Influence of the metal complex-to-peptide linker on the synthesis and properties of bioactive CpMn(CO)3 peptide conjugates

Article abstract of DOI:10.1039/b916907e

By combining organometallic groups and peptides, a large number of conjugates with interesting new biological properties can be prepared. Especially, attachment to cell-penetrating peptides (CPP) that act as efficient cell delivery vehicles has come to the fore. However, the presence of the metal moiety in such systems can interfere with standard conjugate synthesis procedures which therefore need to be optimized for every new compound. In this work, we report on the preparation of six new cymantrene-sC18 peptide bioconjugates that were prepared by solid phase peptide synthesis (SPPS) techniques. The cymantrene complexes were chosen for their different linker to the peptide, to study the influence of the linker group on cellular uptake and cell viability of the conjugates. Interestingly, the attachment of the metal complex leads to a non-standard cleavage of the Rink amide linker used in the SPPS protocol under trifluoroacetic acid (TFA) treatment, resulting in peptide amides that are N-alkylated at the C-terminus. Furthermore, we found that depending on the type of cymantrene moiety attached, the formation of reactive carbocations which result from decomposition of the resin linker is facilitated and can alkylate the metal complex moiety. Both effects were analyzed by MS/MS studies and cleavage mixtures for efficient elimination of this byproduct formation were identified. Moreover, initial biological testing of the cytotoxicity of one of the bioconjugates gave promising results. Concentration-dependent cell viability studies of Cym1-sC18 on human MCF-7 breast adenocarcinoma cells gave an IC₅₀ value of 59.8 (± 6.7) μM and demonstrate their potential in anticancer chemotherapy. The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/b916907e

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Influence of the metal complex-to-peptide linker on the synthesis and
properties of bioactive CpMn(CO)3 peptide conjugates†
Katrin Splith,a Ines Neundorf,*a Wanning Hu,b Harmel W. Peindy N’Dongo,b Vera Vasylyeva,b Klaus Merzb and
Ulrich Schatzschneider*b
Received 19th August 2009, Accepted 8th December 2009
First published as an Advance Article on the web 22nd January 2010
DOI: 10.1039/b916907e
By combining organometallic groups and peptides, a large number of conjugates with interesting new
biological properties can be prepared. Especially, attachment to cell-penetrating peptides (CPP) that act
as efficient cell delivery vehicles has come to the fore. However, the presence of the metal moiety in such
systems can interfere with standard conjugate synthesis procedures which therefore need to be
optimized for every new compound. In this work, we report on the preparation of six new
cymantrene-sC18 peptide bioconjugates that were prepared by solid phase peptide synthesis (SPPS)
techniques. The cymantrene complexes were chosen for their different linker to the peptide, to study the
influence of the linker group on cellular uptake and cell viability of the conjugates. Interestingly, the
attachment of the metal complex leads to a non-standard cleavage of the Rink amide linker used in the
SPPS protocol under trifluoroacetic acid (TFA) treatment, resulting in peptide amides that are
N-alkylated at the C-terminus. Furthermore, we found that depending on the type of cymantrene
moiety attached, the formation of reactive carbocations which result from decomposition of the resin
linker is facilitated and can alkylate the metal complex moiety. Both effects were analyzed by MS/MS
studies and cleavage mixtures for efficient elimination of this byproduct formation were identified.
Moreover, initial biological testing of the cytotoxicity of one of the bioconjugates gave promising
results. Concentration-dependent cell viability studies of Cym1-sC18 on human MCF-7 breast
adenocarcinoma cells gave an IC50 value of 59.8 ( 6.7) mM and demonstrate their potential in
anticancer chemotherapy.
in the amino acid sequence without limitation to terminal sites
Introduction
or side-chain functionalities as in lysines and cysteines. However,
In the last few years, a steadily increasing number of systems with
the synthesis of metal-peptide conjugates often requires special
exciting new biological activity has been prepared by conjugation
conditions different from standard SPPS protocols due to the
of organometallic compounds to bio(macro)molecules.1-3 In par-
potential sensitivity of the metal functional group and overall
ticular, the attachment of metal–organic complexes to peptides
conjugate stability. Generally, there are three strategies to prepare
metal complex-peptide conjugates:13 the first one is to synthesize
peptide-ligand conjugates followed by complexation with metal
ions on a solid support or in solution.14 The second one is to
utilize an organometallic amino acid-like building block during the
SPPS chain elongation reaction.15 Finally, for metal compounds
very sensitive towards SPPS conditions, a post-labeling strategy
can be used, in which the metal–organic moiety is attached after
assembly of the peptide by coupling reactions orthogonal to the
peptide side chain functional groups.16,17
In this work, we describe the synthesis, characterization, and
preliminary study of the biological activity of organometallic
peptide conjugates composed of cymantrene carboxylic acids
with different linkers between the COOH and cyclopentadienyl
groups and sC18, a novel cell-penetrating peptide (CPP).18 CPPs
are delivery vectors for a great variety of different substances
and can help to circumvent the problem of limited transport of
bioactive molecules into their target cells.19-22 These peptides are
able to internalize without the need of a transporter or receptor
molecule and thereby capable of carrying different cargos into
cells. We recently reported the successful and efficient delivery
is a very promising approach since there exist numerous thera-
peutically active peptide sequences and their synthesis can easily
be achieved by means of solid-phase peptide synthesis (SPPS).
Potential applications include use as antitumor4-6 and antibacterial
agents7,8 aswellasradiopharmaceuticals.9-11 Furthermore, surface-
bound ferrocene-peptide conjugates were applied as sensors for
proteins.12
Metal complex-peptide conjugates can conveniently be pre-
pared by solid-phase peptide synthesis (SPPS). This offers several
advantages over solution techniques such as, for instance, the
flexible incorporation of the organometallic group at any position
aInstitut fu¨r Biochemie, Universita¨t Leipzig, Bru¨derstr. 34, D-04103, Leipzig,
Germany. E-mail: neundorf@uni-leipzig.de; Fax: (+)49 341 97-36909
bLehrstuhl fu¨r Anorganische Chemie I - Bioanorganische Chemie, Ruhr-
Univerista¨t Bochum NC 3/74, Universita¨tsstr. 150, D-44801, Bochum,
Germany. E-mail: ulrich.schatzschneider@rub.de; Fax: (+)49 234 32-14378
† Electronic supplementary information (ESI) available: Additional exper-
imental details. CCDC reference numbers 681218, 744989, 744990, and
744991. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/b916907e
2536 | Dalton Trans., 2010, 39, 2536–2545
This journal is
The Royal Society of Chemistry 2010
©

View Online
of nanocrystals, oligonucleotides, and gallium complexes into
cancer as well as primary cells using this strategy.23-25 Cymantrene
itself is an easy to functionalize and robust organometallic
biomarker which is stable in air and water and has been used
for the N-terminal modification of short peptides via SPPS.26
Furthermore, its utility as an IR label of bioactive peptides
was demonstrated since the strong C–O stretching vibrations
are not obscured by signals from the peptide.27 However, by
attachment to these simple short peptides, no cytotoxic effects
on different cancer cell lines could be observed. On the other
hand, conjugation of a cymantrene derivative to a more complex
CPP based on the peptide hormone human calcitonin (hCT) lead
to bioconjugates with promising biologicial activity. A modified
intracellular distribution pattern and significant cytotoxic effects
in the low micromolar concentration range were observed on
MCF-7 human breast cancer cells while the two constituents
themselves were inactive.28 To further study structure–activity
relationships in such a system, we have prepared cymantrene-
peptide bioconjugates based on the novel cell-penetrating peptide
sC18, which consists of residues 106–121 of the C-terminal
region of the cationic antimicrobial peptide cathelicidin (CAP18).
