Interactions of multicationic bis(guanidiniocarbonylpyrrole) receptors with double-stranded nucleic acids: Syntheses, binding studies, and atomic force microscopy imaging

Article abstract of DOI:10.1002/chem.201101544

Compounds 1-3, composed of two guanidiniocarbonylpyrrole moieties linked by oligoamide bridges and differing in number and type of basic groups, were prepared. The sites and degree of protonation of 1-3 depend strongly on the pH value. The interactions of

Full text of DOI:10.1002/chem.201101544

DOI: 10.1002/chem.201101544
Interactions of Multicationic Bis(guanidiniocarbonylpyrrole) Receptors with
Double-Stranded Nucleic Acids: Syntheses, Binding Studies, and Atomic
Force Microscopy Imaging
Karsten Klemm,[a] Marijana Radic´ Stojkovic´,[b] Gordan Horvat,[c] Vladislav Tomisˇic´,[c]
Ivo Piantanida,*[b] and Carsten Schmuck*[a]
Abstract: Compounds 1–3, composed
of two guanidiniocarbonylpyrrole moi-
eties linked by oligoamide bridges and
differing in number and type of basic
groups, were prepared. The sites and
degree of protonation of 1–3 depend
strongly on the pH value. The interac-
tions of these compounds with several
double-stranded (ds) DNA and dsRNA
were investigated by means of UV/Vis
and CD spectroscopy as well as isother-
mal titration microcalorimetry (ITC).
These studies revealed that the binding
of 1–3 to the polynucleotides is driven
by three factors, the presence of ali-
phatic amino groups, the protonation
state of the compounds, and the steric
properties of the polynucleotide bind-
ing site, that is, the shape and structure
of their grooves. The results obtained
by all applied methods consistently in-
dicated that receptors 1–3 bind to the
minor groove of DNA, but, by con-
trast, to the major groove of RNA. Ad-
ditionally, it was shown by atomic force
microscopy (AFM) imaging that upon
interaction of compound 2 with calf
thymus (ct) DNA induced aggregation
of the DNA occurs, leading to pro-
nounced changes in its secondary struc-
ture.
Keywords: DNA · DNA recogni-
tion · groove binding · nucleic acids ·
receptors
chemistry
·
supramolecular
Introduction
tion, minor or major groove binding, external electrostatic
binding) for their interaction with double-stranded (ds)
DNA/RNA.[2] In the past years, first examples of molecules,
which use a combination of two or more different binding
interactions, were developed, often taking Nature as a blue-
print, for example, mimicking interactions found in DNA–
protein complexes. Interestingly, some recent crystal struc-
tures of one of the most abundant bacterial DNA-binding
proteins show selective interactions with DNA that signifi-
cantly depend on the size of its minor groove, thus suggest-
ing that the secondary and ternary structure of the DNA
(specifically, the minor groove geometry) and not just its
base-pair sequence can determine binding specificity.[3]
Moreover, tight fitting of small proteins as well as small
molecules into the minor groove of DNA is quite often ac-
companied by induced conformational changes of DNA,
which can lead to both cooperative and anti-cooperative
binding of other molecules.[2–4]
Molecular recognition of DNA and RNA is an important
area of current research, due to the relevance of such recog-
nition events in medicinal, biochemical, and biological pro-
cesses.[1] In Nature, often proteins are the binding partners,
and specific nucleic acid–protein interactions are important
for the transfer of genetic information. In supramolecular
chemistry, nucleic acid recognition by small molecules
(often called artificial receptors) has been extensively stud-
ied over the years. Most often these small molecules rely on
one dominant non-covalent binding mode (e.g., intercala-
[a] K. Klemm, Prof. Dr. C. Schmuck
Institute for Organic Chemistry
University of Duisburg-Essen
Universitꢀtsstrasse 7, 45141 Essen (Germany)
Fax : (+49)201-183-4259
In contrast to molecules, which interact with DNA, RNA-
targeting molecules are significantly less studied. However,
as there are more than one thousand microRNAs that con-
trol approximately 30% of all genes and that are involved
in most genetic pathways and human diseases, specific inhib-
itors and activators of microRNAs have potential therapeu-
tic applications.[5] Thus, RNA is becoming a major target for
drug discovery. Of particular interest are those RNAs that
fold into complex tertiary structures, because they are often
involved in RNA–RNA or RNA–protein interactions that
have impact on important cellular processes.[6] Therefore,
E-mail: carsten.schmuck@uni-due.de
´
´
[b] Dr. M. Radic Stojkovic, Dr. I. Piantanida
Division of Organic Chemistry and Biochemistry
ˇ
´
“Ruder Boskovic” Institute
¯
P. O. Box 180, 10002 Zagreb (Croatia)
E-mail: Ivo.Piantanida@irb.hr
ˇ ´
[c] G. Horvat, Prof. V. Tomisic
Department of Chemistry
Faculty of Science, University of Zagreb
Horvatovac 102a, 10002 Zagreb (Croatia)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201101544.
1352
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1352 – 1363

FULL PAPER
small molecules, which could selectively report on small
structural differences of DNA as well as RNA secondary or
higher-ordered structures, are of scientific interest as they
might also show potential for future applications as diagnos-
tic and imaging tools in molecular biology and medicine.[7]
In the past three years, we have developed several artifi-
cial nucleic acid receptors in which a guanidiniocarbonylpyr-
role cation is connected through a flexible linker to an aro-
matic moiety.[8,9] Systematic studies of a series of such com-
pounds were undertaken with the intention to i) determine
the minimal size of the aryl moiety necessary for efficient
“anchoring” of the small molecule within dsDNA/RNA by
intercalation, ii) to define structural properties (e.g., length,
flexibility) of the linker between the guanidiniocarbonylpyr-
role cation and the aryl group, which allows efficient binding
of the cation to DNA/RNA after intercalation of the aro-
matic moiety, and iii) to find an appropriate fluorescence
probe to report the binding event in a selective manner.
Until now, pyrene hybrid compounds showed the most
promising results, revealing selective and in some cases se-
quence-specific fluorescence and induced circular dichroism
(ICD) responses to several different DNA or RNA sequen-
ces.[8,9] Inspired by the RNA-specific fluorescence change of
such a pyrene hybrid compound, which was caused by
pyrene excimer formation within the RNA groove,[8] we
have also designed two pyrenyl–guanidiniocarbonylpyrrole
hybrid compounds, in which the linker between the pyrene
and the guanidiniocarbonylpyrrole moiety was significantly
modified by introducing additional aromatic moieties next
to the pyrene (a methyl imidazole and a pyrrole–methyl imi-
dazole unit, respectively).[10] These two heterocycles were
developed by Dervan and coworkers to probe different
DNA sequences upon specific groove binding.[11] Moreover,
these building blocks also significantly rigidified the linker,
whereby a simultaneous intercalation of the pyrene moiety
into the double-stranded nucleic acid and interactions of the
linker within the DNA/RNA grooves could no longer occur.
We found that the incorporated heterocycles were able to
control the accommodation of our molecules within various,
structurally different forms of DNA minor grooves by steric
and hydrogen-bonding effects. Even small differences in the
position of the attached pyrene were reflected in the fluori-
metric responses.[10] Extending this line of research, we here
present a study of a new series of compounds (1–3), consist-
ing of two guanidiniocarbonylpyrroles linked by oligoamide
bridges and differing in both the sterical properties and the
number and site of positive charges. These compounds are
expected to interact with nucleic acids mainly by electrostat-
ic and hydrogen-bonding interactions. The number of charg-
es can easily be controlled by the pH value of the solution,
thus enabling a systematic study of their role in nucleic acid
recognition. By using UV/Vis and CD spectroscopy as well
as isothermal microcalorimetry (ITC) we show that the com-
pounds have a high affinity to nucleic acids, which depends
on the size and shape of the corresponding grooves. More-
over, they also significantly stabilize the double strands and
at the same time induce severe aggregation of the nucleic
acid, as demonstrated by atomic force microscopy (AFM).
Results and Discussion
Compound design and synthesis: We have prepared com-
pounds 1–3 with an ethylene diamine linker separating the
two guanidiniocarbonylpyrrole cations, which function as ef-
ficient binding sites for anions. In addition, compounds 1, 2,
and 3, carry no, two, and four amino groups, respectively.
Therefore, it is possible to accurately and reversibly tune
the number of positively charged groups in the molecule by
variation of the pH value and thus, controlling the binding
properties of the compounds to various DNA/RNAs. More-
over, the guanidiniocarbonylpyrrole cation shows excellent
binding properties for oxyanions due to hydrogen-bonding-
enforced electrostatic interactions.[12] The highly negatively
charged phosphate backbone thus renders polynucleotides
perfect substrates for the positively charged receptor com-
pounds 1–3.
The dicationic compound 1 was prepared according to
Scheme 1 by the use of building block 5, which was coupled
twice to EDA as a linker. Methyl ester 4 was synthesized as
described earlier by us, starting with Cbz-protected aspara-
Scheme 1. Synthesis of the dicationic compound 1 (Cbz=carbobenzyloxy,
Boc=butoxycarbonyl, EDA=ethylenediamine, HCTU=O-(6-chloro-
benzotriazol-1-yl)-N,N,N’,N’-tetramethyluronium hexafluorophosphate,
NMM=N-methylmorpholine, TFA=trifluoroacetic acid).
Chem. Eur. J. 2012, 18, 1352 – 1363
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1353

