Design, synthesis and biological evaluation of (2S,3R,4R,5S,6R)-5-fluoro-6-(hydroxymethyl)-2-aryltetrahydro-2H-pyran-3,4-diols as potent and orally active SGLT dual inhibitors

Article abstract of DOI:10.1016/j.bmcl.2018.09.025

A new series of (2S,3R,4R,5S,6R)-5-fluoro-6-(hydroxymethyl)-2-aryltetrahydro-2H-pyran-3,4-diols as dual inhibitors of sodium glucose co-transporter proteins (SGLTs) were disclosed. Two methods were developed to efficiently synthesize C₅-fluoro-

Full text of DOI:10.1016/j.bmcl.2018.09.025

Accepted Manuscript
Design, Synthesis and Biological Evaluation of (2S,3R,4R,5S,6R)-5-Fluoro-6-
(hydroxymethyl)-2-aryltetrahydro-2H-pyran-3,4-diols as Potent and Orally Ac-
tive SGLT Dual Inhibitors
Guozhang Xu, Michael D. Gaul, Gee-Hong Kuo, Fuyong Du, June Zhi Xu,
Nathaniel Wallace, Simon Hinke, Thomas Kirchner, Jose Silva, Norman D.
Huebert, Seunghun Lee, William Murray, Yin Liang, Keith Demarest
PII:
S0960-894X(18)30756-X
https://doi.org/10.1016/j.bmcl.2018.09.025
BMCL 26046
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
15 July 2018
13 September 2018
17 September 2018
Please cite this article as: Xu, G., Gaul, M.D., Kuo, G-H., Du, F., Zhi Xu, J., Wallace, N., Hinke, S., Kirchner, T.,
Silva, J., Huebert, N.D., Lee, S., Murray, W., Liang, Y., Demarest, K., Design, Synthesis and Biological Evaluation
of (2S,3R,4R,5S,6R)-5-Fluoro-6-(hydroxymethyl)-2-aryltetrahydro-2H-pyran-3,4-diols as Potent and Orally
Active SGLT Dual Inhibitors, Bioorganic & Medicinal Chemistry Letters (2018), doi: https://doi.org/10.1016/
j.bmcl.2018.09.025
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.

ABSTRACT: A new series of (2S,3R,4R,5S,6R)-5-fluoro-6-(hydroxymethyl)-2-
aryltetrahydro-2H-pyran-3,4-diols as dual inhibitors of sodium glucose co-transporter
proteins (SGLTs) were disclosed. Two methods were developed to efficiently synthesize
C5-fluoro-lactones 3 and 4, which are key intermediates to the C5-fluoro-hexose based C-
aryl glucosides. Compound 2b demonstrated potent hSGLT1 and hSGLT2 inhibition
(IC50 = 43 nM for SGLT1 and IC50 = 9 nM for SGLT2). It showed robust inhibition of
blood glucose excursion in oral glucose tolerance test (OGTT) in Sprague Dawley (SD)
rats and exerted pronounced antihyperglycemic effects in db/db mice and high-fat diet-
fed ZDF rats when dosed orally at 10 mg/kg.
Keywords
Diabetes
Glucose transporter
SGLT1 inhibitor
SGLT2 inhibitor
Bioisostere
C-Aryl Glucoside
Corresponding Author. Tel.: +1 215 628 7940
E-mail address: gxu4@its.jnj.com (G. Xu)
Janssen Research & Development
Cardiovascular & Metabolic Disease
SH42-2314C
1400 McKean Road
Spring House, PA 19477

