Design, Synthesis, and Characterization of a Photoactivatable Caged Prodrug of Imatinib

Article abstract of DOI:10.1002/cmdc.201500163

Imatinib is the first protein kinase inhibitor approved for clinical use and is a seminal drug for the concept of targeted therapy. Herein we report on the design, synthesis, photokinetic properties, and in vitro enzymatic evaluation of a photoactivatable caged prodrug of imatinib. This approach allows spatial and temporal control over the activation of imatinib triggered by ultraviolet light. The successful application of the photoactivation concept to this significant kinase inhibitor provides further evidence for the caging technique as a feasible approach in the kinase field. The presented photoactivatable imatinib prodrug will be highly useful as a pharmacological tool to study the impact of imatinib toward biological systems in greater detail. Don′t fence me in! A caged prodrug of imatinib, the first approved kinase inhibitor, was designed, synthesized, and characterized for photokinetic properties. By using a PDGF-Rβ assay, we demonstrated that upon UV irradiation at λ=365nm the biological activity of imatinib is restored.

Full text of DOI:10.1002/cmdc.201500163

DOI: 10.1002/cmdc.201500163
Communications
Design, Synthesis, and Characterization of
a Photoactivatable Caged Prodrug of Imatinib
Melanie Zindler, Boris Pinchuk, Christian Renn, Rebecca Horbert, Alexander Dçbber, and
[a]
Christian Peifer*
Imatinib is the first protein kinase inhibitor approved for clini-
cal use and is a seminal drug for the concept of targeted ther-
apy. Herein we report on the design, synthesis, photokinetic
properties, and in vitro enzymatic evaluation of a photoactivat-
able caged prodrug of imatinib. This approach allows spatial
and temporal control over the activation of imatinib triggered
by ultraviolet light. The successful application of the photoacti-
vation concept to this significant kinase inhibitor provides fur-
ther evidence for the caging technique as a feasible approach
in the kinase field. The presented photoactivatable imatinib
prodrug will be highly useful as a pharmacological tool to
study the impact of imatinib toward biological systems in
greater detail.
can be generated at a defined time point in an irradiated area
[
16,17]
of interest.
The caged prodrug concept is essentially based
on covalently blocking a pharmacophore moiety by a PPG, ren-
[
18]
dering the compound biologically inactive. For this purpose
[
19]
[20]
o-nitrobenzylic and coumarylmethyl derivatives have been
widely used, among others, as PPGs in various biological appli-
[
21–23]
cations.
There are currently only few reports on photoacti-
vatable kinase inhibitors. These include a caged small-molecule
[
24]
[21]
Rho kinase inhibitor, equivalents of Src kinase, and pepti-
[
25]
dic PKA inhibitors. Recently, a study about photo-switchable
[
26]
RET kinase inhibitors was published.
In this study, we set out to design, synthesize, and character-
ize a photoactivatable derivative of imatinib. First, by using
molecular modeling we sought to define a suitable pharmaco-
phoric position in imatinib to covalently attach the PPG. For
this purpose we docked the type-II binder imatinib into the
active site of our in-house PDGF-Rb homology model in the
Since the introduction of the tyrosine kinase inhibitor imati-
[1]
nib (Gleevec/Glivec) into the market in 2001, protein kinase
inhibitors have been successfully established mainly in target-
[
27]
DFG-out conformation. We also chose this receptor tyrosine
[
2]
[28]
ed/personalized therapeutic approaches against cancer. Origi-
kinase, as imatinib potently blocks PDGF-Rb (IC =0.1 mm),
50
[3]
nally, the prototypical type-II inhibitor imatinib was approved
for the therapy of Philadelphia chromosome-positive chronic
and an enzymatic PDGF-Rb assay has been established by our
[
27,29]
research group to test the compounds.
