Labeled 3-aryl-4-indolylmaleimide derivatives and their potential as angiogenic PET biomarkers

Article abstract of DOI:10.1016/j.bmc.2009.12.004

In a continued effort to find a suitable PET tracer for visualization of angiogenic processes, we explored the 3,4-diarylmaleimide family, known to have high affinity and selectivity towards the VEGFR-TKs. One previously reported agent and three new halogen-containing 3,4-diarylmaleimide derivatives were synthesized. The four maleimide derivatives were evaluated for their affinity and selectivity towards the VEGFRs and exhibited promising results. An automated carbon-11 radiolabeling route with a total synthesis time of 50 min successfully labeled the lead compound, resulting in 1.55 ± 0.15 GBq of tracer with a radiochemical yield of 20 ± 2%, 96% radiochemical purity and a SA of 111 ± 22 GBq/μmol (EOB, n = 5). The tracer possessed high stability in in vitro blood stability tests and specific VEGFR-TK binding profiles in intact cell binding experiments. Tracer lipophilicity was evaluated in an n-octanol/phosphate buffer system giving a Log D_7.4 of 1.99 ± 0.04. For the in vivo experiments, two animal models were used. The first was a U87 glioma tumor model, frequently reported in the literature and the second, a newly developed 293/KDR tumor model. Both models were validated for VEGFR-2 expression and used in in vivo biodistribution studies. These studies revealed low accumulation and rapid washout of the tracer from tumor tissue. High accumulation of activity in the liver prompted us to examine the tracer's in vitro stability to liver microsomes, revealing low resistance to P450 metabolism. In spite of encouraging in vitro results, the labeled lead tracer failed to accumulate in VEGFR-2 overexpressing tumors. It is possible that poor resistance to P450 metabolism reduces tracer's circulation leading to low tumor accumulation.

Full text of DOI:10.1016/j.bmc.2009.12.004

Bioorganic & Medicinal Chemistry 18 (2010) 612–620
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Labeled 3-aryl-4-indolylmaleimide derivatives and their potential
as angiogenic PET biomarkers
*
Ohad Ilovich, Hana Billauer, Sharon Dotan, Eyal Mishani
Cyclotron/Radiochemistry Unit, Hadassah Hebrew University Hospital, Jerusalem 91120, Israel
a r t i c l e i n f o
a b s t r a c t
Article history:
In a continued effort to find a suitable PET tracer for visualization of angiogenic processes, we explored
the 3,4-diarylmaleimide family, known to have high affinity and selectivity towards the VEGFR-TKs. One
previously reported agent and three new halogen-containing 3,4-diarylmaleimide derivatives were syn-
thesized. The four maleimide derivatives were evaluated for their affinity and selectivity towards the
VEGFRs and exhibited promising results. An automated carbon-11 radiolabeling route with a total syn-
thesis time of 50 min successfully labeled the lead compound, resulting in 1.55 0.15 GBq of tracer with
Received 21 October 2009
Revised 1 December 2009
Accepted 2 December 2009
Available online 6 December 2009
Keywords:
VEGFR-2
Carbon-11
PET
a radiochemical yield of 20 2%, 96% radiochemical purity and a SA of 111 22 GBq/lmol (EOB, n = 5).
The tracer possessed high stability in in vitro blood stability tests and specific VEGFR-TK binding profiles
in intact cell binding experiments. Tracer lipophilicity was evaluated in an n-octanol/phosphate buffer
system giving a Log D7.4 of 1.99 0.04. For the in vivo experiments, two animal models were used. The
first was a U87 glioma tumor model, frequently reported in the literature and the second, a newly devel-
oped 293/KDR tumor model. Both models were validated for VEGFR-2 expression and used in in vivo bio-
distribution studies. These studies revealed low accumulation and rapid washout of the tracer from
tumor tissue. High accumulation of activity in the liver prompted us to examine the tracer’s in vitro sta-
bility to liver microsomes, revealing low resistance to P450 metabolism. In spite of encouraging in vitro
results, the labeled lead tracer failed to accumulate in VEGFR-2 overexpressing tumors. It is possible that
poor resistance to P450 metabolism reduces tracer’s circulation leading to low tumor accumulation.
Ó 2009 Elsevier Ltd. All rights reserved.
Angiogenesis
1. Introduction
during therapy could prove invaluable for evaluation of disease
progression and therapy efficacy. One approach for obtaining this
Angiogenesis, the formation of new blood vessels from pre-
existing vasculature, is a crucial mechanism for solid tumor growth
and a key player in diseases such as arthritis, psoriasis and macular
degeneration.1 Angiogenesis is regulated by a balance of pro- and
anti-angiogenic growth factors. Cognizant of the key roles angio-
genesis plays in a multitude of pathological conditions, intensive
research endeavors ensued to develop effective anti-angiogenic
therapies.2,3 These therapies include Avastin—a Vascular Endothe-
lial Growth Factor (VEGF) scavenging antibody4 and small mole-
cule Vascular Endothelial Growth Factor Receptor (VEGFR)
tyrosine kinase inhibitors (TKIs) such as sorafenib and sunitinib.5
Until recently, the only available end-points for traditional cyto-
toxic chemotherapy were factors such as tumor volume, time until
relapse and patient survival. With increased understanding of can-
cer-related pathways, novel pharmacodynamic end-points for
evaluating treatment efficacy at early stages are becoming
available.6 In the field of targeted therapies, knowledge regarding
the expression and occupancy of different receptors before and
information is via molecular imaging such as Positron Emission
Tomography (PET).7
Several potential targets for imaging angiogenesis, such as
VEGF, its receptors and integrin
avb3, have been investigated in
the recent past. The RGD peptide motif, targeting integrin
avb3,
is the most studied angiogenic tracer. The labeling of RGD (both
monomers and multimers) has been performed with several iso-
topes and various chemical structural changes were incorporated
aimed at improving tracer pharmacokinetics.8–10 Human studies
have been initiated with some of these labeled peptides.11
The second potential family of targets for imaging angiogenesis
is VEGF and its receptors. VEGFRs are members of the Receptor
Tyrosine Kinase (RTK) superfamily and are composed of three ma-
jor parts: an extracellular domain, a transmembrane region and an
intracellular domain. The extracellular domain contains the ligand
binding site. This domain has been a target for cancer therapy
either directly using peptide antagonists12 and anti-VEGFR anti-
bodies13 or indirectly using anti-VEGF antibodies.14 The intracellu-
lar domain contains the tyrosine kinase domain which, when
activated, initiates signal transduction pathways. This domain
has been targeted using small molecule tyrosine kinase inhibitors
* Corresponding author. Tel.: +972 2 677 7931; fax: +972 2 6421203.
