Meta-iodobenzylguanidine derivatives containing a second guanidine moiety

Article abstract of DOI:10.1016/j.bmc.2004.01.026

Radioiodinated meta-iodobenzylguanidine (MIBG) is used in the diagnosis and therapy of various neuroendocrine tumors. To investigate whether an additional guanidine function in the structure of MIBG will yield analogues that may potentially enhance tumor-to-target ratios, two derivatives - one with a guanidine moiety and another with a guanidinomethyl group at the 4-position of MIBG - were prepared. In the absence of any uptake-1 inhibiting conditions, the uptake of 4-guanidinomethyl-3-[¹³¹I]iodobenzylguanidine ([ ¹³¹I]GMIBG) by SK-N-SH cells in vitro was 1.7±0.1% of input counts, compared to a value of 40.3±1.4% for [¹²⁵I[MIBG suggesting that guanidinomethyl group at the 4-position negated the biological properties of MIBG. On the other hand, 4-guanidino-3-[¹³¹I] iodobenzylguanidine ([¹³¹I]GIBG) had an uptake (5.6±0.3%) that was 12-13% that of [¹²⁵I]MIBG (46.1±2.7%), and the ratio of uptake by control over DMI-treated (nonspecific) cultures was higher for [¹³¹I]GIBG (20.9±0.3) than [¹²⁵I]MIBG itself (15.0±2.7). The exocytosis of [¹³¹I]GIBG and [ ¹²⁵I]MIBG from SK-N-SH cells was similar. The uptake of [ ¹³¹I]GIBG in the mouse target tissues, heart and adrenals, as well as in a number of other tissues was about half that of [¹²⁵I]MIBG. These results suggest that substitution of guanidine functions, especially a guanidinomethyl group, in MIBG structure may not be advantageous.

Full text of DOI:10.1016/j.bmc.2004.01.026

Bioorganic & Medicinal Chemistry 12 (2004) 1649–1656
Meta-iodobenzylguanidine derivatives containing a second
guanidine moiety
Ganesan Vaidyanathan,* Sriram Shankar, Donna J. Affleck, Kevin Alston,
Joseph Norman, Philip Welsh, Holly LeGrand and Michael R. Zalutsky
Department of Radiology, Duke University Medical Center, Durham, NC 27710, USA
Received 15 July 2003; revised 15 July 2003; accepted 17 January 2004
Abstract—Radioiodinated meta-iodobenzylguanidine (MIBG) is used in the diagnosis and therapy of various neuroendocrine
tumors. To investigate whether an additional guanidine function in the structure of MIBG will yield analogues that may potentially
enhance tumor-to-target ratios, two derivatives—one with a guanidine moiety and another with a guanidinomethyl group at the 4-
position of MIBG—were prepared. In the absence of any uptake-1 inhibiting conditions, the uptake of 4-guanidinomethyl-3-
[
131I]iodobenzylguanidine ([131I]GMIBG) by SK-N-SHcells in vitro was 1.7 ꢀ0.1% of input counts, compared to a value of
40.3ꢀ1.4% for [125I[MIBG suggesting that guanidinomethyl group at the 4-position negated the biological properties of MIBG. On
the other hand, 4-guanidino-3-[131I]iodobenzylguanidine ([131I]GIBG) had an uptake (5.6ꢀ0.3%) that was 12–13% that of
[
125I]MIBG (46.1ꢀ2.7%), and the ratio of uptake by control over DMI-treated (nonspecific) cultures was higher for [131I]GIBG
(20.9ꢀ0.3) than [125I]MIBG itself (15.0ꢀ2.7). The exocytosis of [131I]GIBG and [125I]MIBG from SK-N-SHcells was similar. The
uptake of [131I]GIBG in the mouse target tissues, heart and adrenals, as well as in a number of other tissues was about half that of
[
125I]MIBG. These results suggest that substitution of guanidine functions, especially a guanidinomethyl group, in MIBG structure
may not be advantageous.
# 2004 Elsevier Ltd. All rights reserved.
1. Introduction
In our efforts to improve the bioavailability of MIBG,
we recently reported on the structure–activity studies of
several analogues of MIBG.6,7 It was concluded that
introduction of substituents at the 5-position of the ring
is detrimental to the affinity of MIBG binding to the
NET. For example, while a MIBG derivative with a 4-
amino substituent (AIBG) was taken up by the SK-N-
SHhuman neuroblastoma cells to a degree similar to
that for MIBG, the isomeric compound 5-amino-3-
iodobenzylguanidine exhibited only nonspecific binding.
There is not much latitude in altering the guanidine
moiety either.8 Thus to achieve our goal of developing
an optimized MIBG analogue, it was deemed necessary
to make alterations at the 4-position on the ring.
The guanethidine analogue meta-iodobenzylguanidine
(MIBG) is avidly taken up by tissues rich in sympathetic
neurons and neuroendocrine neoplasms via the nor-
epinephrine transporter (NET).1 MIBG, radiolabeled
with 123/131I, is used in the scintigraphic evaluation of
normal and malignant tissues of neuroadrenergic ori-
gin.2,3 High specific activity [131I]MIBG is used in the
therapy of neuroblastoma, pheochromocytoma, and
other neuroendocrine tumors.4 Although radioiodinated
MIBG has been used extensively in diagnostic applica-
tions with success, results from therapy with [131I]MIBG
are not very encouraging, even when combined with
other therapeutic modalities.5 Therapeutic outcome
might be improved if higher tumor-to-background
ratios could be achieved. One possibility is to add polar
substituents to MIBG, creating analogues that may
have lower retention in normal tissues.
It has been reported that in the process of transporting
its substrates, the NET binds to the protonated form of
the amine substrate.9 Because a guanidine moiety will
remain almost exclusively in the protonated form at
physiological pH, we wished to investigate whether the
introduction of an additional guanidine group will have
an additive, if not synergistic, effect on the uptake of
MIBG, in addition to imparting hydrophilicity. It
should be pointed out that a bis-guanidine derivative,
Keywords: Meta-iodobenzylguanidine; Neuroblastoma; Uptake-1;
Norepinephrine transporter.
