Cytometry of chromatin bound Mcm6 and PCNA identifies two states in G1 that are separated functionally by the G1 restriction point1
© Frisa and Jacobberger; licensee BioMed Central Ltd. 2010
Received: 1 September 2009
Accepted: 16 April 2010
Published: 16 April 2010
Cytometric measurements of DNA content and chromatin-bound Mcm2 have demonstrated bimodal patterns of expression in G1. These patterns, the replication licensing function of Mcm proteins, and a correlation between Mcm loading and cell cycle commitment for cells re-entering the cell cycle, led us to test the idea that cells expressing a defined high level of chromatin-bound Mcm6 in G1 are committed - i.e., past the G1 restriction point. We developed a cell-based assay for tightly-bound PCNA (PCNA*) and Mcm6 (Mcm6*), DNA content, and a mitotic marker to clearly define G1, S, G2, and M phases of the cell cycle. hTERT-BJ1, hTERT-RPE-1, and Molt4 cells were extracted with Triton X-100 followed by methanol fixation, stained with antibodies and DAPI, then measured by cytometry.
Bivariate analysis of cytometric data demonstrated complex patterns with distinct clustering for all combinations of the 4 variables. In G1, cells clustered in two groups characterized by low and high Mcm6* expression. Serum starvation and release experiments showed that residence in the high group was in late G1, just prior to S phase. Kinetic experiments, employing serum withdrawal, and stathmokinetic analysis with aphidicolin, mimosine or nocodazole demonstrated that cells with high levels of Mcm6* cycled with the committed phases of the cell cycle (S, G2, and M).
A multivariate assay for Mcm6*, PCNA*, DNA content, and a mitotic marker provides analysis capable of estimating the fraction of pre and post-restriction point G1 cells and supports the idea that there are at least two states in G1 defined by levels of chromatin bound Mcm proteins.
We have been interested in multi-variate cytometry of the mammalian somatic cell cycle. The cell cycle can be analytically divided into a variable number of states based on the correlated levels of biochemical activities provided that the level of each activity changes in a repeating pattern and that the correlated activity patterns are not identical. For the simplest case of any single variable there are essentially 5 states to consider per oscillation: initial basal level (1), increase or decrease from basal (2), maximum or minimum (max, min) (3), and decrease or increase (4) to final basal level (5), which is not biologically equivalent to initial basal. For example, the expression of cyclins E1, A2, and B1 oscillate once per cycle, out of phase with each other. All are at "initial basal" in early G1; begin expression at different times and rise at different rates; reach max at G1/S (E1) or G2/M (A2, B1); decrease in S (E1), prometaphase (A2), or anaphase (B1); are resident at "final basal" approximately in G2 (E1), metaphase (A2), or telophase (B1) [1–4]. These expression patterns are readily discerned by immunofluorescence coupled with DNA content in flow cytometry assays . Thus, measurement of the cyclins, DNA, and a mitotic marker constitute highly informative analyses of cell cycle transition states characteristic of a specific population of somatic cells [e.g., [6–8]].
Because mitosis is characterized by abrupt, specific, sequentially timed proteolysis of substrates of the anaphase promoting complex/cyclosome (APC/C)  and an abrupt increase in kinase activities and phosphorylation of many substrates, a multi-variate cell-based approach to cell cycle analysis subdivides G2 and M in a straightforward manner [e.g., ]. G1 is also characterized by oscillating activities, but has not been analyzed cytometrically in the same manner as M. Here, we wished to extend to G1 this type of multi-variate analysis. One fundamental G1 sub-division is the kinetically defined uncommitted and committed states. The transition between these states has been labeled "R" (restriction point) by Yen and Pardee . The exact biochemical concept or nature of R is unknown, although most investigators would agree with a complex model that integrates signaling (growth and nutrition factors; cell-substrate attachment; cell-cell contact, and cell damage) at a modular level, with an "R" module containing at least the activities of D cyclin/Cdk complexes and the Rb/E2F family of transcription factors [10–14]. For a mathematical model that formally integrates a large body of information, see Novak and Tyson . For a dissenting view see .
Without an exact biochemical definition of "R", it may be possible to quantify cells that are in a pre-R or post-R state (~G1-pm and G1ps subphases of Zetterberg, ) based on measurements of specific Rb family phosphorylations; the levels of E2F transcription factors that are bound to promoters and not bound by Rb family proteins, and the levels of D and E cyclins. Previous work by Juan et al. has shown the feasibility of this approach with an antibody that binds "hypo-phosphorylated" Rb . At this time, a comprehensive analysis would be difficult to achieve since there are 3 Rb family proteins, three D cyclins using 2 cyclin dependent kinases (Cdk4, Cdk6), two E cyclins using Cdk2, and 8 E2F and 2 DP protein subunits that constitute E2F transcription factors as well as at least 5 protein inhibitors of G1 Cdks. Since much of this operates as a function of specific phosphorylations, this might be rendered simpler with probes to specific phosphorylated sites on a subset of these molecules. However, these probes are not generally available and may or may not function well in cytometric assays [e.g., see ].
