Discovery of dihydroisoquinolinone derivatives as novel inhibitors of the p53-MDM2 interaction with a distinct binding mode

Article abstract of DOI:10.1016/j.bmcl.2015.06.058

Blocking the interaction between the p53 tumor suppressor and its regulatory protein MDM2 is a promising therapeutic concept under current investigation in oncology drug research. We report here the discovery of the first representatives of a new class of small molecule inhibitors of this protein-protein interaction: the dihydroisoquinolinones. Starting from an initial hit identified by virtual screening, a derivatization program has resulted in compound 11, a low nanomolar inhibitor of the p53-MDM2 interaction showing significant cellular activity. Initially based on a binding mode hypothesis, this effort was then guided by a X-ray co-crystal structure of MDM2 in complex with one of the synthesized analogs. The X-ray structure revealed an unprecedented binding mode for p53-MDM2 inhibitors.

Full text of DOI:10.1016/j.bmcl.2015.06.058

Bioorganic & Medicinal Chemistry Letters xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of dihydroisoquinolinone derivatives as novel inhibitors
of the p53–MDM2 interaction with a distinct binding mode
François Gessier, Joerg Kallen, Edgar Jacoby, Patrick Chène, Thérèse Stachyra-Valat, Stephan Ruetz,
⇑
Sébastien Jeay, Philipp Holzer, Keiichi Masuya, Pascal Furet
Novartis Institutes for BioMedical Research, WKL-136.P.12, CH-4002 Basel, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
Blocking the interaction between the p53 tumor suppressor and its regulatory protein MDM2 is a
promising therapeutic concept under current investigation in oncology drug research. We report here
the discovery of the first representatives of a new class of small molecule inhibitors of this protein–
protein interaction: the dihydroisoquinolinones. Starting from an initial hit identified by virtual screen-
ing, a derivatization program has resulted in compound 11, a low nanomolar inhibitor of the p53–MDM2
interaction showing significant cellular activity. Initially based on a binding mode hypothesis, this effort
was then guided by a X-ray co-crystal structure of MDM2 in complex with one of the synthesized
analogs. The X-ray structure revealed an unprecedented binding mode for p53–MDM2 inhibitors.
Ó 2015 Elsevier Ltd. All rights reserved.
Received 28 May 2015
Accepted 16 June 2015
Available online xxxx
Keywords:
MDM2
p53
Protein–protein interaction inhibitors
Reactivating the p53 tumor suppressor is an attractive thera-
peutic concept in oncology.1 In particular, in cancer cells overex-
pressing MDM2, the main negative regulator of p53, blocking the
interaction between the two proteins is a strategy under intense
investigation to reactivate the tumor suppressor.2 This line of
research has led to several small molecule p53–MDM2 interaction
inhibitors in clinical evaluation.3 Our group has been involved in
this effort since the early peptide work and so far we have dis-
closed two new classes of potent non peptide p53–MDM2 interac-
tion blockers resulting from our medicinal chemistry program in
this area of anticancer drug research.4–6 In this Letter, we report
the discovery of a third new class of p53–MDM2 inhibitors: the
dihydroisoquinolinones.