Design, Synthesis and Biological Evaluation of (2S,3R,4R,5S,6R)-5-Fluoro-6-
(hydroxymethyl)-2-aryltetrahydro-2H-pyran-3,4-diols as Potent and Orally Active
SGLT Dual Inhibitors
Guozhang Xu*, Michael D. Gaul, Gee-Hong Kuo, Fuyong Du, June Zhi Xu, Nathaniel
Wallace, Simon Hinke, Thomas Kirchner, Jose Silva, Norman D. Huebert, Seunghun
Lee, William Murray, Yin Liang, Keith Demarest
Persistent hyperglycemia is a defining feature of diabetes mellitus (DM), which falls
into two broad etiopathgenic categories-type 1 and type 2 diabetes.1-2 The United States
Centers for Disease Control (CDC) states that in 2015 diabetes affected 30.3 million
people, nearly 9.4% of the total U.S. population.3 The primary focus for management of
T2DM is to achieve recommended levels of glucose control and thereby slow the
progression of microvascular and macrovascular complications.4 Despite the availability
of insulin and many oral antidiabetic agents, therapeutic efficacy is also offset by side
effects such as weight gain and hypoglycemia.5-6 Therefore, the search for additional
therapeutic agents with an improved benefit-risk profile continues.
It is well-known that the human kidney plays a significant role in maintaining glucose
homeostasis of the body.7-8 Sodium-dependent glucose cotransporters (SGLTs) are a
family of glucose transporters that contribute to glucose reabsorption. Two major
isoforms of SGLT family are SGLT1 and SGLT2, among which SGLT2 is a low affinity,
high capacity glucose transporter that is predominately expressed in the S1 and S2
segments of the early proximal tubule of the kidney.9-10 Studies indicated that SGLT2 is
responsible for 90% of renal glucose recovery.11 Sodium glucose transporter 1 (SGLT1)
is a high affinity, low capacity glucose transporter and highly expressed in the proximal
portion of the small intestine. It is also expressed in the S3 segment of the proximal
tubule of the kidney as well as the heart.12-14
There has been a tremendous effort over the past decades to develop selective SGLT2
inhibitors as a new treatment for diabetes via disposal of excess glucose in the urine
through inhibition of glucose reabsorption in the kidney.15 Several SGLT2 selective
inhibitors have been approved by the FDA for the treatment of diabetes. Canagliflozin
(Invokana®) was approved in 2013,16 followed by dapagliflozin17 (Farxiga®) and
empagliflozin18 (Jardiance®). In addition, ipragliflozin,19 luseogliflozin,20 and
tofogliflozin 21have all been approved in Japan. Due to the progressive nature of T2DM, a
potent SGLT1 and SGLT2 dual inhibitor will reduce calorie intake and achieve a more
systematic normoglycemia via two distinct SGLT mediated events: reducing dietary
glucose absorption in the intestine and increasing urinary glucose excretion from the
kidney. Sotagliflozin, a dual SGLT1 and SGLT2 inhibitor, is currently in late-stage
clinical development for diabetes.22-23
Herein, we designed, synthesized, and evaluated the human SGLT1/2 inhibitory
potential of a new series of (2S,3R,4R,5S,6R)-5-fluoro-6-(hydroxymethyl)-2-

aryltetrahydro-2H-pyran-3,4-diols. Further investigation of the structure-activity
relationships of this series resulted in the discovery of 2b, as a potent dual SGLT inhibitor
with IC50 of 43 nM for SGLT1 and IC50 of 9 nM for SGLT2. Compound 2b, when dosed
orally, caused a dose-dependent inhibition of glucose excursion in oral glucose tolerance
test (OGTT) in Sprague Dawley (SD) rats and exerted significant and long-lasting
antihyperglycemic effects in db/db mice and high-fat diet-fed Zucker Diabetic Fatty
(ZDF) rats at 10 mg/kg dosing.
At the beginning of our SGLT1/2 dual inhibitor program, we realized that most of the
reported SGLT inhibitors were derived from C-aryl glucosides (shown as 1 in Figure 1
with a diarylmethane structure as aglycone) due to the relative ease of synthesis. It is well
documented that fluorine can act as an isostere of hydroxy group since the C-F bond
length (1.39 Å) is quite close to the C-O bond length (1.43 Å).24 In addition, early SAR
studies indicated that the C5 site of the C-aryl glucosides is tolerate for structural
modification.25 Our design strategy was therefore directed to the bioisosteric replacement
of C5-hydroxy group with a fluorine atom with the same equatorial configuration as
original hydroxy group to construct a C5-fluoro-hexose-based SGLT inhibitor such as 2
(Ar shown in the blue box represents a generic aglycone moiety). Interestingly, there
were very few literature reports of orally active SGLT1 and SGLT2 dual inhibitors based
on this chemotype.
Figure 1. Design of new SGLT1 and SGLT2 dual inhibitors
A retrosynthetic analysis of target molecule 2 (Figure 2) indicated that the C5-
fluorolactone 3 or 4 with appropriate protecting group such as TBDMS or benzyl (Bn)
was highly desirable as a key intermediate for the target synthesis.

Figure 2. Retrosynthetic analysis of C5-fluoro-hexose-based aryl glucosides
To that end, two methods were developed for the syntheses of these C5-fluoro-lactone
intermediates. As shown in Scheme 1 for the first approach, treatment of commercially
available (2S,3R,4S,5R,6R)-2-(allyloxy)-6-(hydroxymethyl)tetrahydro-2H-pyran-3,4,5-
triol 5 with 3.1 equivalents of benzoyl chloride in pyridine at -35oC afforded C-2,3,6-
tribenzoyl-protected 6 in 73% yield. Conversion of the axial C5-hydroxyl group to the
corresponding equatorial fluoride was accomplished in the presence of
diethylaminosulfur trifluoride (DAST)/dichloromethane (DCM) at -40oC in 70% yield.
The benzoyl-protected 7 was hydrolyzed in NaOMe/MeOH to provide 8, which
subsequently reacted with excess TBDMSOTf in 2,6-lutidine/DCM to afford 9 in
excellent yield over two steps. Removal of the allyl protecting group of 9 was achieved
using a two-step process:26 (1) carbon-carbon double isomerization in the presence of
(PPh3)2RhCl/DABCO under refluxing condition; (2) bond cleavage in the presence of
NMO/OsO4 at room temperature to afford 10 in 87% yield. The resulting lactol 10 was
oxidized in Ac2O/DMSO to provide a key intermediate 3 as a stable white solid in high
yield. To our knowledge, this is the first-time report for the synthesis of this TBDMS-
protected C5-fluoro-lactone. This six-step process also allowed us to synthesize the key
intermediate 3 in large quantity.
Scheme 1. Reagents and conditions: (a) 3.1 eq. BzCl, pyridine, 35oC to room temperature, 73%; (b)
DAST/DCM, -40oC to room temperature, 70% yield; (c) NaOMe, MeOH, room temperature, quantitative
yield; (d) TBDMSOTs, 2,6-lutidine/DCM, 0oC to 4 oC, 96%; (e) cat. (PPh3)3RhCl, DABCO, EtOH/water
(10:1 v/v), 80-100oC, NMO/cat. OsO4, acetone/water (10:1 v/v), 87%: (f) DMSO/Ac2O, 84-88%.

