TY - JOUR
AU - Neumann, Ronald D.
AB - Abstract A parallel binding motif 16mer triplex-forming oligo-nucleotide (TFO) complementary to a polypurine-polypyrimidine target region near the 3′-end of the SupF gene of plasmid pSP189 was labeled with [5- 125 l]dCMP at position 15. Following triplex formation and decay accumulation, radiation-induced site-specific double-strand breaks (DSBs) were produced in the pSP189 SupF gene. Bulk damaged DNA and the isolated site-specific DSB-containing DNA were separately trans-fected into human WI38VA13 cells and allowed to repair prior to recovery and analysis of mutants. Bulk damaged DNA had a relatively low mutation frequency of 2.7 − 10 −3 . In contrast, the isolated linear DNA containing site-specific DSBs had an unusually high mutation frequency of 7.9 − 10 −1 . This was nearly 300-fold greater than that observed for the bulk damaged DNA mixture, and >1.5 − 10 4 -fold greater than background. The mutation spectra displayed a high proportion of deletion mutants targeted to the 125 I binding position within the SupF gene for both bulk damaged DNA and isolated linear DNA. Both spectra were characterized by complex mutations with mixtures of changes. However, mutations recovered from the linear site-specific DSB-containing DNA presented a much higher proportion of complex deletion mutations. Introduction The ability of oligonucleotides to form triple helical structures raises the possibility of developing a new class of pharmacologic agents which act directly on DNA to permanently disrupt or modify gene function ( 1–5 ). Target sequences for binding triple-helix-forming oligonucleotides (TFOs) consist of short polypurine stretches within DNA duplexes ( 6–8 ). Although such sequences are abundant in mammalian genomes, the exquisite sequence specificity of TFOs permit selective binding to unique targets even in genomic DNA ( 3 , 5 , 9–12 ). DNA targeting by TFOs to which mutagenic agents have been attached provides a means by which site-specific mutagenesis might be accomplished in a gene-directed manner. This has recently been demonstrated for TFO-bound psoralen and nitrogen mustards ( 1–3 , 11 ). These studies prove the ability of TFOs to mediate highly precise positioning of mutagenic compounds. However, the lesions induced by chemical mutagens are repaired efficiently, and thus result in relatively low overall mutation frequencies. Consequently, even if TFO-mediated targeting is efficient, modulation of target gene activity will only occur infrequently. In order for TFO-directed mutagenesis to be an effective anti-gene targeting mechanism, the mutagen employed should induce a complex lesion with a concomitant loss of coding information at the damage site. Such lesions are exemplified by multiply damaged sites (MDS) involving DNA double-strand breaks (DSBs). These DSBs are products of ionizing radiation, and are considered to be the most mutagenic and toxic radiation products ( 13–16 ). The majority of low-linear-energy transfer (LET) radiation-induced DNA damage is indirect damage that consists of base modifications and single-strand breaks (SSBs), all of which are repaired efficiently. In contrast, the direct effects of densely ionizing high-LET radiation such as α-particles, is thought to lead to the formation of numerous complex MDS ( 13 , 16–20 ). The most complex member of this class of lesions is a DSB with base modifications, and/or loss, proximal to the break ends. Similar lesions are expected to result when radionuclides which decay by electron capture (EC) and/or internal conversion (IC), decay in close proximity to, or within DNA. Such radionuclides produce extremely charged daughter atoms and emit numerous low-energy Auger electrons resulting in highly localized energy deposition at the decay site, thus producing localized effects similar to high-LET radiation ( 21 , 22 ). When located extracellularly or cytoplasmically, Auger emitters such as 125 I exhibit relatively low radiotoxicity, similar to that of low-LET radiation ( 23 ). However, when 125 I is placed within, or in very close proximity to nuclear DNA, its decay imparts high-LET-type effects (within nanometers of the decay site) and high-LET-like radiotoxicity ( 24 , 25 ). TFOs labeled with 125 I have been shown to induce DSBs in a target DNA duplex molecule with an efficiency approaching one DSB per decay ( 26 ). The nature of the process by which Auger emitters decay and the similarity of the biological effects to those of high-LET radiations, suggests that such DSBs should be of a complex type and thus highly mutagenic. As such, Auger-emitting radionuclides fulfill the criteria discussed above for a mutagenic agent that induces complex DNA lesions including the destructive loss of nucleotides at the damage site. We have examined the possibility of using TFOs labeled with the Auger-emitting radionuclide 125 I as a model system to induce sequence-specific mutagenic disruption of gene function. In order to accomplish this we used the pSP189 human shuttle vector based forward mutation assay described by Paris and Seidman ( 27 ). A short polypyrimidine TFO labeled near the 3′-end with 125 I, and targeted to the SupF gene of the plasmid pSP189 was constructed and used for in vitro damage induction. Following repair in WI-38VA13 human fibroblasts, plasmids were scored for mutations by blue/white screening in Escherichia coli strain MBM7070. Materials and Methods Materials Dulbecco's modified Eagle's medium, fetal bovine serum, non-essential amino acids (10 mM), glutamine (100 mM), penicillin/streptomycin (10 000 U/ml), and the Lipofectamine Plus transfection kit were purchased from Life Technologies (Gaithersburg, MD). Bleomycin was purchased from Sigma (St Louis, MO). Human fibroblast cell line WI-38VA13 was obtained from the American Type Tissue Culture Collection (Manassas, VA). Automated oligonucleotide synthesis reagents were obtained from Glenn Research (Sterling, VA). [5- 125 I]dCTP (2200 Ci/mmol) was purchased from New England Nuclear (Boston, MA). Plasmid pSP189 was a generous gift from Dr Michael Seidman (National Institute