In place of the more intricate and sophisticated approach outlined above, we have asked whether the bimodal distributions shown by bivariate analyses of chromatin-bound minichromosome maintenance (Mcm) proteins and DNA content marks pre- and post-R G1 cells. These patterns were first shown by Friedrich and co-workers  and are based on measuring the residue of Mcm proteins that are left behind after detergent extraction. In support of this idea, Mukherjee at al. have correlated the committed period of G1 with high levels of bound Mcm protein, hyper-phosphorylated Rb, and cyclin E expression . The approach employed by Freidrich et al. (extraction then fixation) was first described by Kurki et al.  for detecting S phase specific 'tightly' bound PCNA that was a subset of total PCNA detected by denaturing fixation. The logic is that extraction of live cells with detergent depletes cells of 'loosely' bound proteins and other molecules that can diffuse out, leaving large structures and tightly bound proteins, including those bound to chromatin. This is also close to the standard methods of creating chromatin pellets in the primary research on replication complexes.
Normal mammalian cell replication involves mechanisms to bias faithful duplication of the genome once per mitotic cell cycle. The sub-system responsible for preventing re-replication consists of a protein complex that is built on origins-of-replication (ORI) and "licensed" for initiation of DNA synthesis. A licensed pre-replication complex (pre-RC) is built on existing complexes of ORC proteins by sequential binding of Cdc6 then Cdt1/RFL-B, which then "loads" Mcm proteins. Licensing is complete after the 6 Mcm proteins are loaded onto chromatin as a functional but inactive hexameric complex. Cdt1 is rate limiting for this process. In replicating cells, licensing occurs as a continuous process from late mitosis, after loss of mitotic Cdk activity, through G1 [20, 21].
In higher eukaryotes somatic cells, the rates of licensing are not known, however, one model for actively dividing cells (based on a zone within the Chinese hamster DHFR locus) suggests that licensing begins in mitosis but origin site specification occurs in G1 . Models for cell cycle re-entry suggest that quiescent cells lack bound Cdc6 and Mcm proteins and therefore license in late G1 [e.g., [18, 23]]. On cell cycle re-entry, Mcm loading appears to be dependent on cyclin E and Cdc7 expression and activities , which peak late in G1. The rate-limiting step for initiation may be unwinding of DNA by the helicase activity of the complex composed of Mcm2 through 7 . Once DNA synthesis has been initiated, irrevocable rescinding of the license occurs by several mechanisms, including removal of Mcm proteins from the chromatin, that prevent reassembly of pre-RC in S phase. Reloading does not occur until the licensing process resumes in mitosis in cycling cells or in G1 in stimulated quiescent cells. This system has been reviewed in detail [e.g., see [20, 21, 24]].
The reported cytometric pattern of Mcm2 and DNA content for CV-1 cells provides a visualization of Mcm binding in G1 and Mcm removal from chromatin during S phase . The expression of detergent resistant Mcm2 was bimodal in G1 with low and high expressing cell clusters. The high expression cluster level was coincident with the G1/S border, suggesting that cells enter S phase only after the bound Mcm complexes have reached a maximum level (max). In the published patterns, the levels in S decreased to a minimum at the S/G2 border consistent with the known cell cycle related expression [20, 21, 24, 25].
Since origins of replication are not well mapped in higher eukaryotes, the timing of licensing and the rate of Mcm loading in actively dividing cells has not been comprehensively studied. The work of Friedrich et al. suggests that G1 cells cluster into two distinct groups with the highest cluster ~10 fold greater than the lower cluster. Since the frequency of cells existing in each cluster is proportional to the time spent at that state, CV-1 cells may pass through two G1 loading periods with a rapid transition between the two states. However, this type of pattern would also be consistent with exponential loading. While over-interpretation is not useful, the quantitative bimodality of G1 cells, if substantiated, creates a biochemically related division of G1 that could be the beginning of subdividing G1 into meaningful biochemical states that would increase the power of multi-variate analyses of the cell cycle.
Since the simplest interpretation of the CV-1 G1 patterns are an early and late subdivision, our goals were to 1) determine whether the same pattern exists in human somatic cells, and 2) determine whether the low and high clusters correlate with pre- and post-restriction point states. The second goal is an experimentally more complicated approach to verifying the early → late sequence, but if true, has the value of giving the measurement of tightly bound Mcm proteins in G1 two meanings. The first is as a measure of the functional state of the Mcm loading sub-system within the cellular/environmental context, and the second is as a surrogate marker for commitment to cell cycle progression.
Because a bivariate analysis of any immunoreactivity versus DNA content leaves an uncertainty at the G1/S interface proportional to the coefficient of variation (CV) of the two measures, we enhanced the probability that cells would be identified as either G1 or S by co-staining for tightly bound PCNA. S phase is associated with the release of the Mcm proteins and PCNA binding to replication forks (reviewed in: [26–28]). Tightly bound PCNA has also been shown by cytometry to correlate with incorporation of BrdU .
