Aromatic nitrogen mustard-based prodrugs: Activity, selectivity, and the mechanism of DNA cross-linking

Article abstract of DOI:10.1002/chem.201400090

Three novel H₂O₂-activated aromatic nitrogen mustard prodrugs (6-8) are reported. These compounds contain a DNA alkylating agent connected to a H₂O₂-responsive trigger by different electron-withdrawing linkers so that they are inactive towards DNA but can be triggered by H₂O₂ to release active species. The activity and selectivity of these compounds towards DNA were investigated by measuring DNA interstrand cross-link (ICL) formation in the presence or absence of H ₂O₂. An electron-withdrawing linker unit, such as a quaternary ammonia salt (6), a carboxyamide (7), and a carbonate group (8), is sufficient to deactivate the aromatic nitrogen mustard resulting in less than 1.5% cross-linking formation. However, H₂O₂ can restore the activity of the effectors by converting a withdrawing group to a donating group, therefore increasing the cross-linking efficiency (>20%). The stability and reaction sites of the ICL products were determined, which revealed that alkylation induced by 7 and 8 not only occurred at the purine sites but also at the pyrimidine site. For the first time, we isolated and characterized the monomer adducts formed between the canonical nucleosides and the aromatic nitrogen mustard (15) which supported that nitrogen mustards reacted with dG, dA, and dC. The activation mechanism was studied by NMR spectroscopic analysis. An in vitro cytotoxicity assay demonstrated that compound 7 with a carboxyamide linker dramatically inhibited the growth of various cancer cells with a GI ₅₀ of less than 1μM, whereas compound 6 with a charged linker did not show any obvious toxicity in all cell lines tested. These data indicated that a neutral carboxyamide linker is preferable for developing nitrogen mustard prodrugs. Our results showed that 7 is a potent anticancer prodrug that can serve as a model compound for further development. We believe these novel aromatic nitrogen mustards will inspire further and effective applications. H₂O₂-triggered alkylating agents: H₂O ₂-triggered DNA cross-linking agents were developed. The activity of the aromatic nitrogen mustard prodrugs is masked by introducing an electron-withdrawing linker unit that sufficiently deactivates the nitrogen mustard, whereas H₂O₂ triggered the conversion of the electron-withdrawing group to a donating group, releasing the active species, which effectively cross-linked DNA (see scheme).

Full text of DOI:10.1002/chem.201400090

DOI: 10.1002/chem.201400090
Full Paper
&
Anticancer Drugs
Aromatic Nitrogen Mustard-Based Prodrugs: Activity, Selectivity,
and the Mechanism of DNA Cross-Linking
Wenbing Chen, Yanyan Han, and Xiaohua Peng*[a]
Abstract: Three novel H2O2-activated aromatic nitrogen
mustard prodrugs (6–8) are reported. These compounds
contain a DNA alkylating agent connected to a H2O2-respon-
sive trigger by different electron-withdrawing linkers so that
they are inactive towards DNA but can be triggered by H2O2
to release active species. The activity and selectivity of these
compounds towards DNA were investigated by measuring
DNA interstrand cross-link (ICL) formation in the presence or
absence of H2O2. An electron-withdrawing linker unit, such
as a quaternary ammonia salt (6), a carboxyamide (7), and
a carbonate group (8), is sufficient to deactivate the aromat-
ic nitrogen mustard resulting in less than 1.5% cross-linking
formation. However, H2O2 can restore the activity of the ef-
fectors by converting a withdrawing group to a donating
group, therefore increasing the cross-linking efficiency
(>20%). The stability and reaction sites of the ICL products
were determined, which revealed that alkylation induced by
7 and 8 not only occurred at the purine sites but also at the
pyrimidine site. For the first time, we isolated and character-
ized the monomer adducts formed between the canonical
nucleosides and the aromatic nitrogen mustard (15) which
supported that nitrogen mustards reacted with dG, dA, and
dC. The activation mechanism was studied by NMR spectro-
scopic analysis. An in vitro cytotoxicity assay demonstrated
that compound 7 with a carboxyamide linker dramatically
inhibited the growth of various cancer cells with a GI50 of
less than 1 mm, whereas compound 6 with a charged linker
did not show any obvious toxicity in all cell lines tested.
These data indicated that a neutral carboxyamide linker is
preferable for developing nitrogen mustard prodrugs. Our
results showed that 7 is a potent anticancer prodrug that
can serve as a model compound for further development.
We believe these novel aromatic nitrogen mustards will in-
spire further and effective applications.
Introduction
H2O2 is generated endogenously. For example, boronate-based
probes have been used for selective detection and imaging of
hydrogen peroxide in cells.[7] Thus, boronate-based DNA cross-
linking agents may lead to a wide variety of new applications
in biology and the life sciences.
Chemical agents capable of inducing interstrand cross-links
(ICL) have attracted growing interest in chemistry, medicine,
and chemical biology. They are used for DNA damage and
repair studies and for nucleic acid detection.[1] Some DNA
cross-linking agents are used for cancer treatment. Over the
past few decades, several research groups have developed
novel chemical methods for inducing ICL formation, such as
photoirradiation,[2] reduction,[3] oxidation,[4] or fluoride induc-
tion.[5] Among these methods, oxidation-induction of ICL for-
mation under physiological conditions is probably least devel-
oped.[4] Recently, Greenberg and Zhou’s group described
a new way for producing ICLs in duplex DNA under mild oxi-
dative conditions by using phenyl selenide precursors.[4a–c] Our
group has shown that aryl boronic acids and esters can be
used as trigger units for developing ROS-activated DNA cross-
linking agents by hydrogen peroxide mediated oxidation.[6]
H2O2-induced ICL formation could be performed in cells as
It is well-known that cancer cells exhibit elevated intrinsic
oxidative stress.[8] Higher levels of H2O2 were found in cancer
cells than that in normal ones. Thus, the H2O2-inducible cross-
linking behaviors of boronate-based agents will provide
a novel strategy for tumor-specific damage, which will allow
the development of selective anticancer prodrugs. H2O2-acti-
vated DNA cross-linking agents should comprise two separate
functional domains: a ROS-accepting moiety (trigger) and a bi-
functional DNA alkylating agent (effector), joined by a “linker
system” in such a way that the reaction of the trigger causes
a large increase in the cytotoxic potency of the effector. The
identification of such compounds requires a systematic design
of the efficient triggers, the suitable linkers, and the effectors.