C. Schmuck, I. Piantanida et al.
gine and a tert-Boc-protected guanidinocarbonylpyrrole car-
boxylic acid.[13] The resulting Cbz- and Boc-protected ester 4
was hydrolyzed with lithium hydroxide in a mixture of THF
and water (THF/water=4:1) to provide the carboxylic acid
5 in a yield of 89%. Coupling with ethylenediamine could
be achieved by the use of HCTU in a mixture of dichloro-
methane and DMF, to obtain compound 6 in 88% yield.
The cleavage of the two Boc protecting groups was achieved
by treatment with TFA and subsequent addition of an aque-
ous hydrochloric acid solution (5%) to obtain the dicationic
compound 1 as the chloride salt in a yield of 40%.
The synthesis of the tetracationic compound 2 (Scheme 2)
also started from building block 4, which was first deprotect-
ed through hydrogenation with Pd/C, followed by a Boc pro-
Scheme 3. Synthesis of the hexacationic compound 3.
the methyl ester 10 with BocO2 in methanol (94%), the
methyl ester 11 was hydrolyzed with lithium hydroxide in a
mixture of THF and water (THF/water=4:1) to obtain the
carboxylic acid 12 in 88% yield. The key step in the synthe-
sis of compound 3, the coupling of 12 with ethylenediamine,
was achieved by activation with PyBOP in a mixture of di-
chloromethane and DMF (DCM/DMF=4:1). Cleavage of
the two Cbz groups in the coupling product 13 by catalytic
hydrogenation in methanol led to the tetra-Boc-protected
compound 14 in a yield of 82%. To obtain the chloride salt
of the hexacationic compound 3, the four Boc groups in 14
were first cleaved with TFA and afterwards treated with an
aqueous solution of HCl (5%; yield, 98%).
Scheme 2. Synthesis of the tetracationic compound 2.
tection with Boc2O. The resulting di-Boc-protected ester 7
was subsequently converted into the acid 8 by use of lithium
hydroxide. For the coupling with the ethylenediamine linker
we tried several coupling reagents (HCTU, (benzotriazol-1-
Physicochemical properties of compounds 1–3 in aqueous
solution: The dicationic compound 1 has only limited solu-
bility in water, thus a stock solution of 1 in DMSO (c(1)=
5ꢂ10ꢀ3 m) was prepared and diluted with water or buffer to
a concentration of c(1)=2ꢂ10ꢀ5 m. The more positively
charged compounds 2 and 3 showed better solubility in pure
water (c(2)=5ꢂ10ꢀ3 m; c(3)=2.5ꢂ10ꢀ3 m) and in the buffer
systems used for the binding studies.
Buffered aqueous solutions of all compounds are stable
for several days. The absorbance of these solutions is pro-
portional to the concentration at least up to 5ꢂ10ꢀ5 m.
Changes of the UV/Vis spectra on temperature increase
were negligible up to 908C and reproducibility of UV/Vis
absorbance upon cooling back to 258C was excellent. These
data indicate that compounds 1–3 do not aggregate under
the experimental conditions. The absorption maxima and
yloxy)tripyrrolidinophosphonium
hexaflourophosphate
(PyBOP)) and different conditions. The best yields were
achieved with HCTU and NMM in a mixture of dichlorome-
thane and DMF (77% yield). After deprotection of 9 with
TFA and subsequent addition of an aqueous solution of HCl
(5%) the tetracationic compound 2 was obtained as the
chloride salt in almost quantitative yield.
For the preparation of the hexacationic compound 3 an-
other building block (compound 10) was used as starting
material as shown in Scheme 3. The methyl ester 10 was syn-
thesized according to a procedure already established in our
group.[14] After Boc protection of the secondary amines in
1354
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1352 – 1363

Multicationic Bis(guanidiniocarbonylpyrrole) Receptors
FULL PAPER
Table 1. Electronic absorption data of compounds 1–3 at pH 5 and 7.
binding of the receptors to the polynucleotides. Similarly,
mixed binding modes were also noticed in CD titrations
(see below). A single type of complex seems to be dominant
at large excesses of DNA/RNA over the receptors, whereas
pH 5[a]
pH 7[b]
lmax
e/103
lmax
e/103
[nm]
[mmolꢀ1 cm2]
[nm]
[mmolꢀ1 cm2]
1
2
3
296
293
293
53.29
40.44
34.00
298
296
295
61.38
36.55
34.35
mixed binding modes appear at higher ratio (r=ACHTNUTRGNE[UNG receptor]/
[polynucleotide]>0.2). Therefore, we processed the titration
data at a ratio r<0.2 by means of the Scatchard equation[15]
to get log(KS) values, as well as the Scatchard ratio n=
[bound receptor]/[polynucleotide phosphate] (lower values
in Table 3). The results indicated a comparable binding af-
[a] Na cacodylate buffer, I=0.01m. [b] Na cacodylate buffer, I=0.01m.
the corresponding molar absorptivities (e) are given in
Table 1.
Table 3. Binding constants (log(KS)) and ratios n[a] ([bound 1–3]/[poly-
nucleotide phosphate]) calculated from the ITC and UV/Vis titrations of
1–3 with ds polynucleotides at pH 7 (Na cacodylate buffer, I=0.01m).
Protonation states: As the absorbance of compounds 1–3 is
pH dependent, their various pKa values could be determined
by means of UV/Vis spectrometry. For the guanidinium
groups we measured pKa values of 6.3 and 6.1 for com-
pounds 1 and 2, respectively. The two primary amino groups
in 2 have a common pKa value of 9.3. Compound 3 showed
three pKa values of 3.8, 6.5, and 8.1 for the two guanidinium
moieties and the two types of amino groups, respectively.
Due to the presence of two additional charges in 3 the pKa
value of the guanidinium groups is significantly decreased
compared to compounds 1 or 2. Therefore, the acylguani-
dine group of compound 1 is only partly protonated at pH 7,
but 2 and 3 are already fully protonated at their amine posi-
tion at this pH. At pH 5, compound 1 exists to more than
95% in the dicationic form, whereas 2 and 3 are expected to
be both tetracations, carrying two types of cationic groups,
an ammonium and a guanidinium cation in 2 and two differ-
ent ammonium cations in 3.
2
3
log(KS)[b]
n[b]
log(KS)[b]
n[b]
ctDNA
5.82
6.28
4.94
6.03
6.06
6.28
5.13
6.12
5.36
6.44
0.24
0.24
0.31
0.30
0.31
0.31
0.15
0.23
0.31
0.30
5.16
5.83
6.00
8.13
5.99
6.55
5.04
6.10
5.58
5.87
0.34
0.28
0.41
0.27
0.29
0.26
0.20
0.23
0.28
0.54
poly(dA)–poly(dT)
poly(A)–poly(U)
poly
poly
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
[a] Accuracy of n=(ꢁ10–30)%, therefore, log(KS) varies within an order
of magnitude. [b] Upper values for log(KS) and n were determined by
ITC, lower values were obtained from UV titration data by the use of
the Scatchard equation.[15]
finity of 1–3 toward most of the studied ds polynucleotides,
with generally higher log(KS) values at pH 5 (log(KS)=6.2–
8.5 (these values as well as data for compound 1 are given
in the Supporting Information; Table S1) than at pH 7 (log
KS =5.0–6.5), except for the binding of 3 to poly(dA)–
poly(dT), which shows a log(KS) of 8.1. This trend is not sur-
prising as all three compounds are higher positively charged
at pH 5 than at pH 7, thus confirming the importance of
electrostatic interactions for polynucleotide binding. Howev-
er, in order to verify the influence of mixed binding modes
on the obtained log(KS) values, it was necessary to apply an
independent method able to determine binding affinities at
high excesses of polynucleotide over receptor. For that pur-
pose we have chosen isothermal titration calorimetry (ITC).
Interactions of 1–3 with polynucleotides in aqueous solution
UV/Vis spectrophotometric titrations: Addition of any ds
polynucleotide to receptors 1–3 resulted in weak hypsochro-
mic and moderate hypochromic shifts of their UV/Vis bands
at l>300 nm, that is, at wavelengths where DNA and RNA
do not absorb. The spectral changes observed in the titra-
tions of 1–3 are summarized in Table 2. In all titrations a sig-
nificant deviation from the isosbestic point was observed at
l>300 nm, especially at pH 5. This points to the coexistence
of at least two different types of complexes formed upon
Table 2. Hypochromic effects (H [%])[a] calculated from the UV/Vis ti-
trations of 1–3 with ds polynucleotides at pH 5 and 7.
Isothermal titration calorimetry: ITC is at present the most
effective tool to characterize the thermodynamic profile of
interactions of small molecules with macromolecules such as
DNA and RNA, providing valuable information about the
enthalpic and entropic contributions to the binding Gibbs
energy. Moreover, for our specific interest, it was possible to
perform titrations under experimental conditions compara-
ble to UV/Vis and CD titrations (i.e., by adding the poly-
nucleotide to the receptor) as well as in “reversed” order
(i.e., by adding the receptor to the polynucleotide). Al-
though the former procedure in most cases confirmed mixed
binding modes, the “reversed” approach allowed to collect a
pH 5[b]
pH 7[c]
1
2
3
1
2
3
ctDNA
poly(dA)–poly(dT)
poly(A)–poly(U)
25.5
43.1
26.0
26.4
30.3
18.5
24.2
23.6
21.0
20.2
–
16.9
32.3
19.0
22.6
40.1
20.0
19.5
27.4
21.4
24.5
16.7
10.7
25.7
21.0
15.5
18.9
23
–
poly
poly
ACHTUNGTRENNUNG
A
–
[a] Titration data were processed according to the Scatchard equation[15]
hypochromic effect calculated by Scatchard for 1–3; H=[Abs
(1–3)ꢀAbs-
(complex)]/Abs(1–3)ꢂ100. [b] Na cacodylate buffer, ionic strength (I)=
;
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
0.01m. [c] Na cacodylate buffer, I=0.01m.
Chem. Eur. J. 2012, 18, 1352 – 1363
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1355