E-mail address: eyalmi@ekmd.huji.ac.il (E. Mishani).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.12.004

O. Ilovich et al. / Bioorg. Med. Chem. 18 (2010) 612–620
613
O
(TKIs). The development of imaging biomarkers targeting VEGF and
O
its receptors has focused on the following approaches: labeling of
anti-VEGF antibodies, labeling of VEGF isoforms and labeling of
small molecule TKIs of VEGFR. A few anti-VEGF antibodies were la-
beled with 124I, 111In, and 89Zr. These antibodies exhibited high var-
iability in their distribution and clearance profiles and their tumor
uptakes were low.10 VEGF isoforms were labeled with 123/125I,
99mTc, 111In, and 64Cu. The iodinated isoforms had low tumor up-
take and high thyroid uptake due to tracer degradation and release
of free iodide.15 The other labeled compounds gave mixed results,
with one of the 64Cu labeled isoforms exhibiting improved tumor
uptake (ꢀ15%ID/g) and correlation between uptake and VEGFR-2
expression.16
O
X
X
1) Oxalyl chloride/THF
2) NaOMe/MeOH
N
H
N
H
1 - X=H (85%)
2 - X=F (82%)
3 - X=I (56%)
4 - X=Br (68%)
O
O
HO
H2N
Several small molecule TKIs were labeled and evaluated as po-
tential tracers for noninvasive monitoring of angiogenic processes.
A family of quinoline-ureas were labeled using fluorine-1817 and
carbon-11.18 Their high Log P values could cause problems such as
low specific tumor binding in vivo and high uptake in non-target tis-
sues. Recently, an anilino-quinoline was labeled with carbon-11.19
Although microPET studies yielded promising results, a stronger
proof-of-concept such as specific and saturable binding in VEGFR-
2 overexpressing tumors and data regarding the biodistribution of
this tracer are lacking. In a continued effort to find a suitable tracer
for evaluation of VEGFR-2 expression, we explored the 3,4-diary-
lmaleimide family20,21 known to have high affinity and selectivity
towards the tyrosine kinase domain of the VEGFR family.
1) DCC/DCM/0oC
2) NH3/Dioxane
80%
O
O
O
O
5
O
O
H
N
O
O
KOtBu/THF
1-4
+
5
X
O
N
H
2. Results and discussion
2.1. Chemistry
O
O
8 - X=H (79%)
9 - X=F (83%)
10 - X=I (66%)
11 - X=Br (55%)
Three derivatives of the previously reported20 VEGFR inhibitor
8, with different halogen substituents were synthesized in order
to allow for the future labeling with different PET radioisotopes.
Synthesis was performed applying general methods described in
the past with minor changes.21,22 Briefly, the indoles were reacted
with oxalyl chloride in an ethereal solution followed by addition of
NaOMe in methanol to give four different indole-oxo-acetic esters
1–4 with yields of 56–85%.22 2-(3,4,5-trimethoxyphenyl)acetic
acid was reacted with DCC and ammonia in a two-step, one-pot,
overnight reaction to produce acetamide 5 with 80% yield. The
acetamide was condensed with each of the four indole-oxo-acetic
esters in THF using KOtBu to furnish the four maleimide deriva-
tives 8–11 with yields of 55–83% (Scheme 1). We found that the
use of a 1:2 molar ratio of acetamide: indole-oxo-acetic esters gave
good yields and little impurities for all four compounds. Lower ra-
tios of indoles resulted in the formation of a multitude of by-prod-
ucts which were extremely hard to purify.
Our first choice for radiolabeling was the non-halogenated
maleimide (8) published in the past. This compound can only be la-
beled using [11C]CH3I chemistry in a general route which could also
be used to label the three other compounds. The synthesis of the
desmethyl precursor molecule was performed using benzyl-pro-
tected phenyl acetic acid 6 which was amidated to give 7 under
the same conditions used for compound 5 (yield 77%) and reacted
with indole 1 to give the protected maleimide 12 with 73% yield.
Compound 12 was deprotected using Pd/C under a hydrogen atmo-
sphere to furnish the desmethyl maleimide precursor 13 (yield
67%) (Scheme 2).
Scheme 1. Synthesis of maleimide derivatives.
three VEGFRs tested. IC50 values of 20–50 nM towards the three
VEGFRs were obtained for compound 8 and good selectivity rela-
tive to the rest of the tested receptors. The presence of a fluorine
atom on the indole ring was expected to decrease electron density,
thus addressing stability problems which caused intramolecular
cyclization during an agarose pellet production, and loss of recep-
tor binding and biological activity.21 However, during the synthesis
of compound 9, some cyclization still occurred, albeit to a lesser
extent relative to the hydrogen analog. Compound 9 showed the
least affinity towards the VEGFRs suggesting that high electron
density in the indole ring is an important factor for receptor affin-
ity. This may be manifested in hydrogen bonding between the sec-
ondary amine and the receptor’s binding site. In addition, 9 had
diminished selectivity towards the VEGFR-2 versus all other recep-
tors with a loss of selectivity (defined as a ratio of 1:20 or less)
relative to Tie-2 and PDGFRa. The conservation of affinity and
selectivity with the iodo (10) and bromo (11) substituted
derivatives suggests tolerance towards lipophilic substitutions at
the 5-indole position in most receptors tested and supports the
hypothesis of the importance of electron density of the maleimide
moiety to achieve good interaction at the VEGFR-2 binding site.
2.3. Radiochemistry
The radiolabeling approach was based on the general C-11
methylation reaction of the free hydroxy group23 and was adapted
for use in a commercial module (GE, Münster, Germany). Carbon-
11 methyl iodide was prepared according to well documented pro-
2.2. Evaluation of IC50 values for inhibition of 10 tyrosine
kinases
The four compounds were evaluated for their affinity and selec-
tivity towards a family of relevant tyrosine kinases (Table 1). All
four compounds showed high affinity and selectivity towards the
cedures.24 11C]CO2 was trapped and reacted with LiAlH4 in the
[
first reactor. THF was evaporated, aqueous HI added and
[
11C]CH3I distilled out to the second reactor which contained

614
O. Ilovich et al. / Bioorg. Med. Chem. 18 (2010) 612–620
O
O
O
HO
O
HO
H2N
BnCl/NaOH
63%
1) DCC/DCM/0 ºC
2) NH3/Dioxane
77%
O
O
O
O
O
OBn
OH
OBn
O
6
7
H
H
N
N
O
O
O
KOtBu/THF
73%
Pd/C
67%
1 + 7
O
O
N
H
N
H
OBn
OH
O
O
12
13
H
N
H
N
O
O
O
O
DMF/NaOH
11
O
O
[
C]MeI
N
H
N
H
11
O
CH3
OH
O
O
11
[
C]8
13
Scheme 2. Precursor synthesis and radiochemical transformation.
Table 1
Evaluation of median inhibitory concentration (IC50) values for inhibition of receptor phosphorylation (n = 3)
Kinase
Maleimidea
8
Selectivity
F-Maleimidea
9
Selectivity
I-Maleimidea 10
Selectivity
Br-Maleimidea 11
Selectivity
EGF-R
ERBB2
IGF1-R
KIT
PDGFR-
PDGFR-b
TIE2
VEGF-R1
VEGF-R2
VEGF-R3
10,133
13,100
6967
1900
943
1867
2633
55
394
510
270
74
36
72
102
2.14
1.00
0.87
>8480
>8480
8533
3467
1800
2667
1567
150
>103
>103
104
42
21
32
1688
15,180
3198
3330
1307
858
3961
69.4
24.89
29.51
148
609
128
133
52
5797
13,316
4241
3219
1156
900
2367
72.4
22.5
21
257
590
188
142
51
40
105
3.2
a
34
19
159
2.78
1.00
1.18
1.83
1.00
0.80
26
22
82
66
1.00
0.93
a
All IC50 values are given in nM.
