Back

Interchain base-stacking and spermine bridges govern ssDNA condensation and reentrant transitions

Mondal, B.; Majumder, S.; Jain, A.; Thirumalai, D.

2026-10-08 biophysics
10.64898/2026.10.03.756434 bioRxiv
Show abstract

Polyvalent-cation-induced condensation and reentrant transition in double-stranded DNA are classically described by backbone electrostatics. Whether the same principles govern single-stranded DNA, whose solvent-exposed bases can interact within and between chains to form complex phases, is unknown. Here, we employ simulations and experiments to show that the tetravalent cation spermine (Spm4+) drives two equilibrium transitions in poly(dT) and poly(dA). The onset of phase separation, resulting in thermodynamic equilibrium between the dispersed low-density phase and high density coacervate, occurs at a low sequence independent Spm4+ concentration and is primarily driven by charge balance between phosphate groups and Spm4+. Strikingly, the condensed phases are stabilized by at most 1.5 (poly dT) and 2.5 (poly dA) bridging Spm4+ per chain. The ssDNA chains condense to form fluid-like networks without chain collapse. The fluidity of chains in the dense phases is confirmed by fluorescence recovery after photobleaching experiments. Both simulations and experiments show that the range of spermine concentration over which phase coexistence persists is sequence dependent and nearly threefold wider in poly(dA). At higher sequence dependent spermine concentrations, a reentrant phase, characterized by dissolution of the coacervates, emerges. The transition to the reentrant phase occurs when the number of bridging Spm4+ molecules vanishes at high sequence dependent spermine concentration. The combination of experiments and simulations show that the two phase transitions in ssDNA are driven by an interplay of stacking and electrostatic interactions.

Matching journals

The top 2 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.