Angelo Aguilar1, Jianfeng Lu1, Liu Liu1, Ding Du1, Denzil Bernard1, Donna McEachern1, Sally Przybranowski1, Xiaoqin Li2, Ruijuan Luo2, Bo Wen2, Duxin Sun2, Hengbang Wang3,4, Jianfeng Wen3,4, Guangfeng Wang3,4, Yifan Zhai3,4, Ming Guo3,4, Dajun Yang3,4,5, Shaomeng Wang1. 1. University of Michigan Comprehensive Cancer Center, and Departments of Internal Medicine, Pharmacology, and Medicinal Chemistry, University of Michigan , 1600 Huron Parkway, Ann Arbor, Michigan 48109, United States. 2. Department of Pharmaceutical Sciences, College of Pharmacy, University of Michigan , Ann Arbor, Michigan 48109, United States. 3. Jiangsu Ascentage Biomed Development Inc. , China Medical City, Taizhou, Jiangsu 225300, China. 4. Suzhou Ascentage Pharma Inc. , Suzhou, Jiangsu 215123, China. 5. Department of Experimental Research, Sun Yat-Sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine , 651 Dongfeng Road East, Guangzhou, China.
Abstract
We previously reported the design of spirooxindoles with two identical substituents at the carbon-2 of the pyrrolidine core as potent MDM2 inhibitors. In this paper we describe an extensive structure-activity relationship study of this class of MDM2 inhibitors, which led to the discovery of 60 (AA-115/APG-115). Compound 60 has a very high affinity to MDM2 (Ki < 1 nM), potent cellular activity, and an excellent oral pharmacokinetic profile. Compound 60 is capable of achieving complete and long-lasting tumor regression in vivo and is currently in phase I clinical trials for cancer treatment.
We previously reported the design of spirooxindoles with two identical substituents at the carbon-2 of the pyrrolidine core as potent MDM2 inhibitors. In this paper we describe an extensive structure-activity relationship study of this class of MDM2 inhibitors, which led to the discovery of 60 (AA-115/APG-115). Compound 60 has a very high affinity to MDM2 (Ki < 1 nM), potent cellular activity, and an excellent oral pharmacokinetic profile. Compound 60 is capable of achieving complete and long-lasting tumor regression in vivo and is currently in phase I clinical trials for cancer treatment.
The tumor suppressor
function of p53 is compromised in essentially
all humancancers. In about half of humancancers, TP53, the gene encoding p53 protein, is mutated or deleted, and this
inactivates the tumor suppressor function of p53.[1] In the remaining 50%, p53 retains its wild-type status
but its function is effectively inhibited by the murine double minute
2 (MDM2) protein through a direct protein–protein interaction.[2−8] Through direct binding, the MDM2 protein blocks the transactivation
domain of p53, transports p53 from the nucleus to the cytoplasm, and
ubiquitinates p53 for proteasomal degradation.[6] Small-molecule inhibitors designed to block the MDM2–p53
interaction (hereafter called MDM2 inhibitors) can liberate the tumor
suppressor function of p53[9−13] and may have a promising therapeutic potential for cancer treatment.
Intense research efforts have resulted in advancement of several potent,
selective, non-peptide MDM2 inhibitors (1–5),[14−20] shown in Figure , into clinical development.
Figure 1
Chemical structures of representative MDM2 inhibitors
in clinical
trials.
Chemical structures of representative MDM2 inhibitors
in clinical
trials.Our laboratory has designed and
optimized spirooxindoles as a new
class of MDM2 inhibitors and has advanced compound 2 (MI-77301)
into clinical development.[18] One shortcoming
of 2 and its earlier analogues that we have found is
that they slowly isomerize in solution.[21,22] To overcome
this stability issue, we recently reported the design of new spirooxindoles
containing two identical substituents at C-2 of the pyrrolidine ring,
which can undergo a rapid and irreversible conversion to a single
diastereoisomer.[21,22] In the present study, we report
an extensive structure–activity relationship (SAR) study for
this class of MDM2 inhibitors. This study has led to the discovery
of 60 (AA-115/APG-115), a highly potent, chemically stable
and efficacious MDM2 inhibitor, which has entered clinical development
for cancer treatment (ClinicalTrials.gov identifier for 60: NCT02935907).
Results and Discussion
Our exploration
started with lead molecule 6(21) (Figure A) that
binds to MDM2 with Ki =
2.9 nM, inhibits the growth of the SJSA-1 cancer cell line with IC50 = 190 nM, and achieves strong tumor growth inhibition but
not complete tumor regression in an SJSA-1 osteosarcoma xenograft
model in mice.[21] In a search for superior
compounds, we have performed extensive modifications of 6.
Figure 2
(A) Chemical structure of compound 6 and five regions
(A–E) where chemical modifications were made. (B) Cocrystal
structure of 2 (light blue spheres, PDB code 5TRF) and model of the
binding mode of 6 (green sticks) in complex with MDM2.
(A) Chemical structure of compound 6 and five regions
(A–E) where chemical modifications were made. (B) Cocrystal
structure of 2 (light blue spheres, PDB code 5TRF) and model of the
binding mode of 6 (green sticks) in complex with MDM2.To guide our chemical modifications,
we modeled the structure of
compound 6 in a complex with MDM2 based upon the cocrystal
structure of compound 2 complexed with humanMDM2 (Figure B).[18] Our model shows that the oxindolephenyl, 3-chloro-2-fluorophenyl,
and cyclohexyl groups of 6 project into the Trp23, Leu26,
and Phe19p53 binding pockets of the MDM2 protein and the 4-hydroxyl
of the N-cyclohexylcarboxamide group forms
a hydrogen bond with Lys94 of MDM2 (Figure B). Guided by this model, we performed extensive
modifications on five regions of the molecule (labeled as sites A–E, Figure A).We began
by replacing the 4-hydroxycyclohexyl group at site D in
compound 6 with a methyl group, giving compound 7. Since compounds 6 and 7 have
similar binding affinities to MDM2, we have used either compound 6 or 7 as the template for modifications of site
A (Table ). Replacement
of the cyclohexyl group at site A with 4-piperidinyl group led to
compound 8, which has no appreciable binding up to 2
μM to MDM2. Methylation or acetylation of the amine group in 8 resulted in compound 9 or 10,
respectively, and these also have very weak affinities for MDM2. Replacement
of the cyclohexyl group at site A in 6 with 4-tetrahydro-2H-pyran generated compound 11, which has a
moderate binding affinity to MDM2 (Ki =
75.9 nM). Introduction of two fluorine substituents at the 4-position
of the cyclohexyl group in compound 6 resulted in compound 12, which has a good affinity to MDM2 (Ki = 9.8 nM) but is 4 times less potent than 6.
However, introduction of two methyl groups at the same 4-position
in 6 yielded compound 15, which has a Ki value of <1 nM to MDM2 and is several times
more potent than 6. Replacement of the cyclohexyl group
with a cyclobutyl group generated compound 13,[21] which is 6 times less potent than 6. Installation of methyl groups at the 3-position of the cyclobutyl
ring led to compound 14, which is equally potent as compound 6 and 7 times more potent than compound 13. Hence,
our SAR data for site A clearly show that a hydrophobic group is highly
desirable for achieving a strong binding affinity to MDM2, and cyclohexyl,
4,4′-dimethylcyclohexyl, and 3,3′-cyclobutyl groups
are the most preferred.