In addition, the overall characteristics of the entire prodrug re-
quire consideration as well. Recently, our group has reported
several arylboronate derivatives that can be activated by hy-
drogen peroxide to release either mechlorethamine or quinone
methides cross-linking DNA.[6] However, these molecules did
not show potent anticancer activity, which might be caused by
the presence of a quarternary ammonia salt linker. It is well-
known that the charged molecules cannot diffuse across cell
[a] Dr. W. Chen, Dr. Y. Han, Prof. X. Peng
Department of Chemistry and Biochemistry
University of Wisconsin-Milwaukee
3210 N. Cramer St, Milwaukee, WI 53211 (USA)
E-mail: pengx@uwm.edu
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201400090.
Chem. Eur. J. 2014, 20, 7410 – 7418
7410
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
membranes, which is not suit-
able for drug development. In
addition, the structures of the
cross-linked DNA adducts were
not determined. Thus, there is
considerable scope for develop-
ing more potent ROS-activated
DNA cross-linking agents and
providing detailed mechanisms
for DNA damages induced by
these compounds.
Scheme 1. Selective DNA cross-linking agents with a ROS-responsive ‘trigger-linker’ system. The activity of the ni-
trogen mustards is masked by introducing an electron-withdrawing ‘linker’ in the prodrug, while reaction of the
trigger unit with ROS converts an electron-withdrawing ‘linker’ to a donating group resulting in a potent ROS-acti-
vated DNA cross-linking agent 2 (‘trigger-linker’ circle changes from black–white to a homogeneous black color).
In this work, we investigated
the effect of boronates on the
activity of aromatic nitrogen
mustards that are the major
class of traditional alkylating
agents that cause cell death by
inducing DNA cross-links.[1a,9]
Some of them are the frequently
used antitumor agents in the
clinic, such as bendamustine,
melphalan, and chlorambucil.
However, like all other antican-
cer agents, their clinical efficacy
has been limited by the toxicity
to normal tissues. One of the ef-
forts to improve the selectivity
of such agents towards cancer
cells is to seek a unique linker-
Scheme 2. Strategy of prodrug design and structures of compounds 6–8.
trigger system that can be cleaved under a tumor-specific mi-
croenvironment (Scheme 1). For example, a series of dinitro-
benzamide nitrogen mustard prodrugs with a nitryl-amidogen
linker were exploited to selectively target hypoxic tumors.[3]
The selectivity was achieved by conversion of an electron-with-
drawing group in the benzene ring to a donating group specif-
ically in cancer cells, which activated the nitrogen mustard. In
ing linker units. However, the reaction of the aryboronates
with H2O2 can activate the aromatic nitrogen mustards by con-
verting a para electron-withdrawing to a donating group.
Thus, such compounds are expected to be potent selective al-
kylating agents towards H2O2.
spite of the advances achieved in this field, the development Results and Discussion
of prodrugs targeting specific tumor cells is still in high
Synthesis
demand. Herein, we report novel aromatic nitrogen mustard
prodrugs with a H2O2-cleavable trigger bonded to three differ-
ent linkers. These compounds showed selective DNA cross-link-
ing ability under ROS-containing conditions and enhanced an-
ticancer activity. In addition, the reaction sites of these nitro-
gen mustards were determined by isolating the alkylating
products as well as by varying the DNA sequences.
To test our hypothesis, we chose to prepare compounds 6–8
containing three different electron-withdrawing linker units in-
cluding a quaternary ammonia salt (6), a carboxyamide (7),
and a carbonate group (8). The synthesis of these prodrugs
was accomplished in a straightforward fashion from the corre-
sponding aromatic nitrogen mustards 10, 13, or 15, respective-
ly (Scheme 3). Quaternary ammonia salt 6 was obtained from
benzyl bromide 11 and compound 10 that was synthesized
from the aniline derivative 9[12] by the Eschweiler–Clarke meth-
ylation. Treatment of benzyl alcohol 12 with isocyanatoben-
zene 13[13] produced 7 with a quantitative yield. The reaction
of benzyl chloroformate 14[7f] with phenol derivative 15[12] re-
sulted in compound 8.
H2O2, a common ROS in cancer cells, is well known to react
with arylboronic ester 3 under physiological conditions to gen-
erate phenol derivative 4 that subsequently releases com-
pound 5.[7d,e,10,11] As the activity of nitrogen mustard depends
on the availability of lone-pair electrons on the nitrogen
atom,[6b] we expect that the cross-linking ability of the aromatic
nitrogen mustard can be tuned by varying the aromatic sub-
stituents. On the basis of this, we designed structure 3 with
the aromatic nitrogen mustard (effector) connected to the aryl-
boronate (trigger) by an electron-withdrawing group (linker)
(Scheme 2). These compounds are expected to be inactive as
the activity of nitrogen mustards is shielded by the withdraw-
DNA cross-linking
To identify the effective linkers capable of masking the toxicity
of the nitrogen mustard and triggering its activity upon H2O2
Chem. Eur. J. 2014, 20, 7410 – 7418
www.chemeurj.org
7411
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
pounds 6–8 (1 mm) (Figure 1,
lanes 2, 4, 6). These data indicat-
ed that the three electron-with-
drawing linkers are sufficient to
deactivate the aromatic nitrogen
mustards. To our delight, addi-
tion of H2O2 greatly increased
the ICL yields to 20% (Figure 1,
lanes 3, 5 and 8). Even at a lower
concentration (300 mm) of 6, the
ICL yield was increased about
five times (10.8%) in comparison
with that without H2O2. In a con-
trol experiment in which only
H2O2 was added, ICL formation
was not observed (Figure 1,
lane 1). Obviously, H2O2 is capa-
ble of triggering the activity of
these prodrugs. As we proposed,
Scheme 3. Synthesis of compounds 6–8.
the conversion of an electron-
withdrawing boron group to
a donating hydroxyl group by
treatment, we first investigated the activity and selectivity of
these compounds by determining their ability to form DNA
ICLs in the presence or absence of H2O2. Initially, a 49-mer DNA
duplex 16 was employed in the DNA cross-linking experiments
that were carried out under physiological conditions (Figure 1).
ICL formation and cross-linking yields were analyzed by dena-
turing polyacrylamide gel electrophoresis (PAGE) with phos-
phorimager analysis (Image Quant 5.2) by taking advantage of
the differing mobility of ICL products and single-stranded DNA.
