Discovery of LX2761, a sodium-dependent glucose cotransporter 1 (SGLT1) Inhibitor restricted to the intestinal lumen, for the treatment of diabetes

Article abstract of DOI:10.1021/acs.jmedchem.6b01541

The increasing number of people afflicted with diabetes throughout the world is a major health issue. Inhibitors of the sodiumdependent glucose cotransporters (SGLT) have appeared as viable therapeutics to control blood glucose levels in diabetic patents. Herein we report the discovery of LX2761, a locally acting SGLT1 inhibitor that is highly potent in vitro and delays intestinal glucose absorption in vivo to improve glycemic control.

Full text of DOI:10.1021/acs.jmedchem.6b01541

Subscriber access provided by University of Newcastle, Australia
Article
Discovery of LX2761, a Sodium-dependent Glucose Cotransporter 1 (SGLT1)
Inhibitor Restricted to the Intestinal Lumen, for the Treatment of Diabetes.
Nicole C Goodwin, Zhi-Ming Ding, Bryce A Harrison, Eric D. Strobel, Angela L. Harris, Melinda Smith,
Andrea Y. Thompson, Wendy Xiong, Faika Mseeh, Debra J Bruce, Damaris Diaz, Suma Gopinathan, Ling
Li, Emily O'Neill, Mary Thiel, Alan G.E. Wilson, Kenneth G Carson, David R. Powell, and David B Rawlins
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 03 Jan 2017
Downloaded from http://pubs.acs.org on January 3, 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 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
Discovery of LX2761, a Sodium-dependent Glucose Cotransporter 1 (SGLT1) Inhibitor Restricted
to the Intestinal Lumen, for the Treatment of Diabetes.
7
8
9
Nicole C. Goodwin,*,† ZhiꢀMing Ding,‡ Bryce A. Harrison,† Eric D. Strobel,† Angela L. Harris,‡ Melinda
Smith,‡ Andrea Y. Thompson,‡ Wendy Xiong,‡ Faika Mseeh,§ Debra J. Bruce,ǁ Damaris Diaz,ǁ Suma
Gopinathan,ǁ Ling Li,ǁ Emily O’Neill,ǁ Mary Thiel,ǁ Alan G. E. Wilson,ǁ Kenneth G. Carson,† David R.
Powell,‡ David B. Rawlins,†
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
† Department of Medicinal Chemistry, Lexicon Pharmaceuticals, 110 Allen Road, Basking Ridge, New
Jersey, 07920.
‡ Department of Metabolism Research, § Department of Pharmaceutical Discovery, and ǁ Department of
Drug Metabolism, Pharmacokinetics, and Toxicology, Lexicon Pharmaceuticals, 8800 Technology
Forest Place, The Woodlands, Texas, 77381.
RECEIVED DATE (will be automatically inserted after manuscript is accepted).
Abstract. The increasing number of people afflicted with diabetes throughout the world is a
major health issue. Inhibitors of the sodiumꢀdependent glucose cotransporters (SGLT) have appeared as
viable therapeutics to control blood glucose levels in diabetic patents. Herein we report the discovery of
LX2761, a locallyꢀacting SGLT1 inhibitor that is highly potent in vitro and delays intestinal glucose
absorption in vivo to improve glycemic control.
Introduction. The prevalence of diabetes has become an increasing concern to the world’s
population. In 2012, the World Health Organization reported that the incidence of the disease was
estimated to be 9% in adults over the age of 18. That number is expected to rise due to population growth,
an aging population, and lifestyle tendencies that have led to an increase in obesity. Diabetes is projected
to be the seventh leading cause of death by 2030.1 It is a metabolic syndrome characterized by
hyperglycemia, which results either from an absolute deficiency in insulin secretion (type 1 diabetes) or
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 48
1
2
3
4
5
6
7
8
9
from resistance to insulin action combined with an inadequate compensatory increase in insulin secretion
(type 2 diabetes). Chronic hyperglycemia is a major risk factor for microvascular complications resulting
in retinopathy, nephropathy, and neuropathy, and contributes to macrovascular complications leading to
cardiovascular disease.2 Current therapies aimed at aggressive glycemic control are often accompanied
with side effects, most notably weight gain and severe hypoglycemia, which may, in turn, be associated
with an increased incidence of cardiovascular events. Additional treatments to aid in glycemic control are
essential to prevent the associated mortality and morbidity of this disease.3,4
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 potential of sodiumꢀdependent glucose cotransporters (SGLTs) as a new family of drug
targets for the treatment of diabetes has been realized over the past few decades.5,6 The SGLT family
consists of several isoforms that actively transport sugars across cellular membranes, a process that is
coupled with sodium ion transport.7,8 SGLT1 is primarily expressed in the gastrointestinal (GI) tract
where it is largely responsible for intestinal absorption of glucose and galactose.9,10 SGLT1 is also
present in the proximal straight tubule of the kidney where it helps to absorbs plasma glucose filtered
from the blood. In contrast, SGLT2 is primarily expressed upstream of SGLT1 in the proximal
convoluted tubule of the kidney.11 Under normal physiologic conditions, greater than 97ꢀ98% of the
plasma glucose filtered in the kidney is reabsorbed, and SGLT2 is responsible for approximately 90% of
this reabsorption. Any glucose not reabsorbed by SGLT2 is reabsorbed by SGLT1 or excreted into the
urine. Humans with inactivating SGLT2 mutations present with significant glucosuria but are otherwise
healthy.12,13 Humans14 and mice15 with inactivating SGLT1 mutations have minimal increase in
glucosuria but exhibit glucoseꢀgalactose malabsorption (GGM),16 a serious GI condition characterized by
diarrhea induced by the inability to absorb these sugars from the gut. The GGM phenotype can be
mitigated through early detection and dietary modifications, and individuals lacking SGLT1 appear to
gradually tolerate the introduction of glucose into their diet as they age.17 No other phenotypes are noted
in individuals with an SGLT1 genetic deficiency.
The focus of industrial programs has been on selective SGLT2 inhibition and several glucoseꢀ
based small molecules have been approved by the FDA for the treatment of diabetes.5,18,19
ACS Paragon Plus Environment

Page 3 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
Dapagliflozin20ꢀ22 and canagliflozin23ꢀ26 were both approved by the FDA in 2013 and empagliflozin27,28
was approved in 2014. In addition, ipragliflozin,29ꢀ31 luseogliflozin,32ꢀ34 and tofogliflozin35ꢀ37 have all been
approved in Japan. (Fig. 1).
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 1. Marketed SGLT2 inhibitors (2016).
Using our proprietary mouse knockout (KO) technology,38,39 we independently confirmed that
both SGLT1 and SGLT2 KO mice have strong antiꢀdiabetic phenotypes.40 Analogous to humans, mice
lacking SGLT2 had improved glycemic control due to an increase in urinary glucose excretion (UGE).
SGLT1 KO mice, when challenged with a glucoseꢀcontaining meal, displayed decreased postprandial
glucose (PPG) excursions, increased cecal glucose, and increased circulating glucagonꢀlike peptideꢀ1
(GLPꢀ1) levels, all consistent with delayed glucose absorption.41 Decreased PPG will improve glycemic
control, and, in addition, increased circulating GLPꢀ1 may further improve glycemic control through the
same mechanisms activated by GLPꢀ1 analogs and by the increased GLPꢀ1 levels that result from
treatment with dipeptidyl peptidase 4 (DPP4) inhibitors such as sitagliptin.42 The added benefit of
inhibiting the SGLT1 pathway compelled us to develop sotagliflozin, also known as LX4211 (1, Fig. 2), a
dual SGLT1 and SGLT2 inhibitor currently in lateꢀstage clinical development for diabetes.43
Sotagliflozin treatment significantly improves glycemic control in patients with diabetes, and this
improvement has been accompanied by a very favorable safety profile. Of particular note, diabetic
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 48
1
2
3
4
patients treated with compound 1 have not exhibited an increased frequency of diarrhea, a side effect that
5
6
might be expected from SGLT1 inhibition.44ꢀ48
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 2. Sotagliflozin (1), a dual SGLT1/2 inhibitor in clinical trials for type 1 and type 2 diabetes.
The potential of enhanced glycemic control through SGLT1 inhibition was attractive because it
would provide an alternative approach to glycemic control that was not dependent on kidney function.
Current SGLT2 selective inhibitors lose efficacy in patients with moderate to severe renal impairment,
which affects 30ꢀ40% of all diabetic patients.49ꢀ51 Although compound 1 at maximal clinical dose works
through this pathway by partially inhibiting intestinal SGLT1,48 a more potent SGLT1 inhibitor that is
restricted to the intestine could potentially be titrated to a dose that provides even greater glycemic control
through this mechanism before GI side effects occur, and would also avoid the glycosuriaꢀrelated side
effects of SGLT2 inhibitors, particularly genital tract infections.
Design of an SGLT1 inhibitor began with the thiomethyl xyloside core of 1 because that
particular anomeric substitution imparted a unique potency at SGLT1 where other xylose and glucose
cores did not.52 To our knowledge at the time, no compound selective for SGLT1 over SGLT2 in vitro
had been reported.53 Therefore, we focused our strategy on obtaining an SGLT1 selectivity profile based
on its in vivo expression where SGLT1, but not SGLT2, is expressed in the GI tract. We set out to design
a compound that would remain within the GI tract due to poor oral bioavailability. When we commenced
our chemistry effort sing this strategy, there were a few reported examples in the patent literature that
described the development of dual SGLT1 and SGLT2 inhibitors that targeted both SGLT1 in the
intestinal tract and both SGLT1 and SGLT2 in the kidney.54ꢀ56 Since then, Kissei has reported their
similar strategy for O-glycosides that are SGLT1 inhibitors.57ꢀ59
ACS Paragon Plus Environment