* Corresponding author. Tel.: +1-919-684-7811; fax: +1-919-684-
7122; e-mail: ganesan.v@duke.edu
0968-0896/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.01.026

1650
G. Vaidyanathan et al. / Bioorg. Med. Chem. 12 (2004) 1649–1656
formed by the substitution of a guanidine moiety on the
guanidine side chain of MIBG, was not a good sub-
strate for NET.8
In this study, we have prepared two MIBG derivatives
with an additional guanidine substituent and evaluated
their biological characteristics. While the derivative with
the guanidine group directly attached to the 4-position
of the ring showed some specificity to NET, the MIBG
analogue with a 4-guanidinomethyl substituent was not
a substrate for the transporter.
2. Results and discussion
2.1. Chemistry
In addition to the fact that NET transports NE and
other amines in their protonated form, it has been pro-
posed that ATP present in storage vesicles binds to the
positively charged catecholamines via its four negative
charges.10 We hypothesized that, in addition to impart-
ing added hydrophilicity, the presence of a second gua-
nidine moiety in MIBG might augment its tumor
binding and/or storage. The substitution of a guanidine
group on the guanidine side chain, resulting in a bis-
guanidine, has been shown to be unfavorable8 and our
earlier studies have shown that the presence of a group
or an atom other than hydrogen at the 5-position in the
ring is also counterproductive. Thus, our target was a
MIBG derivative with a guanidine function at the 4-
position. From a retrosynthetic point of view, it appeared
more facile to prepare a MIBG derivative with a guani-
dinomethyl substituent, instead of one in which the
guanidine group was attached directly to the ring. We
have investigated the effect of a methyl group itself at
the 4-position of MIBG and have shown that 3-iodo-4-
methylbenzylguanidine (MeIBG) retained the biological
properties of MIBG (results will be published else-
where). Because the introduction of a methyl group did
not adversely affect the biological properties of MIBG,
we chose to first prepare the guanidinomethyl derivative.
Scheme 1. Reagents: (a) N-Bromosuccinimide, 1,10-Azobis-(cyclohex-
anecarbonitrile), DCE; (b) N,N0-bis-tert-butyloxycarbonylguanidine,
NaH, DMF; (c) Trifluoroacetic acid/water/triisopropylsilane (95/2.5/
2.5), methylene chloride; (d) Hexamethyditin, Bis-(triphenylpho-
sphine) palladium dichloride, dioxane; (e) i-Radioiodine, NCS ii-c.
directly to the ring (Scheme 2). The starting material, 4-
amino-3-iodobenzylamine was prepared as reported.13
This diamine was converted to the protected bisguani-
dine target 5 using N,N0-bis(tert-butyloxycarbonyl)-
thiourea and Mukaiyama’s reagent in methylene
chloride following a method for guanidinylation repor-
ted by Yong et al.14 The target, 4-guanidinomethyl-3-
iodobenzylguanidine (GIBG), was obtained by the
deprotection of 5 with TFA cocktail. Several attempts
to prepare a tin precursor for use in the preparation of
6a were not successful. Neither palladium-catalyzed
stannylation of 5 nor electrophilic stannylation of the
anion generated from reaction of 5 with butyl lithium
Scheme 1 shows the synthesis of 4-guanidinomethyl-3-
iodobenzylguanidine (GMIBG). Benzylic bromination
of 2-iodo-p-xylene yielded 1,4-bis(bromomethyl)-2-iodo-
benzene,11 which was subjected to guanidinylation using
a procedure reported earlier for the guanidinylation of
benzyl bromides12 to obtain the protected guanidine
derivative 2. The target bisguanidine was generated by
the deprotection of 2 using a cocktail of TFA/water/
triisopropylsilane. Palladium-catalyzed stannylation of
2 yielded the tin precursor 3 and the radiolabeled
GMIBG was obtained from 3 by radioiodination and in
situ deprotection in about 50% radiochemical yield.
GMIBG did not exhibit any significant uptake in SK-N-
SHcells (see below). To investigate whether this might
be due to steric constraints at the active site of the
transporter as a result of the introduction of the rela-
tively bulkier guanidinomethyl group at the 4-position
of MIBG, we embarked on the synthesis of a MIBG
derivative wherein the guanidine moiety is attached
Scheme 2. Reagents: (a) N,N0-Bis-tert-butyloxycarbonylthiourea,
Triethylamine, Mukaiyama reagent; (b) Trifluoroacetic acid/water/
triisopropylsilane. (95/2.5/2.5); (c) N,N0-Bis-tert-butyloxycarbonyl-S-
methylisothiourea, Triethylamine, DMF.

G. Vaidyanathan et al. / Bioorg. Med. Chem. 12 (2004) 1649–1656
1651
yielded the expected N,N0-bis(tert-butyloxycarbonyl)-4-
[N,N0-bis(tert-butyloxycarbonylguanidino)]-3-(trimethyl-
stannyl)benzylguanidine. Assuming that this might be
due to steric hindrance from the bulky bis-Boc-guani-
dino group, N,N0-bis(tert-butyloxycarbonyl)-4-amino-3-
iodobenzylguanidine 7 was prepared first, in the hope
that this could be first stannylated and its 4-amino
group converted subsequently to a protected guanidine
moiety by the method of Yong et al.14 Compound 7
itself was prepared initially from 4-amino-3-iodobenzyl-
amine following the method of Yong et al.14 and using
DMF as the solvent. However, the yields were meager
and the chromatographic isolation was challenging.
Subsequently, 7 was prepared in excellent yields, by
following a literature procedure for guanidinylation,15
by the treatment of 4-amino-3-iodobenzylamine dihydro-
chloride with the commercially available N, N0-bis(tert-
butyloxycarbonyl)]-S-methylisothiourea. Attempts to
convert 7 to N,N0-bis(tert-butyloxycarbonyl)-4-amino-
3-(trimethylstannyl)benzylguanidine using both of the
approaches described above also were futile. Finally,
radioiodinated GIBG was prepared at a no-carrier-
added level in a three-step approach starting from N,N0-
bis(tert-butyloxycarbonyl)-4-aminobenzylguanidine 8,
which itself was prepared from the commercially avail-
able 4-aminobenzylamine following Yong et al.14 and
conducting the reaction in DMF (Scheme 3). Com-
pound 8 was first radioiodinated to yield [131I]7. For
reasons not clear, [131I]7 could not be converted to
[
a value considerably lower (p<0.05) than that for
131I]GMIBG was 1.7ꢀ0.1% of input counts (Fig. 1A)
125I]MIBG (40.3ꢀ1.4%). In fact, it was even less than
[
that for [125I]MIBG (2.1ꢀ0.1%) in cells pretreated with
DMI, a tricyclic antidepressant inhibitor of the uptake-1
pump. We speculate that this loss in NET specificity
may be due to either the charge of the guanidine moiety
and/or the steric bulk of the guanidinomethyl group.
