Alternative DNA Base Pairing through Metal Coordination

Part of the Metal Ions in Life Sciences book series (MILS, volume 10)


Base-pairing in the naturally occurring DNA and RNA oligonucleotide duplexes is based on π-stacking, hydrogen bonding, and shape complementarity between the nucleobases adenine, thymine, guanine, and cytosine as well as on the hydrophobic-hydrophilic balance in aqueous media. This complex system of multiple supramolecular interactions is the product of a long-term evolutionary process and thus highly optimized to serve its biological functions such as information storage and processing. After the successful implementation of automated DNA synthesis, chemists have begun to introduce artificial modifications inside the core of the DNA double helix in order to study various aspects of base pairing, generate new base pairs orthogonal to the natural ones, and equip the biopolymer with entirely new functions. The idea to replace the hydrogen bonding interactions with metal coordination between ligand-like nucleosides and suitable transition metal ions culminated in the development of a plethora of artificial base-pairing systems termed “metal base-pairs” which were shown to strongly enhance the DNA duplex stability. Furthermore, they show great potential for the use of DNA as a molecular wire in nanoscale electronic architectures. Although single electrons have proven to be transmitted by natural DNA over a distance of several base pairs, the high ohmic resistance of unmodified oligonucleotides was identified as a serious obstacle. By exchanging some or all of the Watson-Crick base pairs in DNA with metal complexes, this problem may be solved. In the future, these research efforts are supposed to lead to DNA-like materials with superior conductivity for nano-electronic applications. Other fields of potential application such as DNA-based supramole­cular architecture and catalysis may be strongly influenced by these developments as well. This text is meant to illustrate the basic concepts of metal-base pairing and give an outline over recent developments in this field.


coordination chemistry DNA metal-base pairing nanotechnology 

1 Introduction

1.1 Use of DNA as Intelligent Material

The natural role of DNA is the conservation and propagation of genetic information in each organism. As the result of million years of evolution, it features a maximal density of functionality embedded in its double-helical structure [1].

The interior of the duplex is formed by a parallel stack of pairwise-bonded aromatic nucleobases. Hydrogen bonding, π-stacking, and shape complementarity all play an important role as factors governing the strength and structural integrity of the double strand. Depending on the base-pair sequence and environmental factors such as humidity, salt content, and binding partners, DNA can adopt a number of secondary structures of which the right-handed B-type helix is the predominant form. The latter is characterized by a base-pair distance of 3.4 Å and a helical pitch of 36 degrees per base pair resulting in a complete helix turn approximately every ten base pairs. The structure features two grooves, called the major and minor groove, through which the nucleobases can be accessed and recognized by DNA-binding proteins such as transcription factors in a sequence-dependent manner. The outer surface of the double strand is lined by the negatively-charged sugar-phosphate backbone providing superior water solubility and shielding for the internal hydrophobic π-stack. DNA is a remarkably stable molecule that allows handling in a relatively wide pH range and does not undergo substantial decomposition when heated to a temperature of 100°C for a limited time.

Short DNA strands are routinely synthesized by an automated solid-phase process with full control over the desired sequence. The fully synthetic nature of this process involving no enzymatic reagents allows the introduction of modified building blocks such as artificial nucleosides, unnatural backbones or end-groups regardless of the biological function of DNA. However, the choice of a suitable protecting group strategy complying with the solid-phase synthesis and the subsequent transfer of the products to the aqueous media is required.

Since automated, solid-phase oligonucleotide synthesis is only feasible for the production of relatively short oligonucleotides (in practice up to lengths of about 100 nucleotides), longer DNA strands of hundreds or thousands of base pairs containing artificial building blocks is achievable through the combination of solid-phase DNA synthesis with subsequent enzymatic techniques. Short modified DNA strands can be stitched together and ligated to form longer constructs using proteins called ligases. A new development of recent years is the use of artificial nucleoside triphosphates in the automated polymerase chain reaction (PCR) employing specifically designed or evolved polymerases that are able to process non-natural dNTPs [2].

In the quickly developing field of bottom-up nanotechnology, DNA has become an important building block owing to its superior properties [3]. Several lines of research may benefit from the use of modified DNA structures. In particular, the molecular electronics approach is currently seen as a promising new technology because the ongoing miniaturization of electronic circuits is facing a limit in structure size using the classic silicon-based photolithographic processes [4]. Furthermore, electro-functional organic building blocks such as light-harvesting and charge separating units of future photovoltaic devices might be wired up with such molecular cables.

A great variety of synthetic DNA modifications have been realized in recent years such as fluorescent dyes or surface anchors covalently bound to the 3’ or 5’ ends, outside pointing functionalities protruding from the sides of the double helix and artificial backbone structures dramatically changing the solubility of the DNA constructs (Figure 1) [5]. Most interesting, however, turned out to be the modifications that were introduced right into the middle of the DNA double helix, thus replacing the natural base pairs. Such artificial nucleobases were developed for a range of reasons such as studying the hydrogen bonding and stacking forces that hold together the double helix [6]. Other artificial bases were incorporated into DNA to create new sets of base pairs that are orthogonal to the natural Watson-Crick system and therefore can be seen as an extension of the genetic code [7]. Owing to their wide application in medicinal diagnostics, many artificial bases were developed for the analytical recognition of single nucleotide polymorphisms (SNPs) [8].
Figure 1

Developments in the area of DNA nanotechnology: (a) The “DNA origami” approach makes use of the sequence-specific self-assembly features of DNA and the rational design of inter-duplex connections via cross-links based on the naturally occurring Holliday junction, (b) the covalent attachment of functionalities to the 3’- or the 5’-end(s) of DNA is achievable with contemporary standard methods of automated DNA synthesis, (c) several strategies have been developed for the covalent attachment of functionalities to internal nucleosides protruding out of the DNA duplex, (d) the incorporation of modifications inside the core of the DNA double helix requires the synthesis of artificial nucleobase phosphoramidite building blocks but allows to take direct influence on the process of double strand formation.

Before we restrict our focus on the internal modification of DNA with metal ions coordinated to ligand-like nucleobases, another approach to DNA nanotechnology based on treating DNA as a programmable building material is briefly introduced. The basis for this field, often called “DNA origami”, is that the sequence-dependent hybridization of DNA single strands to double strands and the possibility to generate junctions of several duplexes was recognized as a method for the creation of complex self-assembling 2- and 3-dimensional molecular architectures. This branch of DNA nanotechnology has been pioneered by Seeman [9], followed by reports of complex 2D structures by Winfree et al. [10], Rothemund [11], Yan [12], and Somoza [13]. By going into the 3rd dimension, the topic has recently been brought to another level of sophistication by Shih et al. [14], Gothelf, Kjems, and coworkers [15] (Figure 1a).

1.2 Can DNA Be Used as a Molecular Wire?

While exploring the suitability of DNA as a molecular wire, it turned out that the electron conducting abilities of natural DNA are not sufficient for the use of unmodified double strands in molecular electronic circuits [68]. The literature values for the conductivity actually spread over a great range from 1 to 1 × 107 MΩ. This inconsistency was attributed to the differences in experimental techniques such as the method for contacting the DNA (AFM-based, silicon break junctions, carbon nanotube contacts, etc.), environmental factors (humidity, salt content, buffer), and surface effects [16]. It can also be assumed that the specific sequences of the various DNA strands under investigation play a role for the measured conductivity values.

Astonishingly, the conductance of single excess electrons and positive charges through short patches of DNA is part of biological processes involved in DNA repair [17]. The lossless conductance of many charges, however, over distances of several hundred base pairs does not seem to be a function of DNA that was worth to be optimized by natural evolution. Besides a low ohmic resistance, the transport of a steady current of electrical charges along a molecular wire would also require a high robustness of the oligonucleotide for the successful implementation into a bio-artificial hybrid electronic device [18].

