Dual role of the Cdc7-regulatory protein Dbf4 during yeast meiosis

The Dbf4-dependent Cdc7 kinase (DDK) is essential for chromosome duplication in all eukaryotes, but was proposed to be dispen-sableforyeastpre-meioticDNAreplication.Thisdiscrepancyledus to investigate the role of the unstable Cdc7-regulatory protein Dbf4 in meiosis. We show that, when Dbf4 is depleted at the time of meiotic induction, cells enter the meiotic program but do not rep-licate their chromosomes. Surprisingly when Dbf4 is depleted after the initiation of DNA synthesis, S phase goes to completion, but most cells arrest before anaphase I. Deletion of the cohesin Rec8 suppresses this phenotype, suggesting a distinct role of DDK for meiotic chromosome segregation. As after Cdc5 depletion, a frac-tionofcellsundergoasingleequationaldivisionsuggestingafailure to mono-orient sister kinetochores. Our results demonstrate that Dbf4 is essential for DNA replication during meiosis like in vegetative cells and provide evidence for an additional role in setting up the reductional division of meiosis I. The initiation of eukaryotic DNA replication requires the activation of two Ser/Thr protein kinases, an S phase cyclin-dependent kinase (CDK) 3 and a Dbf4-dependent kinase (DDK), named Cdc7 in budding yeast (1). The Cdc7-Dbf4 kinase, which is conserved from yeast to human (2), is responsible for a late step in replication origin firing, locally and throughout S phase (3, 4).

The Dbf4-dependent Cdc7 kinase (DDK) is essential for chromosome duplication in all eukaryotes, but was proposed to be dispensable for yeast pre-meiotic DNA replication. This discrepancy led us to investigate the role of the unstable Cdc7-regulatory protein Dbf4 in meiosis. We show that, when Dbf4 is depleted at the time of meiotic induction, cells enter the meiotic program but do not replicate their chromosomes. Surprisingly when Dbf4 is depleted after the initiation of DNA synthesis, S phase goes to completion, but most cells arrest before anaphase I. Deletion of the cohesin Rec8 suppresses this phenotype, suggesting a distinct role of DDK for meiotic chromosome segregation. As after Cdc5 depletion, a fraction of cells undergo a single equational division suggesting a failure to mono-orient sister kinetochores. Our results demonstrate that Dbf4 is essential for DNA replication during meiosis like in vegetative cells and provide evidence for an additional role in setting up the reductional division of meiosis I.
The initiation of eukaryotic DNA replication requires the activation of two Ser/Thr protein kinases, an S phase cyclin-dependent kinase (CDK) 3 and a Dbf4-dependent kinase (DDK), named Cdc7 in budding yeast (1). The Cdc7-Dbf4 kinase, which is conserved from yeast to human (2), is responsible for a late step in replication origin firing, locally and throughout S phase (3,4). Cdc7-Dbf4 is also a target and transducer of the ATR-dependent S phase checkpoint in yeast and vertebrates (5,6). Besides its role in origin firing, Cdc7 has also been implicated in induced mutagenesis and meiotic recombination (2). Dbf4 is the Cdc7 kinase regulatory subunit, an unstable protein that accumulates in late G 1 , binds to replication origins, and guides Cdc7 kinase activity toward specific subunits of the pre-initiation complex (7)(8)(9)(10).
Despite its well characterized function for DNA replication in the mitotic cell cycle, little is known on the role of DDK during meiosis. Original experiments using a temperature-sensitive cdc7 allele concluded that it was essential neither for pre-meiotic DNA replication nor for ARS1 origin firing (11,12), leading to the idea that the mechanisms controlling DNA replication could be different in mitosis and meiosis. Hence several conserved initiation proteins (SpCdc18/Cdc6, Mcm2, and Mcm4) were found dispensable for pre-meiotic S phase in fission yeast (13), but more recent results using tighter mutants did not confirm this conclusion (14 -16). Furthermore, the same replication origins are generally used in mitosis and meiosis and acted upon by the same CDK complex in both cell types (17)(18)(19)(20)(21). In fact, Cdc7's dispensability for meiotic replication appeared as an exception and its precise function deserved further examination.