+
[4]
myelogenous leukemia (Ph -CML). In this way, at the molecu-
lar level the drug targets the Abelson murine leukemia viral
The modeled ligand interaction diagram of imatinib in the
ATP binding pocket of PDGF-Rb revealed two prominent posi-
tions to covalently attach a PPG (Figure 1). Namely, the NH
functions of the amide in the benzanilide and the N-arylpyrimi-
dine-2-amine moiety both address key hydrogen bonds to
PDGF-Rb. Moreover, the tight ligand–protein complex suggest-
ed no steric tolerance at these positions within the binding
pocket. In line with this notion, modeling of both N-o-nitro-
benzyl-substituted imatinib derivatives in the active site of
PDGF-Rb did not result in plausible binding modes (not
shown).
[5]
oncogene homolog 1 (ABL1). In addition to ABL1, it was dis-
covered that imatinib also potently inhibits further tyrosine
kinases involved in malignancies including c-KIT and platelet-
[6]
derived growth factor receptor (PDGF-R). Therefore, imatinib
has become an option for the treatment of multiple cancers in-
[7,8]
cluding gastrointestinal tumors (GIST).
Along with the fact
that significant progress has been made in cancer therapy, im-
atinib has also been used in countless experimental in vitro
[
9–12]
and in vivo studies.
Considering the prominent role of this
multi-kinase inhibitor in signal-transduction analysis, it would
be beneficial to develop a photoactivatable imatinib prodrug.
Therefore, we aimed to provide a so-called caged com-
pound that can be activated by irradiation with ultraviolet (UV)
To determine if one of these NH positions in imatinib would
be chemically suitable for subsequent PPG photocleavage, we
first synthesized an imatinib fragment bound to 4,5-dime-
thoxy-2-nitrobenzyl (compound 1, Figure 2; the N-[(2-nitrophe-
nyl)methyl]-N-phenylpyrimidin-2-amine fragment will be inves-
tigated in due course of this project). Next, a 1 mm DMSO solu-
tion of 1 was subjected to irradiation (see the Supporting In-
formation for UV spectrum and details). We found that photo-
induced cleavage of compound 1 at l=365 nm (5.4 W)
resulted in rapid and nearly quantitative cleavage under these
conditions, yielding the parent benzanilide fragment (and the
leaving group 4,5-dimethoxy-2-nitrosobenzaldehyde based on
MS analysis, not shown) within 5 min (Figure 3). We therefore
subsequently focused on the amide NH function in imatinib to
attach the PPG.
[
13]
light. In general, the implementation of a photoremovable
[14]
protecting group (PPG) provides spatial and temporal con-
[15]
trol over the release of a bioactive molecule.
As conse-
quence, significant concentrations of the bioactive compound
[
a] Dr. M. Zindler, B. Pinchuk, C. Renn, R. Horbert, A. Dçbber, Prof. Dr. C. Peifer
Christian Albrechts University of Kiel
Institute of Pharmacy, Gutenbergstr. 76, 24118 Kiel (Germany)
E-mail: cpeifer@pharmazie.uni-kiel.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201500163.
ChemMedChem 2015, 10, 1335 – 1338
1335
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
directly with imatinib and 1-(bro-
momethyl)-4,5-dimethoxy-2-ni-
trobenzene were unsuccessful,
as only the imatinib–piperazine
N-DMNB alkylated product could
be isolated (which was stable
under irradiation; see Support-
ing Information). However, an SN
reaction using the core scaffold
of imatinib 4-methyl-N3-[4-(3-
pyridyl)pyrimidin-2-yl]benzene-
1
,3-diamine 4 with the benzyl-
brominated PPG reagents
DMNB-Br or coumarylbromo-
methyl yielded compounds 5a
and 5b, respectively (Scheme 1).
In turn, introduction of the 4-[(4-
methylpiperazin-1-yl)methyl]ben-
zamide side chain in the final
step yielded both PPG-protected
compounds 2 and 3, respectively
Figure 1. Modeled ligand interaction diagram of imatinib in the active site of a PDGF-Rb homology model in the
DFG-out conformation. Key ligand–protein interactions are shown. The binding mode of imatinib in the closely re-
[
30]
lated kinase c-Src (PDB ID: 2OIQ) is highly similar (not shown).
