Most of the chromosomal DNA is packaged into chromatin. The modulation of chromatin structure is essential to cellular process such as transcription activation and repression, replication, and DNA repair. Alteration in chromatin structure can be achieved through covalent or non-covalent modifications of histories. Non-covalent chromatin modifications consist of mechanisms that require ATP-hydrolysis and ones that do not. ATP-dependent chromatin remodeling is performed by ATPase-containing complexes of the SNF2 super family. There are three major sub-families of remodeling complexes: SWI/SNF, CHD, and ISWI. The diversity of ISWI-based complexes is conserved in organisms. For example, in human the hACF complex consists of hACF1 and human ISWI SNF2h; the CHRAC consists of hACF1, SNF2h, and the p15/p17 dimer; the WICH complex consists of WSTF and SNF2h, and the RSF complex consists of SNF2h and Rsf1. ISWI-based complexes have divergent in vitro as well as in vivo activities. It has been hypothesized that the binding partners regulate ISWI activity differently, therefore resulting in varying in vivo functions of ISWI-based complexes.

The goal of the work described in this dissertation is to investigate if and how the different non-catalytic subunits regulate the activity of the human ISWI protein, SNF2h, in hACF, CHRAC, WICH, and RSF complexes. First the chromatin remodeling parameters of SNF2h were defined. Subsequent comparisons of the remodeling behaviors of the SNF2h-based complexes were made. The remodeling activities of hACF and SNF2h are different in substrate requirement and preference, as well as in the interaction between remodeler and nucleosome. This suggests that the non-catalytic subunit hACF1 alters the remodeling behavior of SNF2h. Further investigation shows that the remodeling behaviors of the CHRAC, WICH and RSF complexes differ from hACF remodeling. CHRAC remodeling requires a shorter length of nucleosomal DNA overhang than hACF, but is similar to hACF in its preference for longer templates. This suggests that the presence of the p15/p17 dimer in CHRAC confers additional regulation on SNF2h activity. In addition, WICH remodeling requires the same length of nucleosomal DNA overhang as CHRAC, but shows reduced preference for longer nucleosomal templates. This suggests that the regulation of SNF2h by WSTF is different from those conferred by hACF and CHRAC subunits. Finally, RSF remodeling resembles that of SNF2h, implying that the regulation of SNF2h remodeling by Rsf1 is different from other complexes. My findings provide evidence in support of the hypothesis that different subunits regulate ISWI activity differently. Moreover, the varied regulation of SNF2h in SNF2h-based complexes might shed light on the different in vivo functions of the complexes.