Functional Characterization of S. cerevisiae Upstream Activating Factor and its subunits

In the baker’s yeast ‘Saccharomyces cerevisiae’, production of all ribosomal RNAs, except one, relies on transcription by RNA-Polymerase I (Pol I). Transcription of the rDNA gene follows the formation of the Pol I pre-initiation complex (Pol I PIC). Over the last decades, numerous studies have provided insight into structure and function of Pol I, as well as its transcription factors Rrn3, TATA-binding protein (TBP), and Core Factor (CF). The fourth transcription factor – the upstream activating factor ‘UAF’ – has been shown to function as a potent enhancer of Pol I transcription initiation, repressor of RNA-Pol-II transcription, and as a maintainer of a native rDNA copy levels. However, to the day this project was started, structural information about this factor was completely missing. With this project, we aimed at providing additional functional insight and structural information about this important complex in order to contribute to a more comprehensive understanding of transcription initiation by Pol I. During this project, expression and purification protocols for UAF subunits and the whole complex were optimized, followed by functional characterization and crystallization attempts. Functional characterization of subunits Uaf30 and the Rrn9 / Rrn10 subcomplex showed, that both – Uaf30 and Rrn9 – play an important role for the interaction with the upstream element. We show that promoterbinding by Rrn9 is likely unspecific and that it relies on a C terminal DNA binding domain. Uaf30, in contrast, shows a significant preference towards an upstream region of the upstream element around position -100 relative to the transcription start site. From this data, we hypothesize, that promoter targeting by UAF may rely on tethering by Uaf30, while Rrn9 cooperatively binds the promoter further downstream in an unspecific manner to strengthen the interaction between UAF and DNA. To challenge these claims, Rrn9-C and Uaf30 deletion mutants were additionally tested in context of the whole UAF complex. In line with the previous results, deletion of Uaf30 leads to a notable loss of specificity, while deletion of Rrn9-C only exhibits reduced overall affinity. A double deletion mutant combines both loss of affinity and specificity. The impact of both deletions on the stimulation of transcription initiation by Pol I was tested via in vitro transcription assays. Interestingly, deletion of Uaf30 almost completely abolishes the transcription enhancement by UAF when compared to the wildtype complex and ΔC Rrn9 mutant. Contrary to expectations, in the latter, while transcription of the wildtype DNA template is less enhanced compared to the wildtype complex, transcription of the control template without an upstream element was significantly enhanced. With the aim of providing structural data, a variety of constructs were subjected to high throughput crystallization screenings. Although isolation of crystallization grade protein was achieved for many of these constructs, including homologs from the baker’s yeasts thermophilic relative - Chaetomium thermophilum, the bottleneck of actual protein crystallization turned out to be a major limiting factor. Only very few crystals were obtained and measured at a synchrotron. Neither did these crystals diffract, nor could they be reproduced.