Conformational and Model-Building Studies of the Hairpin Form of the Mismatched DNA Octamer d(m5C-G-m5C-G-T-G-m5C-G)

Abstract

The hairpin form of the mismatched octamer d(m⁵C-G-m⁵C-G-T-G-m⁵C-G) was studied by means of NMR spectroscopy. In a companion study it is shown that the hairpin form of this DNA fragment consists of a structure with a stem of three Watson-Crick-type base pairs and a loop consisting of only two nucleotides. The non-exchangeable proton resonances were assigned by means of two-dimensional correlation spectroscopy and two-dimensional nuclear Overhauser effect spectroscopy. Proton-proton coupling constants were used for the conformational analysis of the deoxyribose ring and for some of the backbone torsion angles. From the two-dimensional NMR spectra and the coupling-constant analysis it is concluded that: (i) the stem of the hairpin exhibits B-DNA characteristics; (ii) the sugar rings are not conformationally pure, but display a certain amount of conformational flexibility; (iii) the stacking interaction in the stem of the hairpin is elongated from the 3′-side in a more or less regular fashion with the two loop nucleotides; (iv) at the 5′-side of the stem a stacking discontinuity occurs between the stem and the loop; (v) at the 5′-side of the stem the loop is closed by means of a sharp backbone turn which involves unusual γ^t and β⁺ torsion angles in residue dG(6).

The NMR results led to the construction of a hairpin-loop model which was energy-minimized by means of a molecular-mechanics program. The results clearly show that a DNA hairpin-loop structure in which the loop consists of only two nucleotides bridging the minor groove in a straightforward fashion, (i) causes no undue steric strain, and (ii) involves well-known conformational principles throughout the course of the backbone.

The hairpin form of the title compound is compared with the hairpin form of d(A-T-C-C-T- A-T₄-T-A-G-G-A-T), in which the central -T₄- part forms a loop of four nucleotides. Both models display similarities as far as stacking interactions are concerned.     

Structure of a small RNA hairpin.

The hairpin stem-loop form of the RNA oligonucleotide rCGC(UUU)GCG has been studied by NMR spectroscopy. In 10 mM phosphate buffer this RNA molecule forms a unimolecular hairpin with a stem of three base pairs and a loop of three uridines, as judged by both NMR and UV absorbance melting behavior. Distance and torsion angle restraints were determined using homonuclear proton-proton and heteronuclear proton-phosphorus 2-D NMR. These values were used in restrained molecular dynamics to determine the structure of the hairpin. The stem has characteristics of A-form geometry, although distortion from A-form occurs in the 3'-side of the stem, presumably to aid in accommodating the small loop. The loop nucleotides adopt C2'-endo conformations. NOE's strongly suggest stacking of the uracils with the stem, especially the first uracil on the 5'-side of the loop. The reversal of the chain direction in the loop seems to occur between U5 and U6. Loop structures produced by molecular dynamics simulations had a wide range of conformations and did not show stacking of the uracils. A flexible loop with significant dynamics is consistent with all the data.     

Conformational feasibility of a hairpin with two purines in the loop. 5'-d-GGTACIAGTACC-3'

Structural feasibility and conformational requirements for the sequence 5'-d-GGTACIAGTACC-3' to adopt a hairpin loop with I6 and A7 in the loop are studied. It is shown that a hairpin loop containing only two nucleotides can readily be formed without any unusual torsional angles. Stacking is continued on the 5'-side of the loop, with the I6 stacked upon C5. The base A7, on the 3'-side of the loop, can either be partially stacked with I6 or stick outside without stacking. Loop closure can be achieved for both syn and anti conformations of the glycosidic torsions for G8 while maintaining the normal Watson-Crick base pairing with the opposite C5. All torsional angles in the stem fall within the standard B-family of DNA helical structures. The phosphodiesters of the loop have trans,trans conformations. Loop formation might require the torsion about the C4'-C5' bond of G8 to be trans as opposed to the gauche+ observed in B-DNA. These results are discussed in relation to melting temperature studies [Howard et al. (1991) Biochemistry (preceding paper in this issue)] that suggest the formation of very stable hairpin structures for this sequence.     