Our second approach was to synthesize benzyl-protected C5-fluoro-lactone 4, which
started with C5-fluoro-substituted glucose 11 (Scheme 2). All the free hydroxyl groups
were acetylated in the presence of NaOAc/Ac2O. Treatment of 12 with excess HBr in
HOAc afforded α–bromo analog 13. Nucleophilic substitution of 13 with p-tolylthiol
yielded 14 in 47% yield over three steps, which underwent hydrolysis (15), benzylation
(16), and oxidative removal of p-tolylthio with NBS in acetone/water to afford lactol 17.
The resulting lactol 17 was then oxidized under Swern oxidation condition to provide
lactone intermediate 4 in 87%.
Scheme 2. Reagents and conditions: (a) NaOAc, Ac2O 100%; (b) HBr in HOAc; (c) 4-
methylbenzenethiol/Et3N, DCM 47% over two steps; (d) NaOMe, MeOH; (e) NaH, BnBr, n-Bu4NI
95%: (f) NBS, acetone/water 86%; g) DMSO, Ac2O 87%.
With both C5-fluoro-lactones 3 and 4 in hand, we first synthesized targets 2a-2f
containing a fused dihydrobenzofuran as the central moiety of aglycone (Scheme 3).
Bromination of known 2,3-dihydrobenzofuran-4-ol (shown in the box) with PyBr3,
followed by iodination with NIS provided 5-bromo-7-iodo-2,3-dihydrobenzofuran-4-ol
18 in 60% yield over two steps. Interestingly, reversing the order of additions led to the
formation of 7-bromo-5-iodo-2,3-dihydrobenzofuran-4-ol 22. Benzylation of either 18 or
22 with benzyl bromide in the presence of K2CO3 produced 19 or 23 respectively, which
reacted with Turbo Grignard reagent (i-PrMgCl.LiCl) at low temperature to undergo
selective metal/iodo exchange. The resulting dihydrobenzofuran-derivatized lithium
species reacted with aryl aldehyde, followed by reduction with triethylsilane and boron
trifluoride diethyl etherate to generate the corresponding aglycone 20 or 24 in 70%
overall yield. The addition of the aryl lithium species generated by lithium/halogen
exchange of aglycone 20 or 24 to C5-fluoro-lactone 3 generated the corresponding lactol,
which was subjected to further triethylsilane/BF3.Et2O reduction in DCM/acetonitrile at -
30oC to room temperature to provide the corresponding -C-aryl glucoside 21 or 25.
Removal of the benzyl groups of 21 or 25 under the hydrogenolysis condition employing
20% Pd(OH)2 as catalyst afforded the corresponding 2a or 2d. Further methylation gave
the target compounds 2b-c as well as 2e-f.

Scheme 3. Reagents and conditions: (a) PyBr3, MeOH 60%; (b) NIS, acetonitrile 100%; (c)
Benzylbromide/K2CO3, acetone/reflux, 86%; (d) Aryl aldehyde such as 2,3-dihydrobenzo[b][1,4]dioxine-6-
carbaldehyde, i-PrMgCl.LiCl, 70%; (e) Triethylsilane/BF3.Et2O, DCM, 98%: f) BuLi, C5-fluorolactone
3/THF, -78oC; g) Triethylsilane, BF3.Et2O, DCM, acetonitrile, -30oC-0oC, 50% over two steps; h) 20%
Pd(OH)2, H2, EtOAc, MeOH, 94%; i) MeI, K2CO3, acetone, 70%.
Scheme 4 illustrated the construction of bicyclic 6,7-dihydrothieno[3,2-c]pyridine-
derived C5-fluoro-glucosides 2g-o. Starting from 1-(benzyloxy)-2-bromo-4-
(dimethoxymethyl)-5-methylbenzene 26, lithium/bromide exchange with n-butyllithium,
generated aryllithium, which reacted with benzyl-protected C5-fluoro-lactone 4, followed
by acid hydrolysis provided aldehyde 27 in 68% yield over two steps. A second
lithium/bromide exchange of tert-butyl 2-bromo-6,7-dihydrothieno[3,2-c]pyridine-5(4H)-
carboxylate with n-butyllithium, followed by the addition of the lithiated thiophene to
intermediate 27 provided the corresponding 28, which was a mixture of diastereomers.
Without further separation, compound 28 was reduced using triethylsilane and BF3.Et2O
to furnish 2-substituted-4,5,6,7-tetrahydrothieno[3,2-c]pyridine 29 in 65% isolated yield.
Derivatization of 29 with either cyanic bromide or various carbonylchloride yield benzyl-
protected 30. Removal of the benzyl groups with BCl3 in the presence of
pentamethylbenzene gave the desired target molecules such as 2h, 2k, 2m-n respectively.
Methylation of either 2h or 2k with MeI and K2CO3 in acetone afforded the
corresponding analog 2i or 2l. Similarly, analog 2g or 2o (not shown in the scheme 4)
was made starting from 4-bromo-2-(dimethoxymethyl)-1-methylbenzene.

