|
Figure 1. Cyclin E associates with chromatin in LSS after nuclear import. (A) Sperm chromatin was assembled in the presence of cycling LSS at 23°C for 0–2 h (time of assembly shown beneath blots) before spinning through a sucrose cushion to isolate nuclei in duplicate. One nuclear sample was extracted with chromatin extraction buffer and respun to isolate chromatin-associated proteins. Cytosolic, nuclear, and chromatin-associated samples were resolved by SDS-PAGE and analyzed by Western blotting with ORC or cyclin E antibodies. Schematics above blots depict the timing of relevant events including, nuclear import (NI), DNA replication (DNA repl), cyclin E association with chromatin (Cyc E on Chrom), and mitosis (M). The indicated samples were supplemented with 10 μM okadaic acid (OA) or 100 μg/ml cycloheximide (CHX) for 120 min. (B) Samples identical to those in A were supplemented with [α-32P]dCTP. At each time point, the reactions were stopped, and the amount of DNA synthesized in duplicate samples was quantitated as detailed in Materials and Methods.
|
|
Figure 3. To assemble onto chromatin, cyclin E–Cdk2 requires an activity present in HSS that minimally contains ORC and Cdc6. (A) HSS was diluted with XB2 buffer before the addition of λ DNA templates and baculovirus cyclin E–Cdk2 for a 30-min incubation. Assembled chromatin was isolated and analyzed as in Fig. 1. Lane 1, no DNA; lanes 2–6, DNA templates assembled in HSS that was undiluted, or diluted 1:1, 1:3, 1:7, or 1:11 with XB2. (B) HSS was either left untreated (lanes 1–3), heat treated (lane 4), ATP depleted (lane 5), or supplemented with 10 mM MgCl2 (lane 6) before the addition of λ DNA templates (lanes 2–6). Purified baculovirus-expressed cyclin E–Cdk2 was also added to samples in lanes 3–6. Assembled chromatin was isolated and analyzed as above. (C) Individual aliquots of HSS were immunodepleted with antibodies specific to XORC2 (lanes 3 and 6), XCdc6 (lanes 4 and 7), XMCM3 (lane 5), or with beads alone (lane 2). Specific samples were supplemented with purified XORC complex (lanes 6 and 8) or baculovirus-expressed XCdc6 (lanes 7 and 8). All samples included baculovirus-expressed Xcyclin E–Cdk2 and an energy regenerating system. Depleted samples with and without additions were incubated with λ DNA for 30 min, sedimented through a sucrose cushion, and resolved by SDS-PAGE. (D) Western blots of depleted HSS used for assembling chromatin in C. Lane 1, mock depleted; lane 2, ORC2 depleted; lane 3, Cdc6 depleted; lane 4, MCM3 depleted.
|
|
Figure 2. The cyclin E–Cdk2 complex from HSS associates with chromatin with kinetics similar to ORC and Cdc6, but earlier than MCM3. Chromatin was assembled by addition of sperm DNA to HSSs from Xenopus egg extracts, and reactions were stopped at indicated times. Chromatin templates were isolated, resolved by SDS-PAGE, and analyzed by Western blotting with antibodies to Xenopus ORC2, Cdc6, MCM3, and cyclin E (Materials and Methods). Lane 1, no DNA, 30 minutes; lanes 2–5, DNA templates assembled for 0, 5, 10, or 15 min. Later time points showed no additional assembly of ORC, Cdc6, MCM3, or cyclin E–Cdk2.
