5,5-Difluoro- and 5-Fluoro-5-methyl-hexose-based C-Glucosides as potent and orally bioavailable SGLT1 and SGLT2 dual inhibitors

Article abstract of DOI:10.1016/j.bmcl.2020.127387

(2S,3R,4R,5S,6R)-2-Aryl-5,5-difluoro-6-(hydroxymethyl)tetrahydro-2H-pyran-3,4-diols and (2S,3R,4R,5S,6R)-2-aryl-5-fluoro-5-methyl-6-(hydroxymethyl)tetrahydro-2H-pyran-3,4-diols were discovered as dual inhibitors of sodium glucose co-transporter proteins (

Full text of DOI:10.1016/j.bmcl.2020.127387

Journal Pre-proofs
5,5-Difluoro- and 5-Fluoro-5-methyl-hexose-based C-Glucosides as Potent
and Orally Bioavailable SGLT1 and SGLT2 Dual Inhibitors
Guozhang Xu, Fuyong Du, Gee-Hong Kuo, June Zhi Xu, Yin Liang, Keith
Demarest, Michael D. Gaul
PII:
S0960-894X(20)30498-4
DOI:
Reference:
https://doi.org/10.1016/j.bmcl.2020.127387
BMCL 127387
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
16 May 2020
27 June 2020
30 June 2020
Please cite this article as: Xu, G., Du, F., Kuo, G-H., Zhi Xu, J., Liang, Y., Demarest, K., Gaul, M.D., 5,5-
Difluoro- and 5-Fluoro-5-methyl-hexose-based C-Glucosides as Potent and Orally Bioavailable SGLT1 and
SGLT2 Dual Inhibitors, Bioorganic & Medicinal Chemistry Letters (2020), doi: https://doi.org/10.1016/j.bmcl.
2020.127387
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. 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.
© 2020 Elsevier Ltd. All rights reserved.

Corresponding Author
Guozhang Xu, Ph.D.
Tel.: +1 215 628 7940
E-mail address: gxu4@its.jnj.com (G. Xu)
Janssen Research & Development, LLC.
Discovery Sciences
Bldg# 42-2314
1400 McKean Road
Spring House, PA 19002

5,5-Difluoro- and 5-Fluoro-5-methyl-hexose-based C-Glucosides as Potent and
Orally Bioavailable SGLT1 and SGLT2 Dual Inhibitors
Guozhang Xu1*, Fuyong Du2, Gee-Hong Kuo1, June Zhi Xu2, Yin Liang2, Keith
Demarest2, Michael D. Gaul1
1Discovery Sciences and 2Cardiovascular & Metabolic Research, Janssen Research
& Development, L.L.C., Welsh & McKean Roads, Spring House, PA 19477
Type 2 diabetes mellitus (T2DM) is characterized by a reduction in insulin function in
peripheral tissues and an increase in gluconeogenesis in the liver that together result in
hyperglycemia.1-2 Uncontrolled hyperglycemia is associated with an increased risk for
cardiovascular and end-stage kidney disease.3 Currently approved antihyperglycemic
agents act by increasing insulin secretion, enhancing insulin sensitivity, reducing hepatic
glucose production or glucose absorption.4 However, therapeutic efficacy is also offset by
side effects such as weight gain and hypoglycemia.5 In recent years, much attention has
been on excretion of excessive glucose directly into urine, which involves sodium-glucose
co-transporters (SGLTs), for the treatment of hyperglycemia.6
SGLTs are a family of glucose transporter proteins, which rely on the electrochemical
potential for sodium ion to actively transport extracellular glucose into the cytoplasm.7 The
physiological and pathophysiological roles of the two major isoforms, SGLT1 and SGLT2,
are well-documented.8 Since 2013, the FDA has approved multiple SGLT2 selective
inhibitors (known as gliflozins) for the treatment of type 2 diabetes mellitus (T2DM).9-11
Interestingly, SGLT2 knockout or treatment with an SGLT2 selective inhibitor only causes
a fractional glucose excretion (~60%), an effect attenuated by up-regulation of renal
SGLT1.12 Adding SGLT1 inhibition on top of SGLT2 inhibition (e.g. SGLT1 and SGLT2
dual inhibition) is expected to reduce glucose burden by inhibiting the uptake of dietary
glucose (mediated by SGLT1) in the intestine and excreting filtered glucose (mediated by
SGLT2 and SGLT1) via the kidney. Therefore, an orally bioavailable SGLT1/2 dual
inhibitor could not only achieve even better glycemic control and weight loss, but also help
reduce glycemic variability through the day via a non-insulin-dependent mechanism.

