Synthesis and biological effects of new hybrid compounds composed of benzylguanidines and the alkylating group of busulfan on neuroblastoma cells

Article abstract of DOI:10.1016/j.bmcl.2014.04.030

¹³¹Iodine-labelled (meta-iodobenzyl)guanidine ([ ¹³¹I]-mIBG) and busulfan [butane-1,4-diylbis(methanesulfonate)] are well-established pharmaceuticals in neuroblastoma therapy. We report the design, synthesis, and testing of hybrid molecules - mBBG and pBBG - which combine key structural features of (meta-iodobenzyl)guanidine and busulfan: they contain a benzylguanidine moiety for accumulating in neuroblastoma cells via the noradrenaline transporter and, in the meta- or para-position, respectively, one of the two identical alkylating motives of busulfan for killing cells. Uptake and toxicity of hybrids mBBG and pBBG in human neuroblastoma cells compared favorably to their ancestors [¹³¹I]-mIBG and busulfan.

Full text of DOI:10.1016/j.bmcl.2014.04.030

Bioorganic & Medicinal Chemistry Letters 24 (2014) 2728–2733
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and biological effects of new hybrid compounds composed
of benzylguanidines and the alkylating group of busulfan on
neuroblastoma cells
Thomas Hampel a, , Marietta Bruns b, , Melanie Bayer b, Rupert Handgretinger b, Gernot Bruchelt b,
,
⇑
Reinhard Brückner a,
⇑
a Institut für Organische Chemie, Albert-Ludwigs-Universität Freiburg, Albertstraße 21, D-79104 Freiburg, Germany
b Children’s University Hospital, Hoppe-Seyler-Str.1, D-72076 Tübingen, Germany
a r t i c l e i n f o
a b s t r a c t
131Iodine-labelled (meta-iodobenzyl)guanidine ([131I]-mIBG) and busulfan [butane-1,4-diylbis(methane-
sulfonate)] are well-established pharmaceuticals in neuroblastoma therapy. We report the design, syn-
thesis, and testing of hybrid molecules—mBBG and pBBG—which combine key structural features of
(meta-iodobenzyl)guanidine and busulfan: they contain a benzylguanidine moiety for accumulating in
neuroblastoma cells via the noradrenaline transporter and, in the meta- or para-position, respectively,
one of the two identical alkylating motives of busulfan for killing cells. Uptake and toxicity of hybrids
mBBG and pBBG in human neuroblastoma cells compared favorably to their ancestors [131I]-mIBG and
busulfan.
Article history:
Received 24 March 2014
Revised 8 April 2014
Accepted 9 April 2014
Available online 18 April 2014
Keywords:
Neuroblastoma
meta-Iodobenzylguanidine (mIBG)
Alkylating agent
Busulfan
Ó 2014 Elsevier Ltd. All rights reserved.
Neuroblastoma is a malignant tumor of the sympathetic nervous
system in childhood with a poor prognosis in stage IV.1 It
usually originates from the adrenal gland. It is the most abundant
extracranial solid tumor in childhood making up for 8–10% of all
cancers and for 15% of the corresponding mortality.2 Approxi-
mately 1.1 children out of 100,000 under the age of 15 are diag-
nosed with neuroblastoma. 90% of those who are diagnosed are
younger than 10 years, one third is diagnosed in the first year of
life.1,3
A majority of the cytotoxic drugs currently used in cancer ther-
apy acts non-specifically. Such chemotherapeutics4 often cause
severe side effects and/or long-term damages. The latter are espe-
cially severe concerning the treatment of infants. The aim of our
project is to develop cytotoxic compounds with a higher specificity
than their predecessors. This objective shall be reached by target-
ing a particular structure in neuroblastoma cells, namely the nor-
adrenaline transporter. A substance with an enhanced specificity
for such cells of the sympathetic nervous system is (meta-iodoben-
zyl)guanidine (mIBG) [formula: Scheme 1].
