Structure-based design of spiro-oxindoles as potent, specific small-molecule inhibitors of the MDM2-p53 interaction

Article abstract of DOI:10.1021/jm051122a

Potent, specific, non-peptide small-molecule inhibitors of the MDM2-p53 interaction were successfully designed. The most potent inhibitor (MI-63) has a K_i value of 3 nM binding to MDM2 and greater than 10 000-fold selectivity over Bcl-2/Bcl-xL

Full text of DOI:10.1021/jm051122a

3432
J. Med. Chem. 2006, 49, 3432-3435
Letters
Structure-Based Design of Spiro-oxindoles as
Potent, Specific Small-Molecule Inhibitors of the
MDM2-p53 Interaction
Ke Ding,† Yipin Lu,† Zaneta Nikolovska-Coleska,†
Guoping Wang,† Su Qiu,† Sanjeev Shangary,† Wei Gao,†
Dongguang Qin,† Jeanne Stuckey,§ Krzysztof Krajewski,‡
Peter P. Roller,‡ and Shaomeng Wang*,†
Figure 1. Recently reported potent non-peptide inhibitors of the
MDM2-p53 interaction.
ComprehensiVe Cancer Center and Departments of Internal
Medicine, Pharmacology, and Medicinal Chemistry, and
Life Sciences Institute, UniVersity of Michigan, 1500 E. Medical
Center DriVe, Ann Arbor, Michigan 48109, and Laboratory of
Medicinal Chemistry, National Cancer Institute-Frederick,
National Institutes of Health, Frederick, Maryland 21702
ReceiVed NoVember 7, 2005
Abstract: Potent, specific, non-peptide small-molecule inhibitors of
the MDM2-p53 interaction were successfully designed. The most
potent inhibitor (MI-63) has a Ki value of 3 nM binding to MDM2 and
greater than 10 000-fold selectivity over Bcl-2/Bcl-xL proteins. MI-63
is highly effective in activation of p53 function and in inhibition of
cell growth in cancer cells with wild-type p53 status. MI-63 has
excellent specificity over cancer cells with deleted p53 and shows a
minimal toxicity to normal cells.
The tumor suppressor p53 plays a central role in the regulation
of cell cycle, apoptosis, and DNA repair and is an attractive
molecular target for cancer therapy.1-3 Since p53 effectively
suppresses oncogenesis, it is not surprising that in approximately
50% of all human cancers its function has been nullified by
deletions or mutations in the DNA-binding domain of p53.4 In
the remaining 50% of human cancers, p53 retains its wild-type
form, but its activity is effectively inhibited by the MDM2
oncoprotein interacting directly with p53.5 Reactivation of the
p53 function by disruption of the MDM2-p53 interaction using
a non-peptide small-molecule inhibitor is now recognized as a
new and promising strategy for anticancer drug design.6-11
The MDM2-p53 interaction involves a short helix formed
by residues 13-29 of p53 and a hydrophobic cleft in MDM2.12
Peptide-based inhibitors designed on the basis of the sequence
of the p53 domain interacting with MDM2 have achieved very
high binding affinities to MDM2.13 Despite intense research
efforts over the past decade in academia and the pharmaceutical
industry, very few potent, non-peptide small-molecule MDM2
inhibitors have been reported until very recently.6-11 This
highlights the challenge associated with the design of non-
peptide small-molecule inhibitors to target the MDM2-p53
interaction.
