Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (112K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (26)
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Zhou-Sun, B.
Right arrow Articles by Taillandier, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhou-Sun, B.
Right arrow Articles by Taillandier, E.
Social Bookmarking
 Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 1997 Oxford University Press 1782-1787

Footnote

A physico-chemical study of triple helix formation by an oligodeoxythymidylate with N3' -> P5' phosphoramidate linkages

A physico-chemical study of triple helix formation by an oligodeoxythymidylate with N3 ' -> P5 ' phosphoramidate linkages Bei-wen Zhou-Sun, Jian-sheng Sun1, Sergei M. Gryaznov2, Jean Liquier, Thérèse Garestier1, Claude Hélène1 and Eliane Taillandier*

Laboratoire de Spectroscopie Biomoléculaire, CNRS URA 1430, UFR Santé Médecine Biologie Humaine, Université Paris-Nord, 74, rue Marcel Cachin, 93017 Bobigny, France, 1Laboratoire de Biophysique, Muséum National d'Histoire Naturelle, INSERM U201, CNRS URA481, 43, rue Cuvier, 75231 Paris Cedex 05, France and 2Lynx therapeutics, Inc., 3832 Bay Center Place, Hayward, CA 94545, USA

Received December 31, 1996; Revised and Accepted March 19, 1997

ABSTRACT

Non-denaturing gel retardation assay, DNA melting experiments and FTIR spectroscopy were used to characterize the triple helix formed by a 15mer 2'-deoxythymidylate with N3' -> P5' phosphoramidate linkages with its target sequence. The results indicate that: (i) the pentadecadeoxythymidylate with phosphoramidate linkages [dT15(np)] is highly potent to form a triple helix with a dT15-dA15 target duplex through Hoogsteen base-pairing; (ii) it forms a dT15(np)-dA15xdT15(np) triplex with the single-stranded oligo-2'-deoxyadenylate (dA15) without detectable double-helical intermediate; (iii) it does not only form a triple helix on the dT15-dA15 target duplex, but also partially displaces the dT15 strand from the dT15-dA15 duplex to form a dT15(np)-dA15xdT15(np) complex.

INTRODUCTION

Over the past few years, oligonucleotides have been receiving increased attention as a potential new class of pharmacologically active compounds. They can be used to interfere with gene expression in a sequence-specific manner by targeting either mRNA via double helix formation (antisense oligonucleotides or ribozymes) or chromosomal DNA via triple helix formation (antigene oligonucleotides) or proteins (sense oligonucleotides and aptamers) (1 -5 ). These possible applications have raised a great deal of interest in the development of oligonucleotide analogs with increased resistance to nucleases and/or enhanced binding to nucleic acid targets (6 ). The so-called antigene strategy is based on local triple helix formation via the binding of an oligonucleotide to the major groove of double-helical DNA (7 ,8 ). It was demonstrated that local triple helix formation could inhibit transcription in vitro/ex vivo by competing with the binding of transcriptional factors to the promotor, or blocking the initiation or the elongation step of transcription (for review see 3 ). Up to now there is no convincing evidence of an endogenous transcriptional inhibition in vivo by triple helix formation. Nevertheless recent studies which show that the endogenous targets are accessible to antigene oligonucleotides within the chromosomal environment (9 ) are encouraging for further antigene oligonucleotide developments.

The synthesis and some hybridization properties of a new family of oligonucleotide analogs containing N3' -> P5' phosphoramidate linkages have been recently described (10 ). An oligonucleotide phosphoramidate containing thymines and cytosines and/or guanines binds very strongly to double-helical DNA at the targeted oligopurine-oligopyrimidine segment and inhibits in vitro the transcription of the nef gene of the human immunodeficiency virus (HIV) at submicromolar concentration (11 ).