Besides phlorizin, a natural product with modest inhibitory activities against both
SGLT1 and SGLT2,13 several synthetic SGLT1/2 dual inhibitors have been reported
(Figure 1). Sotagliflozin (1, Figure 1) is the first SGLT1/2 dual inhibitor that has received
approval in the European Union for certain patients with T1D.14 Licogliflozin (2, Figure 1)
helped obese patients lose more than 5% of their body weight in 12 weeks and improve key
health measures in those with T2D.15 HM41322 (3, Figure 1) showed a favorable
preclinical profile of postprandial glucose control.16
HO
O
OH
O
OH
O
Cl
O
O
S
HO
HO
OH
HO
OH
OH
OH
SGLT1 IC50 = 36 nM
SGLT2 IC50 = 1.8 nM
SGLT1 IC50 = 210 nM
SGLT2 IC50 = 34.6 nM
Phlorizin
1
O
O
O
O
HO
HO
HO
OH
HO
OH
OH
OH
SGLT1 IC50 = 22 nM
SGLT2 IC50 = 0.5 nM
SGLT1 IC50 = 54.6 nM
SGLT2 IC50 = 5.6 nM
2
3
Figure 1. Representative examples of SGLT1/2 dual inhibitors
These inhibitors exhibit very potent half-maximal inhibitory concentrations (IC50) in the
range of 0.5~5.6 nM against human SGLT2 and IC50 values in the range of 22~54.6 nM
against human SGLT1 in vitro.
We previously reported a series of potent and orally active SGLT1/2 dual inhibitors
based on a mono-fluorine-substituted C-aryl glucoside skeleton (5, Figure 2).17 The
isosteric replacement of the equatorial hydroxy group at the C5 position (4, Figure 2) with
a fluorine atom is well tolerated. We report here further modification of the glucose ring as
well as the aglycone moiety (vide infra). The remaining axial hydrogen at the C5 position
of 5 was replaced with either a fluorine atom or a methyl group, which resulted in a new
series of (2S,3R,4R,5S,6R)-2-aryl-5,5-difluoro-6-(hydroxymethyl)tetrahydro-2H-pyran-

3,4-diols
(6)
and
(2S,3R,4R,5S,6R)-2-aryl-5-fluoro-5-methyl-6-
(hydroxymethyl)tetrahydro-2H-pyran-3,4-diols (7) in which Ar represents a generic
aglycone moiety. Interestingly, there are few reports of orally active SGLT dual inhibitors
based on template 6 or template 7.18
Ar
OH
OH
Ar
OH
3
Ar
OH
OH
2
Current work
Previous work
O
1 O
OH
O
4
5
F
HO
HO
OH
R
HO
F
R = F
R = Me
6
4
5
7
Figure 2. Design of new SGLT1/2 dual inhibitors
The
(hydroxymethyl)tetrahydro-2H-pyran-3,4-diols (6a-6j) is described in Scheme 1.
Treatment of commercially available (2S,3R,4S,5R,6R)-2-(allyloxy)-6-
synthesis
of
the
(2S,3R,4R,5S,6R)-2-aryl-5,5-difluoro-6-
(hydroxymethyl)tetrahydro-2H-pyran-3,4,5-triol with dimethoxymethylbenzene in the
presence of a catalytic amount of D-camphorsulfonic acid afforded (4aR,6S,7R,8R,8aR)-
6-(allyloxy)-2-phenylhexahydropyrano[3,2-d][1,3]dioxine-7,8-diol (9) in 89% yield.
Benzylation of the resulting diol 9 followed by selective reduction of 10 provided alcohol
11 in 57% yield over two steps.

