Perturbation of the Activity of Replication Origin by Meiosis-specific Transcription*

We have determined the activity of all ARSs on the Saccharomyces cerevisiae chromosome VI as chromosomal replication origins in premeiotic S-phase by neutral/neutral two-dimensional gel electrophoresis. The comparison of origin activity of each origin in mitotic and premeiotic S-phase showed that one of the most efficient origins in mitotic S-phase, ARS605, was completely inhibited in premeiotic S-phase. ARS605 is located within the open reading frame of MSH4 gene that is transcribed specifically during an early stage of meiosis. Systematic analysis of relationships between MSH4 transcription and ARS605 origin activity revealed that transcription of MSH4 inhibited the ARS605 origin activity by removing origin recognition complex from ARS605. Deletion of UME6, a transcription factor responsible for repressing MSH4 during mitotic S-phase, resulted in inactivation of ARS605 in mitosis. Our finding is the first demonstration that the transcriptional regulation on the replication origin activity is related to changes in cell physiology. These results may provide insights into changes in replication origin activity in embryonic cell cycle during early developmental stages.

Eukaryotic chromosomes consist of multiple replication units. Each origin of DNA replication is strictly controlled to fire only once in the cell cycle in the fixed sequential order (1,2). This temporal program for origin firing is utilized for the maintenance of genome integrity through DNA replication checkpoint mechanism (3,4). It was shown that a checkpoint effector kinase Rad53 repressed firing of late origins when replication was perturbed by hydroxy urea or methyl methane sulfonate. Furthermore, recent genome-wide studies of eukaryotic chromosomal DNA replication using Saccharomyces cerevisiae have provided us with a kinetic map of progression of DNA replication along the chromosome (5,6). However, these studies were restricted to DNA replication during mitotic cell cycle and little is known about how these multiple replication origins behave under different cell cycle, i.e. meiotic cell cycle.
The decision to enter the meiotic cell cycle is made in G 1 phase and this affects the way in which the G 1 /S transition is controlled. In budding yeast (S. cerevisiae), poor nutrient conditions are the cue to embark on the meiotic cell cycle, which culminates in the production of spores (7). The replication of chromosome is the first detectable cytological event in meiosis. The coordinated synthesis of genomic DNA requires multiple levels of regulation and a large number of gene products. Genetic analysis in S. cerevisiae indicates that the replicative machinery used to synthesize DNA in vegetative cells is also required for the duplication of chromosome in meiosis (8 -10).
In S. cerevisiae, mitotic S-phase and premeiotic S-phase appear to be differently regulated. For example, premeiotic S-phase is 1.5-2 times longer than mitotic S-phase (11). However, replication kinetics as measured by neutral/neutral twodimensional gel electrophoresis of 100-kb segment of chromosome III of S. cerevisiae in premeiotic S-phase are similar to those measured in mitotic S-phase (12). Furthermore, the same study showed that DNA replication of chromosome III initiates at the same origins in meiosis and mitosis. These two conflicting results should be resolved by studying replication of chromosomes other than chromosome III.
Several lines of evidence suggest that premeiotic DNA replication links to the initiation of meiotic recombination events. Borde et al. (13) showed that delaying the replication of a chromosome segment specifically delays the formation of the break at that segment. Cha et al. (14) showed that Spo11, which is required for interhomolog interaction, and Rec8, a meiosisspecific cohesin subunit, are negative and positive regulators for the progression of premeiotic S-phase, respectively. These results suggest that there are many other possible regulations of replication at each chromosomal level that affect the progression of S-phase in meiotic cell cycle.
The chromosome VI of S. cerevisiae is 270 kb in length and the best-characterized chromosome for replicon structure in mitotic cell cycle. It contains ten ARSs, ARS601-609, of which nine ARSs are active as replication origins that comprise five active replicons in mitotic S-phase (1,2,15). Thus, three origins (ARS605, 606, and 607) are used in high frequency, four (ARS601/2, 603, 603.5, and 609) are in intermediate frequency, and two (ARS604 and 608) are in low frequency, less than 5% of cell cycles. Replication of each arm of chromosome starts from one major origin (ARS605 and 607 on the left and the right arm, respectively) and extends sequentially in both directions (1,2,6). During mitotic growth, all of these replication origins are bound by ORC (16,17).
The aim of this study is to determine the replicon structure of the chromosome VI of S. cerevisiae in the premeiotic S-phase and to understand the difference from that of the S-phase replicon at molecular level. Interestingly, we found that activity of one origin ARS605, an efficient origin during mitotic cell cycle, was completely inhibited during premeiotic S-phase. As ARS605 resides within the open reading frame (ORF) 2 of a meiosis-specific gene, MSH4, we tested the possibility that ARS605 was inhibited by transcription. We found that transcription of MSH4 by RNA polymerase II removed ORC from ARS605 during premeiotic S-phase. This is the first demonstration of origin inactivation by transcription that changes depending on the physiology of the cell.

