In this dissertation we discuss H-NS and its connection to nucleoid compaction and organization. Nucleoid formation involves a dramatic reduction in coil volume of the genomic DNA. Four factors are thought to influence coil volume: supercoiling, DNA charge neutralization, macromolecular crowding and DNA deformation by NAPs. This study focuses mainly on the latter two factors, and on their interplay. We investigate both direct and indirect changes in DNA coil volume as a result of H-NS binding to DNA. H-NS / DNA binding is thought to be influenced by the self-association of H-NS, hence DNA self-association (both in bulk and on DNA) has also been investigated.
Chapter 2 focuses on the known cooperative character of H-NS-DNA binding. The molecular origin of the cooperativity is poorly understood. High concentrations of H-NS are known to oligomerize extensively in the absence of DNA. Some models propose that cooperativity is caused by the same protein-protein interactions that cause oligomerization in solution, whereas others propose cooperativity may be induced by the DNA substrate. We have mutated some parts of H-NS we believed to be important in oligomerization to investigate the role of H-NS protein-protein interactions in cooperative DNA binding, and studied the oligomerization and DNA-binding properties of these mutants. The D68VD71V mutant has two aspartic acids in the linker region replaced by valines, making the linker significantly more hydrophobic. The double linker mutation D68VD71V dramatically enhances H-NS oligomerization in solution, and its temperature-dependence is changed as well in vitro. Yet there is only a moderate effect on DNA binding properties, which does point in the direction of enhanced cooperativity, as expected. This suggests that protein-protein interactions have a much larger effect on H-NS self-association in solution than on the DNA binding properties.
Chapter 3 discusses the influence of the bacterial nucleoid protein H-NS on DNA coil sizes in solution, using Light Scattering, for both supercoiled and linear pUC18 DNA. We clearly find H-NS binding: the intensity of light scattered by the DNA coils increased upon the addition of H-NS. But, H-NS did not have a significant effect on the effective hydrodynamic radius of the coils. Our results suggest that under the conditions of our experiment (in particular the buffer conditions: 10mM Sodium Phosphate buffer, pH 7, 100mM NaCl), the H-NS proteins most likely did not cause extensive bridging of dsDNA, since this most certainly would have led to a significant effect on the DNA coil sizes. This absence of bridging in the absence of multivalent cations is consistent with single molecule DNA force measurements performed for similar buffer conditions by other authors. We also find that, although H-NS alone does not have a dramatic effect on DNA coil sizes in solution (for our solution conditions), it does have an interesting synergetic effect on polymer-induced condensation of DNA. Condensation of H-NS/DNA complexes was measured by their sedimentability in solutions of polyethylene glycol (PEG). In the absence of H-NS the critical concentration of PEG needed to condense DNA is approximately 15%, whereas the critical concentration is remarkably lower, about 3.5%, at near saturation concentrations of H-NS.
Chapter 4 is concerned with the effect of binding of H-NS and macromolecular crowding on nucleoid compaction. An osmotic shock method using ampicillin was used to isolate the Escherichia coli nucleoids intact, disrupting the peptidoglycan layer. These nucleoids were stained with DAPI and photographed using confocal microscopy. This showed a decrease in the volume of the isolated nucleoids when polyethylene glycol (PEG) concentrations became higher. The addition of small amounts of H-NS appeared to enhance the compaction due to macromolecular crowding induced by PEG. Remarkably, in the absence of PEG, H-NS did not affect the compaction of the nucleoids even at higher concentrations. The results are consistent with previous experiments done on DNA-binding proteins HU and Sso7d by other research groups. Therefore, our results confirm a general enhancement of macromolecular crowding effect of cytoplasm by nucleoid-associated proteins binding.
In Chapter 5 we focus on an archaeal NAP. Like bacteria, archaea have NAPs that bend DNA and form extended helical protein-DNA fibers. These do not condense the genomic DNA directly, but some NAPS strongly promote DNA condensation by macromolecular crowding, such as the bacterial HU. Using theoretical arguments, we show that this synergy can be explained by the larger diameter and lower net charge density of protein-covered DNA filaments compared to naked DNA. Therefore the effect should be nearly universal in prokaryotes. We illustrate this general effect by demonstrating that Sso7d, a 7 kDa basic DNA-bending protein from the archaeon Sulfolobus Solfataricus, does not significantly condense DNA by itself, using light-scattering to determine coil volumes. However, the Sso7d-coated DNA fibres are much more susceptible to macromolecular crowding-induced condensation. Clearly, if DNA-bending nucleoid proteins fail to condense DNA in dilute solution, this does not mean that they do not contribute to DNA condensation in the context of the crowded living cell.