| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
* Dipartimento di Scienze e Tecnologie Biomediche, Università degli Studi di Udine, 33100 Udine. Italy; and
Laboratoire de Biophysique, Muséum National d’Histoire Naturelle, INSERM U201-CNRS UMR8646, 75231 Paris Cedex 05, France
1Correspondence: Laboratoire de Biophysique, Muséum National d’Histoire Naturelle, INSERM U201-CNRS UMR8646, 43 rue Cuvier, 75231 Paris Cedex 05, France. E-mail: giovanna{at}mnhn.fr
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: peptide nucleic acid • triplex • PPT/HIV-1
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2
). Triplex-forming molecules (TFMs) specifically recognize oligopyrimidine·oligopurine sequences present in duplex DNA and interact with the oligopurine strand by Hoogsteen or reverse Hoogsteen hydrogen bonds. Triplexes are formed when using oligonucleotides or oligonucleotide analogs (TFOs) and peptide nucleic acids (PNAs), which are synthetic oligonucleotide mimics in which the deoxyribose phosphate backbone has been replaced by a N-(2-aminoethyl)glycine polymer to which the nucleobases are attached (for a review, see 3
).
Sequence-specific recognition and targeting of double-stranded DNA have applications in basic research and therapeutics. Modulation of transcription by TFMs has been described in vitro and in cell cultures for several plasmid targets transiently transfected into living cells (4
5
6
7
8
9)
and endogenous genes (10
11
12
13)
. Triplex formation has also been used to induce site-directed mutagenesis and promote recombination in a site-specific manner (14
15
16
17
18
19)
. The first evidence of a triplex-based activity in animals was provided recently (20)
: site-specific genomic modifications have been successfully introduced in somatic cells of adult mice. It is important to understand how this unusual structure interferes with DNA metabolic processes other than transcription, especially DNA replication. So far, direct measurements of the effect of triplexes on DNA replication in cells have not been reported.
Triplexes formed with TFOs were demonstrated to arrest several DNA polymerases in vitro (21
22
23)
. Inhibition of DNA synthesis was observed not only when triplexes blocked the progression of DNA polymerase, but also when the polymerization primer was involved in triplex formation (24)
. In contrast, local duplexes (DNA:DNA complex) formed on single-stranded DNA templates did not act as barriers for replication (25)
. However, duplex-forming PNAs (DNA:PNA complex) have been described to arrest in vitro single-stranded DNA replication (26)
.
The effects of triplexes on purified DNA helicases, enzymes that preceded DNA polymerases in the fully assembled replication holoenzyme, are described (27
28
29)
. The presence of DNA triplexes significantly inhibited DNA unwinding by SV40 large T antigen helicase except when triplex formation involved a TFO with a 3' dangling end (29)
. Duplex-forming PNAs have been described as inhibitors of the UL9 DNA helicase of the herpes simplex virus (30)
.
All these studies concern in vitro experimental systems using purified molecules (DNA polymerases or helicases) and are far from the complexity of the DNA replication machinery present in the cell nucleus. Only one study has shown that an oligonucleotide intercalator conjugate targeted to the origin of replication of SV40 inhibited viral replication in cell culture (31)
. However, the results did not explain whether the inhibition was due to a block of the initiation or the progression of the DNA replication process or to a competition with binding of specific proteins to the origin of replication; involvement of the triplex structure in the observed inhibition was not clearly established.
