It is well known that multivalent cations cause free DNA in solution to condense into nanometer-scale particles with toroidal and rod-like morphologies. However, it has not been shown to what degree kinetic factors (e.g., condensate nucleation) versus thermodynamic factors (e.g., DNA bending energy) determine experimentally observed relative populations of toroids and rods. It is also not clear how multimolecular DNA toroids and rods interconvert in solution. We have conducted a series of condensation studies in which DNA condensate morphology statistics were measured as a function of time and DNA structure. Here, we show that in a typical in vitro DNA condensation reaction, the relative rod population 2 min after the initiation of condensation is substantially greater than that measured after morphological equilibrium is reached (ca. 20 min). This higher population of rods at earlier time points is consistent with theoretical studies that have suggested a favorable kinetic pathway for rod nucleation. By using static DNA loops to alter the kinetics and thermodynamics of condensation, we further demonstrate that reported increases in rod populations associated with decreasing DNA length are primarily due to a change in the thermodynamics of DNA condensation, rather than a change in the kinetics of condensate nucleation or growth. The results presented also reveal that the redistribution of DNA from rods to toroids is mediated through the exchange of DNA strands with solution.

The condensation of nucleic acids into well-defined particles is an integral part of several approaches to artificial cellular delivery. Improvements in the efficiency of nucleic acid delivery in vivo are important for the development of DNA- and RNA-based therapeutics. Presently, most efforts to improve the condensation and delivery of nucleic acids have focused on the synthesis of novel condensing agents. However, short oligonucleotides are not as easy to condense into well-defined particles as gene-length DNA polymers and present particular challenges for discrete particle formation. We describe a novel strategy for improving the condensation and packaging of oligonucleotides that is based on the self-organization of half-sliding complementary oligonucleotides into long duplexes (ca. 2 kb). These non-covalent assemblies possess single-stranded nicks or single-stranded gaps at regular intervals along the duplex backbones. The condensation behavior of nicked- and gapped-DNA duplexes was investigated using several cationic condensing agents. Transmission electron microscopy and light-scattering studies reveal that these DNA duplexes condense much more readily than short duplex oligonucleotides (i.e. 21 bp), and more easily than a 3 kb plasmid DNA. The polymeric condensing agents, poly-l-lysine and polyethylenimine, form condensates with nicked- and gapped-DNA that are significantly smaller than condensates formed by the 3 kb plasmid DNA. These results demonstrate the ability for DNA structure and topology to alter nucleic acid condensation and suggest the potential for the use of this form of DNA in the design of vectors for oligonucleotide and gene delivery. The results presented here also provide new insights into the role of DNA flexibility in condensate formation.

Integration host factor (IHF), a nucleoid-associated protein in bacterial cells, is implicated in a number of chromosomal functions including DNA compaction. IHF binds to all duplex DNA with micromolar affinity and at sequence-specific sites with much higher affinity. IHF is known to induce sharp bends in the helical axis of DNA in both modes of binding, but the role of IHF in controlling DNA condensation within bacterial cells has remained undetermined. Here we demonstrate that IHF influences the morphology of DNA condensed by polyamines in vitro. In the absence of IHF, spermidine and spermine condense DNA primarily into toroidal structures, whereas in the presence of IHF, polyamines condense DNA primarily into rodlike structures. Computer simulations of DNA condensation in the absence and presence of IHF binding lend support to our model in which DNA bending proteins, such as IHF and HU, promote the condensation of DNA into rodlike structures by providing the free energy necessary to bend DNA at the ends of linear bundles of condensed DNA. We propose that a common function of IHF and HU in bacterial cells is to facilitate DNA organization in the nucleoid by the introduction of sharp bends in chromosomal DNA.

Controlling the size and shape of DNA condensates is important in vivo and for the improvement of nonviral gene delivery. Here, we demonstrate that the morphology of DNA condensates, formed under a variety of conditions, is shifted completely from toroids to rods if the bacterial protein HU is present during condensation. HU is a non-sequence-specific DNA binding protein that sharply bends DNA, but alone does not condense DNA into densely packed particles. Less than one HU dimer per 225 bp of DNA is sufficient to completely control condensate morphology when DNA is condensed by spermidine. We propose that rods are favored in the presence of HU because rods contain sharply bent DNA, whereas toroids contain only smoothly bent DNA. The results presented illustrate the utility of naturally derived proteins for controlling the shape of DNA condensates formed in vitro. HU is a highly conserved protein in bacteria that is implicated in the compaction and shaping of nucleoid structure. However, the exact role of HU in chromosome compaction is not well understood. Our demonstration that HU governs DNA condensation in vitro also suggests a mechanism by which HU could act as an architectural protein for bacterial chromosome compaction and organization in vivo.