Quantitation of metal ion and DNA junction binding to the Holliday junction endonuclease Cce1

Cce1 is a magnesium-dependent Holliday junction endonuclease involved in the resolution of recombining mitochondrial DNA in Saccharomyces cerevisiae. Cce1 binds four-way DNA junctions as a dimer, opening the junction into an extended, 4-fold symmetric structure, and resolves junctions by the introduction of paired nicks in opposing strands at the point of strand exchange. In the present study, we have examined the interactions of wild-type Cce1 with a noncleavable four-way DNA junction and metal ions (Mg(2+) and Mn(2+)) using isothermal titration calorimetry, EPR, and gel electrophoresis techniques. Mg(2+) or Mn(2+) ions bind to Cce1 in the absence of DNA junctions with a stoichiometry of two metal ions per Cce1 monomer. Cce1 binds to four-way junctions with a stoichiometry of two Cce1 dimers per junction molecule in the presence of EDTA, and one dimer of Cce1 per junction in 15 mM magnesium. The presence of 15 mM Mg(2+) dramatically reduces the affinity of Cce1 for four-way DNA junctions, by about 900-fold. This allows an estimation of DeltaG degrees for stacking of four-way DNA junction 7 of -4.1 kcal/mol, consistent with the estimate of -3.3 to -4.5 kcal/mol calculated from branch migration and NMR experiments [Overmars and Altona (1997) J. Mol. Biol. 273, 519-524; Panyutin et al. (1995) EMBO J. 14, 1819-1826]. The striking effect of magnesium ions on the affinity of Cce1 binding to the four-way junction is predicted to be a general one for proteins that unfold the stacked X-structure of the Holliday junction on binding.     

Conformation and protein interactions of intramolecular DNA and phosphorothioate four-way junctions

The objectives of this study are to evaluate the structure and protein recognition features of branched DNA four-way junctions in an effort to explore the therapeutic potential of these molecules. The classic immobile DNA 4WJ, J1, is used as a matrix to design novel intramolecular junctions including natural and phosphorothioate bonds. Here we have inserted H2-type mini-hairpins into the helical termini of the arms of J1 to generate four novel intramolecular four-way junctions. Hairpins are inserted to reduce end fraying and effectively eliminate potential nuclease binding sites. We compare the structure and protein recognition features of J1 with four intramolecular four-way junctions: i-J1, i-J1(PS1), i-J1(PS2) and i-J1(PS3). Circular dichroism studies suggest that the secondary structure of each intramolecular 4WJ is composed predominantly of B-form helices. Thermal unfolding studies indicate that intramolecular four-way junctions are significantly more stable than J1. The T_m values of the hairpin four-way junctions are 25.2° to 32.2°C higher than the control, J1. With respect to protein recognition, gel shift assays reveal that the DNA-binding proteins HMGBb1 and HMGB1 bind the hairpin four-way junctions with affinity levels similar to control, J1. To evaluate nuclease resistance, four-way junctions are incubated with DNase I, exonuclease III (Exo III) and T5 exonuclease (T5 Exo). The enzymes probe nucleic acid cleavage that occurs non-specifically (DNase I) and in a 5ʹ→3ʹ (T5 Exo) and 3ʹ→5ʹ direction (Exo III). The nuclease digestion assays clearly show that the intramolecular four-way junctions possess significantly higher nuclease resistance than the control, J1.     

Effects of base mismatches on the structure of the four-way DNA junction

Heteroduplex formation between imperfectly homologous DNA sequences may result in the formation of a four-way junction at which non-Watson-Crick base mismatches are present at the point of strand exchange. This raises the question of the effect of such mismatches on the structure and stability of these potential recombination intermediates. We have constructed a series of four-way DNA junctions containing single-base mismatches, and have studied the structure of the junctions by means of gel electrophoresis and chemical modification. We observed a range of effects on the structure of the junction, ranging from almost total abolition of folding through to normal accommodation into the folded structure. In some cases we observed gel electrophoretic data consistent with a dynamic equilibrium between folded and unfolded conformations, and in general the folded form was favoured at higher concentrations of cation. The effects of single mismatches on the structure of the four-way junction may be summarized in terms of: (1) the nature of the mismatch, where we note a correlation between the thermal stability of a given mismatch and its ability to be accommodated into a folded junction; or (2) the sequence context, where the effect of a given mismatch on the structure of a junction depends on the neighbouring base-pairs. These factors are illustrated by a junction, containing a C.A mismatch, that adopted alternate isomeric conformations dependent upon pH; as the state of protonation of the mispair changed, the structure was altered along with the interaction with neighbouring base-pairs. Most base mismatches may be accommodated into the folded stacked X-conformation of the four-way junction, but many require elevated cation concentration to permit the folding process to proceed. Some mismatches were found to be extremely destabilizing.     

