Radioiodinated N-(2-diethylaminoethyl)benzamide derivatives with high melanoma uptake: Structure-affinity relationships, metabolic fate, and intracellular localization

Article abstract of DOI:10.1021/jm991079p

Several radioiodinated N-(dialkylaminoalkyl)benzamides have been used for planar scintigraphy and single-photon emission computed tomography (SPECT) of melanoma metastases. In a quest for improved melanoma uptake and tissue selectivity, structure-activity studies for N-(2-diethylaminoethyl)benzamides with variation of phenyl substituents were performed using C57B1/6 mice bearing B16 melanoma. Compounds 2 (4-amino-5-bromo-N-(2-diethylaminoethyl)-3-[¹³¹I]iodo-2-methoxybenzamide) and 6 (4-acetamido-N-(2-diethylaminoethyl)-5-[¹³¹I]iodo-2-methoxybenzamide) showed at 6 h post iv injection, for example, melanoma uptake of 16.6 and 23.2% ID/g, respectively (mean values, n = 3). Uptake was 3-5 times higher (P < 0.01) than observed with benzamides known from the literature and was probably facilitated by the relatively slow urinary excretion of 2 or 6. In contrast, analogues lacking either the MeO, Ac, AcNH, or Br substituents exhibited reduced tumor uptake and high urinary excretion of radioactivity in various benzamide metabolites. Uptake of radioiodinated benzamides in B 16 melanoma is not mediated by a specific mechanism such as σ-receptor binding. 2 and 6 exhibited similar melanoma uptake values but quite different σ₁-receptor affinities of K(i) = 0.278 ± 0.018 and 5.19 ± 0.40 μM, respectively. Uptake studies with IMBA (N-(2-diethylaminoethyl)-3-[¹³¹I]-iodo-4-methoxybenzamide) or BZA (N-(2-diethylaminoethyl)-4-[¹³¹I]iodobenzamide) showed that with increasing dose of unlabeled compound the measured uptake of label was unchanged (IMBA) or even enhanced (BZA) while receptor binding of label decreased. Differential and equilibrium density-gradient centrifugation revealed that most of the radioactivity from labeled IMBA was associated with fractions containing melanin granules. Thus, structure-activity studies indicate that blood clearance rates and metabolic stability are the main determinants for benzamide uptake in melanoma. The high uptake and slow clearance of 6 offer considerable potential for melanoma imaging in patients, and this compound may also prove to be useful for radionuclide therapy.

Full text of DOI:10.1021/jm991079p

J . Med. Chem. 2000, 43, 3913-3922
3913
Ra d ioiod in a ted N-(2-Dieth yla m in oeth yl)ben za m id e Der iva tives w ith High
Mela n om a Up ta k e: Str u ctu r e-Affin ity Rela tion sh ip s, Meta bolic F a te, a n d
In tr a cellu la r Loca liza tion
Michael Eisenhut,* William E. Hull,† Ashour Mohammed, Walter Mier, Dorothee Lay,‡ Wilhelm J ust,‡
Karin Gorgas,# Wolf D. Lehmann,† and Uwe Haberkorn
Department of Nuclear Medicine, Center of Biochemistry, and Institute for Anatomy and Cell Biology, University of Heidelberg,
D-69120 Heidelberg, Germany, and Central Spectroscopy Department, German Cancer Research Center (DKFZ),
Heidelberg, Germany
Received May 11, 1999
Several radioiodinated N-(dialkylaminoalkyl)benzamides have been used for planar scintigraphy
and single-photon emission computed tomography (SPECT) of melanoma metastases. In a quest
for improved melanoma uptake and tissue selectivity, structure-activity studies for N-(2-
diethylaminoethyl)benzamides with variation of phenyl substituents were performed using
C57Bl/6 mice bearing B16 melanoma. Compounds 2 (4-amino-5-bromo-N-(2-diethylaminoethyl)-
3-[131I]iodo-2-methoxybenzamide) and 6 (4-acetamido-N-(2-diethylaminoethyl)-5-[131I]iodo-2-
methoxybenzamide) showed at 6 h post iv injection, for example, melanoma uptake of 16.6
and 23.2% ID/g, respectively (mean values, n ) 3). Uptake was 3-5 times higher (P < 0.01)
than observed with benzamides known from the literature and was probably facilitated by the
relatively slow urinary excretion of 2 or 6. In contrast, analogues lacking either the MeO, Ac,
AcNH, or Br substituents exhibited reduced tumor uptake and high urinary excretion of
radioactivity in various benzamide metabolites. Uptake of radioiodinated benzamides in B16
melanoma is not mediated by a specific mechanism such as σ-receptor binding. 2 and 6 exhibited
similar melanoma uptake values but quite different σ1-receptor affinities of Ki ) 0.278 ( 0.018
and 5.19 ( 0.40 µM, respectively. Uptake studies with IMBA (N-(2-diethylaminoethyl)-3-[131I]-
iodo-4-methoxybenzamide) or BZA (N-(2-diethylaminoethyl)-4-[131I]iodobenzamide) showed that
with increasing dose of unlabeled compound the measured uptake of label was unchanged
(IMBA) or even enhanced (BZA) while receptor binding of label decreased. Differential and
equilibrium density-gradient centrifugation revealed that most of the radioactivity from labeled
IMBA was associated with fractions containing melanin granules. Thus, structure-activity
studies indicate that blood clearance rates and metabolic stability are the main determinants
for benzamide uptake in melanoma. The high uptake and slow clearance of 6 offer considerable
potential for melanoma imaging in patients, and this compound may also prove to be useful
for radionuclide therapy.
In tr od u ction
Introduction of a polar phenyl substituent such as
4-MeO in IMBA (N-(2-diethylaminoethyl)-3-[123I]iodo-
4-methoxybenzamide) resulted in faster clearance from
nontarget tissue than obtained with BZA (N-(2-diethyl-
aminoethyl)-4-[123I]iodobenzamide) and improved con-
trast in scintigraphic images of melanoma patients.5 A
subsequent systematic investigation of the influence of
the phenyl substituents HO, MeO, and H2N on mela-
noma uptake and pharmakokinetics in mice led to
several derivatives with characteristics similar to but
not better than those of IMBA.6
In our continuing efforts to improve scintigraphic
tumor contrast by modifying the benzamide structures,
we synthesized a series of compounds which may
enhance our understanding of the influence of phenyl
substituents on melanoma uptake. Some of the new
benzamide derivatives exhibited much higher affinity
for melanoma tissue than observed with the benzamides
known from the literature and our own previous work.
This paper reports the synthesis, characterization, and
biological evaluation of these new compounds with
emphasis on structure-activity relationships (SAR),
The ability of melanoma metastases to disseminate
rapidly means that early detection is an important
diagnostic goal for the successful treatment and fol-
lowup of patients. Therefore, the development of an
effective radiopharmaceutical with melanoma affinity
has been pursued for some time. The discovery of
radioiodinated benzamides which exhibit affinity to
melanocytes led to the development of several N-(2-
dialkylaminoalkyl)-4-iodobenzamide derivatives. Their
remarkable affinity was accidentally observed as uveal
uptake in pigmented experimental animals.1 The evalu-
ation of structural variations in benzamide derivatives
in terms of their biological behavior in melanoma-
bearing mice focused primarily on amide substituents
and the position of radioiodine in the phenyl ring.2-4
* To whom correspondence should be addressed. Tel: ++49-6221-
567720. Fax: ++49-6221-565473. E-mail: michael_eisenhut@med.uni-
heidelberg.de.
† German Cancer Research Center.
‡ Center of Biochemistry.
§ Institute for Anatomy and Cell Biology.
10.1021/jm991079p CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/26/2000

3914 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21
Eisenhut et al.