O
O
O
O
O
O
O
O
HO
HO
a
b
OBn
O
OH
O
OH
OBn
OH
OH
8
9
10
O
O
O
O
O
BnO
HO
O
BnO
d
BnO
c
e
OBn
F
OBn
O
OBn
OBn
F
OBn
OBn
11
12
13
BnO
R1
OH
OBn
O
OH
O
O
O
R'
BnO
BnO
h
g
f
BnO
F
F
F
OBn
F
F
OBn
F
OBn
OBn
OBn
14
15
16
R1
BnO
O
R1
HO
H3CO
O
R1
O
R'
R'
j
k
i
R'
HO
BnO
HO
F
F
F
OH
OBn
OH
F
F
F
OH
OBn
OH
17
6a-6b, 6d, 6f-6j
6c, 6e
O
O
R1
O
O
S
BnO
Br
R1 = Cl, CH3, OCH3
RBr
=
R'
R' =
O
O
Scheme 1. Reagents and conditions: (a) (Dimethoxymethyl)benzene, cat. D-camphorsulfonic acid, CHCl3,
reflux, 89%; (b) BnBr, NaH, DMF, room temperature 75.7% yield; (c) Et3SiH, TfOH, 4A MS, DCM, -78 oC
74.7% yield or NaBH3CN, THF, HCl(g) in diethyl ether 41.7% yield; (d) Ac2O, DMSO, 65 oC or PCC, DCM
95.4%; (e) BAST, DCM 73.5%: (f) PdCl2, NaOAc 95% HOAc or n-Bu3SnH, Pd(PPh3)4, ZnCl2, THF, room
temperature 77.3 %; (g) Ac2O, DMSO, room temperature 87%; (h) n-BuLi, RBr (shown in the box),
o
o
THF/toluene, -78 C; (i) Triethylsilane/BF3.Et2O or TFA, DCM, at 0 C for 1 h, 65%; (j) BCl3, 1,2,3,4,5-
pentamethylbenzene, DCM, at -78 oC for 30 min, 42%; (k) MeI, K2CO3, acetone 90%.

Pyridinium chlorochromate (PCC) oxidation or Albright-Goldman oxidation of 11 and
fluorination of the resulting ketone 12 using BAST (Deoxy-Fluoro®) gave (2R,4R,5R,6S)-
6-(allyloxy)-4,5-bis(benzyloxy)-2-((benzyloxy)methyl)-3,3-difluorotetrahydro-2H-pyran
13 in 69% yield. Selective removal of the allyl protecting group followed by Albright-
Goldman oxidation of 14 provided the C5,5-difluoro-lactone 15 as a key intermediate. The
addition of the aryl lithium species generated by lithium/bromide exchange of the RBr
species shown in Scheme 1 to lactone 15 generates the corresponding lactol (16).
Triethylsilane/BF3.Et2O or triethylsilane/TFA reduction in dichloromethane (DCM) at 0 oC
gave the corresponding β-C-aryl glucoside 17, which was confirmed by the coupling
constant (J = 9.0 ~ 9.6 Hz) between the anomeric hydrogen and the adjacent hydrogen in
1
the H NMR spectra. Mild debenzylation of 17 proceeded at low temperature with a
combination of BCl3 and pentamethylbenzene as a non-Lewis-basic cation scavenger to
afford the corresponding target compound (6a-b, 6d, 6f-6j).19 Methylation of 6b or 6d with
MeI/K2CO3 in acetone provided 6c or 6e, respectively.
The preparation of 5-fluoro-5-methyl C-aryl glucoside derivatives 7a-7e is outlined in
Scheme 2. Addition of an aryl lithium reagent, generated by lithium/bromide exchange of
the RBr species shown in Scheme
1
to (3R,4S,6S)-3,4-bis(benzyloxy)-6-
((benzyloxy)methyl)-5-methylenetetrahydro-2H-pyran-2-one (18),20 followed by
triethylsilane/trimethylsilyl trifluoromethanesulfonate (TMSOTf)-mediated reduction
generates 5-methylene derivative 20. Upjohn dihydroxylation (N-methylmorpholine N-
oxide and a catalytic amount of OsO4) of the methylene group at the C5 position provides
the diol 21, which is then oxidatively cleaved by PhI(OAc)2 to give the corresponding
ketone 22. Addition of methyl lithium to 22 and subsequent DAST-mediated fluorination
of the axial hydroxyl group in 23 affords the precursor 24. Global debenzylation of 24
either by hydrogenolysis catalyzed by Pd(OH)2 on carbon or treatment with
BCl3/pentamethylbenzene affords the target compounds 7a-7e.