Thermodynamic properties of an intramolecular DNA four-way junction

We have investigated the thermodynamic properties of two homologous DNA four-way junctions, J4 and J4M, based on 46-mer linear DNA molecules. J4 and J4M have the same base sequence with the only difference that the latter contains an uncharged methylene-acetal linkage, -O3'-CH2-O5', instead of the phosphodiester linkage, -O3'-PO2-O5'-, between the residues T18 and C19. The comparison of the thermal unfolding of the J4 junction and J4M junction serves to investigate the effect of the uncharged methylene-acetal linkage on the stability of the junction. Our analysis is based on CD, UV absorbance spectroscopy, DSC, and chemical footprinting. The aim is to characterize in detail the structure and stability of the junctions. As demonstrated before by NMR, in the presence of 5 mM MgCl2 +/- 50 mM NaCl, both J4 and J4M form a complete four-way junction. This is now evidenced by protection from OsO4 cleavage (chemical footprinting). We can assume that full base pairing occurs throughout the arms even at the center of the junction. CD spectra suggest that the helices within the junctions adopt the regular B-DNA conformation. Almost identical melting temperatures and unfolding enthalpies are obtained for J4 and J4M both by UV and DSC. Furthermore, the Van't Hoff enthalpy (DeltaHVH) derived from UV melting equals the calorimetric enthalpy (DeltaHcal), which means that the melting process of the structures proceeds in a two-state manner. All results taken together support the conclusion that there are no major conformational and energetic differences between J4 and J4M. The inclusion of the uncharged methylene-acetal group into the junction has no effect on its stability.     

Impact of the third-strand orientation on the thermodynamic stability of the four-way DNA junction

The physical properties of a triple-helical DNA four-way junction J(T2T4) have been characterized by means of UV spectroscopy, CD spectroscopy, and differential scanning calorimetry (DSC). J(T2T4) is another four-way junction that was designed in addition to J(T1T3) (N. Makube and H. H. Klump (2000) Arch. Biochem. Biophys. 377, 31-42) to study the effects of third strands on the stability of the four-way junction with triple-helical arms. The pH titration curves illustrate the sequential folding of single strands to double-helical four-way junctions and finally the binding of third strands to their respective W-C duplexes. CD measurements confirm triplex formation under appropriate pH and ionic strength conditions. The CD spectra also suggest different melting patterns for the triple-helical arms of J(T2T4). The melting temperature as a function of pH or ionic strength characterizes the effect of the third strands on the structural stability. Increased sodium concentration and low pH conditions enhances and stabilizes the overall structure of the junction. The results also indicate that all triplexes in J(T2T4) are formed in the absence of salt and at low pH; however, the junction may, under these conditions, assume a conformation different from the one assumed in the presence of salt. Through the deconvolution of DSC data, the calorimetric enthalpies associated with melting of arms of the junctions were determined. The loops are designed to have the same enthalpic effect on the different arms. The stabilizing effect of the loops is more pronounced when those loops are shifted from arms 1 and 3 in J(T1T3) to arms 2 and 4 in J(T2T4) without changing any of the sequences. Overall, J(T2T4) is slightly more stable than J(T1T3). The differences can be attributed to sequence effects rather than structural effects. All the results illustrate that binding of the third strand in either of the two orientations 5'5'3' (J(T2T4)) or 5'3'3' (J(T1T3)) stabilizes the underlying double-helical four-way junction and its triple-helical arms.     