PCNA and Mcm6 are tightly bound proteins
Fraction of remaining immuno-reactivity
0.6 ± 0.1 (3)
0.4 ± 0.0 (2)
0.4 ± 0.1 (2)
At least 2 bound-Mcm states exist in G1
The data of Friedrich et al.  suggested that there were two distinct clusters in G1 with low and high immunoreactivity for tightly bound Mcm proteins. Figure 1 supports this idea. Figure 1A-C shows bivariate plots of DNA content versus PCNA (1A) for Molt4 cells with clearly delineated G1, S, and G2 clusters as expected from the work of Kurki et al. . Figure 1B shows the expected pattern for Mcm proteins. There is a distinct low fluorescence cluster in G1 (long arrow) that is above the isotype control  and therefore positive for bound Mcm6. There is also a high positive cluster at the apex of the Mcm6 distribution and at the G1/S interface as defined by DNA content (short arrow). Figure 1C shows a bivariate plot for Mcm6 versus PCNA for the same cells using only the G1 and S phases (G2 cells were removed by a G2 gate on a DNA versus PCNA bivariate). This analysis clearly separates late G1 from early S phase. The cell population shown in Figure 1A-C was growing slowly at the time of fixation. Figure 1D shows a plot for Molt4 cells from a different experiment in which the cells were growing rapidly. The high Mcm6 cluster (arrow) is heavily populated in this case whereas the "early" G1 cluster is not. The simplest inference from these data is that the low Mcm6 expression G1 cluster represents an earlier time in G1 and the high Mcm6 expression G1 cluster represents a late time in G1.
The low Mcm6 expression G1 cluster precedes the high expression G1 cluster
Presence of Mcm6 in the low expression G1 cluster
Serum starvation functionally separates the bound-Mcm states of G1
The restriction point (R) divides activated, uncommitted cells from those that are committed to the cell cycle. Cells that have not passed through "R" are not committed and arrest in G1 when serum or nutrients are withdrawn. Cells that are beyond R are committed to the cell cycle and behave like S, G2, and M cells - that is, when serum or nutrients are withdrawn, they complete the current cell cycle and the daughter cells arrest in the next G1. Here, we will refer to the uncommitted cells as G1a and the committed cells as G1p to distinguish them from G1pm and G1ps of Zetterberg  and various uses of G1A and G1B, since each nomenclature is defined by a measurement system.
Because it is apparent that G1a cells lose Mcm6 fluorescence as a function of time in 0.03% serum, and presumably this equates to loss of pre-RC complexes, we needed to address the possibility that tentative G1p cells in both cell lines lost Mcm6 expression - i.e., moved backwards into the low Mcm6 cluster, rather than having progressed into S. To address this, we used cell cycle inhibitors.
Aphidicolin inhibits DNA elongation but not initiation of DNA primer formation by inhibiting DNA polymerase δ/ε , therefore, it blocks the progression of cells beyond early S phase and any cells already in S phase are immobilized. At the same time serum was removed from cycling cultures of BJ1 cells, aphidicolin at 1 μg/ml was added. The results are shown in Figure 7B. Cells in the early S compartment (high Mcm6 and PCNA) decreased during serum starvation but increased during serum starvation in the presence of aphidicolin. Essentially the same results were determined with RPE1 cells (not shown). This indicates that cells within the high Mcm6 G1 cluster moved into early S rather than to the low Mcm6 state.
Nocodazole inhibits polymerization of tubulin and thus prevents formation of the mitotic spindle by inhibiting microtubule elongation. When nocodazole at 50 ng/ml was added to BJ1 cells at the start of serum withdrawal, the low Mcm6 cluster (G1a) did not change (Figure 7C). Cells were prevented from entering G1a by the nocodazole block in M, and they were prevented from leaving by serum withdrawal.
The mimosine, aphidicolin, and nocodazole data confirm the uni-directional flow of cells through G1 phase. When serum is removed, the high Mcm6 cluster G1 cells don't return to a low Mcm6 state, but rather continue on and accumulate in early S in the presence of aphidicolin or mimosine. Cells only enter the low Mcm6 G1 state from mitosis and this entry is blocked in the presence of nocodazole. Therefore, the low Mcm6 state in G1 is equivalent to G1a and the high Mcm6 state in G1 is equivalent to G1p.
