Design of chemically stable, potent, and efficacious MDM2 inhibitors that exploit the retro-mannich ring-opening-cyclization reaction mechanism in spiro-oxindoles

Article abstract of DOI:10.1021/jm501541j

Inhibition of the MDM2-p53 protein-protein interaction is being actively pursued as a new anticancer therapeutic strategy, and spiro-oxindoles have been designed as a class of potent and efficacious small-molecule inhibitors of this interaction (MDM2 inhibitors). Our previous study showed that some of our first-generation spiro-oxindoles undergo a reversible ring-opening-cyclization reaction that, from a single compound in protic solution, results in an equilibrium mixture of four diastereoisomers. By exploiting the ring-opening-cyclization reaction mechanism, we have designed and synthesized a series of second-generation spiro-oxindoles with symmetrical pyrrolidine C2 substitution. These compounds undergo a rapid and irreversible conversion to a single, stable diastereoisomer. Our study has yielded compound 31 (MI-1061), which binds to MDM2 with K_i = 0.16 nM, shows excellent chemical stability, and achieves tumor regression in the SJSA-1 xenograft tumor model in mice.

Full text of DOI:10.1021/jm501541j

Article
pubs.acs.org/jmc
Design of Chemically Stable, Potent, and Efficacious MDM2
Inhibitors That Exploit the Retro-Mannich Ring-Opening-Cyclization
Reaction Mechanism in Spiro-oxindoles
Angelo Aguilar,†,§ Wei Sun,†,§ Liu Liu,† Jianfeng Lu,† Donna McEachern,† Denzil Bernard,†
Jeffrey R. Deschamps,‡ and Shaomeng Wang*,†
†University of Michigan Comprehensive Cancer Center and Departments of Internal Medicine, Pharmacology, and Medicinal
Chemistry, University of Michigan, 1500 East Medical Center Drive, Ann Arbor, Michigan 48109, United States
‡Naval Research Laboratory, Code 6930, 4555 Overlook Avenue, Washington, D.C. 20375, United States
S
* Supporting Information
ABSTRACT: Inhibition of the MDM2−p53 protein−protein interaction is being actively pursued as a new anticancer
therapeutic strategy, and spiro-oxindoles have been designed as a class of potent and efficacious small-molecule inhibitors of this
interaction (MDM2 inhibitors). Our previous study showed that some of our first-generation spiro-oxindoles undergo a
reversible ring-opening-cyclization reaction that, from a single compound in protic solution, results in an equilibrium mixture of
four diastereoisomers. By exploiting the ring-opening-cyclization reaction mechanism, we have designed and synthesized a series
of second-generation spiro-oxindoles with symmetrical pyrrolidine C2 substitution. These compounds undergo a rapid and
irreversible conversion to a single, stable diastereoisomer. Our study has yielded compound 31 (MI-1061), which binds to
MDM2 with Ki = 0.16 nM, shows excellent chemical stability, and achieves tumor regression in the SJSA-1 xenograft tumor
model in mice.
MDM2 and p53 blocks the binding of p53 to targeted DNAs
and transports p53 from the nucleus to the cytoplasm,
rendering p53 ineffective as a transcriptional factor. Con-
sequently, blocking the MDM2−p53 interaction with small-
molecule inhibitors can reactivate the tumor suppressor
function of wild-type p53, and this approach is being pursued
as a new cancer therapeutic strategy.12−17
Using a structure-based approach, our laboratory has
designed and synthesized a spiro-oxindole (1, Figure 1) as an
inhibitor of the MDM2−p53 interaction (MDM2 inhibitor).18
Subsequently, potent and efficacious MDM2 inhibitors in this
family were obtained through extensive optimization,19−22 and
one such compound (SAR405838/MI-77301)23 has been
advanced into clinical development.
In the course of our research, it was discovered that, in protic
solutions, some of the spiro-oxindoles are converted sponta-
neously into four diastereoisomers (Figure 2) which exist in
equilibrium with one another.24 We recently reported a study
of this phenomenon with compound 3 and its analogues
(Figure 1),22,24 and the Roche group, using a different synthetic
INTRODUCTION
■
The powerful tumor suppressor p53 is a transcriptional factor
and plays a key role in preventing tumor development. It is
therefore not surprising that p53 function is compromised in
most if not all human cancers. In approximately 50% of human
cancers, the gene encoding p53 protein is mutated or deleted,
which results in inactivation of its transcriptional activity and
tumor suppressor function.1 In the other half of human cancers,
p53 retains its wild-type status, but its function is inhibited by a
variety of mechanisms.2 One major inhibitory mechanism is
through the direct interaction between p53 and the human
murine double-minute 2 (MDM2) protein.3−8 Overexpression
of MDM2 by either gene amplification or other mechanisms
has been observed in different types of human cancers.9
Furthermore, it has been observed that MDM2 gene
amplification and p53 mutation are mutually exclusive in
human cancers, highlighting a prominent role of MDM2 in
suppressing p53 function.10,11
MDM2 inhibits wild-type p53 function by several mecha-
nisms, which are distinct, but all are mediated through their
direct binding.6 Upon binding, MDM2 ubiquitinates p53 by
functioning as an E3 ligase and promotes proteasomal
degradation of p53. Additionally, the interaction between
Received: October 6, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
IV, Figure 2), which then remain at equilibrium in solution.
After equilibration, the major diastereoisomer was determined
to have configuration IV, in which all the large substituents on
the pyrrolidine ring are trans to one another (Figure 2). This
diastereoisomer IV was isolated and shown to be the most
stable and most biologically active of the diastereoisomers as
MDM2 inhibitors.24
In this paper we report the design, synthesis, and evaluation
of a series of new spiro-oxindoles that exploit the ring-opening-
cyclization mechanism to obtain potent and chemically stable
MDM2 inhibitors. Our study led to the discovery of 31 (MI-
1061), which has excellent stability in solution and displays a
high binding affinity (Ki = 0.16 nM) to MDM2. Significantly,
31 is orally bioavailable and achieves tumor regression in an
SJSA-1 xenograft model in mice.
DESIGN AND CHEMISTRY
■
Figure 1. Previously reported spiro-oxindoles as inhibitors of MDM2−
p53 interaction.
Our study started with the analysis of the proposed mechanism
in Figure 2, in which C2 and C3 of the pyrrolidine ring are
directly involved in and affected by the ring-opening-cyclization
reactions that underlie the isomerization. Because C2 and C3
are both chiral, the mechanism dictates the formation of four
diastereoisomers.16 We initially rationalized that the presence of
two identical substituents at C2 would reduce the number of
diastereoisomers from four to two (Figure 3). We further
predicted that making C2 symmetrical in this way would cause
compounds with cis-configuration to reconfigure themselves
through the retro-Mannich reaction, forming compounds with
trans-configuration (Figure 3). Because compounds with trans-
configuration experience less steric repulsion between the two
bulky phenyl rings, they would be more stable than the
corresponding compounds with cis-configuration and, judging
by our previous binding data, may have superior potency for
binding to MDM2 (Figure 3).
Accordingly, we designed a series of spiro-oxindoles with two
identical substituents at C2 of the pyrrolidine. These included
2,2-dialkyl substituents of various sizes and 2-spirocycloalkyl
substituents with different sized cycloalkyl groups. To
synthesize these new compounds, we applied the same strategy
used to obtain our first-generation spiro-oxindoles.18,22 The key
step, building the spiro-oxindole scaffold, was the asymmetric
1,3-dipolar cycloaddition reaction illustrated in Scheme 1. It
was discovered that the dimethyl ketal derivatives of acetone, 3-
pentanone, or 4-heptanone (8−10) were necessary for the
cycloaddition reactions that produced scaffolds 12A−14A.
Unlike these acyclic ketones, the cyclic ketones undergo the
cycloaddition reaction without prior conversion to their
corresponding dimethyl ketals. Scaffolds 15A−18A were
produced in this way. However, synthesis of scaffold 19A
using either cyclo-octanone or its dimethyl ketal derivative 11
as the substrate of the cycloaddition reaction was unsuccessful,
Figure 2. Proposed isomerization mechanism of spiro-oxindoles.
strategy, also observed the same isomerization in their
preparation of compound 5 (Figure 1).25 Furthermore, it is
likely that this isomerization accounts for the reported
observation of other spiro-oxindole diastereoisomers in co-
crystal structures with MDM2.26−28
The proposed mechanism for the isomerization (Figure 2)
involves a ring-opening retro-Mannich reaction between C2
and C3 of the pyrrolidine ring, generating the transition
intermediate TS.22,25 Reconfiguration of the C2 and C3
pyrrolidine substituents and a subsequent Mannich reaction
cyclization can generate any of the four diastereoisomers (I−
Figure 3. Design of second-generation spiro-oxindoles with symmetrical substituents at the C2 position.
B
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Scheme 1. Asymmetric 1,3-Dipolar Cycloaddition Reaction
Scheme 2. Synthesis of Second-Generation Spiro-oxindoles
presumably due to the steric hindrance associated with the
larger cyclo-octyl ring. The ring-opening of 12A−18A with
trans-4-aminocyclohexanol and oxidative removal of the chiral
auxiliary (1,2-diphenylethanol) on the pyrrolidine nitrogen
from the resulting products (12B−18B) produced the target
molecules (Scheme 2).
Table 1. Ratio of Diastereoisomers after Removal of the
Chiral Auxiliary
RESULTS AND DISCUSSION
■
In the first-generation spiro-oxindoles,22 four diastereoisomers
could be detected after the oxidative removal of the chiral
auxiliary, but in these second-generation spiro-oxindoles, only
two diastereoisomers are possible because only one of the
carbons (C3) involved in the ring-opening-cyclization reaction
remains chiral. Analysis with ultra-performance liquid chroma-
tography (UPLC) of the crude products of the removal of the
chiral auxiliary from compounds 12B−18B (Scheme 2)
revealed a fascinating trend. In the case of the acyclic
pyrrolidine-2,2-dialkyl-substituted spiro-oxindoles (Table 1,
entries 1−3), only one major compound was detected, and it
was determined by NMR comparison (Supporting Information
(SI)) to be the stable diastereoisomer, 12D−14D.
In the case of the pyrrolidine-2-spirocycloalkyl-substituted
spiro-oxindoles (Table 1, entries 4−7), both C- and D-
diastereoisomers were detected. As the ring size increased, the
detected amount of the unstable C-diastereoisomers decreased
from 90% in the case of the cyclobutyl compound 15C to 15%,
23%, and 0% in the cyclopentyl 16C, cyclohexyl 17C, and
cycloheptyl 18C compounds, respectively.