In an attempt to understand the strong binding of antigene oligonucleotides with N3' -> P5' phosphoramidate linkages to the targeted oligopurine-oligopyrimidine DNA sequences, we report here a study of the triple helix formed by a 15mer 2'-deoxythymidylate with phosphoramidate linkages with its target sequence by gel retardation assay, DNA melting experiments and FTIR spectroscopy. Gel retardation assay and DNA melting experiments were carried out in order to determine the thermal stability, the nature and the stoichiometry of the formed complexes. The study by FTIR spectroscopy of an oligonucleotide with N3' -> P5' phosphoramidate linkages is aimed at characterizing the interactions involved upon duplex and triplex formation by such an oligonucleotide. In the present manuscript, the double and triple helices are designated as Y-R and Y-RxH. The letters R and Y or H stand for purine and pyrimidine strands, the symbols - and x indicate Watson-Crick and Hoogsteen hydrogen bonds, respectively.

MATERIALS AND METHODS

Oligonucleotides and polynucleotides

Two 2'-deoxyoligonucleotides (dT15, dA15) and one modified 2'-deoxyoligonucleotide (7c-dA15) containing four dispersed 7-deaza-2'-deoxyadenines (Table 1 ) with phosphodiester linkages (OliGold grade) were purchased from Eurogentech (Belgium). They were ethanol precipitated. The 2'-deoxyoligonucleotide dT15(np) with N3' -> P5' phosphoramidate linkages was synthesized as described earlier (10 ). All these oligonucleotides were purified using a Ultrafree-MC filter (Millipore). Double-stranded polynucleotide poly(dA)-poly(dT) was purchased from Pharmacia and used without further purification. The concentrations were determined by the measurement at 25oC of their absorbance at 260 nm with a Kontron Uvikon 941 spectrophotometer using an extinction coefficient calculated according to a nearest-neighbor model (12 ) for oligonucleotides, and using an extinction coefficient of 6000/M/cm per nucleotide for polynucleotides.

Table 1 Oligonucleotides used in the present study
Oligonucleotides

Backbone

Notation

5'-AAAAAAAAAAAAAAA-3'

PO

dA15

5'-AAAAAAAAAAAAAAA-3'

PO

7c-dA15

5'-TTTTTTTTTTTTTTT-3'

PO

dT15

5'-TTTTTTTTTTTTTTT-3'

PN

dT15np)

PO and PN are abbreviations for the phosphodiester and the N3' -> P5' phosphoramidate linkage, respectively. A stands for 7-deaza-2'-deoxyadenine.

Gel retardation assay

A non-denaturing 10% acrylamide gel (40:1 acrylamide/bis-acrylamide) was prepared with 50 mM HEPES buffer, pH 7.0, without or in presence of 10 mM MgCl2. One strand was labelled at the 5'-end by T4 polynucleotide kinase using [[gamma]-32P]ATP. The labelled strand (10 nM) was completed with its unlabelled analog to 2 [mu]M. Samples prepared with different stoichiometries of strands were incubated at 4oC or room temperature overnight in a 50 mM HEPES buffer, pH 7.0, containing 10% sucrose without or in presence of 10 mM MgCl2. Electrophoresis was performed for 3-4 h with 2 W power at 4oC. Gels were dried, autoradiographed and analyzed on a Molecular Dynamics Phosphorimager.

DNA melting experiments

Double and triple helix stabilities were measured by DNA melting experiments which were carried out on an Uvikon 941 spectrophotometer using quartz cuvettes of 1 cm optical pathlength. The temperature of the cell holder was varied by circulating liquid using a Huber waterbath controlled by aHuber PD415 programmer at a rate of 0.1oC/min. Temperature monitoring was achieved by using a thermocouple in a control cuvette. Samples were dissolved at 1 [mu]M strand concentration in 10 mM sodium cacodylate buffer at pH 7.0 containing 0.1 or 0.4 M sodium chloride. All melting curves of the samples obtained upon cooling or heating were superimposable indicating that the equilibrium was achieved. The melting temperature (Tm) was evaluated as the temperature of half-association (or half-dissociation) of the formed complexes determined by the first derivative of the melting curve.