Earlier results from Friedrich et al. suggested that in CV-1 cells G17 might be subdivided by the expression of low and high levels of chromatin loaded Mcm proteins . To ask whether this subdivision demarked uncommitted and committed G1, we developed a cytometric assay that included DAPI to stain DNA and antibodies reactive with Mcm6, PCNA, phospho-S10-histone H3. PCNA was included to cleanly separate S phase from late G1. It was previously known that G2 cells would have minimal levels of bound Mcm proteins but loading of pre-RCs begins in late mitosis, and therefore, we included a mitotic marker. This 4 color assay clearly showed in three human cell lines (Molt4, BJ1, RPE) that in actively replicating cells, G1 is subdivided into at least two states characterized by low and high levels of loaded Mcm6 (and by inference, the Mcm complex). We showed that the low level expression state is variable with respect to the average or modal levels that are expressed. We suspect that this variability is related to the rate of G1 transit (e.g., Figure 1), and we suspect that the high levels are relatively constant, but neither of these ideas have been tested rigorously. Further, we showed by cytometric analysis of serum stimulation experiments, serum withdrawal experiments, and stathmokinetic experiments that the low level state is correlated with the uncommitted G1 state and that cell cycle commitment is correlated with the high level state. Finally, we observed the lowest expression of bound Mcm6 in mitotic cells (see Additional files 4, 5, and 6). We did not see any evidence of bimodality in mitotic cells, and therefore the pre-RCs that are bound in mitosis are below the level of detection in this assay.
For this study, we have treated G1 as composed of two states defined by levels Mcm levels. However, during this study we observed distributions of Mcm6 in G1 with more structure than this simplified view. At a minimum, we could model G1 as composed of a basal population equal to the mitotic cells with the lowest levels of bound Mcm proteins, a center population with a variable modal level of bound Mcm proteins, and finally a high population with maximum levels of bound Mcm proteins. The mitotic lowest population should set the biological bottom; the variable center population should vary with growth conditions, and the maximum population should define the levels necessary to enter S phase (e.g., see Additional files 5 and 6). Our results concur with Friedrich et al. in that the level of Mcm proteins detected in the high Mcm cluster appear to be equivalent to the highest levels achieved in cell populations fixed and stained for both bound and unbound Mcm proteins. This may mean that at equilibrium, at the G1 → S transition, all the Mcm proteins that can be bound are bound. If that speculation proves to be true, then it would be interesting to know whether the max levels of Mcm proteins fluctuate as a function of growth rate as well as the center population. Support for any of these speculative remarks will require significant further experimentation.
Mukherjee at al.  have correlated high levels of bound Mcm proteins, cdc45, hyper-phosphorylated Rb, and cyclin E expression with the committed period of G1. They showed that synchronized cell populations (starve and release) treated with RNA polymerase II inhibitors were not arrested during a period when most cells would be in late G1, and therefore, these parameters were correlated with cell cycle commitment. These results are in agreement with the findings of Chuang et al. for stimulated quiescent human epithelial cells that assign Cdc6 binding, Mcm loading, and Cdc7 expression to late G1, dependent on cyclin E activity . Our data are in agreement with both of these studies, but further illustrate that even in actively cycling cells there is a measureable period in late G1 in which Mcm proteins are maximally bound to chromatin. Further, our results strongly suggest that this period is universally passed through prior to PCNA loading and the start of S phase in human somatic cells. Our data do not address the conclusion from both papers that loading occurs solely in late G1. We have focused on actively dividing asynchronous cells - an area that cytometry addresses well. We have presented a starve and stimulate experiment in Figure 2 for hTert immortalized human fibroblasts and epithelial cells. As stated in the text associated with Figure 3, our starvation experiments were performed to determine the sequence of Mcm loading from low to high, and it is likely that we have not worked with cells as deeply quiescent as either Mukerjee et al. or Chuang et al.
The low to high sequence of Mcm expression in G1 may seem obvious from primary asynchronous data as shown by Friedrich et al  and more explicitly in this paper (early S phase associates with the high Mcm G1 cluster and late S through mitosis associates with the low Mcm G1 cluster (Figure 1, Additional files 4, 5, 6)). Nevertheless, it was necessary to test the hypothesis. Single static samples of the type presented in Figure 1 could arise from a more complex expression program than "basal → max → basal" and that complexity would become obvious in experiments like that in Figure 2.
Any conclusion about ORI loading rests on the assumption that all or most of the bound Mcm proteins are associated with ORI. In mammalian cells, origins of replication are not well defined (e.g.,  are listed by Schepers and Papior for human cells ), and therefore comprehensive temporal licensing studies have not been done. More Mcm proteins are bound relative to the number of bound ORC complexes and most Mcm proteins do not co-localize with replication forks or with ORC [see references in ]. There is some evidence that much of the excess Mcm loading is related to dormant ORI that respond to replication fork stalling [36, 37]. In contrast to higher eukaryotes, significant progress has been made in yeast toward identifying all origins (ARS in S. cerevesiae) using genome-wide information and technologies [35, 38]. In S. pombe, which has defined and known origins without consensus sequences, there is a distribution of the efficiency with which origins are used during S phase [see references in ]. High frequency origins are used often (e.g., once every two cycles) and lower efficiency origins are used less (e.g., one in ten cycles) [reviewed in ]. There also appears to be a hierarchal order of ORC binding beginning in mitosis according to origin efficiency that appears at least partly dependent on affinity . In S. pombe, evidence suggests that high efficiency origins are bound early and low efficiency origins are bound late and that this order is followed by Mcm loading and subsequently Cdc45 loading . Therefore, at least in actively dividing somatic cells, the possibilities are that low efficiency and/or dormant origins are licensed in the G1p state (maximum loaded Mcm2-7). Alternatives are that no particular origins are licensed during this period or more complex scenarios coupled with transcription promoter sites could be entertained. It seems like genome-wide efforts will be required to determine whether these ideas have any merit. Since cells resident in this cytometrically defined, high Mcm G1 state could be sorted, perhaps FACS coupled with genome-wide technologies [e.g., ] could provide comprehensive answers.
It is clear from our data that Mcm binding occurs throughout G1 in both quiescence-stimulation experiments and in actively cycling populations. It is probable, that in our serum starvation experiments we did not achieve "deep" quiescence (see section: Presence of Mcm6 in the low expression G1 cluster) and that in true G0 cells (e.g., fibroblasts that have been arrested by both density and starvation), most of the binding occurs during late G1. Even in our experiments (see BJ1 cells at 14 h in Figure 2), the Mcm levels of the G1p cells is ~10 fold greater than that of the G1a cells, which would lead one to conclude that most of the loading occurs in late G1 during cell cycle re-entry, whereas it is less clear in actively dividing cells. The Mcm levels are ~4 fold higher in G1p BJ1 than G1a (Figure 2A 0 h), and ~2 fold higher in RPE G1p compared to G1a (Figure 2B 0 h). However in Molt4 the levels of the two states are distinctly different with almost all of the loading occurring in G1p whether or not the population is growing slow or fast. These are also lymphoma cells, and therefore, deregulation of G1 progression is highly probable. However, in other analyses (actively dividing iHUVEC), we have observed clear evidence of non-linear loading in G1, i.e., a large fraction of G1a cells separated from a large fraction of G1p cells as we observed for Molt4 cells (e.g., Figure 1C an 1D). This can be thought of as a switch from a low loading rate to a high loading rate at the time of switching from uncommitted to committed G1.
Overall, our data support the idea that mitotic cells represent the lowest bound Mcm levels in actively cycling cells; that the rate of chromatin loading in uncommitted G1 cells is variable, and that committed G1 cells are characterized by a maximum level achieved just prior to S. The modal level of binding in an asynchronous, actively cycling population appears to be higher in fast replicating populations and lower in slower replicating populations. This supports the idea that the differences in rates of loading between cycling cells and stimulated G0 cells is a continuum and dependent on the rates of synthesis and activation of cyclin E and Cdk2 for re-entry and perhaps D cyclins and Cdk4/6 (E2F activity in both cases) as suggested by Chuang et al. .
Cell lines, culture and fixation
Molt4 (human T cell lymphoma cell line) was obtained from ATCC (CRL-1582); hTERT-BJ1 (BJ1) and hTERT-RPE-1 (RPE) were from Clontech, Mountain View, CA. Molt 4 cells were grown in RPMI Medium 1640, BJ1 cells were grown in 80% Dulbecco's modified Eagle medium (DMEM) and 20% Medium 199 supplemented with 4 mM L-glutamine and 1 mM sodium pyruvate. RPE1cells were grown in D/MEMF-12 containing 2 mM L-glutamine and 0.375% sodium bicarbonate. These media routinely contained 10% fetal bovine serum (FBS) (Cambrex, Charles City, IA) and 50 μg/ml gentamicin sulfate (Fisher, Pittsburgh, PA). Media and supplements were from Gibco, Carlsbad, CA. Cells were grown at 37°C in a humidified atmosphere with 5% CO2.
Cells were serum starved by washing the cells in PBS and then adding the appropriate medium with 0.03% FBS. In some experiments aphidicolin (Alexis, San Diego, CA), mimosine (Sigma, St. Louis, MO), or nocodazole (Sigma) were added at the time of serum withdrawal.
Adherent cells were harvested with trypsin and cell counts were taken on a Coulter Counter (Coulter Electronics, Hialeah, FL). Washed cells were either fixed directly in 90% methanol at -20°C or first detergent extracted (0.5% Triton X-100, 0.2 μg/ml EDTA, 1% BSA in PBS  for 10 min at 4°C before methanol fixation. Detergent solution was added at 40 μl/106 cells. Final cell concentration was 2.5 × 106 cells/ml.
The following antibodies were used for cytometry and immunoblotting: Phycoerythrin-conjugated and unconjugated anti-Mcm6 (PE-Mcm6), clone K1-Mcm6, (BD Biosciences); fluorescein isothiocyanate-conjugated and unconjugated anti-PCNA (FITC-PCNA), clone PC10 (BD Biosciences); BM28 (anti-Mcm2), clone 46, (BD Biosciences); anti-Cdc6, clone DCS-180, (Upstate, Lake Placid, NY); anti-ORC3, clone 1D6, (Upstate); Alexa Fluor 647-conjugated anti-phospho-S10-histone H3 (A647-pHH3) (Cell Signaling, Danvers, MA); FITC-conjugated goat anti-mouse F(ab)2 (BD Biosciences).
Fixed cells were washed and immunostained with primary antibody for 90 min at 4C in PBS with 2% BSA. If the primary antibody was unconjugated, cells were washed twice and immunostained with the appropriate conjugated secondary antibody. DNA was stained with DAPI at 0.25 - 1 μg/106 cells.
Cells were measured with a BD LSR II flow cytometer (BD Biosciences, San Jose, CA). Cells were excited with the 355 nn, 488 nm, and 633 nm lasers. Final filters were 450/50, 530/30, 575/26 and 670/20 nm. Data were analyzed with WinList 3D 6.0 (Verity Software House, Topsham, ME).
Laser scanning cytometry
Cells were grown on 35 mm glass bottom microwell dishes (MatTek, Ashland, MA) that had been treated with FBS for 1 h to coat the glass with fibronectin. After culture, cells were extracted with detergent, fixed in methanol and immunostained as above. Plates were scanned with a Compucyte (Westwood, MA) iCyte using a 405, 488, and 633 nm excitation and 465/40, 530/30, 575/20 and 675/50 nm emission for DAPI, FITC, PE, and A647 fluorescence, respectively.
Electrophoresis and western blotting
Known numbers of cells were washed and lysed directly in 5% SDS lysis buffer (0.137 M NaCl, 2% Nonidet P-40, 5% SDS, 1% sodium deoxycholate, 20 mM Tris, pH 8.0, 2 mM PMSF, 10 μl/ml protease inhibitor cocktail (P8340, Sigma)) or first extracted with detergent as described above, washed twice in 1 ml PBS at 4°C with centrifugation for 1 min at low speed in a swinging bucket microfuge (Fisher, Model 59A) before lysis buffer was added to the pellet.
Equal cell numbers or equal amounts of protein were loaded on 10% polyacrylamide discontinuous mini-gels (Bio-Rad, Hercules, CA) and electrophoresed conventionally . Gels were electrophoretically blotted  onto Immobilon-P membrane (Millipore, Bedford, MA) without methanol for 15 min at 100 V. Antigens were visualized by immunostaining with the appropriate antibody and alkaline phosphatase conjugated secondary antibody (Promega, Madison, WI) using chemiluminescent detection with CDP-Star substrate (Tropix, Bedford, MA). Blots were exposed to X-ray film, developed, then imaged with a BioRad GelDoc EQ. Quantification was performed using Quantity One 4.1.1 software (Bio-Rad).
Alexa Fluor 488
Alexa Fluor 647
anaphase promoting complex/cyclosome
bovine serum albumin
cyclin dependent kinase
fluorescence activated cell sorting
Mcm protein complex
origin recognition complex
origins of replication
proliferating cell nuclear antigen
This work was supported by grants from the National Cancer Institute: R01 CA073413 (JWJ) and P30 CA043703 (Core Facility Support). The funding agency did not have any role in the study design, data collection, analysis, interpretation of the data, writing of the manuscript, or decision to submit the manuscript for publication.
- Hengstschlager M, Braun K, Soucek T, Miloloza A, Hengstschlager-Ottnad E: Cyclin-dependent kinases at the G1-S transition of the mammalian cell cycle. Mutation research. 1999, 436 (1): 1-9. 10.1016/S1383-5742(98)00022-2.View ArticlePubMedGoogle Scholar
- Roussel MF: Key effectors of signal transduction and G1 progression. Advances in cancer research. 1998, 74: 1-24. full_text.View ArticlePubMedGoogle Scholar
- Arellano M, Moreno S: Regulation of CDK/cyclin complexes during the cell cycle. The international journal of biochemistry & cell biology. 1997, 29 (4): 559-573. 10.1016/S1357-2725(96)00178-1.View ArticleGoogle Scholar
- Pines J: Mitosis: a matter of getting rid of the right protein at the right time. Trends in cell biology. 2006, 16 (1): 55-63. 10.1016/j.tcb.2005.11.006.View ArticlePubMedGoogle Scholar
- Darzynkiewicz Z, Gong J, Juan G, Ardelt B, Traganos F: Cytometry of cyclin proteins. Cytometry. 1996, 25 (1): 1-13. 10.1002/(SICI)1097-0320(19960901)25:1<1::AID-CYTO1>3.0.CO;2-N.View ArticlePubMedGoogle Scholar
- Juan G, Li X, Darzynkiewicz Z: Correlation between DNA replication and expression of cyclins A and B1 in individual MOLT-4 cells. Cancer research. 1997, 57 (5): 803-807.PubMedGoogle Scholar
- Juan G, Li X, Darzynkiewicz Z: Phosphorylation of retinoblastoma protein assayed in individual HL-60 cells during their proliferation and differentiation. Experimental cell research. 1998, 244 (1): 83-92. 10.1006/excr.1998.4165.View ArticlePubMedGoogle Scholar
- Jacobberger JW, Frisa PS, Sramkoski RM, Stefan T, Shults KE, Soni DV: A new biomarker for mitotic cells. Cytometry A. 2008, 73 (1): 5-15.View ArticlePubMedGoogle Scholar
- Yen A, Pardee AB: Exponential 3T3 cells escape in mid-G1 from their high serum requirement. Experimental cell research. 1978, 116 (1): 103-113. 10.1016/0014-4827(78)90068-X.View ArticlePubMedGoogle Scholar
- Yung Y, Walker JL, Roberts JM, Assoian RK: A Skp2 autoinduction loop and restriction point control. The Journal of cell biology. 2007, 178 (5): 741-747. 10.1083/jcb.200703034.PubMed CentralView ArticlePubMedGoogle Scholar
- Dimova DK, Dyson NJ: The E2F transcriptional network: old acquaintances with new faces. Oncogene. 2005, 24 (17): 2810-2826. 10.1038/sj.onc.1208612.View ArticlePubMedGoogle Scholar
- Sherr CJ: The Pezcoller lecture: cancer cell cycles revisited. Cancer research. 2000, 60 (14): 3689-3695.PubMedGoogle Scholar
- Planas-Silva MD, Weinberg RA: The restriction point and control of cell proliferation. Current opinion in cell biology. 1997, 9 (6): 768-772. 10.1016/S0955-0674(97)80076-2.View ArticlePubMedGoogle Scholar
- Zetterberg A, Larsson O, Wiman KG: What is the restriction point?. Current opinion in cell biology. 1995, 7 (6): 835-842. 10.1016/0955-0674(95)80067-0.View ArticlePubMedGoogle Scholar
- Novak B, Tyson JJ: A model for restriction point control of the mammalian cell cycle. J Theor Biol. 2004, 230 (4): 563-579. 10.1016/j.jtbi.2004.04.039.View ArticlePubMedGoogle Scholar
- Martinsson HS, Starborg M, Erlandsson F, Zetterberg A: Single cell analysis of G1 check points-the relationship between the restriction point and phosphorylation of pRb. Experimental cell research. 2005, 305 (2): 383-391. 10.1016/j.yexcr.2005.01.023.View ArticlePubMedGoogle Scholar
- Friedrich TD, Bedner E, Darzynkiewicz Z, Lehman JM: Distinct patterns of MCM protein binding in nuclei of S phase and rereplicating SV40-infected monkey kidney cells. Cytometry A. 2005, 68 (1): 10-18.View ArticlePubMedGoogle Scholar
- Mukherjee P, Cao TV, Winter SL, Alexandrow MG: Mammalian MCM loading in late-G(1) coincides with Rb hyperphosphorylation and the transition to post-transcriptional control of progression into S-phase. PloS one. 2009, 4 (5): e5462-10.1371/journal.pone.0005462.PubMed CentralView ArticlePubMedGoogle Scholar
- Kurki P, Ogata K, Tan EM: Monoclonal antibodies to proliferating cell nuclear antigen (PCNA)/cyclin as probes for proliferating cells by immunofluorescence microscopy and flow cytometry. Journal of immunological methods. 1988, 109 (1): 49-59. 10.1016/0022-1759(88)90441-3.View ArticlePubMedGoogle Scholar
- Stillman B: Initiation of DNA replication. The Cell Division Cycle: Controlling when and where cells divide and differentiate. Edited by: Millar J. 2009, The Biomedical & Life Sciences Collection, London: Henry Stewart Talks Ltd, [http://hstalks.com/bio]Google Scholar
- Blow J: Replication licensing. The Cell Division Cycle: Controlling when and where cells divide and differentiate. Edited by: Millar J. 2009, The Biomedical & Life Sciences Collection, London: Henry Stewart Talks Ltd, [http://hstalks.com/bio]Google Scholar
- Sasaki T, Ramanathan S, Okuno Y, Kumagai C, Shaikh SS, Gilbert DM: The Chinese hamster dihydrofolate reductase replication origin decision point follows activation of transcription and suppresses initiation of replication within transcription units. Molecular and cellular biology. 2006, 26 (3): 1051-1062. 10.1128/MCB.26.3.1051-1062.2006.PubMed CentralView ArticlePubMedGoogle Scholar
- Chuang LC, Teixeira LK, Wohlschlegel JA, Henze M, Yates JR, Mendez J, Reed SI: Phosphorylation of Mcm2 by Cdc7 promotes pre-replication complex assembly during cell-cycle re-entry. Molecular cell. 2009, 35 (2): 206-216. 10.1016/j.molcel.2009.06.014.PubMed CentralView ArticlePubMedGoogle Scholar
- DePamphilis ML, Blow JJ, Ghosh S, Saha T, Noguchi K, Vassilev A: Regulating the licensing of DNA replication origins in metazoa. Current opinion in cell biology. 2006, 18 (3): 231-239. 10.1016/j.ceb.2006.04.001.View ArticlePubMedGoogle Scholar
- Maiorano D, Lutzmann M, Mechali M: MCM proteins and DNA replication. Current opinion in cell biology. 2006, 18 (2): 130-136. 10.1016/j.ceb.2006.02.006.View ArticlePubMedGoogle Scholar
- Kearsey SE, Labib K: MCM proteins: evolution, properties, and role in DNA replication. Biochimica et biophysica acta. 1998, 1398 (2): 113-136.View ArticlePubMedGoogle Scholar
- Takeda DY, Dutta A: DNA replication and progression through S phase. Oncogene. 2005, 24 (17): 2827-2843. 10.1038/sj.onc.1208616.View ArticlePubMedGoogle Scholar
- Prasanth SG, Mendez J, Prasanth KV, Stillman B: Dynamics of pre-replication complex proteins during the cell division cycle. Philosophical transactions of the Royal Society of London. 2004, 359 (1441): 7-16. 10.1098/rstb.2003.1360.PubMed CentralView ArticlePubMedGoogle Scholar
- Lohr F, Wenz F, Haas S, Flentje M: Comparison of proliferating cell nuclear antigen (PCNA) staining and BrdUrd-labelling index under different proliferative conditions in vitro by flow cytometry. Cell proliferation. 1995, 28 (2): 93-104. 10.1111/j.1365-2184.1995.tb00058.x.View ArticlePubMedGoogle Scholar
- Landberg G, Roos G: Antibodies to proliferating cell nuclear antigen as S-phase probes in flow cytometric cell cycle analysis. Cancer research. 1991, 51 (17): 4570-4574.PubMedGoogle Scholar
- Landberg G, Roos G: Expression of proliferating cell nuclear antigen (PCNA) and Ki-67 antigen in human malignant hematopoietic cells. Acta Oncol. 1991, 30 (8): 917-921. 10.3109/02841869109088244.View ArticlePubMedGoogle Scholar
- Korcz A, Soprano DR, Soprano KJ: Tyrosine protein kinase expression in long-term quiescent WI-38 cells following growth factor simulation. J Cell Biochem. 1995, 59 (1): 42-52. 10.1002/jcb.240590106.View ArticlePubMedGoogle Scholar
- Durcan TM, Halpin ES, Casaletti L, Vaughan KT, Pierson MR, Woods S, Hinchcliffe EH: Centrosome duplication proceeds during mimosine-induced G1 cell cycle arrest. J Cell Physiol. 2008, 215 (1): 182-191. 10.1002/jcp.21298.PubMed CentralView ArticlePubMedGoogle Scholar
- Dimitrova DS, Todorov IT, Melendy T, Gilbert DM: Mcm2, but not RPA, is a component of the mammalian early G1-phase prereplication complex. The Journal of cell biology. 1999, 146 (4): 709-722. 10.1083/jcb.146.4.709.PubMed CentralView ArticlePubMedGoogle Scholar
- Schepers A, Papior P: Why are we where we are? Understanding replication origins and initiation sites in eukaryotes using ChIP-approaches. Chromosome Res. 2010, 18 (1): 63-77. 10.1007/s10577-009-9087-1.View ArticlePubMedGoogle Scholar
- Ge XQ, Jackson DA, Blow JJ: Dormant origins licensed by excess Mcm2-7 are required for human cells to survive replicative stress. Genes & development. 2007, 21 (24): 3331-3341. 10.1101/gad.457807.View ArticleGoogle Scholar
- Ibarra A, Schwob E, Mendez J: Excess MCM proteins protect human cells from replicative stress by licensing backup origins of replication. Proceedings of the National Academy of Sciences of the United States of America. 2008, 105 (26): 8956-8961. 10.1073/pnas.0803978105.PubMed CentralView ArticlePubMedGoogle Scholar
- Cadoret JC, Meisch F, Hassan-Zadeh V, Luyten I, Guillet C, Duret L, Quesneville H, Prioleau MN: Genome-wide studies highlight indirect links between human replication origins and gene regulation. Proceedings of the National Academy of Sciences of the United States of America. 2008, 105 (41): 15837-15842. 10.1073/pnas.0805208105.PubMed CentralView ArticlePubMedGoogle Scholar
- Wu PY, Nurse P: Establishing the program of origin firing during S phase in fission Yeast. Cell. 2009, 136 (5): 852-864. 10.1016/j.cell.2009.01.017.PubMed CentralView ArticlePubMedGoogle Scholar
- Desprat R, Thierry-Mieg D, Lailler N, Lajugie J, Schildkraut C, Thierry-Mieg J, Bouhassira EE: Predictable dynamic program of timing of DNA replication in human cells. Genome research. 2009, 19 (12): 2288-2299. 10.1101/gr.094060.109.PubMed CentralView ArticlePubMedGoogle Scholar
- Laemmli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970, 227 (5259): 680-685. 10.1038/227680a0.View ArticlePubMedGoogle Scholar
- Towbin H, Staehelin T, Gordon J: Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proceedings of the National Academy of Sciences of the United States of America. 1979, 76 (9): 4350-4354. 10.1073/pnas.76.9.4350.PubMed CentralView ArticlePubMedGoogle Scholar
- Bagwell CB: Hyperlog-a flexible log-like transform for negative, zero, and positive valued data. Cytometry A. 2005, 64 (1): 34-42.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.