5–7 mg of the precursor 13 in a dimethylformamide/NaOH solu-
tion at ꢁ15 °C (Scheme 2). The methylation reaction took place
at 80 °C for 5 min, after which 1:1 CH3CN/H2O was added and
the crude mixture was purified by a built-in RP-HPLC to yield the
final product. The desired fraction was collected into a flask con-
taining water and loaded onto a C-18 cartridge which was washed
with water and eluted with ethanol and saline to give [11C]8 in a
10% ethanol/saline solution. EOB decay corrected radiochemical
yield was 20 2% (n = 5) with a total synthesis time of 50 min.
HPLC analysis of the product solution showed high radiochemical
(ꢀ96%) and chemical purities with a specific activity of 3 0.6 Ci/
1–3 are considered optimal for this purpose. Low lipophilicity
may result in decreased membranal penetration while very high
lipophilicity may result in trapping of the drug inside the mem-
brane and increased non-specific binding. In molecular imaging,
this type of non-specific binding results in accumulation of radio-
activity in the blood pool (due to binding to blood proteins) and
low signal/noise ratios in target tissues. All four compounds have
estimated Log P values of 1.5–2.8, falling well within the optimal
range.
The estimated Log P of compound 8 was calculated to be
ꢀ1.5; this projects a ratio of radioactivity of 30:1 between aque-
ous and octanol phases. In order to compensate for this high ra-
tio and decrease the error in measurement of very small
quantities of radioactivity in the octanol phase, we worked with
a 5:1 ratio of octanol: phosphate buffer (pH 7.4, 100 mM). The
measured Log D7.4 of compound [11C]8 was found to be 1.99
0.04 (n = 4).
lmol decay corrected to EOB.
2.4. Log D
Targeting of the intracellular TK domain requires passage of
the biomarkers through cellular membranes. Log D values of

O. Ilovich et al. / Bioorg. Med. Chem. 18 (2010) 612–620
615
2.5. In vitro blood stability
11C]8 was incubated with blood samples at 37 °C for different
with 1.5–2 ꢂ 106 for 7–21 days for the 293/KDR cells or with
1.5–4 ꢂ 106 for 4–20 days for the U87MG.wt EGFR cells. Cells
were implanted with or without Matrigel. Histopathological con-
firmation of tumor model applicability was performed via evalu-
ation of mitosis as an indication for proliferation, necrosis,
inflammation as indicated by the presence of neutrophils, lym-
phocytes, etc., and vascularity, associated with VEGFR-2 presence.
VEGFR-2 expression in tumors and surrounding vasculature was
estimated using immunohistochemistry (Fig. 2). The most favor-
able results obtained with 21-day U87MG.wt EGFR (1.5 ꢂ 106
cells) and 7-day 293/KDR (2 ꢂ 106 cells) tumors are described in
Table 2.
In addition to the immunohistochemical results, Western blot
analysis was used to evaluate VEGFR-2 expression in the differ-
ent tumors. Both U87MG.wt EGFR and 293/KDR tumors had a
statistically different expression of VEGFR-2 in comparison to
U138 and HEK tumors which are known to express lower levels
of VEGFR-2. No statistical difference was found when comparing
VEGFR-2 expression in the 293/KDR and U87MG.wt EGFR
(Fig. 3).
[
time periods (0, 30 and 60 min). Blood samples were centrifuged
to separate pellet from plasma and the plasma was extracted with
THF/ACN (30:70). The extracted radioactivity was analyzed using
both RP-HPLC (data not shown) and RP-TLC plates where radioac-
tive bands were visualized using a phosphor image plate. The aver-
age extracted radioactivity was 68%, 61% and 62% for 0, 30, and
60 min incubation, respectively. Formation of radioactive metabo-
lites was extremely low (n = 3).
2.6. Specific binding studies of [11C]8 in intact 293/KDR cells
As a candidate bioprobe for tracking changes in expression of
VEGFR-2 in tumors, [11C]8 must offer adequate levels of specific
binding (SB) to the receptor. [11C]8 was incubated with or without
an excess of compound 8 in order to measure non-specific binding
(NSB) and total binding (TB), respectively. Cell-bound radioactivity
was separated from free radioactivity, measured in a c-counter and
SB was calculated by deducting the NSB from the TB. As illustrated
in Figure 1, SB of 72% and 68% of TB was attained at 10 and 20 nM
of [11C]8, respectively.
2.8. [11C]8 Biodistribution studies in tumor-bearing mice
The in vivo distribution of [11C]8 in tumor-bearing mice was as-
sessed at 30 and 60 min post injection. Comparison of the percent-
ages of injected dose per gram of organ obtained for the 293/KDR
and U87MG.wt EGFR models are presented in Table 3. Activity up-
take levels of all organs except for liver declined by approximately
50% between 30 and 60 min post injection. Both tumors under-
went rapid washout of activity between the time points examined
with neither surpassing activity in the blood, indicating a lack of
tracer accumulation in tumor tissue. The highest levels of activity
uptake were observed in the liver and kidneys however while
the kidneys underwent rapid washout of activity, liver activity
was maintained over 60 min, reaching up to a 21-fold ratio with re-
spect to tumors.
2.7. Development of tumor-bearing animal xenograft models
Tumor-bearing animal models were analyzed using several
parameters. Nude mice were inoculated subcutaneously either
2.9. In vitro hepatic microsome stability
[
11C]8 was incubated with hepatic microsomes at 37 °C for dif-
ferent time periods (0, 15, 30 and 60 min). Samples were loaded
onto a celite plug and activity eluted with ethanol. The ethanol
solution was analyzed using RP-HPLC (data not shown). The per-
cent of intact [11C]8 was found to be 96%, 40%, 25%, and 16% at 0,
15, 30 and 45 min, respectively. Two major metabolites were
formed, both highly polar in nature.
Figure 1. Specific binding of [11C]8 to 293/KDR cells. Cells were incubated either
with or without excess of cold compound 8. 10 or 20 nM [11C]8 was added for
incubation, cell bound activity separated and binding quantified using a c-counter.
Specific binding was determined by subtracting the bound activity with an excess
of 8 measurements (blue) from the one without an excess of 8 (red).
Figure 2. Immunohistochemical staining of 293/KDR (A) and U87MG.wt EGFR (B) for the expression of VEGFR-2, confirming VEGFR-2 expression in both tumor models.

616
O. Ilovich et al. / Bioorg. Med. Chem. 18 (2010) 612–620
Table 2
Validation of human tumor xenografts in nude mice, by histopathology and immunohistochemistrya
Tumor model
293/KDR
Volume (cm3)
Necrosis(%)b
Mitosis per HPF
Inflammationc
Vessels per HPF
VEGFR capillaries
VEGFR vessels
VEGFR tumor cells
0.22 0.13
16
9
1.8 0.4
1.5 0.3
4.7 0.6
1.5 0.4
2.3 0.5
2.6 0.2
(n = 11)
(n = 11)
(n = 11)
(n = 4)
(n = 7)
(n = 11)
(n = 11)
(n = 11)
U87MG.wt EGFR
0.09 0.02
33 15
2.8 0.4
1.6 0.3
7.0 1.4
1.9 0.8
5.3 2.4
2.6 0.6
(n = 7)
(n = 7)
(n = 6)
(n = 7)
(n = 3)
(n = 7)
(n = 5)
(n = 7)
a
b
C
Results are presented as mean SEM.
The percentage of necrosis is approximate.