Table 1
Structure–Activity
Relationship
of Pyrrolidine-2-spirocyclohexyl Substituentc
Value
of one experiment.
Reference (21).
Values are average of at least two
tests.
Value
of one experiment.Reference (21).Values are average of at least two
tests.Next, we explored
modifications of the 3-chloro-2-fluorophenyl
and the oxindolephenyl substituents (sites B and C in Figure A, respectively). To improve
solubility, we inserted one or more nitrogen atoms into the aryl rings,
producing compounds 16, 17, 18, and 19 (Table ). Insertion of a nitrogen into the 3-chloro-2-fluorophenyl
substituent (ring B) gave 16, which has a Ki of 77 nM, 26 times weaker than 6. Interestingly,
although the Trp23 pocket is hydrophobic in nature, compounds 17, 18, and 19 containing a pyridinyl
or a pyrimidinyl oxindole group (ring C) are all quite potent with Ki = 13–14 nM. Our data, however, are
consistent with a previous study, which showed that a compound containing
a chloropyridyl oxindole group is a potent MDM2 inhibitor.[23] This previous study also showed that replacing
the oxindole with a thienopyrrolone led to potent MDM2 inhibitors.[23] We therefore synthesized compound 20 containing a thienopyrrolone, which has a Ki of 6.5 nM and is 2-fold less potent than 6.
We found that the thiophene group in 20 can be oxidized
during the oxidative removal of the chiral auxiliary, preventing further
investigation of this compound.
Table 2
SAR of the B and
C Aryl Ringsa
Values are average of at least two
tests.
Values are average of at least two
tests.Our modeling of 6 and the cocrystal structure of 2 (Figure B) showed that the
carbonyl of the amide (site E) forms a hydrogen
bond with His96 of the MDM2 protein but the amino group does not interact
with the MDM2 protein. We evaluated a series of five-membered heteroaromatic
rings as the amide bioisoteres (Table ). Among these replacements, we found that heteroaromatic
rings with carboxylic acid substituents, e.g., 21, 22, 26, and 27 have the best binding
affinities with Ki values between 1 and
3 nM. Unfortunately, these compounds show weak cell growth inhibitory
activity (IC50 > 2 μM, Table ) in the SJSA-1 cancer cells, probably because
the negatively charged acid group in these compounds decreases their
cell permeability. We therefore converted the acid group in compound 21 to an amide but found that the resulting compound 28 is 20 times less potent than 21.
Table 3
Carboxamide (E) Bioisostere Replacementc
Value of only one
experimnent.
NT = not tested.
Values are average of at least
two
experiments.
Value of only one
experimnent.NT = not tested.Values are average of at least
two
experiments.Our modeling
of 6 and the cocrystal structure of 2 also
showed that the 4-hydroxyl on the cyclohexylcarboxamide
substituent of 6 forms a hydrogen bond with Lys94 of
the MDM2 protein (Figure B). We synthesized a series of compounds containing either
a terminal carboxylic acid or a sulfone to investigate the effect
on binding (compounds 29–38) (Table ). These modifications
resulted in a series of compounds with high affinities. In particular,
compounds 31, 32, and 33 bind
to MDM2 with Ki < 1 nM and are more
than 3 times more potent than 6.
Table 4
SAR of
Carboxamide Substituent (D)d
Reference (21).
NT = not tested.
Value of one time testing.
Values are average of at least
two
tests.
Reference (21).NT = not tested.Value of one time testing.Values are average of at least
two
tests.We tested the cell
growth inhibitory activity of compounds 29–38 in the SJSA-1 cell line, and the
data are summarized in Table . Among this series of compounds, 32, 33, and 38 have the best antiproliferative activity with
IC50 values of 0.22, 0.15, and 0.24 μM, respectively.We further evaluated compound 33 for its pharmacodynamic
(PD) effect in the SJSA-1 xenograft tissue. Our PD data showed that
a single, oral administration of 33 at 100 mg/kg is very
effective in inducing upregulation of MDM2, p53, and p21 proteins
at both 3 and 6 h time-points in the SJSA-1 tumor tissue, indicating
strong activation of p53 (Figure A). Upon the basis of its promising PD data, we evaluated
compound 33 for its antitumor efficacy in the SJSA-1
xenograft model (Figure B,C). While compound 33 can effectively retard tumor
growth, it failed to achieve tumor regression. During the course of
our research, we also observed that compound 33 can decompose
in solution, particularly in cell culture media (panels A and B in Scheme ). To analyze the
potential structural features affecting the stability of 33, we determined the stability of compound 31 with a
flexible butanoic acid group, compound 6 with a 4-hydroxylcyclohexyl
group, and 33. Compound 6 had negligible
(<1%) decomposition (panel C in Scheme ). On the other hand, compound 31 showed 18% decomposition after 2 days in cell culture media, which
is greater than that observed for compound 33 (panel
D in Scheme ). On
the basis of these results, we proposed that the carboxylic group
assists in the decomposition of 33.
Figure 3
(A) Pharmacodynamics
(PD) effect of 33 in SJSA-1 tumors
in mice. Mice were given a single, oral dose of the vehicle, or 33 at 100 mg/kg and sacrificed at indicated time points. West
blotting was performed on harvested tumor tissue to probe MDM2, p53,
p21, PAPR, cleaved PARP (cl-PAPR) proteins. (B) Antitumor activity
of 33 in the SJSA-1 osteosarcoma tumor xenograft model
in mice. Compound 33 was administered via oral gavage
at 100 mg/kg daily for 14 days. (C) Body weight change of mice during
administration of 33.
Scheme 1
Proposed Mechanism of Decomposition for 33
(A) Stability of compound 33 in various solutions. Compound 33 was incubated
in CH3CN/H2O, MeOH/H2O, or cell culture
media, and the % composition of 33 in the solutions was
determined daily for 2 days by UPLC (ultraperformance liquid chromatography)
and plotted. (B, C, D) UPLC/MS spectrum of 33, 6, and 31, respectively, after incubation for
2 days in cell culture media. Peaks are labeled with retention time
followed by molecular weight.
(A) Pharmacodynamics
(PD) effect of 33 in SJSA-1 tumors
in mice. Mice were given a single, oral dose of the vehicle, or 33 at 100 mg/kg and sacrificed at indicated time points. West
blotting was performed on harvested tumor tissue to probe MDM2, p53,
p21, PAPR, cleaved PARP (cl-PAPR) proteins. (B) Antitumor activity
of 33 in the SJSA-1 osteosarcoma tumor xenograft model
in mice. Compound 33 was administered via oral gavage
at 100 mg/kg daily for 14 days. (C) Body weight change of mice during
administration of 33.