of Aging, Baltimore, MD). DNA Oligonucleotides were synthesized on an Applied Biosystems 394 DNA/RNA synthesizer and band purified from denaturing 20% PAGE. [5′- 32 P]PSP3 16mer oligonucleotide was produced by standard 5′-end T4 kinase phosphorylation of full-length PSP3 16mer synthesis products ( 28 ). The PSP3 16mer TFO (5′-CCGCCGCCCCCTTCCT-3′) labeled with [5- 125 I]dCTP at position 15 was constructed by primer extension ( 9 ). The 14mer psp3-primer (5′-CCGCC-GCCCCCTTC-3′) and the biotinylated 28mer psp3-template (5′-AGGAAGGGGGCGGCGGTTTTTTTTTBTB-3′) were annealed (50 mM Tris-HCl pH 8.0, 10 mM MgCl 2 , 50 mM NaCl) and the primer was extended with DNA polymerase I Klenow fragment in the presence of [5- 125 I]dCTP (2200 Ci/mmol; dCTP:primer ratio 2:1) and 33 μM TTP. The 125 I-labeled PSP3 16mer was isolated by heat denaturation of the duplex extension products after binding to streptavidin-labeled magnetic Dynabeads (Dynal, Oslo, Norway). Dynabead-bound template oligonucleotides were separated from the mixture in an ice-cold magnet, and unincorporated nucleotides were removed from the 125 I-PSP3 containing-supernatant by Sephadex G-50 spin column chroma-tography. pSP189 plasmid was grown in E.coli strain DH10B (Life Technologies, Gaithersburg, MD) and isolated using the Qiagen Maxiprep kit (Qiagen, Valencia, CA) according to the manufacturer's instructions. DNA triplex formation/damage induction The relative binding efficiency of PSP3 for pSP189 was determined by gel shift assay. Binding reactions (5 μl) containing 30 mM sodium acetate (NaAc) pH 4.45, 10 mM MgCl 2 , 10 nM [5′- 32 P]PSP3, and varying concentrations of Eco RI linearized pSP189, were incubated at 37°C overnight. Binding products were separated from free TFO in 2% agarose vertical gels cast in binding buffer and run at 4°C in 100 mM NaAc pH 4.7, 3 mM MgCl 2 , at 40 V. The relative amount of bound versus free PSP3 was determined by densitometric analysis of phosphor-image autoradiographs using a Fuji BAS-1500 phosphorimager and MacBas imaging software (Fuji, Stanford, CT). Auger effect damage induced by 125 I was targeted to the pSP189 SupF mutagenesis reporter gene by 125 I-PSP3. Triplex formation was accomplished by incubating 44 pmol of topo-isomerase I relaxed pSP189 with 29 pmol of 125 I-PSP3 in binding buffer (30 mM NaAc pH 4.5, 3 mM MgCl 2 ) at 37°C overnight. Unbound TFO was removed from the mixture by CL-4B Sepharose chromatography, and the TFO-bound DNA was stored at -80°C for 60 days to accumulate damage. Bleomycin treatment Linear (form III) DNA containing DSBs with phosphoglycolate (PG)-blocked 3′-ends was prepared by treating supercoiled (form I) pSP189 plasmid DNA with bleomycin. Plasmid DNA was treated as described previously under conditions which have been demonstrated to produce strand breaks exclusively ( 29 ). Briefly, DNA at 150 μg/ml in 12.5 mM Tris-HCl pH 8.0, 300 mM sucrose, 0.0188% Triton X-100, 1.25 mM EDTA, 5 mM MgCl 2 , 7.5 mM β-mercaptoethanol (βME) and 250 μg/ml heat-inactivated BSA, was treated with bleomycin (1 μg/ml) at 37°C for 30 min in the presence of 2 μM ferrous ammonium sulfate. The drug was removed by ethanol precipitation and linearized DNA was band isolated following 1% agarose gel electrophoresis as described previously ( 30 ). Mutagenesis assay DNA damage was assessed for mutagenic effect using a modification of the pSP189 human shuttle vector based forward mutation assay described by Wang et al. ( 3 ). Plasmid DNA was transfected into WI38VA13 human fibroblasts by lipofection. Monolayers were grown in DMEM containing 10% FBS in a 5% CO 2 atmosphere at 37°C. Cells were harvested by trypsini-zation when 80% confluent, and triplicate 35 mm plates were seeded with 4 − 10 5 cells. After overnight incubation under growth conditions, cells were transfected by the following modification of the Lipofectamine Plus Kit transfection protocol. DNA was prepared for transfection by combining 1.0 μg of plasmid in 50 μ1 of serum-free DMEM medium with 50 μl of serum-free DMEM containing 10 μ1 of the Plus reagent followed by incubation at 37°C for 15 min. The DNA/Plus mixture was then mixed with an additional 100 μ1 of serum-free DMEM containing 6 μ1 of Lipofectamine liposome reagent and incubated at 37°C for an additional 15 min. The WI38VA13 cells were washed twice with PBS, overlaid with the DNA-Lipofectamine Plus complex in a total volume of 2.0 ml serum-free medium, and incubated at 37°C for 3 h. Complete medium was added and the cells were incubated for an additional 24 h. Plasmid DNA was recovered by alkaline lysis ( 3 ). Cells were trypsinized, washed, and resuspended in 100 μl suspension buffer (50 mM Tris-HCl pH 8.0, 10 mM EDTA, 100 μg/ml RNase A), mixed with 100 μ1 lysis buffer [0.2 M NaOH, 1% (w/v) SDS] and incubated on ice for 3–5 min, followed by addition of 100 μl neutralization buffer (3 M potassium acetate pH 5.5). After a 15 min incubation at room temperature the mixture was centrifuged for 10 min at 16 000 g and the supernatant was extracted once with phenol-chloroform (1:1). Following ethanol precipitation, the DNA was resuspended in Dpn I reaction buffer (50 mM KAc, 20 mM Tris-Ac, 10 mM MgAc, 1 mM DTT, pH 7.9) and digested with 10 U Dpn I and 100 μg/ml RNase A for 2 h at 37°C. The reaction products were extracted once with phenol-chloroform, ethanol precipitated and resuspended in 10 μ1 TE pH 8.0. Escherichia coli MBM7070 [ lac Z(Am)] (1,2) was transformed by electro-poration in a 0.1 cm cuvette with 1 μ1 of the recovered DNA (Bio-Rad Gene Pulser at 25 μF, 250 W, 1800 V), and plated on LB plates containing 50 μg/ml ampicillin, 100 μg/ml X-Gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) and 1 μM IPTG (isopropyl-β-D-thiogalactopyranoside). Plates were incubated at 37°C overnight and colonies containing mutant plasmids with inactivated SupF tRNA genes unable to suppress the host-cell β-galactosidase gene amber mutation were identified as white colonies, as opposed to blue colonies containing plasmids expressing the wild-type tRNA gene. Sequence analysis Colonies containing pSP189 mutants were subcloned and plasmid midi-preps