The starting points of our previously reported p53–MDM2 inhi-
bitors were identified by structure-based design. More precisely,
they were designed according to the ‘central valine’ concept.5
The latter consists in placing a planar aromatic or hetero-aromatic
core moiety within van der Waals distance of V93, a residue occu-
pying a central position in the p53 binding pocket of MDM2.7 This
provides appropriate exit vectors on the core scaffold to attach
substituents that efficiently occupy the three essential sub-pockets
of the MDM2 cleft involved in the recognition of residues Phe 19,
Trp 23 and Leu 26 of the transactivation domain of p53.8
chemotypes. Following a procedure previously published which
involved a combination of several 2D and 3D virtual screening
methods, around 50,000 compounds from the Novartis compound
collection were selected for testing in our p53–MDM2 program.9
From this effort, compound 1 (Table 1) was identified as an inter-
esting hit inhibiting the p53–MDM2 interaction with an IC50 value
of 0.54
l
M in our TR-FRET biochemical assay and showing a weak
anti proliferative activity of 16.5
l
M at the cellular level.10,11
Compound 1 was considered to be an attractive starting point for
further exploration. To guide the design of analogs, we sought
the most plausible hypothesis for the binding mode of compound
1 in the p53 binding pocket of MDM2. Docking studies, combined
with the pharmacophore knowledge obtained from our previous
classes of p53–MDM2 inhibitors, led to the binding model shown
in Figure 1.12 In this model, conforming to our ‘central valine’ con-
cept, the dihydroisoquinolinone core of the compound in a semi-
boat conformation makes extensive van der Waals contacts with
the side chain of MDM2 residue V93. Such a position of the bicyclic
core in the cavity allows the ethoxy substituents and the chloro-
phenyl ring, in a pseudo-equatorial orientation, to occupy the Trp
23 and Phe 19 sub-pockets, respectively, while the methoxyphenyl
ring forms an aromatic p–p stacking interaction with residue H96
in the Leu 26 sub-pocket.13 In addition, the compound carbonyl
group accepts a hydrogen bond from the imidazole ring of H96.
These interactions with H96 had been observed in crystal struc-
tures of MDM2 in complex with our previous p53–MDM2
inhibitors.6 Moreover, a chlorophenyl moiety filling the Trp 23
sub-pocket is a hallmark of p53–MDM2 inhibitors. This binding
In addition to structure-based design, we have resorted to
virtual screening in our search for new p53–MDM2 inhibitor
⇑
Corresponding author. Tel.: +41 61 696 79 90; fax: +41 61 696 48 70.
E-mail address: pascal.furet@novartis.com (P. Furet).
http://dx.doi.org/10.1016/j.bmcl.2015.06.058
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Gessier, F.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.06.058

2
F. Gessier et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
Table 1
Table 2
Biochemical activity of compounds 1–3 and 5
Biochemical activity of compounds 4 and 6–9
O
O
O
O
O
O
R3
R2
R1
N
N
(R)
R
N
Cl
Cl
(racemic)
Compound
R
TR-FRET
IC50 M)
Compound
R1
R2
R3
TR-FRET
IC50 M)
(l
(
l
4
6
Methyl
0.12 0.02
1
2
3
Methoxy
Cl
H
H
H
Cl
H
H
H
0.54 0.08
1.0 0.1
1.4 0.3
F
F
0.17 0.05
N
N
N
7
0.12 0.01
5
Methyl
H
0.38 0.05
O
N
H
N
N
The IC50 values are averages of at least 2 separate determinations.
8
9
0.015 0.003
0.008 (n = 1)
The IC50 values are averages of at least 2 separate determinations.
positions, it could not satisfactorily explain the preference for para
substituents on the Leu 26 sub-pocket phenyl ring, in particular for
the dimethyl amino substituent present in compound 4. At this
point, we decided to focus our efforts on obtaining a crystal struc-
ture of MDM2 in complex with a representative of our dihydroiso-
quinolinone inhibitors to elucidate their binding mode. Our initial
attempts at co-crystallizing MDM2 with inhibitors were hampered
by their low solubility, a high compound concentration being
required in such experiments. To identify suitable tool compounds
for the co-crystallization trials, solubilizing tags were introduced at
several positions of the inhibitor chemotype. Most of these modifi-
cations, although improving solubility, led to some loss of potency
reducing the chances of obtaining a co-crystal. However, with com-
pound 5 (Table 1) an adequate balance of solubility/potency was
reached for successful co-crystallization. It resulted in the first
crystal structure of MDM2 in complex with one of the dihydroiso-
quinolinone inhibitors.
Figure 1. Initial binding model of compound 1 in the MDM2 cavity. The three
MDM2 sub-pockets are labeled PHE (Phe 19), TRP (Trp 23) and LEU (Leu 26). The
putative interactions with H96, hydrogen bond and
by a dashed line and an arrow, respectively.
p–p stacking, are represented
model was thus consistent with the available crystallographic and
pharmacophore information.