Proposed Mechanism of Decomposition for 33
(A) Stability of compound 33 in various solutions. Compound 33 was incubated
in CH3CN/H2O, MeOH/H2O, or cell culture
media, and the % composition of 33 in the solutions was
determined daily for 2 days by UPLC (ultraperformance liquid chromatography)
and plotted. (B, C, D) UPLC/MS spectrum of 33, 6, and 31, respectively, after incubation for
2 days in cell culture media. Peaks are labeled with retention time
followed by molecular weight.We next performed
further chemical modifications of 33 with the goal to
improve its chemical stability, cellular potency,
and in vivo efficacy.We investigated if replacement of the
cyclohexyl group with an
even more rigid aryl group can prevent the decomposition. A series
of compounds containing an N-arylcarboxamide substituent
were synthesized and tested (Table ). These compounds all display very high binding affinities
to MDM2, superior to that of 6 and equivalent to that
of 33. The cell growth inhibition of SJSA-1 cells by
these compounds was determined (Table ). Compounds 39 (MI-1061)[21] and 44 displayed the best cellular activity
among the compounds in this series and are 3 times more potent than
compound 33 in inhibition of cell growth in the SJSA-1
cell line. Compound 39 was found in our previous study[21] to be very stable in solution. Furthermore, 39 was capable of achieving partial tumor regression in the
SJSA-1 xenograft model as shown in our previous report.[21]
Table 5
SAR of (Aryl acid)carboxamidesd
Value
of one time testing.
NT
= not tested.
Reference (21).
Values are average of at least three
tests.
Value
of one time testing.NT
= not tested.Reference (21).Values are average of at least three
tests.We next combined
all the favorable modifications on sites A and
D and designed and synthesized compound 54 (Scheme ). Compound 54 has a very high binding affinity to MDM2 (IC50 = 2.4 nM, Ki < 1 nM, Scheme ) and potently inhibits cell
growth of SJSA-1 cells with IC50 = 13 nM (Scheme ).
Scheme 2
Design of Compound 54 by Combining the Optimal Substituents
of Sites A and D
Reference (21).
Design of Compound 54 by Combining the Optimal Substituents
of Sites A and D
Reference (21).We assessed the antitumor activity of compound 54 in
the SJSA-1 xenograft model in mice, with compound 2 included
as the control (Figure ). Compound 54 effectively inhibits tumor growth in
a dose-dependent manner compared to the vehicle control. While compound 54 at 50 mg/kg daily dosing effectively retards tumor growth,
at 100 mg/kg it achieves a maximum of 87% tumor regression during
the treatment. However, while compound 2 at 100 mg/kg
achieves complete tumor regression, compound 54 fails
to do so.
Figure 4
(A) Antitumor activity of compound 54, in comparison
to compound 2 in the SJSA-1 osteosarcoma tumor xenograft
model in mice. Compound 54 was administered via oral
gavage at 50 or 100 mg/kg daily for 14 days and compound 2 was administered via oral gavage at 100 mg/kg daily for 14 days.
(B) Changes in body weight of mice during animal dosing.
(A) Antitumor activity of compound 54, in comparison
to compound 2 in the SJSA-1 osteosarcoma tumor xenograft
model in mice. Compound 54 was administered via oral
gavage at 50 or 100 mg/kg daily for 14 days and compound 2 was administered via oral gavage at 100 mg/kg daily for 14 days.
(B) Changes in body weight of mice during animal dosing.Pharmacokinetic (PK) studies in rats (Table ) showed that with
oral administration, compounds 39 and 54 achieve considerably lower Cmax and
AUC values than compound 2, suggesting that further improvement
of the oral PK for compounds 39 and 54 is
needed in order to achieve stronger
in vivo antitumor activity. Since 39 and 54 achieve comparable tumor regression in the SJSA-1 xenograft model
and 39 is less hydrophobic than 54, we have
selected compound 39 for further optimization of its
PK properties and in vivo efficacy.
Table 6
Pharmacokinetics
Table for Comparison
of 39, 54, 56, 59, 60, and 2 in Rats
compound
39
54
56
59
60
2
po dose (mg/kg)
25
25
25
25
25
25
po Cmax (ng/mL)
1553
968
8234
4391
5453
4547
po AUC (h·ng/mL)
6799
10188
73603
35205
39083
17230
po T1/2 (h)
4.0
4.5
4.3
3.9
4.6
1.6
iv dose (mg/kg)
10
10
10
10
10
5
iv AUC (h·ng/mL)
8633
14767
82241
28825
38265
4630
iv CL (L h–1 kg–1)
1.16
0.690
0.119
0.352
0.257
1.03
iv Vss(L/kg)
2.14
2.45
0.734
1.10
1.25
2.98
F (%)
31.5
27.6
35.0
48.6
40.3
74.9
antitumor efficacy
86% regression
87%
regression
no regression
100% regression
100% regression
100% regression
In our attempt to further
improve the oral PK properties of compound 39 while retaining
its high binding affinity to MDM2 and cellular
potency, we investigated the replacement of the benzoic acid with
nonclassical benzoic acid mimetics such as a bicyclo[1.1.1]pentane-1-carboxylic
acid[24] or a bicyclo[2.2.2]octane-1-carboxylic
acid. These efforts led to the design and synthesis of compounds 55 and 56 (Scheme ). Compound 55 binds to MDM2 with a high
affinity (IC50 = 6.4 nM, Ki < 1 nM) but is 5 times less potent than compound 39 in inhibition of cell growth in the SJSA-1 cell line. In comparison,
compound 56 has a high binding affinity to MDM2 (IC50 = 3.7 nM, Ki < 1 nM) and
is as potent as compound 39 in inhibition of cell growth
in the SJSA-1 cell line (Scheme and Table ).
Scheme 3
Replacement of the Benzoic Acid in Compound 39 with
Nonclassical Benzoic Acid Bioisosteres
Table 7
SAR of Benzoic Acid Nonclassical Bioisosteres
and Alkylation of the Pyrrolidine Nitrogenc
Value
from one experiment.
Reference (21).
Values are average of at least two
independent experiments.
Value
from one experiment.Reference (21).Values are average of at least two
independent experiments.A PK study in rats showed that 56 has a higher plasma
exposure than compound 2 based upon both the Cmax (8234 vs 4547 μg/L) and AUC values
(73603 vs 17230 h·μg/L). However, our PD experiment showed
that oral administration of compound 56 at 100 mg/kg
induces only modest p53 activation and no cleavage of PARP and caspase-3
in the SJSA-1 xenograft tumor tissue in mice (Figure A). Consistent with the PD data, compound 56 at 100 mg/kg administered daily for 14 days demonstrates
only a very moderate tumor growth inhibition in the SJSA-1 xenograft
model in mice (Figure A). Hence, the antitumor activity of 56 is much inferior
to that achieved by 2, 39, and 54. Because of its excellent in vitro cellular potency and excellent
plasma exposure, the modest p53 activation and antitumor activity
achieved by compound 56 suggests that this compound has
a poor penetration into the xenograft tumor tissue. Accordingly, we
investigated strategies with which to improve the tissue penetration
while retaining high binding affinity to MDM2, potent cellular activity,
and good oral bioavailability of compound 56.
Figure 6
Pharmacodynamic (PD) analysis of the effect
of MDM2 inhibitors
in SJSA-1 tumors. Mice bearing SJSA-1 tumors were dosed with a single
oral dose of 56, 57, 58, 59, 60 and 2 at 100 mg/kg and the
tumors were harvested at 6 and 24 hours for western blot analysis.