C. Schmuck, I. Piantanida et al.
large set of enthalpy changes at conditions in which the
polynucleotide was in huge excess, and therefore only one
binding mode was present. All ITC experiments were car-
ried out at pH 7. Due to the insufficient solubility of com-
pound 1 at this pH, only titrations of compounds 2 and 3
were performed. As an example, Figure 1 presents the re-
plexes both enthalpic and entropic contributions to the bind-
ing Gibbs energy are favorable, even for DNA and RNA
homopolynucleotides. These findings point out that our re-
ceptors interact with both dsDNA and dsRNA on the basis
of a similar set of binding interactions.
CD spectroscopy: So far, non-covalent interactions at 258C
were studied by monitoring spectroscopic properties of com-
pounds 1–3 upon addition of the polynucleotides. In order
to get an insight into the changes of the polynucleotide
properties induced by binding of the receptors, CD spectros-
copy was chosen as a highly sensitive method to probe con-
formational changes in the secondary structure of the poly-
nucleotides.[16] In addition, achiral molecules may become
chiral when bound to chiral polynucleotides. This induced
CD, which appears within the UV/Vis absorption range of
the bound molecule, can help to get an insight into the
mode of binding of the small molecule to the polynucleo-
tide. For instance, minor groove binding results in a strong
positive ICD band due to the orientation of the molecule
within an angle of approximately 458 in respect to the chiral
axis of the polynucleotide. On the other hand, intercalation
gives only a weak ICD band, in most cases of negative
sign.[17] Moreover, dimerization or even oligomerization of
the receptor through aromatic stacking could be detected by
specific strong bisignate exciton ICD.[17,18] It should be noted
that compounds 1–3 did not show any intrinsic CD spectra
under the experimental conditions.
The CD titrations (see the Supporting Information, Figur-
es S1–S3) showed indeed significant changes in the CD spec-
tra of the polynucleotides, hinting to changes in their secon-
dary structure. These results clearly indicate that the ob-
served changes are mainly controlled by two factors, namely
the type and availability of positive charges and the steric
properties of the minor groove. The poly(dA)–poly(dT)
DNA binding site is much narrower than that of poly-
Figure 1. Calorimetric titration of 2 (c=1.7ꢂ10ꢀ5 m) with ctDNA (c=
9.5ꢂ10ꢀ3 m) in Na cacodylate buffer (pH 7.0, I=0.01m; T=258C. a) Raw
ITC data. b) Dependence of successive enthalpy change per mole of ti-
ACHUTNRGEN(NUG dAdT)2, and the minor groove in polyCAHTUNGTRNEN(UGN dGdC)2 is signifi-
&
trant on ctDNA/2 molar ratio ( =experimental; c=calculated).
cantly sterically hindered by protruding amino groups of
guanine.[19,20]
sults of the microcalorimetric titration of compound 2 with
calf thymus (ct) DNA. In each case the binding was reflect-
ed by exothermic heat changes. The ITC data were fitted ac-
cording to a single-site model by using a non-linear curve
fitting procedure. The results of the ITC experiments are
presented in Tables 3 and Table S2 (see the Supporting In-
formation).
Interactions with poly(dA)–poly(dT): Only compound 1
showed bisignate exciton ICD bands with poly(dA)–
poly(dT), characterized by a positive sign at l=280 nm cou-
pled with a negative band at l=310 nm and isoelliptic
points for r=0.0–0.4 (pointing to one dominant binding
mode). In contrast, compounds 2 and 3 revealed ICD bands
of explicitly mixed binding modes (deviation from isoelliptic
The binding affinities (log(KS) values) obtained from the
“reversed” ITC titrations are in general somewhat lower
than those obtained from the Scatchard analysis of the UV
titrations (Table 3). These differences can be attributed to
the contribution of the mixed binding modes in the UV ex-
periments. Nevertheless, The Scatchard ratios n obtained
from UV and ITC titrations are the same within the error
margins of the methods, thus underlining the consistency of
the data. As can be seen from Table S2 (see the Supporting
Information), for almost all receptor/polynucleotide com-
points), for low ratios r=[2,3]/ACTHUGNETRNNU[G DNA]<0.2 characterized by
negative ICD bands at about l=270 nm coupled with posi-
tive bands at l=290 nm. Contrary, at higher ratios (r>0.2),
a very broad and strong ICD band is formed within the l=
260–320 nm range. As noted before, bisignate exciton ICD
bands point toward formation of higher aggregates than 1:1
complexes.[17] However, the opposite signs of the ICD bands
suggest that the molecules of 1 in these higher aggregates
are oriented in a different way compared to molecules of 2
or 3 (Figure 2).
1356
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1352 – 1363

Multicationic Bis(guanidiniocarbonylpyrrole) Receptors
FULL PAPER
Interactions with polyACHTUNGTRENN(UNG dGdC)2: Addition of compounds 1–3
to poly(dAdT)2 resulted in strong positive ICD bands at l=
AHCTUNGTRENNUNG
280–320 nm (lmax ꢂ305 nm). Isoelliptic points observed for
all three compounds at ratios of r=0.1–0.3 indicate the pres-
ence of a single dominant complex. At higher levels of 1–3
(r>0.3) the formation of additional complexes became ap-
parent. The absence of bisignate bands suggests that the
protruding guanine NH2 groups of poly (dGdC)2 sterically
hamper the formation of higher aggregates within the minor
groove, hence, compounds 1–3 are bound primarily as mon-
omers at low ratios (r<0.3), whereas at higher concentra-
tions non-specific agglomeration seems to occur.
Interactions with RNA (poly(A)–poly(U)): Addition of com-
pound 1 to poly(A)–poly(U) only decreased the intensity of
the CD bands of RNA at lmax =243 and 264 nm, but no ICD
band was observed. Differently, addition of 2 and 3 yielded
ICD bands at l>300 nm. However, the clear deviations
from isoelliptic points indicated aggregation of the receptor
molecules along the RNA groove with different orientations.
The recorded spectra at both pH values are similar, with the
exception that the addition of compound 3 yielded a much
stronger ICD band at l=310 nm at pH 7 than at pH 5
(Figure 4). This possibly results from a more uniform orien-
tation of small molecules with non-protonated guanidinio
groups along the RNA double helix. However, because the
minor groove of dsRNA is shallow and wide (thus unfavora-
Figure 2. CD spectra of poly(dA)–poly(dT) (c=3.0ꢂ10ꢀ5 m) in depend-
ence of the ratio of [1]/
ACHUTGTNRENNUG[poly(dA)–poly(dT)] (top) and [2]/ACHTUNTGREN[NUNG poly(dA)–
poly(dT)] (bottom) at pH 5 in Na cacodylate buffer (I=0.01m).
Interactions with polyACHTNUGRTENNUG(dAdT)2: For alternating polyCAHTUNGTERN(NUGN dAdT)2,
the same differences in signs of ICD bisignate exciton bands
between compounds 1–3 are observed as for poly(dA)–
poly(dT). However, the ICD bands of 2 and 3 at a low ratio
of r=0.1 were much more pronounced in case of alternating
polyACHTUNGTRENNUNG(dAdT)2. This can be related to the easier accommoda-
tion of the compounds within the much broader minor
groove in comparison to the narrow groove of poly(dA)–
poly(dT) (Figure 3).
Figure 4. CD spectra of poly(A)–poly(U) (c=3.0ꢂ10ꢀ5 m) in dependence
Figure 3. CD spectra of poly
(dAdT)2 (c=3.0ꢂ10ꢀ5 m) in dependence of
of the ratio of [3]/
ACHTUNGTREN[NUNG poly(A)–poly(U)] at pH 7 (top) and pH 5 (bottom) in
the ratio [2]/[poly(dAdT)2] at pH 5 in Na cacodylate buffer (I=0.01m).
G
ACHTUNGTRENNUNG
Na cacodylate buffer (I=0.01m).
Chem. Eur. J. 2012, 18, 1352 – 1363
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1357