The inherent properties of DNA four-way junctions: comparing the crystal structures of holliday junctions

Holliday junctions are four-stranded DNA complexes that are formed during recombination and related DNA repair events. Much work has focused on the overall structure and properties of four-way junctions in solution, but we are just now beginning to understand these complexes at the atomic level. The crystal structures of two all-DNA Holliday junctions have been determined recently from the sequences d(CCGGGACCGG) and d(CCGGTACCGG). A detailed comparison of the two structures helps to distinguish distortions of the DNA conformation that are inherent to the cross-overs of the junctions in this crystal system from those that are consequences of the mismatched dG.dA base-pair in the d(CCGGGACCGG) structure. This analysis shows that the junction itself perturbs the sequence-dependent conformational features of the B-DNA duplexes and the associated patterns of hydration in the major and minor grooves only minimally. This supports the idea that a DNA four-way junction can be assembled at relatively low energetic cost. Both structures show a concerted rotation of the adjacent duplex arms relative to B-DNA, and this is discussed in terms of the conserved interactions between the duplexes at the junctions and further down the helical arms. The interactions distant from the strand cross-overs of the junction appear to be significant in defining its macroscopic properties, including the angle relating the stacked duplexes across the junction.     

Direct analysis of Holliday junction resolving enzyme in a DNA origami nanostructure

Holliday junction (HJ) resolution is a fundamental step for completion of homologous recombination. HJ resolving enzymes (resolvases) distort the junction structure upon binding and prior cleavage, raising the possibility that the reactivity of the enzyme can be affected by a particular geometry and topology at the junction. Here, we employed a DNA origami nano-scaffold in which each arm of a HJ was tethered through the base-pair hybridization, allowing us to make the junction core either flexible or inflexible by adjusting the length of the DNA arms. Both flexible and inflexible junctions bound to Bacillus subtilis RecU HJ resolvase, while only the flexible junction was efficiently resolved into two duplexes by this enzyme. This result indicates the importance of the structural malleability of the junction core for the reaction to proceed. Moreover, cleavage preferences of RecU-mediated reaction were addressed by analyzing morphology of the reaction products.     

Holliday junction resolvase in Schizosaccharomyces pombe has identical endonuclease activity to the CCE1 homologue YDC2.

A novel Holliday junction resolving activity has been identified in fractionated cell extracts of the fission yeast Schizosaccharomyces pombe . The enzyme catalyses endonucleolytic cleavage of Holliday junction-containing chi DNA and synthetic four-way DNA junctions. The activity cuts with high specificity a synthetic four-way junction containing a 12 bp core of homologous sequences but has no activity on another four-way junction (with a fixed crossover point), a three-way junction, linear duplex DNA or duplex DNA containing six mismatched nucleotides in the centre. The major cleavage sites map as single nicks in the vicinity of the crossover point, 3' of a thymidine residue. These data indicate that the activity has a strong DNA structure selectivity as well as a limited sequence preference; features similar to the Holliday junction resolving enzymes RuvC of Escherichia coli and the mitochondrial CCE1 (cruciform-cuttingenzyme 1) of Saccharomyces cerevisiae. A putative homologue of CCE1 in S.pombe (YDC2_SCHPO) has been identified through a search of the sequence database. The open reading frame of this gene has been cloned and the encoded protein, YDC2, expressed in E.coli . The purified recombinant YDC2 exhibits Holliday junction resolvase activity and is, therefore, a functional S.pombe homologue of CCE1. The resolvase YDC2 shows the same substrate specificity and produces identical cleavage sites as the activity obtained from S. pombe cells. Both YDC2 and the cellular activity cleave Holliday junctions in both orientations to give nicks that can be ligated in vitro. The partially purified Holliday junction resolving enzyme in fission yeast is biochemically indistinguishable from recombinant YDC2 and appears to be the same protein.     

