Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (96K) 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 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 arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Colombier, C
Right arrow Articles by Leng, M
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Colombier, C
Right arrow Articles by Leng, M
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 1996 Oxford University Press 4519-4525

Footnote

Interstrand cross-linking reaction in triplexes containing a monofunctional transplatin-adduct

Interstrand cross-linking reaction in triplexes containing a monofunctional transplatin-adduct Caroline Colombier , Bernhard Lippert 1 and Marc Leng*

Centre de Biophysique Moléculaire, CNRS, Rue Charles Sadron, 45071 Orléans cedex 2, France and 1 Fachbereich Chemie, Universität Dortmund, 44221 Dortmund , Germany

Received July 24, 1996; Revised and Accepted September 16, 1996

ABSTRACT

Our aim was to determine whether a single transplatin monofunctional adduct, either trans -[Pt(NH 3 ) 2 (dC)Cl] + or trans -[Pt(NH 3 ) 2 (dG)Cl] + within a homopyrimidine oligonucleotide, could further react and form an interstrand cross-link once the platinated oligonucleotide was bound to the complementary duplex. The single monofunctional adduct was located at either the 5 ' end or in the middle of the platinated oligonucleotide. In all the triplexes, specific interstrand cross-links were formed between the platinated Hoogsteen strand and the complementary purine-rich strand. No interstrand cross-links were detected between the platinated oligonucleotides and non-complementary DNA. The yield and the rate of the cross-linking reaction depend upon the nature and location of the monofunctional adducts. Half-lives of the monofunctional adducts within the triplexes were in the range 2-6 h. The potential use of the platinated oligonucleotides to modulate gene expression is discussed.

INTRODUCTION

During recent years, much work has been devoted to the study of nucleic acids triplexes. It has been demonstrated that homopyrimidine oligonucleotides can be targeted to homopurine-homopyrimidine tracts of DNA. The oligonucleotides bind to the major groove of DNA, forming a triple helix. Sequence specificity is achieved by Hoogsteen hydrogen bonding interaction of thymine with A[middot]T and protonated cytosine with G[middot]C base pairs, respectively. These results and others concerning homopurine oligonucleotides, mixed oligonucleotides and modified oligonucleotides have opened a new field of investigation (the so-called anti-gene strategy), largely aimed at the design of sequence-specific gene modulators ( 1 - 6 and references therein).

The formation of a local triple helix can affect gene expression in several ways ( 1 - 6 ). However, a major problem in in vitro and in vivo experiments arises from the stability of the triplexes. Even if the triplexes are stable under physiological conditions, they can be displaced by the replication or the transcription machinery ( 1 - 6 ). One possibility to avoid this displacement is to cross-link the third strand to the duplex. This has been achieved by attaching covalently to the oligonucleotides reagents able to react spontaneously or after light activation with the duplexes ( 2 ). Among the chemical reagents, platinum(II) derivatives have been already used in duplexes ( 7 - 8 ). Gruff and Orgel ( 9 ) demonstrated that binuclear platinum(II) complexes of the form [{ trans -Pt(NH 3 ) 2 Cl} 2 {NH 2 (CH 2 ) n }]Cl 2 react rapidly with oligodeoxynucleotide-5'-phosphorothioates and then with polypurine tracts in triple helical DNA.

Our aim was to determine whether the bifunctional compound trans -diamminedichloroplatinum(II) (transplatin) could be used as a cross-linking reagent in triplexes. Recent results on nucleobases have shown that replacement of a weakly acidic N-H proton in a hydrogen bond between two nucleobases by a metal species of suitable geometry generates metal-modified nucleobase pairs ( 10 - 12 ). The linkage between the two nucleobase pairs becomes considerably stronger as hydrogen bonds are replaced by covalent bonds. In the case of G[middot]C Hoogsteen pairing and transplatin, the platinated pair fits almost exactly the requirement of normal DNA as far as interglycosyl distances are concerned. These results suggested that a monofunctional adduct trans -[Pt(NH 3 ) 2 (dC)Cl] + within the Hoogsteen strand of a triplex is in an ideal position to cross-link the complementary purine-rich strand. They also suggested that the platinated C residue can be replaced by a platinated G residue but the cross-linking reaction requires a switch from the anti to the syn conformation for the platinated G residue. In this paper, we have studied the cross-linking reaction between homopyrimidine oligonucleotides containing a single transplatin monofunctional adduct ( trans -[Pt(NH 3 ) 2 (dC)Cl] + or trans -[Pt(NH 3 ) 2 (dG)Cl] + ) and the complementary duplex. We show that specific interstrand cross-links are formed. The rate and the yield of the cross-linking reaction depend upon the nature and location of the monofunctional adduct residue within the Hoogsteen strand.

