Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (252K) Freely available
Right arrow Supplementary Material
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (7)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Santhosh, U.
Right arrow Articles by Schuster, G. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Santhosh, U.
Right arrow Articles by Schuster, G. B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research, 2003, Vol. 31, No. 19 5692-5699
© 2003 Oxford University Press

Long-distance radical cation reactions in DNA three-way junctions: inter-arm interaction and migration through the junction

U. Santhosh and Gary B. Schuster*

School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, GA 30332, USA

*To whom correspondence should be addressed. Tel: +1 404 894 3300; Fax: +1 404 894 7466; Email: gary.schuster{at}cos.gatech.edu

Received May 20, 2003; Revised and Accepted August 5, 2003


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
DNA three-way junctions (TWJ) are branched molecules having three ‘arms’. We studied long-distance radical cation migration in these assemblies by incorporating anthraquinone (AQ) groups linked by a covalent tether to one strand of one arm of the TWJ. Excitation of the AQ at 350 nm results in one-electron oxidation of the DNA, which generates a base radical cation. This leads to relatively inefficient (compared with duplex DNA) strand cleavage at guanines following piperidine treatment of the irradiated samples. When the AQ is linked to the 5'-terminus of arm III by a flexible tether, gel electrophoretic analysis shows that strand cleavage occurs at the guanines in all three arms. We also investigated a TWJ in which the anthraquinone is specifically intercalated in arm III. In this case, a different pattern of strand cleavage is detected. We conclude that there are at least two mechanisms for long-distance radical cation migration in TWJs: (i) by inefficient charge hopping through the junction; (ii) by a through-space, cross-arm interaction when the AQ is on a flexible tether.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
The one-electron oxidation of DNA creates a radical cation (‘hole’) and results in an irreversible chemical reaction primarily at the 5'-G of Gn (n = 2 or 3) sequences. It is important to understand the mechanism of this process as it is thought to cause mutations that can lead to disease or death (1). Experiments have shown that reaction often occurs at a Gn sequence distant from the site of initial oxidation by migration of a radical cation 200 Å or more (2,3) through the intervening bases of duplex DNA (2,4–7). A thorough investigation of the mechanism of long-distance radical cation migration in DNA indicated that it proceeds according to a process we described as phonon-assisted polaron-like hopping (8–10). In this mechanism, the radical cation is delocalized over several base pairs in a localized structural distortion of the DNA. Thermal activation causes the delocalized radical cation to migrate from site-to-site under the influence of dynamical distortions of the DNA and motions of sodium counter ions and solvating water molecules, which modulate the ionization potential of the constituent bases (11).

Most studies of long-distance radical cation transport in DNA have been carried out using B-form oligonucleotides in solution. In addition, there are reported investigations of charge transfer in guanine quartets (12), DNA triplexes (13,14), DNA/RNA heteroduplexes (15,16) and Z-form DNA (17). However, there have been only a few studies reported of charge transport in DNA structures that contain junctions (18–20), and these have yielded contradictory results.

DNA junctions are branched molecules that arise as intermediates in a number of DNA rearrangement processes. Four-way DNA junctions are formed during homologous recombination, whereas DNA three-way junctions (TWJs) are intermediate structures formed during triplet repeat expansion or deletion (21). These branched DNA structures are substrates for a number of enzymes, which play an important role in the later stages of recombination.

Barton and co-workers examined long-range radical cation migration in four-way junctions using both covalently and non-covalently bound rhodium metallointercalators (19). They reported that light-initiated oxidative reactions at GG steps are observed in each of the four arms of these assemblies when the rhodium metallointercalator is bound to only one arm. From these findings, they conclude that four-way junctions have relatively mobile conformations (19). More recently, Sen and co-workers (20) re-examined long-distance radical cation transport in four-way junctions that are linked to either a rhodium metallointercalator or an anthraquinone (AQ) group as the light-induced one-electron oxidant. They found that reaction at GG steps in each of the four arms occurred when the assembly contained a rhodium metallointercalator, as was observed by Barton, but reaction occurred only at GG steps in the two coaxially stacked arms when AQ was the oxidant. Sen’s investigation of this discrepancy showed that DNA duplexes and four-way junctions that contain rhodium metallointercalators aggregate in solution, which complicates the interpretation of long-distance charge transport reactions that use this oxidant.