R1
BnO
O
R1
BnO
OH
O
O
R'
O
R'
BnO
c
b
a
BnO
BnO
OBn
OBn
OBn
OBn
OBn
OBn
18
20
19
R1
BnO
O
R1
BnO
O
R1
BnO
O
R'
R'
R'
e
d
BnO
Me
HO
BnO
BnO
HOHO
OBn
O
OBn
OBn
OBn
OBn
OBn
23
22
21
R1
BnO
O
R1
HO
R'
O
R'
f
g
BnO
HO
OBn
OH
F
F
OBn
OH
24
7a~7e
o
Scheme 2. Reagents and conditions: (a) RBr (shown in Scheme 1), BuLi, THF, -78 C, 90%; (b) Et3SiH,
TMSOTf, DCM, 60-70% yield; (c) NMO, cat. OsO4, acetone/water, 88% yield; (d) PhI(OAc)2 61%; (e) MeLi,
THF, -78 oC 79.1%: (f) DAST, DCM, 63.7%; (g) Pd(OH)2, H2, EtOAc or BCl3, 1,2,3,4,5-
pentamethylbenzene, DCM, at -78 oC for 30 min, 40-50%.
The in vitro inhibitory activities of newly synthesized compounds for SGLT1 and
SGLT2 were evaluated by monitoring the inhibition of [14C]methyl D-glucopyranoside
(AMG) accumulation in Chinese hamster ovary (CHO) cells expressing human SGLT1 or
SGLT2. The IC50 values against SGLT1 and SGLT2 were summarized in Table 1.
Within the (2S,3R,4R,5S,6R)-2-aryl-5,5-difluoro-6-(hydroxymethyl)tetrahydro-2H-
pyran-3,4-diol series, compound 6b (R1 = Cl) with an ortho-hydroxy group exhibited potent
SGLT2 (IC50 = 4.2 nM) and SGLT1 inhibitory activity (IC50 = 93 nM). Methylation of the
phenol (6c) resulted in a greater than 2-fold loss of potency for SGLT1 (IC50 = 221 nM)
although SGLT2 inhibitory activity was maintained (IC50 = 4.3 nM). The introduction of a
methoxy group at C4 on the central ring (6a) produced an analog with only modest SGLT1
(IC50 = 165 nM) and SGLT2 (IC50 = 24.1 nM) potencies. An analogous trend was observed
when the distal benzothiophene was replaced with 2,3-dihydrobenzo[b][1,4]dioxine.
Compound 6d (R1 = Cl) was more potent than 6f (R1 = MeO) in both SGLT1 and SGLT2

assays. Interestingly, compound 6e with the hydroxyl group replaced by methoxy was
much less active against both SGLT1 and SGLT2. Other distal aryl groups such as
chromane (6g) and p-methoxyphenyl (6i) were well tolerated and displayed potent SGLT1
and SGLT2 inhibitory activities although 2,3-dihydrobenzofuran (6h) showed only modest
SGLT dual activity. The meta-methoxyphenyl 6j was less potent than the para-
methoxyphenyl analog 6i against SGLT1 and SGLT2. All compounds except 6e and 6j
were also evaluated in liver microsome stability assays (Table 1). Compounds 6a, 6b, 6g,
6h, and 6i showed good in vitro stability across species (human, mouse, rat) as exemplified
by more than 50% of test compound remaining after 10 minutes incubation. However, 6c,
6d, and 6f were less stable in either mouse or rat liver microsomes.
Table 1. In vitro potency at SGLT1 and SGLT2 assays and liver microsomal stability
R2
R1
O
R'
HO
F
OH
F
OH
6
Cpd
R1
CH3O
Cl
R2
OH
R’
SGLT1 IC50
(nM)
SGLT2 IC50
(nM)
Microsomal stabilitya
Human/mouse/rat
103/111/98
165
24.1
6a
6b
6c
6d
S
S
S
OH
93 (+/- 38)b
221
4.2 (+/- 2.6)
4.3
120c/106/94
65/37/123
96/74/18
Cl
OCH3
OH
O
O
Cl
49 (+/- 12)
1.4 (+/- 1.4)

