Ru–TAP complexes and DNA: from photo-induced electron ...

rsta.royalsocietypublishing.org

ReviewCite this article:Marcélis L, Moucheron C,Kirsch-De Mesmaeker A. 2013 Ru–TAPcomplexes and DNA: from photo-inducedelectron transfer to gene photo-silencing inliving cells. Phil Trans R Soc A 371: 20120131.http://dx.doi.org/10.1098/rsta.2012.0131

One contribution of 18 to a Discussion MeetingIssue ‘Photoactivatable metal complexes: fromtheory to applications in biotechnology andmedicine’.

Subject Areas:photochemistry, inorganic chemistry,medicinal chemistry

Keywords:ruthenium complexes, oligonucleotides, DNA,electron transfer, photochemistry,photo-cross-linking

Author for correspondence:Andrée Kirsch-De Mesmaekere-mail: [email protected]

Ru–TAP complexes and DNA:from photo-induced electrontransfer to genephoto-silencing in living cellsLionel Marcélis, Cécile Moucheron and

Andrée Kirsch-De Mesmaeker

Chimie Organique et Photochimie, Université librede Bruxelles, CP 160/08, 50 Avenue F.D. Roosevelt,1050 Bruxelles, Belgium

In this review, examples of applications of the photo-induced electron transfer (PET) process betweenphoto-oxidizing Ru–TAP (TAP = 1,4,5,8-tetraazaphen-anthrene) complexes and DNA or oligodeoxynucleo-tides (ODNs) are discussed. Applications using afree Ru–TAP complex (not chemically anchored toan ODN) are first considered. In this case, thePET gives rise to the production of an irreversibleadduct of the Ru complex on a guanine (G) base,with formation of a covalent bond. After absorptionof a second photon, this adduct can generate abi-adduct, whereby the same complex binds toa second G moiety. These bi-adduct formationsare responsible for photo-cross-linking between twostrands of a duplex, each containing a G base,or between two G moieties of a single strandsuch as a telomeric sequence, as demonstrated bypolyacrylamide gel electrophoresis analyses or massspectrometry. Scanning force microscopy also allowsthe detection of such photobridgings with plasmidDNA. Other applications, for example with Ru–ODN, i.e. ODN with chemically anchored Ru–TAPcomplexes, are also discussed. It is shown thatsuch Ru–ODN probes containing a G base intheir own sequences are capable of photo-cross-linking selectively with their targeted complementarysequences, and, in the absence of such targets,they self-photo-inhibit. Such processes are appliedsuccessfully in gene photo-silencing of humanpapillomavirus cancer cells.

2013 The Author(s) Published by the Royal Society. All rights reserved.

on January 10, 2015http://rsta.royalsocietypublishing.org/Downloaded from

http://crossmark.crossref.org/dialog/?doi=10.1098/rsta.2012.0131&domain=pdf&date_stamp=2013-06-17

mailto:[email protected]

http://rsta.royalsocietypublishing.org/

2

rsta.royalsocietypublishing.orgPhilTransRSocA371:20120131

......................................................

1. IntroductionSince the discovery of Pt complexes as anti-cancer drugs [1–5], many other metal complexeshave been studied in the presence of DNA with the aim of developing novel metallic drugswith a lesser toxicity than the Pt-based compounds for future applications [6,7]. Among them,Ru(II) and Ru(III) complexes [8–14] have been the subject of several research studies. By contrast,fewer studies have been reported until now on Ru compounds that are activated under visibleillumination. This possibility represents, of course, a clear advantage over classical drugs, becausethe therapeutic treatment can be triggered by simple illumination with optical fibres, at thetarget site, at a chosen time and with a determined duration. For example, Ru–arene compounds[12,15,16], in which indan ligand (or hydrindene) or chloride ions are photosubstituted bynucleotidic bases, have been examined recently by Sadler et al. With these compounds, the DNAphotodamaging effect is caused by formation of a new complex between the nucleotidic baseand the Ru ion, i.e. by production of a new coordination bond. This type of photoreaction iscompletely different from that of the Ru–TAP (TAP = 1,4,5,8-tetraazaphenanthrene) compounds[17–20]. It has been demonstrated in the past that, when these compounds contain at least twoTAP ligands, they are very oxidizing in their metal-to-ligand charge transfer (MLCT) triplet-excited state, which is able to extract an electron from guanine (G) bases [18]. This photo-inducedelectron transfer (PET) process generates a reduced Ru–TAP complex and oxidized G moiety (seethe reactions scheme in figure 1). This PET, which can be accompanied by a proton transfer, hasbeen examined kinetically in different time domains, from the pico- [21,22] to the microsecond[17,18] time scale. The rate of PET depends upon whether the G base is incorporated in mono- orpolynucleotides. Thus, if the G-donor is included in guanosine monophosphate (GMP), the PETprocess is diffusion controlled, because GMP, in solution, has first to diffuse towards the excitedRu–TAP complex before being oxidized [23,24]. If the G-donor belongs to a polynucleotide inwhich the Ru–TAP complex is intercalated, then the PET process takes place in a few hundredsof picoseconds. Readers interested in the details of these kinetics studies can refer to differentreviews [18,22,24]. In brief, after the PET primary process, as the reduced complex and oxidizedG are transient species, they react according to the scheme in figure 1. First, the back electrontransfer from the reduced complex to the oxidized G, which is thermodynamically favourable,can take place, the kinetics of which also depends on the incorporation of the G base in themononucleotide or in a polynucleotide. Unfortunately, in that case, the absorbed photon willbe unused. Obviously, this process cannot be avoided if there is no other reaction in competition.Actually, as detected by different methods, the reduced and oxidized transient species can alsoreact together and produce an adduct of the Ru complex on the G moiety, the structure of whichis shown in figure 2 [25]. It is important to stress that the occurrence of a covalent bond betweenthe amino group of the G base and the TAP ligand, in the ortho position of its non-chelatednitrogen atom, is characteristic of TAP. This adduct production originates probably from thebasicity of the radical ion (TAP·−) in the reduced complex that can be protonated at pH 7. Indeed,other metallic complexes, even those that are highly photo-oxidizing [26–31], producing a PETwith G bases, are not capable of generating an irreversible adduct of the metallic species on aG base. This constitutes a peculiarity of the Ru–TAP complexes and the basis of their potentialapplications.

In summary, these complexes exhibit a double advantage: their ability to generate a PETprocess and to form an adduct by absorption of one photon. Moreover, it was observed that,by absorption of a second photon by the photo-produced adduct, a second photoreaction takesplace with a second G base belonging to either an oligodeoxynucleotide (ODN) or DNA,which generates a second covalent addition of a G species to the complex; this leads to abi-adduct. Because bi-adducts can be obtained with simple non-derivatized and non-tetheredcomplexes (i.e. free complexes), we first consider this case in the following sections. We illustratethe interest and applications of production of (i) bi-adducts, whose formation requires twophotons and free complexes, and (ii) mono-adducts, whose formation requires one photon andtethered complexes.



3


......................................................

[Ru(TAP)2(L)]2+

+ hv [Ru(TAP)2(L)]2+*

[Ru(TAP)2(L)]2+*

+ G [Ru(TAP)(TAP•–)(L)]1+ + G•+

[Ru(TAP)(TAP•–)(L)]1+ + G•+ [Ru(TAP)

2(L)]2+

+ G adduct of G to the Ru complex

DNA cleavageif G in DNA

(3MLCT excited state)

(PET)

(BET)

Figure 1. Reactions scheme for the PET process with [Ru(TAP)2(L)]2+ complexes (L= a bidentate ligand such asphenanthroline), followed by different possible reactions competing with each other. In this scheme, the protonation ofthe reduced Ru–TAP complex and deprotonation of the G·+ moiety have been omitted. (L, phen or TAP, for example; PET,photo-induced electron transfer; BET, back electron transfer).

NH

N

N

O

NHN

O

H

HHHHOH

OP–O

O

O– NN

NN

Ru

N

N

N

N

N

N

N

N

Figure 2. Structure of a Ru–TAP photo-adduct as determined by nuclear magnetic resonance spectroscopy [25].

Ru

N

NN

N

N

NN

N

N

N

N N

NRuN

N

NN

NN

N

NN

NN

NN

N

(a) (b) (c)

Ru

X

NN

X

X

NN

X

N

N N

N NN

NN

Figure 3. Structure of (a) [Ru(HAT)2(phen)]2+, (b) [Ru(TAP)2(TPAC)]2+ (c) [Ru(phen/TAP)2(PHEHAT)]2+ (X= N for TAP andX= CH for phen).

2. Applications for bi-adduct formation with two absorbed photons and onefree Ru complex

(a) Detection of bi-adductsAs explained above, the conditions to produce a mono-adduct of a G base of a mono- orpolynucleotide to a Ru–TAP complex such as that shown in figure 2 have been fully investigatedpreviously [23,24]. The formation of bi-adducts was first discovered with another complex,[Ru(HAT)2phen]2+ (HAT = 1,4,5,8,9,12-hexaazatriphenylene) (figure 3a) [32]. This compoundbehaves in the same way as the Ru–TAP complexes, thus also leading to a PET process and



4


......................................................

adduct formation. However, the high-pressure liquid chromatography and mass analyses ofan illuminated solution of [Ru(HAT)2phen]2+ in the presence of GMP revealed, in addition tomono-adduct production, the occurrence of a bi-adduct. In electrospray mass spectrometry, themono-adduct is characterized by a mass representing the sum of the mass of the complex plusthat of GMP minus two H atoms [33,34]. Thus, quite logically, the mass of a bi-adduct shouldbe characterized by a total mass comprising: the mass of the complex, plus two GMP moleculesand minus four H atoms. This mass was, indeed, detected with the HAT complex [34], but nofurther structural information could be obtained owing to the difficulty of isolating the bi-adductin sufficient amounts. On the basis of these results, bi-adducts should also be formed with theRu–TAP complexes discussed earlier. However, as the separation of the Ru–TAP photoproductsafter illumination in the presence of the mononucleotide GMP was rather difficult, the easiestexperimental way and best proof of their formation was to go one step further, i.e. to useG-containing ODNs. Indeed, in that case, if there are G bases on the same strand or on differentstrands and that are not too far away from each other, irreversible photo-cross-linking should beproduced between these G units.