C. Schmuck, I. Piantanida et al.
Table 4. DTm values[a] [8C] of various ds polynucleotides with receptors 1–3 at pH 5 and 7 (Na cacodylate
buffer, I=0.01m).
ble for binding of small mole-
cules) but the major groove of
RNA is deep and narrow,[20]
compounds 1–3 most likely
bind to the major groove. In
general, a decrease of the pH
value and accordingly the
number of positive charges of
1–3 results in somewhat stron-
ger ICD bands (with exception
of the 3–RNA complex).
pH 5
(dAdT)2
pH 7
r[b]
poly(dA)–
poly(dT)
poly
N
poly(A)–
poly(U)
poly(dA)–
poly(dT)
poly
A
poly(A)–
poly(U)
0.1
0.2
0.1
0.2
0.1
0.2
2.6
2.7
3.0
0.4/ꢀ0.3[d]
1.9/ꢀ0.8[d]
21.4/ꢀ1.8[d]
23.4[e]
0.2
ꢀ0.1
1.0
2.7
7.3
0.8
0.4
3.8
6.2
11.5
13.7
1.4
1.6
2.0
6.5
11.4
14.7
1
2
3
3.3/12.0[c]
10.0
11.3
18.2
15.1
17.5/24.3[c]
16.0/25.7[c]
26.0
27.5/ꢀ3.3[d]
19.5[c]
28.9[e]
9.4
[a] Error (ꢁ0.5)8C. [b] r=[1–3]/[polynucleotide]. [c] Biphasic thermal denaturation transitions due to the
mixed binding mode. [d] Biphasic transitions: the first transition at Tm =28.58C is attributed to denaturation of
poly(A)–poly(U), the second transition at Tm =80.18C is attributed to denaturation of poly
as poly(A) at pH 5 is mostly protonated and forms ds polynucleotide.[20, 22] [e] The second transition of poly-
(AH+)–poly(AH+) could not be determined due to overlapping with the first transition of poly(A)–poly(U).
(AH+)–poly(AH+)
ACHUTGTNRENNUG ACHTUNGTRENNUNG
Thermal denaturation of ds
polynucleotides: It is well
known that on heating double-
stranded helices of polynucleo-
A
ACHTUNGTRENNUNG
tides dissociate at a well-defined temperature (Tm) into two
single-stranded polynucleotides. Non-covalent binding of
small molecules like 1–3 to ds polynucleotides usually has
some effect on the thermal stability of helices, reflected by
changed (normally increased) Tm values. The difference be-
tween the Tm value of the free polynucleotide and the one
effects for all compounds (data not shown). A more detailed
analysis of the stabilization effects revealed a pronounced
non-linear relation between the DTm values and the ratios
r=[1–3]/[polynucleotide]. Such a behavior suggests satura-
tion of the binding sites at r=0.2–0.3, which agrees well
with the Scatchard ratio n as well as the stoichiometry as de-
termined in the ITC experiments.
of the complex with bound small molecule ACHTUNRGTNEUNG(DTm) is an im-
portant factor for the characterization of small molecule–ds
The degree of stabilization rises from compound 1 to 3,
with a remarkably high effect for the combination of
poly(A)–poly(U) and the highly charged compounds 2
(DTm =23.48C at r=0.2) and 3 (DTm =28.98C at r=0.2).
However, the difference in stabilization between 2 and 3 at
pH 5 is smaller than at pH 7, most likely because the pKa
value of the guanidinium cation in 2 (6.1) is smaller than the
pKa value of the primary amine in compound 3 (6.5). There-
fore, the degree of protonation of compounds 2 and 3 differs
at pH 7, but is almost the same at pH 5.
Most intriguingly, compounds 2 and 3 strongly stabilize
dsRNA at both pH values, giving DTm values somewhat
higher than those determined for dsDNA. Such a preference
of RNA over DNA is quite unusual for compounds that
cannot intercalate (as deduced from the CD data and in line
with the structure of compounds, which do not possess any
large aromatic moiety needed for intercalation), suggesting
the RNA major groove as a possible binding site.
polynucleotide interactions.[21]
Experiments were performed at pH 7 and 5 to probe the
impact of the different protonation states of our receptors
on the stability of the polynucleotides. At pH 7, compound
1 showed only weak stabilizing effects on any of the poly-
nucleotides, whereas, as expected, the higher-charged com-
pounds 2 and 3 induced increased thermal stabilization
(Figure 5 and Table 4). At pH 5, where at least two proton-
ated groups are present in all compounds, the stabilization is
even more pronounced and proportional to the number of
positive charges, that means, increases in the order 1<2<3.
In line with the UV/Vis titrations, these observations under-
line the importance of electrostatic interactions for binding
(Table 4). This was also confirmed by an additional experi-
ment at a five-fold increased ionic strength (cacodylate
buffer, I=0.05m), showing significantly decreased stabilizing
Atomic force microscopy (AFM) measurements of com-
pound 2 with ctDNA: The spectroscopic data clearly show
that especially compounds 2 and 3 interact quite strongly
with polynucleotides, not only stabilizing the double helices
but also inducing changes in their secondary structure. To
get a more detailed insight into these structural changes, the
interaction of 2 with ctDNA was studied by using AFM in
tapping mode on a mica surface. A stock solution of ctDNA
(c=5.0 mgmLꢀ1, not sonicated) in sodium cacodylate buffer
(I=0.2 mm) at pH 7.0 was used, which for all measurements
was diluted to a total DNA concentration of 2.5 mgmLꢀ1 by
adding water or an increasing amount of compound 2. Be-
cause a constant concentration of DNA was employed in all
experiments, the observed changes of the morphology of the
DNA are due to interactions of compound 2 with the poly-
Figure 5. Thermal denaturation curves of poly(A)–poly(U) (c=2ꢂ10ꢀ5 m)
at pH 7 (Na cacodylate buffer, I=0.01m) for a ratio r=0.2 of compounds
~
!
&
*
1–3 to polynucleotide ( =poly(A)–poly(U), =1, =2, =3).
1358
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1352 – 1363

Multicationic Bis(guanidiniocarbonylpyrrole) Receptors
FULL PAPER
Figure 6. a)–d) AFM images of ctDNA (c=2.5 mgmLꢀ1) in Na cacodylate buffer pH 7.0 (I=0.1 mm) on mica with different concentrations of compound
2: 0 (a), 10 (b), 15 (c), and 100 mm (d). e,f) Corresponding height profiles for the lines shown in a) and b), respectively. Size of the AFM images (1.5ꢂ
1.5) mm.
nucleotide and not due to changes of the concentration of
the DNA (which would cause similar effects on AFM). For
each measurement 10 mL of the receptor–DNA mixture
were dropped on a freshly cleaned mica surface, which was
dried by using spin-coating (1 min at 20 rps and 2 min at
100 rps). The AFM image of pure ctDNA is shown in Fig-
ure 6a along with the corresponding height profile. The
strand-like, elongated structure of the DNA is clearly visible
with heights ranging from 0.5 to 1.5 nm, suggesting a mix-
ture of double-stranded DNA and aggregates of dsDNA, as
usually observed in AFM experiments. After the addition of
2 (10 mm) structures similar in shape to the “rope-of-pearls”
structures found in nucleosomes were observed.[23] The
height of the “pearls” along the DNA strand was (2.5–
3.0) nm and hence, significantly larger than found for DNA
alone. This suggests a combination of dsDNA with the
multi-cationic compound 2 wrapped around (see Figure 6b).
Further raising the concentration of compound 2 to 15 mm
resulted in an increase of intra- and intermolecular junc-
tions, simultaneously with the elongated structure of the
DNA becoming more and more compact. At 100 mm only
aggregates of highly condensed DNA were found. These
findings paralleled the changes in the CD spectra of the
polynucleotides, clearly reflecting significant changes of the
conformation of the DNA on addition of compounds 1–3.
Similar effects on the morphology of polynucleotides have
been described in the literature for distamycin A, 4’,6-diami-
dino-2-phenylindole (DAPI), and Hoechst 33258.[24–26]
Discussion of the binding studies: According to the above
results, compounds 1–3 bind strongly to both dsDNA and
dsRNA. Binding is mainly caused by electrostatic interac-
tions. Results from CD experiments, thermal stabilization
effects, and structure–activity analyses strongly suggest bind-
ing of compounds 1–3 to the minor groove of DNA as the
dominant binding mode. However, at equimolar ratios of
[1–3]/ACHTNUTRGNE[NUG DNA] mixed binding modes were also detected, most
likely due to the aggregation of the compounds within the
binding site. Thermal melting studies as well as similar affin-
ities and enthalpy changes obtained for dsRNA and the cor-
responding DNA, indicate a similar set of binding interac-
tions for 1–3, pointing toward the major groove of RNA as
the dominant binding site. Most obviously, the main driving
force for the binding of compounds 1–3 to the polynucleo-
tides is of electrostatic nature, consequently, higher affinities
and stronger binding effects are obtained at weakly acidic
conditions. However, steric effects also play a role, resulting
in different orientations of the compounds within different
DNA and RNA binding sites. Thereby, binding affinities
and thermal stabilization effects were significantly dimin-
ished. AFM experiments further showed that the morpholo-
gy of the DNA changes dramatically upon binding of com-
pound 2. The concentration dependence is visible in both,
the size of the generated structures as well as in their mor-
phology.
Chem. Eur. J. 2012, 18, 1352 – 1363
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1359

