Mcm7, a subunit of the presumptive MCM helicase, modulates its own expression in conjunction with Mcm1.

The Saccharomyces cerevisiae Mcm7 protein is a subunit of the presumed heteromeric MCM helicase that melts origin DNA and unwinds replication forks. Previous work showed that Mcm1 binds constitutively to the MCM7 promoter and regulates MCM7 expression. Here, we identify Mcm7 as a novel cofactor of Mcm1 in the regulation of MCM7 expression. Transcription of MCM7 is increased in the mcm7-1 mutant and decreased in the mcm1-1 mutant, suggesting that Mcm7 modulates its own expression in conjunction with Mcm1. Indeed, Mcm7 stimulates Mcm1 binding to the early cell cycle box upstream of the promoters of MCM7 as well as CDC6 and MCM5. Whereas Mcm1 binds these promoters constitutively, Mcm7 is recruited during late M phase, consistent with Mcm7 playing a direct role in modulating the periodic expression of early cell cycle genes. The multiple roles of Mcm7 in replication initiation, replication elongation, and autoregulation parallel those of the oncoprotein, the large T-antigen of the SV40 virus.

The MCM genes, including MCM1 and the family of related genes, MCM2, MCM3, MCM4, MCM5, MCM6, and MCM7, were originally identified as genes essential for minichromosome maintenance (1,2). However, although all mcm mutants show defects in DNA replication, the molecular basis for their replication defects may be different. Mcm1 (3) belongs to a family of combinatorial transcription factors known as the MADS box proteins found in all eukaryotes (4,5). These proteins regulate the expression of diverse genes by cooperative interaction with multiple cofactors. In multicellular eukaryotes, MADS box proteins such as Agamous, Deficiens, and SRF (6 -9) are homeotic proteins that play an important role in coordinating gene expression and cell proliferation during development (10 -12). In the budding yeast, Mcm1 appears to play an important role in coordinating gene expression during cell cycle progression (13) and in cell type determination (14,15). Mcm1 has been shown to regulate cell cycle-specific genes such as CLB1, CLB2, and SWI5 in association with SFF/Fkh2 (16,17). It also regulates the M/G 1 -specific transcription of early cell cycle genes including CLN3, SWI4, and the DNA replication genes, CDC6 and MCM7 (18), in association with Yox1 and Yhp1 (19). A survey of transcription factors involved in cell cycle-regulated gene expression placed Mcm1 at the top of a serial regulatory scheme (20). Mcm1 also plays a direct role in DNA replication by binding to replication origins (21). A recent study showed that Mcm1 binds multiple sites flanking the minimal functional domains of replication origins in vitro. The association of Mcm1 with replication origins was also demonstrated in vivo by chromatin immunoprecipitation experiments. Thus, Mcm1 may regulate the process of DNA replication as well as the expression of DNA replication genes.
Like Mcm1, the MCM2-7 family of proteins is evolutionarily conserved from yeast to human (2). Several studies suggest that these six proteins form the precursor complex of the hexameric helicase (22)(23)(24), which plays an essential role in the early steps of replication initiation (25)(26)(27). The Mcm2-Mcm7 proteins are assembled as components of the pre-replication chromatin during early G 1 phase. In coordination with this cell cycle-regulated program, all six MCM genes are expressed in a cell cycle-specific manner, with their transcript levels reaching a peak at M/G 1 or G 1 phase (13, 28 -30). However, the notion that the periodic synthesis of these proteins might limit DNA replication to once per cell cycle conflicts with their extreme abundance and constitutive presence in proliferating cells (31)(32)(33)(34). A possible explanation for these contradictory properties is that the MCM2-7 proteins may perform other functions that involve separate pools of these proteins (2). This hypothesis is consistent with the observation that individual members of the MCM2-7 family may undergo various forms of post-translational modifications including phosphorylation (33,35), polyubiquitination (36), and acetylation (37). Moreover, genome wide location studies indicate that individual members of the MCM2-7 family are localized not only at replication origins but also at nonorigin regions and that these proteins may not act in unity (38).
The possibility that Mcm1 may play an indirect role in DNA replication by regulating the expression of DNA replication genes has been investigated (18,29,39). In this study, we confirm that Mcm1 is involved in the transcriptional regulation of MCM7. However, we found that a cofactor of Mcm1 in this regulation is Mcm7, suggesting that Mcm7 may play a dual role in DNA replication by acting as part of the MCM helicase as well as temporally regulating its own expression.