This fragment was recently identified as an effective carrier
peptide for small organic substances like fluorophors and toxic
peptide sequences.18 In addition, different linkers were introduced
between the cymantrene moiety and the peptide to determine
their potential influence on cellular uptake and cytotoxicity of
the conjugates.
Results and discussion
Synthesis of cymantrene carboxylic acids
A total of six different cymantrene carboxylic acids Cym1 to Cym6
were prepared with different linkers between the cyclopentadienyl
ring and the carboxylate group (Scheme 1).
Cym1 was synthesized by Friedel–Crafts acylation of cy-
mantrene with 2-chlorobenzoyl chloride and subsequent cleavage
of the resulting ketone with potassium tert-butoxide following
the procedure of Biehl and Reeves.29 Compounds Cym2 to Cym4
were also prepared by a Friedel–Crafts reaction of cymantrene
with the corresponding cyclic anhydrides in the presence of
aluminium chloride as already reported for Cym2.27 The synthesis
of Cym5 and Cym6, on the other hand, required the five-step
reaction sequence shown in Scheme 2. First, isophthalic or
terephthalic acid was converted to the dimethyl ester by treatment
with thionylchloride in methanol. Then, one methyl ester group
was selectively hydrolyzed by reaction with one equivalent of
potassium hydroxide in methanol. The free carboxylic group
of the monoester was then converted to the carboxylic acid
chloride by heating to reflux in thionylchloride (see ESI†).30,31 The
corresponding acid chloride was reacted with cymantrene in the
presence of aluminium chloride to obtain the functionalized half-
sandwich complexes in reasonable yield after chromatographic
work-up. Finally, the methyl ester was hydrolyzed with sodium
hydroxide in methanol to obtain Cym5 and Cym6 in good yield
and purity.
During the evaluation of different SPPS conditions, we iden-
tified a number of unexpected byproducts that formed after
treatment of the resin with trifluoroacetic acid. Detailed mass
spectrometric studies allowed us to rationalize the formation of
these species. These results will help to improve the preparation
of complex metal-peptide conjugates by solid-phase synthesis
methods and provide new insights into the potential reactivity
of these cytotoxic compounds.
Single crystals suitable for X-ray structure analysis could be
obtained for Cym3 as well as the three isomeric compounds
Cym4, Cym5, and Cym6 with either a 1,2-, 1,3- or 1,4-phenylene
linker. Relevant parameters are listed in Table 1. The molecular
structures are shown in Fig. 1A to 1D, respectively. The metrical
parameters of all four compounds (see Table 2) compare well with
those recently published for Cym2 and the crystal packing is also
Scheme 1 Functionalized cymantrene carboxylic acids Cym1 to Cym6 prepared for peptide conjugation.
The Royal Society of Chemistry 2010 Dalton Trans., 2010, 39, 2536–2545 | 2537
This journal is
©

View Online
Scheme 2 Five-step synthesis of functionalized half-sandwich complexes Cym5 and Cym6 by Friedel–Crafts acylation of cymantrene with carboxylic
acid chlorides.
Fig. 1 Molecular structures of A) Cym3, B) Cym4, and C) Cym5, with ellipsoids drawn at the 50% probability level and D) Cym6 with ellipsoids drawn
at the 30% probability level.
dominated by the formation of hydrogen-bonded dimers via the
carboxylate groups of two adjacent molecules.27 The most notable
feature is the nearly perpendicular arrangement of the five- and
six-membered rings in Cym4 with a torsion angle of 79.2◦ to avoid
close contacts between the two keto groups and the hydrogen
atoms on the carbon centers in ortho position to the connecting
the series of compounds, with torsion angles in the range of 4.3 to
8.5◦.
