Discovery and Optimization of Glucose Uptake Inhibitors

Article abstract of DOI:10.1021/acs.jmedchem.9b02153

Aerobic glycolysis, originally identified by Warburg as a hallmark of cancer, has recently been implicated in immune cell activation and growth. Glucose, the starting material for glycolysis, is transported through the cellular membrane by a family of glucose transporters (GLUTs). Therefore, targeting glucose transporters to regulate aerobic glycolysis is an attractive approach to identify potential therapeutic agents for cancers and autoimmune diseases. Herein, we describe the discovery and optimization of a class of potent, orally bioavailable inhibitors of glucose transporters, targeting both GLUT1 and GLUT3.

Full text of DOI:10.1021/acs.jmedchem.9b02153

Subscriber access provided by BIU Pharmacie | Faculté de Pharmacie, Université Paris V
Article
Discovery and Optimization of Glucose Uptake Inhibitors
Kevin G. Liu, Ji-In Kim, Kellen Olszewski, Anthony M Barsotti, Koi Morris, Christophe
Lamarque, Xuemei Yu, Jack Gaffney, Xiao-Jiang Feng, Jeegar Patel, and Masha Poyurovsky
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b02153 • Publication Date (Web): 13 Apr 2020
Downloaded from pubs.acs.org on April 16, 2020
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 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 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.
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 14
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery and Optimization of Glucose Uptake Inhibitors
Kevin G. Liu,* Ji-In Kim, Kellen Olszewski, Anthony M. Barsotti†, Koi Morris, Christophe
Lamarque, Xuemei Yu, Jack Gaffney, Xiao-Jiang Feng, Jeegar P. Patel, and Masha V. Poyurovsky
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
†
Kadmon Corporation, LLC., 450 East 29th Street, New York, NY 10016, USA. Current address: Inzen Therapeutics,
Inc., 430 East 29th Street, New York, NY 10016, USA
GLUT, GLUT1, GLUT3, glucose transport inhibitor, glucose transport, immunometabolism, inflammation
ABSTRACT: Aerobic glycolysis, originally identified by Warburg as a hallmark of cancer, has recently been implicated in
immune cell activation and growth. Glucose, the starting material for glycolysis, is transported through the cellular
membrane by a family of glucose transporters (GLUTs). Therefore, targeting glucose transporters to regulate aerobic
glycolysis is an attractive approach to identify potential therapeutic agents for cancers and autoimmune diseases. Herein,
we describe the discovery and optimization of a class of potent, orally available inhibitors of glucose transporters, targeting
both GLUT1 and GLUT3.
NH
N
HN
N
N
O
O
N
N
H
15b
GLUT1 IC50 = 242 nM
GLUT3 IC50 = 179 nM
Oral F = 45.5% (mouse) and 49.4% (rat)
INTRODUCTION
growth and present an interesting targets for potential
therapeutic intervention4-6
.
Glucose is the most abundant energy source for cells in
human tissues. Under aerobic conditions, cells normally
metabolize glucose into pyruvate in the cytosol, which is
oxidized in the tricarboxylic acid (TCA) cycle to drive
oxidative phosphorylation (OXPHOS) in the
Elevated rates of glucose consumption and glycolysis
were recently implicated in immune cell activation and
growth. Immunometabolism,7, 8 an emerging field
studying the interplay between the immune system and
metabolism, has attracted considerable attention both for
understanding the phenomenon and searching for new
treatments for autoimmune diseases. Upon activation,
naïve CD4+ T cells rapidly grow, proliferate, and
differentiate into functional subsets including T helper
(Th) cells such as effector Th1, Th2, and Th17 cells which
drive the immune response, and Foxp3+ regulatory T
(Treg) cells which suppress the immune activity. Unlike
their precursor naïve CD4+ T cells which rely
mitochondria. When oxygen is lacking, a condition called
hypoxia, cells are forced to switch to anaerobic glycolysis
where the aerobic glycolysis product pyruvate is instead
converted to lactate. Glycolysis is significantly less
efficient than the OXPHOS pathway, producing an order
of magnitude less ATP per glucose. Despite this, both
cultured cancer cells and tumors typically ferment
glucose to lactate at a high rate even in the presence of
abundant oxygen. This altered metabolism phenomenon,
known as the Warburg effect,1, 2 is a hallmark of cancer
cells. This phenomenon which has been exploited for
cancer imaging and diagnosis with the PET ligand
[18F]fluoro-2-deoxyglucose (18F-FDG), a metabolically
stable glucose analogue. There is still considerable
academic debate about the nature of the survival benefit
aerobic glycolysis phenomenon provides for cancer and
other rapidly proliferating cells. It is, however,
predominantly on OXPHOS, effector CD4+ T cells
primarily rely on aerobic glycolysis for their metabolic
needs and immune functions.9, 10
Glucose transporters (GLUTs) constitute a family of
proteins that facilitate the transport of glucose across
cellular membranes. To date, 14 human GLUT members
(GLUT1-14) have been identified.11 GLUTs have 12
transmembrane domains, and several crystal structures
including human GLUT1 and GLUT3 have been
undisputed that elevation in the rate of glycolysis and
increased glucose uptake,3 are associated with tumor
obtained.12, 13 Given that glucose transport is one of the
rate-limiting steps in glucose metabolism, GLUTs,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 14
especially GLUT1, are overexpressed in cancer and
compensatory mechanisms from GLUT3 following GLUT1
inhibition, we additionally aimed to generate compounds
targeting both isoforms.
1
2
3
4
5
6
7
8
activated immune cells to meet their glucose demands,
and have become an interesting therapeutic target for
cancers and autoimmune diseases.9, 10, 14-16
RESULTS AND DISCUSSION
A number of GLUT (mainly GLUT1) inhibitors have been
reported. Representative GLUT inhibitors are
Dihydrofuropyrimidine and
dihydropyranopyrimidine derivatives. To identify
novel glucose uptake inhibitors, we screened Kadmon’s
compound collection utilizing a high-throughput cell-
based phenotypic assay. In this assay, HT-1080 colorectal
cancer cells were treated with 10 µM oligomycin to inhibit
ATP synthase, blocking mitochondria-derived ATP
production through oxidative phosphorylation and
rendering the cells dependent on glycolytic ATP
production. Glycolytically-derived ATP production was
measured with the Cell Titer-GloTM assay kit (Promega).
HT-1080 cells express both GLUT1 and GLUT3
transporters, so the inhibitory activity of our compounds
in these cells is referred to broadly as “GLUT” inhibition.
Because simply inhibiting ATP production cannot be
linked to inhibition of glucose uptake directly, we
confirmed inhibition of glucose transport, but not other
targets such as downstream glycolysis enzymes, by also
measuring direct glucose uptake, lactate secretion,
glutamine consumption and metabolomic profiling of
treated cells. The results from the above-mentioned
experiments supported the direct blockade of glucose
transport mechanism and were further reinforced by the
in vivo glucose tolerance and imaging experiments
(Olszewski, et al., manuscript in preparation). GLUT
subtype activity was measured using cell lines that
specifically express the GLUT isoforms, e.g., GLUT1
(DLD1); GLUT3 (DLD1-SLC2A1-/-). We initially utilized the
GLUT assay in HT1080 cells to identify hits and lead
chemical series. As the project progressed, we relied
primarily on more informative GLUT1 and GLUT3
subtype assays for lead optimization.
summarized in Chart 1. Compounds 1-3 17-19 have limited
potency (in the µM to mM range) and have been reported
to have additional activity independent of GLUT
inhibition. Compound 4 of peptidic nature and its
analogue have been reported as GLUT inhibitors and co-
crystallized with hGLUT1.20 Compounds 521 and 6 (BAY-
876)22, reported by Siebeneicher and co-workers at Bayer,
represent two classes of GLUT1 sub-type selective
inhibitors. Compound 6 was reported to be a potent and
orally bioavailable GLUT1 selective inhibitor. Very
recently, Waldmann et al., disclosed a potent GLUT1/2/3
inhibitor 7 (Glutor) and its ability to suppress growth of a
number of cancer cell lines. However, no
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
pharmacokinetic or in vivo functional data associated
with this compound was disclosed.23
Chart 1. Known GLUT inhibitors
O
CF3
OH
CF3
O
F
O
O
F3C
HO
O
N
S
1
O
2
WZB117
OH
N
O
O
N
N
N
H
H
N
S
3
O
O
STF-31
N
N
O
H
N
N
N
N
H
F
NH
N
NH
N
O
5
O
N
I
4
OH
F
GLUT-i1
HN
HN
N
EtO
EtO
N
O
N
O
CN
O
O
N
N
N
N
N
H
H
O
F3C
O
III
8
II
NH
N
SAR and optimization
GLUT IC50 = 150 nM
O
N
N
Figure 1. The screening hit and optimization strategy
Our screening efforts led to the identification of
compound 8 (Figure 1) as a potent GLUT inhibitor (IC50 =
HN
NH2
O
F
N
7
Glutor
N
6
BAY-876
150 nM). To elucidate the molecular mechanism of GLUT
inhibition we compared the activity of compound 8 in a
GLUT1 inhibition assay to cytochalasin B at varying
glucose concentrations. Cytochalasin B has been
extensively characterized as a glucose competitive
inhibitor of GLUT1, which functions by binding in the
glucose pocket in the central cavity of the transporter.20, 24
We observed that both cytochalasin B and compound 8
exhibited a similar level of competition with glucose, with
the GLUT1 IC50 for compound 8 shifting from 268 nM at
10 mM glucose to 33 nM in the presence of 0.37 mM
glucose (Figure 2 and Supplementary Table S1). These
As part of our research efforts in metabolomics, we were
interested in studying the inhibition of glucose uptake
and its potential therapeutic indications. While a number
of above-mentioned compounds have been tested in cell-
based assays, little or no in vivo functional data is
available. This is due in part to their poor
pharmacokinetic properties which limit the ability to fully
interrogate the potential therapeutic utility of GLUT
inhibitors. In order to assess the clinical potential of
GLUT targeting, we set out to generate GLUT inhibitors
with optimized safety and PK profiles. To avoid possible
2
ACS Paragon Plus Environment

Page 3 of 14
Journal of Medicinal Chemistry
data suggested that the molecular mechanism of GLUT activity. Although replacement of the quinazoline
1
2
3
4
5
6
7
8
inhibition of glucose import by our compounds was
through direct competition for the glucose binding
pocket of GLUT1.