We started the optimization of compound 1 by the synthesis of
analogs 2 and 3 in which the para-methoxy group of the parent
The X-ray co-crystal structure of the MDM2/compound 5 com-
plex is shown in Figure 2.14 To our surprise, it revealed an unprece-
dented type of binding mode in which the inhibitor bicyclic core
makes hydrophobic contacts with residues I54, F55 and G58 of
compound was replaced by
a para-chloro and meta-chloro
helix
a2 of MDM2 rather than interacts with V93 as proposed in
substituent, respectively. Based on the binding model and the
structure–activity relationships of our previous classes of
p53–MDM2 inhibitors, chloro substituents on the phenyl ring
occupying the Leu 26 sub-pocket were expected to improve
potency. However, compounds 2 and 3 turned out to be around
two-fold less active than 1 in the biochemical assay (Table 1).
These results prompted us to envisage a more systematic investi-
gation of the dihydroisoquinolinone scaffold to further probe the
binding mode hypothesis. A number of derivatives with diverse
variations at the ether positions and various substitutions with
small groups on the phenyl ring assumed to bind in the Leu 26
sub-pocket were prepared. We obtained structure–activity rela-
tionships indicating that the ether positions were rather tolerant
to the diverse substitutions tried with a slight preference for small
aliphatic groups. As for substitutions on the phenyl ring, there was
a clear preference for a para substituent over an ortho or meta one.
At the end of this investigation, the most potent compound
the binding model. This novel type of binding mode appears facil-
itated by a different side chain conformation of F55 compared to
that observed for this residue in all previously reported MDM2
crystal structures available in the PDB database. In the latter, the
v1 torsion angle of F55 corresponds to a t rotamer which allows
the formation of an intramolecular face to edge aromatic interac-
tion with Y56, a neighboring residue. In contrast, in the complex
with compound 5, a g– rotameric state for
v1 brings the side chain
of F55 into closer proximity to the center of the MDM2 cavity
where it can contact the inhibitor. Besides the core interactions,
the three MDM2 sub-pockets determining binding affinity are
occupied by the inhibitor. The deepest one, Trp 23, is filled by
the compound chlorophenyl ring projecting with a pseudo-axial
orientation from the dihydroisoquinolinone core in a semi-boat
conformation. Hence, compound 5 conforms to the usual Trp 23
sub-pocket pharmacophore. However, it is not the case for the
other sub-pockets. Unlike our previous series of inhibitors, 5 does
obtained was 4, having an IC50 value of 0.12 lM in the biochemical
not form an aromatic
p–p stacking interaction with residue H96
assay (Table 2). Although the binding model could largely account
of the Leu 26 sub-pocket. The compound occupies this sub-pocket
for the structure–activity relationships observed at the ether
Please cite this article in press as: Gessier, F.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.06.058

F. Gessier et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
3
Figure 2. Crystal structure of MDM2 in complex with compound 5 (PDB code:
4ZYC). The hydrogen bond between the inhibitor tetrazole ring and H96 appears as
a dashed line. A water mediated hydrogen bond existing between the carbonyl
group of the inhibitor bicyclic core and residue Q79 is represented in the same
manner. The conformation of the side chain of F55 usually observed in crystal
structures of MDM2 is shown in green. The crystal structure indicates that the
MDM2 pocket is occupied by the enantiomer of 5 with the (S) configuration at the
chiral center.
with one of its two ethoxy substituents, the other one making
hydrophobic contacts with the side chain of F55 in addition to
those made by the bicyclic core with this residue. Still, an interac-
tion with H96 is observed but it does not involve the moiety engag-
ing the Leu 26 sub-pocket. In fact, H96 donates a hydrogen bond to
the tetrazole ring of the solubilizing tag attached to the inhibitor
toluyl group which inserts in the Phe 19 sub-pocket. Overall, a
new way of interacting with the p53 binding cleft of MDM2 was
revealed by this crystal structure. It gave a new basis to the opti-
mization of our dihydroisoquinolinone inhibitors.
