مرکزی صفحہ ChemPlusChem DNA Intercalators for Detection of DNA Hybridisation: SCS(MI)-MP2 Calculations and Electrochemical...

DNA Intercalators for Detection of DNA Hybridisation: SCS(MI)-MP2 Calculations and Electrochemical Impedance Spectroscopy

, , , , ,
آپ کو یہ کتاب کتنی پسند ہے؟
فائل کی کوالٹی کیا ہے؟
کوالٹی کا جائزہ لینے کے لیے کتاب ڈاؤن لوڈ کریں
فائل کی کوالٹی کیا ہے؟
جلد:
81
زبان:
english
رسالہ:
ChemPlusChem
DOI:
10.1002/cplu.201600173
Date:
July, 2016
فائل:
PDF, 934 KB
0 comments
 

To post a review, please sign in or sign up
آپ کتاب کا معائنہ کر سکتے ہیں اور اپنے تجربات شیئر کرسکتے ہیں۔ دوسرے قارئین کتابوں کے بارے میں آپ کی رائے میں ہمیشہ دلچسپی رکھیں گے۔ چاہے آپ کو کتاب پسند ہے یا نہیں ، اگر آپ اپنے دیانتدار اور تفصیلی خیالات دیںگے تو لوگوں کو نئی کتابیں ملیںگی جو ان کے لئے صحیح ہیں۔
DOI: 10.1002/cplu.201600173

Full Papers

DNA Intercalators for Detection of DNA Hybridisation:
SCS(MI)–MP2 Calculations and Electrochemical Impedance
Spectroscopy
Daliborka Jambrec,[a] Raoudha Haddad,[a] Anna Lauks,[a] Magdalena Gebala,*[a, b]
Wolfgang Schuhmann,*[a] and Malte Kokoschka*[a]
a single intercalation site, and substantially lower with respect
to the minimum energy needed for binding with ssDNA. In EIS
studies, proflavine did not show any change in the chargetransfer resistance with respect to ssDNA and a decrease with
respect to dsDNA. Calculations showed that 1-pyrenemethylamine has sufficiently high interaction energy to intercalate into
dsDNA, however, the interaction energy towards ssDNA is
close to the minimum required value, suggesting a weak interaction with ssDNA. EIS measurements support the calculations.
A method for the calculation of interaction energies is provided, which can be used to characterise the interaction strength
between new intercalators and DNA before being synthesised.

Quantum mechanical SCS(MI)–MP2/cc-pVTZ calculations predict the strength of proflavine, ellipticine and 1-pyrenemethylamine intercalation into single-stranded (ss) and doublestranded (ds) DNA. The results were compared with experimental results obtained from electrochemical impedance spectroscopy (EIS). Similar interaction energies of ellipticine with
the guanine–cytosine base pair compared to the individual nucleobases guanine and cytosine suggested non-specific binding also to ssDNA. Accordingly, EIS identified ellipticine as
being non-selective and therefore unsuitable for the detection
of DNA hybridisation. The interaction energy of proflavine is
significantly higher than the minimum required energy for

Introduction
Molecular modelling is increasingly used to explain chemical
reaction mechanisms and to investigate the properties of compounds not yet synthesised. Pharmaceutical research already
relies to a large extent on drug candidates identified by means
of molecular modelling. It is therefore ; an interesting question,
to which extent these methods can assist in the development
of electrochemical DNA assays, which are based on the selective intercalation of suitable reporter compounds into dsDNA.
Molecular mechanics calculations, on the one hand, are applicable to large systems such as DNA oligomers including the
environment, due to lower computational demands, whereas
high-level quantum mechanical (QM) calculations (post-SCF
with a reasonably large basis set and counterpoise correction)
can give a more physical description of the interaction itself,
but only for much smaller systems. The latter methods have
been extensively tested and optimised for the treatment of
non-covalent interactions in recent years.[1] Although these
studies included systems of stacked nucleobases, to the best

of our knowledge only a few ab initio studies have been performed on the interaction of intercalators with DNA,[2, 3] as
these systems typically present a size that is still demanding at
the necessary post-SCF levels. However, this level of theory—
being the first that explicitly accounts for electron correlation—has the potential for predicting and functionally explaining the interaction between individual base pairs and intercalators. Several studies applying the significantly cheaper density
functional theory (DFT) can be found, but these results have to
be interpreted carefully as DFT is not able to correctly describe
dispersive interactions. The B3LYP functional in particular has
been shown to fail in the description of stacking interactions.[4, 5] Hobza et al. were able to show that even the X3 LYP
functional specifically designed for this purpose fails for the
description of p-stacking.[6] Functionals and methods that are
regarded as adequate to treat non-covalent systems include
meta-hybrid functionals[5, 7] parameterised in the Truhlar group,
Beckes half-and-half functional BH&H,[8] DFTB,[9, 10] DFT-DCP[11]
and the use of DCACPs[3] which are intended to correct also
the electronic structure of non-covalent systems. Grimme’s
atom-pairwise dispersion correction DFT-D, especially in its
newest version (D3) has been shown in several studies[12, 13] to
produce highly accurate interaction geometries and energies
and is, in our opinion, the most promising method for the
ab initio study of non-covalent interactions of molecules without explicit inclusion of electron correlation. However a causal
determination of interaction energies for new DNA–intercalator
complexes, expected to include dispersion and charge transfer
should include electron correlation, and therefore becomes

[a] D. Jambrec, Dr. R. Haddad, A. Lauks, Dr. M. Gebala,
Prof. Dr. W. Schuhmann, Dr. M. Kokoschka
Analytical Chemistry—Center for Electrochemical Sciences (CES)
Ruhr-Universitt Bochum
Universittstrasse 150, 44780 Bochum (Germany)
E-mail: wolfgang.schuhmann@rub.de
[b] Dr. M. Gebala
Present address:
Department of Biochemistry, Stanford University
Stanford, CA 94305 (USA)
The ORCID identification number(s) for the author(s) of this article can
be found under http://dx.doi.org/10.1002/cplu.201600173.