were prepared (Plasmid Midi-Prep Kit, Qiagen) for cycle sequencing using the Applied Biosystems BigDye Terminators Kit (Perkin Elmer) according to the manufacturer's instructions. The sequencing primer was chosen to bind upstream of the SupF gene near the 3′-end of the β-lactamase gene. Results Mutagenesis assay and gene targeting by triplex forming oligonucleotide PSP3 In order to assess our ability to use radiolabeled TFOs to perform gene-targeted mutagenesis we employed a mutagenesis assay based upon the human shuttle vector pSP189. This plasmid contains the E.coli SupF tyrosine suppressor tRNA gene as the mutagenesis target. Damage to the SupF gene between positions 99 and 180 that results in a mutation after repair and replication in human cells, will disrupt tRNA gene function ( 31 ). Mutants are detected by blue/white screening after plasmid recovery and transformation into E.coli strain MBM7070. Therefore, gene-targeted delivery of the Auger-electron emitting radionuclide 125 I via sequence-specific TFO binding, was studied for mutagenic effect by targeting the 3′-end of the pSP189 SupF gene with the 16mer TFO PSP3. The assay system is depicted in Figure 1 . After formation of triplexes with 125 I-PSP3, pSP189 DNA was frozen and allowed to accumulate damage. The damaged DNA mixture was either transfected directly into WI38VA13 cells, or used for separation of linear DNA containing site-specific DSBs. Isolated linear DNA was separately transfected into WI38VA13 cells. After transfection, the cells were incubated for 24 h to allow repair and replication of the plasmid. Plasmids were recovered by alkaline lysis, treated with Dpn I to remove unreplicated plasmids retaining the bacterial methylation pattern and electroporated into bacteria for blue/white screening. Figure 1 View largeDownload slide Plasmid pSPl89 was exposed to DNA damaging agents and either the bulk Form I, II and III containing DNA mixture or band isolated linear (Form III) DNA was transfected into human Fibroblast WI38VA13 cells and allowed to repair. Repair products were isolated and scored for mutations in E.coli strain MBM7070 [ lac Z(Am)] by blue/white selection on IPTG/X-Gal/Amp LB plates. Figure 1 View largeDownload slide Plasmid pSPl89 was exposed to DNA damaging agents and either the bulk Form I, II and III containing DNA mixture or band isolated linear (Form III) DNA was transfected into human Fibroblast WI38VA13 cells and allowed to repair. Repair products were isolated and scored for mutations in E.coli strain MBM7070 [ lac Z(Am)] by blue/white selection on IPTG/X-Gal/Amp LB plates. The polypyrimidine 16mer TFO PSP3, was designed to recognize and bind in a parallel orientation to positions 167–182 of the SupF gene. The triple helical structure places the [5- 125 I]dCMP residue located at position 15 of the TFO, opposite position 168 of the SupF gene ( Fig. 2 ). The resulting 125 I-C-GC triplet at position 168 would be expected to produce localized DSBs, with the strand break occurring from between 5 bases upstream to 5 bases downstream of position 168 at decreasing probability of break induction as distance from the 125 I binding site increases ( 9 , 26 , 32 ). The chosen 16mer target sequence is not a homopoly-purinic sequence, but includes two thymidine bases at positions three and six (when counting 5′→3′ on the purine strand of the target duplex sequence). The presence of these pyrimidine bases would be expected to destabilize parallel motif TFO binding. Therefore, we initially targeted only the last 10 bases of this target sequence (167–176), however the oligo constructed for this purpose did not bind effectively. Consequently, we sought to increase the TFO Kd and binding efficiency by broadening the TFO target to include the 16 nucleotides from position 167 to 182 of the SupF gene ( Fig. 2 ). Although conventional third strand base pairing rules indicate that the presence of the thymidines within this sequence would prevent triplex formation, stable parallel triplex structures involving G·TA triplets have been reported ( 33 , 34 ). Therefore, PSP3 was designed as a mixed pyrimidine/purine oligo with G at positions three and six. The resulting TFO was capable of binding the pSP189 target sequence in a dose-dependent manner with a Kd of ∼150 nM ( Fig. 3A ). Figure 2 View largeDownload slide Human shuttle vector pSP189. The plasmid contains the bacterial SupF suppressor tRNA gene as the mutagenesis reporter gene, the 3′-end of which has been expanded and numbered in the upper panel according to tRNA convention. The TFO PSP3 is shown aligned with its target sequence. The position of the [5- 125 I]cytosine residue is indicated in bold at position 15 of the TFO. TFO residues and nucleotides within the tRNA gene sequence are shown as uppercase letters. Figure 2 View largeDownload slide Human shuttle vector pSP189. The plasmid contains the bacterial SupF suppressor tRNA gene as the mutagenesis reporter gene, the 3′-end of which has been expanded and numbered in the upper panel according to tRNA convention. The TFO PSP3 is shown aligned with its target sequence. The position of the [5- 125 I]cytosine residue is indicated in bold at position 15 of the TFO. TFO residues and nucleotides within the tRNA gene sequence are shown as uppercase letters. Triplex formation and DNA damage induction These binding conditions were sufficient to permit scale-up of the in vitro binding reaction for production of DNA damaged site specifically by the Auger effect via TFO targeted 125 I decay. The scaled-up binding reaction contained 44 pmol of pSP189 DNA at 50 nM and 29 pmol of 125 I-PSP3 at 30 nM. Following binding, unbound 125 I-PSP3 was removed by Sepharose CL-4B chromatography. The plasmid DNA was recovered, pooled and demonstrated, by agarose gel electrophoresis, to contain stable triplex structures and to be free of unbound 125 I-PSP3 (data not shown). The amount of triplex DNA recovered in the pooled plasmid was determined to be 18.7 μg by scintilation counting. The purified triplex containing plasmid was stored at −70°C for 60 days to accumulate damage. The plasmid DNA