(A, B) PD of 56, 57 and 58 show
modest activation of p53 as seen with the low levels of p53, MDM2,
and p21; (C) Compounds 59, 60 and 2 show robust activation of p53 as seen with increased levels
of p53, MDM2, ubiquitinated MDM2 (ub-MMD2) and p21, and induce strong
apoptosis as seen with increased levels of cleaved PARP (cl-PAPR)
and cleaved caspase-3 (cl-Casp3).
Figure 5
(A) Antitumor
activity of compounds 39 and 56 in the SJSA-1
osteosarcoma xenograft model in mice. (B) Body weight
change of mice during administration period.
(A) Antitumor
activity of compounds 39 and 56 in the SJSA-1
osteosarcoma xenograft model in mice. (B) Body weight
change of mice during administration period.We investigated three strategies to improve the tissue penetration:
(1) decreasing the lipophilicity of the molecule; (2) reducing the
acidity of carboxylic acids; and (3) increasing the basicity of the
nitrogen atom in the pyrrolidine core.[25]Fluorine substitution on aliphatic groups is known to decrease
lipophilicity.[26−28] We therefore synthesized compound 57 with a 4,4-difluorocyclohexyl group at carbon-2 of the pyrrolidine
core (Table ). For
the second strategy, we replaced the carboxylic acid group with an
acyl sulfonamide bioisostere, which resulted in 58 (Table ). For the third strategy,
we installed a methyl or ethyl group on the nitrogen of the pyrrolidine
core, which led to 59 and 60, respectively
(Table ).Compound 57 binds to MDM2 with a high affinity but
is less potent than 56 and is 2 times less potent than 56 in inhibition of cell growth in the SJSA-1 cell line (Table ). Compound 58 is 2 times less potent than 56 in binding
to MDM2 and is 4 times less potent than 56 in inhibition
of cell growth in the SJSA-1 cell line (Table ). The PD study showed that oral administration
of 57 or 58 at 100 mg/kg only has a modest
effect on activation of p53 and induction of cleavage of PARP and
caspase-3 (Figure B).Pharmacodynamic (PD) analysis of the effect
of MDM2 inhibitors
in SJSA-1 tumors. Mice bearing SJSA-1 tumors were dosed with a single
oral dose of 56, 57, 58, 59, 60 and 2 at 100 mg/kg and the
tumors were harvested at 6 and 24 hours for western blot analysis.
(A, B) PD of 56, 57 and 58 show
modest activation of p53 as seen with the low levels of p53, MDM2,
and p21; (C) Compounds 59, 60 and 2 show robust activation of p53 as seen with increased levels
of p53, MDM2, ubiquitinated MDM2 (ub-MMD2) and p21, and induce strong
apoptosis as seen with increased levels of cleaved PARP (cl-PAPR)
and cleaved caspase-3 (cl-Casp3).In comparison, compounds 59 and 60 bind
to MDM2 with a high binding affinity (Ki < 1 nM, Table ). Both compounds potently inhibit cell growth in the SJSA-1 cell
line with IC50 values of 70 and 60 nM, respectively, and
are thus as potent as compound 56 (Table ). PD experiments showed that a single, oral
dose of 59 and 60 at 100 mg/kg achieves
strong p53 activation, with the effect persisting for 24 h in the
SJSA-1 xenograft tumor tissue (Figure C). Compounds 59 and 60 also
achieve strong apoptosis induction in the SJSA-1 xenograft tumor,
based upon cleavage of PARP and caspase-3 (Figure C).We next tested 59 and 60 for the antitumor
activity in the SJSA-1 xenograft model (Figure ). Compound 59 or 60 orally administered at 30 mg/kg daily for 14 days effectively inhibits
tumor growth. Compound 59 or 60 orally administered
at 100 mg/kg daily for 14 days effectively achieves complete and long-lasting
tumor regression. In comparison, compound 56 at 100 mg/kg
inhibits tumor growth only moderately in the same experiment (Figure A). Compounds 56, 59, and 60 cause no or minimal
weight loss during the entire experiment (Figure B).
Figure 7
(A) Antitumor activity of compound 59, 60 in comparison to 56 in the SJSA-1
osteosarcoma xenograft
model in mice. Compound 59 and 60 were administered
at 30 mg/kg or 100 mg/kg daily for 14 days and 56 at
100 mg/kg daily for 14 days. (B) Animal weight changes of mice during
the experiment.
(A) Antitumor activity of compound 59, 60 in comparison to 56 in the SJSA-1
osteosarcoma xenograft
model in mice. Compound 59 and 60 were administered
at 30 mg/kg or 100 mg/kg daily for 14 days and 56 at
100 mg/kg daily for 14 days. (B) Animal weight changes of mice during
the experiment.Our PK studies in rats
with oral administration showed that 59 and 60 achieve a Cmax value similar to that
of 2 but have a 2 times higher
AUC than compound 2 (Table ).We evaluated compounds 56, 59, and 60 for their activity and selectivity
in a number of humancancer cell lines of different tumor types (Table ). Consistent with their mechanism of action,
these three compounds potently inhibit cell growth in cancer cell
lines with wild-type p53 and display outstanding selectivity in cancer
cell lines with deleted p53. Compound 60 is the most
potent compound, achieving IC50 values of 38, 18, and 104
nM in the RS4;11 acute leukemia, LNCaPprostate cancer, and HCT116colon cancer cell lines, respectively.
Table 8
Cell Growth
Inhibition Activity of 56, 59, and 60 in Different Human
Cancer Cell Linesa
cell growth inhibition (IC50)
cell line
tumor type
p53 status
56
59
60
SJSA-1
osteosarcoma
wild-type
89 ± 33 (nM)
70 ± 21(nM)
60 ± 22 (nM)
Saos2
osteosarcoma
null
26.7 ± 5.1 (μM)
25 ± 6 (μM)
22.7 ± 4.7 (μM)
RS4;11
leukemia
wild-type
62 ± 26 (nM)
56 ± 18 (nM)
38 ± 5 (nM)
LNCaP
prostate cancer
wild-type
36 ±
19 (nM)
30 ±
15 (nM)
18 ±
13 (nM)
PC3
prostate cancer
null
12.3 ± 2.5 (μM)
24 ± 5 (μM)
22 ± 7.2 (μM)
HCT116
colon cancer
wild-type
137 ± 31 (nM)
117 ± 33 (nM)
104 ± 36 (nM)
HCT116 p53–/–
colon cancer
knockout
14 ± 2 (μM)
18 ± 8 (μM)
8 ± 1 (μM)
IC50 values are average
of at least three independent experiments.
IC50 values are average
of at least three independent experiments.We next evaluated compound 60 for its
antitumor activity
in the RS4;11 acute leukemia xenograft model in mice. Compound 60 effectively inhibits tumor growth in a dose-dependent manner
and achieves partial tumor regression at 100 mg/kg, daily, oral administration
for 14 days, or at 200 mg/kg weekly dosing for 3 weeks (Figure ). Compound 60 is also well tolerated at all doses and schedules in this experiment.
Figure 8
(A) Antitumor
activity of compound 60 in the RS4;
11 acute lymphoblastic leukemia xenograft model in mice. Compound 60 was administered at 50 mg/kg, 75 mg/kg, or 100 mg/kg daily
for 14 days and 200 mg/kg weekly for 3 weeks. (B) Animal body weight
of mice during administration.