ChemPlusChem 2016, 81, 1 – 10

These are not the final page numbers! ÞÞ

1

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

Full Papers
only possible at a reasonable post-SCF level. We therefore propose a combination of these two techniques, dispersion-corrected density functional theory (DFT-D3) for the calculation of
accurate structures with subsequent determination of the interaction energy using counterpoise-corrected SCS(MI)–MP2
calculations in an economic (cc-pVTZ) basis set.
For their application in electrochemical DNA sensors, intercalators present a useful compromise between improved detection sensitivity and the simplicity of a sensor by avoiding
direct labelling of the target DNA sequence.[14] In electrochemical impedance spectroscopy (EIS)-based hybridisation detection schemes, intercalators are a useful tool for differentiating
between ssDNA and dsDNA.[15] Additionally, they are used for
signal validation to exclude false positives by unspecific modification of the interface.[16] In this study, we used EIS to experimentally challenge previously performed calculations and to
investigate the selectivity of the used intercalators towards
dsDNA.

which, because of the neighbouring methylene group, is more
flexible in favour of out-of-aromatic-plane displacement. Therefore, there is no risk of obtaining both mono- and bis-conjugates upon modification as in the case of proflavine; monosubstitution is expected to have less steric impact on the intercalation geometry.
Quantum mechanical interaction energies
The calculation of QM interaction energies between intercalators and nucleobase pairs is difficult and resource consuming.
It is difficult because many techniques such as Hartree–Fock
(HF) and the semi-empirical methods based on HF do not include electron correlation and are therefore inadequate for
treating the systems of interest. For DFT, the situation is more
complex. As DFT includes certain effects from electron correlation it is in many cases more accurate than HF at similar or
even lower cost. However most GGA and hybrid DFT functionals are not able to recover dispersive interactions, as present in
nucleobase stacking and DNA binding, correctly. Therefore the
second-order Møller–Plesset (MP2) theory was for a long time
the lowest (although computationally expensive) theoretical
level able to correctly describe dispersion-dominated complexes.[22] However, MP2 tends to underestimate intermolecular
distances.[23] Because of the slow convergence towards the
complete basis set (CBS) limit, the inclusion of a counterpoise
correction for the basis set superposition error (BSSE) is necessary even for geometry optimisations, which again increases
the computational costs. Grimme’s atom-pair-wise dispersion
correction for DFT has, on the other hand, been shown to reproduce results of high-level calculations very well and is, because of the significantly lower scaling, applicable to geometry
optimisations of relatively large molecular structures. Geometry
optimisations with dispersion-corrected DFT are quite possible
within the TZVPP basis set, even for large molecules. Due to
the more favourable convergence behaviour of DFT,[24] a counterpoise correction is then not necessary, as this level is already
close to the basis set limit for DFT[24] and Grimme’s dispersion
correction has been parameterised without counterpoise correction using the slightly larger QZVP basis set.[12]
The geometries of nine nucleobase adducts involving the
three intercalators of interest have been calculated using the
GGA functional PBE in combination with the third version of
Grimme’s atom-pairwise dispersion correction (PBE-D3). PBE
has been frequently used for the calculation of H-bonded systems and has been shown to effectively describe hydrogen
bonding[25] as it implicitly accounts for van der Waals attraction.[26] Augmented with Grimme’s atom-pair-wise dispersion
correction, an accurate description of nucleic acids has been
found, which is also superior to that by the more expensive
meta-hybrid functionals M05-2X and M06-2X.[7]
In the proflavine–base pair adduct, the intercalator shows
more overlap with the cytosine than with the guanine. This
does, however, not prove a stronger interaction with cytosine
than with guanine as the electronic structure of the individual
bases might be significantly different from that of the Hbonded base pair.[29] As can be seen from the respective mo-

Results and Discussion
We investigated the interaction of three different intercalators
(Scheme 1) with ssDNA and dsDNA by high-level QM calculations of interaction energies between these three molecules
and a nucleobase pair—either guanine–cytosine (G–C) or adenine–thymine (A–T), depending on the selectivity of the inter-

Scheme 1. Chemical structures of the three intercalators used in this study.

calator—and individual purine and pyrimidine nucleobases.
Proflavine and ellipticine are well-known DNA intercalators[17, 18]
used in DNA assays and intercalation of 1-pyrenemethylamine
into DNA can be expected on the basis of data existing for
closely related analogues.[19, 20]
Proflavine features two amino functions suitable for covalent
modification. According to the crystal structure of a proflavine–DNA adduct, these amino groups are located in the
major groove during intercalation and proflavine is inserted
into DNA in a side-on configuration (PDB ID: 3FT6). Proflavine
can be conjugated through formation of an amide or imide.
This route has been previously followed by us for the modification of proflavine with two molecules of biotin.[16] Ellipticine
lacks amino functions and offers no straightforward means of
conjugation. It exhibits two bulky methyl groups, which were
shown to be located in the middle of the base pair during
side-on intercalation.[21] These methyl groups are likely to
induce structural distortions (buckles) into the DNA structure,
although this is difficult to prove due to base-pair planarity restraints typically used during molecular dynamics (MD) refinement of NMR solution structures or low-resolution X-ray crystal
structures. 1-Pyrenemethylamine has one amino function,

&

ChemPlusChem 2016, 81, 1 – 10

www.chempluschem.org

2

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Full Papers

Figure 1. Molecular structures of intercalator–nucleobase adducts, calculated with dispersion-corrected DFT (PBE-D3/TZVPP) in vacuo. Atoms colours: blue,
carbon; grey, hydrogen; dark blue, nitrogen; red, oxygen. For a more detailed analysis of intercalator–DNA interactions it is necessary to understand nucleobase motif preferences. G–C sites were shown to be the main interaction sites for proflavine.[27] For ellipticine, a preference for G–C base pairs is known and
reflected by its positioning in the crystal structure of an ellipticine–ds[CGATCG] adduct.[21] The pyrene moiety has different intercalation behaviour with a preference for A–T-rich sequences.[28] This preference has to be taken into account for the calculation of interaction energies. Base pair and nucleobase interaction
energies of proflavine and ellipticine were therefore calculated using a G–C adduct, whereas interaction energies for 1-pyrenemethylamine are calculated for
the A–T adduct.

