Discovery of a small molecule MDM2 inhibitor (AMG 232) for treating cancer

Article abstract of DOI:10.1021/jm500627s

We recently reported the discovery of AMG 232 (1), a potent and selective piperidinone inhibitor of the MDM2-p53 protein-protein interaction. Compound 1 is currently being evaluated in human clinical trials for the treatment of cancer. This article provides an overview of its discovery from the de novo design of the piperidinone series to the structure-activity studies leading to the identification of 1. In addition, this article also describes the preclinical pharmacology and pharmacokinetics of 1, along with its drug metabolism and safety assessment.

Full text of DOI:10.1021/jm500627s

Subscriber access provided by UNIV OF UTAH
Drug Annotation
Discovery of a Small Molecule MDM2 Inhibitor (AMG 232) for Treating Cancer
Yosup Rew, and Daqing Sun
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm500627s • Publication Date (Web): 26 Jun 2014
Downloaded from http://pubs.acs.org on June 30, 2014
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 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of a Small Molecule MDM2 Inhibitor
(AMG 232) for Treating Cancer
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
Yosup Rew* and Daqing Sun*
Departments of Therapeutic Discovery, Amgen Inc., 1120 Veterans Boulevard, South San
Francisco, California 94080, United States
KEYWORDS: MDM2, p53, proteinꢀprotein interaction, piperidinone, AMG 232.
ABSTRACT: We recently reported the discovery of AMG 232 (1), a potent and selective
piperidinone inhibitor of the MDM2ꢀp53 proteinꢀprotein interaction. Compound 1 is currently
being evaluated in human clinical trials for the treatment of cancer. This article provides an
overview of its discovery from the de novo design of the piperidinone series to the structureꢀ
activity studies leading to the identification of 1. In addition, this article also describes the
preclinical pharmacology and pharmacokinetics of 1, along with its drug metabolism and safety
assessment.
INTRODUCTION
The p53 tumor suppressor protein induces cell growth arrest and apoptosis in response to
DNA damage or stress,1 and it has been proposed that inactivation of the p53 pathway may be a
required step for tumor survival.2 Approximately 50% of human cancers have p53 mutations
resulting in loss of its function.3 The wildꢀtype p53, present in the remaining 50% of
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 34
1
2
3
4
5
6
7
8
9
malignancies, is downregulated by the MDM2 (human murine double minute 2) oncoprotein via
a direct proteinꢀprotein interaction.2, 4 After binding to p53, MDM2 inhibits p53 activity by
blocking p53 mediated transactivation,5 inducing nuclear export of p53,6 and acting as E3
ubiquitin ligase targeting both itself and p53 for the degradation by the 26S proteasome.7 Since
all of these mechanisms can be blocked by neutralizing the MDM2ꢀp53 interaction, disruption of
the proteinꢀprotein interaction with small molecule MDM2 inhibitors has been considered a
promising strategy to reactivate the p53 pathway in wildꢀtype p53 tumors over the last two
decades.8 To date, a couple of studies with small molecule MDM2 inhibitors have demonstrated
tumor growth inhibition both in vitro and in vivo,9–11 and several small molecules have recently
advanced into clinical trials (Figure 1).12–15
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
2

Page 3 of 34
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 1. Chemical structures and potencies of RG7112 (Roche),13a RG7388 (Roche),13b
SAR299155 (Sanofi),14 and 1.15 aIC50 in the EdU proliferation assay (SJSAꢀ1, 10% human
serum); bKD in the surface plasmon resonance (SPR) spectroscopy binding assay.
Recently, we reported the discovery of a promising piperidinone MDM2 inhibitor, 1
(AMG 232),11,15 which is currently being evaluated in human clinical trials. In this article, we
provide a comprehensive overview of the scientific efforts focused on the discovery of
compound 1: (a) de novo design of piperidinone inhibitors; (b) optimization of piperidinone
series inhibitors via conformational controls and extensive SAR; (c) discovery of 1 via capturing
an additional interaction with Gly58 shelf region; (d) chemical synthesis of 1; (e) in vitro and in
vivo pharmacology; (f) binding mode; and (g) drug metabolism. Finally, this article briefly
describes (h) preclinical pharmacokinetics and human PK prediction, and (i) safety assessment.
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 34
1
2
DE NOVO DESIGN OF PIPERIDINONE INHIBITORS11
3
4
5
6
7
8
9
Our early research in the MDM2 program focused on the optimization of a class of
chromenotriazolopyrimidine inhibitors identified from high throughput screening of a smallꢀ
molecule library (Amgen, Highꢀthroughput screening (HTS) lead; Figure 2).16 To complement
these efforts, we initiated a parallel strategy, a de novo design approach to identify a novel
scaffold of MDM2 inhibitors. In 1996, the crystal structure of the p53 binding domain of human
MDM2 bound to a 15ꢀresidue peptide of human p53 was reported and unveiled that three
hydrophobic residues of p53 (Phe19, Trp23, and Leu26) are essential for key hydrophobic
contacts with the MDM protein.17 Subsequently, these key hydrophobic interactions were
confirmed by the coꢀcrystal structures of small molecule inhibitors with MDM2 (Figure 2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Binding modes of known MDM2 inhibitors, Nutlinꢀ3a,9a Amgen’s HTS Lead,16a and
Oxindole (2).18 MDM2 binding pockets are labeled (by p53 side chain) in blue. aIC50 in the
homogeneous timeꢀresolved fluorescence (HTRF) assay (serum free); bIC50 in the HTRF
biochemical assay (15% human serum).11
ACS Paragon Plus Environment
4