(b) Inter-strand photo-cross-linking of oligodeoxynucleotides via bi-adduct formationIn order to demonstrate unambiguously the formation of a bi-adduct via the occurrence ofphoto-cross-linking between two G-containing oligonucleotides, three types of photo-oxidizingRu complexes have been tested [35]: two with the TAP ligands [Ru(TAP)2phen]2+ and[Ru(TAP)2TPAC]2+ (TPAC = tetrapyridoacridine; figure 3b) and one with the HAT ligand[Ru(HAT)2phen]2+, because a bi-adduct was detected by mass spectrometry in these cases.Those three compounds, although able to produce a PET, interact differently with the ODNduplexes, as demonstrated by emission anisotropy and affinity measurements, as well as bymolecular modelling calculations. [Ru(TAP)2phen]2+ exhibits a non-intercalative binding modeand a rather moderate affinity for the DNA double helix (Kaff = 3.9 × 104 M−1 in the absenceof added NaCl) [36]. Its emission when bound to DNA is polarized to a similar extent to thatof [Ru(phen)3]2+, its homoleptic geometrical analogue, which does not intercalate. The secondcomplex, [Ru(TAP)2TPAC]2+, presents a behaviour characteristic of a metallo-intercalator suchas the well-known [Ru(bpy/phen/TAP)2dppz]2+ or [Ru(phen/TAP)2PHEHAT]2+ (PHEHAT =1,10-phenanthrolino[5,6-b]1,4,5,8,9,12-hexaazatriphenylene) (figure 3c) [21,37–45]. This TPACcomplex displays a large emission anisotropy upon DNA binding, which stems from theintercalation of the extended planar ligand TPAC within the stacking of bases. Its high affinityfor DNA (1.5 × 106 M−1 in the presence of 100 mM NaCl and 100 mM KCl) is of the same orderof magnitude as that for other metallo-intercalators, and the results of molecular modellingcalculations also show the intercalation of the complex within the double helix structure ofa 17-mer oligonucleotide. Its interacting ability with DNA is thus totally different from thatof [Ru(TAP)2phen]2+ and results from the presence of the third ancillary ligand. The thirdcomplex, [Ru(HAT)2phen]2+, was chosen for its previously demonstrated ability to form bi-adducts upon illumination. Its DNA binding affinity (0.6 × 104 M−1 in the presence of 50 mMNaCl or 2.4–3.8 × 105 M−1 without added salt) [32] indicates that it interacts less efficientlywith DNA than complexes containing a classical intercalating ligand such as dppz or PHEHAT.Viscosity measurements with CT-DNA indicate that [Ru(HAT)2phen]2+ induces an increasein viscosity that is slightly lower than that with ethidium bromide [34], and its retention ofemission polarization anisotropy in the presence of DNA is significantly lower than that for aclassical metallo-intercalator [35]. These data indicate a potential intercalation of one HAT ligand,despite the presence of the second HAT, that is much more hindering than the TAP ligands of[Ru(TAP)2TPAC]2+. These observations are in agreement with molecular modelling calculationswhich show that, in contrast to [Ru(TAP)2TPAC]2+, the extended HAT ligand is only slightlyintercalated, the second HAT giving rise to a strong steric hindrance with the DNA backbone andpreventing a deep insertion of the first HAT within the base-pair stack.



5


......................................................

mono-adduct formationdouble-stranded DNA

bi-adduct formation

first photon absorptionhvhv

second photon absorptionG

G

GG

G G

free Ru(II) complex

Figure 4. Schematic of photo-cross-linking between two G-containing complementary ODNs by a free Ru–TAP complex.

The bi-adduct formation with these three mononuclear heteroleptic complexes wasstudied in the presence of complementary G-containing 17-mer ODNs by polyacrylamidegel electrophoresis (PAGE) experiments under denaturing conditions [35]. Thus, if the 5′end of one of the 17-mer ODNs is 32P-labelled and if photo-cross-linking is formed uponillumination of the complex in the presence of the two complementary strands, a retarded bandcorresponding to a duplex should be observed on the gel, even under denaturing conditions,as the two complementary strands are linked by covalent bonds upon photo-cross-linkingformation (schematic representation in figure 4). By contrast, in the absence of photo-cross-linking, only spots migrating as single strands should be visualized. [Ru(TAP)2phen]2+ and[Ru(HAT)2phen]2+ exhibit a similar behaviour in the presence of different 17-mer duplexes (withzero to four base pairs between the two G bases): in both cases, photo-adducts corresponding tothe mono-adduct and to the photo-cross-linking are observed [35]. The mono-adduct is observedupon illumination of both complexes in the presence of each duplex, whereas, by contrast, thephoto-cross-linking appears only for the duplexes in which there is only zero or one base pairbetween the two G bases. The percentage of mono-adduct reaches 40 per cent within the first15 min of illumination, then increases more slowly and even starts decreasing. In parallel, thepercentage of photo-cross-linking increases slowly up to almost 10 per cent. This indicates thatthe mono-adduct has first to be formed before photoreacting to generate the photo-cross-linking.This photo-cross-linking is observed, however, only when zero or one base pair is present betweenthe G bases. For the other examined 17-mer ODNs, the G–G distances are larger, so that thedistortion of the duplex, which would be needed for photo-cross-linking, is also important. Theseexperimental results are in agreement with molecular modelling.

By contrast, for [Ru(TAP)2TPAC]2+, which intercalates and binds strongly to DNA, the photo-cross-linking cannot be detected under conditions of identical loading levels of the duplexby the complex, but appears only by overloading the double helix [35]. Thus, when theintercalative binding mode predominates, no photo-cross-linking is formed. This indicates thatthe intercalation, which induces poor mobility of the complex inside the double helix, preventsaccess to the nitrogen groups of the G bases for the second adduct formation. By contrast, whenthe ODN is overloaded, i.e. when [Ru(TAP)2TPAC]2+ binds to DNA not only by intercalation butalso by adsorption, the photo-cross-linking can be formed.

These results show that, even if complexes fulfil the criteria of a high photo-oxidizing powerand good photoreactivity towards G moieties, they also have to fulfil interaction geometry andmobility criteria to allow the double anchoring process. Of course, the yield of this photo-cross-linking is rather low, because the double anchoring process requires two successive absorptionsof a photon by the same complex.

In order to improve photo-cross-linking, an a priori more suitable complex, the dinuclear[(TAP)2Ru(TPAC)Ru(TAP)2]4+, has been studied [46]. This complex has a photo-oxidizing powersimilar to that of the mononuclear [Ru(TAP)2TPAC]2+ but, because of the presence of the two Ruand the four TAP ligands, the probability of forming more than one photo-adduct on the samecomplex should increase, which should consequently improve the photo-cross-linking efficiency.This complex also exhibits large emission anisotropy upon DNA binding, which is attributednot to an intercalative process, as shown with its mononuclear analogue [Ru(TAP)2TPAC]2+,but rather to a groove binding the large dinuclear moiety, as confirmed by molecular modellingcalculations. [(TAP)2Ru(TPAC)Ru(TAP)2]4+ exhibits a high binding affinity (2.9 × 106 M−1 in the



6


......................................................

Table 1. Percentages of mono-adduct and photo-cross-linking formed with [(TAP)2Ru(TPAC)Ru(TAP)2]4+ and different 17-merduplex sequences; ds, double strand, followed by the number of base pairs between the two Gs (0–6).

sequence distance (Å) photo-adduct (%) photo-cross-linking (%)

ds0 5′TTT TCG TTT TAA ATT TA3′ 3.9 68 6. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3′AAA AGC AAA ATT TAA AT5′. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ds1 5′TTT TTT TCT GAA ATT TA3′

7.6 56 12. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3′AAA AAA AGA CTT TAA AT5′

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ds2 5′TTT TTT TCT AGA ATT TA3′

11.3 47 11. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3′AAA AAA AGA TCT TAA AT5′

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ds3 5′TTT TTT CTT AGA ATT TA3′

14.8 42 25. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3′AAA AAA GAA TCT TAA AT5′

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ds4 5′TTT TTC TTT AGA ATT TA3′

18.1 43 27. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3′AAA AAG AAA TCT TAA AT5′

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ds5 5′TTT TCT TTT AGA ATT TA3′

21.3 47 17. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3′AAA AGA AAA TCT TAA AT5′

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ds6 5′TTT TCT TTT AAG ATT TA3′

24.4 43 17. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3′AAA AGA AAA TTC TAA AT5′

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

presence of 100 mM NaCl and 100 mM KCl), which is mainly attributed to important electrostaticinteractions between the negatively charged phosphate backbone and the 4+ charge of thecomplex. The photo-cross-linking ability of this binuclear complex has been evaluated with aseries of 17-mer duplexes differing by the number of base pairs between the G bases on eachstrand (from zero to six base pairs). As shown in table 1, the percentage of photo-cross-linking ismaximum for the duplexes in which the two Gs are separated by three and four base pairs. Thisis in full agreement with the comparison of the distance between the two Ru reactive centres, andthe distances separating the two G bases in the duplex (figure 5). These results also support theconclusion that each Ru(TAP)2 moiety behaves independently. With the two sequences giving thehighest efficiency for the double-anchoring process, the evolution of the percentage of photo-cross-linking has been measured as a function of the illumination time (table 2). After 4 minillumination, the mono-adduct already represents 30 per cent and the photo-cross-linking only 6per cent of the total radioactivity. With increasing illumination times, the amount of mono-adductfirst increases, then stabilizes, whereas the amount of photo-cross-linking increases steadily up to37 per cent after 30 min illumination. The kinetics analysis indicates that the mono-adduct hasto be formed first, and, afterwards, absorption of light by this mono-adduct gives rise to thedouble anchoring. As expected, this binuclear complex is revealed to be much more efficient atphoto-cross-linking than the mononuclear complexes studied previously.