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

The hairpin formed by d(ATCCTATTTATAGGAT) was studied by means of two-dimensional NMR spectroscopy and conformational analysis. Almost all 1H resonances of the stem region could be assigned, while the 1H and 31P spectra of the loop region were interpreted completely; this includes the stereospecific assignment of the H5' and H5" resonances. The derivation of the detailed loop structure was carried out in a stepwise fashion including some improved and new methods for structure determination from NMR data. In the first step, the mononucleotide structures were examined. The conformational space available to the mononucleotide was scanned systematically by varying the glycosidic torsion angle and pseudorotational parameters. Each generated conformer was tested against the experimental J coupling constants and NOE parameters. In the following stage, the structures of dinucleotides and longer fragments were derived. Inter-residue distances between protons were calculated by means of a procedure in which the simulated NOEs, obtained via a relaxation-matrix approach, were fitted to the experimental NOEs without the introduction of a molecular model. In addition, the backbone torsion angles beta, gamma and epsilon were deduced from homocoupling and heterocoupling constants. These data served as constraints in the next step, in which the loop sequence was subjected to a multi-conformer generation procedure. The resulting structures were tested against the mentioned constraints and disregarded if these constraints were violated. This yielded a family of structures for the loop region, confined to a relatively narrow conformational space. A representative conformation was subsequently docked on a B-type stem which fulfilled the structural constraints (derived from the NMR experiments for the stem region) to yield the hairpin structure. Results obtained from subsequent restrained-molecular-mechanics as well as free-molecular-mechanics calculations are in accordance with those obtained by means of the analysis described above. The structure of the hairpin loop is a compactly folded conformation and the first base of the central TTTA region forms a Hoogsteen T-A pair with the fourth base. This Hoogsteen base pair is stacked upon the sixth base pair of the B-type double-helical stem. The second base of the loop is folded into the minor groove, whereas the third base of the loop is partly stacked on the first and fourth bases. The phosphate backbone exhibits a sharp turn between the third and fourth nucleotides of the loop. The peculiar structure of this hairpin loop is discussed in relation to loop folding in DNA and RNA hairpins and in relation to a general model for loop folding.     

Hairpin structures in DNA containing arabinofuranosylcytosine. A combination of nuclear magnetic resonance and molecular dynamics

Nuclear magnetic resonance (NMR) and model-building studies were carried out on the hairpin form of the octamer d(CGaCTAGCG) (aC = arabinofuranosylcytosine), referred to as the TA compound. The nonexchangeable protons of the TA compound were assigned by means of nuclear Overhauser effect spectroscopy (NOESY) and correlated spectroscopy (COSY). From a detailed analysis of the coupling data and of the NOESY spectra the following conclusions are reached: (i) The hairpin consists of a stem of three Watson-Crick type base pairs, and the two remaining residues, T(4) and dA(5), participate in a loop. (ii) All sugar rings show conformational flexibility although a strong preference for the S-type (C2'-endo) conformer is observed. (iii) The thymine does not stack upon the 3' side of the stem as expected, but swings into the minor groove. (This folding principle of the loop involves an unusual alpha t conformer in residue T(4).) (iv) At the 5'-3' loop-stem junction a stacking discontinuity occurs as a consequence of a sharp turn in that part of the backbone, caused by the unusual beta + and gamma t torsion angles in residue dG(6). (v) The A base slides over the 5' side of the stem to stack upon the aC(3) residue at the 3' side of the stem in an antiparallel fashion. On the basis of J couplings and a set of approximate proton-proton distances from NOE cross peaks, a model for the hairpin was constructed. This model was then refined by using an iterative relaxation matrix approach (IRMA) in combination with restrained molecular dynamics calculations. The resulting final model satisfactorily explains all the distance constraints.     

Structural characterization of d(CAACCCGTTG) and d(CAACGGGTTG) mini-hairpin loops by heteronuclear NMR: the effects of purines versus pyrimidines in DNA hairpins.