Figure 4. (A) Model of MDM2 in complex with
fluorophenyl ring makes an aromatic face to edge interaction with the phenol ring
of Y67. (B) Model of MDM2 in complex with 9 in which the compound pyridyl ring
6
in which the compound
Using a model of compound 4 docked in the MDM2 cavity with
the same binding mode as that of compound 5 in the crystal struc-
ture (Fig. 3), new analogs of the former compound were
designed.15 Examination of the model suggested opportunities to
enhance binding interactions with MDM2 in the Phe 19 sub-pocket
by replacing one of the methyl groups of the dimethylamino sub-
stituent of 4 by a larger group. In particular, this position of the
compound offered an appropriate vector to target Y67, a well
exposed residue of the Phe 19 sub-pocket. On this basis, com-
pounds 6–9 (Table 2) were designed to form aromatic interactions
makes a p–p stacking interaction with the phenol ring of Y67.
Table 3
Biochemical and cellular activity of compounds 10 and 11
O
O
O
O
O
N
N
(S)
(R)
O
(R)
(R)
N
N
N
N
Cl
Cl
11
10
Compound
TR-FRET
IC50 M)
SJSA-1
IC50
SAOS-2
IC50 M)
(
l
(
l
M)
(l
10
11
1.2 0.3
0.0023 0.0004
29.6 (n = 1)
1.2 0.3
>30
12
5
The IC50 values are averages of at least 2 separate determinations.
with the phenol ring of Y67. Modeling indicated that a group of the
benzylic type attached to the para amino substituent of 4 could
interact with the phenol ring of Y67 either in a p–p stacking orien-
tation or a face to edge one. Both types of aryl–aryl interaction are
strengthened by introducing electron withdrawing substituents or
heteroatoms in the ring having as partner an electron rich aromatic
system such as phenol.16 This is why fluorophenyl and pyridine
rings were chosen in compounds 6–9 to interact with Y67. An illus-
tration of these design ideas is given in Figure 4.
Figure 3. Compound 4 docked in the MDM2 pocket. The green arrow indicates the
direction of derivatization to interact with residue Y67 of the Phe 39 sub-pocket.
Please cite this article in press as: Gessier, F.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.06.058

4
F. Gessier et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
HO
O
O
O
O
O
O
O
a
b
c
O
O
O
(R)
(R)
(R)
O
O
O
OH
Cl
12
13
14
15
e
Cl
H2N
CHO
d
NO2
+
NO2
Cl
N
Cl
Cl
Cl
N
NO2
NH2
N
f
g, h
O
O
O
O
N
N
N
(R)
(R)
(R)
O
O
O
O
O
16
17
9
Scheme 1. Reactions and conditions: (a) (S)-butan-2-ol, PPh3, DTAD, DCM, rt, quant.; (b) LiOH, EtOH, water, rt, 2 h, 71%; (c) (COCl)2, DMF, DCM, rt, 1 h, quant.; (d) neat, rt,
3 days, 80%; (e) 14, DCM, rt, 1 h then MeSO3H, 0 °C, 15 min, 47%; (f) Fe, AcOH, AcOEt, water, 80 °C, 1 h, 99%; (g) isonicotinaldehyde, NaBH(OAc)3, AcOH, DCM, rt, 3 h, 82%; (h)
37% HCHO in water, NaBH(OAc)3, AcOH, DCM, rt, 3 h, 54%.
Compounds 6–9 were synthesized and tested in the biochemi-
cal assay (Table 2). The fluorophenyl derivatives 6 and 7 remained
at the same level of potency as the parent compound 4. However,
gratifyingly, a marked improvement of activity by one order of
magnitude was obtained with the pyridyl analogs 8 and 9.17 In fact,
compound 9 represented the first member of the new series reach-
ing single digit nanomolar potency in the biochemical assay.
Encouraged by this result, we decided to separate the two
diastereoisomers of 9. We obtained compounds 10 and 11 shown
in Table 3 together with their activity. Based on the crystal struc-
ture of the analog 5 in complex with MDM2, we attributed the S
configuration to the bicycle chiral center of the most active
diastereoisomer 11. This compound reached low nanomolar bio-
chemical potency, accompanied by significant and specific inhibi-
tion of the proliferation of the p53-dependent SJSA-1 cells in the
low micromolar range.18 With 11, we had reached an interesting
level of cellular potency comparable to that of the reference
chromatography and required the use of chiral chromatography
conditions.