We report here an examination of the long-distance charge transport properties of DNA TWJs using covalently linked AQ groups as the oxidant. It is known that the geometry of TWJs depends on the base sequence at the junction, the presence of divalent cations such as Mg2+ (22–25), the length of the third arm (26) and extra bases that form single-stranded loops at the site of the junction (25–29). We examined perfectly base paired (i.e. no extra bases) TWJs for which atomic force microscopy (AFM) studies (30), oligonucleotide ligation-cyclization analysis (31) and fluorescence resonance energy transfer (FRET) (27) data suggest a pyramidal geometry with inter-arm angles between 60 and 90°. The arms of junctions with this structure are not normally stacked coaxially. We also examined a TWJ that incorporates a two base (T2) single strand bulge, which is a more flexible assembly and thus may form structures where two arms can stack coaxially. Our results show that the site of charge injection in TWJs depends on the way the AQ is tethered to the assembly, which complicates the quantitative comparison of charge transport through junctions and simple duplex DNA.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSION
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Materials
[{gamma}-32P]ATP, [{alpha}-32P]ATP and terminal dinucleotide transferase (TdT) were purchased from Amersham Bioscience. T4 polynucleotide kinase (T4PNK) was purchased from New England Biolabs and stored at –20°C. Unmodified DNA oligomers and AQ-containing complementary oligomers were synthesized as described elsewhere on an Applied Biosystems DNA synthesizer (3,32). The extinction coefficients of the oligomers were calculated using an online biopolymer calculator and their concentrations in solution were determined from the absorbance at 260 nm where the AQ was replaced with adenine in the extinction coefficient determination. Reverse phase HPLC was performed on a Hitachi system using a Dynamax C18 column. Atmospheric pressure electron impact ionization mass spectra of the oligomers were obtained by mass spectrometry. UV melting and cooling curves were recorded on a Cary 1E spectrophotometer equipped with a multi-cell block, temperature controller and sample transport accessory. CD spectra were recorded in JASCO J-720 spectropolarimeter.

Melting temperature (Tm) measurements
Melting temperatures were determined for samples consisting of equimolar amounts of DNA oligomers (2.5 µM) in buffer (1.0 ml, pH 7.0) solutions, consisting of different concentrations of sodium phosphate and magnesium acetate. Samples were placed in cuvettes (1.5 ml capacity, 1.0 cm path length) and sealed with tape to prevent evaporation of water during heating/cooling cycles. The absorbance of the samples was measured at 260 nm as a function of temperature for three consecutive runs: cooling from 90 to 10°C at a rate of 0.5°C/min, heating to 90°C followed by re-cooling at 0.5°C/min. The reported Tm values are the peaks in the first derivative plots obtained by fitting the data with Microcal Origin software.

Cleavage analysis by radiolabeling and polyacrylamide gel electrophoresis (PAGE)
DNA oligonucleotides were radiolabeled according to standard procedures at either the 5'-end or 3'-end with [{gamma}-32P]ATP and bacterial T4PNK for 5'-labeling or with [{alpha}-32P]ATP and TdT for 3'-labeling. Radiolabeled DNA was purified by 20% PAGE at 400 V. Samples for irradiation were prepared by hybridizing a mixture of unlabeled (5.0 or 8.0 µM) and radiolabeled (10 000 c.p.m.) oligonucleotides with AQ or non-AQ complementary strands (5.0 or 8.0 µM) in buffer (pH 7.0) containing different concentrations of sodium phosphate and magnesium acetate (to a total volume of 20.0 µl). Hybridization was achieved by heating the samples at 90°C for 10 min, followed by slow cooling to room temperature overnight. Samples were irradiated at ~30°C for 100 min, unless otherwise specified, in microcentrifuge tubes in a Rayonet photoreactor (Southern New England Ultraviolet Co., Barnsford, CT) equipped with eight 350 nm lamps. After irradiation, the samples were precipitated once with cold ethanol (100 µl), in the presence of glycogen (2 µl, 20 mg/ml), washed with 80% ethanol (2 x 100 µl), dried (Speedvac, low heat) and treated with piperidine (100 µl of a 1 M solution) at 90°C for 30 min. After evaporation of the piperidine (Speedvac, medium heat), lyophilization twice with water (20 µl) and dissolution in dye solution (0.25% w/v bromophenol blue, 30% v/v formamide in water), the samples (3000 c.p.m.) were electrophoresed on a 20% 19:1 acrylamide: bisacrylamide gel containing urea (7 M) at 70 W for 90 min. The gels were dried and the cleavage sites were visualized by autoradiography. Quantification of cleavage bands was performed on a Fuji phosphorimager.

Formation of the TWJ was characterized by their mobilities on 20% non-denaturing PAGE in Tris–borate buffer. Four samples were prepared in sodium phosphate buffer (10 mM, pH 7.0) to a total volume of 10 µl. Sample 1 was DNA(1) labeled with 32P at its 5'-end, (see Fig. 1), sample 2 was a mixture of radioactive labeled DNA(1) and unlabeled DNA(2), sample 3 was a mixture of radioactive labeled DNA(1) and unlabeled DNA(3) and sample 4 was a mixture of all three strands of which only DNA(1) had been radiolabeled. All samples were heated to 90°C and then cooled slowly to room temperature, then 5 µl of dye solution (0.25% w/v bromophenol blue, 0.25% w/v xylene cyanol, 30% v/v glycerol in water) was added to each sample. These samples were electrophoresed on a non-denaturing gel at 400 V for 6 h at 5°C. Gel analysis showed that the mobility of sample 4 is less than those of samples 2 and 3, which were less than that of sample 1. These findings indicate formation of a TWJ.



