The three-dimensional structure of a DNA hairpin in solution two-dimensional NMR studies and structural analysis of d(ATCCTATTTATAGGAT)

Minidumbbell structures formed by ATTCT pentanucleotide repeats in spinocerebellar ataxia type 10

Spinocerebellar ataxia type 10 (SCA10) is a progressive genetic disorder caused by ATTCT pentanucleotide repeat expansions in intron 9 of the ATXN10 gene. ATTCT repeats have been reported to form unwound secondary structures which are likely linked to large-scale repeat expansions. In this study, we performed high-resolution nuclear magnetic resonance spectroscopic investigations on DNA sequences containing two to five ATTCT repeats. Strikingly, we found the first two repeats of all these sequences well folded into highly compact minidumbbell (MDB) structures. The 3D solution structure of the sequence containing two ATTCT repeats was successfully determined, revealing the MDB comprises a regular TTCTA and a quasi TTCT/A pentaloops with extensive stabilizing loop-loop interactions. We further carried out in vitro primer extension assays to examine if the MDB formed in the primer could escape from the proofreading function of DNA polymerase. Results showed that when the MDB was formed at 5-bp or farther away from the priming site, it was able to escape from the proofreading by Klenow fragment of DNA polymerase I and thus retained in the primer. The intriguing structural findings bring about new insights into the origin of genetic instability in SCA10.

High-Resolution Structures of DNA Minidumbbells Comprising Type II Tetraloops with a Purine Minor Groove Residue

Minidumbbell (MDB) is a newly discovered DNA structure formed by native sequences, which serves as a possible structural intermediate causing repeat expansion mutations in the genome and also a functional structural motif in constructing DNA-based molecular switches. Until now, all the reported MDBs containing two adjacent type II tetraloops were formed by pyrimidine-rich sequences 5'-YYYR YYYR-3' (Y and R represent pyrimidine and purine, respectively), wherein the second and sixth residues folded into the minor groove and interacted with each other. In this study, we have conducted a high-resolution nuclear magnetic resonance (NMR) spectroscopic investigation on alternative MDB-forming sequences and discovered that an MDB could also be formed stably with a purine in the minor groove, which has never been observed in any previously reported DNA type II tetraloops. Our refined NMR solution structures of the two MDBs formed by 5'-CTTG CATG-3' and 5'-CTTG CGTG-3' reveal that the sixth purine residue was driven into the minor groove via base-base stacking with the second thymine residue and adenine stacked better than guanine. The results of our present research work expand the sequence criteria for the formation of MDBs and shed light to explore the significance of MDBs.

Molecular Junctions Inspired by Nature: Electrical Conduction through Noncovalent Nanobelts

Charge transport occurs in a range of biomolecular systems, whose structures have covalent and noncovalent bonds. Understanding from these systems have yet to translate into molecular junction devices. We design junctions which have hydrogen-bonds between the edges of a series of prototype noncovalent nanobelts (NCNs) and vary the number of donor-acceptors to study their electrical properties. From frontier molecular orbitals (FMOs) and projected density of state (DOS) calculations, we found these NCN dimer junctions to have low HOMO-LUMO gaps and states at the Fermi level, suggesting these are metallic-like systems. Their conductance properties were studied with nonequilibrium Green's functions density functional theory (NEGF-DFT) and was found to decrease with cooperative H-bonding, that is, the conductance decreased as the alternating donor-acceptors around the nanobelts attenuates to a uniform distribution in the H-bonding arrays. The latter gave the highest conductance of 51.3 × 10^-6 S and the Seebeck coefficient showed n-type (-36 to -39 μV K^-1) behavior, while the lower conductors with alternating H-bonds are p-type (49.7 to 204 μV K^-1). In addition, the NCNs have appreciable binding energies (19.8 to 46.1 kcal mol^-1), implying they could form self-assembled monolayer (SAM) heterojunctions leading to a polymeric network for long-range charge transport.

Sequence Effect on the Formation of DNA Minidumbbells

The DNA minidumbbell (MDB) is a recently identified non-B structure. The reported MDBs contain two TTTA, CCTG, or CTTG type II loops. At present, the knowledge and understanding of the sequence criteria for MDB formation are still limited. In this study, we performed a systematic high-resolution nuclear magnetic resonance (NMR) and native gel study to investigate the effect of sequence variations in tandem repeats on the formation of MDBs. Our NMR results reveal the importance of hydrogen bonds, base-base stacking, and hydrophobic interactions from each of the participating residues. We conclude that in the MDBs formed by tandem repeats, C-G loop-closing base pairs are more stabilizing than T-A loop-closing base pairs, and thymine residues in both the second and third loop positions are more stabilizing than cytosine residues. The results from this study enrich our knowledge on the sequence criteria for the formation of MDBs, paving a path for better exploring their potential roles in biological systems and DNA nanotechnology.

The competing mini-dumbbell mechanism: new insights into CCTG repeat expansion

CCTG repeat expansions in intron 1 of the cellular nucleic acid-binding protein gene are associated with myotonic dystrophy type 2. Recently, we have reported a novel mini-dumbbell (MDB) structure formed by two CCTG or TTTA repeats, which potentially has a critical role in repeat expansions. Here we present a mechanism, called the competing MDB mechanism, to explain how the formation of MDB can lead to efficient mismatch repair (MMR) escape and thus CCTG repeat expansions during DNA replication. In a long tract of CCTG repeats, two competing MDBs can be formed in any segment of three repeats. Fast exchange between these MDBs will make the commonly occupied repeat behave like a mini-loop. Further participations of the 5'- or 3'-flanking repeat in forming competing MDBs will make the mini-loop shift in the 5'- or 3'-direction, thereby providing a pathway for the mini-loop to escape from MMR. To avoid the complications due to the formation of hairpin conformers in longer CCTG repeats, we made use of TTTA repeats as model sequences to demonstrate the formation of competing MDBs and shifting of mini-loop in a long tract of repeating sequence.

Minidumbbell: A New Form of Native DNA Structure

The non-B DNA structures formed by short tandem repeats on the nascent strand during DNA replication have been proposed to be the structural intermediates that lead to repeat expansion mutations. Tetranucleotide TTTA and CCTG repeat expansions have been known to cause reduction in biofilm formation in Staphylococcus aureus and myotonic dystrophy type 2 in human, respectively. In this study, we report the first three-dimensional minidumbbell (MDB) structure formed by natural DNA sequences containing two TTTA or CCTG repeats. The formation of MDB provides possible pathways for strand slippage to occur, which ultimately leads to repair escape and thus expansion mutations. Our result here shows that MDB is a highly compact structure composed of two type II loops. In addition to the typical stabilizing interactions in type II loops, MDB shows extensive stabilizing forces between the two loops, including two distinctive modes of interactions between the minor groove residues. The formation of MDB enriches the structural diversity of natural DNA sequences, reveals the importance of loop-loop interactions in unusual DNA structures, and provides insights into novel mechanistic pathways of DNA repeat expansion mutations.

Anomalous Separation of Small Y-Chromosomal DNA Fragments on Microchip Electrophoresis

We investigated an anomalous DNA separation where two DNA fragments from the human Y-chromosome sY638 (64 bp) and sY592 (65 bp), with only one base pair difference, were separated. This result is abnormal since in a previous study, we found that 5 bp was the minimum difference between two DNA fragments that the microchip electrophoresis system can separate. The formation of a mini-loop in the structure of the DNA fragment of sY638 (64 bp) was strongly expected to be the reason. To investigate this, we synthesized three modified DNA fragments for sY638 (64 bp), and the modifications were in two expected locations for possible mini-loop formation. Later, the separation between sY592 (65 bp) and the three modified fragments of sY638 (64 bp) was not possible. Thus, we conclude that the formation of a mini-loop in the structure of the DNA is the reason behind this anomalous separation.