Synthesis of cymantrene-peptide conjugates
Peptides were synthesized on a Rink amide resin according to the
Fmoc/tBu-strategy on an automatic peptide synthesizer. After
the peptide sequence was assembled, the resin was removed from
the synthesizer and the cymantrene carboxylic acids were coupled
to the N-terminus of the peptides by standard activation with
◦
◦
=
C O group. This angle is reduced to 44.3 in Cym5 and 32.4 in
Cym6, thus decreasing substantially when going from 1,2- to 1,4-
substitution of the phenylene linker. The keto group, on the other
hand, remains essentially coplanar with the Cp ring throughout
2538 | Dalton Trans., 2010, 39, 2536–2545
This journal is
The Royal Society of Chemistry 2010
©

View Online
Table 1 Crystallographic parameters for complexes Cym3 to Cym6
Compound
Cym3
Cym4
Cym5
Cym6
Empirical formula
Formula weight
Dimensions/mm
Crystal system
Space group
C13H11MnO6
318.16
0.30 ¥ 0.30 ¥ 0.20
Monoclinic
P21/c
6.435(3)
19.001(9)
10.920(6)
C16H9MnO6
352.17
0.30 ¥ 0.20 ¥ 0.05
Monoclinic
P21/c
17.398(8)
7.379(3)
C16H9MnO6
352.17
0.20 ¥ 0.20 ¥ 0.10
Monoclinic
P21/c
11.745(2)
10.670(2)
13.424(3)
C16H9MnO6
352.17
0.20 ¥ 0.20 ¥ 0.10
Triclinic
¯
P1
˚
a/A
6.3640(12)
7.0047(13)
18.316(3)
90.874(6)
97.733(5)
114.416(4)
734.4(2)
2
1.593
223(2)
0.927
0.71073
24.99
˚
b/A
˚
c/A
11.634(5)
a (◦)
b (◦)
g (◦)
100.964(9)
103.950(10)
113.63(3)
3
˚
V/A
1310.9(11)
4
1.612
213(2)
1.029
0.71073
25.04
1449.5(11)
4
1.614
213(2)
0.940
0.71073
25.04
1541.1(5)
4
1.518
223(2)
0.892
0.71073
25.00
Z
rc/g cm-3
T/K
m/mm-1
˚
l/A (Mo-Ka)
2Hmax (◦)
Reflections measured
Unique refl./[I > 2s(I)]
Data completeness
Variables
6915
2278/2082
0.982
6822
2520/2163
0.982
6722
2598/2224
0.960
3092
2374/1883
0.917
209
0.0828
0.2196
0.973/-0.690
186
213
245
R (I > 2s(I))
0.0475
0.1203
0.605/-0.675
0.0564
0.1365
0.858/-0.532
0.0556
0.1409
0.671/-0.664
wR [I > 2s(I)]
Largest difference map
-3
˚
peak/hole in e A
Goodness of fit (GOF)
1.134
1.135
1.112
1.019
◦
˚
Table 2 Selected bond lengths [A] and angles ( ) for Cym3 to Cym6
Cym3
Cym4
Cym5
Cym6
Mn1–C11 1.798(4)
Mn1–C12 1.796(3)
Mn1–C13 1.800(4)
C11–O4 1.142(4)
C12–O5 1.144(4)
C13–O6 1.142(4)
Mn1–C14 1.795(4)
Mn1–C15 1.791(3)
Mn1–C16 1.791(4)
C14–O4 1.142(4)
C15–O5 1.143(4)
C16–O6 1.143(5)
Mn1–C1 1.754(4)
Mn1–C2 1.849(4)
Mn1–C3 1.835(4)
C1–O1 1.120(5)
C2–O2 1.179(6)
C3–O3 1.161(5)
Mn1–C1 1.804(7)
Mn1–C2 1.800(9)
Mn1–C3 1.795(8)
C1–O1 1.149(8)
C2–O2 1.163(10)
C3–O3 1.145(9)
Mn1–C1 2.148(3)
Mn1–C2 2.143(3)
Mn1–C3 2.129(3)
Mn1–C4 2.134(3)
Mn1–C5 2.139(3)
C6–O1 1.220(4)
C10–O2 1.217(4)
C10–O3 1.310(4)
C11–Mn1–C12 91.83(15)
C11–Mn1–C13 92.97(16)
C12–Mn1–C13 92.26(15)
Mn1–C1 2.121(3)
Mn1–C2 2.146(3)
Mn1–C3 2.158(3)
Mn1–C4 2.152(3)
Mn1–C5 2.130(3)
C6–O1 1.216(3)
C13–O3 1.228(4)
C13–O2 1.304(4)
C14–Mn1–C15 91.33(16)
C14–Mn1–C16 91.1(2)
C15–Mn1–C16 90.68(16)
Mn1–C4 2.196(3)
Mn1–C5 2.180(3)
Mn1–C6 2.135(3)
Mn1–C7 2.178(3)
Mn1–C8 2.227(3)
C16–O5 1.255(4)
C16–O6 1.256(4)
C9–O4 1.264(4)
Mn1–C4 2.149(8)
Mn1–C5 2.167(7)
Mn1–C6 2.166(7)
Mn1–C7 2.186(8)
Mn1–C8 2.158(7)
C9–O4 1.225(8)
C16–O5 1.289(8)
C16–O6 1.275(8)
C1–Mn1–C2 91.3(3)
C1–Mn1–C3 91.5(4)
C2–Mn1–C3 91.6(4)
C1–Mn1–C2 90.8(2)
C1–Mn1–C3 89.45(18)
C2–Mn1–C3 92.3(3)
1-hydroxybenzotriazole (HOBt) and N,N-diisopropylcar-
bodiimide (DIC). Subsequently, the bioconjugates were
cleaved from the resin with a mixture of 95% TFA, 2.5%
triisopropylsilane (TIS) and 2.5% water. Unexpectedly, additional
peptide conjugates with a molecular weight increased by 106
Da relative to that of the desired compounds were also detected
in high abundance (~60% of total product) in the MALDI
and ESI mass spectra. Recently, such side products were also
reported by Stathopoulos et al. and identified as a resin-,
cleavage mixture-, and sequence-dependent problem.32 The
authors suggested that the molecular weight increase by 106 Da
results from the alternative cleavage of the Rink amide linker in
the position 1 indicated in Fig. 2, leading to the formation of
C-terminally N-alkylated peptide amides. Interestingly, in our
case, this side-product formation seems to be dependent on the
cymantrene group attached to the peptide since we observed no
byproducts during the synthesis of the parent peptide sC18.18
However, we found that this sequence-dependent modification
could be suppressed by using a cleavage mixture containing
1,3-dimethoxybenzene (1,3-DMB), a substance that is reported
to prevent the linker decomposition during peptide cleavage.32
Therefore, cleavage of the cymantrene-peptide bioconjugates
from the resin was repeated using a mixture of 95% TFA and 5%
1,3-DMB, leading to the successful isolation of the Cym1-, Cym2-
and Cym3-sC18 derivatives. Typical MALDI mass spectra and
HPLC traces for Cym1-sC18 are shown in Fig. 3. As demonstrated
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 2536–2545 | 2539
©

View Online
successful coupling of Cym2 to different short carrier peptides
has been described.27 In summary, all these findings suggest two
different effects: first, decomposition of the Rink amide linker
is not only dependent on the peptide sequence, as described by
Stathopoulos et al., but also on the modification of the peptide
with the organometallic complex. In addition, it also seems to
depend on the cymantrene carboxylic acid attached, resulting in
alkylation of the metal complex itself, most likely at the aromatic
linker. To suppress the formation of those byproducts, several
different cleavage cocktails were tested containing either 1,3-
DMB, methanol, 2-methylanisole (2-MA), or 4-methylanisole (4-
MA). All these compounds are related to the carbocations that
might form during cleavage of the resin linker. In Table 4, the
results of the analysis of the resulting products with MALDI- and
ESI-MS as well as analytical RP-HPLC are collected.
Fig. 2 Possible cleavage sites of the Rink amide linker giving rise to the
molecular weight increase determined for the byproducts.
