Phenylindenone isomers as divergent modulators of p38α MAP kinase

Article abstract of DOI:10.1016/j.bmcl.2016.10.001

Two new fluorophenylindenone derivatives were designed as potential p38α MAPK modulators by preserving the key interactions of the vicinal pyridine/fluorophenyl pharmacophore with the enzyme protein. Interestingly, these two fluorophenylindenone isomers showed divergent activities, with compound 6 behaving as an inhibitor and 5 as a putative activator. These results were rationalized by docking studies and molecular dynamics simulations in terms of stabilization of DFG loop, by compound 5 in a conformation more accessible to phosphorylation.

Full text of DOI:10.1016/j.bmcl.2016.10.001

Accepted Manuscript
Phenylindenone Isomers as Divergent Modulators of p38α MAP Kinase
Andrea Cappelli, Chiara Nannicini, Alessia Chelini, Marco Paolino, Germano
Giuliani, Maurizio Anzini, Antonio Giordani, Chiara Sabatini, Gianfranco
Caselli, Laura Mennuni, Francesco Makovec, Gianluca Giorgi, Salvatore
Vomero, Maria C. Menziani
PII:
S0960-894X(16)31029-0
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.10.001
BMCL 24304
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
12 August 2016
29 September 2016
2 October 2016
Please cite this article as: Cappelli, A., Nannicini, C., Chelini, A., Paolino, M., Giuliani, G., Anzini, M., Giordani,
A., Sabatini, C., Caselli, G., Mennuni, L., Makovec, F., Giorgi, G., Vomero, S., Menziani, M.C., Phenylindenone
Isomers as Divergent Modulators of p38α MAP Kinase, Bioorganic & Medicinal Chemistry Letters (2016), doi:
http://dx.doi.org/10.1016/j.bmcl.2016.10.001
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Phenylindenone Isomers as Divergent Modulators of p38α MAP Ki-
nase
Andrea Cappelli,a* Chiara Nannicini,a Alessia Chelini,a Marco Paolino,a Germano Giuliani,a
Maurizio Anzini,a Antonio Giordani,b Chiara Sabatini,b Gianfranco Caselli,b Laura Mennuni,b
Francesco Makovec,b Gianluca Giorgi,a Salvatore Vomero,a and Maria C. Menziani.c
a Dipartimento di Biotecnologie, Chimica e Farmacia and European Research Centre for Drug Discovery and Development,
Università degli Studi di Siena, Via A. Moro 2, 53100 Siena, Italy. b Rottapharm Biotech S.r.l., Via Valosa di Sopra 3, 20900
Monza, Italy. c Dipartimento di Scienze Chimiche e Geologiche, Università degli Studi di Modena e Reggio Emilia, Via
Campi 103, 41125 Modena, Italy.
ABSTRACT: Two new fluorophenylindenone derivatives were designed as potential p38α MAPK modulators by preserv-
ing the key interactions of the vicinal pyridine/fluorophenyl pharmacophore with the enzyme protein. Interestingly, these
two fluorophenylindenone isomers showed divergent activities, with compound 6 behaving as an inhibitor and 5 as a pu-
tative activator. These results were rationalized by docking studies and molecular dynamics simulations in terms of stabi-
lization of DFG loop, by compound 5 in a conformation more accessible to phosphorylation.
level of attention both in the pharmaceutical industry and
in the academia. Thus, the medicinal chemistry work has
produced a large number of inhibitors (see Figure 1 for
some representative structures), and some of them have
been co-crystallized with the enzyme in order to evaluate
the drug-protein interactions in the complexes.11
Keywords: MAPK, kinases, synthesis, docking
* Corresponding author. Tel: +39 0577 234320. E-mail:
andrea.cappelli@unisi.it.
Mitogen-activated protein kinases (MAPK) are protein
kinases that can be classified in several distinct groups
composed of sets of three evolutionarily conserved, se-
quentially acting kinases: MAPK, MAPK kinase (MKK),
and MKK kinase (MAP3K).1-3 The activation of MAP3K
leads to phosphorylation and activation of an MKK, which
in turn stimulates MAPK activity through dual phosphor-
ylation on threonine and tyrosine residues located in the
activation loop. Once activated, MAPK phosphorylate cy-
toplasmic and nuclear target substrates on serine or thre-
onine residues.
p38 proteins are a class of MAPK playing a crucial role
in important cellular processes (i. e. cell differentiation,
apoptosis, and autophagy) and are responsive to stress
stimuli such as heat and osmotic shock, ultraviolet irradi-
ation, and inflammatory cytokines. The p38 MAPK family
includes four different isoforms, namely p38α, p38β, p38γ,
p38δ.4 Among them, p38α has been reported to be largely
expressed in monocytes and macrophages, whereas p38β
is abundant in endothelial cells, p38γ in skeletal muscle,
and p38δ in testes, pancreas, prostate, small intestine and
in certain endocrine tissues.5 Substrates of p38 MAPKs
include a large number of different proteins and a signifi-
cant portion of them is involved in the regulation of gene
expression.6 p38 has been considered an interesting target
for the development of anti-inflammatory drugs.7-10 Thus,
over the past two decades, the development of small mol-
ecule p38α MAPK inhibitors has received an extraordinary
Figure 1. Comparison of the structures of the newly designed
isomeric indenone derivatives 5 and 6 with those of some
selected p38α MAP kinase inhibitors.
The most explored chemical class of p38α MAPK inhibi-
tors is characterized by the presence of a nitrogen-
containing six member heterocyclic moiety (pyridine, py-
rimidine, or equivalents) in a vicinal position to a lipo-
philic (most often a p-fluorophenyl) group. The develop-