MATERIALS AND METHODS
Plasmids and Strains-Plasmids used are: YCp101A (ARS1 CEN5 ADE3), YCp101 (ARS1 CEN5), YCp120 (ARS120 CEN5), and YCp121 (ARS121 CEN5); pSH18 -34 (lexAop-lacZ). pYES.Mcm1-His 6 was constructed by inserting the MCM1 ORF 1 into the pYES2.1 TOPO vector (Invitrogen) to create a galactose-inducible His 6 C-terminal tagged Mcm1 fusion. pMCM1-GST was constructed as follows. The vector * This work was supported in part by National Institutes of Health Grant GM34190. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  pGEX-2T (Amersham Biosciences) was first modified by filling in the ends of the BamHI-cut vector and adding a NdeI 8-mer linker to generate pGEX-2TN. The MCM1 gene was constructed with a NdeI site at the ATG start codon using site-directed mutagenesis. The construct was inserted as a NdeI-ClaI fragment into the pGEX-2TN plasmid at NdeI and SmaI. pMCM7-GST was constructed by cloning the MCM7 ORF in frame to GST within pGEX-2T.
RNA Methods-RNA was purified using the yeast heat-freeze protocol (40). 10 g of total RNA was used for Northern blots (41). Radioactive probes were produced using the STRIP-EZ TM kit (Ambion, Austin, TX). Quantitation of radioactive signals was performed on a Molecular Dynamics Storm 840 PhosphorImager, and quantitative analysis was performed using ImageQuant software.
Protein Purification-Mcm1-His 6 was purified from 500 ml of yeast culture induced for 5 h with 2% galactose. After 5 h of induction, cells were pelleted and Mcm1-His 6 was purified using the Y-PER His 6 purification kit according to the protocol from the manufacturer (Pierce). Peak fractions of Mcm1-His 6 were pooled and dialyzed against 2 liters of minimal Mcm1 band shift buffer (20 mM HEPES-KOH, pH 7.5, 1 mM dithiothreitol, 10% glycerol, 60 mM NaCl, 7 mM MgCl 2 , 1 mM EDTA).
DNase I Footprinting-DNase 1 footprinting was performed as described for the band shift reactions. Purified proteins and DNA were mixed in buffer 1 in a 0.2-ml volume for 20 min. A pre-calibrated amount of DNase I (Sigma) was added in a volume of 0.01 ml for 1 min at room temperature. The reaction was stopped by precipitation in a mixture of ethanol, 0.3 M sodium acetate, and 1 g of linear acrylamide. Precipitate was washed with 70% ethanol, dried, and re-suspended in formamide buffer before loading onto a 6 M urea, 6% polyacrylamide sequencing gel.
Formaldehyde Cross-linking Immunoprecipitation-Formaldehyde cross-linking was performed as described by Hecht et al. (42) with minor modifications. Strains were grown in complete medium in 10 mM phosphate buffer at pH 7.5, to A 600 of 0.6 -1.0. Between 15 and 50 ml of cells were cross-linked and then immunoprecipitated with 1.0 -2.5 l of specific antibodies overnight at 4°C on a shaker. All washing steps were performed in Bio-Rad microspin columns. The collected immune complexes were washed two times in lysis buffer (50 mM HEPES/KOH, pH 7.5, 140 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate), two times in lysis buffer with 0.5 M NaCl, two times in the final wash buffer (10 mM Tris-HCl, pH 8.0, 0.25 M LiCl, 0.75% Nonidet P-40, 0.5% sodium deoxycholate, 1 mM EDTA), and two times in TE (10 mM Tris-HCl, pH 8.0, 1 mM EDTA). Protein-DNA cross-links were reversed overnight at 65°C. DNA samples were purified by the addition of five volumes of PB buffer and then passed through Qiagen PCR purification columns (Qiagen, Valencia, CA). Purified DNA was eluted with 40 l of TE.
PCR Amplification-PCR was carried out using the following conditions: 2 times (94°C for 2.0 min, 96°C for 1.0 min, 54°C for 4.0 min), 24 times (94°C for 1.0 min, 54°C for 1.5 min), and 1 time (70°C for 5.0 min). 5 Ci of [␣-32 P]dATP was added to each PCR set up as a multiplex reaction. Each PCR contains the ATP11 primers (negative control) in addition to primer pairs for the MCM7, MCM5, CDC6 promoters or ARS305, respectively. The primer pair for ATP11 is directed at the coding region of the gene. The sizes of the PCR products for MCM7, MCM5, CDC6, ARS305, and ATP11 are 387, 226, 261, 280, and 412 bp, respectively. Quantitation of radioactive signals was performed on a Molecular Dynamics Storm 840 PhosphorImager, and quantitative analysis was performed using ImageQuant software. Unlike real-time PCR, time points with high signals may be underrepresented in magnitude because these values may have been generated beyond the exponential phase of the PCR reaction.