The inflammation is scored using semi-quantitative grading (0–4).
furan (THF) was distilled over sodium/benzophenone; other
solvents were purchased as anhydrous. 1H NMR spectra were re-
corded on a 300 MHz spectrometer in DMSO-d6 or CDCl3. 1H sig-
nals are reported in ppm. 1H NMR signals are referenced to the
residual proton (7.26 and 2.50 ppm for CDCl3 and DMSO-d6,
respectively) of a deuterated solvent. Mass spectra were obtained
on a spectrometer equipped with CI, EI, and FAB probes and on a
spectrometer equipped with an ESI probe. HRMS results were ob-
tained on MALDI-TOF and ESI mass spectrometers. Flash chroma-
tography was carried out on SiO2 (0.04–0.063 mm). HPLC was
performed on a system with a variable wavelength detector oper-
ating at 254 nm and a radioactivity detector with a NaI crystal. Two
HPLC systems were used: (A) a reversed-phase system employing a
Figure 3. The expression of VEGFR-2 in 293/KDR and U87MG.wt EGFR tumors was
statistically different from HEK (p = 0.03) and U138 (p = 0.009) tumors, respectively.
No statistically difference was found comparing VEGFR-2 expression in the 293/
KDR and U87MG.wt EGFR.
Macherey–Nagel column (C18, Nucleosil, 7
l
m, 250ꢃ 16 mm) and
42% CH3CN/58% acetate buffer 0.1 M/pH 3.8 as eluent, at a flow rate
of 12 mL/min installed in the carbon-11 module. (B) Analysis of
formulated radiotracer was performed on a reversed-phase system
3. Conclusion
using a Waters column (C18,
3.9 mm) and 42% CH3CN/58% acetate buffer 0.1 M/pH 3.8 as sol-
l
Bondapak, 10
l
m, 125A, 300ꢃ
The obtained data in this study indicate that three out of the
four diarylmaleimide derivatives synthesized exhibited high affin-
ity and selectivity towards the VEGF family of receptors. The fluoro
vents at a flow rate of 1 mL/min, rt = 12 min.
Radiosyntheses were carried out on a [11C]CH3I module (GE,
Münster, Germany). Specific radioactivities were determined by
HPLC, using cold mass calibration curves. [11C]CO2 was produced
derivative showed decreased selectivity as compared to PDGFR
a
by the 14N(p, 11C nuclear reaction on nitrogen containing 0.5%
a)
and TIE2. All four derivatives can easily be labeled with carbon-
11 and have appropriate Log P values for PET bioprobe develop-
ment. In addition, the lead compound [11C]8 had specific binding
in cells overexpressing VEGFR-2 and was stable in human blood.
Unfortunately, biodistribution studies performed in tumor-bearing
mice with the labeled lead structure showed neither high accumu-
lation nor retention of the bioprobe in target tissues. The low sta-
bility of these compounds to liver enzyme metabolism may
hamper future development as PET tracers; thus, the maleimide
moiety, known to have high affinity to sulfhydryl functions, may
be at fault, requiring a different connecting group (e.g., triazole)
as replacement.
oxygen, using an 18/9 IBA cyclotron. Bombardment was carried
out for 40 min with a beam of 16-MeV protons. At the end of bom-
bardment (EOB), the target gas was delivered and trapped by a
cryogenic trap in the [11C]CH3I module.
All IC50 studies were performed by ProQinase GmbH (Freiburg,
Germany).
4.2. Chemistry and radiochemistry
4.2.1. Methyl 2-(1H-indol-3-yl)-2-oxoacetate (1)22
Oxalyl chloride (0.86 mL, 10 mmol) was added dropwise to a
solution of 1H-indole (1.17 g, 10 mmol) in diethyl ether (12 mL)
cooled in an ice bath. The reaction mixture kept stirring under the
same conditions for 30 min, cooled to ꢁ65 °C and a 25 wt % solution
of CH3ONa in MeOH (4.7 mL, 23 mmol) was added dropwise. The
reaction mixture was warmed to room temperature, stirred for an
additional 30 min and quenched by adding water (18 mL). The prod-
uct was isolated by filtration, washed with dichloromethane (DCM)
and dried to give 1.73 g of yellow solid (yield 85%). 1H NMR (DMSO-
4. Materials and methods
4.1. General methods
All operations with air- and moisture-sensitive compounds
were performed by the Schlenk techniques under argon atmo-
sphere. All solvents were of analytical grade or better. Tetrahydro-
Table 3
Biodistribution of [11C]8 in 293/KDR and U87MG.wt EGFR tumor-bearing mice
%ID/g (mean SEM)
30 min (n = 10)
60 min (n = 10)
30 min (n = 12)
60 min (n = 10)
Blood
U87MG.wtEGFR
Skin
Bone
Muscle
Liver
0.38 0.04
0.33 0.08
0.41 0.06
0.19 0.03
0.17 0.03
2.64 0.29
2.24 0.23
0.29 0.04
0.17 0.03
0.21 0.04
0.11 0.02
0.09 0.02
1.68 0.51
0.79 0.17
Blood
293/KDR
Skin
Bone
Muscle
Liver
0.36 0.03
0.30 0.03
0.38 0.03
0.15 0.02
0.20 0.02
2.89 0.22
2.29 0.16
0.19 0.04
0.14 0.03
0.14 0.04
0.09 0.02
0.08 0.02
2.95 1.18
0.80 0.27
Kidney
Kidney

O. Ilovich et al. / Bioorg. Med. Chem. 18 (2010) 612–620
617
d6): d 12.40 (br s, 1H), 8.42 (d, J = 3.3, 1H), 8.15 (s, 1H), 7.51 (s, 1H)
7.26 (m, 2H), 3.87 (s, 3H); MS (ESI, m/z): 204 (M++1).
ular sieves (4 Å) was cooled to 0 °C under nitrogen atmosphere. A
solution of potassium tert-butoxide (KOtBu) (1 M, 2.1 mmol) was
added dropwise during which time the reaction color changed to
purple. The reaction mixture was stirred overnight at room tem-
perature and quenched by addition of a saturated solution of
NH4Cl. The THF phase was separated, the aqueous phase extracted
with EtOAc and the organic phases united, dried over Na2SO4 and
the product purified via flash column chromatography using 2%
MeOH in DCM to give 131 mg of orange solid (yield 79%). 1H
NMR (DMSO-d6): d 12.00 (s, 1H), 11.14 (s, 1H), 8.08 (s, 1H), 7.56
(d, J = 7.8, 1H), 7.16 (t, J = 7.5, 1H), 6.84 (m, 3H), 6.45 (d, J = 7.8,
1H), 3.76 (s, 3H), 3.48 (s, 6H); HRMS 379.1322 (M+1, calcd
379.1294).
4.2.2. Methyl 2-(5-fluoro-1H-indol-3-yl)-2-oxoacetate (2)
Material synthesized as 1 with 5-fluoro-1H-indole (810 mg,
6 mmol), diethylether (8 mL), oxalyl chloride (0.52 mL, 6 mmol)
and CH3ONa solution (2.8 mL, 13.7 mmol) to give 1.08 g of yellow
solid (yield 82%). 1H NMR (DMSO-d6): d 8.54 (d, J = 3.6, 1H), 7.4 (dd,
J = 2.4, J = 9.6, 1H), 7.53–7.59 (q, J = 4.5, 1H), 7.12–7.19 (m, 1H),
3.89 (s, 3H); MS (ESI, m/z): 222 (M++1).