Next we examined the effect of varying the carboxamide
substituent. The pyrrolidine-2,2-diethyl-, pyrrolidine-2-spiro-
cyclobutyl-, and pyrrolidine-2-spirocyclohexylspiro-oxindole
scaffolds were selected to further investigate the stability
profiles of these second-generation spiro-oxindoles. The N-(2-
morpholinoethyl)carboxamide, (S)-N-(3,4-dihydroxybutyl)-
C
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
carboxamide, and N-((1S,3S)-3-hydroxy-3-methylcyclobutyl)-
carboxamide substituents were selected because they were used
in previously published first-generation spiro-oxindole MDM2
inhibitors.18−22,24 Combination of the second-generation
scaffolds and the representative carboxamide substituents
produced three series of compounds (21−23, 24−26, and
27−29), shown in Table 2. After removal of the chiral auxiliary
Table 3. Stability of Diastereoisomers of 12−18 in Solution
in MeOH/H2O with 0.1% TFA
Table 2. Ratio of Diastereoisomers after Removal of the
Chiral Auxiliary for Compounds with Varying Carboxamide
Substituents
ratio after HPLC C:D:E
entry
compound
0 h
1 days
2 days
1
12D
0:100:0
0:99:1
0:100:0
0:97:3
0:100:0
0:94:6
2
13D
3
14D
0:99.4:0.6
99.7:0.3:0
2:98:0
0:90:10
96:4:0
0:77:23
93:7:0
4
15C
5
15D
3:97:0
2:98:0
6
16C+16D
16D
73:27:0
0:100:0
31:69:0
0:100:0
0:100:0
0:100:0
0:100:0
0:100:0
0:100:0
0:91:9
0:100:0
0:100:0
0:100:0
0:100:0
0:83:17
7
8
17C+17D
17D
9
10
18D
Table 4. Stability of Diastereoisomers 21−29 in MeOH/H2O
Solution with 0.1% TFA
from 21B−29B, the same trend was observed in each series.
The acyclic pyrrolidine-2,2-diethyl spiro-oxindoles (Table 2,
entries 1, 4, and 7) showed only the stable D-diastereoisomers
in the crude products; the pyrrolidine-2-cyclobutyl compounds
(Table 2, entries 2, 5, and 8) contained the unstable C-
diastereoisomer as the major component in the crude reaction
product, while the pyrrolidine-2-cyclohexyl compounds (Table
2, entries 3, 6, and 9) had the stable D-diastereoisomer as the
major component in the crude product. This trend is consistent
with the results shown in Table 1 for compounds 13, 15, and
17.
To evaluate the stability of the individual products, their
composition was analyzed at different time points after
preparative HPLC purification. The composition relative to
that observed immediately after purification was recorded, and
the data are shown in Tables 3 and 4.
Consistent with the data shown in Tables 1 and 2 obtained
from the crude products, only the D-diastereoisomers were
obtained for the acyclic compounds. However, as the size of the
alkyl substituents increased from dimethyl to diethyl and di-n-
propyl, a second compound (0%, 1%, and 0.6%, respectively)
was detected (Table 3, 0 h, entries 1−3). This did not represent
the unstable C-diastereoisomers but rather was found to be,
after isolation and characterization (SI), the ring-opened
primary amine E (20) formed as shown in Table 3. After 2
ratio after HPLC C:D:E
entry
compound
0 h
1 days
2 days
1
21D
0:98:0.3
96:4:0
0:93:5
92:8:0
0:89:9
89:11:0
0:100:0
0:99.3:0.7
0:99.8:0.2
0:97:3
2
22C
3
22D
0:100:0
53:47:0
0:100:0
0:100:0
100:0:0
0:100:0
35:65:0
0:100:0
0:100:0
99.7:0.3:0
0:100:0
32:68:0
0:100:0
0:100:0
0:99.3:0.7
0:100:0
0:98:2
4
23C+23D
23D
5
6
24D
7
25C
98:2:0
97:3:0
8
25D
0:100:0
0:100:0
0:100:0
0:98:2
0:100:0
0:100:0
0:100:0
0:96:4
9
26C+26D
26D
10
11
12
13
14
15
27D
28C
98:2:0
89:11:0
0:100:0
0:100:0
0:100:0
28D
0:100:0
0:100:0
0:100:0
29C+29D
29D
days in solution the difference in stability became more evident;
the composition of the dimethyl compound 12D remained
unchanged, while 6% and 23% of the ring-opened-elimination
D
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 4. Crystal structure of intermediate with starting cis-configuration and product having the more stable trans-configuration.
product 20 was observed in the solution of the diethyl (13D)
and di-n-propyl (14D) compounds, respectively (Table 3).
Similar ring-opened-elimination products were observed for the
diethyl compounds (21D, 24D, and 27D) with different
carboxamide substituents (Table 4). This indicates that the rate
of decomposition to the ring-opened-elimination product E is
proportional to the size of the alkyl substituent, indicating that
large C2 dialkyl substituents introduce steric hindrance that
slows the recyclization and allows for hydrolysis of the
pyrrolidine-ring-opened imine intermediate (TS in Figure 3)
to yield E.
As indicated above, both diastereoisomers were observed in
the crude reaction product from the C2-spirocycloalkyl
compounds (Table 1, entries 4−7; Table 2, entries 2-3, 5-6,
and 8-9), and we presumed they could be isolated in pure form.
However, pure C-diastereoisomers 15C, 22C, 25C, and 28C,
could be isolated only in the case of the cyclobutyl compounds
(Table 3, entry 4; Table 4, entries 2, 7, and 12). Isolation of the
C-diastereoisomers of the cyclopentyl 16C and cyclohexyl 17C,
23C, 26C, and 29C compounds was unsuccessful. Fractions
collected for the peaks corresponding to the diastereoisomers
16C, 17C, 23C, 26C, and 29C were immediately analyzed by
UPLC and revealed to be mixtures which already showed 27%,
69% (Table 3, entries 6 and 8), 47%, 65%, and 68% (Table 4,
entries 4, 9, and 14) conversion to their corresponding 16D,
17D, 23D, 26D, and 29D diastereoisomers. Isolation of the D-
diastereoisomers with >98% purity (15D−18D, 22D−23D,
25D−26D, 28D−29D) was achieved for all the spirocycloalkyl
compounds (Table 3, entries 5, 7, 9, and 10; Table 4, entries 3,
5, 8, 10, 13, and 15).
cyclohexyl 17C, 23C, 26C, and 29C diastereoisomer fractions
(Table 3, entries 6 and 8; Table 4, entries 4, 9, and 14) had
been completely converted to their corresponding diastereo-
isomers 16D, 17D, 23D, 26D, and 29D. The 15C, 22C, 25C,
and 28C diastereoisomers of the cyclobutyl compounds,
however, proved to be quite stable, showing only 7%, 11%,
3%, and 11% conversion to their corresponding 15D, 22D,
25D, and 28D diastereoisomers after 2 days (Table 3, entry 4;
Table 4, entries 2, 7, and 12). None of the pure D-
diastereoisomers of the cyclic compounds (15D−17D, Table
3, entries 5, 7, and 9; 22D−23D, 25D−26D, 28D−29D, Table
4, entries 3, 5, 8, 10, 13, and 15) revealed any detectable
conversion to the corresponding C-diastereoisomers, and they
maintained >98% purity after 2 days in solution, with the
exception of the cycloheptyl compound 18D (Table 3, entry
10). This compound proved to be unstable, showing 9% and
17% decomposition, respectively, to compound 20 after 1 and 2
days in solution. These results show that no equilibrium exists
between the C- and D-diastereoisomers for these second-
generation spiro-oxindoles and that, if the carbon-2 substituents
are too bulky, the ring-closure reaction simply becomes slower,
resulting in hydrolysis of the iminium intermediate that leads to
compound 20.
To support the proposed cis (C-) and trans (D-
)configurations, we determined the crystal structures of 17B
and 30. The intermediate 17B contains the cis-configuration,
while 30 contains the trans-configuration, and both produce
17D with the trans-configuration. Equivalent NMR data were
obtained for 17D prepared from 17B/C or 30 (Figure 4).
We next directly compared the stability, in solution over 6
days, of representative first-generation and second-generation
spiro-oxindoles (MI-6324 vs 23D, MI-14724 vs 26D, and 3
(MI-888)24 vs 26D) (Table 5). The first-generation spiro-
To assess the possibility of equilibrium between the C- and
D-diastereoisomers, we analyzed the HPLC fractions over 2
days. After 1 day in solution, the cyclopentyl 16C and
E
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 5. Stability Comparison of First-Generation and
Second-Generation Spiro-oxindoles
Table 6. Structure−Activity Relationships of Second-
a
Generation Spiro-oxindoles
binding affinity
IC50 (nM) Ki (nM)
151 15 19
33
SJSA1 cell line
entry
compound
IC50 (μM)
1
12D
13D
14D
15C
15D
16D
17D
18D
20
2
2.9 1.5
5.4 2.1
6.9 1.4
2.0 0.4
1.6 0.4
0.65 0.2
0.19 0.04
2.1 1.9
2
4
3.4 0.5
105 10
234 10
3
775 77
1719 75
149 32
4
5
19
4
6
35
1
3.7 0.1
2.9 0.8
2.8 0.3
7
17.2 4.6
15.6 2.2
>10 000
8
b
9
NT
10
11
12
13
14
15
16
17
18
19
20
21
22
21D
22C
22D
23D
24D
25C
25D
26D
27D
28C
28D
29D
31
167 12
9516 888
589 84
22
2
>10
% composition of compound in
1:1 CH3CN/H2O
1301 121
80 12
1.65 0.27
2.3 0.28
0.30 0.07
>10
entry
compound
23D
0 h
100
1 days
3 days
6 days
21
5
1.7 0.6
23 4.4
1
2
3
4
5
6
100
99.6
100
99.2
100
175 32
9058 863
494 45
26D
100
100
100
1238 118
1.67 0.35
2.08 0.46
0.20 0.18
>10
29D
100
100
100
67
3.0 0.8
15
1026 127
53
6
a
MI-63
98.09
97.45
100
90.79
95.80
99.56
86.9
95.7
98.6
82.7
91.4
97.7
30
6
a
MI-147
3 (MI-888)
118 33
7505 928
396 49
5
a
1.46 0.24
1.68 0.14
0.16 0.02
0.10 0.03
a
Reference 24.
7
20
4
1.7 0.6
4.4 1.1
0.16 0.1
oxindoles 3, MI-147, and MI-63 displayed 2.3%, 8.6%, and
17.3% isomerization, respectively, after 6 days in acetonitrile/
water solution. In direct comparison, the second-generation
compounds 23D, 26D, and 29D showed no detectable change
in composition after 6 days in the same solution. Therefore, it
can be concluded that the second-generation spiro-oxindoles
with the D-configuration possess superior stability to the first-
generation spiro-oxindoles.
a
Mean and standard deviation of at least three independent
experiments. NT = not tested.
b
Hence, our data showed that a symmetrical C2 in the
pyrrolidine in these second-generation spiro-oxindoles results
in destabilization of the C-diastereoisomer which, taking
advantage of the rapid ring-opening-cyclization reaction, is
quickly and irreversibly converted to the D-diastereoisomer.
Next the new spiro-oxindoles were evaluated for their
binding affinities to MDM2. Only the compounds that could be
obtained in pure form, i.e., 15C, 22C, 25C, 28C, 12D−18D,
and 21D−29D, were tested for their binding to MDM2 and
their inhibition of cell growth in the SJSA-1 cancer cell line.
The results are shown in Table 6. Consistent with the data from
the first-generation spiro-oxindoles,22,24 15C, with the unstable
C-configuration, displays a binding affinity (Ki = 234 nM)
inferior to that of its corresponding stable diastereoisomer 15D,
which has Ki = 19 nM. This difference in binding affinities
between the diastereoisomers was maintained after changing
the carboxamide substituents. Thus, 22D, 25D, and 28D have
better binding affinities than their corresponding C-diastereo-
isomers 22C, 25C, and 28C.
We performed computational docking studies to investigate
the binding models of the D-diastereoisomer 15D and C-
diastereoisomer 15C in complex with MDM2 (Figure 5).