EXPERIMENTAL PROCEDURES
Strains-All strains used in this study (Table 1) are isogenic to NKY730 or NKY278. Orc1 protein was tagged by 3ϫ HA epitope tag cassette at its C terminus using the plasmid p306ORC1-HA/C (18). An ⌬ume6 mutant is a deletion mutant that is constructed by insertion of the KanMX cassette (19).
Induction of Sporulation-Cells were grown and sporulated as described (20). In short, a single colony from the YPG (3% glycerol, 2% Bacto peptone, 1% yeast extract) plate was inoculated into 5 ml of SPS presporulation medium (0.5% yeast extract, 1% Bacto peptone, 0.17% yeast nitrogen base without ammonium sulfate and amino acids, 0.05 M potassium phthalate, 1% potassium acetate, 0.5% ammonium sulfate, pH 5.0), and cells were cultured at 30°C overnight. Small amounts of the preculture suspension were then inoculated into 0.5 liter of SPS, and cells were cultured at 30°C to a density of 4 -7 ϫ 10 7 cells/ml. Cells were harvested by centrifugation and washed twice in sterile water and were inoculated into 1 liter of SPM (sporulation medium) (1% potassium acetate, 0.001% polypropylene glycol 2000 in 5-liter flasks), and cells were cultured at 30°C with vigorous aeration.
FACS (Fluorescence-activated Cell Sorter)-The flow cytometric determination of cellular DNA content was performed on FACS calibur (BD Biosciences) as described previously (21). For each histogram, 25,000 cells were analyzed.
Neutral/neutral (N/N) Two-dimensional Agarose Gel Electrophoresis-Mitotic cells were grown to 3 ϫ 10 7 cells/ml at 30°C in YPD and then mixed with an equal volume of the toluene stop solution (95% EtOH, 3% toluene, 20 mM Tris, pH 7.5), and meiotic cells were taken every 15 min from 60 -225 min after cells were transferred to SPM and pooled them. Samples for N/N two-dimensional gel electrophoresis were prepared as described previously (22), and N/N two-dimensional gel electrophoresis was carried out by the method of Brewer and Fangman (23).
Chromatin Immunoprecipitation (ChIP)-Mitotic cells were grown to 1 ϫ 10 7 cells/ml at 23°C in 100 ml of YPD and then fixed with 1% formaldehyde at room temperature for 30 min. Synchronous cultures induced for meiosis were withdrawn every 1 h at 0-5 h and fixed in the same way. Orc1-tagged strains undergo meiosis normally.
Reverse Transcriptase (RT)-PCR-Total RNA was extracted from yeast cells using the RNeasy kit (Qiagen). RT-PCR was performed using the TaKaRa RNA PCR kit (AMV) version 3.0 (TaKaRa) by following the manufacturer's instructions. Regions of MSH4 and ACT1, cDNAs were amplified using the following primer sets: MSH4 5Ј-TTGGCAGTAACTGAACA-TTTAC and 5Ј-CGCTACAAGATTGTGTATTC (359 bp); ACT1 5Ј-TGTTTTGGGTTTGGAATCTG and 5Ј-CCACTT-TCGTCGTATTCTTG (302 bp). Fig. 1, A and B show S. cerevisiae cell growth during the transfer from mitotic to meiotic cell cycle. In response to nitrogen starvation, mitotic cells enter the meiotic cell cycle synchronously at the point in G 1 phase called "Start" to undergo premeiotic DNA synthesis, which is followed by recombination, twice of chromosome division and finally spore formation.

Origin Activities of ARSs on Chromosome VI in Premeiotic S-phase-
To determine the period of the premeiotic S-phase precisely, we assessed the progression of meiotic cell cycle by FACS and by microscopic observation of nuclear division. We used the SK1 strain as a wild type because of its efficient sporulation properties compared with other strains (24). We determined that 90% of cells entered premeiotic S-phase between 1-2 h and finished around 4 -5 h after induction, judging from the FACS profile and kinetics of meiotic divisions (Fig.  1A). The spore formation rate reached 99% (Fig. 1B) 10 h after induction, suggesting that DNA synthesis observed by FACS indeed occurred in the premeiotic S-phase.
To determine the activity of 10 ARSs on chromosome VI as replication origins, we pooled cells every 15 min from 1-4 h after the induction of sporulation and performed the N/N twodimensional gel electrophoresis (Fig. 1C). Because the SK1 strain is genetically distant from the S288C strain, the genomic sequence of which has been determined (24); we newly measured the activity of all ARS during mitotic growth in the SK1 and compared with that of the premeiotic S-phase. Origin activity of nine ARSs in the SK1 is almost same as that of W303 and S288C (2). One exception was that we did not detect the restriction fragment corresponding to ARS601/2, a weak late, firing origin in other two strains. This may be because of a DNA sequence polymorphism between the SK1 and S288C strains.
The activity of origins was similar between the mitotic and premeiotic S-phase except for ARS605 (Fig. 1C). The strong bubble arc detected in the ARS605 region in the mitotic S-phase completely disappeared in the premeiotic S-phase. This result indicated that ARS605 was inactivated in the premeiotic S-phase. ARS605 is located within the ORF of MSH4 (a MutS homolog) and transcribed in early meiosis (25,26). We assumed that the loss of ARS605 activity was because of the MSH4 transcription because it is known that transcription from a strong galactose inducible promoter interferes with origin functions on plasmid (27).
Transcription of MSH4 Interferes with Origin Activity of ARS605 in Meiosis-To confirm our hypothesis, we examined the relationships between MSH4 transcription and ARS605 origin activity in the premeiotic S-phase (Fig. 2).
The transcription of MSH4 was induced at between one and two hours and reached the level of 10-fold of time 0 at three hours after the induction of meiosis, which is in good agreement with the previous data (24). The level of MSH4 transcripts at 3 h was maintained at the same level up to 5 h (data not shown), suggesting that MSH4 was transcribed actively during the premeiotic S-phase.
It was reported previously that the origin function of an ARS1 was inactivated by strong transcription that run through it (27). However, it is not yet clear how the origin function is actually perturbed by transcription at the molecular level. One possible mechanism of inactivation of ARS605 is that transcription disturbs the assembly of prereplication complex, which is the binding of ORC and mini-chromosome maintenance complexes to DNA sequence of the ARS.
We therefore examined whether the ORC binding was affected by transcription using the conventional ChIP method (18). Three primer sets that cover ARS605, ARS607, and a region 4 kb away from ARS1 (non-ARS) were designed. We examined the binding of Orc1, the largest subunit of ORC, during the premeiotic S-phase. Fig. 2B shows that binding of Orc1 to ARS605 was comparable to ARS607 until 1.5 h after the induction but decreased gradually starting from 2 h and reached one tenth of the binding level to ARS607 by 4 h. We further confirmed the decreased binding of Orc1 to ARS605, using ARS1 as a positive control (Fig. 2C) because ARS1 was an active origin in the premeiotic S-phase (28). The decrease in Orc1 binding to ARS605 paralleled negatively with the increase in the transcription of MSH4, suggesting strongly that the removal of Orc1 by transcription had disrupted the origin function of ARS605.
Transcription of MSH4 Interferes with Origin Activity of ARS605 in Mitosis-Induction of the early meiosis-specific gene family including MSH4 is dependent on the transcription factors Ime1 and Ume6 (29). In an ume6 deletion mutant, early meiosis-specific genes are derepressed during vegetative growth (24,29). Therefore we further confirmed the relationships between MSH4 transcription and ARS605 origin activity by introducing a deletion mutation of ume6 to the cell. We  FEBRUARY 16, 2007 • VOLUME 282 • NUMBER 7 assumed that derepression of MSH4 transcription by the ume6 mutation would inhibit ARS605 activity even in the mitotic S-phase. Fig. 3A shows that MSH4 was transcribed at the level comparable to that in the meiosis in the absence of the ume6 function. As expected, N/N two-dimensional gel electrophoresis showed that activity of ARS605 as indicated by bubble arc disappeared completely in the ume6 deletion mutant during mitotic cell cycle (Fig. 3B). The binding of Orc1 to ARS605 was reduced and hardly detectable by ChIP analysis (Fig. 3C). These results confirm that MSH4 transcription interferes with ARS605 activity by removing ORC from its binding site.

Replicon Structure in Meiosis
Mutation in RNA Polymerase II Can Restore Origin Activity-To investigate whether or not the removal of ORC from the origin, ARS605, was because of the actual elongation of transcription through it, we have examined the effect of a temperature sensitive mutation of RNA polymerase II, rpb1-1 (30), on the derepression of MSH4 transcription by ume6 deletion mutation. Surprisingly, the MSH4 transcription in ⌬ume6 mutant cells in the mitotic phase was reduced by the introduction of rpb1-1 even at permissive temperatures (Fig. 3A). Therefore we analyzed the ARS605 activity in a ⌬ume6/rpb1-1 double mutant at permissive temperatures and found that the origin activity of ARS605 was recovered (Fig. 3B). The Orc1 binding to ARS605 was also restored under the same condition (Fig. 3C). These results indicated that the inactivation of the replication origin was caused by the removal of ORC from its binding site, possibly by collision with RNA poymerase II.