To probe the role of the intervening methylene spacer of
the guanidinomethyl moiety in the loss of NET specifi-
city, we investigated the effect of a guanidine group
attached directly to the 4-position of MIBG. As shown
in Figure 1B, [131I]GIBG had an uptake of 5.6ꢀ0.3% of
input counts compared to 46.1ꢀ2.7% for [125I]MIBG
in SK-N-SHcells. Although the uptake of [ 131I]GIBG
was only about 12–13% of [125I]MIBG, it was encoura-
ging to note that the ratio of uptake by control over
DMI-treated (nonspecific uptake) cultures was higher
for [131I]GIBG (20.9ꢀ0.3) compared to that for
[
125I]MIBG (15.0ꢀ2.7; p<0.05). A value of 20 for this
ratio for MIBG in SK-N-SHcells has been reported
previously.17 In addition, other than ouabain, all
uptake-1 blocking conditions reduced the uptake of
[
[
131I]GIBG to a degree similar to that observed with
125I]MIBG. Thus, it appears that, although the abso-
lute uptake of [131I]GIBG is only about a sixth that of
[
131I]5 by treatment with N,N0-bis(tert-butyloxycarb-
onyl)thiourea and Mukaiyama’s reagent in methylene
chloride, even though this reaction was facile in the
conversion of unlabeled 7 to 5 (experimental details not
given). However, treatment of [131I]7 with cyanamide
and subsequent deprotection by in situ treatment with
TFA cocktail yielded [131I]GIBG in about 55% overall
radiochemical yield.
3. Biological evaluation
To investigate the effects of additional guanidine sub-
stitution on the biological behavior of these MIBG
analogues, they were subjected to a standard set of
assays utilized in our laboratories. In these assays, the
uptake of the new tracers and that of MIBG by SK-N-
SHhuman neuroblastoma cells in the presence and
absence of various uptake-116 blocking agents was
determined in a paired-label format. In the absence
of any uptake-1 inhibiting conditions, the uptake of
Figure 1. Figure 1. Uptake of [131I]GMIBG (A) and [131I]GIBG (B) by
SK-N-SHhuman neuroblastoma cells in vitro (open bars). Results
obtained for [125I]MIBG from respective paired-label assays are also
shown (filled bars). The effect of various uptake-1-blocking conditions
is also presented. Cells were incubated as described in the text in the
absence or presence of various uptake-1 inhibitors or at 4 ꢁC, and the
cell-associated radioactivity was determined as a percent of input
radioactivity.
Scheme 3. Reagents: (a) N,N0-Bis-tert-butyloxycarbonylthiourea,
Triethylamine, Mukaiyama reagent; (b) (i)Radioiodine, NCS (ii) cya-
namide; (iii) Trifluoroacetic acid/water/triisopropylsilane (95/2.5/2.5).

1652
G. Vaidyanathan et al. / Bioorg. Med. Chem. 12 (2004) 1649–1656
[
125I]MIBG, the specificity of its uptake via NET is
higher.
SHcells by a reuptake pathway. The observation that
the additional positive charge did not augment the cell
retention may point to the fact that the efflux of MIBG
and its analogues by neuroblastoma cells occurs most
likely via an active carrier-mediated mechanism as pro-
posed by Servidei et al.,22 rather than by a diffusion
pathway. Taken together, GIBG has biological char-
acteristics similar to MIBG albeit with a lower uptake
capacity and/or affinity.
Before undertaking studies in SK-N-SHxenograft
models, we wanted to evaluate [131I]GIBG further in
vitro to investigate whether it held advantages over
MIBG. To augment their intracellular retention, mono-
clonal antibodies which are internalized after binding to
antigens have been labeled with templates containing
positive charges.18,19 In these cases, the ligands were
localized in the lysosomal compartment, the pHof
which is distinctly acidic (ꢂ5). This ensured that the
ligands containing basic moieties will be predominantly
charged, causing them to remain trapped within the
lysosome. In neuroblastomas, MIBG is known to be
stored in the cytoplasm.20,21 Nevertheless, we were
interested in studying if the added guanidine group in
GIBG enhanced retention by SK-N-SHcells in vitro
compared to that seen for MIBG. It has been proposed
that the intracellular levels of MIBG are maintained by
SK-N-SHcells by a dynamic equilibrium, generated by
a rapid reuptake of drug diffusing from the cells.17 If
GIBG remains completely trapped within the tumor
cells due to the presence of an additional guanidine
group in its structure, then one would expect no change
over a period of time in the cell-associated amount of
radioactivity (after accounting for decay) that was taken
up initially. Since it is unlikely that DMI will have a
depleting effect on the drug stored intracellularly,17 the
cell-associated radioactivity should not be affected by
the presence of DMI. As shown in Figure 2, the cell-
associated radioactivity normalized to the initially
bound for the two tracers was comparable at different
time points indicating their similar rate of efflux.
Besides, DMI reduced the retained radioactivity for
both tracers at different time points to a similar degree
suggesting that, like MIBG, GIBG is retained in SK-N-
Finally, a tissue distribution of GIBG was performed in
normal mice in a paired-label format with MIBG to
investigate whether the normal tissue uptake of GIBG
was lower than that of MIBG, which may warrant its
further evaluation in xenograft models. The data pre-
sented in Table 1 shows that the uptake of [131I]GIBG
in the uptake-1-mediated target tissues, heart and adre-
nals, was about 50% of that of [125I]MIBG. Although
not dramatic, the ratio of uptake of [131I]GIBG to
[
125I]MIBG in these tissues increased with time. The
uptake of [131I]GIBG in these tissues, like that of
125I]MIBG, was specific as demonstrated by reduction
[
in the uptake by DMI pretreatment of the mice. The
reduction in heart uptake, about 26% of control
for both tracers, is less than that reported earlier for
[
125I]MIBG (40–50%).23 However, the current study
was performed in a paired-label format and therefore it
can be concluded that the degree of specificity of GIBG
uptake in mouse heart is same as that of MIBG. Com-
pared to [125I]MIBG, the uptake of [131I]GIBG was less
in most of the tissues, with the differences generally
being about a factor of two. A notable difference was in
blood where the uptake of [131I]GIBG was 2- to 4-fold
higher than that of [125I]MIBG. These normal mice tis-
sue distribution results in concert with the observation
that [131I]GIBG uptake in vitro was only one sixth of
MIBG uptake, suggest that achieving higher tumor-to-
tissue ratios with GIBG is not likely. This may indicate
that additional positive charge in MIBG, while not
enhancing, may actually decrease the transport of
MIBG derivatives by NET.