Other Methods-Plasmid stability assay is described by Lei et al. (32). Flow cytometric analysis was performed as described elsewhere (43). ␤-Galactosidase assays were performed as described elsewhere (44).

RESULTS
Dosage Suppression of mcm1-1 by MCM7 and mcm7-1 by MCM1-To investigate the molecular basis of the minichromosome maintenance defect of the transcriptional activator Mcm1, we searched for dosage suppressors that alleviate the Mcm Ϫ defect of mcm1-1. MCM7 was identified as a dosage suppressor that stabilized minichromosomes in the mcm1-1 strain. 2 A tester plasmid YCp101A containing ARS1 is unstable in the mcm1-1 mutant (Fig. 1A). This minichromosome loss defect is complemented by MCM1 and partially suppressed by MCM7 carried on a high copy plasmid. Moreover, MCM1 on a high copy plasmid also partially suppresses the temperaturesensitive growth defect of mcm7-1 (Fig. 1B), which arrests growth at 37°C in late S phase with almost two genome equivalents of DNA (data not shown). This partial suppression of mcm1-1 by MCM7 and mcm7-1 by MCM1 is not gene-specific. Previous work has shown that overexpression of CDC6 suppresses a mcm1 mutation (29). A high copy plasmid containing MCM1 also partially suppresses the temperature-sensitive defects of other mcm mutants such as mcm2-1 and mcm3-10, indicating that MCM1 dosage suppression is not specific to mcm7-1. There are two likely explanations for the partial dosage suppression of mutant phenotypes of mcm1-1 by MCM7 and mcm7-1 by MCM1 that may also explain the relationships between Mcm1 and other replication genes. First, Mcm1 regulates the expression of MCM7. Increasing the dosage of the regulator or the regulated product would increase the level of Mcm7. Second, dosage suppression may result from the stabilization of a complex formed between Mcm1 and Mcm7 by increased concentration of one of the interacting partners. An example is the suppression of the mating defects of mat␣1 and mat␣2 mutants by overexpression of Mcm1 (15).
To explore these possibilities, we compared the transcript levels of MCM7 in the mcm1-1 mutant and wild type strains ( Fig. 1C, a). In parallel, we also compared the transcript level of MCM7 in the mcm7-1 strain (Fig. 1C, b). We found that the MCM7 transcripts are reproducibly reduced by approximately half in the mcm1-1 strain but increased by ϳ2-4-fold in the mcm7-1 mutant (see also Fig. 4, B and C). The effect of mcm7-1 on MCM7 expression is unexpected, suggesting that Mcm7 may also be involved in the regulation of its own transcription. We also investigated the functional relationship between Mcm1 and Mcm7 in the mcm1-1 mcm7-1 double mutant. We found that mcm1-1 partially suppresses the growth defect of mcm7-1 at 34.5°C, supporting the notion that these two mutations may exert compensatory effects (Fig. 1D).
To investigate whether small changes in the transcript level of MCM7 would have a measurable effect on DNA replication, the stability of minichromosomes in diploids containing either one or two copies of MCM7 was examined. Northern blot analysis confirmed that there is a 2-fold difference in the transcript level corresponding to the number of copies of MCM7 in these two strains (data not shown). The stability of three different minichromosomes, each containing a different replication origin, was analyzed. Minichromosomes containing ARS1 and ARS120, but not ARS121, showed an increased loss rate in the diploid containing a single copy of MCM7, compared with the diploid containing two copies of MCM7 (Fig. 1E). This haploinsufficiency of MCM7 in the maintenance of minichromosomes containing selective ARSs is similar to that observed in a heterozygous MCM2/⌬ diploid (32).