View larger version (32K):
[in this window]
[in a new window]
  		Figure 1. Structures of three-way junctions.

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSION
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Structures and characterization of three-way junctions
The structures of the TWJs used in this work are shown in Figure 1. TWJ(1) is composed of three DNA strands, DNA(1, 2, 3), that assemble to form a TWJ having arms I, II and III. DNA(3) contains a covalently linked AQ derivative at its 5'-end tethered with a ‘short’ (n = 2) or ‘long’ (n = 5) chain (see Fig. 2). On assembly of TWJ(1), the AQ is attached to arm III. Arms I and II of TWJ(1) are ‘symmetrical’ around the junction; the sequence of bases in each strand starting at the junction is the same in these two arms, however, the bases are in the opposite order, they run 5' to 3' in one arm and 3' to 5' in its counterpart. Similarly, TWJ(2)n = 2, which incorporates a single strand (T)2 loop at the junction, is formed from DNA(2, 3, 4). TWJ(3), which is identical to TWJ(1) except that a UAQ (see Fig. 2) replaces the AQ as the one-electron oxidant, is formed from DNA(1, 5, 6). The guanines in the TWJs are identified by a subscript that indicates the number of bases from the strand 5'-terminus to that guanine. Arm III contains only isolated guanines, i.e. there are no GG steps in this arm. Arms I and II contain both isolated guanines and GG steps.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 2. Structure of AQ- and UAQ-linked sensitizers.

 
The TWJs used in this work were characterized by their non-denaturing gel behavior, Tm and by CD spectroscopy carried out in solutions containing sodium phosphate and magnesium acetate at various concentrations. The CD spectra show that each TWJ exists predominantly in a B-form DNA global structure under all conditions examined.

The Tm values of the TWJs were measured to assess the effect of structure and buffer on their stability. These data, summarized in Table 1, show that each of the TWJs studied have similar Tm values and, as expected, they are stabilized in the presence of Mg2+. Interestingly, when the AQ derivative in TWJ(1) is linked by a five carbon chain, two melting transitions are observed on the first heating cycle. A ‘normal’ melting of the TWJ at 43.8°C and a low temperature transition at 13.2°C. The low temperature transition is not seen when the AQ derivative is linked to duplex DNA with the short linker.


View this table:
[in this window]
[in a new window]
  		Table 1. Melting temperature (Tm) data obtained from first derivative plots for three-way DNA junctions
 
Photochemistry of the three-way junctions
Previous studies have shown that electronically excited AQ derivatives react with DNA either by hydrogen atom abstraction or by electron transfer, depending on whether the quinone is bound in a groove, intercalated or end capped. Hydrogen abstraction results in non-selective DNA damage (33,34), whereas electron transfer forms a radical cation that migrates through the duplex DNA and reacts with water primarily at the 5'-G of (G)n sequences (n = 2 or 3). Oxidative damage at a base is revealed as strand cleavage by gel electrophoresis and autoradiography after treatment of radiolabeled (32P) samples with piperidine. The experiments we report here were carried out at low conversion (single hit conditions) under which, on average, each DNA assembly is damaged once or not at all. Under these conditions, the amount of strand cleavage at a particular base is proportional to the amount of reaction at that base if it is assumed, which it commonly is, that reaction of water at each G radical cation (G+·) occurs with the same rate as at every G+·. In this circumstance, the amount of strand cleavage observed is proportional to the rate of radical cation migration to and from that guanine (4). A complete analysis of the migration and reaction of the radical cation in a TWJ assembly requires three independent experiments with 32P labeling on each of the three strands that compose the assembly. For TWJ(1), experiments were performed with the radiolabel (indicated by * in Fig. 4) at the 5'-end of DNA(1), the 5'-end of DNA(2) and the 3'-end of DNA(3).



View larger version (20K):
[in this window]
[in a new window]
  		Figure 4. Strand cleavage of TWJ(1).

 
A 5.0 µM solution of TWJ(1)n = 2, which was labeled with 32P at the 3'-terminus of DNA(3), was formed by its hybridization with DNA(1) and DNA(2). This assembly was irradiated at 350 nm (where only the AQ absorbs) for 100 min under a range of salt and buffer concentrations at pH 7.0. The samples were treated with piperidine (90°C, 30 min) and analyzed; a typical result is shown as an autoradiogram in Figure 3; see Supplementary Material. Light-initiated strand cleavage at guanines is observed and this process is largely independent of the buffer conditions. There are several surprising observations. Even though neither arm I nor arm II is expected to be coaxially stacked on arm III, strand cleavage is observed at each of the GG steps of DNA(3) and there is significant strand cleavage at G26, which is an isolated G near the 3'-terminus of arm II. These results are summarized in Figure 4, where the lengths of the arrows indicate approximately the amount of strand cleavage observed at a particular guanine.



