Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (285K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (3)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Slama-Schwok, A.
Right arrow Articles by Vlassov, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Slama-Schwok, A.
Right arrow Articles by Vlassov, V.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research Pages 5142-5151  

Introduction
Materials And Methods
   Materials
   Melting temperatures
   Probing experiments
   Fluorescence measurements
Results
   The Bc oligonucleotide forms hairpins
   Probing experiments
   Probing Bc loops by macrocyclic bis-acridine 1 fluorescence
Discussion
   Bc adopts two hairpin structures at pH 6.0
   Probing Bc loops with the macrocyclic bis-acridine 1
   Existence of a bulge in Bc
   Bnc adopts a single hairpin structure
   Biological relevance of Bc flexibility to the 1731 retrotransposon
Conclusion
Acknowledgements
References
Structural flexibility of a DNA hairpin located in the long terminal repeat of the Drosophila 1731 retrotransposon

Structural flexibility of a DNA hairpin located in the long terminal repeat of the Drosophila 1731 retrotransposon

A. Slama-Schwok*, E. Brossalina1, Y. Demchenko1, M. Best-Belpomme2 and V. Vlassov1

INSERM U310, Institut de Biologie Physico-Chimique, 13 rue Pierre et Marie Curie, 75005 Paris, France, 1Institute of Bioorganic Chemistry, Novosibirsk, Russia and 2INSERM U99, Hopital Mondor, Créteil, France

Received July 24, 1998; Revised and Accepted September 10, 1998

ABSTRACT

The structure of the DNA binding site of the Nuclear single-stranded Binding Factor (NssBF), located in the long terminal repeat of the Drosophila 1731 retrotransposon, was investigated by melting temperature experiments, chemical probing and fluorescence measurements using a macrocyclic bis-acridine. The most probable structure of this element, named Bc, mainly involves two hairpins in equilibrium at pH 6.0 at low concentration. The hairpins differ in their apical loop size; 4 and 8 nt. The structural flexibility of Bc probably derives from the three consecutive CATA repeats complementary to the GTAT nucleotides of the palindrome. In contrast, the Bc complementary strand adopts a single hairpin. Since Bc is implicated in repression of transcription via binding of two specific factors, its structural flexibility could be associated with this process.

INTRODUCTION

Hairpins offer binding sites for specific proteins because they adopt distinct structures (1 and references therein, 2-7). RNA hairpins are generally conserved and often possess a regulatory role (8-11). They are involved in protein translation, in mRNA splicing (12-14,15 and references therein, 16), in retroviral packaging and reverse transcription (17-20).

The existence and function of DNA hairpins are less well documented. They could act at the transcription and replication level (21-23,24 and references therein). DNA hairpins and cruciforms are determinants for topoisomerase II recognition and cleavage (25-28). Hairpins may be associated with methylation processes occurring at the onset of the fragile X syndrome (29-30).

In the present work, we characterize the structure of a DNA element, located in the U3 part of the long terminal repeat (LTR) of the Drosophila 1731 retrotransposon (31). This sequence, named Bc (c stands for coding strand), belongs to the core promoter. It contains a limited palindrome that may form a hairpin (Scheme 1). Bc constitutes the binding site for two nuclear single-stranded binding factors, NssBF and p11 (31-33). Contact points between NssBF and the 1731 LTR are distributed in two distinct domains, 6 and 15 nt long, separated by the 5 nt GAACA (31). The sequence of these 5 nt can be changed without affecting NssBF binding, but the distance between the two domains has to be maintained for efficient protein binding (32). In vitro transcription of the 1731 promoter is repressed by Bc binding protein(s) (31-33). Bc could thus be implicated in transcriptional repression (32). Bc and its complementary strand constitute one of the two major sites of topoisomerase II cleavage (34). The topological state of DNA plays a role in the regulation of transcription of numerous genes (35,36).

The aim of the present work was to determine the most probable structures of the DNA binding site of the factor NssBF, in the light of our previous results (31-34). Melting temperature (Tm) experiments comparing the native sequence with mutated ones were performed. Bc structures were also investigated using probing techniques and fluorescence measurements using a macrocyclic bis-acridine as a hairpin-selective probe (37-39). The secondary structures derived from our data mainly involve two hairpins in equilibrium at pH 6.0. In contrast, the Bc complementary strand adopts a single hairpin structure. The possible role of these structures in the regulation of transcription is discussed.

MATERIALS AND METHODS

Materials

The oligonucleotides, referred to as Bc (c for coding strand), BcI, BcII, H1, H2, Bc2, Bc3, Bc5, Bc6, Bc7 and Bnc (nc for non-coding strand), possess the following sequences, the dashed lines representing deletions (Scheme 1 and Fig. 1):

Bc,5[prime]-TAT GTA ATT TTG TAT GAG AAC ATA CAT ACA TAC ACA TGC A-3[prime];

BcI,5[prime]-TAT GAT ATT TTC TTA GAG AAC ATA CTA AGA ATC ACA TGC A-3[prime];

BcII,5[prime]-TAT GAT ATT TTC TTA GAG AAC TAA GAA TCA TAC ACA TGC A-3[prime];

H1,5[prime]--------------TC TTA GAG AAC ATA CTA AGA----------------3[prime];

H2,5[prime]--------------TC TTA GAG AAC TAA GA----------------------3[prime];

Bc2,5[prime]--------------TG TAT GAG AAC ATA CA----------------------3[prime];

Bc3,5[prime]--------------TG TAT GAG AAC ATA CAT ACA TAC ACA TGC A-3[prime];

Bc5,5[prime]-TAT GTA ATT TTG TAT GAG AAC AT- CAT ACA TAC ACA TGC A-3[prime];

Bc6,5[prime]-TAT GTA-----TG TAT GAG AAC ATA CAT ACA TAC ACA TGC A-3[prime];

Bc7,5[prime]-TAT GTA-----TG TAT GAG AAC ATA CAT ACA TA--------------3[prime];

Bnc,5[prime]-TAT GCA TGT GTA TGT ATG TAT GTT CTC ATA CAA AAT TAC A-3[prime].

They were purchased from Eurogentec or Genosys and purified by HPLC. The samples were ethanol precipitated, then desalted on a Sephadex G-10 or 25 column. The chemicals used were of the highest commercial purity. All aqueous solutions utilized MilliQ water (Millipore). The macrocyclic bis-acridine was a gift of Prof. J.-M. Lehn (37).


