Topoisomerase I regulates open chromatin and controls gene expression in vivo

Abstract

DNA topoisomerases regulate the topological state of the DNA double helix and are key enzymes in the processes of DNA replication, transcription and genome stability. Using the fission yeast model Schizosaccharomyces pombe, we investigate genome wide how DNA topoisomerases I and II affect chromatin dynamics and gene expression in vivo. We show that topoisomerase I activity is directly required for efficient nucleosome disassembly at gene promoter regions. Lack of topoisomerase activity results in increased nucleosome occupancy, perturbed histone modifications and reduced transcription from these promoters. Strong correlative evidence suggests that topoisomerase I cooperates with the ATP-dependent chromatin remodeller Hrp1 in nucleosome disassembly. Our study links topoisomerase activity to the maintenance of open chromatin and regulating transcription in vivo.

Figure 1

Top1 and Top2 are mainly located at intergenic regions (IGRs). (A) ChIP-chip binding profiles for Top1 and Top2 for part of chromosome I in Hu1606 (top1-myc) and Hu1609 (top2-myc) at 30°C. The lower part of the figure shows open reading frames (ORFs) and their orientations. Top1 and Top2 binding in terms of relative enrichment are represented in log 2. ChIP was performed in biological duplicates. (B) Venn diagrams illustrating the similarity of Top1 and Top2 gene binding targets at 5′ IGRs (P=6.03 × 10^–201) and at 3′ IGRs (P=1.44 × 10^–53). Cutoff values of 1.25-fold enrichment were applied. P-values were generated using a hypergeometric probability test in the Genespring 8.0 software. (C) Average gene analysis for Top1 and Top2 binding in relation to RNA transcription levels in Hu0303 (WT) at 30°C. Number of genes (n) in each category is indicated in the figure. Error bars represent 95% confidence intervals.

Figure 2

Topoisomerase activity is required to regulate nucleosome occupancy at promoter regions. (A) ChIP-chip binding profiles for histone H3 C terminal (H3Cter) for part of chromosome I in Hu0303 (WT) and Hu1773 (top1Δtop2-191) at 30°C. The second panel shows the ratio of H3Cter binding in mutant versus WT. The lower part of the figure shows open reading frames (ORFs) and their orientations. H3Cter binding in terms of relative enrichment is represented in log 2. ChIP was performed in biological duplicates. (B) Average gene analysis for the ratio of H3Cter binding in mutant versus WT relative to mRNA levels in Hu0303 (WT) at 30°C. The number of genes (n) in each category is indicated in the figure. Error bars represent 95% confidence intervals. (C) Moving average plot (window size=150 genes, step size=1 gene) of the RNA transcription level ratios in Hu1773 (top1Δtop2-191) versus Hu0303 (WT) plotted as a function of mRNA levels in WT at 30°C. AU, arbitrary units. (D) RNA pol II density. Error bars represent 95% confidence intervals. (E) Venn diagram illustrating the overlaps between the top 500 genes that display increased H3Cter binding and the top 500 genes that display upregulated (P=6.82E−5) or downregulated (P=6.88E−16) mRNA levels in mutant versus WT. The P-values were generated using a hypergeometric probability test using the Genespring 8.0 software.

Figure 3

Topoisomerase activity is required for nucleosome disassembly and maintenance of H3K9ac, supporting high levels of transcription. (A–E) Validation of five gene targets of topoisomerase activity and five control genes (selection based on microarray data) at 30°C. Target gene identities are the following: 1=SPAC1F8.07c; 2=SPAC4F8.07c; 3=SPBC21c3.08c; 4=SPBC1709.05; 5=SPBPB7E8.01. Control gene identities are the following: 1=SPAP14E8.04; 2=SPAC222.06; 3=SPBC11C11.02; 4=SPAC1610.04; 5=SPAC17A5.10. All data in this figure are represented as mean±s.d. The ChIP DNA was analysed by qPCR. (A) ChIP relative enrichment of Top1 in Hu1606 (top1-myc). (B) mRNA levels in Hu0303 (WT) and Hu1773 (top1Δtop2-191). Note that the target and control genes are displayed with different y axis scales due to their differing transcription levels. (C) ChIP relative enrichment of histone H3 C terminal (H3Cter) in Hu0303 (WT) and Hu1773 (top1Δtop2-191). Note that the target and control genes are displayed with different y axis scales due to their differing H3 occupancy. (D) ChIP relative enrichment of H3K9 acetylation (H3K9Ac) normalized to H3Cter in Hu0303 (WT) and Hu1773 (top1Δtop2-191). (E) ChIP relative enrichment of catalytically active covalently bound Top1 after treatment of Hu1606 (top1-myc) with camptothecin (CPT) for 1 min. The IP was performed in triplicate and analysed by qPCR.

Figure 4

Hrp1, Top1 and Top2 share many targets in intergenic regions (IGRs). (A, B) Comparison of Top1 and Top2 in Hu1606 (top1-myc) and Hu1609 (top2-myc) with Hrp1 occupancy at 30°C (Eurogentec spotted microarray platform versus Affymetrix tiling arrays). The correlation coefficients are for Top1 versus Hrp1: R=0.747241868; and for Top2 versus Hrp1: R=0.727174702. (C) Venn diagram illustrating the overlap of Hrp1 with Top1 (P=1.04E−134) and Top2 (P=5.94E−131) occupancy at IGRs. Top1 and Top2 microarray data from the Affymetrix platform has been adapted to the Eurogentec platform using 50th percentile normalization to enable comparison with earlier data (Walfridsson et al, 2007). Cutoff values of two-fold relative enrichment have been applied for Top1 and Top2 binding target genes. Hrp1 binding target genes have been defined as the upper 0.9 percentile. P-values were generated using a hypergeometric probability test in the Genespring 8.0 software. (D) Hierarchical cluster analysis according to the Euclidian distance showing the correlation between increased nucleosome density in Hu1773 (top1Δtop2-191) or hrp1Δ with Top1, Top2 and Hrp1 binding at IGRs at 30°C. The clusters were generated using the TM4 software (MeV_4_2). (E) ChIP relative enrichment of histone H3 C terminal (H3Cter) at promoter regions of Hrp1 targets in Hu0303 (WT) and Hu1773 (top1Δtop2-191) at 30°C. ChIP was performed four times using biological duplicates and analysed by qPCR.

Figure 5

Open reading frames (ORFs) of long genes display Top2 enrichment and, in cells lacking topoisomerase activity, an accumulation of RNA polymerase II. (A) Average gene analysis according to gene length of ChIP-chip binding profile for Top1 in Hu1606 (top1-myc) at 30°C. (B) Average gene analysis according to gene length of ChIP-chip binding profile for Top2 in Hu1609 (top2-myc) at 30°C. (C) Average gene analysis according to gene length of the ratio of ChIP-chip binding profiles for RNA polymerase II in Hu1773 (top1Dtop2-191) versus Hu0303 (WT) at 30°C. RNA polymerase II, Top2 and Top1 occupancy in terms of relative enrichment is represented as a linear ratio. ChIP was performed in biological duplicates. Regions corresponding to ORFs common to all genes in each category are represented by a solid grey background, whereas regions corresponding to ORFs for a fraction of the genes are represented by grey stripes. Gene lengths and number of genes (n) in each category is indicated in the figure. Error bars represent 95% confidence intervals.

Figure 6

A model for how Top1 and Hrp1 may collaborate to maintain open chromatin at active gene promoters. The 5′ end of an active gene is represented, along with proposed nucleosome occupancy and transcription levels, as affected by transcription and the concerted action of Top1 and Hrp1. The model is for genes with both Top1 and Hrp1 at the promoters (236 genes in total; Figure 4C). See main text for further discussion.