was assessed for damage induction by agarose gel electrophoresis and the image was quantified with Gel-Pro image analysis software (Media Cybernetics, Silver Spring, MD). The damaged DNA product was distributed as three forms, supercoiled (Form I), open circular (Form II) and linear (Form Iff) ( Fig. 3B , lane 1). More than 88% of the total DNA product contained strand breaks, while ∼11% was undamaged and ran as supercoiled DNA. The open circular Form II DNA generated by SSB induction comprised 79.6% of the total damaged DNA product, while the linear Form III DNA produced by DSB induction accounted for 8.3% of the total damaged product. In order to determine if site-specific DSBs had been introduced into pSP189, an amount of damaged DNA equivalent to that run in lane 1 was cleaved with the restriction enzyme Pfl M1. The plasmid has a unique restriction site for Pfl M1 opposite and nearly equidistant upstream and downstream from the SupF gene ( Fig. 2 ). Therefore, Pfl M1 cleavage of plasmid molecules that have accumulated DSBs in the SupF gene, results in the production of two DNA fragments of 2861 and 2091 bp respectively. In contrast, plasmids that contain SSBs, or that are otherwise intact, will linearize and run as Form III DNA after electrophoresis. The observation of the 2861 and 2091 bp fragments following Pfl M1 cleavage of the damaged plasmid DNA, demonstrates the production of site-specific DSBs in the SupF gene as a result of 125 I-PSP3 TFO gene targeting ( Fig. 3B , lane 2). The two fragments resulting from site specific DSB induction constitute 5.2% of the DNA in the damaged DNA mixture. Consequently, site specific SupF targeted DSBs represent ∼63% of all DSBs induced in pSP189 by 125 I decay during the course of damage accumulation. Figure 3 View largeDownload slide ( A ) Triplex (T) formation by 125 I-labeled PSP3 oligonucleotide and pSP189. Target sequence binding dose-response with 0, 10, 25, 75, 100 and 150 nM of pSP189 (lanes 1-6 respectively) and 10 nM of 125 I-PSP3 (O). Binding was performed in 100 mM NaAc pH 4.5, 150 mM NaCl and 5 mM MgCl 2 at 37°C overnight. PSP3-bound plasmid was resolved from unbound oligo by electrophoresis in a 2% agarose gel, and autoradiography was performed with a Fuji 1500 phosphorimager. ( B ) Site-specific DSB induction by 125 I-PSP3. Lane 1, 125 I-PSP3-damaged pSP189; lane 2, Pfl M1-cut 125 I-PSP3-damaged pSP189. 125 I-PSP3 triplex formation induces site-specific DSBs in pSP189. DSB induction is indicated by formation of linear pSPl 89 (lane 1, Form HI DNA). Site specificity of 125 I-PSP3-induced DSBs is demonstrated by formation of 2861 and 2091 bp fragments after cutting damaged DNA with Pfl Ml. Figure 3 View largeDownload slide ( A ) Triplex (T) formation by 125 I-labeled PSP3 oligonucleotide and pSP189. Target sequence binding dose-response with 0, 10, 25, 75, 100 and 150 nM of pSP189 (lanes 1-6 respectively) and 10 nM of 125 I-PSP3 (O). Binding was performed in 100 mM NaAc pH 4.5, 150 mM NaCl and 5 mM MgCl 2 at 37°C overnight. PSP3-bound plasmid was resolved from unbound oligo by electrophoresis in a 2% agarose gel, and autoradiography was performed with a Fuji 1500 phosphorimager. ( B ) Site-specific DSB induction by 125 I-PSP3. Lane 1, 125 I-PSP3-damaged pSP189; lane 2, Pfl M1-cut 125 I-PSP3-damaged pSP189. 125 I-PSP3 triplex formation induces site-specific DSBs in pSP189. DSB induction is indicated by formation of linear pSPl 89 (lane 1, Form HI DNA). Site specificity of 125 I-PSP3-induced DSBs is demonstrated by formation of 2861 and 2091 bp fragments after cutting damaged DNA with Pfl Ml. 125 I-labeled TFO mediated damage repair mutagenesis The repairability and mutagenic effect of the damage induced in pSP189 as a consequence of anti- SupF125 I-PSP3 TFO gene targeting was examined in two ways. First, the mutagenicity and overall toxicity (measured as a function of plasmid survival) of all 125 I-PSP3-induced damage was determined by using the bulk damaged DNA mixture depicted in Figure 3B , lane 1, for transfection and repair in human WI38VA13 fibroblasts. Second, the same parameters were examined for the isolated linear DNA component of the mixture which contains SupF site-specific DSBs. In addition, bleomycin, a radiomimetic strand-break-inducing agent with known mutagenic potential in this assay system ( 35 , 36 ), was also used to induce strand-break damage in pSP189 for comparison to the 125 I-PSP3 TFO-induced damage. The mutation frequency for the bulk 125 I-PSP3 TFO-induced damaged DNA mixture was 2.7 − 10 −3 . This frequency is a >50-fold increase over the background (5.2 − 10 −5 ) observed for undamaged pSP189, and approximately equal to that reported for random strand breaks induced by bleomycin ( Table 1 ) ( 35 ). In contrast, the mutation frequency observed for the 125 I-PSP3 TFO-induced site-specific DSBs was 7.9 − 10 −1 , as compared to a mutation frequency of 7.1 − 10 −3 for linear DNA produced by random bleomycin DSB induction. Therefore, the 125 I-induced site-specific DSBs are ′111-fold more mutagenic than the DSBs produced randomly by bleomycin, and have >290-fold higher mutation frequency than that of the bulk 125 I-TFO-induced DNA damage mixture. Furthermore, the site-specific 125 I-TFO-induced DSB mutation frequency is in excess of 1.5 − 10 4 -fold greater than that observed for undamaged super-coiled pSP189 in this system. View largeDownload slide Mutation frequencies and plasmid survival for different forms of pSPl89 following damage induction View largeDownload slide Mutation frequencies and plasmid survival for different forms of pSPl89 following damage induction Plasmid survival in this assay system is a function of a trans-fected plasmid's ability to be repaired, replicated and recovered from the WI38VA13 cells. Consequently, if a lesion in the supF gene is a poor substrate for repair, the plasmid may not be able to replicate prior to recovery. Unreplicated plasmids are eliminated by Dpn I treatment, and do not produce colonies following bacterial transformation. It is also possible that repair of lesions in the SupF gene may result in plasmids suffering