In this paper we reinvestigate the role of Cdc7-Dbf4 during meiosis. Contrary to initial reports, we show that Dbf4 is essential for pre-meiotic DNA replication and that no DNA synthesis occurs in its absence, as observed in mitosis. Furthermore we provide evidence that in addition to its role in origin firing, Dbf4 is also required after S phase to complete the first meiotic division, thus uncovering a novel and unexpected function of DDK for chromosome segregation during meiosis.

EXPERIMENTAL PROCEDURES
Strains and Growth Conditions-The strains used in this work are derived from SK1 background and are described in Table 1. The TET-DBF4 integrative plasmid (ESD560 (10) was cut at the unique EcoRV site to be integrated at URA3 locus of strain E1414 containing one deleted copy of DBF4.
Cells were grown in YEP medium (1% yeast extract, 2% bacto-peptone) supplemented with 50 mg of adenine per liter and 2% dextrose (YEPD). Sporulation was performed as described (22). YPA was 1% yeast extract, 2% bacto-peptone, and 1% potassium acetate. Sporulation medium (SPM) contained 1% potassium acetate with the required amino acid. For synchronous liquid sporulation cells were grown at 30°C in YEPD to late exponential stage (0.8 -1.2 ϫ 10 6 cells/ml) and transferred overnight (no more than 12 h) at 30°C in pre-sporulation medium (YPA) to obtain a concentration of ϳ2 ϫ 10 7 cells/ml. Then cells were washed twice in water and incubated in SPM with vigorous shaking at 30°C. For depletion experiments of Fig. 2, a low concentration of doxycycline (1 g/ml) was added into pre-sporulation medium (YPA) to reduce the level of the expression of the Tet-DBF4 transcript to the wild-type level by the time cells were shifted to sporulation medium. The number of asci (tetrads and dyads) was measured by light microscopy 24 or 48 h after transfer in SPM; an average of 200 cells were counted for each experimental condition.
Centrifugal Elutriation-Cells were grown as described above into one liter of YPA to a concentration of ϳ2 ϫ 10 7 cells/ml. Cells were then resuspended in 50 ml of YPA, sonicated, and pumped into a 40-ml elutriation chamber at about 50 ml/min with a rotor speed of 3000 rpm. Cells were kept at 25°C during the centrifugation and eluted at constant flow decreasing gradually rotor speed. Elutriated G 1 cells were collected, washed twice in water, and dissolved in SPM to a concentration of ϳ3 ϫ 10 7 cells/ml and incubated at 30°C with vigorous shaking.
Analysis of Gene and Protein Expression-Total RNA isolation and Northern analysis were performed as described (23) with 20 g of RNA per lane. RNAs were stained with SybrGold and visualized using an image plate scanner (Typhoon, Amersham Biosciences). DNA probes were prepared by random priming of open reading frame PCR products, amplified from genomic DNA, and labeled according to the NEBlot protocol (New England Biolabs). Protein extracts were prepared from trichloroacetic acid-fixed cells, and analyzed by Western blot as described previously (10).
Fluorescence Microscopy-Cells were fixed 15 min at 25°C in 4% paraformaldehyde added to the medium. Cells were washed twice with potassium phosphate buffer (KPO 4 ) and resuspended in 100 l of phosphate-buffered saline, pH 7.5, containing 4Ј,6-diamidino-2-phenylindol (DAPI) and incubated for at least 30 min. Cells were subsequently washed in phosphate-buffered saline for microscopy. Epifluorescence microscopy was performed with a 63ϫ objective on a Leica DMRA microscope equipped with a CoolSNAP HQ CCD camera (Roper Scientific). MetaMorph Imaging System v.4.5 (Universal Imaging Corp.) was used for image analysis.
Pulsed-field Gel Electrophoresis and DNA Quantification-Yeast cells were embedded in low melting agarose plugs (5 ϫ 10 7 cells/plug), and genomic DNA was extracted as described (24). Yeast chromosomes were separated by PFGE (Gene Navigator, AP Biotec). The gel was stained with SybrGold (1:10,000, Molecular Probes) and scanned with a FluorImager (Typhoon) and signals analyzed with ImageQuant (Amersham Biosciences). The amount of replicated DNA was estimated measuring the re-emergence from the gel wells of fully replicated chromosomes from nine representative chromosomal bands, and the average value was plotted on a graph considering the diploid DNA content as 100%.