In summary, methods were designed to synthesize
MIBG derivatives containing guanidine and guanidino-
methyl substituents at the 4-position. While the intro-
duction of the 4-guanidinomethyl moiety diminished the
affinity of MIBG to, and its transport by, NET, direct
attachment of the guanidine function at the 4-position
yielded a compound that partly retained MIBG char-
acteristics. The results from an in vitro retention assay
and tissue distribution in normal mice suggest that
GIBG may not deserve further consideration as a
radiopharmaceutical for treating tumors that over-
express NET.
4. Experimental
4.1. General
Figure 2. Paired-label exocytosis of [131I]GIBG (*, without DMI; *,
with DMI) and [125I]MIBG (~, without DMI; ~, with DMI) as a
function of time. Cells were allowed to take up the tracers at 37 ꢁC for
2 h. The medium containing the tracers was replaced with fresh med-
ium without or with DMI, and the cell-associated radioactivity was
determined periodically.
All chemicals were purchased from Aldrich unless
[
125I]iodide and sodium
otherwise noted. Sodium
131I]iodide, with specific activities of 2200 Ci/mmol and
[

G. Vaidyanathan et al. / Bioorg. Med. Chem. 12 (2004) 1649–1656
Table 1. Paired-label tissue distribution of [125I]MIBG and [131I]GIBG in normal mice
1653
Tissue
Percent Injected Dose per Grama
1 h Control
1 h DMI
125I]MIBG 131I]GIBG
4 h
24 h
[
125I]MIBG
[
131I]GIBG
[
[
[
125I]MIBG
[
131I]GIBG
[
125I]MIBG
[
131I]GIBG
Liver
5.63ꢀ1.05
3.08ꢀ0.36
4.32ꢀ1.39
3.01ꢀ0.40
4.22ꢀ0.64
2.71ꢀ0.41
5.01ꢀ1.05
1.86ꢀ0.20
2.18ꢀ0.34
4.08ꢀ0.53
2.90ꢀ0.67b
0.13ꢀ0.05
0.84ꢀ0.07
1.56ꢀ0.42b
4.03ꢀ1.43
0.57ꢀ0.13b
0.08ꢀ0.01
6.30ꢀ0.52
3.15ꢀ0.36
4.04ꢀ0.38
8.24ꢀ0.87
2.35ꢀ0.21
2.39ꢀ0.66
5.28ꢀ0.36
2.46ꢀ0.29
0.09ꢀ0.04
1.64ꢀ0.26
0.69ꢀ0.10
5.72ꢀ1.59
0.77ꢀ0.21
0.08ꢀ0.01
3.11ꢀ0.46
5.13ꢀ0.65
2.67ꢀ0.19
3.71ꢀ0.50
2.05ꢀ0.22
1.54ꢀ0.37
4.21ꢀ0.31
1.92ꢀ0.15
0.07ꢀ0.01b
0.99ꢀ0.13
1.74ꢀ0.20
3.33ꢀ1.14
0.83ꢀ0.28b
0.05ꢀ0.00
2.86ꢀ0.28
2.58ꢀ0.25
2.61ꢀ0.28
8.35ꢀ1.30
1.43ꢀ0.22
1.54ꢀ0.66
3.28ꢀ0.52
3.80ꢀ0.73
0.21ꢀ0.08
1.16ꢀ0.66
0.33ꢀ0.04
6.43ꢀ1.61
0.57ꢀ0.08
0.06ꢀ0.01
1.53ꢀ0.12
3.45ꢀ0.57
2.07ꢀ0.61b
4.22ꢀ0.53
0.87ꢀ0.07
1.03ꢀ0.54
1.78ꢀ0.29
3.09ꢀ0.80
0.12ꢀ0.03
0.57ꢀ0.05
1.07ꢀ0.19
3.34ꢀ1.02
0.45ꢀ0.06
0.04ꢀ0.01b
0.70ꢀ0.07
1.04ꢀ0.17
0.82ꢀ0.10
3.39ꢀ0.41
0.52ꢀ0.05
1.36ꢀ0.18
1.33ꢀ0.07
1.44ꢀ0.19
0.22ꢀ0.05
0.36ꢀ0.07
0.08ꢀ0.01
5.35ꢀ1.45
0.15ꢀ0.02
0.01ꢀ0.00
0.50ꢀ0.05
1.27ꢀ0.16
0.59ꢀ0.14
2.13ꢀ0.21
0.38ꢀ0.02
0.98ꢀ0.14
0.89ꢀ0.03
1.03ꢀ0.14
0.06ꢀ0.02
0.21ꢀ0.04
0.26ꢀ0.02
3.20ꢀ0.78
0.12ꢀ0.02
0.01ꢀ0.00
Spleen
Lungs
Heart
Kidneys
Stomach
Sm. Int.
Lg. Int.
Thyroidc
Muscle
Blood
11.10ꢀ3.06
2.18ꢀ0.22
2.86ꢀ0.48
4.51ꢀ0.65
2.68ꢀ0.63
0.08ꢀ0.07
1.46ꢀ0.17
0.62ꢀ0.17
7.78ꢀ3.56
0.62ꢀ0.18
0.09ꢀ0.02
Adrenals
Bone
Brain
a MeanꢀSD (n=5).
b Except for these tissues, the difference in uptake between two preparations statistically significant (p<0.05).
c %ID/organ.
1200 Ci/mmol, respectively, were obtained from Perkin
Elmer Life Sciences (Boston, MA). Unlabeled MIBG
was obtained from Sigma or prepared in house using a
literature protocol.1 The dihydrochloride salt of 4-
amino-3-iodobenzyl amine was prepared as reported.13
on a Hewlett-Packard GC/MS/DS Model HP-5988A
instrument, or on a JEOL SX-102 high resolution mass
spectrometer.