Although the MCM2-MCM7 genes are transcribed periodically, reaching a peak during M/G 1 phase (13,28), their protein products are present constitutively at a constant level throughout the cell cycle (33). To investigate whether reduced MCM7 transcript level or gene dose has an effect on the Mcm7p level in the mcm1-1 mutant and the MCM7/mcm7⌬ diploid, Mcm7 protein levels in mutant and wild-type extracts were compared by Western blotting. There was no measurable difference in the Mcm7p level between the mcm1-1 mutant and parent strain (Fig. 1F). Similarly, no measurable difference in the Mcm7p level was observed in the MCM7/mcm7⌬ and the MCM7/ MCM7 diploid (Fig. 1G). These results suggest that small changes in the MCM7 transcript level and therefore the nascent Mcm7p pool may produce significant effects on the stability of plasmids carrying specific ARSs. However, the bulk of the cellular Mcm7 protein whose accumulation is independent of gene dose or periodic expression has little effect on plasmid stability.
Mcm7 Stimulates Mcm1 Binding to the MCM7 Promoter-Although Mcm7 is not known to be a transcription factor or to bind DNA, Mcm1 has been shown to bind sequences upstream of the MCM7 promoter (18). To investigate whether Mcm7 binds either alone or together with Mcm1, a 60-bp DNA fragment upstream of the MCM7 promoter was used as the binding substrate in an electrophoretic mobility shift assay. Mcm1 tagged either with His 6 ( Fig. 2A, a) or with GST (b, lane 2), and Mcm7-GST proteins ( Fig. 2A, b, lane 1) were purified to greater than 90% homogeneity. Full-length Mcm1 migrates anomalously with an apparent molecular mass of ϳ42 kDa (3). The concentrations of purified proteins were calibrated against BSA in Coomassie-stained polyacrylamide gels. We found that Mcm1-GST and Mcm1-His 6 can be used interchangeably to produce similar results.
To investigate whether this protein-DNA complex results from a specific interaction among Mcm1, Mcm7, and MCM7 promoter DNA, we carried out the binding reaction in the presence of a 50-fold excess of cold competitor DNA (Fig. 2D). To investigate whether the stimulatory effect of Mcm7 is specific for Mcm1, we examined the effect of Mcm7 on the binding of another transcription factor AP2 to its recognition sequence. AP2 binds specifically to the AP2 binding sequence (Fig. 2E, lane 2) (Fig. 2G, a, lane 1, bar to the left) (18). Although Mcm7 alone does not bind MCM7 DNA in a mobility shift assay (Fig. 2F), at a saturating concentration, Mcm7 protects the AT-rich region between the two MCEs ( Fig. 2G, a, lanes 2 and 3, bar to  The MCM7 Promoter Region Is Not Associated with a Replication Origin-Mcm7 is a subunit of the hexameric complex that is associated with replication origins. To investigate whether the association of Mcm7 with its own promoter may be the result of a fortuitous positioning of a replication origin in the promoter region of MCM7, we tested the promoter region of MCM7 for replication origin activity. The 1.6-kb intergenic region between MCM7 and its 5Ј upstream neighbor DER1 was inserted into YIp56, a non-replicating vector, and the resulting plasmid was tested for autonomous replication in a high frequency transformation assay (Fig. 3A). YIp56-MCM7-P DNA yielded only abortive transformants that do not survive restreaking, whereas YIp56-ARS1 yielded true transformants at a high frequency (Ͼ1000 transformants/g DNA). This result suggests that the intergenic region between MCM7 and DER1 does not contain a replication origin. To show a direct correlation between the presence of Mcm1 and Mcm7 at the promoters and their effect on the expression of the early cell cycle genes, we examined the transcript levels of MCM7, MCM5, and CDC6 in the mcm7-1 (Fig. 4B, a) and mcm1-1 strains (Fig. 4B, b). The transcript levels of these genes are increased by 2-4-fold in the mcm7-1 strain (Fig. 4C, a). In contrast, transcript levels of these genes are reduced by about half in the mcm1-1 strain (Fig. 4C, b). These results suggest that the mutant forms of these proteins exert opposing effects; mcm1-1 depresses and mcm7-1 stimulates ECB gene expression. This interpretation is consistent with the partial suppression of mcm7-1 by mcm1-1 in the mcm1-1 mcm7-1 double mutant (Fig. 1D).
Recruitment of Mcm7 to the ECB Promoters Occurs after the Onset of Anaphase-The early cell cycle genes MCM7, MCM5, and CDC6 are expressed periodically during the M/G 1 phase or early G 1 phase. To correlate recruitment of Mcm7 to the MCM7 promoter with MCM7 expression, we followed these activities in a time course traversing the M/G 1 phase. In addition, recruitment of Mcm7 to the promoters of MCM5 and CDC6 was also analyzed. Because Mcm7 is a subunit of the Mcm2-Mcm7 complex that is recruited to replication origins from late M to early S phase (25,26), the association of Mcm7 with ARS305 DNA was used as a control.
