Back

Parallel DNA Holliday Junctions: Myth or Reality? An Atomistic Molecular Dynamics Study

Lemmens, T.; Sponer, J.; Stadlbauer, P.; Krepl, M.

2026-07-16 biophysics
10.64898/2026.07.10.737670 bioRxiv
Show abstract

Holliday junctions (HJs) are key intermediates of homologous recombination and fundamental building blocks in DNA nanotechnology. Although the canonical antiparallel stacked-X conformation has been extensively characterized by X-ray crystallography, whether parallel HJ conformations exist in aqueous solution remains unresolved. Here, we address this question using extensive atomistic molecular dynamics (MD) simulations and replica-exchange umbrella sampling (REUS) free-energy calculations. Starting from canonical antiparallel HJs, standard MD simulations occasionally revealed spontaneous transitions to parallel conformations on the microsecond timescale without disrupting the DNA duplexes or passing through an open junction intermediate. REUS free-energy profiles confirmed the antiparallel state as the global minimum but also identified the parallel conformation as a well-defined local minimum, indicating it is thermodynamically metastable despite an estimated solution population below 1%. The combination of low equilibrium occupancy, microsecond interconversion dynamics, and the surprisingly close structural similarity between antiparallel and parallel junctions provides a plausible explanation for the lack of direct experimental detection. We further found that the free-energy landscape is only weakly affected by branching-point sequence and salt concentration. These results reconcile the apparent absence of experimental evidence for parallel HJs with their structural feasibility in solution and offer a fresh perspective on the historical debate. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=61 SRC="FIGDIR/small/737670v1_ufig1.gif" ALT="Figure 1"> View larger version (24K): org.highwire.dtl.DTLVardef@bb61a0org.highwire.dtl.DTLVardef@673a7org.highwire.dtl.DTLVardef@192f5adorg.highwire.dtl.DTLVardef@13f4941_HPS_FORMAT_FIGEXP M_FIG C_FIG

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.