core with pyrimidine or substituted pyrimidines resulted
in dramatic loss of activity, fusing the pyrimidine ring
with a variety of heterocycles provided alternatives to the
quinazoline ring which retained potency. Some of the
explored heterocycle-fused pyrimidine cores are
summarized in Table 1.
4
10 mM glucose
3
3.33 mM glucose
2
Table 1. Select heterocycle-fused pyrimidine coresa
1.11 mM glucose
1
NH
N
0.37 mM glucose
0.4
9
0.3
0.2
0.1
0.0
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
HN
O
Core
cytochalasin B
compound 8
O
R
N
H
Figure 2. Compound 8 is competitive with glucose. GLUT1
IC50 values for cytochalasin B and compound 8 at
different glucose concentrations. Graph shows
9
GLUT
IC50
(nM)
GLUT
IC50
(nM)
Entr
y
Entry Core
Core
calculated IC50 values +/- the 95% confidence interval.
While being a potent hit as a starting point, this
compound was characterized by suboptimal
*
*
EtO
S
N
N
8
150
9a
9c
200
74
N
*
physicochemical properties, such as poor aqueous
solubility (3.36 µM, 0.01 µM, 0.03 µM at pH 2.0, pH 4.5,
and pH 6.8, respectively), high protein binding (>99.9%),
and poor permeability (Caco-2 Papp 2.0 × 10-6 cm/s). To
follow up on compound 8, we studied the SAR (structure-
activity relationship) and carried out optimization in
three regions of the molecule: phenylpyrazole region I,
phenoxylacetamide II, and quinazoline core III (Figure 2).
For region I, we replaced phenylpyrazole with a variety of
groups such as small monocyclic rings, fused rings, and
those that are structurally similar to phenylpyrazole.
Disappointingly, all the replacements resulted in either a
complete loss or a dramatic reduction of GLUT activity.
Phenylpyrazole was strongly preferred in this region. It is
worth mentioning that the high preference for
N
*
*
*
O
N
N
9b
259
O
N
*
*
*
N
*
*
N
N
N
9d
328
3474
38
9e
9g
9i
146
36
N
N
S
N
*
*
*
N
N
N
N
N
9f
N
*
*
N
*
*
N
N
9h
350
N
N
N
*
N
*
*
*
N
N
N
*
N
9j
1980
282
9k
298
110
N
S
N
*
N
*
phenylpyrazole in this region was maintained with
concomitantchange to other portions of the molecules.
O
*
N
N
N
N
9l
9m
N
*
N
*
Aiming to improve the physicochemical properties and to
enhance drug-like properties, we undertook further
optimization of region II. We first removed the phenyl
ring in order to reduce the total number of aromatic
rings25 or replace the phenyl ring with a saturated ring
system in order to reduce sp2 and to increase sp3
hybridization characteristics of the molecule aiming to
improve the physico-chemical property26 as well as to
potentially improve the potency by incorporating
moieties similar to the glucose ring.26 Unfortunately,
these efforts were not successful as the phenyl group was
shown to be required for GLUT activity. Replacement of
the phenyl ring by heterocycles also led to inactivity. We
then kept the phenyl group, but replaced the amide
moiety with non-amide groups such as acyclic amine or
ether, lactams, heterocycles or heteroaryls with the hope
that these groups can avoid the potential metabolic
liability of the amide group. Once again, these
aIC50 values are average of at least two determinations.
While most of the compounds with heterocyle-fused
pyrimidine cores (9a-m) are expected to have improved
physical properties over quinazoline derivative 8 by
simply having reduced cLogP with the introduction of
additional heteroatoms, compounds 9c and 9l were more
interesting to us due to their increased sp3 hybridization
feature contributed by the dihydrofurano- and
tetrahydropyrido- groups, respectively. The increased sp3
content was expected to translate into less molecular
flatness and enhanced water solubility.
Select SAR of dihydrofuropyrimidine and
dihydropyranopyrimidine derivatives is summarized
Table 2. The data clearly shows that the amide N-H is
required for both GLUT1 and GLUT3 activity as N,N-
disubstituted amides (10e and 10f) are either significantly
less active or inactive. The most potent
dihydrofuropyrimidine derivative identified, 10g,
displayed an IC50 of 37 nM and 12 nM against GLUT1 and
GLUT3, respectively. BAY-876 (6) was also tested in our
replacements resulted in significantly reduced potency (>
5 µM).
Upon further optimization, we found that the quinazoline
core (region III) tolerated modifications without loss of
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 14
assays and was shown to be a potent GLUT1 selective
modest volume of distribution (1.42 and 0.63 L/kg for 10g
inhibitor as reported.22
and 11a, respectively). The lower volume of distribution of
11a may have been due to its higher plasma protein
binding that helped confine the drug in the plasma.
Table 3. ADME and PK profile of 10g and 11aa
1
2
3
4
5
6
7
8
The SAR of 6-membered ring dihydropyranopyrimidine
derivatives in general was consistent to that of their 5-
membered ring dihydrofuranopyrimidine analogues. Two
of the most potent compounds identified are 11a and 11b
(Table 2).
Compound
10g
11a
Solubility @ pH
2.0/4.5/6.8 (µM)b
5.9/0.08/0.03 32/0.75/0.19
Table 2. SAR of dihydrofuranyl and dihydropyranyl
derivativesa
mLM stability (t1/2, min)c 29
29
9
NH
N
NH
N
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
mPPB (%)d
Mouse plasma stabilitye
97.0
99.7
>95%
1870
>95%
1233
HN
N
HN
N
Oral Cmax
(ng/mL)
O
N
O
N
O
O
O
R2
O
R2
N
N
Oral AUC
(ng/mL)
3299
34.0
5982
23.2
R1
R1
11
10
GLUT1 GLUT3
CL
Mouse
Entry
R1
R2
IC50
IC50
(mL/min/kg)
Vdss (L/kg)
t1/2 (h)
PKf
(nM)
(nM)
1.42
2.94
22
0.63
0.92
28
10a
10b
9c
H
H
H
H
H
Et
nPr
cPr
iPr
447
666
289
179
298
238
97
F (%)
B/Pg
0.15
0.10
10c
10d
10e
10f
10g
10h
10i
11a
cBu
cPe
Et
68
aThe in vitro ADME data are average values of at least two
determinations; The in vivo PK parameters are average
values from three animals. bSolubility was determined in
PBS buffer solutions. cMouse liver microsomal t1/2: 2 µM
substrate, 0.5 mg/mL pooled liver microsomes, 1 mM
NAPDH, 37 °C. dMouse plasma protein binding.
ePercentage of compound remaining after 18 h. fiv dose 10
mg/kg, vehicle 10% DMSO/40% PEG400/50% water; po
dose 30 mg/kg, vehicle 0.5% methylcellulose/0.25%
Tween-80 in water. gCompound brain and plasma
exposure ratio measured at 2 h following oral
102
47
1828
686
-(CH2)4-
>10000 1414
H
H
H
H
H
tPe
37
12
CF3CH2
Ph
565
885
48
160
325
33
3-Pe
tBu
11b
36
26
administration.
4809
6
---
---
50 (2b)
(1670b)
Unlike other organs, brain is uniquely dependent on
glucose for survival and function. To avoid potential
toxicities associated with the blockade of neuronal
glucose uptake we aimed to keep compound brain
exposure levels to a minimum. Similar to Lipinski’s
landmark rule of 5 (Ro5) for drug oral bioavailability,27
desirable physicochemical properties for brain
penetration have been extensively studied using
successful CNS drugs and drug candidates.28-31 The GLUT
inhibitors like 10g and 11a fall well outside of the CNS
preferred range as they are characterized by high
molecular weight (~500 Da), high tPSA (>100 Å2), and
high number of hydrogen bond donors (HBDs, 3) and
therefore were not expected to exhibit significant brain
penetration. As predicted, 10g and 11a were found to be
only modestly brain penetrant with measured in vivo
brain and plasma ratios of 0.15 and 0.10, respectively.
aIC50 values are average of at least two determinations.
bReported data in reference 22
.