(c) Intra-strand photo-cross-linking of oligodeoxynucleotides via bi-adduct formationThis photo-cross-linking process would be of great interest in the case of G-rich DNA sequences,which are known to form highly structured architectures called G-quadruplexes [47,48]. Thosestructures play a crucial role in several biological processes such as telomer stabilizationor oncogene activation [49–51]. The folded G-quadruplex conformation thus represents aninteresting pharmaceutical target, and it would be particularly interesting to stabilize and‘freeze’ these architectures via photobridging of two or more Gs. For that purpose, the highphotoreactivity of [(TAP)2Ru(TPAC)Ru(TAP)2]4+ has been examined with the human telomeric



7


......................................................

N

N

N N

N

N

NN

N

N

NN

N

N

NN

N

N

NN

N

TA

CG

T T

T T

A

A A A

CG

Figure 5. Schematic of the inter-strand (duplex ODN) photo-cross-linking with the dinuclear complex[Ru(TAP)2(TPAC)Ru(TAP)2]4+. (Online version in colour.)

Table 2. Evolution of the percentage of mono-adduct and photo-cross-linking with [(TAP)2Ru(TPAC)Ru(TAP)2]4+ as a functionof the illumination time.

irradiation photo-cross-duplex time (min) unreacted (%) photo-adduct (%) linking (%)

ds3 0 100 0 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4 64 30 6. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

8 53 36 11. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

12 45 38 17. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

16 38 38 24. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

30 28 35 37. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ds4 0 100 0 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4 65 30 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

8 52 36 12. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

12 43 38 19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

16 37 38 25. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

30 25 38 37. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

sequence d(T2AG3)4 using gel electrophoresis under denaturing conditions. As shown infigure 6, after only 3 min illumination, 75–80% of the starting sample has reacted (more than90% after 6 min). This photoreaction is extremely rapid compared with the photo-cross-linkingdescribed earlier, with either the mononuclear or dinuclear complexes. For comparison, under



8


......................................................

photo-adduct (a)

–

+

1 2 3 4 5

photo-adduct (b)

photo-adduct (c)

photo-adduct (d )

photo-adduct (e)

photo-adduct ( f )

Figure 6. PAGE results of human telomeric sequence (T2AG3)4 and [Ru(TAP)2(TPAC)Ru(TAP)2]4+. Lane 1, non-illuminatedsequence; lane 2, non-illuminated sequence in the presence of the complex; lanes 3–5, illuminated with an He/Cd laser at442 nm in the presence of the complex for 3, 6 and 9 min, respectively. For lane 3, the approximate percentage for eachphoto-adduct is: (a) 28%, (b) 26%, (c) 16%, (e) 13%, (d,f ) not measured.

1.06 nm

1.31 nm

Figure7. Computer simulationof the intra-strand (human telomeric sequence) photo-cross-linkingwith thedinuclear complex[Ru(TAP)2(TPAC)Ru(TAP)2]4+. (Online version in colour.)

the same conditions, only 18 per cent of the starting sequence has reacted in the presence of[Ru(TAP)2TPAC]2+. Furthermore, the binuclear complex, in contrast to its mononuclear analogue,leads to the observation of five or six new bands, tentatively attributed to the addition of one tosix complexes on the telomeric sequence. As a function of increasing illumination times, the fastermigrating adduct, giving the most intense spot after 3 min illumination, decreases in favour ofthe less migrating and finally non-migrating bands. The occurrence of covalent adducts with thistelomeric sequence has been proven by nano-electrospray ionization analyses on the dialysedsolution after illumination [46]. These analyses, complicated by the presence of a mixture ofproducts, clearly showed the presence of photo-adducts corresponding to the addition of morethan one complex on the telomeric sequence and indicated most probably the appearance ofphoto-cross-linked species inside the telomer. The molecular modelling calculations (figure 7)of the quadruplex photo-cross-linked by [(TAP)2Ru(TPAC)Ru(TAP)2]4+ also supports theseconclusions [46].



9


......................................................

covalently closedcircular (CCC)

open circular (OC)

single-stranded

cleavage

double-stranded

cleavage

linear

Figure 8. Schematic of the transformation of a coiled circular plasmid (or a covalently closed circular plasmid, CCC form) intoan open circular plasmid (OC form, single-strand nick) and linear form (double-strand nick).

Such intramolecular photo-cross-linking would thus prevent the telomer unfolding and,consequently, block the telomerase activity, which could lead to very interesting therapeuticapplications.

(d) Effects of bi-adduct formation with 500-base-pair linear DNA fragments and plasmidDNA as revealed by scanning force microscopy

As outlined earlier, the photo-addition of two G units on one Ru–TAP complex allows thecross-linking of either two ODNs or inside a folded single-strand ODN at the sites of the G bases;this is detected by mass spectrometry or gel electrophoresis. Very interestingly, the effects of theseirreversible photobridgings of DNA can also be examined by scanning force microscopy (SFM)[52]. For this, longer DNA sequences have to be used; therefore, 500-base-pair DNA restrictionfragments and plasmid DNA pUC19 have been examined. It was shown a long time ago thatthe illumination of Ru–TAP complexes in the presence of plasmids (figure 8) induces cleavages,as revealed by the occurrence of open circular and linear forms of the plasmid from agaroseelectrophoresis analysis [53] and from atomic force microscopy detections [54]. It is well knownthat the coiled circular plasmid is transformed, just by one single nick, into its open circular form.It has been demonstrated that these photocleavages by the Ru–TAP complexes originate also fromthe PET process from a G base [23]. On the basis of these results, we should thus complete thephotochemical reactions scheme shown in figure 1 by the addition of a supplementary reaction,subsequent to the PET process, i.e. a step corresponding to DNA cleavages. (The cleavages mightalso appear with the ODNs mentioned earlier, but the resulting mononucleotides or smaller ODNfragments migrate too quickly to be quantified on the gel. Therefore, it is not possible to comparedirectly the amount of cleavages with plasmids and with ODNs.)

Although cleavages can easily be observed by SFM [55], it was not known whether theformation of adducts (mono- and/or bi-adducts) or the consequences of their formationcould also be observed by SFM of the plasmids? For that purpose, Ru–TAP complexesthat intercalate in DNA, i.e. which exhibit a high affinity for DNA, have been chosen andtheir illumination effects directly compared with those of similar intercalating complexesbut which do not induce a PET process. We thus compared the results of illuminationwith [Ru(TAP)2(PHEHAT)]2+ and [Ru(phen)2(PHEHAT)]2+ (figure 3c). Indeed, both complexesintercalate their PHEHAT ligand into the stacking of bases [44,45]. However, as discussed earlier,the complex [Ru(TAP)2(PHEHAT)]2+, having two TAP ligands, gives rise to a PET process andadduct formation, which is not the case for [Ru(phen)2(PHEHAT)]2+. With regard to the plasmids,a non-illuminated or non-enzymatically treated supercoiled plasmid can be detected by SFMand exhibits, for example as shown in figure 9, a precise number of nodes that can be easilycounted from the SFM image. By contrast, an enzymatically treated plasmid, with one nick,exhibits, in most cases, two nodes in its relaxed conformation. The approach used previouslyby Jiang et al. [55] was adopted in this SFM study for characterizing the plasmids. Thus, thenumber of plasmids with a certain number of nodes was counted from the SFM images andplotted as a function of the number of nodes. The resulting curve is illustrated in figure 10,



10


......................................................

n = 7

n = 5

200 nm

Figure 9. An example of the number of nodes of non-illuminated plasmids as detected by SFM.

0.35

0.30

0.25

frac

tion

of m

olec

ules

0.20

0.15

0.10

0.05

00 2 4

nodes per molecule6 8 10 12

Figure 10. The population distribution of plasmids (or fraction of molecules) plotted against the number of nodes for samplesof untreated plasmids (filled squares) and for samples of enzymatically treated plasmids (filled circles, single-strand nick ofplasmids).

and represents the fraction of plasmids or the plasmid population with different numbers ofnodes. From figure 10, the highest population for untreated plasmid molecules has a nodenumber of seven, and that for the enzymatically nicked plasmid molecules in open circularconformation (single nick) has a node number of two. We have to note that, for the SFM images,poly-L-lysine (PLL)-coated mica plates were used because they give quite reproducible results.Furthermore, PLL, which is positively charged in neutral solution, allows kinetic trapping ofthe DNA objects, which gives rise to a projection from three to two dimensions on the micasurface. This explains the presence of overlapping (nodes) in the OC form. After illuminationin the presence of [Ru(phen)2(PHEHAT)]2+, the plasmid’s population changes and is describedby a bimodal distribution of seven and two nodes (figure 11a up to 1 min illumination andfigure 11b for longer illumination times up to 60 min). This is not surprising, because excited[Ru(phen)2(PHEHAT)]2+ sensitizes oxygen into singlet oxygen, which is at the origin of thesecleavages. Most of the Ru(II) complexes or organic intercalators photosensitize oxygen withoccurrence of such cleavages. By contrast, as shown in figure 12a (up to 1 min illumination),the illumination of [Ru(TAP)2(PHEHAT)]2+ induces a decrease in the seven-node population notonly in favour of a two-node population but also in favour of a new population with five nodes.For longer irradiation times (up to 60 min), this five-node population shifts back to a populationwith a higher number of nodes (figure 12b). By comparison with the irradiation effect of the TAP



11


......................................................