An effective method for synchronizing yeast cells during the M/G 1 phase is to deplete cells of Cdc20, an activator of the anaphase-promoting complex using the ⌬cdc20 (GAL-CDC20) strain (46). Fluorescence-activated cell sorting analysis shows that cells that have been arrested and then released from the cdc20 block resume traversing the M and G 1 phase in synchrony up to about 105 min after release (Fig. 5A). Based on DNA content, Ͼ90% of the cells are in M phase 60 min after release from block. At 75 min, ϳ50% of the cells contain one genome-equivalent of DNA, indicating the beginning of the G 1 phase. At 90 and 105 min, Ͼ90% cells are in G 1 phase. Cell morphology indicates that synchronous growth breaks down at about 105 min after release (data not shown). . Anti-HA antibodies were used as a negative control (lane 1). Cross-linked DNA was PCR-amplified for MCM7, MCM5, CDC6, and ATP11 promoter sequences and analyzed in agarose gel. 0.01% and 0.003% of input DNA from log phase cell extract, PCR-amplified for the respective promoter DNA sequences, were loaded in lanes 4 and 5, respectively. B, a, Northern blots of MCM7, MCM5, and CDC6 transcripts in the wild type and mcm7-1 strain at 30 and 37°C. Cells were shifted to 37°C for 5 h for MCM7 and MCM5 and 2 h for CDC6. b, Northern blots of MCM7, MCM5 and CDC6 transcripts in the isogenic wild type and mcm1-1 strain at 30°C. ACT1 and PDA1 transcripts are used as loading controls. C, a, ratios of the transcript levels of MCM7, MCM5, and CDC6 at 37 and 30°C in the mcm7-1 mutant and wild type strains described in B. Transcript levels were normalized against that of ACT1. b, ratios of the transcript levels of MCM5, CDC6, and MCM7 in the mcm1-1 and wild type strain normalized against PDA1 transcripts.
The transcript level of MCM7 was quantified in a Northern blot (Fig. 5B, panel 1) and expressed as a ratio of MCM7 RNA: ACT1 RNA (Fig. 5C, panel 1). The transcript level of MCM7 peaks at two different points, M phase (Fig. 5B, panel 1, 0 min) and late G 1 phase (ϳ105 min). The MCM7 transcript level oscillates as much as 8 -10-fold at its peak and trough during the course of the cell cycle. The association of Mcm7 with the ECB promoters and ARS305 was measured as well by chromatin IP (Fig. 5B, panels 2-5). Cross-linked DNA was PCR-amplified with [␣-32 P]dATP and analyzed by PAGE. The amount of Mcm7 associated with the different promoter DNAs is quantified using a PhosphorImager and plotted as the ratio of the PCR-amplified signal of each of the promoter DNA to that of the ATP11 coding region (Fig. 5C, panels 2-5). The strongest association of Mcm7 with the ECB promoters occurs at the trough of the bimodal expression of MCM7 after the initiation of anaphase. This pattern is reproducibly observed for Mcm7 at the MCM7 (panel 2), MCM5 (panel 3), and CDC6 (panel 4) promoters in independent experiments. In contrast, Mcm7 is recruited to ARS305 in anaphase or telophase (Fig. 5, B and C, panels 5), slightly after its recruitment to the ECB promoters, and remains associated with ARS305 throughout the G 1 phase.