FTIR spectroscopy experiments

The samples were studied in the presence of 0.5 Na+ per nucleotide at pH 7.0. They were prepared in H2O or D2O solutions in ZnSe cells at 25 mM strand concentration. Deuteration experiments were performed by drying the samples under nitrogen and redissolving in D2O (>99.8% purity, purchased from Euriso-Top CEA). FTIR spectra were recorded using a Perkin Elmer 2000 spectrophotometer monitored by the Galaxy GRAMS386 program (1 cm-1 resolution). Spectra were generally obtained by accumulation of 25 scans. Temperature was controlled by a Specac temperature controller.

RESULTS AND DISCUSSION

Gel retardation assay

The experiments presented in Figure 1 A show the formation of the dT15(np)-dA15xdT15(np) and dT15-dA15xdT15(np) triple helices in presence of 10 mM MgCl2. A 7c-dA15 labelled strand containing 7-deaza-2'deoxyadenine residues has been used in lanes 1-5 instead of a dA15 labelled strand in lanes 6-11.


Figure 1. Gel retardation experiments carried out in 50 mM HEPES buffer pH 7, 10 mM MgCl2, at 4oC. The strand concentration of one equimolar strand is 2 [mu]M. The stoichiometry of samples is indicated by ratio. ss, ds and ts stand for single-, double- and triple-stranded complexes, respectively. (A) The purine strand is labelled (dA15* or 7c-dA15*). (B) One pyrimidine strand is labelled as indicated [dT15* or dT15(np)*].

As it can be seen Figure 1 A, lanes 7 and 8, the samples containing dA15 and dT15 with 1:1 and 1:2 stoichiometry respectively exhibit an identical gel mobility which is slower than that of the dA15 single strand (Fig. 1 A, lane 6). These bands migrate similarly to those of the samples containing 7c-dA15 and dT15 with 1:1 or 1:2 stoichiometry (Fig. 1 A, lanes 2 and 3, respectively) in which the modified adenines (7c-dA) were introduced to prevent Hoogsteen-like hydrogen bonds between the purine strand containing 7c-dA and the pyrimidine third strand. Therefore the 1:1 and 1:2 mixtures of dA15 and dT15 form only a dT15-dA15 Watson-Crick double helix under our experimental conditions in agreement with earlier UV studies performed on related oligomers (13 ,14 ) and with the UV data presented in Table 2 .

In contrast, the samples containing dA15 and the dT15(np) oligomer with N3'-P5' phosphoramidate linkages exhibit one band with retarded gel mobility reflecting the formation of the dT15(np)-dA15xdT15(np) triple helix (Fig. 1 A, lanes 10 and 11). In addition, lane 10, for the dA15+dT15(np) mixture with 1:1 stoichiometry, the presence of ~50% of free dA15 single strand and the absence of a band corresponding to the double-stranded species indicate that the formation of the dT15(np)-dA15xdT15(np) triple helix occurs without any detectable duplex intermediate.

On the other hand, upon addition of an equimolar amount of dT15(np) to pre-formed dT15-dA15 Watson-Crick duplex, an intense band is observed which has a slightly faster mobility as compared with that of dT15(np)-dA15xdT15(np) triplex (Fig. 1 A, compare lanes 9 and 11). This band is assigned to the dT15-dA15xdT15(np) triplex. However, another band corresponding to the dT15(np)-dA15xdT15(np) triplex might indicate a partial displacement of the Watson-Crick dT15 strand by the dT15(np) strand as a result of its higher affinity towards the dA15 strand (10 ).

Control experiments have been performed with 1:1 and 1:2 7c-dA15+dT15(np) samples (Fig. 1 A, lanes 4 and 5, respectively). The 7c-dA15 strand was designed to allow formation of a Watson-Crick duplex but not of a triple helix. As expected the band with retarded gel mobility assigned to the dT15(np)-dA15xdT15(np) triplex (Fig. 1 A, lanes 9-11) is absent in both 7c-dA15+dT15(np) samples (Fig. 1 A, lanes 4 and 5). It should be noticed that the dT15(np)-7c-dA15 duplex migrates slower than the dT15-7c-dA15 duplex (Fig. 1 A, compare lanes 4 and 5 with 2 and 3). This may originate from the slower mobility of the oligomer with phosphoramidate linkages than its isosequential analog with phosphodiester linkages and/or of different conformations of these duplexes.