A historical account of Hoogsteen base-pairs in duplex DNA

In 1957, a unique pattern of hydrogen bonding between N3 and O4 on uracil and N7 and N6 on adenine was proposed to explain how poly(rU) strands can associate with poly(rA)-poly(rU) duplexes to form triplexes. Two years later, Karst Hoogsteen visualized such a noncanonical A-T base-pair through X-ray analysis of co-crystals containing 9-methyladenine and 1-methylthymine. Subsequent X-ray analyses of guanine and cytosine derivatives yielded the expected Watson-Crick base-pairing, but those of adenine and thymine (or uridine) did not yield Watson-Crick base-pairs, instead favoring "Hoogsteen" base-pairing. More than two decades ensued without experimental "proof" for A-T Watson-Crick base-pairs, while Hoogsteen base-pairs continued to surface in AT-rich sequences, closing base-pairs of apical loops, in structures of DNA bound to antibiotics and proteins, damaged and chemically modified DNA, and in polymerases that replicate DNA via Hoogsteen pairing. Recently, NMR studies have shown that base-pairs in duplex DNA exist as a dynamic equilibrium between Watson-Crick and Hoogsteen forms. There is now little doubt that Hoogsteen base-pairs exist in significant abundance in genomic DNA, where they can expand the structural and functional versatility of duplex DNA beyond that which can be achieved based only on Watson-Crick base-pairing. Here, we provide a historical account of the discovery and characterization of Hoogsteen base-pairs, hoping that this will inform future studies exploring the occurrence and functional importance of these alternative base-pairs.

Unusual structures of TTTA repeats in icaC gene of Staphylococcus aureus

One and two TTTA repeat expansions have been found in the coding region of icaC gene of Staphylococcus aureus variants which influence the expression of IcaC protein and alter the phenotype. Yet, the mechanism of these small-size TTTA repeat expansions remains unclear. In this study, we performed high-resolution nuclear magnetic resonance spectroscopic studies on TTTA repeats. Our results show that a DNA sequence containing three TTTA repeats can fold into dumbbell structures with a 3' or 5'-overhang. Exchange of these dumbbells makes the sequence behave like a 2-nt TT mini-loop at 25°C. The occurrence of these mini-loop and dumbbell structures in the nascent strand during DNA replication provides possible mechanistic pathways which account for one and two repeat expansions.

Nucleic acid folding determined by mesoscale modeling and NMR spectroscopy: solution structure of d(GCGAAAGC)

Determination of DNA solution structure is a difficult task even with the high-sensitivity method used here based on simulated annealing with 35 restraints/residue (Cryoprobe 750 MHz NMR). The conformations of both the phosphodiester linkages and the dinucleotide segment encompassing the sharp turn in single-stranded DNA are often underdetermined. To obtain higher quality structures of a DNA GNRA loop, 5'-d(GCGAAAGC)-3', we have used a mesoscopic molecular modeling approach, called Biopolymer Chain Elasticity (BCE), to provide reference conformations. By construction, these models are the least deformed hairpin loop conformation derived from canonical B-DNA at the nucleotide level. We have further explored this molecular conformation at the torsion angle level with AMBER molecular mechanics using different possible (epsilon,zeta) constraints to interpret the 31P NMR data. This combined approach yields a more accurate molecular conformation, compatible with all the NMR data, than each method taken separately, NMR/DYANA or BCE/AMBER. In agreement with the principle of minimal deformation of the backbone, the hairpin motif is stabilized by maximal base-stacking interactions on both the 5'- and 3'-sides and by a sheared G.A mismatch base pair between the first and last loop nucleotides. The sharp turn is located between the third and fourth loop nucleotides, and only two torsion angles beta6 and gamma6 deviate strongly with respect to canonical B-DNA structure. Two other torsion angle pairs epsilon3,zeta3 and epsilon5,zeta5 exhibit the newly recognized stable conformation BIIzeta+ (-70 degrees, 140 degrees). This combined approach has proven to be useful for the interpretation of an unusual 31P chemical shift in the 5'-d(GCGAAAGC)-3' hairpin.

Measurement of eight scalar and dipolar couplings for methine-methylene pairs in proteins and nucleic acids

A new 3D, spin-state-selective coherence transfer NMR experiment is described that yields accurate measurements for eight scalar or dipolar couplings within a spin system composed of a methylene adjacent to a methine group. Implementations of the experiment have been optimized for proteins and for nucleic acids. The experiments are demonstrated for Cbeta-Calpha moieties of the third IgG-binding domain from Streptococcal Protein G (GB3) and for C5'-C4' groups in a 24-nt RNA oligomer. Chemical shifts of Calpha, Cbeta and Hbeta (respectively C4', C5' and H5') are dispersed in the three orthogonal dimensions, and the absence of heteronuclear decoupling leads to distinct and well-resolved E.COSY multiplet patterns. In an isotropic sample, the E.COSY displacements correspond to 1J(CalphaHalpha), 2J(CalphaHbeta2)+2J(CalphaHbeta3), 2J(CbetaHalpha), 1J(CbetaHbeta2)+1J(CbetaHbeta3), 1J(CbetaHbeta2)-2J(Hbeta2Hbeta3), 1J(CbetaHbeta3)-2J(Hbeta2Hbeta3), 3J(HalphaHbeta2) and 3J(HalphaHbeta3) for proteins, and 1J(C4'H4'), 2J(C4'H5')+2J(C4'H5"), 2J(C5'H4'), 1J(C5'H5')+1J(C5'H5"), 1J(C5'H5')-2J(H5'H5"), 1J(C5'H5")-2J(H5'H5"), 3J(H4'H5') and 3J(H4'H5") in nucleic acids. The experiment, based on relaxation-optimized spectroscopy, yields best results when applied to residues where the methine-methylene group corresponds to a reasonably isolated spin system, as applies for C, F, Y, W, D, N and H residues in proteins, or the C5'-C4' groups in nucleic acids. Splittings can be measured under either isotropic or weakly aligned conditions, yielding valuable structural information both through the 3J couplings and the one-, two- and three-bond dipolar interactions. Dipolar couplings for 10 out of 13 sidechains in GB3 are found to be in excellent agreement with its X-ray structure, whereas one residue adopts a different backbone geometry, and two residues are subject to extensive chi1 rotamer averaging. The abundance of dipolar couplings can also yield stereospecific assignments of the non-equivalent methylene protons. For the RNA oligomer, dipolar data yielded stereospecific assignments for six out of the eight C5'H2 groups in the loop region of the oligomer, in all cases confirmed by 1J(C5'H5')>1J(C5'H5"), and H5' resonating downfield of H5".

Solution structure of a nitrous acid induced DNA interstrand cross-link

Nitrous acid is a mutagenic agent. It can induce interstrand cross-links in duplex DNA, preferentially at d(CpG) steps: two guanines on opposite strands are linked via a single shared exocyclic imino group. Recent synthetic advances have led to the production of large quantities of such structurally homogenous cross-linked duplex DNA. Here we present the high resolution solution structure of the cross-linked dodecamer [d(GCATCCGGATGC)]2 (the cross-linked guanines are underlined), determined by 2D NMR spectroscopy, distance geometry, restrained molecular dynamics and iterative NOE refinement. The cross-linked guanines form a nearly planar covalently linked 'G:G base pair' with only minor propeller twisting, while the cytidine bases of their normal base pairing partners have been flipped out of the helix and adopt well defined extrahelical positions in the minor groove. On the 5'-side of the cross-link, the minor groove is widened to accommodate these extrahelical bases, and the major groove becomes quite narrow at the cross-link. The cross-linked 'G:G base pair' is well stacked on the spatially adjacent C:G base pairs, particularly on the 3'-side guanines. In addition to providing the first structure of a nitrous acid cross-link in DNA, these studies could be of major importance to the understanding of the mechanisms of nitrous acid cross-linking and mutagenicity, as well as the mechanisms responsible for its repair in intracellular environments. It is also the shortest DNA cross-link structure to be described.