Mcm3 Is Also Recruited to ECB Promoters-To investigate whether other subunits of the MCM2-7 complex are also recruited to the ECB promoters, we carried out chIP using antibodies specific to Mcm7 and Mcm3 (Fig. 6A). Extracts were prepared from cells arrested in nocodazole, ␣ factor, and hydroxyurea, respectively, to produce cell populations that are arrested in different stages of M, G 1 , or S phase. Our results indicate that Mcm3 and Mcm7 have very similar cell cycle localization patterns at respective ECB promoters. Like Mcm7 (Fig.  6B), Mcm3 (Fig. 6C) is localized at the promoters of MCM7 and MCM5 during M and G 1 phase and slightly delayed at the CDC6 promoter. The temporal localization patterns of Mcm7 in cells synchronized by chemicals (Fig. 6) or by Cdc20 depletion (Fig. 5) are consistent within the resolutions of these approaches. These  (29). Transcript levels of these genes reach a peak at M phase and again at late G 1 phase. The bimodal transcription is believed to result from the coordinated action of multiple regulatory factors to fine tune expression of early cell cycle genes in response to multiple transducing signals (50). An examination of the MCM7 promoter sequence indicates that in addition to the tandem ECBs where Mcm1, Yox1, Yhp1 (19), and Mcm7 bind, a conserved early G 1 element known as the SCB (Swi4/6 cell cycle box) is located between the ECBs and the translational start codon. SCBs are regulatory elements for early G 1 gene expression (29,51,52 Mcm7 is recruited to the ECB promoters after the initiation of anaphase upon release from the cdc20 block (Fig. 7A). Recruitment of Mcm7 to replication origins occurs ϳ30 min later at anaphase or telophase (Fig. 7B). Neither Mcm7 alone nor the hexameric complex (53, 54) exhibits DNA binding activity, yet both are recruited by pre-assembled factors: Mcm1 at promoters and other components of the pre-RC at replication origins (55). Both Mcm7 and the MCM complex are released from their respective binding sites after execution of their respective functions (25,26). In both cases, re-assembly at promoters or replication origins is prevented until the next cell cycle.
A number of reports support the view that Mcm7 as well as other subunits of the MCM complex may participate in tran- scription regulation in mammals. Mcm7 directly interacts with the retinoblastoma tumor suppressor protein, for example (56). Mcm7 also interacts with MAT1, the targeting factor for the CAK kinase, substrates of which include the cyclin-dependent kinases as well as components of the transcription machinery such as the C-terminal domain of RNA polymerase II (57). Stat␣1, an interferon-␥-responsive transcription factor, recruits a Mcm3/Mcm5 subcomplex through direct interaction with Mcm5 in interferon-␥-induced gene expression (58). In yet another study, subunits of the MCM complex have been shown to associate with the RNA polymerase II holoenzyme (59). Our results suggest that both Mcm7 and Mcm3 are recruited to the promoters of some ECB genes in a cell cycle-specific manner in the budding yeast. It is possible that members of the Mcm2-Mcm7 family other than Mcm7 may also be involved in the regulation of ECB gene expression.
Subtle Effects in Transcription May Have Significant Effects on Replication Initiation-A 2-fold reduction in the transcript levels of MCM7 has a dramatic effect on minichromosome maintenance. This observation is incongruent with the abundance of Mcm7 (ϳ30,000/cell) (34), which far exceeds the number of replication origins (ϳ400/cell) in a yeast cell (38,60). Furthermore, this extreme abundance of Mcm7 protein level appears not to be influenced by the fluctuation of the MCM7 transcript levels (31). However, dosage effects are common among some of the MCM2-7 proteins (32). That a small change in the total pool of Mcm7 may produce a large effect on DNA replication is best explained if a distinct pool of Mcm7 is targeted for DNA replication. A recent study showed that a small subset of the Mcm3 population is polyubiquitinated during M phase before the establishment of the pre-replication complex (36). In that scenario, a small nascent pool of the Mcm3 protein that is transcribed and translated in a cell cycle-dependent manner may be targeted for replication initiation. Autoregulation of replication initiation factors effectively limits DNA replication to a narrow window by producing a burst of the initiation factors just prior to replication initiation. This strategy is used in the regulation of the E. coli DnaA protein (61) and the SV40 large T antigen (62,63).
Coordination of Gene Expression and Cell Proliferation by Common Factors-There are two issues that a eukaryote cell must contend with in cell growth and cell proliferation: the execution of DNA replication and gene expression, and the coordination of these two processes. Here, we show that Mcm7, a subunit of the putative replicative helicase that is involved in the execution of DNA replication, may have a role in the expression of DNA replication genes. Similarly, Mcm1, which regulates expression of a number of replication initiation factors, also binds replication origins (21). By involving the same protein factors in the execution of both of these processes, cell growth and cell proliferation may be coordinated. A paradigm of this complex scheme can be found in the large T antigen of the SV40 virus, a successful parasite that has evolved to infiltrate the infrastructure of its host (64). Of particular relevance, as a smaller unit, the large T antigen is a transcription factor (65); as a hexamer, it is a replicative helicase (66). Understanding the dynamics between the hexameric MCM helicase and its subunits in the regulation of DNA replication and gene expression will provide insights into the coordination of cell growth and cell proliferation in eukaryotes.