0.35(a) (b)

0 s5 s15 s30 s60 s

1 min5 min15 min30 min60 min

0.30

0.25

0.20

0.15

0.10

0.05

00 2 4 6

nodes per molecule

Ru[(phen)2PHEHAT]2+

PHASE IRu[(phen)2PHEHAT]2+

PHASE II

frac

tion

of m

olec

ules

nodes per molecule

8 10 12 0 2 4 6 8 10 12

Figure 11. Thepopulationdistributionof plasmids after different times of illumination at 445 nmwith [Ru(phen)2(PHEHAT)]2+,(a) up to 60 s illumination, (b) from 1 to 60 min illumination. The direction of the arrows indicates the decrease or increase ofthe peaks.

0.35(a) (b)

0 s5 s15 s30 s60 s

1 min5 min15 min30 min60 min

0.30

0.25

0.20

0.15

0.10

0.05

00 2 4 6

nodes per molecule

Ru[(TAP)2PHEHAT]2+

PHASE IRu[(TAP)2PHEHAT]2+

PHASE II

frac

tion

of m

olec

ules

nodes per molecule8 10 12 0 2 4 6 8 10 12

Figure 12. The population distribution of plasmids after different times of illumination at 455 nmwith [Ru(TAP)2(PHEHAT)]2+,(a) up to 60 s illumination, (b) from 1 to 60 min illumination. The direction of the arrows indicates the decrease or increase ofthe peaks and the shift of the maximum (horizontal).

complex on DNA restriction fragments, indicating an increase in the DNA resistance to bending[52], the five-node population could be attributed to the formation of mono-adducts as detectedabove with the ODNs by PAGE. The new shift of population to higher numbers of nodes forlonger irradiation times could be attributed to irreversible photo-cross-linking between two Gbases of the same plasmid molecule, as explained earlier with the ODNs. In order to test thishypothesis, the illuminated plasmid in the presence of the Ru complex was treated afterwardswith the EcoRI restriction enzyme, which resulted in the images shown in figure 13b,c, in whichone or even several nodes, assigned to photo-cross-linking, are visible. This is quite remarkablewhen compared with the result of the same enzymatic treatment but with a non-illuminatedplasmid sample. Indeed, in that case, one nick gives rise to the formation of completely relaxedlinearized plasmids (figure 13a).

In conclusion, SFM allowed the visualization, for the first time, of the photo-cross-linkingbetween DNA strands, and in this case between different portions of the same plasmid. Themechanical and topological effects of these mono- and bi-adduct formations under illumination,as observed by SFM, should have important implications at the cellular level.



12


......................................................

(a) (b)

(c)

100 nm

100 nm200 nm

Figure 13. (a)Non-illuminatedplasmids treatedwith theenzyme EcoRI (restrictionenzyme for adouble-strandnick at a specificsite). (b,c) Illuminated plasmids with the [Ru(TAP)2(PHEHAT)]2+ complex, treated afterwards with the enzyme EcoRI.

3. Application ofmono-adduct formation owing to one-photon absorption anda tethered Ru complex

For several decades, gene therapy with small ODNs has been the focus of much research andwas believed as the ‘gold strategy’ for the future in the treatment of genetic diseases and cancers.Although some developments have been noted in the literature with promising results [56–60],gene therapy still encounters some limitations owing to different problems such as drug delivery,toxicity of the ODNs and specific targeting.

In this context, the use of ODNs bearing a photoreactive Ru–TAP complex could improve thisstrategy by allowing spatial and temporal activation of the drug and producing longer lastingeffects. The latest developments in this field based on Ru–TAP complexes by using, in this case,one photon absorbed by an ODN tethered complex are discussed below.

(a) Demonstration of photo-cross-linking of Ru–ODN probes with ODNG targetsThe conjugation of ODNs to Ru–TAP complexes is obtained either by using a modified thymineor uracil residue bearing the complex [61,62] or by introducing into the Ru complex one ligandmodified by an oxy-amine group. In this latter case, a highly specific oxime bond is thenachieved in the presence of an ODN bearing an aldehyde function [63,64]. The first experimentswere conducted on Ru–ODN conjugates containing no guanine bases, in order to avoid anyintramolecular photoreaction (the conjugates containing a G base, Ru–ODNG, will be consideredbelow in §3c). The luminescence behaviour of these first Ru–ODNs was analysed by singlephoton-counting experiments, in the absence and the presence of a complementary ODN target.The study of the lifetime of the 3MLCT excited states confirmed that the tethering of an ODNsingle strand to the complex does not introduce any quenching but, on the contrary, induces theoccurrence of a lifetime component longer than the luminescence lifetime of the free complex inaqueous solution. This is attributed to the protection of the excited complex by the single-strandedODN. By contrast, after hybridization of this Ru–ODN with a G-containing complementary targetsequence, luminescence quenching indicates that a photo-induced process effectively takes place.

In order to detect possible photoproducts formed by this quenching, PAGE experiments underdenaturing conditions were performed with Ru–ODN probes (figure 14) in the absence and thepresence of their complementary ODN strands, in which several guanine bases are present andshould, after hybridization, be in close vicinity to the Ru compound of the probe. After irradiation



13


......................................................

–

+

photo-cross-linking

unreacted Ru–ODN

1 2 1 2 1 2 1 2 1 2 1 2 1 2

Figure 14. PAGE experiments with seven different 14-mer duplexes containing radiolabelled Ru–ODNs (black) and their G-containing complementary targets (light grey); lanes 1, non-illuminated; lanes 2, illuminated for 15 min with an He/Cd laser at442 nm.

hν

photo-cross-linking

inhibition of the replication at the site of the photo-cross-linking

no inhibition in dark

Ru–ODNDNA template

Ru–ODN expelled

Figure 15. Schematic of the photo-induced inhibition of the replicative enzyme (DNA polymerase; grey crescent), functioningfrom a DNA template, starting at the primer, and owing to the photo-cross-linking of a Ru–ODN probe (black). In the dark, theenzyme works normally and the Ru–ODN probe is thus expelled.

of this duplex, a retarded band appears on the gel (figure 14), which is attributed to the two ODNstrands (probe and target) that are irreversibly linked. This process is called ‘photo-cross-linking’,as the two strands become covalently linked via the Ru complex after illumination and thereforemigrate together in spite of the denaturing conditions [61].

In the context of a biological application, this photo-cross-linking should alter the replicationprocess of the genetic material. In vitro experiments within this framework have demonstratedthat this photo-process with a Ru–ODN probe leads to complete inhibition of the enzyme DNApolymerase [65]. Indeed, although this enzyme in the absence of or even in the presence ofthe complementary Ru–ODN probe but in the dark is able to produce DNA from the DNAtemplate, after illumination, the elongation of the primer is completely blocked at the site of thephoto-cross-linking (figure 15). Moreover, photo-cross-linking has been proved to be resistant tothe action of exonuclease III [62]. The demonstration of these inhibitions of DNA enzymes in invitro systems represents a crucial milestone that highlights the interest in Ru–ODN conjugates aspotential photoactivatable drugs in gene therapy.



14


......................................................

NN

N N

NN

N

N

NN

N

N

N

NN

N

N

NN

N

NN N

N(a)

(b)

Figure 16. Schematic of the different interaction for (a) [Ru(TAP)2(dppz′)]2+ (anchoring via dppz) and (b)[Ru(TAP)(TAP′)(dppz)]2+ (anchoring via TAP) in a duplex ODN.

(b) Parameters influencing the photo-cross-linkingThe photo-cross-linking process depends on several parameters, such as the site of anchoring ofthe Ru complex, its distance from the G bases and, of course, the number of G units, as well as thenature of the ligands chelated to the ruthenium centre.

As the photoproduct is formed after the PET, factors influencing the electron transfer alterthe yield of photo-cross-linking. As demonstrated in the literature [66–68], stacking of the Gbases decreases the G ionization potential, therefore increasing the rate and efficiency of thePET. This was indeed detected by an increased quenching of luminescence of the excited Ru–TAP complexes. The distance between the Ru complex anchored on the probe sequence andthe guanine in the targeted ODN is also of critical importance for both the PET and the photo-cross-linking. The maximum distance reachable by the complex depends on its linker and on thedirection followed by the tethered complex (3′ or 5′ direction) in the duplex configuration [65]. Forexample, with a previously examined complex and linker [62], a photo-adduct could be formedwith a G up to the fourth base towards the 3′ side and the fifth base towards the 5′ side.

The mode of interaction of the tethered Ru complex with the ODN duplex also influencesthe PET and photo-cross-linking processes, and sometimes in a counterintuitive way. In otherwords, the manner in which the complex [Ru(TAP)2dppz]2+ is attached to the ODN probe,either by the TAP or by the dppz ligand, influences the behaviour of the system differentlyunder illumination [69] (figure 16). In the first case, the complex can intercalate into the stackof bases, and the electron transfer is then enhanced. In the second case, tethering via the dppzligand prevents its intercalation. Under this condition, the PET process decreases; however,the photo-cross-linking yield is higher than with the intercalated complex. This can be easilyexplained. Indeed, in order to give rise to photo-cross-linking, a good relative geometry ofthe reduced complex and oxidized guanine is needed to produce the covalent bond. As theintercalated compound has reduced mobility, it cannot easily adopt a suitable geometry forthe photo-adduct. Unexpectedly, intercalation has, thus, opposite effects on the PET and thephoto-cross-linking yield.