|
|
Figure 6. RXL mutants of Cdc6 show a quantitative defect in their ability to bind to cyclin E, to get phosphorylated by cyclin E–Cdk2, and to sustain replication in Cdc6-depleted extract. (A) LSS was immunodepleted with affinity-purified XCdc6 antibodies conjugated to protein A–Sepharose beads. Depleted samples were supplemented with sperm DNA, an energy regenerating system, α[32P]dCTP, and 1, 5, 10, 20, 30, or 100 nM of either wild-type GST–XCdc6 (♦) or GST-XCdc6 with all three RXL motifs mutated to AXA (▪) (see Materials and Methods for mutant description). Replication was quantitated as indicated in Materials and Methods and plotted as a percentage of undepleted extract, normalizing to 100% rescue in mock-depleted extracts and setting 0% replication as the amount of background counts incorporated after depletion. (B) Purified GST (lane 1), wild-type GST–XCdc6 (lane 2), or triple RXL mutant GST–XCdc6 (lane 3) was incubated with purified baculovirus-expressed cyclin E–Cdk2 in the presence of γ[32P]ATP. Proteins were resolved by SDS-PAGE, and phosphorylated proteins were visualized by autoradiography. Membrane stained with Ponceau S is shown below as a loading control. (C) Purified GST (lane 1), wild-type GST–XCdc6 (lane 2), or triple RXL mutant GST–XCdc6 (lane 3) was incubated with radiolabeled IVT Xcyclin E. After a 30-min incubation, samples were diluted in IP buffer, and GST proteins were precipitated with glutathione–agarose beads and washed. Beads were resuspended in sample buffer, and associated proteins were resolved by SDS-PAGE and visualized by autoradiography. Membrane stained with Ponceau S is shown below as a loading control.
|
|
Figure 4. Purified cyclin E–Cdk2 binds directly to Cdc6. Baculovirus-expressed cyclin E–Cdk2 was incubated for 30 min with an energy regeneration system and purified GST fusion proteins including GST–p21N1–90 (lane 1), GST–p21C87–164 (lane 2), GST–p27 (lane 3), GST–Cdc6N2–168 (lane 4), GST–Cdc6C169–554 (lane 5), GST–XORC1 (lane 6), or GST–hCdc14 (lane 7). Reactions were diluted in IP buffer and bound to glutathione–agarose beads. Beads were washed and resolved by SDS-PAGE, and cyclin E was visualized by Western blotting.
|
|
Figure 5. The MRAIL motif of cyclin E is required for binding of cyclin E to Cdc6, recruitment of cyclin E–Cdk2 to DNA, and replication competence. (A) Schematic of the Xenopus cyclin E protein. The shaded area indicates the cyclin box. Within this region, the specific amino acids mutated for the MLW mutant (*) and the LQ mutant (^). Putative phosphorylation sites are also depicted, proceeded by the amino acid number of the specific serine or threonine residue. (B) Wild-type Xenopus cyclin E (lanes 1, 2, 5, and 6), or cyclin E with mutations in the MLW (lanes 3 and 7) or the LQ (lanes 4 and 8) peptide sequences were radiolabeled by in vitro translation in rabbit reticulocyte lysate. The IVT cyclin E variants were added to bacterially expressed GST–p21N (lanes 1–4) or GST–XCdc6 (lanes 5–8), incubated for 30 min, and diluted in IP buffer. GST–p21- and GST–Cdc6-associated cyclin E–Cdk2 was precipitated with glutathione–agarose beads. Beads were washed and resuspended in sample buffer, and proteins were resolved by SDS-PAGE and visualized by autoradiography. Lanes 9–11 show a matched exposure of the amount of input IVT cyclin E used in the binding experiments. (C) HSS was supplemented with buffer (lanes 1 and 2), with 100 nM GST (lane 3) or increasing doses (30, 60, or 100 nM) of wild type (lanes 4–6), or MRAIL mutant (lane 7–9) GST–Xcyclin E. After preincubating the HSS with the GST proteins, λ DNA templates were added to extracts (lanes 2–9), and assembled chromatin was isolated and resolved by SDS-PAGE and blotted for the presence of endogenous cyclin E. The lack of a cyclin E signal in lanes 5 and 6 indicates that wild-type GST–Xcyclin E can effectively compete away chromatin binding of endogenous cyclin E at the indicated concentrations. (D) LSS was immunodepleted with cyclin E antibodies conjugated to protein A–Sepharose beads. Depleted samples were supplemented with undepleted LSS, increasing concentrations of wild type, or MRAIL mutant GST–Xcyclin E, as noted, before the addition of sperm DNA, an energy regenerating system, and [α-32P]dCTP. Replication was assayed and quantitated in duplicate samples as described in Materials and Methods and plotted as a percentage, normalizing the amount of replication in undepleted LSS to 100%. This corresponds to 1.7 ng/μl of new DNA synthesized from the 2.5 ng/μl of DNA added.