RESULTS
Dbf4 Is Required for Pre-meiotic DNA Replication-We inactivated Cdc7 during meiosis by depleting its unstable regulatory subunit, Dbf4, by shutting off its transcription from a repressible tetracycline promoter (pTET-off) (10). By adding doxycycline to the medium at different times, it is possible to inactivate Cdc7 at different stages of meiotic progression (Fig. 1A). Because Dbf4 degradation is gradual after transcriptional shut-off, doxycycline was added 3 h before shifting the cells to sporula-FIGURE 1. DBF4 is required for pre-meiotic DNA replication. A, experimental approach: cells were grown vegetatively and doxycycline was added at a final concentration of 20 g/ml, 3 h before shifting to SPM or during meiotic progression (delayed Dbf4 depletion) as indicated by arrows. Dbf4 depletion is effective 3 h after doxycycline addition. B, doxycycline does not alter replication kinetics and sporulation efficiency in the wild-type strain E1180 grown in the presence of 20 g/ml of doxycycline (ϩDox). C, meiosis time course of DBF4/DBF4 (E1180), DBF4/ P TET-DBF4 (E1415) pTET-DBF4/pTET-DBF4 (E1418) strains in the presence of doxycycline (DBF4 off ). Sporulation efficiency was measured after 24 h. Northern blot membranes were sequentially hybridized with probes recognizing specific meiotic early (DMC1), middle (SPS1), and late (DIT1) transcripts to monitor meiotic progression. SR1 transcript and rRNA densitometry were used as loading control.
tion medium, to achieve full Dbf4 depletion at the beginning of meiosis. Any potential detrimental effect of doxycycline was ruled out by showing that a wild-type strain progresses through meiosis and sporulates normally whether in the presence or absence of the drug (Fig. 1B). We first determined the effect of shutting off one DBF4 allele in a heterozygous DBF4/pTET-DBF4 strain (Fig. 1C, middle). Northern blot analysis confirmed that DBF4 expression was reduced compared with wild type. Despite this, the kinetics of pre-meiotic DNA replication and the induction of meiosis-specific genes were normal indicating that one copy of DBF4 is sufficient for pre-meiotic DNA replication. However, the percentage of mature tetrads was decreased from 74% in the control strain to only 28%, suggesting that decreased DBF4 dosage has a more pronounced effect on sporulation than on S phase.
When both DBF4 alleles were shut off in a strain homozygous for the pTET-DBF4 construct, DBF4 mRNAs were no longer detectable, and importantly, no DNA synthesis was detected in these cells. This was not due to a failure to enter the meiotic program as early DMC1 transcripts accumulated at normal levels in these cells (Fig. 1C, right, and data not shown). However, middle and late transcripts did not build up and only 5% of asci were formed, suggesting a block in meiotic progression. These data demonstrate that DBF4 is required for pre-meiotic DNA replication.
Although unlikely, it is conceivable that Dbf4 might function in premeiotic S phase independently of Cdc7. To verify that Cdc7 itself is necessary for this process, we sought for temperatures at which DNA replication would be affected in a homozygous cdc7-1 ts strain but not in wild type. Hence we monitored DNA synthesis and sporulation efficiency at 25, 30, and 34°C, using the SK1 genetic background that is less heat-sensitive for meiosis. At 30°C, the cdc7-1 strain progressed more slowly through S phase compared with wild type, while at 34°C premeiotic DNA replication was affected also in the wild-type strain (Fig.  2). This suggests that the cdc7-1 strain is defective for DNA replication FIGURE 2. cdc7-1 mutants have a defect in pre-meiotic DNA replication and sporulation. Wild-type (ES1180) and cdc7-1/cdc7-1 (ES876) diploid strains were grown at 25°C, divided into three cultures, and shifted at 25, 30, and 34°C, respectively, 1 h after sporulation induction. Replication profile was analyzed by FACS analysis and sporulation efficiency was measured 24 h later. but retains some activity at 30°C. Again, sporulation was more affected than DNA replication, being completely defective at 30°C in the cdc7-1 ts strain. Thus both DBF4 and CDC7 seem necessary for DNA replication in meiosis, as during the vegetative cell cycle. Dbf4 Is Required for Completion of the First Meiotic Division-CDC7 was originally reported to be dispensable for pre-meiotic DNA replication but required for synaptonemal complex formation and commitment to recombination, possibly indicative of a role after S phase (11,25). To see whether such a function could be detected for Dbf4, we took advantage of the pTET-DBF4 strain to deplete DBF4 at different times along meiotic progression, which we describe hereafter as delayed Dbf4 depletion (Fig. 1A). Because the kinetics of meiosis depend heavily on the history and environment of the cell, all following experiments