deletions that extend into the essential upstream β-lactamase gene, or the downstream bacterial replication origin. Disruption of either of these sequences, or other sequences critical to replication, will prevent plasmid recovery. Therefore, a reduction in plasmid survival indicates either a highly complex lesion that is difficult to repair, or repair events that disrupt the function of elements outside of the SupF gene which are essential to plasmid replication. In the case of 125 I-PSP3-induced damage, the bulk damaged mixture had a survival of 85.4% with respect to undamaged pSP189. In contrast, the linear DNA component of TFO-mediated damage had a survival of 0.21%. This compares to a survival of 56.3% for DNA linearized by bleomycin. Therefore 125 I-PSP3-induced DSB damage results in a 480-fold reduction in plasmid survival with respect to undamaged plasmid, and 270-fold lower survival than DSBs induced by bleomycin. Distribution of SupF mutations induced by 125 I-labeled TFO targeting Mutant plasmids were recovered from subcloned mutant colonies and sequenced to determine the nature of the mutations. Mutants obtained from the bulk damaged DNA mixture, or isolated linear DNA, consisted of deletions, substitutions, insertions and complex mutations comprising a mixture of these ( Fig. 4 ). In the case of the bulk damaged DNA mixture, 13 (62%) of the 21 mutants sequenced contained deletions ( Fig. 4A ). Of these deletion mutants, four were complex. Three of the complex deletion mutants (3M, 23M, 34M) consisted of one or more deleted bases in addition to base substitutions, while the remaining complex deletion mutant (16M), consisted of a single base deletion plus a 2 base insertion. Base substitutions comprised nine (42%) of the 21 mutants sequenced, seven (78%) of which were multiple base substitutions. Complex mutations were observed in four base substitution mutations, three of which (as indicated above) consisted of base substitutions and deletions, while one consisted of multiple base substitutions plus two base insertions (21M). Insertions of 1–2 bases were observed in four mutants (19%), two of which were complex (16M, 21M). The majority (95%) of mutants obtained from repair of the bulk damaged DNA mixture possessed changes clustered within 5 bases of the 125 I binding position. This is within the strand-break-induction range previously reported for 125 I-labeled TFOs ( 9 ). Of the observed deletions, 69% involved less than 5 bases and had ends within 5 bases of the 125 I binding site at position 168 of the SupF gene. In addition, 70% of all deletions involved multiple bases and 100% of these (9/9) spanned the 125 I binding site, while 75% of the single base deletions were within 5 bases of position 168. Base substitution mutations were more dispersed, with only six of the 24 total observed base substitutions occurring within 5 bases of the 125 I binding site. In contrast, 60% of the observed insertions either spanned, or were within 5 bases of the 125 I binding site. A different distribution of mutations was observed for repair of the isolated linear DNA component of the damaged DNA mixture ( Fig. 4B ). Deletions were observed in 11 (65%) of the 17 mutants sequenced; however, 10 of these mutants were also complex. Eight (73%) of the complex mutations involving deletions also contained base substitutions, of which four (identified as unique via the pSP189 signature sequence) appeared to have undergone a T→C transition at position 167 followed by a 3 base deletion to position 170 (7L, 9L, 25L, 34L). Consequently, this modification occurs in nearly 25% of the mutants recovered from the linear DNA repair reactions. Two of the remaining four complex deletions containing base substitutions consisted of a single base deletion plus a single base substitution (3L, 5L), whereas the other two involved multiple base deletions plus single base substitutions (6L, 15L). The final two complex deletions (1L, 14L) consisted of multiple deletions. In addition, mutants 7L and 15L contained multiple base insertions (9 and 51 bases respectively) at position 124. All of the deletion mutants recovered from the linear DNA repair reactions either spanned the 125 I binding position, or had ends within 4 bases of the 125 I binding position. Furthermore, the deletions resulting from repair of the linear DNA were typically smaller than those recovered following repair of the bulk damaged DNA mixture, with 82% (9/11) of the mutants possessing deletions of 3 bases or less, all of which were clustered within 4 bases of the 125 I binding site. In contrast, only 38% (5/13) of the deletions recovered from the damaged DNA mixture were of 3 bases or less, and of these only four of five were within 5 bases of the 125 I binding site. Figure 4 View largeDownload slide Mutation spectrum of 125 I-PSP3 triplex-induced damage following repair in human WI38VA13 cells. Each mutant is displayed on a separate line with deletions indicated above the SupF sequence and base substitutions and insertions displayed below. Insertions (1–2 bases) are indicated by underscoring, whereas a large insertion position is indicated by a caret. Base substitutions are indicated by unmodified letters. Mutant designation numbers are shown on the right. ( A ) Bulk damaged DNA mixture (M) mutation spectrum. ( B ) Mutation spectrum of the isolated linear (L) site-specific DSB-containing component of the damaged DNA mixture. Figure 4 View largeDownload slide Mutation spectrum of 125 I-PSP3 triplex-induced damage following repair in human WI38VA13 cells. Each mutant is displayed on a separate line with deletions indicated above the SupF sequence and base substitutions and insertions displayed below. Insertions (1–2 bases) are indicated by underscoring, whereas a large insertion position is indicated by a caret. Base substitutions are indicated by unmodified letters. Mutant designation numbers are shown on the right. ( A ) Bulk damaged DNA mixture (M) mutation spectrum. ( B ) Mutation spectrum of the isolated linear (L) site-specific DSB-containing component of the damaged DNA mixture. Base