4.2. Cells and culture conditions
Melting points were determined on a Haake Buchler
apparatus and were uncorrected. High pressure liquid
chromatography was performed using a Beckman Sys-
tem Gold HPLC equipped with a Model 126 program-
mable solvent module, a Model 168 diode array
detector, a Model 170 radioisotope detector, and a
Model 406 analogue interface module. For reversed-
phase chromatography, a Waters XTerra C18 column
(4.6ꢃ250 mm, 5 m) or a Waters Bondapak C18 column
(10 m, 3.9ꢃ300 mm) was used. Normal-phase HPLC
was performed using a 4.6ꢃ250 mm Partisil (10 m) silica
column (Alltech, Deerfield, IL). Analytical TLC was
performed on aluminum-backed sheets (Silica gel 60
F254), and normal-phase column chromatography was
performed using Silica gel 60, both obtained from EM
Science (Gibbstown, NJ). Column chromatographic
fractions were collected using a Gilson model 203 micro
fraction collector (Middleton, WI) or an ISCO Foxy
200 fraction collector (Lincoln, NE), and products
identified by TLC. In some cases, an ISCO UA-6 UV–
vis detector was placed between the column outlet and
the fraction collector to identify fractions. Preparative
thick layer chromatography was performed using 20 ꢃ
20 cm, 1000 m plates (Whatman, Clifton, NJ). Before
applying the sample, the plates were run in ethyl acetate
to clean the plates of any adsorbed impurities. Radio-
activity was measured using a dose calibrator (Capintec,
CRC-7R, USA) for higher amounts and an automated
gamma counter (LKB 1282, Wallac, Finland) for lower
count rates. Proton NMR spectra (300 MHz) were
obtained on a Varian Mercury 300 spectrometer or a
General Electric Midfield GN-300 spectrometer. Che-
mical shifts are reported in d units; solvent peaks were
referenced appropriately. Mass spectra were obtained
The human neuroblastoma cell line SK-N-SH(uptake-1
positive)24 was purchased from the American Type
Culture Collection (Rockville, MD). The incubation
medium (JRHBiosciences, Lenexa, KS) was made by
mixing 440 mL of RPMI 1640, 50 mL of Serum Plus, 5
mL of penicillin-G/streptomycin (5000 U of penicillin
and 5000 mg of streptomycin in 1 mL of 0.85% saline),
and 5 mL of glutamine (200 mM in saline). The cells
were grown at 37 ꢁC in a humidified incubator contain-
ing 5% CO2. Cell viability was evaluated prior to each
binding experiment by trypan blue dye,25 and was 95–
98% for all studies.
4.3. Syntheses of standards and precursors
4.3.1. 1,4-Bis-bromomethyl-2-iodobenzene (1). The title
compound was prepared by modification of a procedure
reported for benzylic bromination.26 The radical initia-
tor 1,10-aza-biscyclohexanenitrile27 (ABCN) was used in
lieu of the traditional aza-bisisobutyronitrile (AIBN).
Thus, a mixture of 2-iodo-p-xylene (233 mg, 1 mmol;
Lancaster, Windham, NH), N-bromosuccinimide (356
mg, 2 mmol), and ABCN (50 mg, 0.2 mmol) in dichloro-
ethane (10 mL) was heated at reflux in the presence of
an incandescent lamp. After 3–4 h, dichloroethane was
evaporated from the reaction mixture and the residue
was taken in ethyl acetate. The ethyl acetate solution
was washed with saturated sodium metabisulfite and
brine, dried with sodium sulfate, and evaporated. The
crude mixture was purified by silica gel chromatography
using 10% ethyl acetate in hexane to obtain 106 mg
(27%) of a white solid: mp 111–113 ꢁC (lit.11 109–
110.5 ꢁC). 1HNMR (CDCl 3) d 4.37 (s, 2H), 4.59 (s, 2H),
7.40 (m, 2H), 7.88 (d, 1H). MS (EI+) m/z: 389 (M+).

1654
G. Vaidyanathan et al. / Bioorg. Med. Chem. 12 (2004) 1649–1656
HRMS (EI+) calcd for C8H779Br81BrI (M+): 389.7937.
Found: 389.7939ꢀ0.0009 (n=4).
Mukaiyama’s reagent (218 mg, 0.85 mmol). The reac-
tion was allowed to stir at room temperature until
completion, as judged by TLC. The methylene chloride
was evaporated, and the residue partitioned between
ethyl acetate and water. The organic layer was dried with
anhydrous sodium sulfate, the solvent was removed on
a rotary evaporator, and the product purified by silica
4.3.2. 1,4-Bis-[(N,N0 -bis-tert-butyloxycarbonyl)guanidi-
nomethyl]-2-iodobenzene (2). To a slurry of sodium
hydride (60% dispersion in mineral oil; 88 mg, 2.2
mmol), in DMF (10 mL) was added 627 mg (2.4 mmol)
of N,N0-bis-tert-butyloxycarbonylguanidine.28 The mix-
ture became homogeneous upon stirring for 5 min. To
this solution was added 390 mg (1 mmol) of 1 in one
portion, and the reaction mixture was stirred for an
additional 10 min. The mixture was partitioned between
ethyl acetate and water, and the aqueous layer was
extracted twice with ethyl acetate. The combined ethyl
acetate solution was washed with brine, dried, and eva-
porated. The crude mixture was subjected to silica gel
chromatography using 15% ethyl acetate in hexane to
obtain 440 mg (59%) of a waxy solid: mp 161–162 ꢁC.
gel chromatography using 15% ethyl acetate in hexane
1
to obtain 67 mg (28%) of an oil: HNMR (CDCl ) d
3
1.5 (s, 18H), 1.55 (s, 18H), 4.62 (br s, 2H), 7.34 (dd, 1H),
7.76 (d, 1H), 8.22 (d, 1H). MS (FAB+) m/z: 733.2
(MH+). HRMS (FAB+) calcd for C29H46IN6O8
(MH+): 733.2422. Found: 733.2422ꢀ0.0011 (n=2).