Interestingly, mixtures I, II, III, VII, IX, and X yielded the
[Cym4-sC18 + 106] byproduct and all mixtures except for I
and II also gave the [Cym4-sC18 + 120] compound. Only when
using mixtures V, VII, VIII, XI, and XII (highlighted in bold
in Table 4) was the desired target molecule formed in significant
yield, as shown by ESI- and MALDI-MS analysis. Variation of the
incubation time did not lead to considerable differences (data not
shown). However, it was still not possible to completely eliminate
the byproduct formation and only with cleavage mixture XII could
the product be obtained in relatively high yield (~70%). It was
purified by preparative HPLC and finally isolated in very high
purity (>99%).
It is interesting to note that during the synthesis of the Cym5-
and Cym6-peptide bioconjugates under the same conditions as
employed for the Cym4-sC18 conjugate, we did not observe any
byproducts at a molecular mass higher by 120, nor when using a
cleavage mixture containing TFA/water (95/5% v/v) nor by the
use of TFA/1,3-DMB (95/5% v/v). This is an additional hint
that the +120 Da byproduct formation depends on the linker of
the metal complex attached to the peptide.
Fig. 3 Typical analytical data for Cym1-sC18: (A) HPLC chromatogram
obtained with a gradient of 10 to 60% acetonitrile–water over 30 min. (B)
MALDI-MS spectrum of the compound. The peak at a retention time of
~5 min in the HPLC chromatogram is the injection peak.
in particular by the HPLC trace in Fig. 3A, the metal complex-
peptide conjugates were obtained in good purity.
In the MALDI-MS analysis of the products, the most intense
signal was identified as [M + H - Mn(CO)3]+, indicating the
loss of the metal tricarbonyl moiety under MALDI conditions
(Fig. 3B). Also, the [M + H]+ and [M + H - (CO)3]+ peaks were
detected. Recently, this phenomenon was also observed for other
cymantrene-peptide bioconjugates.28 However, ESI-MS showed
only the [M + 5H]5+ and [M + 6H]6+ signals of the intact products
(see Table 3).
Surprisingly, cleavage of Cym4-sC18 incorporating a 1,2-
phenylene linker with 1,3-DMB as the scavenger still did not
yield the expected target molecule since mainly a product with
a molecular weight of 120 Da higher than calculated for Cym4-
sC18 was isolated in nearly 90% yield. Furthermore, we detected
additional byproducts at +90 and +226 Da compared to the
desired compound. As shown in Fig. 2, decomposition of the Rink
amide linker is also possible at the positions labeled as 4, 2, and 3,
thus leading to the formation of N-alkylated peptide amides with
a molecular weight increased by 90, 120, and 226 Da, respectively.
Recently, we reported on the successful synthesis and biological
characterisation of Cym4-hCT(18–32)-k7, a cymantrene-peptide
bioconjugate where the metal-containing fragment was coupled
to a branched CPP.28 During the synthesis of this conjugate,
no byproducts resulting from the decomposition of the Rink
amide resin were observed. In addition, we did not observe
byproducts with a molecular weight increased by 120 Da after
coupling of Cym1, Cym2, or Cym3 to the sC18-peptide. Moreover,
Structural characterization of the byproducts by MS/MS studies
Our results suggest that during cymantrene-peptide synthesis, two
different side reactions occur. On the one hand, non-standard
cleavage of the Rink amide linker leads to the formation of C-
terminally N-alkylated peptide amide byproducts and on the other
hand, alkylation of the aromatic phenylene linker between the
cymantrene moiety and the peptide by carbocations resulting from
decomposition of the resin linker occurs. We hypothesized that in
the former case, cleavage of the Rink amide linker at position
1 (see Fig. 2) favors the liberation of the peptide from the resin
without formation of an intermediate carbocation. In contrast,
the byproduct with +120 Da must result from the formation of
a carbocation that can react with the Cym4 moiety, most likely
at the 1,2-phenylene linker. To gather evidence to prove this
hypothesis, we carried out additional MS/MS studies in which
the precursor ions of the +120 Da byproducts were mass-selected
and fragmented (Fig. 4).
The x, y, and z as well as the corresponding a, b, and c series
were observed, pointing to a distribution of cleavage sites as shown
in Fig. 4A. The cymantrene-bearing fragments yielded either the
expected mass peak or those corresponding to a loss of the three
carbonyls or the whole Mn(CO)3 group. Interestingly, the x, y, and
2540 | Dalton Trans., 2010, 39, 2536–2545
This journal is
The Royal Society of Chemistry 2010
©

View Online
Table 3 Analytical data of the cymantrene peptide bioconjugates Cym1- to Cym6-sC18
Peptide
Sequencea
MWcalc.
MALDIexp.
ESIcalc.
ESIexp.
Cym1-sC18
Cym2-sC18
Cym3-sC18
Cym4-sC18
Cym5-sC18
Cym6-sC18
Cym1-GLRKRLRKFRNKIKEK
Cym2-GLRKRLRKFRNKIKEK
Cym3-GLRKRLRKFRNKIKEK
Cym4-GLRKRLRKFRNKIKEK
Cym5-GLRKRLRKFRNKIKEK
Cym6-GLRKRLRKFRNKIKEK
2298.4
2354.5
2368.5
2402.5
2402.5
2402.5
2299.3 [M + H]+
2355.5 [M + H]+
2369.3 [M + H]+
2403.4 [M + H]+
2403.4 [M + H]+
2403.4 [M + H]+
460.7 [M + 5H]5+
471.9 [M + 5H]5+
474.7 [M + 5H]5+
481.5 [M + 5H]5+
481.5 [M + 5H]5+
481.5 [M + 5H]5+
460.8 [M + 5H]5+
472.0 [M + 5H]5+
474.9 [M + 5H]5+
481.7 [M + 5H]5+
481.7 [M + 5H]5+
481.7 [M + 5H]5+
a All peptides are C-terminally amidated.
Table 4 Different scavenger mixtures tested for cleavage of Cym4-sC18
No.
Reagent mixturea
Ratio
Product [%]
[M + 106]+ [%]
[M + 120]+ [%]
Other byproducts [%]
I
II
TFA : TA : TK
90 : 5 : 5
90 : 7 : 3
—
—
—
—
<60
—
<5
<65
—
< 95
< 10
< 15
—
—
—
< 5
—
< 5
< 10
—
—
—
> 5
TFA : TA : EDT
TFA : H2O : TIS
TFA : DMB
TFA : TIS : EDT : H2O
TFA : DMB : TIS
TFA : anisol : H2O : phenol
TFA : H2O : MeOH
TFA : 2-MA : 4-MA
TFA : 2-MA
> 90
> 70
> 10
> 10
> 10
> 30
>10
> 40
> 50
> 30
> 5
III
IV
V
VI
VII
VIII
IX
X
95 : 2.5 : 2.5
95 : 5
94 : 1 : 2.5 : 2.5
92.5 : 5 : 2.5
85 : 5 : 5 : 5
90 : 5 : 5
95 : 2.5 : 2.5
95 : 5
95 : 5
95 : 5
< 15
< 90
< 30
< 90
< 60
< 25
< 55
< 40
< 50
< 25
—
<20
<70
XI
XII
TFA : 4-MA
TFA : H2O
—
a Bold: product was obtained besides byproducts. TFA: trifluoroacetic acid, TA: thioanisol, TK: thiocresol, EDT: 1,2-ethanedithiol, TIS: triisopropylsilane,
DMB: 1,3-dimethoxybenzene, 2-/4-MA: 2- or 4-methylanisol.