4.2.3. Methyl 2-(5-iodo-1H-indol-3-yl)-2-oxoacetate (3)
Material synthesized as 1 with 5-iodo-1H-indole (729 mg,
3 mmol), diethylether (5 mL), oxalyl chloride (0.27 mL, 3 mmol)
and CH3ONa solution (1.4 mL, 6.8 mmol) to give 553 mg of yellow
solid (yield 56%). 1H NMR (DMSO-d6): d 8.5 (s, 1H) 8.46 (d, J = 3.3,
1H), 7.59 (d, J = 8.4, 1H), 7.4 (d, J = 8.7, 1H), 3.89 (s, 3H); MS (ESI, m/
z): 329 (M++1).
4.2.9. 3-(5-Fluoro-1H-indol-3-yl)-4-(3,4,5-trimethoxyphenyl)-
1H-pyrrole-2,5-dione (9)
The title compound was synthesized as described for 8 using
amide 5 (100 mg, 0.44 mmol) and indole 2 (195 mg, 0.88 mmol).
Flash column chromatography using a gradient of 1–2% MeOH in
DCM gave 144 mg of orange solid (yield 83%). 1H NMR (DMSO-
d6): d 12.11 (s, 1H), 11.18 (s, 1H), 8.16 (s, 1H), 7.56 (m, 1H), 7.04
(m, 1H), 6.81 (m, 2H), 6.02 (dd, J = 10.8, J = 2.7, 1H), 3.77 (s, 3H),
3.53 (s, 6H); HRMS 395.1035 (Mꢁ1, calcd 395.1043), mp 258–
259 °C.
4.2.4. Methyl 2-(5-bromo-1H-indol-3-yl)-2-oxoacetate (4)
Material synthesized as 1 with 5-bromo-1H-indole (588 mg,
3 mmol), diethylether (5 mL), oxalyl chloride (0.27 mL, 3 mmol)
and CH3ONa solution (1.4 mL, 6.8 mmol) to give 572 mg of yellow
solid (yield 68%). 1H NMR (DMSO-d6): d 8.51 (d, J = 3.3, 1H), 8.28 (s,
1H), 7.53 (d, J = 8.7, 1H), 7.44 (d, J = 8.7, 1H), 3.89 (s, 3H); MS (ESI,
m/z): 282 (M++1).
4.2.10. 3-(5-Iodo-1H-indol-3-yl)-4-(3,4,5-trimethoxyphenyl)-
1H-pyrrole-2,5-dione (10)
4.2.5. 2-(3,4,5-Trimethoxy-phenyl) acetamide (5)
The title compound was synthesized as described for 8 using
amide 5 (100 mg, 0.44 mmol) and indole 3 (290 mg, 0.88 mmol).
Flash column chromatography using a gradient of 1–2% MeOH in
DCM gave 146 mg of orange solid (yield 66%).1H NMR (DMSO-
d6): d 12.05 (s, 1H), 11.07 (s, 1H), 8.00 (s, 1H), 7.31 (m, 2H), 6.68
(s, 2H), 6.64 (s, 1H), 3.77 (s, 3H), 3.53 (s, 6H); HRMS 503.0128
(Mꢁ1, calcd 503.0104), mp 285 °C (decomposition).
A
solution of 2-(3,4,5-trimethoxyphenyl)acetic acid (1 g,
4.4 mmol) in DCM (30 mL) was added dropwise to a solution of
N,N0-dicyclohexylcarbodiimide (DCC) (1.1 g, 5.3 mmol) in DCM
(15 mL) cooled to 0 °C. The reaction mixture was stirred at 0 °C
for 1 h and a 0.5 M solution of NH3 in dioxane (17.6 mL, 8.8 mmol)
was added and stirring was continued at room temperature over-
night. The reaction mixture was filtered, solvent was evaporated
and the product was purified via column chromatography with
ethyl acetate as eluent to give 790 mg of white solid (yield 80%).
1H NMR (CDCl3): d 6.48 (s, 2H), 5.2 (d, J = 9, 2H), 3.86 (s, 6H),
3.84 (s, 3H), 3.53 (s, 2H); MS (ESI, m/z): 226 (M++1).
4.2.11. 3-(5-Bromo-1H-indol-3-yl)-4-(3,4,5-trimethoxyphenyl)-
1H-pyrrole-2,5-dione (11)
The title compound was synthesized as described for 8 using
amide 5 (100 mg, 0.44 mmol) and indole 4 (248 mg, 0.88 mmol).
Flash column chromatography using a gradient of 1–2% MeOH in
DCM gave 110 mg of orange solid (yield 55%). 1H NMR (DMSO-
d6): d 12.09 (s, 1H), 11.10 (s, 1H), 8.07 (s, 1H), 7.42 (d, J = 8.4,
1H), 7.31 (d, J = 10.5, 1H), 6.70 (s, 2H), 6.37 (s, 1H), 3.75 (s, 3H),
3.46 (s, 6H); HRMS 455.0225 (Mꢁ1, calcd 455.0243), 459 (M++1),
mp 267 °C (decomposition).
4.2.6. 2-(4-Benzyloxy-3,5-dimethoxyphenyl) acetic acid (6)
2-(4-Hydroxy-3,5-dimethoxyphenyl) acetic acid (1 g, 4.7 mmol)
was dissolved in a solution of THF, 2 M NaOH (12.5 mL, 1:1.5) and
benzyl chloride (0.64 mL, 5.6 mmol) was added. The reaction mix-
ture was refluxed overnight, THF was evaporated and the reaction
acidified to pH 1 using concentrated HCl. The white precipitate
which formed was extracted with DCM, the organic phases united
and washed with 5% HCl and brine and dried over Na2SO4. The
product was purified via column chromatography with 2% MeOH
in DCM as eluent to give 0.9 g of white solid (yield 63%). 1H NMR
(CDCl3): d 7.47 (d, J = 9.3, 2H), 7.32 (m, 3H), 6.49 (s, 2H), 4.97 (s,
2H), 3.81 (s, 6H), 3.58 (s, 2H); MS (ESI, m/z): 303 (M++1).
4.2.12. 3-(4-Benzyloxy-3,5-dimethoxyphenyl)-4-(1H-indol-3-
yl)-1H-pyrrole-2,5-dione (12)
The title compound was synthesized as described for 8 using
amide 7 (435 mg, 1.5 mmol), indole 1 (611 mg, 3 mmol), THF
(42 mL), molecular sieves (100 mg) and KOtBu (7 mL, 7 mmol).
Flash column chromatography using 1% MeOH in DCM as eluent
produced 500 mg of orange solid (yield 73%). 1H NMR (DMSO-
d6): d 11.9 (s, 1H), 11.04 (s, 1H), 7.98 (s, 1H), 7.31–7.47 (m, 5H),
7.09 (t, J = 7.8, 1H), 6.74–6.79 (m, 3H), 6.37 (d, 1H), 4.92 (s, 2H),
3.38 (s, 6H); MS (ESI, m/z): 455 (M++1).
4.2.7. 2-(4-Benzyloxy-3,5-dimethoxyphenyl) acetamide (7)
The title compound was synthesized under the same conditions
as 5 using 6 (604 mg, 2 mmol), DCC (500 mg, 2.4 mmol), DCM
(30 mL) and a 0.5 M solution of NH3 in dioxane (8 mL, 4 mmol).