Sch em e 1a
a
Reaction conditions: (a) 2-diethylaminoethylamine, 90 °C/12 h; (b) AcCl, 25 °C/1 h; (c) Tl(TFA)3/TFA, 25 °C/30 min, and I-, 25 °C/5
h or 131I-, 25 °C/5 min; (d) KIO3/I-, 25 °C/30 min or KIO3/131I-, 25 °C/10 min; (e) 2-diethylaminoethylamine, 25 °C/2 h.
where activity refers to melanoma affinity in our
context. In addition, the biological mechanisms which
may influence scintigraphic contrast in clinical studies
were investigated by tracing the metabolic fate of [123I]-
IMBA5 and [123I]BZA.7 Finally, to learn more about the
binding and cellular localization of the benzamides, e.g.,
the importance of σ1-receptors versus melanin, we
performed in vitro receptor binding studies and in vivo
saturation studies and investigated the intracellular
distribution of the benzamides in melanoma cells by
density-gradient centrifugation measurements.
precursor 5 which was thallated and iodinated as
described above. 7 was obtained by adding an aqueous
iodide solution to an equimolar mixture of precursor 4
and KIO3 in HCl. N-Acetylprocainamide 8 was iodinated
with Tl(TFA)3/I- as described above to give 9. 12 was
synthesized by the reaction of 3-methoxybenzoyl chlo-
ride (10) with 2-diethylaminoethylamine to give ben-
zamide 11. The corresponding 2-iodo-5-methoxy deriva-
tive 12 was obtained by iodination with Tl(TFA)3/I- as
described above. 15 was synthesized in a similar man-
ner; the corresponding acid chloride 13 was reacted with
excess 2-diethylaminoethylamine to give the amide 14
which was iodinated with Tl(TFA)3/I-. Purification was
performed by recrystallization until HPLC indicated
>95% purity. The complete and unambiguous structural
identification of the various benzamides was only pos-
Resu lts
Ch em istr y. The syntheses of the benzamides 2, 6,
7, 9, 12, and 15 and their radioiodinated analogues were
accomplished by the routes outlined in Scheme 1. For 2
the commercially available bromopride 1 was regiose-
lectively iodinated by replacing a bis(trifluoroacetyl)-
thallium group generated in situ at the benzamide C3
position with iodine.5,6 Thallation, without separation
of products, and iodination were performed at ambient
temperature. The synthesis of 6 was started with the
aminolysis of methyl 4-amino-2-methoxybenzoate (3)
using excess 2-diethylaminoethylamine. Subsequent
reaction of compound 4 with acetyl chloride afforded
1
sible through the use of both H and 13C NMR (Tables
5 and 6). Although many of the signal assignments
(numbering given in Chart 1) for the one-dimensional
spectra could be tentatively made using chemical shift
predictions, a definitive determination of the phenyl ring
substitution pattern required the a priori positional
assignments for all carbons via long-range couplings to
protons. Therefore, for several compounds the fully 1H-
coupled 13C NMR spectrum was acquired and, where
necessary, 13C NMR spectra with selective 1H decou-

Radioiodinated N-(2-Diethylaminoethyl)benzamides
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21 3915
Ch a r t 1. Numbering for the Signal Assignments in 1H
and 13C NMR Spectra of 2, 6-9, 12, and 15 (Tables 5
and 6)
pling and 1H-1H nuclear Overhauser effects (NOE)
were examined.
For the analogous radioiodination reactions the molar
amounts of the precursors were reduced so that the
complete reaction mixture could be separated on an
analytical RP-18 column. Excellent radiochemical yields
were obtained with the Tl(TFA)3/131I- and KIO3/131I-
methods (70-90%). Specific activities of the radioiodi-
nated benzamides 2-15 were in the range 65-90 Gbq/
µmol.
Biod istr ibu tion Stu d ies. To judge their potential
as tumor markers, the 131I-labeled benzamides 2, 6, 7,
9, 12, and 15 were injected intravenously into C57Bl/6
mice bearing subcutaneously transplanted B16 mela-
noma. Time points of 1 and 6 h after injection were
chosen for determining the distribution of each com-
pound in various organs and tissues. For those benz-
amides with high melanoma uptake, a 24-h time point
was included to judge long-term retention. The biodis-
tribution data are summarized in Table 1.
The benzamides 2 and 6 exhibited the highest mela-
noma uptake and excellent retention. At 1 h the uptake
was 17 and 22% ID/g, respectively, with average mela-
noma/nontarget tissue ratios (M/NTT, selectivity pa-
rameter) of 9 and 10. At 6 h tumor content was
essentially unchanged, and M/NTT increased to 25 and
30, respectively. After 24 h the radioactivity in tumor
decreased to ca. 8 and 16% ID/g while clearance from
other tissues (except liver) was high, yielding M/NTT
values of 132 (2) and 251 (6).
The influence of the phenyl ring substituents on tissue
uptake is reflected by the results for 2 and 6 in
comparison with the other benzamide derivatives in
Table 1. The following structural changes led to a
significant (P < 0.001) absolute decline in melanoma
uptake: (1) 5-Br f 5-I and 3-I f 3-H (2 f 7); (2)
4-AcNH f 4-NH2 (6 f 7); (3) 2-OMe f 2-H (6 f 9); (4)
F igu r e 1. γ-HPLC of urine collected for 4 h after injecting
131I]IMBA (lower panel) or 131I]BZA (upper panel) into
C57Bl/6 mice. Arrows indicate the retention times of the
original compounds.
[
[
4-AcNH f 4-H (6 f 15). Additionally, the M/NTT ratios
(selectivity) also decreased for the changes listed above,
except for 6 f 15.
The benzamides 12-15 were also compared with
IMBA5 to learn whether the positions of radioiodine and
the methoxy group can be optimized. For example, for
the rearrangement 3-I f 2-I and 4-OMe f 5-OMe
(IMBA f 12), melanoma uptake increased slightly at 1
h but decreased slightly at 6 h (one-sided P ) 0.08 in
both cases) while selectivity decreased, especially at 6
h. For the rearrangement to 2-OMe, 5-I (15) there was
little improvement in absolute uptake at 1 h (one-sided
P ) 0.1) and a small improvement at 6 h (one-sided P
) 0.06) compared to IMBA, but selectivity was poorer.
Ur in a r y Meta bolites of Ben za m id es. In view of
the significant differences in the biodistribution of the
benzamides, we investigated the metabolic fate of some
of these compounds (IMBA, BZA, 2, 6, and 15). After iv
injection of a given 131I-labeled benzamide into C57Bl/6
mice, urine was collected for 4 h and analyzed by RP-
HPLC. The elution profiles for 131I-labeled IMBA and
BZA are shown in Figure 1, and the differences observed
indicate that these two compounds follow different
metabolic routes. While IMBA was mainly transformed
Ta ble 1. Biodistribution of Radioiodinated N-(2-Diethylamino)benzamides 2-15 in C57Bl/6 Mice with B16 Melanoma
phenyl substituent position
mean tissue concentrations (% ID/g, n ) 3)a
blood heart lung spleen liver kidney muscle brain
time (h,
compd
2
3
4
5
postinject) M/NTTb
melanoma
2
MeO
I
NH2
Br
I
1
6
24
1
6
24
1
6
1
6
1
9
25
132
10
30
251
6
19
3
3
15
24
14
73
24
129
16.61 ( 1.44 2.32 ( 0.06 1.91 3.16
16.48 ( 0.46 1.86 ( 0.61 0.94 2.07
8.02 ( 0.69 0.19 ( 0.05 0.13 0.26
21.87 ( 0.43 3.30 ( 0.56 2.10 5.53
23.32 ( 0.36 2.56 ( 1.23 1.05 3.02
16.06 ( 1.25 0.21 ( 0.00 0.09 0.25
4.66 ( 0.29 1.93 ( 0.08 1.17 3.12
6.92 ( 0.85 0.85 ( 0.59 0.45 1.06
6.21 ( 1.29 6.50 ( 0.73 3.19 7.98
1.20 ( 0.12 1.51 ( 0.38 0.58 1.82
8.06 ( 1.22 0.90 ( 0.16 0.63 1.41
2.32 ( 0.87 0.33 ( 0.23 0.13 0.41
7.73 ( 1.08 0.46 ( 0.06 0.46 1.15
6.80 ( 1.53 0.14 ( 0.06 0.09 0.19
6.50 ( 0.44 0.20 ( 0.10 0.25 0.69
4.14 ( 1.20 0.04 ( 0.00 0.05 0.09
5.06
1.84
0.38
6.55
2.09
0.17
2.00
1.57
5.62
1.39
1.26
0.31
2.50
0.15
0.39
0.05
11.32
9.61
4.54
8.03
9.86
3.72
4.92
3.43
4.05
0.77
1.23
0.23
0.88
0.26
2.62
0.29
4.53
2.08
0.46
5.09
1.97
0.40
6.89
3.95
6.46
1.22
1.81
0.17
1.71
0.41
0.93
0.17
0.94
0.43
0.01
1.37
0.44
0.03
0.61
0.28
1.81
0.44
0.38
0.08
0.19
0.04
0.19
0.03
0.64
0.15
0.02
0.62
0.19
0.02
0.18
0.08
0.48
0.09
0.15
0.02
0.50
0.04
0.10
0.01
6
MeO
H
AcNH
7
MeO
H
H
I
NH2
AcNH
H
I
9
H
12
I
H
H
I
MeO
6
1
6
1
15
MeO
H
H
I
IMBAc
MeO
H
6
a
b
To simplify the table, standard deviations are shown only for melanoma and blood data. Selectivity parameter calculated as the
mean of 8 melanoma/nontarget tissue ratios. c From ref 5.

3916 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21
Ta ble 2. Excretion of Radioactivity into Urinea
Eisenhut et al.
equation,10 and the results are summarized in Table 3.
The inhibition potency of the benzamides varied con-
siderably. Moderate affinities for σ1-receptors were
obtained with 2, 7, 12, 15, and IMBA. Their Ki values
ranged from 92 nM (15) to 484 nM (7), while 6 and 9
exhibited Ki values of about 5 µM. In general, the
benzamides proved to be relatively poor ligands for σ2-
receptors (Ki > 1 µM), with the exception of 12 (Ki )
347 nM).
compd
4-h urineb
comparison (P value)c
6 (0.01)
2
6
34 ( 6
16 ( 3
71 ( 5
66 ( 5
82 ( 8
15
BZA
IMBA
2 (0.001)
2 (0.005)
2 (<0.001)
15 (0.05)
a
b
NMRI mice. % ID, mean ( SD (3 animals). c P value for
comparison of means (one-sided t-test with equal variance).