Fig. 4 (A) Fragmentation pattern of [Cym4-sC18 + 120] and (B) experimental masses of the different fragments detected by MS/MS.
z series of the byproduct [Cym4-sC18 + 120] showed fragments
with the expected molecular weight whereas the a, b, and c series
had a mass increased by 120 Da. These results further support
the idea that the [Cym4-sC18 + 120] byproduct results from an
alkylation at the aromatic phenylene linker of the cymantrene
complex. We also analyzed the [Cym3-sC18 + 106] byproduct
(data not shown) in which the x, y, and z series showed an increase
of molecular weight by 106 Da as expected while the a, b, and
c series yielded the expected mass providing additional support
that, in this case, the [Cym3-sC18 + 106] byproduct results from
cleavage of the resin at position 1 as depicted in Fig. 2. This
assumption is further supported by the synthesis of a cymantrene-
peptide conjugate that was modified with two Cym4 units (data
not shown). After cleavage from the solid support with 95% TFA
and 5% 1,3-DMB, we observed peaks in the MALDI and ESI
mass spectra with a molecular weight increased by 240 Da. This is
consistent with a modification on both cymantrene units by groups
with a mass of 120 Da each and confirms the assumption that the
modification takes place at the phenylene linker. The fact that both
of the substituents of the phenyl ring are electron-withdrawing
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 2536–2545 | 2541
©

View Online
could explain the preferential attack of the carbocation in the
meta position. As outlined in Fig. 5, reaction at the 1,2-substituted
phenylene ring should be favored due to the fact that it has two,
the 1,3-substituted phenylene ring only one and the 1,4-substituted
phenylene ring no, positions prone to attack by the carbocation.
Fig. 6 (A) Relative cell viability of MCF-7 cells determined with the
resazurin assay after treatment for 24 h with sC18 (white), Cym1 (grey), and
Cym1-sC18 (black). (B) IC50 curve of Cym1-sC18. Data are normalized
to untreated cells (100% viability), for the control samples, cells were
treated with ethanol (70%). All experiments were done with n (number
of experiments) ≥ 2 in triplicate.
Fig. 5 Potential positions for the attack of reactive carbocations on the
linker in A) Cym4-, B) Cym5-, and C) Cym6-sC18 leading to byproducts
with a mass increased by 120 Da as indicated by arrows.
Conclusion
Preliminary tests of biological activity
For a preliminary study of the biological activity, we selected
bioconjugate Cym1-sC18 and tested it on the human breast
adenocarcinoma cancer cell line MCF-7. The results of the
resazurin-based cell viability test are shown in Fig. 6. After
incubation of the cells for 24 h with the sC18 peptide or the
cymantrene carboxylic acid Cym1 alone, no effect could be
observed. This is in line with other recent results on both the
sC18 peptide itself as well as other cymantrene bioconjugates, in
which both the peptide and the organometallic group itself were
found to be inactive.26-28 In contrast, incubation of MCF-7 cells
with the cymantrene-peptide bioconjugate Cym1-sC18 resulted
in a clear concentration-dependent decrease of the cell viability.
From non-linear curve fitting, an IC50 value of (59.8 6.7) mM was
determined for this compound. While other cymantrene-peptide
conjugates with CPPs like Tat and other sequences had no effect on
the human colon carcinoma cell line HT-29 in the concentration
range tested (<100 mM), a more complex Cym4-hCT(18–32)-k7
conjugate exhibited an IC50 of about 30 mM on MCF-7 cells.28
These results suggest that the biological activity depends on both
the cell line as well as the carrier peptide used. In particular,
our model conjugate investigated in this work shows promising
activity and a detailed study of the biological activity of all peptide
conjugates presented here will be the subject of an upcoming study
from our labs.
In this work, six new functionalized cymantrenes were prepared
and conjugated by SPPS to the novel sC18 carrier peptide, which
is made up of residues 106–121 of the C-terminal region of
the cationic antimicrobial peptide cathelicidin (CAP18). Four
of the cymantrenes were additionally characterized by X-ray
structure determination. In the peptide functionalization step, we
found that decomposition of the Rink amide linker can lead, on
one hand, to the formation of C-terminally N-alkylated amide
byproducts and on the other hand, to the formation of reactive
carbocations which undergo further reactions with the aromatic
linker of the attached metal complex. Occurrence of both processes
depends on the peptide sequence as well as the metal complex
attached. The non-standard Rink amide linker cleavage could be
suppressed by addition of 1,3-dimethoxybenzene (1,3-DMB) to
the cleavage mixture while formation of the carbocation could be
reduced by use of a combination of trifluoroacetic acid (TFA)
and water. Preliminary testing of the biological activity of the
Cym1-sC18 bioconjugate on human MCF-7 breast cancer cells
demonstrated the utility of the combination of a functionalized
cymantrene group with the sC18 carrier peptide to generate
organometallic peptide conjugates with promising anticancer
activity. Further studies will be aimed at clarifying the uptake
mechanism, internalization pattern and biological activity of all
six cymantrene-CPP bioconjugates presented here.
2542 | Dalton Trans., 2010, 39, 2536–2545
This journal is
The Royal Society of Chemistry 2010
©

View Online
acid (10 ml). The layers were separated and the aqueous phase
extracted with dichloromethane (3¥ 50 ml). The combined organic
phases were thenwashedwithwater andsaturated aqueous sodium
carbonate solution. The solvent was removed from the organic
phase and the residue purified by column chromatography on silica
using a mixture of n-hexane–ethyl acetate 5 : 3 (v/v) as the eluent
(Rf 0.60). Yield: 54%. 1H-NMR (200 MHz, (CD3)2CO, dppm): 8.43
(s, 1H, HAr), 8.25 (d, 1H, J = 7.8 Hz, HAr), 8.09 (d, 1H, J =
7.8 Hz, HAr), 7.72 (t, 1H, 3J = 7.7 Hz, HAr), 5.74 (s, 2H, Cp), 5.26
(s, 2H, Cp), 3.93 (s, 3H, -OCH3);); 13C-NMR (62.9 MHz, CDCl3,
Experimental
General procedures
Reactions were carried out in oven-dried Schlenk glass-
ware under an atmosphere of pure dinitrogen when nec-
essary. Solvents were dried over molecular sieves and de-
gassed prior to use. Fmoc amino acid derivatives were pur-
chased from Iris Biotech GmbH, CBL, and Novabiochem.