started with the same preculture, which was split only at the time of meiotic induction. Hence the consequences of shutting off DBF4 expression by adding doxycycline at time T 0 , T 2 h , and T 4 h after shift to sporulation medium were compared with a culture expressing DBF4 constitutively (No Dox). Ectopic DBF4 expression did not significantly alter pre-meiotic DNA replication and sporulation by 24 h was quite efficient, giving rise to 94% mature asci (37% dyads ϩ 57% tetrads). When doxycycline was added at T 0 or later, Dbf4 levels decreased but remained high enough to sustain efficient DNA replication in all cultures (Fig. 3, A and C). Sporulation, however, was seriously affected when doxycycline was added at T 0 but not at later times. Dbf4 became undetectable by 3 h, DNA replication was completed by 4 h, but 63% of the cells nevertheless failed to form mature spores. Fluorescence microscopy showed that most cells contained a single nuclear mass, indicative of an arrest before the first meiotic division (Fig. 3B). When doxycycline was added at T 2 h , Dbf4 levels decreased by 4 h, but this did not affect meiotic progression and sporulation. These results suggest that a critical level of Dbf4 is still required after pre-meiotic DNA replication to complete meiosis I but might not be required after this stage.
Delayed Dbf4 Depletion Arrests Cells after Pachytene-To better characterize the arrest induced by delayed Dbf4 depletion, we used a strain that in addition to pTET-DBF4 expresses SPC42-GFP and TUB1-CFP enabling us to follow SPB separation and spindle morphology throughout meiosis. Doxycycline was added at T 0 , which allowed S phase to take place, albeit slightly slower than without depletion (Fig. 4A). As above, most cells arrested before meiosis I, and after 24 h in sporulation medium, only 18% of mature asci were scored (DBF4 off ) compared with 88% when Dbf4 was expressed (DBF4 on ). The remaining cells could be subdivided in five classes depending on the presence of one, two, or four nuclear masses and on the extent of SPB separation (Fig. 4B). In the control culture (DBF4 on ), 85% of the cells became tetranucleated (class V) indicating completion of the two meiotic divisions in most cells. Conversely 60% of the cells (classes II and III) failed to progress beyond anaphase I after Dbf4 depletion (DBF4 off ). Class II (24%) showed duplicated but coalescent Spc42-GFP spots, whereas class III (36%) exhibited separated SPBs with a short spindle characteristic of metaphase I. 16% of the cells (class I) showed a single nuclear mass with a single SPB, probably representing cells that did not replicate. Finally, only 24% of the cells reached the binucleated or tetranucleated stage (classes IV and V) presumably representing cells in which Dbf4 depletion occurred too late. Thus the majority of cells (60%) depleted of Dbf4 after DNA synthesis arrest with separated SPBs (class II and III). This corresponds to a post-pachytene stage as ndt80⌬ mutants, which arrest at pachytene with duplicated but side-by-side spindle pole bodies (26), show only a single Spc42-GFP spot (data not shown).
Arrest Is Not Due to Incomplete DNA Replication or Faulty Recombination-Partial or defective replication is detected during meiosis by a replication checkpoint that stops cell division (21). Since S phase was slightly extended under conditions of delayed Dbf4 depletion, we envisaged the possibility that the meiosis I arrest was a consequence of activating the S phase or recombination checkpoints. Hence we tested whether this arrest was dependent on the DNA damage checkpoint genes RAD17, RAD24, and MEC1 (supplemental Fig. S1) or on the spindle checkpoint gene MAD2 (E2346, data not shown) and found that deletion of neither of these genes bypassed the arrest. To check that DNA replication went to completion, we quantified the amount of chromosomal DNA by fluorography using the more sensitive technique of pulse field gel electrophoresis (24). By this approach, only fully replicated chromosomes migrate in the gel, while replication and recombination intermediates remain trapped in the well. To circumvent the recombination problem we used a strain deleted for SPO11, which encodes the trans-esterase required for double-strand break formation and recombination (27). Chromosomes from spo11 mutants do not undergo meiotic recombination (28), allowing their separation by PFGE. The quantification of individual chromosomes resolved in PFGE showed a precise doubling, indicating that S phase was completed after delayed Dbf4 depletion, albeit slightly later than in the control culture (supplemental Fig. S2). Furthermore, the deletion of SPO11 did not bypass the meiosis I arrest induced by Dbf4 depletion, indicating that it is also not caused by defective or delayed recombination or subsequent DNA repair (Fig. 4C, right).