An mIBG equipped with a radiolabel was first synthesized by
Wieland et al. in 1979.5 In 1984/1985 radio-labelled mIBG was
introduced independently both in Heidelberg6 and Tübingen7 for
scintigraphic imaging of neuroblastoma. Following this discovery
[
123I]-mIBG (b+-emitter,
s1/2 = 13 h) has been used around the
world as a routine diagnosis of neuroblastoma and also for an
ensuing therapy. The major therapeutical agent, which was devel-
oped from the mentioned discovery and which has been used ever
since, is [131I]-mIBG (bꢀ-emitter,
s
1/2 = 8.0 d).8 Radio-labelled ana-
1/2 = 59 d)
1/2 = 7.2 h) were prepared as
logs of [131I]-mIBG containing iodine-1259 (b+-emitter,
or astatine-21110 (mainly
-emitter,
s
a
s
well. They represent therapeutical alternatives for special manifes-
tations of neuroblastoma.9,10
The uptake of mIBG in neuroblastoma cells is the basis for the
aforementioned activities. It was first described in one of our lab-
oratories in cell culture experiments.11 Later it was shown that this
uptake occurs via the noradrenaline transporter.12 mIBG was also
shown to be taken up by OCT-expressing cells (OCT: organic cat-
ionic transporter).13 This process is a well known side effect in
scintigraphy of neuroblastoma.8a,14 Very recently, it turned out
that the uptake of mIBG in OCT-expressing cells can be reduced
by glucocorticoids.15
⇑
Corresponding authors. Tel.: +49 (0) 7071 298 4710; fax: +49 (0) 707 129 5482
Even nonradioactive benzylguanidines—which would not per se
be expected to affect cell life [exceptions: a nitro-substituted mIBG
(Ref. 16) and unsubstituted mIBG (Refs. 17–19; see also the next
(G.B.); tel.: +49 (0) 761 203 6029; fax: +49 (0) 761 203 6100 (R.B.).
E-mail addresses: gernot.bruchelt@med.uni-tuebingen.de (G. Bruchelt),
reinhard.brueckner@organik.chemie.uni-freiburg.de (R. Brückner).
These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.bmcl.2014.04.030
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

T. Hampel et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2728–2733
2729
(a) (meta-Iodobenzyl)guanidine (mIBG)
The simplest nonradioactive mIBG is the parent compound
itself, namely mIBG. Intriguingly, it is active against neuroblastoma
cells: associated effects are a sharp decrease of the ATP/ADP ratio18
and aberrances in glycolysis and oxidative phosphorylation.17–19
Recently, the HR-NBL123 trial of the European SIOP24 Neuroblas-
toma Group demonstrated that busulfan [4;25 formula: Scheme 2
(a)] and melphalan (={4-[N,N-bis(2-chloroethyl)amino]phenyl}ala-
nine) combined make an excellent myeloablative therapy for high-
risk neuroblastoma.26 It is superior to the so-called ‘CEM protocol’,
which is combined treatment with carboplatin, etoposide, and
melphalan. The ‘CEM protocol’ represented the recommended neu-
roblastoma therapy before it was replaced by the combination
busulfan (4)/melphalan.27
What we hoped would turn out to be anti-neuroblastoma
agents were hybrids of the alkylating motive of busulfan [4;25 for-
mula: Scheme 2 (a)] and a benzylguanidine. The alkylating group
should be installed in the benzene ring such that a meta- (mBBG)
or para-substituted benzylguanidine (pBBG) results [formulas:
Scheme 2 (b)]. We hoped that the respective hybrid molecules
(mBBG and pBBG) would combine the cytotoxicity of busulfan
with the ability of (meta-iodobenzyl)- or (para-iodobenzyl)guani-
dines of intruding neuroblastoma cells by means of the noradren-
aline transporter. Neither mBBG nor pBBG contains an iodine
atom. This makes our design novel compared to the Ludeman sys-
tems22 and simpler: introducing two substituents into benzene (as
NH
I
H2N
N
H
mIBG
(b) Octapeptide 1 (Vaidyanathan et al.)
HO
HO
O
O
H
HO
N
N
NH2
NH
N
H
H
O
O
HN
HN
O
NH
S
S
[
131I]
H2N
N
H
O
H
H
N
N
N
H
O
O
O
1
(c) Hybrid molecules from mIBG and an electrophilic functionality (Ludeman et al.)
NH NH
I
I
H2N
N
H2N
N
H
H
Br/Cl
S
S
n = 0-10
2
3
Scheme 1. Benzylguanidine motives in mIBG (a), in octapeptide 121f (b), and in the
iodine-containing alkylating agents 222 and 322 (c).
H2N
NBoc
SMe
BocHN
OH
meta-5
para-5
6
paragraph]—should be modifiable so that they kill neuroblastoma
cells. Such a concept has been pursued by at least two groups:20
Vaidyanathan et al. developed modified—yet always iodine-
containing—analogues of mIBG,21 such as the octapeptide 121f
[Scheme 1]. Its cytotoxic effect is most likely due the redox active
SAS bond. Ludeman et al. proposed ‘A Method for Treating Neuro-
blastoma’ based on a sizable number of mIBGs.22 Each of these
molecules contains an electrophile—in substituents of widely var-
ied structures [e.g., Scheme 1 (c)]—for attacking the mitochondrial
glutathione of tumor cells. Without an exception the Ludeman
compounds display an iodine-substituent in an unvaried mIBG
substructure.16,22
a) DMF
NBoc
O
O
Me
O
O
S
BocHN
BocHN
N
H
S
O
Me
O
OH
meta-7 (88%)
para-7 (72%)
busulfan (4)
b) K2CO3, DMF
NBoc
N
H
O
Me
O
S
(a) Busulfan [butane-1,4-diylbis(methanesulfonate)]
O
O
meta-8 (77%)
para-8 (63%)
O
O
Me
O
S
S
O
Me
O
O
c) SnCl4 or F3C-CO2H ("TFA")
4
NH
(b) This work´s concept for hybrid molecules; pBBG and mBBG
"HX " H2N
N
NH
H
O
Me
S
O
H2N
N
H
O
O
O
Me
S
O
mBBG"•HCl" (92%)[a]
pBBG"•HCl" (74%)[b]
mBBG"•TFA" (quant.)
pBBG"•TFA" (quant.)