C. Schmuck, I. Piantanida et al.
(329 rpm) into the calorimeter reaction cell containing 1.4514 mL of the
receptor solution (cꢂ0.02 mm). “Reverse” titrations were also performed
by adding aliquots of the receptor solution (30ꢂ5 mL, c=0.8 mm) to the
polynucleotide solution (c=0.15 mm). Blank experiments were carried
out to determine the heats of dilution of the compounds and the polynuc-
leotides. All solutions used in the ITC experiments were degassed under
vacuum prior to use to eliminate air bubbles. Each injection generated a
heat burst curve (P in mW versus time). The data were imported to
Origin 7.0 and the area under each peak was determined by integration
to evaluate the heat associated with the injection. The data were correct-
ed for heats of dilution. The resulting data were analyzed by using the
OriginPro 7.5 software according to a model based on a single set of
identical binding sites to estimate the binding constant (KS), the binding
stoichiometry (n) and the enthalpy of binding (DrH8). The reaction Gibbs
energies (DrG8) were calculated by using Equation (1).
Conclusion
In conclusion, we have shown here that two-armed polycat-
ionic artificial receptors strongly interact with both dsDNA
and dsRNA. The most likely binding mode as determined
by ITC and CD spectroscopy is interaction with the minor
groove in DNA and the major groove in RNA. This interac-
tion not only strongly stabilizes the double strands and indu-
ces a significant increase in the melting temperatures, but
also leads to a concentration-dependent condensation of the
nucleic acid as evident from the AFM pictures. Such com-
pounds might be of interest for further development as arti-
ficial transfection vectors for gene delivery.
DrGꢃ ¼ ꢀRTlnðKSÞ
ð1Þ
Entropic contribution to the binding Gibbs energy was calculated by
Equation (2).
Experimental Section
TDrSꢃ ¼ DrHꢃꢀDrGꢃ
ð2Þ
General remarks: Solvents were dried and distilled before use. All ex-
periments were run in oven-dried glassware. The compounds were dried
in high vacuum over orange gel at room temperature overnight unless
otherwise stated. All melting points were measured with a Bꢃchi melting
point B-540 apparatus according to Dr. Tottoli with open-end glass capil-
lary tubes. 1H and 13C NMR spectra were recorded at room temperature
on a Bruker Avance 400, a Bruker DMX 300, or a Bruker DMX 500
spectrometer as mentioned in the text. The chemical shifts are reported
relative to the residual proton signals of the used solvents. Apparent cou-
pling constants are given in Hertz. All IR spectra were measured on a
JASCO FT/IR 430 with a MIRacle ATR from PIKE or on a Perkin–
Elmer FTIR 1600 spectrometer. The MALDI spectra were recorded on a
Bruker Daltonics autoflex II, the ESI spectra were recorded on a Bruker
Daltonics microTOF.
Sample preparation and AFM-imaging: The samples were prepared by
diluting a DNA stock solution (c=5 mgmLꢀ1) in sodium cacodylate
buffer (I=0.2 mm, pH 7.0) with water to
a final concentration of
2.5 mgmLꢀ1. For the mixtures of concentration of 2 of 10 and 15 mm, an
aqueous stock solution of 0.5 mm was used, for the mixture with c(2)=
100 mm the concentration of the stock solution was 5.0 mm. For each mea-
surement 10 mL of the mixed solution were dropped onto a freshly
cleaned mica surface (Plano GmbH). The sample was dried by means of
a spin-coater (1 min at 20 rps and 2 min at 100 rps). For the AFM imag-
ing operating in tapping mode (scan rate 5 mmsꢀ1) with N-type silicon
cantilevers (AC-160TS, Olympus) a NanoDrive controller with an Innova
scannig probe microscope (Veeco, Germany, Mannheim) was used. The
analysis of the AFM-images was carried out by use of the Gwyddion
(version 2.19) software.[28]
Spectroscopic experiments: Electronic absorption spectra were obtained
on a Varian Cary 100 Bio spectrometer, and the CD spectra on a JASCO
J815 spectrophotometer. Hellma precision cells 100-QS (10 mm) made of
SUPRASIL quartz were used. The spectroscopic studies were performed
in aqueous buffer solution (pH 7, sodium cacodylate buffer, I=0.01 and
0.0 m; pH 5, sodium cacodylate buffer, I=0.01m). At the experimental
conditions, the absorbance of compounds 1–3 strictly followed the Lam-
bert–Beer law. Polynucleotides were purchased as noted: poly(A)–
Cbz- and Boc-protected acid 5: Lithium hydroxide (32 mg, 1.30 mmol,
1.5 equiv) was added to
a solution of the methyl ester 4 (460 mg,
0.87 mmol, 1 equiv) in THF/water (4:1; 7 mL). The resulting solution was
stirred at room temperature for 4 h. After complete reaction (tlc) the sol-
vent was evaporated, the resulting residue was dissolved in water, and
dried by lyophilization. The solid was dissolved in water. By adding aque-
ous solution of HCl (5%) till a pH of 5–6 was reached a white solid pre-
cipitates. The solid was filtered off, washed with ice water, and was dried
with phosphorous pentoxide to give 5 (400 mg, 0.77 mmol, 89%) as a col-
orless solid. Rf =0.42 (SiO2, dichloromethane/methanol 9:1); m.p. 2338C;
1H NMR (400 MHz, [D6]DMSO, 258C): d=1.45 (s, 9H; Boc-CH3), 3.49–
3.65 (m, 2H; CH2), 4.15–4.19 (m, 1H; CH), 5.02 (s, 2H; benzyl-CH2),
6.76 (s, 1H; pyrrole-CH), 6.81 (s, 1H; pyrrole-CH), 7.27–7.34 (m, 5H;
aryl-CH), 7.50 (d, 1H; 3J=12.2 Hz, Cbz-NH), 8.51 (s, 1H; guanidine-
NH), 8.58 (s, 1H; guanidine-NH), 9.34 (s, 1H; guanidine-NH), 11.41 ppm
(brs, 1H; pyrrole-NH); 13C NMR (100 MHz, [D6]DMSO, 258C): d=27.8
(Boc-CH3), 40.4 (CH2), 54.3 (CH), 112.1 (pyrrole-CH), 113.7 (pyrrole-
CH), 127.7 (aryl-CH), 128.3 (aryl-CH), 136.9 (Cq), 155.9 (Cq), 158.4
(Cq), 160.1 (Cq), 172.3 ppm (Cq); IR (ATR): n˜ =3302 (brw), 1689 (m),
1621 (m), 1552 (m), 1251 (s), 1146 (s), 1055 cmꢀ1 (s); HRMS (ESI): m/z:
calcd for C23H28N6O8: 515.190 [M]ꢀ; found 515.192.
poly(U), poly(dA)–poly(dT), polyACHTNUTRGENN(GU dAdT)2, polyAHCTUTGNERNN(UGN dGdC)2 (Sigma),
ctDNA (Aldrich). Polynucleotides were dissolved in sodium cacodylate
buffer (I=0.05m, pH 7). ctDNA was additionally sonicated and filtered
through a 0.45 mm filter.[27] Polynucleotide concentration was determined
spectroscopically[27] and given as the concentration of the phosphate
groups. Spectroscopic titrations were performed by adding portions of
polynucleotide solution into the a solution of the compound being stud-
ied. Obtained data were corrected for dilution. Titration data were pro-
cessed by the Scatchard equation.[15] Values for KS and n given in the
Supporting Information all have satisfactory correlation coefficients (r>
0.999). Thermal melting curves for DNA, RNA, and their complexes
with compounds 1–3 were determined by monitoring the absorption
change at l=260 nm as a function of temperature.[21] For each curve, the
absorbance of the ligand was subtracted, and the absorbance scale was
normalized. The Tm values are the midpoints of the transition curves, de-
termined from the maximum of the first derivative and checked graphi-
cally by the tangent method.[21] The DTm values were calculated by sub-
tracting Tm of the free nucleic acid from Tm of the complex. The DTm
values reported here are the averages of at least two measurements, the
error in DTm is (ꢁ0.5)8C.
Di-(Cbz- and Boc-protected) EDA compound 6: Acid
5 (150 mg,
0.29 mmol, 1 equiv) and HCTU (133 mg, 0.32 mmol, 1.1 equiv) were dis-
solved in dichloromethane (15 mL), NMM (95 mL, 0.87 mmol, 3 equiv)
and DMF (5 mL). After addition of ethylenediamine (9 mL, 0.15 mmol,
0.5 equiv), the reaction mixture was stirred at room temperature for 20 h.
The solvent was evaporated and by pouring the residue into ice water
(20 mL) a colorless solid precipitated. This solid was filtered off, washed
with water, and was dried by lyophilization to give the desired product 6
(140 mg, 0.13 mmol, 88%). Rf =0.60 (SiO2, dichloromethane/methanol
9:1); m.p. 2108C; 1H NMR (400 MHz, [D6]DMSO, 258C): d=1.46 (s,
18H; Boc-CH3), 3.11 (s, 4H; CH2), 3.35 (s, 4H; CH2), 4.14–4.17 (m, 2H;
Isothermal titration calorimetry: ITC experiments were performed by
means of an isothermal titration microcalorimeter Microcal VP-ITC (Mi-
croCal, Inc., Northampton, MA, USA) at 25.08C. Origin 7.0 software,
supplied by the manufacturer, was used for data acquisition and manipu-
lation. In the titration experiments aliquots of the polynucleotide (30ꢂ
5 mL, c=1–10 mm) were injected from
a 297 mL rotating syringe
1360
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1352 – 1363