1.3 Metal Coordination of DNA Containing Only Natural Nucleotides

The idea of doping the interior of oligonucleotide duplexes with metal ions in order to yield materials with enhanced conductivity or other interesting electronic effects is not so new [19]. Early experiments to introduce metal ions into the core of a double helix aiming at enhanced electrical conductivity made use of unmodified DNA strands at high pH and metal ions such as Zn(II), Co(II), and Ni(II) [20]. Based on this idea, Lee et al. proposed a model system for a field-effect transistor based on M-DNA [21]. The structure and conductive properties of this so-called M-DNA were, however, controversially discussed and despite some proposed structural models, the exact position of the metal ions inside or around the DNA strands remains rather unclear [19].

More reliable evidence for the binding of metal ions inside DNA duplexes exclusively consisting of natural base pairs was obtained for thymine-rich sequences upon addition of Hg(II) salts. As early as 1952 the effect of Hg(II) on DNA samples was studied. In 1963, the formation and structure of a T-Hg-T metal-base pair, in which each thymine base is deprotonated at the N(3) position, was proposed by Katz (5 in Figure 3, X = CH3; see further below) [22]. Later, this observation was picked up by Buncel et al. and Kuklenyik and Marzilli, who supported the picture of Hg(II) binding to pairs of thymine bases by NMR experiments [23]. Recently, Ono and Togashi picked up this principle again and contributed further studies on the T-Hg-T metal-base pairing [24].

A well documented method for elucidating the interstrand binding of metal ions to oppositely arranged nucleosides is the observation of a duplex-stabilizing effect in terms of an increase in the melting temperature TM (for a more detailed explanation, see below). Whereas TT mismatches are known to strongly destabilize the DNA double helix, the addition of Hg(II) to sequences containing such TT mismatches leads to a significant rise of the thermal duplex stability as observed in the DNA melting profiles. Likewise, Müller, Sigel, and coworkers found that the analogous U-Hg-U metal-base pairs (5 in Figure 3, X = H) can be formed in an RNA context [25]. Recently, Ono et al. reported the C-Ag-C metal-base pair 6 which forms upon addition of Ag(I) to double strands containing CC mismatches [26]. Megger and Müller subsequently found evidence for the formation of parallel-stranded helices containing eight consecutive C-Ag-C base pairs by CD and UV measurements [27]. The fact that TT mismatches bind Hg(II) but no Ag(I) and CC mismatches show an opposite behavior was exploited recently by Willner et al. to create logic AND as well as OR gates based on metal-binding oligonucleotides which are attached to quantum dots [28].

Noteworthy is furthermore the observation of an Au(III) ion bound inside a GC base pair (with guanosine deprotonated at the N(1) position) that was found in a crystallographic study aimed at screening the interaction of an RNA duplex with various metal ions [29]. Some data on the potential for electron conductance through metal-containing oligonucleotides based on natural nucleobases have already been gathered in a small number of experiments. Joseph and Schuster studied the effect of a T-Hg-T base pair on the long-distance radical cation (“hole”) hopping properties but found no significant effect of this metal-base pair on the charge transport [30]. Voityuk however, proposes that a stack of T-Hg-T base pairs might be beneficial for excess electron transfer [31]. Electrical transport through cation-stabilized G-quadruplex DNA was recently reported by Erbe et al. [32].

In an alternative approach, DNA strands bound to surfaces were used as templates for the spatially controlled deposition of metal ions, and subsequent reductive formation of nanoscale wires by thermal soldering or photographic development [33]. Concerning the various roles of metal ions in naturally occurring nucleic acids, the reader may refer to [34] (see also Chapter12 of this book).

Since binding of metal ions to the natural nucleobases was found to be limited to only a small number of metals and Watson-Crick base pairing might interfere with metal binding, a new approach of metal-base pairing starting from designed, artificial nucleobases with metal binding potential was developed.

In these so-called “metal-base pairs”, the natural nucleobase attached to the sugar is replaced by a ligand capable of coordinating a (transition) metal ion in a linear or square-planar fashion [19]. Such a metal-base pair is ideally represented by a flat coordination complex that is integrated into DNA as part of the base pair stack without distorting the shape of the duplex.

1.4 Artificial Metal-Base Pairs

The field of metal-base pairing based on artificial nucleosides was pioneered by Tanaka and Shionoya who studied the coordination chemistry of monomeric ligand-modified nucleosides (14, Figure 2) in aqueous solutions. Although only serving as model compounds, this first work established the foundation of the concept of metal-base pairing [35].
Figure 2

The phenylenediamine, aminophenol, and catechol metal-base pairs. Reprinted from [19a] with permission from Wiley-VCH; copyright 2007.

However, it proved difficult to incorporate the oxidation-sensitive ortho-pheny­lene­diamine, aminophenol, and catechol compounds 1 - 4 into oligonucleotides. Consequently, metal-base pairing inside DNA double strands could not yet be realized in these early experiments. Shortly following this first work, the groups of Schultz and Shionoya both succeeded in using the known pyridine nucleoside P for metal-mediated base pairing inside the DNA double helix. The Shionoya group showed that Ag(I) ions can be coordinated between two (or even three) pyridine nucleosides inside a DNA duplex resulting in the formation of not only double- but also triple-stranded systems (8 in Figure 3, see also Section 2.1) [36]. Schultz et al. developed an unsymmetrical metal-base pair consisting of one pyridine base P and one pyridine ligand equipped with two extra coordinating groups resulting in a family of metal-base pairs denoted as 10 in Figure 3 [37]. Two of the Dipic-Cu-P base pairs (10 with X = O) were successfully incorporated into the dodecamer sequence [d(5’-CGCGDipicATPCGCG)2]. The structure was elucidated by X-ray crystallography showing the Jahn-Teller distorted octahedral coordination of the Cu(II) ions between the P and Dipic ligands forming the metal-base pair and two oxygen atoms of the neighboring nucleotides [38].

Figure 3

The structure of a Watson-Crick GC base pair compared with a selection of the reported metal-base pairs. Only one kind of metal ion is shown in each case and some derivatives are omitted. Since in most cases no or only limited structural data were reported, the real coordination geometries might differ from the structures drawn here. References not mentioned in the text: 5 (X = Br, F, CN) [41], 11 [42], 12 (also reported with a simplified propanediol backbone) [43], 13 [44], and 14 [45]. Reprinted from [19b] with permission from Elsevier; copyright 2010.

Since then, a variety of further metal-base pairs was developed (Figure 3). Only selected examples will be discussed here and the reader may refer to preceding review articles [19,39,40] and the respective original publications mentioned in the caption to Figure 3.

A schematic overview of synthesis and incorporation of ligand-modified nucleosides is given in Figure 4 [46]. Following automated synthesis, deprotection and purification, double strands are obtained by mixing equimolar amounts of matching single strands in an aqueous solution containing an electrolyte such as NaCl of high ionic strength and a buffer to adjust the required pH. Experience has shown that the choice of an appropriate buffer system is crucial for successful formation of the DNA metal-base pairs since a number of buffer components (e.g., tris(hydroxymethyl)-aminomethane, TRIS, and ethylenediamine tetraactetate, EDTA) is known to interfere with metal coordination by forming strong complexes with transition metal ions themselves. The metal-base pairs are finally assembled by adding the required amount of metal ions, which can be monitored by thermal denaturation experiments, UV- or CD-based titrations and mass spectrometry.
Figure 4

Synthesis of ligand-modified nucleosides and incorporation of artificial nucleosides into oligonucleotides by automated solid-phase synthesis using phosphoramidite chemistry. The key step in the synthesis of artificial nucleosides is the coupling of the nucleobase (or the artificial ligand) with the deoxyribose sugar moiety. The examples in Figure 3 show that either a C-N bond or a C-C bond has to be formed and several methods for either case have been reported in the respective publications. The choice of a suitable protecting group strategy for the ligand which is compatible with DNA synthesis but allows for a smooth deprotection of the artificial nucleobases after DNA synthesis in aqueous solution is mandatory. After DNA synthesis, the single strands are cleaved from the solid support, fully deprotected and dissolved in an aqueous buffer. The oligonucleotides may be purified by HPLC, gel electrophoresis, or other techniques, characterized by mass spectrometry and their concentration is estimated by UV spectroscopy (requiring knowledge of the molar extinction coefficient of the modified nucleosides). Reprinted from [19b] with permission from Elsevier; copyright 2010.

In a thermal denaturation experiment, also termed a “melting curve” experiment, the extinction at the absorption maximum of the nucleobases at 260 nm (which is higher for single strands than for the double strand, an effect called hyperchromicity) is plotted against the sample temperature (see Figure 6 below for an example). The transition point of the curve is defined as the melting point TM of the DNA duplex. It is dependent on the length, sequence, and concentration of the DNA as well as the nature and concentration of the additives (salt, buffer). Typical parameters used are: DNA concentration of 1–10 mM, buffer concentration 10 mM (pH around 7, but may depend on the requirements imposed by the kinds of ligand and metal ion), NaCl concentration 50–150 mM (NaClO4 was successfully employed instead in studies using Ag(I) as metal ion under investigation).
Figure 6

Melting curves of the duplexes 16 (a and c) and 17 (b and d). [16] = [17] = 2.0 μM in 10 mM sodium phosphate buffer, 50 mM NaCl (pH 7.0). (a) and (b): no Cu2+; (c) and (d): [Cu2+] = 2.0 μM. Reprinted from Ref. [48] with permission from ACS Publications; copyright 2002.

For the ease of interpretation, the stabilizing effect of the used transition metal ions on the DNA strands containing the ligands in oppositely arranged positions can be expressed as the difference ΔTM between the melting points of the duplex before and after addition of the respective metal ions. When duplex stabilizing effects in terms of ΔTM are compared for the different metal-base pairing systems described in the literature, care has to be taken because the melting temperature TM is not a thermodynamic value. Nevertheless, the fact that a TM value is relatively easy to determine and its descriptive character have made it a key parameter employed for discussing thermal duplex stability.

Deduction of the thermodynamic parameters ΔH and ΔS for describing the base pairing situation upon addition of metal ions requires a thoroughly conducted series of melting point experiments carried out with a number of samples of different DNA concentration and subsequent van’t Hoff analysis. Isothermal titration calorimetry (ITC) [47] may be used as a direct method to determine the thermodynamic para­meters associated with metal complexation, however, to the best of our knowledge this method has never been used by the respective community.

2 Selected Examples

2.1 The Hydroxypyridone Metal-Base Pair

3-Hydroxy-4-pyridone is a heterocyclic, bidentate chelate ligand capable of forming square-planar 2:1 complexes with metal ions such as Cu(II) and octahedral 3:1 complexes with metal ions such as Fe(III). A hydroxypyridone nucleoside H (9 in Figure 3; carrying a methyl group at the 2-position) was synthesized by attachment to the C1’ position of deoxyribose via its nitrogen atom. The synthesis and incorporation into oligonucleotides were reported by Tanaka and Shionoya et al. [48]. Subsequently, addition of Cu(II) to a duplex containing two oppositely arranged hydroxypyridone bases was shown to result in the formation of an H-Cu-H base pair inside the double helix (Figure 5) [68].
Figure 5

The hydro­xy­pyridone metal base pair H-Cu-H. Reprinted from [19b] with permission from Elsevier; copyright 2010.

It was found that the hydroxypyridones lead to a destabilization compared to AT or GC base pairs at the same position in the absence of transition metal ions. Upon addition of Cu(II) ions, however, the duplex underwent a stabilization of ΔTM = 13 K (in the given sequence context, see Figure 6) which was interpreted as formation of one metal base pair inside the DNA double strand [48]. Addition of the same amount of Cu(II) to a similar sequence containing an AT base pair instead of the hydroxypyridones did not result in any change of TM.

The same system was further characterized by UV-based titration studies, CD spectroscopy, and ESI mass spectrometry, all indicating the complexation of one Cu(II) ion inside the double strand. An X-ray structure of the monomeric H-Cu-H base pair 9 confirmed the square-planar coordination of the Cu(II) ion by two hydroxypyridone ligands via their four oxygen atoms [48]. Since this model complex is not embedded within the surrounding DNA structure, the ligands, however, were found to be coordinated in an anti configuration as opposed to the syn configuration that was anticipated for the H-Cu-H metal-base pair inside the DNA double helix by molecular modeling. Recently, the expected syn configuration of the hydroxypyridone ligands around the Cu(II) ion was confirmed by a crystal structure containing two H-Cu-H metal-base pairs 9 inside a modified duplex based on an acyclic three-carbon propylene glycol phosphodiester backbone [49].

In a related study by the Shionoya group aiming at the coordination of soft metal ions such as Pd(II) by ligand-modified nucleobases, two sulfur containing ligands based on hydroxypyridone H in which the two coordinating oxygen atoms were exchanged one by one for a sulfur atom were synthesized and tested for their metal coordinating abilities [50]. Although the formation of Pd(II) and Pt(II) metal-base pairs was successfully achieved using the monomeric nucleosides, the incorporation of these sulfur-containing nucleosides into DNA duplexes has not been reported yet. Section 3 will focus on the incorporation of up to five H-Cu-H metal-base pairs 9 in a way that Cu(II) ions are stacked on top of each other inside the double strands.

Depending on the offered kind of metal ion, the hydroxypyridone ligand is also able to form 3:1 octahedral complexes. The use of Fe(III) was recently shown to result in the formation of triple helical constructs (Figure 7) [51]. Mixing three equivalents of a tetrameric H4 sequence with four equivalents of Fe(III) ions resulted in the formation of a triple helical complex (H4)3Fe4 after two days at 85°C as shown by UV-based titration studies, CD measurements, and ESI mass spectrometric data.
Figure 7

Formation of triple helices. (a) From three H4 strands (consisting of hydroxypridone nucleotides exclusively) and four Fe(III) ions; (b) A metal-containing triple helix can also be formed upon addition of Ag(I) ions to a mixture of duplex [d(3’-T10PT10-5’)•d(3’-A10PA10-5’)] and single strand d(3’-T10PT10-5’). Reprinted from [19b] with permission from Elsevier; copyright 2010.

Noteworthy in the context of triple helix formation is also the earlier work of the Shionoya group using a pyridine ligand P inside the duplex [d(3’-T10PT10-5’)•d(3’-A10PA10-5’)] which was shown to bind another single strand d(3’-T10PT10-5’) upon addition of Ag(I) in fashion of a heterotriplex (P = pyridine in base pair 8, Figures 3 and 7b) [36]. Recently, an observation of parallel triplex stabilization by Ag(I) ions was reported by Jyo et al. [52].

2.2 The Salen Metal-Base Pair

The assembly of most metal-base pairs can be interpreted as a two-component process comprising the DNA duplex containing the preorganized ligands as one reaction partner and the metal ion as the second component.

The situation is different for the salen metal-base pair 15 (S) introduced by Clever, Colborn, and Carell [53]. In contrast to the hydroxypyridone system, the assembly of the salen metal-base pair inside the DNA double helix requires the addition of two components to the duplex containing a pair of salicylic aldehydes at positions facing each other. The first additive is ethylenediamine which reacts with both aldehyde groups of the ligand precursors (salicylic aldehyde) under elimination of two molecules of water to the well known salen ligand (Figure 8). Since the formation of such aromatic imines is reversible in water, an excess of ethylenediamine is added in order to drive the equilibrium to the side of the salen ligand. Subsequently a transition metal ion such as Cu(II), Mn(II) (oxidized to Mn(III) upon complexation), Fe(III), or vanadyl (VO2+) is added and the metal-base pair 15 is formed [53b].
Figure 8

(a) Formation of the salen ligand from perfectly preorganized salicylic aldehyde precursors inside the DNA double helix and subsequent coordination of the metal ion giving the salen-metal-base pair; (b) X-ray structure of the monomeric salen-Cu(II)-base pair and superposition with a Watson-Crick AT base pair. Reprinted from [19b] with permission from Elsevier; copyright 2010.

The structure of the metal-base pairing system was elucidated by X-ray analysis of a monomeric Cu(II) complex of the salen ligand that was formed from the free salicylic acid nucleoside and ethylenediamine [53b,69]. The structure confirms the square-planar coordination of the metal ion (going along with a slight propeller twist between the planes of the two aromatic rings). A superposition of the molecular structure of a natural, hydrogen-bonded Watson–Crick base pair with the salen Cu(II)-base pair is depicted in Figure 8b. The excellent structural agreement between the natural AT base pair and the salen metal-base pair indicated that the chosen molecular design should result in a very good fit of the salen metal-base pair inside the DNA double helix. This assumption could be supported by further experimental results such as CD measurements of salen metal-containing DNA strands [53a].

A comparative study was conducted comparing the ligand depicted in Figure 8 (glycosidic bond para to the carbonyl group) with an isomeric salicylic aldehyde precursor (glycosidic bond para to the hydroxyl group) inside the same sequence context [53b]. In full accordance with the results obtained from the X-ray crystallographic investigations, it was found that the former isomer indeed fits much better inside the DNA duplex compared with the metal-base pair based on the latter isomer. Metal-base pair formation using ethylenediamine and Cu(II) is such a favorable process, however, that even the second isomer is undergoing the reaction inside the DNA, presumably under distortion of the duplex structure.

The mechanistic difference of the metal-base pair formation with the salen system compared to the hydroxypyridone system (and other comparable ligands) could be deduced from the results of thermal denaturation experiments. The same sequence as in the work of Shionoya et al. [48] was used: [d(5’-CACATTASTGTTGTA-3’)·d(3’-GTGTAATSACAACAT-5’)]). Whereas the duplex containing the two salicylic aldehydes S has a melting temperature TM of 40°C (about 10 K destabilization compared to AT), the addition of ethylenediamine and Cu(II) resulted in the formation of the salen Cu(II)-base pair (15 in Figure 3) accompanied by an increase in TM to 82°C. This very large value for ΔTM of +42 K is the highest ever reported increase in melting temperature for a metal-base pairing system (Figure 9) [53a].
Figure 9

Comparison of the melting curves of the sequence d(5’-CACATTASTGTTGTA-3’)·d(3’-GTGTAATSACAACAT-5’) (1) without any additives (39.9°C, solid line); (2) with only ethylenediamine (45.5°C, dashed line); (3) with methylamine and Cu(II) (52.3°C, filled boxes); (4) with only Cu(II) (54.9°C, open boxes); and (5) with ethylenediamine and Cu(II) (82.4°C, crosses) (3 μM DNA, 150 mM NaCl, 10 mM CHES buffer). Reprinted from [19b] with permission from Elsevier; copyright 2010.

The reason for this large duplex stabilization value was elucidated by a series of melting point experiments using different combinations of the single components added to the DNA duplex. If only ethylenediamine was added, the melting point TM was raised by only 5 K, attributed to the formation of the imine cross-link which is reversible in the aqueous environment. If only Cu(II) was added to the duplex, TM was raised by 15 K, a value comparable to the examples of duplex stabilization reported for the other metal-base pairs. The high stabilization of ΔTM = +42 K was achieved only in case, both, ­ethylenediamine and Cu(II) ions, were added to the duplex [53a]. A reasonable explanation is that the formation of the salen system from the salicylic aldehydes and ethylenediamine provides an excellent coordination environment for the metal and the binding of the Cu(II) ion to the tetradentate chelate ligand results in a stabilization of the imine bonds towards hydrolytic cleavage. Both the action of the cross-linking agent ethylenediamine and the coordinated metal ion are responsible for the remarkable duplex stabilization because a metal-stabilized cross-link is formed that is of far higher strength than metal coordination alone (Figure 10). Supporting this hypothesis is further the observation that the addition of Cu(II) and methylamine (instead of ethylenediamine) resulted in a stabilization ΔTM of only 12 K. This value is comparable to the melting temperature obtained for the DNA sample containing Cu(II) alone [53a].
Figure 10

Cooperative assembly of the salen-metal-base pairs in DNA. Reprinted from [19b] with permission from Elsevier; copyright 2010.

In the case of the Cu(II)-containing system, the thermal de- and renaturing profiles (heating, cooling curves) are superimposable. The measurements of the same sequence containing ethylenediamine and Mn(III), however, reproducibly showed a strong hysteresis between the de- and renaturing profiles.

This behavior was explained with the thermal instability of the salen-Mn(III) complex when exposed to temperatures above TM for elongated times. In these experiments, the transition in the heating curve can be assigned to the metal-containing and thus higher melting duplex, whereas the cooling curve shows that the duplex re-hybridizes before reincorporation of the metal (expressed by a lower TM).

After the sample is allowed to spend some time at a temperature below TM, however, the metal is again fully incorporated giving rise to a denaturing profile superimposable with the preceding denaturing curve. This observation allowed the conclusion that the salen Cu(II)-base pair 15 is more stable than the corresponding salen Mn(III)-base pair even at high temperatures. This fact was supported also by the higher ΔTM caused by the addition of Cu(II) (+42 K) in comparison to the addition of Mn(III) (+28 K).

Another form of cooperative binding of Cu(II) ions was observed in the case of duplex d(5’-CACATTSSTGTT GTA-3’)·d(3’-GTGTAASSACAACAT-5’) containing two neighboring salen ligands [53c]. The addition of 0.5 equivalents of Cu(II) results in a 1:1 mixture of metal-free duplexes and duplexes containing two Cu(II) ions as observed by a thermal denaturation study (Figure 11). The binding of the first Cu(II) ion to the duplex was shown to enhance the affinity for binding of the second Cu(II) ion so that all available Cu(II) ions in the solution end up being incorporated pairwise. At any given ratio of Cu(II) ions to DNA duplexes (<2) the reaction mixture thus contains only two species: metal-free duplexes and duplexes containing two metal ions. This model also explains the observation of isosbestic points in the titration curves of all examined double strands containing two or more adjacent metal binding sites such as HH or SS (see Section 3).
Figure 11

Cooperative uptake of metal ions inside duplexes containing two neighboring salen ligands. Addition of 0.5 equivalents of Cu(II) results in a 1:1 mixture of metal-free duplexes and duplexes containing two Cu(II) ions. The observed hysteresis effect is characteristic for the melting of the duplex containing ethylenediamine but no Cu(II) ions [53c]. Reprinted from [19b] with permission from Elsevier; copyright 2010.

2.3 The Triazole Metal-Base Pair

The use of 5-membered nitrogen-containing heterocycles as artificial nucleobases with metal coordinating abilities such as imidazole, triazole, and tetrazole was systematically studied by Müller and coworkers [40,54]. In their systematic studies, the ability to coordinate metal ions such as Ag(I) and Hg(II) (in terms of stability constant and ligand-to-metal ratio) was examined prior to the incorporation of ligands such as 7 into DNA sequences and the results were taken as indicators for predicting the behavior of the same ligand-metal systems inside the DNA context [55]. In addition, DFT calculation and experimental determination of the pKa values of the monomeric nucleosides was applied to estimate the proper pH range that is suitable for the formation of metal-base pairs inside the oligonucleotide environment. Out of the examined 5-membered heterocycles, the incorporation of 1,2,4-triazole Z into DNA was chosen because of its ability to differentiate between Ag(I) and Hg(II) (Figure 12).
Figure 12

Assembly of the Ag(I)-mediated base pair of Müller et al. [54]. Reprinted from [19b] with permission from Elsevier; copyright 2010.

A highly interesting observation was made when the hybridization behavior of the palindromic sequence d(5’-A7Z3T7-3’) containing three neighboring 1,2,4-triazole ligands Z was examined in thermal de-/renaturation experiments (Figure 13) [54]. In the absence of metal ions, the strand forms a hairpin due to the destabilizing effect of the Z bases lacking any ability to engage in mutual hydrogen bonding. The loop region of this hairpin is formed by the central stretch of three Z ligands. Upon addition of Ag(I) ions, however, the oligonucleotides undergo a structural transition and metal binding between pairs of 1,2,4-triazole ligands Z leads to dimerization of the palindromic sequences resulting in a double strand containing three stacked Ag(I) ions. The evidence was brought by UV spectroscopic and MALDI-TOF mass spectrometric experiments. Furthermore it was found that the hairpin exhibits a concentration-independent melting temperature owing to its unimolecular melting behavior. In contrast, the melting curves of the sample containing Ag(I) ions showed a dependence on the oligonucleotide concentration which verifies the existence of a double strand. Subsequently, the proposed structural switching process was supported by fluorescence resonance energy transfer (FRET) experiments with dye-labeled strands of the same sequence and by the observation of an increase in molecular size in dynamic light scattering (DLS) experiments [54]. Recently, the groups around Müller and Sigel have determined the structure of a related palindromic sequence containing three consecutive imidazole-Ag(I)-base pairs 7 by NMR spectroscopy [63]. The results are discussed in the next Section.
Figure 13

Control of hairpin to duplex transition by metal addition [54]. Reprinted from [19b] with permission from Elsevier; copyright 2010.

3 Stacking and Mixing of Metals Inside DNA

After the fundamentals for the formation of metal-base pairs inside the DNA double helix using the discussed set of ligands and metal ions were elucidated, the focus in the field of metal-base pairing has shifted towards the realization of longer metal stacks inside DNA double strands [56]. Whereas the thorough examination of oligonucleotides containing only one metal-base pair is of great importance for determining basic parameters like binding constant and thermal duplex stabilization (as expressed by a ΔTM value), the study of duplexes containing a number of directly stacked metal ions is supposed to contribute to the goal of realizing new DNA-based materials with single molecule electrical conductivity. The first report on consecutive metal stacking inside DNA double strands was given by Kuklenyik and Marzilli who examined the stacking of three Hg(II) ions between consecutive TT mismatches inside the middle of a short oligonucleotide [d(5’-GCGCTTTGCGC-3’)]2 (Figure 14) [23b].
Figure 14

Examples of reported metal stacks inside DNA double helices. The sequences of stretches of natural Watson-Crick base pairs adjacent to the metal-base pair regions have been omitted for clarity. Reprinted from [19b] with permission from Elsevier; copyright 2010.

The extension to longer stretches of stacked metal ions, however, was not successful any more when using the higher homologous sequence [d(5’-GCGC­TTTTGCGC-3’)]2. Instead of duplex formation, the addition of Hg(II) ions to this strand resulted in the arrangement of a hairpin due to metal-mediated intrastrand crosslinking of the terminal T residues of the T4 stretch. In contrast to this, Müller and Sigel et al. recently presented conclusive evidence for the formation of RNA duplexes containing up to six consecutive U-Hg-U base pairs (5, X = H) inside RNA duplexes by performing NOESY-NMR experiments with 15N and 13C labeled oligonucleotides [57].

Using the hydroxypyridone base pair H, Tanaka, Shionoya et al. achieved the stacking of up to five Cu(II) ions inside a duplex formed from the artificial oligonucleotides [d(5’-GHnC-3’)]2 (n = 1 – 5) [58]. The exclusive formation of the double strand [d(5’-GH5C-3’)]2 containing five Cu(II) ions was indicated by UV spectroscopic titration experiments and mass spectrometry. Most interestingly, EPR experiments showed a ferromagnetic interaction between the stacked paramagnetic Cu(II) ions. The EPR experiment also allowed a rough determination of the distance between the metal-base pairs to be 3.7 ± 0.1 Å which is comparable to the distance between the base pairs in natural DNA [58]. A subsequent theoretical DFT study by Mallajosyula and Pati supported the ferromagnetic coupling in this system of stacked H-Cu-H units (albeit assuming a Cu-Cu distance of 3.2 Å with the neighboring metals bridged by two oxygen atoms of the ligands in a {Cu2O2} convex quadrangle structure) [59]. The same theoretical work proposed an antiferro­magnetic coupling for two neighboring Cu(II)-salen-base pairs S-Cu-S. Indeed this was found to be true by a recent EPR study conducted by Schiemann et al. using a DNA double strand containing two adjacent Cu(II)-salen-metal-base pairs [60]. Nakanishi et al. also contributed theoretical studies for both above-mentioned base pairing systems [61].

Use of the salen ligand S introduced in Section 2.2 allowed the incorporation of ten metal ions in a row inside a double strand [53c]. The sequences were designed to consist of non-palindromic GC stretches at their ends in order to achieve preorganization of the ten adjacent pairs of salicylic aldehydes [(5’-CGGCCS10CGCGC -3’)·(3’-GCCGGS10GCGCG-5’)]. The aforementioned differences in duplex stabilization of Cu(II) versus Mn(III) were found to play an important role in the selective and well-controlled formation of double strands containing more than one metal-base pair. Whereas the higher stability of the salen-Cu(II)-base pair with respect to the salen-Mn(III)-base pair is beneficial in systems containing only a single salen ligand, the situation in case of the longer stacks inside DNA follows an opposite trend. Due to the high stability of the salen-Cu(II)-base pair, misfolded, undefined cross-linked or overlapping sequences are anticipated to be formed as kinetic products short after addition of the metal salt to the salen-containing DNA strands. The result was the observation of a large distribution of signals in the ESI mass spectra around the expected value for the duplex containing ten neatly stacked salen-Cu(II) complexes. In contrast, the addition of Mn(III) resulted in the formation of a much cleaner metalated DNA product after the reaction mixture was allowed to equilibrate at room temperature for several days as observed from the high resolution ESI-FTICR mass spectra [53c]. This result was ascribed to the “self-healing” capabilities of the salen-Mn(III)-containing duplexes allowing initially formed kinetic products to transform into one thermodynamic product after some time. Besides mass spectrometry, UV-based titration experiments and CD spectroscopy were employed as analytical methods for the characterization of the multinuclear metal arrangements inside the DNA systems.

By combining both Ag(I) coordination and hydrogen bonding within one base pair consisting of one thymine T and one artificial 1-deazaadenine D, Polonius and Müller succeeded in the stacking of 19 Ag(I) ions inside a 20-mer sequence [62]. The D and A bases are thought to form a doubly hydrogen-bonded Hoogsteen base pair and the addition of Ag(I) ions leads to substitution of one of these hydrogen bonds as illustrated in Figure 14. Likewise to the considerations concerning the effect of the relative stabilities of the salen-Cu(II)- and Mn(III)-base pairs on the formation of neatly metalated duplexes, the rather low duplex stabilization observed for the D-Ag-T base pair was seen as an advantage here as well. Again, the error-free formation of long metal-containing double strands was thought to require structural equilibration to reach the thermodynamically most favored structure. In a joint work, Müller and Sigel et al. recently reported the solution structure determination of a DNA double strand containing three consecutive imidazole-Ag(I)-base pairs 7 by a detailed NMR spectroscopic study involving the direct measurement of the 1J(15N,107/109Ag) coupling constants between the ligands and the metal ions. According to these measurements, the duplex adopts a B-type conformation with only minor deviations in the region of the metal-base pairs [63].

Since stacking of several metal ions of the same kind inside the DNA duplex has now been realized in practice, next steps in the direction of using artificial DNA strands as molecular wires such as single molecule conductivity measurements have to be undertaken. From the viewpoint of information storage and processing with DNA-based systems, the expansion of the genetic code with artificial base pairs is another highly interesting field of research. Also here, metal-base pairing has contributed new ideas since most metal-base pairs offer a binding mode orthogonal to the natural Watson–Crick binding motifs.

Going further in this direction, even the mixing of two different metal ions inside DNA was shown to be possible. Therefore, two ligand-modified nucleosides with orthogonal coordination capabilities for two different kinds of metal ions were chosen and incorporated into artificial oligonucleotides in the form of mixed arrays [64].

The groups around Shionoya and Carell succeeded in combining two different types of ligands which were shown to be able to bind Cu(II) and Hg(II) ions sequence-specifically in the same duplex [65]. The system presented by Shionoya et al. comprised the hydroxypyridone ligand H known for its ability to coordinate Cu(II) and the pyridine ligand P capable of coordinating Hg(II) (Figure 15). The formation of a duplex containing one Hg(II) and two Cu(II) ions was shown by UV- and CD-spectroscopic titration experiments with consecutive addition of both metal ions to the oligonucleotide d(5’-GHPHC-3’) and subsequent characterization of the reaction products by ESI mass spectrometry [65].
Figure 15

A mixed metal array based on the hydroxypyridone H and pyridine P ligands. Reprinted from [19b] with permission from Elsevier; copyright 2010.

Carell et al. synthesized strands containing a predetermined sequence of five salen ligands S with the propensity to bind Cu(II) and five TT mismatches capable of binding Hg(II) (Figure 16). Also in this case, the incorporation of the anticipated number of Cu(II) and Hg(II) ions according to the programmed sequence was indicated by CD-spectroscopic titration studies and the products were indentified by ESI mass spectrometry [65].
Figure 16

A mixed metal array based on the salen S and thymine T ligands. Reprinted from [19b] with permission from Elsevier; copyright 2010.

Several different sequences of Cu(II) and Hg(II) ions could be aligned inside the DNA strands following a rational design and thus the binary pattern that was encoded into the oligonucleotide by the automated solid-phase synthesis could be translated into a determined array of metal ions [65]. In a similar manner, Ono et al. were recently able to incorporate T-Hg-T and C-Ag-C base pairs into the same double strand [66].

4 Conclusion and Future Prospects

The research field of metal-base pairing inside DNA has developed within 10 years from the first attempts to coordinate single metal ions by simple ligand-modified nucleosides to complex oligonucleotide systems such as metal-mediated triple helices and programmable multi-metal stacks [19]. From a structural point of view, the achieved duplex stabilization compared to unmodified oligonucleotides might find interesting applications in the field of DNA nanotechnology, for example by introducing metal-base pairs into DNA origami constructs. Furthermore, the switching of hybridization in the absence or presence of the required metal ions (or a change in their redox states) and the presented possibility to control structural interconversions (i.e., hairpin to duplex transitions) might add switchable and dynamic functions to DNA nanoscale architectures.

From another perspective, the precise positioning of transition metal ions in a three-dimensional space made up by a surrounding oligonucleotide framework could also be a new approach to mimic metalloenzymes with functions such as catalysis, charge separation, or selective guest recognition. The spin-spin interactions between neighboring paramagnetic metal ions such as Cu(II) inside DNA duplexes have already been shown to result in interesting magnetic phenomena. Combined with the potential charge transport capabilities, the newly discovered magnetic properties of the metal-containing oligonucleotides may become of great importance for the emerging field of spintronics. Also an application in future quantum computing devices is imaginable. Furthermore, the substitution of some or all natural base pairs by metal-base pairs may present great potential for the overall enhancement of charge conductivity through oligonucleotide-based materials and so contribute to the area of DNA-based molecular electronics. Further synthetic modification of the ligands and a broader choice of metal ions might be necessary to fine-tune the prospect electronic functions.

Upcoming work has to unambiguously establish the conductive properties of metal-containing DNA strands, e.g., by single molecule measurements in break junctions [16a], single electron transfer studies with charge injectors and traps attached to model duplexes [17] or spectroscopic methods that yield information about charge carrier availability [67].

Keeping in mind the various switching processes, even more sophisticated electronic functions beyond pure charge transport such as processing of logical inputs may be achieved using metal-modified DNA. The first examples going into this direction were introduced by Willner et al. by the application of orthogonal metal-base pairing (T-Hg-T vs. C-Ag-C) in combination with the optical properties of quantum dots as sensors for Hg(II) and Ag(I) ions and as logical AND as well as OR gates [28]. The obtained mixed-metal arrays serve as striking examples for the possibility to incorporate new base pairing principles orthogonal to and alongside with the natural Watson-Crick base pairs into DNA strands. All the discussed and many other works have by now generated a good insight into the factors leading to stable metal-base pairing (complex stability, geometrical prerequisites, compatibility with the natural base pairs) thereby paving the way for future application of artificial metal-base pairs in even more complex nanoscopic constructs.

Notes added in proof

Recently, Shionoya et al. [68] were able to prove the direct conductance of electrons through metal-containing DNA double strands in single-molecule experiments based on the attachment of DNA strands between the ends of single-walled carbon nanotubes. It was found that the conductivity along hydroxypyridone ligand-­containing DNA duplexes can be switched on and off in the presence or absence of metal ions such as Cu(II), respectively. Longer stretches of stacked metal base pairs resulted in enhanced conductivities.

Carell et al. [69] demonstrated recently that metal-salen base pairs can be formed inside DNA double strands by enzymatic DNA polymerization using salicylic ­aldehyde triphosphates, ethylenediamine as reversible cross-linker. A single-crystal X-ray structure of the salen-Cu(II)-containing DNA strand in complex with a fragment of DNA polymerase I from Bacillus stearothermophilus showed that the metal base pair fits perfectly inside the B-DNA-like helical structure.



atomic force microscopy


circular dichroism


2-(cyclohexylamino)ethanesulfonic acid


density functional theory




dynamic light scattering


2’-deoxyribonucleoside 5’-triphosphate




electron paramagnetic resonance


electro spray ionization


fluorescence resonance energy transfer


Fourier-transform ion cyclotron resonance


high performance liquid chromatography


isothermal titration calorimetry


matrix-assisted laser desorption/ionization time-of-flight


nuclear Overhauser effect spectroscopy


polymerase chain reaction




single nucleotide polymorphisms


melting temperature





G.H.C. thanks the “Fonds der Chemischen Industrie” and the “Deutsche Forschungsgemeinschaft” (IRTG 1422 – Metal Sites in Biomolecules) for generous support. This work was supported by grants-in-Aids from MEXT of Japan and the Global COE Program for Chemistry Innovation.


  1. 1.
    H. Lodish, A. Berk, S. L. Zipursky, P. Matzudaira, D. Baltimore, J. Darnell, Molecular Cell Biology, 6th edn, WH Freeman and Co, New York, 2007.Google Scholar
  2. 2.
    (a) S. Jäger, G. Rasched, H. Kornreich-Leshem, M. Engeser, O. Thum, M. Famulok, J. Am. Chem. Soc. 2005, 127, 15071; (b) P. M. E. Gramlich, C. T. Wirges, J. Gierlich, T. Carell, Org. Lett. 2008, 10, 249.Google Scholar
  3. 3.
    (a) F. A. Aldaye, A. Palmer, H. F. Sleiman, Science 2008, 321, 1795; (b) U. Feldkamp, C. M. Niemeyer, Angew. Chem. Int. Ed. 2006, 45, 1856; (c) M. Endo, H. Sugiyama, ChemBioChem 2009, 10, 2420; (d) F. C. Simmel, Angew. Chem. Int. Ed. 2008, 47, 5884; (e) F. A. Aldaye, P. K. Lo, P. Karam, C. K. Mclaughlin, G. Cosa, H. F. Sleiman, Nat. Nanotechnol. 2009, 4, 349; (f) A. Heckel, M. Famulok, Biochimie 2008, 90, 1096.Google Scholar
  4. 4.
    G. E. Moore, Electronics 1965, 38, 114.Google Scholar
  5. 5.
    (a) C. M. Niemeyer, Angew. Chem. Int. Ed. 2001, 40, 4128; (b) K. V. Gothelf, T. H. LaBean, Org. Biomol. Chem. 2005, 3, 4023; (c) J. Wengel, Org. Biomol. Chem. 2004, 2, 277; (d) J. J. Storhoff, C. A. Mirkin, Chem. Rev. 1999, 99, 1849; (e) T. Nguyen, A. Brewer, E. Stulz, Angew. Chem. Int. Ed. 2009, 48, 1974; (f) M. D. Sorensen, M. Petersen, J. Wengel, Chem. Comm. 2003, 2130; (g) D. J. Hurley, Y. Tor, J. Am. Chem. Soc. 1998, 120, 2194.Google Scholar
  6. 6.
    (a) F. Seela, Y. He, J. Org. Chem. 2003, 68, 367; (b) S. A. Benner, Acc. Chem. Res. 2004, 37, 784.Google Scholar
  7. 7.
    A. T. Krueger, E. T. Kool, Chem. Biol. 2009, 16, 242.CrossRefPubMedGoogle Scholar
  8. 8.
    A. Okamoto, K. Kanatani, I. Saito, J. Am. Chem. Soc. 2004, 126, 4820.CrossRefPubMedGoogle Scholar
  9. 9.
    (a) N. C. Seeman, Nature 2003, 421, 427; (b) N. C. Seeman, Int. J. Nanotechnol. 2005, 2, 348.Google Scholar
  10. 10.
    H. T. Maune, S. Han, R. D. Barish, M. Bockrath, W. A. Goddard III, P. W. K. Rothemund, E. Winfree, Nat. Nanotechnol. 2010, 5, 61.CrossRefPubMedGoogle Scholar
  11. 11.
    P. W. K. Rothemund, Nature 2006, 440, 297.CrossRefPubMedGoogle Scholar
  12. 12.
    H. Yan, Science 2004, 306, 2048.CrossRefPubMedGoogle Scholar
  13. 13.
    A. Somoza, Angew. Chem. Int. Ed. 2009, 48, 9406.CrossRefGoogle Scholar
  14. 14.
    S. M. Douglas, H. Dietz, T. Liedl, B. Högberg, F. Graf, W. M. Shih, Nature 2009, 459, 414.CrossRefPubMedGoogle Scholar
  15. 15.
    E. S. Andersen, M. Dong, M. M. Nielsen, K. Jahn, R. Subramani, W. Mamdouh, M. M. Golas, B. Sander, H. Stark, C. L. P. D. Oliveira, J. S. Pedersen, V. Birkedal, F. Besenbacher, K. V. Gothelf, J. Kjems, Nature 2009, 459, 73.CrossRefPubMedGoogle Scholar
  16. 16.
    (a) X. Guo, A. A. Gorodetsky, J. Hone, J. K. Barton, C. Nuckolls, Nat. Nanotechnol. 2008, 3, 163; (b) R. Mas-Balleste, O. Castillo, P. J. Sanz Miguel, D. Olea, J. Gomez-Herrero, F. Zamora, Eur. J. Inorg. Chem. 2009, 2885.Google Scholar
  17. 17.
    (a) C. J. Murphy, M. R. Arkin, T. Jenkins, N. D. Ghatlia, S. H. Bossmann, N. J. Turro, J. K. Barton, Science 1993, 262, 1025; (b) S. Breeger, M. von Meltzer, U. Hennecke, T. Carell, Chem. Eur. J. 2006, 12, 6469; (c) B. Elias, J. C. Genereux, J. K. Barton, Angew. Chem. Int. Ed. 2008, 47, 9067.Google Scholar
  18. 18.
    C. M. Niemeyer, M. Adler, Angew. Chem. Int. Ed. 2002, 41, 3779.CrossRefGoogle Scholar
  19. 19.
    (a) G. H. Clever, C. Kaul, T. Carell, Angew. Chem. Int. Ed. 2007, 46, 6226; (b) G. H. Clever, M. Shionoya, Coord. Chem. Rev. 2010 , 254, 2391.Google Scholar
  20. 20.
    (a) A. Rakitin, P. Aich, C. Papadopoulos, Y. Kobzar, A. S. Vedeneev, J. S. Lee, J. M. Xu, Phys. Rev. Lett. 2001, 86, 3670; (b) P. Aich, S. L. Labiuk, L. W. Tari, L. J. T. Delbaere, W. J. Roesler, K. J. Falk, R. P. Steer, J. S. Lee, J. Mol. Biol. 1999, 294, 477.Google Scholar
  21. 21.
    S. Nokhrin, M. Baru, J. S. Lee, Nanotechnology 2007, 18, 095205.CrossRefGoogle Scholar
  22. 22.
    (a) S. Katz, J. Am. Chem. Soc. 1952, 74, 2238; (b) S. Katz, Biochim. Biophys. Acta 1963, 68, 240.Google Scholar
  23. 23.
    (a) E. Buncel, C. Boone, H. Joly, R. Kumar, A. R. J. Norris, Inorg. Biochem. 1985, 25, 61; (b) Z. Kuklenyik, L. G. Marzilli, Inorg. Chem. 1996, 35, 5654.Google Scholar
  24. 24.
    (a) A. Ono, H. Togashi, Angew. Chem. Int. Ed. 2004, 43, 4300; (b) Y. Miyake, H. Togashi, M. Tashiro, H. Yamaguchi, S. Oda, M. Kudo, Y. Tanaka,Y. Kondo, R. Sawa, T. Fujimoto, T. Machinami, A. Ono, J. Am. Chem. Soc. 2006, 128, 2172; (c) Y. Tanaka, S. Oda, H. Yamaguchi, Y. Kondo, C. Kojima, A. Ono, J. Am. Chem. Soc. 2007, 129, 244.Google Scholar
  25. 25.
    S. Johannsen, S. Paulus, N. Düpre, J. Müller, R. K. O. Sigel, J. Inorg. Biochem. 2008, 102, 1141.CrossRefPubMedGoogle Scholar
  26. 26.
    A. Ono, S. Cao, H. Togashi, M. Tashiro, T. Fujimoto, T. Machinami, S. Oda, Y. Miyake, I. Okamoto, Y. Tanaka, Chem. Comm. 2008, 4825.Google Scholar
  27. 27.
    D. A. Megger, J. Müller, Nucleosides, Nucleotides and Nucleic Acids 2010, 29, 27.CrossRefGoogle Scholar
  28. 28.
    R. Freeman, T. Finder, I. Willner, Angew. Chem. Int. Ed. 2009, 42, 7818.CrossRefGoogle Scholar
  29. 29.
    E. Ennifar, P. Walter, P. Dumas, Nucleic Acids Res. 2003, 31, 2671.CrossRefPubMedGoogle Scholar
  30. 30.
    J. Joseph, G. B. Schuster, Org. Lett. 2007, 9, 1843.CrossRefPubMedGoogle Scholar
  31. 31.
    A. A. Voityuk, J. Phys. Chem. B 2006, 110, 21010.CrossRefPubMedGoogle Scholar
  32. 32.
    S.-P. Liu, S.-H. Weisbrod, Z. Tang, A. Marx, E. Scheer, A. Erbe, Angew. Chem. Int. Ed. 2010, 49, 3313.CrossRefGoogle Scholar