Scheme 4. Reagents and conditions: (a) n-Buli,, then C5-fluoro-lactone 4, THF at -78oC, 2 h; b) 1N HCl,
THF, rt, 2 h, 68% yield two steps; (c) n-BuLi,, THF, then tert-butyl 2-bromo-6,7-dihydrothieno[3,2-
c]pyridine-5(4H)-carboxylate, at -78oC, 2h, 87% (d) Triethylsilane/BF3.Et2O, DCM, at 0oC for 1 h, 65%;
e) R1Br such as cyanic bromide or R1Cl such as ethoxy carbonylchloride/DIEA, DCM; f) BCl3,
pentylmethylbenzene, DCM, at -78oC for 30 mins, 42%; g) MeI, K2CO3, acetone.
The in vitro inhibitory activities of newly synthesized compounds on SGLT1 and
SGLT2 were assessed by monitoring the inhibition of accumulating methyl-α-D-[U-14C]-
glucopyranoside (AMG) in CHO-K1 cells stably expressing human SGLT1 or SGLT2
(See details in supporting material). Triplicate determinations were made for each test
compound and the IC50 values or inhibition activities of the test compounds against
SGLT1 and SGLT2 were determined.
We first investigated a series of 2,3-dihydrobenzo[b][1,4]dioxines and oxathiines with
2,3-dihydrobenzofuran as the central moiety and the data were summarized in Table 1.
The potency of dapagliflozin was also included for comparison. Compound 2a with
ortho-hydroxy group was very potent for both SGLT1 (IC50 = 14 nM) and SGLT2 (IC50 =
3 nM) while methylation of the phenol group (2b) resulted in 3-fold loss of activity for
both SGLT1 and SGLT2. Replacement of distal 2,3-dihydrobenzo[b][1,4]dioxine with
2,3-dihydrobenzo[b][1,4]oxathiine (2c) reduced SGLT1 inhibitory activity (IC50 = 162.7
nM) while SGLT2 activity remained unchanged. Surprisingly, reverse the dihydrofuran
attachment (2d) significantly diminished the inhibitory activity against both SGLT1 and
SGLT2. Methylation of the phenol (2e) improved both SGLT1 (IC50 = 96.4 nM) and
SGLT2 (IC50 = 7.5 nM) activities. Replacement of distal 2,3-dihydrobenzo[b][1,4]dioxine
with 2,3-dihydrobenzo[b][1,4]oxathiine (2f) led to a loss of inhibitory activity, especially
for SGLT2 (IC50 = 62.3 nM). Compounds (2a-c, and 2e) were also tested in liver
microsomes (human, mouse, and rat), 2c and 2e showed poor in vitro stability in rat liver
microsomes with only 12 and 2% remaining respectively.

Table 1. In Vitro Potency at SGLT1 and SGLT2 assays
Cpd
X
Ar
SGLT1 IC50
(nM)
SGLT2 IC50
(nM)
Microsomal
stability (%)a
human/mouse/rat
NDb
14
2
3
Dapa
2a
O
O
S
97/97/64
95/82/66
73/61/12
43
9
2b
2c
2d
162.7
8.3
O
31% at 0.3 uM
26% at 0.3 uM NDb
O
S
96.4
7.5
91/94/2
2e
2f
186.6
62.3
NDb
a Liver microsomal stability assay conditions: 1 µM compound concentration; liver microsomal protein
content: 0.5 mg/ml, cofactor NADPH added, compound remaining at 10 mins time point; solvents: 0.01%
DMSO, 0.05% acetonitrile; positive controls: verapamil, cerivaststin, and warfarin. b ND, not determined.
Since both the substituted-thiophene and benzothiophene have been explored in the
design of SGLT inhibitors and are key structural features in Canagliflizon and
Ipragliflozin,16, 19 we decided to explore a novel bicyclic 6,7-dihydrothieno[3,2-
c]pyridine as the distal aryl group and investigated the effects of modifications of both
central benzene ring and distal aryl motif. The results were presented in Table 2.
Compound 2h with ortho–hydroxy group (R2 = OH) was the most active SGLT2