Multicationic Bis(guanidiniocarbonylpyrrole) Receptors
FULL PAPER
CH), 5.02 (s, 4H; benzyl-CH2), 6.77 (s, 4H; pyrrole-CH), 7.27–7.34 (m,
12H; aryl-CH, NH), 8.07 (s, 2H; NH), 8.40 (s, 2H; NH), 8.55 (s, 2H;
guanidine-NH), 9.32 (s, 2H; guanidine-NH), 10.91 (brs, 2H; guanidine-
NH), 11.29 ppm (brs, 2H; pyrrole-NH); 13C NMR (100 MHz,
[D6]DMSO, 258C): d=27.8 (Boc-CH3), 38.3 (CH2), 40.7 (CH2), 55.2
(CH), 65.6 (benzyl-CH2), 112.1 (pyrrole-CH), 127.7 (aryl-CH), 128.3
(aryl-CH), 136.8 (Cq), 155.8 (Cq), 158.4 (Cq), 160.3 (Cq), 170.0 ppm
(Cq); IR (ATR): n˜ =1720 (m), 1627 (m), 1532 (m), 1239 (s), 1146 (s),
1051 (m), 842 cmꢀ1 (s); MALDI MS: m/z: 1057.1 [M+H]+, 957.1
[MHꢀBoc]+, 857.1 [M+Hꢀ2Boc]+.
Dicationic compound 1: The protected compound 6 (100 mg, 0.10 mmol)
was dissolved in TFA (3 mL) and stirred at room temperature for 2 h.
The TFA was evaporated, the obtained residue was dissolved in aqueous
solution of HCl (5%, 1.5 mL) and lyophilized to yield a colorless solid
(37 mg, 0.04 mmol, 40%). Rf =0.24 (SiO2, ethyl acetate/methanol 9:1+
NEt3 (0.1%); m.p. 2438C; 1H NMR (400 MHz, [D6]DMSO, 258C): d=
3.11 (s, 4H; CH2), 3.56 (s, 4H; CH2), 4.16–4.21 (m, 2H; CH), 4.97–5.06
(m, 4H; benzyl-CH2), 6.85–6.87 (m, 2H; pyrrole-CH), 7.27–7.36 (m,
14H; aryl-CH, pyrrole-CH, NH), 8.10 (s, 2H; NH), 8.41 (s, 8H; NH2),
8.51 (s, 2H; guanidine-NH), 11.69 (s, 2H; guanidine-NH), 12.38 ppm (s,
2H; pyrrole-NH); 13C NMR (100 MHz, [D6]DMSO, 258C): d=38.5
(CH2), 41.0 (CH2), 55.0 (CH), 65.6 (benzyl-CH2), 112.8 (pyrrole-CH),
115.8 (pyrrole-CH), 125.6 (aryl-CH), 127.7 (aryl-CH), 128.3 (aryl-CH),
132.5 (Cq), 136.8 (Cq), 155.4 (Cq), 155.8 (Cq), 159.6 (Cq), 168.9 (Cq),
169.9 ppm (Cq); IR (ATR): n˜ =3061 (brm), 2778 (w), 2359 (w), 1696 (s),
1555 (s), 1255 (s), 1066 cmꢀ1 (m); MALDI: m/z : 857.3 [M+H]+.
(ESI.): m/z: calcd for C20H30N6O8Na: 505.202 [M+Na]+; found 505.202
[M+Na]+, 987.417 [2M+Na]+.
Coupling of 8 with ethylendiamine to the tetra-Boc-protected compound
9: A mixture of the acid 8 (200 mg, 0.42 mmol, 1 equiv), HCTU (191 mg,
0.46 mmol, 1.1 equiv), NMM (138 mL, 1.26 mmol, 3 equiv), and ethylendi-
amine (12 mL, 0.2 mmol, 1/2 equiv) in dichloromethane (20 mL) and
DMF (6 mL) was stirred at room temperature for 40 h. The solvent was
evaporated and the residue taken up in ice water (75 mL). The obtained
white solid was filtered off, washed with water, and lyophilized to yield a
colorless powder (152 mg, 0.15 mmol, 77%). Rf =0.16 (SiO2, dichlorome-
thane/methanol 9:1); m.p. 2158C (decomp); 1H NMR (400 MHz,
[D6]DMSO, 258C): d=1.34 (s, 18H; Boc-CH3), 1.46 (s, 18H; Boc-CH3),
3.09 (s, 4H; EDA-CH2), 3.48–3.51 (m, 4H; CH2), 4.07–4.09 (m, 2H; CH),
6.74–6.77 (m, 6H; pyrrole-CH, Boc-NH), 7.99 (s, 2H; EDA-NH), 8.37 (s,
2H; NH), 8.56 (brs, 2H; guanidine-NH), 9.32 (brs, 2H; guanidine-NH),
10.83 (brs, 2H; guanidine-NH), 11.18 ppm (brs, 2H; pyrrole-NH);
13C NMR (100 MHz, [D6]DMSO, 258C): d=28.1 (Boc-CH3), 38.3 (CH2),
40.8 (CH2), 54.8 (CH), 78.4 (Boc-Cq), 108.0 (pyrrole-CH), 112.0 (pyrrole-
CH), 155.1 (Cq), 158.3 (Cq), 160.2 (Cq), 170.2 ppm (Cq); IR (KBr): n˜ =
3385 (brm), 2979 (m), 2936 (w), 1637 (s), 1547 (s), 1297 (s), 1151 (s),
842 cmꢀ1 (w); HRMS (ESI): m/z: calcd for C42H64N14O14Na: [M+Na]+
1011.462; found 1011.462 [M+Na]+, 889.425 [M+HꢀBoc]+, 789.372
[M+Hꢀ2Boc]+.
Boc-deprotection of 9 to the tetracationic compound 2: A solution of the
Boc-protected compound 9 (140 mg, 0.14 mmol) in TFA (3 mL) was
stirred at room temperature for 2 h. The TFA was evaporated in vacuo,
the obtained residue was dissolved in aqueous solution of HCl (1.5 mL,
5%) and lyophilized to give compound 2 as a colorless solid (100 mg,
Di-Boc-protected methyl ester 7: The Cbz-protected compound 4 (2.15 g,
4.05 mmol) and Pd/C (0.26 g, 10%) were dissolved in methanol (50 mL)
and stirred under hydrogen atmosphere for 2 h at 358C. After complete
reaction (tlc), di-tert-butyldicarbonat (0.97 g, 4.46 mmol, 1.1 equiv) was
added and the reaction mixture was stirred for one additional hour at
358C till the reaction was complete (tlc). The catalyst was filtered off
through a Celite pad and washed thoroughly with methanol. The filtrate
and the combined washing phases were evaporated and lyophilized to
give the methyl ester 7 as a colorless powder (2.01 g, 4.05 mmol, quant.).
Rf =0.48 (SiO2, dichloromethane/methanol 9:1 after Cbz deprotection);
Rf =0.59 (SiO2, dichloromethane/methanol 9:1 after Boc protection);
m.p. 1338C; 1H NMR (400 MHz, [D6]DMSO, 258C): d=1.37 (s, 9H;
Boc-CH3), 1.46 (s, 9H; Boc-CH3), 3.55–3.57 (m, 2H; CH2), 3.61 (s, 3H;
CH3), 4.19–4.21 (m, 1H; CH), 6.75 (s, 1H; pyrrole-CH), 6.80 (s, 1H; pyr-