Page 5 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compounds of interest were required to demonstrate certain characteristics that are briefly
introduced here and will be discussed in detail in the following section. First, the compounds must
display high in vitro potency at SGLT1 in cellular assays. Second, these potent inhibitors must have
efficacy in vivo, specifically in a mouse oral glucose challenge where the critical endpoints were
decreased glucose excursion accompanied by increased cecal glucose and total plasma GLPꢀ1 (tGLPꢀ1)
levels, all of which indicate delayed intestinal glucose absorption from SGLT1 inhibition. Finally, the
compounds must display low systemic exposure determined directly through pharmacokinetic assessment
and indirectly through measurement of low UGE. As all of our compounds were potent inhibitors of
SGLT2 in addition to SGLT1, any significant absorption into the systemic circulation would increase
UGE by inhibiting SGLT1 and SGLT2 in the kidneys, thus negating the in vivo selectivity that we
desired by sequestration of the compound to the GI tract.
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
SAR and In Vitro Activity. In our efforts to find nonꢀabsorbable inhibitors of SGLT1, we chose
to keep the xyloside core of 1 intact while altering physical properties by doing structureꢀactivity
relationships (SAR) on the bisaryl aglycon moiety. As highlighted in Figure 3, chemical exploration was
conducted at three different positions: (1) the substitution “X” at the proximal aryl ring, (2) linker “Y” of
the distal aromatic ring with varying tether lengths “n”, and (3) distal functionality “Z” at the end of the
tether which would be used to modulate the physical properties.
Figure 3. Synthetic strategy to design a compound that is restricted to the lumen starting with the
xyloside core of 1.
Surprisingly and gratifyingly, modifications at these positions provided quick access to
compounds that were extremely potent at human SGLT1 and SGLT2.41 It is worth noting that tuning the
selectivity to significantly favor SGLT1 inhibition over SGLT2 inhibition was never observed in this
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 48
1
2
3
4
5
6
7
8
9
program. Substitution on the proximal aryl ring (Fig. 3, “X”) was explored as either X = Me or Cl. The
methyl substitution at this position was ultimately preferred for its slightly higher potency at SGLT1 but,
more importantly, for its compatibility with synthesis and starting material availability. The linker at the
distal aromatic ring (Fig. 3, “Y”) was explored with either Y = O or CH2. Both of these substitutions
were interchangeable in terms of their potency at SGLT1/2. The most influential position in establishing
potency at hSGLT1 in this series was the functional group (Fig. 3, “Z”) appended to the distal aromatic
ring of the aglycon moiety. In addition to providing potency, installation of polar and charged functional
groups were utilized to limit oral absorption of the compounds.60
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. Representative SAR of nonꢀabsorbable dual SGLT1/2 inhibitors.
A brief summary of substitution at the “Z” position is shown in Table 1. Simple carboxylic acids
(2) were universally not tolerated at hSGLT1. Otherwise, a large variety of functionality was tolerated but
needed to be positioned more than 2 linker atoms away from the aryl ring (3 and 4) to maintain potency.
In addition, combinations of neutral groups were not enough to render these compounds restricted from
systemic circulation (e.g. %F = 18% and 17% for diol 5 and amide 6, respectively). Installation of basic
amines that would have a net charge under physiochemical conditions was beneficial to decrease the
ACS Paragon Plus Environment

Page 7 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
permeability of these compounds (amines 7 – 8, %F = <2%) without sacrificing cellular potency. A
series of compounds that featured amides and charged amines (9 – 11) were designed and proved to be
extremely potent in vitro and exhibited oral bioavailabilities that were a fraction of a percent.
9
Further pharmacodynamic evaluation of these compounds revealed that, for reasons that are
unclear, the bisamides with a basic amine, such as piperidine amides 10 and 11, were more efficacious in
vivo than amines 7 ꢀ 9. In particular, piperidinylamide 11 was identified as an early lead compound and
required scaleꢀup of gram quantities for further in vivo profiling. However, upon scaleꢀup of amide 11, a
synthetic instability of the bisamide motif was observed under acidic conditions in protic solvent. As
shown in Figure 4, exposure of 11 to acidic methanolic solutions initiated a selective methanolysis of the
terminal amide to methyl ester 13 and insignificant amounts of the other potential methanolysis product
14 were observed. This regioselective decomposition was hypothesized to be occurring via oxazolone
intermediate 12 to favor formation of ester 13 in a selective decomposition pathway. In order to confirm
this degradation pathway, 13C NMR was used to detect the presence of the oxazolone intermediate 12.
Bisamide 11 was exposed to catalytic amounts of deuterium chloride (35 wt% in D2O) in deuterated
methanol, and the emergence of a new peak characteristic of an oxazolone (13C δ = 165 ppm) was
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
observed and then followed by conversion to the methyl ester 13 (13C δ = 172 ppm). Although this
hydrolysis was not observed during in vivo PK studies or in forced stability studies with simulated gastric
fluid or simulated intestinal fluids at 50 °C, we felt the stability would complicate future synthetic
chemistry development.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 48
1
2
3
4
Figure 4. Synthetic instability of piperazinylamide 11.
5
6
7
8
9
Consequently, compounds were synthesized that would disfavor or eliminate the formation of the
oxazolone intermediate 12 in order to render these compounds synthetically stable. All compounds
shown in Table 2 maintained low nanomolar in vitro potency against SGLT1 and were thus triaged by
their stability in accelerated degradation conditions (0.5N HCl in 1:1 methanol/water at 50 °C for 72
hours). Compounds that exhibited less than 5% degradation were considered to have acceptable stability.
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 2. Stability of replacements to the hydrolyticallyꢀlabile bisꢀamide motif of 11
Compound 15 removed the carbonyl required to form the oxazolone ring of intermediate 12 and,
as expected, 15 was stable to hydrolysis but lost in vitro activity at hSGLT1. Compounds 16-18 were
designed to alleviate the ThorpeꢀIngold effect, whereby geminal dimethyl substitution promotes
intramolecular cyclization reactions.61,62 In compound 16, replacement of the dimethyl substitution of 11
with a cyclopropyl should increase the bond angle at the carbon and thus kinetically decrease the rate of
oxazolone formation. In 17, the dimethyl substitution was removed altogether, and in 18, an extra
ACS Paragon Plus Environment