Column chromatography with a gradient of 0–4% MeOH in DCM
gave 465 mg of white solid (yield 77%). 1H NMR (CDCl3): d 7.47
(d, J = 9.3, 2H), 7.32 (m, 3H), 6.49 (s, 2H), 5.77 (d, J = 5.7, 2H),
4.94 (s, 2H), 3.77 (s, 6H), 3.46 (s, 2H); MS (ESI, m/z): 302 (M++1).
4.2.13. 3-(4-Hydroxy-3,5-dimethoxyphenyl)-4-(1H-indol-3-yl)-
1H-pyrrole-2,5-dione (13)
Compound 12 (84 mg, 0.185 mmol) was dissolved in
a
MeOH:THF solution (6:1, 14 mL) and Pd/C (10%, 8.5 mg) was
added. The hydrogenation was performed at 4 psi for 2 h. The reac-
tion mixture was filtered and evaporated to give a reddish solid
which was purified via flash column chromatography with 1%
MeOH in DCM as eluent to give 45 mg of red solid (yield 67%).
1H NMR (DMSO-d6): d 11.84 (s, 1H), 10.97 (s, 1H), 8.8 (s, 1H),
4.2.8. 3-(1H-Indol-3-yl)-4-(3,4,5-trimethoxyphenyl)-1H-
pyrrole-2,5-dione (8)21
A stirred solution of the amide 5 (100 mg, 0.44 mmol) and in-
dole 1 (181 mg 0.88 mmol) in dry THF containing 30 mg of molec-

618
O. Ilovich et al. / Bioorg. Med. Chem. 18 (2010) 612–620
7.91 (d, J = 4.5, 1H), 7.42 (d, J = 8.1, 1H), 7.06 (t, J = 7.5, 1H), 6.74–
6.78 (m, 3H), 6.45 (d, J = 7.8, 1H), 3.39 (s, 6H); MS (ESI, m/z): 365
(M++1).
mately 200 lCi (7.4 MBq) were taken from the mother solution,
transferred to a 10 mL V-vial and the volume was corrected to
5 mL with n-octanol. Phosphate buffer (1 mL, pH 7.4, 100 mM)
was added and the solution was vortexed for 2 min. The two
phases were separated and measured for activity. Log D7.4 was cal-
culated as 1.99 0.04 (n = 4).
4.2.14. 3-(1H-Indol-3-yl)-4-(3,5-dimethoxy-4-
[
11C]methoxyphenyl)-1H-pyrrole-2,5-dione ([11C]8)
The synthesis and use of [11C]MeI was published in the past.24
Briefly, [11C]CO2 (1117 72 mCi, 41 2.6 GBq) was trapped at
ꢁ160 °C. The temperature of the cooling trap was increased to
ꢁ20 °C, and the activity was transferred by a stream of argon
4.5. In vitro blood stability
Fresh human blood samples were collected in a Vacutainer™
containing EDTA (K3, BD, USA), divided into 1 mL aliquots and
incubated for 15 min at 37 °C with gentle shaking. Approximately
(20 mL/min) into a reactor containing 300 lL of 0.25 N LiAlH4 in
THF at ꢁ50 °C. After 90 s, the solvent was removed under reduced
pressure. In this manner, more than 70% of the activity was recov-
ered. The reactor temperature was increased to 160 °C, HI was
added and [11C]MeI was distilled (argon flow of 25 mL/min)
through a NaOH column to a second reactor, containing the 5 mg
300 l
Ci of [11C]8 solution was added to blood samples for 0, 30
or 60 min, after which the samples were removed from the incuba-
tor and centrifuged at 3500 rpm for 5 min. The plasma was sepa-
rated from the precipitant and radioactivity levels were
measured for both fractions in a dose calibrator (CRC-35, Capintec,
Canada). 1.2 mL of cold THF/ACN (30:70) solution was added to the
plasma fraction; the plasma was vortexed for 30 s and then centri-
fuged again for 5 min at 10,600 rcf (g) and pellet and plasma were
separated and measured for radioactivity levels. The protein-free
of 13 in a solution of dimethylformamide (600
lL) and NaOH
(5 M, 20
l
L) at ꢁ15 °C. After 1 min of distillation, an average of
502 54 mCi (18.5 2 GBq) (n = 5) was trapped in the second reac-
tor. The reactor was sealed and heated to 80 °C for 5 min. At the
end of the 5 min reaction, the mixture was cooled to 40 °C,
1.2 mL of CH3CN/H2O (1:1) was added and the crude product
was automatically injected to HPLC system A, rt = 14.5 min. The la-
beled product was collected in a flask containing 45 mL of water.
The solution was loaded onto a C-18 cartridge (Waters Sep-Pak
Plus, preactivated with 5 mL ethanol and 10 mL of water) and then
washed with water (4 mL). The product was eluted with 1 mL of
ethanol, followed by 9 mL of saline and collected into the product
vial after a total radiosynthesis time of 50 min, with total activity
of 42 4 mCi (1.5 0.15 GBq) and radiochemical yield of 20 2%
decay corrected to EOB (n = 5). Identification of the products (via
coinjection) and determination of chemical, radiochemical purities
plasma was then filtered through a 0.45 lm filter (Puradisc, What-
man) and samples were both injected into RP-HPLC (results not
shown) and loaded on a TLC plate (Partisil LKC18F silica gel
5 ꢂ 20 cm, 200
lm layer thickness, Whatman). TLC plates were
run with 40:60 H2O:ACN as a mobile phase and exposed for 1 h
to phosphor imager plates (BAS–IP MS 2040 Fuji Photo Film Co.
Ltd) for visualization of radioactive bands (Fig. 4). The plates were
scanned with a BAS reader 3.1 version scanner and analyzed with
TINA 2.10 software.
4.6. Cell culture
(ꢀ96%) and specific activities (3 0.6 Ci/
lmol decay corrected to
EOB) were obtained by reverse-phase HPLC on a C-18 analytical
Human glioma cell line U87MG.wt EGFR and U138 human gli-
oma cell line, were grown as described previously.25 293 Human
kidney cells expressing VEGFR-2 (293/KDR, SibTech, Brookfield,
CT) and 293 Human embryonic kidney (HEK) wt cells were grown
in Dulbecco’s Modified Eagle’s Medium (DMEM; Renium, Israel)
supplemented with 10% fetal calf serum (FCS) and anti-biotics
(105 U/L penicillin and 100 mg/L streptomycin) at 37 °C in 5% CO2.
column in comparison to the standard using system B.
4.3. Evaluation of median inhibitory concentration (IC50) values
for inhibition of 10 tyrosine kinases
A radiometric protein kinase assay (33PanQinaseÒ Activity As-
say) was used for measuring the kinase activity of the 10 protein
kinases. All kinase assays were performed in 96-well FlashPla-
4.7. Specific binding studies of [11C]8 in intact 293/KDR cells
tesTM from Perkin Elmer/NEN (Boston, MA, USA) in a 50
tion volume.
lL reac-
The receptor-binding properties of [11C]8 to VEGFR-2 were
evaluated in a direct radioligand binding assay using 293/KDR cells
(2 ꢂ 106 VEGFR-2/cell). 293/KDR cells were suspended in PBS
(1 ꢂ 106 cells/ml/microtube), and incubated at 4 °C for 30 min with
gentle agitation. To measure non-specific binding (NSB), one set of
microtubes was incubated with excess of non-labeled VEGFR-2
The assay for all enzymes contained 60 mM HEPES–NaOH, pH
7.5, 3 mM MgCl2, 3 mM MnCl2, 3
DTT, 50 g/mL PEG20000, 1 M [
per well). Depending on the kinase, the following substrate pro-
lM Na-orthovanadate, 1.2 mM
-
l
l
c
33P]-ATP (approx. 5 ꢂ 105 cpm
teins were used Tie-2, PDGFR
Tyr)6:2:5:1), EGFR, ERBB2, IGF1-R, KIT, VEGFR-1, VEGFR-2, VEGFR-
3 (Poly(Glu, Tyr)4:1
a and PDGFRb (Poly(Ala, Glu, Lys,
inhibitor (compound 8, 32 lM, 0.5% DMSO, 0.1% ethanol, 30 min).