role-CH), 7.20 (d, 1H; 3J=7.56 Hz, NH), 8.44 (s, 1H; NH), 8.56 (brs,
1H; guanidine-NH), 9.32 (brs, 1H; guanidine-NH), 10.81 (brs, 1H; gua-
nidine-NH), 11.18 ppm (brs, 1H; pyrrole-NH); 13C NMR (100 MHz,
[D6]DMSO, 258C): d=27.8 (Boc-CH3), 28.1 (Boc-CH3), 39.9 (CH2), 51. 9
(CH3), 53.6 (CH), 112.0 (pyrrole-CH), 155.3 (Cq), 158.4 (Cq), 160.3 (Cq),
171.0 ppm (Cq); IR (KBr): n˜ =3388 (brm), 3269 (brm), 2979 (m), 2954
(m), 1726 (s), 1151 (vs), 756 cmꢀ1 (m); HRMS (ESI): m/z: calcd for
C21H32N6O8Na: 519.217 [M+Na]+; found 519.218 [M+Na]+, 1015.451
[2M+Na]+.
1
0.14 mmol, 97%). M.p.>2508C; H NMR (400 MHz, [D6]DMSO, 258C):
d=3.16–3.24 (m, 4H; EDA-CH2), 3.63–3.78 (m, 4H; CH2), 3.95 (d, 2H;
3J=5.52 Hz, CH), 6.95 (d, 2H; 3J=1.88 Hz, pyrrole-CH), 7.53 (d, 2H;
3J=1.76 Hz, pyrrole-CH), 8.40 (s, 8H; NH2), 8.61 (brs, 6H; NH3+), 8.82–
8.84 (m, 4H; amide-NH), 12.03 (s, 2H; guanidine-NH), 12.38 ppm (s,
2H; pyrrole-NH); 13C NMR (100 MHz, [D6]DMSO, 258C): d=38.3
(CH2), 39.5 (CH2), 52.4 (CH), 108.0 (pyrrole-CH), 113.1 (pyrrole-CH),
125.7 (pyrrole-Cq), 132.2 (Cq), 155.5 (Cq), 159.6 (Cq), 160.0 (Cq),
166.8 ppm (Cq); IR (KBr): n˜ =3325 (brm), 3209 (brm), 3098 (brm), 2971
(brm), 2869 (brm), 1699 (s), 1557 (m), 1283 (m), 1200 (m), 754 cmꢀ1 (w);
HRMS (ESI): m/z: calcd for C22H33N14O8: 589.270 [M]+; found 589.273
[M]+.
Boc-protection of the secondary amine of the Cbz- and Boc-protected
methyl ester 10: Triethylamine (0.14 mL, 1.46 mmol, 1.5 equiv) was
added to a solution of the methyl ester 10 (500 mg, 0.97 mmol, 1 equiv)
in methanol (30 mL). Under vigorously stirring a solution of di-tert-butyl-
dicarbonat (230 mg, 1.06 mmol, 1.1 equiv) in methanol (10 mL) was
added dropwise, and the resulting suspension was stirred for 18 h at room
temperature. Afterwards the solvent was evaporated. By adding water a
white solid precipitated. The solid was filtered off, washed with ice water,
and lyophilized giving the desired product 11 (567 mg, 0.92 mmol, 94%)
as a colorless powder. Rf =0.76 (SiO2, dichloromethane/methanol 9:1);
m.p. 1048C; 1H NMR (400 MHz, [D6]DMSO, 258C): d=1.37 (s, 9H;
Boc-CH3), 1.46 (s, 9H; Boc-CH3), 3.57–3.61 (m, 2H; CH2), 3.63 (s, 3H;
CH3), 4.26–4.45 (m, 3H; CH, CH2), 5.04 (s, 2H; benzyl-CH2), 6.02 (s,
1H; pyrrole-CH), 7.10 (brs, 1H; pyrrole-CH), 7.32–7.35 (m, 5H; aryl-
CH), 7.72 (s, 1H; Cbz-NH), 8.33 (brs, 1H; guanidine-NH), 9.41 (brs,
1H; guanidine-NH), 11.30 (brs, 1H; guanidine-NH), 11.72 ppm (brs,
1H; pyrrole-NH); 13C NMR (100 MHz, [D6]DMSO, 258C): d=27.8
(Boc-CH3), 27.9 (Boc-CH3), 31.3 (CH3), 45.6 (CH2), 47.0 (CH2), 52.1
(CH), 65.7 (benzyl-CH2), 79.6 (Boc-Cq), 108.3 (pyrrole-CH), 127.7 (aryl-
CH), 127.8 (aryl-CH), 128.3 (aryl-CH), 136.8 (aryl-Cq), 156.0 (Cq),
171.9 ppm (Cq); IR (KBr): n˜ =3382 (brm), 2977 (m), 1728 (s), 1698 (s),
1153 (s), 1057 (m), 843 cmꢀ1 (w); HRMS (ESI): m/z: calcd for
C29H41N6O9: 617.293 [M]+; found 617.294 [M]+, 517.237 [MꢀBoc]+.
Di-Boc-protected acid 8: Lithium hydroxide (182 mg, 7.40 mmol,
1.5 equiv) was added to a solution of methyl ester 7 (2.00 g, 4.03 mmol,
1 equiv) in THF/water (20 mL, 4.1). The resulting solution was stirred for
4 h at room temperature. After complete reaction (tlc) the solvent was
evaporated and the resulting residue was dissolved in water. By adding
aqueous solution of HCl (5%) till a pH of 5–6 was reached a white solid
precipitates. The solid was filtered off, washed with ice water, and was
dried by lyophilization to give the acid 8 (1.10 g, 2.28 mmol, 57%) as a
colorless solid. Rf =0.20 (SiO2, dichloromethane/methanol 9:1); m.p.
2368C (decomp); 1H NMR (400 MHz, [D6]DMSO, 258C): d=1.37 (s,
9H; Boc-CH3), 1.46 (s, 9H; Boc-CH3), 3.50–3.57 (m, 2H; CH2), 4.09–4.11
(m, 1H; CH), 6.74 (s, 1H; pyrrole-CH), 6.81 (s, 1H; pyrrole-CH), 6.97
(d, 1H; 3J=7.84 Hz, NH), 8.43 (s, 1H; NH), 8.56 (brs, 1H; guanidine-
NH), 9.32 (brs, 1H; guanidine-NH), 10.81 (brs, 1H; guanidine-NH),
11.16 ppm (brs, 1H; pyrrole-NH); 13C NMR (100 MHz, [D6]DMSO,
258C): d=27.8 (Boc-CH3), 28.1 (Boc-CH3), 40.3 (CH2), 53.8 (CH), 78.2
(Boc-Cq), 112.0 (pyrrole-CH), 155.3 (Cq), 158.4 (Cq), 160.1 (Cq),
172.4 ppm (Cq); IR (KBr): n˜ =3371 (brm), 2979 (m), 2936 (w), 1695 (s),
1631 (s), 1557 (s), 1294 (s), 1151 (s), 841 (w), 758 cmꢀ1 (w); HRMS
Cbz- and di-Boc-protected acid 12: Lithium hydroxide (30 mg,
1.22 mmol, 1.5 equiv) was added to a solution of the methyl ester 11
(500 mg, 0.81 mmol, 1 equiv) in THF/water (10 mL, 4:1). The resulting
solution was stirred at room temperature for 5 h. After complete reaction
(tlc) the solvent was evaporated, the resulting residue was dissolved in
Chem. Eur. J. 2012, 18, 1352 – 1363
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1361