The in vitro ADME and in vivo PK profiles of 10g and 11a
are summarized in Table 3. Compound 10g was
characterized by very poor water solubility at pH 6.8 and
pH 4.5 while it was more soluble at lower pH (2.0) due to
its basicity. Compound 10g was also characterized by a
relatively low mouse liver microsomal stability (29 min),
modestly high protein binding (fu = 3%), and good plasma
stability. The 6-membered ring dihydropyranopyrimidine
derivative 11a was more soluble, displaying solubility of 32
µM, 0.75 µM, and 0.19 µM at pH 2.0, pH 4.5, pH 6.8,
respectively. The increased solubility of 11a is presumably
due to the higher degree of flexibility of dihydropyranyl
and 3-pentyl groups present in 11a as compared to their
corresponding groups in 10g (Table 3).
While being potent and orally bioavailable, 10g and 11a
still possessed some undesirable physical properties such
as poor water solubility. We therefore focused our efforts
on other classes of compounds with improved
physicochemical properties.
In the in vivo mouse PK studies, both 10g and 11a were
orally bioavailable. The relatively low oral bioavailability
(22% and 28% for 10g and 11a, respectively) may be due in
part to their relatively high intrinsic clearance and poor
water solubility. Both compounds are characterized by a
4
ACS Paragon Plus Environment

Page 5 of 14
Journal of Medicinal Chemistry
Tetrahydropyridopyrimidine derivatives. To identify dihydropyranopyrimidine derivatives (e.g. 10g and 11a).
1
2
3
4
5
6
7
8
compounds with improved aqueous solubility, we turned
our attention to the tetrahydropyridopyrimidine
derivatives represented by 9l (Table 1). The pKa value of
the piperidinyl nitrogen of 9l is expected to be ~10. Under
physiological pH (7.4), it should be mostly protonated,
increasing the overall compound aqueous solubility.
This compound was also characterized by overall
favorable ADME properties including good liver
microsomal stability in mouse (T1/2 76 min) and rat (T1/2
64 min), low protein binding (83%, mouse), good plasma
stability, and good permeability (Papp A-B 7.1 × 10-6 cm/s).
In the mouse pharmacokinetic studies, 12p displayed high
oral bioavailability (61%), good plasma exposure levels
(AUC 7745 ng/mL @ 30 mg/kg, po) and low brain
penetration (B/P = 0.07). However, in rat, 12p displayed a
less favorable pharmacokinetic profile, which was
characterized by low plasma exposure level (Cmax 399
ng/mL) and low oral bioavailability (20%). Relatively low
rat plasma exposure for 12p necessitated further
optimization in order to identify compounds with
improved PK across multiple species.
The SAR of select tetrahydropyridopyrimidine derivatives
is summarized in Table 4. Several substituents including
H, Me, Et and iPr on the piperidinyl nitrogen (R1) and a
number of groups on the amide nitrogen were explored.
In general, this class of compounds are dual GLUT1 and
GLUT3 inhibitors. Relatively bulkier amide groups (R2)
such as tPe and tBu (e.g. 12e, 12f, 12j, 12p) tended to be
associated with increased GLUT1 and GLUT3 activity and
retained glucose competitive MOA (supplementary table
S1), while relatively smaller groups such as iPr (9l, 12g,
12k, 12m) provided compounds with reduced GLUT1
potency. Among these analogues, 12p displayed an overall
more favorable pharmacological and pharmaceutical
profile. The ADME and pharmacokinetic data of 12p are
summarized in Table 5.
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
Table 5. ADME and PK profile of 12pa
Solubility @ pH
2.0/4.5/6.8 (µM)b
256/189/47
LM stability (t1/2, min)c
mPPB (%)d
76 (mouse), 64 (rat)
83
Mouse plasma stabilitye
Caco-2 Papp (10-6 cm/s)f
PKg
>95%
Table 4. SAR of tetrahydropyridopyrimidine derivativesa
NH
N
A-B 7.1 / B-A 12.4
Mouse
1418
7745
40.6
3.00
1.68
Rat
399
2267
44.3
4.44
1.86
20
Oral Cmax (ng/mL)
Oral AUC (ng/mL)
CL (mL/min/kg)
Vdss (L/kg)
HN
R1
N
N
O
O
R2
N
N
H
12
cpd
R1
R2
GLUT1 GLUT3
t1/2 (h)
IC50
IC50
F (%)
B/Ph
61
(nM)
1266
---
(nM)
0.07
---
9l
Me
Me
Me
Me
Me
Me
Me
H
iPr
867
846
566
391
584
368
391
365
138
121
aThe in vitro ADME data are average values of at least two
determinations; The in vivo PK parameters are average
values from three animals. bSolubility was determined in
PBS buffer solutions. cMouse and rat liver microsomal t1/2:
2 µM substrate, 0.5 mg/mL pooled liver microsomes, 1
mM NAPDH, 37 °C. dMouse plasma protein binding.
ePercentage of compound remaining after 18 h. fApparent
permeability coefficients (Papp) of test compounds at 5 µM
across Caco-2 cell monolayers: apical to basolateral (A-B)
and basolateral to apical (B-A). giv dose 10 mg/kg, vehicle
10% DMSO/40% PEG400/50% water; po dose 30 mg/kg,
vehicle 0.5% methylcellulose/0.25% Tween-80 in water.
hCompound brain and plasma exposure ratio measured at
2 h following oral administration.
12a
12b
12c
12d
12e
12f
iBu
cBu
cPe
3-Pe
tBu
tPe
iPr
763
559
789
359
342
1458
459
441
12g
12h
12i
H
cPe
3-Pe
tBu
iPr
H
12j
H
220
1578
552
58
12k
12l
Et
846
482
619
802
548
412
Incorporation of a basic group to a molecule can in
general help its water solubility, yet higher basicity may
decrease its permeability as the positively charged form of
the molecule resulted from protonation under
Et
tBu
iPr
12m
12n
12o
12p
iPr
iPr
iPr
iPr
2711
1890
1112
iBu
cPe
tBu
physiological conditions tends to interact with negatively
charged lipid membranes, thereby hindering its
permeability.32 The charged form also has reduced
lipophilicity and thus limits its passive diffusion ability
across the gut wall. We therefore tried to modulate the
basicity of the piperidinyl nitrogen with the hope that this
326
aIC50 values are average of at least two determinations.
As expected, 12p had dramatically increased water
solubility (256 µM, 189 µM, and 47 µM at pH 2.0, 4.5, and
6.8, respectively) over dihydrofuropyrimidine or
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 14
mPPB (%)d
Mouse plasma stabilitye
Rat PKf
92
kind of fine-tuning can help enhance oral bioavailability
while maintaining the favorable physical properties.
1
2
3
4
5
6
7
8
>95%
Compound 13a, a fluorinated analogue of 12p, was then
synthesized (Table 6). Fluorine has been extensively used
in drug design to influence compound pKa,
Oral Cmax (ng/mL)
Oral AUC (ng/mL)
CL (mL/min/kg)
Vdss (L/kg)
1875
6720
23.5
2.01
1.95
31
conformation, potency, and membrane permeability.33, 34
A fluorine atom at the beta-position of an amine can
reduce the pKa value by ~1.5 units. We postulated that the
reduced basicity would translate into rat PK
9
t1/2 (h)
improvement. Indeed, as we expected, 13a did display
improved rat PK parameters (Table 7) including plasma
exposure levels (Cmax 1875 ng/mL), clearance (23.5
mL/min/kg), and oral bioavailability (31%) over 12p
(Cmax 399 ng/mL, clearance 44.3 mL/min/kg, oral F
20%). Minimal food effect was observed in dog following
oral administration at 10 mg/kg with 31% and 43% oral
bioavailability, when fed and fasted, respectively.
Compound 13a showed robust linear rat PK profile up to
300 mg/kg dose. Unfortunately, there were also signs of
toxicity at high doses in both mouse and rat. While the
cause of observed toxicity was not fully elucidated, brain
penetration of 13a was found to be 0.15 (brain/plasma
ratio). This level of brain penetration was not optimal for
this target, and may have at least partially contributed to
the observed toxicity.
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
F (%)
aThe in vitro ADME data are average values of at least two
determinations; The in vivo PK parameters are average
values from three animals. bSolubility was determined in
PBS buffer solutions. cMouse and rat liver microsomal t1/2:
2 µM substrate, 0.5 mg/mL pooled liver microsomes, 1
mM NAPDH, 37 °C. dMouse plasma protein binding.
ePercentage of compound remaining after 18 h. fiv dose 10
mg/kg, vehicle 10% DMSO/40% PEG400/50% water; po
dose 30 mg/kg, vehicle 0.5% methylcellulose/0.25%
Tween-80 in water.