The theoretical calculation of interaction energies in dispersion-dominated complexes puts high demands on the
method. To functionally describe the interaction between the
intercalators and base pairs, possibly including polarisation
and charge transfer effects, post-SCF methods including the
exact description of electron correlation are necessary. The first
applicable level is therefore the MP2 theory, which already
scales with N5, where N is the number of basic functions. Due
to the slow convergence behaviour of MP2, large basis sets
and counterpoise correction for the BSSE are necessary. Unfortunately, MP2 tends towards overestimating the interaction
energy of dispersion-bound complexes even (or especially) in
expensive CBS calculations. The oldest way of correcting the
overbinding is by scaling the basis sets. An example is the use
of MP2/6-31G(d0.25)[30] calculations. In principal, CCSD(T) in
a large basis set is the gold standard for the calculation of interaction energies, but can only be applied to the smallest systems. MP2 calculations can, due to similar convergence behaviour, be augmented by a CCSD(T) correction term, representing
the difference between the MP2 and CCSD(T) value in the largest affordable basis set, for example 6-31G(d) or aug-cc-pVDZ.
However, even in a small basis set, CCSD(T) calculations are,
due to the N7 scaling, strongly limited in their application, excluding their use for typical DNA intercalator adducts often involving up to 100 or more atoms. Due to these high demands
in CPU time and memory, great efforts have been put towards
the improvement of MP2 as the “cheapest” method that includes electron correlation and therefore is theoretically capable of correctly describing non-covalent interactions. Starting
from Grimme’s initial SCS–MP2 method,[31] providing an im-

lecular structure in Figure 1, the interaction between proflavine
and guanine is not exclusively one of stacking, but is also
based on H-bonds between an amino group of proflavine and
the carbonyl oxygen of guanine, as well as an amino proton of
guanine and the imino nitrogen atom of proflavine. The interaction between cytosine and proflavine shows stacking on the
one hand and an H-bond interaction between the amino
group of cytosine and the imino nitrogen atom of proflavine
on the other. Ellipticine shows an ideal overlap with the G–C
base pair as its size and shape is similar to that of the G–C
base pair. The longitudinal axis of ellipticine and the base pair
are parallel, and the adduct seems to be locked in this orientation by its bulky methyl groups, located between the Hbonded bases. It has a co-planar orientation in the molecular
adduct with both the G–C base pair and the individual cytosine nucleobase. The geometry of the guanine adduct deviates
from ideal co-planarity due to steric interaction with the
methyl group that takes the place of a ribose moiety. The 1pyrenemethylamine adduct with the A–T base pair exhibits an
ideal co-planar orientation as well as in the adducts with adenine and thymine. An H-bond is formed between the amino
group of the intercalator and the N7 atom of adenine.
1-Pyrenemethylamine is, in its overall length, slightly shorter
than one base pair. However, there is still an overlap between
the nucleobases and the terminal six-membered rings of the
pyrene. The central naphthalene core of the pyrene moiety
overlaps with the H-bond region of the base pair. Pyrene intercalation into DNA might be favoured by the shorter length of
this intercalator, as steric clashes with the DNA backbone are
unlikely.
ChemPlusChem 2016, 81, 1 – 10

www.chempluschem.org

These are not the final page numbers! ÞÞ

3

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

Full Papers
proved description of a wide range of properties including
thermodynamics and kinetics but not non-covalent interactions, different spin-component-scaled methods, specifically
parameterised for the treatment of molecular interactions
(SCS(MI)–MP2) and nucleobase interactions (SCSN–MP2) have
been developed in the groups of Head-Gordon and Hill. Studies found good performance of SCS(MI)–MP2 as well as the
SCSN–MP2 methods.[32, 33] However, as the SCSN method was
parameterised using a set of 10 stacked nucleic acid base
pairs, whereas SCS(MI)–MP2 was parameterised using the less
homogeneous S22 dataset of Hobza,[34] the involvement of organic intercalators led us to the application of the more general SCS(MI)–MP2 method. MP2 results for non-covalent interactions are known to be strongly dependent on the basis set
size, without a systematic improvement with increasing basis
set size. In line with our own experience, the cc-pVTZ basis set
was recommended for the calculation of non-covalent interaction energies with MP2 and with SCS(MI)–MP2, as this combination showed the most balanced description over different
classes of non-covalent interactions (dispersion-controlled, hydrogen-bonded and mixed complexes).[23, 1]
Application of the cc-pVTZ basis set instead of the CBS(T!
Q) extrapolation significantly enlarges the applicability with respect to adduct size and thereby the possibility arises to calculate interaction energies of relatively large intercalator adducts,
with RMS = 0.34 kcal mol1 found for the S22 dataset.[32]
Although these interaction energies characterise the gasphase interaction at a specific point on the potential energy
surface and are not equal to the association energy in water,
they provide valuable information on the proportion of interaction strengths between different intercalators and nucleobases. Additional information about the contribution of entropy in larger systems can only be obtained from MD simulations, which are not possible at the chosen level. A possibility
for the inclusion of entropic effects at an economic QM level,
might be the application of dispersion-corrected semi-empirical methods such as PM6–DH2[35] in MD simulations. However,
these methods still do not reach the accuracy and robustness
of the applied post-SCF approach. For a detailed discussion on
the difference between gas-phase interaction energies and
binding free energies see references [36 and 37].
For proflavine, the calculated interaction energies are significantly higher for its interaction with the G–C base pair than for
its interaction with the individual nucleobases guanine and cytosine. These differences in interaction energy suggest a selectivity for the interaction with dsDNA over ssDNA, which is in
agreement with previous studies.[16] Ellipticine exhibits a very
different behaviour. The interaction towards the G–C base pair
and the individual nucleobases is only slightly different. It can
therefore be expected that ellipticine does not show a high selectivity for dsDNA. 1-Pyrenemethylamine is more similar to
proflavine in its behaviour. The difference between the interaction energies with a base pair or the individual nucleobases
suggests that this intercalator has a stronger interaction with
dsDNA.
DNA intercalation is typically based on the interaction with
two nucleobases, the previous and the subsequent one. The

&

ChemPlusChem 2016, 81, 1 – 10

www.chempluschem.org

model used here for the calculation of interaction energies,
however consists of only one nucleobase or base pair and the
intercalator, which is necessary for the application of a high
theoretical level. It has to be kept in mind, as two nucleobase
interactions contribute to the intercalation, that the intercalated interaction energy should be twice that calculated in the
applied model. Hence, by interacting with dsDNA, it is twice
more likely for the intercalator to follow a G–C or A–T preference than for ssDNA. Therefore it might be necessary to use
the averaged interaction energies towards purine and pyrimidine bases to explain the DNA interaction. For proflavine, the
calculated interaction energy towards the G–C base pair is
23.72 kcal mol1. The base pair stacking energy derived from
high-level CBS(T) calculations varies from 13.0 to 18.4 kcal
mol1, depending on the exact base pair step.[38] However, to
accommodate the intercalator, an intercalation site has to be
formed. This formation not only requires base pair stacking to
be overcome, but also energy penalties that result, for example, from stretching of the backbone. Including these factors,
a theoretical study on the energetics of the formation of intercalation sites suggested an energy of approximately 32 kcal
mol1, necessary for the formation of a single intercalation
site.[39] Taking into account the intercalator interaction with
two base pairs instead of one in our model, the resulting interaction energies are 47.22 kcal mol1 for proflavine,
35.22 kcal mol1 for ellipticine and 39.22 kcal mol1 for 1pyrenemethylamine. On the basis of these values, intercalation
into dsDNA should be energetically favoured for all three molecules.
To the best of our knowledge, no studies have been published on the energetics of the formation of intercalation sites
in ssDNA. We therefore assume this energy to be half of the
value for the formation of intercalation sites in dsDNA leading
to a value of approximately 16 kcal mol1. For the stacking interaction of adenine and thymine, a theoretical value of
12.23 kcal mol1 is available from highly accurate CBS(T) calculations.[34] This value supports the preceding assumption.
The average interaction energies of the intercalators towards
individual bases are 14.62 kcal mol1 for proflavine,
17.23 kcal mol1 for ellipticine and 15.62 kcal mol1 for 1pyrenemethylamine. Taking into account energetic costs of
16 kcal mol1 for the formation of the intercalation site, we
expect unfavourable interactions with ssDNA for ellipticine but
not for proflavine. The interaction energy of 1-pyrenemethylamine is rather close to the assumed minimum required energy
for formation of a single intercalation site with ssDNA, therefore some interaction appears to be possible.
To investigate the effect of covalent modification on the
electronic structure of 1-pyrenemethylamine and proflavine,
the molecular structures of the corresponding mono- and bisamides were calculated using hybrid density functional theory
(B3LYP/TZVPP). A full natural bond orbital (NBO) analysis, including natural population analysis (NPA), was performed to
obtain atomic charges, which were mapped onto the electronic density surface.
In analysing the electrostatic potential of 1-pyrenemethylamine, no strong polarisation of the molecule was observed
4

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Full Papers
formed amide. This is favourable, as it is hard to predict if induced changes that are due to covalent modification of an intercalator, are strengthening or weakening the intercalative interaction. This information is only accessible after synthesis of
the modified intercalator.

(Figure 2). The highest electron density was found at the aromatic system. Even during amidation, no significant changes in
the electrostatic potential of the pyrene moiety were detected.
The aromatic system experiences a negligible reduction of
electron density, whereas a new source of electron density is
introduced by the oxygen atom of the amide carbonyl function in the linker.

Electrochemical impedance measurements
The goal of this part of the study was to experimentally investigate the interaction of intercalators with ss- and dsDNA by
means of EIS and to compare the results with the prediction
from the QM SCS(MI)–MP2/cc-pVTZ calculations as shown
before. The investigation of DNA-modified gold electrode surfaces by means of EIS is based on the interaction between
a negatively charged free-diffusing redox mediator, typically
[Fe(CN)6]3/4, and the DNA-modified electrode in a solution of
a moderate ionic strength.[40]
In order to unambiguously investigate DNA–intercalator interactions experimentally and to compare different intercalators, it is of great importance to use highly reproducible and
reliable Au electrode surfaces. Therefore, we followed each
step of the formation of the DNA-modified surface and its interaction with the intercalator in question using EIS. Initially
we immobilised ssDNA on a gold surface by chemisorption of
a dithiophosphoramidite(DTPA) [(SS)3 form] linker, which forms
three disulfide bonds, assuring highly stable chemisorption. EIS
was performed in an electrolyte solution of comparatively low
ionic strength (5 mm K3[Fe(CN)6] and 5 mm K4[Fe(CN)6] in
10 mm phosphate buffer, PB, 20 mm K2SO4, pH 7.4) where the
negative charge of the DNA strands is not completely counter
balanced, thus substantially modulating the electron transfer
rate of [Fe(CN)6]3/4. This causes an increase of the chargetransfer resistance (Rct) in EIS. Because of the persistence
length of ssDNA it behaves as a flexible coil[41] and relaxes on
the electrode surface, leading to additional steric restriction on
surface accessibility. This step is followed by the immobilisation
of mercaptohexanol (MCH) as a filler molecule that forces the
ssDNA to obtain a more upright orientation that is more favourable for hybridisation with the target DNA. Loosely bound
DNA is removed during this step and the arising gaps are filled
by MCH, which assists the formation of a homogenous self-assembled monolayer.[42] This process leads to easier access of
the negatively charged redox couple to the electrode surface,
as the concentration of negative charges in close proximity to
the surface is reduced. In EIS this is represented by a decrease
in Rct (data not shown). In order to investigate ssDNA–intercalator interactions the ssDNA–MCH electrodes were incubated
with intercalator solution and the obtained surface was investigated by means of EIS. Alternatively, the electrode modification
was continued by hybridisation with the complementary
target DNA strand. Upon hybridisation with the complementary DNA strand, different effects influence the access of redox
mediators to the surface. Namely, the increase in steric hindrance by the additional DNA strand decreases the access of
the redox mediator and contributes to an increase in Rct. On
the other hand, the structure of the formed double helix is
more rigid in comparison to that of the single strand, leading

Figure 2. Electrostatic potential maps of 1-pyrenemethylamine (top) and the
corresponding pyrenemethylethylamide (bottom). Blue regions refer to high
electrostatic potential (low electron density), whereas red areas refer to
a low electrostatic potential (high electron density).