In our study, we designed a cellular system suitable for quantitative determination of site-specific inhibition of DNA replication. An oligopyrimidine·oligopurine sequence suitable for triplex formation (the 16 bp-long HIV-1 ‘polypurine tract’ (PPT) sequence) was introduced in a plasmid downstream of the SV40 origin of replication and replicated DNA was directly quantitated. We demonstrated that the DNA replication process could be inhibited in cells by triple-helical complexes formed with 1) a psoralen-conjugated TFO that was cross-linked to the double-stranded DNA target by photoactivation; 2) a bis-PNA molecule that was noncovalently but tightly bound to the duplex DNA target by strand displacement of the pyrimidine-containing DNA strand[(PNA-(DNA)-PNA triplex]. In contrast, triplex-forming oligonucleotide analogs such as N3'-P5' oligophosphoramidates, which have been shown to arrest the transcription machinery in cells, did not inhibit the DNA replication process. These results open new perspectives for the future design and use of specific modulators of intracellular DNA metabolism.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Plasmids
Plasmids containing two copies of the oligopyrimidine·oligopurine sequence suitable for triplex formation (the polypurine tract sequence of HIV-1, abbreviated as PPT) were obtained by cloning the appropriate oligonucleotides into the vector pSV2neo, a commercial vector (Clontech, Palo Alto, CA) containing the 400 bp core region of the SV40 replication origin. The 37 bp double-stranded oligonucleotides containing the PPT sequence were cloned on both sides of the core origin of replication at the HindIII and AccI restriction sites. The new construct is called pSV/PPT(+) (Fig. 1
A) and its parent plasmid is designated by PSV/PPT(-). Escherichia coli XL1 blue dam+ strain (Stratagene, San Diego, CA) was used as competent cell for transformation. Isolation of clones and characterization of plasmids by DNA dideoxy sequencing were done by standard techniques.
|
The plasmid pSV/PPT(-) without any insert was used as a control for the demonstration of the role of triplex formation in the observed inhibition as well as an internal control in the quantification of replicated DNA by PCR, as described later.
Triplex-forming molecules (TFMs)
The psoralen-modified triplex-forming oligonucleotide (Pso-TFO) was prepared as described previously (32)
; the psoralen derivative was attached to the 5' end. The phosphodiester oligonucleotide was 3'-modified by incorporation of a hexylamino group in order to resist nuclease-mediated degradation. The sequence of the 15TCG TFO directed to the HIV-1/PPT region is 5' TTTTCTTTTGGGGGG 3', where C is 5-methyl cytosine.
PNAs were obtained from PerSeptive Biosystems (Framingham, MA). PNAs were HPLC purified and lyophilized, resuspended in TE buffer, aliquoted, and kept at -20°C. The PNA sequences are 9TC: H-Lys-TTTTCTTTT-CONH2; 9–9TC: H-Lys-TTTTCTTTTOOOTTTTJTTTT-CONH2, where ‘O’ is an 8-amino-3,6-dioxaoctanoic acid linker and J is pseudoisocytosine. Lysine (Lys) was added because it increases PNA solubility. The PNA 9–9TC is a bis-PNA molecule designed to form a PNA-DNA-PNA triplex clamp, as previously reported (33
, 34)
.
Triplex formation
The TFOs were incubated with the target DNA in a 10 mM tris-HCl (pH7.0) buffer containing 50 mM NaCl and 10 mM MgCl2. For covalent triplex induction in vitro and in cells, irradiation with UV light was performed with a 365 nm monochromatic system, as described (32
, 35)
. When irradiation was performed in vitro before cell transfection, the excess of noncovalently associated Pso-TFO was removed by filtration on Sephadex G-50 columns. The amount of adducts was determined by DraI protection and PCR assay. PSV/PPT(+) plasmid containing 100% of covalent triplex at the PPT site (called PSV/PPT(+)XL) was diluted with untreated pSV/PPT(+) plasmid to obtain mixtures with defined percentages of covalent triplex at the PPT site (Fig. 1)
.
PNAs were incubated with the target DNA (pSV/PPT(+) plasmid) in 20 µl TE (10 mM tris-HCl pH 8, 1 mM EDTA) overnight at 37°C before cell transfection. PNA binding was analyzed by DraI protection assay.
Restriction enzyme (DraI) protection assay for triplex detection
A typical assay was performed by incubating for 2 h at 37°C the pSV/PPT(+) plasmid treated or not with the TFM in a final volume of 10 µl in either a low salt-containing buffer (TE buffer: 10 mM tris-HCl pH 8, 1 mM EDTA) or high salt-containing buffer (DraI digestion buffer, named H buffer: 50 mM potassium acetate, 20 mM tris-acetate pH 7.9, 10 mM magnesium acetate). Subsequently, 20 units (1 µl) of the restriction enzyme DraI (New England Biolabs, Beverly, MA) were added in a final volume of 15 µl of digestion buffer. Incubation was continued for 2 h at 37°C. Samples were loaded on a 8% PAGE in TBE and stained with ethidium bromide.