  33. 33.
    (a) A. Houlton, A. R. Pike, M. A. Galindo, B. R. Horrocks, Chem. Comm. 2009, 1797; (b) G. A. Burley, J. Gierlich, M. R. Mofid, H. Nir, S. Tal, Y. Eichen, T. Carell, J. Am. Chem. Soc. 2006, 128, 1398.Google Scholar
  34. 34.
    J. Müller, Metallomics 2010, 2, 318.CrossRefPubMedGoogle Scholar
  35. 35.
    K. Tanaka, M. Shionoya, J. Org. Chem. 1999, 64, 5002.CrossRefGoogle Scholar
  36. 36.
    K. Tanaka, Y. Yamada, M. Shionoya, J. Am. Chem. Soc. 2002, 124, 8802.CrossRefPubMedGoogle Scholar
  37. 37.
    (a) E. Meggers, P. L. Holland, W. B. Tolman, F. E. Romesberg, P. G. Schultz, J. Am. Chem. Soc. 2000, 122, 10714; (b) N. Zimmermann, E. Meggers, P. G. Schultz, Bioorg. Chem. 2004, 32, 13.Google Scholar
  38. 38.
    S. Atwell, E. Meggers, G. Spraggon, P. G. Schultz, J. Am. Chem. Soc. 2001, 123, 364.CrossRefGoogle Scholar
  39. 39.
    K. Tanaka, M. Shionoya, Coord. Chem. Rev. 2007, 251, 2732.CrossRefGoogle Scholar
  40. 40.
    J. Müller, Eur. J. Inorg. Chem. 2008, 3749.Google Scholar
  41. 41.
    I. Okamoto, K. Iwamoto, Y. Watanabe, Y. Miyake, A. Ono, Angew. Chem. Int. Ed. 2009, 48, 1648.CrossRefGoogle Scholar
  42. 42.
    N. Zimmermann, E. Meggers, P. G. Schultz, J. Am. Chem. Soc. 2002, 124, 13684.CrossRefPubMedGoogle Scholar
  43. 43.
    L. Zhang, E. Meggers, J. Am. Chem. Soc. 2005, 127, 74.CrossRefPubMedGoogle Scholar
  44. 44.
    H. Weizman, Y. Tor, J. Am. Chem. Soc. 2001, 123, 3375.CrossRefPubMedGoogle Scholar
  45. 45.
    (a) C. Switzer, S. Sinha, P. H. Kim, B. D. Heuberger, Angew. Chem. Int. Ed. 2005, 44, 1529; (b) C. Switzer, D. Shin, Chem. Commun. 2005, 1342.Google Scholar
  46. 46.
    M. J. Gait (Ed.), Oligonucleotide Synthesis: A Practical Approach, IRL Press, New York, 1990.Google Scholar
  47. 47.
    C. A. Schalley (Ed.), Analytical Methods in Supramolecular Chemistry, Wiley-VCH, Weinheim, 2007.Google Scholar
  48. 48.
    K. Tanaka, A. Tengeiji, T. Kato, N. Toyama, M. Shiro, M. Shionoya, J. Am. Chem. Soc. 2002, 124, 12494.CrossRefPubMedGoogle Scholar
  49. 49.
    (a) M. K. Schlegel, L.-O. Essen, E. Meggers, J. Am. Chem. Soc. 2008, 130, 8158; (b) M. K. Schlegel, L. Zhang, N. Pagano, E. Meggers, Org. Biomol. Chem. 2009, 7, 476.Google Scholar
  50. 50.
    Y. Takezawa, K. Tanaka, M. Yori, S. Tashiro, M. Shiro, M. Shionoya, J. Org. Chem. 2008, 73, 6092.CrossRefPubMedGoogle Scholar
  51. 51.
    Y. Takezawa, W. Maeda, K. Tanaka, M. Shionoya, Angew. Chem. Int. Ed. 2009, 48, 1081.CrossRefGoogle Scholar
  52. 52.
    T. Ihara, T. Ishii, N. Araki, A. W. Wilson, A. Jyo, J. Am. Chem. Soc. 2009, 131, 3826.CrossRefPubMedGoogle Scholar
  53. 53.
    (a) G. H. Clever, K. Polborn, T. Carell, Angew. Chem. Int. Ed. 2005, 44, 7204; (b) G. H. Clever, Y. Söltl, H. Burks, W. Spahl, T. Carell, Chem. Eur. J. 2006, 12, 8708; (c) G. H. Clever, T. Carell, Angew. Chem. Int. Ed. 2007, 46, 250.Google Scholar
  54. 54.
    D. Böhme, N. Düpre, D. A. Megger, J. Müller, Inorg. Chem. 2007, 46, 10144.CrossRefGoogle Scholar
  55. 55.
    (a) J. Müller, D. Böhme, N. Düpre, M. Mehring, F.-A. Polonius, J. Inorg. Biochem. 2007, 101, 470; (b) J. Müller, D. Böhme, P. Lax, M. Morell Cerda, M. Roitzsch, Chem. Eur. J. 2005, 11, 6246.Google Scholar
  56. 56.
    J. K. Klosterman, Y. Yamauchi, M. Fujita, Chem. Soc. Rev. 2009, 38, 1714.CrossRefPubMedGoogle Scholar
  57. 57.
    S. Johannsen, S. Paulus, N. Düpre, J. Müller, R. K. O. Sigel, J. Inorg. Biochem. 2008, 102, 1141.CrossRefPubMedGoogle Scholar
  58. 58.
    K. Tanaka, A. Tengeiji, T. Kato, N. Toyama, M. Shionoya, Science 2003, 299, 1212.CrossRefPubMedGoogle Scholar
  59. 59.
    S. S. Mallajosyula, S. K. Pati, Angew. Chem. Int. Ed. 2009, 48, 4977.CrossRefGoogle Scholar
  60. 60.
    G. H. Clever, S. J. Reitmeier, T. Carell, O. Schiemann, Angew. Chem. Int. Ed. 2010, 49, 4927.CrossRefGoogle Scholar
  61. 61.
    (a) Y. Nakanishi, Y. Kitagawa, Y. Shigeta, T. Saito, T. Matsui, H. Miyachi, T. Kawakami, M. Okumura, K. Yamaguchi, Polyhedron 2009, 28, 1945; (b) T. Matsui, H. Miyachi, Y. Nakanishi, Y. Shigeta, T. Sato, Y. Kitagawa, M. Okumura, K. Hirao, J. Phys. Chem. B. 2009, 113, 12790.Google Scholar
  62. 62.
    F.-A. Polonius, J. Müller, Angew. Chem. Int. Ed. 2007, 46, 5602.CrossRefGoogle Scholar
  63. 63.
    S. Johannsen, N. Megger, D. Böhme, R. K. O. Sigel, J. Müller, Nat. Chem. 2010, 2, 229.CrossRefPubMedGoogle Scholar
  64. 64.
    J. Müller, Nature 2006, 444, 698.CrossRefPubMedGoogle Scholar
  65. 65.
    K. Tanaka, G. H. Clever, Y. Takezawa, Y. Yamada, C. Kaul, M. Shionoya, T. Carell, Nat. Nanotechnol. 2006, 1, 190.CrossRefPubMedGoogle Scholar
  66. 66.
    K. Yanagida, N. Hamochi, K. Sasano, I. Okamoto, A. Ono, Nucleic Acids Symp. Ser. 2007, 51, 179.CrossRefGoogle Scholar
  67. 67.
    (a) K. Kawai, Y. Osakada, M. Fujitsuka, T. Majima, J. Phys. Chem. B 2008, 112, 2144; (b) R. Yamagami, K. Kobayashi, A. Saeki, S. Seki, S. Tagawa, J. Am. Chem. Soc. 2006, 128, 2212.Google Scholar
  68. 68.
    S. Liu, G. H. Clever, Y. Takezawa, M. Kaneko, K. Tanaka, X. Guo, M. Shionoya, Angew. Chem. Int. Ed. 2011, 50, 8886.Google Scholar
  69. 69.
    C. Kaul, M. Müller, M. Wagner, S. Schneider, T. Carell, Nat. Chem. 2011, in press, DOI: 10.1038/nchem.1117.Google Scholar

Copyright information

© Springer Science+Business Media B.V. 2012

Authors and Affiliations

  1. 1.Institute for Inorganic ChemistryGeorg-August University GöttingenGöttingenGermany
  2. 2.Department of Chemistry, Graduate School of ScienceThe University of TokyoTokyoJapan

Personalised recommendations