The electrostatic potential of proflavine (Figure 3) shows
a much stronger polarisation of the molecule with the highest
electron density located on the aromatic nitrogen atom, followed by carbon atoms of the aromatic system. Upon bisamidation the electron density significantly decreases on the aromatic nitrogen atom and the aromatic system in general, with
the highest electron density then being located on the oxygen
atoms of the amides carbonyl function. These changes of the
electronic structure of proflavine most likely affect the p–p
stacking being the basis for the intercalative interaction with
DNA. For 1-pyrenemethylamine, the theoretical results show
that amidation has only very limited effect on the aromatic
system of the intercalator, which can be explained by the
methylene group separating the aromatic system and the

Figure 3. Electrostatic potential maps of proflavine (top) and the corresponding proflavinebismethylethylamide (bottom). Blue regions refer to
high electrostatic potential (low electron density), whereas red areas refer to
a low electrostatic potential (high electron density).

ChemPlusChem 2016, 81, 1 – 10

www.chempluschem.org

These are not the final page numbers! ÞÞ

5

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

Full Papers
to a more upright orientation of strands that facilitates the
access of the redox mediator. Finally, the concentration of negative charges on the surface, which are not entirely balanced
by counter ions, is further increased by the immobilisation of
the complementary DNA strand, causing an increase in the Rct
value. These factors modulate the access of the negatively
charged redox species to the surface and depending on the
amount of dsDNA, a decrease or an increase of the electron
transfer rate prevails. In this assay, we observed a decrease in
the electron transfer rate, which is associated with an increase
in Rct.
The effect of intercalation on the dsDNA and the electrochemical properties of the interface depends on the properties
of the intercalator and the product of intercalation—the
changes imposed on the DNA after intercalation. Generally, intercalators increase the inter-base separation that is accompanied by a lengthening of the dsDNA. Typically, the distance between an intercalator and a base pair is in the range of that
between two adjacent base pairs, which is around 3.4  in BDNA.[43] Depending on the exact structure of the intercalator,
the helix thickness can be modified.
Intercalators with bulky functional groups, or those bearing
large sugar or peptide residues can further increase the thickness of the DNA, whereas sterically less demanding molecules
can, due to the lengthening of the DNA, lead to a narrowing
of the helix. For the molecules used in this study, we expect
a lengthening of the DNA accomplished by unwinding and reduction of the diameter of the helix. This should facilitate
access of the redox couple to the surface, increase the electron
transfer rate and therefore reduce Rct. If the intercalator is positively charged, the interaction can lead to a partial compensation of the negative charge of the DNA. To correctly analyse
the interaction between intercalators and ssDNA versus
dsDNA, separate measurements have to be performed, as a potential interaction between the intercalator and ssDNA might
influence DNA hybridisation.
The Nyquist plots in Figure 4 a show a comparison of EIS for
a ssDNA–MCH-modified electrode before and after incubation
with proflavine. Evidently, there is no change in the Rct value,
suggesting that proflavine does not interact with ssDNA. Figure 4 b shows data plotted for dsDNA before and after incubation with proflavine. The Rct value significantly decreases,
which is in agreement with a favourable interaction between
proflavine and dsDNA upon intercalation, leading to the
above-discussed opening of the electrode surface. Proflavine
behaves as a dsDNA-selective intercalator.[16] Incubation of
a ssDNA/MCH-modified electrode surface in a solution containing 1-pyrenemethylamine gave a small increase in the Rct, (Figure 5 a). By contrast, a clear decrease of Rct can be observed
after incubation of a dsDNA-modified electrode in a 1-pyrenemethylamine-containing solution (Figure 5 b). This decrease,
however, is less pronounced than the one found in the case of
proflavine intercalation. 1-Pyrenemethylamine shows a preference for dsDNA over ssDNA, however, the selectivity is smaller
than for proflavine.
Upon incubation of ssDNA–MCH- and dsDNA-modified electrodes in an ellipticine-containing solution, a significant

&

ChemPlusChem 2016, 81, 1 – 10

www.chempluschem.org

Figure 4. Nyquist plots of a) ssDNA and b) dsDNA before and after incubation in a solution of proflavine (200 mm, 15 min). “After control” refers to
a ssDNA–MCH-modified electrode after incubation with proflavine. Immobilisation of ssDNA was performed by incubation in 10 mm PB containing
450 mm K2SO4 and 1 mm DNA for 2 h. Passivation of the surface was performed overnight in 10 mm MCH solution. Hybridisation was performed by
incubation in 1 mm DNA with 10 mm PB and 450 mm K2SO4 for 1 h. EIS was
performed in 10 mm PB and 20 mm K2SO4 solution containing equimolar
amounts of K4[Fe(CN)6] and K4[Fe(CN)6] (5 mm) at a dc potential of + 220 mV
versus Ag/AgCl/3 m KCl with an ac perturbation amplitude of 10 mVpp and
a frequency range from 30 kHz to 10 mHz.

Figure 5. Nyquist plots recorded for a) ssDNA and b) dsDNA before and after
incubation in a solution of 1-pyrenemethylamine (200 mm, 15 min). “After
control” refers to a ssDNA–MCH-modified electrode after incubation with 1pyrenemethylamine. Immobilisation of ssDNA, passivation, hybridisation and
EIS were performed as described in the legend of Figure 4..

6

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Full Papers
17.2 kcal mol1 for intercalation with dsDNA and ssDNA, respectively, are above the minimum energy required.

change in the Rct value was measured. Whereas Rct increased
after interaction with ssDNA (Figure 6 a), it significantly decreased upon ellipticine intercalation in dsDNA (Figure 6 b). It
is clear that ellipticine strongly interacts with both ssDNA and
dsDNA, hence, it cannot be used to selectively detect the formation of dsDNA using EIS.