The DNA decamers, d(CAACCCGTTG) and d(CAACGGGTTG) were studied in solution by proton and heteronuclear NMR. Under appropriate conditions of pH, temperature, salt concentration and DNA concentration, both decamers form hairpin conformations with similar stabilities [Avizonis and Kearns (1995) Biopolymers, 35, 187-200]. Both decamers adopt mini-hairpin loops, where the first and last four nucleotides are involved in Watson-Crick hydrogen bonding and the central two nucleotides, CC or GG respectively, form the loop. Through the use of proton-proton, proton-phosphorus and natural abundance proton-carbon NMR experiments, backbone torsion angles (beta, gamma and epsilon), sugar puckers and interproton distances were measured. The nucleotides forming the loops of these decamers were found to stack upon one another in an L1 type of loop conformation. Both show gamma tr and unusual beta torsion angles in the loop-closing nucleotide G7, as expected for mini-hairpin loop formation. Our results indicate that the beta and epsilon torsion angles of the fifth and sixth nucleotides that form the loop and the loop-closing nucleotide G7 are not in the standard trans conformation as found in B-DNA. Although the loop structures calculated from NMR-derived constraints are not well defined, the stacking of the bases in the two different hairpins is different. This difference in the base stacking of the loop may provide an explanation as to why the cytosine-containing hairpin is thermodynamically more stable than the guanine-containing hairpin.     

NMR studies on DNA hairpin structures with a five nucleotide loop

One- and two-dimensional NMR experiments have been undertaken to investigate the structure of DNA hairpins with a five nucleotide loop. Analysis of proton NMR spectra suggests that the four hairpin structures examined have some common structural features; B-type conformation in the stem region and the same stacking pattern, 5' (XXX-turn-XX) 3', in the loop region. The phosphorus NMR spectra suggest that the conformational changes in the loop region affect the backbone conformation of the stem duplex.     

On loop folding in nucleic acid hairpin-type structures

In a series of studies, combining NMR, optical melting and T-jump experiments, it was found that DNA hairpins display a maximum stability when the loop part of the molecule comprises four or five nucleotide residues. This is in contrast with the current notion based on RNA hairpin studies, from which it had been established that a maximum hairpin stability is obtained for six or seven residues in the loop. Here we present a structural model to rationalize these observations. This model is based on the notion that to a major extent base stacking interactions determine the stability of nucleic acid conformations. The model predicts that loop folding in RNA is characterized by an extension of the base stacking at the 5'-side of the double helix by five or six bases; the remaining gap can then easily be closed by two nucleotides. Conversely, loop folding in DNA is characterized by extending base stacking at the 3'-side of the double helical stem by two or three residues; again bridging of the remaining gap can then be achieved by one or two nucleotides. As an example of loop folding in RNA the anticodon loop of yeast tRNAPhe is discussed. For the DNA hairpin formed by d(ATCCTAT4TAGGAT) it is shown that the loop structure obtained from molecular mechanics calculations obeys the above worded loop folding principles.     

Solution structure of a DNA double helix incorporating four consecutive non-Watson-Crick base-pairs

A series of DNA 21-mers containing a variety of the 4 x 4 internal loop sequence 5'-CAAG-3'/3'-ACGT-5' were studied using nuclear magnetic resonance (NMR) methodology and distance geometry (DG)/molecular dynamics (MD) approaches. Such oligomers exhibit excellent resolution in the NMR spectra and reveal many unusual NOEs (nuclear Overhauser effect) that allow for the detailed characterization of a DNA hairpin incorporating a track of four different non-Watson-Crick base-pairs in the stem. These include a wobble C.A base-pair, a sheared A.C base-pair, a sheared A.G base-pair, and a wobble G.T base-pair. Significantly different twisting angles were observed between the base-pairs in internal loop that results with excellent intra-strand and inter-strand base stacking within the four consecutive mismatches and the surrounding canonical base-pairs. This explains why it melts at 52 degrees C even though five out of ten base-pairs in the stem adopt non-Watson-Crick pairs. However, the 4 x 4 internal loop still fits into a B-DNA double helix very well without significant change in the backbone torsion angles; only zeta torsion angles between the tandem sheared base-pairs are changed to a great extent from the gauche(-) domain to the trans domain to accommodate the cross-strand base stacking in the internal loop. The observation that several consecutive non-canonical base-pairs can stably co-exist with Watson-Crick base-pairs greatly increases the limited repertoire of irregular DNA folds and reveals the possibility for unusual structural formation in the functionally important genomic regions that have potential to become single-stranded.     

