Microarray deacetylation maps determine genome-wide functions for yeast histone deacetylases

Daniel Robyr, Yuko Suka, Ioannis Xenarios, Siavash K. Kurdistani, Amy Wang, Noriyuki Suka and Michael Grunstein

Cell 109, 437-446 (2002)


Supplemental data    ( 1, 2, 3, 4, 5, 6)
Download acetylation data   (Download data )

Supplemental data


Generating acetylation microarrays. Chromatin fragments from formaldehyde-treated histone deacetylase mutant cells (in that case rpd3D)  and their wild type (WT) isogenic counterparts were immunoprecipitated (Chromatin IP) using highly specific antibodies raised against acetylated histone sites (Suka et al., 2001). DNA from enriched chromatin fragments was purified, amplified by PCR (DNA amplification) and labeled (Klenow) with a fluorophore (Cy3 or Cy5).  Wild type (WT) and mutant sets of labeled DNA were then combined and hybridized to a DNA microarray containing ~6700 intergenic regions ( Intergenic_fragments_coordinates ) (IGRs) (Iyer et al., 2001; Ren et al., 2001). For a given IGR the ratio of the normalized fluorescent intensities between the two probes indicates whether the analyzed lysine residue is hypo- or hyper- acetylated upon deletion of a deacetylase gene (as compared to the WT strain).

 (Top of page)

(pdf file)
Chromosomal display of sites affected by RPD3 at histone H4 K12. Left (L) and right (R) chromosome arms are represented from their telomere end to their centromere (black circle). Each intergenic region is illustrated by a rectangle. Regions whose enrichment is increased (> 1.95 fold) relative to the control are shaded in red. Note that chromosomes are not to scale since ORFs were left out and intergenic regions were arbitrarily given the same width for clarity.

 (Top of page)

(pdf file)
Functional categories found among the Rpd3 and Hda1 target genes based on the MIPS classification.  The number of genes assigned to a given category is listed in the left column. The number of genes whose promoter is deacetylated by Rpd3 (at H4-K12) or Hda1 (at H3-K18) are shown in the "Found" column (within the brackets are indicated the number of genes for which we have data available for a particular category). The probability (P-value) to observe more than k overlapping regions between two data sets was calculated as follow:

where k is the observed overlap ("Found"), n is the number of genes within a category and m is the number of IGRs  hyperacetylated by a factor of 1.95 or more (531 and 467 regions for Rpd3 and Hda1 respectively). N is the total number of regions for which we have data available.

 (Top of page)

(pdf file)
Intergenic and ORF regions identified in HAST domains. Intergenic and ORF fragments are shown along with their genomic coordinates (bp) and the acetylation increase at histone H3-K18 upon HDA1 disruption. The ORF names and descriptions are also indicated. Acetylation fold increases  >1.95 fold are highlighted in orange and ORFs are written in blue in order to distinguish them from intergenic regions. For simplicity, we have chosen to define the domains from the edgemost sequences affected >1.95 fold by hda1D. The size and coordinates of these domains are shown in red.

 (Top of page)

(pdf file)
Hierarchical cluster analysis comparing expression data from 300 mutants or conditions (Hughes et al., 2000) with 149 ORFs located in the HAST domains. Changes of gene expression in high glucose (19 g/L) and low glucose (~0 g/L) are also introduced in the clustering analysis (DeRisi et al., 1997). The scale of gene activity is represented from green (repressed) to red (derepressed). The cluster analysis was performed with Gene Cluster (2.11) and visualized using TreeView (1.50) (Eisen et al., 1998).
   (Top of page)

(pdf file)
IGRs (and some ORFs) at which histone H4 K12 acetylation is most affected by hos2D. Ribosomal protein genes are highlighted in orange. Fold increases in H4 K12 hyperacetylation in a hos2D mutant are shown.

 (Top of page)

Download Data (Data files are text tab delimited)

File containing the coordinates of all intergenic regions used in this study

Normalized average acetylation fold changes in the specified mutant strain versus the isogenic wild type strain (i.e. HDA1-H3K18: acetylation changes at Lysine 18 of histone H3 in hda1D). Please, refer to (Intergenic_fragments_coordinates) for informations regarding the chromsomal location of the intergenic regions (IGRs).

Normalized average acetylation fold changes in the specified mutant strain versus the isogenic wild type strain (i.e. HDA1-H3K18: acetylation changes at Lysine 18 of histone H3 in hda1D).

Relative amount of INPUT DNA (no immunoprecipitation) from the mutant and WT strains are tested as a control for aneuploidy. (Data for rpd3D and hda1D; note that this control experiment was performed once).

 (Top of page)

Grunstein Lab Home Page