(
Scheme 1).
With these caged prodrugs of
imatinib in hand, we next evalu-
ated their photochemical properties. Based on the UV/Vis spec-
tra of the compounds, we irradiated 1 mm solutions of 2 and 3
in DMSO for 12 min with light at a wavelength of 365 nm.
Under these conditions, photocleavage of 3 was very poor,
yielding only ~10% of imatinib (Supporting Information). In
contrast, 2 was efficiently photocleaved within 10 min to pro-
duce imatinib (Figure 5) along with the leaving group 4,5-di-
methoxy-2-nitrosobenzaldehyde as determined by MS analysis
(not shown).
Motivated by these results, we
designed two differently PPG-
caged prodrugs at the amide func-
tion of imatinib that bears the 4,5-
dimethoxy-2-nitrobenzyl (DMNB, 2)
and coumarylmethyl moieties (3),
respectively (Figure 4). Both PPGs
have been reported to be suitable
for caged prodrug applications in
Figure 2. Photoactivatable im-
atinib fragment 1. The com-
pound contains the benzani-
lide scaffold of imatinib which
is caged at the amide with
the 4,5-dimethoxy-2-nitroben-
zyl (DMNB) moiety.
[22]
biological settings.
We next characterized the biological activity of imatinib and
caged prodrug 2 in an enzymatic PDGF-Rb assay without and
with UV irradiation at l=365 nm. In this assay imatinib was
found to potently inhibit PDGF-Rb (IC =0.059 mm); this is in
We aimed to develop a straight-
forward synthetic strategy for the
preparation of the planned caged
prodrugs. Initially, our attempts
50
[
28]
good agreement with the reported value (IC =0.1 mm). In
50
toward base-catalyzed S reactions
line with our modeling data, caged compound 2 was signifi-
N
Figure 3. Time-dependent photoinduced cleavage of caged imatinib frag-
ment 1 to yield benzanilide upon irradiation with UV light (l=365 nm,
Figure 4. Chemical structures of imatinib and designed photoactivatable
caged prodrugs 2 and 3.
5.4 W).
ChemMedChem 2015, 10, 1335 – 1338
www.chemmedchem.org
1336
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
Scheme 1. Synthesis of caged imatinib derivatives 2 (R=DMNB) and 3
R=coumarylmethyl). Reagents and conditions: a) DMNB-Br (for 5a), 4-(bro-
momethyl)-7-methoxychromen-2-one (coumarylbromomethyl) (for 5b),
Li CO , DMF; b) 4-[(4-methylpiperazin-1-yl)methyl]benzoyl chloride, pyridine,
8C!RT. Full synthetic details are provided in the Supporting Information.
(
Figure 6. Biological activity of imatinib and caged prodrug 2 in an in vitro
PDGF-Rb assay without and with irradiation at l=365 nm (5.4 W) for 10 min
2
3
(
ÆSD, n=3, see Supporting Information for details).
0
cess. Thus, we will investigate further PPG systems in order to
optimize the uncaging of imatinib. Our study further validates
the powerful caging technique in the kinase field and provides
a valuable pharmacological tool to study the biological effects
of imatinib in novel settings and in greater detail. Finally, our
approach holds great promise for future applications involving
other kinase inhibitors as well.
Experimental Section
All modeling was performed on a DELL 8 core system. For prepara-
tion, visualization, and building the 3D structures, Maestro (ver-
sion 10.0.013, release 2014-4, Schrçdinger LLC, New York, NY, USA)
was used.
Synthesis of N-(4,5-dimethoxy-2-nitrobenzyl)-N-(4-methyl-3-((4-
Figure 5. Time-dependent photoinduced cleavage of caged imatinib pro-
drug 2 upon irradiation with UV light (l=365 nm, 5.4 W). After 5 min irradia-
tion, 70% of 3 was cleaved to produce imatinib.