O
O
pBBG
NH
Scheme 3. Synthesis of the hybrid molecules pBBG and mBBG by deprotection of
the N,N0-Boc-protected precursors para-8 or meta-8 either by HCl or by TFA in the
final step. Conditions: (a) 5 (1.0 equiv), 6 (1.1 equiv), DMF, room temperature,
16–24 h; 88% (meta-7), 72% (para-7). (b) 7 (1.0 equiv), 4 (3.0–3.1 equiv), K2CO3
(5.1–5.2 equiv), DMF, room temperature, 20 h; 63% (para-8), 77% (meta-8). (c) SnCl4
H2N
N
H
O
O
O
S
O
Me
(2.0–2.4 equiv), EtOAc, room temperature, 3 h; 92% (mBBG’’ꢁHCl’’)[a]
, 74%
mBBG
(pBBG’’ꢁHCl’’)b. (c) F3CCO2H (TFA)/CH2Cl2 1:1 (v:v), room temperature, 1.5 h; quant.
aDifferent batches contained 0–20% of an SN2 product, in which the methanesul-
fonate group was replaced by an unknown nucleophile.[36] bContained 9% of a
SN2-product where the methane sulfonate group was replaced by an unknown
nucleophile.[36]
Scheme 2. The bifunctional alkylating agent busulfan (425) (a). Our concept for a
specific therapy of neuroblastoma (b): hybrid molecules pBBG and mBBG
composed of the alkylating motive of busulfan (4) and of the common benzylgua-
nidine motive both of pIBG and mIBG.

2730
T. Hampel et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2728–2733
in this Letter) rather than introducing three (as in Ref. 22) requires
fewer synthetic steps.
Our syntheses of mBBG and pBBG started with commercially
available N,N0-protected thiourea 628 and (meta-hydroxybenzyl)-
and (para-hydroxybenzyl)amine (meta-5 and para-5), respectively
(Scheme 3). Step 1 consisted of stirring the respective (hydroxy-
benzyl)amines 5 with a slight excess of the guanidinylating agent
6 in DMF at room temperature for 16–24 h. The N,N0-dicarbamoy-
lated benzylguanidines meta-7 (88% yield) and para-7 (72% yield),
respectively, resulted. DMF as a solvent was crucial for attaining
acceptable conversions. The likely reason is that the (hydroxyben-
zyl)amines meta-5 and para-5 were only sparingly soluble in other
solvents, for example, in CH2Cl2.29 Step 2 of the synthetic sequence
of Scheme 3 consisted of treating the respective (meta-hydroxy-
benzyl)- or (para-hydroxybenzyl)guanidine 7 with an excess of
busulfan (4) and K2CO3 in DMF. This delivered the corresponding
N,N0-dicarbamoylated BBG-precursors meta-8 (77% yield) and
para-8 (63% yield).30
The final step of Scheme 3 posed the challenge of removing
the Boc groups from the benzylguanidines moiety of meta- and
para-8 without affecting the sulfonate group. The latter was
not only susceptible to rapid hydrolysis but also to (another)
nucleophilic substitution (cf. the subsequent paragraph).
Whereas we had been able to extract the substrates meta-8
and para-8 of the imminent step from aqueous phases with stan-
dard solvents (e.g., tBuOMe) and to purify them by flash chroma-
tography on silica gel,31 isolating the Boc-free guanidines mBBG
and pBBG was incompatible with either technique. It rather
required a considerable amount of fine-tuning since mBBG and
pBBG were extremely polar.