Efficient search on energy minima for structure prediction of nucleic acid motifs

Structure prediction of non-canonical motifs such as mismatches, extra unmatched nucleotides or internal and hairpin loop structures in nucleic acids is of great importance for understanding the function and design of nucleic acid structures. Systematic conformational analysis of such motifs typically involves the generation of many possible combinations of backbone dihedral torsion angles for a given motif and subsequent energy minimization (EM) and evaluation. Such approach is limited due to the number of dihedral angle combinations that grows very rapidly with the size of the motif. Two conformational search approaches have been developed that allow both an effective crossing of barriers during conformational searches and the computational demand grows much less with system size then search methods that explore all combinations of backbone dihedral torsion angles. In the first search protocol single torsion angles are flipped into favorable states using constraint EM and subsequent relaxation without constraints. The approach is repeated in an iterative manner along the backbone of the structural motif until no further energy improvement is obtained. In case of two test systems, a DNA-trinucleotide loop (sequence: GCA) and a RNA tetraloop (sequence: UUCG), the approach successfully identified low energy states close to experiment for two out of five start structures. In the second method randomly selected combinations of up to six backbone torsion angles are simultaneously flipped into preset ranges by a short constraint EM followed by unconstraint EM and acceptance according to a Metropolis acceptance criterion. This combined stochastic/EM search was even more effective than the single torsion flip approach and selected low energy states for the two test cases in between two and four cases out of five start structures.     

Solution conformation of an RNA hairpin loop.

The hairpin conformation adopted by the RNA sequence 5'GCGAUUUCUGACCGCC3' has been studied by one- and two-dimensional NMR spectroscopy. Exchangeable imino spectra in 60 mM Na+ indicate that the hairpin has a stem of six base pairs (indicated by boldface type) and a loop of three nucleotides. NOESY spectra of nonexchangeable protons confirm the formation of the stem region. The duplex has an A-conformation and contains an A.C apposition; a G.U base pair closes the loop region. The stem nucleotides have C3'-endo sugar conformations, as expected of an A-form duplex, whereas the three loop nucleotides adopt C2'-endo sugar puckers. Stacking within the loop, C8 upon the sugar of U7, stabilizes the structure. The pH dependence of both the exchangeable and nonexchangeable NMR spectra is consistent with the formation of an A+.C base pair, protonated at the N1 position of adenine. The stability of the hairpin was probed by using absorbance melting curves. The hairpin structure with the A+.C base pair is about +2 kcal/mol less stable in free energy at 37 degrees C than the hairpin formed with an A.U pair replacing the A+.C pair.     

Efficient Search on Energy Minima for Structure Prediction of Nucleic Acid Motifs

Abstract

Structure prediction of non-canonical motifs such as mismatches, extra unmatched nucleotides or internal and hairpin loop structures in nucleic acids is of great importance for understanding the function and design of nucleic acid structures. Systematic conformational analysis of such motifs typically involves the generation of many possible combinations of backbone dihedral torsion angles for a given motif and subsequent energy minimization (EM) and evaluation. Such approach is limited due to the number of dihedral angle combinations that grows very rapidly with the size of the motif. Two conformational search approaches have been developed that allow both an effective crossing of barriers during con-formational searches and the computational demand grows much less with system size then search methods that explore all combinations of backbone dihedral torsion angles. In the first search protocol single torsion angles are flipped into favorable states using constraint EM and subsequent relaxation without constraints. The approach is repeated in an iterative manner along the backbone of the structural motif until no further energy improvement is obtained. In case of two test systems, a DNA-trinucleotide loop (sequence: GCA) and a RNA tetraloop (sequence: UUCG), the approach successfully identified low energy states close to experiment for two out of five start structures. In the second method randomly selected combinations of up to six backbone torsion angles are simultaneously flipped into preset ranges by a short constraint EM followed by unconstraint EM and acceptance according to a Metropolis acceptance criterion. This combined stochastic/EM search was even more effective than the single torsion flip approach and selected low energy states for the two test cases in between two and four cases out of five start structures.     