Scheme 1. (Upper) Bc sequence with the palindrome positions indicated by arrows and the nucleotides corresponding to the NssBF binding site shown in red. (Middle) Possible structures of Bc; the numbering arbitrarily begins from the first guanine at the 5[prime] extremity.

Melting temperatures

The Tm measurements were performed with a Kontron 942 spectrophotometer. The temperature within the cell was measured using a temperature sensor. The melting profiles of the oligonucleotides were monitored at 270 and 260 nm and corrected for baseline fluctuations by subtracting the absorbance at 350 nm. Cells of variable path length were used in order not to exceed 1.7 OD. The samples were heated to 70°C for 30 min then gently cooled to 10°C and maintained at this temperature overnight for equilibration before measurement. Deconvolution of the melting derivatives was performed using Origin 3.5 software.

Probing experiments

Samples of 1 µM Bc, labelled with 32P at its 5[prime]-end, were incubated for 40 min at 20°C in the presence of 10% DEPC. Incubation was performed in 30 mM cacodylate buffer, pH 6.0, in the presence of 0.5 mM EDTA (unstructured Bc) or in 100 mM KCl or 10 mM MgCl2 as indicated (structured forms). After the reaction, DEPC was extracted with 0.5 ml diethyl ether and Bc was precipitated in ethanol in the presence of 1 µg/ml DNA carrier and 0.3 M sodium acetate, pH 4.8. The modified Bc was then treated with piperidine (1 M piperidine for 10 min at 100°C). The sample was then precipitated in ethanol and analysed on a 20% denaturating polyacrylamide gel. The reaction of Bc with permanganate was performed in the same buffer for 10 min at 20°C or 15 min at 5°C, as indicated, using [Bc] = 1 µM, [KMnO4] = 15 mg/l. After incubation, the same treatment as above was performed. A similar procedure was used for Bnc but the reaction with permanganate was for 5 min at 20°C. The cleavage reaction with S1 nuclease (1 U/µl) was done under the same conditions.

Fluorescence measurements

The fluorescence measurements were performed with a Fluoromax instrument (Spex) equipped with a Hamamatsu 931 photomultiplier and a thermostated cell holder. The data have been corrected for the photomultiplier response and for variation in absorption at the excitation wavelength. Repeated fluorescence spectra of bound 1 to oligomers induce spectral modifications under continuous 360 nm irradiation due to a photochemical reaction, which was avoided by taking short integration times and small excitation path lengths l = 0.4 cm.

RESULTS

The Bc oligonucleotide forms hairpins

Melting profile of Bc at pH 6.0, 100 mM KCl. The melting profiles of Bc are reversible at very slow heating rates (4-6°C/h). The curve derivatives display a main broad peak at 38.5 ± 0.5°C that remains unchanged at a 150-fold increase in strand concentration; [Bc] = 0.3-53.5 µM (Fig. 1 and Table 1). A comparison of curves obtained at 1 and 53 µM reveals an additional lower transition that depends on Bc concentration. Since the data suggested that Bc might adopt more than one structure, the derivatives have been deconvoluted, assuming a gaussian shape to each peak. Typical calculations are presented in Figure 1, curve 1. They yield a low temperature concentration-dependent peak and two concentration-independent maxima at 37.2 ± 1.2 and 41.9 ± 1.0 (Table 1A). The low temperature transition is attributed to an intermolecular conversion from a double-stranded structure, called a bulged duplex, to two (identical) single-stranded hairpins carrying a loop of sequence L, depicted in Scheme 2.


Scheme 2.

The main derivative peak at 38.5°C, fitted by two gaussian curves, does not depend on Bc concentration and thus melting corresponds to intramolecular folding/unfolding processes. The two intramolecular structures of Bc may correspond to two hairpins. Examination of the Bc sequence reveals that it contains a limited palindrome and three consecutive CATA repeats complementary to the GTAT nucleotides, which allow several hairpin structures. These are divided in two families: a base pairing limited to the palindrome (Scheme 1A) defines the first one. The second one is characterized by more base pairs than those of the palindrome (Scheme 1B). Melting temperature experiments with mutated oligonucleotides were performed to determine the most probable structures of Bc.


Figure 1. (Left) Melting derivatives of Bc and related oligonucleotides. The absorbances were monitored at 260 nm, in 30 mM cacodylate buffer, pH 6.0, 100 mM KCl. Curve 1, [Bc] = 0.8 µM; curve 2, [Bc3] = 1.1 µM; curves 3 and 4, [H1] = [H2] = 4.2 µM, respectively. The fit of curves 1 and 2 by three gaussian peaks is shown as a full line and the individual gaussian peaks as dashed lines. (Right) Deletions relative to native Bc yielding the related oligonucleotides BcI, Bc3, Bc5 and BcII, Bc6, Bc7, presented on Bc structures I and II respectively (top), and H1 and H2, Bc2, drawn on Bc hairpins 1 and 2 respectively (bottom). The double arrows represent base pair permutations. This schematic representation does not intend to show the structures of the mutated oligomers.


Table 1. A. Melting temperatures of Bc and Bnc as a function of concentration
Oligo [Oligo] (µM) Tm (±0.5°C) [Delta]HVHa (kcal/mol) [Delta]SVHa (cal/mol K) Deconvolution into three gaussian peaks ([chi]2)
Bcb,c 0.31 38.3      
0.85 38.4 23.6; 37.7; 41.6 (1.1 × 10-9)
2.3 38.3 25.3: 36.4: 41.5 (5.7 × 10-8)
5.2 38.0 25.3; 35.6; 42.5 (1.8 × 10-9)
9.0 38.8 28.5; 37.3; 41.5 (2.8 × 10-10)
18.1 38.8 29.8; 39.0; 42.2 (1.8 × 10-9)
53.5 39.0  
Bcb,d 1.3 47.6 35.6, 48.0 (2.4 × 10-9)
Bncc,e 1.1 41.0 (23.5 ± 0.7) (74.8 ± 2.0)  
4.4 41.4 (23.8 ± 0.7) (75.4 ± 2.0)
19.5 41.8 (23.2 ± 0.7) (73.6 ± 2.0)
aCalculated according to a two-state model.
bThe data were obtained using a heating rate of 6°C/h.
c30 mM cacodylate buffer, pH 6.0, 100 mM KCl.
d30 mM cacodylate buffer, pH 6.0, [MgCl2] = 10 mM.
eThe data were obtained using a heating rate of 12°C/h.