(c) Photoreactivity of a Ru–ODNG and the ‘seppuku’ processAs mentioned earlier, the first ODN sequences that have been tethered to the photoreactiveRu–TAP complexes were free of G bases, in order to prevent intramolecular photoreaction. Thepresence of a G in the Ru–probe (named Ru–ODNG) could indeed inhibit the photoreactivityof the attached Ru complex with the G bases of the target, which should be in competitionwith the G moiety of the probe. Intra-strand photo-adduct formation should indeed be possible



15


......................................................

photo-cross-linkingproducts

‘seppuku’ adducts

Ru(II)–ODNGG

G

G G

lane

irradiation time

–

+

1 2 3 4 5 6 8 9 12 13 14

0' 5' 10' 15' 30'

11107

60' 0' 5' 10' 15' 30' 60' 30'30'

Figure 17. PAGE experiments with radiolabelled Ru–ODNG conjugates irradiated with an He/Cd laser at 442 nm for increasingirradiation times, in the absence (lanes 1–6) and in the presence (lanes 7–12) of the complementary targets, and in the presenceof two different non-complementary targets (lanes 13 and 14).

G

GGG

GRu–ODN probe

(a) complementary target

photo-cross-linking self-inhibition : ‘seppuku’ product

GG G

hν hν

(b) without target or in the presence of a non- complementary target

Figure 18. Schematic of the competition between the ‘seppuku’ process (b) and the photo-cross-linkingwith a complementarysequence (a).

with a single-stranded ODN, which is more flexible than an ODN duplex. Nevertheless, itturned out that such Ru–ODNG conjugates exhibit a very interesting behaviour [70]. Whenthey are irradiated in the absence of any complementary strand in solution, the luminescenceof the tethered Ru compound is quenched, indicating that a PET occurs between the G baseand the excited complex of the same strand. This PET can be followed by a photo-adductformation, because an additional product is effectively detected with PAGE (figure 17). Thisphotoproduct has an increased electrophoretic mobility when compared with the starting Ru–ODNG, which can be explained by the fact that the intramolecular photoproduct adopts somekind of cyclic shape. Conversely, after irradiation of a Ru–ODNG in the presence of one equivalentof complementary ODN, thus a G-containing target, this cyclic photoproduct is no longerdetected and photo-cross-linking between the two strands of the duplex is obtained, similarto that discussed in §3b. Even more interestingly, when irradiated in the presence of a non-complementary scrambled ODN target, the only photoproduct detected with PAGE correspondsto the intramolecular photoreaction, i.e. a cyclic ODN conformation (figure 17). In conclusion,in the absence of its specific target, the Ru–ODNG self-inhibits (figure 18), preventing in this



16


......................................................

way any side photoreaction with a non-targeted G base. This self-inhibition process triggeredunder illumination was named the ‘seppuku’ process, in reference to the ritual suicides that werecommitted when warriors in Ancient Japan could not fulfil their duties.

The ‘seppuku’ process represents a real improvement in the strategy based on the Ru–ODNs,because it allows very accurate targeting of a given gene sequence without side photoreactions. Asmentioned earlier, in order to ensure an effective photo-cross-linking with the correct sequence,a guanine base in the target sequence has to be present close to the attached complex in theduplex form (see above). In addition to this condition, several other parameters have beenexamined that also influence the competition between the intermolecular photo-cross-linking andthe intramolecular photo-inhibition (seppuku process). The general trend in this competition canbe quantified by the ratio of the percentages of these two processes. Thus, several parametershave been shown to influence this ratio, including salt concentration, temperature, the presenceof an intercalating ligand (see above), and the position and number of mismatches in a non-complementary target [71]. Concerning the influence of salt and temperature, conditions closeto those inside skin cells (33◦C and 50 mM NaCl) increase tremendously the specificity of theRu probe for its targeted sequence. Moreover, the experimental results and molecular modellingperformed on duplexes show that the most distant targeted guanine reachable by the attachedcomplex corresponds to the sixth position on the target strand from the anchored complex on theprobe sequence. The influence of the position of the G base inside the ODN probe is currentlyunder investigation.

All these results highlight the interest in Ru–ODNG conjugates as attractive candidates for thephotoactivated antisense strategy. To confirm their possible use in gene silencing, experimentswere conducted with Ru-ODNGs designed on the basis of the rules established from these studies.

(d) Ru–ODNs in gene-silencing applicationsOn the basis of the above results on these ‘intelligent photoactivatable Ru–ODNs’, their promisingefficacy in gene silencing has been tested with living cells. The choice of a gene to be silencedwas guided by some literature results on oligonucleotide-based therapeutic strategies forhuman papillomavirus (HPV) infections, investigated by the antisense and RNA interferencetechnologies [72,73]. Although some promising results were obtained by the authors, theypresented some limitations, for example, in the duration of the inhibition effects of the geneexpression. The genes responsible for the development of this type of cancer thus constituteexcellent targets for testing the photoactivity of the Ru–ODNs and seppuku strategy. Cancerscaused by HPV represent 5.2 per cent of the world’s cancer burden and include carcinomas of thecervix, penis, vulva/vagina, anus, mouth and oropharynx [74]. The majority of HPV-associatedcancer cases are squamous cell carcinomas, for which treatment under illumination would thusbe convenient.

For HPV-associated cancers, thus representing an attractive model for testing gene-specifictherapy with Ru–ODN probes, it was shown that knockdown of HPV oncogenes E6 and/or E7should result in cancer cell senescence or apoptosis. The E6 protein forms complexes with thep53 protein; this association inhibits its function as an apoptosis regulator, resulting in tumourprogression [75]. Consequently, the role of gene E6 in HPV-mediated carcinogenesis may beto suppress p53-mediated apoptosis [76]. The E6 oncogene is thus an ideal target for targetedgene-silencing therapy in HPV+ cervical cancer. HPV+ SiHa cancer cells containing the gene E6have thus been treated with photoreactive Ru–ODNs to test whether they behave as good Ru–antisense oligonucleotides (Ru-ASOs) using commercial oligofectamine as the transfecting agent[77]. The Ru complex tethered to the 3′ end of the ODN corresponds to [Ru(TAP)2phen]2+ andthe ODN sequence has been chosen so as to correspond to a sequence that is complementaryto a 21-base-pair sequence of E6 and could thus behave as an ASO (sequence (a) of figure 19).Moreover, the ASO sequence had to respond to the criteria elaborated from the earlier-describedresults, i.e. two C bases in positions 3 and 4 of the attachment site of the Ru–ASO complex[62,65], so that the two Gs of the complementary target gene (sequence (b) of figure 19) would



17


......................................................

Ru

3'ATC CAC ATA ATT GAC AGT TTT5'

5'TAG GTG TAT TAA CTG TCA AAA3'

Ru

3'TTT TTT TAG TTA AAT TTA5'

(a)(b)

(c)

Figure 19. The ODNs used with living cells for testing the photo-gene silencing in HPV+ SiHa cancer cells and tested in vitro inPAGE experiments. The Ru–ASO (a), the targeted complementary sequence for the E6 gene (b) and the Ru–nc conjugate (c).

(a)

(b)

(c)

(d)

(e)

( f )

Figure 20. Confocal microscopy: effect of the Ru–ASO treatment under illumination (2.5 h blue light radiation source at380–480 nm, 4 × 24 W intense blue bulbs) on the restoration of production of protein p53, stained by specific p53 antibody(green). (a,b) Untreated cells non-illuminated and illuminated, respectively; (c,d) cells treated by Ru–nc non-illuminated andilluminated, respectively; and (e,f ) cells treated by Ru–ASO non-illuminated and illuminated, respectively. (Online version incolour.)

be close to the photoreactive Ru complex, and one G would be more or less in the middle of theASO sequence so that the seppuku process would be possible. Another Ru–ASO has also beentested and corresponds to a non-complementary sequence (Ru–nc) (sequence (c) of figure 19).When these Ru–ASO and Ru–nc sequences were illuminated as single strands or in the presenceof the 21-mer synthesized complementary target sequence, and analysed afterwards by PAGEexperiments under denaturing conditions, the results were as expected: thus, the seppuku processwith the single-stranded Ru–ASO or with Ru–nc was observed, and no seppuku process wasobserved but photo-cross-linking occurred with the complementary strand.

The treatment of the SiHa cells with the Ru–ASO, followed by visible illumination, induced agrowth inhibition (measured by a metabolic Alamar Blue assay) of 50–60%, 24 h post illumination,when compared with the same samples not subjected to illumination. Very interestingly, the Ru–nc induced only 10–20% growth inhibition under the same conditions of treatment. Moreover,with Ru–ASO, a remarkable decrease of E6 protein expression (more than 60%, measured bywestern blot) was detected 24 h post illumination, whereas with Ru–nc a limited decrease of10–25% was observed. The reduction of E6 protein expression was also able to restore p53expression (measured by confocal microscopy with cells stained for p53 using a specific antibody;figure 20), whereas for the cells treated with Ru–nc, no restoration in p53 expression was observed.



18


......................................................

(a) (b)

(c) (d)

Figure 21. Antiproliferative effect of the Ru–ASO treatment under illumination (2.5 h blue light radiation source at380–480 nm, 4 × 24 W intense blue bulbs) on the organotypic HPV+ cell culture. Antigen used as a marker (Ki-67immunohistochemical staining). (a,b) Cells treated by Ru–nc non-illuminated and illuminated, respectively; (c,d) cells treatedby Ru–ASO non-illuminated and illuminated, respectively. (Online version in colour.)

Moreover, the Ru–ASO treatment under illumination induced antiproliferative effects not only inmonolayer cultures but also in a three-dimensional culture (i.e. in an in vivo environment), namelyorganotypic cultures (figure 21). In agreement with the data on monolayers, Ru–ASO induced,after illumination, a 52 per cent decrease in proliferation of the cells in the organotypic cultures,compared with 12 per cent growth inhibition with Ru–nc (Ki-67 antigen used as a marker ofproliferation).

In conclusion, the Ru–ODNG conjugates able to photo-inhibit themselves in the absence oftarget or with a non-complementary target, prepared and studied under illumination by differenttechniques and under different experimental conditions, turn out to be not only very efficient butalso selective when tested as photoactive Ru–ASO in gene-silencing therapy of HPV+ cervicalcancer cells. This is quite remarkable in spite of the fact that HPV16 E6/E7 target sequences havebeen shown to be not well suited for antisense inhibitors [76].