In order to investigate the partial pyrimidine strand displacement further and to confirm the above results, gel retardation has also been carried out with labelled dT15 or dT15(np) strands (Fig. 1 B, lanes 1-3 and lanes 4-6, respectively). Comparison of lanes 1 and 4 clearly shows that the dT15(np) strand migrates slower than the dT15 strand.

A partial dT15 strand displacement from the dT15-dA15 duplex upon dT15(np) binding is in agreement with the data in Figure 1 B, lane 3. The migration band of the dT15-dA15 duplex is assigned to the retarded band observed upon addition of the labelled dT15 strand to the dA15 strand (Fig. 1 B, lane 2) since it migrates in a similar manner as that of the same duplex but with labelled dA15 strand (data not shown). Upon addition of one equimolar of the dT15(np) oligomer to the dT15-dA15 duplex, three bands are detected (Fig. 1 B, lane 3). The further retarded band is assigned to the dT15-dA15xdT15(np) triplex. The remaining duplex and single strand bands can be interpreted as a result of partial displacement of the dT15 strand upon triplex formation in presence of the dT15(np) oligomer. It should be pointed out that the dT15(np)-dA15xdT15(np) triplex cannot be seen since there is no labelled strand, and it should migrate slightly slower than the dT15-dA15xdT15(np) triplex (Fig. 1 B, compare lanes 5 and 6 with 3, see below).

Finally for the sample containing dA15 and dT15(np) oligomers with 1:1 input ratio, only one retarded band which migrates slightly slower than that of dT15-dA15xdT15(np) triplex is observed (Fig. 1 B, lane 5). This band is therefore assigned to the dT15(np)-dA15xdT15(np) triplex which is again formed without any detectable double-helical intermediate. On the other hand, upon addition of one equimolar of the labelled dT15(np) strand to the pre-formed dT15-dA15 duplex, two bands corresponding to the dT15(np)-dA15xdT15(np) and the dT15-dA15xdT15(np) triplexes are observed (Fig. 1 B, compare lane 6 with lanes 5 and 3) confirming the partial strand displacement.

Experiments carried out in presence of NaCl also show the formation of dT15-dA15xdT15(np) and dT15(np)-dA15xdT15(np) triple helices as well as the dT15 strand displacement (data not shown). However the complexes are less stable and the strand displacement less effective.

DNA melting experiments

Table 2 summarizes the melting temperatures of the samples containing different combinations of two oligomers. The dA15+dT15 sample with 1:1 stoichiometry (duplex) presents a melting temperature of 35.5oC in 100 mM NaCl. The complex with 1:2 stoichiometry obtained by addition of a second dT15 strand has also a monophasic melting profile in 100 mM NaCl and the measured Tm is similar. Therefore, the dT15-dA15xdT15 triplex is not formed under low salt condition in agreement with earlier studies (13 ,14 ). In the presence of 400 mM NaCl, a biphasic melting profile is observed (Fig. 2 , inset). It is noticed that the increase of the optical density (transition amplitude) upon temperature increase in 100 mM NaCl, which corresponds to the melting of the dA15-dT15 duplex is the same for the 1:1 and the 1:2 stoichiometry solutions. However the hyperchromicity is different in both cases, as for the 1:2 sample the absorption contains a contribution of the non base-paired dT15 strand at all temperatures.