DNA transfection
Transient transfection of COS-1 cells was carried out by treatment with DOTAP liposomal transfection reagent (Roche, Nutley, NJ) or by electroporation.
DOTAP-mediated transfection. Plasmids (total volume of 20 µl in 10 mM HEPES pH 7.2, 50 mM NaCl, 10 mM MgCl2) containing or not the TFO were mixed with DOTAP (15 µl+ 35 µl HEPES 10 mM) according to conditions indicated in the instructions. For each single transfection, 
2 x 105 cells were transfected with 400 ng of each plasmid [pSV/PPT(-) and pSV/PPT(+)] and incubated in 48 multiwell plates in 0.5 ml of culture. Final TFO concentrations in the culture medium are indicated in the text and figures.
Electroporation. For each experimental condition, COS-1 cells (5x105) were resuspended in 0.8 ml PBS containing 400–800 ng of each plasmid (with or without the TFM) in a 0.4 cm electrode gap cuvette, electroporated (250 V/cm, 950 microfarads), and incubated in 12-well plates by adding 1 ml of DMEM medium containing 10% fetal bovine serum in 5% CO2 at 37°C. After 24 h, cells were harvested, washed twice with PBS, and plasmid DNA was extracted.
Replicated DNA purification
Plasmid DNA was extracted from COS-1 cells following the standard Hirt procedure. The DNA replicated within the cells was purified from the transfected unreplicated DNA by digestion with Dpn I restriction enzyme (New England Biolabs); this enzyme recognizes only the bacterial dam+ methylation pattern of Dpn I sites and is used to discriminate between replicated and nonreplicated DNA.
Quantification of replicated DNA
A PCR assay was used to quantitate the ratio between replicated DNA from the two plasmids [the pSV/PPT(-) plasmid being 74 bp (=37 bpx2) shorter than pSV/PPT(+); see Fig. 1A
]. One set of two PCR primers (AccDX and AccSX) located in the flanking region of the AccI site (where the 37 bp triplex site has been inserted in the pSV/PPT(+) plasmid) was used. The sequences of the primers were AccDX: 5' CTTACGCATCTGTGCGGTAT 3'; AccSX: 5' CGGTCACAGCTTGTCTGTAA 3'. Amplification products obtained with these primers are a 204 bp fragment from the plasmid pSV/PPT(-) and a 241 (=204+37) bp fragment from the plasmid pSV/PPT(+). Alternatively, another set of primers (HinDX and HinSX) was synthesized on both sides of the restriction site HindIII in order to amplify a fragment of 406 bp from the plasmid pSV/PPT(-) and a fragment of 443 (=406+37) bp from the plasmid pSV/PPT(+). The sequences of the primers were HinDX: 5' TTGGAGGCCTAGGCTTTTGC 3', HinSX: 5' ATGCGAAACGATCCTCATCC 3'. Identical results were obtained with the two set of primers. Conditions for PCR reactions were 1) 50 s at 94°C, 2) 30 s at 60°C, and 3) 30 s at 72°C; 35 cycles of amplification were performed with 1.25 units of Taq polymerase (Perkin-Elmer, Norwalk, CT) in a final volume of 50 µl and 2 mM Mg2+ concentration. The two amplified products were separated using an 8% PAGE in TBE. The ratio between the two PCR products obtained with a set of primers, quantified by PhosphorImager analysis, exactly reflects the initial ratio between the two plasmids [pSV/PPT(-) and pSV/PPT(+)] before amplification, independent of the number of cycles or the presence of any inhibitor of the PCR reaction (36)
.
Plasmidic DNAs isolated from cell cultures and digested by Dpn I restriction enzyme were used as template for PCR reactions. Results are presented as the mean (±SD) of the percentage of DNA replication inhibition of duplicate or triplicate PCR measurements from a representative experiment that was repeated independently at least three times.
For every Dpn I digestion, we verified that the transfected plasmidic DNA was completely removed. A test tube containing the same DNA amount as the one present in the mixture before cell transfection was digested in parallel with the sample tubes and the PCR assay was run; we estimated that the Dpn I digestion was complete when no PCR signal could be detected in the test tube.
