Discovery of a Potent, Selective Renal Sodium-Dependent Glucose Cotransporter 2 (SGLT2) Inhibitor (HSK0935) for the Treatment of Type 2 Diabetes

Article abstract of DOI:10.1021/acs.jmedchem.6b01818

A new class of potent and highly selective SGLT2 inhibitors is disclosed. Compound 31 (HSK0935) demonstrated excellent hSGLT2 inhibition of 1.3 nM and a high hSGLT1/hSGLT2 selectivity of 843-fold. It showed robust urinary glucose excretion in Sprague-Dawley (SD) rats and affected more urinary glucose excretion in Rhesus monkeys. Finally, an efficient synthetic route has been developed featuring a ring-closing cascade reaction to incorporate a double ketal 1-methoxy-6,8-dioxabicyclo[3.2.1]octane ring system.

Full text of DOI:10.1021/acs.jmedchem.6b01818

Subscriber access provided by University of Colorado Boulder
Article
Discovery of a Potent, Selective Renal Sodium-
Dependent Glucose Cotransporter 2 (SGLT2) Inhibitor
(HSK0935) for the Treatment of Type 2 Diabetes
Yao Li, Zongjun Shi, Lei Chen, Suxin Zheng, Sheng Li, Bo
Xu, Zhenhong Liu, Jianyu Liu, Chongyang Deng, and Fei Ye
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 27 Apr 2017
Downloaded from http://pubs.acs.org on April 27, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of a Potent, Selective Renal Sodiumꢀ
Dependent Glucose Cotransporter 2 (SGLT2)
Inhibitor (HSK0935) for the Treatment of Type 2
Diabetes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Yao Li*, Zongjun Shi, Lei Chen, Suxin Zheng, Sheng Li, Bo Xu, Zhenhong Liu, Jianyu Liu,
Chongyang Deng, and Fei Ye
Haisco Pharmaceuticals Group Co. Ltd., 136 Baili Road, Wenjiang district, Chengdu, 611130,
China
ABSTRACT: A new class of potent and highly selective SGLT2 inhibitors is disclosed.
Compound 31 (HSK0935) demonstrated excellent hSGLT2 inhibition of 1.3 nM and a high
hSGLT1/hSGLT2 selectivity of 843 fold. It showed robust urinary glucose excretion in Spragueꢀ
Dawley (SD) rats and affected more urinary glucose excretion in Rhesus monkeys. Finally, an
efficient synthetic route has been developed featuring a ring closing cascade reaction to
incorporate a double ketal 1ꢀmethoxyꢀ6, 8ꢀ dioxabicyclo[3.2.1]octane ring system.
INTRODUCTION
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 42
1
2
3
4
5
6
7
8
9
The incidence of diabetes has been increasing at an alarming rate worldwide.1ꢀ3 About 90% of
diabetes is type 2 diabetes, which is characterized by chronically increased glycemic levels,
insulin resistance (IR) and βꢀcell dysfunction.4, 5 Currently, a wide range of antidiabetic agents
are prescribed for tight glycemic control, but many of them can’t achieve and maintain
satisfactory target glycemic levels. Thus, additional therapies with novel molecular mechanisms
are still in great demand. It is wellꢀknown that the kidneys play an important role in glucose
homeostasis via gluconeogenesis, glucose uptake from the circulation and glucose recovery from
the urine. Sodiumꢀdependent glucose cotransporters (SGLT) are a family of glucose transporters
and contribute to glucose reabsorption. The two most wellꢀknown members of SGLT family are
SGLT1 and SGLT2, which are members of the SLC5A gene family. In the kidney of rats,
approximately 90% of glucose reabsorption has been shown to occur in the site expressing the
lowꢀaffinity, highꢀcapacity SGLT2.5, 6 The remaining 10% of glucose is thought to be recovered
during passage of the filtrate through the site expressing the highꢀaffinity, lowꢀcapacity SGLT1.
Some recent studies indicate that SGLT2 is responsible for 97% of renal glucose recovery.7, 8
The case for selective inhibition of SGLT2 is bolstered by the fact that SGLT2 appears to be
expressed only in the kidney, while SGLT1 is mainly present in the small intestine (the
transporter responsible for absorption of both glucose and galactose) and in the heart (function
unknown). Moreover, individuals with a defective SGLT2 are marked only with significant
glycosuria (as much as 140 g daily) with no other ill effects,9, 10 while individuals expressing a
defective SGLT1 are unable to transport glucose or galactose normally across the intestinal wall,
resulting in the potentially lifeꢀthreatening condition known as glucoseꢀgalactose
malabsorption.11, 12 Therefore, selective SGLT2 inhibitors provide an attractive strategy for the
treatment of type 2 diabetes. Since 2013, leading SGLT2 inhibitors, such as dapagliflozin13 (1),
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 3 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
empagliflozin14 (2) and canagliflozin15 (3), have been approved by the FDA. Ipragliflozin16,
luseoglifozin17 and tofogliflozin18 were approved in Japan in 2014. The glucoseꢀlowering effect
of SGLT2 inhibitors is independent of insulin secretion or insulin action. These marketed SGLT2
inhibitors exhibit a modest reduction in body weight as well as blood pressure, and they are
generally well tolerated and have low risk of hypoglycemia. It is noteworthy that empagliflozin
was reported to show a significant reduction in both cardiovascular risk and cardiovascular death
in a dedicated outcome trial which has rarely been reported with other antidiabetic agents.19 Thus,
in recent years, although there are some concerns on the risks of ketoacidosis,20 SGLT2
inhibitors are considered to have more benefits than risks for diabetic patients, and have become
more and more commonly prescribed.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Selected examples of SGLT2 inhibitors
Over the last decade, many compounds have been reported as SGLT inhibitors. They are
generally divided into two classes, Oꢀglucosides and Cꢀglucosides.17, 21ꢀ22 The natural product,
phlorizin23 (4), is the first nonselective SGLT1/2 dual inhibitor. It was not clinically developed as
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 42
1
2
3
4
5
6
7
8
9
a drug candidate due to its poor metabolic stability to βꢀglucosidases. Optimization of phlorizin
led to the identification of several selective Oꢀaryl glucoside SGLT2 inhibitors such as Tꢀ
1095A24 (5), sergliflozinꢀA25 (6), remogliflozin26 (7) and remogliflozin etabonate27a, b (7a).
However, these Oꢀglucosides are usually associated with degradation by βꢀglucosidases found in
the gut as well. 5a, 6a and 7a were all discontinued after phase II trials for treating type 2
diabetes (7a is still active in phase II trials for the treatment of nonꢀalcoholic steatohepatitis
(NASH)).27c In contrast, Cꢀglucosides are generally more metabolically stable, and exhibit better
oral bioavailability and plasma exposure.28 The leading Cꢀglucosides, dapagliflozin and
canagliflozin, show good in vitro and in vivo potencies against SGLT2. Herein we report the
discovery of compound 31 (HSK0935), a potent and selective SGLT2 inhibitor.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. The hybrid design of the new dioxabicyclo[3.2.1]octane derivatives
RESULTS AND DISCUSSION
Chemistry. The design of SGLT inhibitors began with combination of the rigid and SGLT2
selective dioxabicyclo[3.2.1]octane motif of ertugliflozin29 (8) and the thiomethyl xyloside core
ACS Paragon Plus Environment