4. ConclusionThis review discussed the example of a fundamental study on the synthesis, characterization,photochemistry and photophysics of a special series of Ru–TAP complexes, which led tointeresting applications not only for using these compounds as photochemical tools in DNAtechnology and/or in cellular biology but also for the development of possible futurephotoactivated drugs for selective gene silencing of cancerous cells.

The whole photoactivity is triggered by a simple PET process that was spectroscopicallyanalysed under pulsed laser irradiation, which also allowed kinetic analysis of the forwardand back electron transfer. If this latter process were 100 per cent efficient, of course, theapplications outlined in this review would not have been possible. Further studies are required



19


......................................................

into developing processes that make the back electron transfer negligible and that favour adductformation. This is not that simple, because, as illustrated by the abovementioned differentstudies, this depends on numerous factors such as the affinity and geometry of interaction of thephotoactive Ru complex with DNA. As shown by the results with the Ru–ODNs, depending onthe system, the criteria that should be fulfilled in designing efficient Ru complexes is sometimescounterintuitive. Indeed, one could think a priori that an intercalating complex with a high affinityfor DNA would be more favourable. However, in the case of a complex tethered to an ODN, theintercalation handicaps the formation of a mono-adduct.

The discovery of bi-adduct formation for which two photons have to be used per bi-adductproduction from a free (non-attached) complex could lead to interesting selective or even specificphotoactive molecular tools in DNA technology. Thus, special Ru–TAP complexes, mononuclear,dinuclear or even polynuclear compounds could be tailored to the length separating the G basesthat one would wish to be photo-cross-linked. However, in the case of bi-adduct formation, themechanism of production is still unknown, except for the fact that a photon has to be absorbed bythe mono-adduct that is formed and that, of course, a second G base has to be in the vicinity of thismono-adduct. The mechanism is specially intriguing because the isolated mono-adduct exhibitsno or very weak emission [19]. Kinetic studies under femto-, pico- and nanosecond pulsed laserillumination of the isolated mono-adduct are presently in progress for clarifying the differentprocesses taking place.

The use of Ru–ODNG as Ru–ASO, as demonstrated in the case of inhibition under illuminationof the expression of the gene E6 in HPV+ cancerous cells, is especially interesting. Indeed, it opensthe way to diverse applications of gene silencing corresponding to different types of cancer orillnesses. The main advantage, as shown by the first results, would be the specificity or selectivityof the action, not only because it is triggered exclusively under illumination but also because onlyone gene is targeted. This selectivity should probably be attributed to the ‘seppuku’ effect of theseRu–ASOG conjugates that should lead to fewer off-target effects.

The authors thank the Fonds National pour la Recherche Scientifique (FNRS), the Walloon Region (WALEOproject) and the IAP (Interuniversities Attraction Pool) in Belgium for supporting most of this research. L.M.thanks the FNRS for a fellowship (‘aspirant’ at the FNRS).

References1. Lippert B (ed.) 1999 Cisplatin: chemistry and biochemistry of a leading anticancer drug. Weinheim,

Germany: Wiley-VCH.2. Lippert B. 2007 Platinum nucleobase chemistry. In Progress in inorganic chemistry, vol. 37 (ed.

SJ Lippard). Hoboken, NJ: Wiley.3. Pinto HM, Schornagel JH (eds). 1996 Platinum and other metal coordination compounds in cancer

chemotherapy 2. New York, NY: Plenum.4. Galanski M. 2007 Anticancer platinum complexes. State of the art and future prospects. Anti-

Cancer Agents Med. Chem. 7, 1–2. (doi:10.2174/187152007779314035)5. Szaciłowski K, Macyk W, Drzewiecka-Matuszek A, Brindell M, Stochel G. 2005 Bioinorganic

photochemistry: frontiers and mechanisms. Chem. Rev. 105, 2647–2694. (doi:10.1021/cr030707e)

6. Schatzschneider U. 2010 Photoactivated biological activity of transition-metal complexes. Eur.J. Inorg. Chem. 2010, 1451–1467. (doi:10.1002/ejic.201000003)

7. Schatzschneider U, Niesel J, Ott I, Gust R, Alborzinia H, Wölfl S. 2008 Cellular uptake,cytotoxicity, and metabolic profiling of human cancer cells treated with ruthenium(II)polypyridyl complexes [Ru(bpy)2(NN)]Cl2 with NN = bpy, phen, dpq, dppz, and dppn.ChemMedChem 3, 1104–1109. (doi:10.1002/cmdc.200800039)

8. Ronconi L, Sadler PJ. 2007 Using coordination chemistry to design new medicines. Coord.Chem. Rev. 251, 1633–1648. (doi:10.1016/j.ccr.2006.11.017)

9. Hartinger CG, Zorbas-Seifried S, Jakupec MA, Kynast B, Zorbas H, Keppler BK. 2006From bench to bedside: preclinical and early clinical development of the anticancer agentindazolium trans-[tetrachlorobis(1H-indazole)ruthenate(III)] (KP1019 or FFC14A). J. Inorg.Biochem. 100, 891–904. (doi:10.1016/j.jinorgbio.2006.02.013)


http://dx.doi.org/doi:10.2174/187152007779314035

http://dx.doi.org/doi:10.1021/cr030707e

http://dx.doi.org/doi:10.1021/cr030707e

http://dx.doi.org/doi:10.1002/ejic.201000003

http://dx.doi.org/doi:10.1002/cmdc.200800039

http://dx.doi.org/doi:10.1016/j.ccr.2006.11.017

http://dx.doi.org/doi:10.1016/j.jinorgbio.2006.02.013


20


......................................................

10. Keppler BK (ed.) 1993 Metal complexes in cancer chemotherapy. Weinheim, Germany: VCHPublishers.

11. Jakupec MA, Galanski M, Arion VB, Hartinger CG, Keppler BK. 2008 Antitumour metalcompounds: more than theme and variations. Dalton Trans. 2008, 183–194. (doi:10.1039/B712656P)

12. Hayton TW, Legzdins P, Sharp WB. 2002 Coordination and organometallic chemistry of metal-NO complexes. Chem. Rev. 102, 935–992. (doi:10.1021/cr000074t)

13. Groessl M, Reisner E, Hartinger CG, Eichinger R, Semenova O, Timerbaev AR, Jakupec MA,Arion VB, Keppler BK. 2007 Structure–activity relationships for NAMI-A-type complexes(HL)[trans-RuCl4L(S-dmso)ruthenate(III)] (L = imidazole, indazole, 1,2,4-triazole, 4-amino-1,2,4-triazole, and 1-methyl-1,2,4-triazole): aquation, redox properties, protein binding, andantiproliferative activity. J. Med. Chem. 50, 2185–2193. (doi:10.1021/jm061081y)

14. Alessio E, Mestroni G, Bergamo A, Sava G. 2004 Ruthenium antimetastatic agents. Curr. Top.Med. Chem. 4, 1525–1535.

15. Yan YK, Melchart M, Habtemariam A, Sadler PJ. 2005 Organometallic chemistry, biologyand medicine: ruthenium arene anticancer complexes. Chem. Commun. 2005, 4764–4776.(doi:10.1039/B508531B)

16. Magennis SW, Habtemariam A, Novakova O, Henry JB, Meier S, Parsons S, OswaldIDH, Brabec V, Sadler PJ. 2007 Dual triggering of DNA binding and fluorescence viaphotoactivation of a dinuclear ruthenium(II) arene complex. Inorg. Chem. 46, 5059–5068.(doi:10.1021/ic062111q)

17. Ghesquière J, Le Gac S, Marcélis L, Moucheron C, Kirsch-De Mesmaeker A. 2012 Whatdoes the future hold for photo-oxidizing Ru(II) complexes with polyazaaromatic ligands inmedicinal chemistry? Curr. Top. Med. Chem. 12, 185–196. (doi:10.2174/156802612799079008)

18. Marcélis L, Ghesquière J, Garnir K, Kirsch–De Mesmaeker A, Moucheron C. 2012 Photo-oxidizing Ru(II) complexes and light: targeting biomolecules via photoadditions. Coord. Chem.Rev. 256, 1569–1582. (doi:10.1016/j.ccr.2012.02.012)

19. Jacquet L, Kelly JM, Kirsch-De Mesmaeker A. 1997 Photoaddition of [Ru(tap)2)(bpy)]2+ toDNA: a new mode of covalent attachment of metal complexes to duplex DNA. J. Am. Chem.Soc. 119, 11 763–11 768.

20. Moucheron C. 2009 From cisplatin to photoreactive Ru complexes: targeting DNA forbiomedical applications. New J. Chem. 33, 235–245. (doi:10.1039/b817016a)

21. Elias B et al. 2008 Photooxidation of guanine by a ruthenium dipyridophenazine complexintercalated in a double-stranded polynucleotide monitored directly by picosecond visibleand infrared transient absorption spectroscopy. Chem. Eur. J. 14, 369–375. (doi:10.1002/chem.200700564)

22. Smith JA, George MW, Kelly JM. 2011 Transient spectroscopy of dipyridophenazine metalcomplexes which undergo photo-induced electron transfer with DNA. Coord. Chem. Rev. 255,2666–2675. (doi:10.1016/j.ccr.2011.04.007)

23. Lecomte J-P, Kirsch-De Mesmaeker A, Kelly JM, Tossi AB, Görner H. 1992 Photo-induced electron transfer from nucleotides to ruthenium tris-1,4,5,8-tetraazaphenanthrene:model for photosensitized DNA oxidation. Photochem. Photobiol. 55, 681–689. (doi:10.1111/j.1751-1097.1992.tb08511.x)

24. Elias B, Kirsch-De Mesmaeker A. 2006 Photo-reduction of polyazaaromatic Ru(II)complexes by biomolecules and possible applications. Coord. Chem. Rev. 250, 1627–1641.(doi:10.1016/j.ccr.2005.11.011).