MATERIALS AND METHODS

The oligodeoxyribonucleotides, purchased from Institut Pasteur (Paris) were purified by ion-exchange chromatography on a Pharmacia FPLC system ( 13 ). Their sequences are given in Figure 1 . Endonucleases, alkaline phosphatase and T4 DNA ligase were from Biolabs or Boehringer Mannheim, transplatin was from Johnson Matthey, electrophoresis-grade acrylamide and the other chemicals from Merck. The radioactive products were from Amersham. The plasmid pLYR was constructed by insertion of the duplex (d(GTCAGAAAAAGAAAGAAAAGAAACG).d(CGTTTCTTTTCTTTCTTTTTCTGAC)) into Sma I restriction site of plasmid pSP73 by standard procedures and then transformed in Escherichia coli HB101. The cloning was confirmed by DNA sequencing.


Figure 1 . Sequences of the target duplex (25 bp) and the triplex forming oligonucleotides (19mers). The symbol * indicates that the oligonucleotides (19mer) contain a single monofunctional adduct ( trans- [Pt(NH 3 ) 2 (dC)Cl] + (Pt-C) or trans- [Pt(NH 3 ) 2 (dG)Cl] + (Pt-G). In the case of III*, Pt-C is located at any of the four positions (1,7,11,16). The formula of transplatin ( trans -diamminedichloroplatinum(II)) and of the monofunctional adducts trans- [Pt(NH 3 ) 2 (dC)Cl] + and trans- [Pt(NH 3 ) 2 (dG)Cl] + are also given.

Platination

For sake of clarity, the unplatinated oligonucleotides (19mer) are designated by I, II and III, and the corresponding oligonucleotides containing a single monofunctional adduct I*, II* and III*, respectively. Oligonucleotides I* and II* containing the monofunctional adduct trans -[Pt(NH 3 ) 2 (dG)Cl] + were prepared as previously described ( 14 ) with some modifications. The oligonucleotides I and II (40 [mu]M) were incubated at 37oC for 30 min with trans -[Pt(NH 3 ) 2 Cl(H 2 O)] + (generated by allowing transplatin to react with AgNO 3 overnight) at a platinum/oligonucleotide molar ratio equal to 4, in 10 mM NaClO 4 , 4 mM acetate buffer, pH 3.6. The platination of the oligonucleotide III was performed in 10 mM NaClO 4 . The platinated oligonucleotides containing a single monofunctional adduct were purified by FPLC in a gradient of 0.2-0.6 M NaCl (neutral pH). The purified oligonucleotides were dialyzed against 0.1 M NaCl for 2 h at 4oC. They were stored at -20oC before use. It was verified that all the platinum residues within the platinated oligonucleotides were removed by treatment with 10 mM thiourea for 10 min at 37oC ( 30 ).

Gel retardation assay

The DNA duplex (25 bp) was obtained by hybridization of the two complementary oligonucleotides d(GTCAGAAAAAGAAAGAAAAGAAACG) (called purine-rich strand) and d(CGTTTCTTTTCTTTCTTTTTCTGAC) (called pyrimidine-rich strand) in 150 mM NaClO 4 , 2 mM phosphate buffer pH 6.5. To the double-stranded DNA (c = 0.6 [mu]M) was added an excess (arbitrarily two equivalents) of the platinated or unplatinated Hoogsteen strands and the mixtures were incubated in 150 mM NaClO 4 , 5 mM Mg(ClO 4 ) 2 , 5 mM acetate buffer pH 5.5, at 4oC for 1 h. The mixtures were analysed by gel electrophoresis under native conditions (10% polyacrylamide, 50 mM 2-[N-morpholino]ethanesulfonic acid (MES) buffer, adjusted at pH 5.5 with NaOH, 5 mM MgCl 2 ).

Interstrand cross-linking reaction

The mixtures containing the target double-stranded oligonucleotide (25 bp) and the Hoogsteen strands prepared as for the gel retardation assays, were incubated in 150 mM NaClO 4 , 5 mM Mg(ClO 4 ) 2 , 5 mM acetate buffer, pH 5 or phosphate buffer (pH range 6-7), at 37oC. Aliquots were withdrawn at various time intervals and analysed by gel electrophoresis under denaturing conditions (24% polyacrylamide/8 M urea). Quantitation of the gel bands was done on a Molecular Dynamics PhosphorImager using ImageQuant software version 3.3 for data processing.

Mapping of the interstrand cross-links

The products containing the interstrand cross-links were purified on a 10% polyacrylamide gel under native conditions (pH 8). The location of the platinated bases in the purine-rich strand was determined by dimethylsulfate (DMS) footprinting experiments as previously described ( 15 - 16 , 19 ).

Reaction between the Hoogsteen strands and plasmid DNA

The plasmids (pSP73, pLYR) were digested by endonuclease Hind III. The linear DNAs (c ~ 0.1 [mu]M) were incubated respectively with two equivalent of 32 P-labeled oligonucleotides I* or II* in the conditions described above for the interstrand cross-linking reaction. The products of the reaction were analysed by gel electrophoresis [1.4% agarose gel, 40 mM tetraacetic acid (TAE) buffer, pH 8.1].

RESULTS

Several single-stranded oligonucleotides (19mer) (see Fig. 1 for their sequences) were reacted with transplatin in conditions that the platinated oligonucleotides contained a single monofunctional adduct, either trans -[Pt(NH 3 ) 2 (dC)Cl] + or trans - [Pt(NH 3 ) 2 (dG)Cl] + . After addition of the platinated oligonucleotides to the complementary duplex (25 bp), the stability of the triplexes, the cross-linking reaction and the nature of the interstrand cross-links were studied. The first part of the work deals with the oligonucleotides containing a single monofunctional trans- [Pt(NH 3 ) 2 (dG)Cl] + adduct and the second part with the oligonucleotides containing a single monofunctional trans- [Pt(NH 3 ) 2 (dC)Cl] + adduct.

Monofunctional trans- [Pt(NH 3 ) 2 (dG)Cl] + adduct

Stability of the triplexes . The first step was to determine whether the presence of the monofunctional adducts within the Hoogsteen strand prevented the formation of the triplexes. The double helix (25 bp) and the homopyrimidine oligonucleotides (19mer) containing or omitting a single monofunctional trans- [Pt(NH 3 ) 2 (dG)Cl] + adduct were mixed, incubated in 150 mM NaClO 4 for 1 h at 4oC and then analysed by gel-retardation assays at 37oC and pH 5.5. As shown in Figure 2 , the triplex with platinated oligonucleotide I* (the platinated G is at the 5' end of the oligonucleotide) is stable whereas triplex with the platinated oligonucleotide II* (the platinated G is at the seventh position from the 5' end) is partially dissociated. In the same experimental conditions but at 20oC, free duplexes were not detected (not shown). As the pH was increased, all the triplexes were partially dissociated and completely dissociated at pH 7 (not shown).


Figure 2 . Recognition of the the target duplex by the platinated or unplatinated Hoogsteen strand (19mer). Gel retardation assay showing binding of the 19mers to the duplex (25 bp). The purine-rich strand (25mer) was 32 P-labeled at the 5' end. After hybridization with its complementary strand, the duplex was mixed with the different 19mers, respectively, as described in Materials and Methods. The mixtures were analysed by gel electrophoresis under native conditions. The symbols above the lanes stand for: ss, single-stranded purine-rich oligonucleotide (25mer), ds, target duplex (25 bp), ds + I and ds + I*, the target duplex plus the oligonucleotide I or I*, ds + II and ds + II*, the target duplex plus the oligonucleotide II or II*. The buffer for the electrophoresis was 50 mM MES adjusted to pH 5.5 with NaOH, 5 mM MgCl 2 . The gel was run at 37oC.
Rate of the interstrand cross-linking reaction. The triplexes in which the purine-rich strand was 32 P-labeled at the 5' end, were incubated at 37oC and in the pH range 5-7. As a function of time, aliquots were withdrawn and analysed by gel electrophoresis under denaturing conditions. The results relative to platinated oligonucleotide I* are shown in Figure 3 (left). At pH 5 and time = 0 of incubation, only one band corresponding to the purine-rich strand is present. After longer times of incubation, the intensity of this band decreases whereas three new slowly migrating bands (a major one and two minor ones ; the sum of the intensities of the two minor bands over the sum of the intensities of the three bands is 10-15%) appear. The [tau] 1/2 ( monofunctional adduct within the triplex) is ~6 h (Fig. 3 , right). At pH 6, the results were similar but fewer cross-links (70%) are formed (Fig. 3 , right). No cross-links were formed at pH 7.