NDd
99/97/26
95/85/57
100/91/70
90/97/58
ND
O
Cl
OCH3
OCH3
OCH3
Cl
OCH3
OH
37% inh. at 0.3
78% inh. at 0.3
6e
6f
µM
895
µM
O
O
26.3
O
O
OH
96
708
1.3
17.3
6g
6h
6i
OH
O
O
OH
34 (+/- 6.7)
112
1.1 (+/- 0.28)
2.6
Cl
OH
6j
O
aLiver microsomal stability assay conditions: 1 µM compound concentration; liver microsomal protein
content: 0.5 mg/ml, cofactor NADPH was added, percentage of compound remaining at 10-minute time point;
solvents: 0.01% DMSO, 0.05% acetonitrile; positive control: verapamil, cerivastatin, and warfarin. bStandard
c
d
derivation (+/- SD) was added when multiple batches were tested; T1/2 (human) > 180 mins; ND, not
determined.
The in vitro cell-based SGLT inhibitory activities and liver microsomal stability data for
(2S,3R,4R,5S,6R)-2-aryl-5-fluoro-5-methyl-6-(hydroxymethyl)tetrahydro-2H-pyran-3,4-
diols (7a-7e) are summarized in Table 2. Compound 7a displayed potent SGLT2 activity
(IC50 = 2.1 nM), however was less potent against SGLT1 (IC50 = 269 nM). Unlike the 5,5-
difluoro-C-aryl glucoside 6f which showed modest SGLT1 (IC50 = 895 nM) and SGLT2
(IC50 = 26 nM) activities, 7b exhibited potent SGLT2 (IC50 = 4.3 nM) and SGLT1 (IC50 =
37 nM) inhibitory activity. Surprisingly, 7d and 7e were less potent than the corresponding
5,5-difluoro-C-aryl glucoside counterpart 6b and 6g in both SGLT assays. In the liver
microsomal stability assays, compounds 7a and 7b were stable across species (human,
mouse, and rat).

We selected compounds (6b, 6g, 6i and 7b) with a balanced SGLT1 and SGLT2
inhibitory profile (10 nM < IC50 < 100 nM for SGLT1 and IC50 < 10 nM for SGLT2) as
well as good liver microsomal stability (more than 50% remaining after 10 minutes
incubation) for pharmacokinetic (PK) studies in Sprague Dawley (SD) rats.
Table 2. In vitro potency at SGLT1 and SGLT2 assays and liver microsomal stability
R2
R1
O
R'
HO
OH
F
OH
7
Cpd
R1
R2
R’
SGLT1 IC50
(nM)
SGLT2 IC50
(nM)
Microsomal stabilitya
Human/mouse/rat
74/96/56
269 (+/- 43)b
2.1 (+/- 0.49)
O
Cl
OH
OH
7a
7b
7c
7d
7e
O
O
OCH3
OCH3
Cl
37
4.3
103/98/101
NDc
O
O
OCH3
OH
25% inh. at 0.3 59% inh. at 0.3
µM
287
µM
O
10.2
89/105/15
S
OCH3
OH
31% inh. at 0.3 67% inh. at 0.3
µM µM
aLiver microsomal stability assay conditions: 1 M compound concentration; liver microsomal protein
ND
O
content: 0.5 mg/ml, cofactor NADPH was added, percentage of compound remaining at 10 minute time point;
solvents: 0.01% DMSO, 0.05% acetonitrile; positive control: verapamil, cerivastatin, and warfarin. bStandard
derivation (+/- SD) was added when multiple batches were tested; cND, not determined.