1-Hydroxybenzotriazole (HOBt) and 4-(2¢,4¢-dimethoxyphenyl-
Fmoc-aminomethyl)-phenoxymethyl-linked polystyrene (Rink
amide) were obtained from Novabiochem. 1,2-Ethanedithiol
(EDT), thioanisol (TA), thiocresol (TK), piperidine, and triiso-
propylsilane (TIS) were purchased from Fluka. Dichloromethane
(DCM) and N,N-dimethylformamide (DMF) were Biosolve prod-
ucts and 2-methylanisole (2-MA) and 4-methylanisole (4-MA)
from Acros. N,N-diisopropylcarbodiimide (DIC) was obtained
from Iris Biotech GmbH, anisole, trifluoroacetic acid (TFA),
and 1,3-dimethoxybenzene (DMB) from Sigma-Aldrich. Phenol
was purchased from Riedel-de Ha¨en and gradient degree high-
performance liquid chromatography (HPLC) solvents acetonitrile
(ACN) from Chromanorm and methanol from LiChrosolv. All
other chemicals were obtained from commercial sources and
used without further purification. NMR spectra were recorded
on Bruker DPX 200, DPX 250, or DRX 400 spectrometers (1H
at 200.13 and 400.13 MHz, respectively; 13C at 50.33, 62.90 and
100.62 MHz). Chemical shifts d in ppm indicate a downfield shift
relative to tetramethylsilane (TMS) and were referenced relative
to the signal of the solvent.33 Coupling constants J are given
in Hz. Individual peaks are marked as singlet (s), doublet (d),
triplet (t) quartet (q), or multiplet (m). Mass spectra of small
molecules were measured on Bruker Esquire 6000 (ESI) and VG
Autospec (EI, FAB) instruments, only characteristic fragments
are given for the most abundant isotope peak. The solvent flow
rate for the ESI measurements was 4 ml min-1 with a nebulizer
pressure of 10 psi and a dry gas flow rate of 5 l min-1 at a dry gas
temperature of 300 ◦C. IR spectra were recorded on pure solid
samples with a Bruker Tensor 27 IR spectrometer equipped with a
Pike MIRacle Micro ATR accessory. The elemental composition
of the compounds was determined with a VarioEL analyzer from
Elementar Analysensysteme GmbH.
3
3
=
=
=
d
ppm): 222.76 (C O), 191.38 (C O), 166.14 (C O), 138.05 (CAr),
133.64 (CAr), 132.26 (CHAr), 131.22 (CHAr), 130.46 (CHAr), 127.91
(CHAr), 97.98 (Cp), 91.18 (Cp), 89.69 (Cp), 86.46 (Cp), 82.72 (Cp),
53.72 (-OCH3); FAB-MS: m/z = 367.0 [M + H]+; IR (ATR, cm-1):
2019, 1919, 1715, 1641, 1271, 1163; Elemental analysis (%): calc.
for C17H11MnO6: C 55.76, H 3.03 found: C 55.39, H 2.61.
The Cym6 methyl ester was prepared by a similar procedure
using 8 instead of 4 and with a slightly different chromatographic
work-up. For this compound, n-hexane–ethyl acetate 5 : 2 (v/v)
was used as the eluent (Rf 0.45) to give the product as a yellow
1
solid. Yield: 45%. H-NMR (400 MHz, CDCl3, dppm): 8.16 (d,
2H, 3J = 8.0 Hz, HAr), 7.82 (d, 2H, 3J = 8.0 Hz, HAr), 5.49 (s, 2H,
Cp), 4.93 (s, 2H, Cp), 3.96 (s, 3H, -OCH3); 13C-NMR (100.6 MHz,
=
=
=
CDCl3, dppm): 222.72 (C O), 191.89 (C O), 166.22 (C O), 141.65
(CAr), 133.53 (CAr), 129.97 (CHAr), 127.98 (CHAr), 91.15 (Cp), 88.30
(Cp), 84.19 (Cp), 52.63 (-OCH3); FAB-MS: m/z = 367.0 [M +
H]+; IR (ATR, cm-1): 2020.12, 1962.93, 1921.38, 1721.82, 1643.55,
1278.93, 1112.89; Elemental analysis (%): calc. for C17H11MnO6:
C 55.76, H 3.03, found: C 55.13, H 3.21.
Cym5 and Cym6. An aqueous solution of sodium hydroxide
(1 M, 10 ml) was added to Cym5 or Cym6 methyl ester (200 mg,
0.55 mmol) in methanol (50 ml) and stirred at room temperature.
The reaction was monitored with TLC (silica, n-hexane–ethyl ac-
etate 5 : 2 v/v) until the disappearance of the starting material was
complete. Then, the solvent was removed and water (200 ml) added
to the residue. The solution was washed with dichloromethane
(3¥ 50 ml), and acidified with concentrated hydrochloric acid to
pH 1. The precipitate was taken up in ethyl acetate and dried over
magnesium sulfate. After removal of the solvent, Cym5 and Cym6
were obtained as yellow solids. Yields: 85% for both compounds.
Cym5. 1H-NMR (200 MHz, CD3OD, dppm): 8.28 (s, 2H, HAr),
7.94 (s, 1H, HAr), 7.66 (s, 1H, HAr), 5.66 (s, 2H, Cp), 5.14 (s,
Synthesis of cymantrene keto carboxylic acids
2H, Cp);); 13C-NMR (62.9 MHz, (CD3)2SO, dppm): 224.33 (C O),
The three-step synthesis of the terephthalic and isophthalic acid
monomethyl ester chlorides 4 and 8 used in the Friedel–Crafts
acylations is described in the ESI.†
=
=
=
191.79 (C O), 170.35 (C O), 139.28 (CAr), 134.58 (CAr), 132.02
(CHAr), 131.34 (CHAr), 130.30 (CHAr), 128.73 (CHAr), 97.20 (Cp),
92.69 (Cp), 91.12 (Cp); ESI-MS (negative): m/z = 351.0 [M -
H]-; IR (ATR, cm-1): 2557, 2024, 1945, 1692, 1649, 1307, 1265;
Elemental analysis (%): calc. for C16H9MnO6: C 54.57, H 2.58,
found: C 54.24, H 2.60.