Page 5 of 34
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 3. Early hits from de novo design.11b aCompounds are racemic; bIC50 in the HTRF
biochemical assay (serum free).
Analysis of the binding modes of these known MDM2 inhibitors promoted us to design
several new scaffolds based on a common structural element of two aromatic groups held in
close proximity by a rigid cyclic core. These efforts resulted in the identification of multiple hits
including a morpholinone series, as illustrated by compound 6, with an IC50 of 5 ꢀM in the
HTRF biochemical assay (Figure 3). Adding a benzyl group at the C2 position in 6 improved
potency about 2.5ꢀfold (7 vs. 6 in Figure 4). At a glance, we presumed that three aryl groups of 7
contact the corresponding key binding clefts in the MDM2 protein. However, the coꢀcrystal
structure of 7 with MDM2 revealed that only two hydrophobic Trp23(p53) and Phe19(p53) pockets
are filled by the pꢀbromophenyl rings, respectively and the Leu26(p53) pocket remains unoccupied
(Figure 5a). The benzyl group employs an edgeꢀface aromatic contact with the Phe55 residue on
a shallow hydrophobic shelf region.
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 34
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 4. Discovery of the novel piperidinone lead 11 from the early hit 6.11 MDM2 binding
pockets are labeled (by p53 side chain) in blue. aCompounds are racemic; bIC50 in the HTRF
biochemical assay (serum free).
As one of the strategies to fill the Leu26(p53) pocket, an attempt to change the substitution
pattern on the C6ꢀaryl ring in 6 in order to adopt a different binding mode resulted in a great
success (Figure 4). An inꢀhouse crystal structure of oxindole 2 (Figure 5b, transparent white)
shows that the Leu26(p53) pocket is filled by mꢀchlorophenyl group, which also makes a πꢀπ
stacking arrangement with His96 simultaneously. Since the oxindole 2 was one of the most
potent MDM2 inhibitors at that time, we tried to induce a similar binding mode within the
ACS Paragon Plus Environment
6

Page 7 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
morpholinone chemotype. It was proposed that this reoriented binding mode could be induced by
shifting a halogen at the paraꢀposition in the C6 phenyl group of the morpholinone to the meta
position (8 vs. 7 in Figure 4) so that the aromatic ring substitution pattern of this chemotype
would be identical to that of 2 (Figure 5b, cyan).
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)
(b)
Figure 5. (a) Xꢀray coꢀcrystal structure of the active isomer in 7 bound to MDM2 (cyan, PDB
4JV7). MDM2 binding pockets are labeled (by p53 side chain) in white. MDM2 residues are
labeled in yellow. Coꢀcrystallized water molecules are depicted as red spheres.11b (b) Model of
the morpholinone core (Nꢀmethyl, (S)ꢀCH3 at C2, cyan) in the proposed alternative binding mode
superimposed upon the coꢀcrystal structure of 2 (transparent white, the electron density of the
pendant morpholinone is not fully resolved, PDB 4JVR).11b
This hypothesis was soon verified by the SAR results of the synthesized compounds 8–
11 (Figure 4). Changing the halogen substitution of the C6 aryl from para to meta and the Nꢀ
alkyl group from methyl to cyclopropylmethyl led to compound 8, showing a retained potency
relative to 7. Incorporation of an acetic acid fragment at the C2 position resulted in a 6ꢀfold
increase of potency (9 vs. 8). The improved potency can be rationalized by a strong electrostatic
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 34
1
2
3
4
5
6
7
8
9
interaction between the acetic acid functionality appended at C2 and the His96 residue of
MDM2, which is later observed in the coꢀcrystal structure of 12 with MDM2. Notably, changing
the stereochemistry of the C6ꢀaromatic ring was well tolerated within the morpholinone core (10
vs. 9). Further modifications by inverting the stereochemical configuration of the acetic acid
substituent and switching the morpholinone core to a piperidinone, led to the identification of
transꢀpiperidinone lead 11 with improved potency.
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)
(b)
Figure 6. (a) Model of proposed binding mode of piperidinone series lead 11. MDM2 binding
pockets labeled (by p53 side chain) in white, H96 labeled in yellow.11a (b) Coꢀcrystal structure of
12 bound to human MDM2 (17–111) at 1.8 Å resolution (yellow, PDB 4ERE). MDM2 binding
pockets labeled (by p53 side chain) in white, H96 labeled in yellow. Coꢀcrystallized water
molecules are shown in red.11a
The binding mode of 11 was predicted from docking studies. As shown in Figure 6a, it
was proposed that the C5 mꢀCl phenyl group fills the Leu26(p53) pocket, while the deepest
Trp23(p53) pocket is occupied by the C6 pꢀCl Phe group. The cyclopropylmethyl group occupies
the Phe19(p53) pocket and the carboxylate anion forms an electrostatic interaction with the His96
imidazole side chain of MDM2. The binding mode of 11 was consistent with subsequent SAR
ACS Paragon Plus Environment
8

Page 9 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and confirmed by the coꢀcrystal structure of its piperidinone analogue 12 (Figure 6b). Improved
potency of piperidinone 11 over morpholinones 9 and 10 was presumably due to the alleviation
of the electrostatically repulsive interaction between the carboxylic acid and the oxygen
embedded in the morpholinone ring by replacing the morpholinone core oxygen with a
methylene.
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
OPTIMIZATION
OF
PIPERIDINONE
SERIES
INHIBITORS
VIA
CONFORMATIONAL CONTROLS AND EXTENSIVE SAR11a
To maximize interaction with MDM2, the Nꢀcyclopropylmethyl group in 11 requires a
downward orientation into the Phe19(p53) pocket (Figure 6a and Figure 7). However, this syn
orientation shown in Figure 7 destabilizes the desired binding conformer. To stabilize the
conformation required to position the Nꢀalkyl substituent into the Phe19(p53) pocket, we decided
to introduce an additional hydrophilic substituent at the methylene adjacent to the ring nitrogen.
This exercise improved the binding affinity to MDM2 significantly by “directing” the highly
flexible Nꢀsubstituent into the Phe19(p53) pocket. Therefore, compound 12, possessing a directing
group of tꢀbutyl ester in the Nꢀalkyl fragment, was 7–8 times more potent than 11 in both the
biochemical assay (IC50 for 12 of 4.2 nM vs. IC50 for 11 of 34 nM) and the cell proliferation
assay (IC50 for 12 of 0.48 ꢀM vs. IC50 for 11 of 3.4 ꢀM) (Figure 8). The coꢀcrystal structure of
12 bound to MDM2 confirms that the ethyl group reaches into the Phe19 (p53) pocket and the tꢀ
butyl ester projects toward a solventꢀexposed region. In addition, the C5 mꢀCl phenyl group
engages in a faceꢀtoꢀface, πꢀstacking interaction with His96 while it fills the Leu26(p53) pocket
(Figure 6b).
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 34
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 7. Interconversion of two cyclopropylmethyl conformations. MDM2 binding pockets are
labeled (by p53 side chain) in blue.11a
Figure 8. Optimization of potency and in vitro stability via a combination of conformational
control and extensive SAR.11a aIC50 in the HTRF biochemical assay (serum free); bIC50 in the
EdU proliferation assay (SJSAꢀ1, 10% human serum); cKD in the SPR spectroscopy binding
assay.
ACS Paragon Plus Environment
10