In conclusion, improvements to a virtual screening hit have led
to the first potent representatives of a new class of inhibitors of the
p53–MDM2 interaction based on a dihydroisoquinolinone scaffold.
A crystal structure shows that the new inhibitors adopt a different
binding mode in the MDM2 cavity compared to all the inhibitors
reported in the literature. The potent inhibitor 11 has been the
starting point of another round of medicinal chemistry optimiza-
tion that has culminated in NVP-CGM097, a compound currently
under evaluation in a phase I clinical trial for cancer patients.19,20
References and notes
1. Selivanova, G. FEBS Lett. 2014, 588, 2628.
2. For recent reviews on the topic see: (a) Lv, P.-C.; Sun, J.; Zhu, H.-L. Curr. Med.
Chem. 2015, 22, 618; (b) Nag, S.; Zhang, X.; Wang, M.-H.; Srivenugopal, K. S.;
Wang, W.; Zhang, R. Curr. Med. Chem. 2014, 21, 553; (c) Wade, M.; Li, Y. C.;
Wahl, G. M. Nat. Rev. Cancer 2013, 13, 83; (d) Li, Q.; Lozano, G. Clin. Cancer Res.
2013, 19, 34; (e) Carry, J. C.; García-Echeverría, C. Bioorg. Med. Chem. Lett. 2013,
23, 2480; (f) Zak, K.; Pecak, A.; Rys, B.; Wladyka, B.; Doemling, A.; Weber, L.;
Holak, T. A.; Dubin, G. Expert Opin. Ther. Pat. 2013, 23, 425; (g) Wang, S.; Zhao,
Y.; Bernard, D.; Aguilar, A.; Kumar, S. Top. Med. Chem. 2012, 8, 57; (h) Capdevila,
J.; Cervantes, A.; Tabernero, J. Drugs Future 2012, 37, 273; (i) Ray-Coquard, L.;
Blay, J. Y.; Italiano, A.; LeCesne, A.; Penel, N.; Zhi, J.; Heil, F.; Rueger, R.; Graves,
B.; Ding, M.; Geho, D.; Middelton, S. A.; Vassilev, L. T.; Nichols, G. L.; Bui, B. N.
Lancet Oncol. 2012, 13, 1133; (j) Macchiarulo, A.; Giacchè, N.; Carotti, A.;
Moretti, F.; Pellicciari, R. Med. Chem. Commun. 2011, 2, 455; (k) Cheok, C. F.;
Verma, C. S.; Baselga, J.; Lane, D. P. Nat. Rev. Clin. Oncol. 2011, 8, 25; (l) Millard,
M.; Pathania, D.; Grande, F.; Xu, S.; Neamati, N. Curr. Pharm. Des. 2011, 17, 536;
(m) Brown, C. J.; Cheok, C. F.; Verma, C. S.; Lane, D. P. Trends Pharmacol. Sci.
2011, 32, 53; (n) Vu, B. T.; Vassilev, L. In Current Topics in Microbiology and
Immunology (Small Molecules Inhibitors of Protein–Protein Interactions); Vassilev,
L., Fry, D., Eds.; ; Springer: Berlin Heidelberg, 2011; Vol. 348, pp 151–172; (o)
Weber, L. Expert Opin. Ther. Pat. 2010, 20, 179.
3. (a) Zhao, Y.; Aguilar, A.; Bernard, D.; Wang, S. J. Med. Chem. 2015; (b) Hoe, K. K.;
Verma, C. S.; Lane, D. P. Nat. Rev. Drug Disc. 2014, 13, 217; (c) Rew, Y.; Sun, D. J.
Med. Chem. 2014, 57, 6332.
4. García-Echeverría, C.; Chène, P.; Blommers, M. J. J.; Furet, P. J. Med. Chem. 2000,
43, 3205.
5. Furet, P.; Chène, P.; De Pover, A.; Stachyra-Valat, T.; Hergovich Lisztwan, J.;
Kallen, J.; Masuya, K. Bioorg. Med. Chem. Lett. 2012, 22, 3498.