(pyridine-3-yl)pyrimidine-2-yl)amino)phenyl)-4-((4-methylpipera-
zin-1-yl)methyl)benzamide (2): Compound 5a (330 mg, 700 mmol)
was dissolved in dry pyridine (20 mL). The reaction mixture was
cooled to 08C, and 4-[(4-methylpiperazin-1-yl)methyl]benzoyl chlo-
ride (288 mg, 1.00 mmol) was added in portions under a nitrogen
atmosphere. The solution was stirred for 3 h at room temperature,
cantly less biologically active without UV irradiation (2 PDGF-
Rb IC =5.8 mm). In contrast, testing 2 in the assay irradiated
50
at l=365 nm (5.4 W) for 10 min revealed an inhibition of
PDGF-Rb (IC =0.089 mm) similar to that of native imatinib
and H O (100 mL) was added. The mixture was extracted with
2
EtOAc (3”100 mL), and the combined organic layers were dried
over anhydrous Na SO . The solvent was removed under vacuum,
50
(
Figure 6). Notably, the residual biological activity of caged pro-
2
4
and the residue was purified by flash chromatography (SiO2 re-
drug 2 results from minor impurities of uncaged imatinib in
the sample (~1% based on HPLC analysis). Thus, at total com-
pound concentrations >10 mm in the assay, the inhibitory ef-
fects on PDGF-Rb cannot be differentiated. However, their dif-
ference regarding blockage of PDGF-Rb is still significant in the
range of typical in vitro assay concentrations (0.01–1 mm).
In conclusion, we have demonstrated that a caged prodrug
of the kinase inhibitor imatinib could be successfully generated
and characterized. Compound 2 was photoactivated by UV
light at a wavelength of 365 nm within 10 min in an in vitro
PDGF-Rb assay to restore the biological activity of the parent
imatinib. In contrast to imatinib, 1 can be almost quantitatively
uncaged within 3 min. This indicates a significant influence of
the leaving group (active inhibitor) toward the uncaging pro-
versed phase, MeOH/H O) to give compound 2 as a pale-yellow
2
1
solid (40 mg, yield: 8%). H NMR (300 MHz, [D ]DMSO): d=2.09 (s,
6
3
2
H, pip-CH ), 2.14 (s, 3H, CH ), 2.21 (mc, 8H, pip-H-2,3,5,6), 3.33 (s,
3 3
H, pip-CH ), 3.74 (s, 3H, OCH ), 3.82 (s, 3H, OCH ), 5.38 (bs, 2H,
2
3
3
3
4
CH ), 6.79 (dd, 1H, J=8.1 Hz, J=2.1 Hz, ph-H-6), 7.03 (s, 1H, nitro-
benz-H-6), 7.05 (d, 1H, J=8.1 Hz, ph-H-5), 7.13 (d, 2H, J=8.2 Hz,
benz-H-3,5), 7.34 (d, 2H, J=8.2 Hz, benz-H-2,6), 7.44 (d, 1H, J=
2
3
3
3
3
3
3
5
.2 Hz, pyrim-H-5), 7.49 (ddd, 1H, J=8.0 Hz, J=4.0 Hz, pyr-H-5),
4
7
.61 (s, 1H, nitrobenz-H-3), 7.62 (d, 1H, J=2.0 Hz, ph-H-2), 8.36
3
4
3
(
td, 1H, J=8.0 Hz, J=2.0 Hz, pyr-H-4), 8.44 (d, 1H, J=5.1 Hz,
4
pyrim-H-6), 8.69 (d, 1H, J=4.0 Hz, pyr-H-6), 8.80 (s, 1H, pyrim-NH),
4
13
9
.23 ppm (d, 1H, J=1.7 Hz, pyr-H-2);
C NMR (75.5 MHz,
[
D ]DMSO): d=17.5 (q, CH ), 45.6 (q, pip-CH ), 50.6 (t, CH ), 52.3,
6
3
3
2
54.6 (d, pip-C-2,3,5,6), 55.8, 55.9 (q, OCH ), 61.4 (t, pip-CH ), 107.9
3
2
(d, pyrim-C-5), 108.2 (d, nitrobenz-C-3), 110.5 (d, nitrobenz-C-6),
ChemMedChem 2015, 10, 1335 – 1338
www.chemmedchem.org
1337
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
1
22.5 (d, ph-C-2), 122.9 (d, ph-C-6), 123.7 (d, pyr-C-5), 127.3 (s, ni-
[10] R. Steri, J. Achenbach, D. Steinhilber, M. Schubert-Zsilavecz, E. Proschak,
Biochem. Pharmacol. 2012, 83, 1674–1681.