Effect of sequence on the conformation of DNA holliday junctions

Structures of the DNA sequences d(CCGGCGCCGG) and d(CCAGTACbr(5)UGG) are presented here as four-way Holliday junctions in their compact stacked-X forms, with antiparallel alignment of the DNA strands. Thus, the ACC-trinucleotide motif, previously identified as important for stabilizing the junction, is now extended to PuCPy, where Pu is either an adenine or guanine, and Py is either a cytosine, 5-methylcytosine, or 5-bromouracil but not thymine nucleotide. We see that both sequence and base substituents affect the geometry of the junction in terms of the interduplex angle as well as the previously defined conformational variables, J(roll) (the rotation of the stacked duplexes about their respective helical axis) and J(slide) (the translational displacement of the stacked duplexes along their respective helical axis). The structures of the GCC and parent ACC containing junctions fall into a distinct conformational class that is relatively undistorted in terms of J(slide) and J(roll), with interduplex angles of 40-43 degrees. The substituted ACbr(5)U structure, however, is more akin to that of the distorted methylated ACm(5)C containing junction, with J(slide) (>or=2.3 A) and a similar J(roll) (164 degrees) opening the major groove-side of the junction, but shows a reduced interduplex angle. In contrast, the analogous d(CCAGTACTGG) sequence has to date been crystallized only as resolved B-DNA duplexes. This suggests that there is an electronic effect of substituents at the pyrimidine Py position on the stability of four-stranded junctions. The single-crystal structures presented here, therefore, show how sequence affects the detailed geometry, and subsequently, the associated stability and conformational dynamics of the Holliday junction.     

Tertiary Motifs Revealed in Analyses of Higher-Order RNA Junctions

RNA junctions are secondary-structure elements formed when three or more helices come together. They are present in diverse RNA molecules with various fundamental functions in the cell. To better understand the intricate architecture of three-dimensional (3D) RNAs, we analyze currently solved 3D RNA junctions in terms of base-pair interactions and 3D configurations. First, we study base-pair interaction diagrams for solved RNA junctions with 5 to 10 helices and discuss common features. Second, we compare these higher-order junctions to those containing 3 or 4 helices and identify global motif patterns such as coaxial stacking and parallel and perpendicular helical configurations. These analyses show that higher-order junctions organize their helical components in parallel and helical configurations similar to lower-order junctions. Their sub-junctions also resemble local helical configurations found in three- and four-way junctions and are stabilized by similar long-range interaction preferences such as A-minor interactions. Furthermore, loop regions within junctions are high in adenine but low in cytosine, and in agreement with previous studies, we suggest that coaxial stacking between helices likely forms when the common single-stranded loop is small in size; however, other factors such as stacking interactions involving noncanonical base pairs and proteins can greatly determine or disrupt coaxial stacking. Finally, we introduce the ribo-base interactions: when combined with the along-groove packing motif, these ribo-base interactions form novel motifs involved in perpendicular helix-helix interactions. Overall, these analyses suggest recurrent tertiary motifs that stabilize junction architecture, pack helices, and help form helical configurations that occur as sub-elements of larger junction networks. The frequent occurrence of similar helical motifs suggest nature's finite and perhaps limited repertoire of RNA helical conformation preferences. More generally, studies of RNA junctions and tertiary building blocks can ultimately help in the difficult task of RNA 3D structure prediction.     

A four-way junction with triple-helical arms: design, characterization, and stability

The formation of the four-way junction containing four triple-helical arms has been demonstrated using chemical methods (polyacrylamide gel electrophoresis and chemical footprinting using OsO(4) as a probe) and physical methods (UV absorbance melting and DSC). The junction J(T1T3) was assembled from two 20-mer purine strands and two 44-mer pyrimidine strands. To determine the contribution of the different arms to the stability of the complete structure of J(T1T3), the junction was compared to two simplified substructures, J(T1) and J(T3), respectively. Common to these complexes is the underlying double-helical four-way junction Js. Addition of Na(+) had a profound effect on stabilizing and subsequently folding the junctions into the stacked X-structures. The following results support the structure present: (i) The native polyacrylamide electrophoresis exhibits only a single band(s) corresponding to one species present when all four single strands are mixed in equal amounts. (ii) OsO(4) modifications were investigated at pH 5.0 and in the presence of 10 mM Mg(2+) and 100 mM Na(+). There is no cleavage of thymine residues at the branch point and throughout the structure. (iii) The thermal unfolding of J(T1) and J(T3) illustrates that the triple-helical arms are more stable than the double-helical arms which are contained in these junctions and that J(T1T3) with four triple-helical arms is slightly more stable than J(T1) and J(T3). (iv) The calorimetric transition enthalpies determined for the arms of J(T1T3) are comparable to those associated with the unfolding of its corresponding arms in J(T1) and J(T3). The results also illustrate that the formation of the junctions is not restricted by the pH, [Na(+)], sequence composition of the arms, and/or the loop position.     