Screening several Lewis acids, solvents, and solvent mixtures
we found trifluoroacetic acid (TFA) or SnCl4 suitable for decarba-
moylating meta-8 and para-8 while tolerating their methanesul-
fonate group. In the former case we stirred the respective
precursor meta- or para-8 in CH2Cl2/TFA (1:1, v:v) at room tem-
perature for 1.5 h.32 Then we evaporated all volatiles (i.e., the
solvent and excess TFA as well as isobutene and CO2, which
are stoichiometric by-products of Boc cleavage) in vacuo. Ini-
tially, this seemed to be a reliable procedure for isolating the
expected products mBBG ꢁTFA and pBBG ꢁTFA in quantitative
yields. Later we realized that TFA could be removed from mBBG
ꢁTFA and pBBG ꢁTFA in vacuo to some extent. This phenomenon
was a surprise. It was unveiled by preparing several batches of
mBBG’’ꢁTFA’’ and pBBG’’ꢁTFA’’ (for the meaning of the quotation
marks: cf. below), respectively, and by finding that the integrals
of the respective F3CCO2HꢁBBG or F3CCO2HꢁBBG 13C NMR reso-
nances (400 or 500 MHz, d6-DMSO) varied relative to the other
13C NMR integrals. The exact composition of the respective ‘tri-
fluoroacetates’ remaining unknown, we depict the latter using
quotation marks, that is, as mBBG’’ꢁTFA’’ and as pBBG’’ꢁTFA’’,
respectively.33
Treatment of the benzylguanidine dicarbamates meta- or para-8
with SnCl4 (2.0–2.4 equiv) in ethyl acetate at room temperature for
3 h34 removed the Boc groups as efficiently as deprotection in TFA/
CH2Cl2 (vide supra). However, we suspected that rests of Sn(IV) in
the resulting hydrochlorides mBBG’’ꢁHCl’’ and pBBG’’ꢁHCl’’,35
respectively, might disturb the upcoming cell tests. Therefore, we
freed the crude reaction mixtures from excess SnCl4 in vacuo. At
0 °C we quenched the residue with H2O [+SnCl4 ? SnO2(s)]. The
resulting solution was extracted with a mixture of iPrOH and CHCl3
(1:3, v:v). The combined extracts were concentrated to provide
mBBG’’ꢁHCl’’ and pBBG’’ꢁHCl’’ in up to 92% and 74% yield,
respectively.36
Figure 1. Top (a) uptake of [3H]-NA (0.1
lCi; final concentration 0.1 lM) into
different human neuroblastoma cell lines after a 15 min incubation period at 37 °C
in pure PBS++41 and in the presence of unlabelled mIBG [final concentration (f.c.):
100
tion 0.1
PBS++41 and in the presence of unlabelled mIBG, pBBG, mBBG and busulfan (4),
l
M] n = 3, mean SD. Bottom (b) uptake of [3H]-NA (0.1
lCi; final concentra-
lM) into SK-N-SH cells after a 15 min incubation period at 37 °C in pure
respectively (final concentration: each 100
mean SD).
lM); n = 3 independent experiments,
cells. First, the uptake of pBBG and mBBG into human neuroblas-
toma cell lines SK-N-SH, LS, SiMa and Kelly was analyzed by com-
petitive uptake experiments with [3H]-noradrenaline {[3H]-NA}.
Cell suspensions were prepared and adjusted to 1 ꢂ 106 cells/mL.
[3H]-noradrenaline {norepinephrine-(levo[7,3H]); [3H]-NA; specific
activity: 12.1 lCi/mmol; Perkin–Elmer; final concentration in the
incubation mixture: 1 ꢂ 10–7 M} as well as the other compounds
used in the experiment (final concentration 1 ꢂ 10–4 M) were
added in 10 lL portions, respectively. After incubation at 37 °C
for 15 min 10 mL ice-cold phosphate buffered saline (PBS40) con-
taining Ca2+ and Mg2+ (PBS++41) was added and the suspensions
were centrifuged at 500g for 5 min. This procedure was repeated
twice. Finally the radioactivity in the cell pellets was measured
after lysis with 500 lL distilled water and addition of 10 mL liquid
scintillation cocktail (OptiPhase Super Mix, Wallac). It is known
that among these four human neuroblastoma cell lines SK-N-SH
cells express the noradrenaline transporter. Figure 1a shows that
SK-N-SH cells have a much higher uptake of [3H]-noradrenaline
than cells of the other neuroblastoma cell lines. The surplus of
unlabelled mIBG strongly inhibits this uptake process. A significant
uptake inhibition in SK-N-SH cells was also obtained by pBBG,
and—to a lesser extent—by mBBG, whereas busulfan [4;25 formula:
Scheme 2 (a)] did not show an inhibitory effect at all (Fig. 1b).
Among the substances synthesized, mBBG’’ꢁTFA’’ and
pBBG’’ꢁHCl’’ were most soluble in aqueous medium and were
therefore used for the following experiments with neuroblastoma

T. Hampel et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2728–2733
2731
Figure 2. Concentration-dependent influence of mIBG, pBBG, mBBG, and busulfan (4) on cell proliferation/vitality on four different human neuroblastoma cells compared to
untreated controls, set as 100%. MTT-test; mean SD, n = 3–5 independent experiments.