Page 9 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
methylene was inserted in expectation that the sixꢀmembered oxazinone ring should have a relatively
slower rate of formation than the corresponding fiveꢀmembered oxazolone ring. All three of these
compounds did indeed exhibit slower rates of acidic hydrolysis, however, none were sufficiently stable
for the purposes of this program.
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 last approach examined the electronic nature of the terminal carbonyl. As shown in Figure
5A, we hypothesized that acidic conditions could enforce a boat conformation through a bridging
intramolecular hydrogen bond. The nitrogen lone pairs would no longer be in conjugation with the
carbonyl and thus increase the electrophilicity of the amide and its susceptibility to nucleophilic attack.
Piperidine amide 19 was expected to be stable because it cannot adopt the conformation shown in Figure
5A. To our surprise, it was just as unstable as piperazinylamide 11, thus causing us to amend our
hypothesis. A simple steric argument is shown in Figure 5B. The cyclic amide has an unfavorable
interaction with the gemꢀdimethyl group if the nitrogen lone pair retains some sp2ꢀcharacter in
conjugation with the carbonyl. Rotation about the Nꢀcarbonyl bond would alleviate this steric interaction
and remove the sp2 character of the nitrogen, thus causing the carbonyl to be more reactive and
susceptible to nucleophilic attack of the neighboring amide. Therefore, if the amide were to be either an
uncyclized tertiary amide (20) or a secondary amide (21 and 22), it should be stable under acidic
conditions. Gratifyingly, compounds 20 – 22 not only exhibited stability past 72 hours in our accelerated
hydrolysis conditions, but they retained also excellent potency at both hSGLT1 and hSGLT2.
Figure 5. Destabilizing conformations of cyclic amides. (A) Acidꢀpromoted conformational change of
piperidinylamide engages nitrogen lone pair and destabilizes the carbonyl. (B) Unfavorable steric
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 48
1
2
3
4
5
6
interaction of piperidinylꢀamide with gemꢀdimethyl group favors a conformation that removes the
nitrogen lone pair from conjugation with the carbonyl.
7
8
9
The stable compounds from Table 2 were compared in a single dose oral glucose tolerance test
(OGTT). Compound 21 ultimately emerged as a very potent, chemicallyꢀstable, and luminallyꢀrestricted
molecule that achieved proof of concept in OGTT at very low doses. Based on the PK and PD data that is
discussed in detail in the following section, Nꢀ(1ꢀ((2ꢀ(dimethylamino)ethyl)amino)ꢀ2ꢀmethylꢀ1ꢀ
oxopropanꢀ2ꢀyl)ꢀ4ꢀ(4ꢀ(2ꢀmethylꢀ5ꢀ((2S,3R,4R,5S,6R)ꢀ3,4,5ꢀtrihydroxyꢀ6ꢀ(methylthio)tetrahydroꢀ2Hꢀ
pyranꢀ2ꢀyl)benzyl)phenyl)butanamide (21) was nominated as preclinical candidate LX2761.63
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
Results and Discussion.
In order to assess the potency of these SGLT1 inhibitors in an oral glucose challenge, male C57
mice were fed a lowꢀfat diet (10% lard fat, 35% sucrose) for six days prior to dosing the SGLT1 inhibitors
by oral gavage once daily for four consecutive days. At 22 hours after the last dose of inhibitor and at 16
hours after food was removed, the mice were challenged with a glucoseꢀcontaining meal administered by
oral gavage. Blood samples were collected via retroꢀorbital bleeding at various time points within the
first hour postꢀchallenge and the test subjects were necropsied at the end of that hour. Necropsies
provided measurement of cecal glucose and cecal pH levels while blood sampling provided blood glucose
and plasma tGLPꢀ1 levels.
The results from the oral glucose challenge are shown below in Figure 6. Mice treated with 21
exhibited a significant reduction in glucose excursion compared to vehicle and a significant increase in
postprandial circulating levels of tGLPꢀ1. As expected with a delayed glucose absorption that is
mechanistic of SGLT1 inhibition in the intestine, mice treated with 21 exhibited a significant increase in
cecal glucose levels and decrease in cecal pH due to fermentation of glucose to short chain fatty acids in
the cecum. The magnitude of these effects recapitulates the magnitudes observed in our previously
reported studies with SGLT1 KO mice.41
ACS Paragon Plus Environment

Page 11 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
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 6. Compound 21 delays intestinal glucose absorption and increases postprandial tGLP-1
levels in mice. Individual healthy mice received either vehicle (ꢀ) or 21 at a dose of 1.5 mg/kg (21, ꢀ)
by oral gavage for 4 consecutive days and then, 22 hours later after a 16 hour fast, received a glucoseꢀ
containing meal challenge by oral gavage. A) Blood samples were collected through retroꢀorbital
bleeding at 0, 10 30 and 60 minutes after meal challenge, and glucose AUC was calculated for each
glucose excursion. At 60 minutes after the meal challenge (23 hours after the dose of compound or
vehicle), mice were necropsied; blood samples collected at that time were used to measure B) plasma
tGLPꢀ1 levels, and cecal contents collected at that time were analyzed for C) total cecal glucose and D)
cecal pH. * Different from vehicle, P < 0.001.
Compound 21 was designed to remain in the intestine after oral delivery to inhibit SGLT1 locally
without affecting the SGLT1/2 mechanism in the kidney. The results of mouse pharmacokinetic (PK)
studies are summarized below in Figure 7 using the structurally similar and orally bioavailable compound
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 48
1
2
3
4
5
6
7
8
9
1 as an internal reference. After oral administration at 10 mg/kg, 21 exhibited limited bioavailability with
negligible concentrations detected in the plasma. Figure 8 shows the striking difference of plasma
compound levels over time after oral dosing of compounds 21 and 1, which in contrast affected high
exposure and bioavailability in mice (Cmax of 43 and 4073 nM, %F of <2 and 98%, respectively). Even
upon escalating the oral dose of 21, plasma exposure of the compound still remained very low.
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 7. Pharmacokinetic data comparing compounds 21 and 1 at 1mg/kg IV and 10 mg/kg (PO).
Figure 8. Plasma concentrationꢀtime course comparing compounds 21 and 1 at 10 and 30 mg/kg after
oral administration.
Lastly, in order to further support the hypothesis that 21 was restricted to the intestines and not
absorbed into the systemic circulation, this compound was compared to 1 for its ability to increase UGE
in mice. If 21 was present in the systemic circulation, it would be expected to increase UGE due to its
ACS Paragon Plus Environment

Page 13 of 48
Journal of Medicinal Chemistry
1
2
potent inhibitory activity at both SGLT1 and SGLT2 in the kidneys64 As shown in Figure 9, 21 had little
effect on UGE when measured for 2 days after mice received a single 1.5 mg/kg dose (day 0) of 21 by
oral gavage. As with the PK discussed previously, it is striking to compare this to 1 at the same dose.
During the 24 hours after compound dosing, 1 increased UGE by 453 ± 77 mg/day while 21 increased
UGE by a trivial 4 ± 9 mg/day.
3
4
5
6
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 9. Urinary glucose excretion comparing a single 1.5 mg/kg dose of 21 and 1 before (day 0) and
after (days 1 and 2) administration by oral gavage.
Chemistry. The medicinal chemistry synthesis of xyloside bisamide 21 is shown in Scheme 1.
Synthesis of aryl chloride aglycon 26 began with the formation of Weinreb amide 24 of 2ꢀmethylꢀ5ꢀ
bromobenzoic acid 23 through an acid chloride intermediate. Addition of pꢀchlorophenylmagnesium
chloride into Weinreb amide 24 provided the corresponding bisaryl ketone 25, which was then reduced
using triethylsilane in the presence of BF3·OEt2 to furnish the aglycon 26. Lithium/bromide exchange of
26 at −78 ºC with nꢀbutyllithium generated the aryl lithium species which was subsequently added to the
magnesium salt of morpholinoamide 2765 to cleanly yield ketone 28. A diastereoselective reduction of
the ketone under Luche conditions gave alcohol 29 in >20:1 ratio for the newly created diastereomeric
center. Refluxing alcohol 29 in aqueous acetic acid resulted in a concomitant acetonide deprotection and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 48
1
2
3
4
5
6
7
8
9
rearrangement to the pyran, which was then peracetylated to provide the crystalline tetracetate 30.
Diastereoselective installation of the thiomethyl betaꢀanomer of thioxyloside 31 was accomplished in a
twoꢀstep, oneꢀpot procedure. The anomeric acetate mixture of 30 was subjected to TMSOTf to generate
the transient oxocarbenium ion. The neighboring acetate at the 2ꢀposition blocks the α face of the sugar,
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
allowing for a nucleophilic attack by thiourea to occur only from the β face. The intermediate thiourea
adduct was then converted to the thiomethyl group by exposure to methyl iodide followed by N,N-
diisopropylethylamine (DIPEA). The order of the addition of these reagents influences the efficiency of
the reaction and suggests that the methyl iodide activates the sulfur prior to the DIPEAꢀpromoted
decomposition of the thiourea intermediate. Aryl chloride 31 was coupled with methyl butꢀ3ꢀenoate
using conditions previously reported by Fu66 to afford the trans olefin 32 in 60% yield. Hydrogenation of
the double bond of 32 followed by hydrolysis of the ester and acetate groups on the sugar provided acid
34. Finally, HATU coupling of 34 with 2ꢀaminoꢀNꢀ(2ꢀ(dimethylamino)ethyl)ꢀ2ꢀmethylpropanamide
provided 21.
Scheme 1. Synthesis of 21a
ACS Paragon Plus Environment