(A) Antitumor
activity of compound 60 in the RS4;
11 acute lymphoblastic leukemia xenograft model in mice. Compound 60 was administered at 50 mg/kg, 75 mg/kg, or 100 mg/kg daily
for 14 days and 200 mg/kg weekly for 3 weeks. (B) Animal body weight
of mice during administration.We evaluated the chemical stability of compound 60 in three different solutions and found that this compound is very
stable in solutions for a period of 7 days (Figure ). In comparison, compound 2 slowly isomerizes in solutions and retains only 93–96% purity
after 7 days (Figure ).
Figure 9
Stability comparison of compounds 2 and 60 in (A) 1:1 CH3CN to H2O; (B) 1:1 MeOH to H2O; (C) cell culture media.
Stability comparison of compounds 2 and 60 in (A) 1:1 CH3CN to H2O; (B) 1:1 MeOH to H2O; (C) cell culture media.
Chemistry
To explore achiral substituents in carbon-2 of
the pyrrolidine
core, the compounds in Table were synthesized using the method[21] outlined in Scheme . The azomethine ylide generated in situ by reaction of the chiral
amine (62) with the cyclic ketone (63) was
allowed to react with 61 in a 1,3-dipolar cycloaddition
reaction, generating the chiral spirooxindoles (6a–15a). Reaction of this morpholine (6a–15a) with an aliphatic amine led to the open-ring compound
(6b–15b). Upon oxidation with ceric
ammonium nitrate (CAN), 6c–15c were
formed as a mixture of two diastereoisomers. “Aging”
in acetonitrile/water resulted in formation of the single diastereoisomers 6–15.
Reagents and conditions:
(a)
toluene, reflux, 2 h; (b) R1NH2, THF, reflux
overnight; (c) CH3CN/H2O (1:1), CAN, 15 min;
(d) CH3CN/H2O, overnight.Scheme shows the
method by which the B and C aryl rings were replaced. First the intermediates 16b–20b were synthesized by condensation
of the pyrrolidones (16a–20a) with
an arylaldehyde. Subsequently, the compounds (16–20) in Table were obtained by using the intermediates (16b–20b), which are equivalent to 61, in the synthetic
method outlined in Scheme .
Scheme 5
Synthesis of Compounds 16–20
Reagents and conditions: (a)
MeOH, piperidine, reflux.
Synthesis of Compounds 16–20
Reagents and conditions: (a)
MeOH, piperidine, reflux.Compounds containing
various five-membered heteroaromatic rings
as bioisostere replacements for the carboxamide group E were synthesized
as outlined in Schemes and 7. The N-methylpyrrolidine
(65) was generated by reductive amination of 64 with paraformaldehyde and NaBH(OAc)3. Hydrolysis of the
methyl ester (65) was followed by conversion of the acid
(66) to the corresponding carboxamide (67). Refluxing of 67 with Lawesson’s reagent produced
the thioamide (68) which was reacted with an α-bromoketone
in a Hantzsch thiazole synthesis reaction, generating the thiazoles 69 and 70.[29] Basic
hydrolysis of the esters produced the acid analogues 21 and 27. Compound 21 was further elaborated
by amide coupling, generating the carboxamide (28).
Scheme 6
Synthesis of Compounds 21, 27, and 28
Reagents and conditions: (a)
DCM/AcOH (1:1), paraformaldehyde, NaBH(OAc)3; (b) THF/MeOH/H2O, LiOH·H2O; (c) DCE, CDI, DIEA, 50 °C
30 min, then NH4OH; (d) DCM, Lawesson’s reagent,
reflux overnight; (e) (1) THF, BrCH2COR, Et3N, reflux overnight, (2) DME, TFAA, pyridine, −20 °C
to rt, 30 min; (f) THF/MeOH/H2O (1:1:1), NaOH, rt, prep-HPLC.
Scheme 7
Synthesis of Compounds 22–26
Reagents and conditions: (a)
DCE, CDI, DIEA, DMAP, 50 °C for 30 min, then 71, 74, 76, 78, and reflux; (b) (1)
THF, Deoxo-Fluor, −20 °C for 30 min, (2) BrCCl3, DBU at rt overnight; (c) THF/H2O (2:1), LiOH·H2O, rt for 30 min, then preparative HPLC; (d) pyridine, 100
°C for 4 h; (e) DCM, PPh3, I2, Et3N, rt; (f) DCM, Lawesson’s reagent, reflux overnight.
Synthesis of Compounds 21, 27, and 28
Reagents and conditions: (a)
DCM/AcOH (1:1), paraformaldehyde, NaBH(OAc)3; (b) THF/MeOH/H2O, LiOH·H2O; (c) DCE, CDI, DIEA, 50 °C
30 min, then NH4OH; (d) DCM, Lawesson’s reagent,
reflux overnight; (e) (1) THF, BrCH2COR, Et3N, reflux overnight, (2) DME, TFAA, pyridine, −20 °C
to rt, 30 min; (f) THF/MeOH/H2O (1:1:1), NaOH, rt, prep-HPLC.
Synthesis of Compounds 22–26
Reagents and conditions: (a)
DCE, CDI, DIEA, DMAP, 50 °C for 30 min, then 71, 74, 76, 78, and reflux; (b) (1)
THF, Deoxo-Fluor, −20 °C for 30 min, (2) BrCCl3, DBU at rt overnight; (c) THF/H2O (2:1), LiOH·H2O, rt for 30 min, then preparative HPLC; (d) pyridine, 100
°C for 4 h; (e) DCM, PPh3, I2, Et3N, rt; (f) DCM, Lawesson’s reagent, reflux overnight.Several five-membered heteroaromatic rings were formed
from the
carboxylic acid (66). The carboxylic acid (66) was activated with CDI (1,1′-carbonyldiimidazole),
and this was followed by addition of commercially available compounds 71, 74, 76, 78, resulting
in the formation of intermediates 72, 75, 77, 79, respectively. Cyclization to
the respective five-membered heteroaromatic compounds (73,[30]23,[31]24,[32]25, 80) was accomplished under the conditions outlined
in Scheme . The carboxylic
acid compounds 22 and 26 were produced by
hydrolysis of the esters in compounds 73 and 80, respectively.The compounds in Tables and 5 were obtained
by the synthesis
outlined in Scheme . Activation of the carboxylic acid (29) by HATU (1-[Bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxide
hexafluorophosphate) followed by addition of an alkylamine gave the N-alkylcarboxamide intermediates 30, 36, 38, 31–35, 37, 55, and 56. When N-arylamines were used, HATU failed to accomplish the coupling,
but the coupling was successful when the acid was activated with Ph2POCl [17] and produced the N-arylcarboxamides 39–48, 50, and 51. Basic hydrolysis of the esters
produced the carboxylic acid compounds listed in Tables and 5. Further elaboration of the acids (56 to give 58; 39 to give 49 and 52; and 44 to give 53) was accomplished by
CDI activation of the carboxylic acid followed by reaction with the
respective amine.
Scheme 8
Synthesis of Compounds in Tables and 5
Reagents
and conditions: (a)
DCM, HATU, DIEA, alkylamine; (b) THF/MeOH/H2O (1:1:1),
LiOH·H2O for alkyl esters or NaOH for aryl esters;
(c) DCM, Ph2POCl, DIEA 30 min, then arylamine and DMAP;
(d) DCE, CDI, DIEA, DMAP, 50 °C for 30 min, then R2NH2 and reflux.