4.3.6. 4-Guanidino-3-iodobenzylguanidine (6). N,N0-(bis-
tert-butyloxycarbonyl)-4-(N,N0-bis-tert-butyloxycarbonyl-
guanidino)-3-iodobenzylguanidine (5; 20.6 mg, 0.03 mmol)
was treated with a 95:2.5:2.5 (v/v/v) mixture of tri-
fluoroacetic acid:water:triisopropylsilane (685 mL). After
stirring at room temperature for 70 min, the solvents
were evaporated using an argon stream. The residual
solvents were removed by co-evaporating once with
methanol and thrice with ethyl acetate (1 mL each). The
1HNMR (CDCl ) d 1.28 (s, 9H), 1.42 (s, 9H), 1.46 (s,
3
9H), 1.50 (s, 9H), 5.08 (s, 2H), 5.16 (s, 2H), 6.86 (dd,
1H), 7.22 (dd, 1H), 7.78 (d, 1H), 9.42 (br m, 4H). MS
(FAB+) m/z: 747 (MH+), 647. HRMS (FAB+)
calcd for C30H48IN6O8 (MH+): 747.2578. Found:
747.2608ꢀ0.0007 (n=2).
residue was dried under high vacuum to obtain 6 as a film
1
in almost quantitative yield. HNMR (CD OD) d 4.46
3
4.3.3. 1,4-Bis-[(N, N0-bis-tert-butyloxycarbonyl)guanidi-
nomethyl]-2-(trimethylstannyl)benzene (3). A mixture of
2 (44 mg, 0.06 mmol), hexamethylditin (163 mg, 0.5
mmol), and bis(triphenylphosphine)palladium dichlor-
ide (16 mg, 0.02 mmol) in 2 mL of dioxane was refluxed
for 1 h. The mixture was filtered through a bed of
Celite, which was washed with ethyl acetate. The com-
bined filtrate was concentrated and chromatographed
using 15% ethyl acetate in hexane to yield 15 mg (32%)
of an oil. An analytically pure sample was obtained by
the preparative TLC of the above oil: 1HNMR (CDCl 3)
d 0.34 (s, 9H[ 119Sn-H, d]), 1.26 (s, 9H), 1.37 (s, 9H),
1.42 (s, 9H), 1.48 (s, 9H), 5.14 (s, 2H), 5.20 (s, 2H),
6.86 (dd, 1H), 7.16 (dd, 1H), 7.40 (d, 1H), 9.42 (br
m, 4H). MS (FAB+) m/z: cluster peaks around 785
(MH+), 769, 685. HRMS (FAB+) calcd for
C33H57N6O1816Sn (MH+): 781.3255. Found: 781.3246ꢀ
0.0020 (n=3).
(s, 2H), 7.44 (m, 2H), 7.96 (dd, 1H). MS (FAB+) m/z: 333
(MH+), 289, 274. HRMS (FAB+) calcd for C9H14IN6
(MH+): 333.0325. Found: 333.0317ꢀ0.0001 (n=2).
4.3.7. (N,N0-bis-tert-butyloxycarbonyl)-4-amino-3-iodo-
benzylguanidine (7). Method A. To a mixture of 4-
amino-3-iodobenzylamine dihydrochloride (80 mg, 0.25
mmol) and triethylamine (240 mL) in DMF (120 mL)
was added 1, 3-bis-(tert-butyloxycarbonyl)-2-methyl-2-
thiopseudourea (88 mg, 0.3 mmol), and the mixture was
stirred at room temperature for 14 h. The mixture was
partitioned between ethyl acetate and water, and the
combined organic layer was dried and concentrated.
The crude mixture was chromatographed using 15%
ethyl acetate in hexane to obtain 101 mg (82%) of a
1
foam: HNMR (CDCl ) d 1.47 (s, 9H), 1.53 (s, 9H),
3
4.50 (d, 2H), 6.70 (d, 1H), 7.10 (dd, 1H), 7.59 (d, 1H).
MS (FAB+) m/z: 491 (MH+). HRMS (FAB+) calcd for
C18H28IN4O4 (MH+): 491.1155. Found: 491.1155ꢀ
0.0006 (n=2). Method B. The title compound 7 was
isolated in 52% yield by following a procedure similar
to that described above for the preparation of 5 with the
following changes: one equivalent of Mukaiyama’s
reagent and DMF, in lieu of methylene chloride as the
solvent, were used.
4.3.4. 1,4-Bis-(guanidinomethyl)-2-iodobenzene (4). To
34 mg (0.04 mmol) of 2 in a 1/2-dram vial was added a
95/2.5/2.5 (v/v/v) cocktail of TFA/water/triisopropylsi-
lane (1 mL) and the mixture was stirred at room temp-
erature for 3–4 h. The solvents were evaporated and any
residual solvent was removed by coevaporating three
times each with 1 mL methanol and methylene chloride.
The residue was dried to obtain 4 as a film in almost
1
4.3.8. (N,N0-bis-tert-butyloxycarbonyl)-4-aminobenzyl-
guanidine (8). To a solution of 4-aminobenzyl amine
(61 mg, 0.5 mmol) in anhydrous DMF (Pierce, Rock-
ford, IL; 165 mL) was added N,N0-bis(tert-butyloxy-
carbonyl)thiourea (165 mg, 0.6 mmol) and triethylamine
(154 mL, 1.1 mmol). A suspension of Mukaiyama’s
reagent (153 mg, 0.5 mmol) in DMF (330 mL) was
added dropwise to the above mixture. An additional 430
mL of DMF was used to transfer the sedimented
Mukaiyama’s reagent to the reaction flask. The reaction
mixture was stirred at room temperature for 1.5 h and
then partitioned between water and ethyl acetate. The
quantitative yield. HNMR (CD OD) d 4.40 (s, 2H),
3
4.41 (s, 2H), 7.44 (m, 2H), 7.96 (d, 1H). MS (FAB+) m/z:
347 (MH+). HRMS (FAB+) calcd for C10H16IN6
(MH+): 347.0481. Found: 347.0482ꢀ0.0001 (n=2).
4.3.5. (N,N0-bis-tert-butyloxycarbonyl)-4-(N,N0-bis-tert-
butyloxycarbonylguanidino)-3-iodobenzylguanidine (5).