The PK data are summarized in Table 3. Administration of a 2.0 mg/kg dose
intravenously or a 10 mg/kg dose orally to rat revealed that 6g and 7b had very high
clearance (Cl = 75.6 ml/min/kg for 6g and 72.0 ml/min/kg for 7b, respectively) while 6b
exhibited low in vivo clearance, a higher Cmax, and good oral bioavailability (F = 46.8%).
Compound 6i had low clearance, but also low oral bioavailability (F = 17.4%). The high
clearance seen for 6g could be due to metabolism as only 57% remained upon 10 minutes
incubation in rat liver microsomes. Compound 6g was more stable in mouse liver
microsomes, thus it was further examined in a mouse PK study (Table 3). Compound 6g
had relatively low clearance, a high Cmax of 4680 ng/ml, and oral bioavailability of 52.2%
in C57BL mice.
Table 3. Pharmacokinetic profiles of selected SGLT1/2 dual inhibitorsa
Compound Species
CL
Vss
t1/2 (iv)
(h)
AUC0-inf
Cmax
Tmax
(h)
F%
(ml/min/Kg)
(L/Kg)
(h*ng/mL)
(ng/mL)
rat
rat
3.92
75.6
19.0
10.9
72.0
0.91
3.85
0.96
1.89
2.41
3.71
2.28
0.83
4.85
1.52
18601
524
5245
258
0.31
0.25
1.0
46.8
22.3
52.2
17.4
21.8
6b
6g
6g
6i
mouse
rat
9158
2701
ND
4680
549
0.42
0.25
rat
274
7b
aCompounds were dosed in male C57 black mice at 5 mg/kg IV and 20 mg/kg PO and male Sprague
Dawley rats at 2 mg/kg IV and 10 mg/kg PO. 20% HPbCD was vehicle for iv dosing and 0.5% Methocel
was the vehicle for po dosing. Compounds were formulated with 20% HPbCD for iv dosing and 0.5%
Methocel for po dosing, respectively.
With balanced SGLT 1/2 potencies and reasonable rat PK profiles, compounds 6b,
6g, and 6i were next studied in an oral glucose tolerance test (OGTT) in SD-rats. In this
experiment, overnight fasted rats were orally dosed with either vehicle or compound
(at 1 and 10 mg/kg) 60 minutes prior to OGTT. The percentage (%) of blood glucose

area under the curve (AUC) versus vehicle control, the percent reduction of blood
glucose AUC, and urinary glucose excretion (UGE) over a 4-hr period post OGTT for
6b, 6g, and 6i are summarized in Table 4.
Table 4. Effect of SGLT1/2 inhibitors on glucose control and UGE in SD-rats
Compound
Dose (mg/kg)
Reduction of blood
glucose AUC (%)a
16.0 ± 2.0
UGE (grams/4-hrs)
1
10
1
0.001 ± 0.00
0.005 ± 0.002
0.002 ± 0.001
0.2 ± 0.02
6b
6b
6g
6g
6i
41.4 ± 3.0
38.3 ± 2.1
10
1
60.7 ± 1.7
38.4 ± 2.4
0.099 ± 0.037
0.276 ± 0.032
10
62.8 ± 1.4
6i
aVehicle was 0.5% Methocel from Sigma. All compounds were formulated in 0.5% Methocel.
Data was expressed as mean ± standard error of mean. N=8 in each group. The data was
statistically analyzed using one-way ANOVA-Dunnett’s analysis.
As shown in Table 4, compound 6b was least potent in reducing blood glucose
excursion in rats compared to compounds 6g and 6i. Moreover, UGE for compound 6b
was modest suggesting that the blood glucose reduction was almost completely
mediated through SGLT1 inhibition in this model. In contrast, compounds 6g and 6i
showed robust dose-dependent efficacy in lowering blood glucose excursion during
OGTT, indicating that both compounds worked through dual inhibition of SGLT1/2 in
vivo. Given that compound 6g had a good in vitro safety profile (data are shown in the
supporting material) and oral bioavailability in mice as shown in Table 3, it was further
evaluated in a separate OGTT study relative to the marketed SGLT2 inhibitor
dapagliflozin (Dapa). After an overnight fast, rats were administered a single oral dose
of Dapa (1 mg/kg corresponding to the clinical dose) or 6g (1 mg/kg or 10 mg/kg) and
2g/kg glucose (60 mins after compound dosing). The effect of 6g or Dapa on blood
glucose excursion and urinary glucose excretion are illustrated in Figure 3. Compared
to vehicle-treated animals, treatment with 6g significantly reduced blood glucose levels
during OGTT in a dose-dependent manner. The reduction of blood glucose was superior
to Dapa at a dose of 1 mg/kg (Figure 3, panel A). Blood glucose excursion AUC was

calculated using blood glucose levels at multiple time points (0-120 mins) during
OGTT. Compared to vehicle-treated rats, compound 6g significantly lowered blood
glucose AUC in a dose-dependent manner (Figure 3, panel B; one-way ANOVA
analysis, P<0.001). As shown in panel C (Figure 3) the urinary glucose excretion was
measured 0-4 hours after glucose challenge. Compound 6g did not increase UGE at the
1 mg/kg dose, but it substantially increased UGE at 10 mg/kg (one-way ANOVA
analysis, P<0.001 vs. vehicle). This in vivo result suggests that compound 6g may only
inhibit SGLT1 at 1 mg/kg but inhibit both SGLT1 and SGLT2 at the 10 mg/kg dose.
(A)
Vehicle
Compound 6g-10 mg/kg
Dapa-1 mg/kg
Compund 6g-1 mg/kg
250
200
150
100
50
-30
0
30
60
90
120
Time of OGTT (min)
(B)