The selectivity of the ICL formation induced by compounds
6–8 towards H2O2 is illustrated in Figure 1. In the absence of
H2O2, very few DNA ICLs (1.0–1.5%) were observed for com-
H2O2 produces the phenol derivatives greatly increasing the
electron density of mustard nitrogen and therefore facilitating
the ICL formation.
Subsequently, we investigated the effect of drug concentra-
tion, ratio of drug to H2O2, and pH value. In the presence of
H2O2, DNA cross-linking induced by compounds 6–8 was ob-
served at a concentration as low as 10 mm (about 2.4% ICL
yield) and 1 mm of drugs resulted in 20–26% of cross-linked
DNA (see Figure S1 and Table S1 in the Supporting Informa-
tion). It is known that a single unrepaired ICL is sufficient to kill
an eukaryotic cell and approximately 40 unrepaired ICLs can
kill a mammalian cell.[14] So we believe that these compounds
could be potent selective alkylating agents. A higher ratio of
H2O2 to drug (from 1:1 to 2:1) resulted in more efficient ICL for-
mation (see Table S2 and Figure S2 in the Supporting Informa-
tion). The H2O2-inducible activity of 7 and 8 is pH-dependent.
The most efficient ICL formation was observed under physio-
logical pH (7.0–7.5). Acidic conditions (pH 5 and 6) and basic
conditions (pH 8–9) resulted in lower ICL yields (see Table S3
and Figure S3 in the Supporting Information). It is likely that
acidic conditions suppressed oxidative deboronation[15] and
the activity of the nitrogen mustards,[16] while basic conditions
increased the coordination of boronate ester and boronic acid
with hydroxide anion, resulting in a negative charge in the
boron group, which slightly blocked the reaction of boron
with H2O2.[10]
The time course study of ICL formation induced by com-
pounds 7 and 8 showed that the DNA cross-linking formation
was completed within 24 h (Figure 2). The ICL growth followed
first-order kinetics with a rate constant (kICL) of 2.25Æ0.10ꢁ
10À5 sÀ1 for 7 and 2.69Æ0.25ꢁ10À5 sÀ1 for 8, respectively.
It is well-known that ICL induced by nitrogen mustards
mainly occurred at N-7 of dG,[17] thus the stability and reaction
sites of the ICL products were examined to provide further in-
sight into the reactivity of compounds 7 and 8. We isolated
Figure 1. Selective DNA cross-linking ability of 6–8 towards H2O2. Phosphor-
image autoradiogram of denaturing PAGE analysis of the cross-linking reac-
tion of DNA duplex 16 in the presence of 6–8 (1.0 mm for lanes 1–6 and 8;
0.3 mm for lane 7) with or without H2O2 (all reactions were carried out at
room temperature). Lane 9, DNA marker.
Chem. Eur. J. 2014, 20, 7410 – 7418
www.chemeurj.org
7412
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 3. Phosphorimage autoradiogram of 20% denaturing PAGE analysis
of the isolated ICL product and single-stranded DNA upon heating in piperi-
dine or phosphate buffer (The ICL product and single-stranded DNA were
isolated from reaction of duplex 16 with 7 or 8 in the presence of H2O2, 16a
was radiolabeled at the 5’-terminus). Lanes 1-3: single-stranded DNA in-
duced by 7/H2O2; lanes 4-6: ICL induced by 7/H2O2; lanes 7–9: single-strand-
ed DNA induced by 8/H2O2; lanes 10-12: ICL induced by 8/H2O2; lanes 1, 4,
7, and 10: control (no treatment); lanes 2, 5, 8, and 11: treated by heating at
908C in buffer (pH 7.0); lanes 3, 6, 9, and 12: treated by heating at 908C in
piperidine; lane 13: G+A sequencing of 16a; lane 14: Fe·EDTA
Figure 2. Rate of ICL growth from 16 upon treatment with 7/H2O2 (a) or 8/
(EDTA=ethylenediaminetetraacetate) treatment of 16a.
H2O2 (b).
the ICL products and monoalkylated single-stranded DNA,
which were heated in a phosphate buffer (pH 7) or in 1.0m pi-
peridine (908C). Piperidine is known to induce cleavage with
N-7 alkylated purines according to the Maxam–Gilbert se-
quencing procedure.[4c,18] The DNA ICLs were completely de-
stroyed after heating for 30 min, which led to obvious cleavage
bands at dGs and dAs, such as G97, G96, G90, A89, G71, G58,
A57, G52, G44, G40, A39-37, A31, G27, G22, A18, A15, A14, and
G6 (Figure 3 and Figures S4–S6). Our results indicated that the
alkylation induced by 7 and 8 mainly occurred at the purine
sites, which is consistent with previous observation. Similar
cleavage patterns were observed with single-stranded DNA
which showed that in addition to ICL formation, compounds 7
and 8 also induced an intrastrand cross-link that is deleterious
to cells as well.
yield obtained from duplex 17 (21-mer) than from 16 (49-mer)
is shorter DNA sequences. To the best of our knowledge, this
is the first report that nitrogen mustard derivatives could also
cross-link pyrimidines.[17]
Nucleoside alkylation selectivity
To investigate whether the ICL could occur with pyrimidines,
we tested the ICL reaction with duplex 17 containing dCs/dTs
in one strand and dGs/dAs in the complimentary strand. Obvi-
ous ICL formation was observed when duplex 17 was treated
with 7/H2O2 (7.5%) or 8/H2O2 (11%). Although the ICL yields
with 17 were not as high as that with 16, these results clearly
showed that ICL induced by 7 or 8 not only occurred at purine
sites but also at pyrimidine sites (see Figure S7 in the Support-
ing Information). One of the possible reasons for the lower ICL
To gather the direct proof for the DNA cross-linking sites and
to examine the reactivity of these drugs towards DNA, we per-
formed a monomer reaction by treating the activated drug
with four canonical nucleosides (dA, dC, dG, and dT) in
a DMSO/PBS (pH 8) solution. Since the reaction of 8 with H2O2
releases compound 15 that is the direct DNA cross-linking
agent, 15 was employed in the monomer reaction. After
a two-day reaction, the new products were observed with dG,
dC, and dA but not with dT. After separation and purification,
Chem. Eur. J. 2014, 20, 7410 – 7418
www.chemeurj.org
7413
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
which underwent an intramolecular nucleophilic sub-
stitution reaction to form highly electrophilic inter-
mediate 15A. It is well known that N-7 of dG is the
most active alkylation site among all nucleosides.
