4.11 –
.01 (m, 1H), 3.92 (d, J = 9.5 Hz, 1H), 3.36 (t, J = 9.3 Hz, 1H), 1.26 (d, J = 6.2 Hz,
+
H). LRMS (ESI, m/z): 573 [M+Na] .
(1S,3'R,4'S,5'S,6'R)-6-(4-ethylbenzyl)-6'-methyl-3',4',5',6'-tetrahydro-3H-
spiro[isobenzofuran-1,2'-pyran]-3',4',5'-triol (10). Compound 10 was prepared
from 54 and 1-bromo-4-ethylbenzene according to the procedure described for
compound 9 in 38% yield (3 steps) as a white solid. HPLC purity: 96.43%. m.p.
1
1
7
–
36–138 °C; H NMR (400 MHz, Methanol-d ) δ 7.24 – 7.17 (m, 2H), 7.16 (s, 1H),
4
.11 (d, J = 1.8 Hz, 4H), 5.16 – 5.03 (m, 2H), 3.97 (s, 2H), 3.91 – 3.81 (m, 1H), 3.77
3.66 (m, 2H), 3.18 – 3.10 (m, 1H), 2.59 (q, J = 7.6 Hz, 2H), 1.24 – 1.16 (m, 6H).
1
3
C NMR (125 MHz, Methanol-d ) δ 143.2, 142.7, 140.3, 139.7, 139.6, 131.2, 129.9,
4
1
28.9, 123.4, 121.8, 111.5, 77.4, 76.1, 75.2, 73.5, 71.4, 42.2, 29.4, 18.2, 16.3. LRMS

ACCEPTED MANUSCRIPT
+
+
(
ESI) m/z: 371 [M+H] ; HRMS (ESI) cacld for C H O Na [M+Na] : 393.1672;
22 26 5
found: 393.1668.
(1S,3'R,4'S,5'S,6'R)-6'-methyl-6-(4-propylbenzyl)-3',4',5',6'-tetrahydro-3H-spiro[
isobenzofuran-1,2'-pyran]-3',4',5'-triol (11). Compound 11 was prepared from 54
and 1-bromo-4-propylbenzene according to the procedure described for compound 9
1
in 50% yield (3 steps) as a white solid. HPLC purity: 97.18%. mp 135–138 °C; H
NMR (400 MHz, Methanol-d ) δ 7.25 – 7.18 (m, 2H), 7.17 (s, 1H), 7.10 (q, J = 8.2
4
Hz, 4H), 5.14 – 5.05 (m, 2H), 3.97 (s, 2H), 3.92 – 3.80 (m, 1H), 3.77 – 3.67 (m, 2H),
3
3
1
7
.19 – 3.10 (m, 1H), 2.60 – 2.48 (m, 2H), 1.69 – 1.54 (m, 2H), 1.20 (d, J = 6.3 Hz,
1
3
H), 0.92 (t, J = 7.4 Hz, 3H). C NMR (100 MHz, Methanol-d ) δ 142.7, 141.5,
4
40.3, 139.7, 139.7, 131.2 129.8, 129.6, 123.4, 121.8, 111.5, 77.4, 76.1, 75.2, 73.5,
+
1.4, 42.2, 38.7, 25.8, 18.2, 14.1. LRMS (ESI) m/z: 385 [M+H] ; HRMS (ESI) cacld
+
for C H O Na [M+Na] : 407.1829; found: 407.1829.
2
3
28
5
(
1S,3'R,4'S,5'S,6'R)-6-(4-isopropylbenzyl)-6'-methyl-3',4',5',6'-tetrahydro-3H-spi
ro[isobenzofuran-1,2'-pyran]-3',4',5'-triol (12). Compound 12 was prepared from
4 and 1-bromo-4-isopropylbenzene according to the procedure described for
5
compound 9 in 56% yield (3 steps) as a white solid. HPLC purity: 95.26%. mp
1
1
7
2
34–136 °C; H NMR (400 MHz, Methanol-d ) δ 7.25 – 7.18 (m, 2H), 7.17 (s, 1H),
4
.13 (s, 4H), 5.14 – 5.05 (m, 2H), 3.97 (s, 2H), 3.91 – 3.82 (m, 1H), 3.78 – 3.67 (m,
13
H), 3.18 – 3.10 (m, 1H), 2.91 – 2.80 (m, 1H), 1.24 – 1.18 (m, 9H). C NMR (100
MHz, Methanol-d ) δ 147.8, 142.7, 140.3, 139.8, 139.7, 131.2, 129.9, 127.4, 123.4,
4
1
3
4
21.8, 111.5, 77.4, 76.1, 75.2, 73.5, 71.4, 42.2, 35.0, 24.5, 18.2. LRMS (ESI) m/z:
+
+
85 [M+H] ; HRMS (ESI) cacld for C H O Na [M+Na] : 407.1829; found:
2
3
28
5
07.1825.
(
1S,3'R,4'S,5'S,6'R)-6-(4-methoxybenzyl)-6'-methyl-3',4',5',6'-tetrahydro-3H-spir
o[isobenzofuran-1,2'-pyran]-3',4',5'-triol (13). Compound 56e was prepared from
4 and 1-bromo-4-methoxybenzene according to the procedure described for 56a in
5