Delayed Dbf4 Depletion in the spo11⌬ Strain Produces Viable Spores-Despite the lack of recombination, spo11 mutants are able to complete meiosis and produce tetrads, even though the spores are non-viable due to random segregation of homologues at the first division (28,29). Depletion of Dbf4 in spo11 mutants produced only 17% of dyads, but surprisingly 87% of spores from these dyads were viable (Fig. 3C, right). Remarkably these spores turned out to be MATa/MAT␣ diploids, indicating that they were the product of an equational division. This result suggests that delayed Dbf4 depletion prevents mono-orientation of sister kinetochores at meiosis I, a phenotype reminiscent of that produced by the depletion of the polo-like kinase Cdc5 (30,31).
Deletion of REC8 Suppresses DBF4-dependent Arrest-The absence of Cdc5, like that of Dbf4, leads to a normal progression of the early pTET-DBF4, SPC42-GFP, TUB1-CFP diploid strain (E2377) was synchronized by centrifugal elutriation and put in SPM without (DBF4 on ) or with 20 g/ml doxycycline (DBF4 off ). A, FACS analysis and sporulation efficiency. B, cytological classification of arrest phenotypes after 24 h in SPM, without (DBF4 on ) or with 20 g/ml doxycycline (DBF4 off ). Cells were scored according to SPBs, spindle and nuclear morphologies as: mononucleated cells with unduplicated SPB (I), mononucleated cells with duplicated, but coalesced, SPBs (II), mononucleated cells with short spindles and two separated SPBs (III), binucleated cells with elongated spindle (IV), and tetranucleated cells (V). Bar ϭ 2 m. C, the proportion of cells having either one (F), two (f), or four (OE) masses of DNA was estimated by microscopic examination of DAPIstained cells. Spore viability was measured by dissecting isolated dyads on YPD plates and scoring colony formation after 1 week. The number of dissected dyads is indicated in parentheses. meiotic events but causes a failure at anaphase I. cdc5 mutants accumulate hypophosphorylated forms of the meiotic cohesin Rec8, which fails to be degraded (30,31) and prevents the proper resolution of chiasmata and the separation of homologues during meiosis I (32). In addition, Cdc5 depletion prevents the association of the monopolin Mam1 with kinetochores, causing an abnormal bi-orientation of sister chromatids at meiosis I. Considering the similar phenotypes induced by Cdc5 and Dbf4 depletion, we investigated whether a persistent cohesion could be responsible for the Dbf4-dependent arrest. If, as for Cdc5, loss of Dbf4 interfered with both cohesin cleavage and chromatid mono-orientation, we would expect chromosomes to fully segregate in Dbf4-depleted spo11 rec8 cells. Deletion of REC8 should allow cells to undergo a first division, but not efficient sporulation, as seen previously for rec8 mutants (33). This prediction was entirely satisfied by the experiment. The formation of bi-nucleated cells (meiosis I) in spo11⌬ rec8⌬ cells was no longer affected by the depletion of Dbf4 (Fig. 5A). This indicates that cells depleted for Dbf4 after S phase arrest before anaphase I mainly because they bi-orient their sister chromatids without being able to dissolve the cohesive links between them (Fig. 5B).

DISCUSSION
The meiotic division is a differentiation program precisely controlled by the action of protein kinases (34). Regulatory circuits check the occurrence of the specific meiotic division (meiose I) that should be coordinated with DNA replication and recombination to produce the correct segregation of homologous chromosomes into the gametes (35).
The present study demonstrates that Cdc7-Dbf4 is required during the early stages of meiosis to initiate DNA replication. These data contrast with earlier findings that suggested that CDC7 was not required for pre-meiotic DNA replication. This discrepancy could be due to the fact that initial investigations on meiotic functions of CDC7 used temperature-sensitive alleles (11). As meiosis itself is inhibited at elevated temperatures, the results obtained were likely to suffer a strong experimental bias. To avoid this caveat we inactivated Cdc7 by depleting its unstable regulatory subunit Dbf4. Using this approach we demonstrate that pre-meiotic DNA replication depends on Cdc7/Dbf4 activity (DDK). In budding yeast, pre-meiotic DNA replication is triggered by the same S phase CDKs that act in mitosis (18,20,21). These data combined with our results strongly suggest that mitosis and meiosis share a common mechanism for the activation of the origins of DNA replication.