The expression of σ-receptors in B16 melanoma cells
was investigated by saturating tumor cell membranes
with [3H]-(+)-pentazocine and [3H]DTG. The cell mem-
branes (P2 pellet) were obtained from C57Bl/6 mice with
subcutanously transplanted B16 melanoma. To avoid
complications with necrotic tissue, cells were harvested
from small tumors weighing 0.1-0.25 g. Tissue affinity
for 131I-labeled benzamides was tested with [131I]2 and
found to be consistent with the uptake data listed in
Table 1: i.e., 14.2% ID/g at 1 h after injection. The Kd
values of [3H]-(+)-pentazocine (σ1) and [3H]DTG/dex-
trallorphan (σ2) obtained with B16 melanoma mem-
branes were 109 ( 33 nM (Bmax ) 23200 ( 4200 fmol/
mg) and 101 ( 25 nM (Bmax ) 18700 ( 2670 fmol/mg),
respectively.
Ta ble 3. Receptor Binding Data and Partition Coefficients for
Benzamides 2-15a
Ki (nM)
compd
σ1-receptorb
σ2-receptorc
log P,d pH 7.4
2
6
7
9
278 ( 18
5190 ( 397
484 ( 27
4645 ( 508
455 ( 35
92 ( 8
2350 ( 65
115000 ( 18200
21400 ( 17800
148000 ( 30800
347 ( 23
1.81 ( 0.11
1.78 ( 0.01
1.62 ( 0.02
1.34 ( 0.09
1.16 ( 0.02
1.26 ( 0.02
1.13 ( 0.01
12
15
IMBA
1020 ( 91
249 ( 17
2.95 ( 0.58
6520 ( 360
46.5 ( 2.1
haloperidol
a
b
Results represent mean ( SD for 2-4 measurements. Com-
petition assay with guinea pig brain membranes and [3H]-(+)-
pentazocine as ligand. c Rat liver membranes with [3H]DTG as
d
ligand. Octanol/water partition coefficient.
In Vivo Sa tu r a tion Exp er im en ts. Additional in-
formation about the binding of radioiodinated benza-
mides in B16 cells as a function of specific activity was
obtained by injecting C57Bl/6 mice with a constant
amount of radioiodinated IMBA or BZA together with
increasing amounts of the unlabeled compound. The
biodistribution data for melanoma and blood are sum-
marized in Table 4. The melanoma content of labeled
IMBA showed little change with decreasing specific
activity, while the uptake of labeled BZA actually
increased with the addition of unlabeled compound and
with time (6 h vs 1 h). Increasing amounts of unlabeled
IMBA and BZA injected into the animals altered the
M/NTT ratios (specificity) differently. The ratios in-
creased using BZA and were reduced with IMBA. The
reduction of IMBA specificity passed, however, not
below the ratios obtained with no-carrier-added [131I]-
BZA.
In tr a cellu la r Distr ibu tion of [131I]IMBA in B16
Mela n om a . After exposure to [131I]IMBA excised tumor
tissue was homogenized, and the homogenate was
fractionated by differential centrifugation (Figure 2A).
A sample of the unfractionated homogenate as well as
pellets 2 and 3 obtained by differential centrifugation
and sedimentation at 3500g and 14000g, respectively,
were further analyzed by equilibrium density-gradient
centrifugation (Figure 2B). The measured distribution
of radioactivity revealed that the majority of [131I]IMBA
(more than 60% in fractions 1-4) sedimented to a high
equilibrium density (1.26 g/mL), beyond that of peroxi-
somes which usually are recovered at densities of 1.22-
1.23 g/mL. This result indicates that [131I]IMBA cosed-
iments with melanin granules which are known to
exhibit buoyant densities of about 1.26 g/mL in a sucrose
gradient.11 Finally, the organelles were recovered from
the peak fractions with the highest content of [131I]-
IMBA, and ultrastructural analysis was performed.
Figure 2C shows the subcellular composition of fraction
2, demonstrating that the majority of structures present
were melanin granules.
to a more polar metabolite, BZA was converted to a more
lipophilic compound. The urinary metabolite pattern for
15 was nearly identical to that for IMBA. In contrast,
the highly accumulating benzamides 2 and 6 were
excreted to a lesser extent. As summarized in Table 2,
the amount of urinary radioactivity collected within 4
h after injection was lowest with 6 and 2: i.e., only about
one-fourth or one-half, respectively, of the excretion
observed with benzamide 15, BZA, or IMBA. For 2 and
6 HPLC analysis of the urine samples revealed that the
radioactivity was almost exclusively in the form of [131I]-
iodide.
The main metabolites of BZA and IMBA were identi-
fied by injecting 1 mg of the unlabeled benzamide into
C57Bl/6 mice. Urine was collected for 4 h postinjection,
and desalted samples were analyzed by negative-ion
ESI-MS/MS. The [I]- fragment at m/z 126.9 was used
for a parent-ion scan to detect only iodinated species.
The main metabolites of IMBA were identified as the
sulfate (m/z 441) and glucuronate (m/z 537) of N-(2-
diethylaminoethyl)-4-hydroxy-3-iodobenzamide. 4-Iodo-
hippuric acid (m/z 304) and an unknown metabolite (m/z
483) were detected using BZA.
P a r tition Coefficien ts. The lipophilicities were
measured by partitioning no-carrier-added radioiodi-
nated benzamides 2, 6, 7, 9, 12, and 15 and IMBA in
an 1-octanol/buffer system at pH 7.4. The results are
summarized in Table 3. The log P values fall into two
groups with ranges of 1.6-1.9 (2, 6, and 7) and 1.1-1.3
(9, 12, 15, and IMBA).
Recep tor Bin d in g Stu d ies. The IC50 values for the
binding of unlabeled benzamides 2-15 to σ-receptors
were determined by means of competitive binding
assays, using guinea pig brain membranes (σ1-receptors)
with the ligand [3H]-(+)-pentazocine8 and rat liver
membranes (σ2-receptors) with [3H]DTG.9 The apparent
Kd values for the radioligands were 6.44 ( 0.53 nM (σ1)
and 23.7 ( 2.0 nM (σ2), respectively. Ki values for the
benzamides were calculated using the Cheng-Prusoff

Radioiodinated N-(2-Diethylaminoethyl)benzamides
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21 3917
Ta ble 4. Influence of Carrier Addition on Melanoma Uptake and Blood Values of Labeled Benzamidea
melanoma (% ID/g)
blood (% ID/g)
M/NTTb
6 h
total BA injected
(µmol/mouse)
compd
IMBA
1 h
6 h
1 h
6 h
1 h
1.8 × 10-6
1.5 × 10-3
1.5
6.50 ( 0.44
4.86 ( 0.56
5.41 ( 0.43
5.67 ( 0.45
10.02 ( 0.48c
11.60 ( 0.60d
4.14 ( 1.20
5.27 ( 0.49
3.72 ( 0.54
9.32 ( 0.21
15.47 ( 0.73c
14.58 ( 1.47d
0.20 ( 0.10
1.03 ( 0.04
0.78 ( 0.26
0.65 ( 0.04
1.06 ( 0.36
1.04 ( 0.19
0.04 ( 0.00
0.24 ( 0.08
0.29 ( 0.11
0.17 ( 0.03
0.15 ( 0.03
0.10 ( 0.04
24
10
10
3
4
5
129
88
48
27
50
45
BZA
3.9 × 10-7
2.0 × 10-3
1.7
a
d
Mean (n ) 3). b Selectivity parameter calculated as the mean of 8 melanoma/nontarget tissue ratios. c P < 0.01. P < 0.05 (in comparison
with no-carrier-added injection).
Discu ssion
Thus, we find that the nature and position of phenyl
substituents in the benzamide derivatives have a sig-
nificant influence on their biodistribution and useful-
ness as melanoma markers. Possible mechanisms for
these SAR effects could involve (1) efficiency of transport
to the target tissue or binding site, (2) relative affinities
at target and nontarget binding sites, or (3) rates of
metabolism and renal clearance which influence intra-
vasal bioavailability. On the basis of the large amount
of data available from previous investigations,5,6 it is
likely that the third group of mechanisms is primarily
responsible for the observed effects. From Tables 1 and
2 we see that 6, which exhibits the highest uptake in
tumor and several other tissues, also has the lowest
clearance of radioactivity into the urine. For 2 mela-
noma uptake is somewhat lower while renal clearance
is somewhat higher. In both of these cases urinary
radioactivity was found to be primarily [131I]iodide (I-).