Page 15 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
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
aReagents and conditions: (a) (COCl)2, DMF (cat.), CH2Cl2; MeNH(OMe)·HCl, Et3N, CH2Cl2; 99% yield
(b) (4ꢀchlorophenyl)magnesium chloride, THF, 0 °C; 99% yield (c) Et3SiH, BF3·OEt2, CH3CN, 60 °C;
62% yield (d) nꢀBuLi, THF, −78 °C, then 27/tꢀBuMgCl, THF, 0 °C to 23 °C; 70% yield (e) NaBH4,
CeCl3·7H2O, MeOH; 99% yield (f) HOAc/H2O, 100 °C; Ac2O, Et3N, CH3CN; 92% yield (g) TMSOTf,
thiourea, dioxane, 80 °C; MeI then DIPEA, 23 °C; 60% yield (h) methyl butꢀ3ꢀenoate, Pd2dba3, [tꢀ
Bu3PH]BF4, Cy2NMe, NMP, 160 °C, microwave irradiation; 60% yield (i) H2, Pd/C; 94% yield (j) LiOH,
MeOH/H2O; 99% yield (k) 2ꢀaminoꢀNꢀ(2ꢀ(dimethylamino)ethyl)ꢀ2ꢀmethylpropanamide hydrochloride,
HATu, DIPEA, DMF; 40% yield after prep HPLC.
Conclusion. The discovery of preclinical candidate 21 for the selective inhibition of SGLT1
has been disclosed for the first time. 21 is a chemically stable and very potent inhibitor against SGLT1
and SGLT2 in vitro but displays specific pharmacology attributed to SGLT1 inhibition in the GI tract.
We expect that this strategy could provide a promising potential therapeutic for the treatment of diabetes,
particularly in patients with renal impairment who cannot benefit from the renal mechanism of selective
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 48
1
2
3
4
5
6
SGLT2 inhibitors. Further results of the pharmacology and development of this compound for
therapeutic use will be reported in subsequent publications.
7
8
9
Experimental Section.
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
Chemical Methods. All reactions were conducted under a static atmosphere of argon or
nitrogen and stirred magnetically unless otherwise noted. Reagents, starting materials, and solvents were
purchased from commercial suppliers and used as received. Flash column chromatography was carried
out using preꢀpacked silica gel columns from Biotage or ISCO, or by slurry preparation using EMD silica
1
gel 60 (particle size 0.040ꢀ0.063 mm). H and 13C NMR spectra were collected on Bruker ARX300,
DRX400 or DPX400 NMR spectrometers. Chemical shifts are reported in parts per million (ppm) relative
to tetramethylsilane in δꢀunits, and coupling constants (Jꢀvalues) are given in hertz (Hz). Data are
reported in the following format: chemical shift, multiplicity, coupling constants, and assignment.
Reactions were monitored by TLC using 0.25 mm E. Merck silica gel plates (60 F254) and were
visualized with UV light. Analytical HPLC spectra were collected on Shimadzu HPLC systems equipped
with a UV detector measuring absorbance at 220 and 254 nm. Mass spectra were obtained on Waters ZQ
or ZMD LCMS systems equipped with an autoꢀsampler, ELSD detector, a UV detector measuring
absorbance at 220 and 254 nm, and a mass detector. High resolution mass spectra were obtained on a
Waters LCT Premier XE Micromass® MS Technologies instrument equipped with an autoꢀsampler. All
compounds submitted for in vitro testing were >95% purity (HPLC) and those for in vivo testing were
>98% purity (HPLC).
(5-Bromo-2-methylphenyl)(4-chlorophenyl)methanone (25). 2ꢀMethylꢀ5ꢀbromobenzoic acid
(23, 26.0 g, 121 mmol) and oxalyl chloride (13.2 mL, 152 mmol) were suspended in 520 mL of CH2Cl2.
A catalytic amount of DMF (0.5 mL) was added in small portions and the reaction was stirred at room
temperature until the reaction became homogenous. The volatiles were removed in vacuo. The crude
material was dissolved in 200 mL of CH2Cl2 and N,Oꢀdimethylhydroxylamine hydrochloride (23.6 g, 242
mmol) was added. The reaction was cooled to 0 °C and triethylamine (55 mL, 399 mmol) was slowly
ACS Paragon Plus Environment

Page 17 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
added. Upon completion of addition of triethylamine, the reaction was warmed to room temperature and
stirred overnight. The reaction was quenched with 50% saturated aqueous NaHSO4. The aqueous layer
was extracted twice with CH2Cl2. The combined organic layers were washed with brine, dried over
Na2SO4, filtered, and the solvent removed in vacuo. The resulting Weinreb amide (24, 31.3 g, 99%
yield) was used without further purification in the next step.
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 Weinreb amide (24, 31.3g, 121 mmol) was taken up in 250 mL of dry THF. 4ꢀ
Chloromagnesium bromide (1M in Et2O, 182 mL, 182 mmol) was added at room temperature, and the
reaction stirred for 2 hours. If the reaction was not complete, additional Grignard reagent was added until
LCMS indicated reaction completion. The reaction was quenched with a solution of saturated aqueous
NH4Cl/brine (1:1 v:v) and extracted twice with EtOAc. The combined organic layers were washed with
brine, dried over MgSO4, filtered, and the solvent removed in vacuo. (5ꢀBromoꢀ2ꢀmethylphenyl)(4ꢀ
chlorophenyl)methanone (25, 37.0 g, 99% yield) was used without further purification in the next step.
1H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 7.74 (d, J=8.3 Hz, 2 H), 7.53 (dd, J=8.1, 2.0 Hz, 1 H),
7.46 (d, J=8.3 Hz, 2 H), 7.42 (d, J=2.0 Hz, 1 H), 7.18 (d, J=8.1 Hz, 1 H), 2.26 (s, 3 H). GCMS (CH4ꢀCI)
[M+H] + = 309.
4-Bromo-2-(4-chlorobenzyl)-1-methylbenzene
(26).
(5ꢀBromoꢀ2ꢀmethylphenyl)(4ꢀ
chlorophenyl)methanone (25, 37.0 g, 121 mmol) and triethylsilane (77.3 mL, 484 mmol) were dissolved
in 300 mL of CH3CN and cooled to 0 °C. BF3·OEt2 (91 mL, 726 mmol) was added and the reaction was
heated to 60 °C for 2 hours. GCMS was used to monitor the reaction. Upon completion, the reaction was
cooled to 0 °C and quenched with 500 mL saturated aqueous NaHCO3. The aqueous phase was extracted
twice with EtOAc. The combined organic layers were washed with H2O and brine, dried over Na2SO4,
filtered, and the solvent removed in vacuo. The crude solid was slurried in 20% EtOAc/hexanes and
passed over a silica plug to remove residual salts. Concentration of the filtrate provided the title
compound as a white solid (22.0 g, 62% yield). 1H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 7.22 (d,
J=2.0 Hz, 1 H), 7.21 ꢀ 7.31 (m, 2 H), 7.04 (d, J=8.3 Hz, 2 H), 7.04 (d, J=8.1 Hz, 2 H), 3.91 (s, 2 H), 2.17
(s, 3 H). GCMS (CH4ꢀCI) [M+H] + = 295.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 48
1
2
3
4
(3-(4-Chlorobenzyl)-4-methylphenyl)((3aS,5R,6S,6aS)-6-hydroxy-2,2-
5
6
7
8
dimethyltetrahydrofuro[2,3-d][1,3]dioxol-5-yl)methanone (28). To a solution of ((3aS,5R,6S,6aS)ꢀ6ꢀ
hydroxyꢀ2,2ꢀdimethyltetrahydrofuro[2,3ꢀd][1,3]dioxolꢀ5ꢀyl)(morpholino)methanone (27, 25.3 g, 92.6
mmol) in THF (200 mL) under nitrogen at 0 °C was added tertꢀbutylmagnesium chloride (1 M in THF,
100 mL, 100 mmol). This solution was stirred at 0 °C for 30 minutes. Meanwhile, a solution of 4ꢀ
bromoꢀ2ꢀ(4ꢀchlorobenzyl)ꢀ1ꢀmethylbenzene (26, 32.9 g, 111.1 mmol) in THF (330 mL) under nitrogen
was cooled to ꢀ78 °C. nꢀButyllithium (2.5 M in hexanes, 48 mL, 120 mmol) was added dropwise via
syringe and stirred for 10 min. The magnesium alkoxide solution was transferred via cannula into the
aryllithium solution at ꢀ78 °C. The reaction was stirred for 30 min at ꢀ78 °C, allowed to warm to room
temperature and stirred for 60 min, quenched with 500 mL of a 1:1 (v:v) solution of saturated aqueous
NH4Cl/brine. The aqueous layer was extracted two times with 300 mL of EtOAc. The combined organic
layers were washed with brine, dried over MgSO4, filtered, and concentrated under vacuum. The crude
residue was taken up in 100 mL of EtOAc and heated until most of the solids dissolved. Hexanes (250
mL) was added and the flask was chilled in an ice bath for two hours. The white precipitate was filtered
off and washed with 20% EtOAc/hexane, providing the title compound as a white solid (26.09 g, 70%
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
1
yield). H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 7.88 (dd, J=7.8, 1.8 Hz, 1 H), 7.76 (d, J=1.5 Hz,
1 H), 7.29 (d, J=8.1 Hz, 1 H), 7.26 (d, J=8.3 Hz, 2 H), 7.05 (d, J=8.3 Hz, 2 H), 6.08 (d, J=3.8 Hz, 1 H),
5.28 (d, J=2.8 Hz, 1 H), 4.59 (d, J=3.5 Hz, 1 H), 4.57 (t, J=3.2 Hz, 1 H), 4.01 (s, 2 H), 3.06 (d, J=4.0 Hz,
1 H), 2.30 (s, 3 H), 1.46, (s, 3 H), 1.37 (s, 3 H). MS (ES+) [M+H]+ = 403.
(3aS,5S,6R,6aS)-5-((S)-(3-(4-Chlorobenzyl)-4-methylphenyl)(hydroxy)-methyl)-2,2-
dimethyltetrahydrofuro[2,3-d][1,3]dioxol-6-ol
(29).
(3ꢀ(4ꢀChlorobenzyl)ꢀ4ꢀ
methylphenyl)((3aS,5R,6S,6aS)ꢀ6ꢀhydroxyꢀ2,2ꢀdimethyltetrahydrofuro[2,3ꢀd][1,3]dioxolꢀ5ꢀyl)methanone
(28, 26.1 g, 64.9 mmol) and CeCl3‧7H2O (29.0 g, 77.9 mmol) were suspended in 520 mL of MeOH.
Sodium borohydride (982 mg, 26.0 mmol, dissolved in 10 mL of 1N aqueous NaOH) was added and the
reactants slowly went into solution over about 5 minutes. Another 100 mg (2.6 mmol) of sodium
ACS Paragon Plus Environment

Page 19 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
borohydride was added to push the reaction to completion. The reaction was stirred for 10 minutes and
quenched with 500 mL of saturated aqueous NH4Cl. Most of the MeOH was removed in vacuo and the
residual solvents were diluted with a 1:1 (v:v) solution of saturated aqueous NH4Cl:brine. The aqueous
layer was extracted three times with 500 mL of EtOAc. The combined organic layers were washed with
brine, dried over MgSO4, filtered, and concentrated under vacuum. The crude product was used without
further purification in the next step (26.2 g, 99% yield, >10:1 d.r.). 1H NMR (400 MHz,
CHLOROFORMꢀd) δ ppm 7.14 ꢀ 7.31 (m, 5 H), 7.04 (d, J=8.3 Hz, 2 H), 6.04 (d, J=3.8 Hz, 1 H), 5.24 (t,
J=3.4 Hz, 1 H), 4.51 (d, J=3.8 Hz, 1 H), 4.14 ꢀ 4.21 (m, 2 H), 4.04 (d, J=1.5 Hz, 1 H), 3.97 (s, 2 H), 2.77
(d, J=3.0 Hz, 1 H), 2.20 ꢀ 2.27 (m, 3 H), 1.46 (s, 3 H), 1.33 (s, 3 H). MS (ES+) [M+NH4]+ = 422.
(3S,4R,5S,6S)-6-(3-(4-Chlorobenzyl)-4-methylphenyl)tetrahydro-2H-pyran-2,3,4,5-tetrayl
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
tetraacetate (30). (3aS,5S,6R,6aS)ꢀ5ꢀ((S)ꢀ(3ꢀ(4ꢀChlorobenzyl)ꢀ4ꢀmethylphenyl)(hydroxy)ꢀmethyl)ꢀ2,2ꢀ
dimethyltetrahydrofuro[2,3ꢀd][1,3]dioxolꢀ6ꢀol (29, 26.2 g, 64.8 mmol) was suspended in 150 mL H2O
and 150 mL glacial acetic acid. The reaction was heated to 100 °C for 7 hours. The solvents were
removed in vacuo and the crude residue was azeotroped three times from toluene. The crude material was
placed on the high vacuum overnight and used in the next step without further purification. The crude
material was dissolved in 350 mL of CH3CN. Triethylamine (57.5 mL, 414 mmol) and acetic anhydride
(46.0 mL, 414 mmol) were added, followed by a catalytic amount of DMAP (100 mg). The reaction was
stirred at room temperature for 1 hour. About 200 mL of CH3CN was removed in vacuo and the
remainder was diluted with 600 mL of EtOAc. The organic layer was washed twice with 50% saturated
aqueous NaHSO4. The acidic aqueous layers were backextracted with 300 mL of EtOAc. The combined
organic layers were washed with brine, dried over MgSO4, filtered, and concentrated under vacuum. The
crude residue was azeotroped twice from toluene and once from hexanes to provide the title compound as
1
an easily transferable beige solid (34.0 g, 92% yield, mixture of α and β anomers). H NMR (400 MHz,
CHLOROFORMꢀd) δ ppm 7.24 (d, J=8.3 Hz, 2 H), 7.13 ꢀ 7.21 (m, 2H), 7.09 (s, 1 H), 7.01 (d, J=8.3 Hz,
2 H), 6.47 (d, J=3.5 Hz, 1 Ha), 5.89 (d, J=8.3 Hz, 1 Hb), 5.59 (t, J=9.8 Hz, 1 Ha), 5.37 (t, J=9.6 Hz, 1
Hb), 5.23 ꢀ 5.31 (m, 1 Ha + 1 Hb), 5.19 (t, J=9.6 Hz, 1 Hb), 5.14 (t, J=9.7 Hz, 1 Ha), 4.82 (d, J=10.1 Hz,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 48
1
2
3
4
5
6
7
8
1 Ha), 4.51 (d, J=9.9 Hz, 1 Hb), 3.94 (s, 2 H), 2.21 (s, 3 Ha), 2.20 (s, 3 Ha), 2.19 (s, 3 Hb), 2.11 (s, 3 Hb),
2.07 (s, 3 Hb), 2.06 (s, 3 Ha), 2.04 (s, 3 Ha), 2.03 (s, 3 Hb), 1.79 (s, 3 Ha), 1.77 (s, 3 Hb). MS (ES+)
[M+NH4]+ = 550.
9
(2S,3S,4R,5S,6R)-2-(3-(4-Chlorobenzyl)-4-methylphenyl)-6-(methylthio)tetrahydro-2H-
pyran-3,4,5-triyl triacetate (31). Trimethylsilyl trifluoromethanesulfonate (19.7 mL, 108.5 mmol) was
added to a solution of (3S,4R,5S,6S)ꢀ6ꢀ(3ꢀ(4ꢀchlorobenzyl)ꢀ4ꢀmethylphenyl)tetrahydroꢀ2Hꢀpyranꢀ2,3,4,5ꢀ
tetrayl tetraacetate (30, 33.9 g, 63.8 mmol) and thiourea (9.71 g, 128 mmol) in 340 mL of dioxane. The
reaction was heated to 80 °C for two hours, at which point LCMS analysis revealed the reaction was
stalled. Additional TMSOTf was added (2 mL, 10.8 mmol) and the reaction stirred for 1 hour at 80°C.
The reaction was cooled to room temperature. Sequential addition of methyl iodide (11.9 mL, 191 mmol)
followed by DIPEA (55.6 mL, 319 mmol) was performed, allowing the reaction to stir for 18 hours. 500
mL of H2O was slowly added to quench the reaction. The aqueous layer was extracted two times with
300 mL EtOAc. The combined organic layers were washed with saturated aqueous NaHSO4 and brine,
dried over MgSO4, filtered, and concentrated under vacuum. The crude solid was slurried in 300 mL of
MeOH. Sonication resulted in the precipitation of a light beige precipitate, which was filtered and
washed with cold MeOH. The filtrate was concentrated and the slurry procedure repeated once more to
provide and combined with the first batch. The product was isolated as pure beta anomer as a light beige
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
solid (20.4 g, 60% yield). H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 7.24 (d, J=8.6 Hz, 2 H), 7.10 ꢀ
7.18 (m, 2 H), 7.05 (s, 1 H), 7.00 (d, J=8.6 Hz, 2 H), 5.34 (dd, J=9.6 Hz, 1 H), 5.21 (dd, J=9.6 Hz, 1 H),
5.12 (dd, J=9.6 Hz, 1 H), 4.53 (d, J=9.9 Hz, 1 H), 4.39 (d, J=9.9 Hz, 1 H), 3.86 ꢀ 4.00 (m, 2 H), 2.19 (s, 3
H), 2.17 (s, 3 H), 2.10 (s, 3 H), 2.01 (s, 3 H), 1.76 (s, 3 H). MS (ES+) [M+NH4]+ = 538.
(2S,3S,4R,5S,6R)-2-(3-(4-((E)-4-Methoxy-4-oxobut-1-en-1-yl)benzyl)-4-methylphenyl)-6-
(methylthio)tetrahydro-2H-pyran-3,4,5-triyl triacetate (32). A microwave vial was charged with
(2S,3S,4R,5S,6R)ꢀ2ꢀ(3ꢀ(4ꢀchlorobenzyl)ꢀ4ꢀmethylphenyl)ꢀ6ꢀ(methylthio)tetrahydroꢀ2Hꢀpyranꢀ3,4,5ꢀtriyl
triacetate (31, 1.04 g, 2.0 mmol), methyl butꢀ3ꢀenoate (600 mg, 6.0 mmol), Pd2dba3 (183 mg, 0.20 mmol),
tri(tertꢀbutyl)phosphonium tetrafluoroborate (235 mg, 0.80 mmol), dicyclohexylmethylamine (1.27 mL,
ACS Paragon Plus Environment

Page 21 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
6.0 mmol), and Nꢀmethylpyrrolidinone (10 mL). The reaction was heated in the microwave at 160 °C for
20 min. The reaction was filtered over Celite with excess EtOAc. The organic layer was washed with
H2O, saturated aqueous NaHSO4, and brine. It was dried with MgSO4 and concentrated in vacuo. Silica
gel flash chromatography (gradient 10ꢀ50% EtOAc/hexanes) provided Heck adduct 32 as a light yellow
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
solid (700 mg, 60% yield). Minor amounts of isomerized olefin were observed in the 1H NMR. H NMR
(400 MHz, CHLOROFORMꢀd) δ ppm 7.28 ꢀ 7.31 (m, 2 H), 6.97 ꢀ 7.19 (m, 5 H), 6.46 (d, J=15.9 Hz, 1
H), 6.25 (dt, J=15.9, 7.1 Hz, 1 H), 5.33 (dd, J=9.6 Hz, 1 H), 5.21 (dd, J=9.6 Hz, 1 H), 5.12 (dd, J=9.6 Hz,
1 H), 4.52 (d, J=9.6 Hz, 1 H), 4.39 (d, J=9.6 Hz, 1 H), 3.87 ꢀ 4.01 (m, 2 H), 3.72 (s, 3 H), 3.24 (dd, J=7.1,
1.3 Hz, 2 H), 2.21 (s, 3 H), 2.17 (s, 3 H), 2.10 (s, 3 H), 2.01 (s, 3 H), 1.75 (s, 3 H). MS (ES+) [M+NH4]+ =
602.
(2S,3S,4R,5S,6R)-2-(3-(4-(4-Methoxy-4-oxobutyl)benzyl)-4-methylphenyl)-6-
(methylthio)tetrahydro-2H-pyran-3,4,5-triyl triacetate (33).
(2S,3S,4R,5S,6R)ꢀ2ꢀ(3ꢀ(4ꢀ((E)ꢀ4ꢀ
Methoxyꢀ4ꢀoxobutꢀ1ꢀenꢀ1ꢀyl)benzyl)ꢀ4ꢀmethylphenyl)ꢀ6ꢀ(methylthio)tetrahydroꢀ2Hꢀpyranꢀ3,4,5ꢀtriyl
triacetate (32, 1.74 g, 3.0 mmol) was dissolved in a 1:1 (v:v) THF/MeOH solution. Pd/C (10% wet, 174
1
mg) was added and the reaction hydrogenated at 40 psi for 3 hours. The reaction was monitored by H
NMR. Upon completion, the reaction was filtered over Celite with excess MeOH. Removal of solvents
in vacuo provided the product as a light yellow solid (1.65 g, 94% yield). 1H NMR (400 MHz,
CHLOROFORMꢀd) δ ppm 7.11 ꢀ 7.20 (m, 2 H), 7.07 (t, J=7.8 Hz, 3 H), 6.99 (d, J=8.1 Hz, 2 H), 5.33
(dd, J=9.6 Hz, 1 H), 5.21 (dd, J=9.6 Hz, 1 H), 5.12 (dd, J=9.6 Hz, 1 H), 4.52 (d, J=9.9 Hz, 1 H), 4.39 (d,
J=9.9 Hz, 1 H), 3.85 ꢀ 4.00 (m, 2 H), 3.67 (s, 3 H), 2.61 (t, J=7.6 Hz, 2 H), 2.33 (t, J=7.5 Hz, 2 H), 2.21
(s, 3 H), 2.18 (s, 3 H), 2.10 (s, 3 H), 2.01 (s, 3 H), 1.93 (quin, J=7.6 Hz, 2 H), 1.75 (s, 3 H). MS (ES+)
[M+NH4]+ = 604.
4-(4-(2-Methyl-5-((2S,3R,4R,5S,6R)-3,4,5-trihydroxy-6-(methylthio)tetrahydro-2H-pyran-2-
yl)benzyl)phenyl)butanoic acid (34).
(2S,3S,4R,5S,6R)ꢀ2ꢀ(3ꢀ(4ꢀ(4ꢀMethoxyꢀ4ꢀoxobutyl)benzyl)ꢀ4ꢀ
methylphenyl)ꢀ6ꢀ(methylthio)tetrahydroꢀ2Hꢀpyranꢀ3,4,5ꢀtriyl triacetate (33, 1.65 g, 2.81 mmol) was
dissolved in a MeOH/THF/H2O solution (25 mL, 2:1:2 vol. ratio). Lithium hydroxide (674 mg, 28.1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 48
1
2
3
4
5
6
7
8
9
mmol) was added and the reaction stirred at room temperature for 1 hour. The reaction was acidified to
pH = 1ꢀ2 with saturated aqueous NaHSO4. The acidic aqueous layer was extracted three times with
EtOAc. The combined organic layers were washed with brine, dried over MgSO4, filtered, and
concentrated in vacuo. The crude product was rotovapped down once from hexanes to provide 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
1
product as white transferrable solid (1.27 g, 99% yield). H NMR (400 MHz, DMSOꢀd6) δ ppm 11.99 (s,
1 H), 6.96 ꢀ 7.16 (m, 7 H), 4.39 (d, J=9.6 Hz, 1 H), 4.12 (d, J=9.1 Hz, 1 H), 3.90 (s, 2 H), 3.34 —3.51 (m,
3 H), 2.53 (t, J=7.3 Hz, 2 H), 2.19 (t, J=7.3 Hz, 2 H), 2.17 (s, 3 H), 2.03 (s, 3 H), 1.76 (quin, J=7.6 Hz, 2
H). MS (ES+) [M+NH4]+ = 464.
N-(1-((2-(dimethylamino)ethyl)amino)-2-methyl-1-oxopropan-2-yl)-4-(4-(2-methyl-5-
((2S,3R,4R,5S,6R)-3,4,5-trihydroxy-6-(methylthio)tetrahydro-2H-pyran-2-
yl)benzyl)phenyl)butanamide
(21).
4ꢀ(4ꢀ(2ꢀMethylꢀ5ꢀ((2S,3R,4R,5S,6R)ꢀ3,4,5ꢀtrihydroxyꢀ6ꢀ
(methylthio)tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)benzyl)phenyl) butanoic acid (34, 157 mg, 0.35 mmol), 2ꢀaminoꢀ
Nꢀ(2ꢀ(dimethylamino)ethyl)ꢀ2ꢀmethylpropanamide hydrochloride (73 mg, 0.53 mmol), HATU (161 mg,
0.42 mmol), and DIPEA (0.15 mL, 1.06 mmol) were combined in DMF (2 mL) and stirred for 2 hours at
room temperature. The reaction was quenched with saturated aqueous NaHCO3 and extracted twice with
EtOAc. The combined organic layers were washed with brine, dried with MgSO4, filtered, and
concentrated under vacuum. The residue was purified by preparative HPLC (C18 30 x 100 mm column,
5ꢀ100% CH3CN/10 mM aqueous ammonium formate, 45 mL/min) to provide the title compound 21 as
1
the formate salt after lyophilization (75 mg, 40% yield). H NMR (400 MHz, MeOHꢀd4) δ ppm 8.52 (s, 1
H), 7.12 ꢀ 7.21 (m, 3 H), 7.03 ꢀ 7.12 (m, 4 H), 4.39 (d, J=9.6 Hz, 1 H), 4.13 (d, J=9.3 Hz, 1 H), 3.96 (s, 2
H), 3.51 (t, J=5.6 Hz, 2 H), 3.33 ꢀ 3.47 (m, 3 H), 3.07 (t, J=4.8 Hz, 2 H), 2.79 (s, 6 H), 2.60 (t, J=7.6 Hz,
2 H), 2.21 (s, 3 H), 2.22 (t, J=7.6 Hz, 2 H), 2.14 (s, 3 H), 1.88 (quin, J=7.5 Hz, 2 H), 1.41 (s, 6 H). MS
(ES+) [M+H]+ = 602.
Biological Methods (In Vitro). Human SGLT1 Inhibition Assay. Human sodium/glucose coꢀ
transporter type 1 (SGLT1; accession number NP_000334; GI: 4507031) was cloned into pIRESpuro2
vector for mammalian expression (construct: HAꢀSGLT1ꢀpIRESpuro2). HEK293 cells were transfected
ACS Paragon Plus Environment