Superposition of the modeled complex structure for 15D with
the co-crystal structure of 4 complexed with MDM223 shows
that the cyclobutyl in 15D and neopentyl in 4 occupy the
Phe19 pocket, the same oxindole aryl group in both
compounds projects into the Trp23 pocket, and the same 3-
chloro-2-fluorophenyl group in both compounds occupies the
Leu26 pocket, respectively (Figure 5A). Interestingly but
Figure 5. (A) Modeled binding mode of 15D (yellow) and (B)
modeled binding mode of 15C (green), superimposed upon the co-
crystal structure of 4 (pink) in MDM2 protein. Amino acid residues of
the MDM2 protein are labeled in red, and the binding pockets are
labeled in white. Docked models were obtained with GOLD, and
figures were generated using Pymol.
consistent with our previous modeling prediction for MI-
219,20 the predicted binding model for 15C shows that its
cyclobutyl and 3-chloro-2-fluorophenyl substituents are re-
versed, as compared to the binding model for 15D, projecting
into the Leu26 and Phe19 pockets, respectively. In the binding
models for 15D and 4, these compounds benefit from several
interactions with MDM2 protein: (1) their 3-chloro-2-fluoro-
phenyl substituent in the Leu26 pocket has π−π stacking with
His96; (2) the carbonyl group of the pyrrolidine carboxamide
forms a hydrogen bond with His96 and; and (3) refolding of
the N-terminal of MDM2 provides additional hydrophobic
interactions with the 3-chloro-2-fluorophenyl substituent in the
Leu26 pocket.23 These interactions are not present in the
binding model for 15C and likely contribute to the superior
F
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 6. Stability comparison of 3, 17D, and 31 in (A) 1:1 CH3CN/H2O, (B) 1:1 MeOH/H2O, and (C) cell culture media.
MDM2 binding affinities for the D-diastereoisomers over their
C-diastereoisomers.
that 17D has excellent chemical stability in different solutions
and has a superior stability compared to 3.
For a potent and specific MDM2 inhibitor, its cellular activity
should be dependent on the activation of wild-type p53. Hence,
17D was assessed for its cell activity and specificity in the HCT-
116 p53+/+ colon cancer cell line and its p53 knockout isogenic
HCT-116 p53−/− cell line (Table 7). In a cell growth assay,
Compounds 12D−18D, which possess the stable D-
configuration, all display potent binding affinities with Ki values
ranging from 2.8 to 105 nM (Table 6). In the acyclic
modifications, the binding affinities are improved by 5-fold,
from 19 to 3.4 nM, by increasing the substituent size from
dimethyl in 12D to diethyl in 13D (Table 6, entries 1 and 2),
but further increase in substituent size, to di-n-propyl in 14D,
results in a 30-fold decrease in binding. The cyclic compounds
15D to 18D tolerate a substituent as large as a cycloheptyl
group (18D). The spirocyclohexyl- and spirocycloheptyl-
containing compounds 17D and 18D display the best Ki
values (2.9 and 2.8 nM, respectively) in this series.
Table 7. Cell Growth Inhibition in HCT-116 p53+/+ and
a
HCT-116 p53−/− Cell Lines
cell growth inhibition, IC50 (μM)
entry compound HCT116 p53+/+ cell line HCT116 p53−/− cell line
1
2
17D
31
0.32 0.09
0.25 0.05
>10
>10
As expected, varying the carboxamide substituent did not
change the trend. Independent of the carboxamide substituents,
an improvement in binding affinities was observed going from
the spirocyclobutyl (22D, 25D, and 28D) to the 2,2-diethyl
(21D, 24D, and 27D) to the spirocyclohexyl (23D, 26D, 29D)
compounds, respectively. However, in all cases, compounds
13D, 15D, and 17D, with the trans-4-hydroxycyclohexyl
carboxamide substituent, showed higher binding affinities
than their analogues having other carboxamide substituents.
We evaluated these new compounds for their cell growth
inhibitory activity in the SJSA-1 cell line, which has wild-type
p53 and overexpression of MDM2 protein due to MDM2 gene
amplification. In general, all compounds with high binding
affinities (Ki values in the low single digit nanomolar to sub-
nanomolar) have sub-micromolar IC50 values in inhibition of
cell growth (e.g., 16D, 17D, 23D, 26D, and 29D), with the
exception of compounds 13D and 18D, which have IC50 = 5.4
and 2.1 μM, respectively. One factor that may contribute to
their modest cellular potencies is that 13D and 18D are less
stable in solution (Table 3, entries 2 and 10) showing
decomposition to compound 20, which does not bind to
MDM2. Other compounds with Ki > 10 nM to MDM2, such as
12D, 14D, 15C, 15D, 22C, 22D, 25C, 25D, and 28C, all have
IC50 values in the low micromolar range, highlighting that very
high affinities to MDM2 are necessary to achieve potent cell
growth inhibition activity in the SJSA-1 cell line.
a
Mean and standard deviation of at least three runs.
17D has IC50 = 0.32 μM in the HCT-116 p53+/+ cell line but
shows an IC50 > 10 μM in the isogenic HCT-116 p53−/− cell
line, thus displaying high selectivity.
Based on its high binding affinity to MDM2, potent cell
growth inhibitory activity, and excellent chemical stability, 17D
was evaluated in pharmacodynamic (PD) and efficacy experi-
ments in vivo in SCID mice bearing SJSA-1 osteosarcoma
xenografts. A single oral 100 mg/kg dose of 17D effectively
activates p53, as evidenced by robust accumulation of p53
protein itself, and MDM2 and p21 proteins, two p53-targeted
gene products (Figure 7A). However, at 100 mg/kg 17D
induces a minimal amount of PARP cleavage, indicative of
modest apoptosis induction in the SJSA-1 tumor tissue.
Consistent with the PD data, while 17D effectively inhibited
tumor growth at 200 mg/kg with daily, oral administration for
2 weeks, it failed to achieve tumor regression (Figure 7B). Our
in vitro and in vivo data indicated that while 17D represents a
promising lead compound, it needs to be further optimized for
potency and efficacy for therapeutic applications.
To further improve 17D as an MDM2 inhibitor, we focused
our modifications on its 4-hydroxycyclohexyl group (Figure 8)
since our previous studies have shown that this site plays a key
role in modulation of binding affinity to MDM2, in vitro
cellular potency, and in vivo efficacy.19,21,22 In the co-crystal
structure of compound 4 complexed with MDM2, the 4-
hydroxycyclohexyl group forms a hydrogen bonding interaction
with a lsyine residue in MDM2.23 Therefore, we replaced the 4-
hydroxycyclohexyl group with a benzoic acid group to enhance
the interactions with the lysine residue, which yielded 31
(Figure 8). Of note, the benzoic acid group at this site was also
employed by the Roche group for their design of MDM2
inhibitors.25,29,30 Compound 31 binds to MDM2 with Ki = 0.16
nM. In a cell growth assay, 31 achieves IC50 = 100 and 250 nM
in the SJSA-1 and HCT-116 p53+/+ cell lines, respectively, and
Among these new spiro-oxindoles, compound 17D possessed
the most potent binding affinity for MDM2 and growth
inhibition of SJSA-1 cell line and was therefore selected for
further evaluation.
We further evaluated the stability of 17D, together with 3, in
several solvents and found that 17D shows no detectable
isomerization over a period of 7 days in solution, whereas 3 has
2−3% isomerization in CH3CN/H2O (Figure 6A) and MeOH/
H2O (Figure 6B) but more than 10% isomerization in cell
culture media (Figure 6C) over the same period. We concluded
G
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
when administered orally daily for 14 days at 100 mg/kg. At the
end of treatment (day 25), the average tumor volume for the
eight mice treated with 31 was reduced to 12 mm3 from 94
mm3 at the start of treatment, an 88% regression, and three of
the mice had no detectable tumor. In comparison, 17D only
inhibited tumor growth but did not induce tumor regression
when dosed at 200 mg/kg, daily for 14 days when compared to
the control group. All the mice treated with 17D or 31 suffered
no weight loss and did not show any signs of toxicity during or
after the treatment (SI).
CONCLUSION
■
In summary, we have designed and synthesized a new
generation of spiro-oxindoles as MDM2 inhibitors, which
exploit the ring-opening-cyclization reaction mechanism of this
class of compounds. These second-generation compounds have
a symmetrical pyrrolidine C2 position. Our data demonstrated
that these second-generation spiro-oxindoles can be converted,
rapidly and irreversibly, to the configuration preferred for
binding to MDM2. To determine the optimal C2 substituents,
a series of compounds was prepared with symmetrical acyclic
and cyclic substituents at the C2 position of the pyrrolidine. It
was found that increasing the size of the substituents in these
compounds increases the rate of conversion to the trans-
configured, stable D-diastereoisomers. Among these new spiro-
oxindoles, compound 31 binds to MDM2 with a high affinity
(Ki = 0.16 nM), potently activates p53, and induces apoptosis
in the SJSA-1 xenograft tumor tissue in mice. Compound 31 is
capable of achieving tumor regression in the SJSA-1 xenograft
tumor model in mice with oral administration. Importantly,
compound 31 has excellent chemical stability in solution,
overcoming a limitation observed for the first-generation spiro-
oxindole MDM2 inhibitors.
Figure 7. (A) In vivo pharmacodynamics comparison using a single
dose of 100 mg/kg of 17D or 31 in SCID mice bearing SJSA-1 tumor
xenograft. (B) Efficacy of 17D in a SJSA-1 xenograft model. (C)
Efficacy of 31 in an SJSA-1 xenograft model.
EXPERIMENTAL SECTION
■
General Information. Unless otherwise stated, all reactions were
performed under a nitrogen atmosphere in dry solvents under
anhydrous conditions, and all commercial reagents were used as
supplied without further purification. NMR spectra were obtained on a
1
Figure 8. Design of a new analogue of 17D to further improve binding
to MDM2.
Bruker 300 UltraShield spectrometer at a H frequency of 300 MHz
and 13C frequency of 75 MHz. Chemical shifts (δ) are reported in
parts per million (ppm) relative to an internal standard. The final
products were purified on a preparative HPLC (Waters 2545,
Quaternary Gradient Module) with a SunFire Prep C18 OBD 5 μm
50 × 100 mm reverse-phase column. The mobile phase was a gradient
of solvent A (0.1% TFA in water) and solvent B (0.1% TFA in
MeOH) at a flow rate of 40 mL/min and 1%/2 min increase of solvent
B. CH3CN as solvent B failed to separate the diastereoisomers. All
final compounds have purity ≥95% as determined by Waters
ACQUITY UPLC using reverse-phase column (SunFire, C18, 5 μm,
4.6 × 150 mm) and a solvent gradient of A (0.1% of TFA in water)
and solvent D (0.1% of TFA in MeOH).
Synthesis and characterization of compound 20, 22C, 25C, 28C,
21D−29D, and 30 are described in the SI. The procedure for
monitoring the stability of 3, 17D, and 31 in cell culture media is the
same as described previously.24
3,3-Dimethoxypentane (9). p-Toluenesulfonic acid (PTSA) (29
mg, cat) was added to a solution of 3-pentanone (10 g, 116 mmol),
trimethyl orthoformate (30.8 g, 290 mmol) in methanol (41 mL).