Relative stabilities of DNA three-way, four-way and five-way junctions (multi-helix junction loops): unpaired nucleotides can be stabilizing or destabilizing.

Competition binding and UV melting studies of a DNA model system consisting of three, four or five mutually complementary oligonucleotides demonstrate that unpaired bases at the branch point stabilize three- and five-way junction loops but destabilize four-way junctions. The inclusion of unpaired nucleotides permits the assembly of five-way DNA junction complexes (5WJ) having as few as seven basepairs per arm from five mutually complementary oligonucleotides. Previous work showed that 5WJ, having eight basepairs per arm but lacking unpaired bases, could not be assembled [Wang, Y.L., Mueller, J.E., Kemper, B. and Seeman, N.C. (1991) Biochemistry, 30, 5667-5674]. Competition binding experiments demonstrate that four-way junctions (4WJ) are more stable than three-way junctions (3WJ), when no unpaired bases are included at the branch point, but less stable when unpaired bases are present at the junction. 5WJ complexes are in all cases less stable than 4WJ or 3WJ complexes. UV melting curves confirm the relative stabilities of these junctions. These results provide qualitative guidelines for improving the way in which multi-helix junction loops are handled in secondary structure prediction programs, especially for single-stranded nucleic acids having primary sequences that can form alternative structures comprising different types of junctions.     

Analysis of conformational changes in the DNA junction-resolving enzyme T7 endonuclease I on binding a four-way junction using EPR

The four-way (Holliday) DNA junction is the central intermediate in homologous recombination. It is ultimately resolved into two nicked-duplex species by the action of a junction-resolving enzyme. These enzymes are highly selective for the structure of branched DNA, yet as a class these proteins impose significant distortion on their target junctions. Bacteriophage T7 endonuclease I selectively binds and cleaves DNA four-way junctions. The protein is an extremely stable dimer, comprising two globular domains joined by a β-strand bridge with each active site including amino acids from both polypeptides. The crystal structure of endonuclease I has been solved both as free protein and in complex with a DNA junction, showing that the protein, as well as the junction, becomes distorted on binding. We have therefore used site-specific spin-labeling in conjunction with EPR distance measurements to analyze induced fit in the binding of endonuclease I to a DNA four-way junction. The results support the change in protein structure as it binds to the junction. In addition, we have examined the structure of wild type and catalytically inactive mutants alone and in complex with DNA. We demonstrate the presence of hitherto undefined metastable conformational states within endonuclease I, showing how these states can be influenced by DNA-junction binding or mutations within the active sites. In addition, we demonstrate a previously unobserved instability in the N-terminal α1-helix upon active site mutation. These studies reveal that structural changes in both DNA and protein occur in the action of this junction-resolving enzyme.     

Substrate recognition and catalysis by the Holliday junction resolving enzyme Hje

Two archaeal Holliday junction resolving enzymes, Holliday junction cleavage (Hjc) and Holliday junction endonuclease (Hje), have been characterized. Both are members of a nuclease superfamily that includes the type II restriction enzymes, although their DNA cleaving activity is highly specific for four-way junction structure and not nucleic acid sequence. Despite 28% sequence identity, Hje and Hjc cleave junctions with distinct cutting patterns--they cut different strands of a four-way junction, at different distances from the junction centre. We report the high-resolution crystal structure of Hje from Sulfolobus solfataricus. The structure provides a basis to explain the differences in substrate specificity of Hje and Hjc, which result from changes in dimer organization, and suggests a viral origin for the Hje gene. Structural and biochemical data support the modelling of an Hje:DNA junction complex, highlighting a flexible loop that interacts intimately with the junction centre. A highly conserved serine residue on this loop is shown to be essential for the enzyme's activity, suggesting a novel variation of the nuclease active site. The loop may act as a conformational switch, ensuring that the active site is completed only on binding a four-way junction, thus explaining the exquisite specificity of these enzymes.     