)
To measure the total binding (TB), a matching set of microtubes
The IC50 values were measured by testing 12 concentrations of
compounds in triplicate. As a parameter for assay quality, the Z0-
factor for the low and high controls of each assay plate (n = 8)
was used. The final DMSO concentration in the assay was 1%. The
reaction cocktails were incubated at 30 °C for 80 min. The reaction
was incubated with 0.5% DMSO and 0.1% ethanol vehicle. [11C]8
(3.15 Ci/lmol, 0.5% ethanol final concn) was then added to the
microtubes at two concentrations (10, 20 nM) for 30 min. For each
concentration of [11C]8 that was added to the cells, a correspond-
ing control sample, consisting of cell-free PBS, was incubated with
the same concentration of the labeled inhibitor so as to calculate
the cell-independent retention of activity by the test microtubes.
Cell-bound radioactivity was separated from free radioactivity by
centrifugation at 2700g, 4 °C for 5 min, followed by washing with
PBS. Microtubes were measured for their radioactive content using
was stopped with 50
washed two times with 200
l
L of 2% H3PO4, plates were aspirated and
l
L of 0.9% NaCl. Incorporation of 33Pi
was determined with a microplate scintillation counter (Microbeta
Trilux, Wallac). All assays were performed with a BeckmanCoulter/
Sagian robotic system.
a c
-counter (1480 Wizard 300). Values obtained (in cpm) were cor-
4.4. Log D7.4 evaluation
rected for the cell-independent microtube retention of activity by
deducting the signals of the control samples. Specific binding
(SB) was calculated by subtracting the NSB from the corresponding
TB at each concentration.26
For this specific experiment, [11C]8 was eluted from the C18
sep-pak with n-octanol (5 mL) instead of ethanol/saline. Approxi-

O. Ilovich et al. / Bioorg. Med. Chem. 18 (2010) 612–620
619
ment with a pressure cooker for 12 min in a Nuclear Decloaker
solution (pH 9). The tumor sections were incubated with pri-
mary antibody specific for VEGFR-2 (rabbit monoclonal anti-
VEGFR-2, 1:100, Cell Signaling Technology, Boston, MA) for 1 h
in a humidifier chamber. Slides were washed in PBS and incu-
bated for 5 min with Zymed DAB staining kit. Negative control
staining was carried out by omitting primary antibody. Finally,
tissue sections were counterstained with hematoxylin and
coverslipped.
4.11. VEGFR extraction from tumors
Tumors were collected from mice, placed immediately in liquid
nitrogen and kept at ꢁ170 °C until VEGFR extraction. Prior to the
extraction, tumors were weighed and homogenized on ice (Omni
tip, US) in 2 mL/100 mg of lysis buffer (50 mM Tris–HCl; 1% NP-
40; 0.25% Na-deoxycholate; 150 mM NaCl, 1 mM EDTA; 1 mM
NaF; 1 mM Na3VO4 and 1% protease inhibitor). After homogeniza-
tion samples were centrifuged (15 min; 4 °C; 14,000 rpm) to allow
separation of pellet and supernatant. The supernatant (containing
VEGFR-2) was diluted (3:1) with Laemmli sample buffer (Bio-
Rad, Israel) and boiled for 10 min at 100 °C to allow for western
blot analysis. Protein amounts of the samples were measured by
Qubit fluorometer.
Figure 4. TLC of the radioactive metabolites extracted from blood samples at 0, 30,
and 60 min (left to right, respectively) showing only one major spot.
4.12. Western blot
Equal amounts of protein lysates (300 lg) were loaded and sep-
4.8. Animal studies
arated by sodium dodecyl sulfate polyacrylamide gel (SDS–PAGE,
Bio-Rad) and electrophoretically transferred to a nitrocellulose
membrane. The membrane was blocked in 5% low fat dehydrated
milk (BD) in TBST buffer (10 mM Tris–HCl, pH 7.4, 0.05% Tween
20) for 120 min, and then probed overnight at 4 °C with the follow-
ing primary antibodies diluted in blocking TBST: (I) rabbit poly-
clonal anti-Flk-1 (VEGFR-2), 1:1000 dilution (Santa Cruz
Biotechnology Inc.) and (II) mouse monoclonal anti-b-catenin,
1:200 dilution (Santa Cruz Biotechnology Inc.). The membrane
was washed thoroughly with TBST and incubated for 1 h at room
Nude Hsd-Athymic Nude-nu male mice (7–8 weeks, 25–30 g)
were obtained from Harlan Industries, Inc. All animal studies were
conducted under a protocol approved by the Research Animal Eth-
ics Committee of the Hebrew University of Jerusalem and in accor-
dance with its guidelines. Mice were allowed to acclimate in the
animal facility for at least 3 days prior to their inoculation with tu-
mor cells. They were routinely kept in 12-h light/dark cycles and
provided with food and water ad libitum.
temperature, with
a
horseradish peroxidase—conjugated
4.9. Development of tumor-bearing animal xenograft models
of 293/KDR and human glioma U87MG.wt EGFR
anti-mouse IgG (1:5000 dilution in blocking TBST). Finally, the
membrane was washed in TBST (4 ꢂ 5 min), and immunoreactive
proteins were visualized using an enhanced chemiluminescence
(ECL, PIR34080 pierce) detection reagent, and exposed to a FUGI
medical X-ray film (FUJIFILM Corporation, Tokyo).
The films were scanned with a BAS reader 3.1 version scanner
and analyzed with TINA 2.10 software. The intensity of each VEG-
FR-2 band was normalized to the intensity of the corresponding
b-catenin band to correct for differences in the content of total pro-
tein loaded.
293 Human kidney expressing VEGFR-2 (293/KDR), 293 Human
Embryonic Kidney (HEK) wt, human glioma U87MG.wt EGFR and
U138MG cells were cultured in vitro as described above. Subcon-
fluent cultures were released from the flasks with trypsin, counted
with a hemocytometer using trypan blue and suspended in DMEM
medium. U87MG.wt EGFR cell suspensions were supplemented
with Matrigel (20% vol/vol; BD Biosciences, Biological Industries,
Israel) and thoroughly mixed prior to their injection into mice.
Mice were inoculated subcutaneously with 1.5–2 ꢂ 106 for 7–
21 days for the 293/KDR cells, with 5 ꢂ 106 for 7–10 days for the
HEK cells, with 1.5–4 ꢂ 106 for 4–20 days for the U87MG.wt EGFR
cells and with 2–2.5 ꢂ 106 for 7–13 days for the U138 cells. All
4.13. [11C]8 Biodistribution studies in tumor-bearing mice
Biodistribution studies were carried out with Nude Hsd-Athy-
mic Nude-nu mice (10–12 weeks old) carrying U87MG.wt EGFR
or 293/KDR tumors grown as described above.
inoculations were in a volume of 100 lL.