Tm of mutated sequences relative to Bc. Nine mutated sequences relative to wild-type Bc were designed, named BcI, BcII, H1, H2, Bc2, Bc3, Bc5, Bc6 and Bc7 (Fig. 1). BcI is an analogue of Bc, in which 3 bp from the second and third CATA repeats are permuted in order to selectively stabilize the eight-membered loop hairpin (structure I or 1). Similarly, BcII differs from the native Bc by permutations in the first and second repeats and only allows four-membered loop hairpin (structure II or 2) formation. H1 and H2 derive from BcI and BcII by deletion of the extremities, thus modelling hairpins 1 and 2 respectively. Bc2 differs from hairpin 2 by deletion of the dangling ends (fragment 8-23 of Bc). Bc3, deleted in the 5[prime]-end compared with Bc, corresponds to the binding site of NssBF (31). A21, which is one of the important residues for protein binding, is deleted in Bc5 compared with Bc (41). The deletion of A4TTT7 in Bc6 relative to Bc was designed since these residues are inserted within the second TATG motif, complementary to the CATA repeats. Bc7 derives from Bc6 by deletion of the dangling 3[prime]-end. The melting derivatives of the mutated sequences are compared with that of Bc in Figure 1 and Table 1B.

Table 1B. Melting temperatures of the mutated sequences BcI, BcII, Bc2, H1, H2, Bc3, Bc5, Bc6 and Bc7
Sequence [Oligo] (µM) Tm (±0.5°C)
BcIc,e 0.8 29.3
2.7 30.0
5.1 29.8
16.4 29.4
BcId,e 1.3 47.4
BcIIc,e 0.5 33.1
0.8 33.2
5.0 33.5
14.0 33.0
BcIId,e 1.3 47.5
Bc2c,e 1.0 49.4
4.2 48.9
10 49.2
40 49.1
H1c,e 1.0 34.8
4.2 34.0
9.4 34.1
34.0 33.8
H2c,e 1.0 41.0
4.2 40.0
7.3 40.1
17.0 40.1
Bc3b,c 1.0 38.5
2.4 38.6
9.6 38.9
23.0 38.7
Bc3b,d 1.3 50.8
Bc5c,e 1.1 36, 43
7.0 36, 44
Bc6c,e 0.3 59.8
1.3 59.6
5.7 60.3
17.4 59.4
Bc6d,e 1.3 66.0
Bc7c,e 0.5 62.3
1.3 62.2
5.0 62.7
aCalculated according to a two-state model.
bThe data were obtained using a heating rate of 6°C/h.
c30 mM cacodylate buffer, pH 6.0, 100 mM KCl.
d30 mM cacodylate buffer, pH 6.0, [MgCl2] = 10 mM.
eThe data were obtained using a heating rate of 12°C/h.

The Bc3 melting curve derivative, presented in Figure 1, curve 2, is as broad as that of Bc, with a main peak at 38.6 ± 0.5°C and a shoulder around 42-44°C. The Bc3 derivative can be deconvoluted into two concentration-independent peaks at 37.2 ± 1.1 and 42.4 ± 0.9°C, as for Bc, together with a concentration-dependent gaussian curve at lower temperature. Bc3 differs from Bc by a 5[prime]-end deletion that does not allow base pairing in the lower stem. The similarity in Tm values of Bc and Bc3 suggests that native Bc mainly adopts limited base paired hairpins of family A structures at 100 mM KCl.

BcI, BcII, H1 and H2 present a single concentration-independent transition. Their melting temperatures are listed in Table 1B. The single concentration-independent peaks of these hairpins contrast with the two peaks of Bc. This result is anticipated from the design of BcI, BcII, H1 and H2, each of which is expected to mimic a single hairpin (Fig. 1).

We now hypothesize that, at low concentration, Bc mainly adopts both hairpins 1 and 2, as suggested by the deconvolution and Bc3 results (Fig. 1). This assumption implies that the sum of BcI and BcII melting derivatives should mimic that of Bc. However, the base permutations within the mutated sequences could alter their stability compared with hairpins 1 and 2 (below). Figure 22A shows that the sum of the BcI and BcII derivatives can model that of Bc, assuming a 5°C relative translation (the use of the sum of BcI and BcII derives from the averaged weight of each gaussian curve obtained by deconvolution). Data at higher Bc concentrations can be similarly fitted by the sum of BcI and BcII provided a 5°C relative translation is applied (data not shown).


Figure 2. (A) Fit of Bc melting derivative (crosses and lower abscissa) by the derivative of the sum of BcI and BcII melting curves (dashed line and upper abscissa); [Bc] = [BcI] = [BcII] = 0.8 µM, 30 mM cacodylate buffer, pH 6.0, 100 mM KCl. (B) Comparison of Bc with BcI and BcII at 10 mM MgCl2, 30 mM cacodylate buffer, pH 6.0; [Bc] = [BcI] = [BcII] = 1.3 µM. From bottom to top: +, BcI; ×, BcII; diamonds, Bc. The fits of BcI (34.7 and 47.9, [chi]2 = 8.5 × 10-10), BcII (36.1, 47.8,[chi]2 = 9.3 × 10-9) and Bc (35.6, 48.0, [chi]2 = 2.4 × 10-9) derivatives by two gaussian curves are represented as full lines. The derivative of the sum of the BcI and BcII (experimental) melting profiles is shown as a dashed and dotted line.
   A, B

   C, D

Figure 3. (A) Probing of Bc with DEPC. Lanes 1 and 4, Bc treated with piperidine; lanes 2 and 5, Bc modified with DEPC, 0.5 mM EDTA; lanes 3 and 6, Bc modified with DEPC at 100 mM KCl. Fragments of Bc sequence are shown on the left and right sides of the autoradiograph. (B) Analysis of Bc modified with permanganate. Lane 1, Bc treated with permanganate in 0.5 mM EDTA; lane 2, Bc modified with permanganate at 100 mM KCl. Equilibrium structures of Bc at 100 mM KCl (C) or at 10 mM MgCl2 (D), 30 mM cacodylate buffer, pH 6, 20°C. Red arrows indicate cleavage by KMNO4, the size of the arrow varies with the extent of cleavage. Blue arrows represent DEPC cleavage sites and green arrows S1 nuclease sites. Dots show the bases protected from cleavage with DEPC (blue) or permanganate (red). The size of the dots is proportional to the extent of protection.