Page 11 of 34
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 9. Conformers and relative energies (kcal/mol) for (a) C3 (R)ꢀmonosubstituted and (b)
C3ꢀdisubstituted systems.11a a Relative energy profile for anti and gauche conformations is
calculated by the density functional theory (DFT) B3LYP/6ꢀ31g* method in Jaguar.11a
A closer look at the two aryl groups in the coꢀcrystal structure of 12 indicates that the
transꢀC5 and C6 aryl groups have a gaucheꢀlike orientation in the presence of MDM2 (Figure
6b). However, quantum mechanical calculations on the simple core structure (Nꢀmethyl, (R)ꢀCH3
at C3; Figure 9a) suggest that in the free state, the more stable conformer instead possesses an
antiꢀlike orientation of the C5 and C6 aryls (ꢁE = −0.5 kcal/mol), which is less favorable for
binding. On the other hand, incorporation of a methyl group at the C3 position substantially
destabilizes the undesired anti conformation via a 1,3ꢀsteric interaction with the C5 aryl group,
leading to a dominant gauche conformation (>98%; ꢁE = 2.4 kcal/mol, Figure 9b). This
prediction translated well into potency increase of C3ꢀmethylated analogues.
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 34
1
2
3
4
5
6
7
8
9
Compound 13 was approximately 2–3 times more potent than the corresponding C3 desꢀ
methyl analogue 12 in the HTRF biochemical assay (IC50 for 13 of 2.2 nM vs. IC50 for 12 of 4.2
nM) and the SJSAꢀ1 proliferation assay (IC50 for 13 of 0.19 ꢀM vs. IC50 for 12 of 0.48 ꢀM)
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
(Figure 8). The experimental conformational analysis of 12 and 13, using the H NMR vicinal
coupling constants between the C5 and C6 methine protons at two different temperatures, was
also consistent with the theoretical predictions. Compound 13 was the first piperidinone inhibitor
demonstrating a doseꢀdependent in vivo antitumor efficacy in the SJSAꢀ1 osteosarcoma
xenograft mouse model with an ED50 of 118 mg/kg.
Having identified a core structure with highly populated and wellꢀdefined freeꢀstate
conformation suited for optimal binding, more efforts were made to further enhance potency and
improve pharmacokinetic properties via optimization of the pendent tꢀbutyl ester in 13. As
mentioned, the coꢀcrystal structure of 12 strongly suggests that any replacement of the tꢀbutyl
ester group would project outward to the solvent region and not interact with MDM2. As
predicted, a variety of functional groups is tolerated in this area. Replacing the tꢀbutyl ester in 13
with a hydroxyethyl moiety produced the secondary alcohol 14 with high potency and low
intrinsic clearance in human hepatocytes (Figure 8). In a SPR spectroscopy binding assay, the KD
for 14 and 13 was measured as 0.4 nM and 1.0 nM, respectively. Compound 14 was consistently
more potent than 13 in the HTRF biochemical assay and the SJSAꢀ1 cell proliferation assay as
well (Figure 8). Compound 14 was also evaluated for its ability to inhibit tumor growth in the
SJSAꢀ1 xenograft mouse model. As expected, compound 14 significantly inhibited tumor growth
at 150 and 200 mg/kg QD, compared to the vehicle, with an ED50 of 78 mg/kg. Compound 14
was found to have considerably reduced liability in the human pregnane X receptor (hPXR)
assay, cytochrome P450 (CYP) inhibition assay, and time dependent inhibition (TDI) assay for
ACS Paragon Plus Environment
12