trobenz-C-2), 128.1 (d, benz-C-3,5), 128.3 (d, benz-C-2,6), 129.5 (s,
ph-C-1), 130.5 (d, ph-C-5), 131.9 (s, pyr-C-3), 134.1 (d, pyr-C-4),
[
[
[
11] P. La RosØe, M. W. Deininger, Semin. Hematol. 2010, 47, 335–343.
12] A. C. Dar, M. S. Lopez, K. M. Shokat, Chem. Biol. 2008, 15, 1015–1022.
13] W. A. Velema, W. Szymanski, B. L. Feringa, J. Am. Chem. Soc. 2014, 136,
1
34.3 (s, benz-C-4), 138.3 (s, ph-C-4), 139.9 (s, ph-C-3), 140.3 (s,
benz-C-1), 140.7 (s, nitrobenz-C-1), 147.3 (s, nitrobenz-C-5), 148.0
d, pyr-C-2), 151.4 (d, pyr-C-6), 152.9 (s, nitrobenz-C-4), 159.2 (d,
2178–2191.
(
[
[
14] G. C. Ellis-Davies, Nat. Methods 2007, 4, 619–628.
15] H. M. Lee, D. R. Larson, D. S. Lawrence, ACS Chem. Biol. 2009, 4, 409–
pyrim-H-6), 160.6 (s, pyrim-C-4), 161.5 (s, pyrim-C-2), 169.9 ppm (s,
C=O); IR (ATR): n˜ =2836 (CÀH), 1643 (C=O), 1574, 1330 (NO ),
427.
2
+
À1
7
95 cm (arom); MS (ESI, MeOH): m/z (%)=689.1 (100) [M+H] .
[16] G. Mayer, A. Heckel, Angew. Chem. Int. Ed. 2006, 45, 4900–4921; Angew.
Chem. 2006, 118, 5020–5042.
17] C. Brieke, F. Rohrbach, A. Gottschalk, G. Mayer, A. Heckel, Angew. Chem.
Further experimental details can be found in the Supporting Infor-
mation.
[
Int. Ed. 2012, 51, 8446–8476; Angew. Chem. 2012, 124, 8572–8604.
ˇ
[
18] P. Klµn, T. Solomek, C. G. Bochet, A. Blanc, R. Givens, M. Rubina, V. Popik,
A. Kostikov, J. Wirz, Chem. Rev. 2013, 113, 119–191.
19] A. Deiters, Curr. Opin. Chem. Biol. 2009, 13, 678–686.
20] R. S. Givens, M. Rubina, J. Wirz, Photochem. Photobiol. Sci. 2012, 11,
Acknowledgements
[
[
This research project was supported by the German Research So-
ciety (DFG) grant PE 1605/2-1. We thank Dr. Ulrich Girreser and
Martin Schütt (Institute of Pharmacy in Kiel, Germany) for excel-
lent technical and analytical assistance.
4
72–488.
[21] H. Li, J.-M. Hah, D. S. Lawrence, J. Am. Chem. Soc. 2008, 130, 10474–
0475.
[
[
1
22] G. Dormµn, G. D. Prestwich, Trends Biotechnol. 2000, 18, 64–77.