The effect of TFMs on the PCR reaction was also tested. Replicated DNA from cells untreated by TFM was mixed with an amount of TFM corresponding to a 100% TFM delivery and our standard PCR reaction was run; under these conditions, no effect of TFM could be detected. These data allowed us to exclude a TFM effect during the PCR procedure and evaluate its actual effect on cellular replication.
We evaluated the precision of the PCR assay. We performed PCR amplification to determine the amount of cross-links (XL) at the PPT sites on the pSV/PPT(+) plasmid after in vitro treatment with Pso-TFO and irradiation; the AccDX and AccSX primers bind on each side of the XL site. Under the conditions used for amplification, a complete inhibition of PCR was observed indicating a 100% cross-linked pSV/PPT(+). Then we mixed the plasmids pSV/PPT(-), pSV/PPT(+), and the 100% cross-linked pSV/PPT(+) (called pSV/PPT(+)XL) in various ratios (1/0:1; 1/0.25:0.75; 1/0.5:0.5; 1/0.75:0.25; 1/1:0). These mixtures were amplified by PCR as described above. As expected, the pSV/PPT(+)XL plasmid did not give rise to any PCR product, only the non-cross-linked pSV/PPT(+) yielded an amplification product. The ratio between the two amplified products (204 and 241 bp) strictly reflects the ratio between the pSV/PPT(-) and the non-cross-linked pSV/PPT(+) plasmids (respectively) present in the initial mixture, in agreement with competitive PCR principles.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 37)
, was cloned on both sides of the SV40 replication origin in order to evaluate whether a triplex structure flanking the origin of replication could inhibit DNA replication (Fig. 1A
). Different TFMs (TFOs or PNAs) were tested in this replication cellular assay. The plasmid containing the two PPT sequences [plasmid called pSV/PPT(+)] was cotransfected together with an equal amount of the plasmid without the PPT inserts [pSV/PPT(-)] in the presence or absence of various TFMs. The two plasmids are replicated by the DNA replication machinery of COS-1 cells. The replicated DNA of both plasmids was isolated and the ratio between the amounts of replicated DNA from the two plasmids was measured (see Materials and Methods). This system is able to quantify the inhibitory effect of triplex formation on pSV/PPT(+) replication. In fact, the plasmid pSV/PPT(-) constituted an internal control for the demonstration of triplex involvement in the observed inhibition of DNA replication, since it contains exactly the same sequence as the pSV/PPT(+) plasmid except for the 37 bp inserts containing the PPT target for the TFMs downstream the origin of replication. If treatment with TFM induces selective inhibition of replication of the pSV/PPT(+) plasmid compared with the pSV/PPT(-) plasmid, it can be concluded that the TFM acts by binding to the PPT sequence because any putative interaction of the TFM with cellular components other than the inserted target sequence is identical for both plasmids. The use of such a modified target sequence with the same TFM is aimed at providing direct evidence for the TFM mechanism of action whereas targeting the wild-type sequence and changing the TFM sequence, as commonly done, are equivocal because all other potential interactions of the TFM within the cell are also modified.
To determine the ratio between the two replicated plasmid DNAs (different in length by only 74 bp=37 bpx2), we performed PCR amplifications using one set of primers (AccDX and AccSX) located in the flanking region of the cloning site AccI, used to introduce the 37 bp triplex insert in the pSV/PPT(+) plasmid (Fig. 1A
). Alternatively, we used a set of primers (HinDX and HinSX) located in the flanking region of the other cloning site HindIII used here. This technique allowed us to coamplify the two DNA species derived from pSV/PPT(+) and pSV/PPT(-) in the same tube using the same set of primers. The amplified products could be analyzed simultaneously (8% PAGE) because the PCR product derived from the pSV/PPT(+), containing the PPT insert, is 37 bp longer than the product derived from pSV/PPT(-) (Fig. 1B
). The ratio between the two PCR products measures the ratio between the two replicated plasmids (36)
and therefore allowed us to quantify the effect on DNA replication of a triplex formed on the PPT sequence.
