Conserved features of cohesin binding along fission yeast chromosomes

High-resolution analysis of cohesin localization on fission yeast chromosomes reveals that several determinants, previously thought to be organism-specific, come together to shape overall distribution.


(b)
Maximal distance analysis between neighboring cohesin peaks. 10,000 random cohesin patterns, distributed among the convergent sites along fission yeast chromosome 2, were created using a bootstrapping approach. The number and widths of the distributed cohesin peaks were those from the observed binding pattern, as assigned in Figure S1. The maximal distance between neighboring cohesin peaks in each of the random distributions was determined and is shown in the histogram below. The red line indicates the maximal distance between two neighboring cohesin-bound convergent sites in the actual observed cohesin pattern.

Figure S3
Higher expression levels of convergent gene pairs flanking cohesin binding sites. The distribution of relative mRNA levels from genes flanking cohesin-bound and cohesin-free convergent sites along chromosome 2 is shown. Boxes indicate boundaries of the 25 th to the 75 th percentile. The median is shown as a bold black line, whiskers extend to 1.5 times the interquartile range, outliers are marked as circles. A Wilcoxon signed ranks test suggests that genes flanking cohesin-bound convergent sites are more strongly expressed than genes flanking cohesin-free convergent sites (p=0.011).
The cohesin loader Mis4/Ssl3 overlaps with tRNA and ribosomal protein genes. (a) Binding profile of Mis4 and Ssl3 along fission yeast chromosome 2. Chromatin immunoprecipitation was performed against epitope-tagged Mis4-Pk 9 and Ssl3-Pk 9 from exponentially proliferating cells. As a control, cells without epitope-tagged protein were grown under identical conditions and processed in parallel for chromatin immunoprecipitation with an α-Pk antibody. The map shows an overlay of Mis4 (green), Ssl3 (blue) and the untagged control (purple), as described in Figure S1. Peaks were assigned as detailed in Figure S1, but the threshold for local maxima was a signal intensity of 0.5. The untagged control sample yielded trace amounts of immunoprecipitated DNA, which after amplification led to several strong peaks often in intergenic low complexity regions. Several of these peaks overlapped with Mis4/Ssl3 peaks, which were excluded from the analysis. Peaks in low complexity regions were not observed in chromatin immunoprecipitates of cohesin subunits (compare Figure S1). (b)

(c)
Similar Mis4 localization in exponentially growing and in G1 arrested cells. The Mis4 localization pattern along chromosome 2 in exponentially growing cells is compared to the pattern observed in cells arrested in G1 using the cdc10-129 temperature sensitive mutation. (a) Exponentially growing cdc10-129 cells were shifted to the restrictive temperature of 37°C for 3.5 hours. G1 arrest was confirmed by flow cytometry of DNA content. (b) Chromatin immunoprecipitation was performed against epitope-tagged Mis4-Pk 9 from the G1 arrested cells 3.5 hours after temperature shift. The association pattern along chromosome 2 is shown (brown) in comparison to that observed in exponentially growing cells that are predominantly in the G2 phase of the cell cycle (geen, data reproduced from Figure S4).

(b)
Mis4/Ssl3 binding sites correlate with strongly expressed genes. This graph compares the relative mRNA levels of ribosomal protein genes, Mis4-bound genes other than ribosomal protein genes, and all remaining genes that are not associated with Mis4 along chromosome 2. tRNA genes were excluded from this analysis. Box boundaries mark the 25 th to 75 th percentile surrounding the median (bold line). Whiskers extend to 1.5 times the interquartile range. Outliers are indicated as circles. Wilcoxon signed ranks tests suggest that Mis4-bound genes are significantly more strongly expressed than non-bound genes.
The cohesin pattern along chromosome arms remains qualitatively unchanged during cohesin removal in mitosis. Cells were arrested in G2 and synchronously released into mitotic progression. Chromatin immunoprecipitation was performed against the Rad21-Pk 9 cohesin subunit. A 100 kb region on the left arm of chromosome 2 in G2-arrested cells is compared to the patterns at the indicated timepoints after release when the majority of cells were in metaphase and in anaphase, respectively (compare Figure 7).

Table S1
Yeast strains used in this study.