Interaction of the dye ethidium bromide with DNA containing guanine repeats.

DNA containing one or more copies of the motifs repeated in telomere sequences has unusual conformational properties. The isolated sequence from the protozoan Oxytricha, dT4G4 has the potential to form tetramers in the presence of sodium or potassium ions. We report here that these tetramers bind ethidium tightly, with an interaction that fulfills several criteria for an intercalative mechanism in the G sequence. By contrast, the 4-fold tandem repeat of this subunit, d(T4G4)4, does not interact specifically with ethidium in the presence of Na+. This difference might have a simple structural basis: the tetramer of dT4G4 forms a stack of four G-quartets in the presence of Na+ or K+, whereas the constraint imposed by the T4 "tethers" in the repeat d(T4G4)4 allows only two layers to form in the presence of Na+. In the presence of sufficient K+, the latter can partially form a four-layer G-quartet structure, which interacts with ethidium. This idea is supported by analysis of a "relaxed" sequence, dT4G4(T7G4)3, which allows formation of four G-quartets and binds ethidium in the presence of Na+ as well as K+. Ethidium (and intercalators generally) should thus be able to retard or inhibit the action of telomerase in the presence of K+.     

Adenine affects the structure and stability of telomeric sequences.

Adenine occurs in the strand containing repeated G clusters in the telomeric DNA of a variety of organisms, including that of humans. The role of adenine has been investigated by constructing two sets of oligonucleotides each with one, two, or four copies of the telomeric sequence dTTTAGGG together with a control sequence in which T replaces the A residue, dTTTTGGG. Comparison of the stability and spectral properties of these two sequences in the presence of Na+ or K+ affords a basis for defining the role of adenine in these structures. In Na+, the A residue stabilizes the structure formed by each oligomer significantly, presumably by a base-pairing interaction with T. In K+, by contrast, there is little difference in stability. In two- and four-copy oligomers, the A sequence has a different structure from its T analog, as detected by CD spectroscopy. In the presence of either Na+ or K+, the tetraplexes of A and T interact with intercalators.     

The solution structure of d(G(4)T(4)G(3))(2): a bimolecular G-quadruplex with a novel fold

The G-rich 11-mer oligonucleotide d(G(4)T(4)G(3)) forms a bimolecular G-quadruplex in the presence of sodium ions with a topology that is distinct from the folds of the closely related and well-characterized sequences d(G(4)T(4)G(4)) and d(G(3)T(4)G(3)). The solution structure of d(G(4)T(4)G(3))(2) has been determined using a combination of NMR spectroscopy and restrained molecular dynamics calculations. d(G(4)T(4)G(3))(2) forms an asymmetric dimeric fold-back structure consisting of three stacked G-quartets. The two T(4) loops that span diagonally across the outer faces of the G-quartets assume different conformations. The glycosidic torsion angle conformations of the guanine bases are 5'-syn-anti-syn-anti-(T(4) loop)-anti-syn-anti in one strand and 5'-syn-anti-syn-anti-(T(4) loop)-syn-anti-syn in the other strand. The guanine bases of the two outer G-quartets exhibit a clockwise donor-acceptor hydrogen-bonding directionality, while those of the middle G-quartet exhibit the anti-clockwise directionality. The topology of this G-quadruplex, like other bimolecular fold-back structures with diagonal loops, places each strand of the G-quartet region next to a neighboring parallel and an anti-parallel strand. The two guanine residues not involved in G-quartet formation, G4 and G12 (i.e. the fourth guanine base of one strand and the first guanine base of the other strand), adopt distinct conformations. G4 is stacked on top of an adjacent G-quartet, and this base-stacking continues along with the bases of the loop residues T5 and T6. G12 is orientated away from the core of G-quartets; stacked on the T7 base and apparently involved in hydrogen-bonding interactions with the phosphodiester group of this same residue. The cation-dependent folding of the d(G(4)T(4)G(3))(2) quadruplex structure is distinct from that observed for similar sequences. While both d(G(4)T(4)G(4)) and d(G(3)T(4)G(3)) form bimolecular, diagonally looped G-quadruplex structures in the presence of Na(+), K(+) and NH(4)(+), we have observed this folding to be favored for d(G(4)T(4)G(3)) in the presence of Na(+), but not in the presence of K(+) or NH(4)(+). The structure of d(G(4)T(4)G(3))(2) exhibits a "slipped-loop" element that is similar to what has been proposed for structural intermediates in the folding pathway of some G-quadruplexes, and therefore provides support for the feasibility of these proposed transient structures in G-quadruplex formation.     

A comparative study on the interaction of acridine and synthetic bis-acridine with G-quadruplex structure

DNA from the telomeres contains a stretch of simple tandemly repeated sequences in which clusters of G residues alternate with clusters of T/A sequences along one DNA strand. Model telomeric G-clusters form four-stranded structures in presence of Na(I), K(I) and NH(4)(I) ions. Electrophoretic and spectroscopic studies were made with the telomeric related sequences d(T6G16) or d(G4T2G4T2G4T2G4). It was noticed earlier that G-quadruplex may either be inter-molecular, or intra-molecular, or a mixture of both. CD spectral characteristics of various G-quadruplex DNA suggests that the CD maximum at 293 nm corresponds to that of an intra-molecular G-quadruplex structure or hairpin dimers. Fluorescence titration studies also show that acridine and the bis-acridine are interacting with G-quadruplex DNA and destabilize the K(I)-quadruplex structure more efficiently than the quadruplex formed by NH(4)(I) ion. Among the two drugs studied, acridine is more capable of breaking the G-quadruplex structure than bis-acridine. This result is further confirmed by the CD experiments.     

Functional role and affinity of inorganic cations in stabilizing the tetrameric structure of the KcsA K+ channel

Crystal structures of the tetrameric KcsA K+ channel reveal seven distinct binding sites for K+ ions within the central pore formed at the fourfold rotational symmetry axis. Coordination of an individual K+ ion by eight protein oxygen atoms within the selectivity filter suggests that ion-subunit bridging by cation-oxygen interactions contributes to structural stability of the tetramer. To test this hypothesis, we examined the effect of inorganic cations on the temperature dependence of the KcsA tetramer as monitored by SDS-PAGE. Inorganic cations known to permeate or strongly block K+ channels (K+, Rb+, Cs+, Tl+, NH4+, Ba2+, and Sr2+) confer tetramer stability at higher temperatures (T0.5 range = 87 degrees C to >99 degrees C) than impermeant cations and weak blockers (Li+, Na+, Tris+, choline+; T0.5 range = 59 degrees C to 77 degrees C). Titration of K+, Ba2+, and other stabilizing cations protects against rapid loss of KcsA tetramer observed in 100 mM choline Cl at 90 degrees C. Tetramer protection titrations of K+, Rb+, Cs+, Tl+, and NH4+ at 85 degrees C or 90 degrees C exhibit apparent Hill coefficients (N) ranging from 1.7 to 3.3 and affinity constants (K0.5) ranging from 1.1 to 9.6 mM. Ba2+ and Sr2+ titrations exhibit apparent one-site behavior (N congruent with 1) with K0.5 values of 210 nM and 11 microM, respectively. At 95 degrees C in the presence of 5 mM K+, titration of Li+ or Na+ destabilizes the tetramer with K0.5 values of 57 mM and 109 mM, respectively. We conclude that specific binding interactions of inorganic cations with the selectivity filter are an important determinant of tetramer stability of KscA.     

d(G3T4G4) forms unusual dimeric G-quadruplex structure with the same general fold in the presence of K+, Na+ or NH4+ ions

We have recently communicated that DNA oligonucleotide d(G(3)T(4)G(4)) forms a dimeric G-quadruplex in the presence of K(+) ions [J. Am. Chem. Soc.2003, 125, 7866-7871]. The high-resolution NMR structure of d(G(3)T(4)G(4))(2) G-quadruplex exhibits G-quadruplex core consisting of three stacked G-quartets. The two overhanging G3 and G11 residues are located at the opposite sides of the end G-quartets and are not involved in G-quartet formation. d(G(3)T(4)G(4))(2) G-quadruplex represents the first bimolecular G-quadruplex where end G-quartets are spanned by diagonal (T4-T7) as well as edge-type loops (T15-T18). Three of the G-rich strands are parallel while one is anti-parallel. The G12-G22 strand demonstrates a sharp reversal in strand direction between residues G19 and G20 that is accommodated with the leap over the middle G-quartet. The reversal in strand direction is achieved without any extra intervening residues. Here we furthermore examined the influence of different monovalent cations on the folding of d(G(3)T(4)G(4)). The resolved imino and aromatic proton resonances as well as (sequential) NOE connectivity patterns showed only minor differences in key intra- and interquartet NOE intensities in the presence of K(+), Na(+) and NH(4)(+) ions, which were consistent with subtle structural differences while retaining the same folding topology of d(G(3)T(4)G(4))(2) G-quadruplex.     