Bc2 corresponds to Bc hairpin 2 without the permutations included in H2, thus enabling us to check the validity of the above hypothesis. Bc2 melts in a single transition observed at 49.2 ± 0.5°C, which is 9°C higher than the Tm of H2 (Table 1B). This result emphasizes the destabilizing effect of such base permutations and supports the assumption that the sum of the BcI and BcII derivatives should be lower than that of Bc.

Bc2 melts at 7°C higher than Bc hairpin 2. The same 7°C gap exists between H2 and BcII and yet these hairpins differ by an identical deletion of the dangling ends. Thus the melting temperature difference between Bc2 and Bc hairpin 2 is likely to arise from the dangling ends deletion.

H2 melting (compared with hairpin 2) probably results from a balance between a decreased stability due to base permutations and an increased stability resulting from the dangling ends deletion. Thus the H2 Tm at 40.3 ± 0.5°C fits Bc hairpin 2, peaking at 41.9 ± 1.0°C (Fig. 1, curve 4). The same reasoning holds for hairpin 1: deletion of the dangling ends in H1 increases the Tm by 4°C compared with BcI and the base pairs permutations in H1 relative to hairpin 1 are probably destabilizing. Thus, the H1 Tm at 34.5 ± 0.5°C is quite close to the calculated Tm of hairpin 1 at 37.2 ± 1.2°C (Fig. 1, curve 3). Additional evidence for the validity of the analysis is given by the correct fit of the Bc derivative by the sum of H1 and H2 (data not shown).

The derivative of the Bc5 melting curve exhibits two broad peaks, both independent of Bc5 concentration, reflecting the existence of two hairpin conformers. These data emphasize the structural flexibility of Bc5 and, indirectly, that of native Bc.

Bc6 and Bc7 melting profiles exhibit more cooperativity compared with that of Bc. The concentration-independent transition of Bc6 is 23°C higher than that of Bc. Only hairpins of family B could be changed by this A4TTT7 deletion: hairpins I or II would yield a larger contiguous base paired stem in Bc6 compared with that of Bc, more stable than if interrupted by a bulge. The higher melting temperature of Bc6 relative to Bc hints at a highly destabilizing effect of a bulge. Although the A4TTT7 deletion could induce a structural rearrangement in Bc6, these data suggest an ability of the Bc sequence to form bulged hairpins. The Tm values of Bc6 and Bc7 differ by only 2°C; this difference probably arises from the contribution of the dangling extremities of Bc6 relative to Bc7.

The melting results show that Bc is likely to adopt two hairpins in slow equilibrium at low concentration and 100 mM KCl, pH 6.0. The probing experiments presented below were performed to confirm this hypothesis.


Figure 4. (A) Bc cleavage by DEPC in 10 mM Mg2+ or in 100 mM KCl. Quantification of the data of Figure 3 and DEPC probing performed in the presence of Mg2+ at 20°C (data not shown). The ordinates are a white to black scale; 0 and 256 pixels correspond to 100% white and 100% black respectively. The abscissa shows the position of a given nucleotide from the first G at the 5[prime]-end. Filled and open circles, data obtained at 100 mM KCl and 10 mM MgCl2 respectively. One representative blank at pH 6.0, 0.5 mM EDTA is shown as filled squares. (B) Comparison of Bc modification with KMnO4 in the presence of Mg2+ and 100 mM KCl. Quantification of the probing data of Figure 3 (100 mM KCl) and modifications performed in 10 mM MgCl2 for 15 min, filled and open diamonds respectively.

Effect of Mg2+ on Bc, BcI, BcII and Bc3 melting. The structures adopted by Bc may depend on the ionic strength; i.e. Mg2+ could stabilize the extended structures of family B. Melting data for Bc, BcI and BcII are shown in Figure 22B. For BcI and BcII the main transition is observed at 47.5 ± 0.7 and 47.4 ± 0.7°C, respectively, whereas the Bc derivative peaks at 47.6 ± 0.7°C. These derivatives all present a more pronounced low lying transition than at 100 mM KCl. The sum of BcI and BcII reproduces the Tm of the Bc main and low transitions without any relative translation (Discussion).


The Tm of native Bc is 47.6°C, whereas the Tm of Bc3 is 50.8°C (Table 1). This difference between the two sequences at 10 mM Mg2+, which is not observed at 100 mM KCl, points to a change in Bc structure. The lower Tm value of Bc could be compatible with the lower stem and bulge formation, which is avoided in Bc3 due to its 5[prime]-end deletion relative to Bc. The energetic cost of the bulge is expected to be large, which may be consistent with the 18°C difference between the Tm of Bc and Bc6.

The Tm data strongly suggest that Bc could adopt both hairpins I and II at 10 mM Mg2+; this assumption was cross-checked using probing experiments.

Probing experiments

Probing of Bc at 100 mM KCl, pH 6.0. Probing of Bc was performed using the chemicals diethylpyrocarbonate (DEPC) and potassium permanganate as well as the enzyme nuclease S1. These experiments were done under the same ionic conditions as used in the Tm studies. Typical gels are presented in Figure 33 and quantified in Figure 44, comparing folded Bc at 100 mM KCl (lanes 3 and 6) with unstructured Bc in the presence of 0.5 mM EDTA (lanes 2 and 5). Independent Tm experiments confirmed that Bc is unstructured at 20°C under the latter ionic conditions(Tm = 6°C; data not shown). Nucleotides G9, A11, G13 and A23 are strongly protected against DEPC cleavage (Fig. 44A). Residues A14, G15, A16 and A17 display enhanced reactivity to DEPC (compare lanes 2 and 3). In contrast, adenine A21 is partially protected against DEPC, whereas A25 and A27 are somewhat more reactive than A23. Residues A3 and A4 have similar reactivity in folded and unfolded Bc. In the folded molecule, T1, T2 and T5-T7 display enhanced permanganate cleavage. T8 is partially protected, probably due to its location in the stem of the structure. The single-stranded nature of the 5[prime]-part of Bc is confirmed by efficient cleavage by nuclease S1 at G1 (Fig. 33C and data not shown).