C. Schmuck, I. Piantanida et al.
water, and dried by lyophilization. The resulting solid was dissolved in
water. By adding aqueous solution of HCl (5%) till a pH of 5–6 was
reached a colorless solid precipitates. The solid was filtered off, washed
with ice water, and was dried with phosphorous pentoxide to give the
acid 12 (430 mg, 0.71 mmol, 88%) as a colorless powder. Rf =0.50 (SiO2,
dichloromethane/methanol 9:1); m.p. 1878C; 1H NMR (400 MHz,
[D6]DMSO, 258C): d=1.38 (s, 9H; Boc-CH3), 1.43 (s, 9H; Boc-CH3),
3.59–3.63 (m, 2H; CH2), 4.25–4.33 (m, 2H; CH2), 4.38–4.48 (m, 1H;
CH), 5.04 (s, 2H; benzyl-CH2), 5.96 (s, 1H; pyrrole-CH), 6.96 (brs, 1H;
pyrrole-CH), 7.32–7.35 (m, 5H; aryl-CH), 7.51 (s, 1H; Cbz-NH), 8.55
(brs, 1H; guanidine-NH), 9.12 (brs, 1H; guanidine-NH), 11.46 ppm (brs,
1H; pyrrole-NH); 13C NMR (100 MHz, [D6]DMSO, 258C): d=27.9
(Boc-CH3), 27.9 (Boc-CH3), 43.1 (CH2), 47.2 (CH2), 52.9 (CH), 65.5
(benzyl-CH2), 79.4 (Boc-Cq), 108.0 (pyrrole-CH), 114.8 (pyrrole-CH),
127.7 (aryl-CH), 127.8 (aryl-CH), 128.3 (aryl-CH), 136.9 (aryl-Cq), 150.8
(Cq), 156.0 (Cq), 172.2 ppm (Cq); IR (KBr): n˜ =3353 (brm), 2978 (m),
1686 (s), 1146 (s), 1056 (w), 756 cmꢀ1 (w); HRMS (ESI): m/z: calcd for
C28H39N6O9: 603.277 [M+H]+; found 603.277 [M+H]+, 503.224
[M+HꢀBoc]+.
Coupling of 12 with ethylendiamine to the tetra-Boc-, di-Cbz-protected
compound 13: PyBOP (188 mg, 0.36 mmol, 1.1 equiv), N-methylmorpho-
line (108 mL, 0.99 mmol, 3 equiv), DMF (4 mL), and ethylenediamine
(9.6 mL, 0.16 mmol, 0.5 equiv) were added to a solution of the acid 12
(200 mg, 0.33 mmol, 1 equiv) in dichloromethane (15 mL). The solution
was stirred for 29 h at room temperature. After the solvent was evaporat-
ed and water was added a colorless solid precipitated. The suspension
was extracted with ethyl acetate (200 mL). The combined organic phases
were dried with magnesium sulphate, evaporated, and lyophilized. The
resulting solid was purified with MPLC (RP18, gradient; 10 to 90%
methanol in water within 90 min) and dried by lyophilization to yield a
colorless powder (100 mg, 0.08 mmol, 51%). Rf =0.19 (RP18, water/
methanol 3:7); m.p. 1858C (decomp); 1H NMR (400 MHz, [D6]DMSO,
258C): d=1.35 (s, 18H; Boc-CH3), 1.43 (s, 18H; Boc-CH3), 3.12 (s, 4H;
EDA-CH2), 3.46–3.50 (m, 4H; CH2), 4.24–4.30 (m, 4H; CH2), 4.36–4.40
(m, 2H; CH), 5.05 (s, 4H; benzyl-CH2), 5.94 (s, 2H; pyrrole-CH), 6.97
(brs, 2H; pyrrole-CH), 7.32–7.35 (m, 12H; aryl-CH, Cbz-NH), 8.05 (s,
2H; EDA-NH), 8.56 (brs, 2H; guanidine-NH), 9.12 (brs, 2H; guanidine-
NH), 10.84 (s, 2H; guanidine-NH), 11.48 ppm (brs, 2H; pyrrole-NH);
13C NMR (100 MHz, [D6]DMSO, 258C): d=27.9 (Boc-CH3), 27.9 (Boc-
CH3), 38.3 (CH2), 47.5 (CH2), 48.7 (CH2), 53.9 (CH), 65.6 (benzyl-CH2),
79.4 (Boc-Cq), 108.1 (pyrrole-CH), 114.7 (pyrrole-CH), 127.6 (aryl-CH),
127.8 (aryl-CH), 128.3 (aryl-CH), 136.8 (aryl-Cq), 152.1 (Cq), 155.8 (Cq),
170.1 ppm (Cq); IR (KBr): n˜ =3392 (brm), 2977 (w), 1675 (m), 1631 (m),
1244 (m), 1153 (m), 844 cmꢀ1 (w); HRMS (ESI): m/z: calcd for
C58H81N14O16: 1229.595 [M+H]+; found 1229.594 [M+H]+.
5%), and lyophilized to give the hexacationic compound 3 (48 mg,
0.062 mmol, 98%) as a slightly yellow solid. M.p.>2508C; 1H NMR
(400 MHz, [D6]DMSO, 258C): d=3.29 (s, 4H; EDA-CH2), 3.46–3.50 (m,
4H; CH2), 4.30–4.38 (m, 6H; CH, CH2), 6.53 (s, 2H; pyrrole-CH), 7.62
(s, 2H; pyrrole-CH), 8.46 (s, 6H; NH3+), 8.72–8.85 (m, 8H; NH2+), 9.17
(s, 2H; amide-NH), 9.89 (brs, 4H; NH2+), 12.04 (s, 2H; guanidine-NH),
12.24 ppm (s, 2H; pyrrole-NH); 13C NMR (100 MHz, [D6]DMSO, 258C):
d=38.6 (CH2), 42.9 (CH2), 46.4 (CH2), 49.6 (CH), 110.2 (pyrrole-CH),
116.1 (pyrrole-CH), 155.6 (Cq), 159.6 (Cq), 165.5 ppm (Cq); IR (KBr):
n˜ =3397 (brm), 2934 (m), 2865 (m), 2375 (m), 1685 (s), 1243 (m),
809 cmꢀ1 (w); HRMS (ESI): m/z: calcd for C22H36N14O4Na: 583.294
[M+Na]+; found 583.294 [M+Na]+.
Acknowledgements
C.S. thanks the DFG (Deutsche Forschungsgemeinschaft) and the Fonds
of the Chemical Industry for ongoing financial support of his work. I.P.
and V.T. thank the Ministry of Science, Education and Sports of Croatia
for financial support of this study (Projects 098-0982914-2918 and 119-
1191342-2960). We also acknowledge the support from a bilateral funding
project between the DAAD and the Ministry of Science, Education and
Sport of Croatia. We thank Johannes Hofmann for help with the HPLC
analysis and Dr. Hans-Gert Korth (university of Duisburg–Essen) for
helpful comments on the manuscript.
[1] R. B. Silverman, The Organic Chemistry of Drug Design and Drug
Action, Elsevier, New York, 2004.
[2] a) Small Molecule DNA and RNA Binders (Eds.: M. Demeunynck,
C. Bailly, W. D. Wilson), Wiley-VCH, Weinheim, 2002; b) M. J.
Hannon, Chem. Soc. Rev. 2007, 36, 280–295.
[3] S. Stella, D. Cascio, R. C. Johnson, Genes Dev. 2010, 24, 814–826.
[4] R. Rohs, S. M. West, P. Liu, B. Honig, Curr. Opin. Struct. Biol. 2009,
19, 171–177.
[5] A. Deiters, AAPS J. 2010, 12, 51–60.
[6] J. R. Thomas, P. J. Hergenrother, Chem. Rev. 2008, 108, 1171–1224.
[7] E. Trinquet, G. Mathis, Mol. BioSyst. 2006, 2, 380–387.
´
[8] L. Hernandez-Folgado, C. Schmuck, S. Tomic, I. Piantanida, Bioorg.
Med. Chem. Lett. 2008, 18, 2977–2981.
´
´
[9] L. Hernandez-Folgado, D. Baretic, I. Piantanida, M. Marjanovic, M.
Kralj, T. Rehm, C. Schmuck, Chem. Eur. J. 2010, 16, 3036–3056.
´
´
[10] K. Grçger, D. Baretic, I. Piantanida, M. Marjanovic, M. Kralj, M.
´
Grabar, S. Tomic, C. Schmuck, Org. Biomol. Chem. 2011, 9, 198–
209.
Cbz-deprotection of 13: The Cbz-protected compound 13 (100 mg,
0.08 mmol) and Pd/C (10 mg, 10%) were dissolved in methanol (25 mL)
and acetic acid (100 mL). The suspension was stirred under an hydrogen
atmosphere for 5 h at 408C. After complete reaction (tlc) the catalyst
was filtered off through a Celite pad and washed thoroughly with metha-
nol (100 mL). The filtrate and the combined washing phases were evapo-
rated and lyophilized to give the compound 14 (68 mg, 0.07 mmol, 82%)
as a colorless powder. Rf =0.27 (SiO2, dichloromethane/methanol 9:1);
m.p.>2508C; 1H NMR (400 MHz, [D6]DMSO, 258C): d=1.35 (s, 18H;
Boc-CH3), 1.45 (s, 18H; Boc-CH3), 3.17 (s, 4H; EDA-CH2), 3.37–3.39
(m, 6H; CH2, CH), 4.32–4.41 (m, 4H; CH2), 5.98 (d, 2H; 3J=3.68 Hz,
pyrrole-CH), 6.95 (d, 2H; 3J=3.16 Hz, pyrrole-CH), 8.20 (s, 2H; EDA-
NH), 8.55 (brs, 2H; guanidine-NH), 9.16 (brs, 2H; guanidine-NH),
11.61 ppm (brs, 2H; guanidine-NH); 13C NMR (100 MHz, [D6]DMSO,
258C): d=27.9 (Boc-CH3), 27.9 (Boc-CH3), 38.3 (CH2), 46.5 (CH2), 48.7
(CH2), 49.8 (CH), 79.4 (Boc-Cq), 108.1 (pyrrole-CH), 114.7 (pyrrole-
CH), 152.1 (Cq), 155.8 (Cq), 170.1 ppm (Cq); IR (KBr): n˜ =3398 (brm),
2976 (w), 1679 (m), 1629 (m), 1243 (m), 1154 cmꢀ1 (m); HRMS (ESI): m/
z: calcd for C42H69N14O12: 961.521 [M+H]+; found 961.522 [M+H]+.
[11] P. B. Dervan, Bioorg. Med. Chem. 2001, 9, 2215–2235.
[12] C. Schmuck, M. Schwegmann, Org. Lett. 2005, 7, 3517–3520.
[13] C. Schmuck, L. Geiger, Chem. Commun. 2005, 772–774.
[14] C. Schmuck, L. Geiger, J. Am. Chem. Soc. 2005, 127, 10486–10487.
[15] a) G. Scatchard, Ann. N. Y. Acad. Sci. 1949, 51, 660–672; b) J. D.
McGhee, P. H. von Hippel, J. Mol. Biol. 1976, 103, 679–684.
[16] A. Rodger, B. Nordꢄn, in Circular Dichroism and Linear Dichroism,
Oxford University Press, New York, 1997, Chapter 2.
[17] M. Eriksson, B. Nordꢄn, Methods Enzymol. 2001, 340, 68–98.
[18] a) B. Nordꢄn, F. Tjerneld, Biopolymers 1982, 21, 1713–1734; b) R.
Lyng, A. Rodger, B. Nordꢄn, Biopolymers 1991, 31, 1709–1820;
c) P. E. Schipper, B. Nordꢄn, F. Tjerneld, Chem. Phys. Lett. 1980, 70,
17–21.
[19] W. D. Wilson, Y. H. Wang, C. R. Krishnamoorthy, J. C. Smith, Bio-
chemistry 1985, 24, 3991–3999.
[20] C. R. Cantor P. R. Schimmel, Biophysical Chemistry, Vol. 3, WH
Freeman, San Francisco, 1980, pp. 1109–1181.
[21] J. L. Mergny, L. Lacroix, Oligonucleotides 2003, 13, 515–537.
ˇ
ˇ ´
´
´
´
[22] G. Malojcic, I. Piantanida, M. Marinic, M. Zinic, M. Marjanovic, M.
Boc-deprotection of 14 to the hexacationic compound 3: A solution of
the Boc-protected compound 14 (65 mg, 0.063 mmol) in TFA (4 mL) was
stirred at room temperature for 2 h. The TFA was evaporated in vacuo,
the obtained residue was dissolved in aqueous solution of HCl (1.5 mL,
´
Kralj, K. Pavelic, H. J. Schneider, Org. Biomol. Chem. 2005, 3,
4373–4381.
[23] M. Sato, K. Ura, K. I. Hohmura, F. Tokumasu, S. Yoshimura, F. Ha-
naoka, K. Takeyasu, FEBS Lett. 1999, 452, 267–271.
1362
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1352 – 1363

Multicationic Bis(guanidiniocarbonylpyrrole) Receptors
FULL PAPER
ˇ
[24] M. Saito, M. Kobayashi, S. Iwabuchi, Y. Morita, J. Biochem. 2004,
136, 813–823.
[25] K. Utsuno, M. Tsuboi, S. Katsumata, T. Iwamoto, Chem. Pharm.
Bull. 2002, 50, 216–219.
[28] P. Klapetek, D. Necas, Gwyddion—Free SPM (AFM, SNOM/
NSOM, STM, MFM) data analysis software. http://www.gwyddion.-
net (accessed 2010).
[26] Y. Li, U. Yildiz, K. Mꢃllen, F. Grçhn, Biomacromolecules 2009, 10,
530–540.
[27] a) J. B. Chaires, N. Dattagupta, D. M. Crothers, Biochemistry 1982,
Received: May 19, 2011
Revised: September 15, 2011
Published online: December 23, 2011
ˇ
´
21, 3927–3932; b) B. S. Palm, I. Piantanida, M. Zinic, H. J. Schneid-
er, J. Chem. Soc. Perkin Trans. 2 2000, 385–392.
Chem. Eur. J. 2012, 18, 1352 – 1363
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1363