View larger version (24K):
[in this window]
[in a new window]
  		Figure 3. Autoradiogram of TWJ(1)n = 2 labeled on DNA(3) with 32P (3000 c.p.m.) at the 3'-end. Samples containing 5 µM duplex at different salt concentrations (pH 7) were irradiated at 350 nm. Lane 1 is an unirradiated control in 10 mM sodium phosphate buffer solution. Lane 2 is same as lane 1 but irradiated for 100 min. Lanes 3–6 are for samples containing (lane 3) 10 mM sodium phosphate and 1 mM magnesium acetate, (lane 4) 100 mM sodium phosphate, (lane 5) 100 mM sodium phosphate and 1 mM magnesium acetate and (lane 6) 100 mM sodium phosphate and 10 mM magnesium acetate, respectively. All samples were treated with piperidine at 90°C for 30 min after irradiation.

 
We performed a control experiment designed to determine if reaction of the guanines of TWJ(1) are intramolecular or if formation of a diffusible oxidant such as singlet oxygen (35) might lead to an intermolecular reaction. None of the three DNA strands that compose TWJ(4) (Fig. 5) are complementary to any strand of TWJ(1). A 10 mM sodium phosphate buffer solution containing 5.0 µM radiolabeled TWJ(4), which does not contain a covalently linked AQ, and 5.0 µM unlabeled TWJ(1), which does contain a covalently linked AQ, was irradiated for 100 min at 350 nm. After irradiation, the solution was treated with piperidine and analyzed. No strand cleavage could be detected at any base of the labeled strand of TWJ(4), which shows that there are no intermolecular reactions of TWJ(1) under these conditions.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 5. Structure of TWJ(4).

 
The analysis of radical cation migration in TWJ(1)n = 2 was completed by irradiating samples that were radiolabeled at the 3'-end of DNA(1) or the 5'-end of DNA(2). The results are shown schematically in Figure 4. In addition to the strand cleavage sites identified from irradiation of TWJ(1)n = 2 when DNA(3) is labeled, strand cleavage is detected at every GG step of the TWJ, regardless of which arm contains the step. Irradiation of TWJ(1)n = 2 gives strand cleavage at isolated G4, G7, G9 and G26 of DNA(3), but not at G34 of DNA(1) or G2 of DNA(2). The GG steps of TWJ(1)n = 2 in arms I and II react, but only the isolated G of arm II is cleaved. Similarly, cleavage at GG32 and at GG26, which are in arm I, is measurably less efficient than for GG12 and for GG6, which are in arm II. These findings indicate that an asymmetry in the structure of TWJ(1) is reflected in reactions of isolated guanines near the ends of arms I and II, and less so in reactions observed at the GG steps.

Irradiation of TWJ(1)n = 5 gives a somewhat different result. When DNA(3) is radiolabeled, enhanced cleavage is observed at isolated G26. When DNA(2) is radiolabeled, irradiation of TWJ(1)n = 5 gives considerably less cleavage at the GG steps in arm I and no detectable cleavage at isolated G2. Apparently, the pattern of observed strand cleavage is affected less by the asymmetry of the TWJ structure for TWJ(1)n = 2 than it is for TWJ(1)n = 5.

A duplex DNA sample was prepared having the same base pair sequence as in arms II and III of TWJ(1) in order to determine if the damage observed at the isolated guanines is a consequence of the TWJ or due to the specific sequence of bases. A 5.0 µM solution of the duplex oligonucleotide formed from DNA(3) and DNA(7), radiolabeled at its 3'-end (see Fig. 6), was irradiated at 350 nm. Strand cleavage in the duplex is clearly detectable after 10 min of irradiation. Significantly, cleavage is observed at G4, G7 and G9, each of which is located 5' to an A (a reactivity pattern that is commonly observed) (33), and cleavage is detected at GG14 and GG19, but, in contrast to TWJ(1), no significant cleavage is detected at G26, even after 100 min of irradiation. This finding shows that the reaction seen at G26 of TWJ(1) is a property of the TWJ and not of the base sequence of arm II.



View larger version (10K):
[in this window]
[in a new window]
  		Figure 6. Duplex DNA mimicking arms II and III of TWJ(1).

 
TWJ(2)n = 2 contains a single strand TT loop at the junction site. Analysis of TWJ structures reveals that these looped junctions are more flexible than the perfect junction, and we assessed the consequence of this increased flexibility on long-distance radical cation migration. A sample of TWJ(2) was prepared that contained a radiolabel at the 3'-terminus of DNA(3). Samples were irradiated in buffer solutions that contained sodium phosphate and magnesium acetate over a range of concentrations. After irradiation and treatment with piperidine, the samples were analyzed by gel electrophoresis (see Supplementary Material). The results are shown schematically in Figure 7. In the absence of magnesium, when the TWJ is most flexible, there is a similar amount of cleavage at G26 as with TWJ(1), but the amount of cleavage at this position increases when Mg2+ is present. Clearly, the flexibility and stacking arrangement of TWJs affects the amount of strand cleavage observed at G26 of DNA(3).



View larger version (13K):
[in this window]
[in a new window]
  		Figure 7. Strand cleavage of TWJ(2).

 
In TWJ(1)n = 2 or 5 and TWJ(2)n = 2, the AQ group of the AQ is linked to the 5'-oxygen atom of the terminal deoxyribose by a tether that imparts flexibility to the structure of the conjugate. The AQ group is presumed to be end capped, i.e. close to and parallel with the terminal base pair of the duplex (4), intercalated or bound in a groove. However, the data indicate that the instantaneous position of the AQ is determined by a dynamic equilibrium that is affected by the length of the tether. In certain structural variants of TWJ(1) and TWJ(2) arm I or II may be close enough to the terminus of arm III that, with an extended conformation of the tether, the AQ group could intercalate or end cap arm I or II. In this circumstance, what appears to be a long-distance reaction [cleavage at G26 of DNA(3), for example] could actually be a local reaction of an AQ somehow associated with arm II. To assess this possibility, we prepared and examined TWJ(3), which contains a UAQ (see Fig. 2) in place of the AQ sensitizer. Previous work in this laboratory has shown that UAQ, which has an AQ group linked through a one atom tether to the 2'-oxygen atom of a ribose, is capable of sensitizing long-distance strand cleavage in duplex DNA (3). Rather than being end capped or groove bound, the AQ group of UAQ-containing duplexes is primarily intercalated at the 3'-side of the base to which it is bound. Consequently, it is less likely that the AQ of a UAQ-containing TWJ can reach across the structure and interact directly with a ‘distant’ arm.

TWJ(3) is identical in base sequence to TWJ(1) except that the AQ bound to the 5'-terminus of DNA(3) is removed and replaced at T26 of DNA(2) by a UAQ bound near to the 3'-terminus of DNA(5). Irradiation and analysis of TWJ(3) having the 3'-terminus of DNA(6) radiolabeled reveals the pattern of reactivity shown schematically in Figure 8 (see Supplementary Material). Strand cleavage is detected at isolated guanines G4, G7 and G9 and at GG steps GG14 and GG19, but, in contrast to TWJ(1), not at G26. This finding shows that specific linkage of the AQ group at the terminus of arm III affects the reaction selectivity near the terminus of arm II, which indicates some inter-arm interaction leading to strand cleavage that is independent of the path through the junction.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 8. Strand cleavage of TWJ(3).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSION
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Our study of long-distance radical cation migration in TWJs expands the recently reported work on four-way junctions (19,20). A key question raised in the study of four-way junctions concerns the ability of a radical cation to migrate from one arm to another that is not stacked coaxially with the first. Barton and co-workers, using a rhodium metallointercalator as oxidant, reported that this is possible (19), while Sen and co-workers, using an AQ oxidant, suggested that it is not (20). The results reported here indicate that for TWJs the answer is not simply yes or no.

Different geometries for perfect TWJs have been suggested by different investigators (23,27,36–38). These include: (i) a 3-fold symmetric ‘Y’ shape with inter-arm angles of 120°; (ii) a 2-fold symmetric ‘Y’ shape with an acute angle of 60° between two arms and 150° between the other two adjacent arms; (iii) an asymmetric ‘Y’ with three different inter-arm angles; (iv) a ‘T’-shaped form with two inter-arm angles of 90°; (v) a trigonal pyramid with three 60° inter-arm angles; (vi) a pyramidal form with three inter-arm angles of 90°. A recent AFM study on TWJs has shown that they are not flat but have a pyramid-like shape (30) and that the mean angle between the arms can vary from 57 to 80°. Apart from in the T-shaped structure, which appears to be unlikely, the arms of TWJs will not normally stack coaxially.

In order to put the charge migration results in TWJs in context, it is important to note that the efficiency of strand cleavage from irradiation of the AQ-linked TWJ(1) is considerably less than is typically observed for duplex DNA. This conclusion is rooted in the observation that 100 min of irradiation is required to obtain significant cleavage at GG steps of TWJ(1) compared with just 10 min for the duplex equivalent to arms III + II. We reported previously that the quantum yield of strand cleavage for duplex DNA with an AQ linked to a 5'-end having a leading AAA sequence is ~4.5% (39). Unfortunately, the structure of TWJ(1) does not lend itself to measurement of cleavage efficiency (low conversion is required and there are too many cleavage sites that react similarly to permit separation and analysis of products). Nevertheless, it is clear that the efficiency of reaction of TWJ(1) is more than 10 times less than that expected for duplex DNA. This reduction in efficiency indicates that processes that usually play a relatively minor role in duplex DNA reactions, such as strand cleavage at a G located 5' to an A, may become more important in TWJ(1). In fact, this is just what is observed: in TWJ(1) strand cleavage at G4, G7 and G9 of DNA(3) is more efficient than at GG14 and GG19, which suggests that the junction is a bottleneck to radical cation migration.

The most surprising result from irradiation of TWJ(1) is efficient strand cleavage on arm II at G26 of DNA(3), particularly when the linker with n = 5 is irradiated. G26 is an ‘isolated’ guanine that is not 5' to an A. Efficient strand cleavage in duplex DNA is not seen at similar sites. Significantly, G34 of DNA(1) and G2 of DNA(2), which are isolated guanines in ‘equivalent’ locations on arm I of TWJ(1), do not react and the GG steps in arm I are consistently less reactive than their counterparts in arm II. In addition, if charge injection into TWJ(1) occurred exclusively at the end of arm III, where the AQ is attached, then the strand cleavage pattern should not depend on whether n = 2 or 5. The observation that the length of this linker affects the reaction outcome requires that there be at least two modes of charge injection, one of which we associate with cross-arm interactions of the AQ.

Figure 9 is a schematic representation of two possible structures for TWJ(1). Of course, in reality there is a dynamic mixture of many structures that includes the two presented in Figure 9 among them, but the observations we report can be accommodated by consideration of those shown. In isomer A, the AQ group is stacked on arm III, whereas in isomer B the AQ is stacked on or intercalated in arm II, which, because of the asymmetrical pyramidal structure of the TWJ, is suggested to be closer to the AQ-labeled terminus of arm III than is arm I. The longer linker (n = 5) gives more isomer B. This proposal accommodates the asymmetrical reactivity of TWJ(1) and may account for the unusual strand cleavage at the isolated guanines at the end of this arm if the remote AQ binds nearby.



View larger version (9K):
[in this window]
[in a new window]
  		Figure 9. Three-way junction structures.

 
We prepared TWJ(2) and TWJ(3) to test this hypothesis. In the latter case, UAQ replaces AQ as the one-electron oxidant. It is on a shorter tether and its location, intercalated 3' to the base to which it is bound, should make it less able to participate in cross-arm reactions. This seems to be precisely the case because irradiation of TWJ(3) does not result in detectable strand cleavage at the isolated guanine near the end of arm II. Significantly, strand cleavage is observed at GG14 and GG19 of DNA(6), which indicates that long-distance charge migration can occur from arm III to arm II even though they are not normally stacked coaxially. However, the reaction efficiency of TWJ(3) is also much lower than is typical of duplex DNA and the reaction at G4, G7 and G9 of DNA(6) is about the same as that seen at G13G and G18G, which indicates again that the junction presents a barrier to radical cation migration.

Similar conclusions are reached from examination of TWJ(2), which differs from TWJ(1) by the presence of two unpaired thymine bases at the junction. The presence of such additional bases changes the structure of the TWJ from pyramidal to the more extended T-shaped structure. Gel electrophoresis studies have demonstrated that these bulged junctions have greater stability and can undergo ion-induced conformational change (25,40). The results of irradiation of TWJ(2) are subtly different from those of TWJ(1). The junction still presents a barrier to radical cation migration, since the reactivity is similar to TWJ(1) and G4 and G7 are at least as reactive as GG14 and GG19 of DNA(3). Strand cleavage is detected at the isolated guanine of arm II, but it is far less efficient than it is for TWJ(1) and it increases when Mg2+ is present in the solution. These findings show that the efficiency and path for radical cation migration depends upon certain structural details of the TWJ assembly.


   CONCLUSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
SUPPLEMENTARY MATERIAL
 REFERENCES
 
Analysis of the migration of radical cations in assemblies containing TWJs is complex because two paths operate. One path appears to proceed through the junction and presumably requires at least transient coaxial stacking of the junction arms. We conclude that this path has low efficiency because much longer irradiation times are required to achieve comparable reaction than for corresponding duplex DNA. The second path bypasses the junction and occurs as a result of a through-space interaction of an AQ linked by a flexible tether to arm III with another arm of the TWJ. The evidence supporting the existence of this second path is circumstantial but convincing: the degree of cross-arm reaction depends on the length and flexibility of the tether and the way the AQ is bound to the DNA. The existence of two paths for long-distance reaction complicates the analysis of radical cation transport through the junction and no quantitative conclusions concerning the nature of radical cation transfer through DNA junctions can be reached. However, these results shed light on the previous reports concerning four-way junctions and on radical cation migration in DNA generally. It is possible that cross-arm interactions can also occur in some four-way junctions, which may explain the different results obtained with metallointercalators (19) and AQs (20) as the electron acceptor.


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 SUPPLEMENTARY MATERIAL
REFERENCES
 
Supplementary Material is available at NAR Online.