Page 13 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CYP3A4 isoform, as compared to 13. The most significant improvement of 14 over 13 was its
projected human clearance (CL) (0.03 L/h/kg for 14 vs. 0.55 L/h/kg for 13) along with a long
halfꢀlife (>12 h for 14 vs. 0.28 h for 13), which resulted from greatly enhanced intrinsic
clearance in human hepatocytes (3.0 ꢀL/min per 106 cells for 14 vs. 26 ꢀL/min per 106 cells for
13).19 The coꢀcrystal structure of 14 with MDM2 (Figure 10a) confirms that it occupies the three
critical binding pockets of MDM2 in a manner consistent with the previous coꢀcrystal structure
of 12 (Figure 9b). Additionally, the C3 methyl group is found to have no direct interaction with
the protein and project toward the solvent, supporting our proposal that the potency increase
from C3 methylation (13 vs. 12) is conformationally induced and independent of protein
contacts.
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
DISCOVERY OF 1 VIA CAPTURING AN ADDITIONAL INTERACTION WITH
GLY58 “SHELF” REGION15
Further analysis of the coꢀcrystal structure of 14 suggests that exploration of a previously
underutilized region (G58 “shelf” region in Figure 10a) on the MDM2 surface might improve the
binding affinity of an inhibitor towards MDM2. When the secondary alcohol moiety in 14 was
replaced with a methyl sulfonamide, the resultant sulfonamide 15 had slightly improved
biochemical potency but weaker cellular potency (IC50 for 15 of 0.24 ꢀM vs. IC50 for 14 of 0.068
ꢀM), most likely due to the poor permeability (Figure 11). Interestingly, alignment of 15 to coꢀ
crystal structures of 14 and related piperidinone inhibitors illustrates that the methyl moiety in
the sulfonamide side chain lies in the Gly58 “shelf” region bordered on the edge by F55, N59,
and M63 (Figure 10b).
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 34
1
2
3
4
5
6
(a)
(b)
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 10. (a) Coꢀcrystal structure of 14 bound to human MDM2 (17ꢀ111) at 2.0 Å resolution
(green, PDB 4ERF) and depicting proposed electrostatic interaction with G58 “shelf” region
(circled).11a, 15 MDM2 binding pockets labeled (by p53 side chain) in white. MDM2 residues,
H96, G58, M62, N59, and F55 are labeled in yellow. Coꢀcrystallized water molecules are shown
in red. (b) Proposed binding mode of sulfonamide 15.20
G58 "Shelf"
G58 "Shelf"
G58 "Shelf"
G58 "Shelf"(?)
O
O
O
S
HN
O
O
O
O
N
O
N
O
O
N
O
S
O
N
O
O
S
OH
O
O
S
OH
OH
N
OH
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
1
16
Biochemical IC50 0.2 nMa
17
15
Biochemical IC50 0.6 nMa
Cellular IC50
Biochemical IC50 0.1 nMa
Biochemical IC50 0.7 nMa
Cellular IC50
KD (SPR)
0.0091 ꢀMb
0.045 nMc
b
0.24 ꢀMb
N/D
Cellular IC50
KD (SPR)
0.005 ꢀMb
N/D
Cellular IC50
KD (SPR)
0.003
M
ꢀ
0.040 nMc
KD (SPR)
hHep CLint 6.3 ꢀM/min per 106 cells
rHep CLint 10 ꢀM/min per 106 cells
fu in rat plasma 1.2
hHep CLint 16 ꢀM/min per 106 cells
rHep CLint 22 ꢀM/min per 106 cells
fu in rat plasma 1.1
hHep CLint 11 ꢀM/min per 106 cells
6
ꢀ
rHep CLint 21 M/min per 10 cells
fu in rat plasma 0.4
Rat PKd
Rat PKd
Rat PKd
CL (L/h/kg) / t1/2 (h) 0.66 / 2.3
CL (L/h/kg) / t1/2 (h) 1.9 / 1.3
CL (L/h/kg) / t1/2 (h) 0.47 / 3.4
Figure 11. Discovery of 1 via identification and optimization of an additional interaction with
Gly58 shelf region.15 aIC50 in the HTRF biochemical assay (serum free); bIC50 in the EdU
proliferation assay (SJSAꢀ1, 10% human serum); cKD in the SPR spectroscopy binding assay;
dRat PK (iv, 0.5 mg/kg)
ACS Paragon Plus Environment
14

Page 15 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
Once the interaction with this shallow predominantly hydrophobic cleft adjacent to the
Phe19(p53) pocket was optimized within the sulfonamide piperidinone series, we were able to
discover highly potent MDM2 inhibitors such as the cyclopropyl sulfonamide 16 with singleꢀ
digit nanomolar potency in cell proliferation assay (EdU IC50 = 5.3 nM) (Figure 11). However,
these sulfonamide inhibitors were found to have moderate to high in vivo clearance in rat and
relatively high intrinsic clearance in human hepatocytes (i.e. 16 ꢀL/min per 106 cells for 16),
which was used for the human clearance prediction described below in the section of human PK
prediction.19 Encouraged by the increased potency imparted by the sulfonamide functionality, we
set out to explore the less polar and potentially more stable sulfone moiety to retain the favorable
interaction with the glycine shelf and improve the metabolic stability. Gratifyingly, replacement
of the sulfonamide group with a sulfone moiety provided a series of extremely potent MDM2
inhibitor exemplified as 17 in Figure 11. More importantly, many of these compounds maintain
favorable cellular potency and have low intrinsic clearance in human hepatocytes and in vivo
clearance in rat. Further optimization of Nꢀalkyl substituent interacting with the Phe19(p53) pocket
led to the discovery of 1 (Figure 11). With its combination of high potency and excellent
pharmacokinetics, compound 1 was chosen for further development.
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
BINDING MODE15
To understand specific interactions with MDM2, the binding mode of 1 was predicted
from the coꢀcrystal structures of 17 and related analogues (Figure 17).15 Compound 1 occupies
the three critical binding pockets of Leu26(p53), Trp23(p53), and Phe19(p53). The C5 aryl group fills
the Leu26(p53) pocket and engages in a faceꢀtoꢀface πꢀstacking interaction with His96, while the
C6 aryl group reaches deep into the Trp23(p53) binding cavity. The isopropyl group is directed
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 34
1
2
3
4
5
6
7
8
9
into the Phe19(p53) pocket by the conformational constraint induced by the sulfone side chain.
The carboxylate functionality forms a strong hydrogen bond (2.7 Å) with the imidazole on the
His96 side chain of MDM2. In addition, the sulfone moiety properly projects the isopropyl group
to the glycine shelf region to maximize the hydrophobic contact with the neighboring proteins
while the sulfone itself is situated ca. 3.5 Å from the αꢀcarbon of G58, suggesting a CH•••O type
interaction with this residue.
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
Phe19
M62
H96
N59
G58
Trp23
Leu26
F55
Figure 12. Binding mode of 1 based on the coꢀcrystal structures of 17 bound to human MDM2
(17–111)(PDB code: 4OAS).15 MDM2 binding pockets are labeled (by p53 side chain) in white.
MDM2 residues H96, G58, M62, N59, and F55 are labeled in yellow.
Furthermore, superimposition of the single crystal structure of 1 onto the coꢀcrystal
structure of 17 shows that the binding conformation of 1 is essentially identical to the
conformation of 1 crystallized in the absence of protein. This suggests that 1 has the optimal
ACS Paragon Plus Environment
16