Structure of a U.U pair within a conserved ribosomal RNA hairpin.

A conserved hairpin corresponding to nt 1057-1081 of large subunit rRNA (Escherichia coli numbering) is part of a domain targeted by antibiotics and ribosomal protein L11. The stem of the hairpin contains a U.U juxtaposition, found as either U.U or U.C in virtually all rRNA sequences. This hairpin has been synthesized and most of the aromatic and sugar protons were assigned by two-dimensional proton NMR. Distances and sugar puckers deduced from the NMR data were combined with restrained molecular dynamics calculations to deduce structural features of the hairpin. The two U residues are stacked in the helix, form one NH3-O4 hydrogen bond and require an extended backbone conformation (trans alpha and gamma) at one of the U nucleotides. The hairpin loop, UAGAAGC closed by a U-A pair, is the same size as tRNA anticodon loops, but not as well ordered.     

Comparison of the B- and Z-form hairpin loop structures formed by d(CG)5T4(CG)5

The partially self-complementary synthetic DNA oligonucleotide d(CG)5T4(CG)5 has been studied by using 1H and 31P NMR and circular dichroism. Results show that, under low-salt conditions (120 mM NaCl buffer), an intramolecular hairpin loop exists in which the double-helical stem region is B-form and the thymidine loop residues have predominantly southern (C2'-endo) sugar conformations. The thymidine glycosidic torsion angles are intermediate between syn and anti or exist as an equilibrium mixture of residues in the two extremes. NOESY data indicate that the structure of the loop region is very similar to that found for d(CG)2T4(CG)2 [Hare, D. R., & Reid, B. R. (1986) Biochemistry 25, 5341-5350]. Under high-salt conditions (6 M NaClO4 buffer), the dominant form (approximately equal to 85%) is an intramolecular hairpin structure in which the stem region forms a Z-form double helix. As in the B-form, the loop thymidine residues are intermediate between the syn and anti conformations or exist as an equilibrium mixture of the two, but the thymidine sugar conformations differ in that they are biased toward northern (C3'-endo) conformations.     

Structural similarities and differences between H1- and H2-family DNA minihairpin loops: NMR studies of octameric minihairpins

TheDNA sequences 5′-d(CGC-AC-GCG)-3′ (HPAC), 5′-d(CGC-AA-GCG)-3′ (HPAA), 5′-d(CGC-TC-GCG)-3′ (HPTC), and 5′-d(CGC-CT-GCG)-3′ (HPCT), were studied by means of nmr spectroscopy. At low DNA concentration and no added salt all four molecules adopt a minihairpin structure, containing three Watson–Crick base pairs and a two-residue loop. The structure of the HPAC hairpin is based on quantitative distance restraints, derived by a full relaxation matrix approach (iterative relaxation matrix approach), together with torsion angles obtained from coupling constant analysis. The loop folding is of the H1-family type, characterized by continuous 3′-5′ stacking of the loop bases on the duplex stem. The structure of the HPAA hairpin is similar to that of HPAC, but is more flexible and has a lower thermodynamic stability (T_m 326 K vs 320 K). According to “weakly” distance-constrained simulations in water on the HPAC minihairpin, the typical H1-family loop folding remains intact during the simulation. However, residue-based R factors of simulated nuclear Overhauser effect spectroscopy spectra, free molecular dynamics simulations in vacuo, and unusual chemical shift profiles indicate partial destacking of the loop bases at temperatures below the overall melting midpoint. The dynamic nature of the loop bases gives insight into the geometrical tolerances of stacking between bases in H1-family minihairpin loops. The HPTC and HPCT minihairpins, both containing a pyrimidine base at the first position in the loop, adopt a H2-family type folding, in which the first loop base is loosely bound in the minor groove and the second loop base is stacked upon the helix stem. The thermal stability for these two hairpins corresponds to 327–329 K, but depends on local base sequence. Preference for the type of folding depends on a single substitution from a pyrimidine (H2 family) to a purine (H1 family) at the first position of the miniloop and is explained by differences in base stacking energies, steric size, and the number of possible candidates for hydrogen bonds in the minor groove. In view of newly collected data, previous models of the H1-family and H2-family hairpins had to be revised and are now compatible with the reported HPTC and HPAC structures. The structural difference between the refined structure of HPAC and HPTC show that a conversion between H1-family and H2-family hairpins is geometrically possible by a simple pivot point rotation of 270° along two torsion angles, thereby swiveling the first loop base from a stacked position in a H1-family folding toward a position in the minor groove in a H2-family folding. The second loop residue subsequently shifts to the position of the first base in a concerted fashion. © 1998 John Wiley & Sons, Inc. Biopoly 46: 375–393, 1998     