6. Vaupel, A.; Bold, G.; De Pover, A.; Stachyra-Valat, T.; Hergovich Lisztwan, J.;
Kallen, J.; Masuya, K.; Furet, P. Bioorg. Med. Chem. Lett. 2014, 24, 2110.
7. To avoid confusion the one letter code is used to name the amino acid residues
of MDM2 while the three letter code is used for those of p53.
p53–MDM2 inhibitor Nutlin-3a (IC50 = 1.9 lM in the SJSA-1 assay).
A representative synthesis for compound 9 is described in
Scheme 1. The commercially available methyl 2-(4-hydroxy-3-
methoxyphenyl)acetate was subjected to a Mitsunobu reaction
with (S)-butan-2-ol using DTAD and triphenylphosphine as
reagents to deliver the sec-butyl ether 12 as its pure (R)-enan-
tiomer. The methyl ester functionality of 12 was hydrolyzed with
LiOH into the corresponding carboxylic acid 13 which reacted with
oxalyl chloride to give the acyl chloride 14 in quantitative yields.
The cyclo-addition of 14 with the imine 15, obtained from the con-
densation of the commercially available 4-chlorobenzaldehyde and
4-nitroaniline, led to the dihydroisoquinolinone analog 16 with no
stereo-selectivity observed for the newly created asymmetric cen-
ter. The nitro group of 16 was reduced with iron following classical
reaction conditions to give the corresponding aniline 17 in nearly
quantitative yields. Finally, the aniline functionality underwent
two consecutive reductive aminations with isonicotinaldehyde
and formaldehyde, respectively, using NaBH(OAc)3 as a reducing
agent to deliver compound 9 as a mixture of two diastereoisomers.
At that point, the separation of the diastereoisomer pair could not
be achieved by classical normal phase or reversed phase
Please cite this article in press as: Gessier, F.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.06.058

F. Gessier et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
5
8. Kussie, P. H.; Gorina, S.; Marechal, V.; Elenbaas, B.; Moreau, J.; Levine, A. J.;
Pavletich, N. P. Science 1996, 274, 948.
9. Jacoby, E.; Boettcher, A.; Mayr, L. M.; Brown, N.; Jenkins, J. L.; Kallen, J.;
Engeloch, C.; Schopfer, U.; Furet, P.; Masuya, K.; Lisztwan, J. Methods Mol. Biol.
2009, 575, 173.
conformational energy penalty to binding of a bit more than 1 kcal/mol can
be compensated by a few favorable van der Waals contacts with the binding
site.
14. Kallen, J. et al. The details of this X-ray structure determination (solved at
1.95 Å resolution) will be published elsewhere. The coordinates have been
deposited with PDB ID code 4ZYC.
10. For
a detailed description of the TR-FRET (Time Resolved Fluorescence
Resonance Energy Tranfer) biochemical assay used see: Berghausen, J.;
Buschmann, N.; Furet, P.; Gessier, F.; Hergovich Lisztwan, J.; Holzer, P.;
Jacoby, E.; Kallen, J.; Masuya, K.; Pissot Soldermann, C.; Ren, H.; Stutz, S. PCT
Int. Appl. WO 2011076786, 2011; Chem. Abstr. 2011, 155, 152537. In this assay
the donor fluorophore is MDM2 (amino acid residues 2-188) tagged with a C-
terminal biotin moiety in combination with a Europium labeled streptavidin.
The acceptor fluorophore is a p53 derived peptide (amino acid sequence 18–26
of p53: TFSDLWKLL) labeled with the fluorescent dye Cy5. The IC50 values
given are means of at least two measurements. For reference, the p53–MDM2
15. Compound 4 was aligned on 5 in the crystal structure shown in Figure 2 and
after removal of 5 the resulting complex was energy minimized using the
procedure described in note 12.
16. Salonen, L. M.; Ellermann, M.; Diederich, F. Angew. Chem., Int. Ed. 2011, 50,
4808.