Page 17 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
MDM2 binding with minimal conformational cost and the sulfone side chain is bound in the
extended low energy conformation. The favorable interactions and binding conformations of 17
with MDM2 observed in the coꢀcrystal structures greatly contribute to its high binding affinity
toward MDM2.
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 SYNTHESIS OF 115, 21
Asymmetric synthesis of 1 is described in Scheme 1. Overall, it was prepared in 12
linear steps beginning with 3ꢀchlorophenylacetic acid 18.
FriedelꢀCrafts acylation of
chlorobenzene with the acid chloride of 18 gave ketone 19. Conjugate addition of 19 to methyl
methacrylate provided keto ester 20 in 95% yield as a racemic mixture of diastereomers.
Catalytic hydrogenation of 20 in the presence of 0.4% RuCl2[(S)ꢀxylBINAP][(S)ꢀDAIPEN] with
40% sodium tertꢀbutoxide in 2ꢀpropanol afforded compound 21 in 92% ee as a mixture of
diastereomers at Cꢀ2. This dynamic kinetic resolution is the key step in establishing the correct
absolute configuration of the piperidinone core of 1.
The isopropyl ester of 21 was hydrolyzed to carboxylic acid with lithium hydroxide and
cyclized to lactone 22 in the presence of pyridinium pꢀtoluenesulfonate (PPTS) under DeanꢀStark
conditions in 81% overall yield from 20. Lactone 22 was stereoselectively alkylated with allyl
bromide to afford the disubstituted lactone 23 in 94% yield with a diastereomeric ratio of greater
than 99:1. The enantiomeric excess of crystalline lactone 23 could be further improved to greater
than 98% by recrystallization from heptane.
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 34
1
2
Scheme 1.a Asymmetric Synthesis of 1
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
a
Reagents and conditions: (a) chlorobenzene, AlCl3, SOCl2, 89%; (b) methyl methacrylate, tꢀ
BuOK, THF, 95%; (c) 0.4% RuCl2[(S)ꢀxylBINAP][(S)ꢀDAIPEN], tꢀBuONa, 2ꢀpropanol, H2 (50
psi), 45 h; (d) LiOH, THF/MeOH/H2O; (e) PPTS, toluene, reflux, 81% from 20; (f) LiHMDS,
allyl bromide, THF, −30 to −5 °C, 94%; (g) recrystallization from heptane, 80%; (h) LꢀValinol,
neat, 100 °C, 100%; (i) Tf2O, lutidine, −50 °C; (j) LiOH, THF/H2O then recrystallization from
ACS Paragon Plus Environment
18

Page 19 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
cyclohexane, 75% from 24; (k) TsOH, toluene, reflux, 100%; (l) 2ꢀpropanethiol, K2CO3, DMF,
50 °C, and then mCPBA, 0 °C, 90% over 2 steps; (m) RuCl3, NaIO4, EtOAc/CH3CN/H2O, then
recrystallization from EtOH/H2O, 93%.
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
Addition of Lꢀvalinol to lactone 23 resulted in ringꢀopened amide 24 in quantitative yield,
and cyclization of 24 in the presence of trifluoromethanesulfonic anhydride gave intermediate
25. In this reaction, the triflate of the primary alcohol is rapidly displaced by the amide oxygen to
give an oxazoline ring. The nitrogen of the oxazoline then displaces the secondary triflate in a
SN2ꢀtype fashion to give bicyclic oxazolinium triflate 25. The subsequent basic hydrolysis of 25
provided lactam 26 which was purified by column chromatography and recrystallization.
Acidꢀcatalyzed cyclization of 26 resulted in the formation of oxazolinium tosylate 27
which was opened with 2ꢀpropanethiol in the presence of lithium hydroxide. Oxidative workup
afforded sulfone 28 as the penultimate intermediate in 90% yield from 26. Finally, ruthenium
chloride/sodium periodate oxidatively cleaved the terminal alkene to form the desired carboxylic
acid in 93% yield. Recrystallization from ethanol/water provided 1 with greater than 99%
chemical purity. This reliable and scalable route provided multiꢀhundred gram quantities of 1 for
its preclinical studies.
IN VITRO AND IN VIVO PHARMACOLOGY15,22
Compound 1 significantly inhibits the human MDM2ꢀp53 interaction in the biochemical
HTRFꢀbased assay (IC50 = 0.6 nM), but has no inhibition against MDMX at concentrations up to
10 ꢀM. It binds to the human MDM2 in the SPR spectroscopy binding assay with a KD of 0.045
nM. In the EdU proliferation assay, the molecule substantially suppresses growth of MDM2ꢀ
amplified SJSAꢀ1 osteosarcoma tumor cells (IC50 = 9.1 nM). Furthermore, compound 1 potently
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 34
1
2
3
4
5
6
inhibits proliferation of nonꢀMDM2ꢀamplified HCT116 colorectal cells in the BrdU assay (IC50
= 10 nM).
7
8
9
The cellular selectivity of 1 was evaluated by examining its effect on the proliferation of
cultured HCT116 wildꢀtype p53 (p53wt) and p53ꢀdeficient (p53−/−) tumor cells. As shown in
Figure 13, the molecule significantly inhibited the growth of wildꢀtype p53 cells (IC50 = 10 nM)
in the BrdU proliferation assay, but had no growth inhibition of p53ꢀdeficient cells at
concentrations up to 10 ꢀM (IC50 > 25 ꢀM). The cellular selectivity between the deficient and
wildꢀtype p53 lines was greater than 2500ꢀfold, indicating that 1 has p53 dependent cell activity
exclusively.
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 13. Cell activity of 1 is p53 dependent. In HCT116 p53wt and p53−/− cells, the percentage
of BrdU positive cells was measured 16 h postꢀcompound treatment by flow cytometry. DMSO
control was designated as 0% inhibition.
To understand the effect of 1 on p53 pathway signaling in vivo, we measured levels of
p21 mRNA induction, a transcriptional target and pharmacodynamic readout for p53 activity.23
Treatment of SJSAꢀ1 led to robust induction of p21 over 24 hours, with a peak p21 induction of
ACS Paragon Plus Environment
20