Sequence specific DNA binding of Ets-1 transcription factor: molecular dynamics study on the Ets domain--DNA complexes

Molecular dynamics (MD) simulations for Ets-1 ETS domain-DNA complexes were performed to investigate the mechanism of sequence-specific recognition of the GGAA DNA core by the ETS domain. Employing the crystal structure of the Ets-1 ETS domain-DNA complex as a starting structure we carried out MD simulations of: (i). the complex between Ets-1 ETS domain and a 14 base-pair DNA containing GGAA core sequence (ETS-GGAA); (ii). the complex between the ETS domain and a DNA having single base-pair mutation, GGAG sequence (ETS-GGAG); and (iii). the 14 base-pair DNA alone (GGAA). Comparative analyses of the MD structures of ETS-GGAA and ETS-GGAG reveal that the DNA bending angles and the ETS domain-DNA phosphate interactions are similar in these complexes. These results support that the GGAA core sequence is distinguished from the mutated GGAG sequence by a direct readout mechanism in the Ets-1 ETS domain-DNA complex. Further analyses of the direct contacts in the interface between the helix-3 region of Ets-1 and the major groove of the core DNA sequence clearly show that the highly conserved arginine residues, Arg391 and Arg394, play a critical role in binding to the GGAA core sequence. These arginine residues make bidentate contacts with the nucleobases of GG dinucleotides in GGAA core sequence. In ETS-GGAA, the hydroxyl group of Tyr395 is hydrogen bonded to N7 nitrogen of A(3) (the third adenosine in the GGAA core), while the hydroxyl group makes a contact with N4 nitrogen of C(4') (the complementary nucleotide of the fourth guanosine G(4) in the GGAG sequence) in the ETS-GGAG complex. We have found that this difference in behavior of Tyr395 results in the relatively large motion of helix-3 in the ETS-GGAG complex, causing the collapse of bidentate contacts between Arg391/Arg394 and the GG dinucleotides in the GGAG sequence.     

An archaeal Holliday junction resolving enzyme from Sulfolobus solfataricus exhibits unique properties

The rearrangement and repair of DNA by homologous recombination often involves the creation of Holliday junctions, which must be cleaved by junction-specific endonucleases to yield recombinant duplex DNA products. Holliday junction resolving enzymes are a ubiquitous class of proteins with diverse structural and mechanistic characteristics. We have characterised an endonuclease (Hje) from the thermophilic crenarchaeote Sulfolobus solfataricus that exhibits a high degree of specificity for Holliday junctions via an apparently novel mechanism. Hje resolves four-way DNA junctions by the introduction of paired nicks in a reaction that is independent of the local nucleotide sequence, but is restricted solely to strands that are continuous in the stacked-X form of the junction. Three-way DNA junctions are cleaved only when the presence of a bulge in one strand allows the junction to stack in an analogous manner to four-way junctions. These properties differentiate Hje from all other known junction resolving enzymes.     

Holliday junction resolution is modulated by archaeal chromatin components in vitro.

The Holliday junction-resolving enzyme Hjc is conserved in the archaea and probably plays a role analogous to that of Escherichia coli RuvC in the pathway of homologous recombination. Hjc specifically recognizes four-way DNA junctions, cleaving them without sequence preference to generate recombinant DNA duplex products. Hjc imposes an X-shaped global conformation on the bound DNA junction and distorts base stacking around the point of cleavage, three nucleotides 3' of the junction center. We show that Hjc is autoinhibitory under single turnover assay conditions and that this can be relieved by the addition of either competitor duplex DNA or the architectural double-stranded DNA-binding protein Sso7d (i.e. by approximating in vivo conditions more closely). Using a combination of isothermal titration calorimetry and fluorescent resonance energy transfer, we demonstrate that multiple Hjc dimers can bind to each synthetic four-way junction and provide evidence for significant distortion of the junction structure at high protein:DNA ratios. Analysis of crystal packing interactions in the crystal structure of Hjc suggests a molecular basis for this autoinhibition. The wider implications of these findings for the quantitative study of DNA-protein interactions is discussed.     