To a mixture of free base of 4-amino-3-iodobenzylamine
(87 mg, 0.35 mmol), N,N0-bis(tert-butyloxycarbonyl)-
thiourea29 (213 mg, 0.77 mmol), and triethylamine (150
mg, 1.48 mmol) in 10 mL methylene chloride was added

G. Vaidyanathan et al. / Bioorg. Med. Chem. 12 (2004) 1649–1656
1655
combined organic layer was washed with brine, dried,
and concentrated. The product was isolated by silica
gel chromatography using a stepwise gradient of 10%
through 50% ethyl acetate in hexane to obtain 108 mg
(60%) of a yellow oil. An analytical sample was
obtained as a foamy solid by further preparative
TLC: mp 125–128 ꢁC. 1HNMR (CDCl 3) d 1.44 (s, 9H),
1.50 (s, 9H), 3.65 (br s, 2H), 4.45 (s, 2H), 6.60 (dd, 2H),
7.08 (dd, 2H), 8.4 (s, 1H), 11.45 (br s, 1H). MS
(FAB+) m/z: 365 (MH+). HRMS (FAB+) calcd for
C18H29N6O4 (MH+): 365.2189. Found: 365.2204ꢀ0.0004
(n=2).
HPLC column (XTerra) and eluted isocratically at 1
mL/min with 5% THF in 5 mM ammonium bicarbo-
nate, pH10.3. The fractions corresponding to 6a
(tR=14 min) were isolated in 60% radiochemical yield
(for this step). Radioactivity from reversed-phase HPLC
fractions in all cases was concentrated using a C18
solid-phase cartridge.
4.5. Paired-label in vitro uptake of MIBG and analogues
by SK-N-SH cells
SK-N-SHcells were plated in 6-well plates at an initial
density of 5ꢃ105 cells per well in 3 mL of medium, and
incubated for 24 h. The medium was removed by
aspiration, 100 nCi each of [125I]MIBG and either
4.4. Radiochemistry
4.4.1. 1,4-Bis-(guanidinomethyl)-2-[131I]iodobenzene (4a).
To a 1/2-dram vial containing 100 mg of 3 was added
1–2 mL of 131I (about 1 mCi) in 0.1N NaOHfollowed by
5 mL of a 3:1 (v/v) mixture of acetic acid and 30% (w/v)
H2O2. The contents of the vial were sonicated for 30 s
and the solvents were evaporated with a stream of
argon. To the residual radioactivity was added 100 mL
of a 95/2.5/2.5 mixture of TFA/water/triisopropylsilane.
The vial was vortexed and left at room temperature for
10 min. Most of the solvents were evaporated under a
flow of argon, and any residual solvents were removed
by coevaporating with chloroform and methanol in that
order (each 3ꢃ25 mL). The radioactivity was recon-
stituted in 25 mL methanol and injected onto a reversed-
phase column (Bondapak) that was eluted isocratically
with 99:1 0.2 M ammonium dihydrogen phosphate, pH
7:THF at a flow rate of 1 mL/min. The product 4a
(tR=14 min) was isolated in an overall radiochemical
yield of 50%.
[
131I]GMIBG or [131I]GIBG were added to each well in
3 mL of medium, and the cells were allowed to take up
the tracers for 2 h at 37 ꢁC. The specificity and energy
dependence of tracer uptake was determined using DMI
and ouabain, respectively. For this, the initial medium
was removed and the cells were incubated with 3 mL
each of either 1.5 mM DMI or 1 mM ouabain for 30
min. The medium was removed and tracers were added
as above and incubated for 2 h at 37 ꢁC. The energy
dependence of tracer uptake also was ascertained by
determining the effect of temperature. For this, the ori-
ginal medium was removed and fresh medium that has
been cooled to 4 ꢁC was added. Then tracers (100 nCi
each) in 10 mL of the medium were added and the cells
were incubated for 2 h at 4 ꢁC. Finally, the effect of
norepinephrine and MIBG was ascertained by removing
the original medium and co-incubating the cells at 37 ꢁC
for 2 h with the tracers (100 nCi in 10 mL each) and 3
mL each of either 50 mM norepinephrine or 10 mM
MIBG. At the end of the 2 h incubation period, the cells
were solubilized by incubation with 500 mL of 0.5 N
NaOHfor 30 min at room temperature and then
removed with cotton swabs. The cell-bound radio-
activity was counted along with input standards using a
dual-channel gamma counter. Three to six replicates
were performed for each of the uptake-1-blocking con-
ditions.
4.4.2. 4-Guanidino-3-[131I]iodobenzylguanidine (6a). To a
solution of 8 in methanol containing 1% acetic acid
(0.18 mg per mL; 50 mL) was added a solution of N-
chlorosuccinimide in methanol (1 mg/mL; 10 mL) fol-
lowed by 1–3 mL of 131I in 0.1 N NaOH(1–2 mCi). The
vial was vortexed and the reaction was allowed to pro-
ceed at room temperature for 30 min. The reaction
mixture was injected onto a normal phase HPLC col-
umn eluted at 1 mL/min with hexane (A) and ethyl
acetate (B), each containing 0.2% acetic acid. The col-
umn was initially eluted isocratically with 93:7 A:B for
20 min, then the percent composition of solvent B
increased linearly to 90 over 15 min. The fractions
corresponding to [131I]7 (tR=15 min) were collected,
and a radiochemical yield greater than 90% was
obtained. The solvents from these fractions were eva-
porated and any residual acetic acid was removed by
coevaporating with ethyl acetate (3 ꢃ 100 mL). To the
residual radioactivity was added 1-5 mg of cyanamide in
20 mL methanol. The vial was vortexed, and the metha-
nol was evaporated with a stream of argon. The vial was
heated at 100 ꢁC in oil bath for 10 min. Trifluoroacetic
acid (100 mL) was added to the reaction vial and the vial
was heated for another 5 min. Trifluoroacetic acid was
removed with an argon stream, and any residual acid
present was removed by coevaporating with 3ꢃ25 mL of
methanol. Finally, the radioactivity was dissolved in 25
mL of methanol and injected onto a reversed-phase
4.6. Paired-label exocytosis of
125I]MIBG from SK-N-SH cells
[
131I]GIBG and
[
Cells were added to 6-well plates at a density of ꢂ4ꢃ105
cells per well per 3 mL of medium and incubated at
37 ꢁC for 24 h. The medium was then replaced with
fresh medium containing ꢂ1 mCi of each tracer in a
total volume of 3 mL per well. After incubating the cells
with the radioactivity for 2 h at 37 ꢁC, the medium was
removed and 3 mL/well of either fresh medium or 1.5
mM desipramine in medium was added. The cell-bound
radioactivity was determined at 0, 2, 4, 8, 24, 48, 72, and
96 h after the initial uptake. The medium was aspirated
at the end of each period and the cells were washed
twice with 0.5 mL each of PBS. The cells were solubi-
lized by incubating with 500 mL of 0.5 N NaOHfor 30
min and then removed with cotton swabs. The cell-
bound radioactivity was counted using a dual-channel
gamma counter. The assay was performed in quad-
ruplicate for each time point.