Page 21 of 34
Journal of Medicinal Chemistry
1
2
3
4
30ꢀfold occurring at 4 hours post dose (Figure 14). This confirms that 1 effectively activates the
5
6
p53 signaling in human tumors in vivo.
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 14. Pharmacodynamic study: treatment with 1 caused time and concentration dependent
induction of p21 mRNA in SJSAꢀ1 tumor xenografts. *p<0.05. Female athymic nude mice were
implanted subcutaneously with 5 × 106 SJSAꢀ1 cells. When tumors reached ~175 mm3, 50 mg/kg
of 1 or vehicle was administered orally once daily (QD) for 4 days. Mice were sacrificed on day
4 at 1, 2, 4, 8, and 24 hours postꢀdose (n = 5/group). Tumors were immediately removed and
snapꢀfrozen. p21 mRNA levels were measured by quantitative RTꢀPCR. Tumors treated with
vehicle served as a negative control and indicated the baseline p21 mRNA level. Data are
represented as mean p21 fold induction over vehicle and error bars represent SEM of data from
five mice. Concentrations in plasma were analyzed by LC/MS/MS, and are plotted against the
right yꢀaxis.
To assess the ability of 1 to inhibit tumor growth in vivo, efficacy studies were performed
in xenograft models of wildꢀtype p53 human cancer cells. In the MDM2ꢀamplified SJSAꢀ1
osteosarcoma model, compound 1 displayed robust tumor growth inhibition compared to the
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 34
1
2
3
4
5
6
7
8
9
vehicle, with an ED50 of 9.1 mg/kg QD (Figure 15). Unbound EC50 associated with ED50 was
calculated as 2.8 nM.24 Regression was observed with the two higher doses of 30 mg/kg and 60
mg/kg QD. The 60 mg/kg of dose caused complete tumor regression in 10 of 12 mice. No body
weight loss was observed with any of the treatment groups.
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 15. * p<0.05. Treatment with 1 inhibited the growth of SJSAꢀ1 tumors in vivo. SJSAꢀ1
cells (5 × 106) were implanted subcutaneously into female athymic nude mice. Treatment with
vehicle or 1 at 7.5, 15, 30 or 60 mg/kg QD by oral gavage began on day 11 when tumors had
reached ~200 mm3 (n=12/group). Tumor sizes were measured twice per week. Data represent
mean tumor volumes and the error bars represent SEM of data from 12 mice.
Compound 1 was also evaluated in the nonꢀMDM2ꢀamplified HCTꢀ116 colorectal
carcinoma xenograft model. Treatment with 1 twice a day dosing (BID) caused a doseꢀdependent
tumor growth inhibition with an ED50 of 16 mg/kg (Figure 16). The highest dose of 100 mg/kg
BID resulted in tumor stasis (100% TDI) and no body weight loss was observed with any of the
treatment groups. Tumor growth inhibition with QD dosing was comparable to that of BID
dosing at all dose levels.22
ACS Paragon Plus Environment
22

Page 23 of 34
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 16. * p<0.0001. Treatment with 1 inhibited the growth of HCTꢀ116 tumors in vivo. HCTꢀ
116 colorectal carcinoma cells (5 × 106) were implanted subcutaneously into female athymic
nude mice. Treatment with vehicle or 1 at 25, 50 or 100 mg/kg BID dosing by oral gavage began
on day 11 when tumors had reached ~200 mm3 (n=10/group). Tumor sizes were measured twice
per week. Data represent mean tumor volumes and the error bars represent SEM of data from 10
mice.
DRUG METABOLISM25
The metabolism of 1 labeled with carbonꢀ14 (Figure 17) was studied in hepatocytes from
rat, dog, monkey, and human. Glucuronidation was found to be the primary metabolic pathway
in hepatocytes from all species including human. The formation rate of acyl glucuronide (M1 in
Figure 17) was determined to be in the order of dog > monkey > human > rat hepatocytes, which
was consistent with the published data.26 In an in vivo metabolism study, about 8% of
administered dose was recovered as unchanged parent molecule in rat bile and feces over a
period of 24 hours when [14C]ꢀ1 was orally dosed in bile cannulated rats. In this study, acyl
glucuronide was identified as a major metabolite along with two monohydroxyl metabolites. In
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 34
1
2
3
4
vitro and in vivo data suggest that metabolism is the major route of clearance and
5
6
glucuronidation is the major metabolic pathway for 1.
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 17. Structures of [14C]ꢀ1 and major metabolite acyl glucuronide (M1).
PRECLINICAL PHARMACOKINETICS AND HUMAN PK PREDICTION25
The preclinical pharmacokinetics of 1 were investigated in mouse, rat, cynomolgus
monkey, and beagle dog following singleꢀdose intravenous (IV) or oral administration. The
molecule was found to have a low clearance and good oral bioavailability in mouse, rat and
cynomolgus monkey, but high clearance and low oral bioavailability in beagle dog. Given the
different pharmacokinetic properties of 1 in the preclinical species, in vitro−in vivo extrapolation
of hepatocyte stability data19 was explored to predict the human clearance. As a result, good in
vivoꢀin vitro correlation was observed in preclinical species when the hepatic intrinsic clearance
values from preclinical species were scaled up to predict in vivo clearance. Using this method, 1
was predicted to have low clearance and a long halfꢀlife in humans.
SAFETY ASSESSMENT
ACS Paragon Plus Environment
24