Thus, dG in DNA easily reacts with 15A generating
monoalkylated single-stranded adduct 2b. Subse-
quently, other nucleotides (dG, dA, or dC) in the com-
plementary strand react with 21b producing DNA
ICLs (Scheme 5). N-7 alkylated dG adducts are labile
under basic conditions or in the presence of strong
nucleophiles such as piperidine. For example, a degly-
cosylation of 21a,b easily occurred generating 18
and sugar residues 22a,b or 23a,b under these con-
ditions (Scheme 5). Compound 23b further induced
strand cleavage after b-elimination. Therefore, cleav-
age bands were observed at dGs when monoalkyl-
ated ODNs or ICLs were heated in 1.0m piperidine
(Figure 3).
Scheme 4. Reactions of mononucleosides with compound 15.
An NH2-N6 alkylated dA adduct 20 might result
from the rearrangement of the N-1 adduct 24a by
a proposed mechanism shown in Scheme 6.[4c,19a,d,e] It was re-
ported that the basicity of the ring nitrogen of N9-substituted
adenine decreased in the order N1>N7>N3.[20] Thus, alkyla-
tion with 15A preferentially occurred at N1 of dA providing
adduct 24b, which further reacted with other nucleobases in
the complementary strand to form DNA ICLs. Meanwhile, the
N1-alkylated adduct 24a,b can rearrange to the N6-alkylade-
nine 25a,b, which is facilitated under basic conditions
(Scheme 6).[19] However, we could not exclude that the DNA al-
kylation might also occur at N7 or N3 of dA in DNA.[19c]
There are very few reports about the reaction of nitrogen
mustards with pyrimidine nucleosides.[17] We found that the ar-
omatic nitrogen mustard 15 can react with dC but
not with dT. The structure of the monomer adduct
19 suggested that the alkylation of dC occurred at
N4 (Scheme 4).
18 (34%), 19 (24%), and 20 (24%) were obtained and charac-
terized by NMR spectroscopy and HRMS (Scheme 4). Collective-
ly, our results suggested that compounds 7 and 8 could alkyl-
ate dC, dG, and dA upon hydrogen peroxide treatment. Thus,
we conclude that the ICL induced by the aromatic nitrogen
mustards occurred with both purines and pyrimidines.
On the basis of the monomer reaction, the mechanism for
DNA alkylation was proposed (Schemes 5 and 6). We propose
that compound 18 resulted from a deglycosylation of the cor-
responding N7 adduct of dG (21a) (Scheme 5). A similar mech-
anism was proposed for the DNA ICL reaction induced by com-
pound 8. Treatment of the inactive 8 with H2O2 generated 15
Determination of the activation mechanism by
NMR spectroscopic analysis
NMR spectroscopic analysis of the monomer reac-
tions was used to determine the activation mecha-
nism of compounds 7 and 8 by H2O2 (Scheme 7). The
reaction of 7 or 8 with H2O2 was carried out in a mix-
ture of 1 mm deuterated potassium phosphate buffer
(pH 8.0) (50 mL) and [D6]DMSO (450 mL). Without addi-
tion of H2O2, hydrolysis of boronate ester 8 easily oc-
curred in the phosphate buffer yielding boronic acid
8A and pinacol. This was evident by the appearance
of C4’,6’ÀH (8A, d=7.78–7.77 ppm) and C8’ÀH (pinacol,
d=1.05 ppm) (see Figure S8 in the Supporting Infor-
mation). After the addition of H2O2, oxidative
deboronation occurred leading to the formation of
8C and 4,4,5,5-tetramethyl[1,3,2]dioxaborolan-2-ol (B)
(d=1.13 ppm, C8’ÀH), which was further hydrolyzed
to pinacol. The formation of the direct alkylating
Scheme 5. Mechanism for DNA alkylation at dG and deglycosylation of N7-alkylated dG
adduct under basic conditions.
agent 15 was indicated by the appearance of d=
Chem. Eur. J. 2014, 20, 7410 – 7418
www.chemeurj.org
7414
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The linker units developed are
suitable for the future design of
ROS-activated DNA cross-linking
agents.
DNA cross-links towards differ-
ent ROS
Finally, we investigated the activ-
ity of compounds 7 and 8 to-
wards other ROS including tert-
butylhydroperoxide (TBHP), hy-
pochlorite (OClÀ), hydroxyl radi-
C
cal (HO ), tert-butoxy radical
(tBuO ), and superoxide (O2À).
C
Among these, hydrogen perox-
ide is the most efficient ROS that
can trigger the activity of these
prodrugs, while TBHP, OClÀ, and
superoxide also slightly activate
them (Figure 4). In the presence
of H2O2, these compounds in-
duced about 20% ICL formation,
while less than 5% ICL yields
were observed with OClÀ, TBHP,
and superoxide). No ICL forma-
tion was induced by hydroxyl
radical and tert-butoxy radical.
The selectivity of arylboronate
esters towards hydrogen perox-
ide is consistent with previous
reports.[6]
Scheme 6. DNA alkylation induced by 8 occurred at dA.
Cytotoxicity assay
As two representatives, the cyto-
toxicity of compounds 6 and 7
were further evaluated in biolog-
Scheme 7. Activation of 7 and 8 by H2O2.
7.11–6.68 ppm for C2,3,5,6ÀH and d=3.61–3.56 for C7,8ÀH. The
quinone methide intermediate D rapidly hydrolyzed to 8E
which was evidenced by appearance of C1’ÀH (d=4.33 ppm)
and C3’,4’,6’,7’ÀH (d=7.11–6.68 ppm). The formation of 8E was
proved by an authentic sample obtained from the reaction of
12 with H2O2, the peaks of which matched exactly with that of
8E and pinacol (see Figure S9 in the Supporting Information).
The conversion of 8 to 8E was so fast that about 73% of 8E
was generated within 30 min, and more than 93% of 8 was
consumed within 2 h. These data showed that the aryl-
boronates developed in this work are efficient H2O2-responsive
trigger units. Different from 8, formation of 8E was not ob-
served with 7 upon treatment with H2O2. Instead, the final
product 7E was produced by the reaction of D and 9 possibly
due to the presence of a more nucleophilic amino group or
from an intramolecular reaction of 7C (see Figure S10 in the
Supporting Information). Overall, the prodrugs developed in
this work are sensitive to H2O2 under physiological conditions.