Both passive diffusion and transporter-mediated
processes are major mechanisms for brain permeability.35
The brain penetration of the GLUT inhibitors reported
here is likely mainly driven by passive diffusion due to
their relatively high lipophilicity and lack of active
Table 6. Additional tetrahydropyridopyrimidine
derivatives
transporter-recognized moieties. Several approaches were
explored to lower brain penetration, such as
NH
F
*
*
*
*
N
incorporating a carboxylic acid group or significantly
increasing the hydrophilicity of the molecule with the
hope that the carboxylate (formed in vivo) or the
increased molecular polarity would reduce their ability to
get into the brain through the passive diffusion
mechanism. However, in line with the above described
SAR exploration, compounds generated through these
changes either totally lost GLUT activity or were not
orally bioavailable.
13
14
15
16
R3
=
HN
N
O
O
O
*
*
*
R3
N
N
O
17
18
19
O
R4
N
H
R4
=
*
*
*
a
b
c
Cpd
GLUT1 GLUT3 Caco-2
B/Pb
0.15
IC50
(nM)
ERa
IC50
(nM)
Xenobiotics are either blocked from entering the brain by
tight junctions between brain vascular endothelial cells
(BVECs) of the BBB or are rapidly pumped out from the
BVECs by ATP-dependent transporter proteins such as P-
glycoprotein (P-gp) and Breast Cancer Resistance Protein
(BCRP). Compounds designed to act as substrates for
these efflux transporters in order to lower brain
13a
14b
14c
15a
15b
15c
16c
17a
18a
19a
105
72
232
211
7.33/6.23
11.97/4.52 0.10
140
259
242
914
1048
161
548
287
179
238
392
155
175
275
10.52/5.60 0.062
4.58/6.09 0.10
11.19/4.33
0.050
penetration have been reported.36 We tried to adapt this
approach by designing and identifying compounds with
efflux property as measured by in vitro caco-2 assay. Caco-
2 cells express efflux transporters and have been widely
used as a model of the intestinal or brain barrier.37
13.20/3.97 0.046
7.11/4.83
4.25/3.53
0.15
0.073
143
128
12.26/3.01 0.062
7.20/2.94 0.064
We then designed and synthesized a number of
compounds aiming to marginally increase polarity of the
molecule to facilitate efflux and to slightly reduce the
basicity of piperidinyl nitrogen to help rat PK (Table 6).
For the amide region, smaller and less hydrophobic
groups including iPr and 1-methylcyclopropyl were
introduced. For the piperidinyl group, slightly electron-
withdrawing groups were chosen to reduce the basicity of
the nitrogen either through hyperconjugation (e.g.
cyclopropyl group, 14) or through inductive effect (e.g.
aEfflux ratio calculated by Papp(B-A)/Papp(A-B). bCompound
brain and plasma exposure ratio at 1 h following oral
administration at 100 mg/kg in CD-1 mice.
Table 7. ADME and PK profile of 13aa
Solubility @ pH
2.0/4.5/6.8 (µM)b
305/212/1.1
LM stability (t1/2, min)c
100 (mouse), 60 (rat)
6
ACS Paragon Plus Environment

Page 7 of 14
Journal of Medicinal Chemistry
methoxyethyl, 18). With the exception of 15c and 16c, no appreciable activity when tested at 10 μM in a safety
1
2
3
4
5
6
7
8
these compounds maintained good GLUT1 and GLUT3
activity and when tested, 14c and 15b maintained glucose
competitive MOA, suggesting that the binding site was
retained by all the compounds across our SAR efforts
(Supplementary Table S1). The efflux ratios of these
compounds were in the range of approximately 1 to 4. As
a result, compounds with reduced brain penetration were
identified. For example, 14c, 15b, 15c, 18a, and 19a all
displayed a brain/plasma ratio below 0.07. The B/P ratios
were plotted against Caco-2 efflux ratio values (Figure 3)
in order to see their relationship. These data indicated
that a small degree of correlation between brain
penetration and Caco-2 efflux ratio exists for the
compounds examined.
screening panel that included 31 receptors, ion channels,
and transporters.
Table 8. ADME and PK profile of 15ba
Solubility @ pH
2.0/4.5/6.8 (µM)b
291/256/1.4
LM stability (t1/2, min)c
PPB (%)d
70 (m), 63 (r), 58 (d), 116 (h)
79.8 (m), 98.9 (r), 99.0 (d),
99.3 (h)
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Mouse plasma stabilitye
PKf
>95%
Mouse
2525
5890
40
Rat
1675
6813
37
Oral Cmax (ng/mL)
Oral AUC (ng/mL)
CL (mL/min/kg)
Vdss (L/kg)
0.2
1.70
4.51
2.59
49.4
---
0.15
B/P
t1/2 (h)
0.785
45.4
0.05
0.1
F (%)
B/Pb
0.05
0
aThe in vitro ADME data are average values of at least two
determinations; The in vivo PK parameters are average
values from three animals. bSolubility was determined in
PBS buffer solutions. cMouse, rat, dog, and human liver
microsomal t1/2: 2 µM substrate, 0.5 mg/mL pooled liver
microsomes, 1 mM NAPDH, 37 °C. dMouse, rat, dog, and
human plasma protein binding. ePercentage of compound
remaining after 18 h. fiv dose 10 mg/kg, vehicle 10%
DMSO/40% PEG400/50% water; po dose 30 mg/kg,
vehicle 0.5% methylcellulose/0.25% Tween-80 in water.
0.00
1.00
2.00
3.00
4.00
5.00
Caco-2 ER
Figure 3. Plot of brain and plasma exposure ratio (B/P) at
1 h following oral administration at 100 mg/kg in CD-1
mice versus Caco-2 efflux ratio.
Most of the compounds in Table 6 displayed high plasma
exposure levels in mice following oral administration
(data not shown). A single 100 mg/kg dose of compounds
14c, 15b and 17a was well tolerated without obvious
adverse effects. Compound 15b was chosen to be further
profiled due to its overall favorable properties. The profile
of 15b is summarized in Table 8. While being very soluble
at lower pH (> 200 µM at pH 2.0 and pH 4.5), 15b
displayed much lower solubility at pH 6.8 (1.4 µM). The
compound is characterized by modest to good liver
microsomal and plasma stability across all species. In PK
studies, 15b showed good oral bioavailability in both
mouse (45.4%) and rat (49.4%) at 30 mg/kg dose. Linear
PK was observed in rats up to 300 mg/kg without obvious
adverse effects. In dog PK study at 10 mg/kg orally, 15b
displayed a modest food effect with a 19% (fed) and a 29%
(fasted) oral bioavailability. It is worth noting that the PK
studies were typically done with methylcellulose/tween-
80 standard formulation for the compounds. A number of
other commonly used oral formulations, e.g.
When dosed orally, a number of compounds including 8,
10g, and 15b showed in vivo efficacy in tumor xenograft
models as well as models of autoimmune conditions at
doses devoid of overt toxicity (data not shown). These
data further support the notion that pharmacological
blockade of glucose transport in vivo may have
therapeutic benefits across multiple disease areas, and
will be presented elsewhere.
CHEMISTRY
The synthesis of boronic esters 23 and THP protected 4-
aminophenylpyrazole 26 is depicted in Scheme 1. These
are necessary intermediates for the synthesis of the final
compounds. The chloroacetamides 21 were easily
synthesized from amines 20 and 2-chloroacetyl chloride.
Alkylation of 3-bromophenol with 21 followed by Pd-
mediated coupling with bis(pinacolato)diboron afforded
boronic esters 23. Alternatively, boronic esters 23 were
prepared by direct alkylation of 3-(4,4,5,5-tetramethyl-
1,3,2-dioxaborolan-2-yl)phenol with 21. THP protected 4-
aminophenylpyrazole 26 was synthesized in two steps
from readily available 24.
cyclodextrins, were explored and it was found that 2-
hydroxypropyl-β-cyclodextrin significantly enhanced oral
bioavailability and plasma exposure levels of 15b.
However, with this PK-enhanced formulation, a mild and
transient toxicity was observed following oral
The synthesis of 8, 9a-f, and 9h-k (Scheme 2) was
accomplished by starting with fused 1,4-
administration of 15b at 100 mg/kg dose in mice.
Safety profiling of 15b showed low potential for QT
prolongation with an estimated IC50 greater than 10 μM in
the manual patch clamp hERG channel assay. It also had
dichloropyrimidines 27 (with the fused ring designated as
W) that are either commercially or synthetically available.
Nucleophilic substitution with 4-(1H-pyrazol-4-yl)
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
aniline38 exclusively or predominantly occurred at the 4-
Page 8 of 14
NH
N
NH
N
1
2
3
4
5
6
7
8
position of the pyrimidine ring to provide free pyrazole
28. The free pyrazole 28 successfully went through Suzuki
coupling with 23 in a microwave oven at 180 °C for 2 h to
provide the final compounds in good yields. Alternatively,
free pyrazole 28 was protected with Boc or THP
Cl
a
HN
N
N
HN
b
W
N
O
N
Cl
N
W
W
O
R2
N
N
Cl
protecting groups (29 and 30) and Suzuki coupling was
then carried out under more conventional reaction
conditions (e.g. 100 °C, overnight). These synthetic routes
in general worked well for the synthesis of majority of the
final compounds with a few exceptions including 9g.