In addition to the replication function, we show that Dbf4 has a second role during meiosis. Indeed, in the absence of Dbf4 cells arrest before the first meiotic division suggesting a possible role of Dbf4 in setting up the specific chromosome segregation during meiosis. Two experimental observations confirmed this hypothesis. First, although in the absence of Dbf4 most of the cells did not carry out anaphase, a few cells eventually divided their nuclei. However, they produced only two diploid spores instead of four haploid gametes. This indicated that an equational, instead of the normal reductional division, occurred. Second, deletion of Rec8 restored the nuclear division despite the absence of Dbf4, suggesting that the loss of Dbf4 may cause persistence of Rec8 cohesion. On the basis of these observations, we propose the following model (Fig. 5B). The absence of Dbf4 causes both kinetochore bipolar orientation and cohesion persistence between sister chromatides. Therefore, spindle fibers pull chromosomes to the opposite cell poles, while Rec8 cohesion persists, causing an arrest of chromosome separation. This arrest is bypassed by the deletion of REC8 that allows chromosomes to separate. This model can account for the arrest observed in the absence of Dbf4 and the failure of the subsequent anaphase. Our results thus give evidence that Dbf4 plays an unexpected but important role for the first meiotic division.
This new meiotic function for Dbf4 is in agreement with literature reports as paralogs and orthologs of DBF4 and CDC7 have been reported to play distinct biological functions in other organisms. In fission yeast, a second Cdc7-Dbf4-related complex (Spo4-Spo6) is required for chromosome segregation during the second meiotic division but dispensable in mitosis (36). A role distinct to replication initiation has been ascribed to the fission yeast Cdc7 ortholog Hsk1, specifically for the formation of centromeric heterochromatin and sister FIGURE 5. Deletion of REC8 bypasses the arrest caused by Dbf4 depletion. A, spo11⌬, rec8⌬, pTET-DBF4 strain (E2663) was shifted to SPM with (DBF4 on ) or without doxycycline (DBF4 off ). Cells having two (F) or four DNA masses (f) were scored by DAPI staining. B, proposed model for chromosome segregation and kinetochore attachments (arrows) following Dbf4 depletion alone (left), in combination with spo11⌬ (middle), or in combination with spo11⌬ and rec8⌬. In the absence of Dbf4 (left panel) and Spo11 (middle panel) bipolar orientation of sister kinetochores and persistent cohesion impairs chromosome segregation. Deletion of REC8 (right panel) allows sister chromosomes to separate. chromatid cohesion (37,38). A Dbf4-related protein, Drf1, involved in the progression of both S and M phases as well as in checkpoint responses, was isolated in human cells (39,40) and in Xenopus (41). It has also been proposed that the Xenopus homolog of Dbf4 is required for heart development independently of its function in DNA replication (42). No Cdc7 or Dbf4 paralogs have been described in budding yeast so far, making plausible a dual role for Dbf4 in meiosis.
The phenotypes that we described for the loss of Dbf4 function (i.e. persistent cohesion and kinetochore bi-orientation) show a strong resemblance with cdc5/polo mutant phenotypes during meiosis (30,31). We propose that Dbf4 might be necessary for Cdc5/Polo function. Pololike kinases are characterized by the presence of two phospho-dependent substrate-targeting modules, the polo boxes, in their non-catalytic C-terminal domain (43). Polo boxes recognize S-[pS/pT]-P/X motifs in pre-phosphorylated substrates, and by doing so, both unmask the catalytic domain and recruit the kinase to specific subcellular structures. Thus the biological activity of polo kinases is believed to rely on "priming kinases" that pre-phosphorylate their substrates. The Cdc7 kinase is known to preferentially phosphorylate S-S/T-P motifs. Because the meiotic phenotypes of depleting Cdc5 and Dbf4 are so similar, it is tempting to speculate that Cdc7-Dbf4 may act as the priming kinase for Cdc5 providing a link between DNA replication and chromosome segregation during meiosis.