Preparation of Cym5 methyl ester
Aluminium chloride (333.8 mg, 2.5 mmol, Fluka, ≥99.0%) and
anhydrous dichloromethane (20 ml) were placed in a 50 ml
three-necked Schlenk-flask equipped with a reflux condenser and
internal thermometer. The solution was cooled to 0 to 5 ◦C with ice
while stirring. Then, solid 4 (243.3 mg, 1.23 mmol) was added in
small portions followed by cymantrene (250 mg, 1.23 mmol) while
the temperature was maintained at 0 to 5 ◦C. After 1 h, the reaction
mixture was allowed to warm to room temperature and stirring
continued overnight. The reaction mixture was then poured into
a mixture of ice/water (100 ml) and concentrated hydrochloric
Cym6. 1H-NMR (200 MHz, (CD3)2CO, dppm): 8.19 (s, 2H,
HAr), 7.94 (s, 2H, HAr), 5.76 (s, 2H, Cp), 5.25 (s, 2H, Cp); 13C-NMR
=
=
(62.9 MHz, (CD3)2SO, dppm): 223.44 (C O), 191.21 (C O), 166.21
(C O), 140.47 (CAr), 133.96 (CHAr), 90.94 (Cp), 89.20 (Cp), 85.81
=
(Cp) 85.13 (Cp); ESI-MS(negative): m/z = 350.9 [M - H]-; IR
(ATR, cm-1): 2543, 2028, 1928, 1683, 1632, 1378, 1285; Elemental
analysis (%): calc. for C16H9MnO6: C 54.57, H 2.58, found: C
54.69, H 3.02.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 2536–2545 | 2543
©

View Online
Synthesis of cymantrene-peptide conjugates
inactivated fetal bovine serum (FBS) (v/v) and 2 mM L-glutamine
(Q) was used. The effect of cymantrene peptide conjugate Cym1-
sC18 on the cell viability was examined using a resazurin-based in
vitro toxicology assay. MCF-7 cells were seeded in 96-well plates at
40.000 cells per well. Having reached 80% confluency, the medium
was removed and cells were incubated for 24 h at 37 ◦C with
the test substance at different concentrations. Negative controls
were incubated with cell culture medium only. After 24 h, the cells
were washed and incubated for 2 h with 10% resazurin in cell
medium without fetal bovine serum (v/v) at 37 ◦C. As a positive
control, cells treated with 70% ethanol were used since this is
known to be highly toxic to cells. The fluorescence was measured
using a Spectrafluor plus multiwell reader (Tecan) at 595 nm with
excitation at 550 nm.
sC18 was synthesized by automated solid-phase peptide synthesis
(SPPS) on a Rink amide resin (30 mg, resin loading 0.45 mmol g-1)
using the Fmoc/tBu-strategy on a multiple synthesizer (Syro, Mul-
tiSynTech, Bochum, Germany) as described elsewhere.34 Manual
coupling of cymantrene acids Cym1 to Cym6 to the N-terminus
was achieved by activation with 5 eq. HOBt/DIC. Peptides were
cleaved from the resin with TFA under addition of different
scavenger mixtures (see Table 4) for 1 to 3 h and precipitated by
addition of ice-cold diethylether. The compounds were analyzed
by reversed-phase (RP) HPLC, matrix-assisted laser desorption
ionization time of flight (MALDI-ToF) mass spectrometry (MS)
and electrospray ionization (ESI) MS. ESI ion trap measurements
were performed on a Bruker HCT mass spectrometer. MALDI-
ToF mass spectrometry was carried out on a Bruker Daltonics
Ultraflex III instrument in reflection mode. Analytical RP-HPLC
was done on a Merck-Hitachi system with a Grace Vydac 218TP54
X-ray crystallographic data collection and refinement of structures
Cym3 to Cym6
˚
A single crystal of either Cym3, Cym4, Cym5, or Cym6 was
coated with perfluoropolyether, picked up with a glass fiber,
and immediately mounted in the nitrogen cold stream of the
diffractometer. Intensity data were collected at 213(2) or 223(2) K
column (4.6 ¥ 250 mm; 5 mm; 300 A) using a linear gradient of
10 to 60% of acetonitrile/0.08% TFA in water/0.1% TFA over
30 min and a flow rate of 0.6 ml min-1 or a linear gradient
of 40 to 90% of methanol/0.08% TFA in water/0.1% TFA
over 30 min at the same flow rate. The functionalized peptides
were purified by semi-preparative or preparative RP-HPLC on a
Shimadzu Chromatopac system using the same binary elution
system as used in the analytical HPLC. In the first case, a
˚
using graphite monochromated Mo-Ka radiation (l = 0.71073 A).
Final cell constants were obtained from a least squares fit of
a subset of a few thousand strong reflections. Data collection
was performed by hemisphere runs taking frames at 0.3◦ in
w on a Bruker AXS CCD 1000 diffractometer. The program
SADABS was used to account for absorption.35 The SHELXL-
97 software package was used for solution, refinement, and
artwork of the structures.36 The structures were readily solved
by Patterson methods and difference Fourier techniques. All non-
hydrogen atoms were refined anisotropically and hydrogen atoms
were placed at calculated positions and refined as riding atoms
with isotropic displacement parameters. Note that for Cym6, the
data completeness is low due to crystal decomposition during
the measurement. CCDC numbers: 681218, 744989, 744990, and
744991.†
˚
Vydac 218TP510 C18 column (250 ¥ 4.6 mm; 5 mm; 300 A)
with a linear gradient running from 20 to 60% water/0.1%
TFA in acetonitrile/0.08% TFA over 50 min or from 40 to
80% water/0.1% TFA in acetonitrile/0.08% TFA over 50 min
was used. Preparative HPLC was carried out utilizing a Vydac
˚
218TP1022 C18 column (250 ¥ 20 mm; 10 mm; 300 A) and the same
linear gradients as described above. Fractions containing peptide
conjugates were collected and analyzed by analytical HPLC and
MALDI-ToF MS. Pure fractions were combined and frozen at
-80 ◦C followed by subsequent lyophilization. Analytical data for
all peptides is collected in Table 3.
Acknowledgements
MS/MS
The authors thank R. Reppich-Sacher for help with the MALDI
and ESI mass spectrometry, K. Lo¨bner for help with the cell
cultures, and N. Zehethofer (Ultrasensitive Protein Detection
Unit, Leipzig University) for help with the MS/MS analysis. This
work was supported by the Deutsche Forschungsgemeinschaft
(DFG) within the project FOR 630 “Biological function of
organometallic compounds”. I. N. and U. S. thank Prof. Dr
A. G. Beck-Sickinger (Leipzig) and Prof. Dr N. Metzler-Nolte
(Bochum) for generous access to all facilities of the respective
institutes.