   ACKNOWLEDGEMENTS
 
We thank Dr Sriram Kanvah of Georgia Institute of Technology for his assistance in the preparation of DNA samples and for a critical reading of this manuscript. This work was supported by the National Science Foundation and by the Vassar Woolley Foundation, for which we are grateful.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 SUPPLEMENTARY MATERIAL
 REFERENCES
 

  1. Burrows,C.J. and Muller,J.G. (1998) Oxidative nucleobase modifications leading to strand scission. Chem. Rev., 98, 1109–1154.[CrossRef][Web of Science][Medline]

  2. Nunez,M., Hall,D.B. and Barton,J.K. (1999) Long-range oxidative damage to DNA: effects of distance and sequence. Chem. Biol., 6, 85–97.[Medline]

  3. Ly,D., Sanii,L. and Schuster,G.B. (1999) Charge transport in DNA: internally-linked anthraquinones support a mechanism requiring dynamic structural distortion. J. Am. Chem. Soc., 12, 9400–9410.

  4. Schuster,G.B. (2000) Long-range charge transfer in DNA: transient structural distortions control the distance dependence. Acc. Chem. Res., 33, 253–260.[CrossRef][Web of Science][Medline]

  5. Nakatani,K., Sando,S. and Saito,I. (2000) Recognition of a single guanine bulge by 2-acylamino-1,8-naphthyridine. J. Am. Chem. Soc., 122, 2172–2177.

  6. Giese,B. (2000) Long-distance charge transport: the hopping mechanism. Acc. Chem. Res., 33, 631–636.[CrossRef][Web of Science][Medline]

  7. Wagenknecht,H.A., Rajski,S.R., Pascaly,M., Stemp,E.D. and Barton,J.K. (2001) Direct observation of radical intermediates in protein-dependent DNA charge transport. J. Am. Chem. Soc., 123, 4400–4407. [CrossRef][Web of Science][Medline]

  8. Henderson,P.T., Jones,D., Hampikian,G., Kan,Y. and Schuster,G.B. (1999) Long-distance charge transport in duplex DNA: the phonon-assisted polaron-like hopping mechanism. Proc. Natl Acad. Sci. USA, 96, 8353–8358.[Abstract/Free Full Text]

  9. Wan,C.Z., Fiebig,T., Schiemann,O., Barton,J.K. and Zewail,A.H. (2000) Femtosecond direct observation of charge transfer between bases in DNA. Proc. Natl Acad. Sci. USA, 97, 14052–14055.[Abstract/Free Full Text]

  10. Kendrick,T. and Giese,B. (2002) Charge transfer through DNA triggered by site selective injection into adenine. Chem. Commun., 18, 2016–2017.[CrossRef]

  11. Barnett,R.N., Cleveland,C.L., Joy,A., Landman,U. and Schuster,G.B. (2001) Charge migration in DNA: ion-gated transport. Science, 294, 567–571.[Abstract/Free Full Text]

  12. Szalai,V.A., Singer,M.J. and Thorp,H.H. (2000) Site-specific probing of oxidative reactivity and telomerase function using 7,8-dihydro-8-oxoguanine in telomeric DNA. J. Am. Chem. Soc., 122, 4524–4525.

  13. Kan,Y. and Schuster,G.B. (1999) Radical cation transport and reaction in triplex DNA: long-range guanine damage. J. Am. Chem. Soc., 11607–11614.

  14. Nunez,M. and Barton,J.K. (2000) Probing DNA charge transport with metallointercalators. Curr. Opin. Chem. Biol., 4, 199–206.[CrossRef][Web of Science][Medline]

  15. Odom,D.T. and Barton,J.K. (2001) Long-range oxidative damage in DNA/RNA duplexes. Biochemistry, 40, 8727–8737.[CrossRef][Medline]

  16. Sartor,V., Henderson,P.T. and Schuster,G.B. (1999) Radical cation transport and reaction in RNA/DNA hybrid duplexes: effect of global structure on reactivity. J. Am. Chem. Soc., 121, 11027–11033.

  17. Abdou,I.M., Sartor,V., Cao,H. and Schuster,G.B. (2001) Long-distance radical cation migration in Z-form DNA. J. Am. Chem. Soc., 123, 6696–6697.[CrossRef][Web of Science][Medline]

  18. Fahlman,R.P. and Sen,D. (2002) DNA conformational switches as sensitive electronic sensors of analytes. J. Am. Chem. Soc., 124, 4610–4616.[CrossRef][Web of Science][Medline]

  19. Odom,D.T., Dill,E.A. and Barton,J.K. (2001) Charge transport through DNA four-way junctions. Nucleic Acids Res., 29, 2026–2033.[Abstract/Free Full Text]

  20. Fahlman,R.P., Sharma,R.D. and Sen,D. (2002) The charge conduction properties of DNA Holliday junctions depend critically on the identity of the tethered photooxidant. J. Am. Chem. Soc., 124, 12477–12485.[CrossRef][Web of Science][Medline]

  21. Lilley,D.M. (2000) Structures of helical junctions in nucleic acids. Q. Rev. Biophys., 33, 109–159.[CrossRef][Web of Science][Medline]

  22. Ma,R.I., Kallenbach,N.R., Sheardy,R.D., Petrillo,M.L. and Seeman,N.C. (1986) Three-arm nucleic acid junctions are flexible. Nucleic Acids Res., 14, 9745–9753.[Abstract/Free Full Text]