Page 23 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with the human HAꢀSGLT1ꢀpIRESpuro2 vector and the stably transfected cell line was established in the
presence of 0.5 ꢁg/mL of puromycin. Human HAꢀSGLT1 cells were maintained in DMEM media
containing 10% fetal bovine serum, 2 M Lꢀglutamine, 100 units penicllin/mL, 0.1 mg/mL streptomycin,
and 0.5 ꢁg/mL puromycin. The HEK293 cells expressing the human HAꢀSGLT1 were seeded in 384
well plates (30,000 cells/well) in DMEM media containing 10% fetal bovine serum, 2 M Lꢀglutamine,
100 units penicllin/mL, 0.1 mg/mL streptomycin, and 0.5 ꢁg/mL puromycin, then incubated overnight
at 37 °C, 5% CO2. Cells were then washed with uptake buffer (140 mM NaCl, 2 mM KCl, 1 mM CaCl2, 1
mM MgCl2, 10 mM HEPES, 5 mM Tris, 1 mg/mL bovine serum albumin (BSA), pH 7.3). Twenty
microliters of uptake buffer with or without testing compounds were added to the cells. Then, 20
microliters of uptake buffer containing 14CꢀAMG (100 nCi) were also added to cells. The cell plates were
incubated at 37 °C, 5% CO2 for 1ꢀ2 hours. After washing the cells with uptake buffer, scintillation fluid
was added (40 microliters/well) and 14CꢀAMG uptake was measured by counting radioactivity using a
scintillation coulter (TopCoulter NXT; Packard Instruments). The calculation of the IC50 is performed
using XLFit4 software, Model 205 (ID Business Solutions Inc., Bridgewater, NJ 08807) for Microsoft
Excel.
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
Human SGLT2 Inhibition Assay. Human sodium/glucose coꢀtransporter type 2 (SGLT2;
accession number P31639; GI:400337) was cloned into pIRESpuro2 vector for mammalian expression
(construct: HAꢀSGLT2ꢀpIRESpuro2). HEK293 cells were transfected with the human HAꢀSGLT2ꢀ
pIRESpuro2 vector and the stably transfected cell line was established in the presence of 0.5 ꢁg/mL of
puromycin. Human HAꢀSGLT2 cells were maintained in DMEM media containing 10% fetal bovine
serum, 2 M L-glutamine, 100 units penicllin/mL, 0.1 mg/mL streptomycin, and 0.5 ꢁg/mL
puromycin. The HEK293 cells expressing the human HAꢀSGLT2 were seeded in 384 well plates
(30,000 cells/well) in DMEM media containing 10% fetal bovine serum, 2 M L-glutamine, 100
units penicllin/mL, 0.1 mg/mL streptomycin, and 0.5 ꢁg/mL puromycin, then incubated
overnight at 37 °C, 5% CO2. Cells were then washed with uptake buffer (140 mM NaCl, 2 mM KCl, 1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 48
1
2
3
4
5
6
7
8
9
mM CaCl2, 1 mM MgCl2, 10 mM HEPES, 5 mM Tris, 1 mg/mL bovine serum albumin (BSA), pH 7.3).
Twenty microliters of uptake buffer with or without testing compounds were added to the cells. Then, 20
microliters of uptake buffer containing 14CꢀAMG (100 nCi) were added to the cells. The cell plates were
incubated at 37°C, 5% CO2 for 1ꢀ2 hours. After washing the cells with uptake buffer, scintillation fluid
was added (40 microliters/well) and 14CꢀAMG uptake was measured by counting radioactivity using a
scintillation coulter (TopCoulter NXT; Packard Instruments). The calculation of the IC50 is performed
using XLFit4 software, Model 205 for Microsoft Excel.
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
Biological Methods (In Vivo). All studies were performed at Lexicon Pharmaceuticals, Inc., in
strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of
the National Institutes of Health. The protocols for all studies were approved by the Lexicon Institutional
Animal Care and Use Committee (OLAW Assurance Number, A4152ꢀ01; AAALAC International
Accreditation Number, 001025). General methods for mouse care have been described by us
previously.40,59 C57BL/6ꢀTyrcꢀBrd (C57) mice obtained from an inꢀhouse colony were fed either standard
rodent chow diet (5010, LabDiet, PMI Nutrition International, St. Louis, MO) containing 23% calories as
fat or low fat diet (LFD; D12451, Research Diets, New Brunswick, NJ) containing 10% calories as lard
fat and 35% calories as sucrose. Mice were maintained in a temperatureꢀcontrolled environment on a
fixed 12 hour light/12 hour dark cycle and with free access to water and food. In all in vivo studies
presented here, all compounds were administered by oral gavage in a volume of no more than 10 mL/kg,
and the vehicle was always aqueous 0.1% v/v Tween 80.
Urinary Glucose Excretion Study in C57 Mice. At 15 weeks of age, male C57 mice were
individually housed in metabolic cages (Nalge Nunc International, Rochester, NY) and fed powdered
lowꢀfat diet (LFD, 10% lard fat, 35% sucrose) homogenized in water at a 2:1 (wt:wt) ratio. This paste
diet prevented contamination of urine with crumbled diet. After an acclimatization period, each mouse
received, by oral gavage, a single dose of either vehicle or vehicle containing either 21 or 1. Complete
24ꢀhour urine samples were collected during the 24 hours after dosing (day 1) and also on day 2. After
the volume of each 24ꢀhour urine collection was recorded, the urine sample was centrifuged and analyzed
ACS Paragon Plus Environment

Page 25 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
for glucose concentration (Cobas Integra 400 Clinical Chemistry Autoanalyzer, Roche Diagnostics,
Indianapolis, IN).
7
8
9
Glucose-containing Meal Challenge Study in C57 Mice. LFDꢀfed, adult male C57 mice were
randomized into treatment groups by body weight. Each mouse received either vehicle or 21 by oral
gavage and then, 22 hours later after a 16 hour fast, received a meal consisting of LFD supplemented with
glucose. This meal was prepared, as described previously41: 50 g of LFD powder and 9.4 g of glucose
were added to water at 60 °C; the final volume was 94 mL and the meal was kept until used. The mice
received 25 ml/kg of this meal (9.2 g/kg glucose, 2.5 g/kg protein, 0.6 g/kg fat) by oral gavage. Blood
samples were collected through retroꢀorbital bleeding at 0, 10, 30 and 60 minutes after meal challenge for
assessment of glucose excursion. In addition, blood samples collected 60 minutes after meal challenge
were used to measure circulating levels of total glucagonꢀlike peptide 1 (tGLPꢀ1). Mice were necropsied
at the end of the meal challenge (23 hours after the dose of compound or vehicle) and cecal contents were
collected, weighed and frozen at ꢀ20 °C for later analysis. For batched analysis, samples were thawed on
wet ice, and MilliQ water was added to each sample at the ratio (w/v) of 1:5. The samples were then
homogenized by beadꢀbeating using a Mini Beadbeater (Biospec Products) at high speed for one minute.
The homogenized samples were centrifuged at 4 °C at 3750 rpm (3200 g) for 25 minutes. The supernatant
was removed and the volume was recorded. A 25 µL aliquot of supernatant was assayed for glucose
concentration by Roche Cobas Integra 400 Clinical Chemistry Autoanalyzer. Total glucose content (mg)
was estimated by multiplying the glucose concentration by the volume of the centrifuged supernatant. The
remaining supernatant was thawed to room temperature for pH measurement, which was performed using
a Mettler Toledo FE20 pH Meter.
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
Analysis of Blood Samples. Whole blood glucose was measured using an AccuꢀChek Aviva
glucometer (Roche Diagnostics, Indianapolis, IN). Plasma GLPꢀ1 levels were measured using the
GlucagonꢀLike Peptideꢀ1 Total ELISA Kit (catalog #EZGLP1Tꢀ36K, Millipore, St. Charles, MO) exactly
as described previously.41
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 48
1
2
3
4
5
6
7
8
9
Pharmacokinetics and Data Analysis. Adult male C57 mice (25ꢀ35 g) were maintained on chow
diet with free access to water and were conscious throughout the study. For bolus intravenous
administration (IV), 21 and 1 were dissolved in 0.1% Tweenꢀ80 to form a clear solution and was then
administered at a dose of 1 mg/kg. For oral administration, 21 was provided by gavage at a dose of 10
mg/kg or 30 mg/kg, while 1 was provided by gavage at a dose of 30 mg/kg. Following intravenous
injection, serial blood samples were collected in EDTAꢀcontaining tubes through 6 hours, and following
oral administration, serial blood samples were collected through 24 hours. The plasma fraction was
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
o
o
immediately separated by centrifugation at 4 C, and then stored at ꢀ20 C until sample analysis.
Bioanalysis of plasma samples for quantitating plasma concentrations of compounds 21 and 1 was
conducted using liquid chromatography/mass spectroscopy. All plasma concentration versus time data
for 21 were analyzed using nonꢀcompartmental model 200 (for oral administration) and 201 (for IV
administration) of WinNonlin (version 5.0, Pharsight, Inc. Mountain View, CA) as described
previously.59 The halfꢀlife during the terminal phase was calculated from the elimination rate
constant (λ) determined by the linear regression analysis of the logꢀlinear part of the plasma
concentration curve. The AUC0ꢀt was calculated using linear up/log down trapezoidal method up
to the last measured concentration at time t. AUC0ꢀ∞ was calculated as AUC0ꢀt + Ct/λ. Clearance
(CL) was calculated by dose/AUC0ꢀ∞. Other pharmacokinetic parameters included plasma peak
concentration (Cmax), the time of Cmax (Tmax) and volume of distribution at steady state (Vss).
Associated Content
Supporting Information. Synthetic methods and characterization for compounds in Tables 1
and 2. The Supporting Information is available free of charge on the ACS Publications website.
Author Information
Corresponding Author.
ACS Paragon Plus Environment