inhibitor with IC50 = 2.6 nM and also showed significant SGLT1 inhibitory activity
among the first three 6,7-dihydrothieno[3,2-c]pyridine-5-(4H)-carbonitriles (2g-i).
Methylation of this phenol group resulted in a 17-fold loss of SGLT1 activity while
slightly lost its SGLT2 potency (2i with SGLT1 IC50 = 1421 nM and SGLT2 IC50 = 5.3
nM). Similar trend was also observed for the 2-thiazol-amide derivatives (2j-l) although
all these compounds were less active against SGLT2. In addition, the distal aryl motif
also tolerated other functional groups. Urea 2m and carbamate 2n exhibited modest
activity for both SGLT1 and SGLT2, however, amide 2o without ortho-hydroxy group on
the central benzene moiety resulted in significant loss of hSGLT1 and SGLT2 activities.
Compounds (2h, 2k, 2m-n) were also tested in liver microsomes (human, mouse, and
rat), all of which showed good in vitro stability across species with 66~95% remaining
after 10 mins incubation.
Table 2. In Vitro Potency at SGLT1 and SGLT2 assays
R1
R2
Cpd
SGLT1 IC50 (nM)
SGLT2 IC50 (nM) Microsomal
stability (%)a
human/mouse/rat
-CN
H
29% at 0.3 uM
86
76% at 0.3 uM
NDb
2g
2h
2i
-CN
OH
OCH3
H
2.6
95/91/79
NDb
NDb
-CN
1421
5.3
2-Thiazol-(CO)-
2-Thiazol-(CO)-
2-Thiazol-(CO)-
1-Pyrrolidino-(CO)-
C2H5O-(CO)-
Cyclic-C5H9-(CO)-
39% at 0.3 uM
100.9
73% at 0.3 uM
2j
OH
OCH3
OH
OH
H
36
69/66/91
NDb
2k
2l
376.1
23.2
40.4
13.6
71/74/82
80/86/78
NDb
2m
2n
20
156
36
49% at 0.3 uM
59% at 0.3 uM
a Liver microsomal stability assay conditions: 1 µM compound concentration; liver microsomal protein
content: 0.5 mg/ml, cofactor NADPH added, compound remaining at 10 mins time point; solvents: 0.01%
DMSO, 0.05% CAN; positive controls: verapamil, cerivaststin, and warfarin. b Not determined.
Since complete inhibition of intestinal SGLT1 might not be desirable due to the
concern of glucose-galactose malabsorption,27 we select compounds (2a-b, and 2h) with a
balanced SGLT1 and SGLT2 inhibitory profile (IC50 of less than 10 nM for SGLT2 and
IC50 of 100 nM for SGLT1) as well as good in vitro liver microsomal stability across
multiple species (50% remaining after 10 mins incubation) for further in vivo

pharmacokinetic studies in SD rats. Administration of a single 2.0 mg/kg dose
intravenously or a single 10 mg/kg dose orally revealed that 2a and 2h had low Cmax (168
ng/ml for 2a and 13.7 ng/ml for 2h respectively) and very poor oral bioavailability in SD
rats (Table 3). Compound 2h also had very high clearance, which could be due to the
phase 2 metabolism of phenol moiety in the central benzene ring. Unlike 2a and 2h,
compound 2b was notable for rapid absorption and had higher Cmax of 990 ng/ml with an
oral bioavailablity of 48.2% in SD rats. The high clearance for 2b in rat seems to be
species-related. In higher species such as monkey (Table 3), 2b had a high volume of
distribution at steady-state about (four times total body water considering monkey total
body water of 0.693 L/kg) and a very moderate systemic clearance of 15.5 ml/min/kg
given monkey’s heptatic blood flow is 43.6 ml/min/kg. After 10 mg/kg oral
administration, plasma concentration of 2b reached Cmax of 1487 ng/ml after 1.33 hours
and had an oral bioavailability of 56%.
Table 3. Pharmacokinetic properties of three SGLT dual inhibitors after oral
and intravenous administration to rats and monkeysa
Parameters
2a
rat
iv
2
2a
2b
rat
iv
2b
2h
rat
iv
2h
2b
monkey
iv
2
2b
po
10
po
10
po
10
po
10
Dose (mg/kg)
T1/2 (h)
2
2
0.31
NDb 0.41
1.44
0.99
1.44
7.9
5.6
CL (ml/min/kg)
Vdss (L/kg)
28.1
1.22
50.9
1.07
149
15.5
3.05
3.79
Cmax (ng/ml)
168
990
13.7
1487
1.33
Tmax (h)
0.25
0.30
2.20
44.6
5.0
AUC(0-inf) (ng.h/ml)
F (%)
307
4.0
1600
48.2
6089
56
a PK experiments were conducted in male Sprague-Dawley rats or male cynomolgus monkeys (n = 3
per group). iv formulation: 20% HPbCD.; po formulation: 0.5% Methocel in saline. b ND not
determined.
To understand more on the species-related metabolic stability of 2b and potential
implication to its human blood clearance, we further investigated it in different
cryopreserved hepatocytes and the data were summarized in Table 4.
Table 4. In vitro clearance studies of 2b in cryopreserved hepatocytesa
Compound Parameters
human
5.4
SD-rat
1.2
dog
8.0
monkey
2.2
T1/2 (h)
2b
CL (ml/min/kg) 9.76
T1/2 (h) 0.7
51.42
0.8
9.29
5.2
25.30
1.0
2b
Diclofenac