From these data, two structures for Bc can be proposed, depicted in Scheme 1A: a hairpin with a 6 bp stem and an 8 nt loop (hairpin 1) and a structure with a similar stem but a 4 nt loop (hairpin 2). The results show that these structures are in equilibrium, indicated by the partial protection of A21, A25 and A27 to a greater extent and strong protection of the region G9-G13. The structural flexibility of Bc probably occurs via slippage of one extremity by one CATA repeat, relative to the other end, which allows, on average, the formation of 6 bp restricted to the upper stem. This slippage is probably a dynamic process. The data cannot exclude transient contributions of alternative structures involving a more extended base pairing, like hairpins I and II.

The data obtained at pH 6.0, 100 mM KCl, strongly suggest that Bc involves a dynamic equilibrium between two hairpins of family A. The question then arises, could Mg2+ stabilize extended structures of family B?

   A

   B

Figure 5. (A) Modification of Bnc with KMnO4. Lane 1, Bnc treated with piperidine; lane 2, Bnc reacted with KMnO4 in 0.5 mM EDTA; lane 3, as lane 2 but in 30 mM cacodylate buffer, pH 6.0, 10 mM MgCl2. Fragments of the sequence are also shown. (B) Proposed structure of Bnc from (A).

Probing of Bc in the presence of 10 mM MgCl2 at pH 6.0. A comparison of Bc probing by DEPC and KMnO4 at 100 mM KCl and 10 mM Mg2+ is shown in Figure 4. These results differ from those obtained at 100 mM KCl in the 5[prime]-end of the sequence. Cleavage of nucleotide T2 is very weak compared with that at T5 and T6 (Fig. 4B), in contrast to that obtained at 100 mM KCl, where all the thymine residues T1-T7 are equally cleaved. Similarly, A3 and A4 are both efficiently cleaved at 100 mM KCl, whereas A3 is somewhat protected compared with A4 at 10 mM MgCl2 (Figs 3C and D and 4A).


The data in Figure 44 support the existence of a lower stem and a bulge at 10 mM Mg2+, pH 6.0. The partial protection of A19, T20 and A21 at 20°C is concomitant with a strong protection of A23, A25 and A27. The latter residues are indeed located in the upper stem of hairpin I and in the lower stem of hairpin II. These data strongly suggest that, in the presence of magnesium ions at pH 6.0, Bc adopts two predominant structures in equilibrium, hairpins I and II, depicted in Scheme 1B.

Probing Bnc in the presence of 10 mM Mg2+. Typical probing experiments of the Bc complementary strand, Bnc, using permanganate, are shown in Figure 5A. These experiments test whether Bnc could adopt a single hairpin or two hairpins in equilibrium, similarly to Bc. The enhanced cleavage at residues T23, T24 and T26, compared with the control lane, is most obvious. The 5[prime]-end residues T7, T9 and T11 follow a similar trend. In contrast, T13, T15, T17, T19, T21 and T29 are relatively protected. From these data, a single structure of Bnc can be proposed, depicted in Figure 5B. This hairpin carries a four-membered loop, shown by the strong modification at T23, T24 and T26. The Bnc stem is defined by marked protection of the base paired nucleotides T15, T19, T21 and T29. The less pronounced protection of T13 and T17 relative to T15, T19, T21 and T29 is consistent with their location, being the first base pair of the lower and upper stems, respectively.

Probing Bc loops by macrocyclic bis-acridine 1 fluorescence

The macrocycle bis-acridine 1 selectively binds to model hairpins compared with other DNA structures such as a double helix or `unstructured' single-stranded oligomers (38). Furthermore, this compound presents interesting fluorescence properties that discriminate between single-stranded and double-stranded oligomers (38,39). The macrocycle is thus expected to bind preferentially to loop sites as compared with other single-stranded subunits and should allow probing of the structural flexibility of Bc by comparison with mutated hairpins.

The fluorescence spectra of the bis-acridine 1 bound to Bc, BcI and BcII are shown in Figure 6 (a 1:1 Bc-macrocycle complex has been shown by independent measurements of the stoichiometry; data not shown). The broad and unstructured maximum of free 1 is modified in the presence of Bc: the spectra present two vibronic bands, the higher one peaking at 442 nm, in agreement with a single-stranded, adenine-rich environment (38). The macrocycle bound to either Bc3 or Bc5 presents the same fluorescence yield as in the presence of Bc ([Phi]/[Phi]0 = 1.40 ± 0.04). In contrast, the spectrum of 1 bound to BcII displays a higher intensity than that to Bc ([Phi]/[Phi]0 = 1.75 ± 0.06), but comparable within experimental error to that of Bc6 and Bc7 ([Phi]/[Phi]0 = 1.80 ± 0.06 and 1.85 ± 0.06, respectively). The relative yield of 1 complexed to BcI or H1 is lower than that of Bc ([Phi]/[Phi]0 = 0.94 ± 0.06). These data are obtained only at low concentrations of BcII, Bc6 and Bc7 (0.4-1.0 µM); at higher concentrations (2.5-9.0 µM), the spectra present typical shoulders at 472 nm, in agreement with the bulged duplex structures adopted by these sequences (38,39; data not shown). The similar yield of the first 1 equivalent, whether bound to either BcI or H1, shows that the macrocycle first binds to the loop, the only common single-stranded site on these hairpins.

Why should the fluorescence yields in the presence of Bc be different from those with BcII, Bc6, Bc7, BcI and H1? BcII is designed as a mimic of hairpin II, whereas BcI models hairpin I. The apparent discrepancy between 1 relative fluorescence yield on Bc and on the mutated sequences is explained if the macrocycle binds and probes both 4 and 8 nt hairpins loops. Indeed, [Phi]/[Phi]0 (Bc) = 1.40 is intermediate between [Phi]/[Phi]0 (BcI) = 0.94 and [Phi]/[Phi]0 (BcII) = 1.75 and the macrocycle spectrum bound to Bc can be modeled by the sum of 1 complexed to BcI and BcII (Fig. 6).