Structures and stabilities of small DNA dumbbells with Watson-Crick and Hoogsteen base pairs

The structures and stabilities of cyclic DNA octamers of different sequences have been studied by NMR and CD spectroscopy and by restrained molecular dynamics. At low oligonucleotide concentrations, some of these molecules form stable monomeric structures consisting of a short stem of two base pairs connected by two mini-loops of two residues. To our knowledge, these dumbbell-like structures are the smallest observed to date. The relative stabilities of these cyclic dumbbells have been established by studying their melting transitions. Dumbbells made up purely of GC stems are more stable than those consisting purely of AT base pairs. The order of the base pairs closing the loops also has an important effect on the stabilities of these structures. The NMR data indicate that there are significant differences between the solution structures of dumbbells with G-C base pairs in the stem compared to those with A-T base pairs. In the case of dumbbells with G-C base pairs, the residues in the stem form a short segment of a BDNA helix stabilized by two Watson-Crick base pairs. In contrast, in the case of d<pCATTCATT>, the stem is formed by two A-T base pairs with the glycosidic angles of the adenine bases in a syn conformation, most probably forming Hoogsteen base pairs. Although the conformations of the loop residues are not very well defined, the thymine residues at the first position of the loop are observed to fold back into the minor groove of the stem.

Protonation of deoxycytidine residues in dC4 tetraloops: UV spectrophotometric study of dC10 and d(A14C4T14)

It is shown that component analysis could be applied to study the UV difference spectra of cytidine oligomers and hairpin oligonucleotides with cytidines in the loop region in order to account for the melting and titration results in terms of cytidine stacking and protonation. Upon acid titration, the dC(10) oligomer undergoes cooperative conformational transition at pH 6.3 accompanied by protonation and formation of the i-structure with half of the residues protonated. The stability of the hemiprotonated structure increases with decreasing pH, the i-structure persisting still in the region of pH<pK of cytidine. An UV difference spectrum that reflects the stacking/unstacking of hemiprotonated cytidine residues was acquired from the melting and titration experiments of the dC(10) oligomer and used to describe the behavior of the dC(4) loop of the hairpin oligonucleotide d(A(14)C(4)T(14)). It is shown that upon titration, the 50% level of protonation of the deoxycytidine tetraloop is attained at pH 5.0. Simultaneously, the stacking interactions of cytidine residues reach the maximum at this pH with two residues stacked, and thereafter decline again. Only marginal stabilization of the oligomer hairpin (DeltaT(m)=1.5 degrees C) is found to accompany the formation of this single hemiprotonated dC.dC(+) base pair. We propose that at pH 5 the cytidines of the dC(4) loop form a hemiprotonated dC.dC(+) pair stacked with the last dA.dT base pair of the hairpin stem.

Proton NMR studies of 5'-d-(TC)(3) (CT)(3) (AG)(3)-3'--a paperclip triplex: the structural relevance of turns

In this study, we present the results of structural analysis of an 18-mer DNA 5'-T(1)C(2)T(3)C(4)T(5)C(6)C(7)T(8)C(9)T(10)C(11)T(12)A(13)G(14)A(15)G(16)A(17)G(18)-3' by proton nuclear magnetic resonance (NMR) spectroscopy and molecular modeling. The NMR data are consistent with characteristics for triple helical structures of DNA: downfield shifting of resonance signals, typical for the H3(+) resonances of Hoogsteen-paired cytosines; pH dependence of these H3(+) resonance; and observed nuclear Overhauser effects consistent with Hoogsteen and Watson-Crick basepairing. A three-dimensional model for the triplex is developed based on data obtained from two-dimensional NMR studies and molecular modeling. We find that this DNA forms an intramolecular "paperclip" pyrimidine-purine-pyrimidine triple helix. The central triads resemble typical Hoogsteen and Watson-Crick basepairing. The triads at each end region can be viewed as hairpin turns stabilized by a third base. One of these turns is comprised of a hairpin turn in the Watson-Crick basepairing portion of the 18-mer with the third base coming from the Hoogsteen pairing strand. The other turn is comprised of two bases from the continuous pyrimidine portion of the 18-mer, stabilized by a hydrogen-bond from a purine. This "triad" has well defined structure as indicated by the number of nuclear Overhauser effects and is shown to play a critical role in stabilizing triplex formation of the internal triads.

The first example of a Hoogsteen base-paired DNA duplex in dynamic equilibrium with a Watson-Crick base-paired duplex--a structural (NMR), kinetic and thermodynamic study

A single-point substitution of the O4' oxygen by a CH2 group at the sugar residue of A6 (i.e. 2'-deoxyaristeromycin moiety) in a self-complementary DNA duplex, 5'-d(C1G2C3G4A5A6T7T8C9G10C11G12)2(-3), has been shown to steer the fully Watson-Crick basepaired DNA duplex (1A), akin to the native counterpart, to a doubly A6:T7 Hoogsteen basepaired (1B) B-type DNA duplex, resulting in a dynamic equilibrium of (1A)<==>(1B): Keq = k1/k(-1) = 0.56+/-0.08. The dynamic conversion of the fully Watson-Crick basepaired (1A) to the partly Hoogsteen basepaired (1B) structure is marginally kinetically and thermodynamically disfavoured [k1 (298K) = 3.9 0.8 sec(-1); deltaHdegrees++ = 164+/-14 kJ/mol; -TdeltaS degrees++ (298K) = -92 kJ/mol giving a deltaG degrees++ 298 of 72 kJ/mol. Ea (k1) = 167 14 kJ/mol] compared to the reverse conversion of the Hoogsteen (1B) to the Watson-Crick (1A) structure [k-1 (298K) = 7.0 0.6 sec-1, deltaH degrees++ = 153 13 kJ/mol; -TdeltaSdegrees++ (298K) = -82 kJ/mol giving a deltaGdegrees++(298) of 71 kJ/mol. Ea (k-1) = 155 13 kJ/mol]. Acomparison of deltaGdegrees++(298) of the forward (k1) and backward (k-1) conversions, (1A)<==>(1B), shows that there is ca 1 kJ/mol preference for the Watson-Crick (1A) over the double Hoogsteen basepaired (1B) DNA duplex, thus giving an equilibrium ratio of almost 2:1 in favour of the fully Watson-Crick basepaired duplex. The chemical environments of the two interconverting DNA duplexes are very different as evident from their widely separated sets of chemical shifts connected by temperature-dependent exchange peaks in the NOESY and ROESY spectra. The fully Watson-Crick basepaired structure (1A) is based on a total of 127 intra, 97 inter and 17 cross-strand distance constraints per strand, whereas the double A6:T7 Hoogsteen basepaired (1B) structure is based on 114 intra, 92 inter and 15 cross-strand distance constraints, giving an average of 22 and 20 NOE distance constraints per residue and strand, respectively. In addition, 55 NMR-derived backbone dihedral constraints per strand were used for both structures. The main effect of the Hoogsteen basepairs in (1B) on the overall structure is a narrowing of the minor groove and a corresponding widening of the major groove. The Hoogsteen basepairing at the central A6:T7 basepairs in (1B) has enforced a syn conformation on the glycosyl torsion of the 2'-deoxyaristeromycin moiety, A6, as a result of substitution of the endocyclic 4'-oxygen in the natural sugar with a methylene group in A6. A comparison of the Watson-Crick basepaired duplex (1A) to the Hoogsteen basepaired duplex (1B) shows that only a few changes, mainly in alpha, sigma and gamma torsions, in the sugar-phosphate backbone seem to be necessary to accommodate the Hoogsteen basepair.

Multiple four-stranded conformations of human telomere sequence d(CCCTAA) in solution

By detailed NMR analysis of a human telomere repeating unit, d(CCCTAA), we have found that three distinct tetramers, each of which consists of four symmetric single-strands, slowly exchange in a slightly acidic solution. Our new finding is a novel i-motif topology (T:-form) where T4 is intercalated between C1 and C2 of the other duplex. The other two tetramers have a topology where C1 is intercalated between C2 and C3 of the other parallel duplex, resulting in the non-stacking T4 residues (R-form), and a topology where C1 is stacked between C3 and T4 of the other duplex (S:-form). From the NMR denaturation profile, the R-form is the most stable of the three structures in the temperature range of 15-50 degrees C, the S:-form the second and the T:-form the least stable. The thermodynamic parameters indicate that the T-form is the most enthalpically driven and entropically opposed, and its population is increased with decreasing temperature. The T-form structure determined by restrained molecular dynamics calculation suggests that inter-strand van der Waals contacts in the narrow grooves should contribute to the enthalpic stabilization of the T-form.

Stable sheared A.C pair in DNA hairpins

Single-residue d(Pu1NPu2) (Pu1.Pu2=G.A, G.G or A.A) hairpin loops can be stably closed by sheared purine.purine pairs. These special motifs have been found in several important biological systems. We now extend these loop-closing base-pairs to a sheared purine. pyrimidine (A.C) pair at a neutral pH condition. High-resolution NMR spectroscopy, distance geometry, and molecular dynamics methods were used to study d(GTACANCGTAC) oligomers. Numerous idiosyncratic nuclear Overhauser enhancements, especially those across the A.C base-pair between C4NH2left and right arrow AH1', C4NH2left and right arrow AH2, and CH5left and right arrow AH2 proton pairs, clearly define the novel sheared nature of the closing A.C base-pair. This novel base-pair is possibly present in several biological systems and in two single-stranded DNA aptamers selected from oligonucleotide libraries.

Cross-strand purine-pyrimidine stack and sheared purine.pyrimidine pairing in the human HIV-1 reverse transcriptase inhibitors

Cross-strand homo purine-purine (G-G or A-A) stacks and sheared purine.purine pairing have been found to be important motifs in nucleic acid duplex structures. We now report novel cross-strand purine-pyrimidine (A-C) and hetero purine-purine (G-A) stacks that are established from a sheared purine.pyrimidine (A.C) pair adjacent to a sheared G.A pair in the 5'-AA/GC-3' sequence. This "internal loop" sequence is conserved in two families of single-stranded DNA inhibitors of the reverse transcriptase of type 1 human immunodeficiency virus. The distorted backbone of these inhibitors, resulting from the unique helical twists and kinks in the 5'-AA/GC-3' sequence, may be responsible for the increased affinities of these single-stranded DNA inhibitors as compared with other regular B-form duplex substrates. Two simple rules have been generalized to account for all reported cross-strand stacks.

Structure determination and analysis of helix parameters in the DNA decamer d(CATGGCCATG)2 comparison of results from NMR and crystallography

The solution structure of the DNA decamer (CATGGCCATG)2 has been determined by NMR spectroscopy and restrained molecular dynamic and distance geometry calculations. The restrainted data set includes interproton distances and torsion angles for the deoxyribose sugar ring which were obtained by nuclear Overhauser enhancement intensities and quantitative simulation of cross-peaks from double quantum filtered correlation spectroscopy. The backbone torsion angles were constrained using experimental data from NOE cross-peaks, 1H-1H and 1H-31P-coupling constants. The NMR structure and the crystal structure of the DNA decamer deviates from the structure of the canonical form of B-DNA in a number of observable characteristics. Particularly, both structures display a specific pattern of stacking interaction in the central GGC base triplet. Furthermore, a specific local conformation of the TG/CA base-pair step is present in NMR and crystal structure, highlighting the unusually high flexibility of this DNA duplex part. The solution structure of the TG/CA base-pair step obtained by our high resolution NMR study is characterized by a positive roll angle, whereas in crystal this base-pair step tends to adopt remarkably high twist angles.

Comparative structural analysis by [1H,31P]-NMR and restrained molecular dynamics of two DNA hairpins from a strong DNA topoisomerase II cleavage site

The structural analysis of two single-stranded DNAs d(AGCTTATCATCGATAAGCT) (ATC-19) and d(AGCTTATCGATGATAAGCT) (GAT-19) was performed by NMR and restrained molecular dynamics. These oligonucleotides reproduce the 15-33 segment of phage pBR322 DNA, which contains a strong cleavage site for topoisomerase II coupled to the antitumor drugs VP-16 and ellipticine. Because of their partial palindromic nature, the two oligonucleotides ATC-19 and GAT-19 may fold back into stable hairpin structures, consisting of a stem of eight base-pairs and a loop of three residues. NMR assignments and conformational parameters were determined from combined 2D NOESY, COSY and 1H-31P spectra. Conformations of ATC-19 and GAT-19 hairpins were calculated using the X-PLOR 3.1 program. Structures were generated through simulated annealing procedures starting from 50 structures with randomized torsion angles. A good convergence was observed for ATC-19 molecules, while no consensus was found for GAT-19. Within the GAT-19 loop, the base stacking was poor and no hydrogen bond could be detected. In contrast, ATC-19 displayed a well-defined three residue loop stabilized by both extensive base stackings and hydrogen bonding between the N3 atom of the adenine ring and the amino group of the cytosine ring. The results confirm our earlier ATC-19 structure obtained by a completely different calculation procedure (JUMNA) and the higher thermal stability of ATC-19 compared to GAT-19. Moreover, due to its mismatched base-pair, the ATC-19 loop may be better described as a single residue loop rather than a three residue loop. Comparison of this loop to those containing sheared purine.purine base-pairs revealed striking resemblances, particularly on the backbone angle combination. Finally, the differences observed between the ATC-19 and GAT-19 structures could help toward understanding the sequential cleavage of DNA strands by topoisomerase II.

Molecular structure of two crystal forms of cyclic triadenylic acid at 1A resolution

The three dimensional structures of cyclic deoxytriadenylic acid, c-d(ApApAp), from two different trigonal crystal forms (space groups P3 and R32) have been determined by x-ray diffraction analysis at 1A resolution. Both structures were solved by direct methods and refined by anisotropic least squares refinement to R-factors of 0.109 and 0.137 for the P3 and R32 forms, respectively. In both crystal forms, each of the two independent c-d(ApApAp) molecules sits on the crystallographic 3-fold axis. All four independent c-d(ApApAp) molecules have similar backbone conformations. The deoxyriboses are in the S-type pucker with pseudorotation angles ranging from 156.7 degrees to 168.6 degrees and the bases have anti glycosyl torsion angles (chi falling in two ranges, one at -104.3 degrees and the other ranging from -141.0 degrees to -143.8 degrees). In the R32 form, a hexahydrated cobalt(II) ion is found to coordinate through bridging water molecules to N1, N3, and N7 atoms of three adjacent adenines and oxygen atoms of phosphates. Comparison with other structures of cyclic oligonucleotides indicates that the sugar adopts N-type pucker in cyclic dinucleotides and S-type pucker in cyclic trinucleotides, regardless whether the sugar is a ribose or a deoxyribose.