[
11C]8 (197 16
lCi; 1.1 0.3 Ci/lmol; 225 11 lL) in a 10%
4.10. Histology and immunohistochemistry
ethanol/saline solution was injected via the lateral vein. At allotted
time points (30, 60 min), blood was drawn from the orbital sinus;
mice were sacrificed by cervical dislocation and selected organs
(tumor, skin, bone, liver, and kidney) were excised, weighed and
Histological and immunohistochemical validation of tumor
models was performed by Patho-Lab Ltd (Rehovot, Israel). Tu-
mors were generated as described above. Tumor specimens were
fixed in 4% formalin and delivered to Patho-Lab Ltd for further
treatment. Briefly, tumor samples were paraffin embedded, serial
paraffin sections of the tumors were deparaffinized and rehy-
drated, and endogenous peroxidase activity was quenched by
immersing the sections in 3% H2O2 for 10 min, followed by treat-
measured for their radioactive content using a
c-counter (1480
Wizard 300). Distribution of activity was calculated as the percent-
age of injected dose per gram of organ (%ID/g). Activity uptake
ratios of various organs were calculated by dividing the corre-
sponding calculated percentages of injected dose per gram of
organ.

620
O. Ilovich et al. / Bioorg. Med. Chem. 18 (2010) 612–620
7. Van de Wiele, C.; Boersma, H.; Dierckx, R. A.; De Spiegeleer, B.; Van Waarde, A.;
4.14. In vitro hepatic microsome stability
Elsinga, P. H. Curr. Pharm. Des. 2008, 14, 3340.
8. Decristoforo, C.; Hernandez Gonzalez, I.; Carlsen, J.; Rupprich, M.; Huisman, M.;
Virgolini, I.; Wester, H. J.; Haubner, R. Eur. J. Nucl. Med. Mol. Imaging 2008, 35,
1507.
9. Beer, A. J.; Schwaiger, M. Cancer Metastasis Rev. 2008, 27, 631.
10. Cai, W.; Chen, X. J. Nucl. Med. 2008, 49, 113S.
11. Kenny, L. M.; Coombes, R. C.; Oulie, I.; Contractor, K. B.; Miller, M.; Spinks, T. J.;
McParland, B.; Cohen, P. S.; Hui, A. M.; Palmieri, C.; Osman, S.; Glaser, M.;
Turton, D.; Al-Nahhas, A.; Aboagye, E. O. J. Nucl. Med. 2008, 49, 879.
12. Erdag, B.; Balcioglu, K. B.; Kumbasar, A.; Celikbicak, O.; Zeder-Lutz, G.;
Altschuh, D.; Salih, B.; Baysal, K. Mol. Biotechnol. 2007, 35, 51.
13. Youssoufian, H.; Hicklin, D. J.; Rowinsky, E. K. Clin. Cancer Res. 2007, 13, 5544s.
14. Panares, R. L.; Garcia, A. A. Exp. Rev. Anticancer Ther. 2007, 7, 433.
15. Cai, W.; Chen, X. Front. Biosci. 2007, 12, 4267.
Metabolism was evaluated by incubating the tracer [2–3 mCi
(74-111 GBq) in 100 lL ethanol] with 0.5 mg/mL mouse hepatic
microsomes (BD Biosciences) in PBS pH 7.4 at 37 °C for 5 min.
The reaction was started with the addition of 0.25 mM NADPH fol-
lowed by incubation in 37 °C for 0, 15, 30, and 60 min. At each time
point, 300–600 lL of the mixture were removed and loaded onto a
2 cm celite plug. The plug was dried and eluted with 2 mL of eth-
anol. The ethanol solution was analyzed by RP-HPLC using system
B.19,27
16. Chen, K.; Cai, W.; Li, Z. B.; Wang, H.; Chen, X. Mol. Imaging Biol. 2009, 11, 15.
17. Ilovich, O.; Jacobson, O.; Aviv, Y.; Litchi, A.; Chisin, R.; Mishani, E. Bioorg. Med.
Chem. 2008, 16, 4242.
Acknowledgments
18. Ilovich, O.; Åberg, O.; Långström, B.; Mishani, E. J. Labelled Compd. Radiopharm.
2009, 52, 151.
19. Samen, E.; Thorell, J. O.; Lu, L.; Tegnebratt, T.; Holmgren, L.; Stone-Elander, S.
Eur. J. Nucl. Med. Mol. Imaging 2009, 36, 1283.
20. Peifer, C.; Krasowski, A.; Hammerle, N.; Kohlbacher, O.; Dannhardt, G.; Totzke,
F.; Schachtele, C.; Laufer, S. J. Med. Chem. 2006, 49, 7549.
21. Peifer, C.; Stoiber, T.; Unger, E.; Totzke, F.; Schachtele, C.; Marme, D.; Brenk, R.;
Klebe, G.; Schollmeyer, D.; Dannhardt, G. J. Med. Chem. 2006, 49, 1271.
22. Faul, M. M.; Winneroski, L. L.; Krumrich, C. A. J. Org. Chem. 1998, 63, 6053.
23. Lin, K. S.; Ding, Y. S. Chirality 2004, 16, 475.
We would like to thank Dr. Orit Jacobson for her help with the
radiochemical experiments. This research was supported by the Is-
rael Science Foundation (Grant #445/07 to E.M.). Ohad Ilovich’s re-
search was generously supported by the Hoffman Leadership and
Responsibility fund, at the Hebrew University.
References and notes
24. Crouzel, C.; Långström, B.; Pike, V. W.; Coenen, H. H. Int. J. Radiat. Appl. Inst. Part
A Appl. Radiat. Isot. 1987, 38, 601.
25. Abourbeh, G.; Dissoki, S.; Jacobson, O.; Litchi, A.; Ben Daniel, R.; Laki, D.;
Levitzki, A.; Mishani, E. Nucl. Med. Biol. 2007, 34, 55.
26. de Lemos Rieper, C.; Galle, P.; Svenson, M.; Pedersen, B. K.; Hansen, M. B. J.
Immunol. Methods 2009, 350, 46.
27. Lee, S. Y.; Choe, Y. S.; Kim, D. H.; Park, B. N.; Kim, S. E.; Choi, Y.; Lee, K. H.; Lee, J.;
Kim, B. T. Nucl. Med. Biol. 2001, 28, 391.
1. Pandya, N. M.; Dhalla, N. S.; Santani, D. D. Vasc. Pharmacol. 2006, 44, 265.
2. Levitzki, A.; Mishani, E. Annu. Rev. Biochem. 2006, 75, 93.
3. Fujita, Y.; Abe, R.; Shimizu, H. Curr. Pharm. Des. 2008, 14, 3820.
4. Chase, J. L. Pharmacotherapy 2008, 28, 23S.
5. Pytel, D.; Sliwinski, T.; Poplawski, T.; Ferriola, D.; Majsterek, I. Anticancer Agents
Med. Chem. 2009, 9, 66.
6. Serkova, N. J.; Garg, K.; Bradshaw-Pierce, E. L. Rec. Pat. Anticancer Drug Discovery
2009, 4, 35.