Figure 6. Comparison of the fluorescence spectra of one macrocycle equivalent bound to Bc or mutated sequences: 1-Bc (full line), 1-BcII (broken line) and 1-BcI (dotted line); the sum of the spectra of the macrocycle bound to BcI and BcII is shown as crosses. Excitation wavelength [lambda] = 362 nm, slits 1.15 nm, [1] = 1 µM, 30 mM cacodylate buffer, pH 6.0, 100 mM KCl. (Inset) Comparison of free 1 fluorescence (dashed and dotted line) with that of one macrocycle per Bc strand (solid line).


DISCUSSION

Bc adopts two hairpin structures at pH 6.0

The Tm of Bc is 38.5 ± 0.5°C. The corresponding peak derivative is concentration-independent in the range 0.3-53.5 µM, which indicates intramolecular folding (Table 1). Bc migrates as a single band, faster than the duplex formed by Bc and its complementary strand; this holds in the range [Bc] = 10-10-10-6 M (data not shown). Thus Bc folding is intramolecular. There is evidence of a distinct intermolecular transition at high concentrations or in the presence of magnesium ions (Fig. 22B and Table 1), indicating that Bc does form a bulged duplex under certain experimental conditions. Bc3 bulged duplex always melts at a lower temperature than that of Bc, consistent with the difference in sequence lengths (Fig. 1 and data not shown). Bc6 and Bc7, which possess the longest contiguous stem, form a bulged duplex at low strand concentration. The thermodynamic parameters of the Bc7 and Bc duplex to hairpin interconversions were calculated from the Tm concentration dependence: [Delta]HVH(Bc7) = 68 ± 8 kcal/mol and [Delta]SVH(Bc7) = 196 ± 25 eu; [Delta]HVH(Bc) = 86 ± 6 kcal/mol and [Delta]SVH(Bc) = 262 ± 15 eu (the latter are calculated from the deconvoluted low transition as a function of concentration; Table 1).

The Tm curve derivative of Bc shows a main concentration-independent peak at 38.5°C, together with a shoulder around 42-44°C (Fig. 1, curve 1). This hints that more than one hairpin exists. Bc3, which lacks the 5[prime] extremity compared with Bc, gives similar results (Fig. 1, curve 2). Both profiles are best fitted by three gaussian curves, two of which are obtained at the same temperature, 37.2 and 42°C, for both sequences. Thus we hypothesize that, at 100 mM KCl and low concentration, Bc mainly adopts two hairpins of family A among the possible structures depicted in Scheme 1. BcI and BcII Tm should model Bc melting if Bc adopts both hairpins 1 and 2. However, permutations of their stem compared with the native structures 1 and 2 could modify their Tm. [Calculation of the stem free energy difference due to permutations is only[Delta][Delta]G = 0.1 kcal/mol (40), but the loop free energy may also differ.] Indeed, comparison of the Bc2 and H2 Tm points to a 9°C decrease owing to base permutations. This is consistent with the need for a relative shift in the sum of the BcI and BcII curves to model the Bc derivative (Fig. 22A). Thus the Bc sequence is likely to adopt mainly hairpins 1 and 2 in equal proportions according to the averaged weight of the two (concentration-independent) gaussian curves. These structures slowly interconvert, probably on the timescale of seconds to minutes, in order to account for the very slow heating rates needed for fully reversible profiles.

Further supporting evidence for the two Bc conformers is given by the partial protection from DEPC cleavage of A21, consistent with an equilibrium between 4 and 8 nt loop hairpins. Moreover, the strong protection of A23 is expected, since this base is located in the stem of the two structures of Scheme 1A. The equilibrium probably involves slippage of four bases, corresponding to one CATA repeat. Both probing and melting results at 100 mM KCl support the structural flexibility of Bc that is suggested from inspection of its sequence.

Probing Bc loops with the macrocyclic bis-acridine 1

The fluorescence maximum of the bound macrocycle to Bc is characteristic of a single-stranded, adenine-rich site (38). Probing of Bc by the macrocycle can be deduced from the fluorescence yields obtained with Bc compared with mutated sequences. The same 1 fluorescence yield, measured in the presence of BcI and H1, implies that the first macrocycle binds to the only common single-stranded site of these sequences: the loop. The fluorescence spectrum of 1 on Bc can be modelled by the sum of the bound macrocycle to BcI and BcII. This shows that the macrocycle binds to both 4 and 8 nt loops of the native hairpins, in agreement with the equilibrium between two Bc hairpins determined by the melting and probing experiments. The lower fluorescence yield of bound 1 to BcI relative to that observed with BcII probably reflects less base stacking in the 4 nt loop than in the 8 nt one. A correlation between the macrocycle fluorescence yield and the extent of base stacking has been previously noted to account for the higher yield of 1 bound to an A3 hairpin loop relative to that on the trimer A3 (38). Therefore, the variation in the macrocycle fluorescence yields allows probing of the structural difference between the hairpin I and II loops, suggesting that macrocycle 1 could be a structural probe for complex sequences bearing loops subunits.

Existence of a bulge in Bc

The melting curves of Bc6 and Bc7 compared with that of Bc first suggest the possible formation of bulged hairpins. These structures (Scheme 1B) indeed exist, stabilized by Mg2+, shown by a partial protection of A3 and a strong one of T2. These two residues are located in the lower stem. Extensive cleavage of the bulge residues T5 and T6 is also observed. The equilibrium between hairpins I and II is suggested by partial protection of residues A19, T20 and A21 combined with strong protection of A23, A25 and A27. The melting temperature results, showing a great similarity between Bc and BcI + BcII derivative profiles (Fig. 22B) support this attribution. The lower Tm of Bc compared with that of Bc3 is indeed consistent with a high energetic cost of the bulge.

Bc hairpins I and II possess a four-membered bulge. The occurrence of such a bulge is frequent in RNA hairpins. Bulges induce bending of DNA and RNA duplexes (41-49) in a direction away from the bulge-containing strand, providing an opening in the opposite groove. Bulge-induced kinks have been suggested to facilitate recognition of the opposite strand sequence by cognate proteins. The larger recognition site of NssBF, (TACA)3, lies in the 3[prime]-part of Bc, precisely opposite the bulge (31,32). Moreover, Bc3 does not bind NssBF, although it corresponds to the binding site of this protein on Bc (31,32). Deletion of the 5[prime]-end in Bc3 relative to Bc does not allow formation of a bulge and a lower stem. This modification of Bc3 overall structure compared with that of Bc might be responsible for the lack of NssBF binding on Bc3.