After standing overnight, the reaction was poured into ice and
extracted with Et2O. The combined ether extracts were washed with
saturated sodium bicarbonate and brine, dried over sodium sulfate, and
filtered, and then the solvent was removed by rotary evaporation using
mild vacuum to produce 9 (11.2 g, 73% yield) as a colorless liquid. 1H
NMR (300 MHz, CDCl3) δ ppm 3.15 (s, 6H), 1.59 (q, J = 7.5 Hz,
has IC50 > 10 000 nM in the p53 knockout cell line HCT-116
p53−/− cell line. Thus, 31 has a higher binding affinity to
MDM2 and better cell growth inhibitory activity than 17D.
Chemical stability testing further showed that compound 31
has excellent stability in three different solutions (Figure 6A−
C).
We next evaluated 17D and 31 together for their ability to
activate p53 in a PD experiment in the SJSA-1 tumor tissue
harvested from mice treated with a single, oral dose of each
compound at 100 mg/kg (Figure 7A). Both compounds
effectively activated p53 in the SJSA-1 tumor tissue, leading to
accumulation of p53, MDM2, and p21 proteins. Compound 31
effectively induced robust cleavage of PARP in the tumor,
indicative of strong apoptosis induction, but 17D had a minimal
effect on PARP cleavage. Encouraged by the strong p53
activation and apoptosis induction by 31 in the SJSA-1 tumor
tissue, we next tested 31 for its efficacy in the SJSA-1 xenograft
model. Consistent with the strong p53 activation and apoptosis
induction in the tumor tissue, 31 demonstrated strong
antitumor activity and achieved significant tumor regression
H
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
4H), 0.82 (t, J = 7.5 Hz, 6H); 13C NMR (75 MHz, CDCl3) δ ppm
104.27, 47.73(2C), 24.43(2C), 7.98(2C).
4,4-Dimethoxyheptane (10). Starting with 4-heptanone, com-
(3′S,4′R,7′S,8′S,8a′R)-6″-Chloro-8′-(3-chloro-2-fluoro-
phenyl)-3′,4′-diphenyl-3′,4′,8′,8a′-tetrahydro-1′H-dispiro-
[cyclobutane-1,6′-pyrrolo[2,1-c][1,4]oxazine-7′,3″-indoline]-
1′,2″-dione (15A). Starting with cyclobutanone, compound 15A
(684 mg, 69% yield) was prepared according to the procedure
described for the preparation of 12A. 1H NMR (300 MHz, CDCl3) δ
ppm 9.12 (s, 1H), 7.37−7.05 (m, 11H), 6.96−6.73 (m, 5H), 6.15 (d, J
= 3.6 Hz, 1H), 4.96 (d, J = 3.7 Hz, 1H), 4.74 (d, J = 9.2 Hz, 1H), 4.59
(d, J = 9.2 Hz, 1H), 3.01−2.86 (m, 1H), 2.68 (q, J = 9.34 Hz, 1H),
2.12 (q, J = 9.7 Hz, 1H), 1.73−1.46 (m, 2H), 1.36−1.15 (m, 1H); 13C
pound 10 (10.52 g, 74% yield) was prepared according to the
1
procedure described for the preparation of 9. H NMR (300 MHz,
CDCl3) δ ppm 3.14 (s, 6H), 1.59−1.49 (m, 4H), 1.35−1.19 (m, 4H),
0.92 (t, J = 7.3 Hz, 6H); 13C NMR (75 MHz, CDCl3) δ ppm 103.45,
47.79(2C), 35.03(2C), 17.23(2C), 14.56(2C).
1,1-Dimethoxycyclooctane (11). Starting with cyclo-octanone,
compound 11 (2.23 g, 82% yield) was prepared according to the
NMR (75 MHz, CDCl3) δ ppm 178.27, 169.84, 156.63 (d, JC−F
=
1
procedure described for the preparation of 9. H NMR (300 MHz,
249.92 Hz), 142.50, 137.66, 136.09, 134.88, 130.20, 128.74, 128.52,
128.49, 128.28, 127.80, 127.58, 127.40 (d, JC−F = 2.70 Hz), 126.30,
125.65, 124.48 (d, JC−F = 13.11 Hz), 124.26 (d, JC−F = 4.08 Hz),
122.51, 121.50 (d, JC−F = 19.15 Hz), 111.15, 81.99, 74.00, 62.67, 60.99,
58.25, 46.55, 32.72, 28.22, 14.49; ESI-MS m/z 613.17 (M+H)+.
(3′S,4′R,7′S,8′S,8a′R)-6″-Chloro-8′-(3-chloro-2-fluoro-
phenyl)-3′,4′-diphenyl-3′,4′,8′,8a′-tetrahydro-1′H-dispiro-
[cyclopentane-1,6′-pyrrolo[2,1-c][1,4]oxazine-7′,3″-indoline]-
1′,2″-dione (16A). In a round-bottom flask, (E)-6-chloro-3-(3-
chloro-2-fluorobenzylidene)indolin-2-one (6, 500 mg, 1.62 mmol),
(5R,6S)-5,6-diphenyl-2-morpholinone (7, 492 mg, 1.94 mmol), and
cyclopentanone (409 mg, 4.86 mmol) were suspended in toluene (10
mL) and heated at reflux at an oil bath temperature of 140 °C. After
heating at reflux for 2 h, the reaction was allowed to cool to room
temperature, and the solvent was removed by rotary evaportion. The
crude product was purified by column chromatography (the
compound was eluted with 100% CH2Cl2) to give 562 mg (55%
CDCl3) δ ppm 3.14 (s, 6H), 1.82−1.73 (m, 4H), 1.56 (br. s, 10H);
13C NMR (75 MHz, CDCl3) δ ppm 103.95, 47.81(2C), 30.48(2C),
28.31(2C), 24.68, 21.44(2C).
(3S,3′S,4′R,8′S,8a′R)-6-Chloro-8′-(3-chloro-2-fluorophenyl)-
6′,6′-dimethyl-3′,4′-diphenyl-3′,4′,8′,8a′-tetrahydrospiro-
[indoline-3,7′-pyrrolo[2,1-c][1,4]oxazine]-1′,2(6′H)-dione
(12A). In a sealed tube, (E)-6-chloro-3-(3-chloro-2-fluoro-
benzylidene)indolin-2-one (6, 500 mg, 1.62 mmol), (5R,6S)-5,6-
diphenyl-2-morpholinone (7, 492 mg, 1.94 mmol), and 2,2-
dimethoxypropane (506 mg, 4.86 mmol) were suspended in a mixture
of toluene (7 mL) and THF (0.7 mL) and heated to 140 °C. After 1 h,
the reaction was allowed to cool to room temperature, and the solvent
was removed by rotary evaporation. The crude product was purified by
column chromatography (the compound eluted with 100% dichloro-
methane) to give 133 mg (14% yield) of 12A as a light yellow solid. 1H
NMR (300 MHz, CDCl3) δ ppm 7.63 (br s, 1H), 7.59−7.02 (m,
14H), 6.86 (t, J = 8.0 Hz, 1H), 6.70 (d, J = 1.6 Hz, 1H), 5.72 (d, J =
4.7 Hz, 1H), 5.08 (d, J = 8.8 Hz, 1H), 4.95−4.81 (m, 2H), 1.46 (s,
3H), 0.72 (s, 3H); 13C NMR (75 MHz, CDCl3) δ ppm 178.03, 172.51,
156.72 (d, JC−F = 251.38 Hz), 142.75, 137.97, 135.03, 135.00, 129.87,
129.55, 129.40, 128.91, 128.50, 128.29, 128.17, 127.87, 125.03 (d, JC−F
= 12.89 Hz), 124.31 (d, JC−F = 4.69 Hz), 124.20, 122.36, 121.19 (d,
JC−F = 19.41 Hz), 110.06, 84.35, 69.20, 66.34, 61.82, 60.37, 42.02,
24.75, 20.74; ESI-MS m/z 601.33 (M+H)+.
1
yield) of 16A as a yellow solid. H NMR (300 MHz, CDCl3) δ ppm
8.98 (s, 1H), 7.28−7.01 (m, 12H), 6.96−6.77 (m, 4H), 6.44 (d, J = 4.2
Hz, 1H), 5.03 (d, J = 9.5 Hz, 1H), 4.87 (d, J = 4.4 Hz, 1H), 4.79 (d, J
= 9.5 Hz, 1H), 2.41−2.26 (m, 1H), 2.22−2.07 (m, 1H), 1.66−1.12 (m,
6H); 13C NMR (75 MHz, CDCl3) δ ppm 180.08, 171.03, 156.64 (d,
JC−F = 248.76 Hz), 142.37, 138.85, 136.05, 134.47, 130.02, 128.54,
128.48, 128.41, 128.26, 128.21, 127.68, 127.48 (d, JC−F = 3.21 Hz),
126.79, 126.27, 125.30 (d, JC−F = 13.47 Hz), 124.33 (d, JC−F = 4.40
Hz), 121.90, 121.53 (d, JC−F = 19.02 Hz), 110.76, 82.04, 81.26, 64.57,
60.33, 59.43, 49.27, 36.21, 28.99, 23.88, 23.69; ESI-MS m/z 627.16 (M
+H)+.
(3S,3′S,4′R,8′S,8a′R)-6-Chloro-8′-(3-chloro-2-fluorophenyl)-
6′,6′-diethyl-3′,4′-diphenyl-3′,4′,8′,8a′-tetrahydrospiro-
[indoline-3,7′-pyrrolo[2,1-c][1,4]oxazine]-1′,2(6′H)-dione
(13A). Starting with 3,3-dimethoxypentane (9), compound 13A (166
mg, 16% yield) was prepared according to the procedure described for
(3′S,4′R,7′S,8′S,8a′R)-6″-Chloro-8′-(3-chloro-2-fluoro-
phenyl)-3′,4′-diphenyl-3′,4′,8′,8a′-tetrahydro-1′H-dispiro-
[cyclohexane-1,6′-pyrrolo[2,1-c][1,4]oxazine-7′,3″-indoline]-
1′,2″-dione (17A). Starting with cyclohexanone, compound 17A
(665 mg, 64% yield) was prepared according to the procedure
described for the preparation of 16A. 1H NMR (300 MHz, CDCl3) δ
ppm 9.24 (s, 1H), 7.79 (t, J = 6.4 Hz, 1H), 7.35−7.03 (m, 11H), 6.94
(d, J = 1.4 Hz, 1H), 6.80 (d, J = 6.7 Hz, 2H), 6.61 (dd, J = 1.4, 8.2 Hz,
1H), 6.29 (d, J = 8.2 Hz, 1H), 5.53 (d, J = 11.3 Hz, 1H), 4.95 (d, J =
1.4 Hz, 1H), 4.68 (d, J = 11.3 Hz, 1H), 2.50 (d, J = 12.6 Hz, 1H), 2.07
(d, J = 12.1 Hz, 1H), 1.53−1.01 (m, 7H), 0.93−0.70 (m, 1H); 13C
1
the preparation of 12A. H NMR (300 MHz, CDCl3) δ ppm 8.76 (s,
1H), 7.54 (t, J = 6.7 Hz, 1H), 7.29−7.02 (m, 10H), 6.98−6.80 (m,
4H), 6.62 (dd, J = 1.7, 8.2 Hz, 1H), 6.39 (d, J = 8.3 Hz, 1H), 5.22 (d, J
= 10.8 Hz, 1H), 4.97 (d, J = 3.3 Hz, 1H), 4.68 (d, J = 10.9 Hz, 1H),
2.51−2.33 (m, 1H), 1.93−1.66 (m, 2H), 1.58−1.41 (m, 1H), 0.64 (t, J
= 7.4 Hz, 3H), 0.58 (t, J = 7.1 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ
ppm 182.54, 170.19, 156.84 (d, JC−F = 247.24 Hz), 142.44, 139.19,
136.48, 134.29, 130.01, 128.76, 128.70, 128.17, 128.03, 127.87, 127.22,
126.63, 126.44, 126.40, 125.91 (d, JC−F = 14.75 Hz), 124.56 (d, JC−F
=
4.50 Hz), 121.38, 121.33 (d, JC−F = 18.72 Hz), 110.52, 81.36, 75.59,
66.97, 60.16, 58.29, 52.07, 27.61, 23.99, 8.95, 8.29; ESI-MS m/z
629.00 (M+H)+.
NMR (75 MHz, CDCl3) δ ppm 183.39, 170.16, 156.79 (d, JC−F
=
246.90 Hz), 142.81, 139.07, 136.64, 134.25, 129.95, 129.12, 128.69,
128.11, 127.50, 127.03, 126.58, 126.41, 126.21, 126.01, 125.58, 124.69
(d, JC−F = 4.04 Hz), 121.43, 121.21, 110.62, 80.82, 74.37, 68.50, 61.01,
59.25, 51.78, 36.88, 29.98, 25.30, 22.62, 22.36; ESI-MS m/z 641.08 (M
+H)+.
(3S,3′S,4′R,8′S,8a′R)-6-Chloro-8′-(3-chloro-2-fluorophenyl)-
3′,4′-diphenyl-6′,6′-di-n-propyl-3′,4′,8′,8a′-tetrahydrospiro-
[indoline-3,7′-pyrrolo[2,1-c][1,4]oxazine]-1′,2(6′H)-dione
(14A). Starting with 4,4-dimethoxyheptane (10), compound 14A (266
mg, 25% yield) was prepared according to the procedure described for
(3′S,4′R,7′S,8′S,8a′R)-6″-Chloro-8′-(3-chloro-2-fluoro-
phenyl)-3′,4′-diphenyl-3′,4′,8′,8a′-tetrahydro-1′H-dispiro-
[cycloheptane-1,6′-pyrrolo[2,1-c][1,4]oxazine-7′,3″-indoline]-
1′,2″-dione (18A). Starting with cycloheptanone, compound 18A
(315 mg, 30% yield) was prepared according to the procedure
described for the preparation of 16A. In this case the reaction was
heated for just 30−60 min, a longer reaction time produces lower
1
the preparation of 12A. H NMR (300 MHz, CDCl3) δ ppm 9.16 (s,
1H), 7.58 (t, J = 6.7 Hz, 1H), 7.28−7.05 (m, 10H), 6.97 (d, J = 3.1 Hz,
1H), 6.91−6.78 (m, 3H), 6.64 (dd, J = 1.9, 8.2 Hz, 1H), 6.39 (d, J =
8.3 Hz, 1H), 5.25 (d, J = 10.9 Hz, 1H), 4.97 (d, J = 3.4 Hz, 1H), 4.69
(d, J = 10.9 Hz, 1H), 2.27 (t, J = 12.1 Hz, 1H), 1.87−1.67 (m, 2H),
1.48−1.34 (m, 1H), 1.33−1.17 (m, 1H), 1.16−0.99 (m, 1H), 0.94−
0.66 (m, 8H); 13C NMR (75 MHz, CDCl3) δ ppm 182.90, 170.18,
156.83 (d, JC−F = 247.50 Hz), 142.56, 139.13, 136.50, 134.31, 130.00,
128.66, 128.60, 128.16, 128.05, 127.85, 127.21, 126.53, 126.39, 126.27
1
yields. H NMR (300 MHz, CDCl3) δ ppm 8.42 (s, 1H), 7.30−7.11
(m, 8H), 7.11−7.00 (m, 4H), 6.92 (t, J = 7.9 Hz, 1H), 6.83 (d, J = 1.3
Hz, 1H), 6.79−6.67 (m, 2H), 6.57 (d, J = 3.7 Hz, 1H), 5.01 (d, J =
10.0 Hz, 1H), 4.96 (d, J = 3.89 Hz, 1H), 4.73 (d, J = 9.9 Hz, 1H),
2.44−2.26 (m, 2H), 1.80−0.77 (m, 10H); 13C NMR (75 MHz,
CDCl3) δ ppm 180.42, 171.02, 156.73 (d, JC−F = 248.72 Hz), 142.04,
138.73, 136.22, 134.33, 129.96, 129.11, 128.80, 128.33 (d, JC−F = 2.37
Hz), 128.25, 128.11, 127.58, 127.16 (d, JC−F = 2.77 Hz), 126.61,
(d, JC−F = 2.89 Hz), 125.94 (d, JC−F = 14.61 Hz), 124.61 (d, JC−F
=
4.24 Hz), 121.42, 121.35 (d, JC−F = 18.59 Hz), 110.59, 81.21, 75.16,
67.20, 60.21, 58.29, 51.86, 38.18, 34.47, 17.92, 17.09, 15.29, 14.66;
ESI-MS m/z 657.08 (M+H)+.
I
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
125.94, 125.45 (d, JC−F = 13.66 Hz), 124.35 (d, JC−F = 4.48 Hz),
121.71, 121.49 (d, JC−F = 19.14 Hz), 110.60, 81.90, 77.38, 66.65, 59.72,
59.12, 48.96, 37.05, 32.44, 30.12, 29.77, 24.54, 22.68; ESI-MS m/z
655.25 (M+H)+.
3.70−3.55 (m, 1H), 3.49−3.36 (m, 1H), 2.52 (t, J = 13.0 Hz, 1H),
2.18 (t, J = 13.0 Hz, 1H), 2.10−1.68 (m, 6H), 1.66−1.53 (m, 1H),
1.53−1.11 (m, 7H), 1.11−0.86 (m, 2H), 0.81 (t, J = 6.8 Hz, 3H),
0.75−0.57 (m, 1H); 13C NMR (75 MHz, CD3OD) δ ppm 178.45,
167.82, 157.78 (d, JC−F = 249.65 Hz), 144.71, 136.92, 132.38, 128.98,
(3R,4′S,5′R)-6-Chloro-4′-(3-chloro-2-fluorophenyl)-N-
((1R,4R)-4-hydroxycyclohexyl)-2′,2′-dimethyl-2-oxospiro-
[indoline-3,3′-pyrrolidine]-5′-carboxamide (12D, TFA Salt).
trans-4-Aminocyclohexanol (127 mg, 1.11 mmol) was added to
solution of 12A (133 mg, 0.22 mmol) in THF (3 mL) and heated at
reflux. After heating at reflux overnight, the reaction was cooled to
room temperature, and the solvent was removed by rotary
evaporation. The crude 12B was purified by column chromatography
to produce 107 mg (68% yield) of 12B. Cerium ammonium nitrate
(162 mg, 0.296 mmol) was added to a solution of the resulting 12B
(107 mg, 0.148 mmol) in MeCN (3 mL) and stirred for 5 min at room
temperature, and then H2O (3 mL) was added. After the reaction was
stirred for an additional 10 min, it was quenched with saturated
sodium bicarbonate, brine was added, and the solution was extracted
with EtOAc. The EtOAc solution was dried over sodium sulfate and
filtered through Celite, and the solvent was removed by rotary
evaporation to produce crude 12D. The crude material was dissolved
in a 3:1 mixture of MeOH:H2O that was acidified with TFA, and this
solution was immediately analyzed by UPLC (data shown in Table 1)
and purified by preparative HPLC (mobile phase was a gradient of
solvent A (0.1% of TFA in H2O) and solvent B (0.1% of TFA in
MeOH) at a flow rate of 40 mL/min and 1%/2 min increase of solvent
B). The combined fractions of the pure compound were immediately
analyzed by UPLC (data shown in Table 3, “0 h” column) and
concentrated by rotary evaporation. The concentrate was redissolved
in a minimum amount of MeCN, H2O was added, and the solution
was frozen and lyophilized to produce the TFA salt of 12D (45 mg,
128.91, 126.42 (d, JC−F = 4.81 Hz), 124.06, 123.50, 122.15 (d, JC−F
=
19.28 Hz), 121.72 (d, JC−F = 11.61 Hz), 111.75, 75.43, 69.91, 66.76,
61.83, 36.07, 35.06, 34.47, 34.33, 34.26, 30.98, 30.77, 17.65, 17.16,
14.60, 14.39; ESI-MS m/z 576.50 (M+H)+.
(3′S,4′S,5′R)-6″-Chloro-4′-(3-chloro-2-fluorophenyl)-N-
((1R,4R)-4-hydroxycyclohexyl)-2″-oxodispiro[cyclobutane-
1,2′-pyrrolidine-3′,3″-indoline]-5′-carboxamide (15C, TFA
Salt). Starting with 15B (493 mg), compound 15C was prepared
according to the procedure described for the preparation of 12D. The
crude sample contains 15C and 15D (Table 1, entry 4). One-third of
the crude was purified by preparative HPLC, which separated the
isomers 15C (81 mg, 55% yield) and 15D (9 mg, 6% yield), which
1
were stable enough to be characterized. 15C: H NMR (300 MHz,
CD3OD) δ ppm 8.28 (d, J = 7.7 Hz, 1H), 7.42−7.32 (m, 1H), 7.24−
7.16 (m, 1H), 7.12−7.03 (m, 1H), 6.96−6.89 (m, 3H), 4.82 (d, J =
11.1 Hz, 1H), 4.44 (d, J = 11.2 Hz, 1H), 3.73−3.57 (m, 1H), 3.54−
3.40 (m, 1H), 2.67−2.52 (m, 2H), 2.42−2.26 (m, 1H), 2.25−1.77 (m,
5H), 1.72−1.50 (m, 2H), 1.39−1.20 (m, 3H), 1.19−1.01 (m, 1H); 13C
NMR (75 MHz, CD3OD) δ ppm 178.26, 166.49, 157.72 (d, JC−F
=
248.78 Hz), 145.52, 136.79, 132.02, 128.60 (d, JC−F = 2.52 Hz),
128.53, 126.07 (d, JC−F = 4.82 Hz), 124.20, 123.88 (d, JC−F = 13.56
Hz), 123.02, 122.53 (d, JC−F = 18.31 Hz), 112.14, 73.33, 70.01, 63.09,
61.70, 34.42, 34.37, 31.00, 30.95, 30.14, 15.47; ESI-MS m/z 532.50 (M
+H)+.
(3′R,4′S,5′R)-6″-Chloro-4′-(3-chloro-2-fluorophenyl)-N-
((1R,4R)-4-hydroxycyclohexyl)-2″-oxodispiro[cyclobutane-
1,2′-pyrrolidine-3′,3″-indoline]-5′-carboxamide (15D, TFA
Salt). Isomerization of 15C to produce 15D was accomplished by
refluxing a solution of 15C (20 mg) in a 3:1 mixture of MeOH:H2O
with 0.1% TFA. After 2 days, the reaction was cooled, purified by
preparative HPLC, concentrated, redisolved in water, frozen, and
1
47% yield) as a white powder. H NMR (300 MHz, CD3OD) δ ppm
8.16 (d, J = 7.7 Hz, 1H), 7.67−7.59 (m, 1H), 7.51 (dd, J = 2.1, 8.2 Hz,
1H), 7.43−7.35 (m, 1H), 7.22−7.13 (m, 1H), 7.11 (dd, J = 1.9, 8.2
Hz, 1H), 6.80 (d, J = 1.9 Hz, 1H), 5.09 (d, J = 11.2 Hz, 1H), 4.83 (d, J
= 11.3 Hz, 1H), 3.69−3.53 (m 1H), 3.50−3.35 (m, 1H), 2.08−1.84
(m, 5H), 1.84−1.73 (m, 1H), 1.65−1.51 (m, 1H), 1.41 (s, 3H), 1.39−
1.06 (m, 3H), 1.04−0.87 (m, 1H); 13C NMR (75 MHz, CD3OD) δ
ppm 178.50, 167.15, 157.75 (d, JC−F = 249.27 Hz), 145.27, 137.16,
132.41, 128.94 (d, JC−F = 2.15 Hz), 128.69, 126.51 (d, JC−F = 4.6 Hz),
123.57, 122.27 (d, JC−F = 19.05 Hz), 122.10, 121.95 (d, JC−F = 12.41
Hz), 111.86, 69.95, 69.91, 66.66, 61.69, 47.23, 34.32, 34.26, 30.92,
30.78, 29.84, 24.62, 22.36; ESI-MS m/z 520.67 (M+H)+.