25. Jacquet L, Kelly JM, Kirsch-De Mesmaeker A. 1995 Photoadduct between tris(1,4,5,8-tetraazaphenanthrene)ruthenium(II) and guanosine monophosphate: a model for a new modeof covalent binding of metal complexes to DNA. J. Chem. Soc. Chem. Commun. 9, 913–914.(doi:10.1039/C39950000913)

26. Elias B, Shao F, Barton JK. 2008 Charge migration along the DNA duplex: hole versus electrontransport. J. Am. Chem. Soc. 130, 1152–1153. (doi:10.1021/ja710358p)

27. Shao F, Elias B, Lu W, Barton JK. 2007 Synthesis and characterization of iridium(III)cyclometalated complexes with oligonucleotides: insights into redox reactions with DNA.Inorg. Chem. 46, 10 187–10 199. (doi:10.1021/ic7014012)

28. Shao F, Barton JK. 2007 Long-range electron and hole transport through DNA with tetheredcyclometalated iridium(III) complexes. J. Am. Chem. Soc. 129, 14 733–14 738. (doi:10.1021/ja0752437)


http://dx.doi.org/doi:10.1039/B712656P

http://dx.doi.org/doi:10.1039/B712656P

http://dx.doi.org/doi:10.1021/cr000074t

http://dx.doi.org/doi:10.1021/jm061081y

http://dx.doi.org/doi:10.1039/B508531B

http://dx.doi.org/doi:10.1021/ic062111q



http://dx.doi.org/doi:10.1039/b817016a

http://dx.doi.org/doi:10.1002/chem.200700564



http://dx.doi.org/doi:10.1111/j.1751-1097.1992.tb08511.x

http://dx.doi.org/doi:10.1111/j.1751-1097.1992.tb08511.x


http://dx.doi.org/doi:10.1039/C39950000913

http://dx.doi.org/doi:10.1021/ja710358p

http://dx.doi.org/doi:10.1021/ic7014012

http://dx.doi.org/doi:10.1021/ja0752437

http://dx.doi.org/doi:10.1021/ja0752437


21


......................................................

29. Brunner J, Barton JK. 2006 Targeting DNA mismatches with rhodium intercalatorsfunctionalized with a cell-penetrating peptide. Biochemistry 45, 12 295–12 302. (doi:10.1021/bi061198o)

30. Williams TT, Dohno C, Stemp EDA, Barton JK. 2004 Effects of the photooxidant on DNA-mediated charge transport. J. Am. Chem. Soc. 126, 8148–8158. (doi:10.1021/ja049869y)

31. Delaney S, Barton JK. 2003 Charge transport in DNA duplex/quadruplex conjugates.Biochemistry 42, 14 159–14 165. (doi:10.1021/bi0351965)

32. Blasius R, Moucheron C, Kirsch-De Mesmaeker A. 2004 Photoadducts of metallic compoundswith nucleic acids: role played by the photoelectron transfer process and by the TAP and HATligand in the Ru complexes. Eur. J. Inorg. Chem. 2004, 3971–3979. (doi:10.1002/ejic.200400489)

33. Jacquet L, Kelly JM, Kirsch-De Mesmaeker A. 1999 Formation of a covalently-linkedbimetallic compound upon irradiation of tris(1,4,5,8-tetraazaphenanthrene) ruthenium(II) inthe presence of 5′-guanosine-monophosphate. Inorg. Chem. Commun. 2, 135–138. (doi:10.1016/S1387-7003(99)00030-1)

34. Blasius R, Nierengarten H, Constant MLJ, Defrancq E, Dumy P, Dorsselaer AV, MoucheronC, Kirsch-De Mesmaeker A. 2005 Photoreaction of [Ru(hat)2phen]2+ with guanosine-5′-monophosphate and DNA: formation of new types of photoadducts. Chem. Eur. J. 11,1507–1517. (doi:10.1002/chem.200400591)

35. Ghizdavu L et al. 2009 Oxidizing Ru(II) complexes as irreversible and specific photo-cross-linking agents of oligonucleotide duplexes. Inorg. Chem. 48, 10 988–10 994. (doi:10.1021/ic901007w)

36. Del Guerzo A, Kirsch-De Mesmaeker A. 2002 Novel DNA sensor for guanine content. Inorg.Chem. 41, 938–945. (doi:10.1021/ic0108944)

37. Lincoln P, Norden B. 1998 DNA binding geometries of ruthenium(II) complexes with 1,10-phenanthroline and 2,2′-bipyridine ligands studied with linear dichroism spectroscopy.Borderline cases of intercalation. J. Phys. Chem. B 102, 9583–9594. (doi:10.1021/jp9824914)

38. Friedman AE, Chambron JC, Sauvage JP, Turro NJ, Barton JK. 1990 Molecular ‘light switch’for DNA: Ru(bpy)2(dppz)2(2+). J. Am. Chem. Soc. 112, 4960–4962. (doi:10.1021/ja00168a052)

39. Jenkins Y, Friedman AE, Turro NJ, Barton JK. 1992 Characterization of dipyridophenazinecomplexes of ruthenium(II): the light switch effect as a function of nucleic-acid sequence andconformation. Biochemistry 31, 10 809–10 816. (doi:10.1021/bi00159a023)

40. Lincoln P, Broo A, Norden B. 1996 Diastereomeric DNA-binding geometries ofintercalated ruthenium(II) trischelates probed by linear dichroism: [Ru(phen)(2)dppz](2+)and [Ru(phen)(2)bdppz](2+). J. Am. Chem. Soc. 118, 2644–2653. (doi:10.1021/ja953363l)

41. Coates CG et al. 2001 Picosecond time-resolved resonance Raman probing of the light-switchstates of [Ru(Phen)2dppz]2+. J. Phys. Chem. B 105, 12 653–12 664. (doi:10.1021/jp0127115)

42. Zeglis BM, Pierre VC, Barton JK. 2007 Metallo-intercalators and metallo-insertors. Chem.Commun. 44, 4565–4579. (doi:10.1039/B710949K)

43. Ortmans I, Elias B, Kelly JM, Moucheron C, Kirsch-De Mesmaeker A. 2004 Ru(TAP)2(dppz)2+:a DNA intercalating complex, which luminesces strongly in water and undergoes photo-induced proton-coupled electron transfer with guanosine-5′-monophosphate. Dalton Trans.2004, 668–676. (doi:10.1039/b313213g)

44. Moucheron C, Kirsch-De Mesmaeker A, Choua S. 1997 Photophysics of Ru(phen)2(PHEHAT)2+: a novel ‘light switch’ for DNA and photo-oxidant for mononucleotides. Inorg.Chem. 36, 584–592. (doi:10.1021/ic9609315)

45. Moucheron C, Kirsch-De Mesmaeker A. 1998 New DNA-binding ruthenium(II) complexes asphotoreagents for mononucleotides and DNA. J. Phys. Org. Chem. 11, 577–583. (doi:10.1002/(SICI)1099-1395(199808/09)11:8/9577::AID-POC53>3.0.CO;2-X)

46. Rickling S, Ghisdavu L, Pierard F, Gerbaux P, Surin M, Murat P, Defrancq E, MoucheronC, Kirsch-De Mesmaeker A. 2010 A rigid dinuclear ruthenium(II) complex as an efficientphotoactive agent for bridging two guanine bases of a duplex or quadruplex oligonucleotide.Chem. Eur. J. 16, 3951–3961. (doi:10.1002/chem.200902817)

47. Davis JT. 2004 G-quartets 40 years later: from 5′-GMP to molecular biology andsupramolecular chemistry. Angew. Chem. Int. Ed. 43, 668–698. (doi:10.1002/anie.200300589)

48. Neidle S, Balasubramanian S. 2006 Quadruplex nucleic acids. Cambridge, UK: Royal Society ofChemistry.

49. Zaug AJ, Podell ER, Cech TR. 2005 Human POT1 disrupts telomeric G-quadruplexes allowingtelomerase extension in vitro. Proc. Natl Acad. Sci. USA 102, 10 864–10 869. (doi:10.1073/pnas.0504744102)


http://dx.doi.org/doi:10.1021/bi061198o

http://dx.doi.org/doi:10.1021/bi061198o

http://dx.doi.org/doi:10.1021/ja049869y

http://dx.doi.org/doi:10.1021/bi0351965


http://dx.doi.org/doi:10.1016/S1387-7003(99)00030-1

http://dx.doi.org/doi:10.1016/S1387-7003(99)00030-1


http://dx.doi.org/doi:10.1021/ic901007w

http://dx.doi.org/doi:10.1021/ic901007w


http://dx.doi.org/doi:10.1021/jp9824914

http://dx.doi.org/doi:10.1021/ja00168a052

http://dx.doi.org/doi:10.1021/bi00159a023

http://dx.doi.org/doi:10.1021/ja953363l

http://dx.doi.org/doi:10.1021/jp0127115

http://dx.doi.org/doi:10.1039/B710949K

http://dx.doi.org/doi:10.1039/b313213g


http://dx.doi.org/doi:10.1002/(SICI)1099-1395(199808/09)11:8/9577::AID-POC53$>$3.0.CO;2-X

http://dx.doi.org/doi:10.1002/(SICI)1099-1395(199808/09)11:8/9577::AID-POC53$>$3.0.CO;2-X


http://dx.doi.org/doi:10.1002/anie.200300589

http://dx.doi.org/doi:10.1073/pnas.0504744102

http://dx.doi.org/doi:10.1073/pnas.0504744102


22


......................................................