Figure 4. ICL formation induced by 7–8 (1 mm) upon treatment with various
&
&
ROS. : 7; : 8.
Chem. Eur. J. 2014, 20, 7410 – 7418
www.chemeurj.org
7415
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ical systems. Initially, their ability for inhibiting cancer cell
growth was determined at a single dose of 10 mm in the cell
lines of leukemia, non-small cell lung cancer, colon cancer, CNS
cancer, melanoma, ovarian cancer, renal cancer, prostate
cancer, and breast cancer. Compound 6 did not show obvious
toxicity in all cell lines tested, while compound 7 induced sig-
nificant growth inhibition of most cancer cell lines (see Fig-
ure S11–S12 in the Supporting Information). The growth per-
centage of most cell lines treated with 7 was less than 50%.
These data indicated that a neutral carboxyamide linker is pref-
erable for developing ROS-activated prodrugs. The toxicity of 7
was further evaluated against the 60 cell line panel at five con-
centration levels to determine the GI50. The GI50 of 7 in most
cancer cell lines is less than 1 mm (Table 1). In particular, higher
toxicities by 7 were observed towards leukemia, lung cancer,
breast cancer, and renal cancer. It was reported that these cell
lines have high intracellular concentrations of ROS that can
more efficiently activate the ROS-activated prodrugs.[8] Our re-
sults showed that 7 is a potent anticancer prodrug that can be
used as a lead compound for further development.
and selectivity have been determined by measuring DNA ICLs
with or without H2O2 as well as by evaluating their ability to in-
hibit cancer cell growth. Less than 2% of ICL products were
observed with these prodrugs in the absence of H2O2 while
the ICL yields increased to more than 20% with the addition of
H2O2. The stability and reaction sites of the ICL products were
also determined. We, for the first time, reported that alkylation
induced by nitrogen mustards not only occurred at the purine
sites but also at pyrimidine sites. This is supported by isolation
and determination of the monomer adducts formed between
nitrogen mustard analogue 15 and three canonical nucleosides
dA, dC, and dG. Compound 6 with a charged linker showed
less toxicity than those containing a neutral linker unit. Among
the prodrugs tested, compound 7 with a carboxyamide linker
displayed the highest cytotoxicity with a GI50 of less than 1 mm
for most cell lines tested. Our results showed that H2O2-activat-
ed DNA cross-linking agent 7 is a potent anticancer prodrug
that can be used as a lead compound for further development.
In addition, a carboxyamide linker is preferable for developing
ROS-activated prodrugs. This study reveals a novel way of cre-
ating targeted anticancer prodrugs to improve the therapeutic
effectiveness and selectivity of current anticancer agents.
Table 1. Cytotoxicities induced by 7.
Tumor type
Cell Line
GI50 [mm]
Experimental Section
CCRF-CEM
HL-60(TB)
MOLT-4
SR
A549/ATCC
HOP-92
NCI-H23
NCI-H460
NCI-H522
SF-268
0.63
0.62
1.73
0.44
3.21
0.39
2.05
0.53
2.64
1.73
1.80
2.20
0.73
1.8
3.86
1.89
2.26
2.72
0.57
0.48
1.43
2.50
2.19
0.65
1.08
N1,N1-Bis(2-chloroethyl)benzene-1,4-diamine (10)
leukemia
A mixture of 9 (2.5 g, 15 mmol) and sodium borohydride (1.4 g,
37.5 mmol) was added to a solution of formalin (3.7 g, 37.5 mmol)
and conc. H2SO4 (3 mL) in THF. The resulting mixture was stirred at
room temperature for 1.5 h. After being quenched with NaOH so-
lution, extracted with dichloromethane, and washed with water,
the mixture was dried over Na2SO4 followed by filtration, and the
residue was purified by column chromatography (hexane/ethyl
acetate=6:1) to afford 10 (75% yield). 1H NMR (CDCl3, 300 MHz):
d=2.88 (s, 6H), 3.62 (d, J=3.6, 8H), 6.73–6.81 ppm (m, 4H);
13C NMR (CDCl3, 75 MHz): d=40.9, 42.1, 54.5, 115.3, 115.9 ppm;
HRMS-ES: m/z: calcd for C12H19N2Cl2: 261.0290 [M+H]+; found:
261.0920.
non-small cell lung cancer
SF-295
SF-539
CNS
SNB-75
LOX IMVI
M14
UACC-257
UACC-62
786-0
ACHN
CAKI-1
RXF 393
UO-31
DU-145
MCF7
melanoma
4-[Bis(2-chloroethyl)amino]-N,N-dimethyl-N-[4-(4,4,5,5-tetra-
methyl-1,3,2-dioxaborolan-2-yl)benzyl]benzenaminium bro-
mide (6)
renal
Compound 10 (520 mg, 2 mmol) and 11 (600 mg, 2 mmol) were
mixed in acetonitrile (10 mL) at room temperature and the mixture
was stirred overnight. After removal of the solvent, ether (5 mL)
and CH2Cl2 (5 mL) were added and the product was filtered and
washed with more ether (10 mL). Compound 6 (1.0 g, 89% yield);
1H NMR (CDCl3, 300 MHz): d=1.28 (s, 12H), 3.51 (s, 6H), 3.76–3.81
(m, 8H), 4.98 (s, 2H), 6.88 (d, J=9.3, 2H), 7.10 (d, J=8.1, 2H), 7.58–
7.62 ppm (m, 4H); 13C NMR (CDCl3, 75 MHz): d=25.1, 41.6, 52.0,
53.3, 72.0, 84.5, 112.4, 123.1, 131.8, 132.5, 134.1, 134.8, 147.7 ppm;
HRMS-ES: m/z: calcd for C25H36BN2O2Cl2: 477.2241 [MÀBr]+; found:
477.2262.