Figure 2. Structure-based design of potent small-molecule inhibitors
to mimic Phe19, Trp23, Leu26, and Leu22 residues in p53. (A) X-ray
structure of the p53-MDM2 complex shows that in addition to Phe19,
Trp23, and Leu26, Leu22 in p53 was also shown to play a role for the
interaction between p53 and MDM2. MDM2 binding site is color-coded
according to the cavity depth (buried regions are coded in yellow, and
solvent-exposed regions are coded in blue), and the side chains of four
key residues in p53 are depicted purple in stick model. (B, C) Predicted
binding models of 1 and 4 complexed with MDM2 using the GOLD
program. For both compounds, carbons are in white, nitrogens in blue,
chloride in green, and oxygens in red. Hydrogen bonds are depicted
with a dashed yellow line. (D) Superposition of compound 4 to the
p53 peptide conformation in the crystal structure of p53 peptide in
complex with MDM2. Four residues, Phe19, Leu22, Trp23, and Leu26,
in p53 are colored in purple. For compound 2, the same colors are
used to color-code atoms as in panel C.
nM binding to the recombinant human MDM2 protein in our
fluorescence-polarization-based (FP-based) binding assay.9 Com-
pound 1 is effective in inhibition of cell growth in cancer cells
with wild-type p53 and has selectivity over cancer cells with
deleted p53 and a minimal toxicity to normal cells.9 Compound
1 thus represents a promising lead compound for further
optimization. Herein, we wish to report our further structure-
based optimization for this class of compounds as inhibitors of
the MDM2-p53 interaction.
We have recently reported our structure-based de novo design
of spiro-oxindoles as a class of potent, non-peptide small-
molecule inhibitors of the MDM2-p53 interaction.9 We showed
that the most potent inhibitor 1 (Figure 1) has a Ki value of 86
X-ray crystallography revealed that the interaction between
p53 and MDM2 involves primarily three hydrophobic residues
(Phe19, Trp23, and Leu26) from a short R-helix in p53 and a
small but deep hydrophobic pocket in MDM2 (Figure 2A).12
Those recently reported potent non-peptide inhibitors of the
* Corresponding author. Phone: 734-615-0362. Fax: 734-647-9647.
E-mail: shaomeng@umich.edu.
† Comprehensive Cancer Center and Departments of Internal Medicine,
Pharmacology, and Medicinal Chemistry, University of Michigan.
§ Life Sciences Institute, University of Michigan.
‡ National Cancer Institute-Frederick.
10.1021/jm051122a CCC: $33.50 © 2006 American Chemical Society
Published on Web 05/13/2006

Letters
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 12 3433
Scheme 1. Synthesis of compounds 4, 8, and 10a
Figure 3. Designed new inhibitors of the MDM2-p53 interaction.
MDM2-p53 interaction, including the spiro-oxindole (1),9
Nutlin-3 (2),8 and the benzodiazepinedione 310 (Figure 1), are
all designed to mimic the Phe19, Trp23, and Leu26 residues in
p53 and their interactions with MDM2.
a Reagents and conditions: (a) 4 Å molecule sieves, toluene; (b)
2-morpholin-4-yl-ethylamine/THF(or 1-(2-aminoethyl)-piperidine/THF), RT;
(c) Pb(OAc)4, CH2Cl2-MeOH (1:1), 0 °C.
Although those non-peptide inhibitors achieve high binding
affinities to MDM2, they are still significantly less potent than
the most potent peptide-based inhibitors. Compound 1, for
example, is 100 times less potent than the most potent peptide-
based inhibitor in our binding assay.9 This suggests that there
may be additional interaction between MDM2 and those peptide-
based inhibitors13 that is not captured by these non-peptide
inhibitors. Analysis of the X-ray structure12 of the MDM2-
p53 complex showed that in addition to the Phe19, Trp23, and
Leu26 residues in p53, a fourth residue, Leu22, also appears to
play an important role in the overall interaction between p53
and MDM2 (Figure 2A), a suggestion that finds support in
results from mutation analysis14 and alanine scanning of p53
peptides.15 Since compound 1 was designed to mimic Phe19,
Trp23, and Leu26 residues in p53, we reasoned that further
optimization of compound 1 to capture the additional interaction
between Leu22 in p53 and MDM2 should yield new inhibitors
with higher affinity.
To test this idea, we superimposed the modeled structure9 of
compound 1 bound to MDM2 on the crystal structure12 of the
MDM2-p53 complex and showed that the 5′-carbonyl carbon
of compound 1 is within 2 Å of the backbone carbonyl carbon
of Leu22 in p53. This suggests that an appropriate group could
be attached to this site in compound 1 to mimic Leu22 in p53.
Unlike Phe19, Trp23, and Leu26, Leu22 is not buried deeply
inside the binding pocket in MDM2 and is in fact partially
exposed to solvent in the crystal structure.12 This provides an
opportunity to deploy a chemical group containing some polar
moieties at this position not only to mimic the hydrophobic
interaction of Leu22 in p53 with MDM2 but also to improve
physiochemical properties of the resulting compounds such as
aqueous solubility.
Figure 4. Competitive binding curves of small-molecule inhibitors to
recombinant human MDM2 protein as determined using a fluorescence-
polarization-based binding assay.
charged carboxylic group of the Glu17 in p53 in the crystal
structure.12 Hence, compound 4 not only mimics the four
hydrophobic residues (Phe19, Leu22, Trp23, Leu26) in p53
(Figure 2D) for their interaction with MDM2 but also may
capture the interaction between Glu17 in p53 and Lys90 in
MDM2. Indeed, the GOLD program predicted that compound
4 binds more tightly to MDM2 than compound 1 based upon
the predicted binding models for these two compounds.
Compound 4 was synthesized stereospecifically via a route
(Scheme 1) in which the key step is an asymmetric 1,3-dipolar
cycloaddition reaction.9,17 Compound 4 was determined to bind
to MDM2 with a Ki of 13 nM in our FP-based binding assay
(Figure 4). Consistent with the modeling prediction, compound
4 is 6 times more potent than 1, providing support for our design
strategy.
Modeling suggests that the 4′-(m-chlorophenyl) ring in
compounds 1 and 4 occupies a hydrophobic pocket in MDM2
(Figure 2B,C) and mimics Phe19 in p53 (Figure 2A,D). Fluorine
substitution on a phenyl ring has been used frequently as an
effective strategy to increase the metabolic stability of drug
molecules, and we have sought to investigate the effect on
binding to MDM2 of a fluoro substituent on the 4′-(m-
chlorophenyl) in 1. Compounds 5, 6, and 7 (Figure 3), in which
the fluoro substituent is, respectively, at the 2, 4, or 5 position
of the 4′-(m-chlorophenyl) ring, were synthesized using our
published method9,17 and tested for their binding affinities to
MDM2. As can be seen from Figure 4, a fluorine substitution
at C4 (compound 6) results in a 2-fold reduction in binding
affinity to MDM2, but substitution at C2 (compound 5) or C5
(compound 7) improves the binding affinity of the resulting
compounds by a factor of 2. With compound 4 as the template,
On the basis of these considerations, we have designed
compound 4 (Figure 3), in which the N,N-dimethylamine in 1
is replaced by a 2-morpholin-4-yl-ethylamine group, a chemical
group that has been extensively used in drug design. The binding
model predicted by the GOLD program16 for compound 4
(Figure 2C,D) shows that the carbon atoms in the morpholine
ring, together with the two carbon linker between the morpholine
ring and the 5′-carbonyl group in compound 4 closely mimic
Leu22 in p53 in its interaction with MDM2. In addition, the
oxygen atom in the morpholine ring is in close proximity to
the positively charged amine group in Lys90 in MDM2,
suggesting the possibility of hydrogen bonding. Indeed, the
positively charged amino group in Lys90 in MDM2 was shown
to have a strong charge-charge interaction with the negatively

3434 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 12
Letters
Figure 6. Western blotting analysis of p53, MDM2, and p21cip1/waf1
proteins in LNCaP and PC-3 prostate cancer cells when treated with
compounds 8 and 9 for 24 hours.
Figure 5. Inhibition of cell growth as determined by WST assay in
prostate cancer LNCaP cells with wild-type p53.
cellular specificity (Supporting Information). To evaluate its
therapeutic potential further, we have tested 8 in normal prostate
epithelial cells and determined that it has minimal toxicity to
normal prostate epithelial cells at concentrations as high as 5
µM.
compound 8, with the 2-morpholin-4-yl-ethylamine group, was
designed and synthesized. Compound 8 was determined to bind
to MDM2 with a Ki value of 3 nM (Figure 4).
Compound 8 contains four chiral centers and modeling
predicts that the binding of 8 to MDM2 is stereospecific. To
test this prediction, we have synthesized compound 9 (Figure
3), an enantiomer of 8, and determined that 9 has a Ki of 4.0
µM binding to MDM2 (Figure 4). Hence, 9 is 1000-times less
potent than 8, confirming stereospecific binding of 8 to MDM2.
To investigate the importance of the oxygen atom in the
morpholine ring in compound 8 for binding to MDM2, we have
designed and synthesized compound 10, in which the oxygen
atom was replaced by a carbon atom. Compound 10 has a Ki
value of 39 nM, 13-times less potent than compound 8,
confirming the significant contribution of this oxygen atom for
binding to MDM2.
Nutlin-3 was probably the most potent, cell-permeable non-
peptide inhibitor of the MDM2-p53 interaction reported
to date.8 It was determined that Nutlin-3 has a Ki value of
36 nM binding to MDM2 in our binding assay (Figure 4).
Hence, compound 8 is 12 times more potent than Nutlin-3
and 2000-times more potent than the natural p53 peptide (Figure
4).
These inhibitors were designed to mimic the short R-helix in
p53 and to compete with p53 for interaction with MDM2. This
kind of interaction, involving a surface pocket in one protein
and a short R-helix in its binding partner is also observed in
the interactions of the anti-apoptotic proteins Bcl-2/Bcl-xL with
the pro-apoptotic Bcl-2 members such as Bid, Bad, Bak, and
Bax proteins.18 Using our established and sensitive FP-based
assays for Bcl-2 and Bcl-xL proteins (Supporting Information),
we have investigated whether 8 binds to these two proteins. It
was found that 8 does not show a significant binding to either
Bcl-2 or to Bcl-xL protein at concentrations as high as 50 µM,
revealing a more than 10 000-fold specificity for MDM2 protein
over Bcl-2/Bcl-xL proteins.
Potent, specific, and cell-permeable small-molecule inhibitors
of the MDM2-p53 interaction are predicted to inhibit cell
growth effectively in cancer cells with wild-type p53.7-9
Furthermore such an inhibitor should have a minimal activity
in cancer cells with mutated or deleted p53. Using Nutlin-3 as
a positive control, we have evaluated 4 and 8 for their ability
to inhibit cell growth in LNCaP prostate cancer cells with wild-
type p53.19 Compounds 4 and 8 and Nutlin-3 are indeed
effective in inhibition of cell growth and achieve IC50 values
of 800, 280, and 1500 nM, respectively, in inhibition of
cell growth in LNCaP cells (Figure 5). Consistent with its higher
binding affinity to MDM2, 8 is 3-times more potent than 4
and 5-times more potent than Nutlin-3 in the cell growth assay.
To test the cellular specificity of 8, we have evaluated it in
human prostate cancer PC-3 cells with deleted p5319 and found
that it has an IC50 value of 18 µM, thus displaying an excellent
A potent small-molecule inhibitor of the MDM2-p53
interaction would be expected to activate p53, resulting in an
increase in the levels of p53 in cells with wild-type p53 but not
in cells with mutated or deleted p53. In addition, activation of
p53 should lead to induction of p21 cyclin-dependent kinase
inhibitor 1 (p21cip1/waf) and MDM2, two p53 targeted genes, and
result in an increase in the levels of p21cip1/waf and MDM2
proteins.8 To test these predictions, we have examined the levels
of p53, MDM2, and p21cip1/waf proteins by Western blotting
analysis in LNCaP and PC-3 cancer cells when treated with 8
and 9. Consistent with predictions, compound 8 causes a dose-
dependent increase in the levels of p53, MDM2, and p21cip1/waf
proteins in LNCaP cells (Figure 6), indicating a strong activation
of p53. In direct comparison, 9 with a weak binding affinity to
MDM2 has a minimal effect at concentrations as high as 10
µM. Since MDM2 protein is an E3 ubiquitin ligase and is known
to effectively degrade p53 protein upon binding,5 the very high
levels of p53 and MDM2 protein in LNCaP cells treated with
8 suggest that MDM2 protein is unable to degrade p53 protein
when blocked by 8. Furthermore, both 8 and 9 do not increase
the levels of p53, MDM2, and p21cip1/waf proteins in PC-3 cells
with deleted p53 status (Figure 6), indicative of the lack of p53
activation.
In summary, we have successfully designed a class of high-
affinity, specific, cell-permeable, non-peptide small-molecule
inhibitors of the MDM2-p53 interaction using structure-based
design strategy. The most potent inhibitor (8), which was named
MI-63 (MDM2 inhibitor 63), has a Ki value of 3 nM binding to
MDM2 and is more than 2000-times more potent than the
natural p53 peptide (residues 13-29). MI-63 displays a high
stereospecificity and is very selective in blocking the MDM2-
p53 interaction rather than the Bcl-2(Bcl-xL)-Bid interaction.
MI-63 is highly effective in activation of p53 function and in
inhibition of cell growth in LNCaP prostate cancer cells with
wild-type p53. In addition, MI-63 shows an excellent selectivity
over PC-3 prostate cancer cells with deleted p53 and has
minimal toxicity to normal cells. In view of all these observa-
tions, MI-63 can be regarded as a specific, promising, non-
peptide small-molecule inhibitor of the MDM2-p53 interaction.
This present study provides a clear example that structure-based
strategy can be successfully employed in the design of potent,
specific, drug-like, non-peptide small-molecule inhibitors to
target protein-protein interactions.
Acknowledgment. We are grateful for the financial support
from the National Cancer Institute, National Institutes of Health
(Grants R01CA121279, P50CA069568, P50CA097248, and
P30CA046592), and from Ascenta Therapeutics Inc.

Letters
Supporting Information Available: An experimental section
including the information on the synthesis and chemical data for
compounds 4-10, molecular modeling methods and results for 8,
the experimental procedure for the fluorescence polarization-based
binding assay, and details on the cellular assays and results. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 12 3435
(11) Yin, H.; Lee, G.-I.; Park, H.-S.; Payne, G. A.; Rodriguez, J. M.;
Sebti, S. M.; Hamilton, A. D. Terphenyl-based helical mimetics that
disrupt the p53/HDM2 Interaction. Angew. Chem., Int. Ed. 2005, 44,
2704-2707.
(12) Kussie, P. H.; Gorina, S.; Marechal, V.; Elenbaas, B.; Moreau, J.;
Levine, A. J.; Pavletich, N. P. Structure of the MDM2 oncoprotein
bound to the p53 tumor suppressor transactivation domain. Science
1996, 274, 948-953.
(13) Garcia-Echeverria, C.; Chene, P.; Blommers, M. J. 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.
(14) Lin, J.; Chen, B.; Elenbass, B.; Levine, A. J. Several hydrophobic
amino acids in the p53 amino-terminal domain are required for
transcriptional activation, binding to mdm-2 and the adenovirus 5
E1B 55-kD protein. Genes DeV. 1994, 8, 1235-1246.
(15) Picksley, S. M.; Vojtesek, B.; Sparks, A.; Lane, D. P. Immunochemi-
cal analysis of the interaction of p53 with MDM2;- -fine mapping
of the MDM2 binding site on p53 using synthetic peptides. Oncogene
1994, 9, 2523-2529.
(16) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.
Development and validation of a genetic algorithm for flexible
docking. J. Mol. Biol. 1997, 267, 727-748.
References
(1) Levine, A. J. p53, the cellular gatekeeper for growth and division.
Cell 1997, 88, 323-331.
(2) Vogelstein, B.; Lane, D.; Levine, A. J. Surfing the p53 network.
Nature 2000, 408, 6810, 307-310.
(3) Vousden, K. H.; Lu, X. Live or let die: the cell’s response to p53.
Nat. ReV. Cancer 2002, 2, 594-604.
(4) Hainaut, P.; Hollstein, M. p53 and human cancer: the first ten
thousand mutations. AdV. Cancer Res. 2000, 77, 81-137.
(5) Wu, X.; Bayle, J. H.; Olson, D.; Levine, A. J. The p53-mdm-2
autoregulatory feedback loop. Genes DeV. 1993, 7 (7A), 1126-1132.
(6) Chene, P. Inhibiting the p53-MDM2 interaction: an important target
for cancer therapy. Nat. ReV. Cancer 2003, 3, 102-109.
(7) Vassilev, L. T. p53 activation by small molecules: application in
oncology. J. Med. Chem. 2005, 48, 4491-4499.
(17) Ding, K.; Wang, G.; Deschamps, J. R.; Parrish, D. A.; Wang, S.
Synthesis of spirooxindoles via asymmetric 1,3-dipolar cyclo-
addition. Tetrahedron Lett. 2005, 46, 5949-5951.
(8) 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.
(9) Ding, K.; Lu, Y.; Nikolovska-Koleska, Z.; Qiu, S.; Ding Y.; Gao,
W.; Stuckey, J.; Roller, P. P., Tomita, Y.; Deschamps, J. R.; Wang,
S. Structure-based design of potent non-peptide MDM2 inhibitors.
J. Am. Chem. Soc. 2005, 127, 10130-10131.
(10) Grasberger, B. L.; Lu, T.; Schubert, C.; Parks, D. J.; Carver, T. E.;
Koblish, H. K.; Cummings, M. D.; LaFrance, L. V.; Milkiewicz, K.
L.; Calvo, R. R.; Maguire, D.; Lattanze, J.; Franks, C. F.; Zhao, S.;
Ramachandren, K.; Bylebyl, G. R.; Zhang, M.; Manthey, C. L.;
Petrella, E. C.; Pantoliano, M. W.; Deckman, I. C.; Spurlino, J. C.;
Maroney, A. C.; Tomczuk, B. E.; Molloy, C. J.; Bone, R. F.
Discovery and cocrystal structure of benzodiazepinedione HDM2
antagonists that activate p53 in cells. J. Med. Chem. 2005, 48, 909-
912.
(18) Oltersdorf, T.; Elmore, S. W.; Shoemaker, A. R.; Armstrong, R. C.;
Augeri, D. J.; Belli, B. A.; Bruncko, M.; Deckwerth, T. L.; Dinges,
J.; Hajduk, P. J.; Joseph, M. K.; Kitada, S.; Korsmeyer, S. J.; Kunzer,
A. R.; Letai, A.; Li, C.; Mitten, M. J.; Nettesheim, D. G.; Ng, S.;
Nimmer, P. M.; O’Connor, J. M.; Oleksijew, A.; Petros, A. M.; Reed,
J. C.; Shen, W.; Tahir, S. K.; Thompson, C. B.; Tomaselli, K. J.;
Wang, B.; Wendt, M. D.; Zhang, H.; Fesik, S. W.; Rosenberg, S. H.
An inhibitor of Bcl-2 family proteins induces regression of solid
tumours. Nature 2005, 435, 677-681.
(19) Fan, R.; Kumaravel, T. S.; Jalali, F.; Marrano, P.; Squire, J. A.;
Bristow, R. G. Defective DNA strand break repair after DNA damage
in prostate cancer cells: implications for genetic instability and
prostate cancer progression. Cancer Res. 2004, 64, 8526-8533.
JM051122A