Figure 3 . Interstrand cross-linking reaction within the triplex containing the monofunctional adduct trans- [Pt(NH 3 ) 2 (dG)Cl] + . ( Left ) Autoradiogram of a denaturing 24% polyacrylamide gel. The purine-rich oligonucleotide (25mer) was labeled with 32 P. The mixture of the target duplex and the oligonucleotide I* were incubated in 5 mM Mg(ClO 4 ) 2 , 150 mM NaClO 4 , 5 mM acetate buffer, pH 5 and at 37oC. The reaction times (h) are indicated above the lanes. ( Right ) Kinetics of interstrand cross-links formation at pH 5 ([squf]), 6 (s) and 7 (-) (at pH 6 and 7, phosphate buffer instead of acetate buffer was used). The percentages of interstrand cross-links were calculated from the ratios of the sums of the intensities of the slowly moving bands over the sum of the intensities of the all the bands on the corresponding autoradiograms.

Similar experiments were repeated with platinated oligonucleotide II*. After 24 h incubation at 37oC, ~7% of interstrand cross-links were formed. An explanation for this low yield is that the triplex was in part dissociated (Fig. 2 ). In favor of this explanation was the fact that the yield increased up to ~20% when the triplex was incubated at 30oC for 24 h.

Both sets of experiments were also done with the triplexes containing the labeled pyrimidine-rich strand. No interstrand cross-links were detected by gel electrophoresis under denaturing conditions which confirms the formation of the cross-links between the Hoogsteen strand and the purine-rich strand. Location of the interstrand cross-links. The platinated triplexes were incubated for 24 h, at pH 5 and at 37oC (third strand I*) or at 30oC (third strand II*). The cross-linked triplexes were separated from the non-cross-linked triplexes by gel electrophoresis under native conditions (buffer TBE, pH 8), conditions in which the non-cross-linked triplexes were dissociated. The cross-linked triplexes were eluted from the gel and then reacted with dimethylsulfate (DMS).

With regard to the triplex with third strand I*, at pH 8.5 DMS reacts with all the G residues but G5 within the 25mer purine-rich strand (Fig. 4 ). This shows that the main cross-link is between the platinated G (strand I*) and G5 (purine-rich strand). It is likely that the minor bands (Fig. 3 , left) are due to the cross-links between the platinated G and the two A residues adjacent to G5 on the 3' and 5' sides, but this has not been definitively proved. At pH 7, all the other G residues are less reactive with DMS. It means that the platinated Hoogsteen strand is paired to the duplex even at neutral pH and thus the triplexes are strongly stabilized by the interstrand cross-links.


Figure 4 . Dimethylsulfate footprinting of the cross-linked triplexes. Autoradiogram of a denaturing 24% polyacrylamide gel showing the reactivity of DMS with the guanine residues within the purine-rich strand of the cross-linked triplexes, the third strand being the oligonucleotide I*. The purine-rich strand was 32 P-labeled at the 5' end. The reaction with DMS was done at pH 8.5 or 7. Prior to the piperidine cleavage step, the samples were incubated in 0.2 M NaCN (basic pH) at 50oC overnight to remove the platinum residues. The arrow, on the right of the figure, indicates the platinated G residue. Lanes G+A and G: Maxam-Gilbert-specific reaction products.

In the case of the triplex with third strand II*, at pH 8.5 DMS reacts with all the G residues except G11 whereas at pH 7, the reactivity of all the G residues is greatly reduced (not shown). Specificity of the cross-linking reaction . To determine whether other sites including a single G residue within a DNA molecule could react with the platinated Hoogsteen strand, the following experiment was done. The double-stranded oligonucleotide (25 bp) was inserted into the plasmid DNA pSP73 (2464 bp). The DNAs with and without the insert, were cleaved by endonuclease Hin dIII (the insert is at 33 bp from the 3' end). The linear DNAs were mixed respectively with the Hoogsteen strand I* 32 P-labeled at the 5' end, incubated at pH 5 or 7 at 4oC for 1 h and then for 24 h at 37oC. The mixtures were analysed on an agarose gel under native conditions. As shown in Figure 5 , radioactivity is only detected at the level of the DNA with the insert. In addition, it was verified that the yields of the cross-linking reaction between the oligonucleotide I* and the duplex (25 bp) free or inserted into the plasmid, were of the same order of magnitude (the plasmid was first cleaved by Hin dIII, 32 P-labeled at the 3' end and incubated with the oligonucleotide I*. Then Eco RV was added to the mixture. The cross-linking reaction was analysed by gel electrophoresis under denaturing conditions. The yield of the cross-linking reaction was estimated from the relative intensities of the two bands corresponding to the cross-linked and non-cross-linked fragments, respectively (not shown)).


Figure 5 . Specificity of the cross-linking reaction. The linear pSP73 DNAs (0.1 [mu]M) either containing or omitting the insert [the purine-rich oligonucleotide (25mer)] were incubated with the 32 P-5'-end-labeled oligonucleotide I* (2 equivalents) in 150 mM NaClO 4 , 5 mM Mg(ClO 4 ) 2 , 5 mM acetate buffer, pH 5 or phosphate buffer, pH 7, at 37oC for 24 h. The products of the reaction were separated by gel electrophoresis (1.4% agarose gel, buffer 40 mM TAE, pH 8.1) and revealed by autoradiography (left) or by fluorescence after staining with ethidium bromide (right).

A similar experiment was done with platinated oligonucleotide II*. Interstrand cross-links were only formed with the inserted target. Intra-strand cross-linking reaction in platinated single-stranded oligonucleotides. The stability of the triplexes depends upon the pH of the solution and the concentration of the free oligonucleotides containing a monofunctional adduct. As a function of time during the cross-linking reaction, the concentration of the free oligonucleotides containing a monofunctional adduct decreased because of the closure of the monofunctional adducts into intrastrand cross-links. This reaction is well-documented ( 17 - 18 ). In the pH range 5-7, we found that the [tau] 1/2 of the monofunctional adducts within the oligonucleotides I* and II* were equal to 5 +- 0.5 h in agreement with the published results ( 19 - 22 ). This competitive intrastrand cross-linking reaction within the free platinated Hoogsteen strand can lead to a decrease of the interstrand cross-linking reaction yield within the triplexes.

Monofunctional trans- [Pt(NH 3 ) 2 (dC)Cl] + adduct

All the experiments just described were also done using oligonucleotide III* as Hoogsteen strand, which contains a single monofunctional trans- [Pt(NH 3 ) 2 (dC)Cl] + adduct.

The presence of the adduct destabilized the triplex. At 37oC and pH 5.5, the presence of two bands of equal intensities in the gel-retardation assays (not shown) indicated a partial dissociation of the triplex. Nevertheless, the cross-linking reaction occurred during the incubation of the triplex (Fig. 6 , left). There are three slowly migrating bands which indicates that at least three different interstrand cross-links are formed. The cross-linking reaction rapidly reaches a plateau, the [tau] 1/2 (monofunctional adduct within the triplex) being ~2 h. After 8 h incubation, addition of oligonucleotide III* to the mixture resulted in the formation of more interstrand cross-links. As discussed in the case of oligonucleotide II*, the low yield of the cross-linking reaction is due, at least in part, to the dissociation of the triplex and to the formation of intrastrand cross-links within free oligonucleotide III*.


Figure 6 . Interstrand cross-linking reaction within the triplex containing the monofunctional adduct trans- [Pt(NH 3 ) 2 (dC)Cl] + . ( Left ) Autoradiogram of a denaturing 24% polyacrylamide gel. ( Right ) Kinetics of interstrand cross-links formation at pH 5. For experimental conditions see legend to Figure 3.