Table 2 . Characteristics of the melting curves of different samples in a buffer containing 10 mM sodium cacodylate, pH 7.0, 0.1 M sodium chloride
Strands

A15+dT15

7c-dA15+dT15

7c-dA15+dT15(np)

dA15+dT15(np)

Stoichiometry

1:1

1:2

1:1

1:1

1:1

1:2

Tm (+-0.5oC)

35.5

36.5

29.0

32.0

45.0

45.5

[Delta]A (260 nm)

0.06

0.06

0.03

0.03

0.06

0.12

Hypochromism [Delta]A/A (260 nm)

23%

17%

13%

13%

20%

30%

The strand concentration of one equimolar of oligomer is 1 [mu]M. [Delta]A(260 nm) represents the amplitude of the melting transition as measured by the difference of absorbance at 260 nm after and before the transition.


Figure 2. The melting curve (bold line) and its first derivative (thin line) of the sample containing three oligomers dT15, dA15 and dT15(np) with 1:1:1 stoichiometry in 10 mM sodium cacodylate, pH 7.0, 0.1 M sodium chloride. The strand concentration is 1 [mu]M. Inset shows the melting curve of dT15-dA15xdT15 in 10 mM sodium cacodylate, pH 7.0, 0.4 M sodium chloride.Duplexes have been formed by substitution of one of the strands in the original dA15-dT15 duplex either by a 7c-dA15 strand or by a dT15(np) strand. Melting of the 7c-dA15+dT15 (1:1) complex occurs at a lower temperature (Tm = 29oC), which can be ascribed to the effect of 7-deaza-adenines. In the latter duplex, replacement of the dT15 strand by a dT15(np) strand with N3' -> P5' phosphoramidate linkages [7c-dA15+dT15(np) with 1:1 ratio] slightly increases the thermal stability (Tm = 32oC) of the sample.

The dA15+dT15(np) 1:1 complex presents a monophasic melting profile with a significatively higher melting temperature (Tm = 45oC) than for the duplexes discussed above [dA15+dT15, 7c-dA15+dT15, 7c-dA15+dT15(np)]. Addition of an extra amount of dT15(np) to the dA15+dT15(np) sample (final stoichiometry 1:2) does not change the monophasic melting temperature. The complexes formed in the 1:1 and 1:2 mixtures are thus identical. However, and in contrast to what was observed for the dA15+dT15 samples, the addition of extra dT15(np) to the 1:1 dA15+dT15(np) mixture up to a 1:2 ratio induces an increase in the transition amplitude which is doubled for the 1:2 stoichiometry as compared with that of 1:1. This shows that the formed complexes are the same in both cases, but their amount is double for 1:2 stoichiometry, indicative of a triple helix formed by one dA15 and two dT15(np) strands, in agreement with gel retardation experiments. However the hyperchromism is again different since there is 50% dA15 single strand in the 1:1 mixture.

Addition of a modified dT15(np) strand to a dA15-dT15 duplex shows a partial displacement of the dT15 strand of the target duplex by the dT15(np) strand and formation of a dT15(np)-dA15xdT15(np) triple-stranded structure. This has been shown by comparing complexes with two different dT15 input ratios. The melting curve of the sample containing three oligomers dT15, dA15 and dT15(np) with 1:1:1 stoichiometry is shown in Figure 2 . The sample is prepared by addition of the dT15(np) strand to the pre-formed dT15-dA15 duplex solution. Two transitions are observed at 36 and 45oC, corresponding respectively to the melting of a dT15-dA15 duplex and of a dT15(np)-dA15xdT15(np) triplex (see above). When the stoichiometry of the sample is 0.8:1:1, an identical profile of the melting curve (Tm values and transition amplitude) is obtained. The overall decrease of absorbance (~0.02) corresponds to -20% absorbance of the missing dT15 strand (data not shown). This indicates the non-involvement of at least 20% of dT15 strand in the formed complexes, and thus implies a partial strand displacement of the oligomer with natural phosphodiester linkages, dT15, from the dT15-dA15 duplex by the dT15(np) oligomer with N3' -> P5' phosphoramidate linkages upon triplex formation.

