Single-Base Resolution Sequence-Directed Nucleosome Mapping

Nucleosome positioning along the DNA sequence is often viewed as a combination of two distinct aspects – translational positioning and rotational positioning. With low-resolution ( 10–20 bases) mapping techniques such as MNase digestion of chromatin, only the approximate translational position of the nucleosomes is determined. Rotational positioning can be evaluated only when the outward/inward orientations of base pairs are specified, on the surface of the histone octamers. Translational positioning with one-base accuracy would provide orientations of the bases with ~348 uncertainty. Accurate nucleosome positioning is thus reduced to determination of the central base of the nucleosome DNA, which provides both translational and rotational positioning. Experimental derivation of nucleosome positions with such accuracy is problematic. In a recent study, a new technique offering single-base resolution was suggested based on the chemical modification of engineered histones. This technique was applied to the whole genome of yeast, although it remains to be seen how well it would perform in specific cases of detailed nucleosome positioning. The original version of the technique failed to reveal alternative nucleosome positions separated by ~10 bases. The highest-accuracy mapping of nucleosome DNA is achieved by crystallization of the nucleosomes reconstituted on specific sequences. It is limited only by the atomic resolution of the x-ray structures, giving a rotational error on the order of only a few degrees. The bases located on the exterior of the DNA in the nucleosome are accessible for interactions with various regulatory proteins, while those bases located on the interface between DNA and histones are inaccessible for these interactions. The crucial dependence of such interactions on the rotational setting of DNA in the nucleosome is well documented. The rotational orientation of the TATA box strongly influences the TATA-binding protein (TBP) binding, and the same is observed for thyroid hormone response element, glucocorticoid response element, transcription factor NF1, and p53 factor. In the last three cases the switch from the inaccessible to the accessible condition is achieved by changing the rotational setting of DNA by 1808. High-resolution nucleosome mapping routines are thus indispensable for the detailed structural analysis of the regulation of gene expression and of other processes involving the nucleosomes. The first computational procedure that provided highresolution mapping was suggested three decades ago. It was based on the 10–11 base periodicity of the chromatin DNA sequences, with the repeating pattern (xAAAxxTTTx)n. [12] The nucleosome maps appeared as clusters of peaks, separated by 10–11 bases. Each peak corresponded to one of several alternative translational nucleosome positions, each with the same rotational setAbstract : Frequently used nucleosome mapping techniques, both experimental and computational ones, pursue a rather simple task – to determine whether a given sequence segment is likely to belong to a nucleosome, thus measuring the nucleosome occupancy along the sequence. A more ambitious task is to determine the position with high resolution, so that not only the approximate translational position of the nucleosome on DNA would be known, but also the rotational setting of the DNA. The rotational setting is important to know since the binding of various transcription factors to the nucleosome DNA crucially depends on the accessibility of the respective recognition sequences. The only experimental technique that provides the highest possible accuracy of the positioning is crystallization of the nucleosomes reconstituted on specific sequences, with subsequent solving of their structures from x-ray diffraction data. Two computational approaches to the accurate positioning of nucleosomes, based on energy calculations and on sequence pattern directed mapping, are currently in different stages of progress.