substitutions were observed in 10 (59%) of the 17 DSB mutants sequenced, and four were complex involving deletions as indicated above. In contrast to the bulk damaged DNA mixture, only one of the 10 base substitution mutants obtained from the repair of linear DNA involved multiple base substitutions. Also, 50% (5/10) of the base substitution mutants involved G→T transversions and were clustered around positions 159 and 160. Interestingly, none of the base substitutions occurred within 5 bases of the 125 I binding position at residue 168. Also, unlike the mutants recovered from the bulk damaged DNA mixture, two multiple base insertions were observed as part of complex deletion mutants recovered from linear DNA repair. However, both insertions occurred at position 124, well away from the 125 I binding site. The distribution of mutations observed for the repair of the bulk damaged DNA mixture and repair of the linear DNA component containing site-specific DSBs is summarized in Table 2 . View largeDownload slide Distribution of mutations in bulk 125 I-PSP3-damaged DNA versus isolated site-specific DSB-containing 125 I-PSP3 linearized DNA View largeDownload slide Distribution of mutations in bulk 125 I-PSP3-damaged DNA versus isolated site-specific DSB-containing 125 I-PSP3 linearized DNA Discussion The results presented here illustrate the possibility of using TFOs labeled with Auger-electron emitting radionuclides to induce site-specific mutations and disrupt gene function. Labeling of the TFO PSP3 at position 15 with [ 125 I]dCTP allowed us to take advantage of the highly localized energy deposition produced by Auger-electron emitter decay to induce site-specific DSBs in the SupF gene of pSP189 in vitro . The 125 I-TFO-induced DSBs were found to be highly mutagenic upon repair in the normal human fibroblast WI38VA13 cell line with a mutation frequency of nearly 80%. Since site-specific DSBs were estimated to represent >63% of the DSBs in the 125 I-TFO linearized DNA, this suggests that the majority of mutants were produced by site-specific DSBs and that <20% of the mutants recovered from the linear DNA transfections were due to non-targeted damage. The mutation frequency observed for 125 I-TFO-induced DSBs is nearly 7-fold greater than that reported for the repair of a 5′-psoralen-conjugated anti-parallel TFO targeted to crosslink SupF positions 166 and 167, and >2.5-fold greater than the mutation frequency observed for the same lesion repaired in nucleotide excision repair defective XPV cells ( 37 ). These results suggest that the Auger-emitter-induced DSBs are more effective mutagenic lesions than crosslinks produced by a chemical mutagen, even under conditions for which the chemically induced lesion is poorly repaired. In contrast, the Auger-emitter-induced DSBs would be expected to mutagenically alter the target sequence with an incidence of nearly 1:1, making Auger emitters a more effective anti-gene agent than chemical mutagens. DSBs with blocked 3′-ends similar to those produced by ionizing radiation were generated with the radiomimetic drug bleomycin. We observed a mutation frequency of 7.1 − 10 −3 for this lesion, which is in good agreement with previously reported results in this system ( 35 ). Therefore, 125 I-TFO-induced DSBs are >100-fold more mutagenic than the bleomycin-induced DSBs. Although part of this disparity is likely due to the random nature of the bleomycin-induced DSBs as opposed to the SupF targeted nature of the 125 I-TFO-induced DSBs, this probably does not fully account for the difference, since non-ligatable site-specific DSBs induced in pUC19 with restriction enzymes only produce mutation frequencies of up to 5 − 10 −3 following repair in normal human cell extracts and do not exceed 1.8 − 10 −1 even when repaired in ataxia-telangiectasia cell extracts ( 38 , 39 ). In these systems the mutation frequencies and plasmid survivals were similar to that reported here for bleomycin-linearized DNA and the bulk 125 I-TFO-induced damage mixture. This is in contrast to the low survival of plasmid linearized by Auger-emitter-induced DSBs, suggesting that either the kinetics of Auger-emitter-induced DSB repair are considerably slower than that of DNA ends possessing protruding ends or 3′-end blocking groups, or that Auger-emitter-induced DSB damage processing may predominantly result in unrecoverable products (i.e. large deletions), or elimination of the substrate altogether. The latter two possibilities are consistent with the observation that 125 I-induced mutants of the HPRT gene in CHO cells and human cells predominantly consist of large deletions ( 40 , 41 ). However, the former possibility may also play a role in the low survival of 125 I-TFO linearized plasmids in light of the observation by Gu et al. that site-specific DSB plasmid constructs with ends blocked by 3′-phos-phoglycolates were rejoined 30–100 times slower than substrates with 3′-OH ends, although recovery of both constructs was similar and at least 10-fold greater than that observed here for 125 I-TFO linearized plasmids ( 42 , 43 ). The mutation frequency observed for the bulk 125 I-TFO-damaged pSP189 DNA is essentially the same as that reported previously for SSBs produced randomly by bleomycin ( 35 ). This relatively low mutation frequency indicates that the nonspecific radiolysis damage in the bulk 125 I-TFO-damaged pSP189 DNA mixture, which constitutes the majority of the damage as judged by the proportion of Form II DNA, was relatively innocuous in comparison to the site-specific DSBs. The mutation spectra obtained for the bulk 125 I-TFO-damaged DNA mixture and the isolated DSB-containing component differ most markedly in their relative proportions of complex deletions. Only four of the 13 deletion mutants recovered from the bulk damaged DNA mixture were complex as compared to 10 of 11 deletion mutants recovered from the isolated DSB component of the mixture. In addition, four of the 10 complex deletions obtained from the DSB-containing DNA consisted of a C→T transition followed by a 3 bp deletion which may constitute a low frequency signature mutation for 125 I-TFO-induced DSBs. The recovery of complex deletion mutants from the 125 I-TFO linearized DNA component of the mixture is consistent with the observation that 125 I-induced DSBs constitute multiple breaks in each DNA strand such that the DSB is typically accompanied by the deletion of several base pairs ( 44–46 ). Furthermore, 125 I decays in DNA produce between three and six SSBs per DSB, at least one-third of which are due to indirect effects ( 47–50 ). This may account for the base substitutions associated with the deletions recovered from the DSB-containing DNA, as well as those observed in the mutants recovered from the bulk damaged DNA mixture, since SSBs have been shown to be responsible for base substitutions ( 35 , 51 ). Interestingly, none of the deletions recovered from either the bulk damaged DNA mixture, or the isolated DSB containing DNA, was bordered by short direct repeat sequences which is a feature associated with non-homologous end joining (NHEJ) mechanisms ( 35 , 52–56 ). This is in contrast to deletions produced by X-rays, bleomycin, restriction enzyme DSB repair, and site-specific chemically defined DSB constructs containing 3′-ends blocked by phosphoglycolate ( 35 , 39 , 43 , 52 , 53 , 57–59 ). These results suggest that NHEJ processing of 125 I-TFO-induced DSBs may proceed somewhat differently than for many of the other DSB lesions that have been examined to date. In those cases in which NHEJ does not result in direct rejoining of DSB ends, repair involves unwinding of the DSB ends followed by single-strand annealing at a region of microhomology, digestion of unpaired single-stranded regions and ligation, resulting in a deletion bordered by a short direct repeat sequence ( 60–62 ). Consequently, except in those cases where only one or two nucleotides are deleted near the break site, the high degree of complexity observed in the mutation spectrum for 125 I-TFO-induced DSBs would seem to argue against direct end joining being responsible for the absence of direct repeat sequences bordering the deletions. However, if the single-strand annealing process plays a role in formation of the deletions, it must proceed slightly differently than for many other DSB types, since none of the 125 I-TFO-induced deletions was bordered by direct repeats. These results further support the supposition that 125 I-TFO-induced DSBs may be more chemically complex than typical DSBs produced by other agents. Elucidation of the exact chemical nature of the DSBs produced by the Auger effect would greatly enhance our understanding of the biochemical requirements of their repair. The current work models the mutagenic effect of site-specific Auger-emitter-induced DNA damage using an 125 I-TFO to target plasmid DNA in vitro under non-physiologic binding conditions. Although the long half-life of 125 I along with the artificial TFO binding conditions used here preclude the use of the current model system for in vivo anti-gene targeting, recently described oligonucleotide chemical modifications which stabilize triplex formation in vivo , in conjunction with short half-life Auger-emitters may permit such an approach. Consequently, the results presented here illustrate that Auger-electron-emitting radionuclides may be highly effective mutagens for site-specific disruption of gene function when linked to TFOs as a delivery vehicle. Work is currently in progress to identify improved target sequences and develop improved TFOs with sufficient binding efficiency to conduct in vivo targeting experiments. Acknowledgments We would like to thank Dr Michael Seidman for the pSP189 shuttle vector, the MBM7070 E.coli indicator strain, and helpful discussions. We also thank George Poy for assistance with DNA sequencing. We are especially thankful to Dr Ludwig Feinendegen for his careful reading of the manuscript, and for many stimulating and helpful discussions. References 1 Havre P.A.,  Glazer P.M.. ,  J. Virol. ,  1993, vol.  67 (pg.  7324- 7331) PubMed  2 Havre P.A.,  Gunther E.J.,  Gasparro F.P.,  Glazer P.M.. ,  Proc. Natl. Acad. Sci. USA ,  1993, vol.  90 (pg.  7879- 7883) CrossRef Search ADS   3 Wang G.,  Levy D.D.,  Seidman M.M.,  Glazer P.M.. ,  Mol. Cell Biol. ,  1995, vol.  15 (pg.  1759- 1768) CrossRef Search ADS PubMed  4 Vasquez K.M.,  Wensel T.G.,  Hogan M.E.,  Wilson J.H.. ,  Biochemistry ,  1996, vol.  35 (pg.  10712- 10719) CrossRef Search ADS PubMed  5 Majumdar A.,  Khorlin A.,  Dyatkina N.,  Lin F.L.M.,  Powell J.,  Liu J.,  Feiz Z.Z.,  Khripine Y.,  Watanabe K.A.,  George J.,  Glazer P.M.,  Seidman M.M.. ,  Nature Genet. ,  1998, vol.  20 (pg.  212- 214) CrossRef Search ADS   6 Moser H.E.,  Dervan P.B.. ,  Science ,  1987, vol.  238 (pg.  645- 650) CrossRef Search ADS PubMed  7 Sun J.S.,  Garestier T.,  Helene C.. ,  Curr. Opin. Struct. Biol. ,  1996, vol.  6 (pg.  327- 333) CrossRef Search ADS PubMed  8 Escude C.,  Nguyen C.H.,  Kukreti S.,  Janin Y.,  Sun J.S.,  Bisagni E.,  Garestier T.,  Helene C.. ,  Proc. Natl. Acad. Sci. USA ,  1998, vol.  95 (pg.  3591- 3596) CrossRef Search ADS   9 Panyutin I.G.,  Neumann R.D.. ,  Acta Oncol ,  1996, vol.  35 (pg.  817- 823) CrossRef Search ADS PubMed  10 Wang G.,  Seidman M.M.,  Glazer P.M.. ,  Science ,  1996, vol.  271 (pg.  802- 805) CrossRef Search ADS PubMed  11 Belousov E.S.,  Afonina I.A.,  Kutyavin I.V.,  Gall A.A.,  Reed M.W.,  Gamper H.B.,  Wydro R.M.,  Meyer R.B.. ,  Nucleic Acids Res. ,  1998, vol.  26 (pg.  1324- 1328) CrossRef Search ADS PubMed  12 Vasquez K.M.,  Wang G.,  Havre P.A.,  Glazer P.M.. ,  Nucleic Acids Res. ,  1999, vol.  27 (pg.  1176- 1181) CrossRef Search ADS PubMed  13 Ward J.F.. ,  Int. J. Radiat. Biol. ,  1994, vol.  66 (pg.  427- 432) CrossRef Search ADS PubMed  14 Chaudhry M.A.,  Weinfeld M.. ,  Nucleic Acids Res. ,  1995, vol.  23 (pg.  3805- 3809) CrossRef Search ADS PubMed  15 Chaudhry M.A.,  Weinfeld M.. ,  J. Biol. Chem. ,  1997, vol.  272 (pg.  15650- 15655) CrossRef Search ADS PubMed  16 Harrison L.,  Hatahet Z.,  Purmal A.A.,  Wallace S.S.. ,  Nucleic Acids Res. ,  1998, vol.  26 (pg.  932- 941) CrossRef Search ADS PubMed  17 Ward J.F.,  Jones G.D.D.,  