Page 5 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of sotagliflozin30 (9). To our knowledge when this project was initiated, ertugliflozin and
sotagliflozin were in phase II clinical trials for type 2 diabetes (now both are in phase III clinical
trials). Both compounds showed good results and possessed novel structures. Based on the
skeleton of the dual SGLT1/2 inhibitor sotagliflozin, a new class of highly selective SGLT2
inhibitors incorporating a structurally new 1ꢀmethoxyꢀ6,8ꢀdioxabicyclo[3.2.1]octane ring system
featured by a cascade ring closing reaction to form the double ketal Oꢀglycoside were designed
by introducing the dioxabicyclo[3.2.1]octane motif of ertugliflozin (Figure 2). The synthetic
methods are described as follows.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The known carboxylic acid 1031 was first converted to the corresponding acid chloride
(Scheme 1) and then directly reacted with sodium 2ꢀthioxopyridinꢀ1(2H)ꢀolate to furnish the
Barton ester 11. Compound 11 then underwent photochemical decarboxylation with a 500 W
halogen lamp and the radical intermediate was either trapped with MeSSMe to give
methylthioether 12 or trapped with its pyridylꢀ2ꢀthio radical intermediate to provide
pyridylthioether 13.32 Finally debenzylation reactions of compounds 12ꢀ13 were carried out in
the presence of BCl3 to furnish the thioether compounds 14ꢀ15, respectively.
Scheme 1. Synthesis of derivatives of 6, 8ꢀdioxabicyclo[3.2.1]octane with a thioether moiety
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 42
1
2
3
4
Reagents and conditions: (a) (COCl)2, DMF, rt to reflux, 2 h; (b) sodium 2ꢀthioxopyridinꢀ1(2H)ꢀ
5
6
olate, DCM, 0 ℃, 2 h; (c) 500 W halogen lamp, MeSSMe, 0 ℃, 1.5 h; (d) BCl3, DCM, ꢀ78 ℃,
7
8
9
1.5–2 h; (e) 500 W halogen lamp, DCM, 0 ℃, 1.5 h.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Target compounds 31ꢀ42 were synthesized from key intermediates 21aꢀd by one of three
strategies. As shown in Route 1 (Scheme 2), the known Cꢀglycosides33 16aꢀb were first iodinated
with triphenylphosphine and iodine to give the iodo intermediates 17aꢀb which underwent
dehydroiodination to provide alkenes 18aꢀb.34 Epoxidation of compounds 18aꢀb with mCPBA
led to epoxides 19a-b35 which underwent epoxide opening in the presence of trifluoroacetic acid
to provide the hydroxyl diketone intermediates 20a-b. The crude hydroxyl diketone
intermediates underwent a ring closure cascade in 2 M NaOH to produce the key 1ꢀhydroxyꢀ6,8ꢀ
dioxabicyclo[3.2.1]octane skeleton of intermediates 21a-b.36
Scheme 2. Route 1 synthesis of intermediate 21
ACS Paragon Plus Environment

Page 7 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
Reagents and conditions: (a) PPh3, imidazole, I2, toluene, 70 ℃, 3–5 h; (b) NaH, DMF, rt, 3 h; (c)
NaHCO3, mCPBA, DCM, rt, 2.5–4 h; (d) TFA, THF/H2O, rt, overnight; then 2 M NaOH, rt, 2 h.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Alternatively as shown in Route 2 (Scheme 3), hydroxyl diketone 20a was also prepared in
three simple steps. The dihydroxyl compound 24a was prepared by reacting the freshly prepared
Grignard reagent of bromide 23a with 3,4,5,6ꢀtetraꢀOꢀbenzylꢀDꢀglucopyranose 22. The resulting
dihydroxy compound 24a was oxidized using trifluoroacetic anhydride in DMSO to provide
diketone 25a. Since benzylꢀprotected primary alcohols are more susceptible to deprotection
under acidic conditions, selective debenzylation of 25a with BCl3 produced alcohol 20a in
acceptable yield.
Scheme 3. Route 2 synthesis of intermediate 21
Reagents and conditions: (a) Mg, I2, THF, 60 ℃, 3 h; (b) TFAA, DMSO, Et3N, DCM, ꢀ78 ℃, 2
h; (c) BCl3, DCM, ꢀ78 ℃, 3 h; (d) MeONa, MeOH, rt, 2 h.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 42
1
2
3
4
5
6
7
8
Furthermore, as shown in Route 3 (Scheme 4), selective debenzylation and acetylation of the
dihydroxy intermediates 24a,cꢀd led to the triacetoxy intermediates 26a,cꢀd. These were
hydrolyzed, selectively silylated and oxidized to diketone compounds 29a,cꢀd. Treatment of
29a,c-d with TBAF produced the desired intermediates 21a,cꢀd. The chemistry as described in
Route 3 allowed a more scalable synthesis for kilogram production.37
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Route 3 synthesis of intermediate 21
Reagents and conditions: (a) Mg, I2, THF, 60 ℃to 70 ℃, 3 h; (b) Ac2O, pꢀTsOH, 70 ℃, 2–3 h;
(c) K2CO3, MeOH, rt, 2 h; (d) TBSCl, imidazole, DCM, rt, monitored by TLC; (e) DMSO,
TFAA, Et3N, DCM, ꢀ78 ℃, 2 h; (f) Bu4NF, THF, ꢀ10 ℃to ꢀ5 ℃, 1 h.
With key intermediates 21a-d ready in hand, compounds 31-42 were synthesized (Scheme 5)
from compounds 30a-k respectively by debenzylation. Compounds 30a-k could be obtained by
alkylation of intermediates 21aꢀd.
ACS Paragon Plus Environment

Page 9 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
Scheme 5. Synthesis of compounds 31ꢀ42
7
8
9
OR1
R2
Cl
OR1
R2
OEt
Cl
Cl
O
O
HO
O
O
O
O
O
O
c
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
e
R3
NCH2C
OBn
BnO
41%
OBn
OH
BnO
HO
OBn
a
21a R1 = Et, R2 = H;
21b R1 = Et, R2 = F;
OBn
OH
38
30a 90% R1 = Et, R2 = H, R3 = Me;
30b 83% R1 = Et, R2 = H, R3 = Et;
30c 44% R1 = Et, R2 = H, R3 = CH2CN;
21c R1
21d R1
=
=
, R2 = H;
, R2 = H;
O
d
1%
O
30d 90% R1 = Et, R2 = F, R3 = Me;
30e 96% R1 = Et, R2 = F, R3 = Et;
30f 11% R1 = Et, R2 = F, R3 = iPr;
OEt
b
OR1
O
Cl
O
O
O
O
30g 27% R1
30h 80% R1
=
=
O
R2 = H, R3 = Me;
O
,
F2HC
OH
HO
R2 = H, R3 = Me;
,
OH
OBn
BnO
39
O
30i 47% R1 = Et
OBn
30j 35% R1
30k 56% R1
=
=
O
O
OR1
R2
Cl
b
O
O
O
R3
OH
HO
OR1
Cl
OH
O
O
O
31 (HSK0935) 85% R1 = Et, R2 = H, R3 = Me;
32 81% R1 = Et, R2 = H, R3 = Et;
F2HC
OH
HO
33 94% R1 = Et, R2 = F, R3 = Me;
34 86% R1 = Et, R2 = F, R3 = Et;
35 78% R1 = Et, R2 = F, R3 = iPr;
OH
40 90% R1 = Et
41 63% R1
42 63% R1
=
=
O
36 55% R1
37 95% R1
=
=
R2 = H, R3 = Me;
,
O
O
R2 = H, R3 = Me;
,
O
Reagents and conditions: (a) BrF2CPO3(Et)2, KOH, MeCN/H2O, ꢀ5 ℃ to rt, 1–3 h; (b) H2, 10%
Pd/C, 1,2ꢀdichlorobenzene, THF/MeOH, rt, 1–5 h; (c) alkylation reagents, NaH, DMF, rt, 1–3 h;
(d) H2, 10% Pd/C, THF/MeOH, rt, 5 h; (e) BCl3, DCM, ꢀ78 ℃, 4 h.
As shown in Scheme 6, compounds 45 and 49 were prepared from the common intermediate
ethyl ester 43 which was prepared by reacting the key intermediate 21a with ethyl 2ꢀ
bromoacetate. Reduction of compound 43 with LiBH4 gave hydroxyethyl compound 44 which
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 42
1
2
3
4
5
6
was debenzylated to produce compound 45. The cyclopropyl compound 49 was prepared from
alcohol 44 in four standard steps: iodination, elimination, cyclopropanation and debenzylation.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 6. Synthesis of compounds 45 and 49
Reagents and conditions: (a) BrCH2COOEt, NaH, DMF, rt, 3 h; (b) LiBH4, THF, rt, 5 h; (c) H2,
10% Pd/C, 1,2ꢀdichlorobenzene, THF/H2O, rt, 2 h; (d) PPh3, imidazole, I2, toluene, 70 ℃, 1 h;
(e) NaH, DMF, rt, 1 h; (f) Et2Zn, CH2I2, DCM, 0 ℃ to rt, 24 h; (g) H2, 10% Pd/C, 1,2ꢀ
dichlorobenzene, THF/MeOH, rt, 1 h.
Biological Evaluation. As shown in Table 1, compounds 14 and 15 where X = S and R1 =
methyl or 2ꢀpyridyl were found to have only moderate inhibitory activity (IC50 = 13.9 and 28.4
nM, respectively). However, when the sulfur atom of compound 14 was replaced by an oxygen
atom, compound 31 exhibited 10ꢀfold stronger inhibitory activity (IC50 = 1.3 nM) than
compound 14. Based on this observation, the X group was kept as an oxygen atom and the R1
ACS Paragon Plus Environment