Bnc adopts a single hairpin structure

The Bc complementary strand, Bnc, also forms a hairpin, which melts at a higher temperature than Bc (Tm = 41.4 ± 1.0°C) (Table 1A). This may result from the change in the stem closing pair from GC in Bc to CG in Bnc or from a difference in adenine/thymine content of their loops (50 and references therein). Probing data indicate that Bnc adopts a single hairpin structure bearing a four-membered loop TCTT, shown by the similar and enhanced cleavage pattern of T23, T24 and T26 (Fig. 5). The dangling 3[prime]-end residues, T7, T9 and T11, follow a similar trend. Protection of T15, T17, T21 and T29 indicates the stem residues. The results show that Bnc does not possess the structural flexibility of its complementary sequence.

Biological relevance of Bc flexibility to the 1731 retrotransposon

1731 is a mobile retrovirus-like element, framed by two LTRs of U3-R-U5 architecture, both being unidirectional promoters (31-34). The Bc sequence is located in the U3 region of both LTRs, between positions -110 and -73 in the 5[prime] LTR, thus just upstream of the transcription start (+1). The structural flexibility of Bc may be associated with its function. The regulatory transcription factors NssBF and p11 only bind to the Bc strand (31-33). DNA hairpins have been shown to be involved in transcription and replication processes (21-24) and in topoisomerase II recognition and cleavage (25-28). Moreover, one of the main cleavage sites of topoisomerase II is precisely located in the Bc sequence within the whole LTR (34), indicating that this enzyme is probably required for Bc extrusion and function. Thus Bc slippage suggests that structural and topological parameters contribute to modulate the rate of transcription and transposition of 1731 retrotransposon (51).

CONCLUSION

The binding site of the transcription factor NssBF, named Bc, mainly adopts two hairpin structures in slow equilibrium at pH 6.0. The predominant structures of Bc, derived from probing experiments, consist of 4 and 8 nt loop hairpins, but these techniques cannot discount minor or transient contributions of alternative structures. Mutated hairpins carrying 4 and 8 nt loops can mimic its melting profile. A macrocyclic bis-acridine has been used to probe the structural flexibility of Bc compared with 4 and 8 nt loop model hairpins. The slow equilibrium between these hairpins probably involves slippage of 4 nt, originating from the presence of three CATA repeats complementary to the GTAT nucleotides in the Bc sequence.

The presence of magnesium ions affects Bc structures probed at pH 6.0. Chemical probing shows modifications at the lower stem level, consistent with two bulged hairpins in equilibrium. A comparison of Bc melting with that of a mutated sequence lacking the 5[prime]-end is also consistent with lower stem and bulge formation in the presence of magnesium ions.

In contrast, the Bc complementary strand adopts a single hairpin structure carrying a 4 nt loop. Two transcriptional factors bind only to the Bc strand within the whole retrotransposon 1731 LTR, suggesting that Bc structural flexibility could be associated with transcription.

ACKNOWLEDGEMENTS

The authors thank Prof. Lehn and Drs Vigneron and Teulade-Fichou for the gift of the macrocyclic bis-acridine. Dr Ouali and Prof. Westhof are acknowledged for helpful discussions, Drs Buckingham and Hui Bon Hoa for their critical reading of the manuscript and T. Bataille for the drawings. This work was supported by INTAS grant 95-IN-RU-653.

REFERENCES

1. Sierzputowka-Gracz,H., McKenzie,R.A. and Theil,E.C. (1995)Nucleic Acids Res., 23, 146-153.

2. Kikinis,Z., Eisenstein,R.S., Bettany,A.E. and Munro,H.N. (1995)Nucleic Acids Res., 23, 4190-4195. MEDLINE Abstract

3. Witherell,G.W., Gott,J.M. and Uhlenbeck,O.C. (1991)Prog. Nucleic Acids Res., 40, 185.

4. Borer,P.N., Lin,Y., Roggenbuck,M.W., Gott,J.M., Uhlenbeck,O.C. and Pelezer,I. (1995) Biochemistry, 34, 6488-6503. MEDLINE Abstract

5. Hall,K.B. and Stump,W.T. (1992) Nucleic Acids Res., 20, 4283-4290. MEDLINE Abstract

6. Oubridge,C., Ito,N., Evans,P.R., Teo,C.H. and Nagai,K. (1994) Nature, 272, 432-438. MEDLINE Abstract

7. Williams,A.S. and Marzluff,W.F. (1995) Nucleic Acids Res., 23, 654-662. MEDLINE Abstract

8. Hehl,A., Vassella,E., Braun,R. and Roditi,I. (1994) Proc. Natl Acad. Sci. USA, 91, 370-374. MEDLINE Abstract

9. Persson,C., Wagner,E.G.H. and Nordström,K. (1990) EMBO J., 9, 3767-3775. MEDLINE Abstract

10. Hjalt,T.A.H. and Wagner,E.G.H. (1995) Nucleic Acids Res., 23, 580-587. MEDLINE Abstract

11. Marino,J.P., Gregorian,R.S.G., Csankovszki,G. and Crothers,D.M. (1995) Science, 268, 1446-1454.

12. Mitas,M., Yu,A., Dill,J., Kamp,T.J., Chambers,E.J. and Haworth,I.S. (1995) Nucleic Acids Res., 23, 1050-1059. MEDLINE Abstract

13. Bruzik,J.P., Van Doren,K., Hirsch,D. and Steitz,J.A. (1988) Nature, 335, 559-562. MEDLINE Abstract

14. Gallinaro,H., Domenjoud,L. and Jacob,M. (1994) J. Mol. Biol., 240, 206-225.

15. White,S.A., Li,H. and Rauch,M.E. (1995) J. Biomol. Struct. Dyn., 13, 285-299. MEDLINE Abstract

16. Philippe,C., Bénard,L., Potier,C., Westhof,E., Ehresmann,B. and Ehresmann,C. (1995) Nucleic Acids Res., 23, 18-28. MEDLINE Abstract