Novel DNA superstructures formed by telomere-like oligomers.

DNA oligomers containing three or more contiguous guanines form tetrastranded parallel complexes, G4-DNA, in the presence of alkali cations. However, oligomers that have a single multi-guanine motif at their 3' or 5' end, with a guanine as the terminal base, also form higher order products. Thus, the oligomer T8G3T forms a unique G4-DNA product at neutral pH in the presence of Na+, K+, or Rb+; however, its isomeric counterpart T9G3 in K+ or Rb+ generates an additional ladder of products of substantially lower gel mobility. We show that these larger complexes contain, respectively, 8, 12, or 16 distinct strands of oligomer. The octamer structure formed by T9G3 assembles in moderate salt at room temperature and melts around 60 degrees C in 100 mM KCl. Methylation protection experiments suggest a nested head-to-tail superstructure containing two tetraplexes bonded front-to-back via G quartets formed by out-of-register guanines. Naturally occurring chromosomal telomeres, which all have guanines at their 3' termini, may be able to form these superstructures.     

Cleavage of stem-and-loop structure DNA by bleomycin. Reaction on the bacteriophage G4 origin of complementary strand synthesis

The cleavage by bleomycin-Fe(II) complex in the presence of dithiothreitol of 3'-or 5'-end-labeled DNA from the region of the bacteriophage G4 origin of complementary strand synthesis was investigated by using the DNA-sequencing technique. Bleomycin cleaved a single-stranded DNA substrate preferentially at inverted repeat sequences, which potentially form stem-and-loop structures, while it cleaved double-stranded DNA substrates with different specificity. The results support the formation of three adjoining stem-and-loop structures in the region of the phage G4 origin of complementary strand synthesis under the low-salt conditions used and suggest a difference in the form of the double helix between the stem and the double-stranded DNA fragment. Bleomycin appears to be a useful reagent for searching stem-and-loop structures. The results may also contribute to the understanding of the mode of action of bleomycin as an antitumor antibiotic.     

Examining the contribution of a dA+dT element to the conformation of Escherichia coli integration host factor-DNA complexes.

DNA binding proteins that induce structural changes in DNA are common in both prokaryotes and eukaryotes. Integration host factor (IHF) is a multi-functional DNA binding and bending protein of Escherichia coli that can mediate protein-protein and protein-DNA interactions by bending DNA. Previously we have shown that the presence of a dA+dT element 5'-proximal to an IHF consensus sequence can affect the binding of IHF to a particular site. In this study the contribution of various sequence elements to the formation of IHF-DNA complexes was examined. We show that IHF bends DNA more when it binds to a site containing a dA+dT element upstream of its core consensus element than to a site lacking a dA+dT element. We demonstrate that IHF can be specifically crosslinked to DNA with binding sites either containing or lacking this dA+dT element. These results indicate the importance of flanking DNA and a dA+dT element in the binding and bending of a site by IHF.     

Affinity and sequence specificity of DNA binding and site selection for primer synthesis by Escherichia coli primase

Primase is an essential DNA replication enzyme in Escherichia coli and responsible for primer synthesis during lagging strand DNA replication. Although the interaction of primase with single-stranded DNA plays an important role in primer RNA and Okazaki fragment synthesis, the mechanism of DNA binding and site selection for primer synthesis remains unknown. We have analyzed the energetics of DNA binding and the mechanism of site selection for the initiation of primer RNA synthesis on the lagging strand of the replication fork. Quantitative analysis of DNA binding by primase was carried out using a number of oligonucleotide sequences: oligo(dT)(25) and a 30 bp oligonucleotide derived from bacteriophage G4 origin (G4ori-wt). Primase bound both sequences with moderate affinity (K(d) = 1.2-1.4 x 10(-)(7) M); however, binding was stronger for G4ori-wt. G4ori-wt contained a CTG trinucleotide, which is a preferred site for initiation of primer synthesis. Analysis of DNA binding isotherms derived from primase binding to the oligonucleotide sequences by fluorescence anisotropy indicated that primase bound to DNA as a dimer, and this finding was further substantiated by electrophoretic mobility shift assays (EMSAs) and UV cross-linking of the primase-DNA complex. Dissection of the energetics involved in the primase-DNA interaction revealed a higher affinity of primase for DNA sequences containing the CTG triplet. This sequence preference of primase may likely be responsible for the initiation of primer synthesis in the CTG triplet sites in the E. coli lagging strand as well as in the origin of replication of bacteriophage G4.     

Preferential binding of quinolones to DNA with alternating G,C / A,T sequences: a spectroscopic study

The binding of quinolones, nalidixic acid (Nal), oxolinic acid (Oxo) with double stranded polynucleotides was undertaken by using UV-melting, UV-Vis absorption, fluorescence and CD spectroscopic techniques. The binding of Nal or Oxo to the polynucleotides under low-salt buffer conditions were determined for poly (dA).(dT), poly [d(A-T)], poly (dG).(dC), poly [d(G-C)] and E. coli DNA. The fluorescence data were analyzed using a previously established two step mechanism with two different DNA-Drug complexes [Rajeswari et al., Biochemistry 26, 6825-31 (1987)]. The first complex [DN](1) with a binding constant K(1), is formed where the interactions are 'nonspecific' and complex [DN](2) with a binding constant K(2), is formed where the interactions are "specific" which involve (additional) hydrophobic type of interactions like 'stacking' of the drug and the overall association constant is represented as K(=K(1)K(2)). The order of binding for Nal and Oxo is: poly [d(G-C)] > poly [d(A- T)] > E. coli > poly (dG).(dC) > poly (dA).(dT). Interaction of quinolones seems to be preferential in the alternating G, C or A, T stretches of DNA than those of non-alternating. Within any alternating or non-alternating in DNA sequences the G, C rich sequences have distinctly greater binding than A, T sequences. The overall association constant data (K) reveal higher binding of Oxo to DNA compared to Nal to any given polynucleotide investigated; which also explains the higher antibacterial potency of Oxo. Changes in the absorption difference spectra and in circular dichroic spectra also manifest these results. As the melting temperatures of the polynucleotides were only marginally raised in presence of the quinolone, we rule out the possibility of 'classical intercalation' of the drug. Amino group of guanine facilitates the binding of quinolones and therefore has the greater binding with the DNA. However, poly (dG).(dC) is known to exist in 'A' conformation which is not adopted by quinolones as in the case of poly (dA).(dT). Present results suggest that Nal or Oxo bind to DNA in a non-classical fashion which is partially stacking in nature.     

fd gene 5 protein binds to double-stranded polydeoxyribonucleotides poly(dA.dT) and poly[d(A-T).d(A-T)]

Circular dichroism (CD) data indicated that fd gene 5 protein (G5P) formed complexes with double-stranded poly(dA.dT) and poly[d(A-T).d(A-T)]. CD spectra of both polymers at wavelengths above 255 nm were altered upon protein binding. These spectral changes differed from those caused by strand separation. In addition, the tyrosyl 228-nm CD band of G5P decreased more than 65% upon binding of the protein to these double-stranded polymers. This reduction was significantly greater than that observed for binding to single-stranded poly(dA), poly(dT), and poly[d(A-T)] but was similar to that observed for binding of the protein to double-stranded RNA [Gray, C.W., Page, G.A., & Gray, D.M. (1984) J. Mol. Biol. 175, 553-559]. The decrease in melting temperature caused by the protein was twice as great for poly[d(A-T).d(A-T)] as for poly(dA.dT) in 5 mM tris(hydroxymethyl)aminomethane hydrochloride (Tris-HCl), pH 7. Upon heat denaturation of the poly(dA.dT)-G5P complex, CD spectra showed that single-stranded poly(dA) and poly(dT) formed complexes with the protein. The binding of gene 5 protein lowered the melting temperature of poly(dA.dT) by 10 degrees C in 5 mM Tris-HCl, pH 7, but after reducing the binding to the double-stranded form of the polymer by the addition of 0.1 M Na+, the melting temperature was lowered by approximately 30 degrees C. Since increasing the salt concentration decreases the affinity of G5P for the poly(dA) and poly(dT) single strands and increases the stability of the double-stranded polymer, the ability of the gene 5 protein to destabilize poly(dA.dT) appeared to be significantly affected by its binding to the double-stranded form of the polymer.     

Isolation and construction of mutants of the G4 minus strand origin: analysis of their in vivo activity

An active, rifampicin-resistant primase-dependent bacteriophage G4 origin of complementary DNA strand synthesis has been cloned as a 274 bp fragment into the filamentous phase M13 and its secondary structure altered by deletion and insertion. It has been found that the entire 136 bp G4 intergenic region containing the secondary structure loops I and III is necessary for rifampicin-resistant conversion of SS----RF DNA in vivo. The secondary structures, however, can be widely separated by insertion between them of both random DNA sequences, and sequences that form strong additional secondary structure configurations and the origins still retain activity. Primase therefore probably recognises two DNA domains on loops I and III, the physical separation of which is not important.     

Probing minor groove hydrogen bonding interactions between RB69 DNA polymerase and DNA

Minor groove hydrogen bonding (HB) interactions between DNA polymerases (pols) and N3 of purines or O2 of pyrimidines have been proposed to be essential for DNA synthesis from results obtained using various nucleoside analogues lacking the N3 or O2 contacts that interfered with primer extension. Because there has been no direct structural evidence to support this proposal, we decided to evaluate the contribution of minor groove HB interactions with family B pols. We have used RB69 DNA pol and 3-deaza-2'-deoxyadenosine (3DA), an analogue of 2-deoxyadenosine, which has the same HB pattern opposite T but with N3 replaced with a carbon atom. We then determined pre-steady-state kinetic parameters for the insertion of dAMP opposite dT using primer/templates (P/T)-containing 3DA. We also determined three structures of ternary complexes with 3DA at various positions in the duplex DNA substrate. We found that the incorporation efficiency of dAMP opposite dT decreased 10(2)-10(3)-fold even when only one minor groove HB interaction was missing. Our structures show that the HB pattern and base pair geometry of 3DA/dT is exactly the same as those of dA/dT, which makes 3DA an optimal analogue for probing minor groove HB interactions between a DNA polymerase and a nucleobase. In addition, our structures provide a rationale for the observed 10(2)-10(3)-fold decrease in the rate of nucleotide incorporation. The minor groove HB interactions between position n - 2 of the primer strand and RB69pol fix the rotomer conformations of the K706 and D621 side chains, as well as the position of metal ion A and its coordinating ligands, so that they are in the optinal orientation for DNA synthesis.     

The cohering telomeres of Oxytricha.

We have studied the process by which purified Oxytricha macronuclear DNA associates with itself to form large aggregates. The various macronuclear DNA molecules all have the same terminal or telomeric DNA sequences that are shown below. 5' C4A4C4A4C4--mean length----G4T4G4T4G4T4G4T4G4 G4T4G4T4G4T4G4T4G4-----2.4 kb------C4A4C4A4C4. When incubated at high concentrations, these telomeric sequences cohere with one another to form an unusual structure--one that is quite different from any DNA structure so far described. The evidence for this is the following: 1) These sequences cohere albeit slowly, in the presence of relatively high concentrations of Na+, and no other cation tested. This contrasts with the rapid coherence of complementary single-chain terminals of normal DNA (sticky ends) which occurs in the presence of any cation tested. 2) If the cohered form is transferred into buffers containing a special cation, K+, it becomes much more resistant to dissociation by heating. We estimate that K+ increases the thermal stability by 25 degrees or more. The only precedent known (to us) for a cation-specific stabilization is that seen in the quadruplex structure formed by poly I. The thermal stability of double helical macronuclear DNA depends on the cation concentration, but not the cation type. Limited treatment with specific nucleases show that the 3' and 5'-ended strands are essential for the formation of the cohering structure. Once in the cohered form, the telomeric sequences are protected from the action of nucleases. Coherence is inhibited by specific, but not by non-specific, synthetic oligomers, and by short telomeric fragments with or without their terminal single chains. We conclude that the coherence occurs by the formation of a novel condensed structure that involves the terminal nucleotides in three or four chains.     