Page 27 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
*(NCG) Current address: GlaxoSmithKline, 709 Swedeland Rd, King of Prussia, PA 19406. Eꢀ
mail: nicole.c.goodwin@gsk.com. Tel: +1 (610) 270ꢀ4260
7
8
9
Notes. All authors contributed equally to this work. D.R.P., S.G., and A.G.E.W are current
employees of Lexicon Pharmaceuticals and own stock. N.C.G., Z.M.D., B.A.H., A.L.H., M.S., A.Y.T.,
W.X., F.M., D.J.B, E.O., M.T., R.B.V., and D.B.R. were employees of Lexicon Pharmaceuticals at the
time of studies and own stock. E.D.S., D.D., L.L., E.O., and K.G.C. were employees of Lexicon
Pharmaceuticals at the time of the study and declare no competing interests. The authors wish to
acknowledge Kristi Boehm, MS, ELS for her help in figure preparation.
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
Abbreviations Used
SGLT, sodiumꢀdependent glucose cotransporter; GI, gastrointestinal; GGM, glucoseꢀgalactose
malabsorption; KO, knockꢀout; FDA, Food & Drug Administration; UGE, urinary glucose excretion;
PPG, postprandial glucose; GLPꢀ1, glucagonꢀlike peptide 1; tGLPꢀ1, total GLPꢀ1; SAR, structureꢀactivity
relationship; AMG, alphaꢀmethylglucopyranoside; IC50, half maximal (50%) inhibitory concentration;
mSGLT, mouse SGLT; hSGLT, human SGLT; DIO, dietꢀinduced obese; OGTT, oral glucose tolerance
test; NMR, nuclear magnetic resonance; HCl, hydrochloric acid; PK, pharmacokinetic; BF3·OEt2, boron
trifluoride diethyl etherate; TMSOTf, trimethylsilyl trifluoromethanesulfonate; DIPEA, N,Nꢀ
diisopropylethylamine; HATU, 1ꢀ[bis(dimethylamino)methylene]ꢀ1Hꢀ1,2,3ꢀtriazolo[4,5ꢀb]pyridinium 3ꢀ
oxid hexafluorophosphate
References
(1) Diabetes Fact Sheet, www.who.int/mediacentre/factsheets/fs312/en; accessed March 6, 2016.
(2) Fowler, M. J. Microvascular and Macrovascular Complications of Diabetes. Clin. Diabetes
2008, 26, 77ꢀ82.
(3) Morrish, N .J.; Wang, S. L.; Stevens, L. K.; Fuller, J. H.; Keen, H. Mortality and Causes of
Death in the WHO Multinational Study of Vascular Disease in Diabetes. Diabetologia 2001, 44, S14ꢀ
S21.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 48
1
2
3
4
5
6
7
8
(4) Tancredi, M.; Rosengren, A.; Svensson, A. M.; Kosiborod, M.; Pivodic, A.; Gudbjornsdottir, S.;
Wedel, H.; Clements, M.; Dahlqvist, S.; Lind, M. Excess Mortality Among Persons with Type 2
Diabetes. N. Eng. J. Med. 2015, 373, 1720ꢀ1732.
9
(5) Mudaliar, S.; Polidori, D.; Zambrowicz, B.; Henry, R. R. SodiumꢀGlucose Cotransporter
Inhibitors: Effects on Renal and Intestinal Glucose Transport: From Bench to Bedside. Diabetes Care
2015, 38, 2344ꢀ2353.
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
(6) Marsenic, O. Glucose Control by the Kidney: an Emerging Target in Diabetes. Am. J. Kidney
Dis. 2009, 53, 875ꢀ883.
(7) Wright, E. M. Renal Na+ꢀGlucose Cotransporters. Am. J. Physiol. Renal. Physiol. 2001, 280,
F10ꢀF18.
(8) Nishimura, M.; Naito, S. Tissueꢀspecific mRNA Expression Profiles of Human ATPꢀbinding
Cassette and Solute Carrier Transporter Superfamilies. Drug Metab. Pharmacokinet. 2005, 20, 452ꢀ477.
(9) Turk, E.; Zwbel, B.; Mundlos, S.; Dyer, J.; Wright, E. M. Glucose/galactose Malabsorption
Caused by a Defect in the Na+/glucose Cotransporter. Nature 1991, 350, 354ꢀ356.
(10)
Martin, M. G.; Turk, E.; Lostao, M. P.; Kerner, C.; Wright, W. M. Defects in
Na+/glucose Cotransporter (SGLT1) Trafficking and Function Cause Glucoseꢀgalactose Malabsorption.
Nat. Genet. 1996, 12, 216ꢀ220.
(11)
Kanai, Y.; Lee, W. S.; You, G.; Brown, D.; Hediger, M. A. The Human Kidney Low
Affinity Na+/glucose Cotransporter SGLT2: Delineation of the Major Renal Reabsorptive Mechanism for
Dꢀglucose. J. Clin. Invest. 1994, 93, 397ꢀ404.
(12)
Van den Heuvel, L. P.; Assink, K.; Willemsen, M.; Monnens, L. Autosomal Recessive
Renal Glucosuria Attributable to a Mutation in the Sodium Glucose Cotransporter (SGLT2). Hum.
Genet. 2002, 111, 544ꢀ547.
(13)
Santer, R.; Calado, J. Familial Renal Glucosuria and SGLT2: From a Mendelian Trait to
a Therapeutic Target. Clin. J. Am. Soc. Nephrol. 2010, 5, 133ꢀ141.
ACS Paragon Plus Environment

Page 29 of 48
Journal of Medicinal Chemistry
1
2
3
4
(14)
Lam, J. T.; Martin, M. G.; Turk, E.; Hirayama, B.A.; Bosshard, N. U.; Steinmann, B,;
5
6
Wright, E. M. Missense Mutations in SGLT1 Cause Glucoseꢀgalactose Malabsorpion by Trafficking
7
8
Defects. Biochim. Biophys. Acta 1999, 1453, 297ꢀ303.
9
(15)
Gorboulev, V.; Schurmann, A,; Vallon, V.; Kipp, H.; Jaschke, A.; Klessen, D.; Friedrich,
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
A.; Scherneck, S.; Rieg, T.; Cunard, R.; VeyhlꢀWichmann, M.; Srinivasan, A.; Balen, D.; Breljak, D.;
Rexhepal, R.; Parker, H. E.; Gribble, F. M.; Reiman, F.; Lang, F.; Wiese, S.; Sabolic, I.; Sendtner, M.;
Koepsell, H. Na(+)ꢀDꢀglucose Cotransporter SGLT1 is Pivotal for Intestinal Glucose Absorption and
Glucoseꢀdependent Incretin Secretion. Diabetes 2012, 61, 187ꢀ196.
(16)
Wright, E. M.; Turk, E.; Martin, M. G. Molecular Basis for Glucoseꢀgalactose
Malabsorption. Cell Biochem. Biophys. 2002, 36, 115ꢀ121.
(17)
Xin, B.; Wang, H. Multiple Sequence Variations in SLC5A1G are Associated with
Glucoseꢀgalactose Malabsorption in Large Cohort of Old Order Amish. Clin. Genet. 2011, 79, 86ꢀ91.
(18)
Mechanism for the Pharmacotherapy of Diabetes Mellitus Type 2. J. Med. Chem. 2009, 52, 1785ꢀ1794.
(19) Washburn, W. N. Evolution of Sodium Glucose CoꢀTransporter 2 Inhibitors as Antiꢀ
Diabetic Agents. Expert Opin. Ther. Pat. 2009, 19, 1485ꢀ1499.
Washburn, W. N. Development of the Renal Glucose Reabsorption Inhibitors: a New
(20)
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. Discovery of Dapagliflozin: a Potent, Selective Renal Sodiumꢀdependent Glucose
Cotransporter 2 (SGLT2) Inhibitor for the Treatment of Type 2 Diabetes. J. Med. Chem. 2008, 51, 1145–
1149.
(21)
Deshpande, P. P.; Ellsworth, B. A.; Singh, J.; Denzel, T. W.; Lai, C.; Chiajen, G.;
Randazzo, M. E.; Gougoutas, J. Z. Methods of Producing CꢀAryl Glucoside SGLT2 Inhibitors. U.S. Pat.
Appl. Pub. US2004138439, 2004.
ACS Paragon Plus Environment