Conclusion
This study underlines the possibility to predict interaction energies of intercalators with ssDNA and dsDNA by using QM calculations. Although interaction energies cannot be directly related to experimentally accessible free energies, they provide
quantitative information about the attraction between intercalator and nucleobase in the calculated situation. Due to the difficulties in theoretically describing non-covalent interactions,
a well-applicable, albeit high, theoretical level was chosen with
counterpoise-corrected SCS(MI)–MP2/cc-pVTZ//PBE-D3/TZVPP.
This theoretical level enables application to systems including
larger intercalators than proflavine.
The calculated interaction energies suggest an interaction of
proflavine, ellipticine and 1-pyrenemethylamine with dsDNA,
as they are larger than the energy of around 32 kcal mol1 necessary for the formation of an intercalation site. This coincides
with the experimental EIS results, which shows significant
changes in the Rct values for all intercalators upon incubation
with dsDNA-modified electrode surfaces. For proflavine and 1pyrenemethylamine, the QM interaction energies with a base
pair are higher than those involving the individual nucleobases, with a larger difference observed for proflavine. Also from
the experimental results, proflavine was shown to be a dsDNAselective intercalator, whereas 1-pyrenemethylamine showed
a minor effect also upon incubation of a ssDNA-modified electrode surface.
For ellipticine, no significant differences in the interaction
energies towards base pairs or individual bases were detected.
This suggests that intercalation with dsDNA is not preferred
over that with ssDNA and hence no selectivity for the detection of DNA hybridisation is expected. QM calculations showed
that the influence of conjugation on the electronic structure of
the intercalator is significantly lower for 1-pyrenemethylamine
as compared to proflavine. This is, on the one hand, a result of
the bis-conjugation of proflavine, and on the other hand, of
the better separation between the aromatic system of 1-pyrenemethylamine and the site of modification by the methylene
group. As a general conclusion it can be added that QM calculations on intercalator–DNA adducts are feasible with a speed
and accuracy that allows their seamless integration into the
process of design, synthesis and analysis of DNA intercalators.
This would prevent the synthesis of non-selective intercalators
and at least decrease the need for experimental investigations
and the tedious preparation of sensitive DNA-modified surfaces for selecting optimal intercalators for the unequivocal detection of DNA hybridisation.

Figure 6. Nyquist plots recorded for a) ssDNA and b) dsDNA before and after
incubation in a solution of ellipticine (200 mm, 15 min). “After control” refers
to a ssDNA–MCH-modified electrode after incubation with ellipticine. Immobilisation of ssDNA, passivation, hybridisation and EIS were performed as described in the legend of Figure 4.

From the EIS results, it is apparent that both proflavine and
1-pyrenemethylamine intercalate with dsDNA, with proflavine
participating in slightly stronger interactions. However, whereas proflavine does not interact with ssDNA, 1-pyrenemethylamine exhibits a small effect on the Rct value upon incubation
with a ssDNA-modified surface. These results are in perfect
agreement with the QM SCS(MI)–MP2/cc-pVTZ calculations.
Based on these calculations, both proflavine and 1-pyrenemethylamine have interaction energies well above the theoretical
value required for a single intercalation site of 32 kcal mol1,
with values for proflavine and 1-pyrenemethylamine of 47.2
and 39.2 kcal mol1, respectively. Furthermore, according to
the assumption of a minimum energy needed for the formation of an intercalation site in ssDNA of 16 kcal mol1, proflavine is, with 14.6 kcal mol1 average nucleobase interaction
energy, clearly below this minimum energy; 1-pyrenemethylamine is, with 15.6 kcal mol1, still below but rather close to
this energy. This might explain our observation of a small interaction of 1-pyrenemethylamine with ssDNA and the absence
of any interaction for proflavine. Additionally, the experimental
results also support the calculations for the case of ellipticine,
for which an obvious interaction was observed with both the
ss- and the dsDNA. The interaction energies of 35.2 and
ChemPlusChem 2016, 81, 1 – 10

www.chempluschem.org

These are not the final page numbers! ÞÞ

Experimental Section
Computational details
All QM calculations were performed by using Gaussian 09
(Rev. A.02)[44] or Turbomole (Ver. 6.3.1)[45] Graphical representations

7

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

Full Papers
were produced
(Ver. 0.99)[47]

with

Gaussview

(Ver. 5.0.9)[46]

and

PyMOL

Electrode preparation
Polycrystalline gold electrodes were purchased from CH Instruments (Austin, TX) and polished with diamond slurries ranging
from 3 to 0.1 mm on polishing cloths. Before each experiment, the
electrodes were cleaned electrochemically by cyclic voltammetry in
0.5 m H2SO4 (+ 0.6 to + 2.2 V vs. a Pb/PbF2 reference electrode,[54]
0.1 V s1) until stable voltammograms were obtained.

Molecular structures of the intercalators
Molecular structures shown in Table 1 were calculated using DFT
with the B3LYP[48] hybrid functional as implemented in Gaussian 09
in the def2-TZVPP basis set[24] in vacuo. The def2-TZVPP basis set
was obtained from the EMSL website Basis Set Exchange https://
bse.pnl.gov/bse/portal.[49]

Immobilisation of ssDNA
ssDNA was immobilised by immersion of an electrode into a solution of ssDNA (1 mm in 10 mm PB, 450 mm K2SO4, pH 7.4) for 2 h at
37 8C in a shaking incubator (Thermomixer, HLC BioTech, Bovenden, Germany). In order to remove the non-specifically bound DNA
from the electrode surface, electrodes were subsequently rinsed
with buffer (10 mm PB, 450 mm K2SO4).

Table 1. Molecular interaction energies [kcal mol1] between a base pair
versus the individual nucleobases and different intercalators calculated at
the SCS(MI)–MP2/cc-pVTZ level including counterpoise correction. Proflavine and ellipticine interact with G–C, 1-pyrenemethylamine interacts
with A–T.
DNA

Proflavine

Ellipticine

1-Pyrenemethylamine

Passivation with 1-mercaptohexanol

Base pair
Purine
Pyrimidine
Average (Pu/Py)

23.72
18.64
10.59
14.62

17.61
16.87
17.59
17.23

19.51
15.34
15.90
15.62

ssDNA-modified electrodes were further modified with MCH by
overnight incubation (at 37 8C in a Thermomixer) to allow for passivation of the surface. The electrode was subsequently rinsed with
ethanol and water.

Hybridisation of DNA

Interaction geometries and energies

DNA was hybridised by immersion of the electrode into the target
strand solution (1 mm in 10 mm PB, 450 mm K2SO4, pH 7.4) for 1 h
at 37 8C in a Thermomixer. The electrode was subsequently rinsed
with buffer (10 mm PB, 450 mm K2SO4).

The geometries of intercalator–DNA adducts were optimised in vacuo, using DFT with the PBE[50] functional, which was corrected for
dispersive interactions by a third version of Grimme’s atom-pairwise dispersion correction (D3).[12] Interaction energies were subsequently calculated with the spin-component-scaled second-order
Møller Plesset theory, optimised for molecular interactions,
SCS(MI)–MP2,[32] using the resolution-of-the-identity approach and
the frozen-core approximation. For all calculations, the cc-pVTZ
basis set was used and the results were corrected for the BSSE applying the counterpoise method according to Boys and Bernardi.[51]
The scaling factors for SCS(MI)–MP2 with the used cc-pVTZ basis
set were c*os = 0.17 and c*ss = 1.75.

