J Cell Biol
April 7, 1997;
A role for Cdk2 kinase in negatively regulating DNA replication during S phase of the cell cycle.
Using cell-free extracts made from Xenopus eggs, we show that cdk2
-cyclin E and A kinases play an important role in negatively regulating DNA replication. Specifically, we demonstrate that the cdk2
kinase concentration surrounding chromatin in extracts increases 200-fold once the chromatin is assembled into nuclei. Further, we find that if the cdk2
-cyclin E or A concentration in egg
cytosol is increased 16-fold before the addition of sperm
chromatin, the chromatin fails to initiate DNA replication once assembled into nuclei. This demonstrates that cdk2
-cyclin E or A can negatively regulate DNA replication. With respect to how this negative regulation occurs, we show that high levels of cdk2
-cyclin E do not block the association of the protein complex ORC with sperm
chromatin but do prevent association of MCM3
, a protein essential for replication. Importantly, we find that MCM3
that is prebound to chromatin does not dissociate when cdk2
-cyclin E levels are increased. Taken together our results strongly suggest that during the embryonic cell cycle, the low concentrations of cdk2
-cyclin E present in the cytosol after mitosis and before nuclear formation allow proteins essential for potentiating DNA replication to bind to chromatin, and that the high concentration of cdk2
-cyclin E within nuclei prevents MCM
from reassociating with chromatin after replication. This situation could serve, in part, to limit DNA replication to a single round per cell cycle.
[+] show captions
References [+] :
Figure 2. High concentrations of cdk2–cyclin E inhibits both DNA replication and the binding of MCM3 to chromatin. (A) Purified egg cytosol containing 1 μM of added cdk2–cyclin E was preincubated for 30 min. After this incubation, membrane and sperm (1,000/μl) were added to the extract. As shown, by using phase optics (left), these sperm formed intact nuclei, and by fluorescent staining (right), the DNA decondensed. (B) Interphase cytosol was preincubated alone (− cdk2–cyclin E), or with 1 μM of cdk2– cyclin E (cdk2–Cyclin E) for 30 min. After this membrane and 1,000 sperm/μl were added. The autoradiogram shows 32P incorporated into DNA during 15 min pulses starting at indicated times after sperm addition. In the absence of cyclin E, replication occurred normally, while in the presence of cyclin E it was strongly inhibited at all time points. (C) cdk2–cyclin E was added to interphase cytosol to the final concentrations indicated. After a 30-min incubation, 1,000 sperm/μl were added to each reaction, the reactions were incubated for another 30 min, and then the sperm were pelleted. The pellets were resuspended in SDS-PAGE sample buffer. MCM3 bound to the sperm chromatin was determined by Western blotting using anti-MCM3 antibody. As shown, MCM3 binding to chromatin was inhibited as the concentration of cdk2–cyclin E increased from 0.12 to 1.0 μM. (D) Interphase cytosol was first incubated with 1,000 sperm/μl for 30 min. 1 μM of cdk2–cyclin E was then added together with membrane. Photographs show a typical nucleus that formed under these conditions. (E) Interphase cytosol was first incubated with 1,000 sperm/μl for 30 min. After this, membrane was added with or without 1 μM of cdk2–cyclin E and DNA replication was assayed as in B. Replication in both extracts was identical, demonstrating that late addition of cdk2–cyclin E does not inhibit chromatin from replicating. (F) Sperm was added to cytosol to a final concentration of 1,000 sperm/μl. After a 30-min incubation, the indicated amounts of cdk2–cyclin E were added, and the reaction was incubated for another 30 min. At the end of the incubation the sperm were pelleted and assayed for MCM3 by Western blotting as described in C. As shown, late addition of cdk2–cyclin E does not inhibit binding of MCM3 to chromatin. Bars: (A and D) 10 μm.
Figure 3. High concentration of cdk2–cyclin E doesn't inhibit ORC2 from binding to chromatin. 500 nM of cdk2–cyclin E was added either before (early cdk2–cyclin E) or 30 min after (late cdk2– cyclin E) sperm addition. Chromatin fractions were extracted with ELB containing 0.1% NP-40 and spun through a sucrose cushion containing 0.1% NP-40. The pellet fractions were analyzed for MCM3 and ORC2 content by Western blotting with specific antibodies. MCM3 binding to chromatin is inhibited by early addition of cdk2–cyclin E, while ORC2 binding is unaffected.
Figure 4. Inhibition of MCM3 binding requires active cdk2– cyclin E kinase activity. (A) 500 nM of cdk2–cyclin E was first incubated with interphase cytosol for 30 min (cdk2–cyclin E). The reaction was then split into two parts, and cip (final 600 nM) was added to one half (+ cip). After a 15-min incubation, sperm was added (final 5,000/μl) to both, and the reactions were carried out for another 30 min. Chromatin-associated MCM3 was analyzed by Western blotting using anti-MCM3 antibody. Control shows the amount of MCM3 bound to chromatin without cdk2– cyclin E treatment. When the kinase activity of cdk2–cyclin E was blocked by cip, it could no longer inhibit MCM3 binding. (B) Interphase cytosol was incubated either alone (control) or with indicated concentrations of cdk2-K33R–cyclin E for 30 min. After this incubation, sperm chromatin was added to 5,000/μl. After being incubated for another 30 min, the sperm was pelleted. MCM3 bound to the sperm chromatin was assayed as in A. cdk2K33R–cyclin E, which is a kinase inactive complex, does not inhibit MCM3 from binding to chromatin.