Milligan J.R.. ,  Radiat. Protect. Dosim. ,  1994, vol.  52 (pg.  271- 276) 18 Chaudhry M.A.,  Weinfeld M.. ,  J. Mol. Biol. ,  1995, vol.  249 (pg.  914- 922) CrossRef Search ADS PubMed  19 Ward J.F.. ,  Radiat. Res. ,  1995, vol.  142 (pg.  362- 368) CrossRef Search ADS PubMed  20 Wallace S.S.. ,  Radial Res. ,  1998, vol.  150 (pg.  60- 79) CrossRef Search ADS   21 Howell R.W.. ,  Med. Phys. ,  1992, vol.  19 (pg.  1371- 1383) CrossRef Search ADS PubMed  22 Sastry K.S.R.. ,  Med. Phys. ,  1992, vol.  19 (pg.  1361- 1370) CrossRef Search ADS PubMed  23 Rao D.V.,  Narra V.R.,  Howell R.W.,  Sastry K.S.R.. ,  Radiat. Res. ,  1990, vol.  124 (pg.  188- 193) CrossRef Search ADS PubMed  24 Rao D.V.,  Howell R.W.,  Narra V.R.,  Govelitz G.F.,  Sastry K.S.R.. ,  Lancet ,  1989, vol.  2 (pg.  650- 653) CrossRef Search ADS PubMed  25 Howell R.W.,  Rao D.V.,  Hou D.Y.,  Narra V.R.,  Sastry K.S.R.. ,  Radiat. Res. ,  1991, vol.  128 (pg.  282- 292) CrossRef Search ADS PubMed  26 Panyutin I.G.,  Neumann R.D.. ,  Nucleic Acids Res. ,  1994, vol.  22 (pg.  4979- 4982) CrossRef Search ADS PubMed  27 Parris C.N.,  Seidman M.M.. ,  Gene ,  1992, vol.  117 (pg.  1- 5) CrossRef Search ADS PubMed  28 Sambrook J.,  Fritsch E.F.,  Maniatis T.. ,  Molecular Cloning: A Laboratory Manual ,  1989 Plainview, NY Cold Spring Harbor Laboratory Press 29 Winters T.A.,  Weinfeld M.,  Jorgensen T.J.. ,  Nucleic Acids Res. ,  1992, vol.  20 (pg.  2573- 2580) CrossRef Search ADS PubMed  30 Dar M.E.,  Winters T.A.,  Jorgensen T.J.. ,  Mutat. Res. ,  1997, vol.  384 (pg.  169- 179) CrossRef Search ADS PubMed  31 Kraemer K.H.,  Seidman M.M.. ,  Mutat. Res. ,  1989, vol.  220 (pg.  61- 72) CrossRef Search ADS PubMed  32 Panyutin I.G.,  Neumann R.D.. ,  Nucleic Acids Res. ,  1997, vol.  25 (pg.  883- 887) CrossRef Search ADS PubMed  33 Griffin L.C.,  Dervan P.B.. ,  Science ,  1989, vol.  245 (pg.  967- 971) CrossRef Search ADS PubMed  34 Yoon K.,  Hobbs C.A.,  Koch J.,  Sardaro M.,  Kutny R.,  Weis A.L.. ,  Proc. Natl. Acad. Sci. USA ,  1992, vol.  89 (pg.  3840- 3844) CrossRef Search ADS   35 Dar M.E.,  Jorgensen T.J.. ,  Nucleic Acids Res. ,  1995, vol.  23 (pg.  3224- 3230) CrossRef Search ADS PubMed  36 Petukhova G.,  Stratton S.,  Sung P.. ,  Nature ,  1998, vol.  393 (pg.  91- 94) CrossRef Search ADS PubMed  37 Dolganov G.M.,  Maser R.S.,  Novikov A.,  Tosto L.,  Chong S.,  Bressan D.A.,  Petrini J.H.. ,  Mol. Cell Biol. ,  1996, vol.  16 (pg.  4832- 4841) CrossRef Search ADS PubMed  38 North P.,  Ganesh A.,  Thacker J.. ,  Nucleic Acids Res. ,  1990, vol.  18 (pg.  6205- 6210) CrossRef Search ADS PubMed  39 Ganesh A.,  North P.,  Thacker J.. ,  Mutat. Res. ,  1993, vol.  299 (pg.  251- 259) CrossRef Search ADS PubMed  40 Gibbs R.A.,  Camakaris J.,  Hodgson G.S.,  Martin R.F.. ,  Int. J. Radiat. Biol. ,  1987, vol.  51 (pg.  193- 199) 41 Whaley J.M.,  Kassis A.I.,  Kinsey B.M.,  Adelstein S.J.,  Little J.B.. ,  Int. J. Radiat. Biol. ,  1990, vol.  57 (pg.  1087- 1103) CrossRef Search ADS PubMed  42 Gu X.-Y.,  Bennett R.A.O.,  Povirk L.F.. ,  J. Biol. Chem. ,  1996, vol.  271 (pg.  19660- 19663) CrossRef Search ADS PubMed  43 Bennett R.A.,  Gu X.-Y.,  Povirk L.F.. ,  Int. J. Radiat. Biol. ,  1996, vol.  70 (pg.  623- 636) CrossRef Search ADS PubMed  44 Martin R.F.,  Haseltine W.A.. ,  Science ,  1981, vol.  213 (pg.  896- 898) CrossRef Search ADS PubMed  45 Martin R.F.,  Holmes N.. ,  Nature ,  1983, vol.  302 (pg.  452- 454) CrossRef Search ADS PubMed  46 Nikjoo H.,  Martin R.F.,  Charlton D.E.,  Terrissol M.,  Kandaiya S.,  Lobachevsky P.. ,  Acta Oncol. ,  1996, vol.  35 (pg.  849- 856) CrossRef Search ADS PubMed  47 Painter R.B.,  Young B.R.,  Burki H.J.. ,  Proc. Natl. Acad. Sci. USA ,  1974, vol.  71 (pg.  4836- 4838) CrossRef Search ADS   48 Feinendegen L.E.,  Henneberg P.,  Tisljar-Lentulis G.. ,  Curr. Top. Radiat. Res. Quart. ,  1977, vol.  12 (pg.  436- 452) 49 Sundell-Bergman S.,  Johanson K.J.. ,  Radiat. Environ. Biophys. ,  1980, vol.  18 (pg.  239- 248) CrossRef Search ADS PubMed  50 Pomplun E.,  Terrissol M.,  Demonchy M.. ,  Acta Oncol. ,  1996, vol.  35 (pg.  857- 862) CrossRef Search ADS PubMed  51 Seidman M.M.,  Bredberg A.,  Seetharam S.,  Kraemer K.H.. ,  Proc. Natl. Acad. Sci. USA ,  1987, vol.  84 (pg.  4944- 4948) CrossRef Search ADS   52 Roth D.B.,  Wilson J.H.. ,  Mol. Cell Biol. ,  1986, vol.  6 (pg.  4295- 4304) CrossRef Search ADS PubMed  53 Thacker J.,  Chalk J.,  Ganesh A.,  North P.. ,  Nucleic Acids Res. ,  1992, vol.  20 (pg.  6183- 6188) CrossRef Search ADS PubMed  54 King J.S.,  Valcarcel E.R.,  Rufer J.T.,  Phillips J.W.,  Morgan W.F.. ,  Nucleic Acids Res. ,  1993, vol.  21 (pg.  1055- 1059) CrossRef Search ADS PubMed  55 Nicolas A.L.,  Munz P.L.,  Young C.S.H.. ,  Nucleic Acids Res. ,  1995, vol.  23 (pg.  1036- 1043) CrossRef Search ADS PubMed  56 Mason R.M.,  Thacker J.,  Fairman M.P.. ,  Nucleic Acids Res. ,  1996, vol.  24 (pg.  4946- 4953) CrossRef Search ADS PubMed  57 Sikpi M.O.,  Dry S.M.,  Freedman M.L.,  Lurie A.G.. ,  Int. J. Radiat. Biol. ,  1992, vol.  62 (pg.  555- 562) CrossRef Search ADS PubMed  58 Morris T.,  Thacker J.. ,  Proc. Natl. Acad. Sci. USA ,  1993, vol.  90 (pg.  1392- 1396) CrossRef Search ADS   59 Lehman C.W.,  Trautman J.K.,  Carroll D.. ,  Nucleic Acids Res. ,  1994, vol.  22 (pg.  434- 442) CrossRef Search ADS PubMed  60 Jeggo P.A.. ,  Radiat. Res. ,  1998, vol.  150 (pg.  S80- S91) CrossRef Search ADS PubMed  61 Dynan W.S.,  Yoo S.. ,  Nucleic Acids Res. ,  1998, vol.  26 (pg.  1551- 1559) CrossRef Search ADS PubMed  62 Zheng S.X.,  Newton G.L.,  Gonick G.,  Fahey R.C.,  Ward J.F.. ,  Radiat. Res. ,  1988, vol.  114 (pg.  11- 27) CrossRef Search ADS PubMed  © 1999 Oxford University Press
TI - Gene targeted DNA double-strand break induction by 125 l-labeled triplex-forming oligonucleotides is highly mutagenic following repair in human cells
JF - Nucleic Acids Research
DO - 10.1093/nar/27.21.4282
DA - 1999-11-01
UR - https://www.deepdyve.com/lp/oxford-university-press/gene-targeted-dna-double-strand-break-induction-by-125-l-labeled-9kL8hKUSaM
SP - 4282
EP - 4290
VL - 27
IS - 21
DP - DeepDyve
ER -