Diclofenac
Midazolam T1/2 (h)
Midazolam CL (ml/min/kg) 20.24
CL (ml/min/kg) 20.88
61.41
0.5
67.32
13.13
0.5
30.85
36.90
1.4
43.45
1.0
a Cryopreserved hepatocytes stability assay conditions: 1 µM compound concentration; liver
hepatocytes: 0.5 million cells/ml; timepoints: 0, 0.5, 1, 2, 3, 4 h; positive controls: diclofenac (control for
phase 2 metabolism), midazolam (control for phase 1 metabolism) at 1 µM.
In the studies, Diclofenac and midazolam were chosen as internal standards, 2b had
high clearance in rat with calculated clearance of 51.42 ml/min/kg, which is consistent
with its in vivo clearance of 50.9 ml/min/kg. Compound 2b had a modest clearance in
monkey hepatocytes (CL = 25.30 ml/min/kg), which was slightly higher than the in vivo
finding (CL = 15.5 ml/min/kg). In both dog and human hepatocytes, 2b had low clearance.
Compound 2b was further evaluated in cell-based SGLT functional assays in CHO-K1
cells expressing rat SGLT1 and SGLT2 and its IC50 values for SGLT1 and SGLT2 were 68
nM and 16 nM respectively, which are slightly less potent than its human SGLT1 and
SGLT2 values. With its good in vitro potency and in vivo PK profiles, compound 2b was
next studied in an oral glucose tolerance test (OGTT) in SD rats.28 As shown in Figure 3, a
single oral dose of 2b (1 and 10 mg/kg) resulted in a dose-dependent inhibition of glucose
excursion (Figure 3, A). The reduction of glucose AUC reached 32.4 % versus vehicle by
the administration of 2b at 1 mg/kg dosing, which was more pronounced than that of
Dapagliflozin at the same dose (22.3%) (Figure 3, B).
A)
2 5 0
2 0 0
V e h icle
1 5 0
2 b -1 m g/kg
2 b -1 0 m g/kg
1 0 0
D a p a-1 m g /kg
5 0
0
0
3 0
6 0
9 0
1 2 0
T im e(m in s) o f O G T T
B)

B lo o d G lu c o s e E x c u rs io n A U C o f S D R a ts
2 5 0 0 0
2 0 0 0 0
1 5 0 0 0
1 0 0 0 0
5 0 0 0
0
****
****
****
Ve h ic le
1 m g /k g
1 0 m g /k g
D a p a -1 m g /k g
C o m p o u n d 2 b
O n e -W a y A N O V A -D u n n e tt's a n a ly s is
****: P < 0 .0 0 0 1 v s V e h ic le ;
Figure 3. Effect of compound 2b on post-prandial glucose control in SD rats (n=8). SD rats were
treated with either vehicle, Dapa, or compound 2b 60 min prior to OGTT (2 g/kg glucose
challenge). Blood glucose levels were determined at 0, 30, 60, and 120 min after glucose
administration (A) and blood glucose excursion AUC (B) was calculated based on the blood
glucose levels at timepoints of OGTT. Compared with vehicle group, compound 2b
treatments significantly reduced blood glucose levels in a dose-dependent fashion (****:
P<0.0001, one-way ANOVA followed by Dunnett’s multiple comparison test).
Furthermore, db/db mice and ZDF rats, both of which are rodent models of
diabetes, were used to assess the antihyperglycemic effect of compound 2b.28 As
shown in Figure 2, a single oral dose (10 mg/kg) of compound 2b in db/db mice
significantly lowered the fed blood glucose levels up to 24 h and total 24-hrs blood
glucose AUC compared with vehicle treatment (Figure 4, A and B) indicating its
remarkable antihyperglycemic efficacy. The SGLT2 inhibitory effect of 2b was
evidenced by increased urine glucose excretion (UGE) over 24 h in db/db mice
(Figure 4, C). It is interesting to observe that 2b induced UGE0-24h is lower than with
a selective SGLT2 inhibitor (data not shown) due to decreased glucose absorption by
SGLT1 inhibition at intestinal lumen.

A)
B)
C)
2b -10 m pk
V e h icle
6 0 0
4 0 0
2 0 0
0
***
*
***
***
***
***
***
0
4
8
1 2
1 6
2 0
2 4
H o u rs P o s t-D o s e
1 5 0 0 0
1 0 0 0 0
5 0 0 0
0
****
-5 7 .4 %
Ve h ic le
2 b -1 0 m p k

0 .8
0 .6
0 .4
0 .2
0 .0
*
Ve h ic le
2 b -1 0 m p k
Figure 4. Anti-hyperglycemic effect of compound 2b in db/db mice (n=8). The mice were treated with either
vehicle or compound 2b at 10 mg/kg dose. Fed blood glucose levels were determined at multiple timepoints
after the animals were treated (A). Total blood glucose AUC of 24-hrs was calculated based on the fed blood
glucose levels at multiple timepoints (B). Total urine glucose excretion (UGE) was calculated based on urine
glucose concentrations and urine volumes at multiple timepoints (C). Compared with vehicle group, a single
dose of compound 2b treatment significantly reduced the fed blood glucose levels and increased urine glucose
excretion in db/db mice (mean±se, *: P<0.05, ***: P<0.001, ****: P<0.0001. T test).
As indicated in Figure 5, a single dose of 2b at 10 mg/kg was orally administrated to
ZDF rats and their blood glucose level was monitored for 24 hours.28 It significantly
decreased fed blood glucose levels of ZDF rats at all timepoints including at 24 h (Figure 5,
A). It also significantly reduced total fed blood glucose AUC (Figure 5, B).
A)