Attempts to synthesize the dichloropyrimidine 27 core of
9g failed due to its stability issue. Compound 9g was then
synthesized through a different and slightly lengthier
route that is depicted in Scheme 3.
R1
28
27
8, 9a-f, 9h-k
c
R2
N
d
f
R2
N
9
N
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
N
R1
R1
N
N
N
N
e
O
N
W
O
R2
W
N
N
Cl
Scheme 1. Synthesis of boronic esters and THP protected
4-aminophenylpyrazole intermediatesa
R1
31,
32,
R1 = R2 = Boc
R1 = H, R2 = THP
29,
30,
R1 = R2 = Boc
R1 = H, R2 = THP
O
O
R2
aReagents and conditions: (a) 4-(1H-pyrazol-4-yl)aniline,
iPr2NEt, DMF, 90-110 °C, 3h; (b) 23, Pd(PPh3)4, 9:1
dioxane/water, Na2CO3, microwave, 180 °C, 2h; (c) Boc2O,
DMAP, TEA, DCM, 50 °C, 1.5 h; (d) iPr2NEt, n-BuOH, 90-
110 °C, 16 h; (e) Pd(dppf)Cl2, K2CO3, dioxane/water, 100 °C,
16 h; (f) HCl in dioxane or TFA/DCM, rt, 2 h.
a
b
Br
O
R2
HN
R1
20
Cl
R2
N
N
R1
R1
21
22
d
c
O
O
B
O
R2
O
N
Another general synthesis of the
R1
tetrahydropyridopyrimidine derivatives is depicted in
Scheme 5. This general synthesis allows efficient
modification of the amide region of the molecule. The
synthesis started with double Michael addition of amine
48 to ethyl acrylate affording 49 followed by
intramolecular Claisen condensation of 49 providing 50.
Subsequent cyclization with urea followed by
chlorination provided substituted piperidine fused 1,4-
dichloropyrimidine 52, which further underwent Suzuki
coupling followed by deprotection to provide the final
compounds.
23
THP
N
N
H
THP
N
N
N
N
O
f
O
e
B
B
O
O
H2N
24
25
26
aReagents and conditions: (a) 2-chloroacetyl chloride,
Na2CO3, water, rt, 1 h; (b) 3-bromophenol, K2CO3, DMF,
80 °C, 12h; (c) 4,4,4',4',5,5,5',5'-octamethyl-2,2'-bi(1,3,2-
dioxaborolane, Pd(dppf)Cl2, AcOK, dioxane, 85 °C, 16 h;
(d) 3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenol,
K2CO3, DMF, 80 °C, 12 h; (e) DHP, p-TsOH, DCM, 20 °C, 4
h; (f) 4-bromoaniline, Pd(dppf)Cl2, K2CO3, dioxane/water,
100 °C, 16 h.
Scheme 3. Synthesis of 9ga
One general synthesis of the tetrahydropyridopyrimidine
derivatives is depicted in Scheme 4. This synthetic route
allows efficient derivatization on the piperidinyl nitrogen.
The synthesis started with commercially available Boc
protected 1,4-dichloropyrimidine 44 followed by aromatic
nucleophilic substitution, Suzuki coupling, and
deprotection to provide free piperidine 47, which was
then derivatized through alkylation, reductive amination,
or acylation.
Scheme 2. Synthesis of 8, 9a-f, 9h-ka
8
ACS Paragon Plus Environment

Page 9 of 14
1
2
3
Journal of Medicinal Chemistry
O
O
aReagents and conditions: (a) 26, iPr2NEt, n-BuOH, 100
O
N
OMe
NH
N
NH2
NH
°C, 16 h; (b) 23, Pd(dppf)Cl2, dioxane/water, 100 °C, 16 h;
(c) HCl/dioxane/water, rt, 15 h; (d) aldehyde, NaBH3CN,
HOAc, MeOH, rt, 16 h; (e) R’Br, DMF, rt, 16 h.
N
OMe
a
b
c
NH2
R
O
O
O
O
4
5
Scheme 5. Second general synthesis of
33
34
35
Br
tetrahydropyridopyrimidine derivativesa
6
7
R'
e
g
HN
N
O
O
N
N
N
a
b
O
N
N
8
9
N
EtO
N
OEt
R NH2
OEt
O
R'
R
48
49
50
O
O
e
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
36
37
R = OH
R = Cl
38
39
R = Me
R = H
THP
d
f
Cl
OH
N
N
R'
Br
c
R'
d
N
N
N
N
h
HN
N
Cl
THP
N
N
N
OH
THP
N
N
HN
51
52
N
N
O
N
N
O
O
N
O
O
R
N
O
40
41
42
R = Me
R = H
i
THP
NH
N
HN
N
HN
N
N
N
f
R'
R'
N
N
N
N
O
O
R
j
HN
Cl
N
H
HN
53
54
NH
N
N
N
O
N
N
O
O
N
N
O
H
N
N
H
43
9g
HN
aReagents and conditions: (a) 3-methoxybenzoyl chloride,
TEA, DCM, rt, 16 h; (b) NH3, MeOH, rt, 16 h; (c) NaOH,
iPrOH, 90 °C, 3 h; (d) POCl3, iPr2NEt, 80 °C, 14 h; (e) 4-
bromoaniline, THF, 60 °C, 16 h; (f) BBr3, 35 °C, 16 h; (g)
methyl 2-bromoacetate, K2CO3, DMF, 40 °C, 1 h; (h) 24,
Pd(dppf)Cl2, dioxane/water, 100 °C, 16 h; (i) LiOH,
THF/water, 15 °C, 16 h; (j) HCl in dioxane, 15 °C, 1 h.
g
R'
N
N
O
O
R
N
N
H
14c, 15b, 16c
aReagents and conditions: (a) ethyl acrylate, EtOH, rt, 72
h; (b) LiHMDS, THF, rt, 2 h; (c) Urea, NaOMe, MeOH, 80
°C, 16 h; (d) POCl3, iPr2NEt, 90 °C, 16 h; (e) 26, iPr2NEt, n-
BuOH, 100 °C, 16 h; (f) 23, Pd(dppf)Cl2, dioxane/water, 100
°C, 16 h; (g) HCl/dioxane/water, rt, 15 h.
Scheme 4. General synthesis of
tetrahydropyridopyrimidine derivativesa
THP
N
Both versions of the general synthesis depicted in
Schemes 4 and 5 worked well to introduce variations in
different regions of the molecule and were used to
prepare gram quantities of material. However, these
schemes are not ideal for the synthesis of large quantities
of GLP or GMP grade material, due to some undesirable
or low-yielding steps such as the Pd-mediated coupling
and the reductive amination. For individual compounds,
a more efficient synthesis was then developed. In Scheme
6, a scale-up synthesis of 15b is presented, and it
represents another general synthesis of this class of
compounds. The key feature of this efficient convergent
synthesis is that the amide moiety was pre-installed in
amidine 58, and the tetrahydropyridopyrimidine core was
synthesized via a [3+3] cyclization. This robust synthesis
allows less effort for purification of intermediates and has
been used for greater than 100 g scales.
Cl
N
N
a
b
Boc
N
N
HN
N
Cl
Boc
44
45
N
N
Cl
NH
N
THP
N
N
HN
N
c
HN
HN
N
O
Boc
O
R
N
N
O
N
H
O
R
N
N
H
46
47
NH
N
Scheme 6. Synthesis of 15ba
HN
N
d or e
R'
N
N
O
O
R
N
H
9l, 9m, 12a-p, 13a, 15a, 15c, 17a, 18a, 19a
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 14
O
O
made to minimize animal suffering, to reduce the number
of animals.
a
b
NC
O
1
2
Cl
N
H
H2N
55
N
H
56
57
Synthesis of 2-(3-(4-((4-(1H-Pyrazol-4-yl)phenyl)amino)-
6-cyclobutyl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-2-
yl)phenoxy)-N-(1-methylcyclopropyl)acetamide (15b)
3
4
5
NH
O
c
O
H2N
N
H
Step 1. 2-Chloro-N-(1-methylcyclopropyl)acetamide (56).
58
6
7
8
9
To a mixture of 1-methylcyclopropan-1-amine
O
O
O
d
e
hydrochloride 55 (200.0 g, 1.86 mol) and triethylamine
(564.4 g, 5.58 mol) in DCM (1.0 L) was added 2-
chloroacetyl chloride (220.6 g, 1.95 mol) drop-wise at -10
°C. The reaction mixture was then warmed to 20 °C and
stirred at 20 °C for 12 h, poured into water, and washed
with HCl (1.0 M, 4.0L). The combined organic layers were
dried with anhydrous Na2SO4, filtered, and concentrated
under vacuum to give the title compound (198.0 g, 72%)
as a brown solid. 1H NMR (400 MHz, DMSO-d6) δ 8.42 (s,
1H), 3.93 (s, 2H), 1.27 (s, 3H), 0.62 – 0.53 (m, 4H). HRMS
m/z calculated for C6H10ClNO [M + H+]: 148.0524, found
148.0529.