Synthesis of Compounds in Tables and 5
Reagents
and conditions: (a)
DCM, HATU, DIEA, alkylamine; (b) THF/MeOH/H2O (1:1:1),
LiOH·H2O for alkyl esters or NaOH for aryl esters;
(c) DCM, Ph2POCl, DIEA 30 min, then arylamine and DMAP;
(d) DCE, CDI, DIEA, DMAP, 50 °C for 30 min, then R2NH2 and reflux.Reductive amination
of 56 with paraformaldehyde or
acetaldehyde produced 59 and 60 as shown
in Scheme .
Scheme 9
Synthesis
of Compound 59 and 60
Reagents
and conditions: (a)
DCM/AcOH (1:1), paraformaldehyde or acetaldehyde, NaBH(OAc)3, rt overnight.
Synthesis
of Compound 59 and 60
Reagents
and conditions: (a)
DCM/AcOH (1:1), paraformaldehyde or acetaldehyde, NaBH(OAc)3, rt overnight.
Conclusion
In
this study, we have performed an extensive SAR study on spirooxindoles
containing two substituents at the carbon-2 of the pyrrolidine core
as MDM2 inhibitors. Extensive modifications in five different regions
of the molecule have led to discovery of a number of potent and highly
efficacious MDM2 inhibitors. In particular, compound 60 has a very high binding affinity to MDM2 (Ki < 1 nM), potently activates wild-type p53, inhibits cell
growth with low nanomolar IC50 values in humancancer cell
lines carrying wild-type p53, and demonstrates an outstanding cellular
selectivity over humancancer cell lines with deleted p53. Compound 60 is very stable in solutions, has excellent oral pharmacokinetics,
and effectively activates p53 in the SJSA-1 xenograft tumor tissue
in mice following a single oral administration. Significantly, 60 achieves complete and long-lasting regression of the SJSA-1
xenograft tumors in mice and demonstrates strong antitumor activity
in the RS4;11 acute leukemia model at well tolerated dose schedules.
Compound 60 has been advanced into phase I clinical development
for the treatment of humancancer.
Experimental
Section
General Information
Unless otherwise stated, all commercial
reagents were used as supplied without further purification and all
reactions were performed under a nitrogen atmosphere in dry solvents
under anhydrous conditions. NMR spectra were obtained on either a
Bruker 300 UltraShield spectrometer at a 1H frequency of
300 MHz and 13C frequency of 75 MHz or Bruker 400 Ascend
spectrometer at a 1H frequency of 400 MHz and 13C frequency of 100 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 mm ×
100 mm reverse phase column. The mobile phase was a gradient of solvent
A (H2O with 0.1% TFA) and solvent B (MeOH with 0.1% TFA
or MeCN with 0.1% TFA) at a flow rate of 40 mL/min and 1%/2 min increase
of solvent B or a flow rate of 60 mL/min and 1%/min increase of solvent
B. All final compounds have purity of ≥95% as determined by
Waters ACQUITY UPLC using reverse phase column (SunFire, C18-5 μm,
4.6 mm × 150 mm) and a solvent gradient of A (H2O
with 0.1% of TFA) and solvent D (MeOH with 0.1% of TFA) or A (H2O with 0.1% of TFA) and solvent B (CH3CN with 0.1%
of TFA). ESI mass spectrum analysis was performed on a Thermo-Scientific
LCQ Fleet mass spectrometer. No PAINS liability was found in any of
the compounds presented as determined by analysis in the online filter
SmartsFilter (http://pasilla.health.unm.edu/tomcat/biocomp/smartsfilter).Compounds 6, 13, 29, 39 and intermediate 64 were reported
previously.[21] Compounds 16b–20b were synthesized as reported previously.[23]Compounds 7–20 were synthesized
according to our reported method[21] which
is described below for the synthesis of compound 6.
In a round-bottom
flask, (E)-6-chloro-3-(3-chloro-2-fluorobenzylidene)indolin-2-one
(500 mg, 1.62 mmol), (5R,6S)-5,6-diphenyl-2-morpholinone
(492 mg, 1.94 mmol), and cyclohexanone (4.86 mmol) were suspended
in toluene (10 mL) and heated at reflux for 2 h at 140 °C. Then
the reaction was allowed to cool to room temperature and the solvent
was removed by rotoevaporation. The crude product was purified by
column chromatography (the compound was eluted with 100% DCM) to give
665 mg (64% yield) of 6a as a pale yellow solid. trans-4-Aminocyclohexanol (475 mg, 4.13 mmol) was added
to a solution of 6a (530 mg, 0.826 mmol) in THF (20 mL)
and heated at reflux overnight. Then the reaction was cooled to room
temperature and the solvent was removed by rotoevaporation. The crude 6b was purified by column chromatography to produce 224 mg
of 6b. Cerium ammonium nitrate (324 mg, 0.592 mmol) was
added to a solution of the resulting 6b (224 mg, 0.296
mmol) in MeCN (6 mL) and stirred for 5 min at room temperature, then
H2O (6 mL) was added. After stirring the reaction 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, filtered through Celite and
the solvent was removed by rotoevaporation to produce crude 6c as a mixture of diastereisomers. The crude material was
dissolved in a 3:1 mixture of MeOH/H2O that was acidified
with TFA and aged. After 2 days this solution was purified by preparative
HPLC. The combined fractions of the pure compound were concentrated
and then redissolved in a minimum amount of MeCN, H2O was
added, and the solution was frozen and lyophilized to produce the
TFA salt of 6 (116 mg, 70% yield) as a white powder.
Chemical data were reported previously.[21]
Ethyl 3-bromo-2-oxopropanoate (48
mg, 0.244 mmol) and triethylamine (0.033 mL, 0.244 mmol), were added
to a solution of 68 (30 mg, 0.061 mmol) in THF (3 mL)
and refluxed overnight. Then the reaction was cooled, and the solvent
was removed. The crude product was redissolved in DME (3 mL), and
to this solution, at −20 °C, was added trifluoracetic
anhydride (0.025 mL, 0.183 mmol) and pyridine (0.025 mL, 0.305 mmol),
and then the reaction was allowed to warm to rt. After 30 min, saturated
sodium bicarbonate solution was added and the reaction was extracted
with EtOAc. The EtOAc solution was dried over sodium sulfate, filtered,
and purified by column chromatography to produce 18 mg of compound 69. Lithium hydroxide monohydrate (50 mg, 1.19 mmol) was added
to a solution of 69 (18 mg, 0.031 mmol) in THF/MeOH/H2O (1:1:1, 3 mL). After 2 h the reaction was quenched with
ammonium chloride, and brine was added and extracted with EtOAc. The
EtOAc solution was dried over sodium sulfate, filtered, concentrated
and the crude was purified by preparative HPLC to produce compound 21 (TFA salt). 1H NMR (300 MHz, CD3OD)
δ ppm 8.34 (s, 1H), 7.70 (t, 1H, J = 6.8 Hz),
7.53 (d, 1H, J = 7.1 Hz), 7.27 (t, 1H, J = 7.5 Hz), 7.13–7.03 (m, 2H), 6.73 (d, 1H, J = 1.8 Hz), 5.78–5.59 (m, 1H), 4.81–4.68 (m, 1H), 3.10
(s, 3H), 2.51–2.38 (m, 1H), 2.25 (d, 1H, J = 15.1 Hz), 2.13–1.94 (m, 1H), 1.87–1.10 (m, 7H);
ESI-MS m/z 560.33 (M + H)+.