6 0 0
4 0 0
2 0 0
0
V e h icle
2 b -10 m pk
*
***
***
***
***
***
***
0
4
8
1 2
1 6
2 0
2 4
H o u rs P o s t-D o s e
B)
1 5 0 0 0
1 0 0 0 0
5 0 0 0
0
****
-5 6 .8 %
Ve h ic le
2 b -1 0 m p k

Figure 5. Anti-hyperglycemic effect of compound 2b in ZDF rats (n=8). ZDF rats were treated with either
vehicle or compound 2b at 10 mg/kg dose. Fed blood glucose levels were determined at multiple timepoints
after the animals were treated (A). Total fed blood glucose AUC was calculated based on the fed blood glucose
levels measured at multiple timepoints (B) and 24-hrs UGE was measured based on the urine glucose
concentrations and urine volumes at multiple timepoints post dosing (mean±se, *: P<0.05, ***: P<0.001, ****:
P<0.0001. T test).
Because of its excellent in vivo PD profiles, compound 2b was assessed against 55
different GPCRs, kinases, enzymes, and ion channels and it showed minimal activity (less
than 21% inhibition) at a concentration of 10 µM; No activity was observed against the
hERG channel at 50 µM. It was also tested in AMES II mutagenicity assay (TA98 strains)
and in vitro micronucleus assay. Unlike some compounds from C-aryl glucoside class,29
compound 2b was negative under both assays.
In summary, we successfully designed and synthesized a new series of
(2S,3R,4R,5S,6R)-5-fluoro-6-(hydroxymethyl)-2-aryltetrahydro-2H-pyran-3,4-diols as
SGLT dual inhibitors. More importantly, two methods were developed to efficiently
synthesize C5-fluoro-lactones 3 and 4, key intermediates for diverse C5-fluoro-hexose based
C-aryl glucosides. The study of structure-activity relationships of these novel SGLT dual
inhibitors resulted in the identification of 2b with potent SGLT1 and SGLT2 activities
(hSGLT1 IC50 = 43 nM, hSGLT2 IC50 = 9 nM), which is orally active in db/db mouse and
ZDF rat models of diabetes and has several encouraging preclinical data profiles.

Acknowledgements
Thanks to the Discovery Science ADME and Analytical teams for their contributions to this
work. We would like to thank Mark Macielag for reviewing this manuscript.
References and notes
1. Stumvoll, M.; Goldstein, B. J.; van Haeften, T. W. Lancet 2005, 365, 1333.
2. American Diabetes Association. Diabetes Care 2004, 27 (suppl. 1), 5.
3. Diabetes Fact Sheet. https://www.cdc.gov/diabetes/pdfs/data/statistics/national-
diabetes-statistics-report.pdf (accessed December 21, 2017)
4. Fowler, M. J. Clin. Diabetes 2008, 26, 77.
5. Davidson, M. B.; Peter, A. L. Am. J. Med. 1997, 102, 99.
6. Davis, S. N. J. Diabetes Complications 2004, 18, 367.
7. Marsenic, O. Am. J. Kidney Dis. 2009, 53, 875.
8. DeFronzo, R. A.; Davidson, J. A.; Del Prato, S. Diabetes Obes. Metab. 2012, 14, 5.
9. Kanai, Y.; Lee, W. S.; You, G.; Brown, D.; Hediger, M. A. J. Clin. Invest. 1994, 93,
397.
10. Wright, E. M.; Hirayama, B. A.; Loo, D. F. J. Intern. Med. 2007, 261, 32.
11. Wang, X. X.; Levi, J.; Luo, Y.; Myakala, K.; Herman-Edelstein, M.; Qiu, L.; Wang,
D.; Peng, Y.; Grenz, A.; Lucia, S.; Dobrinskikh, E.; D’Agati, V. D.; Koepsell, H.;
Kopp, J. Rosenberg, A. Z.; Levi, M. J. Biol. Chem. 2017, 292, 5335.
12. Turk, E.; Zabel, B.; Mundlos, S.; Dyer, J.; Wright, E. M. Nature 1991, 350, 354.
13. Martin, M. G.; Turk, E.; Lostao, M. P.; Kerner, C.; Wright, E. M. Nat. Genet. 1996,
12, 216.
14. Lee, W. S.; Kanai, Y.; Wells, R. G.; Hediger, M. A. J. Biol. Chem. 1994, 269, 12032.
15. Chao, E. C.; Henry, R. R. Nat Rev Drug Discov. 2010, 9, 551.
16. Nomura, S.; Sakamaki, S.; Hongu, M.; Kawanishi, E.; Koga, Y.; Sakamoto, T.;
Yamamoto, Y.; Ueta, K.; Kimata, H.; Nakayama, K.; Tsuda-Tsukimoto, M. J. Med.
Chem. 2010, 53, 6355.
17. Meng, W.; Ellsworth, B. A.; Nirschl, A. A.; McCann, P. J.; Patel, M.; Girotra, R. N.;
Wu, G.; Sher, P. M.; Morrison, E. P.; Biller, S. A.; Zahler, R.; Deshpande, P. P.;
Pullockaran, A.; Hagan, D. L.; Morgan, N.; Taylor, J. R.; Obermeier, M. T.;
Humphreys, W. G.; Khanna, A.; Discenza, L.; Robertson, J. G.; Wang, A.; Han, S.;
Wetterau, J. R.; Janovitz, E. B.; Flint, O. P.; Whaley, J. M.; Washburn, W. N. J. Med.
Chem. 2008, 51, 1145.
18. Grempler, R.; Thomas, L.; Eckhard, M.; Himmelsbach, F.; Sauer, A.; Sharp, D. E.;
Bakker, R. A.; Mark, M.; Klein, T.; Eickelmann, P. Diabetes, Obes. Metab. 2012, 14,
83.