Page 25 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Possible effects of 1 on enzyme activity and binding to receptors, were evaluated in the
CEREP in vitro pharmacology assay panel. It was found to have no inhibition on any control
specific binding at a concentration of 10 µM. In the ScanMax profiling assay, compound 1 was
associated with only one hit out of 392 nonꢀmutated kinases tested (protein kinase D2 (PRKD2),
69% inhibition at 10 ꢀM). These data indicate that 1 is a highly selective inhibitor of MDM2.
The cardiovascular safety profile of 1 was assessed through a combination of in vitro (hERG
assay and isolated rabbit heart assay) and in vivo studies (rat telemetry study). These studies
indicated no cardiovascular liability at exposure or concentration levels significantly higher than
pharmacologically active exposures and concentrations. For example, in a singleꢀdose rat
cardiovascular telemetry study, no significant changes in cardiovascular variables were detected
up to 250 mg/kg.
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
Compound 1 was a weak competitive inhibitor of CYP2C8 with an IC50 of 8.5 ꢁM and
not a potent inhibitor of CYP1A2, 2C9, 2C19, 2D6, 2E1 or 3A4, with IC50 of >30 ꢁM, indicating
low potential to affect clearance of coꢀadministrated drugs by competitive inhibition of human
CYP isoforms. PXR activation by 1 in vitro was evaluated using a PXR reporter gene assay. At 2
and 10 µM, 1 activated human PXR at 18 and 21% of positive control rifampicin at 10 µM,
respectively. CYP induction assay was carried out with human hepatocytes from donors, the
changes in mRNA level and activity of CYP3A4, CYP2B6, and CYP1A2 were minor at 1 µM of
1. Overall potential for 1 to increase clearance of CYP3A4 substrates is considered low due to
the relatively low exposure at a projected clinically efficacious dose. In vivo nonclinical
toxicology profile of 1 was evaluated in the preclinical species where no safety findings
precluded progression of 1 to clinical development.
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 34
1
2
3
4
CONCLUSION
5
6
7
8
Combination of de novo design, Xꢀray crystallography, molecular modeling, and iterative
medicinal chemistry, resulted in the discovery of piperidinone MDM2 inhibitors. Further
optimization of this series through conformational controls, extensive SAR studies, and
capturing an additional interaction with Gly58 “shelf” region led to the discovery of 1, which is a
highly potent MDM2 inhibitor in clinical trials. Compound 1 has favorable in vitro and in vivo
toxicology profiles. Moreover, this molecule is projected to have low clearance and a long halfꢀ
life in humans.
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
AUTHOR INFORMATION
Corresponding Authors
* Y.R.: (address) 1120 Veterans Blvd., South San Francisco, CA 94080; (phone, office) 650ꢀ
244ꢀ2682; (eꢀmail) yrew@amgen.com
* D.S.: (address) 1120 Veterans Blvd., South San Francisco, CA 94080; (phone, office) 650ꢀ244ꢀ
2195; (eꢀmail) daqings@amgen.com
ABBREVIATIONS USED
BID, twice a day dosing; BrdU, 5ꢀbromoꢀ2ꢀdeoxyuridine; CL, clearance; CYP, Cytochrome
P450; DMF, N,Nꢀdiemethylformamide; DMSO, dimethylsulfoxide; dr, diastereoselectivity ratio;
EC50, half maximal effective concentration; ED50, median effective dose; EdU, 5ꢀethynylꢀ2´ꢀ
deoxyuridine; ee, enantiomeric excess; er, enantioselectivity ratio; EtOAc, ethyl acetate; EtOH,
ethanol; fu, fraction unbound; hPXR, Human Pregnane X Receptor; HTRF, homogeneous timeꢀ
resolved fluorescence; HTS, Highꢀthroughput screening; IC50, half maximal inhibitory
ACS Paragon Plus Environment
26