17. Konings,D.A.M., Nash,M.A., Maizel,J.V. and Arlinghaus,R.B. (1992) J. Virol., 66, 632-640.

18. Wang,G.H. and Seeger,C. (1993) J. Virol., 67, 6507-6512. MEDLINE Abstract

19. Knaus,T. and Nassal,M. (1993) Nucleic Acids Res., 21, 3967-3975. MEDLINE Abstract

20. Yang,S. and Temin,H.M. (1994) EMBO J., 13, 713-726. MEDLINE Abstract

21. Wilson,K.S. and von Hippel,P.H. (1995) Proc. Natl Acad. Sci. USA, 92, 8793-8797. MEDLINE Abstract

22. McMurray,C.T., Wilson,W.D. and Douglass,J.O. (1991) Proc. Natl Acad. Sci. USA, 88, 666-670. MEDLINE Abstract

23. Gacy,A.M. and McMurray,C.T. (1994) Biochemistry, 33, 11951-11959. MEDLINE Abstract

24. Miao,D.M., Honda,Y., Tanaka,K., Higashi,A., Nakumara,T., Tagushi,Y., Sakai,H., Komano,T. and Bagdasarian,M. (1993) Nucleic Acids Res., 21, 4900-4903. MEDLINE Abstract

25. Froelich-Ammon,S.J., Gates,K.C. and Osheroff,N. (1994) J. Biol. Chem., 269, 7719-7725. MEDLINE Abstract

26. Pognan,F. and Paoletti,C. (1992) Biochimie, 74, 1019-1023. MEDLINE Abstract

27. Amir-Aslani,A., Mauffret,O., Bittoun,P., Sourgen,F., Monnot,M., Lescot,E. and Fermidjian,S. (1995) Nucleic Acids Res., 23, 3850-3857. MEDLINE Abstract

28. Hanke,J.H., Hambor,J.E. and Kavathas,P. (1995) J. Mol. Biol., 246, 63-73. MEDLINE Abstract

29. Chen,X., Mariappan,S.S.V., Catati,P., Ratlife,R., Moyzis,R.K., Laayoun,A., Smith,S.S., Bradbury,E.M. and Gupta,G. (1995) Proc. Natl Acad. Sci. USA, 92, 5199-5203. MEDLINE Abstract

30. Mariappan,S.V.S., Garcia,A.E. and Gupta,G. (1996) Nucleic Acids Res., 24, 775-783.

31. Fourcade-Peronnet,F., Codani-Simonart,S. and Best-Belpomme,M. (1992) J. Virol., 66, 1682-1687. MEDLINE Abstract

32. Lacoste,J. and Fourcade-Peronnet,F. (1995) FEBS Lett., 357, 283-286. MEDLINE Abstract

33. Lacoste,J., Codani-Simonart,S., Best-Belpomme,M. and Peronnet,F. (1995) Nucleic Acids Res., 23, 5073-5079. MEDLINE Abstract

34. Nahon,E., Best-Belpomme,M. and Saucier,J.-M. (1993) Eur. J. Biochem., 218, 95-102. MEDLINE Abstract

35. Dunaway,M. and Ostrander,E.A. (1993) Nature, 361, 746-748. MEDLINE Abstract

36. Dai,X., Greizerstein,M.B., Nadas-Chinni,K. and Rothman-Denes,L.B. (1997) Proc. Natl Acad. Sci. USA, 94, 2174-2179. MEDLINE Abstract

37. Teulade-Fichou,M.-P., Vigneron,J.-P. and Lehn,J.-M. (1995)J. Supramol. Chem., 5, 139-147.

38. Slama-Schwok,A., Teulade-Fichou,M.-P., Vigneron,J.-P., Taillandier,E. and Lehn,J.-M. (1995) J. Am. Chem. Soc., 117, 6822-6830.

39. Slama-Schwok,A., Peronnet,F., Hantz-Brachet,E., Taillandier,E., Teulade-Fichou,M.-P., Vigneron,J.-P., Best-Belpomme,M. and Lehn,J.-M. (1997) Nucleic Acid Res., 25, 2574-2581. MEDLINE Abstract

40. Breslauer,K.J., Frank,R., Bloecker,H. and Marky,L.A. (1986)Proc. Natl Acad. Sci. USA, 83, 3746-3750. MEDLINE Abstract

41. Bhattachayya,A. and Lilley,D.M.J. (1988) Nucleic Acids Res., 17, 6821-6840.

42. Joshua-Tor,L., Rabinovich,D., Hope,H., Frolow,F., Appella,E. and Sussman,J.L. (1988) Nature, 334, 82-84 and 85-86.

43. Rice,J.A. and Crothers,D.M. (1989) Biochemistry, 28, 4512-4516. MEDLINE Abstract

44. Hsieh,C.H. and Griffith,J.D. (1989) Proc. Natl Acad. Sci. USA, 86, 4833-4837. MEDLINE Abstract

45. Bhattachayya,A., Murchle,A.I.H. and Lilley,D.M.J. (1990) Nature, 343, 484-487.

46. Rosen,M.A., Live,D. and Patel,D.J. (1992) Biochemistry, 31, 4004-4014 and 4015-4026. MEDLINE Abstract

47. Wang,Y.H., Barker,P. and Griffith,J.D. (1992) J. Biol. Chem., 267, 4911-4915. MEDLINE Abstract

48. Aboul-ela,F., Murchie,A.I.H., Homans,S.W. and Lilley,D.M.J. (1993) J. Mol. Biol., 229, 173-188. MEDLINE Abstract

49. Gohlke,C., Murchie,A.I.H., Lilley,D.M.J. and Clegg,R.M. (1994)Proc. Natl Acad. Sci. USA, 91, 11660-11664. MEDLINE Abstract

50. Paner,T.M., Ricelli,P.V., Owczarzy,A.S., Benight,A.S. (1996) Biopolymers, 39, 779-793. MEDLINE Abstract

51. van Gent,D.C., Mizuuchi,K. and Gellert M. (1996) Science, 271, 1592-1594. MEDLINE Abstract

*To whom correspondence should be addressed. Tel: +33 1 43 25 26 09; Fax: +33 1 43 29 80 88; Email: schwok@ibpc.fr

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 3 Nov 1998
Copyright©Oxford University Press, 1998.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
M. Kaushik and S. Kukreti
Structural polymorphism exhibited by a quasipalindrome present in the locus control region (LCR) of the human {beta}-globin gene cluster
Nucleic Acids Res., July 19, 2006; 34(12): 3511 - 3522.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (285K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (3)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Slama-Schwok, A.
Right arrow Articles by Vlassov, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Slama-Schwok, A.
Right arrow Articles by Vlassov, V.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?