After purification of the triplexes containing an interstrand cross-link, the location of the interstrand cross-links was probed by means of DMS. The reactivity of the G residues within the purine-rich strand is largely decreased as compared with the same G residues within the duplex (Fig. 7 ). This suggests that whatever the position of the monofunctional adduct in the oligonucleotide, it further reacts to form interstrand cross-links within the triplex. It was not attempted to compare the reactivity of the C residues in the two steps of the reaction (formation of monofunctional adducts, formation of interstrand cross-links). However, to make sure that the C residue at the 5' end of oligonucleotide III did not play a major role, the experiment was repeated with an 18mer which had the same sequence as the 19mer III but no C residue at the 5' end. The results (not shown) confirmed the formation of interstrand cross-links at the three sites.


Figure 7 . Dimethylsulfate footprinting of the cross-linked triplexes. Autoradiogram of a denaturing 24% polyacrylamide gel showing the reactivity of DMS with the guanine residues within the purine-rich strand of the cross-linked triplexes, the third strand being the oligonucleotide III* target. The purine-rich strand was 32 P-labeled at the 5' end. The reaction with DMS was done at pH 8.5. Prior to the piperidine cleavage step, the samples were incubated in 0.2 M NaCN (basic pH) at 50oC overnight. The arrows, on the right of the figure, indicate the platinated G residues. Lanes G+A and G: Maxam-Gilbert-specific reaction products.

Finally, the specificity of the cross-linking reaction was proved by comparing the binding of oligonucleotide III* to pSP73 DNA with and without the insert. Interstrand cross-links were only formed with DNA containing the insert.

DISCUSSION

The purpose of this work was to study in triplexes the cross-linking reaction between a Watson-Crick duplex and its complementary Hoogsteen homopyrimidine strand which contained a single transplatin-monofunctional adduct.

The Hoogsteen strand was either the 19mer d(CT 5 CT 3 CT 4 CT 3 ) or the same oligonucleotide in which one C residue (C1 or C7) was replaced by one G residue. The reaction with transplatin was carried out where there was a single monofunctional adduct per oligonucleotide ( trans- [Pt(NH 3 ) 2 (dC)Cl] + at one of the 4 positions 1, 7, 11 or 16, trans- [Pt(NH 3 ) 2 (dG)Cl] + at position 1 or 7, respectively). In the case of oligonucleotides I and II which contain only one G residue, the platination was done at acidic pH (below the pK of C residues) which prevented reaction with the C residues ( 19 , 21 ). In the case of oligonucleotide III which contain several C residues (but no G residues), all the C residues appeared to react with transplatin. Ion-exchange chromatography allows separation of oligonucleotides bearing one adduct from those bearing several adducts but does not allow separation of oligonucleotides III* modified at one of the four C residues. It is not yet possible to platinate only one C or G residue within an oligonucleotide containing several C or G residues. This should no longer be a problem since two very recent and independent reports ( 23 , 24 ) have shown that site-specifically platinated oligonucleotides can be prepared by using automated solid-phase synthesis.

The platinated oligonucleotides are able to bind to the complementary duplex and form interstrand cross-links. The reaction is specific in the sense that no interstrand cross-links were formed with DNA which did not contain the complementary duplex. The platination of the oligonucleotides does not interfere with a parallel orientation of the Hoogsteen strand with respect to the purine-rich strand as shown by the location of the interstrand cross-links. The interstrand cross-links are only formed with the residues within the purine-rich strand, these residues being almost exclusively G residues. A systematic study has still to be done but it is likely that the cross-linking reaction could occur with an A residue if the G[middot]C base pair complementary to the monofunctional adduct were replaced by a A[middot]T base pair. The interstrand cross-links stabilize the triplexes as shown by the protection of the G residues within the purine-rich strand to DMS at pH 7 (in the same experimental conditions, the unplatinated Hoogsteen strand does not bind to the duplex).

The cross-linking reaction is the fastest with the strand containing the monofunctional adduct trans- [Pt(NH 3 ) 2 (dC)Cl] + ([tau] 1/2 ~2 h). However, the apparent yield of interstrand cross-links is only ~35%. Two reasons can explain this low yield. One is that the monofunctional adduct, which prevents the formation of an hydrogen bond, destabilizes the triplex (in our experimental conditions, half-triplex was dissociated). The other reason is a suicide reaction which occurs simultaneously with the cross-linking reaction. This reaction concerns the closure of the monofunctional adduct into intrastrand cross-link within the free platinated Hoogsteen oligonucleotide ([tau] 1/2 ( trans- [Pt(NH 3 ) 2 (dC)Cl] + ) ~5 h). The consequence is a decrease of the concentration of this free oligonucleotide and thus a possible dissociation of the Hoogsteen strand from the triplex.