DISCUSSION
Through the chromosome-wide investigation of replicon dynamics during the mitotic and premeiotic S-phase, we found that the activity of an efficient origin in the mitotic S-phase, ARS605, was completely lost in the premeiotic S-phase (Fig.  1C). The inhibition was correlated with the induction of transcription of a meiosis-specific gene MSH4 after the onset of presporulation cell growth (Fig. 2). We have also demonstrated that Orc1 binding to ARS605 origin was reduced specifically in parallel with the increase in MSH4 transcription. Because ARS605 is located within the ORF of MSH4, we tested the possibility that the origin activity is inversely correlated with the transcription that runs through the origin region and removes ORC complex that is essential for the origin activity. As expected, the activity of ARS605 was completely dependent on the UME6 gene function that regulated MSH4 negatively during mitotic cell and positively during meiotic cell cycle (Fig. 3). Furthermore, we showed that RNA polymerase was required for the origin inactivation (Fig. 3). A similar result has been reported by an artificial system, where the gal promoter fused to ARS1 does abolish origin activity when transcription is induced by galactose, although no molecular mechanism is indicated (27). Here, we demonstrated clearly that replication origin activity was negatively regulated through transcription, which interfered the binding of ORC complex to the replication origin either directly by running of RNA polymerase or indirectly by transcription-associated events like changes in DNA topology or chromatin structure. Based on these results, we propose a model for the inactivation of ARS activity by transcription (Fig. 4).
Previous studies have shown that the length of S-phase is different between the mitotic and meiotic cell cycle, whereas elongation rate of replication as well as initiation activity of all ARS on the chromosome III is invariable between the two cell cycles. Present finding of the loss of one of the three active origins on the chromosome VI may reduce the rate of replication of the chromosome specifically in meiotic cycle. If inverse relationships between transcription of genes and initiation of replication are general features of the replication origins of the eukaryotic cells, origins located within the transcription units should affect the S-phase progression, dependent on physiological conditions, that affects transcription of those particular genes. It is therefore interesting to determine replication origins that are located within the transcription units. In particular, transcriptions specific for the meiotic cell cycle should be examined carefully for the presence of ORC binding sites, ACS, as possible origins of replication that affect S-phase progression during meiotic cell growth.
Chromosomes in the pre-blastula Xenopus embryonic cells initiate replication at many evenly spaced sites distributed at random throughout the chromosomes (31). And it was shown that reduction in number and specifications of origins are first FIGURE 3. The MSH4 transcription, ARS605 initiation/replication activity, and Orc1 binding to ARS605 in mutant strains, ⌬ume6, and ⌬ume6/rpb1-1. A, levels of MSH4 and ACT1 mRNA in asynchronous vegetative growth in wild type (NKY278) and mutants (SKY10010 and SKY10011) at 23°C. 21 cycles of PCR were performed for both MSH4 and ACT1, and PCR products were quantitated as in shown in Fig. 2B. Results were normalized using ACT1 values. B, the N/N twodimensional gel electrophoresis of the replication origin activity of ARS605. A diagram shows the 4.4 kb BamHI-XhoI fragment examined by the N/N two-dimensional gel electrophoresis. The position of the MSH4 ORF and ARS605 are indicated. Samples were taken from wild-type cells and mutant cells, wild type (left), ⌬ume6 mutant (middle), and ⌬ume6/rpb1-1 double mutant (right), in asynchronous vegetative growth at 23°C and analyzed as shown in Fig. 1C. The bubble arc is indicated by an arrow. C, the physical association of Orc1 with ARS605 in the two mutants. Samples were also taken from cells in vegetative growth at 23°C. Levels of Orc1 binding were measured and analyzed as in Fig. 2. Bars in the bottom graph are standard errors of two experiments. The initiation of DNA replication requires the stepwise association of several proteins at the origin of replication. When the initiator binding site, ACS, resides in ORF, the origin can be activated in the absence of transcription. Switch from active to inactive state of replication origin is accompanied by removal of ORC complex from origin by transcription. In the absence of transcription, the origin is active and fired normally (A). When the transcription takes place, the origin function is prevented (inactive) as the result of dissociation of Orc1 from the origin probably either directly by the collision with RNA polymerase II or indirectly by transcription associated events like changes in DNA topology or chromatin structure (B). observed at the time of the mid-blastula transition (32,33). The mechanisms behind the origin specification remain poorly understood. However, it is well known that mid-blastula is accompanied by the increases in transcription of zygotic genes. Similar inverse correlation between origin activation and induction of transcription proposed in Fig. 4 may explain the origin specification during early development of Xenopus embryonic cells.