Next, the cytotoxic, that is, antiproliferative effects of pBBG and
mBBG on neuroblastoma cells compared to mIBG and busulfan
[4;25 formula: Scheme 2] were investigated using the MTT-test
according to Mosmann.37 After an incubation period of 72 h, pBBG
and—to a lesser extent—mBBG were more effective in a dose-
By analyzing the influence of mIBG on cell metabolism, Loes-
berg et al. observed that mIBG enhanced the rate of glycolysis
(enhanced lactate formation) in neuroblastoma and lymphosar-
coma cells.18 This observation is of special interest since many
tumor cells, among them neuroblastoma cells, preferentially
metabolize glucose to lactate instead of CO2 and H2O even in the
presence of oxygen (‘aerobic’ glycolysis, Warburg effects).38,39
Based on the assumption that one mol glucose can theoretically
be metabolized at the most to two mol lactate (100%), all three
neuroblastoma cell lines used in the experiments shown in Figure 3
followed significantly the Warburg characteristics: LS: 81 19%;
SiMa: 98 2%; Kelly: 74 26% glucose converted to lactate (mean
SD; n = 3). We were interested whether pBBG and mBBG behave
similar to mIBG concerning their influence on the glucose metab-
olism. For this purpose, glucose consumption and lactate produc-
tion were measured with commercial assays (Roche, Mannheim,
dependent manner (f.c. 10–100 lM) than mIBG or even busulfan
(4), indicating their powerful potential for neuroblastoma treat-
ment (Fig. 2). Surprisingly, SK-N-SH cells which incooperate [3H]-
noradrenaline much better than LS, Kelly and SiMa cells, were
affected in the MTT-assay to a lesser extent by mIBG, pBBG, mBBG,
and busulfan (4). Compared to the other cells, SK-N-SH proliferate
much slower (doubling time approx. 30 h compared to 20–24 h in
the case of LS, Kelly, and SiMa). This may explain the latter discrep-
ancy. Further experiments with systematic time-dependent uptake
proliferation, and cytotoxic studies should clarify this unexpected
effect.

2732
T. Hampel et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2728–2733
Figure 3. Influence of mIBG, pBBG, mBBG, and busulfan (4) on glycolysis (glucose consumption and lactate production) in fast proliferating neuroblastoma after 12 hours
incubation period at 37 °C in PBS++41 containing 10 mM glucose. Fractional changes in glucose consumption (top) and lactate production (bottom) in the presence of mIBG,
pBBG, mBBG and busulfan were related to the untreated control cells, set as 1.0; mean SD, n = 3 independent experiments. Further information: see text.
Germany) in the supernatant of approx. 500.000 seeded cells in 24
well plates (2 mL PBS++ 41
containing approx. 10 mM glucose) dur-
positive and NAT-negative neuroblastoma cells, and potential ther-
apeutic applications.
,
ing a 12 h incubation period at 37 °C. Fractional changes in glucose
consumption and lactate production in the presence of mIBG,
pBBG, mBBG, and busulfan [4;25 formula: Scheme 2 (a)] were
related to the untreated control cells (Fig. 3). In these experiments,
it could be confirmed that mIBG enhanced glucose consumption
and lactate production also in the three investigated neuroblas-
toma cell lines. Most characteristically, mBBG and pBBG caused
also enhanced lactate production (glucose consumption was only
significantly enhanced in LS cells), indicating that the new synthe-
sized compounds still retained features of mIBG after iodine
replacement by the alkylating group. In contrast, these effects on
glycolysis were not observed after treatment with busulfan (4).
Further experiments should clarify in which way glycolysis is influ-
enced by mIBG, mBBG, and pBBG on a molecular level and whether
these features contribute to their cytotoxic effects and could be
used for therapy.
In summary the present study discloses the syntheses of salts of
mBBG and pBBG [formulas: Scheme 2 (b)], that is, of hybrid mole-
cules combining the benzylguanidine moiety of mIBG [formula:
Scheme 1 (a)] with the methanesulfonate moiety of busulfan [4;
formula: Scheme 2 (a)]. The N,N0-dicarbamoylated precursors
meta-8 and para-8 of mBBG and pBBG, respectively, were obtained
in two steps from the (hydroxybenzyl)amines meta-5 or para-5,
thiourea 6, and busulfan (4; Scheme 3); their overall yields were
68% and 45%, respectively. Boc removal from the dicarbamates
meta-8 and para-8 was straightforward in trifluoroacetic acid
(TFA)/CH2Cl2 (1:1, v:v) but at best tedious if not outright trouble-
some using SnCl4. In the former case we obtained the trifluoroace-
tates of mBBG and pBBG, in the latter case contaminated[36]
hydrochlorides. We showed that mBBG and pBBG retain features
of mIBG, that is, uptake via the NAT and an enhancing effect on gly-
colysis in human neuroblastoma cells. Gratifyingly, mBBG and
pBBG were even more cytotoxic towards these cells than their
ancestors mIBG and busulfan (4). Further studies shall clarify the
mechanism of action of mBBG and pBBG. They will also address
their effect on glycolysis, the time-dependent uptake into NAT-
Acknowledgments
T.H. is indebted both to the Konrad-Adenauer-Stiftung and to
Novartis AG (Basel) for PhD scholarships. In addition, he is grateful
to BASF SE (Ludwigshafen) for continuous support. Furthermore,
we thank the Förderverein für krebskranke Kinder Tübingen e.V.
for the financial support of M.B.