Page 11 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
group was varied to explore the SAR. Initially, the small R1 groups such as ethyl, cyanomethyl,
difluoromethyl and cyclopropyl (compounds 32, 38, 40 and 49) were all well tolerated (IC50 =
2.5, 1.1, 1.0, and 4.4 nM, respectively). However, when R1 was elongated one more atom than
ethyl, such as hydroxyethyl (compound 45), the inhibitory potency was found to decrease to 31.0
nM. After exploration of R1, the R2, R3 and R4 substitutions on the bisaryl aglycon were further
investigated. The replacement of the chlorine atom with a hydrogen atom at the R2 position
decreased the SGLT2 inhibitory potency from 1.3 nM (compound 31) to 5.0 nM (compound 39).
Introduction of a fluorine atom at the R3 position was found to decrease the potencies as well
(compounds 33ꢀ35). Finally, when R4 group was modified from the smaller ethyl group
(compound 31) to larger groups, such as tetrahyrofuranꢀ3ꢀyl (compound 37) and 2ꢀ
cyclopropoxyethyl group (compound 36), the inhibitory potencies decreased to 5.9 nM and 10.8
nM, respectively. A similar trend was observed for compounds 40ꢀ42 as well.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. In Vitro Inhibitory Activity of hSGLT2a
hSGLT2
Compound
X
R1
R2
R3
R4
IC50 (nM)
13.9
S
S
CH3
Cl
Cl
H
H
Et
Et
14
15
28.4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 42
1
2
3
4
5
6
7
8
9
O
O
O
O
O
O
O
O
O
O
CH3
CH2CH3
CH2CN
CHF2
Cl
Cl
Cl
Cl
Cl
Cl
H
H
H
H
H
H
H
H
F
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
1.3
2.5
31
32
38
40
49
45
39
33
34
35
1.1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1.0
Cyclopropyl
CH2CH2OH
CH3
4.4
31.0
5.0
CH3
Cl
Cl
Cl
5.9
CH2CH3
iꢀPr
F
22.3
6.8
F
O
O
O
O
CH3
CH3
Cl
Cl
Cl
Cl
H
H
H
H
5.9
10.8
5.9
37
36
41
42
CHF2
CHF2
3.4
1.5–2.6
2.2
1
3
ACS Paragon Plus Environment

Page 13 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
2.4
1.5
8
9
7
8
9
aIn vitro human SGLT2 inhibition activities of compounds were determined at the
same laboratory by evaluating the sodiumꢀdependent uptake of methylꢀαꢀDꢀ[Uꢀ
14C]glucopyranoside in Chinese hamster ovary (CHO) cells stably expressing human
SGLT2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compounds 14, 15, 31 and 40 were further evaluated for selectivity against hSGLT1 by
measuring their in vitro hSGLT1 inhibitory activities (Table 2). Compounds 31 and 40 showed
better selectivity profiles. It is noteworthy that the selectivity of compound 31 was even better
than that of compound 1 (dapagliflozin). However, compounds 14 and 15 showed moderate
selectivities (32 and 39 fold respectively). This observation indicated that the replacement of the
sulfur atom at the X position with an oxygen atom could improve not only the inhibitory potency
of hSGLT2 but also the selectivity of hSGLT2 over hSGLT1.
Table 2. In vitro inhibitory activity and selectivitya
hSGLT2
IC50 (nM)
13.9
hSGLT1
IC50 (nM)
453
selectivity
Compound
hSGLT1/ hSGLT2
32
39
14
15
28.4
1082
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 42
1
2
3
4
5
6
7
8
9
1.3
1096
235
629
265
392
34.6
843
235
31
40
1
1.0
1.5–2.6
2.2
242–419
120
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
2.4
163
8
1.5
23
9
aIn vitro human SGLT1 and SGLT2 inhibition activities of compounds were determined
at the same laboratory by evaluating the sodiumꢀdependent uptake of methylꢀαꢀDꢀ[Uꢀ
14C]glucopyranoside in Chinese hamster ovary (CHO) cells stably expressing human
SGLT1 or SGLT2.
Compounds 31 and 40 with their better in vitro activities and greater selectivity for SGLT2
were further investigated for pharmacokinetic (PK) studies in SpragueꢀDawley (SD) rats. After
single oral or intravenous doses of compounds 31 and 40, blood samples were taken at planned
time points and the unchanged drugs in plasma were quantified using a LCꢀMS/MS method.
Major PK parameters were calculated using Winnolin software, version 6.3. As shown in Table 3,
compounds 31 and 40 presented acceptable oral PK profiles. Compound 31 showed better
systemic exposure and oral bioavailability than compound 40 (76% and 46%, respectively). And
the low CL demonstrated that the test compounds were metabolized and excreted very slowly.
Table 3. Pharmacokinetic parameters of compounds 31 and 40 in SD rats
ACS Paragon Plus Environment