1
lyophilized to produce 15D as a white powder. H NMR (300 MHz,
CD3OD) δ ppm 8.19 (d, J = 7.6 Hz, 1H), 7.64 (dd, J = 1.2, 8.2 Hz,
1H), 7.57−7.48 (m, 1H), 7.44−7.35 (m, 1H), 7.19−7.10 (m, 2H),
6.87 (d, J = 1.6 Hz, 1H), 4.95 (d, J = 10.9 Hz, 1H), 4.48 (d, J = 10.9
Hz, 1H), 3.71−3.55 (m, 1H), 3.50−3.36 (m, 1H), 3.10−2.82 (m, 2H),
2.63−2.46 (m, 1H), 2.11−1.72 (m, 5H), 1.69−1.54 (m, 1H), 1.49−
1.12 (m, 4H), 1.06−0.85 (m, 1H); 13C NMR (75 MHz, CD3OD) δ
ppm 178.06, 167.26, 157.67 (d, JC−F = 248.91 Hz), 145.33, 137.35,
132.44, 128.35, 127.90, 126.61 (d, JC−F = 4.92 Hz), 123.94, 122.36 (d,
JC−F = 18.92 Hz), 122.19 (d, JC−F = 12.81 Hz), 121.97, 112.16, 72.28,
69.92, 64.97, 62.52, 47.33, 47.29, 34.34, 34.29, 31.14, 30.96, 30.84,
28.67, 14.94; ESI-MS m/z 532.42 (M+H)+.
(3R,4′S,5′R)-6-Chloro-4′-(3-chloro-2-fluorophenyl)-2′,2′-di-
ethyl-N-((1R,4R)-4-hydroxycyclohexyl)-2-oxospiro[indoline-
3,3′-pyrrolidine]-5′-carboxamide (13D, TFA Salt). Starting with
13B (69 mg), compound 13D (23 mg, 38% yield) was prepared
1
according to the procedure described for the preparation of 12D. H
NMR (300 MHz, CD3OD) δ ppm 8.13 (d, J = 7.7 Hz, 1H), 7.67−7.58
(m, 1H), 7.54 (dd, J = 2.2, 8.3 Hz, 1H), 7.43−7.34 (m, 1H), 7.22−
7.12 (m, 1H), 7.09 (dd, J = 1.9, 8.3 Hz, 1H), 6.75 (d, J = 1.9 Hz, 1H),
5.05 (d, J = 11.2 Hz, 1H), 4.81 (d, J = 11.3 Hz, 1H), 3.70−3.54 (m,
1H), 3.49−3.35 (m, 1H), 2.77−2.59 (m, 1H), 2.42−2.24 (m, 1H),
2.24−2.09 (m, 1H), 2.09−1.94 (m, 1H), 1.93−1.83 (m, 2H), 1.83−
1.71 (m, 1H), 1.63−1.50 (m, 1H), 1.41−0.82 (m, 7H), 0.57 (t, J = 7.6
Hz, 3H); 13C NMR (75 MHz, CD3OD) δ ppm 178.36, 167.49, 157.79
(d, JC−F = 249.25 Hz), 144.68, 136.95, 132.44, 129.05, 128.87, 126.45
(d, JC−F = 4.77 Hz), 124.00, 123.50, 122.17 (d, JC−F = 19.39 Hz),
121.45 (d, JC−F = 12.15 Hz), 111.76, 77.69, 76.20, 69.88, 66.54, 61.71,
34.30, 34.22, 30.93, 30.70, 29.93, 26.00, 25.23, 7.82, 7.38; ESI-MS m/z
548.42 (M+H)+.
(3′R,4′S,5′R)-6″-Chloro-4′-(3-chloro-2-fluorophenyl)-N-
((1R,4R)-4-hydroxycyclohexyl)-2″-oxodispiro[cyclopentane-
1,2′-pyrrolidine-3′,3″-indoline]-5′-carboxamide (16D, TFA
Salt). Starting with 16B (274 mg), compound 16D was prepared
according to the procedure described for the preparation of 12D. The
crude sample contains 16C and 16D (Table 1, entry 5). Preparative
HPLC separated the isomers; however, the fractions corresponding to
16C already revealed some conversion to 16D (Table 3, entry 6).
Because of the rapid isomerization, isolation of pure 16C (mixture: 37
mg, 15% yield) was not possible, and only 16D (123 mg, 50% yield)
1
was characterized. 16D: H NMR (300 MHz, CD3OD) δ ppm 8.24
(d, J = 7.7 Hz, 1H), 7.62−7.50 (m, 2H), 7.43−7.34 (m, 1H), 7.19−
7.12 (m, 1H), 7.10 (dd, J = 2.0, 8.2 Hz, 1H), 6.81 (d, J = 1.9 Hz, 1H),
5.07 9d, J = 11.0 Hz, 1H), 4.69 (d, J = 11.1 Hz, 1H), 3.70−3.56 (m,
1H), 3.49−3.36 (m, 1H), 2.81−2.68 (m, 1H), 2.53−2.39 (m, 1H),
2.24−2.11 (m, 1H), 2.11−1.44 (m, 9H), 1.37−1.17 (m, 3H), 1.03−
0.86 (m, 1H); 13C NMR (75 MHz, CD3OD) δ ppm 178.90, 167.28,
157.70 (d, JC−F = 249.06 Hz), 145.40, 137.21, 132.38, 128.82 (d, JC−F
= 2.25 Hz), 128.39, 126.57 (d, JC−F = 4.90 Hz), 123.67, 122.33 (d, JC−F
= 19.10 Hz), 122.20 (d, JC−F = 12.68 Hz), 122.06, 111.97, 80.49,
69.92, 65.76, 62.14, 49.80, 48.32, 34.88, 34.73, 34.34, 34.27, 30.97,
30.82, 24.78, 24.31; ESI-MS m/z 546.33 (M+H)+.
(3R,4′S,5′R)-6-Chloro-4′-(3-chloro-2-fluorophenyl)-N-
((1r,4R)-4-hydroxycyclohexyl)-2-oxo-2′,2′-di-n-propylspiro-
[indoline-3,3′-pyrrolidine]-5′-carboxamide (14D, TFA salt).
Starting with 14B (174 mg), compound 14D (71 mg, 46% yield)
was prepared according to the procedure described for the preparation
1
of 12D. H NMR (300 MHz, CD3OD) δ ppm 8.15 (d, J = 7.3 Hz,
1H), 7.61 (t, J = 6.7 Hz, 1H), 7.52 (dd, J = 1.6, 8.3 Hz, 1H), 7.38 (t, J
= 7.2 Hz, 1H), 7.16 (t, J = 8.1 Hz, 1H), 7.10 (dd, J = 1.4, 8.1 Hz, 1H),
6.75 (s, 1H), 5.01 (d, J = 11.1 Hz, 1H), 4.81 (d, J = 11.2 Hz, 1H),
J
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(3′R,4′S,5′R)-6″-Chloro-4′-(3-chloro-2-fluorophenyl)-N-
((1R,4R)-4-hydroxycyclohexyl)-2″-oxodispiro[cyclohexane-
1,2′-pyrrolidine-3′,3″-indoline]-5′-carboxamide (17D, TFA
Salt). Starting with 17B (224 mg), compound 17D was prepared
according to the procedure described for the preparation of 12D. The
crude sample contains 17C and 17D (Table 1, entry 6). Preparative
HPLC separated the isomers; however, the fractions corresponding to
17C already displayed conversion to 17D (Table 3, entry 8). Because
of the rapid isomerization, isolation of pure 17C (mixture: 49 mg, 25%
yield) was not possible, and only 17D (67 mg, 34% yield) was
characterized. 17D: 1H NMR (300 MHz, CD3OD) δ ppm 8.18 (d, J =
7.5 Hz, 1H), 7.69−7.59 (m, 1H), 7.48 (dd, J = 2.1, 8.2 Hz, 1H), 7.44−
7.34 (m, 1H), 7.21−7.13 (m, 1H), 7.10 (dd, J = 1.8, 8.2 Hz, 1H), 6.79
(d, J = 1.8 Hz, 1H), 5.09 (d, J = 11.1 Hz, 1H), 4.78 (d, J = 11.2 Hz,
1H), 3.72−3.54 (m, 1H), 3.50−3.35 (m, 1H), 2.85 (d, J = 9.8 Hz,
1H), 2.19 (d, J = 12.5 Hz, 1H), 2.02−1.82 (m, 5H), 1.82−1.69 (m,
3H), 1.67−1.13 (m, 7H), 1.02−0.82 (m, 1H); 13C NMR (75 MHz,
CD3OD) δ ppm 178.22, 167.68, 157.78 (d, JC−F = 249.08 Hz), 145.27,
19.19 Hz), 120.43(2C), 111.55, 72.58, 68.95, 63.25, 46.92, 32.45,
31.95, 25.77, 23.36, 22.18; ESI-MS m/z 582.17 (M+H)+.
Diastereomer Composition Analysis. The crude products from
the removal of the chiral auxiliary from 12B−18B and 21B−29B,
separately, were dissolved in 3:1 MeOH:H2O and 0.1% TFA. An
aliquot (250 μL) was immediately analyzed by analytical reverse-phase
UPLC (mobile phase gradient of MeOH:H2O), and their
compositions are shown in Tables 1 and 2. The prepared crude
solutions were immediately purified by reverse-phase preparative
HPLC (mobile phase gradient of MeOH:H2O). Immediately upon
collecting the fraction corresponding to the C- or D-diastereomers, an
aliquot (250 μL) was taken and immediately analyzed by analytical
reverse-phase UPLC. The same samples were then analyzed each day
to determine their change in composition, and the data are shown in
Tables 3 and 4.
Preparation of Samples for Analysis of Stability in Cell
Culture Medium. 322,24 or 17D or 31 (3.0 mg) was dissolved in cell
growth medium (3 mL) containing 10% FBS, and the solution was
incubated at 37 °C. At approximately the same time each day, 0.25 mL
of the solution was taken into a 1.5 mL microcentrifuge tube, and
MeCN was added to make a total volume of 1 mL. The sample was
sonicated in a water bath for 2 min and then centrifuged at 14 000 rpm
for 10 min. The supernatant aliquot (250 μL) was taken and mixed
with 250 μL of water. The final sample was immediately analyzed by
analytical reverse-phase UPLC.
Fluorescence Prolarization (FP)-Based Protein Binding
Assay. The binding affinity of MDM2 inhibitors was determined by
an optimized, sensitive, and quantitative FP-based binding assay, using
a recombinant human His-tagged MDM2 protein (residues 1−118)
and a FAM tagged p53-based peptide as the fluorescent probe.
The design of the fluorescent probe was based upon a previously
reported high affinity p53-based peptidomimetic compound (5-FAM-
βAla-βAla-Phe-Met-Aib-pTyr-(6-Cl-LTrp)-Glu-Ac3c-Leu-Asn-
NH2).31 This tagged peptide was named as PMDM6-F. The
equilibrium dissociation constant (Kd) of PMDM6-F to the MDM2
protein was determined to be 1.4 0.3 nM by monitoring the total
fluorescence polarization of mixtures composed with the fluorescent
probe at a fixed concentration and the MDM2 protein with increasing
concentrations up to full saturation. Fluorescence polarization values
were measured using the Infinite M-1000 plate reader (Tecan U.S.,
Research Triangle Park, NC) in Microfluor 1 96-well, black, round-
bottom plates (Thermo Scientific). In the saturation experiments, 1
nM of PMDM6-F and increasing concentrations of proteins were
added to each well to a final volume of 125 μL in the assay buffer (100
mM potassium phosphate, pH 7.5, 100 μg/mL bovine γ-globulin,
0.02% sodium azide (Invitrogen), with 0.01% Triton X-100 and 4%
DMSO). Plates were mixed and incubated at room temperature for 30
min with gentle shaking to ensure equilibrium. The polarization values
in millipolarization units (mP) were measured at an excitation
wavelength of 485 nm and an emission wavelength of 530 nm. The Kd
value was then calculated by fitting the sigmoidal dose-dependent FP
increases as a function of protein concentrations using Graphpad
Prism 6.0 software (Graphpad Software, San Diego, CA).
IC50 and Ki values of tested compounds were determined in a dose-
dependent competitive binding experiment. Mixtures of 5 μL of the
tested compound with different concentrations in DMSO and 120 μL
of pre-incubated protein/fluorescent probe complex with fixed
concentrations in the assay buffer (100 mM potassium phosphate,
pH 7.5, 100 μg/mL bovine γ-globulin, 0.02% sodium azide, with 0.01%
Triton X-100) were added into assay plates and incubated at room
temperature for 30 min with gentle shaking. Final concentrations of
the protein and fluorescent probe in the competitive assays were 10
and 1 nM, respectively, and final DMSO concentration was 4%.
Negative controls containing protein/fluorescent probe complex only
(equivalent to 0% inhibition), and positive controls containing free
fluorescent probe only (equivalent to 100% inhibition), were included
in each assay plate. FP values were measured as described above. IC50
values were determined by nonlinear regression fitting of the sigmoidal
dose-dependent FP decreases as a function of total compound
concentrations using Graphpad Prism 6.0 software (Graphpad
137.11, 132.38, 129.50 (d, JC−F = 2.50 Hz), 128.89, 126.49 (d, JC−F
=
4.89 Hz), 123.42, 122.42, 122.22 (d, JC−F = 19.24 Hz), 121.92 (d, JC−F
= 12.12 Hz), 111.78, 73.30, 69.88, 67.72, 62.25, 46.75, 46.70, 34.31,
34.23, 32.10, 31.35, 30.98, 30.73, 25.34, 23.11, 21.69; ESI-MS m/z
560.58 (M+H)+.
(3′R,4′S,5′R)-6″-Chloro-4′-(3-chloro-2-fluorophenyl)-N-
((1R,4R)-4-hydroxycyclohexyl)-2″-oxodispiro[cycloheptane-
1,2′-pyrrolidine-3′,3″-indoline]-5′-carboxamide (18D, TFA
Salt). Starting with 18B (146 mg), compound 18D (56 mg, 43%
yield) was prepared according to the procedure described for the
preparation of 12D. The crude sample contains only 18D (Table 1,
1
entry 7). H NMR (300 MHz, CD3OD) δ ppm 8.22 (d, J = 7.6 Hz,
1H), 7.66−7.53 (m, 2H), 7.43−7.34 (m, 1H), 7.19−7.11 (m, 1H),
7.08 (dd, J = 1.9, 8.2 Hz, 1H), 6.79 (d, J = 1H), 5.10 (d, J = 11.3 Hz,
1H), 4.84 (d, J = 11.2 Hz, 1H), 3.69−3.53 (m, 1H), 3.49−3.35 (m,
1H), 3.13−2.99 (m, 1H), 2.41 (dd, J = 9.6, 15.6 Hz, 2H), 1.95−1.83
(m, 2H), 1.83−1.66 (m, 4H), 1.66−1.39 (m, 5H), 1.35−0.83 (m, 6H);
13C NMR (75 MHz, CD3OD) δ ppm 178.43, 167.16, 157.78 (d, JC−F
= 249.13 Hz), 145.18, 137.17, 132.37, 129.09 (d, JC−F = 2.68 Hz),
128.83, 126.44 (d, JC−F = 4.87 Hz), 123.45, 122.25 (d, JC−F = 19.13
Hz), 121.84, 121.69 (d, JC−F = 12.44 Hz), 111.81, 77.13, 69.89, 67.35,
61.75, 47.46, 47.41, 37.17, 35.11, 34.30, 34.23, 32.33, 31.29, 30.92,
30.80, 25.18, 23.65; ESI-MS m/z 574.33 (M+H)+.
4-((3′R,4′S,5′R)-6″-Chloro-4′-(3-chloro-2-fluorophenyl)-2″-
oxodispiro[cyclohexane-1,2′-pyrrolidine-3′,3″-indoline]-5′-
carboxamido)benzoic Acid (31, TFA Salt). CDI (525 mg, 3.24
mmol), DIEA (0.941 mL, 5.4 mmol), and DMAP (catalytic) were
added to a solution of the carboxylic acid intermediate 30 (500 mg,
1.08 mmol) dissolved in 1,2-dichloroethane, and the resulting solution
was heated to 40 °C. After 30 min, methyl 4-aminobenzoate (816 mg,
5.4 mmol) was added to the reaction, which was then heated at reflux.
After heating at reflux overnight, the solvent was removed, and the
crude product was purified by column chromatography to give 265 mg
(41% yield) of intermediate 31-methylester as a white solid. The
resulting methyl ester intermediate, 31-methylester (265 mg, 0.44
mmol), was dissolved in THF (10 mL) and MeOH (5 mL), and then
LiOH·H2O (56 mg, 1.33 mmol) and NaOH (53 mg, 1.33 mmol) were
added, followed by H2O (10 mL). After 2 h, TLC showed the reaction
was complete, and then 3 mL of TFA was added, the mixture was
stirred briefly, and the solvent was evaporated. The resulting oil was
redissolved in 3:1 MeOH:H2O and then purified by preparative
HPLC. The mobile phase was a gradient flow of solvent A (H2O with
0.1% TFA) and solvent B (MeOH with 0.1% TFA). The eluent was
lyophilized to give 235 mg (31% yield, 2 steps) of the title compound
31 as a white solid. 1H NMR (300 MHz, CD3OD) δ ppm 7.98 (d, J =
8.7 Hz, 2H), 7.75−7.61 (m, 3H), 7.52 (dd, J = 2.4, 8.2 Hz, 1H), 7.39−
7.29 (m, 1H), 7.16 (t, J = 8.0 Hz, 1H), 7.09 (dd, J = 1.8, 8.2 Hz, 1H),
6.78 (d, J = 7.8 Hz, 1H), 5.21 (d, J = 10.6 Hz, 1H), 4.94 (d, J = 10.6
Hz, 1H), 2.73 (d, J = 10.32 Hz, 1H), 2.11 (d, J = 13.6 Hz, 1H), 2.01−
1.47 (m, 6H), 1.32−1.09 (m, 2H); 13C NMR (75 MHz, CD3OD) δ
ppm 178.60, 169.24, 169.13, 157.71 (d, JC−F = 249.13 Hz), 145.23,
142.83, 136.68, 131.97, 131.91(2C), 129.54 (d, JC−F = 2.62 Hz),
129.03, 128.14, 126.23 (d, JC−F = 4.65 Hz), 123.17, 122.21 (d, JC−F
=
K
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Software, San Diego, CA). Ki values of competitive inhibitors were
obtained directly by nonlinear regression fitting as well, based upon
the Kd values of the probe to different proteins and concentrations of
the proteins and probes in the competitive assays.32,33
Notes
The authors declare the following competing financial
interest(s): Shaomeng Wang, Angelo Aguilar, Wei Sun, Liu
Liu, Jianfeng Lu, and Donna McEachern are inventors on the
MDM2 inhibitors described in this study, which have been
licensed by Ascenta and Sanofi from the University of
Michigan. Shaomeng Wang has also received research funding
from Ascenta and Sanofi and owns stocks in Ascenta.
Cell Growth Inhibition Assay. The SJSA-1 osteosarcoma tumor
cell line was purchased from the American Type Culture Collection
(ATCC). Cells were seeded in 96-well flat bottom cell culture plates at
a density of (3−4) × 103 cells/well, grown overnight, and then
incubated with compounds at different concentrations. The rate of cell
growth inhibition after treatment with different concentrations of a
compound was determined by assaying with a lactate dehydrogen-
asebased WST-8 assay (WST-8; Dojindo Molecular Technologies Inc.,
Gaithersburg, MD). WST-8 solution was added to each well to a final
concentration of 10%, and then the plates were incubated at 37 °C for
2−3 h. The absorbance of the samples was measured at 450 nm using
a TECAN ULTRA reader. The concentration of a compound that
inhibited cell growth by 50% (IC50) was calculated by comparing
absorbance of the untreated cells with the treated cells.22
ACKNOWLEDGMENTS
■
This work was supported by grants from the NCI/NIH
R01CA121279, University of Michigan Cancer Center grant
P30CA046592, Sanofi, and Ascenta Therapeutics Inc. We thank
Dr. G. W. A. Milne for editing of the manuscript.
Pharmacodynamic (PD) Study in SJSA-1 Xenograft Model.
Xenograft tumors were developed by subcutaneous injection of 5 ×
106 of SJSA-1 cancer cells with 50% Matrigel on the dorsal side of
SCID mice (from Charles River). When the tumors reached an
average volume of ∼100 mm3, mice were treated with a single dose of
vehicle control, 17D at 100 mg/kg, or 31 at 100 mg/kg, all via oral
gavage. Mice were sacrificed at indicated time points and tumors
harvested, which were then analyzed by Western blotting for p53
activation and apoptosis.