Intercalation
Intercalation was performed by immersion of the electrode into
a solution of the intercalator (200 mm in 10 mm PB, 20 mm K2SO4)
for 15 min. The electrode was subsequently rinsed with buffer
(10 mm PB, 450 mm K2SO4).

Electrochemical impedance spectroscopy
All electrochemical measurements were made in an electrochemical cell consisting of the modified gold working electrode (diameter 2 mm), a platinum auxiliary electrode and an in-house-made
Pb/PbF2 reference electrode (in 5 m KF). The potential of the Pb/
PbF2 electrode was 600 mV versus Ag/AgCl/3 m KCl. This reference electrode was used in order to avoid the presence of chlorides in the solution. EIS was used for the characterisation of each
step of the surface modification. Measurements were performed in
5 mm K3[Fe(CN)6] and 5 mm K4[Fe(CN)6] prepared in 10 mm PB containing 20 mm K2SO4 (pH 7.4). The equilibrium potential of the
redox couple (+ 830 mV vs. Pb/PbF2) was applied as the dc potential that was superimposed with an ac perturbation of 5 mVpp amplitude. A modulation frequency between 30 kHz and 10 mHz was
used. Measurements were performed at room temperature.

Electrostatic potential of modified intercalators
The molecular structures of intercalators and their adducts were
calculated using DFT with the B3LYP functional as implemented in
Gaussian 09, together with the def2-TZVPP basis set. After geometry optimisation, a single-point calculation with full NBO analysis[52]
including NPA[53] was performed at the same level. The electrostatic
potential (0.05 to 0.05) was then mapped on the electronic density surface (isoval = 0.0004) using Gaussview (Ver. 5.0.9).

Materials and reagents
DNA oligonucleotides were purchased from FRIZ Biochem (Neuried, Germany). Intercalators were purchased from Sigma–Aldrich.
The surface-immobilised DNA probe was 5’-TGCGGATAACACAGTCACCTTTTTTTT-(dithiophosphoramidite)3-OH-3’; the target DNA
had the sequence 5’-AGGTGACTGTGTTATCCGCA-3’. All solutions
were prepared in ultra-pure water (Siemens Ultra Clear water purification system). DMSO and buffer salts for PB were purchased
from Sigma–Aldrich.

&

ChemPlusChem 2016, 81, 1 – 10

www.chempluschem.org

Acknowledgements
Financial support by the BMBF within the framework of the project InnoEMat “eDx” (FKZ:13XP5007C) and the Center for Electrochemical Sciences (CES) funded by the European Commission
8

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Full Papers
and the state of North Rhine-Westphalia (NRW) within the framework of the HighTech.NRW program is gratefully acknowledged.
Tarik Nabil Abdulazim is acknowledged for his contribution
during a laboratory practical.

[31]
[32]
[33]
[34]
[35]

Keywords: DNA intercalation · electrochemical impedance
spectroscopy · molecular modelling · post-Hartree–Fock ·
quantum mechanical interaction energy

[36]
[37]
[38]

[1] P. Hobza, Acc. Chem. Res. 2012, 45, 663 – 672.
[2] a) D. Řeha, M. Kabelč, F. Ryjček, J. Šponer, J. E. Šponer, M. Elstner, S.
Suhai, P. Hobza, J. Am. Chem. Soc. 2002, 124, 3366 – 3376; b) D. A. Bondarev, W. J. Skawinski, C. A. Venanzi, J. Phys. Chem. B 2000, 104, 815 –
822.
[3] I.-C. Lin, U. Rothlisberger, Phys. Chem. Chem. Phys. 2008, 10, 2730 – 2734.
[4] P. Hobza, J. Šponer, T. Reschel, J. Comput. Chem. 1995, 16, 1315 – 1325.
[5] Y. Zhao, D. G. Truhlar, Phys. Chem. Chem. Phys. 2005, 7, 2701 – 2705.
[6] J. Černý, P. Hobza, Phys. Chem. Chem. Phys. 2005, 7, 1624 – 1626.
[7] E. G. Hohenstein, S. T. Chill, C. D. Sherrill, J. Chem. Theory Comput. 2008,
4, 1996 – 2000.
[8] M. P. Waller, A. Robertazzi, J. A. Platts, D. E. Hibbs, P. A. Williams, J.
Comput. Chem. 2006, 27, 491 – 504.
[9] M. Elstner, P. Hobza, T. Frauenheim, S. Suhai, E. Kaxiras, J. Chem. Phys.
2001, 114, 5149.
[10] L. Zhechkov, T. Heine, S. Patchkovskii, G. Seifert, H. A. Duarte, J. Chem.
Theory Comput. 2005, 1, 841 – 847.
[11] S. O. Nilsson Lill, J. Phys. Chem. A 2009, 113, 10321 – 10326.
[12] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132,
154104.
[13] K. E. Riley, J. Vondršek, P. Hobza, Phys. Chem. Chem. Phys. 2007, 9,
5555.
[14] E. M. Regan, A. J. Hallett, L. C. Wong, I. Q. Saeed, E. E. Langdon-Jones,
N. J. Buurma, S. J. Pope, P. Estrela, Electrochim. Acta 2014, 128, 10 – 15.
[15] M. Ge˛bala, L. Stoica, D. Guschin, L. Stratmann, G. Hartwich, W. Schuhmann, Electrochem. Commun. 2010, 12, 684 – 688.
[16] M. Gebala, L. Stoica, S. Neugebauer, W. Schuhmann, Electroanalysis
2009, 21, 325 – 331.
[17] H. M. Berman, P. R. Young, Annu. Rev. Biophys. Bioeng. 1981, 10, 87 – 114.
[18] K. W. Kohn, M. J. Waring, D. Glaubiger, C. A. Friedman, Cancer Res. 1975,
35, 71 – 76.
[19] K. W. Bair, R. L. Tuttle, V. C. Knick, M. Cory, D. D. McKee, J. Med. Chem.
1990, 33, 2385 – 2393.
[20] E. Grueso, R. Prado-Gotor, Chem. Phys. 2010, 373, 186 – 192.
[21] A. Canals, M. Purciolas, J. Aymam, M. Coll, Acta Crystallogr. Sect. D
2005, 61, 1009 – 1012.
[22] K. E. Riley, J. A. Platts, J. Řezč, P. Hobza, J. G. Hill, J. Phys. Chem. A 2012,
116, 4159 – 4169.
[23] K. E. Riley, M. Pitoňk, J. Černý, P. Hobza, J. Chem. Theory Comput. 2010,
6, 66 – 80.
[24] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297 – 3305.
[25] J. Ireta, J. Neugebauer, M. Scheffler, J. Phys. Chem. A 2004, 108, 5692 –
5698.
[26] S. Grimme, J. Comput. Chem. 2004, 25, 1463 – 1473.
[27] N. Yoshida, Y. Kiyota, F. Hirata, J. Mol. Liq. 2011, 159, 83 – 92.
[28] J. Ren, J. B. Chaires, Biochemistry 1999, 38, 16067 – 16075.
[29] A. A. Golubeva, A. I. Krylov, Phys. Chem. Chem. Phys. 2009, 11, 1303 –
1311.
[30] a) J. Šponer, J. Leszczyński, P. Hobza, J. Phys. Chem. 1996, 100, 5590 –
5596; b) P. Hobza, A. Mehlhorn, P. Črsky, R. Zahradnk, J. Mol. Struct.
1986, 138, 387 – 399; c) P. Hobza, J. Šponer, M. Polšek, J. Am. Chem.
Soc. 1995, 117, 792 – 798.