References and notes
1. Maris, J. M. N. Eng. J. Med. 2010, 362, 2202.
2. (a)Neuroblastoma; Cheung, N.-K. V., Cohn, S. L., Eds.; Springer: Heidelberg,
Berlin, New York, 2005; (b) Park, J. R.; Eggert, A.; Caron, H. Hematol. Oncol. Clin.
North Am. 2010, 24, 65.
3. (a) Powell, J. E.; Estève, J.; Mann, J. R.; Parker, L.; Frappaz, D.; Michaelis, J.; Kerbl,
R.; Mutz, I. D.; Stiller, C. A. Lancet 1998, 352, 682; (b) Matthay, K. K.; Villablanca,
J. G.; Seeger, R. C.; Stram, D. O.; Harris, R. E.; Ramsay, N. K.; Swift, P.; Shimada,
H.; Black, C. T.; Brodeur, G. M.; Gerbing, R. B.; Reynolds, C. P. N. Eng. J. Med.
1999, 341, 1165; (c) Berthold, F.; Boos, J.; Burdach, S.; Erttmann, R.; Henze, G.;
Hermann, J.; Klingebiel, T.; Kremens, B.; Schilling, F. H.; Schrappe, M.; Simon,
T.; Hero, B. Lancet Oncol. 2005, 6, 649.
4. (a) Papac, R. J. Yale J. Biol. Med. 2001, 74, 391; (b) Chabner, B. A.; Roberts, T. G.
Nat. Rev. Cancer 2005, 5, 65.
5. Wieland, D. M.; Swanson, D. P.; Brown, L. E.; Beierwaltes, W. H. J. Nucl. Med.
1979, 20, 155.
6. Kimmig, B.; Brandeis, W. E.; Eisenhut, M.; Bubeck, B.; Hermann, H. J.; zum
Winkel, K. J. Nucl. Med. 1984, 25, 773.
7. Treuner, J.; Feine, U.; Niethammer, D.; Müller-Schaumburg, W.; Meinke, J.;
Eibach, E.; Dopfer, R.; Klingebiel, T.; Grumbach, S. Lancet 1984, 323, 333.
8. (a) Treuner, J.; Klingebiel, T.; Feine, U.; Buck, J.; Bruchelt, G.; Dopfer, R.; Girgert,
R.; Müller-Schauenburg, W.; Meinke, J.; Kaiser, W.; Niehthammer, D. Pediatr.
Hematol. Oncol. 1986, 3, 205; (b) Baierwaltes, W. H. Int. J. Rad. Appl. Instrum. B
1987, 14, 183; (c) Troncone, L.; Rufini, V. Anticancer Res. 1997, 17, 1823; (d)
Kaltsas, G. A.; Mukherjee, J. J.; Grossman, A. B. Ann. Oncol. 2001, 12, 47.
9. Hoefnagel, C. A.; Smelts, L.; Voûte, P. A.; de Kraku, J. J. Nucl. Med. 1991, 32, 3612.
10. Vaidyananthan, G.; Strickland, D. K.; Zalutsky, M. R. Int. J. Cancer 1994, 57, 908.
11. Buck, J.; Bruchelt, G.; Girgert, R.; Treuner, J.; Niethammer, D. Cancer Res. 1985,
45, 6366.
12. Pacholczyk, T.; Blakely, R. D.; Amara, S. G. Nature 1991, 350, 350.
13. Bayer, M.; Kuçi, Z.; Schömig, E.; Gründemann, D.; Dittmann, H.; Handgretinger,
R.; Bruchelt, G. Nucl. Med. Biol. 2009, 36, 287.
14. Bonnin, F.; Lumbroso, J.; Tennenbaum, F.; Harmann, O.; Parmentier, C. J. Nucl.
Med. 1994, 35, 803.

T. Hampel et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2728–2733
2733
15. Bayer, M. S.; Schmitt, J.; Glaesmann, K.; Pichler, B.; Handgretinger, R.; Bruchelt,
G. ‘Inhibition of Organic Cation Transporter 3 for a More Selective Uptake of
meta-Iodobenzylguanidine in Neuroblastoma’’, 4th Int. Tübingen-Symposium on
Pediatric Solid Tumors, February 16–18, 2012 (Abstract).
on the concentration and/or the temperature (we did not probe such effects,
though).