ment of this family of compounds has been stimulated by
On the other hand, isomeric indenone 5 failed in show-
ing an inhibitory activity at both the ATP concentrations
tested. Surprisingly on the contrary, at the ATP concen-
tration corresponding to the Km value, compound 5 ap-
peared to increase MAPK14 phosphorylation, thus behav-
ing as a partial activator (activation effect around 40-50%
at 10 μM, Figure 3).
the interesting results obtained with compound
1
(SB203580) and its congeners by SmithKline Beecham.12
The exploration of the chemical space around this struc-
ture has produced inhibitors showing potency in the
nanomolar range and the discovery of 4-azaindole deriva-
tives 4a,b.13
In order to evaluate the effects of further core modifica-
tions on the interaction with p38α, fluorophenylindenone
derivatives 5 and 6 (Figure 1) were designed by replacing
the imidazole heterocyclic core of 1 with the indenone
scaffold. The central pharmacophore consisting in the vic-
inal pyridine/fluorophenyl system was conserved so that
the pyridine nitrogen could behave as the hydrogen bond
acceptor for the NH of Met-109 and the 4-fluorophenyl
moiety could occupy the hydrophobic back pocket. On
the other hand, the indenone moiety of 5 and 6 could find
different accommodations in the phosphate/sugar region
of the binding site.
Figure 3. MAPK14 (p38α) phosphorylation in the presence of
compound 5 and different ATP concentrations, 100 μM or
500 μM (close to the Km value). Data are expressed as mean ±
sem.
Compounds 5 and 6 were synthesized as de-
scribed in the Supporting Information, characterized by
crystallography [CCDC 1487925 (5), 1487926 (6), see Sup-
porting Information], and tested for their potential modu-
latory activity on p38α MAPK enzyme at Life Technolo-
gies Ltd.14 [SelectScreen Kinase Profiling, Protein Ser-
ine/Threonine Kinase MAPK14 (p38α) direct Z-LITE ki-
nase assays platform, for the experimental details see
http://www.lifetechnologies.com/selectscreen] by using
two different ATP concentrations, namely 500 μM (close
to the Km value) and 100 μM.15
In order to verify this surprising result, compound 5
was re-sampled and re-assayed in the same test system
and the results confirmed those obtained in the prelimi-
nary assay. In fact, the top three concentrations induced a
significant increase of phosphorylation. This phenome-
non could be caused by compound-dependent activation
of the kinase, or by undetected assay interference of some
sort. In order to rule out fluorescence interferences and
development reaction interferences, control wells were
used and compound 5 did not show any evidence of inter-
ference. On the other hand, the activation effect appeared
to be almost negligible at 100 μM ATP, suggesting the im-
portance of the ATP concentration in revealing the activa-
tion features of indenone derivative 5.
In the enzymatic studies, indenone derivative 6 showed
significant inhibitory properties, with a potency in the
micromolar range that appeared to be dependent by the
concentration of ATP used in the enzymatic assay (Figure
2).
Thus, the preliminary results of the enzymatic studies
performed with compound 5 was surprising because this
indenone isomer of the p38α MAPK inhibitor 6 appeared
to behave as activator.
Owing to these intriguing results, docking studies and
molecular dynamics simulations were performed in order
to rationalize the apparently divergent behavior obtained
with the isomeric indenone derivatives. Docking of com-
pounds 5 and 6 to p38α MAPK was performed on the ba-
sis of the structural similarity of these compounds with
the 4-azaindole inhibitor 4b (type I inhibitor),11 the crystal
structure of which, in complex with p38α MAPK, has been
resolved (PDB code :1OZ1, Figure 4).13
Figure 2. MAPK14 (p38α) phosphorylation in the presence of
compound 6 and different ATP concentrations, 100 μM or
500 μM (close to the Km value). Data are expressed as mean ±
sem.
The X-ray crystal structure of the complex showed that
the pyridine nitrogen of the 4-azaindole inhibitor 4b be-
haved as the hydrogen bond acceptor for the NH of
Met109, the 4-fluorophenyl moiety occupied the hydro-
phobic back pocket called hydrophobic region I (HR-I),
and the 4-azaindole moiety was capable of direct hydro-
gen bonding the terminal nitrogen of Lys53. The hydro-
gen bond with Lys53 was claimed to be required for the
In particular, with the ATP concentration close to the
Km value the concentration-effect curve for compound 6
is slightly shifted to right compared with that observed
with a lower ATP concentration. Accordingly the IC50 val-
ue in the presence of 500 µM ATP was slightly but signifi-
cantly higher (1.35 μM vs. 0.899 μM, for 500 and 100 µM
ATP, respectively).