Page 15 of 42
Journal of Medicinal Chemistry
1
2
3
4
Compound 31
Compound 40
5
6
7
8
9
p.o.
10
i.v.
1
p.o.
10
i.v.
1
Dose (mg/kg)
Cmax (ng/mL)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1826
758
5824
2.8
AUC 0→∞ (ng·h/mL) 15284
2007
3.4
1274
3.1
t1/2 (h)
3.6
CL (mL/min/kg)
F (%)
8.82
13.2
76
46
As shown in Figure 3A, single oral administrations of compounds 31 and 40 of 1, 3, and 10
mg/kg to SD rats induced doseꢀdependent urinary glucose excretion, resulting in a more than
1000ꢀfold elevation in glucosuria relative to the vehicle control. The ability of compound 31 to
increase urinary glucose excretion was similar to that of canagliflozin. Compound 40 was found
to be more efficacious than compound 31 and canagliflozin in this experiment. Afterwards, a
single dose of compounds 31 or 40 of 25 mg/kg was orally administrated to Rhesus monkeys. As
indicated in Figure 3B, compounds 31 and 40 could increase urinary glucose excretion robustly,
which was superior to that of canagliflozin. In addition, both compounds 31 and 40 could
maintain a comparatively high urinary glucose excretion level for 4 to 5 days similar to
dapagliflozin, while canagliflozin could maintain a high level of excretion for less than 3 days.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 42
1
2
3
4
These results indicated that both compounds 31 and 40 are potent SGLT2 inhibitors and exhibit a
5
6
long duration of action.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3A
3
10 mg/kg
3 mg/kg
1 mg/kg
2
1
0
3
31
40
Figure 3B
Monkey urinary glucose excretion
12
10
8
31
40
3
1
Vehicle
6
4
2
0
Time (h)
ACS Paragon Plus Environment

Page 17 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 3: Effects of oral administration of selected compounds on urinary glucose excretion in
SD rats and Rhesus monkeys. Animals were fasted overnight and orally administrated vehicle or
selected compounds at 1, 3, 10 mg/kg for SD rats (n = 3) and 25 mg/kg for Rhesus monkeys (n =
2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Furthermore, streptozotocin (STZ) and highꢀfat diet (HFD) induced Institute of Cancer
Research (ICR) mice were used to assess the antihyperglycemic effect in an oral glucose
tolerance test (OGTT). A single oral dose of compounds 31 or 40 (10 mg/kg) reduced the blood
glucose excursion when the compounds were administrated half an hour before an oral glucose
challenge. The area under the curve of blood glucoseꢀtime (AUC0ꢀ120min) showed that compounds
31 and 40 had statistically significant antihyperglycemic efficacies (Figure 4) compared with
vehicle treatment (P<0.001).
22
20
18
∗∗∗
16
14
12
10
8
∗∗∗
∗∗∗
6
Figure 4: Effects of SGLT2 inhibitors during an OGTT in STZ and HFD induced ICR mice.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 42
1
2
3
4
5
6
7
8
9
Animals were fasted overnight and orally given vehicle or SGLT2 inhibitors (10 mg/kg). All
animals received an oral glucose load (1 g/kg) half an hour later; areas under the curve of blood
glucoseꢀtime (0ꢀ120 min) are shown. Values are means ±SDs (n = 10). ***, P < 0.001.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Due to the good in vivo profiles, compounds 31 and 40 were selected for preclinical
development. They were tested for their ability to inhibit the hERG channel as part of the
evaluation of these compounds. Compounds 31 and 40 were not found to block the hERG
channel at concentrations up to 30 ꢁM in the manual wholeꢀcell patchꢀclamp assay. In a single
oral dose toxicology study in SD rats, the Maximum Tolerated Doses (MTDs) of the two
compounds were over 1000 mg/kg. Compound 31 was also evaluated in a 28ꢀday repeatꢀdose
toxicology study in Beagle dogs.38 It was found that compound 31 was well tolerated up to 300
mg/kg without any mortality or severe untoward effects being noted.
CONCLUSION
We have disclosed a new class of potent and highly selective SGLT2 inhibitors that feature a
novel double ketal dioxabicyclo[3.2.1]octane scaffold. Compound 31, which is currently
undergoing further preꢀclinical development for the treatment of type 2 diabetes, increased
urinary glucose excretion efficaciously and exhibited a long duration of action in Rhesus
monkeys.
EXPERIMENTAL SECTION
In Vitro hSGLTs Uptake Assays. The CHO cells stably expressing human SGLT2 or
SGLT1 were seeded into 96ꢀwell plates (Corning, NY) at a density of 30,000 cells/well and
ACS Paragon Plus Environment

Page 19 of 42
Journal of Medicinal Chemistry
1
2
3
4
incubated for 48 h in a 5% CO2 atmosphere at 37 ℃ in growth medium (1 : 1 Fꢀ12/DMEM
5
6
7
8
media and 10% FBS). The culture medium was removed and the cells were washed three times
with 200 ꢁL of KRH solution (120 mM NaCl, 4.7 mM KCl, 1.2 mM MgCl2, 2.2 mM CaCl2, and
10 mM HEPES, pH 7.4). Then incubated in KRH solution containing 3 ꢁM methylꢀαꢀDꢀ[Uꢀ14C]ꢀ
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
glucopyranoside ([14C]AMG) in the absence or presence of inhibitors for up to 120 min at 37 ℃.
After that, the KRH solution was removed and the wells were rinsed three times with 200 ꢁL of
iceꢀcold KRH solution. The cells were lysed in 0.1% sodium dodecyl sulfate (Sigma). After 24 h,
plates were quantitated in a TopCount (PerkinꢀElmer) for counting of radioactive [14C]AMG.
The percent effect of compounds to inhibit AMG uptake was calculated by comparing counts per
minute (CPM) in inhibitorꢀcontaining wells with in DMSO wells. The IC50 values were fitted to
a sigmoidal doseꢀresponse model using Origin 8.0 software.
Oral Glucose-Tolerance Test in STZ and HFD Induced Mice. Male ICR mice of 4 weeks
of age were obtained from Vital River (Beijing, China), fed with high fat diet (15% fat by
weight). All mice were given access to food and water ad libitum. At 7 weeks of age, all mice
were injected with a single dose of 50 mg/kg of streptozotocin (STZ). The diet induced obesity
(DIO) mice at 8 weeks of age (n = 10/group) were randomly assigned to treatment groups and
fasted overnight. Mice were then treated orally with vehicle (0.5% methylcellulose) or 10 mg/kg
of 3, 31, and 40 respectively. Half an hour after treatment (t = 0 min), blood was taken via tail
nick and glucose concentration was measured with a glucometer. Also at 30 minutes after
treatment, the mice were orally challenged with glucose (1 g/kg, 10 mL/kg). Blood samples for
glucose measurement were obtained at 15, 30, 60, and 120 min after the glucose load. The blood
glucose excursion profile from t = 0 to t = 120 min was used to integrate an area under the curve
(AUC) for each treatment.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 42
1
2
3
4
5
6
7
8
9
In Vivo Pharmacokinetics Evaluation in SD Rats. Male SpragueꢀDawley rats of 8 weeks of
age were obtained from Vital River (Beijing, China). Compounds were administered
intravenously to three male rats at 1 mg/kg or orally at 10 mg/kg dose. Blood samples (∼200 ꢁL)
were collected from each animal via the jugularꢀvein cannula at 0, 5 (iv only), 15, and 30 min
and at 1, 2, 4, 8, 12 and 24 h after dosing. Plasma was separated from the blood by centrifugation
at 5000 g for 15 min at 4 ℃ and stored in a freezer (ꢀ80 ℃). All samples were analyzed for the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
test compounds by LCꢀMS/MS (ABI 4000Q Trap). The concentration data were analyzed with a
standard noncompartmental method with Kinetica (version 4.1.1, InnaPhase Corporation, PA).
Urinary Glucose Excretion in Normal SD Rats and Rhesus Monkeys. Normal male
SpragueꢀDawley rats of 8 weeks of age were obtained from Vital River (Beijing, China). Male
SD rats were overnightꢀfasted and divided into 3 rats per group. The rats were treated orally with
a single dose of compounds in different doses (1, 3, and 10 mg/kg) or vehicle respectively. Half
an hour later, the rats were orally challenged with a 50% glucose solution at a dose of 2 g/kg.
The animals were refed at 1 h after glucose challenge, and urine was collected at 24 h. After the
urine volumes were measured, the urine samples were measured for glucose levels by using a
DRIꢀCHEM 3500s (FUJI, Tokyo, Japan).
Male Rhesus monkeys of 2 years age were obtained from Sichuan Primed BioꢀTech Group
Co., Ltd (Chengdu, China). The animals were overnightꢀfasted and divided into 2 monkeys per
group. The monkeys were treated orally with a single dose of the compounds in 25 mg/kg or
vehicle respectively. Half an hour later, the monkeys were orally challenged with a 50% glucose
solution at a dose of 2 g/kg. The monkeys were refed at 1 h after glucose challenge, and urine
over 24 h was collected. After the urine volumes were measured, the urine samples were
measured for glucose levels by using a DRIꢀCHEM 3500s (FUJI, Tokyo, Japan).
ACS Paragon Plus Environment