prostate
breast
MDA-MB-468
Conclusion
We have developed a potent ROS-activated anticancer prodrug
that shows high cytotoxicity. The aromatic nitrogen mustard
was employed as the effector, the activity of which is masked
in the prodrugs by connecting to an arylboronate by an elec-
tron-withdrawing linker. Such nontoxic prodrugs can be acti-
vated by H2O2 that converts the electron-withdrawing boron-
ate group to a donating hydroxyl group. The activation mecha-
nism was studied by NMR spectroscopic analysis. The activity
4-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)benzyl 4-
[bis(2-chloroethyl)amino]phenylcarbamate (7)
Compound 12 (468 mg, 2 mmol) and TEA (323 mL, 2.5 mmol) were
dissolved in dichloromethane (20 mL) at 08C under N2, and then
Chem. Eur. J. 2014, 20, 7410 – 7418
www.chemeurj.org
7416
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
a solution of 13 (645 mg, 2.5 mmol) in dichloromethane (5 mL)
was added dropwise. After stirring at the same temperature for
3 h, the solvent was evaporated, and the residue was purified by
column chromatography (hexane/ethyl acetate=5:1) to afford 7
(2R,3S,5R)-5-(6-{2-[(2-Chloroethyl)(4-hydroxyphenyl)amino]
ethylamino}-9H-purin-9-yl)-2-(hydroxymethyl)tetrahydrofur-
an-3-ol (20)
1
1
Yield: 24%; H NMR ([D6]DMSO, 500 MHz): 2.36–2.38 (m, 1H), 2.58–
(95% yield). H NMR (CDCl3, 300 MHz): d=1.37 (s, 12H), 3.63 (t, J=
2.61 (m, 1H), 3.47–3.48 (m, 2H), 3.52–3.53 (m, 1H), 3.53–3.55 (m,
2H), 3.59–3.60 (m, 1H), 3.61–3.64 (m, 2H), 3.89–3.90 (m, 1H), 4.40–
4.41 (m, 1H), 4.51–4.52 (m, 2H), 5.10 (s, 1H), 5.50 (s, 1H), 6.32 (t,
J=6.0, 1H), 6.57 (d, J=8.5, 2H), 6.61 (d, J=8.5, 2H), 8.31 (s, 1H),
8.72 (s, 1H), 8.95 (brs, 1H), 9.85 ppm (brs, 1H); 13C NMR (CDCl3,
125 MHz): d=42.1, 48.8, 49.0, 52.5, 54.2, 61.6, 70.8, 84.5, 88.7,
116.2, 116.3, 117.3, 119.4, 139.8, 142.9, 146.6, 148.3, 150.4,
151.1 ppm; HRMS-ESI/APCI: m/z: calcd for C20H27ClN6O4: 449.1699
[M+H]+; found: 449.1704.
6.3, 4H), 3.72 (t, J=6.3, 4H), 6.51 (brs, 1H), 6.67 (d, J=8.7, 2H),
7.26 (brs, 1H), 7.41 (d, J=7.8, 2H), 7.42 (d, J=7.8, 2H), 7.84 ppm
(d, J=7.8, 2H); 13C NMR (CDCl3, 75 MHz): d=24.9, 40.5, 53.8, 66.8,
83.9, 112.8, 121.5, 127.3, 135.0, 139.2 ppm; HRMS-ES: m/z: calcd for
C24H32BN2O4Cl2: 493.1827 [M+H]+; found: 493.1841.
4-[Bis(2-chloroethyl)amino]phenyl 4-(4,4,5,5-tetramethyl-
1,3,2-dioxaborolan-2-yl)benzyl carbonate (8)
A solution of 14 (466 mg, 2 mmol) in dichloromethane (2 mL) was
added dropwise to a mixture of 15 (1.2 g, 4 mmol), dry pyridine
(320 mL, 4 mmol), and 4-dimethylaminopyridine (DMAP) (24 mg,
0.2 mmol) in dichloromethane (50 mL) at 08C. After stirring for
13 h, 1n HCl (aq.) was added to the resulting mixture and organic
phase was separated and the water phase was extracted with di-
chloromethane. The combined organic phase was dried over
Na2SO4, evaporated and purified by column chromatography
Acknowledgements
We are grateful for the financial support for this research from
the National Cancer Institute (grant: 1R15A152914–01), Chronic
Lymphocytic Leukemia Research Consortium (grant: 2 PO1 CA
81534), the Great Milwaukee Foundation (Shaw Scientist
Award), and the UWM Research Growth Initiative (grant:
RGI101X234).
(hexane/ethyl acetate=5:1) to afford
8
(33% yield). 1H NMR
([D6]DMSO, 300 MHz): d=1.31 (s, 12H), 3.72 (s, 8H), 5.27 (s, 2H),
6.76 (d, J=9.0, 2H), 7.06 (d, J=9.0, 2H), 7.45 (d, J=8.1, 2H),
7.72 ppm (d, J=8.1, 2H); 13C NMR (CDCl3, 75 MHz): d=25.1, 41.5,
52.7, 69.8, 84.2, 122.7, 122.4, 127.8, 135.1, 138.9, 142.3, 144.9,
154.0 ppm; HRMS-ES: m/z: calcd for C24H32BNO5Cl2: 494.1667
[M+H]+; found: 494.1657.
Keywords: anticancer drugs · aromatic nitrogen mustard ·
DNA · cross-linking · hydrogen peroxide
[1] a) D. M. Noll, T. M. Mason, P. S. Miller, Chem. Rev. 2006, 106, 277–301;
b) M. Weng, Y. Zheng, V. Jasti, E. Champeil, M. Tomasz, Y. Wang, A. Basu,
M. Tang, Nucleic Acids Res. 2010, 38, 6976–6984; c) X. Peng, M. M.
Greenberg, Nucleic Acids Res. 2008, 36, e31; d) X. Peng, A. Ghosh, B. V.
Houten, M. M. Greenberg, Biochemistry 2010, 49, 11–19; e) J. Wang, H.
Cao, C. You, B. Yuan, S. Gupta, C. Nishigori, L. J. Niedernhofer, P. J.
Brooks, Y. Wang, Nucleic Acids Res. 2012, 40, 7368–7374; f) Y. Jiang, H.
Hong, H. Cao, Y. Wang, Biochemistry 2007, 46, 12757–12763; g) Q.
Zhang, Y. Wang, J. Am. Chem. Soc. 2003, 125, 12795–12802; h) H. Sun,
X. Peng, Bioconjugate Chem. 2013, 24, 1226–1234.
Reactions of mononucleosides with compound 15
Compound 15 (47 mg, 0.2 mmol) and dA, dC, dG, or dT (0.1 mmol)
were dissolved in a mixture of DMSO/H2O (500/50 mL) respectively,
followed by the addition of PBS (pH 8.0, 5 mL, 1m). The mixtures
were shaken for 2 days at room temperature. After evaporation of
the solvent, the residues were purified by column chromatography
(dichloromethane/CH3OH=8:1–4:1) to afford the adducts.
