Session A46: Invited Session: DNA Physics and Chromatin Organization

Dynamics of the Competition Between Nucleosome Unwrapping and DNA Binding Proteins

In eukaryotic organisms DNA is tightly wrapped into nucleosomes. This bears the question how this DNA can be accessed in order to be copied, transcribed, or repaired. A process that allows access to the DNA is transient unwrapping of the DNA from the histone proteins. We have developed a quantitative model of this unwrapping process which we calibrate by comparison to nucleosome unzipping experiments by the Wang group. We then apply this model to quantitatively explain the dynamics of transcription factor binding within nucleosomal DNA. In this context, it has been well known that nucleosomes reduce the affinity for transcription factors to binding sites covered by the nucleosome. It has been assumed that this is due to a reduction in on-rate since a transcription factor can only bind when a rare thermal fluctuation of the nucleosome makes the DNA accessible. However, recent experimental data surprisingly shows that the off-rate of transcription factors is also strongly affected in the presence of a nucleosome. The application of our nucleosome unwrapping free energy landscape demonstrates that this increase in off-rate by several orders of magnitude is a consequence of a competition between partial binding events of dimeric transcription factors and the nucleosome.   

Hydrodynamics

The talk will present a hydrodynamic description of large-scale cooperative movement of chromatin that have been observed by particle tracking methods. The results of the hydrodynamic description will be compared with the tracking data. Passive thermal fluctuations and active ``scalar'' events - associated with local chromatin condensation - are found to dominate cooperative motion at shorter length scales while active ``vector events'' - associated with chromatin remodeling - driving transverse hydrodynamic modes dominate at large length scales.   

A physical analysis of nucleosome positioning

The first level of genome packaging in eukaryotic cells involves the formation of dense nucleosome arrays, with DNA coverage near 90\% in yeasts. A high nucleosome coverage is essential for cells, e.g. to prevent cryptic transcription, and the local positions of specific nucleosomes can play an important role in gene regulation. It is known that in vivo nucleosome positions are affected by a complex mix of passive and active mechanisms, including sequence-specific histone-DNA binding, nucleosome-nucleosome interactions, ATP-dependent remodeling enzymes, transcription, and DNA replication. Yet, the statistical distribution of nucleosome positions is extremely well described by simple physical models that treat the chromatin fiber as an interacting one-dimensional gas. I will discuss how can we interpret this surprising observation from a mechanistic perspective. I will also discuss the kinetics of the interacting gas model, which is pertinent to the question of how cells achieve the high nucleosome coverage within a short time, e.g. after DNA replication.   

DNA-protein recognition and sequence-dependent variations of DNA conformational properties

Parameters of B-DNA, the major form of the double helix, depend on its sequence. This dependence can contribute to the recognition of specific DNA sequences by proteins. Here we try to analyze this contribution quantitatively. In the first approach to this goal we used experimental data on the sequence dependence of DNA bending rigidity and its helical repeat. The solution data on these parameters of B-DNA were derived from the experiments on cyclization of short DNA fragments with specially designed sequences. The data allowed calculating the sequence variations of DNA bending energy, as well as the variations of the energy of torsional deformation of the double helix associated with a protein binding. The results show that DNA conformational parameters can have very limited influence on the sequence specificity of protein binding. In the second approach we analyzed the experimental data on the binding affinity of the nucleosome core with DNA fragments of different sequences. The conclusions derived in these two approaches are in a good agreement with one another.   

Polymer Models of Looping in Genome

Regions of the chromosome that are many thousands of bases away often have to come in contact for turning some genes on or off. To discover the specific and highly regulated contacts, we need to have a good statistical description of the strength of non-specific ones. We model the chromatin fiber as a potentially heterogeneous polymer and discuss how to use such a model in the analysis of chromatin conformation capture data.