17. The reason why the fluorophenyl analogs did not lead to a gain in activity is not
clear. Docking based on the crystal structure suggests that the fluorophenyl
rings can make good van der Waals contacts with Y67 or possibly the adjacent
residue M62. In the spirit of our design hypothesis, one could conclude that
introduction of heteroatoms in the ring interacting with the phenol ring of Y67
inhibitor Nutlin-3A has an IC50 of 0.01 lM in this assay.
11. Our cellular assay measures the ability of compounds to inhibit the
proliferation of SJSA-1 cells. These are p53 positive cancer cells in which the
MDM2 gene is amplified. For control, inhibition of the proliferation of the p53-
null SAOS-2 cells is also measured. A detailed description of the cellular SJSA-1
and SAOS-2 proliferation assays is given in: Furet, P.; Guagnano, V.; Holzer, P.;
Kallen, J.; Liao, L.; Mah, R.; Mao, L.; Masuya, K.; Schlapbach, A.; Stutz, S.; Vaupel,
A. PCT Int. Appl. WO 201311105. Compound 1 inhibits the proliferation of
is more efficient than electron withdrawing substituents to produce
favorable aryl–aryl contact.
a
18. In line with our previous experience, at least two orders of magnitudes in
potency are lost going from biochemical to cellular setting. As the
a
a
consequence of the p53–MDM2 auto-regulatory loop existing in cells, the
accumulation of p53 caused by the inhibitors triggers an increase of cellular
MDM2 levels which reduces inhibitor potency. See for example: Lahav, G. In
Cellular Oscillatory Mechanisms; Maroto, M., Monk, N. A. M., Eds.; Landes
Bioscience and Springer Science, 2008; pp 28–38. The fact that in cells, the
inhibitors compete with full length p53 instead of the truncated form used in
the biochemical assay is also likely to contribute to this loss of potency.
19. Jeay, S.; Berghausen, B.; Buschmann, N.; Chène, P. Cozens, R; Erdmann, D.;
Ferretti, S.; Furet, P.; Gabriel, T.; Gessier, F.; Graus-Porta, D.; Hofmann; F.;
Holzer, P.; Ito, M.; Jacoby, E.; Jensen, M.; Kallen, J.; Lang, M.; Lisztwan, J.;
Murakami, M.; Pissot-Soldermann, C.; Ruetz, S.; Rynn, C.; Sterker, D.; Stutz, S.;
Valat, T.; Wiesmann, M.; Masuya, K. Discovery of NVP-CGM097, A Highly
Potent and Optimized Small Molecule of MDM2 Under Evaluation in a Phase I
Clinical Trial. In: Proceedings of the 105th Annual Meeting of the American
Association for Cancer Research; 2014 Apr 5–9; San Diego. Philadelphia (PA):
AACR; Cancer Res. 2014; 74: Abstract nr 1797.
SJSA-1 and SAOS-2 cells with IC50 values of 16.6 lM and 19.6 lM, respectively.
Compounds with micromolar/submicromolar biochemical activity such as 1 do
not usually show a significant specific antiproliferative effect on SJSA-1 cells.
Nanomolar biochemical activity is required for this.
12. Modeling and docking was performed with a version of MacroModel enhanced
for graphics by A. Dietrich. MacroModel: Mohamadi, F.; Richards, N. G. J.;
Guida, W. C.; Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.;
Still, W. C. J. Comput. Chem. 1990, 11, 440. The compounds were manually
constructed and docked in the MDM2 pocket (PDB structure 4OQ3) and the
resulting ligand–protein complexes energy-minimized using the AMBER⁄/H2O/
GBSA force field.
13. Ab initio calculations in Jaguar (Schrödinger Inc.) at the B3LYP/6-31G⁄⁄ level
indicate that in the lowest energy conformation of 1, the chlorophenyl ring is
pseudo-axial. The pseudo-equatorial conformation is less stable by
1.6 kcal/mol. Since 1 is not a highly potent inhibitor, we judged plausible
20. The optimization of compound 11 towards a clinical candidate will be reported
elsewhere.
that it does not bind to MDM2 in its lowest energy conformation.
A
Please cite this article in press as: Gessier, F.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.06.058