Page 21 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Ethics Statement All the animal experimental studies were performed in accordance with the
Guide for the Care and Use of Laboratory Animals, formulated by the State Council of the
People's Republic of China. And it was approved by the Office of the Experimental Animal
Management Committee of Sichuan Province.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General Procedures All purchased starting materials were used without further purification.
13
19
1H NMR, C NMR and F NMR spectra were acquired on a Bruker Avanceꢀ400 spectrometer
(400 MHz) or a Bruker Avanceꢀ300 spectrometer (300 MHz), with tetramethylsilane (TMS) as
an internal standard; chemical shifts are expressed in parts per million (ppm, δ units). Mass
spectra were obtained on a Finnigan LCQAd instrument (ESI) and Agilent 6120 (APCI). Most
masses were reported as those of the protonated parent ions. Preparative column chromatography
was performed using Yantai Huanghai 200–300 mesh silica. Highꢀresolution mass spectra
(HRMS) were recorded on an Apex III 7.0 T FTMS from Bruker Daltonics, Inc. (U.S.). High
Performance Liquid Chromatography (HPLC) was performed on an Agilent1260
(chromatographic column (Agilent Zorbax SBꢀC18 4.6×100mm, 3.5ꢁm)). Rotation was
performed on Automatic Polarimeter SGWꢀ3. Melting point was obtained on an OptiMelt
MPA100. All compounds submitted for in vitro testing were >95% purity (HPLC) and those for
in vivo testing were >98% purity (HPLC).
General Procedure A: for the preparation of compounds 14, 15 and 38, taking the preparation
for compound 14 as an example.
(1R,2S,3R,4R,5S)-5-(4-Chloro-3-(4-ethoxybenzyl)phenyl)-1-(methylthio)-6,8-
dioxabicyclo[3.2.1]octane-2,3,4-triol (14). To a solution of 12 (640.0 mg, 0.89 mmol) in DCM
(5 mL), was added boron trichloride (17.7 mL, 1 M in DCM, 17.70 mmol) at ꢀ78 ℃. Upon
completion of the addition, the mixture was stirred for 2 h. The resulting mixture was quenched
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 42
1
2
3
4
5
6
7
8
9
by the dropwise addition of a solution of DCM and MeOH (20 mL, v / v = 1 : 1). Water (20 mL)
was then added. The water layer was extracted with DCM (20 mL × 2). The combined organic
extracts were dried (Na2SO4), filtered and concentrated. The crude product was purified by
column chromatography (DCM / MeOH (v / v) = 40 : 1 ~ 30 : 1, Rf = 0.1) to afford compound
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
14 as a white solid (55.0 mg, yield 14%, HPLC: 95.72% (210 nm)). H NMR (400 MHz,
CD3OD): δ 7.90–7.84 (m, 2 H, ArH), 7.54 (d, J = 8.3 Hz, 1 H, ArH), 7.12 (d, J = 8.5 Hz, 2 H,
ArH), 6.85 (d, J = 8.3 Hz, 2 H, ArH), 5.63 (d, J = 6.6 Hz, 1 H, CH), 4.67 (dd, J = 6.6, 4.5 Hz, 1
H, CH), 4.41 (d, J = 4.5 Hz, 1 H, CH), 4.10 (s, 2 H, CH2), 4.04–3.95 (m, 3 H, CH2, CH), 3.74 (d,
13
J = 12.2 Hz, 1 H, CH), 2.02 (s, 3 H, CH3), 1.37 (t, J = 7.0 Hz, 3 H, CH3); C NMR (101 MHz,
CD3OD): δ 196.4, 157.7, 139.9, 139.4, 134.6, 130.7, 130.7, 129.6, 129.5, 127.5, 114.3, 95.1, 80.9,
79.1, 78.7, 63.1, 37.7, 13.8, 8.3. LCꢀMS m/z (ESI): 475.0 [M + Na]+, HRMS: calc.
C22H29ClNO6S (M + NH4)+ 470.1404, found 470.1394.
(1R,2S,3R,4R,5S)-5-(4-Chloro-3-(4-ethoxybenzyl)phenyl)-1-(pyridin-2-ylthio)-6,8-
dioxabicyclo[3.2.1]octane-2,3,4-triol (15). Compound 13 was treated following general
procedure A to afford compound 15 as a white solid (20.0 mg, yield 23%, HPLC: 96.31% (210
nm)). 1H NMR (400 MHz, CD3OD): δ 8.38 (dt, J = 4.9, 1.4 Hz, 1 H, pyridine ArH), 7.57 (dd, J =
5.0, 1.5 Hz, 2 H, ArH), 7.34–7.15 (m, 4 H, ArH), 7.00 (d, J = 8.7 Hz, 2 H, ArH), 6.73 (d, J = 8.7
Hz, 2 H, ArH), 4.32 (d, J = 7.9 Hz, 1 H, CH), 3.94 (d, J = 2.6 Hz, 2 H, CH2), 3.90 (q, J = 7.0 Hz,
2 H, CH2), 3.61 (d, J = 8.1 Hz, 2 H, CH, CH), 3.56 (t, J = 7.9 Hz, 1 H, CH), 3.47 (d, J = 7.6 Hz,
1 H, CH), 1.27 (t, J = 7.0 Hz, 3 H, CH3); 13C NMR (101 MHz, DMSOꢀd6): δ 157.4, 154.7, 150.3,
138.4, 137.6, 137.3, 133.5, 131.5, 130.1, 129.5, 129.1, 127.6, 126.7, 122.8, 114.8, 108.8, 92.6,
77.2, 76.3, 74.6, 69.0, 63.4, 38.1, 15.1. LCꢀMS m/z (ESI): 516.0 [M + H]+, HRMS: calc.
C26H30ClN2O6S (M + NH4)+ 533.1513, found 533.1525.
ACS Paragon Plus Environment