Blood Glucose AUC of SD Rats
25000
20000
15000
10000
5000
0
***
***
***
e
o
g
g
g
l
k
k
k
/
c
/
/
i
g
g
g
h
e
m
m
m
V
1
-
a
1
0
-
1
-
g
p
6
g
a
6
d
D
d
n
n
u
u
o
p
p
m
m
o
o
C
C
(C)
4-hr UGE of SD Rats
0.4
0.3
0.2
0.1
0.0
***
***
e
g
g
k
g
k
l
k
c
i
/
/
/
g
g
g
h
e
m
m
m
V
1
1
-
0
-
1
-
a
p
a
g
6
g
6
d
D
n
d
n
u
o
u
o
p
p
m
o
C
Figure 3. Effect of compound 6g or Dapa on post-prandial glucose control in SD rats (n=8) at 0, 30,
60, and 120 min after glucose administration (panel A), blood glucose excursion AUC (panel
B), and urinary glucose excretion measured during 0-4 hrs period after glucose challenge
(panel C). (***: P<0.001, one-way ANOVA followed by Dunnett’s multiple comparison

test).
We further examined the antihyperglycemic effect of 6g in diabetic mice. As shown
in Figure 4, a single oral dose (10 mg/kg) of compound 6g in db/db mice significantly
lowered the fed blood glucose levels up to 24 h (panel A). Compared to the vehicle-
treated group, compound 6g had a strong trend of increasing 24-hour UGE in db/db
mice (panel B, unpaired T test. P = 0.0621). The pronounced antihyperglycemic effect
of 6g reflected a combination of decreased glucose absorption by SGLT1 inhibition at
the intestinal lumen and increased urinary glucose excretion mediated by SGLT2
inhibition.
A:
600
400
*
200
*
*
Vehicle
Compound 6g-10 mg/kg
*
*
*
*
0
0
4
8
2
6
0
4
1
1
2
2
Hours Post-Dose
B:

24-hr UGE of db/db mice
0.8
0.6
0.4
0.2
0.0
)
r
h
-
4
2
/
g
(
E
G
U
e
g
l
k
c
i
/
g
h
e
m
V
0
1
-
g
6
d
n
u
o
p
m
o
C
Figure 4. Anti-hyperglycemic effect of compound 6g in db/db mice. Male db/db mice were treated with
either vehicle or compound at 10 mg/kg. Fed blood glucose levels (panel A) and 24-hr UGE (panel B) (mean
± SE, one-way ANOVA analysis: p<0.05 vs vehicle control).
We previously reported that a C5-fluoro-aryl glucoside derivative (compound 2b17)
was a potent SGLT1/2 dual inhibitor. Our present results showed that compound 6g
had the comparable anti-diabetic efficacy to our early lead 2b in both OGTT model of
rats and diabetic model of db/db mice. The discovery of compound 6g has provided us
with more selections in the candidates’ pool of anti-diabetic molecules to be further
evaluated and developed.
In conclusion, we successfully designed and synthesized a new series of
(2S,3R,4R,5S,6R)-2-aryl-5,5-difluoro-6-(hydroxymethyl)tetrahydro-2H-pyran-3,4-
diols and (2S,3R,4R,5S,6R)-2-aryl-5-fluoro-5-methyl-6-(hydroxymethyl)tetrahydro-
2H-pyran-3,4-diols as potent SGLT dual inhibitors. Several compounds (6b, 6g, and
6i) in the C5,5-difluoro-aryl glucoside series demonstrated excellent postprandial
glucose control in an OGTT test in SD-rats. Furthermore, compound 6g was orally
active in a db/db mouse model of diabetes.