New insight into the recognition of branched DNA structure by junction-resolving enzymes

Junction-resolving enzymes are nucleases that exhibit structural selectivity for the four-way (Holliday) junction in DNA. In general, these enzymes both recognize and distort the structure of the junction. New insight into the molecular recognition processes has been provided by two recent co-crystal structures of resolving enzymes bound to four-way DNA junctions in highly contrasting ways. T4 endonuclease VII binds the junction in an open conformation to an approximately flat binding surface whereas T7 endonuclease I envelops the junction, which retains a much more three-dimensional structure. Both proteins make contacts with the DNA backbone over an extensive area in order to generate structural specificity. The comparison highlights the versatility of Holliday junction resolution, and extracts some general principles of recognition.     

Structures of helical junctions in nucleic acids

Our knowledge of the architectural principles of nucleic acid junctions has seen significant recent advances. The conformation of DNA junctions is now well understood, and this provides a new basis for the analysis of important structural elements in RNA. The most significant new data have come from X-ray crystallography of four-way DNA junctions; incidentally showing the great importance of serendipity in science, since none of the three groups had deliberately set out to crystallise a junction. Fortunately the results confirm, and of course extend, the earlier conformational studies of DNA junctions in almost every detail. This is important, because it means that these methods can be applied with greater confidence to new systems, especially in RNA. Methods like FRET, chemical probing and even the humble polyacrylamide gel can be rapid and very powerful, allowing the examination of a large number of sequence variants relatively quickly. Molecular modelling in conjunction with experiments is also a very important component of the general approach. Ultimately crystallography provides the gold standard for structural analysis, but the other, simple approaches have considerable value along the way. At the beginning of this review I suggested two simple folding principles for branched nucleic acids, and it is instructive to review these in the light of recent data. In brief, these were the tendency for pairwise coaxial stacking of helical arms, and the importance of metal ion interactions in the induction of folding. We see that both are important in a wide range of systems, both in DNA and RNA. The premier example is the four-way DNA junction, which undergoes metal ion-induced folding into the stacked X-structure that is based on coaxial stacking of arms. As in many systems, there are two alternative ways to achieve this depending on the choice of stacking partners. Recent data reveal that both forms often exist in a dynamic equilibrium, and that the relative stability of the two conformers depends upon base sequence extending a significant distance from the junction. The three-way junction has provided a good test of the folding principles. Perfect three-way (3H) DNA junctions seem to defy these principles in that they appear reluctant to undergo coaxial stacking of arms, and exhibit little change in conformation with addition of metal ions. Modelling suggests that such a junction is stereochemically constrained in an extended conformation. However, upon inclusion of a few additional base pairs at the centre (to create a 3HS2 junction for example) the additional stereochemical flexibility allows two arms to undergo coaxial stacking. Such a junction exhibits all the properties consistent with the general folding principles, with ion-induced folding into a form based on pairwise coaxial stacking of arms in one of two different conformers. The three-way junction is therefore very much the exception that proves the rule. It is instructive to compare the folding of corresponding species in DNA and RNA, where we find both similarities and differences. The RNA four-way junction can adopt a structure that is globally similar to the stacked X-structure (Duckett et al. 1995a), and the crystal structure of the DNAzyme shows that the stacked X-conformation can include one helical pair in the A-conformation (Nowakowski et al. 1999). However, modelling suggests that the juxtaposition of strands and grooves will be less satisfactory in RNA, and the higher magnesium ion concentration required to fold the RNA junction indicates a lower stability of the antiparallel form. Perhaps the biggest difference between the properties of the DNA and RNA four-way junctions is the lack of an unstacked structure at low salt concentrations for the RNA species. This clearly reflects a major difference in the electrostatic interactions in the RNA junction. In general the folding of branched DNA provides some good indications on the likely folding of the corresponding RNA species, but caution is required in making the extrapolation because the two polymers are significantly different. A number of studies point to the flexibility and malleability of branched nucleic acids, and this turns out to have particular significance in their interactions with proteins. Proteins such as the DNA junction-resolving enzymes exhibit considerable selectivity for the structure of their substrates, which is still not understood at a molecular level. Despite this, it appears to be universally true that these proteins distort the global, and in some cases at least the local, structure of the junctions. The somewhat perplexing result is that the proteins appear to distort the very property that they recognise. In general it seems that four-way DNA junctions are opened to one extent or another by interaction with proteins. (ABSTRACT TRUNCATED)