FTIR spectroscopy: triple helix formation

Triple helix formation can be characterized by the FTIR spectra recorded in D2O solutions in the spectral domain between 1750 and 1550 cm-1 presented in Figure 3 . We show here the spectra ofdA15+dT15(np) samples with 1:1 and 1:2 stoichiometry (Fig. 3 c and d) and of the complex between poly(dT)-poly(dA) and dT15(np) prepared so as to contain a 1:1:1 stoichiometry (Fig. 3 e). Their spectral profiles are similar to that of the canonical triple helix dT15-dA15xdT15 (Fig. 3 b) which is easily stabilized at 0.5 Na+ per nucleotide (15 ,16 ).


Figure 3. FTIR spectra recorded in D2O (pH 7.0) in the presence of 0.5 Na+/phosphate, in the spectral range 1750-1550 cm-1 (at 25oC): (a) double helix dT15-dA15; (b) triple helix dT15-dA15xdT15; (c) 50% triple helix dT15(np)-dA15xdT15(np) + 50% dA15; (d) triple helix dT15(np)-dA15xdT15(np); (e) triple helix poly(dT)-poly(dA)xdT15(np).

For comparison, the spectrum of the dT15-dA15 double helix is shown on Figure 3 a. The spectral range 1750-1550 cm-1 presented contains absorption bands assigned to in-plane double-bond stretching vibrations of bases. These bands are known to be particularly sensitive to base-pairing interactions. The four strong bands observed in the dT15-dA15 double helix spectrum (Fig. 3 a) have been respectively assigned to vibrational modes of thymine (C2=O2 stretching, 1695 cm-1; C4=O4 stretching, 1662 cm-1; ring C=C and C=N stretching, 1641cm-1), and adenine (ring C=C and C=N stretching, ND2 scissoring, 1622 cm-1) (17 ).

Formation of the canonical triple helix dT15-dA15xdT15 is characterized by significant decreases in the intensities of the 1622 cm-1 adenine vibration band, of the 1641 cm-1 thymine ring vibration band and a slight down-shift of the C4=O4 thymine band (1662 -> 1659 cm-1; Fig. 3 b) (18 ). Similarly, thedA15+dT15(np) spectrum with 1:2 stoichiometry (Fig. 3 d) shows that the relative intensities of the adenine absorption and of the thymine ring vibration absorption are strongly decreased, indicative of triplex formation. Moreover, the C2=O2 and C4=O4 carbonyl bands are further shifted to lower wave numbers. In the case of the complex prepared with poly(dT)-poly(dA) and a third dT15(np) strand with 1:1:1 stoichiometry (Fig. 3 e) the decrease of the relative intensities of both adenine and thymine vibrations also reflects the formation of a triple helix. The broader profiles of the carbonyl absorptions in the poly(dT)-poly(dA)xdT15(np) triplex spectrum and their location (1690 and 1656 cm-1) at positions intermediate between the corresponding absorptions of the dT15-dA15xdT15 (1696 and 1659 cm-1; Fig. 3 b) and of the dT15(np)-dA15xdT15(np) (1685 and 1651 cm-1; Fig. 3 d) triplexes reflect the overlap of contributions of dT15(np) and dT15 strands.

The spectrum of the dA15+dT15(np) sample with 1:1 stoichiometry exhibits two carbonyl bands with the same wavenumbers (Fig. 3 c) as those of the dT15(np)-dA15xdT15(np) triplex (Fig. 3 d) and an additional band at 1628 cm-1 which corresponds to free adenine. It is known that this band is down-shifted when A-T base pairs are formed through either Watson-Crick, reverse Watson-Crick or Hoogsteen hydrogen bonds (18 -20 ). In the present study, if this additional band originates from a contribution of free dA15 strand, the spectral characteristics of the sample containingdA15+dT15(np) with 1:1 stoichiometry (Fig. 3 c) could be interpreted according to the following scheme: dA15 + dT15(np) -> 50% dT15(np)-dA15 xdT15(np) + 50% dA15. In other words, dA15+dT15(np) with 1:1 stoichiometry directly forms a triplex and not a duplex.