General Procedure A: Amide Coupling to Compound 66
CDI (3 equiv), DIEA (5 equiv), DMAP (1 equiv) were added
to a suspension of compound 66 (1 equiv) in 1,2-dichloroethane
and heated at 50 °C. After 30 min, the respective amine (5 equiv)
was added and the reaction was refluxed overnight. Then the solvent
was removed and the crude product was purified to produce the carboxamide.
Compound 72 was produced
according to general procedure A. At −20 °C, Deoxo-Fluor
(50 μL) was added to a solution of 72 (51 mg) in
THF (3 mL). After 30 min bromotrichloromethane (500 μL)
and DBU (400 μL) were added and the reaction was allowed to
warm to room temperature. After 5 h the reaction was quenched with
saturated sodium bicarbonate (stirred for 5 min) and then extracted
with EtOAc. The EtOAc solution was dried over sodium sulfate, filtered,
and purified by column chromatography to produce 34 mg of compound 73. Lithium hydroxide monohydrate (60 mg, 1.43 mmol) was added
to a solution of 73 (34 mg, 0.061 mmol) in THF/H2O (1:1, 4 mL). After 2 h the reaction was quenched with TFA,
concentrated and the crude was redissolved in 3:1 MeOH/H2O and purified by HPLC to produce 22 (TFA salt) as a
white powder. 1H NMR (400 MHz, CD3OD) δ
ppm 8.50 (s, 1H), 7.62–7.53 (m, 2H), 7.24 (t, 1H, J = 7.3 Hz), 7.07 (dd, 1H, J = 1.7, 8.2 Hz), 7.03
(t, 1H, J = 8.1 Hz), 6.73 (d, 1H, J = 1.8 Hz), 5.34 (d, 1H, J = 10.3 Hz), 5.02 (d,
1H, J = 11.6 Hz), 2.97 (s, 3H), 2.33 (d, 1H, J = 13.1 Hz), 2.15 (d, 1H, J = 14.1 Hz),
1.95–1.85 (m, 1H), 1.74–1.44 (m, 5H), 1.38–1.26
(m, 1H), 1.19–1.05 (m, 1H); ESI-MS m/z 544.42 (M + H)+.
Starting with 44 mg of compound 66, 25 mg of 79 was obtained using general procedure
A. Cyclization to generate 80 (13 mg) was accomplished
according to the same procedure used to obtain 24. Lithium
hydroxide monohydrate (30 mg, 0.715 mmol) was added to a solution
of 80 (13 mg, 0.023 mmol) in THF/H2O (1:1,
2 mL). After 2 h the reaction was quenched with TFA, concentrated
and the crude was redissolved in 3:1 MeOH/H2O and purified
by HPLC to produce 26 (TFA salt) as a white powder. 1H NMR (400 MHz, CD3OD) δ ppm 7.60 (t, 1H, J = 7.2 Hz), 7.52 (dd, 1H, J = 2.7, 8.1
Hz), 7.21 (t, 1H, J = 8.3 Hz), 7.07–6.98 (m,
2H), 6.70 (d, 1H, J = 1.9 Hz), 5.07 (d, 1H, J = 10.8 Hz), 4.91 (d, 1H, J = 11.7 Hz),
2.83 (s, 3H), 2.59 (s, 3H), 2.31–2.23 (m, 1H), 2.12–2.04
(m, 1H), 1.85–1.02 (m, 8H); ESI-MS m/z 558.75 (M + H)+.
Starting with acid 21 in place
of 66, compound 28 was produced using general
procedure A. Crude 28 was purified by HPLC to produce 28 as its TFA salt. 1H NMR (400 MHz, CD3OD) δ ppm 8.11 (s, 1H), 7.75 (t, 1H, J = 7.2
Hz), 7.42 (d, 1H, J = 8.1 Hz), 7.26 (t, 1H, J = 8.46 Hz), 7.09 (t, 1H, J = 7.83), 7.01
(dd, 1H, J = 1.7, 8.1 Hz), 6.71 (d, 1H, J = 1.8 Hz), 3.00 (s, 3H), 2.36–2.16 (m, 2H), 2.09–2.00
(m, 1H), 1.91–1.52 (m, 5H), 1.38–1.24 (m, 1H), 1.21–1.06
(m, 1H); ESI-MS m/z 559.42 (M +
H)+.
General Procedure B: Amide Coupling of Alkylamines
HATU (2 equiv) and DIEA (4 equiv) were added to a suspension of
acid
compound 29 (1 equiv) in DCM. After 10 min, alkylamine
(4 equiv) was added. After 12 h, the reaction was concentrated and
purified by column chromatography to produce alkyl amide compounds.
General Procedure C: Hydrolysis of Alkyl Esters
LiOH·H2O (3 equiv) was added to a solution of ester (1 equiv) in
THF/MeOH/H2O (1:1:1). After 2 h, TFA was added and the
solution was concentrated, redissolved in MeOH/H2O (3:1),
and purified by reverse phase preparative HPLC. The pure fractions
were combined and lyophilized to give the carboxylic acid products
(TFA salt) as a white powder.
General Procedure D: Amide
Coupling of Arylamines
Ph2POCl (3 equiv) and DIEA
(5 equiv) were added to a suspension
of compound 29 (1 equiv) in DCM. After 30 min, arylamine
(4 equiv) and DMAP (1 equiv) were added. After 12 h, the reaction
was concentrated and purified by column chromatography to produce
arylamide compounds.
General Procedure E: Hydrolysis of Aryl Esters
NaOH
(3 equiv) was added to a solution of ester (1 equiv) in THF/MeOH/H2O (1:1:1). After 2 h, TFA was added and the reaction was concentrated,
redissolved in MeOH/H2O (3:1), and purified by reverse
phase preparative HPLC. The pure fractions were combined and lyophilized
to give the carboxylic acid products as the TFA salt as a white powder.