|
|
Figure 7. Replication elongation is required for cyclin E accumulation on chromatin. Cycling LSS extracts were incubated with sperm DNA for the indicated times in the absence (lanes 1–6) or the presence (lanes 7 and 8) of aphidicolin (Aphid; 40 μg/ml) before isolating chromatin templates by sedimentation and resolving chromatin-associated proteins by SDS-PAGE. Top shows Western blots for cyclin E and Cdc6, which remain bound to chromatin in varying amounts throughout DNA replication (DNA rep). Later time points showed no additional assembly of cyclin E onto chromatin in aphidicolin-treated samples. Bottom shows IP kinase assays of samples identical to those above. Anti–cyclin B antibodies conjugated to protein A–Sepharose beads were used to immunoprecipitate cyclin B, and associated kinase activity was assayed by in vitro phosphorylation of histone H1 in the presence of γ[32P]ATP. The peak in cyclin B kinase activity indicates that the extracts are in mitosis (M).
|
|
Figure 10. Model of the cell cycle–regulated association of cyclin E–Cdk2 with chromatin and its effects on DNA replication and rereplication control. In a first phase, cyclin E–Cdk2 is recruited to origins of DNA replication by ORC, Cdc6, and possibly an unknown factor (X?). In this conformation, with MCMs bound, DNA replication is initiated. In a second phase, dependent on the progression of replication forks, multiple molecules of cyclin E accumulate on chromatin, blocking re-replication. In a final phase, cyclin E is hyperphosphorylated by cyclin B–Cdc2 and stripped from chromatin in mitosis. Rebinding of cyclin E to chromatin in interphase is possible only after dephosphorylation by Cdc14 or a related phosphatase (Discussion).
|
|
Figure 8. Specific mitotic kinases are capable of phosphorylating cyclin E and displacing cyclin E–Cdk2 from chromatin; Cdc14 can oppose phosphorylation by these kinases. (A) Sperm chromatin assembled in interphase LSS (in the presence of cycloheximide) for 1 h was subsequently treated with buffer (lane 1), 1 U of MAP kinase (lane 2), cyclin B–Cdc2 (lane 3), Plk1 (lane 4), or 10 μM okadaic acid (lane 5) for 10 min. Chromatin was extracted, and associated proteins were resolved by SDS-PAGE and Western blotted for the presence of cyclin E or ORC. (B) Purified GST–Xcylin E was incubated with buffer (lanes 1 and 2), MAP kinase (lanes 3 and 4), cyclin B–Cdc2 (lanes 5 and 6), Plk1 (lanes 7 and 8), or cyclin E–Cdk2 (lane 9–10) in the presence of γ[32P]ATP. After 30 min, 2 μM GST–Cdc14 was added to indicated samples (lanes 2, 4, 6, 8, and 10), and all samples were incubated for an additional 30 min. Reactions were resolved by SDS-PAGE, and phosphorylated GST–cyclin E was visualized by autoradiography.
|
|
Figure 9. Cdc14 reverses the inability of mitotic hyperphosphorylated cyclin E to bind to chromatin. (A) Interphase extract (lanes 1–4) or mitotic extract stabilized by the addition of nondestructible cyclin B (lanes 5–8) was supplemented with buffer (lanes 1 and 4), 10 μM okadaic acid (OA; lanes 2 and 6), 1 μM GST–Cdc14 (lanes 3 and 7), or both okadaic acid and Cdc14 (lanes 4 and 8) and incubated at 23°C for 30 min. Reactions were stopped by adding sample buffer, proteins were resolved by SDS-PAGE, and Western blots were performed with cyclin E antibodies to detect the various phosphorylated forms of cyclin E. (B) Baculovirus-expressed Xenopus cyclin E–Cdk2 in an autohyperphosphorylated form was mixed with buffer (lane 2), increasing concentrations of the CIP phosphatase (lanes 3–5), or increasing concentrations of GST–Cdc14 (lanes 6–8) for 30 min. Untreated HSS (lane 1) and treated samples were resolved by SDS-PAGE and analyzed by Western blotting with antibodies to cyclin E (top). The samples in lanes 2–8 were incubated with λ DNA templates and a small amount of HSS (bottom). Assembled chromatin was isolated by sedimentation, and proteins were resolved by SDS-PAGE and analyzed by Western blotting with anti–cyclin E antibodies. The sample in lane 1 is HSS that was not treated (NT).
|