N
O
NH2
EtO
N
OEt
OEt
59
60
61
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
f
g
N
NH
O
N
N
H
62
Cl
N
N
N
O
h
15b
O
N
H
63
aReagents and conditions: (a) 2-chloroacetyl chloride,
TEA, DCM, -5 °C to 20 °C, 2 h, 72%; (b) 3-cyanophenol,
K2CO3, CH3CN, 10 °C to 80 °C, 24 h, 64%; (c) i. AcCl,
EtOH, 0 °C to 20 °C, 12 h; ii. NH3, EtOH, 20 °C to 25 °C, 11
h; (d) ethyl acrylate, EtOH, 20 °C, 72 h, 72%; (e) LiHMDS,
THF, 20 °C, 2 h, 98%; (f) 58, NaOMe, 80 °C, 9 h, 60%; (g)
SOCl2, DMF, 0 °C to 20 °C, 9.5 h, 52%; (h) 4-(1H-pyrazol-
4-yl)aniline, NMP, 145 °C, 5 h, 51%.
Step 2. 2-(3-Cyanophenoxy)-N-(1-
methylcyclopropyl)acetamide (57). To a mixture of 3-
hydroxybenzonitrile (152.0 g, 1.28 mol) and K2CO3 (352.8
g, 2.55 mol) in CH3CN (2.0 L) was added 56 (198.3 g, 1.34
mol) in portions. The mixture was stirred at 80 °C for 10
h, concentrated, and extracted with EtOAc (2.0 L × 3).
The combined organic phases were dried with anhydrous
Na2SO4, filtered, and concentrated to give a crude black
solid. The crude product was triturated with EtOAc and
MTBE/EtOAc to give the title compound as a brown solid
(105 g). The mother liquor was concentrated and purified
by chromatography to provide additional title compound
(84.0 g). Total yield: 189 g, 64%. 1H NMR (400 MHz,
CDCl3) δ 7.43 (m, 1H), 7.31 (m, 1H), 7.19 – 7.16 (m, 2H),
6.82 (s, 1H), 4.43 (s, 2H), 1.41 (s, 3H), 0.81 – 0.68 (m, 4H).
HRMS m/z calculated for C13H14N2O2 [M + H+]: 231.1128,
found 231.1132.
CONCLUSIONS
In summary, a class of potent and orally bioavailable
inhibitors of glucose transport, targeting GLUT1/3, have
been identified. SAR and optimization were carried out to
improve the ADME, PK, and the brain penetration profile.
Compound 15b was identified with good GLUT1/3
potency, low brain penetration, and favorable ADME and
PK profile across several different species and can serve as
a valuable tool molecule to study GLUT functions in vivo.
Step 3. 2-(3-Carbamimidoylphenoxy)-N-(1-
methylcyclopropyl)acetamide (58). To a mixture of 57
(189.0 g, 0.821 mol) in EtOH (1.0 L) was added acetyl
chloride (773.2 g, 9.85 mol) drop-wise at 0 °C. The
mixture was stirred at 20 °C for 12 h and was concentrated
and was then stirred with NH3 in EtOH (2 L) for 10 h. The
mixture was concentrated to provide a residue which was
triturated with EtOH (300 mL) to afford the title
compound (242.0 g) as a yellow solid. The crude material
was used directly for in the next step reaction without
further purification. 1H NMR (400 MHz, DMSO-d6) δ 9.33
(b, 3H), 8.46 (s, 1H), 7.53 (t, J = 8.0 Hz, 1H), 7.45-7.43 (m,
2H), 7.30-7.28 (m, 1H), 4.53 (s, 2H), 1.28 (s, 3H), 0.66-0.64
(m, 2H), 0.56-0.53 (m, 2H). HRMS m/z calculated for
C13H17N3O2 [M + H+]: 248.1394, found 248.1395.
EXPERIMENTAL SECTION
General. All solvents and reagents were obtained
commercially and used as received. 1H NMR spectra were
recorded on a Bruker instrument (300 MHz, 400 MHz, or
500 MHz) in the cited deuterated solvents. Chemical
shifts are given in ppm, and coupling constants are in
hertz. All final compounds were purified by flash
chromatography using 220-400 mesh silica gel or reverse-
phase HPLC with CH3CN/water as the solvents. Thin-
layer chromatography was done on silica gel 60 F-254
(0.25-mm thickness) plates. Visualization was
accomplished with UV light and/or 10%
Step 4. Diethyl 3,3'-(cyclobutylazanediyl)dipropionate
(60).
phosphomolybdic acid in ethanol. LC-MS experiments
were done on a Shimadzu LCMS-2020 system. Compound
purity was determined by a LC-MS or HPLC with 230 nm
and 254 nm wavelengths. All final compounds reported
here have purity ≥ 95%. All animal experiments
A mixture of cyclobutanamine 59 (300.0 g, 4.22 mol) and
ethyl acrylate (1.24 kg, 12.39 mol) in EtOH (900 mL) was
stirred at 30 °C for 72 h. The reaction mixture was
concentrated under vacuum. The residual was purified by
flash chromatography (SiO2, petroleum ether: ethyl
acetate = 1:0 to 1:2) to give the title compound (823.0 g,
performed in the manuscript were conducted in
compliance with institutional guidelines. All efforts were
10
ACS Paragon Plus Environment

Page 11 of 14
Journal of Medicinal Chemistry
71.9%) as a yellow oil. 1H NMR (400 MHz, CDCl3) δ 4.03
yl)phenoxy)-N-(1-methylcyclopropyl)acetamide (15b). A
1
2
3
4
5
6
7
8
(q, 4 H), 2.99 – 2.96 (m, 1 H) 2.66 (t, J = 7.2 Hz, 4 H), 2.32
(t, J = 7.6 Hz, 4 H), 1.95 – 1.91 (m, 2 H), 1.77 - 1.74 (m, 2 H),
1.53 - 1.48 (m, 2 H), 1.17 (t, J = 7.2 Hz, 6 H). HRMS m/z
calculated for C14H25NO4 [M + H+]: 272.1856, found
272.1863.
mixture of 63 (145.0 g, 0.340 mol), 4-(1H-pyrazol-4-
yl)aniline (54.1 g, 0.340 mol), and N-methyl-2-pyrrolidone
(580 mL) was stirred at 145 °C for 5 h, diluted with CH3CN
(1.50 L), and filtered. The filter cake was dissolved with
Conc. HCl (500 ml) and MeOH (3 L). The resulting
solution was concentrated to give a crude material which
was purified by prep-HPLC to give the title compound as
a yellow HCl salt solid (102.2 g, 51.4%). 1H NMR (400
MHz, CD3OD) δ 8.12 (s, 2H), 7.88 (d, J = 8.0 Hz, 1H), 7.83
(m, 1H), 7.73 (m, 4H), 7.49 (t, J = 8.0 Hz, 1H), 7.26 (dd, J =
8.0,and 2.0 Hz, 1H), 4.52 (s, 2H), 4.38 (s, 2H), 4.08 - 3.99
(m, 1H), 3.64 (s, 2H), 3.36 (t, J = 5.2 Hz, 2H), 2.54- 2.50 (m,
4H), 2.02 - 1.95 (m, 2H), 1.35 (s, 3H), 0.72 (m, 2H), 0.62 (m,
2H); 13C NMR (101 MHz, DMSO-d6) δ 168.24, 159.75,
158.39, 157.19, 156.52, 137.15, 136.65, 131.10, 130.20, 129.36,
125.61, 123.64, 121.41, 121.22, 118.38, 114.33, 105.66, 67.42,
58.56, 45.05, 44.47, 28.63, 27.23, 25.40, 25.35, 23.03, 14.01.
HRMS m/z calculated for C32H36N7O2 [M + H+]: 550.2925,
found 550.2912.
Step 5. Ethyl 1-cyclobutyl-4-oxopiperidine-3-carboxylate
(61). To LiHMDS in THF (1.0 M, 1.11 L) was added a
solution of 60 (200.0 g, 0.727 mol) in THF (400 mL)
under N2 atmosphere. The mixture was stirred at 20 °C
for 2 h, quenched with saturated NH4Cl (2.00 L), and
extracted with ethyl acetate (2.00 L × 2). The combined
organic phases were dried by Na2SO4 and concentrated in
vacuo to provide the title compound (163.0 g, 98.2%) as a
brown oil which was used directly in the next step
reaction without further purification. 1H NMR (400 MHz,
DMSO-d6) δ 4.21-4.09 (m, 2H), 3.53-3.50 (m, 1H), 2.90-
2.69 (m, 4H), 2.48-2.30 (m, 3H), 2.00-1.98 (m, 2H), 1.83-
1.77 (m, 2H), 1.65-1.61 (m, 2H), 1.24-1.17 (m, 3H). HRMS
m/z calculated for C12H19NO3 [M + H+]: 226.1438, found
226.1437.