Page 27 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
concentration; IV, intravenous; LiHMDS, lithium bis(trimethylsilyl)amide; MeOH, Methanol;
MDM2, murine double minute 2; NMR, Nuclear magnetic resonance; PPTS, pyridinium pꢀ
toluenesulfonate; PK, pharmacokinetics; PXR, pregnane X receptor; QD, once a day dosing; RTꢀ
PCR, reverse transcription polymerase chain reaction; SAR, structure activity relationship; SEM,
standard error of mean; SN2, substitution nucleophilic (biꢀmolecular); SPR, Surface Plasmon
Resonance; tꢀBuOK, potassium tertꢀbutoxide; tꢀBuONa, sodium tertꢀbutoxide; TDI, time
dependent inhibition; Tf2O, Trifluoromethanesulfonic anhydride; THF, tetrahydrofuran; TsOH,
4ꢀtoluenesulfonic acid.
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
REFERENCES
(1) Vazquez, A.; Bond, E. E.; Levine, A. J.; Bond, G. L., The genetics of the p53 pathway,
apoptosis and cancer therapy. Nat. Rev. Drug Discovery 2008, 7, 979ꢀ987.
(2) Vogelstein, B.; Lane, D.; Levine, A. J., Surfing the p53 network. Nature (London) 2000, 408,
307ꢀ310.
(3) (a) Hollstein, M.; Sidransky, D.; Vogelstein, B.; Harris, C. C., p53 Mutations in human
cancers. Science (Washington, D.C.) 1991, 253, 49ꢀ53; (b) Soussi, T.; Dehouche, K.; Beroud, C.,
p53 Webꢀsite and analysis of p53 gene mutations in human cancer: forging a link between
epidemiology and carcinogenesis. Hum. Mutat. 2000, 15, 105ꢀ113.
(4) (a) Chene, P., Inhibiting the p53ꢀMDM2 interaction: An important target for cancer therapy.
Nature Rev. Cancer 2003, 3, 102ꢀ109; (b) Oliner, J. D.; Kinzler, K. W.; Meltzer, P. S.; George,
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 34
1
2
3
4
D. L.; Vogelstein, B., Amplification of a gene encoding a p53ꢀassociated protein in human
5
6
sarcomas. Nature (London) 1992, 358, 80ꢀ83.
7
8
9
(5) (a) Momand, J.; Zambetti, G. P.; Olson, D. C.; George, D.; Levine, A. J., The MDMꢀ2
oncogene product forms a complex with the p53 protein and inhibits p53ꢀmediated
transactivation. Cell 1992, 69, 1237ꢀ1245; (b) Oliner, J. D.; Pietenpol, J. A.; Thiagalingam, S.;
Gyuris, J.; Kinzler, K. W.; Vogelstein, B., Oncoprotein MDM2 conceals the activation domain
of tumor suppressor p53. Nature (London) 1993, 362, 857ꢀ860.
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) (a) Roth, J.; Dobbelstein, M.; Freedman, D. A.; Shenk, T.; Levine, A. J., Nucleoꢀcytoplasmic
shuttling of the hdm2 oncoprotein regulates the levels of the p53 protein via a pathway used by
the human immunodeficiency virus rev protein. EMBO J. 1998, 17, 554ꢀ564; (b) Tao, W.;
Levine, A. J., Nucleocytoplasmic shuttling of oncoprotein Hdm2 is required for Hdm2ꢀmediated
degradation of p53. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 3077ꢀ3080.
(7) (a) Haupt, Y.; Maya, R.; Kazaz, A.; Oren, M., MDM2 promotes the rapid degradation of p53.
Nature (London) 1997, 387, 296ꢀ299; (b) Kubbutat, M. H. G.; Vousden, k. H., Proteolytic
cleavage of human p53 by calpain: a potential regulator of protein stability. Mol. Cell. Biol.
1997, 17, 460ꢀ468.
(8) Vassilev, L. T., MDM2 inhibitors for cancer therapy. Trends Mol. Med. 2007, 13, 23ꢀ31.
(9) (a) Vassilev, L. T.; Vu, B. T.; Graves, B.; Carvajal, D.; Podlaski, F.; Filipovic, Z.; Kong, N.;
Kammlott, U.; Lukacs, C.; Klein, C.; Fotouhi, N.; Liu, E. A., In vivo activation of the p53
pathway by smallꢀmolecule antagonists of MDM2. Science (Washington, D.C.) 2004, 303, 844ꢀ
848; (b) Tovar, C.; Rosinski, J.; Filipovic, Z.; Higgins, B.; Kolinsky, K.; Hilton, H.; Zhao, X.;
ACS Paragon Plus Environment
28

Page 29 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Vu, B. T.; Qing, W.; Packman, K.; Myklebost, O.; Heimbrook, D. C.; Vassilev, L. T., Smallꢀ
molecule MDM2 antagonists reveal aberrant p53 signaling in cancer: implications for therapy.
Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 1888ꢀ1893.
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
(10) (a) Ding, K.; Lu, Y.; NikolovskaꢀColeska, Z.; Qiu, S.; Ding, Y.; Gao, W.; Stuckey, J.;
Krajewski, K.; Roller, P. P.; Tomita, Y.; Parrish, D. A.; Deschamps, J. R.; Wang, S., Structureꢀ
based design of potent nonꢀpeptide MDM2 inhibitors. J. Am. Chem. Soc. 2005, 127, 10130ꢀ
10131; (b) Ding, K.; Lu, Y.; NikolovskaꢀColeska, Z.; Wang, G.; Qiu, S.; Shangary, S.; Gao, W.;
Qin, D.; Stuckey, J.; Krajewski, K.; Roller, P. P.; Wang, S., Structureꢀbased design of spiroꢀ
oxindoles as potent, specific smallꢀmolecule inhibitors of the MDM2ꢀp53 interaction. J. Med.
Chem. 2006, 49, 3432ꢀ3435; (c) Shangary, S.; Qin, D.; McEachern, D.; Liu, M.; Miller, R. S.;
Qiu, S.; NikolovskaꢀColeska, Z.; Ding, K.; Wang, G.; Chen, J.; Bernard, D.; Zhang, J.; Lu, Y.;
Gu, Q.; Shah, R. B.; Pienta, K. J.; Ling, X.; Kang, S.; Guo, M.; Sun, Y.; Yang, D.; Wang, S.,
Temporal activation of p53 by a specific MDM2 inhibitor is selectively toxic to tumors and leads
to complete tumor growth inhibition. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 3933ꢀ3938; (d)
Yu, S.; Qin, D.; Shangary, S.; Chen, J.; Wang, G.; Ding, K.; McEachern, D.; Qiu, S.;
NikolovskaꢀColeska, Z.; Miller, R.; Kang, S.; Yang, D.; Wang, S., Potent and orally active
smallꢀmolecule inhibitors of the MDM2ꢀp53 interaction. J. Med. Chem. 2009, 52, 7970ꢀ7973.
(11) (a) Rew, Y.; Sun, D.; GonzalezꢀLopez De Turiso, F.; Bartberger, M. D.; Beck, H. P.;
Canon, J.; Chen, A.; Chow, D.; Deignan, J.; Fox, B. M.; Gustin, D.; Huang, X.; Jiang, M.; Jiao,
X.; Jin, L.; Kayser, F.; Kopecky, D. J.; Li, Y.; Lo, M.ꢀC.; Long, A. M.; Michelsen, K.; Oliner, J.
D.; Osgood, T.; Ragains, M.; Saiki, A. Y.; Schneider, S.; Toteva, M.; Yakowec, P.; Yan, X.; Ye,
Q.; Yu, D.; Zhao, X.; Zhou, J.; Medina, J. C.; Olson, S. H., Structureꢀbased design of novel
inhibitors of the MDM2ꢀp53 interaction. J. Med. Chem. 2012, 55, 4936ꢀ4954; (b) Gonzalezꢀ
ACS Paragon Plus Environment
29