1656
G. Vaidyanathan et al. / Bioorg. Med. Chem. 12 (2004) 1649–1656
P. C.; Slade, S. K.; Zalutsky, M. R. Bioconjugate Chem.
2001, 12, 798.
4.7. Biodistribution in normal mice
8. Wieland, D.M., In Radiopharmaceuticals; Fritzberg, A. R.,
(Ed.). CRC Press: Boca Raton, 1986, Vol. 1, p 117.
9. Bonisch, H.; Bruss, M. New York Acad. Sci. 1994, 733,
193.
10. Blaschko, H.; Born, G. V. R.; D’iorio, A.; Eade, N. R. J.
Physiol. 1956, 133, 548.
11. Huang, C. C.; Korn, N.; Counsell, R. E. J. Med. Chem.
1979, 22, 449.
12. Vaidyanathan, G.; Zalutsky, M. R. J. Org. Chem. 1997,
62, 4867.
13. Lee, H.; Inbasekaran, M. N.; Wieland, D. M.; Sherman,
P. S.; Fisher, S. J.; Mangner, T. J.; Rogers, W. L.; Clin-
thorne, N. H. J. Nucl. Med. 1986, 27, 256.
14. Yong, Y. F.; Kowalski, J. A.; Lipton, M. A. J. Org.
Chem. 1997, 62, 1540.
15. Nicolaou, K. C.; Trujillo, J. I.; Jandeleit, B.; Chibale, K.;
Rosenfeld, M.; Diefenbach, B.; Cheresh, D. A.; Good-
man, S. L. Bioorg. Med. Chem. 1998, 6, 1185.
16. Sisson, J. C.; Shapiro, B.; Meyers, L.; Mallette, S.; Man-
gner, T. J.; Wieland, D. M.; Glowniak, J. V.; Sherman, P.;
Beierwaltes, W. H. J Nucl. Med. 1987, 28, 1625.
17. Smets, L. A.; Janssen, M.; Metwally, E.; Loesberg, C.
Biochem. Pharmacol. 1990, 39, 1959.
The study utilized male BALB/c mice weighing about 25
g with groups of five mice for each time point. About 5
mCi each of [125I]MIBG and [131I]GIBG in 100 mL of
PBS was injected via the tail vein to three groups of five
animals and the animals were killed at 1, 4, and 24 h
post-injection. To determine the specificity of uptake,
another group of 5 animals was pretreated with DMI
(10 mg/kg; ip) 30 min before injecting the tracers and
the tissue distribution was performed 1 h later. The mice
were killed by an overdose of halothane and tissues of
interest were isolated, washed, blot-dried, and weighed.
The tissues, along with 5% dose standards, were coun-
ted for 131I and 125I using a dual-label program in an
automated gamma counter. The tissue radioactivity
levels were expressed as percent injected dose per gram
of tissue (%ID/g) unless otherwise specified. The statis-
tical significance in the difference between the uptake of
125I and 131I in each tissue was calculated by the Stu-
dents paired t-test using the Excel computer program
(Microsoft).
18. Reist, C. J.; Archer, G. E.; Wikstrand, C. J.; Bigner,
D. D.; Zalutsky, M. R. Cancer Res. 1997, 57, 1510.
19. Vaidyanathan, G.; Affleck, D. J.; Li, J.; Welsh, P.;
Zalutsky, M. R. Bioconjugate Chem. 2001, 12, 798.
20. Clerc, J.; Halpern, S.; Fourre, C.; Omri, F.; Briancon, C.;
Jeusset, J.; Fragu, P. J. Nucl. Med. 1993, 34, 1565.
21. Sisson, J. C.; Shulkin, B. L. Q. J. Nucl. Med. 1999, 43,
217.
Acknowledgements
This work was supported in part by Grants CA78417,
CA42324, and CA91927 from the National Institutes of
Health.
22. Servidei, T.; Iavarone, A.; Lasorella, A.; Mastrangelo, S.;
Riccardi, R. Eur. J. Cancer 1995, 31A, 591.
23. Vaidyanathan, G.; Zhao, X. G.; Strickland, D. K.;
Zalutsky, M. R. J. Nucl. Med. 1997, 38, 330.
24. Biedler, J. N.; Helson, L.; Spengler, B. A. Cancer Res.
1973, 33, 2643.
References and notes
1. Wieland, D. M.; Wu, J.; Brown, L. E.; Mangner, T. J.;
Swanson, D. P.; Beierwaltes, W. H. J. Nucl. Med. 1980,
21, 349.
2. Patel, A. D.; Iskandrian, A. E. J. Nucl. Cardiol. 2002, 9,
75.
25. Kaltenbach, J. P.; Kaltenbach, M. H.; Lyons, W. B. Exp.
Cell Res. 1958, 15, 112.
3. Kaltsas, G. A.; Mukherjee, J. J.; Grossman, A. B. Ann.
Oncol. 2001, 2, S47.
4. Tepmongkol, S.; Heyman, S. Med. Ped. Oncol. 1999, 32,
427.
5. Matthay, K. K.; DeSantes, K.; Hasegawa, B.; Huberty, J.;
Hattner, R. S.; Ablin, A.; Reynolds, C. P.; Seeger, R. C.;
Weinberg, V. K.; Price, D. J. Clin. Oncol. 1998, 16, 229.
6. Vaidyanathan, G.; Shankar, S.; Zalutsky, M. R. Bio-
conjugate Chem. 2001, 12, 786.
26. Sneath, R. L., Jr.; Wright, J. E.; Soloway, A. H.; O’Keefe,
S. M.; Dey, A. S.; Smolnycki, W. D. J. Med. Chem. 1976,
19, 1290.
27. Keck, G. E.; Burnett, D. A. J. Org. Chem. 1987, 52, 2958.
28. Prepared following a procedure reported30 or obtained
from Aldrich.
29. Iwanowicz, E.; Poss, M. A.; Lin, J. Synth. Commun. 1993,
23, 1443.
30. Dodd, D. S.; Kozikowski, A. P. Tetrahedron Lett. 1994,
35, 977.
7. Vaidyanathan, G.; Shankar, S.; Affleck, D. J.; Welsh,