These results are consistent with the hypothesis that
in the absence of efficient metabolic pathways renal
clearance is slow, resulting in high intravasal bioavail-
ability, high uptake, and relatively slow clearance from
all tissues but especially from melanoma. The relatively
high partition coefficients (lipophilicity) observed for 2
and 6 (ca. 1.8, see Table 3) may also contribute to their
favorable uptake.
In contrast, benzamide derivatives which can be
efficiently metabolized may show more rapid clearance
from tissues and higher radioactivity excretion rates.
Metabolism has been proven for BZA (conversion to
4-iodohippuric acid by amide bond cleavage and forma-
tion of the glycine conjugate), IMBA (conversion to the
4-sulfate ester and 4-glucuronate), and 15 (metabolites
not shown but similar to those of IMBA). Thus, metabo-
lism is likely to be responsible for the relatively rapid
nontarget tissue clearance observed for 15 and IMBA
in Table 1 (especially from liver, spleen, and kidney) and
the high renal clearance (Table 2) observed for all three
compounds. Furthermore, blood levels at 1 or 6 h for
the compounds listed in Table 2 were inversely cor-
related with the 4-h urinary excretion data. All of these
results point to intravascular concentration as a major
determinant of melanoma uptake. This conclusion is
supported by the results of Figure 3, where melanoma
tissue concentrations (radioactivity) at 1 h postinjection
are plotted against blood concentrations for all com-
pounds in Table 1. With the exception of compound 9,
the data show a modest linear correlation (r ) 0.85),
indicating that melanoma uptake is roughly propor-
tional to the concentration of benzamide circulating in
the blood. Thus, differences in the biodistribution data
for the benzamide derivatives of Table 1 reflect differ-
ences in the routes of metabolic degradation and the
For the radioiodinated N-(2-diethylaminoethyl)ben-
zamide derivatives discussed up to now in the literature,
diverse values of melanoma uptake have been observed.
In our current study of B16 melanoma in the C57Bl/6
mouse (Table 1), benzamides 2 and 6 exhibited the
highest uptake at 1 and 6 h postinjection with a 3-5-
fold increase compared to radioiodinated IMBA, a
recently developed benzamide with improved scinti-
graphic contrast.5 2 and 6 also have the advantage of
slow clearance from melanoma, with tumor levels
decreasing from 6 to 24 h postinjection by only about
50% or 30%, respectively. Thus, tumor selectivity,
expressed as the mean of the eight measured melanoma/
nontarget tissue ratios (M/NTT), increased with time,
reaching values of about 130 or 250, respectively, 1 day
after injection.
Although IMBA exhibits only moderate melanoma
uptake, comparable to that observed for several other
benzamides in this study, it has the advantage of much
faster clearance from nontarget tissues, including the
liver, so that M/NTT reaches about 130 after only 6 h.
Thus, IMBA can provide the highest scintigraphic
contrast for melanoma metastases5 at early times
postinjection, but 2 or 6 can provide a significant
improvement in sensitivity while reaching similar con-
trast ratios ca. 24 h after injection. A possible disad-
vantage of 2 and 6 is their relatively high uptake in the
liver and slow clearance from that organ (melanoma/
liver ratios at 24 h are only 1.8 and 4.3, respectively).
Role of Meta bolism a n d Ren a l Clea r a n ce. In the
course of investigations aimed at improving the biologi-
cal characteristics of IMBA, derivatives with the iodo
and methoxy substituents at different positions in the
phenyl ring have been prepared. Thus, 2-methoxy-5-
iodophenyl substitution (15) led to a small increase in
melanoma uptake (19% at 1 h and 64% at 6 h postin-
jection), but selectivity in terms of melanoma/nontarget
tissue ratios decreased, except for liver and muscle. The
opposite arrangement of phenyl substituents (2-iodo-5-
methoxy in 12) led to uptake and selectivity character-
istics that were similar to those of IMBA at 1 h but
significantly poorer at 6 h. On the other hand, when
the phenyl substituents of 2 or 6 were altered or
removed, absolute uptake into melanoma was reduced
to levels similar to those observed for IMBA, and in
some cases selectivity was also decreased. For example,
loss of the acetyl group on the amine function at C4 (7)
reduced uptake and to some extent selectivity. Loss of
the methoxy group at C2 and the halogen at C5 (9)
reduced uptake and increased clearance from tumor
with a dramatic loss of selectivity at 6 h postinjection.

3918 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21
Eisenhut et al.
F igu r e 3. Correlation plot for benzamide uptake in melanoma
tissue vs blood concentration at 1 h after injection of radioio-
dinated 2, 6, 7, 9, 12, 15, and IMBA. Each data point
represents one animal and compound. The linear regression
(r ) 0.85) is for all data points except those for 9.
other benzamides. Uptake at 1 h postinjection is similar
for tumor, lung, spleen, liver, and kidney and about
equal to the blood level; uptake is lower in heart, muscle,
and expecially brain. At 6 h all tissue levels decrease
in parallel with the blood level by about a factor of 4-5.
Thus, 9 exhibits no selectivity for melanoma compared
to several other important organs, and the data suggest
that there is a rapid equilibrium between extravascular
tissue and the intravascular volume.
Role of σ-Recep tor s. Radioiodinated N-(2-diethyl-
aminoethyl)benzamide derivatives have been described
to exhibit high affinity for the σ1-receptor2,12 expressed
by several tumors.13 As summarized in Table 3 the
benzamides investigated here showed high to moderate
σ1-receptor affinities in vitro (Ki range: 92-5190 nM)
which, however, do not correlate with the melanoma
uptake values in Table 1 (range: 4.7-23.3% ID/g at 1
h postinjection). The N-acetylated derivative 6 exhibited
the lowest σ1 affinity but the highest uptake. Except for
12, affinities for σ2-receptors were at least 1 order of
magnitude lower and also did not correlate with mela-
noma uptake. Thus, the inherent affinity of a given
benzamide for σ-receptors is not an important determi-
nant of melanoma uptake.
The expression of σ-receptors in the tumor model
studied was evaluated by saturating B16 melanoma
membranes with [3H]-(+)-pentazocine (σ1) and [3H]DTG/
dextrallorphan (σ2). The Kd values of both receptors were
found to be comparable with the Kd values found for the
amelanotic melanoma A375.13 However, the binding
capacities in terms of the Bmax values for both receptor
types in B16 melanoma were 5-7 times higher than for
the A375 tumor. We interpret this difference to be due
to the additional interaction of [3H]-(+)-pentazocine and
[3H]DTG with the melanin present in B16 melanoma.
In a recent clinical trial using [123I]BZA,7 which also
exhibits high affinity for σ1-receptors,2 positive scinti-
graphic imaging of melanoma and metastases was
achieved with specific activities as low as 0.006 GBq/
µmol. For comparison, commercially available radiop-
harmaceuticals used for receptor imaging have specific
activities of typically 15 or 200 GBq/µmol for [111In]-
pentetreotide or [123I]IBZM, respectively. Therefore,
saturation studies for B16 melanoma in the C57Bl/6
mouse model have been performed by injecting [131I]-
F igu r e 2. A: Distribution of radioactivity obtained by dif-
ferential centrifugation of melanoma tissue homogenate. Pel-
lets: (1) nuclei, debris; (2) peroxisomes, mitochondria, melanin
granules; (3) lysosomes, melanin granules; (4) microsomes;
(supernatant) soluble proteins, etc. B: Distribution of radio-
activity in fractions obtained from density-gradient centrifuga-
tion (30-70% Nycodenz gradient) of pellets 2 and 3 obtained
from differential centrifugation; melanosomes sediment in
fractions 2-3. C: Subcellular composition of fraction 2 showing
melanin granules.
rates of tissue and renal clearance of metabolites, which
compete with melanoma uptake by reducing intravas-
cular concentrations.
However, in Figure 3 we find that compound 9, which
lacks substituents at C2 and C5, represents a unique
exception to the pattern of behavior exhibited by the

Radioiodinated N-(2-Diethylaminoethyl)benzamides
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21 3919
123I]IMBA5 or [123I]BZA7 is based on the relatively rapid
IMBA and [131I]BZA with specific activities ranging from
90 GBq/µmol (no carrier added) to 0.02 MBq/µmol
(carrier added). As shown in Table 4 1- and 6-h tumor
levels of [131I]IMBA remained almost unchanged (non-
saturable) while the specificity (M/NTT) was reduced
to some extent, not going below the values of BZA (see
below). For [131I]BZA a significant increase in melanoma
uptake with decreasing specific activity and with time
was observed. An increase at 6 h postinjection has also
been detected with other radioiodinated benzamides.5,6
In contrast to [131I]IMBA the selectivity (M/NTT) of [131I]-
BZA increased with decreasing specific activity. These
positive results at very low specific activities are con-
sistent with the clinical studies mentioned above.
It is also interesting to note that radioiodinated
1-(iodopropen-2-yl)-4-[(4-cyanophenoxy)methyl]piperi-
dine shows both high B16 melanoma uptake and high
affinity for σ-receptors. However, melanoma uptake was
not suppressed by DuP 734, a σ-receptor blocking agent,
but was in fact doubled.14 Thus, all of the results
discussed in this section indicate that binding to σ-re-
ceptors is not an important mechanism for melanoma
uptake of the compounds investigated here.
