Metal-mediated base pairing in DNA involving the artificial nucleobase imidazole-4-carboxylate

Article abstract of DOI:10.1016/j.jinorgbio.2018.10.013

The use of imidazole-4-carboxylate (X) as an artificial nucleobase in metal-mediated base pairing is reported. Towards this end, the corresponding deoxyribonucleoside was synthesized and structurally characterized as its sodium salt (sodium 1,2-dideoxy-1-(4-carboxyimidazol-1-yl)-D-ribofuranose). The deoxyribonucleoside was incorporated into different DNA duplexes (parallel-stranded and antiparallel-stranded), and their Cu(II)- and Ag(I)-binding behavior was investigated. It was shown that both X–Cu(II)–X and X–Ag(I)–X base pairs can be formed, with the former being more stabilizing than the latter. The formation of an X–Cu(II)–X base pair is accompanied by an increase in the duplex melting temperature of approximately 20 °C for antiparallel-stranded duplexes and of 12 °C for the parallel-stranded duplex under investigation. Imidazole-4-carboxylate represents the first imidazole-based nucleoside for Cu(II)-mediated base pairing. Moreover, it is the smallest nucleoside known to form stable Cu(II)-mediated base pairs. Structures of the X–Cu(II)–X and X–Ag(I)–X base pairs are proposed, too, based on molecular structures obtained using the model nucleobase 1-benzyl-1H-imidazole-4-carboxylate.

Full text of DOI:10.1016/j.jinorgbio.2018.10.013

Accepted Manuscript
Metal-mediated base pairing in DNA involving the artificial
nucleobase imidazole-4-carboxylate
Nikolas Sandmann, Denise Defayay, Alexander Hepp, Jens
Müller
PII:
S0162-0134(18)30509-9
DOI:
Reference:
https://doi.org/10.1016/j.jinorgbio.2018.10.013
JIB 10587
To appear in:
Journal of Inorganic Biochemistry
Received date:
Revised date:
Accepted date:
31 August 2018
18 October 2018
29 October 2018
Please cite this article as: Nikolas Sandmann, Denise Defayay, Alexander Hepp, Jens
Müller , Metal-mediated base pairing in DNA involving the artificial nucleobase
imidazole-4-carboxylate. Jib (2018), https://doi.org/10.1016/j.jinorgbio.2018.10.013
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

ACCEPTED MANUSCRIPT
Metal-mediated base pairing in DNA involving the artificial nucleobase imidazole-4-
carboxylate
Nikolas Sandmann, Denise Defayay, Alexander Hepp, Jens Müller*
Westfälische Wilhelms-Universität Münster, Institut für Anorganische und Analytische
Chemie, Corrensstraße 28/30, 48149 Münster, Germany.
*Corresponding author: E-mail: mueller.j@uni-muenster.de; Fax: +49 251 83 36007; Tel: +49
251 83 36006.
1

ACCEPTED MANUSCRIPT
Abstract
The use of imidazole-4-carboxylate (X) as an artificial nucleobase in metal-mediated base pairing is
reported. Towards this end, the corresponding deoxyribonucleoside was synthesized and structurally
characterized as its sodium salt (sodium 1,2-dideoxy-1-(4-carboxyimidazol-1-yl)-D-ribofuranose). The
deoxyribonucleoside was incorporated into different DNA duplexes (parallel-stranded and
antiparallel-stranded), and their Cu(II)- and Ag(I)-binding behavior was investigated. It was shown that
both X–Cu(II)–X and X–Ag(I)–X base pairs can be formed, with the former being more stabilizing than
the latter. The formation of an X–Cu(II)–X base pair is accompanied by an increase in the duplex
melting temperature of approximately 20 °C for antiparallel-stranded duplexes and of 12 °C for the
parallel-stranded duplex under investigation. Imidazole-4-carboxylate represents the first imidazole-
based nucleoside for Cu(II)-mediated base pairing. Moreover, it is the smallest nucleoside known to
form stable Cu(II)-mediated base pairs. Structures of the X–Cu(II)–X and X–Ag(I)–X base pairs are
proposed, too, based on molecular structures obtained using the model nucleobase 1-benzyl-1H-
imidazole-4-carboxylate.
Keywords
DNA, imidazole, metal-mediated base pair, silver, copper
2

ACCEPTED MANUSCRIPT
1. Introduction
The formation of stable nucleic acid duplexes is inevitably dependent on the presence of cations to
stabilize the anionic sugar phosphate backbone via electrostatic interactions. In biological systems,
alkaline and earth alkaline metal ions regularly serve this purpose [1]. In addition, nucleobases can
also bind transition metal ions in a site-specific manner. One of the best-established systems in this
respect is the T–Hg(II)–T base pair (T, thymine), in which a mercury(II) ions cross-links two
deprotonated thymine residues from opposite strands [2-4]. Similarly, a C:C mispair (C, cytosine) is
known to accommodate one Ag(I) ion to form a C–Ag(I)–C base pair [5]. In fact, DNA duplexes can be
created that comprise Ag(I)-mediated base pairs only [6-10]. The site-specific incorporation of metal
ions into nucleic acid structures has opened an entire new field of applied bioinorganic chemistry [11].
Such conjugates of self-assembling nucleic acids and functionality-bearing metal ions have already
been applied in numerous contexts, including the detection of single-nucleotide polymorphisms [12-
14], the modification of the charge-transfer properties of DNA [15-20], the generation of DNA-
templated metal nanoclusters [21], and more [22]. Interestingly, several reports also indicate that
metal-mediated base pairs can be introduced into nucleic acids enzymatically [23-27].
The scope of metal-mediated base pairing has been significantly extended by the introduction of
artificial metal-binding nucleosides into the nucleic acids [28, 29]. In addition to Ag(I) and Hg(II) ions
mentioned above, this also allowed the use of Cu(I) [30], Cu(II) [31-34], Ni(II) [35, 36], Zn(II) [37], Pd(II)
[38], Mn(III) [39], and others, in metal-mediated base pairs. Even unusual DNA topologies such as
three-way junctions, quadruplexes, and other higher-order structures can be stabilized by suitably
designed metal-binding nucleic acid components [40-42]. Yet most artificial metal-mediated base
pairs appear to include Ag(I) [9, 10, 43-46]. One of the best investigated Ag(I)-mediated base pairs
involves imidazole as artificial nucleobase [47-50]. In addition, a variety of synthetic nucleobases for
metal-mediated base pairing have been derived from imidazole, e.g. by attaching alkyl groups to
3

ACCEPTED MANUSCRIPT
modify the basicity of the ligand [51] or by adding a second binding site and hence generating a
bidentate ligand [52].
In this paper, we report for the first time the development of an imidazole-based artificial nucleoside
for Cu(II)-mediated base pairing. An inspection of previously reported Cu(II)-mediated base pairs has
shown that the most stable ones comprise two identical nucleobases, each one of which contains one
neutral and one anionic donor atom. The neutral atom can be oxygen [31] or nitrogen [16, 39, 53, 54],
whereas the anionic one persistently is oxygen. In line with these considerations and to ensure an
appropriate bite angle of the intended imidazole-based bidentate ligand, we chose to investigate
imidazole-4-carboxylate as a ligand for metal-mediated base pairing (Chart 1a).
4