ABBREVIATIONS USED
■
C2, carbon-2; CAN, cerium ammonium nitrate; CDI, carbonyl-
diimidazole; DIEA, N,N-diisopropylethylamine; DMAP,
dimethylaminopyridine; EDCI, 1-ethyl-(3-(dimethylamino)-
propyl)carbodiimide; EtOAc, ethyl acetate; HOBt, hydroxy-
benzotriazole; TFA, trifluoroacetic acid; THF, tetrahydrofuran;
UPLC, ultra-performance liquid chromatography
For Western blot analysis, tumor tissues were lysed in ice-cold RIPA
buffer: 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1
mM EGTA, 1% sodium deoxycholate, 2.5 mM sodium pyrophosphate,
1 mM β-glycerophosphate, 1 mM sodium orthovanadate, and 1 μg/
mL leupeptin. The expressions of indicated proteins in the whole cell
lysates were detected by Western blot analysis using the following
antibodies: anti-p53 (OP43, Calbiochem), anti-MDM2 (sc-965, Santa
Cruz), anti-p21 (556431, BD Biosciences), anti-PARP (9542, Cell
Signaling Technology), anti-caspase-3 (AAP-113, Stressgen Biore-
agents), and HRP-conjugated anti-GAPDH (sc-5778, Santa Cruz).22
In Vivo Efficacy Study. Xenograft tumors (one tumor per mouse)
were developed by subcutaneous injection of 5 × 106 of SJSA-1 cancer
cells with 50% Matrigel on the dorsal side of SCID mice, purchased
from Charles River. When tumors reached an average volume of ∼100
mm3, mice were randomized into different groups of eight and were
treated with vehicle control, 17D at 200 mg/kg, 31 at 100 mg/kg daily
for 14 days via oral gavage. Tumor sizes and animal weights were
measured 3 times per week during treatment. Data are presented as
REFERENCES
■
(1) Hainaut, P.; Hollstein, M. p53 and human cancer: the first ten
thousand mutations. Adv. Cancer Res. 2000, 77, 81−137.
(2) Vousden, K. H.; Lane, D. P. p53 in health and disease. Nat. Rev.
Mol. Cell Biol. 2007, 8, 275−283.
(3) Momand, J.; Zambetti, G. P.; Olson, D. C.; George, D.; Levine,
A. J. The mdm-2 oncogene product forms a complex with the p53
protein and inhibits p53-mediated transactivation. Cell 1992, 69,
1237−1245.
(4) Freedman, D. A.; Wu, L.; Levine, A. J. Functions of the MDM2
oncoprotein. Cell. Mol. Life Sci. 1999, 55, 96−107.
(5) Bond, G. L.; Hu, W. W.; Levine, A. J. MDM2 is a central node in
the p53 pathway: 12 years and counting. Curr. Cancer Drug Targets
2005, 5, 3−8.
(6) Wu, X.; Bayle, J. H.; Olson, D.; Levine, A. J. The p53-mdm-2
autoregulatory feedback loop. Genes Dev. 1993, 7, 1126−1132.
(7) Juven-Gershon, T.; Oren, M. Mdm2: the ups and downs. Mol.
Med. 1999, 5, 71−83.
(8) Momand, J.; Wu, H. H.; Dasgupta, G. MDM2 - master regulator
of the p53 tumor suppressor protein. Gene 2000, 242, 15−29.
(9) Momand, J.; Jung, D.; Wilczynski, S.; Niland, J. The MDM2 gene
amplification database. Nucleic Acids Res. 1998, 26, 3453−3459.
(10) Shvarts, A.; Steegenga, W. T.; Riteco, N.; van Laar, T.; Dekker,
P.; Bazuine, M.; van Ham, R. C.; van der Houven van Oordt, W.;
Hateboer, G.; van der Eb, A. J.; Jochemsen, A. G. MDMX: a novel
p53-binding protein with some functional properties of MDM2.
EMBO J. 1996, 15, 5349−5357.
mean tumor volumes
SEM. Statistical analyses were performed
using two-way ANOVA and unpaired two-tailed t test, using Prism
(version 4.0, GraphPad, La Jolla, CA). P < 0.05 was considered
statistically significant.
ASSOCIATED CONTENT
■
S
* Supporting Information
Diastereoisomer assignment, detailed synthesis of compounds,
animal weights in efficacy experiments, crystal structures of
compounds 17B and 30, and a molecular string file (CSV) for
compounds reported in this study and their associated
biochemical binding data and cell growth inhibitory activity.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(11) Oliner, J. D.; Kinzler, K. W.; Meltzer, P. S.; George, D. L.;
Vogelstein, B. Amplification of a gene encoding a p53-associated
protein in human sarcomas. Nature 1992, 358, 80−83.
(12) 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 small-molecule
antagonists of MDM2. Science 2004, 303, 844−848.
AUTHOR INFORMATION
■
(13) Vassilev, L. T. p53 Activation by small molecules: application in
oncology. J. Med. Chem. 2005, 48, 4491−4499.
Corresponding Author
(14) Shangary, S.; Wang, S. Targeting the MDM2−p53 interaction
for cancer therapy. Clin. Cancer Res. 2008, 14, 5318−5324.
(15) Wang, S.; Zhao, Y.; Bernard, D.; Aguilar, A.; Kumar, S.
Targeting the MDM2−p53 Protein-Protein Interaction for New
Cancer Therapeutics. Top. Med. Chem. 2012, 8, 57−80.
*S.W.: E-mail shaomeng@umich.edu, phone 734-615-0362, fax
734-647-9647.
Author Contributions
§A.A. and W.S. contributed equally.
L
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(16) Zhao, Y.; Bernard, D.; Wang, S. Small Molecule inhibitors of
MDM2−p53 and MDMX−p53 interaction as new cancer therapeutics.
BioDiscovery 2013, 8.
(17) Carry, J. C.; Garcia-Echeverria, C. Inhibitors of the p53/hdm2
protein-protein interaction-path to the clinic. Bioorg. Med. Chem. Lett.
2013, 23, 2480−2485.
(18) Ding, K.; Wang, G. P.; Deschamps, J. R.; Parrish, D. A.; Wang,
S. M. Synthesis of spirooxindoles via asymmetric 1,3-dipolar
cycloaddition. Tetrahedron Lett. 2005, 46, 5949−5951.
(19) Ding, K.; Lu, Y.; Nikolovska-Coleska, Z.; Wang, G.; Qiu, S.;
Shangary, S.; Gao, W.; Qin, D.; Stuckey, J.; Krajewski, K.; Roller, P. P.;
Wang, S. Structure-based design of spiro-oxindoles as potent, specific
small-molecule inhibitors of the MDM2−p53 interaction. J. Med.
Chem. 2006, 49, 3432−3435.
RO2468 for Potential Clinical Development. ACS Med. Chem. Lett.
2014, 5, 124−127.
(31) Garcia-Echeverria, C.; Chene, P.; Blommers, M. J.; Furet, P.
Discovery of potent antagonists of the interaction between human
double minute 2 and tumor suppressor p53. J. Med. Chem. 2000, 43,
3205−3208.
(32) Wang, Z. X. An exact mathematical expression for describing
competitive binding of two different ligands to a protein molecule.
FEBS Lett. 1995, 360, 111−114.
(33) Zhang, R.; Mayhood, T.; Lipari, P.; Wang, Y.; Durkin, J.; Syto,
R.; Gesell, J.; McNemar, C.; Windsor, W. Fluorescence polarization
assay and inhibitor design for MDM2/p53 interaction. Anal. Biochem.
2004, 331, 138−146.
(20) Shangary, S.; Qin, D.; McEachern, D.; Liu, M.; Miller, R. S.;
Qiu, S.; Nikolovska-Coleska, 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, 3933−3938.
(21) Yu, S.; Qin, D.; Shangary, S.; Chen, J.; Wang, G.; Ding, K.;
McEachern, D.; Qiu, S.; Nikolovska-Coleska, Z.; Miller, R.; Kang, S.;
Yang, D.; Wang, S. Potent and orally active small-molecule inhibitors
of the MDM2−p53 interaction. J. Med. Chem. 2009, 52, 7970−7973.
(22) Zhao, Y.; Yu, S.; Sun, W.; Liu, L.; Lu, J.; McEachern, D.;
Shargary, S.; Bernard, D.; Li, X.; Zhao, T.; Zou, P.; Sun, D.; Wang, S. A
potent small-molecule inhibitor of the MDM2−p53 interaction (MI-
888) achieved complete and durable tumor regression in mice. J. Med.
Chem. 2013, 56, 5553−5561.
(23) Wang, S.; Sun, W.; Zhao, Y.; McEachern, D.; Meaux, I.; Barriere,
C.; Stuckey, J. A.; Meagher, J. L.; Bai, L.; Liu, L.; Hoffman-Luca, C. G.;
Lu, J.; Shangary, S.; Yu, S.; Bernard, D.; Aguilar, A.; Dos-Santos, O.;
Besret, L.; Guerif, S.; Pannier, P.; Gorge-Bernat, D.; Debussche, L.
SAR405838: An Optimized Inhibitor of MDM2−p53 Interaction That
Induces Complete and Durable Tumor Regression. Cancer Res. 2014,
74, 5855−5865.
(24) Zhao, Y.; Liu, L.; Sun, W.; Lu, J.; McEachern, D.; Li, X.; Yu, S.;
Bernard, D.; Ochsenbein, P.; Ferey, V.; Carry, J. C.; Deschamps, J. R.;
Sun, D.; Wang, S. Diastereomeric spirooxindoles as highly potent and
efficacious MDM2 inhibitors. J. Am. Chem. Soc. 2013, 135, 7223−
7234.
(25) Shu, L. H.; Li, Z. Z.; Gu, C.; Fishlock, D. Synthesis of a
Spiroindolinone Pyrrolidinecarboxamide MDM2 Antagonist. Org.
Process Res. Dev. 2013, 17, 247−256.
(26) Popowicz, G. M.; Czarna, A.; Wolf, S.; Wang, K.; Wang, W.;
Domling, A.; Holak, T. A. Structures of low molecular weight
inhibitors bound to MDMX and MDM2 reveal new approaches for
p53-MDMX/MDM2 antagonist drug discovery. Cell Cycle 2010, 9,
1104−1111.
(27) Popowicz, G. M.; Domling, A.; Holak, T. A. The structure-based
design of Mdm2/Mdmx−p53 inhibitors gets serious. Angew. Chem.,
Int. Ed. 2011, 50, 2680−2688.
(28) Gonzalez-Lopez de Turiso, F.; Sun, D.; Rew, Y.; Bartberger, M.
D.; Beck, H. P.; Canon, J.; Chen, A.; Chow, D.; Correll, T. L.; Huang,
X.; Julian, L. D.; Kayser, F.; Lo, M. C.; Long, A. M.; McMinn, D.;
Oliner, J. D.; Osgood, T.; Powers, J. P.; Saiki, A. Y.; Schneider, S.;
Shaffer, P.; Xiao, S. H.; Yakowec, P.; Yan, X.; Ye, Q.; Yu, D.; Zhao, X.;
Zhou, J.; Medina, J. C.; Olson, S. H. Rational design and binding mode
duality of MDM2−p53 inhibitors. J. Med. Chem. 2013, 56, 4053−4070.
(29) Ding, Q.; Zhang, Z.; Liu, J. J.; Jiang, N.; Zhang, J.; Ross, T. M.;
Chu, X. J.; Bartkovitz, D.; Podlaski, F.; Janson, C.; Tovar, C.; Filipovic,
Z. M.; Higgins, B.; Glenn, K.; Packman, K.; Vassilev, L. T.; Graves, B.
Discovery of RG7388, a potent and selective p53-MDM2 inhibitor in
clinical development. J. Med. Chem. 2013, 56, 5979−5983.
(30) Zhang, Z.; Chu, X. J.; Liu, J. J.; Ding, Q.; Zhang, J.; Bartkovitz,
D.; Jiang, N.; Karnachi, P.; So, S. S.; Tovar, C.; Filipovic, Z. M.;
Higgins, B.; Glenn, K.; Packman, K.; Vassilev, L.; Graves, B. Discovery
of Potent and Orally Active p53-MDM2 Inhibitors RO5353 and
M
dx.doi.org/10.1021/jm501541j | J. Med. Chem. XXXX, XXX, XXX−XXX