Figure 4. FTIR spectra recorded in D2O (pH 7.0) in the presence of 0.5 Na+/phosphate, in the spectral range 1750-1550 cm-1 (at 25oC): (a) double helix 7c-dA15-dT15; (b) double helix 7c-dA15-dT15(np).

In an attempt to further check this hypothesis, the duplex formed by dT15(np) and 7c-dA15 (in which the presence of 7-deaza-adenine prevents triple helix formation) has been studied by FTIR (Fig. 4 b). For comparison, the spectrum of the dT15-7c-dA15 duplex is shown in Figure 4 a. The Watson-Crick double helix dT15-dA15 (Fig. 3 a) is characterized by spectral shifts of the ring vibration bands of adenine (1628 -> 1622 cm-1) and of thymine (1633 -> 1641 cm-1), and changes in relative band intensities of the thymine bands, as compared with the single-stranded deoxyadenine and deoxythymine strands (18 ). The Watson-Crick double helix dT15(np)-7c-dA15 is formed with similar spectral signatures and the presence of four 7-deaza-adenines gives two additional shoulders at 1605 and 1560 cm-1 (which are also detected in the single-stranded 7c-dA15 oligomer, spectrum not shown). For the sample containing dT15(np)+7c-dA15 with 1:1 stoichiometry (Fig. 4 b), the down-shifted ring vibration mode of 7c-dA15 (1628 -> 1622 cm-1) and the up-shifted ring vibration mode of dT15(np) (1633 -> 1637 cm-1) indicate the formation of the duplex. This spectrum is different from that of dT15(np)+dA15 with 1:1 stoichiometry. Thus dT15(np)+dA15 with 1:1 stoichiometry directly forms a dT15(np)-dA15xdT15(np) triplex, in agreement with gel retardation assay and DNA melting experiments. It is worth noticing that significant down shifts of dT15(np) C=O bands observed in the triplex spectrum are also observed in the dT15(np)-7c-dA15 duplex spectrum.

Base carbonyl vibrations are known to be sensitive to hydrogen bonding (17 ). When an A-T base pair is formed with reverse Watson-Crick hydrogen bonds, the C2=O2 band is down shifted ( -> 1685 cm-1) and the C4=O4 band is up-shifted ( -> 1668 cm-1) (19 ). However, it is not the case for the dT15(np)-7c-dA15 duplex. When the A-T base pair is formed with reverse Hoogsteen hydrogen bonds, the C2=O2 band is up-shifted ( -> 1712 cm-1) whereas the C4=O4 band is not significantly shifted (16 ). This does not correspond to the wavenumber shifts in the triplex formed by dT15(np) with dA15 or poly(dT)-poly(dA). Therefore we can conclude that the duplex and the triplex containing dT15(np) are formed with Watson-Crick and for the third strand Hoogsteen hydrogen bonds.

CONCLUSION

In summary, three techniques (gel retardation, DNA melting and FTIR spectroscopy) were used to investigate triple helix formation by a 15mer deoxythymidylate with N3' -> P5' phosphoramidate linkages on its target sequence. Together, they have consistently determined the nature and the stoichiometry of the formed complexes, as well as the interactions involved in each complex. The results clearly indicate that: (i) the pentadecadeoxythymidylate with phosphoramidate linkages is highly potent to form a triple helix with its target sequence through Hoogsteen base pairing; (ii) it forms a triplex with the single-stranded pentadecadeoxyadenylate without any detectable double-helical intermediate; (iii) it does not only form a triple helix on the dT15-dA15 target duplex, but also displaces, at least partially, the dT15 strand from dT15-dA15 to form a new complex, the dT15(np)-dA15x dT15(np) triplex.