ACCEPTED MANUSCRIPT
2. Experimental
1-Benzyl-4-iodoimidazole and Hoffer’s chloro sugar were synthesized according to literature
procedures [55, 56]. Single-crystal X-ray diffraction data were collected with graphite-
monochromated Mo Kα radiation (λ = 0.71073 Å) on a Bruker D8 Venture diffractometer. The
structures were solved by direct methods and were refined by full-matrix, least squares on F2 by using
the SHELXTL and SHELXL-97 programs [57]. Crystallographic data are listed in Table 1. Oligonucleotides
were synthesized [58]. The oligonucleotides were deprotected and cleaved from the solid support by
incubation in 0.1 M NaOH at 55 °C for 14 h. They were purified by denaturing PAGE [gel solution: 7 M
urea, 1 TBE buffer, 18% or 14% (depending on the length of the oligonucleotides)
polyacrylamide/bisacrylamide (29:1); loading buffer: 11.8 M urea, 42 mM Tris-HCl (pH 7.5), 0.83 mM
EDTA (pH 8.0), 8% sucrose] and desalted with NAP 10 columns. Afterwards, they were additionally
incubated in 0.1 M NaOH at 55 °C for 14 h, neutralized with 2 M aqueous triethylammonium acetate
buffer (pH 7.0) and desalted with NAP 10 columns. The desalted oligonucleotides were characterized
by MALDI-ToF mass spectrometry using a 3-hydroxypicolinic acid / ammonium citrate matrix (Fig. S1
– S8, Supplementary data). For quantification of the oligonucleotides, a molar extinction coefficient
ε260 of 0.04 M–1 cm−1 was used for 10–. 1H and 13C NMR spectra were recorded on Bruker Avance(I) 400
and Avance(III) 400 instruments at 300 K. NMR spectra were referenced to residual solvent peaks
(DMSO-d6, CD3OD, CD2Cl2), to tetramethylsilane (CDCl3) or to 3-(trimethylsilyl)propionate (D2O). HRMS
ESI spectra were recorded on an Orbitrap LTQ XXL (Nanospray) from Thermo Scientific. UV and CD
(circular dichroism) spectra were recorded using aqueous solutions comprising 1 M oligonucleotide
duplex, 150 mM NaClO4 (except for duplexes IV and IVr, where 500 mM were used) and 5 mM buffer
(2-(cyclohexylamino)ethanesulfonic acid (CHES) at pH 9.0 for the measurements with Cu(NO3)2 and 3-
(N-morpholino)propanesulfonic acid (MOPS) at pH 6.8 for the measurements with AgNO3. UV spectra
were recorded on a Cary100 Bio instrument between 5 and 75 °C at 1 °C min–1. In the UV melting
profiles, the absorbance at 260 nm was normalized according to Anorm = (A – Amin) / (Amax – Amin).
Melting temperatures (Tm) were determined by applying a Gauss fit to the first derivative of the
5

ACCEPTED MANUSCRIPT
respective melting profile. CD spectra were recorded at 5 °C on JASCO J-815 instrument followed by
smoothing and manual baseline correction.
Synthesis of methyl 1-benzyl-1H-imidazole-4-carboxylate (4). Isopropylmagnesium chloride (2 M in
THF, 2.25 mL, 4.50 mmol, 1.1 equiv.) was added dropwise to a solution of 1-benzyl-4-iodo-1H-
imidazole (3) (1.16 g, 4.08 mmol, 1.0 equiv.) in dry THF (30 mL). After stirring at ambient temperature
for 0.5 h, gaseous CO2 was bubbled through the solution for 5 min. The solid obtained by removing
the solvent was dissolved in methanol (50 mL). After cautious addition of thionyl chloride (1.04 mL,
14.3 mmol, 3.5 equiv.), the solution was refluxed for 48 h. The solution was cooled to ambient
temperature, the solvent was removed in vacuo, and the residue was dissolved in water (20 mL). After
neutralization with aqueous NaOH (1 M), the aqueous layer was extracted with ethyl acetate (3  100
mL) and the combined organic layers were dried (MgSO4). After purification by column
chromatography (SiO2, ethyl acetate), compound 4 was isolated as a white solid (565 mg, 2.61 mmol,
64%). HRMS ESI m/z: [M+Na]+ 239.0787 (calcd. 239.0796). Elemental analysis (%): found: C 66.5, H 5.3,
N 13.1; calcd. for C12H12N2O2: C 66.7, H 5.6, N 13.0. 1H NMR (400 MHz, DMSO-d6),  / ppm: 7.96 (d, 1.3
Hz, 1H, H2), 7.89 (d, 1.3 Hz, 1H, H5), 7.41–7.23 (m, 5H, benzyl), 5.24 (s, 2H, CH2), 3.72 (s, 3H, CH3). 13C
NMR (101 MHz, DMSO-d6),  / ppm: 162.5 (C=O), 138.7 (C2), 136.9 (C4), 132.2 (benzyl, ipso), 128.7
(benzyl, meta), 127.9 (benzyl, para), 127.9 (benzyl, ortho), 126.2 (C5), 50.8 (CH3), 49.8 (CH2).
Synthesis of sodium 1-benzyl-1H-imidazole-4-carboxylate (Na(5)). A solution of 4 (450 mg, 2.08 mmol,
1.0 equiv.) in aqueous NaOH (0.1 M, 22.9 mL, 1.1 equiv.) was stirred for 12 h at 55 °C. After removal
of the solvent in vacuo, the product was recrystallized from ethanol to yield the sodium salt of 5– as
colorless crystals (466 mg, 2.08 mmol, 100%). HRMS ESI m/z: [M+H]+ 225.0634 (calcd. 225.0640).
Elemental analysis (%): found: C 58.8, H 3.8, N 12.3; calcd. for C11H9N2NaO2: C 58.9, H 4.1, N 12.5. 1H
NMR (400 MHz, D2O, pD 7.7),  / ppm: 7.67 (d, 1.5 Hz, 1H, H2), 7.54 (d, 1.5 Hz, 1H, H5), 7.43–7.33 (m,
3H, benzyl)), 7.27 (dd, 7.7 Hz, 1.8 Hz, 2H, benzyl)), 5.16 (s, 2H, CH2). 13C NMR (101 MHz, D2O, pD 7.7),
6

ACCEPTED MANUSCRIPT
 / ppm: 168.3 (C=O), 135.9 (C2), 135.6 (C4), 134.1 (benzyl, ipso), 126.7 (benzyl, meta), 126.0 (benzyl,
para), 125.3 (benzyl, ortho), 121.6 (C5), 48.3 (CH2).
Synthesis of the Cu(II) complex [Cu(bnimc)2]. An aqueous solution (1 mL) of Cu(OAc)2 ∙ H2O (22 mg, 0.11
mmol, 0.5 equiv.) was added to an aqueous solution (10 mL) of ligand Na(5) (50 mg, 0.22 mmol, 1.0
equiv.). After 0.5 h, the blue precipitate was filtered, washed with water (10 mL), ethanol (5 mL) and
diethylether (10 mL) and dried (56 mg, 0.11 mmol, 50%). Single crystals of [Cu(bnimc)2(CH3OH)2]
suitable for X-ray diffraction analysis were obtained by vapor diffusion crystallization in methanol with
diethylether as antisolvent. HRMS ESI m/z: [M+Na]+ 488.0515 (calcd. 488.0522). Elemental analysis
(%): found: C 54.0, H 4.4, N 11.5; calcd. for C22H18CuN4O4 · 1.5 H2O: C 53.6, H 4.3, N 11.4.
Synthesis of the Ag(I) complex [Ag4(bnimc)4] · 2 H2O. An aqueous solution (1 mL) of AgClO4 (50 mg,
0.22 mmol, 1.0 equiv.) was added to an aqueous solution (10 mL) of ligand Na(5) (50 mg, 0.22 mmol,
1.0 equiv.). The colorless precipitate was filtered, washed with water (10 mL), ethanol (5 mL) and
diethylether (10 mL) and dried (35 mg, 0.10 mmol, 45%). Single crystals suitable for X-ray diffraction
analysis were obtained by vapor diffusion crystallization in acetonitrile with diethylether as
antisolvent. HRMS ESI m/z: [Ag(bnimc)2]– 509.0349 (calcd. 509.0379). Elemental analysis (%): found:
C 37.4, H 2.5, N 8.0; calcd. for C11H9AgN2O2 · 2.5 H2O: C 37.3, H 4.0, N 7.9. 1H NMR (400 MHz, CDCl3), 
/ ppm: 8.08 (s, 1H, H2), 7.67 (s, 1H, H5), 7.40–7.27 (m, 5H, benzyl), 5.26 (s, 2H, CH2). 13C NMR (101
MHz, CDCl3),  / ppm: 165.5 (C=O), 138.2 (C2), 137.1 (benzyl), 136.4 (C4), 128.7 (benzyl), 128.0
(benzyl), 127.7 (benzyl), 122.5 (C4), 50.3 (CH2).
Synthesis of 3,5-di-O-p-toluoyl-1,2-dideoxy-1-(4-(methylcarboxy)imidazol-1-yl)-D-ribofuranose (6).
Methyl 1H-imidazole-4-carboxylate (500 mg, 3.96 mmol, 1.0 equiv.) was suspended in dry THF
(30 mL). Sodium hydride (60% in mineral oil, 238 mg, 5.95 mmol, 1.5 equiv.) was added at 0 °C. After
0.5 h of stirring, Hoffer’s chloro sugar (2.29 g, 5.95 mmol, 1.5 equiv.) was added in three portions over
1 h. The reaction mixture was stirred at ambient temperature overnight, and the solvent was removed
in vacuo. The residue was dissolved in CH2Cl2 (100 mL) and washed with water (3  25 mL). The organic
7

ACCEPTED MANUSCRIPT
layer was dried (MgSO4) and the solvent removed in vacuo. After purification by column
chromatography (cyclohexane : ethyl acetate 2:1  ethyl acetate), compound 6 was obtained as a
white solid (965 mg, 2.02 mmol, 51%). HRMS ESI m/z: [M+Na]+ 501.1649 (calcd. 501.1638). Elemental
analysis (%): found: C 65.1, H 5.4, N 5.8; calcd. for C26H26N2O7: C 65.3, H 5.5, N 5.9. 1H NMR (400 MHz,
CDCl3),  / ppm: 7.94–7.90 (m, 2H, Tol), 7.89–7.85 (m, 2H, Tol), 7.76 (d, 1.4 Hz, 1H, H5), 7.70 (d, 1.4 Hz,
1H, H2), 7.28–7.21 (m, 4H, Tol), 6.13 (dd, 8.1 Hz, 5.6 Hz, 1H, H1′), 5.65 (dt, 6.1 Hz, 2.4 Hz, 1H, H3′), 4.62
(d, 3.7 Hz, 2H, H5’, H5′′), 4.58 (td, 3.7 Hz, 2.4 Hz, 1H, H4′), 3.83 (s, 3H, OCH3), 2.75 (ddd, 14.2 Hz, 5.7
Hz, 2.4 Hz, 1H, H2′), 2.64 (ddd, 14.2 Hz, 8.1 Hz, 6.2 Hz, 1H, H2′′), 2.42 (s, 3H, CH3), 2.40 (s, 3H, CH3). 13C
NMR (101 MHz, CDCl3),  / ppm: 166.1 (C=O (Tol)), 165.8 (C=O (Tol)), 162.8 (C=O), 144.6 (Tol), 144.2
(Tol), 136.2 (C2), 134.4 (C4), 129.7 (Tol), 129.6 (Tol), 129.3 (Tol), 129.3 (Tol), 126.5 (Tol), 126.2 (Tol),
122.5 (C5), 86.6 (C1′), 83.1 (C4′), 74.8 (C3′), 63.8 (C5′), 51.6 (OCH3), 39.6 (C2′), 21.7 (CH3), 21.6 (CH3).
Synthesis of 1,2-dideoxy-1-(4-(methylcarboxy)-1H-imidazol-1-yl)-D-ribofuranose (7). Compound 6
(1.14 g, 2.37 mmol, 1.0 equiv.) was suspended in a solution of sodium methoxide (282 mg, 5.22 mmol,
2.2 equiv.) in methanol (30 mL) and stirred at ambient temperature for 2 h. The solvent was removed
in vacuo. After purification of the residue by column chromatography (CH2Cl2 : methanol 10:1),
compound 7 was obtained as an off-white solid (466 mg, 1.92 mmol, 81%). This compound was
structurally characterized by single crystal X-ray diffraction analysis (Fig. S11, Supplementary data).
HRMS ESI m/z: [M+H]+ 243.0989 (calcd. 243.0981). Elemental analysis (%): found: C 49.5, H 6.0, N 11.6;
1
calcd. for C10H14N2O5: C 49.6, H 5.8, N 11.6. H NMR (400 MHz, CD3OD),  / ppm: 8.08 (d, 1.4 Hz, 1H,
H5), 8.02 (d, 1.4 Hz, 1H, H2), 6.16 (t, 6.5 Hz, 1H, H1′), 4.50 (dt, 5.6 Hz, 3.8 Hz, 1H, H3′), 4.02 (q, 3.8 Hz,
1H, H4′), 3.88 (s, 3H, OCH3), 3.81–3.69 (m, 2H, H5’, H5′′), 2.55–2.44 (m, 2H, H2’, H2′′). 13C NMR (101
MHz, CD3OD),  / ppm: 164.4 (C=O), 138.7 (C2), 134.1 (C4), 124.9 (C5), 89.4 (C4′), 88.2 (C1′), 72.4 (C3′),
63.0 (C5′), 52.1 (OCH3), 42.9 (C2′).
Synthesis
of
5-O-(4,4’-dimethoxytrityl)-1,2-dideoxy-1-(4-(methylcarboxy)-1H-imidazol-1-yl)-D-
ribofuranose (8). 4,4’-Dimethoxytrityl chloride (787 mg, 2.33 mmol, 1.2 equiv.) was added to a solution
8

ACCEPTED MANUSCRIPT
of compound 7 (466 mg, 1.94 mmol, 1.0 equiv.) in dry pyridine (20 mL) in the presence of catalytic
amounts of 4’-(dimethylamino)pyridine and stirred at ambient temperature for 2 h. The solution was
diluted with CH2Cl2 (50 mL) and washed with saturated aqueous NaHCO3 (3  20 mL). The organic layer
was dried (MgSO4) and the solvent removed in vacuo. After purification by column chromatography
(cyclohexane : ethyl acetate : triethylamine 2:1:0.03  ethyl acetate : methanol : triethylamine
1:0.05:0.03), compound 8 was obtained as a white foam (943 mg, 1.70 mmol, 88%). HRMS ESI m/z:
[M+Na]+ 567.2140 (calcd. 567.2107). Elemental analysis (%): found: C 67.5, H 6.1, N 4.9; calcd. for
C31H32N2O7 · 0.5 H2O: C 67.3, H 6.0, N 5.1. 1H NMR (400 MHz, CD2Cl2),  / ppm: 7.74 (d, 1.4 Hz, 1H, H5),
7.62 (d, 1.3 Hz, 1H, H2), 7.43–7.38 (m, 2H, DMT), 7.32–7.25 (m, 6H, DMT), 7.24–7.18 (m, 1H, DMT),
6.85–6.80 (m, 4H, DMT), 6.03 (t, 6.4 Hz, 1H, H1′), 4.56 (dt, 5.9 Hz, 3.6 Hz, 1H, H3′), 4.16–4.06 (m, 1H,
H4′), 3.77 (s, 9H, OCH3), 3.33 (dd, 10.3 Hz, 4.8 Hz, 1H, H5′ or H5’’), 3.24 (dd, 10.3 Hz, 4.2 Hz, 1H, H5’ or
H5′′), 2.52–2.37 (m, 2H, H2′, H2′’). 13C NMR (101 MHz, CD2Cl2),  / ppm: 163.3 (C=O), 159.1 (DMT),
145.2 (DMT), 137.0 (C2), 136.1 (DMT), 136.0 (DMT), 134.2 (C4), 130.4 (DMT), 128.4 (DMT), 128.3
(DMT), 127.2 (DMT), 123.4 (C5), 113.6 (DMT), 86.9 (C1′), 86.8 (C4′), 72.4 (C3′), 64.3 (C5′), 55.6 (OCH3
(DMT)), 51.7 (OCH3), 42.2 (C2′).
Synthesis of 2-cyanoethyl (5-O-(4,4’-dimethoxytrityl)-1,2,3-trideoxy-1-(4-(methylcarboxy)-1H-
imidazol-1-yl)-D-ribofuranos-3-yl) diisopropylphosphoramidite (9). N,N-Diisopropylethylamine (144
L, 0.827 mmol, 4.6 equiv.) was added to a solution of compound 8 (100 mg, 0.181 mmol, 1.0 equiv.)
in dry CH2Cl2 (2 mL). Cyanoethyl-N,N-diisopropylchlorophosphoramidite (56 L, 0.25 mmol, 1.4 equiv.)
was added, and the solution was stirred at ambient temperature for 2 h. After purification by column
chromatography (cyclohexane : ethyl acetate : triethylamine 1:1:0.01), compound 9 was obtained as
a mixture of diastereomers as a white foam (121 mg, 0.162 mmol, 90%). HRMS ESI m/z: [M+Na]+
767.3184 (calcd. 767.3186). 1H NMR (400 MHz, CD2Cl2),  / ppm: 7.82–7.72 (m, 1H, H5), 7.64 (m, 1H,
H2), 7.45–7.38 (m, 2H, DMT), 7.35–7.19 (m, 7H, DMT), 6.90–6.78 (m, 4H, DMT), 6.08–5.97 (m, 1H,
H1′), 4.68–4.53 (m, 1H, H3′), 4.30–4.16 (m, 1H, H4′), 3.78 (s, 9H, OCH3), 3.75–3.54 (m, 4H, OCH2, CH),
9

ACCEPTED MANUSCRIPT
3.34–3.20 (m, 2H, H5′, H5′’), 2.67–2.57 (m, 1H, H2′), 2.57–2.52 (m, 1H, H2′′), 2.52–2.36 (m, 2H, CH2CN),
1.22–1.05 (m, 12H, CH3). 31P NMR (162 MHz, CD2Cl2),  / ppm: 148.9, 148.8.
Synthesis of sodium 1,2-dideoxy-1-(4-carboxyimidazol-1-yl)-D-ribofuranose (Na(10)). A solution of
compound 7 (100 mg, 0.413 mmol, 1.0 equiv.) in aqueous NaOH (0.1 M, 41 mL, 1.0 equiv.) was stirred
at ambient temperature for 3 d. After removal of the solvent in vacuo, the product was obtained as a
white solid (103 mg, 0.412 mmol, 100%). This compound was structurally characterized by single
crystal X-ray diffraction analysis (Fig. S12, Supplementary data). HRMS ESI m/z: [M+H]+ 251.0638
(calcd. 251.0644). Elemental analysis (%): found: C 38.9, H 4.7, N 10.0; calcd. for C9H1N2NaO5 · 1.5 H2O:
C 39.0, H 5.1, N 10.1. 1H NMR (400 MHz, D2O, pD 7.6),  / ppm: 7.85 (d, 1.3 Hz, 1H, H2), 7.72 (d, 1.4
Hz, 1H, H5), 6.16 (t, 6.5 Hz, 1H, H1′), 4.52 (dt, 5.9 Hz, 3.9 Hz, 1H, H3′), 4.08 (dt, 5.3 Hz, 3.8 Hz, 1H, H4′),
3.77 (dd, 12.3 Hz, 4.0 Hz, 1H, H5′ or H5’’), 3.68 (dd, 12.3 Hz, 5.5 Hz, 1H, H5’ or H5′′), 2.62–2.43 (m, 2H,
H2′, H2′′). 13C NMR (101 MHz, D2O, pD 7.6),  / ppm: 170.4 (C=O), 138.4 (C4), 137.0 (C2), 121.1 (C5),
86.9 (C4′), 86.2 (C1′), 70.9 (C3′), 61.6 (C5′), 39.9 (C2′).
10

ACCEPTED MANUSCRIPT
3. Results and Discussion
3.1 Synthesis of model nucleobase and nucleoside
To investigate the applicability of imidazole-4-carboxylate (imc–) as a ligand in metal-mediated base
pairing, a model nucleobase was synthesized first. In model nucleobases, the carbohydrate moiety is
formally replaced by an alkyl group, leaving only the nucleobase donor atoms available for metal ion
coordination. Moreover, this approach simplifies the NMR spectra and makes crystallization more
likely. Model nucleobases have been prominently used in deciphering the binding mode of cisplatin
[59, 60], but have also been highly valuable in determining possible geometries of metal-mediated
base pairs [39, 61-67]. We therefore decided to investigate the coordination chemistry of 1-benzyl-
1H-imidazole-4-carboxylate (bnimc–, Chart 1b). Its synthesis is shown in Scheme 1. Based on literature
procedures, 1-benzyl-4-iodoimidazole 3 was obtained from imidazole in a three-step procedure.
Initially, imidazole was iodinated to give 4,5-diiodiimidazole 1 [68]. Subsequent deprotonation and
alkylation gave 1-benzyl-4,5-diiodo-1H-imidazole 2 [55]. Using a Grignard reaction, the 5-iodo
substituent was selectively removed to give 1-benzyl-4-iodoimidazole 3 [55]. In a second Grignard
reaction with gaseous CO2, the remaining iodo substituent was replaced by a carboxylate group, which
was transformed into the respective methyl ester without prior purification, yielding methyl 1-benzyl-
1H-imidazole-4-carboxylate 4. Finally saponification quantitatively gave the sodium salt of 1-benzyl-
1H-imidazole-4-carboxylate (Na(5), Na(bnimc)) in a clean reaction.
To obtain the phosphoramidite required for an automated solid-phase synthesis of oligonucleotides
bearing an imidazole-4-carboxylate-containing deoxyribonucleoside, commercially available methyl
1H-imidazole-4-carboxylate was deprotonated and reacted with Hoffer’s chloro sugar to give the p-
toluoyl-protected nucleoside 6 (Scheme 2). Removal of the p-toluoyl protecting groups gave the free
nucleoside 7, still bearing a methyl group at the carboxylate (as required during solid-phase
oligonucleotide synthesis). Consecutive introduction of DMT (4,4’-dimethoxytrityl) and CEDIP
(cyanoethyl N,N-diisopropyl phosphoramidite) protecting groups at the primary and secondary OH
11

ACCEPTED MANUSCRIPT
groups finally gave the required phosphoramidite 9. In a separate reaction, the methyl ether 7 was
saponified using aqueous NaOH to give the sodium salt of the free nucleoside 10–.
3.2 Metal complexation behavior of the model nucleobase
Various metal complexes of derivatives of imidazole-4-carboxylate (imc–) have been reported in the
past, including Cu(II) complexes. The Cu(II) complex of 1-methylimidazole-4-carboxylate (mimc–)
[Cu(mimc)2(OH2)2] adopts a distorted octahedral coordination geometry with trans-oriented ligands
[69]. When using unsubstituted imidazole-4-carboxylate, a planar complex [Cu(imc)2] is obtained,
likewise displaying a trans orientation of the ligands [70, 71]. Interestingly, complexes involving other
metal ions also show ligands in a cis orientation, as for example observed in [Zn(imc)2(OH2)2] [72]. The
ability to form complexes both with cis- and trans-orientation of the ligand is of importance in the
context of metal-mediated base pairing in antiparallel- and parallel-stranded DNA (vide infra).
Our studies using the model nucleobase bnimc– confirm the preference of Cu(II) complexes to adopt
a trans orientation of the ligands. The complex [Cu(bnimc)2(CH3OH)2], obtained by combining aqueous
solutions of Na(5) and Cu(II) acetate and subsequent recrystallization of the blue precipitate from
methanol, contains two axially coordinated methanol ligands in addition to the bnimc– ligands. Fig. 1
shows the molecular structure of [Cu(bnimc)2(CH3OH)2], and Table 2 lists representative bond lengths
and angles. Assuming a dissociation of the weakly bound methanol ligands, the planar complex of
Cu(II) and imidazole-4-carboxylate should be ideally suited for an incorporation as a metal-mediated
base pair into an oligonucleotide duplex. In such a base pair, the N1 atoms bonded to the benzyl
substituents in the model nucleobase would be involved in the glycosidic bonds. To enable a trans
orientation of the N1 atoms as observed in the crystal structure, the use of a parallel-stranded DNA
duplex would be advisable, as most parallel-stranded nucleic acid duplexes display a trans orientation
of the glycosidic bonds [73].
12

ACCEPTED MANUSCRIPT
As numerous metal-mediated base pairs have been reported using Ag(I) ions as central metal ions
incorporated into the base pairs, we also investigated the formation of an Ag(I) complex of the model
nucleobase bnimc–. As would be required for metal-mediated base pair formation, two bnimc–
moieties combine with one Ag(I) ion. In the resulting [Ag4(bnimc)4] complex, two such [Ag(bnimc)2]–
units dimerize via two additional Ag(I) ions. Fig. 2 shows the molecular structure of [Ag4(bnimc)4]. Each
individual [Ag(bnimc)2]– unit of the tetranuclear complex represents the structure of the anticipated
Ag(I)-mediated base pair. One of these units is indicated by a frame in Fig. 2. The N1 atoms are in a
cisoid arrangement, which is in agreement with the requirements for an incorporation as a metal-
mediated base pair within a B-DNA double helix [74]. Table 3 lists representative interatomic distances
and bond angles. It is interesting to note that the Ag(I) ion is located asymmetrically in-between the
two bnimc– ligands, as particularly evident from the two different Ag1–O bond lengths (Ag1–O1i,
2.599(2) Å; Ag1–O1i1, 2.746(2) Å). The latter bond is rather long, but several precedents exist for such
long Ag–O distances [75, 76]. Moreover, the bond length is well within the sum of the van der Waals
radii of the elements involved [77]. In [Ag4(bnimc)4], two [Ag(bnimc)2]– entities are bridged via two
Ag(I) ions. In the resulting carboxylate-bridged dimer, a short Ag2–Ag2a distance of 2.9507(5) Å is
observed, indicating the possibility of argentophilic interactions [78]. In the crystal structure,
additional short Ag–Ag contacts are present between [Ag4(bnimc)4] moieties of neighboring planes
(Ag1–Ag2, 2.9366(5) Å, symmetry operation for Ag2: 1+x, y, z), leading to the formation of a
supramolecular network (not shown).
3.3 Characterization of the artificial nucleoside
Prior to any metal complexation study performed in water, the acidity constants of the respective
ligand need to be established to rule out a possible competition of protonation and metalation events.
1
Towards this end, the H NMR chemical shifts of the free nucleoside 10– were monitored in a pD-
dependent fashion (Fig. 3). A least-squares fit of the data indicates the presence of two pKa values at
13

ACCEPTED MANUSCRIPT
1.95  0.05 and 5.36  0.02 (Table S1, Supplementary data) [79], assigned to the deprotonation of the
carboxylic acid and the imidazole moiety, respectively. Hence, the artificial nucleobase is fully
deprotonated above ca. pH 6, suggesting that competing protonation events do not take place under
neutral and alkaline conditions.
During the preparation of the free nucleoside 10– required for the pKa determination, single crystals
of its sodium salt were obtained that were suitable for a structure determination by X-ray diffraction
analysis. Fig. 4 shows the molecular structure of 10–. A structure including the sodium ion is given in
the Supplementary data (Fig. S12). The 1,2-dideoxy-1-(4-carboxyimidazol-1-yl)-D-ribofuranose adopts
a C3’-endo conformation (Table S2, Supplementary data). The artificial nucleobase is oriented anti
with respect to the carbohydrate, with the designated metal binding sites N3i and O1i pointing in the
direction of the center of a hypothetical double helix. Hence, the structure corroborates the
applicability of this artificial nucleoside in metal-mediated base pairing.
3.4 Formation of metal-mediated base pairs
To investigate the capability of imidazole-4-carboxylate to engage in metal-mediated base pairing,
four DNA oligonucleotide duplexes were synthesized and evaluated (Table 4). Duplexes I – III contain
one X:X pair each (X = 1,2-dideoxy-1-(4-carboxyimidazol-1-yl)-D-ribofuranose), whereas duplex IV
comprises two X:X pairs. Moreover, the former duplexes are antiparallel-stranded, whereas a parallel
strand arrangement is adopted in the latter case. For comparison, reference duplexes Ir – IVr were
investigated, too. In these duplexes, the X:X pairs were formally replaced by G:C (duplexes Ir and IIr)
or A:T (duplexes IIIr and IVr) base pairs, to make the duplexes devoid of designated metal-binding sites
(G, guanine; A, adenine).
The formation of a metal-mediated base pair is typically accompanied by an increase in thermal
stability, caused by the formation of coordinate bonds and evident from a rise in the melting
14

ACCEPTED MANUSCRIPT
temperature Tm [22]. Accordingly, Tm was determined for all duplexes under investigation by means
of temperature-dependent UV spectroscopy. The data obtained for duplex I will be described in detail
in the following, whereas the corresponding data for duplexes II – IV and for all reference duplexes
are given in the Supplementary data. Figure 5 shows the melting curves of duplex I in the presence of
increasing amounts of Cu(II). As can nicely be seen, the initial addition of Cu(II) leads to a large increase
in Tm. As expected for metal-mediated base pair formation, the addition of more than one Cu(II) per
X:X pair does not influence the thermal stability any further. In case of reference duplex Ir without the
designated Cu(II)-binding site, no significant change in Tm is observed upon the addition of Cu(II) (Table
5 and Supplementary data, Fig. S13). Hence, it can be concluded that in duplex I one Cu(I) binds per
X:X pair to give an X–Cu(II)–X base pair. The formation of this base pair is accompanied by a significant
increase of the melting temperature Tm from 22.0 °C to 41.9 °C (Tm = 19.9 °C). According to a CD
spectroscopic study, the B-DNA conformation of duplex I is little affected by the formation of the X–
Cu(II)–X base pair (Supplementary data, Fig. S14a). Similar results were obtained for duplexes II – IV
and the corresponding reference duplexes (Supplementary data, Figs. S15 – S20). Table 5 summarizes
the increase in melting temperature upon the formation of X–Cu(II)–X base pairs for all four duplexes,
Chart 2 shows the proposed structure of the metal-mediated base pair.
It is worth highlighting some peculiarities of duplex IV, which is the only parallel-stranded duplex
among the investigated ones. As a result of its less favored parallel strand orientation, its melting
temperature is lower than those of the antiparallel-stranded duplexes. Nonetheless, the incorporation
of Cu(II) to form X–Cu(II)–X base pairs is accompanied by a large increase in thermal stability. In fact,
the thermal stability of duplex IV after formation of the X–Cu(II)–X base pairs is essentially identical to
that of the reference duplex IVr. In other words, the destabilization evoked by the introduction of X:X
mispairs is compensated by their involvement in metal-mediated base pair formation. The thermal
stabilization of ca. 11.7 °C per X–Cu(II)–X base pair is much smaller than that observed for the
antiparallel-stranded duplexes I – III though. This finding is unexpected, because for geometric
considerations duplex IV must contain the metal-mediated pairs with a transoid orientation of the
15

ACCEPTED MANUSCRIPT
glycosidic bonds (Chart 2b) [73]. This orientation was found to be the preferred one outside the DNA
context, as evident from the crystal structure of the model nucleobase complex (Fig. 1). Presumably,
the transoid X–Cu(II)–X base pair does not perfectly match the geometry of a parallel-stranded duplex.
This assumption is corroborated by the observation of a change in the CD spectrum of duplex IV upon
the formation of the Cu(II)-mediated base pairs (Supplementary data, Fig. S19b).
As the unsubstituted imidazole nucleobase has been reported as an excellent Ag(I)-binding ligand, we
also investigated the applicability of imidazole-4-carboxylate in Ag(I)-mediated base pairing. Again,
duplexes I – IV were investigated. Fig. 6 shows the melting curves of duplex I in the presence of
increasing amounts of Ag(I). As had already been observed upon the addition of Cu(II), a significant
rise in Tm is found when one Ag(I) is added per X:X pair. The addition of excess Ag(I) does not lead to
significant further increase in the melting temperature. According to the CD spectra, the B-DNA
conformation is not affected by the presence of Ag(I) (Supplementary data, Fig. S14b). Moreover, Ag(I)
does not influence the thermal stability of the reference duplex Ir (Supplementary data, Fig. S21).
Hence, it can be reasoned that an X–Ag(I)–X base pair is formed in duplex I. The same conclusion can
be drawn for duplexes II – IV (Supplementary data, Fig. S22 – S27). As can be seen from a comparison
of the melting temperatures of the duplexes in the presence of Ag(I) and Cu(II), respectively, X–Cu(II)–
X are significantly more stabilizing than X–Ag(I)–X base pairs (Table 5).
4. Summary and conclusions
For the first time, an imidazole-base artificial nucleoside has been used to generate a Cu(II)-mediated
base pair. By applying imidazole-4-carboxylate (X), highly stabilizing X–Cu(II)–X base pairs were formed
in a variety of different DNA duplexes. Notably, these X–Cu(II)–X base pairs are compatible with both
antiparallel-stranded and parallel-stranded duplexes. In the former case, the formation of such a base
pair is accompanied by an increase in duplex melting temperature Tm of approximately 20 °C, whereas
16

ACCEPTED MANUSCRIPT
Tm increases by only ~12 °C per X–Cu(II)–X base pair in the latter case. The X–Cu(II)–X base pairs are
more stabilizing than the corresponding X–Ag(I)–X base pairs in the same duplex sequences. Hence,
the design principle of providing one neutral and one anionic donor site to guarantee the formation
of an overall neutral X–Cu(II)–X base pair was applied successfully. Imidazole-4-carboxylate is the
smallest Cu(II)-binding nucleobase reported to date. Due to its small aromatic surface, its  stacking
interactions with neighboring nucleobases do not significantly contribute to duplex stability, as
evident from the low melting temperatures in the absence of Ag(I) or Cu(II) ions. Hence, the large
increase in Tm upon the formation X–Cu(II)–X base pairs can be attributed to the generation of the
coordinate bonds. Due to its simplicity, imidazole-4-carboxylate is expected to become a prominent
artificial nucleobase for metal-mediated base pairing.
Abbreviations
A
adenine
bnimc–
C
1-benzyl-1H-imidazole-4-carboxylate
cytosine
CD
circular dichroism
CEDIP
CHES
DMT
G
cyanoethyl N,N-diisopropyl phosphoramidite
2-(cyclohexylamino)ethanesulfonic acid
4,4’-dimethoxytrityl
guanine
imc–
1H-imidazole-4-carboxylate
17

ACCEPTED MANUSCRIPT
mimc–
1-methyl-1H-imidazole-4-carboxylate
3-(N-morpholino)propanesulfonic acid
thymine
MOPS
T
Tm
X
melting temperature
imidazole-4-carboxylate
Appendix A. Supplementary data
CCDC 1863218-1863221 contain the supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Supplementary data to this article can be found online at
http://dx.doi.org/10.1016/j.jinorgbio.2018.XX.XXX.
References
1.
2.
3.
4.
J. Müller, Metallomics 2 (2010) 318-327.
S. Katz, Nature 194 (1962) 569.
Z. Kuklenyik, L. G. Marzilli, Inorg. Chem. 35 (1996) 5654-5662.
H. Yamaguchi, J. Šebera, J. Kondo, S. Oda, T. Komuro, T. Kawamura, T. Dairaku, Y. Kondo, I.
Okamoto, A. Ono, J. V. Burda, C. Kojima, V. Sychrovský, Y. Tanaka, Nucleic Acids Res. 42
(2014) 4094-4099.
5.
T. Dairaku, K. Furuita, H. Sato, J. Šebera, K. Nakashima, J. Kondo, D. Yamanaka, Y. Kondo, I.
Okamoto, A. Ono, V. Sychrovský, C. Kojima, Y. Tanaka, Chem. Eur. J. 22 (2016) 13028-13031.
18

ACCEPTED MANUSCRIPT
6.
7.
J. Kondo, Y. Tada, T. Dairaku, Y. Hattori, H. Saneyoshi, A. Ono, Y. Tanaka, Nat. Chem. 9 (2017)
956-960.
D. A. Megger, C. Fonseca Guerra, J. Hoffmann, B. Brutschy, F. M. Bickelhaupt, J. Müller,
Chem. Eur. J. 17 (2011) 6533-6544.
8.
9.
F.-A. Polonius, J. Müller, Angew. Chem. Int. Ed. 46 (2007) 5602-5604.
N. Santamaría-Díaz, J. M. Méndez-Arriaga, J. M. Salas, M. A. Galindo, Angew. Chem. Int. Ed.
55 (2016) 6170-6174.
10.
11.
J. M. Méndez-Arriaga, C. R. Maldonado, J. A. Dobado, M. A. Galindo, Chem. Eur. J. 24 (2018)
4583-4589.
Y. Takezawa, M. Shionoya, J. Müller, in Comprehensive Supramolecular Chemistry II, ed. J. L.
Atwood, Elsevier Ltd., Oxford, 2017, vol. 4, pp. 259-293.
B. Jash, J. Müller, Eur. J. Inorg. Chem. (2017) 3857-3861.
Y.-W. Lin, H.-T. Ho, C.-C. Huang, H.-T. Chang, Nucleic Acids Res. 36 (2008) e123.
H. Torigoe, A. Ono, T. Kozasa, Transition Metal Chem. 36 (2011) 131-144.
J. Joseph, G. B. Schuster, Org. Lett. 9 (2007) 1843-1846.
T. Ehrenschwender, W. Schmucker, C. Wellner, T. Augenstein, P. Carl, J. Harmer, F. Breher,
H.-A. Wagenknecht, Chem. Eur. J. 19 (2013) 12547-12552.
S. Liu, G. H. Clever, Y. Takezawa, M. Kaneko, K. Tanaka, X. Guo, M. Shionoya, Angew. Chem.
Int. Ed. 50 (2011) 8886-8890.
12.
13.
14.
15.
16.
17.
18.
19.
20.
S. Hensel, K. Eckey, P. Scharf, N. Megger, U. Karst, J. Müller, Chem. Eur. J. 23 (2017) 10244-
10248.
J. C. Léon, Z. She, A. Kamal, M. H. Shamsi, J. Müller, H.-B. Kraatz, Angew. Chem. Int. Ed. 56
(2017) 6098-6102.
E. Toomey, J. Xu, S. Vecchioni, L. Rothschild, S. Wind, G. E. Fernandes, J. Phys. Chem. C 120
(2016) 7804-7809.
19

ACCEPTED MANUSCRIPT
21.
J. C. Léon, G. González-Abradelo, C. A. Strassert, J. Müller, Chem. Eur. J. 24 (2018) 8320-
8234.
22.
23.
B. Jash, J. Müller, Chem. Eur. J. 23 (2017) 17166-17178.
T. Kobayashi, Y. Takezawa, A. Sakamoto, M. Shionoya, Chem. Commun. 52 (2016) 3762-
3765.
24.
T. Funai, J. Nakamura, Y. Miyazaki, R. Kiriu, O. Nakagawa, S.-i. Wada, A. Ono, H. Urata,
Angew. Chem. Int. Ed. 53 (2014) 6624-6627.
25.
26.
E.-K. Kim, C. Switzer, ChemBioChem 14 (2013) 2403-2407.
T. Funai, Y. Miyazaki, M. Aotani, E. Yamaguchi, O. Nakagawa, S.-i. Wada, H. Torigoe, A. Ono,
H. Urata, Angew. Chem. Int. Ed. 51 (2012) 6464-6466.
27.
28.
29.
30.
31.
32.
C. Kaul, M. Müller, M. Wagner, S. Schneider, T. Carell, Nat. Chem. 3 (2011) 794-800.
Y. Takezawa, J. Müller, M. Shionoya, Chem. Lett. 46 (2017) 622-633.
G. H. Clever, C. Kaul, T. Carell, Angew. Chem. Int. Ed. 46 (2007) 6226-6236.
B. Jash, J. Müller, Angew. Chem. Int. Ed. 57 (2018) 9524-9527.
K. Tanaka, A. Tengeiji, T. Kato, N. Toyama, M. Shionoya, Science 299 (2003) 1212-1213.
K. Tanaka, G. H. Clever, Y. Takezawa, Y. Yamada, C. Kaul, M. Shionoya, T. Carell, Nat.
Nanotechnol. 1 (2006) 190-194.
33.
34.
35.
36.
37.
38.
39.
40.
S. Atwell, E. Meggers, G. Spraggon, P. G. Schultz, J. Am. Chem. Soc. 123 (2001) 12364-12367.
H. Weizman, Y. Tor, J. Am. Chem. Soc. 123 (2001) 3375-3376.
C. Switzer, S. Sinha, P. H. Kim, B. D. Heuberger, Angew. Chem. Int. Ed. 44 (2005) 1529-1532.
C. Switzer, D. Shin, Chem. Commun. (2005) 1342-1344.
B. Jash, J. Müller, J. Inorg. Biochem. 186 (2018) 301-306.
S. K. Maity, T. Lönnberg, Chem. Eur. J. 24 (2018) 1274-1277.
G. H. Clever, Y. Söltl, H. Burks, W. Spahl, T. Carell, Chem. Eur. J. 12 (2006) 8708-8718.
Y. Takezawa, S. Yoneda, J.-L. H. A. Duprey, T. Nakama, M. Shionoya, Chem. Sci. 7 (2016)
3006-3010.
20

ACCEPTED MANUSCRIPT
41.
42.
D. M. Engelhard, J. Nowack, G. H. Clever, Angew. Chem. Int. Ed. 56 (2017) 11640-11644.
H. Yang, C. K. McLaughlin, F. A. Aldaye, G. D. Hamblin, A. Z. Rys, I. Rouiller, H. F. Sleiman, Nat.
Chem. 1 (2009) 390-396.
43.
X. Guo, P. Leonard, S. A. Ingale, J. Liu, H. Mei, M. Sieg, F. Seela, Chem. Eur. J. 24 (2018) 8883-
8892.
44.
45.
46.
47.
48.
H. Zhao, P. Leonard, X. Guo, H. Yang, F. Seela, Chem. Eur. J. 23 (2017) 5529-5540.
H. Yang, H. Mei, F. Seela, Chem. Eur. J. 21 (2015) 10207-10219.
N. Zimmermann, E. Meggers, P. G. Schultz, J. Am. Chem. Soc. 124 (2002) 13684-13685.
S. Johannsen, N. Megger, D. Böhme, R. K. O. Sigel, J. Müller, Nat. Chem. 2 (2010) 229-234.
S. Kumbhar, S. Johannsen, R. K. O. Sigel, M. P. Waller, J. Müller, J. Inorg. Biochem. 127 (2013)
203-210.
49.
50.
51.
52.
53.
54.
K. Schweizer, J. C. Léon, B. J. Ravoo, J. Müller, J. Inorg. Biochem. 160 (2016) 256-263.
X. Tan, S. Litau, X. Zhang, J. Müller, Langmuir 31 (2015) 11305-11310.
S. Hensel, N. Megger, K. Schweizer, J. Müller, Beilstein J. Org. Chem. 10 (2014) 2139-2144.
K. Schweizer, J. Kösters, J. Müller, J. Biol. Inorg. Chem. 20 (2015) 895-903.
L. Zhang, E. Meggers, J. Am. Chem. Soc. 127 (2005) 74-75.
M. Su, M. Tomás-Gamasa, S. Serdjukow, P. Mayer, T. Carell, Chem. Commun. 50 (2014) 409-
411.
55.
C. J. Lovely, H. Du, R. Sivappa, M. R. Bhandari, Y. He, H. V. R. Dias, J. Org. Chem. 72 (2007)
3741-3749.
56.
57.
58.
59.
60.
V. Rolland, M. Kotera, J. Lhomme, Synth. Commun. 27 (1997) 3505-3511.
G. M. Sheldrick, Acta Cryst. A64 (2008) 112-122.
P. Scharf, B. Jash, J. A. Kuriappan, M. P. Waller, J. Müller, Chem. Eur. J. 22 (2016) 295-301.
B. Lippert, G. Raudaschl, C. J. L. Lock, P. Pilon, Inorg. Chim. Acta 93 (1984) 43-50.
H. Schöllhorn, G. Raudaschl-Sieber, G. Müller, U. Thewalt, B. Lippert, J. Am. Chem. Soc. 107
(1985) 5932-5937.
21

ACCEPTED MANUSCRIPT
61.
62.
D. A. Megger, J. Kösters, A. Hepp, J. Müller, Eur. J. Inorg. Chem. (2010) 4859-4864.
K. Seubert, D. Böhme, J. Kösters, W.-Z. Shen, E. Freisinger, J. Müller, Z. Anorg. Allg. Chem.
638 (2012) 1761-1767.
63.
C. Radunsky, D. A. Megger, A. Hepp, J. Kösters, E. Freisinger, J. Müller, Z. Anorg. Allg. Chem.
639 (2013) 1621-1627.
64.
65.
66.
67.
I. Sinha, J. Kösters, A. Hepp, J. Müller, Dalton Trans. 42 (2013) 16080-16089.
T. Richters, O. Krug, J. Kösters, A. Hepp, J. Müller, Chem. Eur. J. 20 (2014) 7811-7818.
I. Sinha, A. Hepp, J. Kösters, J. Müller, J. Inorg. Biochem. 153 (2015) 355-360.
S. Mandal, C. Wang, R. K. Prajapati, J. Kösters, S. Verma, L. Chi, J. Müller, Inorg. Chem. 55
(2016) 7041-7050.
68.
69.
70.
C. J. Lovely, H. Du, H. V. R. Dias, Heterocycles 60 (2003) 1-7.
L.-J. Yang, W.-T. Song, Y.-H. Luo, B.-W. Sun, Inorg. Nano-Metal Chem. 47 (2017) 493-499.
M. Kondo, E. Shimizu, T. Horiba, H. Tanaka, Y. Fuwa, K. Nabari, K. Unoura, T. Naito, K. Maeda,
F. Uchida, Chem. Lett. 32 (2003) 944-945.
71.
Y.-G. Sun, M.-Y. Guo, G. Xiong, F. Ding, L. Wang, B. Jiang, M.-C. Zhu, E.-J. Gao, F. Verpoort, J.
Coord. Chem. 63 (2010) 4188-4200.
72.
73.
74.
75.
W. Shuai, S. Cai, S. Zheng, Acta Crystallogr. E67 (2011) m897.
J. Müller, Beilstein J. Org. Chem. 13 (2017) 2671-2681.
J. Müller, Eur. J. Inorg. Chem. (2008) 3749-3763.
M. McCann, R. Curran, M. Ben-Shoshan, V. McKee, A. A. Tahir, M. Devereux, K. Kavanagh,
B.v S. Creaven, A. Kellett, Dalton Trans. 41 (2012) 6516-6527.
L. Brammer, M. D. Burgard, M. D. Eddleston, C. S. Rodger, N. P. Rath, H. Adams
CrystEngComm 4 (2002) 239-248.
76.
77.
78.
79.
A. Bondi, J. Phys. Chem. 68 (1964) 441-451.
H. Schmidbaur, A. Schier, Angew. Chem. Int. Ed. 54 (2015) 746-784.
R. Tribolet, H. Sigel, Eur. J. Biochem. 163 (1987) 353-363.
22

ACCEPTED MANUSCRIPT
80.
81.
B. Jash, J. Neugebauer, J. Müller, Inorg. Chim. Acta 452 (2016) 181-187.
K. Petrovec, B. J. Ravoo, J. Müller, Chem. Commun. 48 (2012) 11844-11846.
23

ACCEPTED MANUSCRIPT
Table 1. Crystallographic data for [Cu(bnimc)2(CH3OH)2], [Ag4(bnimc)4] · 4 H2O, compound 7 and Na(10).
[Cu(bnimc)2(CH3OH)2]
[Ag4(bnimc)4] · 4 H2O
7
Na(10)
Empirical formula
C24H26CuN4O6
530.03
C22H22Ag2N4O6
654.17
C20H28N4O10
484.46
C9H11N2NaO5
250.19
Formula weight
Crystal system
Space group
a / Å
Triclinic
Monoclinic
P21/c
Orthorhombic
P212121
5.16500(10)
10.0259(2)
43.0366(8)
90
Orthorhombic
P212121
5.6142(5)
9.2098(8)
20.0500(16)
90
P–1
5.2023(12)
10.525(3)
11.995(4)
110.268(14)
93.096(12)
103.517(8)
592.4(3)
1
7.7354(5)
23.0010(15)
12.7547(8)
90
b / Å
c / Å
 / °
93.639(2)
90
90
90
 / °
90
90
 / °
V / Å3
Z
2264.8(3)
4
2228.60(7)
4
1036.70(15)
4
24

ACCEPTED MANUSCRIPT
calcd. / g cm–1
1.49
1.0
1.92
1.8
1.60
1.0
1.44
0.2

µ(Mo-K) / mm–1
Crystal size / mm
Temperature / K
0.24 × 0.09 × 0.02
100(2)
0.63 × 0.06 × 0.06
100(2)
0.16 × 0.14 × 0.02
100(2)
0.24 × 0.06 × 0.03
100(2)
2.25, 27.85
–6:6, –13:13, –15:15
6846, 2765
2395
2.39, 30.10
–10:10, –32:32, –17:18
32065, 6637
4952
4.11, 76.51
–6:6, –12:11, –52:53
15192, 4483
4268
2.43, 27.94
–7:7, –12:12, –26:26
14305, 2488
2336

min, max / °
Dataset
Tot., uniq. Data
Observed data [I > 2σ(I)]
Nref, Npar
2765, 162
6637, 323
4483, 313
2488, 159
Flack parameter
R, wR2, S [I > 2σ(I)]
Resd. dens. min. and max. / e Å–3
n/a
n/a
0.04(6)
0.03(13)
0.0387, 0.1200, 0.897
0.474, –0.327
0.0347, 0.0713, 1.012
0.824, –0.591
0.0317, 0.0915, 0.735
0.193, –0.291
0.0301, 0.0692, 1.089
0.254, –0.228
25

ACCEPTED MANUSCRIPT
Table 2. Selected bond lengths and angles in [Cu(bnimc)2(CH3OH)2].
Bond length / Å
1.972(2)
Bond angle / °
Cu1–N3i
Cu1–O1i
Cu1–O1
N3i–Cu1–O1i
N3i–Cu1–O1
O1i–Cu1–O1
83.64(7)
88.66(7)
91.11(6)
1.969(2)
2.509(2)
26

ACCEPTED MANUSCRIPT
Table 3. Selected bond lengths and angles in [Ag4(bnimc)4].
Bond length / Å
2.174(2)
Bond angle / °
158.52(9)
70.06(8)
Ag1–N3i
N3i–Ag1–N3i1
N3i–Ag1–O1i
Ag1–N3i1
Ag1–O1i
2.157(2)
2.599(2)
N3i1–Ag1–O1i
O2i1–Ag2–O1i1a
O1i–Ag2a–Ag2
129.56(8)
159.62(7)
148.03(5)
Ag1–O1i1
Ag2–O2i1
Ag2–Ag2a
Ag2a–O1i
Ag2a–O1i1
2.746(2)
2.193(2)
2.9507(5)
2.308(2)
2.241(2)
27

ACCEPTED MANUSCRIPT
Table 4. Oligonucleotide duplexes investigated in this study (X = imidazole-4-carboxylic acid).
[M+H]+ / Da
Sequence
Chemical formula
Entry
Calcd.
Found a
5'–d(GAG GGT XTG AAA G)–3'
3'-d(CTC CCA XAC TTT C)-5'
C129H159N56O75P12
C123H161N38O79P12
C130H160N59O74P12
C123H162N39O78P12
C147H187N50O91P14
C147H185N56O87P14
C148H188N53O90P14
C147H186N57O86P14
C259H334N64O176P25
C255H313N119O134P25
ODN1
ODN2
4064
3806
4064
3805
Duplex I
Duplex Ir
Duplex II
Duplex IIr
Duplex III
Duplex IIIr
Duplex IV
Duplex IVr
5'-d(GAG GGT GTG AAA G)-3'
[80]
3'-d(CTC CCA CAC TTT C)-5'
5'-d(CAC ATT AXT GTT GTA)-3'
ODN3
ODN4
ODN5
ODN6
ODN7
ODN8
4542
4560
4581
4559
7930
7959
4543
4561
4581
4559
7931
7962
3'-d(GTG TAA TXA CAA CAT)-5'
5'-d(CAC ATT AGT GTT GTA)-3'
3'-d(GTG TAA TCA CAA CAT)-5'
5'–d(TTT GTT TGT TTG XTT GTT TTT TTT TT)–3'
3'–d(AAA CAA ACA AAC XAA CAA AAA AAA AA)–5'
5'-d(TTT GTT TGT TTG TTT GTT TTT TTT TT)-3'
3'-d(AAA CAA ACA AAC AAA CAA AAA AAA AA)-5'
5'-d(AAA AAA AAA XTA ATT TTX AAT ATT T)-3'
5'–d(TTT TTT TTT XAT TAA AAX TTA TAA A)–3'
5'-d(AAA AAA AAA ATA ATT TTA AAT ATT T)-3'
5'–d(TTT TTT TTT TAT TAA AAT TTA TAA A)–3'
[81]
C248H309N92O145P24
C248H314N77O155P24
ODN9
7638
7593
7643
7596
ODN10
[37]
a The MALDI-ToF mass spectra are given in the Supplementary data (Fig. S1 – S10) unless the duplexes have been reported previously.
28

ACCEPTED MANUSCRIPT
Table 5. Melting temperature (Tm / °C) of all duplexes under investigation prior to and after the
addition of one metal ion per X:X pair. The change in melting temperature (Tm / °C) is given, too.
Cu(II) a
Tm (Cu)
Ag(I) b
Duplex
Tm (no Cu)
22.0
Tm (no Ag)
20.2
Tm (Ag)
Tm
19.9
0.1
Tm
17.4
I
41.9
50.1
43.0
47.0
61.2
57.6
38.1
37.6
37.6
Ir
50.0
51.5 [80]
24.8
52.7 [80]
40.1
1.2 [80]
15.3
II
24.2
18.8
–0.3
14.9
0.0
IIr
III
IIIr
IV
IVr
47.3
49.1
50.4
1.3
46.3
48.7
57.9
9.2
57.6
56.7 [81]
16.0
57.0 [81]
37.4
0.3 [81]
21.4
14.6
23.5
0.0
37.6
39.6
40.5
0.9
a Data acquired at pH 9.0; b Data acquired at pH 6.8.
29