In Vitro Studies of Nucleosome Positioning and Stability at the PHO5 and PHO8 Promoters in Saccharomyces cerevisiae

The PHO5 and PHO8 genes in yeast provide typical examples for the role of chromatin in promoter regulation. Both genes are regulated by the same transcriptional activator, Pho4, which initiates nucleosome remodeling and transcriptional activation. In spite of this co-regulation, there are important differences in gene activity and in the way promoter chromatin undergoes chromatin remodeling. First, PHO5 belongs to one of the most strongly induced genes in yeast being 10-fold more active than the PHO8 gene (Oshima, 1997; Barbaric et al., 1992). Second, chromatin remodeling at the PHO5 promoter affects four nucleosomes (Almer et al., 1986), whereas only two nucleosomes are afffected at the PHO8 promoter (Barbaric et al., 1992). Third, neither the histone acetyl transferase Gcn5 nor chromatin remodeling complex Swi/Snf seem to be critically required for chromatin remodeling at the PHO5 promoter (Barbaric et al., 2001; Reinke and Hörz, 2003; Dhasarathy and Kladde, 2005; Neef and Kladde, 2003). At the PHO8 promoter, on the other hand, absence of Swi/Snf results in the complete loss of chromatin remodeling under inducing conditions. Furthermore, Gcn5 is required for full remodeling and transcriptional activation at this promoter (Gregory et al., 1999). Ever since these differences were recognized there have been speculations about the underlying reasons. This work shows that these discrepancies are not a direct consequence of the position or strength of the UASp elements driving the activation of transcription. Instead, these differences result from different stabilities of the two promoter chromatin structures. The basis for these results was the development of a competitive yeast in vitro assembly technique in which differences in nucleosome stability between promoter regions could be directly compared. This technique originated from a yeast in vitro chromatin assembly system that generated the characteristic PHO5 promoter chromatin structre (Korber and Hörz, 2004). As shown here, this system also assembles the native PHO8 promoter nucleosome pattern. Using the competitive assembly system it was shown that the PHO8 promoter has greater nucleosome positioning power, and that the properly positioned nucleosomes are more stable than at the PHO5 promoter. This provided for the first time evidence for the correlation of inherently more stable chromatin with stricter co-factor requirements. Remarkably, the positioning information for the in vitro assembly of the native PHO5 and PHO8 promoter chromatin patterns was specific to the yeast extract. Salt gradient dialysis or Drosophila embryo extract assemblies did not support the proper nucleosome positioning. However, nucleosomes in chromatin generated in these systems could be shifted to their in vivo-like positions by the addition of yeast extract. This indicates that the nucleosome positioning mechanisms in vitro are uncoupled from the nucleosome loading machinery. The nucleosome positioning at the PHO5 and PHO8 promoters was energy dependent suggesting a role of chromatin remodeling machines in generation of the repressed promoter chromatin structure. In spite of this, the chromatin remodeling machines Swi/Snf, Isw1, Isw2 and Chd1 were dispensable nucleosome positioning at both promoters.