19. Imamura, M.; Nakanishi, K.; Suzuki, T.; Ikegai, K.; Shiraki, R.; Ogiyama, T.;
Murakami, T.; Kurosaki, E.; Noda, A.; Kobayashi, Y.; Yokota, M.; Koide, T.;
Kosakai, K.; Ohkura, Y.; Takeuchi, M.; Tomiyama, H.; Ohta, M. Bioorg. Med. Chem.
2012, 20, 3263.
20. Kakinuma, H.; Oi, T.; Hashimoto-Tsuchiya, Y.; Arai, M.; Kawakita, Y.; Fukasawa,
Y.; Iida, I.; Hagima, N.; TTakeuchi, H.; Chino, Y.; Asami, J.; Okumura-Kitajima, L.;
Io, F.; Yamamoto, D.; Miyata, N.; Takahashi, T.; Uchida, S.; Yamamoto, K. J. Med.
Chem. 2010, 53, 3247.
21. Suzuki, M.; Honda, K.; Fukazawa, M.; Ozawa, K.; Hagita, H.; Kawai, T. Takeda, M.;
YYata, T.; Kawai, M.; Fukuzawa, T.; Kobayashi, T.; Sato, T.; Kawabe, Y.; Ikeda, S.
J. Pharmacol. Exp. Ther. 2012, 341, 692.
22. Goodwin, N. C.; Mabon, R.; Harrison, B. A.; Shadoan, M. K.; Almstead, Z. Y.; Xie,
Y.; Healy, J.; Buhring, L. M.; DaCosta, C. M.; Bardenhagen, J.; Mseeh, F.; Liu, Q.;
Nouraldeen, A.; Wilson, A. G. E.; Kimball, S. D.; Powell, D. R.; Rawlins, D. B. J.
Med. Chem. 2009, 52, 6201.
23. Sands, A. T.; Zambrowicz, B. P.; Rosenstock, J. Diabetes Care 2015, 38, 1181.
24. Lima, L. M.; Barreiro, E. J. Curr. Med. Chem. 2005, 12, 23.
25. Robinson, R. P.; Mascitti, V.; Boustany-Kari, C. M.; Carr, C. L.; Foley, P. M.;
Kimoto, E.; Leiniger, M. T.; Lowe, A.; Klenotic, M. K.; MacDonald, J. I.; Maguire,
R. J.; Masterson, V. M.; Maurer, T. S.; Miao, Z.; Patel, J. D.; Preville, C.; Reese, M.
R.; She, L.; Steppan, C. M.; Thuma, B. A.; Zhu, T. Bioorg. Med. Chem. Lett. 2010,
20, 1569.
26. a) Corey, E. J.; Suggs, W. J. J. Org. Chem. 1973, 38, 3224; b) Lamberth, C.;
Bednarski, M. D. Tetrahedron Lett. 1991, 32, 7369
27. Wright, E. M. Am. J. Physiol. 1998, 275, G879.
28. All reported studies on rats and mice were carried out in accordance with the
recommendations in the National Institutes of Health Guide for the Care and Use of
Laboratory Animals. The protocols for all studies performed at Janssen
Pharmaceuticals were approved by the Johnson & Johnson Animal Care and Use
Committee. Rats or mice were housed 2 animals per cage in a temperature-controlled
room with 12-hour light/dark cycle. They are allowed ad libitum access to water and
food. For details, please see the supporting material.
29. Washburn, W.; Meng, W. Chem. Abstr. 2006, 144, 305154.

Highlights
 A new series of SGLT1 and SGLT2 dual inhibitors were disclosed.
 Two methods were developed to efficiently synthesize C5-fluoro-lactones 3 and 4.
 Lead compound 2b demonstrated potent hSGLT1 and hSGLT2 inhibition.
 Lead compound 2b showed robust in vivo efficacy in rodent models of diabetes.