Kinetics of conformational fluctuations in DNA hairpin-loops

The kinetics of DNA hairpin-loop fluctuations has been investigated by using a combination of fluorescence energy transfer and fluorescence correlation spectroscopy. We measure the chemical rates and the activation energies associated with the opening and the closing of the hairpin for different sizes and sequences of the loop and for various salt concentrations. The rate of unzipping of the hairpin stem is essentially independent of the characteristics of the loop, whereas the rate of closing varies greatly with the loop length and sequence. The closing rate scales with the loop length, with an exponent 2.6 +/- 0.3. The closing rate is increased at higher salt concentrations. For hairpin closing, a loop of adenosine repeats leads to smaller rates and higher activation energies than a loop with thymine repeats.

An intramolecular i-motif: the solution structure and base-pair opening kinetics of d(5mCCT3CCT3ACCT3CC)

We present a high-definition structure of d(5mCCT3CCT3ACCT3CC), a DNA sequence which resembles a four-times repeat of the C-rich strand of telomeres and centromeres. The structure is monomeric. The CC stretches form four hemi-protonated C.C base-pairs, belonging to two parallel-stranded duplexes which intercalate head-to-tail into an i-motif core. The four grooves of the core are similar to those observed previously in i-motif tetramers, with P-P distances around 0.9 nm and 1.4 nm for the narrow and wide grooves, respectively. At 0 degrees C, the structure is formed even at pH 7, despite the required protonation of cytidine pairs, suggesting that it may be biologically relevant.The intercalation topology of the i-motif core is read off the pattern of inter-residue cross-peaks along each groove: between H1' protons across the narrow grooves, and between amino and H2' protons across the wide grooves. In the hemi-protonated C.C pairs, the imino proton is shared equally between the two bases, as shown by the equal intensities of the NOESY cross-peaks between the imino proton and the two cis amino protons of the pair. Short inter-sugar distances and the direction of CH1' bonds are consistent with CH1'...O4' hydrogen bonds across the narrow grooves, as suggested by Berger et al. (1996). Proc. Natl. Acad. Sci. USA, 93, 12116-12121. At one extremity of the i-motif core, the T3A linker loops across one of the two wide grooves. It extends the core by stacking of A11, which also forms a strongly propeller-twisted reverse-Hoogsteen pair with T8. At the other extremity, the two T3 linkers loop side by side across the two narrow grooves, extending the core by stacking of a T5.T16 pair which connects the two linkers. In this T.T pair between parallel strands, the hydrogen bonds are from imino proton to O4, and the base-pair lifetime is 6 ms at 0 degrees C. The structures of segments 1 to 7 and 12 to 18, which form the i-motif core and the T3 loops, are related by a 2-fold pseudo-symmetry: the geometries and environment are so similar that the NOESY spectra are barely resolved. These various interactions illustrate how linker sequences may affect the stability, intercalation topology and folding pattern of the intramolecular i-motif.

Solution structure and base pair opening kinetics of the i-motif dimer of d(5mCCTTTACC): a noncanonical structure with possible roles in chromosome stability

BACKGROUND

Repetitive cytosine-rich DNA sequences have been identified in telomeres and centromeres of eukaryotic chromosomes. These sequences play a role in maintaining chromosome stability during replication and may be involved in chromosome pairing during meiosis. The C-rich repeats can fold into an 'i-motif' structure, in which two parallel-stranded duplexes with hemiprotonated C.C+ pairs are intercalated. Previous NMR studies of naturally occurring repeats have produced poor NMR spectra. This led us to investigate oligonucleotides, based on natural sequences, to produce higher quality spectra and thus provide further information as to the structure and possible biological function of the i-motif.

RESULTS

NMR spectroscopy has shown that d(5mCCTTTACC) forms an i-motif dimer of symmetry-related and intercalated folded strands. The high-definition structure is computed on the basis of the build-up rates of 29 intraresidue and 35 interresidue nuclear Overhauser effect (NOE) connectivities. The i-motif core includes intercalated interstrand C.C+ pairs stacked in the order 2*.8/1.7*/1*.7/2.8* (where one strand is distinguished by an asterisk and the numbers relate to the base positions within the repeat). The TTTA sequences form two loops which span the two wide grooves on opposite sides of the i-motif core; the i-motif core is extended at both ends by the stacking of A6 onto C2.C8+. The lifetimes of pairs C2.C8+ and 5mC1.C7+ are 1 ms and 1 s, respectively, at 15 degrees C. Anomalous exchange properties of the T3 imino proton indicate hydrogen bonding to A6 N7 via a water bridge. The d(5mCCTTTTCC) deoxyoligonucleotide, in which position 6 is occupied by a thymidine instead of an adenine, also forms a symmetric i-motif dimer. However, in this structure the two TTTT loops are located on the same side of the i-motif core and the C.C+ pairs are formed by equivalent cytidines stacked in the order 8*.8/1.1*/7*.7/2.2*.

CONCLUSIONS

Oligodeoxynucleotides containing two C-rich repeats can fold and dimerize into an i-motif. The change of folding topology resulting from the substitution of a single nucleoside emphasizes the influence of the loop residues on the i-motif structure formed by two folded strands.

Structural features of the DNA hairpin d(ATCCTA-GTTA-TAGGAT): formation of a G-A base pair in the loop

The three-dimensional structure of the hairpin formed by d(ATCCTA-GTTA-TAGGAT) has been determined by means of two-dimensional NMR studies, distance geometry and molecular dynamics calculations. The first and the last residues of the tetraloop of this hairpin form a sheared G-A base pair on top of the six Watson-Crick base pairs in the stem. The glycosidic torsion angles of the guanine and adenine residues in the G-A base pair reside in the anti and high- anti domain ( approximately -60 degrees ) respectively. Several dihedral angles in the loop adopt non-standard values to accommodate this base pair. The first and second residue in the loop are stacked in a more or less normal helical fashion; the fourth loop residue also stacks upon the stem, while the third residue is directed away from the loop region. The loop structure can be classified as a so-called type-I loop, in which the bases at the 5'-end of the loop stack in a continuous fashion. In this situation, loop stability is unlikely to depend heavily on the nature of the unpaired bases in the loop. Moreover, the present study indicates that the influence of the polarity of a closing A.T pair is much less significant than that of a closing C.G base pair.

Hairpin loops consisting of single adenine residues closed by sheared A.A and G.G pairs formed by the DNA triplets AAA and GAG: solution structure of the d(GTACAAAGTAC) hairpin

The DNA undecamers GTACAAAGTAC (AAA 11-mer) and GTACGAGGTAC (GAG 11-mer) have been studied in solution by high-resolution NMR spectroscopy. Both duplexes form stable hairpins containing single deoxyadenosine loops and stems containing five base-pairs that are closed at the loop end by sheared AxA and GxC pairs, respectively. These molecules thus contain new AAA and GAG loop turn motifs. All protons, including the chiral H5'/H5" protons of the loop residues, were assigned using NOESY, DQF-COSY and heteronuclear 1H-31P COSY experiments. The backbone torsion angles were constrained using experimental data from NOE crosspeaks, three-bond 1H-1H coupling constants and four-bond 1H-31P coupling constants and four-bond 1H-31P coupling constants. The AAA and GAG 11-mers form similar structures in solution. The detailed structure of the AAA 11-mer was determined by the combined use of NMR, distance geometry and energy minimization methods. This structure exhibits good stacking of the loop adenosine base on the closing 5Ax7A sheared pair, with the 6A base stacking on the 5A base and the 6A deoxyribose stacking with the 7A base. All sugars in the AAA 11-mer hairpin adopt the typical DNA C2'-endo conformation and a sharp backbone turn occurs between residues 6A and 7A. This loop turn is brought about mainly by a change in the backbone phosphate torsion angles from zeta(g-) alpha(g-) to zeta(g+) alphat(g+) at the turn. The gamma torsion angle of residue 7A in the closing sheared pair also changes from gauche+ to trans. In Pu1NPu2 loop turns of the GCA, AAA and GAG types, the chemical shift of the H4' proton of the loop deoxyribose depends on the nature of Pu2; this reflects the stacking of the loop sugar on the Pu2 base and the different ring current effects of A or G in this position.

A single G-to-C change causes human centromere TGGAA repeats to fold back into hairpins

Recently, we established that satellite III (TGGAA)n tandem repeats, which occur at the centromeres of human chromosomes, pair with themselves to form an unusual "self-complementary" antiparallel duplex containing (GGA)2 motifs in which two unpaired guanines from opposite strands intercalate between sheared G.A base pairs. In separate studies, we have also established that the GCA triplet does not form bimolecular (GCA)2 motifs but instead promotes the formation of hairpins containing a GCA-turn motif in which the loop contains a single cytidine closed by a sheared G.A pair. Since TGCAA is the most frequent variant of TGGAA found in satellite III repeats, we reasoned that the potential of this variant to form GCA-turn miniloop fold-back structures might be an important factor in modulating the local structure in natural (TGGAA)n repeats. We report here the NMR-derived solution structure of the heptadecadeoxynucleotide (G)TGGAATGCAATGGAA(C) in which a central TGCAA pentamer is flanked by two TGGAA pentamers. This 17-mer forms a rather unusual and very stable hairpin structure containing eight base pairs in the stem, only four of which are Watson-Crick pairs, and a loop consisting of a single cytidine residue. The stem contains a (GGA)2 motif with intercalative 14G/4G stacking between two sheared G.A base pairs; the loop end of the stem consists of a sheared 8G.10A closing pair with the cytosine base of the 9C loop stacked on 8G. The remarkable stability of this unusual hairpin structure (Tm = 63 degrees C) suggests that it probably plays an important role in modulating the folding of satellite III (TGGAA)n repeats at the centromere.

Parallel processing of chemical information in a local area network--III. Using genetic algorithms for conformational analysis of biomacromolecules

Multi-dimensional nuclear magnetic resonance experiments are an excellent means of revealing the three-dimensional structure of biomacromolecules in solution. However, the search space in the conformational analysis of biomacromolecules, using multi-dimensional NMR data, is huge and complex. This calls for global optimization techniques with good sampling properties. This paper describes a genetic algorithm that optimizes the fit between (simulated) experimental two-dimensional Nuclear Overhauser Effect spectra and the corresponding calculated spectra for trial structures. This is a very computational intensive procedure. Speed-up of performance is achieved by parallelizing the algorithm, i.e. creating small subpopulations of trial structures, each of which can be processed on different processors. Good sampling behavior is obtained by initializing each subpopulation with its own random seed and the introduction of a migration operator. The latter replaces the best performing individual from one subpopulation with the worst performing individual from another subpopulation after a predetermined number of generations. A parallel genetic algorithm for the conformational analysis of nucleic acids is developed using the software package HYDRA. It is demonstrated that, for the data sets used in the study, a considerable reduction in computation time is obtained for the parallel genetic algorithm as compared to a sequential implementation, while the same optimal solutions are found.

Similar conformations of hairpins with TTT and TTTT sequences: NMR and molecular modeling evidence for T.T base pairs in the TTTT hairpin

The conformations of the d[G(1)C(2)G(3)C(4)-T(a)T(b)T(c)T(d)-G(5)C(6)G(7)C(8)] (T4) and d[G(1)C(2)G(3)C(4)-T(a)T(b)T(c)-G(5)C(6)G(7)C(8)] (T3) DNA hairpins have been studied. The 1H and 31P signals of the two hairpins have been nearly completely assigned by means of two-dimensional NMR spectroscopy in D2O (NOESY (two-dimensional nuclear Overhauser effect and exchange spectroscopy) at mixing times of 5, 50, 100, 300 and 500 ms, double-quantum-filtered correlation spectroscopy (DQF-COSY) and 1H-31P reverse chemical shift correlation (RCSC), and one-dimensional NOE spectra in 90% H2O. Conformational analysis using distance geometry (DG), molecular mechanics (MM) and molecular dynamics (MD) gave model conformations, which were evaluated by comparison of experimental and simulated 2D NOESY spectra. For the T4 sequence in T4, both NMR data and modeling indicated a T(a).T(d) wobble base pair. Although two types of T(a).T(d) base pairs are possible, the one with T(a)NH-T(d)O4 and T(a)O2-T(d)NH H-bonds was calculated to be more stable. Because the T(a).T(d) base pair of T4 extends the stem, there are only two residues (T(b) and T(c) in the loop. Although there are three residues in the T3 loop, the T(c) base projects into the solvent. The resulting conformational models have very similar loop folding patterns (FP): the bases of the two adjacent residues that begin the loop [T(b)T(c) of T4 and T(a)T(b) or T3] have a minor groove/major groove orientation with the first residue each having a trans alpha torsion angle; and the phosphodiester group that links the residues at the 3' end of the loop and the 5' top of the stem [T(c)pT(d) of T4 and T(c)pG(5) of T3] has a gauche+, gauche+ zeta,alpha conformation with a trans gamma angle for the second residue in both. These or similar features appear to be present in most of the few other hairpins studied previously by conformational methods. Thus, we believe that the conformations of the loops in T3 and T4 hairpins have greater similarities than previously recognized.

Asymmetric structure of five and six membered DNA hairpin loops

The tertiary structure of nucleic acid hairpins was elucidated by means of the accessibility of the single-strand-specific nuclease from mung bean. This molecular probe has proven especially useful in determining details of the structural arrangement of the nucleotides within a loop. In this study 3'-labeling is introduced to complement previously used 5'-labeling in order to assess and to exclude possible artifacts of the method. Both labeling procedures result in mutually consistent cleavage patterns. Therefore, methodological artifacts can be excluded and the potential of the nuclease as structural probe is increased. DNA hairpins with five and six membered loops reveal an asymmetric loop structure with a sharp bend of the phosphate-ribose backbone between the second and third nucleotide on the 3'-side of a loop. These hairpin structures differ from smaller loops with 3 or 4 members, which reveal this type of bend between the first and second 3' nucleotide, and resemble with respect to the asymmetry anticodon loops of tRNA.

NMR studies of DNA duplexes singly cross-linked by different synthetic linkers

Molecular modelling studies resulted in the design of a variety of non-nucleotidic covalent linkers to bridge the 3'-end of the (+)-strand and the 5'-end of the (-)-strand in DNA duplexes. Three of these linkers were synthesized and used to prepare singly cross-linked duplexes d(GTGGAATTC)-linker-d(GAATTCCAC). Linker I is an assembly of a propylene-, a phosphate- and a second propylene-group and is thought to mimic the backbone of two nucleotides. Linkers II and III consist of five and six ethyleneglycol units, respectively. The melting temperatures of the cross-linked duplexes are 65 degrees C for I and 73 degrees C for II and III, as compared with 36 degrees C for the corresponding non-linked nonadeoxynucleotide duplex. The three cross-linked duplexes were structurally characterized by nuclear magnetic resonance spectroscopy. The 1H and 31P resonance assignments in the DNA stem were obtained using standard methods. For the resonance assignment of the linker protons, two-dimensional 1H-31P heteronuclear COSY and two-quantum-experiments were used. Distance geometry calculations with NOE-derived distance constraints were performed and the resulting structures were energy-minimized. In duplex I, the nucleotides flanking the propylene-phosphate-propylene-linker do not form a Watson-Crick base pair, whereas in duplexes II and III the entire DNA stem is in a B-type double helix conformation.

Slow conformational exchange in DNA minihairpin loops: a conformational study of the circular dumbbell d<pCGC-TT-GCG-TT>

In recent years various examples of highly stable two-residue hairpin loops (miniloops) in DNA have been encountered. As the detailed structure and stability of miniloops appear to be determined not only by the nature and sequence of the two bases in the loop, but also by the closing base pair, it is desirable to carry out in-depth studies of especially designed small model DNA compounds. Therefore, a circular DNA dumbbell-like molecule is tailored to consist of a stem of three Watson-Crick base pairs, flanked on each side by a minihairpin loop. The resulting circular DNA decamer 5'-d<pCGC-TT-GCG-TT>-3' (I) is studied in solution by means of nmr spectroscopy. At a temperature of 269 K the molecule occurs in a 50/50 mixture of two dumbbell structures (denoted L2L2 and L2L4). L2L2 contains three Watson-Crick C-G base pairs and two two-residue loops (H2-family type) in opposite parts of the molecule. On raising the temperature from 269 to 314 K, the L2L4 conformer becomes increasingly dominant (95% at 314 K). This conformer has a partially disrupted closing G-C base pair in the 5'-GTTC-3' loop with only one remaining solvent-accessible hydrogen bond between NH alpha of the cytosine C(1) and O6 of the guanine G(8), whereas the opposite 5'-CTTG-3' loop remains stable. The disruption of the C(1)-G(8) base pair in the L2L4 form is correlated with the presence of a syn orientation for the C(1) base at the 5'-3' loop-stem junction in the 5'-GTTC-3' loop. The two conformers, L2L2 and L2L4, occur in slow equilibrium (2-20 s-1). Moderate line broadening of specific 1H, 13C, and 31P resonances of residues C(1), G(8), T(9), and T(10) at low temperatures, due to chemical exchange between L2L2 and L2L4, show that the interconversion from an anti to syn conformer in residue C(1) has a small local effect on the structure of the dumbbell. T1 relaxation measurements, chemical-shift considerations, and complete band-shape calculations of the exchange process of the G(8) imino proton reveal a possibility for the existence of multiconformational states in the anti-syn equilibrium.

A comparison of the hairpin stability of the palindromic d(CGCG(A/T)4CGCG) oligonucleotides

The palindromic deoxyribonucleotides 5'-CGCGA-TATCGCG-3' and 5'-CGCGTTAACGCG-3' have been characterized by 1H NMR spectroscopy. The NMR data identified both B-DNA duplex conformations and hairpin conformations, the latter with loop regions consisting of the four central nucleotides. The resonances of the various conformations were assigned by use of two-dimensional NMR methods. The relative stability of the various conformations was investigated as a function of temperature, ionic strength and nucleotide concentration. The duplexes were found to be stabilized at high ionic strength and at low temperature, while the hairpins were stabilized at low ionic strength and at medium temperature. The thermodynamics of the duplex-hairpin and the hairpin-random coil transitions were examined, and compared to the other two oligonucleotide in the palindromic d(CGCG(A/T)4CGCG) oligonucleotide family. The relative stabilities of the duplex conformations with respect to the random coil conformations are similar for the d(CGCGAATTCGCG), d(CGCGATATCGCG) and d(CGCGTATACGCG) oligonucleotides. The duplex conformation of d(CGCGTTAACGCG) is less stable. The hairpin of d(CGCGTTAACGCG) seems also to be less stable relative to the random coil conformation than in the case of the other oligonucleotides at an equal oligonucleotide concentration. A cruciform intermediate between the duplex and hairpin conformations is suggested to explain some discrepancies observed in this work in case of the d(CGCGTTAACGCG) oligonucleotide. This is similar to what has been reported for the d(CGCGTATACGCG) oligonucleotide.

Solution structure of the Na+ form of the dimeric guanine quadruplex [d(G3T4G3)]2

The solution structure of the DNA quadruplex formed by the association of two strands of the DNA oligonucleotide, d(G3T4G3), in NaCl solution has been determined by 1H two-dimensional NMR techniques, full relaxation matrix calculations and restrained molecular dynamics. The refined structure incorporates the sequences 5'-G1sG2AG3AT4AT5AT6AT7AG8sG9AG10A-3' and 5'-G11sG12AG13AT14AT15AT16AT17AG18sG19sG20A-3' (where S and A denote syn and anti, respectively) in a three-quartet, diagonal-looped structure that we [Strahan, G. D., Shafer, R. H. & Keniry, M. A. (1994) Nucleic Acids Res. 22, 5447-5455] and others [Smith, F. W., Lau, F. W. & Feigon, J. (1994) Proc. Natl. Acad. Sci. USA 91, 10546-10550] have described. The loop structure is compact and incorporates many of the features found in duplex hairpin loops including base stacking, intraloop hydrogen bonding and extensive van der Waals' interactions. The first and third loop thymines stack over the outermost G-quartet and are also associated by hydrogen bonding. The second and the fourth loop thymines fold inwards in order to enhance van der Waals' interactions. The unexpected sequential syn-syn deoxyguanosines in the quadruplex stem appear to be a direct consequence of the way DNA oligonucleotides fold and the subsequent search for the most stable loop structure. The implications of loop sequence and length on the structure of quadruplexes are discussed.

Refinement of the solution structure of a branched DNA three-way junction

We have refined the structure of the DNA Three-Way Junction complex, TWJ-TC, described in the companion paper by quantitative analysis of two 2D NOESY spectra (mixing times 60 and 200 ms) obtained in D2O solution. NOESY crosspeak intensities extracted from these spectra were used in two kinds of refinement procedure: 1) distance-restrained energy minimization (EM) and molecular dynamics (MD) and 2) full relaxation matrix back calculation refinement. The global geometry of the refined model is very similar to that of a published, preliminary model (Leontis, 1993). Two of the helical arms of the junction are stacked. These are Helix 1, defined by basepairs S1-G1/S3-C12 through S1-C5/S3-G8 and Helix 2, which comprises basepairs S1-C6/S2-G5 through S1-G10/S2-G1. The third helical arm (Helix 3), comprised of basepairs S2-C6/S3-G5 through S2-C10/S3-G1 extends almost perpendicularly from the axis defined by Helices 1 and 2. The bases S1-C5 and S1-C6 of Strand 1 are continuously stacked across the junction region. The conformation of this strand is close to that of B-form DNA along its entire length, including the S1-C5 to S1-C6 dinucleotide step at the junction. The two unpaired bases S3-T6 and S3-C7 lie outside of the junction along the minor groove of Helix 1 and largely exposed to solvent. Analysis of the refined structure reveals that the glycosidic bond of S3-T6 exists in the syn conformation, allowing the methyl group of this residue to contact the hydrophobic surface of the minor groove of Helix 1, at S3-G11. The helical parameters of the three helical arms of the structure exhibit only weak deviations from typical values for right-handed B-form DNA. Unusual dihedral angles are only observed for the sugarphosphate backbone joining the "hinge" residues, S2-G5 and S2-C6, and S3-G5 through S3-G8. The glycosidic bond of S3-G8also lies within the syn range, allowing favorable Watson-Crick base-pairing interactions with Si -C5. The stability of this structure was checked in 39 ps molecular dynamic simulation at 330 K in water. The structure of TWJ-TC retained the geometrical features mentioned above at the end of the simulation period. The final R(1/6)-factor of the refined structure is 5%.