30. The 13C NMR spectrum (100 MHz, CDCl3) of a given batch of compound para-8
in CDCl3 was registered two times. To our considerable surprise one spectrum
showed two NCO2C(CH3)3 resonances (d = 153.2 and 163.7 ppm) for the two
NCO2C(CH3)3 groups (which is what we expected) while the other spectrum
showed a single NCO2C(CH3)3 resonance (d = 153.2 ppm) whereas the signal at
d = 163.7 ppm had disappeared. Under the latter conditions the spectrum
showed two dramatically weakened NCO2C(CH3)3 resonances. We attributed
these phenomena tentatively to the quadrupole moments of the three nitrogen
atoms of the guanidine moiety.
16. Nonradioactive mIBGs without electrophilic groups like those depicted in
Scheme 1c (but in some instances containing
a nitro group, which is a
potentially cytotoxic group) were described, too. Vaidyanathan, G.; Shankar, S.;
Affleck, D. J.; Slade, S. A.; Welsh, P.; Zalutsky, M. R. J. Labelled Compd.
Radiopharm. 2012, 44, 324. pharmacological data have been published yet.
17. Cornelissen, J.; Wanders, R. J.; Van den Bogert, C.; Van Kuilenburg, A. B.;
Elzinga, L.; Voûte, P. A.; Van Gennip, A. H. Eur. J. Cancer 1995, 31A, 582.
18. Loesberg, C.; Van Rooij, H.; Nooijen, W. J.; Meijer, A. J.; Smets, L. A. Int. J. Cancer
1990, 46, 276.
31. Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923.
32. Procedure: Kaul, M.; Parhi, A. K.; Zhang, Y.; LaVoie, E. J.; Tuske, S.; Arnold, E.;
Kerrigan, J. E.; Pilch, D. S. J. Med. Chem. 2012, 55, 10160.
19. Cornelissen, J.; Wanders, R. J.; Van Gennip, A. H.; Van den Bogert, C.; Voûte, P.
A.; Van Kuilenburg, A. B. Biochem. Pharmacol. 1995, 49, 471.
33. Sometimes impurities formed during the deprotection of compound meta-8
20. Vaidyanathan, G. Q. J. Nucl. Med. Mol. Imaging 2008, 52, 351.
21. (a) Vaidyanathan, G.; Shankar, S.; Affleck, D. J.; Alston, K.; Norman, J.; Welsh, P.;
LeGrand, H.; Zalutsky, M. R. Bioorg. Med. Chem. 2004, 12, 1649; (b) Boyd, M.;
Mairs, R. J.; Keith, W. N.; Ross, S. C.; Welsh, P.; Akabani, G.; Owens, J.;
Vaidyanathan, G.; Carruthers, R.; Dorrens, J.; Zalutsky, M. R. J. Gene Med. 2004,
6, 937; (c) Vaidyanathan, G.; Welsh, P. C.; Vitorello, K. C.; Snyder, S.; Friedman,
H. S.; Zalutsky, M. R. Eur. J. Nucl. Med. Mol. Imaging 2004, 31, 1362; (d)
Vaidyanathan, G.; Affleck, D. J.; Alston, K. L.; Welsh, P.; Zalutsky, M. R. Nucl.
Med. Commun. 2004, 25, 947; (e) Vaidyanathan, G.; Affleck, D. J.; Alston, K. L.;
Zhao, X. G.; Hens, M.; Hunter, D. H.; Babich, J.; Zalutsky, M. R. Bioorg. Med.
Chem. 2007, 15, 3430; (f) Vaidyanathan, G.; Affleck, D. J.; Norman, J.; O’Dorisio,
S.; Zalutsky, M. R. Bioconjugate Chem. 2007, 18, 2122; (g) Vaidyanathan, G.;
White, B.; Affleck, D. J.; McDougald, D.; Zalutsky, M. R. Nucl. Med. Biol. 2011, 38,
77.
with TFA. These impurities displayed a
1H NMR triplet at d = 4.46 ppm
(500 MHz, d6-DMSO) and a 13C NMR resonance at d = 68.3 ppm.
34. Procedure: Miel, H.; Rault, S. Tetrahedron Lett. 1997, 38, 7865.
35. How much HCl the desired guanidines mBBGꢁHCl and pBBGꢁHCl contained was
not ascertained experimentally. However, we had discovered circumstantial
evidence that the analogous guanidines mBBGꢁTFA and pBBGꢁTFA lost TFA to
some extent, when these compounds were dried in vacuo (cf. main body of the
text). By analogy, HCl losses from mBBGꢁHCl and from pBBGꢁHCl must be
considered as possibilities until the contrary is proved. Therefore, we denote
these guanidine specimens as mBBG’’ꢁHCl’’ and pBBG’’ꢁHCl’’, respectively,
attempting to indicate that they might contain less than a full equiv of HCl.