HATU (616 mg, 1.62 mmol) and DIEA
(0.550 mL, 3.24 mmol) were added to a suspension of acid 29 (500 mg, 1.08 mmol) in DCM (15 mL) and stirred. After 10 min, methyl
4-aminobicyclo[2.2.2]octane-1-carboxylate (396 mg, 2.16 mmol)
and DMAP (132 mg, 1.08 mmol) were added to the reaction. After overnight,
the solvent was removed in vacuo and the crude was purified by column
chromatography to give 549 mg of interemediate 56-ester. LiOH·H2O (110 mg, 2.62 mmol) and sodium hydroxide
(105 mg, 2.62 mmol) were added to a solution of interemidiate 56-ester (549 mg, 0.873 mmol) dissolved in a mixture of THF
(3 mL), H2O (3 mL), and MeOH (3 mL). After the hydrolysis
was complete, as determined by TLC, the reaction was quenched with
TFA (3 mL) and stirred. After 5 min, the solution was concentrated
in vacuo (not to dryness) and the resulting oil was redissolved in
CH3CN and H2O (1:1) and the solution was purified
by preparative HPLC. The purified fractions were combined, concentrated
in vacuo, redissolved in H2O, frozen, and lyophilized to
give 56 (TFA salt) as a white powder. 1H NMR
(300 MHz, CD3OD) δ ppm 7.63 (t, 1H, J = 6.84 Hz), 7.45 (d, 1H, J = 6.76 Hz), 7.35 (t,
1H, J = 7.21 Hz), 7.18–7.04 (m, 2H), 6.77
(dd, 1H, J = 1.26 Hz), 4.68 (d, 1H, J = 10.61 Hz), 2.73–2.48 (m, 1H), 2.16–1.98 (m, 1H),
1.98–1.43 (m, 18H), 1.27–1.02 (m, 2H); 13C NMR (100 MHz, CD3OD) δ ppm 180.79, 178.22, 167.86,
157.78 (d, JC–F = 248.9 Hz), 145.23, 137.07, 132.32,
129.39, 128.99, 126.45 (d, JC–F = 4.9 Hz), 123.39, 122.50, 122.17, 122.15 (d, JC–F = 19.3 Hz), 111.77, 73.22, 67.72, 62.58, 52.82,
46.96 (d, JC–F = 3.5 Hz), 38.96,
32.17, 31.36, 30.41(3C), 29.45(3C), 25.35, 23.08, 21.74; ESI-MS m/z 614.92 (M + H)+.
CDI (49 mg, 0.303 mmol), DIEA (88 μL,
0.505 mmol), and DMAP (cat.) were added to a solution of 56 (62 mg, 0.101 mmol) in 1,2-dichloroethane (10 mL), and the reaction
was heated to 40 °C. After 20 min, methanesulfonamide (96
mg, 1.01 mmol) was added and the reaction was refluxed. After overnight,
the sovent was removed in vacuo and the crude was purified by prepartive
HPLC to give 58 (TFA salt) as a white solid. 1H NMR (300 MHz, CD3OD) δ ppm 7.64 (t, 1H, J = 7.23 Hz), 7.45 (dd, 1H, J = 1.93, 8.22
Hz), 7.36 (t, 1H, J = 7.23 Hz), 7.18–7.04
(m, 2H), 6.77 (d, 1H, J = 1.66 Hz), 4.69 (d, 1H, J = 10.70 Hz), 3.19 (s, 3H), 2.75–2.52 (m, 1H), 2.21–1.99
(m, 1H), 1.99–1.44 (m, 17H), 1.41–1.27 (m, 1H), 1.27–1.03
(m, 2H); ESI-MS m/z 691.42 (M +
H)+.
Intermediate 64(21) (1.0 g, 2.09 mmol) was dissolved in a mixture
of DCM (10 mL) and AcOH (10 mL). Paraformaldehyde (1.13 g, 37.62 mmol)
was added, and after 10 min NaBH(OAc)3 (4.43 g, 20.9 mmol)
was added in portions. After 4 h, as determined by TLC, the reaction
was complete. Saturated NH4Cl (aq) was added to the reaction
which was then extracted with EtOAc. The EtOAc solution was washed
with NaOH (1 M) and brine, then dried over Na2SO4, filtered, concentrated, and purified by column chromatography to
give 764 mg of intermediate 65. Sodium hydroxide (320
mg, 8.0 mmol) was added to a solution of intermediate 65 (764 mg, 1.6 mmol) dissolved in a mixture of THF (3 mL), H2O (3 mL), and MeOH (3 mL), and the solution was heated to 45–50
°C. After hydrolysis was complete, as determined by TLC, the
reaction was cooled, neutralized with 2 M HCl, and extracted with
EtOAc. The EtOAc solution was dried over Na2SO4, filtered, and concentrated to give the acid intermediate 66 as a solid that was used without further purification.
CDI (315 mg, 1.94 mmol) and DIEA (0.450 mL, 2.59 mmol) were added
to a solution of 66 (300 mg, 0.648 mmol) in THF (15 mL)
and heated to 50 °C. After 30 min ammonium hydroxide was was
added. After standing overnight, H2O was added and the
reaction was extracted with EtOAc. The EtOAc solution was dried over
sodium sulfate, filtered, concentrated, and purified by column chromatography
to give 240 mg of compound 67. Lawesson’s reagent
(102 mg, 0.252 mmol) was added to a solution of compound 67 (60 mg, 0.126 mmol) in DCM (3 mL) and refluxed overnight. Then the
reaction was cooled, the solvent was removed, and the crude was column
chromatographed to produce 60 mg of 68. 1H
NMR (400 MHz, CD3OD) δ ppm 7.66 (ddd, 1H, J = 1.4, 6.4, 7.9 Hz), 7.53 (dd, 1H, J =
2.5, 8.3 Hz), 7.38 (t, 1H, J = 7.3 Hz), 7.15 (t,
1H, J = 8.0 Hz), 7.10 (dd, 1H, J = 1.9, 8.2 Hz), 6.77 (d, 1H, J = 1.9 Hz), 5.48
(br s, 1H), 4.71 (d, 1H, J = 11.2 Hz), 3.37 (s, 3H),
2.38 (d, 1H, J = 14.2 Hz), 2.19–2.02 (m, 1H),
1.99–1.85 (m, 1H), 1.83–1.61 (m, 4H), 1.50–1.36
(m, 1H), 1.35–1.16 (m, 2H); ESI-MS m/z 492.67 (M + H)+.
Fluorescence Polarization-Based
(FP-Based) Protein Binding Assay,
Cell Growth Inhibition Assay, Pharmacodynamics (PD) Study in SJSA-1
Xenograft Model, and in Vivo Efficacy Study
These assays
were performed according to the previously reported procedures.[21] All the in vivo studies were performed under
an animal protocol (PRO00005315) approved by the University Committee
on Use and Care of Animals of the University of Michigan, in accordance
with the recommendations in the Guide for the Care and Use of Laboratory
Animals of the National Institutes of Health.
Computational Methods
Docking studies were performed
using our previously reported structure of MDM2 in complex with 2 (PDB code 5TRF)[18] with the GOLD[33,34] program (version 5.2). For each genetic algorithm (GA) run, a maximum
number of 200 000 operations were performed on a population
of five islands of 100 individuals. Operator weights for crossover,
mutation, and migration were set to 95, 95, and 10, respectively,
and the Chemscore[35] fitness function was
used to evaluate the docked conformations. Docking was terminated
after 10 runs for each ligand, and the top 10 conformations were saved
for analysis of the predicted binding modes.
Authors: Lukasz Skalniak; Aleksandra Twarda-Clapa; Constantinos G Neochoritis; Ewa Surmiak; Monika Machula; Aneta Wisniewska; Beata Labuzek; Ameena M Ali; Sylwia Krzanik; Grzegorz Dubin; Matthew Groves; Alexander Dömling; Tad A Holak Journal: FEBS J Date: 2019-02-16 Impact factor: 5.542
Authors: Matthias R Bauer; Paolo Di Fruscia; Simon C C Lucas; Iacovos N Michaelides; Jennifer E Nelson; R Ian Storer; Benjamin C Whitehurst Journal: RSC Med Chem Date: 2021-01-07