ChemPlusChem 2016, 81, 1 – 10

www.chempluschem.org

These are not the final page numbers! ÞÞ

[39]
[40]
[41]
[42]
[43]
[44]

[45]

[46]
[47]
[48]
[49]

[50]

[51]
[52]
[53]
[54]

S. Grimme, J. Chem. Phys. 2003, 118, 9095.
R. A. Distasio, Jr., M. Head-Gordon, Mol. Phys. 2007, 105, 1073 – 1083.
J. G. Hill, J. A. Platts, J. Chem. Theory Comput. 2007, 3, 80 – 85.
P. Jurečka, J. Šponer, J. Černý, P. Hobza, Phys. Chem. Chem. Phys. 2006,
8, 1985.
M. Korth, M. Pitoňk, J. Řezč, P. Hobza, J. Chem. Theory Comput. 2010,
6, 344 – 352.
J. Šponer, J. E. Šponer, A. Mldek, P. Jurečka, P. Banš, M. Otyepka, Biopolymers 2013, 99, 978 – 988.
J. Šponer, J. E. Šponer, A. Mldek, P. Banš, P. Jurečka, M. Otyepka, Methods 2013, 64, 3 – 11.
J. Šponer, P. Jurečka, I. Marchan, F. J. Luque, M. Orozco, P. Hobza, Chem.
Eur. J. 2006, 12, 2854 – 2865.
M. Trieb, Nucleic Acids Res. 2004, 32, 4696 – 4703.
M. Gebala, W. Schuhmann, Phys. Chem. Chem. Phys. 2012, 14, 14933 –
14942.
R. R. Netz, D. Andelman, Phys. Rep. 2003, 380, 1 – 95.
D. Jambrec, M. Gebala, F. La Mantia, W. Schuhmann, Angew. Chem. Int.
Ed. 2015, 54, 15064 – 15068; Angew. Chem. 2015, 127, 15278 – 15283.
M. Hogan, N. Dattagupta, D. M. Crothers, Biochemistry 1979, 18, 280 –
288.
Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
a) TURBOMOLE V6.3.1 2012, University of Karlsruhe and Forschungszentrum Karlsruhe GmbH, 1989 – 2007, TURBOMOLE GmbH, since 2007,
available from http://www.turbomole.com; b) R. Ahlrichs, M. Bar, M.
Haser, H. Horn, C. Kolmel, Chem. Phys. Lett. 1989, 162, 165 – 169.
GaussView, Version 5, R. Dennington, T. Keith, J. Millam, Semichem Inc.,
Shawnee Mission, KS, 2009.
W. DeLano, The PyMOL Molecular Graphics System, DeLano Scientific,
Palo Alto, CA, 2002.
A. D. Becke, J. Chem. Phys. 1993, 98, 5648 – 5652.
a) D. Feller, J. Comput. Chem. 1996, 17, 1571 – 1586; b) K. L. Schuchardt,
B. T. Didier, T. Elsethagen, L. S. Sun, V. Gurumoorthi, J. Chase, J. Li, T. L.
Windus, J. Chem. Inf. Model. 2007, 47, 1045 – 1052.
a) J. P. Perdew, Y. Wang, Phys. Rev. B 1992, 45, 13244 – 13249; b) J. M.
Tao, J. P. Perdew, V. N. Staroverov, G. E. Scuseria, Phys. Rev. Lett. 2003, 91,
146401.
S. F. Boys, F. Bernardi, Mol. Phys. 1970, 19, 553.
A. E. Reed, F. Weinhold, J. Chem. Phys. 1983, 78, 4066 – 4073.
A. E. Reed, R. B. Weinstock, F. Weinhold, J. Chem. Phys. 1985, 83, 735 –
746.
M. Pasta, A. Battistel, F. La Mantia, Electrochem. Commun. 2012, 20,
145 – 148.

Manuscript received: April 5, 2016
Revised: May 18, 2016
Accepted Article published: May 19, 2016
Final Article published: && &&, 0000

9

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

FULL PAPERS
D. Jambrec, R. Haddad, A. Lauks,
M. Gebala,* W. Schuhmann,*
M. Kokoschka*
&& – &&
DNA Intercalators for Detection of
DNA Hybridisation: SCS(MI)–MP2
Calculations and Electrochemical
Impedance Spectroscopy

&

ChemPlusChem 2016, 81, 1 – 10

A model sandwich: The prediction of
interaction energies using quantum mechanical SCS(MI)–MP2/cc-pVTZ calculations is presented. These calculations
can be used to predict the strength of
interactions between intercalators and
DNA, allowing the evaluation of the

probability of intercalation and the selectivity for dsDNA. Electrochemical impedance spectroscopy was used for
comparison and the experimental results are in good agreement with the
calculated predictions.

10

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chempluschem.org

ÝÝ These are not the final page numbers!