regioselective positioning of the pyridine and the 4-
fluorophenyl groups.16
As expected, fluorophenylindenone derivatives 5 and 6
(Figure 5) maintained the key interactions of type I inhib-
itors. In particular, a hydrogen bond is established be-
tween pyridine nitrogen and the amide proton of Met109,
and significant hydrophobic interactions involve the 4-
fluorophenyl group with hydrophobic region I. HR-I is
located at the back of the ATP binding site, and is defined
by Ala51, Lys53, Leu75, Ile84, Leu104, Thr106 and Leu167.11
Most importantly, compound 6 also accepts a hydrogen
bond from Lys53 (Figure 5). On the contrary, in com-
pound 5 the oxygen acceptor of H-bond is located at the
opposite side, preventing the achievement of this interac-
tion (Figure 5), which is necessary to engage the inhibitor
in loco.13,16 This may trigger the different behaviour of the
two ligands in the enzymatic essay. In fact, a marked con-
formational change upon binding of compound 5 is ob-
served for Arg173, which moves away from Tyr35 of the
Gly–loop (aa res. 31-37) and establishes a network of
strong electrostatic interactions with Asp168 and Glu178
of the DFG loop. Consequently, the binding site is more
accessible, being the gateway of the binding pocket
around 3 Å larger, the DFG motif is stabilized in its active
DFG-in conformation and the phosphorylation sites more
accessible (Figure 6).
Figure 4. 4-Azaindole inhibitor 4b bound to p38α MAPK
(PDB code: 1OZ1).13
Therefore, in the docking of compounds 5 and 6 we as-
sumed that the ligands target the ATP binding site of the
enzyme in its active form. The active form of the enzyme
is characterized by an open conformation of the activa-
tion loop (DFG-loop), which is usually referred to as DFG
“in” on the basis of the position of the Phe169 residue be-
longing to the conserved triad Asp168-Phe169-Gly170 at
the beginning of the activation loop.17-19
In conclusion, two new fluorophenylindenone deriva-
tives (5 and 6) were designed from p38α MAPK inhibitor 1
by replacing its imidazole heterocyclic core with
indenone scaffold and conserving the central
pharmacophore (i. e. the vicinal pyridine/fluorophenyl
system). In our working hypothesis, the pyridine nitrogen
of 5 and 6 could behave as the hydrogen bond acceptor
for the NH of Met-109, the 4-fluorophenyl moiety could
occupy the hydrophobic back pocket, and the indenone
moiety could find different accommodations in the phos-
phate/sugar region of the binding site. The
fluorophenylindenone isomers 5 and 6 were synthesized,
characterized by X-ray crystallography, and tested for
their potential modulatory activity on p38α MAPK en-
zyme by using two different ATP concentrations. In these
studies, compound 6 behaved as an inhibitor, whereas
isomeric indenone 5, at the ATP concentration corre-
sponding to the Km value, appeared to behave as an acti-
vator. This unprecedented result was rationalized by
docking studies, which showed that compounds 5 and 6
retained the typical interactions of type I inhibitors, as
expected. The main difference was the absence of the hy-
drogen bond interaction from Lys53 in the complexes of
activator 5 that produced a remarkable conformational
change resulting in the stabilization of DFG loop in a con-
formation more accessible to phosphorylation.
Figure 5. Docking of compound 5 (top) and 6 (bottom) to
p38α. DFG loop is in the “in” conformation (binding mode of
type I inhibitors).