Figure 5. Nuclei assembled in the presence of high cdk2–cyclin E fail to initiate DNA replication when the cytosol is diluted. (A) Interphase cytosol was first incubated with 1 μM of cdk2–cyclin E for 30 min. After this preincubation, sperm chromatin (1,000 sperm/μl) and membrane were added. Aliquots were removed and incubated with [32P]dATP to determine early replication (left). As expected, high cdk2–cyclin E concentration blocked replication. The remainder of the mixture was incubated for a further 60 min and then diluted with 4 vol of fresh extract containing both cytosol and membrane but lacking cdk2–cyclin E. The diluted reaction was then divided in half, and new sperm chromatin (1,000/μl) was added to one half. Radioactively labeled dATP was then added to both reactions, and DNA replication was assayed after a further 60-min incubation. Nuclei assembled in the presence of high cdk2–cyclin E concentrations failed to replicate after dilution of the extract (− new sperm) while new nuclei added to such a diluted extract replicated normally (+ new sperm). (B) After dilution, an aliquot was taken from the sample “+ new sperm,” and bio-dUTP was added. After 1 h of incubation, the nuclei were spun onto a coverslip, stained with straptavidine-conjugated Texas red, and mounted with Hoechst. Five nuclei were visualized in this field (left), and four of them had bio-dUTP incorporated (right).
Figure 6. Cyclin A inhibits MCM3 binding to chromatin and blocks DNA replication. (A) Interphase cytosol was preincubated alone (− CYCLIN A) or with 66 nM of cyclin A–GST fusion protein for 30 min. After this preincubation, membrane and sperm (1,000 sperm/μl) were added. Equal volumes of these extracts were removed at the indicated times and labeled with [32P]dATP for 15 min. In extracts lacking added cyclin A, replication occurred normally, while in the presence of cyclin A replication was strongly inhibited. (B) Interphase cytosol was preincubated with or without 66 nM of cyclin A-GST for 30 min. Sperm nuclei were then added to 5,000 sperm/μl. After a further 30-min incubation, the samples were diluted fivefold with ELB and the sperm chromatin separated from the cytosol by centrifugation through a sucrose cushion. The cytosol (S) and chromatin (P) fractions were analyzed for MCM3 content by Western blotting with antiMCM3 antibody. Preincubation of cytosol with cyclin A strongly inhibited the subsequent association of MCM3 with chromatin. (C) Interphase cytosol was first incubated with 1,000 sperm/μl for 30 min. This reaction was then divided in half, and 66 nM cyclin A was added to one aliquot (+ CYCLIN A). Membrane was then added to both halves, and DNA replication was assayed as in (A). Under these conditions cyclin A did not inhibit DNA replication. (D) Interphase cytosol was incubated with 5,000 sperm/μl for 30 min. Either ELB buffer (− CYCLIN A) or 67 nM of cyclin A-GST (+ CYCLIN A) was then added to the reactions. After a further 30-min incubation, samples were collected and analyzed as in B. Under these conditions late addition of cyclin A did not prevent MCM3 from binding to sperm chromatin. (E) The samples were treated as in B above, however, instead of pelleting the sperm after incubation, the chromatin-bound MCM3 was visualized by staining with anti-MCM3 antibody, followed by rhodamine-labeled anti–rabbit IgG. Early addition of cyclin A (+ CYCLIN A) blocked the association of MCM3 with chromatin when compared to controls lacking cyclin A (− CYCLIN A).
Figure 7. Comparison between the inhibitor and licensing models. In the inhibitor model, proteins required for potentiating DNA replication (shaded circles) bind to DNA when cdk2 activity is low. In Xenopus eggs this occurs at the end of mitosis before nuclear formation, when cdk2–cyclin E activity is dilute (1 and 2). In somatic cells this would occur during early G1 when cdk2–cyclin E is inactive. After nuclear formation, cdk2–cyclin E (black squares) is rapidly transported into nuclei and accumulates (4). Replication potentiating proteins which enter the nucleus (black circles) are prevented from associating with DNA due to the high nuclear concentration of cdk2–cyclin E. However, the nuclear cdk2 does not displace potentiating proteins which are prebound to DNA. Similarly in somatic cells, activation of nuclear cdk2–cyclin E or A kinases during late G1 would block any further potentiation of DNA for replication. During DNA replication the potentiating proteins are displaced from DNA and prevented from rebinding DNA by the presence of high concentrations of cdk2 kinase activity (5 and 6). In the licensing model, proteins (shaded circles) required for DNA replication associate with DNA at the end of mitosis (1 and 2). Because these proteins cannot enter the nucleus, enclosure of the DNA within the nucleus blocks further licensing (3 and 4). As a result of DNA replication the licensing proteins are converted to an inactive state (5 and 6).
Yeast origin recognition complex functions in transcription silencing and DNA replication.