50. Bejugam M, Sewitz S, Shirude PS, Rodriguez R, Shahid R, Balasubramanian S. 2007Trisubstituted isoalloxazines as a new class of G-quadruplex binding ligands: small moleculeregulation of C-kit oncogene expression. J. Am. Chem. Soc. 129, 12 926–12 927. (doi:10.1021/ja075881p)

51. Duquette M, Pham P, Goodman MF, Maizels N. 2005 AID binds to transcription-inducedstructures in c-MYC that map to regions associated with translocation and hypermutation.Oncogene 24, 5791–5798. (doi:10.1038/sj.onc.1208746)

52. Vanderlinden W, Blunt M, David CC, Moucheron C, Kirsch-De Mesmaeker A, DeFeyter S. 2012 Mesoscale DNA structural changes on binding and photo-reaction with[Ru(TAP)2(PHEHAT)]2+. J. Am. Chem. Soc. 134, 10 214–102 21. (doi:10.1021/ja303091q)

53. Kelly JM, Feeney MM, Tossi AB, Lecomte J-P, Kirsch-De Mesmaeker A. 1990 Interactionof tetra-azaphenanthrene ruthenium complexes with DNA and oligonucleotides.A photophysical and photochemical investigation. Anti-cancer Drug Design 5, 69–75.

54. Uji-i H, Foubert P, Schryver FD, De Feyter S, Gicquel E, Etoc A, Moucheron C, Kirsch-De Mesmaeker A. 2006 [Ru(TAP)3]2+-photosensitized DNA cleavage studied by atomicforce microscopy and gel electrophoresis: a comparative study. Chem. Eur. J. 12, 758–762.(doi:10.1002/chem.200500419)

55. Jiang Y, Ke C, Mieczkowski PA, Marszalek PE. 2007 Detecting ultraviolet damage insingle DNA molecules by atomic force microscopy. Biophys. J. 93, 1758–1767. (doi:10.1529/biophysj.107.108209)

56. Edelstein ML, Abedi MR, Wixon J. 2007 Gene therapy clinical trials worldwide to 2007: anupdate. J. Gene Med. 9, 833–842. (doi:10.1002/jgm.1100)

57. Cavazzana-Calvo M et al. 2000 Gene therapy of human severe combined immunodeficiency(SCID)-X1 disease. Science 288, 669–672.

58. Thomas CE, Ehrhardt A, Kay MA. 2003 Progress and problems with the use of viral vectorsfor gene therapy. Nat. Rev. Genet. 4, 346–358. (doi:10.1038/nrg1066)

59. Anderson WH. 1998 Human gene therapy. Nature 392, 25–30.60. Tamm I, Dorken B, Hartmann G. 2001 Antisense therapy in oncology: new hope for an old

idea? Lancet 358, 489–497. (doi:10.1016/S0140-6736(01)05629-X)61. Ortmans I, Content S, Boutonnet N, Kirsch-De Mesmaeker A, Bannwarth W, Constant

J-F, Defrancq E, Lhomme J. 1999 Ru-labeled oligonucleotides for photoinduced reactionson targeted DNA guanines. Chem. Eur. J. 5, 2712–2721. (doi:10.1002/(SICI)1521-3765(19990903)5:92712::AID-CHEM2712>3.0.CO;2-P)

62. Lentzen O, Constant J-F, Defrancq E, Prevost M, Schumm S, Moucheron C, Dumy P, Kirsch-De Mesmaeker A. 2003 Photocrosslinking in ruthenium-labelled duplex oligonucleotides.ChemBioChem 4, 195–202. (doi:10.1002/cbic.200390031)

63. Deroo S, Defrancq E, Moucheron C, Kirsch-De Mesmaeker A, Dumy P. 2003 Synthesis of anoxyamino-containing phenanthroline derivate for the efficient preparation of phenanthrolineoligonucleotide oxime conjugates. Tetrahedron Lett. 44, 8379–8382. (doi:10.1016/j.tetlet.2003.09.128)

64. Villien M, Deroo S, Gicquel E, Defrancq E, Moucheron C, Kirsch-De Mesmaeker A, Dumy P.2007 The oxime bound formation as an efficient tool for the conjugation of Ru complexes tooligonucleotides and peptides. Tetrahedron 63, 11 299–11 306. (doi:10.1016/j.tet.2007.08.088)

65. Lentzen O, Defrancq E, Constant J-F, Schumm S, Garcia-Fresnadillo D, Moucheron C, DumyP, Kirsch-De Mesmaeker A. 2004 Determination of DNA guanine sites forming photo-adducts with Ru(II)-labeled oligonucleotides: DNA polymerase inhibition by the resultingphoto-crosslinking. J. Biol. Inorg. Chem. 9, 100–108. (doi:10.1007/s00775-003-0502-3)

66. Saito I, Nakamura T, Nakatani K, Yoshioka Y, Yamaguchi K, Sugiyama H. 1998 Mappingof the hot spots for DNA damage by one-electron oxidation: efficacy of GG doublets andGGG triplets as a trap in long-range hole migration. J. Am. Chem. Soc. 120, 12 686–12 687.(doi:10.1021/ja981888i)

67. Saito I, Takayama M, Sugiyama H, Nakatani K, Tsuchida A, Yamamoto M. 1995 PhotoinducedDNA cleavage via electron transfer: demonstration that guanine residues located 5′ toguanine are the most electron-donating sites. J. Am. Chem. Soc. 117, 6406–6407. (doi:10.1021/ja00128a050)

68. Saito I, Nakamura T, Nakatani K. 2000 Mapping of highest occupied molecular orbitalsof duplex DNA by cobalt-mediated guanine oxidation. J. Am. Chem. Soc. 122, 3001–3006.(doi:10.1021/ja993891n)




http://dx.doi.org/doi:10.1038/sj.onc.1208746

http://dx.doi.org/doi:10.1021/ja303091q


http://dx.doi.org/doi:10.1529/biophysj.107.108209

http://dx.doi.org/doi:10.1529/biophysj.107.108209

http://dx.doi.org/doi:10.1002/jgm.1100

http://dx.doi.org/doi:10.1038/nrg1066

http://dx.doi.org/doi:10.1016/S0140-6736(01)05629-X

http://dx.doi.org/doi:10.1002/(SICI)1521-3765(19990903)5:92712::AID-CHEM2712$>$3.0.CO;2-P

http://dx.doi.org/doi:10.1002/(SICI)1521-3765(19990903)5:92712::AID-CHEM2712$>$3.0.CO;2-P

http://dx.doi.org/doi:10.1002/cbic.200390031

http://dx.doi.org/doi:10.1016/j.tetlet.2003.09.128

http://dx.doi.org/doi:10.1016/j.tetlet.2003.09.128

http://dx.doi.org/doi:10.1016/j.tet.2007.08.088

http://dx.doi.org/doi:10.1007/s00775-003-0502-3

http://dx.doi.org/doi:10.1021/ja981888i



http://dx.doi.org/doi:10.1021/ja993891n


23


......................................................

69. Le Gac S, Foucart M, Gerbaux P, Defrancq E, Moucheron C, Kirsch-De Mesmaeker A. 2010Photo-reactive RuII-oligonucleotide conjugates: influence of an intercalating ligand on theinter- and intra-strand photo-ligation processes. Dalton Trans. 39, 9672–9683. (doi:10.1039/C0DT00355G)

70. Le Gac S, Rickling S, Gerbaux P, Defrancq E, Moucheron C, Kirsch-De Mesmaeker A. 2009A photoreactive ruthenium(II) complex tethered to a guanine-containing oligonucleotide: abiomolecular tool that behaves as a ‘Seppuku molecule’. Angew. Chem. Int. Ed. 48, 1122–1125.(doi:10.1002/anie.200804503)

71. Le Gac S, Surin M, Defrancq E, Moucheron C, Kirsch-De Mesmaeker A. 2013 What arethe parameters controlling inter- versus intra-strand DNA photodamage with Ru-TAPoligonucleotides? Eur. J. Inorg. Chem. 2013, 208–216. (doi:10.1002/ejic.201201019)

72. Butz K, Hengstermann TRA, Denk C, Scheffner M, Hoppe-Seyle F. 2003 siRNA targeting ofthe viral E6 oncogene efficiently kills human papillomavirus-positive cancer cells. Oncogene22, 5938–5945. (doi:10.1038/sj.onc.1206894)

73. Zhang J, Hua Z. 2004 Targeted gene silencing by small interfering RNA-based knock-downtechnology. Curr. Pharm. Biotechnol. 5, 1–7. (doi:10.2174/1389201043489558)

74. Parkin DM. 2006 The global health burden of infection-associated cancers in the year 2002.Int. J. Cancer 118, 3030–3044. (doi:10.1002/ijc.21731)

75. Lechner MS, Mack DH, Finicle AB, Crook T, Vousden KH, Laimins LA. 1992 Humanpapillomavirus E6 proteins bind p53 in vivo and abrogate p53-mediated repression oftranscription. EMBO J. 11, 3045–3052.

76. Thomas M, Pim D, Banks L. 1999 The role of the E6-p53 interaction in the molecularpathogenesis of HPV. Oncogene 18, 7690–7700.

77. Rechner A et al. 2012 Ruthenium oligonucleotides, targeting HPV16 E6 oncogene, inhibit thegrowth of cervical cancer cells under illumination by a mechanism involving p53. Gene Ther.18, 435–443. (doi:10.1038/gt.2012.54)


http://dx.doi.org/doi:10.1039/C0DT00355G

http://dx.doi.org/doi:10.1039/C0DT00355G

http://dx.doi.org/doi:10.1002/anie.200804503


http://dx.doi.org/doi:10.1038/sj.onc.1206894


http://dx.doi.org/doi:10.1002/ijc.21731

http://dx.doi.org/doi:10.1038/gt.2012.54


Ru–TAP complexes and DNA: from photo-induced electron ...

Documents

Transcript of Ru–TAP complexes and DNA: from photo-induced electron ...