23] D. Geissler, W. Kresse, B. Wiesner, J. Bendig, H. Kettnemann, V. Hagen,
Chembiochem 2003, 4, 162–170.
[
24] A. R. Morckel, H. Lusic, L. Farzana, J. A. Yoder, A. Deiiters, N. M. Nascone-
Keywords: caged prodrugs · imatinib · inhibitors · kinases ·
PDGF-Rb · photoactivation
Yoder, Development 2012, 139, 437–442.
[
[
25] K. Curley, D. S. Lawrence, Pharmacol. Ther. 1999, 82, 347–354.
26] R. Ferreira, J. R. Nilsson, C. Solano, J. AndrØasson, M. Grøtli, Sci. Rep.
2015, 5, 9769.
[
[
[
[
[
1] K. K. Parmar, R. S. King, Cancer Pract. 2001, 9, 263–265.
2] S. J. Baker, E. P. Reddy, Mt. Sinai J. Med. 2010, 77, 573–586.
3] A. C. Dar, K. M. Shokat, Annu. Rev. Biochem. 2011, 80, 769–795.
4] H. M. Kantarjian, M. Talpaz, Semin. Oncol. 2001, 28, 9–18.
5] A. Virgili, M. Koptyra, Y. Dasgupta, E. Glodkowska-Mrowka, T. Stoklosa,
E. P. Nacheva, T. Skorski, Cancer Res. 2011, 71, 5381–5386.
[27] R. Horbert, B. Pinchuk, E. Johannes, J. Schlosser, D. Schmidt, D. Cappel,
F. Totzke, C. Schachtele, C. Peifer, J. Med. Chem. 2015, 58, 170–182.
[28] C. Sourbier, C. J. Ricketts, S. Matsumoto, D. R. Crooks, P. J. Liao, P. Z.
Mannes, Y. Yang, M. H. Wei, G. Srivastava, S. Ghosh, V. Chen, C. D. Vocke,
M. Merino, R. Srinivasan, M. C. Krishna, J. B. Mitchell, A. M. Pendergast,
T. A. Rouault, L. Neckers, W. M. Linehan, Cancer Cell 2014, 26, 840–850.
[29] B. Pinchuk, E. Johannes, S. Gul, J. Schlosser, C. Schaechtele, F. Totzke, C.
Peifer, Mar. Drugs 2013, 11, 3209–3223.
[
6] M. Baccarani, D. Cilloni, M. Rondoni, E. Ottaviani, F. Messa, S. Merante,
M. Tiribelli, F. Buccisano, N. Testoni, E. Gottardi, A. de Vivo, E. Giugliano,
I. Iacobucci, S. Paolini, S. Soverini, G. Rosti, F. Rancati, C. Astolfi, F. Pane,
G. Saglio, G. Martinelli, Hematol. J. 2007, 92, 1173–1179.
[30] M. A. Seeliger, B. Nagar, F. Frank, X. Cao, M. N. Henderson, J. Kuriyan,
Structure 2007, 15, 299–311.
[
7] a) J. Dorn, H. Spatz, M. Schmieder, T. F. Barth, A. Blatz, D. Henne-Bruns,
U. Knippschild, K. Kramer, BMC Cancer 2010, 10, 350; b) K. Søreide,
O. M. Sandvik, J. A. Søreide, E. Gudlaugsson, K. Mangseth, H. K. Haug-
land, Clin. Transl. Oncol. 2012, 14, 619–629.
[
[
8] K. Adekola, M. Agulnik, Curr. Oncol. Rep. 2012, 14, 327–332.
9] R. Capdeville, E. Buchdunger, J. Zimmermann, A. Matter, Nat. Rev. Drug
Discovery 2002, 1, 493–502.
Received: April 13, 2015
Published online on June 15, 2015
ChemMedChem 2015, 10, 1335 – 1338
www.chemmedchem.org
1338
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim