DNA methylation and histone H3 lysine 9 methylation (H3K9me) are hallmarks of heterochromatin in plants and mammals and are successfully maintained across generations. The biochemical and structural basis for this maintenance is poorly understood. The maintenance DNA methyltransferase from Zea mays, ZMET2, recognizes H3K9me2 via a chromodomain (CD) and bromo-adjacent homology domain (BAH), which flank the catalytic domain. Here, we show that the H3K9me mark allosterically stimulates ZMET2 activity and also substantially increases the specificity of ZMET2 for hemimethylated versus unmethylated DNA. Interestingly, dinucleosomes are the preferred ZMET2 substrate, with DNA methylation targeted primarily to linker DNA. Further, while the CD stabilizes ZMET2 binding, the BAH domain is critical for catalysis and serves as an allosteric regulator of the enzyme. Negative stain electron microscopy shows a single ZMET2 molecule bridging two nucleosomes within a dinucleosome, consistent with the biochemical analyses. We propose a model in which CD-mediated binding and allosteric activation via BAH/H3K9me interactions coupled with nucleosome bridging allows ZMET2 to correctly read the chromatin context and thereby faithfully maintain DNA methylation and reinforce heterochromatic elements across the genome.

Chromatin structure regulates the accessibility of DNA in several crucial nuclear processes, including transcription, replication and recombination. The ISWI-family of chromatin remodeling complexes translationally repositions histone octamers on DNA. One member of this family, the ATP-dependent chromatin assembly factor (ACF), generates regularly spaced nucleosomes in vitro and is required for transcriptional silencing in vivo.

To dissect ACF mechanism, we used a FRET based approach to follow the movement of DNA with respect to the histone octamer in real time. We find that ACF generates at least one transient intermediate in which part of the DNA has moved across the histone octamer. The rate of ACF remodeling is sensitive to the length of the linker DNA such that shortening the linker DNA below 60 bp progressively slows down both ATP hydrolysis and remodeling. As a result, ACF generates and maintains nucleosome spacing by constantly moving nucleosomes towards longer flanking DNA faster than shorter flanking DNA. ACF thus creates a dynamic equilibrium between different nucleosome positions such that the centered position is most highly populated.

To understand how ACF can rapidly sample both sides of a nucleosome to accomplish bidirectional movement, my collaborators and I employed a myriad of techniques to conclude that ACF functions as a dimeric molecular motor. We found that ACF binds nucleosomes in a cooperative manner, and that this oligomeric state is required for back and forth remodeling of the nucleosome. Additionally, the ATPase subunit energetically couples different nucleotide states with distinct reaction intermediates, and thus we observe that the physical interactions with the nucleosome change as a function of the ATP reaction cycle. Lastly, we found that ACF utilizes ATP in three distinct ways, and that translocation is characterized by distinct pauses during movement of DNA relative to the histone octamer.

In summary, these data provide a mechanistic basis for the directionality of nucleosome repositioning and explain the centering activity of ACF. Futhermore, the dimeric architecture of ACF allows for the rapid and processive bidirectional nucleosome movement required for nucleosome spacing.

The primary unit of eukaryotic DNA packaging is a nucleosome, which contains ~150 bp of DNA wrapped around an octamer of histone proteins. This packaging restricts accessibility for transcription, replication, repair, and recombination. ATP-dependent chromatin remodeling complexes play a key role in regulating this accessibility. The catalytic subunits of these complexes use ATP, to alter histone-DNA and histone-histone interactions. Interestingly, the ATPase subunits alone carry out most of the remodeling activities of the entire complexes. This raises the question, of what are the functions of the non-catalytic subunits. Two hypotheses are: (i) non-catalytic subunits enhance the biochemical activity of the catalytic subunit, and (ii) non-catalytic subunits facilitate targeted localization.

In Chapter 2, mechanistic roles of the non-catalytic subunits were elucidated. Comparing the hSWI/SNF complex with Brg1, its ATPase subunit, revealed that the non-catalytic subunits do not simply promote an active conformation of Brg1, but rather, enhance discrete steps in the remodeling reaction. Minimal complexes containing, Baf155, Baf170, and Ini1 were sufficient to lower the KM for nucleosomes, suggesting that these non-catalytic subunits help stabilize nucleosome binding. Overall, the data reveal that non-catalytic subunits have important mechanistic roles, including distinguishing between nucleotide states.

In Chapter 3, other roles of non-catalytic subunits were identified, following a novel acetylation on the RSC complex, the yeast homolog of hSWI/SNF. Rsc4, the acetylated subunit, contains two bromodomains (BDs). BD1 has been previously shown to bind the acetylation mark when BD2 is not bound to acetylated H3K14 tails. However, we find that acetylation of Rsc4 does not modulate RSC's ability to recognize H3-acetylated nucleosomes. Instead, we found that high Rsc4 acetylation levels as maintained by BD1, are critical for resistance to DNA damage, especially in the absence of another remodeler, INO80. Further, cells lacking Rsc4 acetylation and NHP10, a unique INO80 subunit, have delayed S-phase progression and HU sensitivity, revealing a role for the RSC complex and its acetylation in resistance to replication stress. This study suggests that non-catalytic subunits can specify remodeler participation in cellular processes, distinct from enzymatic functions.

Eukaryotic genomes are packaged into chromatin: a highly heterogeneous structure composed of nucleic acids and proteins. This packaging controls access to the underlying DNA sequences and, as a result plays a critical role in nearly all genomic processes. The primary molecular structure of chromatin is the nucleosome: ~147 bp of DNA wrapped around a core of histone proteins. Both the location and status of nucleosomes in the genome are critical for the proper packaging of chromatin. As a result, cells have evolved several sophisticated molecular machines to disrupt or modify nucleosomes to achieve specific packaging states. A critical member of these machines are the SWI2/SNF2 superfamily of ATP-dependent chromatin remodeling enzymes, which are DNA translocases that harness the energy of ATP hydrolysis to physically disrupt nucleosomes. Because of their central role in control nucleosome structure throughout the genome, remodelers play roles in virtually all DNA-dependent process, but the precise mechanisms of how remodelers disrupt nucleosomes and how this disruption is coupled to other molecular events remains very poorly understood. In this thesis we focus on understanding the remodeling mechanisms of two subfamilies of SWI2/SNF2 remodelers that slide nucleosomes: INO80 and ISWI. To understand how these and other SWI2/SNF2 ATP-dependent remodelers might cooperate with nuclear machinery to enable biological processes, we first review our broad understanding of remodeling mechanism as it compares to another molecular motor that disrupts nucleosomes: RNA polymerase. We then speculate on how these two distinct families cooperate to accomplish transcription on chromatin templates. After this, we set out to uncover elements of nucleosome that control remodeler activity and identify a conserved surface of the nucleosome known as the acidic patch that is required to activate both ISWI and INO80 family remodelers. Using a combination of biochemical and biophysical assays, we show that this surface activates remodeling by these two families after they bind the nucleosome. For the ISWI remodeler SNF2h, the acidic patch activates remodeling by serving as a landing pad for the binding of autoinhibitory domains while INO80 uses a separate mechanism. We then solve the near-atomic CryoEM structure of SNF2h bound to the nucleosome. Unexpectedly, we find that SNF2h binding in an activated state asymmetrically distorts the histone core of the nucleosome and that this may be important in regulating the activity of the enzyme. Finally, we test the hypothesis that by measuring remodeling activity of nucleosomes with site-specific restraints in the histone core. We find that specifically restraining histone dynamics in locations across all 4 histone proteins inhibits SNF2h-mediated nucleosome sliding. Taken together, these results suggest that remodelers rely on the structure and dynamics of both the DNA and protein components of the nucleosome to accomplish their activities.

The fundamental basis of gene regulation is the physical accessibility of genetic material to downstream machinery that convert DNA into proteins. In eukaryotes, accessibility to the genome is ultimately regulated by the structure of the nucleosome—the fundamental unit of chromatin that is comprised of 147 base pairs of DNA tightly wrapped around a core of eight histone proteins. Of the many factors that affect nucleosome structure, ATP-dependent chromatin remodeling enzymes are unique in their ability to convert the chemical energy of ATP into physical changes in the nucleosome that would otherwise be energetically disfavorable. Of these remodelers, the INO80 remodeling enzyme is a 15-subunit complex from yeast that is unique both in its modular structure and its critical roles at sites of genomic instability such as double strand breaks and replication forks. However, very little is known about how the INO80 machinery is assembled or how it achieves specificity in nucleosome movement. In the first half of this thesis, we explore the potential roles of Rvb1 and Rvb2 (Rvbs), two poorly understood AAA+ ATPases that are integral components of the INO80 complex, as well as other multi-subunit complexes in the cell. While the Rvbs were traditionally thought of as DNA helicases, we have new evidence for their role as protein chaperones. We discover the first protein client of Rvbs, a small insertion region of the Ino80 ATPase (Ino80INS), which robustly stimulates the ATPase activity of the Rvbs while promoting dodecamerization of the normally hexameric Rvbs. Using a combination of structural methods including crosslinking mass spectrometry, native mass spectrometry, cryo-EM and integrative modeling, we find that two Ino80INS molecules bind asymmetrically and adjacent to one another along the dodecamerization interface, resulting in a conformationally flexible dodecamer that collapses into hexamers upon ATP addition. Our results demonstrate the chaperone-like potential of Rvb1/Rvb2 and suggest a model where binding of multiple clients like Ino80 stimulates ATP-driven cycling between hexamers and dodecamers, providing iterative opportunities for correct subunit assembly. In the second half of the thesis, we explore the molecular mechanism of nucleosome movement by INO80. While it is known that INO80 can center nucleosomes on a short piece of DNA, it is unknown how the motor accomplishes this task. Using a combination of enzymology and single molecule FRET, we find that INO80 exhibits a switch-like response to flanking DNA length, largely regulated by subcomplex of INO80 called the Nhp10 module. We also find that remodeling by INO80 is extremely processive: once sliding is initiated, INO80 moves the nucleosome rapidly at least 20 bp without re-assessing flanking DNA length, and can change the direction of nucleosome sliding without dissociation. Taken together, these findings provide unprecedented insight into INO80’s lifetime in the nucleus: from its assembly by Rvb1/Rvb2 to its action on chromatin.

Chromatin remodelers are molecular motors that reposition and restructure nucleosomes to regulate gene expression and genome organization. The human ATP-dependent chromatin assembly factor (ACF) is a dimeric chromatin remodeler that evenly spaces nucleosomes to promote the formation of silent heterochromatin. Two copies of its ATPase subunit, SNF2h, bind opposite sides of a nucleosome, but how the two protomers work together remains poorly understood.

Central to the nucleosome spacing activity of ACF is the ability of SNF2h to equalize the length of DNA flanking either side of a nucleosome. SNF2h senses flanking DNA length via its conserved HAND-SANT-SLIDE (HSS) domain, yet it is unclear 1) how this interaction contributes to remodeling, and 2) whether or how the HSS communicates with the ATPase domain.

In this work, we address these questions with a combination of biochemical, biophysical and protein engineering methods. Using covalently connected SNF2h dimers we show that dimerization accelerates remodeling and that the HSS contributes to communication between protomers in trans. Moreover, we find that the state of the ATPase domain influences the function of the HSS domain within the same protomer in cis. Using thermodynamic analyses and new FRET-based assays, we further examine how nucleotide state regulates interactions between the HSS domain and the nucleosome. We identify a large, nucleotide-dependent conformational change in SNF2h: in one conformation the HSS binds flanking DNA, and in another conformation the HSS retracts to engage the nucleosome core. Finally, we examine the role of a recently identified regulatory motif, termed NegC, in nucleosome remodeling by SNF2h. We find that mutating NegC impairs flanking DNA length sensing, abolishes mononucleosome centering, and reduces the dependence of SNF2h activity on the histone H4 tail.

Based on these new results and previously published work, we propose a model in which DNA length sensing and translocation are performed by two distinct conformational states of SNF2h. We speculate that the ATP-dependent switch between DNA length-sensing and translocation-competent forms of the enzyme represents a rate-limiting regulatory step that enables nucleosome spacing by ACF. Such separation of function suggests that these two activities could be independently regulated to fine-tune remodeling outcomes.