Mechanistical studies of two functions of HMGB1 protein : facilitation of nucleosome sliding and translocation from the nucleus to the extracellular medium

Bonaldi, Tiziana (2004). Mechanistical studies of two functions of HMGB1 protein : facilitation of nucleosome sliding and translocation from the nucleus to the extracellular medium. PhD thesis The Open University.

DOI: https://doi.org/10.21954/ou.ro.0000f6f0

Abstract

This thesis develops two lines of investigation that focus on HMGBl protein from different points of view. In Section 3, acétylation of lysine is identified as the molecular switch that controls movement of HMGBl from the nucleus to the cytoplasm and Section 4 investigates its role in chromatin remodelling. High Mobility Group protein HMGBl is a chromatin component that, when leaked out by necrotic cells, triggers inflammation. HMGBl can also be secreted by activated monocytes and macrophages and fimctions as a late mediator of inflammation. Secretion of a nuclear protein must require a tightly controlled relocation programme. It was found that in all cells HMGBl shuttles actively between nucleus and cytoplasm due to two nuclear localisation signals (NLSs) and two nuclear export signals (NESs) within its sequence. Analysis of HMGBl samples, extracted under specific conditions, using iso-electric focussing/SDS 2- dimensional gels, followed by mass spectrometry, demonstrated that HMGBl is extensively acetylated (up to 10 modification sites per molecule) and that acétylation of the two NLSs inhibits nuclear import, causing cytoplasmic accumulation of the protein. This provides the first step in directing the protein to the secretion pathway, since cytosolic HMGBl is then concentrated by default into secretory lysosomes and finally secreted when monocytic cells receive an appropriate second signal. Multiple acétylation of lysine sidechains was thereby defined as the molecular switch redirecting the nuclear protein HMGBl to the cytoplasm and subsequently to secretion.
The second project investigated the potential role of HMGBl in nucleosome sliding. Nucleosome remodelling complexes containing the ATP-ase ISWl, such as ACF, contribute to chromatin remodelling by converting chemical energy into the sliding of nucleosomes on DNA. ISWl interacts with DNA at the sites of its entry into the nucleosome, where it alters histone/DNA interactions that may lead to the relocation of DNA relative to the associated histone octamer. This work elaborated on this concept: if the rate-limiting step in nucleosome sliding is the distortion of linker DNA, a protein that can generate and/or stabilise such distortions might facilitate sliding. HMGBl is able to transiently bend DNA, so is a good candidate to help “lubricate” nucleosome sliding. Using gel-shift assays, it was found that transient interaction of HMGBl with nucleosomal DNA, at sites overlapping with ISWl binding sites, enhanced ACF-induced sliding. The ACFl subunit of ACF was required to render the remodelling process sensitive to HMGBl action since no enhancement was detected when the sliding assay was performed using ISWl alone. In contrast, an HMGBl derivative lacking the acidic tail that interacts with chromatin in a more dynamic way, had a strong inhibitory effect on sliding. These data suggest that HMGBl is able to increase chromatin ‘fluidity’ by generating strategic DNA bends, or ‘bulges’, which are profitably used by ACF to induce nucleosome sliding. Moreover they support a “local loop” model of sliding and identify HMGBl as a potential regulator of ATP-dependent nucleosome remodelling processes. This work was published in the EMBO Journal on 16th December 2002.