PDF, English - main document Restricted access: Repository staff only until 22 July 2025. Login+Download (36MB) | Lizenz: Structural and biochemical studies of ATP-dependent chromatin remodelers from the INO80/SWR1 family by Jungblut, Anna underlies the terms of Creative Commons Attribution-NoDerivatives 4.0

Abstract

The genome of a eukaryotic cell is packaged by histone proteins into nucleosomes. ATP-dependent chromatin remodeling enzymes shape this first level of chromatin organisation and play a fundamental role in regulating access of nuclear factors to DNA. Eukaryotes possess four evolutionary conserved families of chromatin remodelers. The INO80/SWR1 family members assemble into mega-Dalton sized, multi-subunit complexes and configure the nucleosome landscape at gene promoters, origins of replications and sites of DNA damage. In Saccharomyces cerevisisae (S.c.), the 15-subunit INO80 complex translocates DNA through ATP hydrolysis along the histone core leading to positioning of +1 nucleosomes downstream of promoter DNA. Subsequently, the 14-subunit S.c. SWR1 complex exchanges the canonical histone H2A/H2B dimers with the variant H2A.Z/H2B through ATP hydrolysis. H2A.Z containing +1 nucleosomes are proposed to serve as a gateway for gene transcription. Despite of this importance, the precise molecular mechanisms underlying these nucleosome remodeling reactions are poorly understood. In addition, it remains unclear what defines the substrate specificity and the different reaction outcomes of INO80 and SWR1 remodelers which in principle share a common subunit composition including nuclear actin and Actin-related proteins (Arps). To address these questions, I characterized the structural basis of nucleosome remodeling by INO80 and SWR1 complexes by using a combination of cryo-electron microscopy (cryoEM) and biochemistry. In the first part of this thesis, I present the mechanistic basis of hexasome remodeling by the INO80 complex. Hexasomes are non-canonical nucleosomes lacking one copy of H2A/H2B and are a hallmark of actively transcribed genes. My findings demonstrate how INO80 identifies non-canonical DNA and histone marks of hexasomes that result from the absence of H2A/H2B. INO80 is activated by directly sensing an exposed H3/H4 histone interface, operating independently of the H2A/H2B acidic patch. This is enabled through a spin-rotated binding mode of the Ino80 motor domain while its Arp module remains tethered to the now unwrapped linker DNA. These findings shed light on how the absence of H2A/H2B creates altered substrates for chromatin remodeling, thereby creating a different, however unexplored layer of energy-mediated chromatin regulation. In the second part, I present the first structural and biochemical study of SWR1 from Chaetomium thermophilum (C.t.). I identified the subunit composition of C. t. SWR1 and implemented a recombinant expression and purification strategy for both S.c. and C.t. SWR1 that enabled subsequent cryoEM studies. In addition, I investigated the structure of the Arp module by using an integrative structural biology approach building on cryoEM, AI-powered AlphaFold structure predictions and crosslinking mass spectrometry. Thereby, I determined an almost complete model of the Apo SWR1 complex in which the Arp module is flexibly tethered to the core of SWR1. A kink in the Arp module enables histone tail recognition of the Yaf9 subunit, while recognition of linker DNA might occur by switching SWR1 into an open conformation. Histone tail recognition and tethering to linker DNA may specify SWR1 activity at +1 nucleosomes. INO80 and SWR1 share similar architectures, yet drive distinct functional outcomes. My thesis provides a structural framework for understanding the underlying molecular mechanisms.