Quantification of triplex-induced inhibition of DNA replication
To evaluate the potency of our system for studying DNA replication inhibition, the pSV/PPT(+) plasmid was covalently modified specifically at the PPT sites: the 15TCG TFO conjugated to a psoralen molecule (Pso-15TCG) was used for triplex-directed covalent modification of pSV/PPT(+) plasmid under UV irradiation (see Materials and Methods). The Pso-15TCG oligonucleotide was previously demonstrated to form a covalent complex with the PPT target region after photoactivation at 365 nm (32
, 35)
. This reaction involves the formation of a cross-link between the two DNA strands at the 5'-TpA-3' sequence present at the duplex-triplex junction (Fig. 1A
). The presence of cross-links formed on the pSV/PPT(+) plasmid at the TpA sequences of the PPT targets flanking the SV40 origin of replication should prevent any opening of the double helix and its replication in cells. On the other hand, the plasmid pSV/PPT(-) does not contain any target for the TFO and should be replicated as in the absence of the TFO.
We evaluated the effect of cross-links present at the triplex sites on DNA replication in the nucleus (Fig. 1)
. We prepared mixtures with determined amounts of cross-links containing the plasmids pSV/PPT(-), pSV/PPT(+), and the 100% cross-linked pSV/PPT(+) (called pSV/PPT(+)XL) in various ratios (1/1:0; 1/0.75:0.25; 1/0.5:0.5; 1/0.25:0.75; 1/0:1) and COS-1 cells were transfected with these various mixtures. The low molecular weight DNA was extracted 24 h after transfection, digested with Dpn I to eliminate the transfected DNA, and intracellular replicated DNA was quantitated by PCR (Fig. 1B
). The percentage of DNA replication inhibition was reported as a function of the percentage of cross-linked pSV/PPT(+)XL plasmid present in the transfected mixture. As shown in Fig. 1C
, the extent of DNA replication in the cells was linearly related to the quantity of intact (not cross-linked) pSV/PPT(+) in the initial transfected mixture. These results demonstrate that the covalent triplexes generated in vitro by photoactivation of the psoralen moiety of the TFO-conjugate can produce an inhibition of intracellular plasmid DNA replication 24 h after transfection and that our experimental system is appropriate for quantitative evaluation of the inhibitory activity of TFM directed against the oligopyrimidine·oligopurine PPT sequence on intracellular DNA replication.
Replication inhibition induced by TFO-psoralen conjugates
We then tested the ability of TFO-psoralen conjugates to act as inhibitors of DNA replication when a noncovalent triplex was transfected into cells and the covalent triplex was formed directly within cells by UV irradiation. The experiments were performed not only with the Pso-15TCG oligonucleotide containing a phosphodiester backbone (po), but also with a modified one in which the phosphodiester linkages were changed to N3'-P5' phosphoramidate ones (np). Oligophosphoramidates have been demonstrated to form strong triplexes with the PPT target sequence and to inhibit DNA transcription elongation in vitro and in cells (13
, 38)
.
An equimolar mixture of plasmids pSV/PPT(+) and pSV/PPT(-) was incubated with increasing concentrations of Pso-15TCG oligonucleotide and transfected into COS-1 cells using cationic lipids, and UV irradiation was performed 2 h later. The DNA was extracted after 24 h and quantified with the PCR procedure described above. The Pso-15TCG (po) and (np) oligonucleotides were able to selectively inhibit the replication process of the plasmid pSV/PPT(+) in a dose-dependent manner (Fig. 2
). The phosphoroamidate TFO was more efficient as an inhibitor of DNA replication than the phosphodiester one: 60% inhibition was obtained with 0.2 µM of np TFO vs. 10% for po TFO. This result confirms the higher efficiency and stability of the triple-helical structure formed by the phosphoroamidate oligonucleotides, already described in vitro and in physiological intracellular conditions through measurements of transcription inhibition.
|
Experiments were also performed in the absence of irradiation to evaluate the inhibitory effect on replication of noncovalent triplexes formed with TFO-psoralen conjugates. No effect on DNA replication within the cells could be detected in the absence of photoadduct formation even at the highest concentration tested, which produced 70% inhibition after irradiation. In our experimental system, neither the natural phosphodiester (po) oligonucleotide nor the phosphoramidate (np) oligonucleotide interfered with the replication machinery unless the noncovalent triplex was converted into a covalent one by photo-induced cross-linking of psoralen at the triplex binding site.