Page 23 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(1S,2S,3S,4R,5S)-2,3,4-Tris(benzyloxy)-5-(4-chloro-3-(4-ethoxybenzyl)phenyl)-6,8-
ioxabicyclo[3.2.1]octan-1-ol (21a). A solution of tetraꢀnꢀbutylammonium fluoride (1.2 kg) in
THF (3 L) was added dropwise to a solution of 29a in THF (6 L) at ꢀ10 ℃. After the addition
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
was complete the temperature was allowed to reach ꢀ5 ℃. The mixture was stirred for 1 h,
poured into water (3 L) and extracted with EA (3 L × 3). The combined organic extracts were
washed with brine (5 L), dried (Na2SO4) and concentrated. The crude product was purified by
column chromatography (PE / EA (v / v) = 10 : 1, Rf = 0.2) to give compound 21a as a light
25
yellow solid (480.0 g, yield 33% for 3 steps, HPLC: 96.15% (210 nm)). M.P.: 135.9 ℃, [α]D
=
1
ꢀ44 (c = 0.1, MeOH). H NMR (300 MHz, CDCl3): δ 7.14–7.37 (m, 16 H, ArH), 7.08 (d, J = 6
Hz, 2 H, ArH), 6.88 (d, J = 6 Hz, 2 H, ArH), 6.77 (d, J = 9 Hz, 2 H, ArH), 4.80–5.01 (m, 4 H,
CH2), 4.43 (d, J = 6 Hz, 1 H, CH), 4.30 (d, J = 9 Hz, 1 H, CH), 3.89–4.03 (m, 8 H, CH, CH2),
3.68 (d, J = 6 Hz, 1 H, CH2), 3.54 (d, J = 9 Hz, 1 H, CH), 1.41 (t, J = 6 Hz, 3 H, CH3). LCꢀMS
m/z (APCI): 715.0 [M + Na]+.
(1S,2S,3R,4R)-2,3,4,6-Tetrakis(benzyloxy)-1-(4-chloro-3-(4-ethoxybenzyl)phenyl)hexane-
1,5-diol (24a). To a suspension of Mg powder (88.0 g, 3.66 mol) and iodine (a catalytic amount)
in anhydrous THF (200 mL) was slowly added 23a (1.1 kg, 3.36 mol) in anhydrous THF (1 L).
The reaction mixture was kept at 60 ℃for 1 h. In the meantime isopropylmagnesiumbromide (1
M in THF, 2.77 mol) was added dropwise to the solution of 22 (1.5 kg, 2.77 mol) in anhydrous
THF (2 L) at 0 ℃. The freshly prepared Grignard reagent was then added dropwise to the
solution of 22 at 0 ℃. The reaction mixture was kept at 60 ℃ for 2 h and was quenched by the
addition of 2 M hydrochloric acid at 0 ℃. The mixture was extracted with EA (1.2 L × 3). The
combined organic extracts were washed with brine (3 L), dried (Na2SO4) and concentrated to
give crude product 24a as a light yellow syrup. 1H NMR (400 MHz, CDCl3): δ 7.40–7.23 (m, 24
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 42
1
2
3
4
5
6
7
8
H), 7.19 (dd, J = 7.2, 2.1 Hz, 3 H), 4.90 (ddd, J = 32.4, 16.3, 7.0 Hz, 4 H), 4.73–4.51 (m, 5 H),
4.07–3.91 (m, 4 H), 3.82–3.64 (m, 4 H), 3.58 (d, J = 11.3 Hz, 1 H), 3.46 (d, J = 11.3 Hz, 1 H),
1.27 (d, J = 12.5 Hz, 3 H). LCꢀMS m/z (APCI): 787.3 [M + H]+.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(2S,3R,4R,5S,6R)-3,4,5-Tris(benzyloxy)-6-(4-chloro-3-(4-ethoxybenzyl)phenyl)hexane
1,2,6-triyl triacetate (26a). Crude 24a (2.18 kg) was dissolved in acetic anhydride (5 L) and pꢀ
TsOH (590.0 g) was added at 0 ℃. The mixture was stirred for 3 h at 70 ℃. The reaction
mixture was cooled to room temperature and poured into ice water (3 L). The aqueous layer was
extracted with EA (2 L × 3). The combined organic extracts were washed with sodium
bicarbonate solution (2 L × 2) and water (2 L × 1), dried (Na2SO4) and concentrated. The residue
(2.28 kg) was directly used for the next step without purification. 1H NMR (400 MHz, CDCl3): δ
7.37–7.15 (m, 15 H), 7.09–6.93 (m, 5 H), 6.77–6.67 (m, 2 H), 4.77–4.42 (m, 6 H), 4.34–4.15 (m,
2 H), 4.13–3.74 (m, 8 H), 3.61 (m, 1 H), 2.04–1.87 (m, 9 H), 1.36 (t, J = 7.0 Hz, 3 H). LCꢀMS
m/z (APCI): 823.3 [M + H]+.
(3R,4R,5S)-3,4,5-Tris(benzyloxy)-6-(4-chloro-3-(4-ethoxybenzyl)phenyl)hexane-1,2,6-
triol (27a). Potassium carbonate (1.5 kg) was added in batches to a solution of 26a (2.28 kg) in
MeOH (4.7 L) and water (0.2 L) at 0 ℃. The mixture was allowed to reach room temperature
and stirred for 2 h. The resulting mixture was filtered and the filtrate was poured into saturated
NH4Cl solution (5 L) and concentrated. The residue was extracted with EA (2 L × 3). The
combined organic extracts were washed with brine (2 L), dried (Na2SO4) and concentrated. The
crude product was purified by column chromatography (PE / EA (v / v) = 4 : 1, Rf = 0.1) to give
1
compound 27a as a light yellow syrup (1.5 kg, yield 70% for 3 steps). H NMR (400 MHz,
CDCl3): δ 7.39–7.17 (m, 15 H), 7.12–6.96 (m, 5 H), 6.81–6.66 (m, 2 H), 4.60–4.43 (m, 5 H),
4.22 (d, J = 10.8 Hz, 1 H), 4.12 (q, J = 7.1 Hz, 1 H), 3.92 (ddd, J = 12.5, 12.0, 7.4 Hz, 6 H), 3.73
ACS Paragon Plus Environment

Page 25 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
(td, J = 6.3, 3.3 Hz, 2 H), 3.64 (dd, J = 11.5, 4.4 Hz, 1 H), 3.42 (dd, J = 9.1, 5.0 Hz, 1 H), 1.28–
1.21 (m, 3 H). LCꢀMS m/z (ESI): 697.0 [M + H]+.
7
8
9
(2S,3R,4R)-2,3,4-Tris(benzyloxy)-6-((tert-butyldimethylsilyl)oxy)-1-(4-chloro-3-(4-ethoxy
benzyl)phenyl)hexane-1,5-diol (28a). 27a (1.5 kg, 2.10 mol) was treated under a nitrogen
atmosphere with TBSCl (366.0 g, 2.31 mol), imidazole (266.0 g, 3.15 mol) and DCM (4 L) at
room temperature. After completion of reaction, the mixture was poured into ice water (4 L). The
mixture was extracted with DCM (2 L × 2). The organic layer was dried (Na2SO4), concentrated,
and the residue was directly used for the next step without purification.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(2R,3R,4S)-2,3,4-Tris(benzyloxy)-6-((tert-butyldimethylsilyl)oxy)-1-(4-chloro-3-(4-
ethoxybenzyl)phenyl)hexane-1,5-dione (29a). A solution of dimethylsulfoxide (971 mL) in
DCM (2.8 L) was added dropwise to a solution of trifluoroacetic anhydride (1424 mL) in DCM
(1.4 L) at ꢀ78 ℃. 28a in DCM (2.3 L) was then added. After stirring for 1 h at ꢀ78 ℃,
triethylamine (2530 mL, 18.20 mol) in DCM (1.4 L) was added, the temperature was allowed to
reach room temperature and the mixture was poured into water (3 L). The mixture was extracted
with DCM (4 L × 2). The organic layer was dried (Na2SO4), concentrated, and the residue was
directly used for the next step without purification.
(1S,2S,3R,4R,5S)-2,3,4-Tris(benzyloxy)-5-(4-chloro-3-(4-ethoxybenzyl)phenyl)-1-
methoxy-6,8-dioxabicyclo[3.2.1]octane (30a). To a solution of 21a (330.0 g, 0.48 mol) in DMF
(1.5 L) was added iodomethane (101.0 g, 0.72 mol) under nitrogen atmosphere at 0 ℃. Then
NaH (14.0 g, 0.61 mol) was added in batches and the mixture was stirred for 1 h at room
temperature under nitrogen atmosphere. The resulting mixture was adjusted to pH = 7 with
saturated NH4Cl solution. The water phase was extracted with EA (2 L × 3). The combined
organic extracts were washed with brine (5 L × 1) and water (5 L × 1), dried (Na2SO4) and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 42
1
2
3
4
5
6
7
8
9
concentrated. The crude product was purified by column chromatography (PE / EA (v / v) = 15 :
1, Rf = 0.2) to give compound 30a as a light yellow oil (303.0 g, yield 90%). 1H NMR (300 MHz,
CDCl3): δ 7.13–7.35 (m, 16 H, ArH), 7.06 (d, J = 6 Hz, 2 H, ArH), 6.85 (d, J = 6 Hz, 2 H, ArH),
6.75 (d, J = 9 Hz, 2 H, ArH), 4.74–4.95 (m, 4 H, CH2), 4.27 (d, J = 9 Hz, 1 H, CH2), 4.20 (d, J =
9 Hz, 1 H, CH2), 3.80–4.01 (m, 8 H, CH, CH2), 3.62 (m, 1 H, CH), 3.43 (s, 3 H, OCH3), 1.38 (t,
J = 6 Hz, 3 H, CH3). LCꢀMS m/z (APCI): 707.2 [M + H]+.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(1R,2S,3R,4R,5S)-2,3,4-Tris(benzyloxy)-5-(4-chloro-3-(4-ethoxybenzyl)phenyl)-1-
(difluoromethoxy)-6,8-dioxabicyclo[3.2.1]octane (30i). To a solution of 21a (12.5 g, 18.06
mmol) in acetonitrile (150 mL) and water (25 mL) was added potassium hydroxide (12.1 g,
216.07 mmol) at ꢀ5 ℃ in an iceꢀwater bath. Bromodifluoromethane diethyl phosphate (14.3 g,
53.56 mmol) was added and the reaction mixture was stirred for 3 h at room temperature. The
resulting mixture was quenched with water (100 mL) and extracted with EA (200 mL × 3). The
combined organic extracts were washed with saturated NH4Cl solution (150 mL × 3) and water
(100 mL × 2), dried (Na2SO4), filtered and concentrated. The crude product was purified by
column chromatography (PE / EA (v / v) = 30 : 1, Rf = 0.2) to afford compound 30i as a light
yellow oil (6.3 g, yield 47%, HPLC: 96.30% (210 nm)). [α]D = ꢀ37.5 (c = 0.1, MeOH). 1H
25
NMR (400 MHz, CDCl3): δ 7.39–7.27 (m, 11 H, ArH), 7.25–7.10 (m, 5 H, ArH), 7.05 (t, J = 8.6
Hz, 2 H, ArH), 6.83 (m, 2 H, ArH), 6.78–6.71 (m, 2 H, ArH), 6.71–6.37 (t, J = 67.6 Hz, 1 H,
CF2H), 4.95 (d, J = 11.1 Hz, 1 H, CH), 4.80 (dd, J = 33.1, 11.0 Hz, 3 H, CH,CH2), 4.45 (d, J =
8.7 Hz, 1 H, CH), 4.26 (d, J = 10.9 Hz, 1 H, CH), 4.09–3.82 (m, 7 H, CH, CH2), 3.78 (d, J = 8.6
Hz, 1 H, CH), 3.66 (d, J = 7.8 Hz, 1 H, CH), 1.38 (t, J = 7.0 Hz, 3 H, CH3). 19F NMR (376 MHz,
CDCl3): δ ꢀ79.51, ꢀ79.97, ꢀ84.03, ꢀ84.49. LCꢀMS m/z (APCI): 743.2 [M + H]+.
ACS Paragon Plus Environment

Page 27 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
General Procedure B: The preparation of compounds 31-37, 40-42, 45, 49, followed the
method for compound 31.
7
8
9
(1S,2S,3R,4R,5S)-5-(4-Chloro-3-(4-ethoxybenzyl)phenyl)-1-methoxy-6,8-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
dioxabicyclo[3.2.1]octane-2,3,4-triol (31). Compound 30a (275.0 g, 0.39 mol) was dissolved in
MeOH / THF (1.3 L, v / v = 1 : 1), 1,2ꢀdichlorobenzene (573.0 g, 3.91 mol) and 10% Pd/C
(220.0 g) were added and the mixture was stirred for 16 h under hydrogen atmosphere at room
temperature. The reaction mixture was filtered and washed with MeOH/DCM (1 L, v / v = 1 : 1).
The organic filtrate was concentrated. The crude product was purified by column
chromatography (DCM / MeOH (v / v) = 20 : 1, Rf = 0.1) to give compound 31 as a white solid
(144.0 g, yield 85%, HPLC: 98.50% (210 nm)). M.P.: 47.6 ℃. [α]D25 = ꢀ23 (c = 0.1, MeOH). 1H
NMR (400 MHz, CDCl3): δ 7.38 (d, J = 1.7 Hz, 1 H), 7.28 (d, J = 8.3 Hz, 1 H), 7.23 (d, J = 1.9
Hz, 1 H), 7.03 (d, J = 8.6 Hz, 2 H), 6.73 (d, J = 8.6 Hz, 2 H), 4.04 (s, 3 H, OH), 4.01 – 3.95 (m, 3
H), 3.90 (q, J = 7.0 Hz, 2 H), 3.85 (d, J = 8.0 Hz, 1 H), 3.74 (t, J = 8.0 Hz, 1 H), 3.66 (d, J = 7.8
Hz, 1 H), 3.61 (d, J = 7.9 Hz, 1 H), 3.32 (s, 3 H), 1.32 (t, J = 7.0 Hz, 3 H); 13C NMR (101 MHz,
CD3OD): δ 157.5, 138.5, 136.8, 133.8, 131.3, 129.5, 128.8, 128.5, 125.4, 114.1, 106.3, 106.1,
78.1, 75.9, 72.6, 65.3, 63.1, 50.0, 37.9, 13.8. LCꢀMS m/z (ESI): 459.0 [M + Na]+, HRMS: calc.
C22H29ClNO7 (M + NH4)+ 454.1633, found 454.1623.
(1S,2S,3R,4R,5S)-5-(4-Chloro-3-(4-ethoxybenzyl)phenyl)-1-ethoxy-6,8-
dioxabicyclo[3.2.1]octane-2,3,4-triol (32). Prepared from compound 30b following general
procedure B to afford compound 32 as a white solid (0.2 g, yield 83%, HPLC: 95.17% (210 nm)).
1H NMR (400 MHz, CD3OD): δ 7.40 (s, 1 H, ArH), 7.38–7.33 (m, 2 H, ArH), 7.08 (d, J = 8.7 Hz,
2 H, ArH), 6.79 (d, J = 8.7 Hz, 2 H, ArH), 4.10 (d, J = 7.9 Hz, 1 H, CH), 4.02 (s, 2 H, CH2), 3.97
(q, J = 7.0 Hz, 2 H, CH2), 3.85–3.78 (m, 2 H, CH), 3.69 (m, 1 H, CH), 3.63–3.57 (m, 2 H, CH2),
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 42
1
2
3
4
5
6
7
8
9
3.52 (d, J = 7.9 Hz, 1 H, CH), 1.34 (t, J = 7.9 Hz, 3 H, CH3), 1.19 (t, J = 7.0 Hz, 3 H, CH3); 13C
NMR (101 MHz, CD3OD): δ 157.5, 138.5, 136.8, 133.8, 131.3, 129.5, 128.8, 128.5, 125.4, 114.1,
106.3, 106.0, 78.1, 75.9, 72.6, 65.3, 63.1, 49.9, 37.9, 14.5, 13.8. LCꢀMS m/z (ESI): 405.0 [M ꢀ
EtO]+, HRMS: calc. C23H31ClNO7 (M + NH4)+ 468.1789, found 468.1778.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(1S,2S,3R,4R,5S)-5-(4-Chloro-3-(4-ethoxy-3-fluorobenzyl)phenyl)-1-methoxy-6,8-
dioxabicyclo[3.2.1]octane-2,3,4-triol (33). Prepared from compound 30d following general
procedure method B to afford compound 33 as a white solid (0.2 g, yield 94%, HPLC: 97.81%
(210 nm)). M.P.: 47.9 ℃. [α]D25 = ꢀ22.5 (c = 0.1, MeOH). 1H NMR (400 MHz, CD3OD): δ 7.43
(s, 1 H, ArH), 7.42–7.36 (m, 2 H, ArH), 6.99–6.87 (m, 3 H, ArH), 4.11–4.02 (m, 5 H, CH,CH2),
3.84 (m, 1 H, CH), 3.68–3.56 (m, 2 H, CH2), 3.53 (d, J = 7.9 Hz, 1 H), 3.46 (s, 3 H, OCH3), 1.38
19
13
(t, J = 7.0 Hz, 3 H, CH3); F NMR (376 MHz, CD3OD): δ ꢀ136.78, C NMR (101 MHz,
CD3OD): δ 153.7–151.2 (ArCF), 145.2, 137.8, 136.9, 132.7, 128.9, 128.7, 125.8, 124.3, 116.1,
115.9, 114.9, 106.3, 106.1, 78.0, 75.9, 72.6, 65.3, 64.7, 50.1, 37.8, 13.8. LCꢀMS m/z (ESI): 423.0
[M ꢀ MeO]+, HRMS: calc. C22H28ClFNO7 (M + NH4) + 472.1538, found 472.1527.
(1S,2S,3R,4R,5S)-5-(4-Chloro-3-(4-ethoxy-3-fluorobenzyl)phenyl)-1-ethoxy-6,8-
dioxabicyclo[3.2.1]octane-2,3,4-triol (34). Prepared from compound 30e following general
procedure B to afford compound 34 as a white solid (0.2 g, yield 86%, HPLC: 98.81% (210 nm)).
1H NMR (400 MHz, CD3OD): δ 7.42 (s, 1 H, ArH), 7.41–7.35 (m, 2 H, ArH), 6.96 (t, J = 8.3 Hz,
1 H, ArH), 6.90 (d, J = 10.4 Hz, 2 H, ArH), 4.12–4.08 (m, 1 H, CH), 4.07 (m, 4 H, CH2), 3.85–
3.80 (m, 2 H, CH2), 3.75–3.66 (m, 1 H, CH), 3.64–3.57 (m, 2 H, CH2), 3.52 (d, J = 7.9 Hz, 1 H,
19
CH), 1.38 (t, J = 7.0 Hz, 3 H, CH3), 1.20 (t, J = 7.1 Hz, 3 H, CH3); F NMR (376 MHz,
CD3OD): δ ꢀ136.78; 13C NMR (101 MHz, CD3OD): δ 153.7–151.3 (ArCF), 145.2, 137.7, 137.0,
133.8, 132.8, 128.9, 128.6, 125.8, 124.3, 116.0, 114.8, 106.2, 106.0, 78.0, 75.9, 73.0, 65.9, 64.7,
ACS Paragon Plus Environment

Page 29 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
58.9, 37.8, 14.5, 13.8. LCꢀMS m/z (ESI): 423.0 [M ꢀ EtO]+, HRMS: calc. C23H30ClFNO7 (M +
NH4)+ 486.1695, found 486.1683.
6
7
8
9
(1S,2S,3R,4R,5S)-5-(4-Chloro-3-(4-ethoxy-3-fluorobenzyl)phenyl)-1-isopropoxy-6,8-
dioxabicyclo[3.2.1]octane-2,3,4-triol (35). Compound 30f was treated following general
procedure B to afford compound 35 as a colorless syrup. (25.0 mg, yield 78%, HPLC: 97.94%
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
(210 nm)). H NMR (400 MHz, CD3OD): δ 7.38 (d, J = 7.6 Hz, 3 H, ArH), 7.02–6.79 (m, 3 H,
ArH), 4.21 (dt, J = 12.3, 6.1 Hz, 1 H, CH), 4.13–3.97 (m, 5 H, CH2, CH2, CH), 3.75 (dd, J = 8.2,
1.7 Hz, 1 H, CH), 3.65–3.54 (m, 2 H, CH, CH), 3.49 (d, J = 7.9 Hz, 1 H, CH), 1.38 (t, J = 7.0 Hz,
13
3 H, CH3), 1.19 (d, J = 6.1 Hz, 3H, CH3), 1.16 (d, J = 6.2 Hz, 3 H, CH3), C NMR (101 MHz,
CDCl3) δ 153.8–151.3 (ArCF), 145.3, 138.1, 135.8, 134.7, 132.3, 129.4, 128.7, 125.7, 124.4,
116.6, 116.5, 114.9, 106.0, 105.4, 78.2, 75.6, 74.6, 67.1, 65.0, 38.4, 24.1, 23.8, 14.8. LCꢀMS m/z
(ESI): 423.1 [M ꢀ OCH(CH3)2]+, HRMS: calc. C24H32ClFNO7 (M + NH4)+ 500.1851, found
500.1838.
(1S,2S,3R,4R,5S)-5-(4-Chloro-3-(4-(2-cyclopropoxyethoxy)benzyl)phenyl)-1-methoxy-
6,8-dioxabicyclo[3.2.1]octane-2,3,4-triol (36). Compound 30g was treated following general
procedure B to compound 36 as a white solid (30.0 mg, yield 55%, HPLC: 95.40% (210 nm)).
M.P.: 48.5 ℃. [α]D25 = ꢀ25 (c = 0.1, MeOH). 1H NMR (400 MHz, CDCl3): δ 7.33–7.41 (m, 3 H,
ArH), 7.09 (d, J = 8.7 Hz, 2 H, ArH), 6.83–6.72 (m, 2 H, ArH), 4.11–4.05 (m, 5 H, CH, CH2),
3.95 (dd, J = 8.2, 1.8 Hz, 1 H, CH), 3.85–3.83 (m, 2 H, CH2), 3.79–3.71 (m, 3 H, CH2), 3.47 (s, 3
13
H, OCH3), 3.39 (m, 1 H, CH), 0.63–0.60 (m, 2 H, CH2), 0.50–0.47 (m, 2 H, CH2); C NMR
(101 MHz, CD3OD): δ 157.3, 138.4, 136.8, 133.8, 131.7, 129.5, 128.8, 128.5, 125.5, 114.2,
106.3, 106.05, 78.1, 75.9, 72.6, 68.8, 66.9, 65.3, 53.0,49.9, 37.9, 4.5. LCꢀMS m/z (ESI): 515.0
[M + Na]+, HRMS: calc. C25H33ClNO8 (M + NH4)+ 510.1895, found 510.1880.
ACS Paragon Plus Environment