The highest yield of the cross-linking reaction is obtained with the monofunctional trans- [Pt(NH 3 ) 2 (dG)Cl] + adduct at the 5' end of the Hoogsteen strand (oligonucleotide I*) and the lowest with the same adduct in the middle of the strand (oligonucleotide II*). As discussed, the yield depends upon the stability of the complexes and the competitive suicide reaction of the monofunctional adducts. To a first approximation, the rates of the suicide reactions within the free platinated single-stranded oligonucleotides (I*, II* and III*) are independent of the nature and location of the monofunctional adduct. In our experimental conditions, the triplex with oligonucleotide I* is stable. The rate of the interstrand cross-linking reaction ([tau] 1/2 ~6 h) is three-fold slower than that with trans- [Pt(NH 3 ) 2 (dC)Cl] + . It means that the anti-syn rotation of trans- [Pt(NH 3 ) 2 (dG)Cl] + required for the reaction is not a major obstacle. Interestingly, the anti-syn rotation of trans- [Pt(NH 3 ) 2 (dG)Cl] + at position 7 is also possible. The yield of the reaction was too low to get an estimate of the [tau] 1/2 at 37oC. However, this yield was increased up to 20% at 30oC and thus the dissociation of the triplex rather than a slow cross-linking reaction is the main reason for a low yield at 37oC. These results also suggest that both trans- [Pt(NH 3 ) 2 (dG)Cl] + and trans- [Pt(NH 3 ) 2 (dC)Cl] + ) adducts should be able to form interstrand cross-links when the Hoogsteen strand is a platinated homopurine oligonucleotide.

Can the yield and the rate of the cross-linking reaction be increased? It is clear that the monofunctional adducts decrease the affinity of the Hoogsteen strand for the duplex and subsequently decrease the yield of the reaction. Several papers have described ways to increase the binding strength of the Hoogsteen strand ( 25 - 27 and references therein) and we are currently investigating some of them. The mechanism of the interstrand cross-linking reaction is still unknown. Two main pathways exist for the closure of the monofunctional lesions into bifunctional lesions ( 29 ). One pathway is a direct nucleophilic attack of the G residue within the purine-rich strand on the platinum residue. There are only few examples of such a pathway in platinated nucleic acids ( 19 , 21 , 22 , 28 ). A major point of interest of this pathway is that in principle, the reaction can complete in a few minutes. The more general pathway proceeds through solvent-associated intermediates and the rate-limiting step is hydrolysis of the chloride ion of the monofunctional adduct ( 17 , 20 , 29 ). From the values of the [tau] 1/2 (monofunctional adducts in the triplexes) in the range 2-6 h, we cannot choose one of the two mechanisms. Whatever the mechanism, assays for accelerating the cross-linking reactions are in progress by replacing transplatin by transplatin derivatives with various leaving or non-leaving groups.

Concerning the anti-gene strategy, in addition to the suicide reaction of the free platinated oligonucleotides, another problem specific to platinated oligonucleotides is that the cells contain several compounds such as glutathione or proteins which react strongly with platinum(II) complexes ( 17 , 20 , 30 ). We have verified that glutathione or thiourea react poorly with platinum(II) residues within the triplexes whereas they do react with the monofunctional adducts within the single-stranded oligonucleotides. Although several problems still exist, it seems possible that the platinated oligonucleotides could serve to improve the efficiency of the anti-gene oligonucleotides in modulation of gene expression and also as a tool in molecular biology.

ACKNOWLEDGEMENTS

We are indebted to Dr M. Boudvillain and J. M. Malinge for helpful discussions. This work was supported in part by la Ligue contre le Cancer, l'Association pour la Recherche sur le Cancer, l'Agence Nationale de Recherche sur le Sida and Deutsche Forschungsgemeinshaft.