[2] a) P. Wang, R. Liu, X. Wu, H. Ma, X. Cao, P. Zhou, J. Zhang, X. Weng, X.
Zhang, J. Qi, X. Zhou, L. Weng, J. Am. Chem. Soc. 2003, 125, 1116 – 1117;
b) S. N. Richter, S. Maggi, S. C. Mels, M. Palumbo, M. Freccero, J. Am.
Chem. Soc. 2004, 126, 13973–13979; c) S. Colloredo-Mels, F. Doria, D.
Verga, M. Freccero, J. Org. Chem. 2006, 71, 3889–3895; d) D. Verga, M.
Nadai, F. Doria, C. Percivalle, M. D. Antonio, M. Palumbo, S. N. Richter,
M. Freccero, J. Am. Chem. Soc. 2010, 132, 14625–14637; e) C. Percivalle,
A. L. Rosa, D. Verga, F. Doria, M. Mella, M. Palumbo, M. D. Antonio, M.
Freccero, J. Org. Chem. 2011, 76, 3096–3106; f) Y. Kuang, H. Sun, J. C.
Blain, X. Peng, Chem. Eur. J. 2012, 18, 12609–12613; g) F. Doria, S. N.
Richter, M. Nadai, S. Colloredo-Mels, M. Mella, M. Palumbo, M. Freccero,
J. Med. Chem. 2007, 50, 6570–6579.
2-Amino-7-{2-[(2-chloroethyl)(4-hydroxyphenyl)amino]ethyl}-
1H-purin-6(7H)one (18)
1
Yield: 34%; H NMR ([D6]DMSO, 300 MHz): d=3.43 (d, J=6.6, 2H),
3.60 (d, J=6.6, 4H), 4.26 (d, J=6.6, 2H), 6.15 (s, 2H), 6.63 (d, J=
9.3, 2H), 6.70 (d, J=9.3, 2H), 7.82 (s, 1H), 8.68 (s, 1H), 10.79 ppm
(s, 1H); 13C NMR (CDCl3, 125 MHz): d=41.8, 44.5, 52.4, 53.3, 108.3,
115.4, 116.4, 140.4, 144.2, 149.5, 153.1 ppm; HRMS-ESI/APCI: m/z:
calcd for C15H17ClN6O2: 425.1586 [M+H]+; found: 425.1593.
[3] a) G. J. Atwell, S. Yang, F. B. Pruijn, S. M. Pullen, A. Hogg, A. V. Patterson,
W. R. Wilson, W. A. Denny, J. Med. Chem. 2007, 50, 1197–1212; b) R. S.
Singleton, C. P. Guise, D. M. Ferry, S. M. Pullen, M. J. Dorie, J. M. Brown,
A. V. Patterson, W. R. Wilson, Cancer Res. 2009, 69, 3884; c) A. V. Patter-
son, D. M. Ferry, S. J. Edmunds, Y. Gu, R. S. Singleton, K. Patel, S. M.
Pullen, K. O. Hicks, S. P. Syddall, G. J. Atwell, S. Yang, W. A. Denny, W. R.
Wilson, Clin Cancer Res. 2007, 13, 3922; d) M. Di Antonio, F. Doria, M.
Mella, D. Merli, A. Profumo, M. Freccero, J. Org. Chem. 2007, 72, 8354–
8360.
[4] a) I. S. Hong, M. M. Greenberg, J. Am. Chem. Soc. 2005, 127, 10510–
10511; b) I. S. Hong, H. Ding, M. M. Greenberg, J. Am. Chem. Soc. 2006,
128, 485–491; c) X. Peng, I. S. Hong, H. Li, M. M. Seidman, M. M. Green-
berg, J. Am. Chem. Soc. 2008, 130, 10299–10306; d) X. Weng, L. Ren, L.
Weng, J. Huang, S. Zhu, X. Zhou, L. Weng, Angew. Chem. 2007, 119,
8166–8169; Angew. Chem. Int. Ed. 2007, 46, 8020–8023.
4-{2-[(2-Chloroethyl)(4-hydroxyphenyl)amino]ethylamino}-1-
[(2R,4S,5R)-4-hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-
yl]pyrimidin-2(1H)one (19)
Yield: 24%; 1H NMR ([D6]DMSO, 500 MHz): d=2.03–2.06 (m, 1H),
2.17–2.19 (m, 1H), 3.46–3.51 (m, 4H), 3.53–3.59 (m, 2H), 3.64–3.65
(m, 2H), 3.82 (d, J=3.0, 1H), 4.07–4.10 (m, 2H), 4.21 (s, 1H), 5.13 (s,
1H), 5.33 (d, J=4.5, 1H), 6.06 (dd, J=6.0, J=9.0, 1H), 6.62 (d, J=
9.0, 2H), 6.70 (d, J=9.0, 2H), 7.89 (s, 1H), 8.73 (s, 1H), 9.64 ppm
(brs, 1H); 13C NMR (CDCl3, 125 MHz): d=41.2, 41.9, 47.5, 49.7, 52.9,
61.2, 61.7, 70.2, 86.8, 88.4, 114.7, 115.0, 116.3, 116.4, 140.3, 140.9,
148.7, 149.8, 158.5, 170.9 ppm; HRMS-ESI/APCI: m/z: calcd for
C19H26ClN4O5: 349.1174 [M+H]+; found: 349.1189.
Chem. Eur. J. 2014, 20, 7410 – 7418
www.chemeurj.org
7417
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[5] a) W. F. Veldhuyzen, P. Pande, S. E. Rokita, J. Am. Chem. Soc. 2003, 125,
14005–14013; b) H. Wang, S. E. Rokita, Angew. Chem. 2010, 122, 6093–
6096; Angew. Chem. Int. Ed. 2010, 49, 5957–5960; c) J. Wu, R. Huang, T.
Wang, X. Zhao, W. Zhang, X. Weng, T. Tian, X. Zhou, Org. Biomol. Chem.
2013, 11, 2365–2369.
[6] a) C. Sheng, Y. Wang, X. Peng, Chem. Eur. J. 2012, 18, 3850–3854; b) Y.
Kuang, K. Balakrishnan, V. Gandhi, X. Peng, J. Am. Chem. Soc. 2011, 133,
19278–19281; c) S. Cao, Y. Wang, X. Peng, J. Org. Chem. 2014, 79, 501–
508; d) S. Cao, X. Peng, Curr. Org. Chem. 2014, 18, 70–85.