Acknowledgements
The authors are grateful for the financial support from Janssen Research & Development
L.L.C. and would like to thank the Discovery Sciences ADME and Analytical teams for their
contributions to this work and Dr. 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. Fowler, M. J. Clin. Diabetes 2008, 26, 77.
4. a) Lebovitz, H. Nature Reviews Endocrinology 2011, 7, 408-419; b) Hossain, M.A.;
Pervin, R. Nutritional and Therapeutic Interventions for Diabetes and Metabolic
Syndrome, Elsevier, 2018, pp. 455-473.
5. Davis, S. N. J. Diabetes Complications 2004, 18, 367.
6. Goldenberg, R. M.; Gilbert, J. D.; Hramiak, I. M.; Woo, V. C.; Zinman, B. Diabetes,
Obesity & Metabolism. 2019, 21, 2189-2191.
7. Wright, E. M.; Hirayama, B. A.; Loo, D. F. J. Intern. Med. 2007, 261, 32.
8. Kanai, Y.; Lee, W. S.; You, G.; Brown, D.; Hediger, M. A. J. Clin. Invest. 1994, 93,
397.
9. 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.
10. 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.
11. Grempler, R.; Thomas, L.; Eckhard, M.; Himmelsbach, F.; Sauer, A.; Sharp, D. E.;
Bakker, R. A.; Mark, M.; Klein, T.; Eickelmann, P. Diabetes, Obesity & Metabolism.
2012, 14, 83.
12. Abdul-Ghani, M. A.; deFronzo, R. A.; Norton, L. Diabetes 2013, 62, 3324-3328.

13. Dominguez, R.; Jessica, A.; Rieg, T. Diabetes, Obesity & Metabolism 2019, 21(S2),
43-52.
14. a) 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. b) Sims, H.; Smith, K. H.; Bramlage, P.; Minguet, J.
Diabetic Medicine 2018, 35, 1037-1048.
15. He Y.; Haynes W.; Meyers C. D.; Mendonza A. E; Ma S.; Bachman E.; Haynes W.;
Meyers C. D; Amer A.; Zhang Y.; Mahling P.; Chutkow W.; Bachman E. Diabetes,
Obesity & Metabolism 2019, 21, 1311-1321.
16. Hang, L. K.; Don, L. S.; Namdu, K.; Hyun, S. K.; Hoon, K. Y.; Soo, S. S. The Korean
Journal Physiology & Pharmacology, 2019, 23, 55-62.
17. Xu, G.; Gaul, M. D.; Kuo, G.; Du, F.; Xu, J. Z.; Wallace, N.; Hinke, S.; Kirchner, T.;
Silva, J.; Huebert, N.; Lee. S.; Murray, W.; Liang, Y.; Demarest, K. Bioorg. Med.
Chem. Lett. 2018, 28, 3446-3453.
18. Chen, Z.; Wang, R.; Qing, F. Tetrahedron Letters (2012) 53 (17), 2171-2176.
19. Okaya, S.; Okuyama, K.; Okano, K.; Tokuyama, H. Organic Synthesis 2016, 93, 63-
74.
20. Plewe, M.; Sandhoff, K.; Schmidt, R. R. Journal of Carbohydrate Chemistry 1992,
11, 881-890.

ABSTRACT: (2S,3R,4R,5S,6R)-2-Aryl-5,5-difluoro-6-(hydroxymethyl)tetrahydro-2H-
pyran-3,4-diols and (2S,3R,4R,5S,6R)-2-aryl-5-fluoro-5-methyl-6-
(hydroxymethyl)tetrahydro-2H-pyran-3,4-diols were discovered as dual inhibitors of
sodium glucose co-transporter proteins (e.g. SGLT1 and SGLT2) through rational drug
design, efficient synthesis, and in vitro and in vivo evaluation. Compound 6g demonstrated
potent dual inhibitory activities (IC50 = 96 nM for SGLT1 and IC50 = 1.3 nM for SGLT2).
It showed robust inhibition of blood glucose excursion in an oral glucose tolerance test
(OGTT) in Sprague Dawley (SD) rats when dosed at both 1 mg/kg and 10 mg/kg orally. It
also demonstrated postprandial glucose control in db/db mice when dosed orally at 10
mg/kg.
HO
O
O
O
HO
F
OH
F
6g
SGLT1 IC50 = 96 nM
SGLT2 IC50 = 1.3 nM
OH