Lyophilized byproducts were dissolved in acetonitrile–water (1 : 1)
and diluted 1 : 100. Full scan and product ion spectra were
acquired on a LTQ Orbitrap XL ETD (Thermo Scientific) mass
spectrometer in the positive ion mode. The sample solutions were
infused into the mass spectrometer using a built-in syringe pump
at a flow rate of 2 ml min-1. The following general parameters were
used: spray voltage 1.6 kV, capillary temperature 200 ◦C, and tube
lens voltage 125 V. Collision energies were chosen from 10 to 50 eV
and optimized to achieve the desired level of fragmentation. Data
acquisition and analysis was performed using Xcalibur software
(version 2.0.7, Thermo Scientific). The mass spectrometer was
calibrated externally using the manufacturer’s calibration standard
mixture.
References
1 G. Jaouen, Bioorganometallics, Wiley-VCH, Weinheim, 2006.
2 D. R. van Staveren and N. Metzler-Nolte, Chem. Rev., 2004, 104, 5931.
3 N. Metzler-Nolte, Chimia, 2007, 61, 736.
4 F. Noor, A. Wu¨stholz, R. Kinscherf and N. Metzler-Nolte, Angew.
Chem., 2005, 117, 2481.
5 S. Mukhopadhyay, C. M. Barnes, A. Haskel, S. M. Short, K. R. Barnes
and S. J. Lippard, Bioconjugate Chem., 2008, 19, 39.
6 M. A. Neukamm, A. Pinto and N. Metzler-Nolte, Chem. Commun.,
2008, 232.
Cell culture and resazurin-based cell viability assay
MCF-7 human breast adenocarcinoma cells were used for pre-
liminary cytotoxicity studies. They were grown to confluency at
37 ◦C and 5% CO2 in a humidified atmosphere in 75 cm2 cell culture
flasks. DMEM/Ham’s F12 medium supplemented with 10% heat-
2544 | Dalton Trans., 2010, 39, 2536–2545
This journal is
The Royal Society of Chemistry 2010
©

View Online
7 J. T. Chantson, M. V. Verga Falzacappa, S. Crovella and N. Metzler-
Nolte, J. Organomet. Chem., 2005, 690, 4564.
8 J. T. Chantson, M. V. V. Falzacappa, S. Crovella and N. Metzler-Nolte,
ChemMedChem, 2006, 1, 1268.
9 D. Zwanziger and A. G. Beck-Sickinger, Curr. Pharm. Des., 2008, 14,
2385.
10 N. Agorastos, L. Borsig, A. Renard, P. Antoni, G. Viola, B. Spingler,
P. Kurz and R. Alberto, Chem.–Eur. J., 2007, 13, 3842.
11 A. F. Armstrong, N. Oakley, S. Parker, P. W. Causey, J. Lemon,
A. Capretta, C. Zimmerman, J. Joyal, F. Appoh, J. Zubieta,
J. W. Babich, G. Singh and J. F. Valliant, Chem. Commun., 2008,
5532.
22 K. M. Stewart, K. L. Horton and S. O. Kelley, Org. Biomol. Chem.,
2008, 6, 2242.
23 R. Rennert, I. Neundorf, H. G. Jahnke, P. Suchowerskyj, P. Dournaud,
A. Robitzki and A. G. Beck-Sickinger, ChemMedChem, 2008, 3, 241.
24 C. Walther, K. Meyer, R. Rennert and I. Neundorf, Bioconjugate Chem.,
2008, 19, 2346.
25 C. Walther, I. Ott, R. Gust and I. Neundorf, Biopolymers, 2009, 92,
445.
26 H. W. Peindy N’Dongo, I. Ott, R. Gust and U. Schatzschneider,
J. Organomet. Chem., 2009, 694, 823.
27 H. W. Peindy N’Dongo, I. Neundorf, K. Merz and U. Schatzschneider,
J. Inorg. Biochem., 2008, 102, 2114.
12 K. A. Mahmoud and H. B. Kraatz, Chem.–Eur. J., 2007, 13, 5885.
13 G. Dirscherl and B. Ko¨nig, Eur. J. Org. Chem., 2008, 597.
14 S. I. Kirin, P. Du¨bon, T. Weyhermu¨ller, E. Bill and N. Metzler-Nolte,
Inorg. Chem., 2005, 44, 5405.
28 I. Neundorf, J. Hoyer, K. Splith, R. Rennert, H. W. Peindy N’Dongo
and U. Schatzschneider, Chem. Commun., 2008, 5604.
29 E. R. Biehl and P. C. Reeves, Synthesis, 1973, 360.
30 R. A. Firestone, N. S. Maciejewicz and B. G. Christensen, J. Org. Chem.,
1974, 39, 3384.
31 E.-D. Chenot, D. Bernardi, A. Comel and G. Kirsch, Synth. Commun.,
2007, 37, 483.
32 P. Stathopoulos, S. Papas and V. Tsikaris, J. Pept. Sci., 2006, 12, 227.
33 H. E. Gottlieb, V. Kotlyar and A. Nudelman, J. Org. Chem., 1997, 62,
7512.
15 L. Barisic, V. Rapic and N. Metzler-Nolte, Eur. J. Inorg. Chem., 2006,
4019.
16 U. Hoffmanns and N. Metzler-Nolte, Bioconjugate Chem., 2006, 17,
204.
17 H. Pfeiffer, A. Rojas, J. Niesel and U. Schatzschneider, Dalton Trans.,
2009, 4292.
18 I. Neundorf, R. Rennert, J. Hoyer, F. Schramm, K. Lo¨bner, I. Kitanovic
and S. Wo¨lfl, Pharmaceuticals, 2009, 2, 49.
34 B. Rist, M. Entzeroth and A. G. Beck-Sickinger, J. Med. Chem., 1998,
41, 117.
¨
19 U. Langel, Handbook of Cell-Penetrating Peptides, 2nd edn, CRC Press,
35 G. M. Sheldrick, SADABS, Program for Empirical Correction of Area
Detector Data, (1996), Universita¨t Go¨ttingen, Germany.
36 G. M. Sheldrick, SHELXL-97, Program for Crystal Structure Refine-
ment, (1997), Universita¨t Go¨ttingen, Germany.
Boca Raton, 2006.
20 A. Joliot and A. Prochiantz, Nat. Cell Biol., 2004, 6, 189.
¨
21 M. Ma¨e and U. Langel, Curr. Opin. Pharmacol., 2006, 6, 509.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 2536–2545 | 2545
©