[
clearance from nontarget tissue and the resulting high
contrast for melanoma imaging within a few hours after
injection. Of the new compounds presented here, 6 offers
the highest potential for clinical use; it features highest
melanoma uptake, slowest tumor washout, metabolic
stability, and a moderate rate of clearance from most
normal tissues (slow clearance from liver). This allows
scintigraphic imaging with both high sensitivity (count
rates) and high contrast to be achieved at 24 h postin-
jection, for example. In addition, at the uptake levels
attained by 6, radionuclide therapy with this agent may
be feasible.
Ma ter ia ls a n d Meth od s
All commercially available chemicals were used without
further purification. These include the following precursors:
4-amino-5-bromo-N-(2-diethylaminoethyl)-2-methoxybenza-
mide (1) (bromopride), 4-acetamido-N-(2-diethylaminoethyl)-
benzamide (8) (N-acetylprocainamide), 3-methoxybenzoyl chlo-
ride (10) and 2-methoxybenzoyl chloride (13) (all from Sigma
Aldrich, Deisenhofen, Germany); 4-amino-2-methoxybenzoate
(3) prepared from 4-amino-2-hydroxybenzoic acid17 (Lancaster;
Mu¨hlheim, Germany). 131I- with a specific activity of 90 GBq/
µmol in 0.02 M NaOH was obtained from Amersham Int.
(England). IMBA (N-(2-diethylaminoethyl)-3-iodo-4-methoxy-
benzamide), BZA (N-(2-diethylaminoethyl)-4-iodobenzamide),
Role of Mela n in . The intracellular distribution of
[
131I]IMBA in the mouse model was investigated by
analyzing homogenized B16 melanoma tissue by dif-
ferential and density-gradient centrifugation tech-
niques. In fractions obtained from differential centrifu-
gation (Figure 2A) radioactivity peaked in pellet 3
(lysosomes, melanin granules). Following density-gradi-
ent centrifugation (Figure 2B) >60% of radioactivity was
associated with fractions containing melanin granules,
and an electron microscopy image (Figure 2C) showed
dark-colored melanin granules in fraction 2 (Figure 2B).
These studies demonstrated that there was no signifi-
cant amount of membrane-bound radioactivity, and
analysis of tissue extracts by HPLC indicated only the
parent compound [131I]IMBA, and not its metabolites,
was responsible for the intracellular distribution ob-
served.
[
131I]BZA and [131I]IMBA were prepared as described previ-
ously.5 [3H]-2-Dimethylallyl-5,9-dimethyl-2′-hydroxybenzomor-
phan ([3H]-(+)-pentazocine; 1036 GBq/mmol) and [3H]-1,3-di-
o-tolylguanidine ([3H]DTG; 1147 GBq/mmol) were obtained
from NEN (Zaventem, Belgium). Melting points (mp) were
determined in open capillary tubes using an electrothermal
apparatus and are uncorrected. HPLC was carried out on
LATEK systems equipped with variable UV (Latek, Heidel-
berg) and γ-radiation detectors (Berthold, Wildbad). The
reverse-phase HPLC columns (270 × 4 mm) used were
Nucleosil C18 5 µm (Macherey & Nagel, Du¨ren). The solvents
used throughout for HPLC measurements consisted of MeOH
and 0.9% TRIS buffer adjusted to pH 2.6 with H3PO4. The
detectors were equipped with a C-R5A dual-channel integrator
(Shimadzu, Duisburg).
Confirmation of the structures of the synthesized benz-
amides was obtained with the following analytical methods:
MALDI-TOF mass spectrometry (MALDI I and III TOF
spectrometer from Kratos/Shimadzu, Duisburg); elemental
analysis (Microanalytical Laboratory, Chemistry Department,
University of Heidelberg); 1H and 13C NMR spectroscopy at
7.0 T (Bruker AM-300 spectrometer, Rheinstetten, Germany).
Signal assignments for the one-dimensional spectra could be
made directly using chemical shift predictions based on (a) the
empirical increment rules incorporated in the SpecTool soft-
ware (version 2.1, Chemical Concepts, Weinheim, Germany)
and (b) the experimental database in Bruker’s SpecEdit
software.
Additional support for the binding of radioiodinated
benzamides to melanin was recently obtained by others
using B16/C3 melanoma cells.15 Here, cellular uptake
of [131I]BZA was clearly dependent on the amount of
melanin present in the cells.
Su m m a r y
The results of the experiments presented here provide
strong evidence that the uptake of radioiodinated benz-
amides in B16 melanoma is not mediated by a specific
uptake mechanism involving σ-receptors, for example,
but is mainly controlled by the available vascular
concentration of agent and diffusion or efficient trans-
port into the melanoma cells. The binding sites are
melanosomes, and the binding capacity appears to be
nonsaturable, at least up to an injected dose of ca. 2
µmol/mouse. Similar characteristics were observed for
other weak bases which concentrate in melanin-con-
taining cells.16 The SARs investigated here involve
primarily metabolic stability and tissue and renal
clearances which determine the local bioavailability of
the diffusable benzamides in melanoma, their absolute
uptake in tumor, and their differential clearance from
tumor and nontarget tissues. The diagnostic value of
ESI-MS analyses of the urine metabolites were performed
using a triple-quadrupole instrument (TSQ 7000 from Finni-
gan, San J ose, CA) equipped with a nanoelectrospray source
(EMBL, Heidelberg). Samples were sprayed at (600 V from
gold-coated glass capillaries prepared in-house using a micro-
capillary puller (Sutter Instruments, type 87-B).18 Argon was
used as a collision gas at a nominal pressure of 2 mTorr.
4-Am in o-5-b r om o-N-(2-d iet h yla m in oet h yl)-3-iod o-2-
m eth oxyben za m id e (2). 218 mg (0. 0.633 mmol) of 4-amino-
5-bromo-N-(2-diethylaminoethyl)-2-methoxybenzamide (1) was
dissolved in 6 mL of TFA and reacted at ambient temperature
with a 1.2-fold molar amount of Tl(TFA)3. After 30 min 98.2
mg of NaI (0.655 mmol) in H2O was added. The mixture was
stirred at ambient temperature until the solid NaI dissolved
and reacted with the thallated intermediate product. After 5
h TFA was evaporated and the residue dissolved in 6 mL of
H2O. The pH was raised above 12 with 10 M NaOH, and the

3920 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21
Ta ble 5. 1H NMR Data for Benzamides 2-15 as Free Bases in CDCl3
Eisenhut et al.
a
1H NMR chemical shifts in ppm relative to TMS (multiplicity)
positionb
C2
2, 30 °C
6, 30 °C
7, 27 °C
8, 23 °C
9,c 30 °C
12, 27 °C
15, 27 °C
7.712 (d)
7.608 (d)
8.281 (d)
C3
8.224 (s, br)
6.274 (s)
7.694 (d)
6.728 (d)
C4
C5
C6
N8
C9
C10
C12
C13
4-NH2
4-NHAc (NH)
(CH3)
OMe
6.681 (dd)
7.688 (dd)
7.608 (d)
7.712 (d)
6.977 (t, br)
3.465 (dt)
2.639 (t)
2.559 (q)
1.033 (t)
8.300 (d, br)e
7.640 (dd)
6.881 (t, br)
3.446 (dt)
2.632 (t)
8.216 (s)
8.579 (s)
8.311 (t, br)
3.496 (dt)
2.620 (t)
2.564 (q)
1.041 (t)
8.455 (s)
8.137 (t)
3.477 (dt)
2.607 (t)
2.556 (q)
1.034 (t)
4.424 (s)
6.985 (d)
6.665 (br)
3.530 (dt)
2.721 (t)
2.625 (q)
1.061 (t)
8.476 (d)
8.301 (t, br)
3.507 (dt)
2.633 (t)
2.575 (q)
1.045 (t)
7.960 (t, br)
3.510 (dt)
2.641 (t)
2.582 (q)
1.034 (t)
2.557 (q)
1.031 (t)
4.994 (s, br)
7.623 (s, br)
2.270 (s)
3.960 (s)
8.308 (s, br)
2.186 (s)
7.578 (s, br)
2.254 (s)
3.803 (s)
3.868 (s)
3.794 (s)
3.924 (s)
J couplingsd
8.7 (2, 3)
8.59 (5, 6)
2.04 (2, 6)
8.75 (3, 4)
3.0 (4, 6)
8.68 (3, 4)
2.41 (4, 6)
a
b
Free bases measured at 300 MHz at the temperatures given. See Chart 1 for numbering. c For HCl salt at 45 °C, N11 is protonated
and positively charged, giving a broad resonance at 11.45 ppm and the following characteristic shifts for the side chain: 8.955 (t, N8);
d
3.866 (dt, C9); 3.296 (dt, C10); 3.204 (dq, C12); 1.409 (t, C13). In Hz with coupling partners in parentheses; couplings in the amide side
chain are typically: 5.0 (8,9), 6.2 (9,10), 7.13 (12,13) in the free base and 4.4 (8,9), 6.0 (9,10), 4.5 (10,11), 4.5 (11,12), 7.27 (12,13) in the
salt form. e Broad due to hindered rotation of the NHAc group.
solution was extracted twice with Et2O. The organic phase was
dried over Na2SO4 and evaporated. The residue was dissolved
in methanolic HCl and evaporated. The oily residue crytallized
slowly. Recrystallization from CH2Cl2/AcOEt yielded 249 mg
of benzamide 2 (HCl salt) (78%). Mp: 128 °C. MALDI-MS: m/z
469.9, 471.9 (M + H)+. Anal. (C14H22BrClIN3O2) C, H, N.
4-Acet a m id o-N-(2-d iet h yla m in oet h yl)-5-iod o-2-m et h -
oxyben za m id e (6). 1.4 g (7.73 mmol) of methyl 4-amino-2-
methoxybenzoate (3) (mp 157 °C; lit. mp 155-156 °C17) and
4.5 g (37.8 mmol) of 2-diethylaminoethylamine were heated
at 90 °C for 12 h. Excess amine was evaporated under reduced
pressure. Addition of 5 mL of 1 M NaOH (H2O) and extraction
with CHCl3 afforded a slightly yellow oil after evaporation.
Compound 4 was precipitated as the HCl salt in Et2O (1.2 g,
54%). Mp: 164 °C (lit. mp 153 °C 19). MALDI-MS: m/z 266.4
(M + H)+. 400 mg (1.33 mmol) of compound 4 was transferred
to the free base and dissolved in 10 mL of CH3CN. Acetylation
was performed at ambient temperature by the dropwise
addition of 115 mg (1.46 mmol) of acetyl chloride. After 1 h
the solvent was evaporated under reduced pressure. Addition
of 3 mL of 1 M NaOH (H2O) and extraction with CHCl3
afforded compound 5 as a white, HPLC-pure solid (410 mg,
ca. 100%). MALDI-MS: m/z 308.2 (M + H)+. 350 mg (1.14
mmol) of compound 5 was dissolved in 20 mL of TFA and
reacted with Tl(TFA)3/I- as described for 2. The benzamide 6
was precipitated as the free base and recrystallized from
AcOEt/hexane (1/4) (245 mg, 50%). Mp: 120 °C. MALDI-MS:
m/z 434.3 (M + H)+. Anal. (C16H24IN3O3) C, H, N.
4-Am in o-N-(2-d iet h yla m in oet h yl)-5-iod o-2-m et h oxy-
ben za m id e (7). 250 mg (0.828 mmol) of 4-amino-N-(2-diethyl-
aminoethyl)-2-methoxybenzamide (HCl salt of 4) was dissolved
in 17 mL of 1 N HCl (H2O) and mixed with 3.3 mL of 0.5 M
KIO3 (H2O). To this solution was added 1.65 mL of 1 M NaI
(H2O) in 100-µL portions within 30 min. The resulting mixture
was treated with 1.65 mL of 1 M Na2S2O5 (H2O) for 10 min
and extracted with Et2O after raising the pH above 12. After
evaporation 7 was obtained as a slightly yellow oil which
crystallized overnight. The solid was recrystallized from AcOEt
(254 mg, 78%). Mp: 158 °C. MALDI-MS: m/z 392.2 (M + H)+.
Anal. (C14H22IN3O2) C, H, N.
(10) was dissolved in 15 mL of CH3CN and reacted with 293
mg (2.53 mmol) of 2-diethylaminoethylamine at ambient
temperature for 2 h. Solvent and excess amine were evapo-
rated under reduced pressure. Addition of 5 mL of 2 M NaOH
(H2O), extraction with CHCl3, and evaporation of volatile
components afforded compound 11 as a colorless oil (590 mg,
ca. 100%). MALDI-MS: m/z 251.2 (M + H)+. 500 mg (2.00
mmol) of the precursor 11 was dissolved in 40 mL of TFA and
reacted with Tl(TFA)3/I- as described for 2. The crude product
was precipitated as the HCl salt and recrystallized from AcOEt
(759 mg, 80%). Mp: 178 °C. MALDI-MS: m/z 377.3 (M + H)+.
Anal. (C14H22ClIN2O2) C, H, N.
N -(2-D ie t h y la m in o e t h y l)-5-io d o -2-m e t h o x y b e n z -
a m id e (15). 450 mg (2.64 mmol) of 2-methoxybenzoyl chloride
(13) and 452 mg (3.9 mmol) 2-diethylaminoethylamine were
reacted under comparable conditions, as described for 12, to
give compound 14 as a slightly yellow oil (665 mg, ca. 100%).
MALDI-MS: m/z 251.2 (M + H)+. 400 mg (1.6 mmol) of
compound 14 was dissolved in 32 mL of TFA and reacted with
Tl(TFA)3/I- as described for 2. The benzamide 15 was precipi-
tated as the HCl salt and recrystallized from AcOEt (428 mg,
65%). Mp: 194-6 °C dec. MALDI-MS: m/z 377.3 (M + H)+.
Anal. (C14H22ClIN2O2) C, H, N.
Ra d ioiod in a tion of 2, 6, 9, 12, a n d 15 (Tl(TF A)3/[131I]-
iod id e m et h od ). The radiolabeling procedure has been
described previously in detail.5,6 Briefly, 20 µL of a 10 mM TFA
solution of the corresponding precursor benzamide (1, 5, 8, 11
and 14) was mixed with 10 µL of 10 mM Tl(TFA)3 dissolved
in TFA. After 10 min the solution was mixed with previously
evaporated [131I]iodide. After 5 min TFA was evaporated, and
the resulting reaction mixture was separated on an analytical
RP18 HPLC column using the following solvent gradient:
0-100% methanol in Tris buffer over 30 min at 0.7 mL/min.
The collected eluate was evaporated to dryness, redissolved
in 0.9% NaCl, neutralized and sterile-filtered. The radiochemi-
cal yields were 70-80% (isolated), and the specific activities
of the no-carrier-added [131I]benzamide preparations ranged
between 70 and 90 GBq/µmol (calculated).
Ra d ioiod in a tion of 7 (KIO3/[131I]iod id e m eth od ). Radio-
labeling of 7 with 131I was performed essentially as described
earlier.20 Briefly, 20 µL of a 10 mM solution of the precursor
4 (1 N HCl) was mixed with 5 µL of 50 mM KIO3 (H2O) and
30 MBq [131I]iodide (0.02 M NaOH). After 10 min the reaction
mixture was chromatographed by analytical HPLC (RP18
column) using a 0-100% methanol in TRIS buffer gradient
over 30 min at 0.7 mL/min. The collected eluate was evapo-
rated to dryness, redissolved in 0.9% NaCl (H2O), neutralized
and sterile-filtered. The radiochemical yield was 90% (isolated),
4-Ace t a m id o-N -(2-d ie t h yla m in oe t h yl)-3-iod ob e n z-
a m id e (9). 500 mg (1.8 mmol) of 4-acetamido-N-(2-diethyl-
aminoethyl)benzamide (8) was dissolved in 36 mL TFA and
reacted with Tl(TFA)3/I- as described for 2. Benzamide 9 was
precipitated as the HCl salt and recrystallized from AcOEt
(594 mg, 75%). Mp: 197-200 °C. MALDI-MS: m/z 404.4 (M
+ H)+. Anal. (C15H23ClIN3O2) C, H, N.