REFERENCES

1 Cheng, Y-K. and Pettitt, B.M. (1992) Prog. Biophys. Mol. Biol., 58, 225-257. MEDLINE Abstract

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

3 Mirkin, S.M. and Frank-Kamenetskii, M.D. (1994) Annu. Rev. Biophys. Biomol. Struct., 23, 541-576. MEDLINE Abstract

4 Nielsen, P.E. (1995) Annu. Rev. Biophys. Biomol. Struct., 24, 167-183. MEDLINE Abstract

5 Plum, G.E., Pilch, D.S., Singleton, S.F. and Breslauer, K.J. (1995) Annu. Rev. Biophys. Biomol. Struct., 24, 319-350. MEDLINE Abstract

6 Sun, J.S., Garestier, T. and Hélène, C. (1996) Curr. Opin. Struct. Biol., 6, 327-333. MEDLINE Abstract

7 Vlassov, V.V., Gorn, V.V., Ivanova, E.M., Kazakov, S.A. and Mamaev, S.V. (1983) FEBS Lett., 162, 286-289.

8 Chu, B.C.F. and Orgel, L.E. (1990) Nucleic Acids Res., 18, 5163-5171.

9 Gruff, E.S. and Orgel, L.E. (1991) Nucleic Acids Res., 19, 6849-6854. MEDLINE Abstract

10 Dieter-Wurm, I., Sabat, M. and Lippert, B. (1992) J. Am. Chem. Soc., 114, 357-359.

11 Krizanovic, O., Sabat, M., Beyerle-Pfnür, R. and Lippert, B. (1993) J. Am. Chem. Soc., 115, 5538-5548.

12 Schreiber, A., Lüth, M.S., Erxleben, A., Fusch, E.C. and Lippert, B. (1996) J. Am. Chem. Soc., 118, 4124-4132.

13 Anin, M.F. and Leng, M. (1990) Nucleic Acids Res., 18, 4395-4400. MEDLINE Abstract

14 Brabec, V., Reedijk, J. and Leng, M. (1992) Biochemistry, 31, 12397-12402. MEDLINE Abstract

15 Maxam, A.M. and Gilbert, W. (1977) Proc. Natl. Acad. Sci. USA, 74, 560-564. MEDLINE Abstract

16 Lemaire, M.A., Schwartz, A., Rahmouni, A.R. and Leng, M. (1991) Proc. Natl. Acad. Sci. USA, 88, 1982-1985.

17 Lepre, C.A. and Lippard, S.J. (1990) In Eckstein, F. and Lilley, D.M.J. (eds), Nucleic Acids and Molecular Biology, Springer, Berlin, Germany, Vol. 4, pp. 9-38.

18 Reedijk, J. (1987) Pure Appl. Chem., 59, 181-192.

19 Comess, K.M., Costello, C.E. and Lippard, S.J. (1990) Biochemistry, 29, 2102-2110. MEDLINE Abstract

20 Bancroft, D.P., Lepre, C.A. and Lippard, S.J. (1990) J. Am. Chem. Soc., 112, 6860-6871.

21 Dalbiès, R., Payet, D. and Leng, M. (1994) Proc. Natl. Acad. Sci. USA, 91, 8147-8151.

22 Dalbiès, R., Boudvillain, M. and Leng, M. (1995) Nucleic Acids Res., 23, 949-953.

23 Schliepe, J., Berghoff, U., Lippert, B. and Cech, D. (1996) Angew. Chem., 108, 705-707.

24 Manchanda, R., Dunham, S.U. and Lippard, S.J. (1996) J. Am. Chem. Soc., 118, 5144-5145.

25 Wang, S.H. and Kool, E.T. (1995) Biochemistry, 34, 4125-4132.

26 Priestley, E.S. and Dervan, P.B. (1995) J. Am. Chem. Soc., 117, 4761-4765.

27 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.

28 Boudvillain, M., Dalbiès, R., Aussourd, C. and Leng, M. (1995) Nucleic Acids Res., 23, 2381-2388. MEDLINE Abstract

29 Lippard, S.J. and Berg, J.M. (1994) in Principles of Bioinorganic Chemistry, University Science Books, Mill Valley, USA.

30 Eastman, A. and Barry, M.A. (1987) Biochemistry, 26, 3303-3307. MEDLINE Abstract

Return

* To whom correspondence should be addressed. Tel: +33 238 51 55 84; Fax +33 238 63 15 17; Email: leng@cnrs-orleans.fr
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Print PDF (96K) 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 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 arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Colombier, C
Right arrow Articles by Leng, M
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Colombier, C
Right arrow Articles by Leng, M
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?