Two-base DNA hairpin-loop structures in vivo

In vitro studies have revealed that DNA hairpin-loops usually contain four unpaired bases. However, a small subset of sequences can form two-base loops. We have previously described an in vivo assay that is sensitive to tight loop formation and have set out to test whether DNA sequences known to form two-base loops in vitro also form tight loops in vivo. It is shown that the sequences 5'dCNNG and 5'dTNNA behave as predicted if they favour two-base loop formation in vivo, a result that is consistent with previously described in vitro studies. The ability of specific DNA sequences to form tight loops in vivo has implications for their potential to form transient structures involved in gene regulation, recombination and mutagenesis.

Determination of the backbone torsion angle epsilon in nucleic acids

The multiplet structure of cross peaks in double-quantum-filtered COSY NMR spectra is analysed for those resonances that include passive heteronuclear couplings. Interestingly, the cross peak involving the sugar-ring protons H2' and H3' in nucleic acids display an E. COSY-type appearance exclusively when the backbone torsion angle epsilon (C4'-C3'-O3'-P) adopts a gauche(-) conformation. This observation allows an unambiguous analysis of the conformation around epsilon, without the knowledge of 3JCP coupling constants.

Conformational polymorphism in telomeric structures: loop orientation and interloop pairing in d(G4TnG4)

Sequence repeats constituting the telomeric regions of chromosomes are known to adopt a variety of unusual structures, consisting of a G tetraplex stem and short stretches of thymines or thymines and adenines forming loops over the stem. Detailed model building and molecular mechanics studies have been carried out for these telomeric sequences to elucidate different types of loop orientations and possible conformations of thymines in the loop. The model building studies indicate that a minimum of two thymines have to be interspersed between guanine stretches to form folded-back structures with loops across adjacent strands in a G tetraplex (both over the small as well as large groove), while the minimum number of thymines required to build a loop across the diagonal strands in a G tetraplex is three. For two repeat sequences, these hairpins, resulting from different types of folding, can dimerize in three distinct ways--i.e., with loops across adjacent strands and on same side, with loops across adjacent strands and on opposite sides, and with loops across diagonal strands and on opposite sides--to form hairpin dimer structures. Energy minimization studies indicate that all possible hairpin dimers have very similar total energy values, though different structures are stabilized by different types of interactions. When the two loops are on the same side, in the hairpin dimer structures of d(G4TnG4), the thymines form favorably stacked tetrads in the loop region and there is interloop hydrogen bonding involving two hydrogen bonds for each thymine-thymine pair. Our molecular mechanics calculations on various folded-back as well as parallel tetraplex structures of these telomeric sequences provide a theoretical rationale for the experimentally observed feature that the presence of intervening thymine stretches stabilizes folded-back structures, while isolated stretches of guanines adopt a parallel tetraplex structure.

The solution structure of the circular trinucleotide cr(GpGpGp) determined by NMR and molecular mechanics calculation

The 3'-5' circular trinucleotide cr(GpGpGp) was studied by means of 1D and 2D high resolution NMR techniques and molecular mechanics calculations. Analysis of the J-couplings, obtained from the 1H and 13C-NMR spectra, allowed the determination of the conformation of the sugar rings and of the 'circular' phosphate backbone. In the course of the investigations it was found that the Karplus-equation most recently parametrized for the CCOP J-coupling constants could not account for the measured J(C4'P) of 11.1 Hz and a new parametrization for both HCOP and CCOP coupling constants is therefore presented. Subsequent analysis of the coupling constants yielded 'fixed' values for the torsion angles beta and delta (with beta = 178 degrees and delta = 139 degrees). The value of the latter angle corresponds to an S-type sugar conformation. The torsion angles gamma and epsilon are involved in a rapid equilibrium in which they are converted between the gauche(+) and trans and between the trans and gauche(-) domain respectively. We show that the occurrence of epsilon in the gauche(-) domain necessitates S-type sugar conformations. Given the aforementioned values for beta, gamma, delta and epsilon the ring closure constraints for the ring, formed by the phosphate backbone can only be fulfilled if alpha and zeta adopt some special values. After energy minimization with the CHARMm force field only two combinations of alpha and zeta result in energetically favourable structures, i.e. the combination alpha (t)/zeta(g-) in case gamma is in a gauche(+) and epsilon is in a trans conformation, and the combination alpha (t)/zeta (g+) for the combination gamma (t)/epsilon (g-). The results are discussed in relation to earlier findings obtained for cd(ApAp) and cr(GpGp), the latter molecule being a regulator of the synthesis of cellulose in Acetobacter xylinum.

Molecular modelling of the 3-D structure of RNA tetraloops with different nucleotide sequences

One surprisingly common element of RNA secondary structure consists of a hairpin capped by a four-base loop (or the tetraloop). Recently the 3-D structures of two RNA-tetraloops have been determined by NMR-studies. Both structures have a similar architecture: the first and the last bases of the loop form a hydrogen bonded pair which is stacked on the stem base pair. We have analysed the ability of tetraloops, with the other combinations of the first and the fourth bases, to adopt such a 'diloop' conformation using computer modelling. The analysis has shown that the 'diloop' conformation has many covalent and steric constraints which give a possibility for reliable structural predictions. As a result, a set of the tetraloop 3-D structures in which hydrogen bonded pairing of the first and the last bases does not cause covalent and steric hindrances has been selected. In most cases several predicted 3-D structures corresponded to one tetraloop sequence. Taking into consideration the folding pathway of RNA hairpins we have resolved this ambiguity and predicted the most probable 3-D structure for every possible nucleotide sequence of the tetraloop. On the basis of these results a conclusion has been drawn on the possible reasons of the tetraloop phylogenetic preference.

Conformational differences between bulged pyrimidines (C-C) and purines (A-A, I-I) at the branch point of three-stranded DNA junctions

We have synthesized DNA oligomers that can combine to form three-way junctions containing six base pairs in each stem and two unpaired bases at the branch point. Gel electrophoresis experiments indicate that the oligomers form stable complexes with equimolar stoichiometry. Using two- and three-dimensional proton nuclear magnetic resonance spectroscopy, we have completed nonexchangeable proton chemical shift assignments for three junctions which differ only in the identity of the unpaired bases (C-C, A-A, or I-I) at the branch point. Our results indicate that unpaired pyrimidines at the branch point of junctions behave differently than do unpaired purines. In a junction with two unpaired cytidines, the 5' base loops out from the molecule to lie along the minor groove of the preceding duplex stem of the junction. The 3' unpaired cytidine also demonstrates an unusual pattern of NOE connectivities with detected cross peaks to the subsequent base in the 3' direction. Junctions with unpaired purines at the branch point exhibit different behavior. Our data suggests that in these molecules the unpaired bases participate in stacking interactions among themselves and with the neighboring bases in the molecule. Despite these differences, the NOE patterns from each junction suggest the presence of a preferred, pair-wise stacking between two of the helices within the molecule. The structural differences between bulge-pyrimidine and bulge-purine junctions are discussed in light of the functional significance unpaired bases might have in the structure and dynamics of multistranded DNA junctions and, by extension, to junctions within cellular RNAs.