Multiple modes of chromatin configuration at natural meiotic recombination hot spots in fission yeast

Abstract

The ade6-M26 meiotic recombination hot spot of fission yeast is defined by a cyclic AMP-responsive element (CRE)-like heptanucleotide sequence, 5'-ATGACGT-3', which acts as a binding site for the Atf1/Pcr1 heterodimeric transcription factor required for hot spot activation. We previously demonstrated that the local chromatin around the M26 sequence motif alters to exhibit higher sensitivity to micrococcal nuclease before the initiation of meiotic recombination. In this study, we have examined whether or not such alterations in chromatin occur at natural meiotic DNA double-strand break (DSB) sites in Schizosaccharomyces pombe. At one of the most prominent DSB sites, mbs1 (meiotic break site 1), the chromatin structure has a constitutively accessible configuration at or near the DSB sites. The establishment of the open chromatin state and DSB formation are independent of the CRE-binding transcription factor, Atf1. Analysis of the chromatin configuration at CRE-dependent DSB sites revealed both differences from and similarities to mbs1. For example, the tdh1+ locus, which harbors a CRE consensus sequence near the DSB site, shows a meiotically induced open chromatin configuration, similar to ade6-M26. In contrast, the cds1+ locus is similar to mbs1 in that it exhibits a constitutive open configuration. Importantly, Atf1 is required for the open chromatin formation in both tdh1+ and cds1+. These results suggest that CRE-dependent meiotic chromatin changes are intrinsic processes related to DSB formation in fission yeast meiosis. In addition, the results suggest that the chromatin configuration in natural meiotic recombination hot spots can be classified into at least three distinct categories: (i) an Atf1-CRE-independent constitutively open chromatin configuration, (ii) an Atf1-CRE-dependent meiotically induced open chromatin configuration, and (iii) an Atf1-CRE-dependent constitutively open chromatin configuration.

FIG. 1.

The chromatin structure around the DSB sites at the mbs1 locus shows a relatively open state in mitosis and meiosis. (A) Diploid strain D20 (wild type [WT]) was cultured in MM-NH₄Cl medium (mitosis lanes). Cells were then transferred to MM medium lacking nitrogen and cultured further for 4 h (meiosis lanes). Chromatin isolated from the cells was digested with MNase and analyzed as described previously (17). To analyze the meiotic DSBs, haploid pat1-114 rad50s strains PKH138 (wild type) and PKH163 (atf1Δ) were cultured, and DNA was prepared as described in Materials and Methods. DNA samples from MNase-digested chromatins and synchronous meioses were digested with SpeI and analyzed in the same gel. Lane N, MNase-digested naked S. pombe genome DNA. Meiotic DSBs are indicated by dotted lines. Thick lines indicate MNase-sensitive regions. (B) MNase-digested chromatin DNA from diploid strains D20 (wild type) and WSP779 (atf1Δ) were analyzed in same gel to compare the chromatin structure.

FIG. 2.

Meiotic chromatin remodeling in the tdh1⁺ locus depends on Atf1. MNase-digested chromatin DNAs from diploid strains D20 (wild type [WT]) and WSP779 (atf1Δ) were analyzed as in Fig. 1. The atf1Δ samples were slightly overdigested, since the atf1Δ mutant is more sensitive to Zymolyase treatment that allows increased permeation of MNase. The vertical and the horizontal arrows indicate the tdh1⁺ ORF and the position of the CRE sequences, respectively. Lane M, marker λ, EcoT14I digested (Takara); lane N, MNase-digested naked S. pombe genome DNA.

FIG. 3.

The chromatin structure around ade6-M26 and tdh1⁺ in mitosis and meiosis in DSB-defective mutants. Meiotic DSB formation is critically impaired in rec6Δ, rec7Δ, rec8Δ, rec10Δ, rec12Δ, rec14Δ, rec15Δ, and mei4Δ mutants (1, 4, 6, 41). We examined the chromatin structure around ade6-M26 (A) and tdh1⁺ (B) in these mutants. The M26 diploid strains D20 (wild type [WT]), D55 (rec7Δ), D68 (rec8Δ), D69 (rec10Δ), D12 (rec12Δ), D70 (rec15Δ), and D66 (mei4Δ) were cultured as in Fig. 1. Isolation of the chromatin fraction and treatment with MNase were done as described in Materials and Methods. Southern blot analysis was performed according to the method of Mizuno et al. (17). Lane N, partial digestion of naked DNA with MNase.

FIG. 4.

The transcription of tdh1⁺ and cds1⁺ is not dependent on Atf1 and is not induced during meiosis. The diploid strains D20 (wild type [WT]) and WSP779 (atf1Δ) were cultured as in Fig. 1. The cells were harvested at the indicated time points after the medium shift. Preparation of total RNA and Northern blot analysis were performed as described in Materials and Methods. rRNA was detected by ethidium bromide staining as a loading control.

FIG. 5.

Formation of meiotic DSBs in the tdh1⁺ locus depends on Rec12 and the Atf1-Pcr1 complex. (A) Haploid pat1-114 rad50s strains PKH138 (wild type [WT]), PKH118 (rec12Δ), PKH163 (atf1Δ), and PKH160 (pcr1Δ) were cultured, and DNA was prepared as described in Materials and Methods. tdh1⁺ ORF and CRE sequences are indicated as in Fig. 2. An arrowhead indicates the DSB site. The genome-wide DSBs in the same samples were analyzed by pulsed-field gel electrophoresis. (B) DNA samples from the MNase-digested chromatins and synchronous meioses in Fig. 1 were digested with ApaLI and AflII and analyzed in the same gel. DSBs are indicated by a dotted line; tdh1⁺ ORF and CRE sequences are indicated as in Fig. 2.

FIG. 6.

Rec12 and Rad32 proteins bind to meiotic DSB sites ade6-M26, tdh1⁺ (CRE-mediated hot spots), and mbs1 (CRE-independent hot spot) in meiosis. (A) The binding of Rec12-Flag and Rad32-Flag proteins to the ade6 locus carrying the M26 or M375 allele was examined. The haploid pat1-114 rad50s strains PKH50 (rec12-flag ade6-M26), PKH52 (rad32-flag ade6-M26), PKH338 (rec12-flag ade6-M375), and PKH339 (rad32-flag ade6-M375) were cultured to induce meiosis (0 and 4 h) and fixed with formaldehyde. ChIP analysis was performed as described in Materials and Methods. The binding of Rec12-Flag and Rad32-Flag to the ade6, tdh1⁺, and mbs1 loci was detected using PCR. Whole genomic DNA from the 1% input sample was amplified at the same time. (B) ChIP efficiency in the M26 and M375 alleles was quantified by real-time PCR analysis as described in Materials and Methods. ChIP efficiency was calculated as IP sample/input material and represented as IP (percent) Error bars represent standard deviations. (C) The binding of Rec12-Flag protein to the ade6, tdh1⁺, and mbs1 loci was examined in wild-type and atf1Δ strains. The haploid ade6-M26 pat1-114 rad50s rec12-flag strains PKH50 (wild type) and PKH114 (atf1Δ) were cultured to induce meiosis (0, 4, and 6 h) and fixed with formaldehyde. The binding of Rec12-Flag to ade6, tdh1⁺, and mbs1 loci was detected as for panel A. (D) Quantification of ChIP efficiency from panel C. Error bars represent standard deviations.

FIG. 7.

The chromatin structure at the CRE sequence in the cds1⁺ locus demonstrates a constitutively open state in an Atf1-dependent manner. (A) Meiotic DSB and chromatin structures were analyzed in the same gel. The ORF of cds1⁺ and its associated CRE sequence are indicated as in Fig. 2. Lane N, MNase-digested naked S. pombe genome DNA. Meiotic DSBs are indicated by dotted lines. (B) The chromatin structures in diploid strains D20 (wild type [WT]), WSP779 (atf1Δ), and D74 (cds1-2) were compared as for Fig. 1. The CRE sequence is indicated by an arrowhead.