Figure 6. 3D structure of complexes between the compound 5 (left) and 6 (right) and p38α MAPK, as resulted from the molecu-
lar dynamics simulation.
11.
Cecchetti, V. Chem. Med . Chem. 2015, 10, 957.
12. Clerk, A.; Sugden, P. H. FEBS Lett. 1998, 426, 93 and
references cited therein.
Astolfi, A.; Iraci, N.; Manfroni, G.; Barreca, M. L.;
Acknowledgment
The authors are grateful to Dr. Francesco Berrettini (CIADS,
Università di Siena) for the X-ray data collection and to Prof.
Alessandro Donati (Università di Siena) for the NMR experiments
performed with compound 6.
13.
Trejo, A.; Arzeno, H.; Browner, M.; Chanda, S.; Cheng,
S.; Comer, D. D.; Dalrymple, S. A.; Dunten, P.; Lafargue, J.;
Lovejoy, B.; Freire-Moar, J.; Lim, J.; Mcintosh, J.; Miller, J.; Papp,
E.; Reuter, D.; Roberts, R.; Sanpablo, F.; Saunders, J.; Song, K.;
Villasenor, A.; Warren, S. D.; Welch, M.; Weller, P.; Whiteley, P.
E.; Zeng, L.; Goldstein, D. M. J. Med. Chem. 2003, 46, 4702.
14.
15.
16.
http://www.lifetechnologies.com
Szafranska, A. E.; Dalby, K. N. FEBS J. 2005, 272, 4631.
Wrobleski, S. T.; Doweyko, A. Curr. Top. Med. Chem.
References and notes
1.
Pearson, G.; Robinson, F.; Gibson, T. B.; Xu, B. E.;
2005, 5, 1005.
Karandikar, M.; Berman, K.; Cobb, M. H. Endocr. Rev. 2001, 22,
153.
17.
Vogtherr, M.; Saxena, K.; Hoelder, S.; Grimme, S.; Betz,
M.; Schieborr, U.; Pescatore, B.; Robin, M.; Delarbre, L.; Langer,
T.; Wendt, K. U.; Schwalbe, H. Angew. Chem., Int. Ed. 2006, 45,
993.
2.
Lin, S.C., Han, J. H. J. Biol. Chem. 1996, 271, 17920.
3. Kumar, S.; McDonnell, P. C.; Gum, R. J.; Hand, A. T.;
Lee, J. K.; Young, P. R. Biochem. Biophys. Res. Commun. 1997,
235, 533.
Jiang, Y.; Chen, C. H.; Li, Z. J.; Gou, W.; Gegner, J. A.;
18.
Zuccotto, F.; Ardini, E.; Casale, E.; Angiolini, M. J. Med.
Chem. 2010, 53, 2681.
19.
Filomia, F.; De Rienzo, F.; Menziani, M. C. Bioorg. Med.
4.
Saklatvala, J. Curr. Opin. Pharmacol. 2004, 4, 372.
Chem. 2010, 18, 6805.
5.
Hale, K. K.; Trollinger, D.; Rihanek, M.; Manthey. C. L.
J. Immunol. 1999, 162, 4246.
6.
7.
8.
Igea, A.; Nebreda, A. R. Cancer Res. 2015, 75, 3997.
Norman, P. Expert Opin. Investig. Drugs 2015, 24, 383.
Fisk, M.; Gajendragadkar, P. R.; Mäki-Petäijä, K. M.;
Supporting Information
Experimental details for the synthesis and the characteriza-
tion of target compounds and their intermediates. The exper-
imental procedures used in molecular modeling studies.
Wilkinson, I. B.; Cheriyan, J. Am. J. Cardiovasc. Drugs 2014, 14,
155.
9.
Foster, M. L.; Halley, F.; Souness, J. E. Drug News
Perspect. 2000, 13, 488.
10.
Boehm, J. C.; Adams, J. L. Expert Opin. Ther. Pat. 2000,
10, 25.

O
O
F
F
N
N
INHIBITOR
ACTIVATOR
5