DiTucci, S. T. Phillips, Angew. Chem. 2012, 124, 12879–12882; Angew.
Chem. Int. Ed. 2012, 51, 12707–12710.
[12] H. B. Zhang, J. J. Xue, X. L. Zhao, D. G. Liu, Y. Li, Chin. Chem. Lett. 2009,
20, 680–683.
[13] N. Kapuriyaa, K. Kapuriya, H. Dong, X. Zhang, T. C. Chou, Y. T. Chen, T. C.
Lee, W. C. Lee, T. H. Tsai, Y. Naliapara, T. L. Su, Bioorg. Med. Chem. 2009,
17, 1264–1275.
[14] a) N. Magana-Schwencke, J. A. Henriques, R. Chanet, E. Moustacchi,
Proc. Natl. Acad. Sci. USA 1982, 79, 1722–1726; b) P. D. Lawley, D. H.
Phillips, Mutat. Res. 1996, 355, 13–40.
[7] a) E. W. Miller, A. E. Albers, A. Pralle, E. Y. Isacoff, C. J. Chang, J. Am.
Chem. Soc. 2005, 127, 16652–16659; b) B. C. Dickinson, C. J. Chang, J.
Am. Chem. Soc. 2008, 130, 9638–9639; c) E. W. Miller, O. Tulyathan, E. Y.
Isacoff, C. J. Chang, Nat. Chem. Biol. 2007, 3, 263–267; d) L. C. Lo, C. Y.
Chu, Chem. Commun. 2003, 2728–2729; e) D. Srikun, E. W. Miller, D. W.
Domaille, C. J. Chang, J. Am. Chem. Soc. 2008, 130, 4596–4597; f) C.
Chung, D. Srikun, C. S. Lim, C. J. Chang, B. R. Cho, Chem. Commun.
2011, 47, 9618–9620.
[8] a) T. P. Szatrowski, C. F. Nathan, Cancer Res. 1991, 51, 794–798; b) S.
Toyokuni, K. Okamoto, J. Yodoi, H. Hiai, FEBS Lett. 1995, 358, 1–3; c) P. T.
Schumacker, Cancer Cell 2006, 10, 175–176; d) S. D. Lim, C. Sun, J. D.
Lambeth, F. Marshall, M. Amin, L. Chung, J. A. Petros, R. S. Arnold, Pros-
tate 2005, 62, 200–207; e) M. Zieba, M. Suwalski, S. Kwiatkowska, G. Pia-
secka, I. Grzelewska-Rzymowska, R. Stolarek, D. Nowak, Respir. Med.
2000, 94, 800–805; f) N. S. Brown, R. Bicknell, Breast Cancer Res. 2001,
3, 323–327; g) D. Trachootham, H. Zhang, W. Zhang, L. Feng, M. Du, Y.
Zhou, Z. Chen, H. Pelicano, W. Plunkett, W. G. Wierda, M. J. Keating, P.
Huang, Blood 2008, 112, 1912–1922; h) G. K. Veni, D. B. Rao, M. Kumar,
B. Usha, M. Krishma, T. R. Rao, Rec. Res. Sci. Tech. 2011, 3, 55–58.
[9] S. R. Rajski, R. M. Williams, Chem. Rev. 1998, 98, 2723–2795.
[10] D. G. Hall, Structure, Properties, and Preparation of Boronic Acid Deriva-
tives. Overview of Their Reactions and Applications, in Boronic Acids:
Preparation and Applications in Organic Synthesis and Medicine, Wiley-
VCH, Weinheim, 2006.
[15] a) H. G. Kuivila, J. Am. Chem. Soc. 1954, 76, 870–874; b) H. G. Kuivila,
A. G. Armour, J. Am. Chem. Soc. 1957, 79, 5659–5662.
[16] A. Polavarapu, J. A. Stillabower, S. G. W. Stubblefield, W. M. Taylor, J. Org.
Chem. 2012, 77, 5914–5921.
[17] a) K. Hemmimki, S. Kallama, Reactions of nitrogen mustards with DNA in
Carcinogenicity of Alkylating Cytostatic Drugs (Eds: D. Schmahl and J. M.
Kaldor), IARC Publication No. 78, Oxford University Press, Oxford,
pp. 55–70, 1987; b) O. P. Pieper, L. C. Erickson, Carcinogenesis 1990, 11,
1739–1746; c) P. Brookes, P. D. Lawley, Biochem. J. 1961, 80, 496–503.
[18] a) A. M. Maxam, W. A. Gilbert, Proc. Natl. Acad. Sci. USA 1977, 74, 560–
564; b) K. Haraguchi, M. O. Delaney, C. J. Wiederholt, A. Sambandam, Z.
Hantosi, M. M. Greenberg, J. Am. Chem. Soc. 2002, 124, 3263–3269.
[19] a) J. W. Jones, R. Robins, J. Am. Chem. Soc. 1963, 85, 193–201; b) P. D.
Lawley, P. Brookes, Biochem. J. 1963, 89, 127–138; c) C. J. Burrows, J. G.
Muller, Chem. Rev. 1998, 98, 1109–1151; d) M. Freccero, R. Gandolfi, M.
Sarzi-Amade, J. Org. Chem. 2003, 68, 6411–6423; e) C. Percivalle, F.
Doria, M. Freccero, Current Organic Chemistry 2014, 18, 19–43.
ˇ
ˇ
[20] a) J. E. Sponer, J. Leszczynski, F. Glahꢂ, B. Lippert, J. Sponer, Inorg. Chem.
2001, 40, 3269–3278; b) G. Kampf, L. E. Kapinos, R. Griesser, B. Lippert,
H. Sigel, J. Chem. Soc. Perkin Trans. 1 2002, 2, 1320–1327; c) C.-J. Chang,
J. D. Gomes, S. R. Byrn, J. Org. Chem. 1983, 48, 5160–5164.
[11] a) J. L. Major Jourden, S. M. Cohen, Angew. Chem. 2010, 122, 6947–
6949; Angew. Chem. Int. Ed. 2010, 49, 6795–6797; b) G. G. Lewis, M. J.
Received: January 8, 2014
Published online on May 7, 2014
Chem. Eur. J. 2014, 20, 7410 – 7418
www.chemeurj.org
7418
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim