Robust generation of lead compounds for protein-protein interactions by computational and MCR chemistry: P53/Hdm2 antagonists

Article abstract of DOI:10.1002/anie.201001343

The parallel discovery of multiple scaffolds useful to antagonize the cancer-relevant proteinprotein interaction p53/Hdm2 is described. The new method is based on the tightly interwoven interplay of multicomponent reaction chemistry, structural biology, c

Full text of DOI:10.1002/anie.201001343

Communications
DOI: 10.1002/anie.201001343
Multicomponent Reactions
Robust Generation of Lead Compounds for Protein–Protein
Interactions by Computational and MCR Chemistry:
p53/Hdm2 Antagonists**
Anna Czarna, Barbara Beck, Stuti Srivastava, Grzegorz M. Popowicz, Siglinde Wolf,
Yijun Huang, Michal Bista, Tad A. Holak, and Alexander Dꢀmling*
In memory of Ivar Ugi
The discovery of a lead compound is an essential process in
early drug discovery, hopefully eventually resulting in a
clinical candidate and a drug for the treatment of a disease.
Besides affinity and selectivity for the target, however, other
target-unrelated compound properties are equally important
for the fate of a drug candidate, for example, water solubility,
lipophilicity, and molecular weight since they determine
important aspects such as oral bioavailability, dosing sched-
ule, and side effects. The parallel discovery and early
development of several leads is therefore now pursued
whenever possible, an approach that takes into account the
high attrition rate of early drug discovery projects. Currently,
hits as starting points for medicinal chemistry optimizations
are mostly found by high-throughput screening (HTS)
campaigns and to a much lesser extent by structure-based
approaches including fragment-based and computational
drug discovery. For certain target classes, however, HTS
often yields very low numbers of hits.[1] For example, protein–
protein interactions (PPIs) are notoriously difficult to hit with
druglike small molecules.[2] This has been assigned to the
unusual structure, topology, and flexibility of protein–protein
interfaces.[3] The fact that several drugs targeting PPIs have
recently entered the clinical development clearly shows that
certain PPIs, for example, between Bcl-x and XIAP, can be
efficiently targeted by small molecules.[4] Herein, we describe
a complementary process that led to the parallel discovery of
several compounds belonging to seven different scaffold
classes, amenable to synthesis by efficient multicomponent
reaction (MCR) chemistry in just one step, which antagonize
the PPI between the transcription factor p53 and its negative
regulator Hdm2.
Protein–protein interactions are often mediated by only a
few key amino acid side chains and the terms “hot spot” and
“anchor” have been introduced for such locally constrained
areas and amino acids on the surface of interacting pro-
teins.[2,5,8] In a first approximation, the depth into which a
specific amino acid side chain of the donor protein is buried in
the acceptor protein is indicative of its energetic importance.
We reasoned that this “anchor” amino acid side chain might
serve as a reasonable starting point for the design of
(ant)agonists of a PPI. Thus, we use this particular amino
acid side chain as an initial anchor in virtual libraries of low-
molecular-weight scaffolds. Virtual compounds containing
anchor side chains are selected for synthesis and screening
based on their docking into the PPI interface. The starting
point for the docking/energy minimization procedure is
chosen in such a way that an overlap between the anchor
and the template amino acid side chain is ensured. In order to
rapidly test these ideas, we chose an efficient and fast but
versatile synthetic approach, MCR (Figure 1).[6] MCR allows
for the assembly of many diverse and complex scaffolds in a
one-step/one-pot manner, thus saving time and resources, and
potentially increasing the success rate of lead discovery.
The PPI interface of p53/Hdm2 has been characterized in
molecular detail by X-ray structure analysis.[7] It relies on the
steric complementarity between the Hdm2 cleft and the
hydrophobic face of the p53 a-helix and, in particular, on a
triad of p53 amino acids Phe19, Trp23, and Leu26, which
insert deep into the Hdm2 cleft (Figure 1A in the Supporting
Information). We chose the indole side chain of Trp23 as the
anchor residue for three reasons: 1) it is the central amino
acid of the triad, thus facilitating addressing by the antago-
nists the crucial Phe19 and Leu26 binding sites; 2) it is deeply
buried in Hdm2; and 3) it also features, in addition to
extensive van der Waals contacts, a hydrogen bond to the
Leu54 backbone carbonyl oxygen of Hdm2. In fact, calcu-
lation of the solvent-accessible surface areas of all amino
acids in the p53/Hdm2 interaction ranks Trp23 the highest
(Trp23 > Phe19 > Leu26).[8] Next, from our in-house database
of several hundred MCR scaffolds, we selected forty MCR
scaffolds to create virtual compound libraries.[6] By design,
each of the compounds incorporated the anchor. We used
indole and bioistosteric 4-chlorophenyl derivatives[24] sup-
plied with the corresponding functional groups as anchors to
[*] Dr. B. Beck, Dr. S. Srivastava, Y. Huang, Prof. A. Dꢀmling
Departments of Pharmaceutical Sciences and Chemistry
University of Pittsburgh, Biomedical Science Tower 3
3501 Fifth Avenue, Pittsburgh, PA 15261 (USA)
Fax: (+1)412-383-5298
E-mail: asd30@pitt.edu
Homepage: http://www.pitt.edu/~asd30/Doemling_lab.html
Dr. A. Czarna, G. M. Popowicz, S. Wolf, M. Bista, Prof. T. A. Holak
Max-Planck-Institut fꢁr Biochemie, Martinsried (Germany)
[**] MCR: multicomponent reaction. We thank Dr. Ulli Rothweiler for
fruitful discussion. This work was supported by the NIH (grant
1R21M087617-01 to A.D.) and the Deutsche Krebshilfe (grant
108354 to T.H.) and is part of an NCI-RAND program. We thank
Haixia Liu for the synthesis of PB3.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201001343.
5352
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 5352 –5356

Angewandte
Chemie
Figure 1. Graphical representation of the workflow for the rapid generation of low-molecular-weight (ant)agonists of PPIs. The process relies on
high-resolution structural data of the target (1), the presence and identification of a highly buried amino acid (2), a fragmentation/anchor
generation step (3), virtual chemistry employing the anchor and based on multicomponent reactions (4), constrained docking forcing the anchor
fragment into the binding site (5), and synthesis and screening (6). (FG=functional group.)
act as starting materials for the MCRs (Supporting Informa-
tion, Figure 1B); for example, we used unsubstituted indole
and oxindole to perform a Friedel–Crafts-type alkylation and
an Ugi four-component condensation, respectively (Support-
ing Information, Table 1).
The virtual library of compounds comprising all possible
stereoisomers was generated using REACTOR software.[9]
Different aliphatic and aromatic starting materials to repre-
sent different shapes and electrostatic features were chosen.
The virtual libraries incorporating the anchor side chain were
docked into a rigid model of the p53 binding site in the Hdm2
receptor (PDB identifier: 1YCR) using the modeling/docking
software package Moloc.[7,10] Assuming that the anchor
residue predefines the binding site of the molecules and in
order to avoid nonproductive docking geometries, we forced
the anchor part (indole or 4-chlorophenyl) of the compounds
to overlap with the respective Trp23 anchor as a starting point
Figure 2. Alignment of the anchor p53 residues Phe19, Trp23, and
for energy minimization. As an encouraging evidence for the
Leu26 (cyan sticks) bound to Hdm2 (dark pink cartoon, PDB ID:
1YCR) with a van Leusen imidazole PB12 based antagonist (pink
sticks) bound to Hdm2 (light pink cartoon, PDB ID: 3LBK) and the
anchor approach we solved the crystal structure of the MCR
compound PB12 bound to Hdm2 (Figure 2).[11] This structure
shows an almost perfect alignment of the Trp23 indole anchor
corresponding predicted docking geometry of PB12 (grey sticks).
Angew. Chem. Int. Ed. 2010, 49, 5352 –5356
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5353

Communications
Table 1: NMR spectroscopically determined activities of predicted,
synthesized, and tested compounds and known optimized compounds
(fluorescent polarization (FP) assay).[a]
Compound Kd [mm]
Compound Kd [mm]
>100[b]
PB1
40Æ15[b] PB7
syn-PB2
anti-PB2
PB2a
PB3
PB4
PB5
3Æ1[c]
PB8
60Æ20[b,c]
40Æ10[b] PB9
60Æ30[b]
2Æ1[c]
PB10
PB11
5Æ2[c] (11Æ0.5)
20Æ7[b,c]
0.8Æ0.4[c]
30Æ10[b] PB12
1.1[c] (0.7Æ0.04)
0.09Æ0.03[c] (0.04Æ0.01)
0.01Æ0.01[c] (0.003Æ0.002)
30Æ10[c]
nutlin-3[d]
PB6
precipitation MI-63[d]
[a] Errors for the FP values are determined by fit confidence. The
experimental error of Kd measured by NMR was assumed to be ca. 30%.
The techniques used for [b] and [c] gave the same values within the error
boundaries. [b] Binary titration of the ligand with apo-15N-Hdm2 using
15N-HSQC.[22] [c] AIDA experiment.[23] [d] Nutlin-3 and MI-63 are de-
scribed in Ref. [24] and Ref. [11,25], respectively; the Kd values are our
measurements, in agreement with Ref. [11,24,25,26].
Scheme 1. Predicted and found p53/Hdm2 antagonists based on
multiple MCR scaffolds (the anchor moiety is marked gray, PB=
Pittsburgh).
with the indole moiety of PB12 (see Scheme 1) with an
RMSD (indole) of 0.34 ꢀ.
From the highest-ranking docked compounds, which
addressed all three binding sites of p53 (Phe19, Trp23, and
Leu26), we chose several compounds for synthesis and
screening based on several features (Scheme 1). First, the
compounds were chosen so as to belong to different types of
scaffolds to ensure variation in backbones and chemistry.
Second, we chose reactions for which the starting materials
were commercially available or readily accessible by syn-
thesis. As a third criterion, we preferentially chose MCRs we
had used before.[6a] Products were synthesized according to
two different Ugi MCR variations (PB3, PB5),[12] the van
Leusen three-component reaction (PB1, PB12),[13] the Orru
three-component reaction (PB2) followed by an amidation
step (PB11),[14] the Passerini three-component reaction
(PB9),[15] the Betti three-component reaction (PB8),[16] the
Staudinger three-component reaction (PB6),[17] an MCR that
formed an amidinosulfone amide (PB7),[18] the Gewald three-
component reaction to an aminothiophene (PB10),[19] and an
a-amidoalkylation (PB4).[20] Among the eleven different
MCR series prepared, seven scaffolds showed binding
activities less than 60 mm (Table 1).
Figure 3. Binding models of four p53/Hdm2 antagonists by constrained
docking into PDB 1YCR: 5-indoloimidazole PB1; imidazoline PB2; thiohy-
dantoinimide PB3; and 2-aminothienyl-3-acetamide PB10. Several rim key
amino acids of Hdm2 are shown as sticks and green surface and
numbered. The docking models are in accordance with the observed
Hdm2 HSQC NMR shifts obtained from titrations. All geometries show
the receptor in the same orientation as in Figure 2.
otherwise cannot be assessed in high-throughput assays.
Importantly, use of NMR-based screening allowed for
determination of both compound affinity to Hdm2 and the
ability of a compound to dissociate the preformed p53/Hdm2
complex. Other important questions about the compound
properties can also be answered with an NMR-based
approach. These include: Is the compound sufficiently water
soluble? Does the compound bind to Hdm2 or to p53? Does
the compound cause precipitation of any of the proteins?
Does the compound lead to major conformational changes in
the binding protein? What is the binding site of the compound
on the protein surface?
Representative docking geometries of the predicted p53/
Hdm2 antagonist compounds are shown in Figure 3 and in
Figure 2 of the Supporting Information. Next, these com-
pounds were screened for their ability to bind to Hdm2 and to
antagonize the p53/Hdm2 PPI. We chose NMR spectroscopy
for this test as it provides a wealth of information that
All of this information is instrumental in order to optimize
a compound series not only for potency and selectivity but
also for, for example, water solubility and protein binding.
5354
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 5352 –5356

Angewandte
Chemie
Moreover the usage of screening techniques based on
independent physical methods prevents from discovering
PAIN compounds (PAIN: Pan Assay Interference) and
covalent binders.[21] The different compounds were tested
for binding to Hdm2 by performing a series of NMR titrations
with isotopically enriched 15N-Hdm2.[22] Strong binding of a
compound to its target is indicated by signal splitting in a
heteronuclear single quantum coherence (HSQC) spectrum,
whereas a signal shift indicates weaker binding. Figure 4A–C
shows NMR data of Hdm2 titrated with PB12 as an example
of slow chemical exchange and Figure 3 of the Supporting
Information shows similar spectra with PB2 diastereomers as
an example of fast chemical exchange. Additionally we used
our recently introduced antagonist-induced dissociation assay
(AIDA), a fast one-dimensional NMR method to determine
Kd values (Figure 4C and Supporting Information,
Figure 4),[23] which are in agreement with the fluorescence
surface area (TPSA, Supporting Information, Table 2), most
probably resulting in low water solubility, an unfavorable
quality of compounds designed to become drugs. According
to our binding model of PB2, the methyl ester group is not
involved in binding, but rather remains exposed to the solvent
(Figure 3). Therefore, we reasoned that derivatization of the
ester bond by amidation could potentially improve water
solubility and binding affinity. Indeed, conversion of PB2
derivative PB2a to amide PB11 (Scheme 2) improved the
potency of the imidazoline scaffold as well as its water
solubility, as judged by the NMR experiments and the TPSA
(Table 1 and Supporting Information, Table 2 and Figure 6).
polarization data shown in Figure 4D.
NMR
From the HSQC spectra we calculated the Kd
values
for syn-PB2 and anti-PB2 to be (3 Æ 1) mm and ca. 40 mm,
respectively. Table 1 shows a selection of other Hdm2 binding
compounds with their Kd values, ranging from 60 mm down to
less than 1 mm. Due to the inherent requirements of the Hdm2
binding pocket, the PB compounds are highly hydrophobic
and most of the active PB compounds show a low total polar
Scheme 2. One-pot synthesis of imidazoline PB2a by an Orru-3CR and
optimization by amidation yielding derivative PB11 with improved
affinity and water solubility (TBD=1,5,7-triazabicyclo[4.4.0]dec-5-ene).
We have introduced a new process for predicting and
producing antagonists of the cancer-relevant p53/Hdm2
protein–protein interaction. This approach differs substan-
tially from currently used techniques, including high-through-
put screening, fragment-based approaches, structure-based
design or computational screening of compound libraries.
Classical fragment-based drug discovery approaches, for
example, are strong in detecting weakly, however efficiently
binding small molecular units which serve as starting points to
assemble more-potent druglike compounds, however with no
straightforward synthetic pathway.[27] Herein, MCR chemistry
was used to synthesize more-potent leads starting from a
fragment (= anchor). Thus a seamless pathway for the
optimization of affinity and other important drug properties
is predefined. Nutlin-3, for example, is synthesized in a
sequential synthesis of more than eight steps whereas PB11 is
accessible in only two steps from commercially available
starting materials.[28] The compounds collected in Scheme 1
are druglike and can be optimized regarding potency and
physicochemical properties (Supporting Information,
Figure 6). Certainly, this new approach, successful for one
target example, demands further validation with different
types of PPIs. Extensive optimization and in vitro and in vivo
testing of optimized compounds of the different scaffolds
reported herein are under way and will be reported in due
course.
Figure 4. Screening of the activity of PB12. A) 1H-15N HSQC spectra of
Hdm2 (1–125) titrated with increasing amounts of PB12. Red spec-
trum: apo-Hdm2; green spectrum: approximately 40% saturation of
Hdm2 with PB12 (the spectrum shows slow chemical exchange that is
typical for strong interactions with submicromolar Kd values); blue
spectrum: Hdm2 fully saturated with PB12. B) Perturbation plot of
PB12. C) AIDA experiment: the concentration of p53 released from the
p53/Hdm2 complex by the antagonist is proportional to the height of
the HN indole peak of W23(p53). Bottom: downfield-shifted NMR
signals of the p53/Hdm2 complex (20 mm); middle: approximately
70% of dissociation of the complex upon addition of PB12 to the
complex in a 1:1 molar ratio; top: signals of free p53. D) Study of
PB12 by the fluorescence polarization assay.[11,26]
Received: March 5, 2010
Revised: May 4, 2010
Published online: June 23, 2010
Angew. Chem. Int. Ed. 2010, 49, 5352 –5356
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5355

Communications
[14] D. van Leusen, Org. React. 2001, 57, 417 – 666.
Keywords: antagonists · drug discovery ·
multicomponent reactions · p53/Hdm2 interaction ·
protein–protein interactions
.
[15] a) R. S. Bon, C. Hong, M. J. Bouma, R. F. Schmitz, F. J. J.
de Kanter, M. Lutz, A. L. Spek, R. V. A. Orru, Org. Lett. 2003,
5, 3759 – 3762; b) R. S. Bon, N. E. Sprenkels, M. M. Koningstein,
R. F. Schmitz, F. J. J. Kanter, A. Dꢁmling, M. B. Groen, R. V. A.
Orru, Org. Biomol. Chem. 2008, 6, 130 – 137.
[16] M. Passerini, T. Bonciani, Gazz. Chim. Ital. 1933, 63, 138 – 144.
[17] M. Betti, Gazz. Chim. Ital. 1900, 30, 306 – 316.
[18] H. Staudinger, Justus Liebigs Ann. Chem. 1907, 356, 51 – 123.
[19] a) K. Gewald, E. Schinke, H. Bꢁttcher, Chem. Ber. 1966, 99, 94 –
100; b) Y. Huang, A. Dꢁmling, Mol. Divers. 2010, DOI:
10.1007s11030-010-9229-6.
[1] a) R. W. Spencer, Biotech. Bioeng. Comb. Chem. 1998, 61, 61 –
67; b) E. Jacoby, A. Boettcher, L. M. Mayr, N. Brown, J. L.
Jenkins, J. Kallen, C. Engeloch, U. Schopfer, P. Furet, K. Masuya,
J. Lisztwan in Chemogenomics (Ed.: E. Jacoby), Springer,
Heidelberg, 2009, pp. 173 – 194 (Series: Methods in Molecular
Biology, Vol. 575).
[2] J. A. Wells, C. L. McClendon, Nature 2007, 450, 1001 – 1009.
[3] I. M. A. Nooren, J. M. Thornton, EMBO J. 2003, 22, 3466 – 3492.
[4] a) C. Tse et al., Cancer Res. 2008, 68, 3421 – 3428; b) K. Zobel, L.
Wang, E. Varfolomeev, M. C. Franklin, L. O. Elliott, H. J.
Wallweber, D. C. Okawa, J. A. Flygare, D. Vucic, W. J. Fair-
brother, K. Deshayes, ACS Chem. Biol. 2006, 1, 525 – 533.
[5] a) A. A. Bogan, K. S. Thorn, J. Mol. Biol. 1998, 280, 1 – 9; b) I. S.
Moreira, P. A. Fernandes, M. J. Ramos, Proteins Struct. Funct.
Genet. 2007, 68, 803 – 812.
[6] a) A. Dꢁmling, Chem. Rev. 2006, 106, 17 – 89; b) R. V. A. Orru,
M. de Greef, Synthesis 2003, 1471 – 1499; c) L. Weber, K. Illgen,
M. Almstetter, Synlett 1999, 366 – 374; d) C. Hulme, Y.-S. Lee,
Mol. Diversity 2008, 12, 1 – 15; e) I. Ugi, A. Dꢁmling, W. Hꢁrl,
Endeavour 1994, 18, 115 – 122; f) Multicomponent Reactions
(Eds: J. Zhu, H. Bienaymꢂ), Wiley-VCH, Weinheim, 2005; g) B.
Ganem, Acc. Chem. Res. 2009, 42, 463 – 472; h) J. D. Sunderhaus,
S. F. Martin, Chem. Eur. J. 2009, 15, 1300 – 1308.
[7] P. H. Kussie, S. Gorina, V. Marechal, B. Elenbaas, J. Moreau,
A. J. Levine, N. P. Pavletich, Science 1996, 274, 948 – 953.
[8] L. Meireles, A. Dꢁmling, C. J. Camacho, Nucl. Acids Res. 2010,
DOI:10.1093/nar/gkq502.
[9] G. Pirok, N. Mate, J. Varga, J. Szegezdi, M. Vargyas, S. Dorant, F.
Csizmadia, J. Chem. Inf. Model. 2006, 46, 563 – 568.
[10] a) P. R. Gerber, K. Mꢃller, J. Comput.-Aided Mol. Des. 1995, 9,
251 – 268; b) P. R. Gerber, J. Comput.-Aided Mol. Des. 1998, 12,
37 – 51.
[11] G. M. Popowicz, A. Czarna, S. Wolf, K. Wang, W. Wang, A.
Dꢁmling, T. A. Holak, Cell Cycle 2010, 9, 1104 – 1111.
[12] I. Ugi in Isonitrile Chemistry (Ed.: I. Ugi), Academic Press, New
York, 1971, pp. 149 – 155.
[20] R. Bossio, S. Marcaccini, R. Pepino, Tetrahedron Lett. 1995, 36,
2325 – 2326.
[21] a) J. B. Baell, G. A. Holloway, J. Med. Chem. 2010, 53, 2719 –
2740; b) D. Reed, Y. Shen, A. Shelat, A. Arnold, A. Ferreira, F.
Zhu, N. Mills, D. Smithson, C. Regi, D. Bashford, S. Cicero, B.
Schulman, A. G. Jochemsen, K. Guy, M. A. Dyer,
J. Biol. Chem. 2010, DOI: 10.1074/jbc.M109.056747.
[22] a) S. B. Shuker, P. J. Hajduk, R. P. Meadows, S. W. Fesik, Science
1996, 274, 1531 – 1534; b) M. Pellecchia, I. Bertini, D. Cowburn,
C. Dalvit, E. Giralt, W. Jahnke, T. L. James, S. W. Homans, H.
Kessler, C. Luchinat, B. Meyer, H. Oschkinat, J. Peng, H.
Schwalbe, G. Siegal, Nat. Rev. Drug Discovery 2008, 7, 738 – 745.
[23] a) M. Krajewski, U. Rothweiler, L. DꢅSilva, S. Majumdar, C.
Klein, T. A. Holak, J. Med. Chem. 2007, 50, 4382 – 4387; b) M.
Bista, K. Kowalska, W. Janczyk, A. Dꢁmling, T. Holak, J. Am.
Chem. Soc. 2009, 131, 7500 – 7501.
[24] a) L. T. Vassilev, B. T. Vu, B. Graves, D. Carvajal, F. Podlaski, Z.
Filipovic, N. Kong, U. Kammlott, C. Lukacs, C. Klein, N. Fotouhi,
E. A. Liu, Science 2004, 306, 844 – 848; b) B. L. Grassberger,
et al., J. Med. Chem. 2005, 48, 909 – 912.
[25] K. Ding, Y. Lu, Z. Nikolovska-Coleska, G. Wang, S. Qiu, S.
Shangary, W. Gao, D. Qin, J. Stuckey, K. Krajewski, P. P. Roller,
S. Wang, J. Med. Chem. 2006, 49, 3432 – 3435.
[26] A. Czarna, G. Popowicz, A. Pecak, S. Wolf, G. Dubin, T. Holak,
Cell Cycle 2009, 8, 1176 – 1184.
[27] C. W. Murray, D. C. Rees, Nat. Chem. 2009, 1, 187 – 192.
[28] a) D. C. Fry, S. D. Emerson, S. Palme, B. T. Vu, C.-M. Liu, F.
Podlaski, J. Biomol. NMR 2004, 30, 163 – 173; b) Z. Wang, J.
Malgorzata, T. Lambros, S. Ferguson, R. Goodnow, J. Pharm.
Biomed. Anal. 2007, 45, 720 – 729.
[13] C. Kalinski, M. Umkehrer, S. Gonnard, N. Jꢄger, G. Ross, W.
Hiller, Tetrahedron Lett. 2006, 47, 2041 – 2044.
5356
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 5352 –5356