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
GLUT, GLUT1 and GLUT3 assays
Step 6. 2-(3-(6-Cyclobutyl-4-oxo-3,4,5,6,7,8-
GLUT1 and GLUT3 assays were performed using a
modified version of the pairwise assay described by
Siebeneicher et al.22. Briefly, GLUT1-dependent DLD1
wild-type and GLUT3-dependent DLD1-SLC2A1−/−
(Horizon Discovery) were maintained in RPMI 1640 cell
culture medium containing 10% fetal bovine serum, 1%
penicillin-streptomycin and 10 mM HEPES in a
humidified incubator with 5% CO2 at 37° C. The day
before the assay, the cells were seeded in 90 L of this
medium in 96-well plates at a density of 50,000 cells/well
and allowed to attach overnight. The day of the assay, 10
μL of culture medium containing GLUT inhibitors and
oligomycin was added to each well to a final
concentration of 10 μM oligomycin and 0.06% DMSO.
The plates were returned to the incubator for 90 minutes
and then ATP levels were determined using the
CellTiter-Glo® Luminescent Cell Viability Assay
(Promega). For GLUT HT-1080 assay, the procedure was
the same except the cells were seeded at 40,000 cells/well.
hexahydropyrido[4,3-d]pyrimidin-2-yl)phenoxy)-N-(1-
methylcyclopropyl)acetamide (62). To a mixture of 61
(163.0 g, 0.724 mol) and 58 (119.3 g, 0.482 mol) in MeOH
(1.6 L) was added NaOMe (104.2 g, 1.93 mol). The mixture
was stirred at 80 °C for 9 h and was concentrated under a
reduced pressure. The residual was diluted with H2O (2.0
L) and neutralized with HCl. The mixture was filtrated
and the cake was dried in vacuo to provide the title
compound (118.7 g, 60.1%) as a yellow solid. 1H NMR (400
MHz, DMSO-d6) δ 12.58 (s, 1H), 8.33 (s, 1H), 7.68 (d, J =
8.0 Hz, 1H), 7.64 (s, 1H), 7.41 (t, J = 8.0, 1H), 7.14 (dd, J =
8.0, and 1.6 Hz, 1H), 4.47 (s, 2H), 3.16 (s, 2H), 2.89 - 2.70
(m, 1H), 2.66 - 2.64 (m, 2H), 2.55 - 2.54 (m, 2H), 2.06 -
2.05 (m, 2H), 1.86 - 1.83 (m, 2H), 1.69 - 1.67 (m, 2H), 1.29
(s, 3H), 0.67 – 0.64 (m, 2H), 0.64 – 0.55 (m, 2H). HRMS
m/z calculated for C23H28N4O3 [M + H+]: 409.2234, found
409.2250.
Step 7. 2-(3-(4-Chloro-6-cyclobutyl-5,6,7,8-
tetrahydropyrido[4,3-d]pyrimidin-2-yl)phenoxy)-N-(1-
methylcyclopropyl)acetamide (63). A mixture of 62 (116.0
g, 0.284 mol) was dissolved in DMF (1.50 L) and was
cooled to 0 °C. To the mixture was added SOCl2 (50.7 g,
0.426 mol). The resulting mixture was stirred at 0 °C for
30 min followed by 20 °C for 9 h. The mixture was
filtrated and the solid collected was washed with cold
saturated sodium bicarbonate solution (1.0 L) and water.
The solid was redissolved in dichloromethane, dried over
Na2SO4, and concentrated in vacuo to afford the title
compound (70.0 g, 57.7% yield) as a brown solid. 1H NMR
(400 MHz, CDCl3) δ 8.07 (d, J = 8.4 Hz, 1H), 7.98 (d, J =
2.4 Hz, 1H), 7.40 (t, J = 8.0 Hz, 1H), 7.00 (dd, J = 8.4, 2.0
Hz, 1H), 6.91 (s, 1H), 4.51 (s, 2H), 3.53 (s, 2H), 3.07-3.02 (m,
3H), 2.73 (t, J = 6.0 Hz, 2H), 2.19-2.17 (m, 2H), 2.02-1.96
(m, 2H), 1.81-1.74 (m, 2H), 1.43 (s, 3H), 0.84-0.81 (m, 2H),
0.72-0.68 (m, 2H). HRMS m/z calculated for
Glucose competition assay
The assay was performed according to the standard
GLUT1 (DLD-1) assay protocol, except that the cells were
washed once in 100 µL PBS prior to addition of media
containing the test compounds, oligomycin (10 µM) and
the glucose at either 10, 3.33, 1.11 or 0.37 mM
concentration.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on
the ACS Publications website at DOI: xxx.
Details of the synthetic procedures and analytical data for
intermediates and final compounds 9a-m, 10a-i, 11a, 11b,
12a-p, 13a, 14b, 14c, 15a, 15c, 16c, 17a, 18a, 19a (S1-S46);
Table S1. GLUT1 IC50 values (nM) in the presence of
varying glucose concentrations (S47); Table S2. Eurofins
safety panel screen of 15b (S48).
C23H27ClN4O2 [M + H+]: 427.1895, found 447.1897.
Step 8. 2-(3-(4-((4-(1H-Pyrazol-4-yl)phenyl)amino)-6-
cyclobutyl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-2-
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 14
Metabolism and Lymphocyte Function. Science 2013,
342, 1242454.
Molecular formula strings are available separately in
comma-separated values file format (CSV).
1
2
3
4
5
6
7
8
8. Hotamisligil, G. S. Foundations of
Immunometabolism and Implications for Metabolic
Health and Disease. Immunity 2017, 47, 406-420.
AUTHOR INFORMATION
Corresponding Author
9. Palmer, C. S.; Ostrowski, M.; Balderson, B.; Christian,
N.; Crowe, S. M. Glucose Metabolism Regulates T
Cell Activation, Differentiation, and Functions. Front.
Immunol. 2015, 6, 1-6.
*K.G.L.: e-mail, kevin.liu@kadmon.com.
ACKNOLEDGMENT
9
10. Macintyre, A. N.; Gerriets, V. A.; Nichols, A. G.;
Michalek, R. D.; Rudolph, M. C.; Deoliveira, D.;
Anderson, S. M.; Abel, E. D.; Chen, B. J.; Hale, L. P.;
Rathmell, J. C. The Glucose Transporter Glut1 Is
Selectively Essential for Cd4 T Cell Activation and
Effector Function. Cell Metab. 2014, 20, 61-72.
The authors thank Professor Joshua D. Rabinowitz of
Princeton University for his valuable insights and
suggestions and Yong Li, Yafeng Cao, Binglin Wang, Xian
Xiao, and others involved at WuXi AppTec (Wuhan) for
their synthetic support.
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
11. Navale, A. M.; Paranjape, A. N. Glucose Transporters:
Physiological and Pathological Roles. Biophys. Rev.
2016, 8, 5-9.
ABBREVIATIONS USED
ADME, adsorption, distribution, metabolism, and excre-
tion; Ac, acetyl; AUC, area under the curve; BCRP, breast
cancer resistance protein; BBB, blood-brain barrier; CNS,
central nervous system; CL, clearance; Da, Dalton; DCM,
dichloromethane; DHP, 3,4-dihydropyran; DMAP, 4-
(dimethylamino)pyridine; DMF, N,N-
dimethylformamide; GLUT, glucose transporters; IC50,
inhibitory concentration at half maximal effect; 18F-FDG,
[18F]fluoro-2-deoxyglucose; fu, unbound fraction; iv,
intravenous; LM, liver microsomes; LiHMDS, lithium
bis(trimethylsilyl)amide; mM, millimolar; µM, mi-
cromolar; OXPHOS, oxidative phosphorylation; P-gp, P-
glycoprotein; Pd(dppf)Cl2, [1,1′-
bis(diphenylphosphino)ferrocene]dichloropalladium(II);
PK, pharmacokinetic; po, oral administration; RO5, rule
of five (Lipinski); SAR, structure−activity relationship;
TCA, tricarboxylic acid; Th, T helper; THP, 2-
tetrahydropyranyl; p-TsOH, p-Toluenesulfonic acid; Vdss,
volume of distribution at steady state.
12. Deng, D.; Xu, C.; Sun, P.; Wu, J.; Yan, C.; Hu, M.; Yan,
N. Crystal Structure of the Human Glucose
Transporter Glut1. Nature 2014, 510, 121-125.
13. Deng, D.; Sun, P.; Yan, C.; Ke, M.; Jiang, X.; Xiong, L.;
Ren, W.; Hirata, K.; Yamamoto, M.; Fan, S.; Yan, N.
Molecular Basis of Ligand Recognition and Transport
by Glucose Transporters. Nature 2015, 526, 391-396.
14. Adekola, K.; Rosen, S. T.; Shanmugam, M. Glucose
Transporters in Cancer Metabolism. Curr. Opin.