NMR structure and dynamics of the RNA-binding site for the histone mRNA stem-loop binding protein

The 3' end of replication-dependent histone mRNAs terminate in a conserved sequence containing a stem-loop. This 26-nt sequence is the binding site for a protein, stem-loop binding protein (SLBP), that is involved in multiple aspects of histone mRNA metabolism and regulation. We have determined the structure of the 26-nt sequence by multidimensional NMR spectroscopy. There is a 16-nt stem-loop motif, with a conserved 6-bp stem and a 4-nt loop. The loop is closed by a conserved U.A base pair that terminates the canonical A-form stem. The pyrimidine-rich 4-nt loop, UUUC, is well organized with the three uridines stacking on the helix, and the fourth base extending across the major groove into the solvent. The flanking nucleotides at the base of the hairpin stem do not assume a unique conformation, despite the fact that the 5' flanking nucleotides are a critical component of the SLBP binding site.     

Conformational analysis of the dinucleotides 5'-methylphospho-N6-dimethyladenylyl-uridine (mpm62A-U) and 5'methylphospho-uridylyl-N6-dimethyladenosine (mpU-m62A) and of the trinucleotide U-m62A-U. A nuclear magnetic resonance and circular dichroic study

NMR and CD studies were carried out on the dinucleotides 5'-methylphospho-N6-dimethyladenylyl-uridine (mpm62-U) and 5'-methylphospho-uridylyl-N6-dimethyladenosine (mpU-m62A) and on the trinucleotide U-m62A-U. A detailed comparison is given of the conformational features of mpm62A-U and mpU-m62A with the corresponding 5'-nonphosphorylated dinucleotides m62A-U and U-m62A, respectively. The behaviour of the trinucleotide U-m62A-U is compared with the properties of the constituent dinucleotides U-m62A and mpm62A-U. Chemical-shift and CD data were used to determine the amount of stacking interactions. For each compound NMR spectra were recorded at two or three sample concentrations in order to separate intermolecular and intramolecular base-base interactions. The coupling constants of the ribose ring are interpreted in terms of the N/S equilibrium, and population distributions along the backbone angles beta, gamma and epsilon are presented. The combined data indicate a strong similarity between mpm62A-U and m62A-U both in degree and in mode of stacking. In contrast, the existence of different types of stacking interactions in mpU-m62A and U-m62A is suggested in order to explain the NMR and CD data. It is concluded that dinucleoside bisphosphates serve better as a model for the behaviour of trinucleotides than dinucleoside monophosphates. The trinucleotide U-m62A-U adopts a regular single-stranded stacked RNA structure with preference for N-type ribose and gamma+ and beta t backbone torsion angles. The difference in behaviour between the U-m62A- part of U-m62A-U and the dimer U-m62A is seen as a typical example of conformational transmission.     

Studies on the structure and stabilizing factor of the CUUCGG hairpin RNA using chemically synthesized oligonucleotides.

A tridecaribonucleotide, r-UGAGCUUCGGCUC, and two analogues r(UGAGC)d(UUCG)r(GCUC) and r-UGAGCUUCIGCUC, which form a hairpin structure with a four-base-paired stem and a UUCG loop, were synthesized by the solid-phase phosphoramidite method. Properties of these three oligomers and d-TGAGCTTCGGCTC, the DNA analogue, were studied by UV, CD and NMR spectroscopy. The melting temperature (Tm) data suggest that the 2'-hydroxy1 groups and the 2-amino group of guanosine in the loop (9G) stabilize the CUUCGG hairpin which is known to have an unusually high Tm. NMR studies show that this 9G takes a syn conformation and the phosphodiester backbone has a turn at 9G-10G which is a junction of the stem and loop.