N -(2-D ie t h y la m in o e t h y l)-2-io d o -5-m e t h o x y b e n z -
a m id e (12). 400 mg (2.3 mmol) of 3-methoxybenzoyl chloride

Radioiodinated N-(2-Diethylaminoethyl)benzamides
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21 3921
a
Ta ble 6. 13C NMR Data for Benzamides 2-15 as Free Bases in CDCl3
13C NMR chemical shifts in ppm relative to TMS (multiplicity)
6, 30 °C 7, 27 °C 8, 23 °C 12, 27 °C
9,c 30 °C
119.11 (s) 114.26 (s) 129.90 (s) 131.85 (s)
127.81 (d) 138.26 (d)
119.24 (d)
141.11 (s) 140.70 (s)
119.24 (d) 120.64 (d, br) e 159.79 (s)
positionb
C1
C2
C3
C4
C5
C6
C7
C9
C10
C12
C13
4-NHAc (CO)
(CH3)
OMe
2, 30 °C
117.48 (s)
15, 27 °C
124.02 (s)
143.13 (s)
80.77 (s)
158.18 (s)
80.94 (s)
147.96 (s)
102.73 (s)
135.52 (d)
163.43 (s)
37.64 (t)
51.53 (t)
46.58 (t)
11.66 (q)
158.58 (s)
103.95 (d)
141.44 (s)
77.81 (s)
141.55 (d)
163.35 (s)
37.64 (t)
51.48 (t)
46.76 (t)
12.16 (q)
168.62 (s)
25.17 (q)
56.07 (q)
159.28 (s)
96.79 (d)
150.30 (s)
73.49 (s)
142.64 (d)
164.04 (s)
37.52 (t)
51.66 (t)
47.78 (t)
12.15 (q)
157.44 (s)
113.67 (d)
140.94 (d)
83.56 (s)
140.69 (d)
163.64 (s)
37.63 (t)
51.41 (t)
46.72 (t)
12.05 (q)
89.31 (s, br) e 140.53 (d)
117.69 (d)
127.81 (d) 127.29 (d)
166.91 (s) 165.08 (s)
114.11 (d)
169.07 (s)
37.19 (t)
51.32 (t)
46.71 (t)
11.45 (q)
37.30 (t)
51.18 (t)
46.74 (t)
12.05 (q)
37.37 (t)
51.22 (t)
46.81 (t)
12.07 (q)
168.91 (s) 168.34 (s)
24.51 (q)
24.91 (q)
61.89 (q)
55.73 (q)
55.52 (q)
55.96 (q)
d
1J CH
168.9 (C6)
165.6 (C3)
168.9 (C6)
139.0 (C9)
156.4 (C3)
166.7 (C6)
167.4 (C2)
161.0 (C6)
ca. 169 (C5)
166.4 (C3)
162.0 (C4)
161.3 (C6)
160.5 (C3)
166.6 (C4)
168.8 (C6)
ca. 135 (C10)
132.0 (C12)
125.3 (C13)
128.7 (MeCO)
145.8 (OMe)
2J CH
1.9 (C1, H6)
4.0 (C5, H6)
3.9 (C7, NH)
1.9 (C1, H6)
4.0 (C2, H3)
4.0 (C4, H3)
6.3 (CO, Me)
2.5 (CO, 4-NH)
1.9 (C1, H6)
5.4 (C3, 4-NH2)
2.2 (C4, H3)
6.5 (C5, 4-NH2)
3.3 (C5, H6)
2.3 (C2, H3)
ca. 1 (C1, H6)
ca. 1 (C3, H4)
3.0 (C5, H6)
3.0 (C5, H4)
3J CH
4.0 (C2, 2-OMe)
7.3 (C2, H6)
6.0 (C3, 4-NH2)
8.5 (C4, H6)
6.2 (C5, 4-NH2)
2.8 (C7, H9)
6.1 (C1, H3)
9.7 (C2, H6)
5.2 (C3, NHAc)
9.6 (C4, H6)
5.5 (C1, H3)
9.2 (C4, H6)
8.0 (C5, H3)
9.0 (C4, H2)
9.0 (C4, H6)
7.1 (C2, H6)
8.0 (C1, H5)
6.9 (C6, H2)
7.3 (C1, H3)
9.0 (C2, H6)
10.6 (C2, H4)
5.5 (C4, H6)
5.4 (C6, H4)
5.5 (C1, H3)
7.6 (C4, H6)
10.7 (C5, H3)
6.8 (C6, H4)
4.8 (C7, H6)
4J CH
1.8 (C3, H6)
1.0 (C3, H6)
0.8 (C2, H5)
1.4 (C6, H3)
ca. 1 (C1, H4)
ca. 1 (C3, H6)
a
b
Measured at 75.47 MHz at the temperatures given. See Chart 1 for numbering. c For HCl salt (45 °C) protonation at N11 leads to
d
characteristic changes in side chain shifts: 35.27 (t, C9); 52.75 (t, C10); 48.29 (t, C12); 8.62 (q, C13). In Hz with coupling partners in
parentheses. e Broad due to hindered rotation of NHAc group.
and the specific activity of the 131I labeled 7 was 90 GBq/µmol
(calculated).
Oct a n ol/Bu ffer P a r t it ion Coefficien t s. The octanol/
buffer partition coefficients were determined with HPLC-
purified (no-carrier-added), radioiodinated benzamides. 10 µL
of the HPLC eluate was transferred into 1 mL of 0.067 M
phosphate buffer of pH 7.4. After the addition of 1 mL of
1-octanol the tubes were vortexed until equilibrium was
reached. The emulsions were centrifuged and separated.
Radioactivity in aliquots of each phase (100 µL) were measured
in a γ-counter, and the partition coefficients are expressed as
log P ) log(cpmoctanol/cpmbuffer).
σ1-Recep tor Bin d in g Assa ys. The in vitro σ1 binding
affinities of unlabeled benzamides were determined in a
competition assay using guinea pig brain membranes and the
high-affinity ligand [3H]-(+)-pentazocine.8 The membranes
were prepared from guinea pig brains (minus cerebella) as
previously described.8 Fifteen concentrations of unlabeled
benzamides ranging from 10-10 to 10-3 M and protein samples
(0.15 mg of membrane protein) were incubated with 5 nM [3H]-
(+)-pentazocine in a total volume of 0.25 mL of 50 mM Tris-
HCI, pH 8.0. Incubations were carried out for 120 min at 25
°C. The equilibrium binding constants (Kd and Bmax) for the
radioligand binding were determined using saturation binding
assays with each assay tube containing 0.6-40 nM concentra-
tions of [3H]-(+)-pentazocine and 0.15 mg of protein in the
same buffer volume as described above. Nonspecific binding
was determined in the presence of 10 µM haloperidol and used
for the correction of binding data. All assays were terminated
by dilution with 5 mL of ice-cold 10 mM Tris-HCI, pH 8.0,
and the solutions were filtered through glass fiber filters
(Whatman GF/B; presoaked in 0.5% polyethylenimine for 30
min at 25 °C). Filters were then washed twice with 5 mL of
Biod istr ibu tion Stu d ies. The animal experiments were
performed in compliance with the German Animal Protection
Laws (Permission 37-9185.81/79/95, Reg.-Pra¨sidium, Karlsru-
he). Biodistribution time-course studies were performed with
C57Bl/6 mice bearing the B16 murine melanoma. The tumor
cells (DKFZ, Heidelberg) were washed with PBS and trans-
planted subcutaneously into the left flank by injecting 0.5 ×
106 cells (0.1 mL). After 10-14 days each animal (20-29 g)
was given an iv tail-vein injection of one of the 131I-labeled
benzamide derivatives (2-3 MBq dose). The injected volume
was determined by weighing the syringe before and after
injection. At specific times postinjection, the animals were
weighed, sacrificed by cervical dislocation and dissected.
Organs or tissues were blotted dry, when appropriate, and
weighed. Radioactivities in tissues and in standards of the
injected dose were determined with a γ-counter. The results
(Table 1) are expressed as % ID/g tissue. In addition, a
selectivity factor representing the mean melanoma/nontarget
tissue ratio, M/NTT (averaged over all tissues examined), was
calculated.
HP LC a n d ESI-MS/MS of Ur in a r y Meta bolites. 15 MBq
of 131I-labeled IMBA, BZA, 2, 6 or 9 was intravenously injected
into NMRI mice (29-33 g). After 4 h the excreted urine was
collected and combined with the urine aspirated from the
bladder. Radioactivity in the urine was counted, and the
samples were chromatographed by RP-HPLC. In separate
experiments a 1-mg sample of IMBA or BZA was injected into
animals. The 4-h urine samples were desalted and used for
negative-ion ESI-MS/MS.

3922 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21
Eisenhut et al.
(2) J ohn, C. S.; Bowen, W. D.; Saga, T.; Kinuya, S.; Vilner, B. J .;
Baumgold, J .; Paik, C. H.; Reba, R. C.; Neuman, R. D.; Varma,
V. M.; McAfee J . G. A malignant melanoma imaging agent.
Synthesis, characterization, in vitro binding and biodistribution
of 4-[125I]iodo-(2-piperidinylaminoethyl)benzamide. J . Nucl. Med.
1993, 34, 2169-2175.
(3) Moreau, M. F.; Michelot, J .; Papon, J . M.; Bayle, M.; Labarre,
P. G.; Madelmont, J . C.; Parry, D. F.; Boire, J . Y. P.; Moins, N.;
Seguin, H.; Veyre, A. J .; Mauclaire L. Synthesis, radiolabeling,
and preliminary evaluation in mice of some (N-diethylaminoet-
hyl)-4-iodobenzamide derivatives as melanoma imaging agents.
Nucl. Med. Biol. 1995, 22, 737-747.
(4) Brandau, W.; Niehoff, T.; Pulawski, P.; J onas, M.; Dutschka, K.;
Sciuk, J .; Coenen, H. H.; Schober, O. Structure distribution
relationship of I-123 iodobenzamides as tracers for the detection
of melanotic melanoma. J . Nucl. Med. 1996, 37, 1865-1871.
(5) Nicholl, C.; Mohammed, A.; Hull, W. E.; Bubeck, B.; Eisenhut,
M. Pharmacokinetics of 123I-IMBA for melanoma imaging. J .
Nucl. Med. 1997, 38, 127-133.
(6) Mohammed, A.; Nicholl, C.; Titsch U.; Eisenhut, M. Radioiodi-
nated N-(alkylaminoalkyl)-substituted 4-methoxy-, 4-hydroxy-,
and 4-aminobenzamides: biological investigations for the im-
provement of melanoma-imaging agents. Nucl. Med. Biol. 1997,
24, 373-380.
(7) Michelot, J . M.; Moreau, M. F. C.; Veyre, A. J .; Bonafous, J . F.;
Bacin, F. J .; Madelmont, J . C.; Bussiere, F.; Souteyrand, P. A.;
Mauclaire, L. P.; Chossat, F. M.; Papon, J . M.; Labarre, P. G.;
Kauffmann, P.; Plagne, R. J . Phase II scintigraphic clinical trial
of malignant melanoma with iodine-123-N-(2-diethylaminoethyl-
4-iodobenzamide). J . Nucl. Med. 1993, 34, 1260-1266.
(8) Bowen, W. D.; de Costa, B. R.; Hellewell, S. B.; Walker J . M.;
Rice, K. C. [3H]-(+)-Pentazocine: a potent and highly selective
benzomorphan-based probe for sigma-1 receptors. Mol. Neurop-
harmacol. 1993, 3, 117-126.
(9) Hellewell S. B.; Bruce A.; Feinstein G.; Orringer J .; Williams
W.; Bowen W. D. Rat liver and kidney contain high densities of
σ1 and σ2 receptors: characterization by ligand binding and
photoaffinity labeling. Eur. J . Pharmacol.-Mol. Pharmacol. Sect.
1994, 268, 9-18.
(10) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition
constant (KI) and the concentration of inhibitor which causes
50% inhibition (IC50) of an enzymatic reaction. Biochem. Phar-
macol. 1973, 22, 3099-3108.
(11) Menon, I. A.; Haberman, H. F. Isolation of melanin granules.
Methods Enzymol. 1974, 31 (Part A), 389-394.
(12) J ohn, C. S.; Bowen, W. D.; Fisher, S. J .; Lim, B. B.; Geyer, B.
C.; Vilner, B. J .; Wahl, R. L. Synthesis, in vitro pharmacologic
characterization, and preclinical evaluation of N-[2-(1′-piperidi-
nyl)ethyl]-3-[125I]iodo-4-methoxybenzamide (P[125I]MBA) for
imaging breast cancer. Nucl. Med. Biol. 1999, 26, 377-382.
(13) Vilner, B. J .; J ohn, C. S.; Bowen, W. D. Sigma-1 and sigma-2
receptors are expressed in a wide variety of human and rodent
tumor cell lines. Cancer Res. 1995, 55, 408-413.
(14) Waterhouse, R. N.; Chapman, J .; Izard, B.; Donald, A.; Belbin,
K.; O’Brian, J . C.; Collier, T. L. Examination of four 123I-labeled
piperidine-based sigma receptor ligands as potential melanoma
imaging agents: Initial studies in mouse tumor models. Nucl.
Med. Biol. 1997, 24, 587-593.
(15) Dittmann, H.; Coenen, H. H.; Zolzer F.; Dutschka K.; Brandau
W.; Streffer C. In vitro studies on the cellular uptake of
melanoma imaging aminoalkyl-iodobenzamide derivatives (ABA).
Nucl. Med. Biol. 1999, 26, 51-56.
(16) El Gamoussi, R.; Threadgill, M. D.; Prade, M.; Stratford, I. J .;
Guichard, M. Relationship between the melanin content of a
human melanoma cell line and its radiosensitivity and uptake
of pimonidazole. Cancer Chemother. Pharmacol. 1993, 31, 277-
282.
(17) Murakami, M.; Inukai, N.; Koda A.; Nakano K. An improved
synthesis of metoclopramide. Chem. Pharm. Bull. 1971, 19,
1696-1699.
(18) Wilm, M.; Mann, M. Analytical properties of the nanoelectro-
spray ion-source. Anal. Chem. 1996, 68, 1-8.
(19) Thominet, M. L. Compositions for treatment of emesis and
behavior disturbances. U.S. Patent 3,219,528, 1965; Chem. Abstr.
1966, 64, 12609h.
(20) Bubeck, B.; Eisenhut, M.; Heimke, U.; zum Winkel, K. Melanoma
affine radiopharmaceuticals. I. A comparative study of 131I-
labeled quinoline and tyrosine derivatives. Eur. J . Nucl. Med.
1981, 6, 227-233.
ice-cold 10 mM Tris-HCI, pH 8.0, and counted in Hionic-Fluor
cocktail (Packard, Groningen, Netherlands). The corresponding
IC50, Kd and Bmax values were determined using SigmaPlot
software (SigmaPlot 4.0, SPSS Inc., Chicago, IL) and were used
for the calculation of the apparent Ki values.10
σ2-Recep tor Bin d in g Assa ys. Rat liver membranes were
prepared from the livers of male Sprague-Dawley rats as
previously described.9 The σ2-receptors were labeled as previ-
ously described9 using [3H]DTG as ligand in the presence of 1
µM dextrallorphan to mask σ1-receptors. Competition assays
were performed with 15 concentrations of unlabeled benza-
mides ranging from 10-10 to 10-3 M and protein samples (0.10
mg of membrane protein) in 50 mM Tris-HCI, pH 8.0, for 120
min at 25 °C in a volume of 0.25 mL. The equilibrium binding
constants (Kd and Bmax) for the radioligand binding were
determined as described above using saturation binding assays
with each assay tube containing 0.6-80 nM concentrations of
[3H]-(+)-pentazocine and 0.1 mg of protein in the buffer volume
described above. Nonspecific binding was determined in the
presence of 10 µM haloperidol. All other manipulations and
data analysis were performed as described above for the σ1-
receptor assay.
Sa tu r a tion Assa ys for B16 Mem br a n es. B16 melanoma
membranes were prepared from subcutaneously transplanted
B16 cells (2 × 106 cells/mouse) grown for 10 days in C57Bl/6
mice. The preparation conditions of the membranes were
adapted to the procedures described earlier.8,9 Binding studies
for σ1- and σ2-receptors in B16 melanoma membranes were
carried out under the conditions described above. Twelve
concentrations of the radioligands [3H]-(+)-pentazocine and
[3H]DTG (in the presence of 1 µM dextrallorphan) in the range
1-80 nM were applied. Nonspecific binding was determined
in the presence of 10 µM haloperidol. All other manipulations
and data analysis were performed as described above for the
σ1-receptor assay.
Den sity-Gr a d ien t Ultr a cen tr ifu ga tion of B16 Hom o-
gen a te a n d Ultr a str u ctu r a l An a lysis. One hour after
injection of radiolabeled benzamide, the B16 tumor was
dissected and homogenized at 4 °C using a glass-Teflon Potter
homogenizer rotating at 800 min-1. Differential centrifugation
of the homogenated tissue produced four pellets which con-
sisted of crudely separated cell organelles, as outlined in
Figure 2A. Subcellular fractionation of the B16 homogenate
was carried out essentially as described previously for the
separation of a rat liver light mitochondrial fraction.21 Briefly,
pellets 2 and 3 obtained from differential centrifugation were
layered on top of a density gradient made up of 30-70% (w/v)
Nycodenz overlayed with a reversed sucrose gradient of 8-0%
(w/v) in 10 mM glycylglycine (pH 7.4), 1 mM EDTA and 0.1%
ethanol. Centrifugation was performed in a vertical rotor at
30000g for 90 min at 4 °C. Fractions were recovered from
bottom to top and counted in
a γ-counter. For electron
microscopy organelle fractions were diluted, pelleted, and fixed
with 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4).
Subsequently the fixed organelle pellets were embedded in
Epon as decribed previously.22 Ultrathin sections were viewed
in an EM 10 electron microscope (Zeiss, Oberkochen, Ger-
many).
Ack n ow led gm en t. A postdoctoral grant for A.M.
and a research grant, both received from the Deutsche
Forschungsgemeinschaft (DFG), are gratefully acknowl-
edged. We also thank Dr. K. C. Rice (NIH, Bethesda,
MD) as well Dr. E.-M. Gutknecht, P. Weber, and Dr. S.
Hauptmann (Roche-Pharma, Basel, Switzerland) for
generous gifts of dextrallorphan.
(21) Hartl, F. U.; J ust, W. W. Integral membrane polypeptides of rat
liver peroxysomes: Topology and response to different metabolic
states. Arch. Biochem. Biophys. 1987, 255, 109-119.
(22) Gorgas, K. Serial section analysis of mouse hepatic peroxysomes.
Anat. Embryol. 1985, 172, 21-32.
Refer en ces
(1) Michelot, J . M.; Moreau, M. F. C.; Labarre, P. G.; Madelmont,
J . C.; Veyre, A. J .; Papon, J . M.; Parry, D. F.; Bonafous, J . F.;
Boire, J . Y. P.; Desplanches, G. G.; Bertrand, S. J .; Meyniel, G.
Synthesis and evaluation of new 125I radiopharmaceuticals as
potential tracers for malignant melanomas. J . Nucl. Med. 1991,
32, 1573-1580.
J M991079P