Iron(II)-ethylenediaminetetraacetic acid catalyzed cleavage of RNA and DNA oligonucleotides: similar reactivity toward single- and double-stranded forms

Fe(II)-EDTA catalyzes the cleavage of nucleic acids with little or no base-sequence specificity. We have now studied the preference of this reagent in catalyzing the cleavage of single- versus double-stranded nucleic acid structures. Three RNA and two DNA molecules, each expected to contain both single- and double-stranded regions, were synthesized and their structures characterized by enzymatic digestion using secondary structure specific nucleases. Fe(II)-EDTA catalyzed nearly uniform strand scission along the entire length of each molecule; no correlation with secondary structure was observed. The homopolymer sequence dA30:dT30, embedded in a mixed-sequence context to promote exact register of the homopolymer tract, was cleaved to an extent similar to that of flanking sequences. The reactions were relatively insensitive to K+, Na+, and Mg2+ in the range 10-100 mM and were quenched by Tris-HCl buffer. We conclude that the Fe(II)-EDTA-catalyzed strand scission reaction does not discriminate between typical single- and double-stranded regions, which simplifies the interpretation of experiments in which the reaction is used to probe the tertiary structure of RNA molecules [Latham, J. A., & Cech, T. R. (1989) Science 245, 276-282].     

Inhibition of DNA replication by berenil in plasmids containing Poly(dA)poly(dT) sequences

The effect of berenil on plasmid DNA replication was studied on pBR322-derived plasmids containing poly(dA)poly(dT) sequences. In comparison to the parental plasmid pBR322, plasmid pKH47 harboring 100 bp of poly(dA)poly(dT) at the PvuII site showed a decrease in plasmid yield in the presence of berenil. This effect was also observed in pVL26, a related plasmid in which the location of the poly(dA)poly(dT) region had been shifted to the EcoRV site in pBR322. [(3)H]Thymidine incorporation experiments indicated that DNA synthesis may be affected in these plasmids in the presence of the drug. Bromodeoxyuridine incorporation experiments coupled to Cs(2)SO(4) equilibrium density gradient centrifugation indicated that the lower plasmid yield was due to an inhibition of DNA replication by berenil. We have also found that berenil induces DNA degradation in plasmids containing the homopolymer. Our studies strongly suggest that the effect of berenil on plasmid replication and DNA stability results from its binding to the poly(dA)poly(dT) region present in these plasmids. Moreover, we have found a correlation between the position of the poly(dA)poly(dT) region and this inhibitory effect. Thus, plasmid pKH47, containing the poly(dA)poly(dT) region most proximal to the origin of pBR322 replication, was most severely affected.     

Long (dA)n.(dT)n tracts can form intramolecular triplexes under superhelical stress.

Plasmids containing long tracts of (dA)n.(dT)n have been prepared and their conformations examined in linear and supercoiled DNA using a series of chemical and enzymic probes which are known to be sensitive to unusual DNA structures. Under superhelical stress and in the presence of magnesium the sequence T69.A69 adopts a conformation at pH 8.0 consistent with the formation of an intramolecular DNA triplex. Site specific cleavage of the supercoiled plasmid by single-strand specific nucleases occurs within the A.T insert; the 5'-end of the purine strand is sensitive to reaction with diethylpyrocarbonate while the central 5-6 bases of the pyrimidine strand are reactive to osmium tetroxide. By contrast shorter inserts of A33.T33 and A23.T23 do not appear to form unusual structures.     

Monovalent cation-induced structure of telomeric DNA: the G-quartet model

We have investigated the structures formed by oligonucleotides composed of two or four repeats of the telomeric sequences from Oxytricha and Tetrahymena. The Oxytricha four-repeat molecule (d(T4G4)4 = Oxy-4) forms structures with increased electrophoretic mobility in nondenaturing gels containing Na+, K+, or Cs+, but not in gels containing Li+ or no added salt. Formation of the folded structure results in protection of a set of dG's from methylation by dimethyl sulfate. Efficient UV-induced cross-links are observed in Oxy-4 and the related sequence from Tetrahymena (d(T2G4)4 = Tet-4), and join thymidine residues in different repeats. Models proposed to account for these data involve G-quartets, hydrogen-bonded structures formed from four guanosine residues in a square-planar array. We propose that the G-quartet structure must be dealt with in vivo by the telomere replication machinery.