The effects of the linker-DNA on linker histone positioning: a single-pair FRET study from mono to trichromatosomes

Abstract

The nucleosome consists of a core complex of two copies each of four histone proteins wrapped by about 1.65 turns of DNA. The DNA arms entering and leaving the core are known as linker-DNA (L-DNA) arms. The linker histone, H1, associates with the nucleosome at the region bounded by the two ends of the DNA leaving the core. The nucleosome in conjunction with the H1 forms the chromatosome, the smallest repeating unit of the chromatin. How does H1 associate with the nucleosome? Are there any contributions from the L-DNA stretches flanking the core to H1 association? Does the length or the sequence of the L-DNA affect the way H1 associates with the nucleosome? Does the binding mode of H1 affect the higher-order structuring of the chromatin? The work presented in this thesis aims to address these questions. Chromatosomes were reconstituted from core histone octamers (Xenopus laevis), linker histone of subtype H1.0b (X. laevis), and 226 bp DNA containing the strongly positioning Widom 601 sequence. The linker histone was labelled with the fluorophore Alexa 488 on the globular domain or the upstream end of the C-terminal tail domain (CTD), and the DNA was labelled with Alexa 594 on either one or the other L-DNA arm. Single-pair FRET (Forster Resonance Energy Transfer) spectroscopy was used to measure the proximity of the globular domain or the CTD to one or the other L-DNA. First, it was determined how the length and the sequences of the L-DNA flanking the H1 affect the location of the H1 on the nucleosome. It was observed that the structured globular domain of the H1 was proximal to L-DNA arms that contained an A-tract consisting of eleven contiguous adenines. However, the globular domain was not proximal to either a purely GC tract or a mixed sequence having a 64% AT content. The fluorophore on the disordered CTD was equidistant from both the L-DNAs, despite sequence variations. Distances extracted from the single-pair FRET measurements were used to build models of A-tract-containing nucleosomes. It was observed that the globular domain of the H1 associated in an on-dyad fashion on the nucleosomes, and two conserved arginine residues on the globular domain of H1 were proximal to the characteristically narrow minor groove of the A-tract. To experimentally check whether the A-tract recognition was mediated via the minor grooves as observed in the models, or via hydrophobic interactions with the thymine methyl groups, a nucleosome was reconstituted with two A-tracts, one complementary to thymine and the other complementary to methyl-group-lacking deoxy-uridine. The globular domain showed similar proximity to both the A-tracts, proving the role of the A-tract minor groove in its recognition.  Finally, I studied how linker histones compact two types of trinucleosomes containing A-tracts flanking the first and the third nucleosome either on the inner or the outer L-DNA. The extent of compaction was measured by single-pair FRET between the two inner L-DNAs, one joining the first and second, and the other joining the second and third nucleosomes. Based on the mononucleosome models, linker histones associating with the first and the third nucleosome are expected to be oriented towards the A-tracts on the outer or the inner L-DNAs. Both the trinucleosome types were highly compacted in the presence of linker histones. However, there was no difference in the extent of compaction between the two types of trinucleosomes. Overall, this thesis shows a new mode of DNA sequence recognition by the linker histone that may affect the compaction of AT- and A-tract-rich heterochromatin. However, trinucleosome compaction by linker histones was not observed to be affected by the location of A-tracts.