36. mBBG’’ꢁHCl’’ and pBBG’’ꢁHCl’’ were obtained together with up to 20% and 9%,
respectively, of an impurity. These impurities showed characteristic triplets at
dH = 3.71 ppm (400 MHz, d6-DMSO). The distinctness of this value from the 1
H
22. (a) Ludeman, S. M.; Gamcsik, M. P.; Driscoll, T. A.; Springer, J. B.; Colvin, O. M.;
Adams, D. J.; Base, K. WO2008121322, 2008; (b) Ludeman, S. M.; Gamcsik, M.
P.; Driscoll, T. A.; Springer, J. B.; Colvin, O. M.; Adams, D. J.; Base, K.
US20100104628, 2010.
23. HR-NBL1: High Risk Neuroblastoma Study.
24. SIOP: International Society of Pedriatic Oncology.
NMR chemical shifts of nPrCH2Cl (d = 3.62 ppm) and nPrCH2OH (d = 3.38 ppm)
in d6-DMSO (400 MHz;Abraham, R. J.; Byrne, J. J.; Griffiths, L.; Perez, M. Magn.
Reson. Chem. 2006, 44, 491) seems to rule out that the methanesulfonate group
in compounds mBBG and pBBG had been exchanged by SN2 attacks of a
chloride ion or H2O. Moreover, the 1H NMR shifts (500 MHz, d6-DMSO) of
C6H5CH2NH2 (d = 3.72 ppm) or 2-MeO–C6H4–CH2NH2 (d = 3.71 ppm) published
by Swain, C. J.; Seward, E. M.; Cascieri, M. A.; Fang, T. M.; Herbert, R.;
MacIntyre, D. E.; Merchant, K. J.; Owen, S. N.; Owens, A. P.; Sabin, V.; Teall, M.;
VanNiel, M. B.; Williams, B. J.; Sadowski, S.; Strader, C.; Ball, R. G.; Baker, R. J.
Med. Chem. 1995, 38, 4793 are virtually identical with that of the mentioned
impurities. At this point we held the latter for the respective benzylamines,
which would be released from compounds mBBG and pBBG hydrolytically.
However, we observed one decomposition of compound mBBG, which falsified
this interpretation: The signal of CH2OSO2CH3 (d = 4.27 ppm) decreased while
the triplet of the impurity (d = 3.71 ppm) increased (400 MHz 1H NMR analysis,
d6-DMSO); concomitantly the signals of both CH2OSO2CH3 (d = 70.1 ppm) and
CH2OSO2CH3 (d = 36.6 ppm) decreased and the signal of the impurity
(d = 45.2 ppm) increased (125 MHz 13C NMR analysis, d6-DMSO)—but the
intensity of the signal of the guanidine–carbon remained unchanged.
37. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
25. Iwamoto, T.; Hiraku, Y.; Oikawa, S.; Mizutani, H.; Kojima, M.; Kawanishi, S.
Cancer Sci. 2004, 95, 454. and references cited therein.
26. (a) Ladenstein, R. ‘Busulphan–Melphalan as a Myeloablative Therapy (MAT) for
High Risk Neuroblastoma: Results from the HR-NBL1/SIOPEN Trial’, 2011 ASCO
Annual Meeting, June 5, 2011 (Abstract); (b) Ladenstein, R. ‘Results from the
HR-NBL1/SIOPEN Trial’, 4th Int. Tübingen-Symposium on Pediatric Solid Tumors,
February 16–18, 2012 (Abstract).
27. Grupp, S. A.; Asgharzadeh, S.; Gregory, Y. A. Biol. Blood Marrow Transplant.
2012, 18, S92. and references cited therein.
28. Hammerschmidt, F.; Kvaternik, H.; Schweifer, A.; Mereiter, K.; Aigner, R. M.
Synthesis 2012, 3387.
29. The 1H NMR spectra (300–500 MHz) of the (hydroxybenzyl)guanidine meta-7
in CDCl3 changed both with the concentration of meta-7 and with the
temperature during signal acquisition. Sample concentrations could be
(unintentionally) (1) such that the chemically distinct NCO2C(CH3)3 groups
displayed a single singlet rather than two singlets and/or (2) such that peaks of
the benzyl group were shifted by about as much as the ‘spread’ of an ortho
coupling. The interference of these variations was attributed to differential
intermolecular hydrogen bonding. As a consequence, the chemical shifts in all
benzylguanidines of this study (7, 8, mBBG, and pBBG) might depend likewise
38. Niewisch, M.; Kuci, Z.; Wolburg, H.; Sautter, M.; Krampen, L.; Deubzer, B.;
Handgretinger, R.; Bruchelt, G. Cell Physiol. Biochem. 2012, 29, 373.
39. Gatenby, R. A.; Gilles, R. J. Nat. Rev. Cancer 2004, 4, 891.
40. ‘PBS’ is for ‘phosphate-buffered saline’ and usually refers to a salt solution
without Ca2+ or Mg2+
.
41. ‘PBS++’ designates PBS,40 to which both Ca2+ and Mg2+ were added.