  23. Duckett,D.R. and Lilley,D.M. (1990) The three-way DNA junction is a Y-shaped molecule in which there is no helix-helix stacking. EMBO J., 9, 1659–1664.[Web of Science][Medline]

  24. Lu,M., Guo,Q., Seeman,N.C. and Kallenbach,N.R. (1989) DNase I cleavage of branched DNA molecules. J. Biol. Chem., 264, 20851–20854.[Abstract/Free Full Text]

  25. Leontis,N.B., Kwok,W. and Newman,J.S. (1991) Stability and structure of three-way DNA junctions containing unpaired nucleotides. Nucleic Acids Res., 19, 759–766.[Abstract/Free Full Text]

  26. Zhong,M. and Kallenbach,N.R. (1993) Conformation and thermodynamics of DNA "necks". Models for three-arm branch formation in a duplex. J. Mol. Biol., 230, 766–778.[CrossRef][Web of Science][Medline]

  27. Stuhmeier,F., Welch,J.B., Murchie,A.I., Lilley,D.M. and Clegg,R.M. (1997) Global structure of three-way DNA junctions with and without additional unpaired bases: a fluorescence resonance energy transfer analysis. Biochemistry, 36, 13530–13538.[CrossRef][Medline]

  28. Shlyakhtenko,L.S., Appella,E., Harrington,R.E., Kutyavin,I. and Lyubchenko,Y.L. (1994) Structure of three-way DNA junctions. 2. Effects of extra bases and mismatches. J. Biomol. Struct. Dyn., 12, 131–143.[Web of Science][Medline]

  29. Rosen,M.A. and Patel,D.J. (1993) Structural features of a three-stranded DNA junction containing a C-C junctional bulge. Biochemistry, 32, 6576–6587.[CrossRef][Medline]

  30. Shlyakhtenko,L.S., Potaman,V.N., Sinden,R.R., Gall,A.A. and Lyubchenko,Y.L. (2000) Structure and dynamics of three-way DNA junctions: atomic force microscopy studies. Nucleic Acids Res., 28, 3472–3477.[Abstract/Free Full Text]

  31. Lyubchenko,Y.L., Shlyakhtenko,L.S., Appella,E. and Harrington,R.E. (1993) CA runs increase DNA flexibility in the complex of lambda Cro protein with the OR3 site. Biochemistry, 32, 4121–4127.[CrossRef][Medline]

  32. Santhosh,U. and Schuster,G.B. (2002) Long-range charge transport in duplex DNA: anthraquinone sensitization results are independent of terminal ionic distribution. J. Am. Chem. Soc., 124, 10986–10987.[CrossRef][Web of Science][Medline]

  33. Armitage,B.A., Yu,C., Devadoss,C. and Schuster,G.B. (1994) Cationic anthraquinone derivatives as catalytic DNA photonucleases: mechanisms for DNA damage and quinone recycling. J. Am. Chem. Soc., 116, 9847–9859.

  34. Breslin,D.T. and Schuster,G.B. (1996) Anthraquinone photonucleases: mechanisms for GG-selective and nonselective cleavage of double-stranded DNA. J. Am. Chem. Soc., 118, 2311–2319.

  35. Rogers,M.A.J. and Snowden,P.T. (1982) Lifetime of singlet oxygen in liquid water as determined by time-resolved luminescence measurements. J. Am. Chem. Soc., 104, 5541–5543.

  36. Guo,Q., Lu,M., Churchill,M.E., Tullius,T.D. and Kallenbach,N.R. (1990) Asymmetric structure of a three-arm DNA junction. Biochemistry, 29, 10927–10934.[CrossRef][Medline]

  37. Lu,M., Guo,Q. and Kallenbach,N.R. (1991) Effect of sequence on the structure of three-arm DNA junctions. Biochemistry, 30, 5815–5820.[CrossRef][Medline]

  38. Shlyakhtenko,L.S., Rekesh,D., Lindsay,S.M., Kutyavin,I., Appella,E., Harrington,R.E. and Lyubchenko,Y.L. (1994) Structure of three-way DNA junctions. 1. Non-planar DNA geometry. J. Biomol. Struct. Dyn., 11, 1175–1189.[Web of Science][Medline]

  39. Sanii,L. and Schuster,G.B. (2000) Long-distance charge transport in DNA: sequence-dependent radical cation injection efficiency. J. Am. Chem. Soc., 122, 11545–11546.

  40. Welch,J.B., Duckett,D.R. and Lilley,D.M. (1993) Structures of bulged three-way DNA junctions. Nucleic Acids Res., 21, 4548–4555.[Abstract/Free Full Text]


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 (252K) Freely available
Right arrow Supplementary Material
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (7)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Santhosh, U.
Right arrow Articles by Schuster, G. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Santhosh, U.
Right arrow Articles by Schuster, G. B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?