REFERENCES

1 Hélène,C. and Toulmé,J.J. (1990) Biochem. Biophys. Acta., 1049, 99-125.

2 Hélène,C. (1994) Eur. J. Cancer, 30, 1721-1726.

3 Thuong,N.T. and Hélène,C. (1993) Angew. Chem. Int. Ed. Engl., 32, 666-690.

4 Stein,C.A. and Cheng,Y.C. (1993) Science, 261, 1004-1012. MEDLINE Abstract

5 Wagner,R.W. (1994) Nature, 372, 333-335. MEDLINE Abstract

6 Uhlmann,E. and Peyman,A. (1990) Chem. Rev., 90, 543-584.

7 Le Doan,T., Perrouault,L., Praseuth,D., Habhoub,N., Decout,J.L., Thuong,N.T., Lhomme,J. and Hélène,C. (1987) Nucleic Acids Res., 15, 7749-7760. MEDLINE Abstract

8 Moser,H. and Dervan,P.B. (1987) Science, 238, 645-650. MEDLINE Abstract

9 Giovannageli,C., Divialcco,S., Labrousse,V., Gryaznov,S., Charneau,P. and Hélène,C. (1997) Proc. Natl. Acad. Sci. USA, 94, 79-84.

10 Gryaznov,S.M. and Chen,J.K. (1994) J. Am. Chem. Soc., 116, 3143-3144.

11 Escudé,C., Giovannangeli,C., Sun,J.S., Lloyd,D.H., Chen,J.K., Gryaznov,S.M., Garestier,T. and Hélène,C. (1996) Proc. Natl. Acad. Sci. USA, 93, 4365-4369. MEDLINE Abstract

12 Cantor,C.R. and Warshaw,M.M. (1970) Biopolymers, 9, 1059-1077. MEDLINE Abstract

13 Pilch,D.S., Brousseau,R. and Shafer,R.D. (1990) Nucleic Acids Res., 18, 5743-5750. MEDLINE Abstract

14 Thomas,T. and Thomas,T.J. (1993) Biochemistry, 32, 14068-14074. MEDLINE Abstract

15 Liquier,J., Coffinier,P., Firon,M. and Taillandier,E. (1991) J. Biomol. Struct. Dyn., 3, 437-445.

16 Dagneaux,C., Liquier,J. and Taillandier,E. (1995) Biochemistry, 34, 14815-14818. MEDLINE Abstract

17 Tsuboi,M. (1969) In Bram,E.G.Jr (ed.), Applied Spectroscopy Reviews. M. Dekker, New York, Vol. 3, pp. 45-90.

18 Liquier,J. and Taillandier,E. (1996) In Mantsch,H.H. and Chapman,D.J. (ed.), Infrared Spectroscopy of Biomolecules Wiley Publishing, pp. 131-158.

19 Fritzsche,H., Akhebat,A., Taillandier,E., Rippe,K. and Jovin,T.M. (1993) Nucleic Acids Res., 21, 5085-5091. MEDLINE Abstract

20 Escudé,C., Mohammadi,S., Sun,J.S., Nguyen,C.H., Bisagni,E., Liquier J., Taillandier,E., Garestier,T. and Hélène,C. (1996) Chem. Biol., 3, 57-65. MEDLINE Abstract


Return

* To whom correspondence should be addressed. Tel: +33 1 48 38 76 90; Fax: +33 1 48 37 74 43
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
Nucleic Acids ResHome page
S. Ivanov, Y. Alekseev, J.-R. Bertrand, C. Malvy, and M. B. Gottikh
Formation of stable triplexes between purine RNA and pyrimidine oligodeoxyxylonucleotides
Nucleic Acids Res., July 15, 2003; 31(14): 4256 - 4263.
[Abstract] [Full Text] [PDF]


Home page
Nucleic Acids ResHome page
J. Liu, R.-H. Xu, Y.-X. Jin, and D.-B. Wang
Triplex targeting of human PDGF-B (c-sis, proto-oncogene) promoter specifically inhibits factors binding and PDGF-B transcription
Nucleic Acids Res., February 1, 2001; 29(3): 783 - 791.
[Abstract] [Full Text] [PDF]


This Article
Right arrow Abstract Freely available
Right arrow Print PDF (112K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (26)
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Zhou-Sun, B.
Right arrow Articles by Taillandier, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhou-Sun, B.
Right arrow Articles by Taillandier, E.
Social Bookmarking
 Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?