Oncol. 2012, 24, 650-654.
15. Granchi, C.; Fancelli, D.; Minutolo, F. An Update on
Therapeutic Opportunities Offered by Cancer
Glycolytic Metabolism. Bioorg. Med. Chem. Lett. 2014,
24, 4915-4925.
16. Barron, C. C.; Bilan, P. J.; Tsakiridis, T.; Tsiani, E.
Facilitative Glucose Transporters: Implications for
Cancer Detection, Prognosis and Treatment.
Metabolism 2016, 65, 124-139.
REFERENCES
17. Liu, Y.; Cao, Y.; Zhang, W.; Bergmeier, S.; Qian, Y.;
Akbar, H.; Colvin, R.; Ding, J.; Tong, L.; Wu, S.;
Hines, J.; Chen, X. A Small-Molecule Inhibitor of
Glucose Transporter 1 Downregulates Glycolysis,
Induces Cell-Cycle Arrest, and Inhibits Cancer Cell
Growth in Vitro and in Vivo. Mol. Cancer Ther. 2012,
11, 1672-1682.
1. Warburg, O. U¨Ber Den Stoffwechsel Der
Carcinomzelle. Naturwissenschaften 1924, 12, 1131-1137.
2. Warburg, O. On the Origin of Cancer Cells. Science
1956, 123, 309-314.
3. Altenberg, B.; Greulich, K. O. Genes of Glycolysis Are
Ubiquitously Overexpressed in 24 Cancer Classes.
Genomics 2004, 84, 1014-1020.
18. Wang, D.; Chu, P. C.; Yang, C. N.; Yan, R.; Chuang, Y.
C.; Kulp, S. K.; Chen, C. S. Development of a Novel
Class of Glucose Transporter Inhibitors. J. Med.
Chem. 2012, 55, 3827-3836.
4. Upadhyay, M.; Samal, J.; Kandpal, M.; Singh, O. V.;
Vivekanandan, P. The Warburg Effect: Insights from
the Past Decade. Pharmacol. Ther. 2013, 137, 318-330.
19. Chan, D. A.; Sutphin, P. D.; Nguyen, P.; Turcotte, S.;
Lai, E. W.; Banh, A.; Reynolds, G. E.; Chi, J. T.; Wu, J.;
Solow-Cordero, D. E.; Bonnet, M.; Flanagan, J. U.;
Bouley, D. M.; Graves, E. E.; Denny, W. A.; Hay, M.
P.; Giaccia, A. J. Targeting Glut1 and the Warburg
Effect in Renal Cell Carcinoma by Chemical Synthetic
Lethality. Sci. Transl. Med. 2011, 3, 94ra70.
5. DeBerardinis, R. J.; Lum, J. J.; Hatzivassiliou, G.;
Thompson, C. B. The Biology of Cancer: Metabolic
Reprogramming Fuels Cell Growth and Proliferation.
Cell Metab. 2008, 7, 11-20.
6. Vander Heiden, M. G.; Cantley, L. C.; Thompson, C.
B. Understanding the Warburg Effect: The Metabolic
Requirements of Cell Proliferation. Science 2009, 324,
1029-1033.
20. Kapoor, K.; Finer-Moore, J. S.; Pedersen, B. P.;
Caboni, L.; Waight, A.; Hillig, R. C.; Bringmann, P.;
Heisler, I.; Muller, T.; Siebeneicher, H.; Stroud, R. M.
7. Pearce, E. L.; Poffenberger, M. C.; Chang, C.-H.;
Jones, R. G. Fueling Immunity: Insights into
12
ACS Paragon Plus Environment

Page 13 of 14
Journal of Medicinal Chemistry
Mechanism of Inhibition of Human Glucose
33. Gillis, E. P.; Eastman, K. J.; Hill, M. D.; Donnelly, D. J.;
Meanwell, N. A. Applications of Fluorine in
Medicinal Chemistry. J. Med. Chem. 2015, 58, 8315-
8359.
1
2
3
4
5
6
7
8
9
Transporter Glut1 Is Conserved between Cytochalasin
B and Phenylalanine Amides. Proc. Natl. Acad. Sci. U.
S. A. 2016, 113, 4711-4716.
21. Siebeneicher, H.; Bauser, M.; Buchmann, B.; Heisler,
I.; Muller, T.; Neuhaus, R.; Rehwinkel, H.; Telser, J.;
Zorn, L. Identification of Novel Glut Inhibitors.
Bioorg. Med. Chem. Lett. 2016, 26, 1732-1737.
34. Meanwell, N. A. Fluorine and Fluorinated Motifs in
the Design and Application of Bioisosteres for Drug
Design. J. Med. Chem. 2018, 61, 5822-5880.
35. Sugano, K.; Kansy, M.; Artursson, P.; Avdeef, A.;
Bendels, S.; Di, L.; Ecker, G. F.; Faller, B.; Fischer, H.;
Gerebtzoff, G.; Lennernaes, H.; Senner, F.
Coexistence of Passive and Carrier-Mediated
Processes in Drug Transport. Nat. Rev. Drug. Discov.
2010, 9, 597-614.
22. Siebeneicher, H.; Cleve, A.; Rehwinkel, H.; Neuhaus,
R.; Heisler, I.; Muller, T.; Bauser, M.; Buchmann, B.
Identification and Optimization of the First Highly
Selective Glut1 Inhibitor Bay-876. ChemMedChem
2016, 11, 2261-2271.
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
23. Reckzeh, E. S.; Karageorgis, G.; Schwalfenberg, M.;
Ceballos, J.; Nowacki, J.; Stroet, M. C. M.; Binici, A.;
Knauer, L.; Brand, S.; Choidas, A.; Strohmann, C.;
Ziegler, S.; Waldmann, H. Inhibition of Glucose
Transporters and Glutaminase Synergistically Impairs
Tumor Cell Growth. Cell. Chem. Biol. 2019, 26, 1214-
1228 e1225.
36. Bagal, S.; Bungay, P. Restricting Cns Penetration of
Drugs to Minimise Adverse Events: Role of Drug
Transporters. Drug Discov. Today Technol. 2014, 12,
e79-85.
37. Volpe, D. A. Drug-Permeability and Transporter
Assays in Caco-2 and Mdck Cell Lines. Future Med.
Chem. 2011, 3, 2063-2077.
24. Sergeant, S.; Kim, H. D. Inhibition of 3-O-
Methylglucose Transport in Human Erythrocytes by
Forskolin. J. Biol. Chem. 1985, 260, 14677-14682.
38. Olszewski, K.; Kim, J.-I.; Poyurovsky, M.; Liu, K.;
Barsotti, A.; Morris, K. Preparation of Fused
Pyrimidines as Glucose Uptake Inhibitors.
WO2016210330A1, 2016.
25. Ritchie, T. J.; Macdonald, S. J. The Impact of
Aromatic Ring Count on Compound Developability--
Are Too Many Aromatic Rings a Liability in Drug
Design? Drug Discov. Today 2009, 14, 1011-1020.
26. Lovering, F.; Bikker, J.; Humblet, C. Escape from
Flatland: Increasing Saturation as an Approach to
Improving Clinical Success. J. Med. Chem. 2009, 52,
6752-6756.
27. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney,
P. J. Experimental and Computational Approaches to
Estimate Solubility and Permeability in Drug
Discovery and Development Settings. Adv. Drug.
Deliv. Rev. 2001, 46, 3-26.
28. Hitchcock, S. A.; Pennington, L. D. Structure-Brain
Exposure Relationships. J. Med. Chem. 2006, 49,
7559-7583.
29. Wager, T. T.; Hou, X.; Verhoest, P. R.; Villalobos, A.
Moving Beyond Rules: The Development of a Central
Nervous System Multiparameter Optimization (cns
mpo) Approach to Enable Alignment of Druglike
Properties. ACS Chem. Neurosci. 2010, 1, 435-449.
30. Ghose, A. K.; Herbertz, T.; Hudkins, R. L.; Dorsey, B.
D.; Mallamo, J. P. Knowledge-Based, Central Nervous
System (Cns) Lead Selection and Lead Optimization
for Cns Drug Discovery. ACS Chem. Neurosci. 2012, 3,
50-68.
31. Rankovic, Z. Cns Physicochemical Property Space
Shaped by a Diverse Set of Molecules with
Experimentally Determined Exposure in the Mouse
Brain. J. Med. Chem. 2017, 60, 5943-5954.
32. Palm, K.; Luthman, K.; Ros, J.; Grasjo, J.; Artursson, P.
Effect of Molecular Charge on Intestinal Epithelial
Drug Transport: Ph-Dependent Transport of Cationic
Drugs. J. Pharmacol. Exp. Ther. 1999, 291, 435-443.
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 14
Table of Contents graphic
1
NH
N
2
3
4
5
6
HN
7
N
N
O
8
9
O
N
N
H
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15b
GLUT1 IC50 = 242 nM
GLUT3 IC50 = 179 nM
Oral F = 45.5% (mouse) and 49.4% (rat)
14
ACS Paragon Plus Environment