UCSF

The eukaryotic genome is organized and regulated in the form of nucleosomes - the fundamental unit of chromatin composed of ~147 base pairs of DNA wrapped around a histone octamer core. Accessing the DNA wrapped around nucleosomes is central to all DNA-based processes. A class of enzymes known as ATP-dependent chromatin remodelers uses the energy from ATP hydrolysis to catalyze transformations to the nucleosome that alter DNA accessibility. This thesis explores the unique mechanisms of INO80, a 15-subunit remodeler that transforms the nucleosome by sliding a histone octamer core along DNA. INO80 has been linked to various biological processes including DNA damage repair, replication, and transcription. How INO80 uses its biochemical activity to influence nucleosome organization at these various genomic loci remains an open question. This thesis will use biochemical and biophysical techniques to explore INO80's mechanism, specifically its (1) hexasome sliding activity and (2) assembly. In Chapter 2, we explore INO80's activity on hexasomes, subnucleosomal particles missing an H2A-H2B dimer that can form, for example in the wake of transcription. Prior to this work, our lab showed that INO80 prefers to slide hexasomes over nucleosomes. Here, we explore this mechanism further by using a combination of cryo-EM and biochemistry using (sub)nucleosomal substrates with DNA containing single base gaps. We solved high-resolution structures of S. cerevisiae INO80 bound to a hexasome and nucleosome. Surprisingly, we found that INO80 reorients to position its core Ino80 ATPase subunit at superhelical location (SHL) -2 on a hexasome compared to engaging with the nucleosome at SHL-6. This nearly 180º rotation positions INO80 with dramatically different interactions between its core subunits and the nucleosomal DNA and histones. Interestingly, we show that single base gaps, which block DNA translocation, near SHL-2 inhibit INO80's activity on both hexasomes and nucleosomes. We propose a model where INO80 slides all (sub)nucleosomal substrates most efficiently from SHL-2 with INO80 first rotating on a nucleosome before sliding. This model could explain why INO80 slides hexasomes faster than nucleosomes. In Chapter 3, we explored INO80's assembly mechanism. Prior work from our lab showed that Rvb1 and Rvb2, two AAA+ ATPase subunits of INO80, display chaperone-like activity where they recognize a unique insertion domain of the core Ino80 ATPase, Ino80INS, as a protein client. This suggested that the Rvbs may play a role in INO80 assembly. Here, we explored this role further by developing a co-expression system of Rvb1, Rvb2, and a truncated form of Ino80 containing its core ATPase and insertion domains. We show that this complex is functional for ATP hydrolysis, yet the core ATPase domain of Ino80 appears to inhibit the stimulatory effect of Ino80INS alone. In addition, we show that other INO80 core subunits do not stimulate Rvb hydrolysis and we explore conformational changes of the Rvbs upon interactions with clients. Overall, this thesis explores two aspects of INO80's mechanism that add to the overall understanding of a versatile chromatin remodeler.