19th-banner-rev.gif

Book of Abstracts: Albany 2009

category image Albany 2009
Conversation 16
June 16-20 2009
© Adenine Press (2008)

The Dynamic Chromatin Structure of Transcriptionally Active Yeast Genes

A number of years ago, inspired by the work of Dr. Bob Simpson [1], we developed a model system in yeast to study the events that occur when a gene is activated for transcription [2, 3]. This involves the purification from yeast cells of native plasmid chromatin containing a model gene expressed at basal or activated levels. In essence, we isolate a gene in its basal or transcriptionally activated native chromatin structure. We have compared these chromatin structures using methods originally developed to elucidate reconstituted chromatin structures. Our studies of two model genes, CUP1 and HIS3, have revealed that activation correlates with movements of nucleosomes and remodelling of nucleosomes over the entire gene, not just at the promoter [4-6].

HIS3 encodes an enzyme required for histidine metabolism and is induced by amino acid starvation. HIS3 is activated by Gcn4p and regulated by the Gcn5p histone acetyltransferase (HAT) in the SAGA complex and the Esa1p HAT in the NuA4 complex, as well as by the SWI/SNF and RSC ATP-dependent remodelling machines [7, 8]. We have demonstrated that HIS3 plasmid chromatin exists in two alternative structural states which, for simplicity, are referred to as remodelled and unremodelled chromatin [5]. HIS3 plasmid chromatin purified from uninduced cells is predominantly composed of fully supercoiled chromatin that is generally protected from cleavage by restriction enzymes, indicating that it has a canonical chromatin structure. In contrast, induced chromatin is predominantly composed of remodelled chromatin, characterised by a much reduced level of negative supercoiling, decreased compaction and increased sensitivity to restriction enzymes, indicating a highly accessible chromatin structure. The formation of remodelled chromatin requires both the Gcn4p activator and the SWI/SNF remodelling machine.

We have addressed the roles of the SWI/SNF and Isw1 remodelling machines in determining the positions of nucleosomes in HIS3 chromatin [6]. We used the "monomer extension" procedure [9] to map nucleosome positions in our yeast chromatin. In this method, the DNA from fully trimmed nucleosome core particles prepared from plasmid chromatin is mapped on the gene sequence using a primer extension approach, in which the core particle DNA acts as primer. This method can resolve complex chromatin structures, including overlapping nucleosome positions, which appear to be the rule rather than the exception. In contrast, indirect end-labelling, the traditional method for mapping nucleosomes in native chromatin, cannot detect overlapping positions and therefore yields only a simplified, low resolution, nucleosome map. The presence of overlapping positions indicates that native chromatin structures are highly heterogeneous, since nucleosomes cannot physically overlap. Although monomer extension revealed a large number of alternative, overlapping positions on the HIS3 gene, the nucleosome spacing is highly regular. We conclude that the HIS3 gene is organised into one of several alternative overlapping arrays of nucleosomes. In basal HIS3 chromatin, there is a dominant array, but this array loses its dominance in activated chromatin. Disruption of the dominant array requires both the Gcn4p activator and the SWI/SNF remodelling machine.

We propose that Gcn4p and SWI/SNF direct the mobilisation of nucleosomes over the entire HIS3 gene, apparently involving the coordinated shunting of nucleosomes from one array of positions to another, always maintaining the nucleosome spacing characteristic of yeast cells. We suggest that the net effect of nucleosome mobilisation might be to provide windows of opportunity for transcription initiation and elongation factors to access the underlying DNA, as the nucleosomes are shunted back and forth. Thus, the interplay between various remodelling machines is expected to create a highly dynamic chromatin structure.

Currently, we are ascertaining whether our findings for yeast plasmid chromatin can be extrapolated to the chromosome and to the entire yeast genome, using a high throughput sequencing approach.

References and Footnotes
  1. Thoma, F., Bergman, L.W., and Simpson, R.T. (1984). J. Mol. Biol. 177, 715-733.
  2. Alfieri, J.A., and Clark, D.J. (1999). Methods Enzymol. 304, 35-49.
  3. Kim, Y.J., Shen, C.-H., and Clark, D.J. (2004). Methods 33, 59-67.
  4. Shen, C.-H., Leblanc, B.P., Alfieri, J.A., and Clark, D.J. (2001). Mol. Cell. Biol. 21, 534-547.
  5. Kim, Y., and Clark, D.J. (2002). Proc. Natl. Acad. Sci. USA 99, 15381-15386.
  6. Kim, Y., McLaughlin, N. B., Lindstrom, K., Tsukiyama, T., and Clark, D. J. (2006). Mol. Cell. Biol. 26, 8607-8622.
  7. Natarajan, K., Jackson, B.M., Zhou, H., Winston, F., and Hinnebusch, A.G. (1999). Mol. Cell 4, 657-664.
  8. Reid, J.L., Iyer, V.R., Brown, P.O., and Struhl, K. (2000). Mol. Cell 6, 1297-1307.
  9. Yenidunya, A., Davey, C., Clark, D. J., Felsenfeld, G., and Allan, J. (1994). J. Mol. Biol. 237, 401-414.

David J. Clark

Laboratory of Molecular Growth Regulation
NICHHD, National Institutes of Health
Building 6 Room 2A14
Bethesda MD 20892-2426

Tel.: 301 496 6966
Fax: 301 480 1907
email David Clark