Replication inhibition induced by triplex-forming PNA
Oligopyrimidine PNAs are known to form stable PNA-DNA-PNA triplexes involving double helix invasion and strand displacement reactions; therefore, we tested the ability of triplex-forming PNAs to inhibit the replication process with the same experimental system used for the TFOs.
Two different PNA molecules were targeted to the PPT sequence: the 9TC and the 9–9TC (see Materials and Methods). The 9TC and the 9–9TC recognize nine contiguous bases (A4GA4) of the PPT sequence and their binding should involve DNA opening with formation of a PNA-DNA-PNA triple helix with the (A4GA4) sequence and a single-stranded oligopyrimidine region (see Fig. 4
, top). The 9TC should form a triplex on the (A4GA4) symmetric sequence involving two PNA molecules bound in opposite orientations. The 9–9TC is a bis-PNA that contains two PNA sequences of nine monomers linked to each other by a flexible linker; this molecule should bind as a clamp to the same A4GA4 sequence as the 9TC PNA by forming both Watson-Crick and Hoogsteen hydrogen bonds. Furthermore, the Hoogsteen strand of 9–9TC contains pseudoisocytosine instead of cytosine in order to eliminate the pH dependence of binding due to the requirement for cytosine protonation to form Hoogsteen hydrogen bonds (33
, 34)
. Bis-PNAs have been shown to have a better capacity for strand invasion and greater stability than a simple PNA. The triplex-forming PNAs are designed to bind to the lagging strand of the replication fork on both sides of the origin of replication at the PPT region.
|
To characterize the binding of PNAs to the target PPT, we used an assay based on the sequence-specific inhibition of restriction enzyme cleavage by triplex formation (39)
. We incubated the pSV/PPT(+) plasmid with the different PNAs in a low-salt buffer (TE buffer; see Materials and Methods) at 37°C, when the buffer required for enzyme activity was added. The plasmid was digested with the restriction enzyme DraI that recognizes a site overlapping the triplex site over 3 bp (5' TTT
AAA 3'). In the absence of PNA, complete DraI cleavage of pSV/PPT(+) plasmid produced nine fragments (Fig. 3
, lane 1; the shortest 19 bp fragment was not visualized in the gel). In the presence of PNA, three fragments (505, 987, 1556 bp) disappeared and one longer (3048 bp) appeared; cleavage by the restriction endonuclease was inhibited at the PPT sites, thus demonstrating the binding of PNA to its target (Fig. 3B
, lanes 2, 3). The specificity of binding was demonstrated by the absence of inhibition of DraI cleavage at other DraI sites of the plasmid. The same experiment was performed by changing the incubation conditions of plasmid and PNAs: incubation was performed directly in the high-salt digestion buffer (H buffer: 50 mM potassium acetate pH 7.9, 20 mM tris-acetate, 10 mM magnesium acetate). No inhibition of DraI activity was observed under these conditions (Fig. 3B
, lanes 4, 5). These results are consistent with the formation of a PNA-DNA-PNA complex previously reported involving a strand displacement reaction after a local DNA opening of the PPT sequence (as depicted on Fig. 4
), which is favored at low salt concentration. This triple-helical structure differs from that observed with TFOs, which bind in the major groove of the preexisting duplex.
|
An equimolar mixture of plasmids pSV/PPT(+) and pSV/PPT(-) was incubated in a low-salt buffer (TE) in the presence of different concentrations of PNAs, 9–9TC and 9TC at 37°C (see Materials and Methods). Then, the mixture was electroporated into COS-1 cells and DNA replication inhibition measured as already described. Figure 4
describes the inhibitory effect of the bis-PNA 9–9 TC on DNA replication, whereas 9TC PNA did not exhibit any inhibition up to 0.8 µM. These results indicate that DNA replication can be inhibited by a triplex-forming bis-PNA. However, the stable complex formed in vitro might be partially dissociated when transfected into cells, since only 60% inhibition was observed after 24 h at 0.8 µM of bis-PNA, a concentration at which the DraI cleavage assay revealed 100% triplex formation in vitro in the low-salt buffer. We used a 9 bp sequence as a target in our experiments. The partial dissociation of the complex formed by the bis-PNA might be slowed down considerably when longer sequences are targeted.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). In our system, a small amount of repair might be occurring that is undetectable considering the sensitivity of our PCR assay. But our results support the mutation pattern resulting from noncovalent triplex formation obtained in isolated cells and in vivo (16
, 20
, 40)
: single base insertions or deletions within a stretch of eight G·C base pairs are mainly observed (80% of the mutations). These data are consistent with strand slippage events by inhibition or stalling of the DNA polymerase that can occur during DNA repair synthesis but also during DNA replication.
Noncovalent triplexes formed with TFOs targeted to the HIV-1/PPT sequence have exhibited a significant inhibitory effect on transcription elongation (9
, 13
, 38)
but had no effect on DNA replication in our experimental system. This result suggests that replication is much more difficult to block than transcription. For transcription elongation, it has been discussed that inhibition can be achieved whatever the target orientation (38)
; how this parameter will influence replication inhibition remains to be determined. The quantitative assay established in the present work will help and facilitate the evaluation of potential candidate molecules that could specifically interfere with cellular DNA replication.
In addition, the results presented here demonstrate that the potential effect of TFMs on the replication machinery must be considered and evaluated when these TFMs are used for modulation of gene expression. Their binding to a specific region of genomic DNA could modulate not only the transcriptional activity of the target gene, but also DNA replication of the same region (as is the case for PNAs and Pso-TFOs on irradiation), with potential effects on cell viability or other cellular processes such as recombination or repair.
The efficiency of PNAs as antigene molecules is limited by the necessity to open the DNA target sequence in order to form a stable triplex. To bypass this limitation and to evaluate the intrinsic ability of triplexes formed with PNA to interfere with the targeted biological process of interest in the cellular environment, PNA-DNA-PNA complexes were preformed under low salt conditions and then transfected into cells (as in the present study). Various factors have been described to influence PNA strand invasion into duplex DNA at physiological salt concentrations, such as negative supercoiling (41)
or transcriptional activity associated with the presence of a transcription ‘bubble’ (42)
. Since DNA replication also involves a replication bubble and DNA opening, it is likely that the replication process could increase the efficiency of triple helix formation by PNA on genomic targets and may be suitable for replication inhibition. A few studies have described the use of PNAs to target endogenous or integrated genes and their ability to induce the expected biological response in cell cultures, e.g., site-directed mutagenesis at the triplex site (16)
, transcription activation (12)
, and transcription inhibition (43)
. Progress has recently been made in overcoming the limited ability of PNAs to reach cell nuclei, particularly by using a covalent linkage to peptides (43
, 44)
. This set of results clearly demonstrates that PNA binding to DNA target sequences within cell nuclei might not be as difficult to achieve as sometimes believed.
The possibility of blocking replication in cells by triplex formation described here (with bis-PNAs, or Pso-TFO conjugates on irradiation) opens interesting avenues to evaluate the cellular consequences of inhibiting the replication process at specific sites on selected chromosomes.
|
ACKNOWLEDGMENTS |
|---|
Received for publication June 4, 2001.
Revision received August 16, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. S. Keating, N. Toor, P. S. Perlman, and A. M. Pyle A structural analysis of the group II intron active site and implications for the spliceosome RNA, January 1, 2010; 16(1): 1 - 9. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Duca, P. Vekhoff, K. Oussedik, L. Halby, and P. B. Arimondo The triple helix: 50 years later, the outcome Nucleic Acids Res., September 1, 2008; 36(16): 5123 - 5138. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.-H. Kim, P. E. Nielsen, and P. M. Glazer Site-directed gene mutation at mixed sequence targets by psoralen-conjugated pseudo-complementary peptide nucleic acids Nucleic Acids Res., December 3, 2007; 35(22): 7604 - 7613. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Ge, J. E. Heinonen, M. G. Svahn, A. J. Mohamed, K. E. Lundin, and C. I. E. Smith Zorro locked nucleic acid induces sequence-specific gene silencing FASEB J, June 1, 2007; 21(8): 1902 - 1914. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A. Vetcher and R. D. Wells Sticky DNA Formation in Vivo Alters the Plasmid Dimer/Monomer Ratio J. Biol. Chem., February 20, 2004; 279(8): 6434 - 6443. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
