The stereochemistry of a four-way DNA junction: a theoretical study

Abstract

The stereochemical conformation of the four-way helical junction in DNA (the Holliday junction; the postulated central intermediate of genetic recombination) has been analysed, using molecular mechanical computer modelling. A version of the AMBER program package was employed, that had been modified to include the influence of counterions and a global optimisation procedure. Starting from an extended planar structure, the conformation was varied in order to minimise the energy, and we discuss three structures obtained by this procedure. One structure is closely related to a square-planar cross, in which there is no stacking interaction between the four double helical stems. This structure is probably closely similar to that observed experimentally in the absence of cations. The remaining two structures are based on related, yet distinct, conformations, in which there is pairwise coaxial stacking of neighbouring stems. In these structures, the four DNA stems adopt the form of two quasi-continuous helices, in which base stacking is very similar to that found in standard B-DNA geometry. The two stacked helices so formed are not aligned parallel to each other, but subtend an angle of approximately 60 degrees. The strands that exchange between one stacked helix and the other are disposed about the smaller angle of the cross (i.e. 60 degrees rather than 120 degrees), generating an approximately antiparallel alignment of DNA sequences. This structure is precisely the stacked X-structure proposed on the basis of experimental data. The calculations indicate distortions from standard B-DNA conformation that are required to adopt the stacked X-structure; a widening of the minor groove at the junction, and reorientation of the central phosphate groups of the exchanging strands. An important feature of the stacked X-structure is that it presents two structurally distinct sides. These may be recognised differently by enzymes, providing a rationalisation for the points of cleavage by Holliday resolvases.

Kinking of DNA and RNA helices by bulged nucleotides observed by fluorescence resonance energy transfer.

“…Varying the and T angles in this way, as 0 increases, generates a folded juxtaposition of the two stems equivalent to that of the two quasicontinuous helices of the four-way DNA junction when 0 = 1200 and = T = 90°. As we (19,36,37) and others (38,39) have discussed, this orientation of two helices is electrostatically and geometrically favorable for closely juxtaposed DNA molecules.…”

Folding DNA to create nanoscale shapes and patterns

“…The source of the inter-helix gap is unknown, it may be electrostatic repulsion between helices (as first, to my knowledge, suggested by Rizal Hariadi), or detailed geometry of the crossovers (free crossovers, when not constrained by adjacent crossovers in a multi-crossover molecule, assume an angle of approximately 60 degrees 32,33 ). It remains for the gap to be measured on different substrates, or in solution, or by a different imaging technique such as TEM, or for it to be measured as a function of salt concentration which might be expected to change the gap by changing the screening of electrostatic interactions.…”

Binding of Endonuclease VII to Cruciform DNA

“…The three-dimensional structure of a cruciform DNA in solution was predicted by model building as an X-shaped scissors like arrangement of the arms in space with a small angle of 60°between two pairs of arms and a compensatory wide angle of 120°between the other two pairs of arms (19,21). Binding of endo-VII to cruciform DNA was reported to cause a structural change from an Xshaped molecule into a planar ϩ-shaped molecule (22).…”