Model of Cdk1 function in controlling DNA replication in vertebrate cells.

Effects of a Cdk2 deletion on cdk1as cells. (A) Growth curve of cdk1as/cdk2^−/− cells compared to the indicated controls. (B) Cell cycle distribution of the indicated cell lines, calculated from BrdU pulse labeling, phosphohistone H3 staining, and PI staining experiments. (C) Cell cycle analysis of cdk1as/cdk2^−/− cells, following synchronization in G1 phase by elutriation. (D) Cell cycle analysis of cdk1as/cdk2^−/− and cdk1as/cdk2^−/+ cells treated for 8 h with the indicated concentration of 1NM-PP1. The control was not exposed to 1NM-PP1. (E) Centrosomes detected by γ-tubulin immunofluorescence (green color) and counterstained with DAPI (blue color) in fixed cdk1as and cdk1as/cdk2^−/− cells (bar, 10 μm), at the indicated times after 1 μM or 10 μM 1NM-PP1 addition. (F) Average number of centrosomes per cell (n = 50) plotted against time after 10 μM 1NM-PP1 addition.

Causal relationship of Geminin destruction and origin licensing in mitotic HeLa cells after Cdk1 inhibition. (A) Chromatin fractionation of Nocodazole (Noc) arrested cells with the indicated additions. MG132 was added to mitotic cells after a 12-h incubation in Nocodazole. 50 μM Roscovitine was added 1 h later, and the cells were incubated for an additional 2 h. Shown are the immunoblots of soluble and chromatin fractions with the indicated antisera. (B) Degradation of APC/C substrates Cdc20 and Aurora kinase A after Roscovitine treatment of Nocodazole arrested HeLa cells. Cell lysates before or 2 h after 50 μM Roscovitine treatment were analyzed by immunoblotting with the indicated antibodies. (C) Chromatin fractionation of Nocodazole-arrested HeLa cells after transient expression of the indicated cDNAs. The cells were pre-synchronized by a double thymidine block, and transient transfections of either GFP or the destruction box deficient mutants of cyclin B or Geminin were performed, after release from the second thymidine block. Nocodazole was added 7 h after the release, and samples were taken after 12 h further incubation. The soluble and chromatin fractions were probed by immunoblotting with the indicated antisera. Note the absence of Cdt1 and Mcm3 in the Roscovitine treated samples containing the indestructible Geminin mutant (lane 12).

Effects of Cdk1 inhibition on DNA replication and cyclin synthesis in cdk1as/cdk2^−/− and cdk1as/cdk2^−/+ cells. (A) Cdk1 inhibition does not prevent S phase in cdk1as/cdk2^−/+ cells. Cells synchronized in G1 by elutriation were incubated in 10 μM 1NM-PP1, and samples were taken every 2 h for cell cycle analysis, as in Fig. 1 D. (B) Cdk1 inhibition in cdk1as/cdk2^−/− cells inhibits entry into S phase. The same experiment as in A was performed. (C) Inhibition of Cdk1 blocks the initiation of DNA synthesis in cdk1as/cdk2^−/− cells. Asynchronous cultures of cdk1as/Cdk2^−/+ or ^−/− cells were exposed to a 10-min pulse of BrdU and subjected to cell cycle analysis as in Fig. 1 C (top panels). BrdU labeled cells were chased with BrdU-free medium containing 10 μM 1NM-PP1 for a further 7 h (pulse-chase analysis shown in bottom panel). The X-axis of the dot blot shows the content of DNA on a linear scale; the Y axis shows the BrdU uptake on a log scale. Cells in the gate area represent the G1 fraction, of which the percentage is shown. (D) Time course of cyclin synthesis. Cells synchronized in G1 phase were incubated in the presence or absence of 10 μM 1NM-PP1. Cell extracts were analyzed by SDS-PAGE and immunoblotting at the indicated time points.

Mechanisms of endoreplication control during G2 phase. (A) HeLa cells were transfected with scrambled siRNA (lanes 1–3) or Geminin siRNA (lanes 4–6) and synchronized by a double thymidine block. 8 h after release we added 50 μM Roscovitine (lane 1 and 4), 10 μM MG132 (lane 2 and 5), or both (lane 3 and 6). After a further 2 h of incubation we prepared soluble (left) and chromatin-enriched (right) fractions and analyzed for the presence of the indicated proteins by immunoblotting. (B) Diagram of the synchronization protocol used to induce endoreplication in G2 cells. (C) Left panel: DNA content analysis of the same cells as described in A. Right panel: a part of these cells were washed in excess medium and cultured for a further 12 h, before DNA content analysis. The percentage of cells with DNA content greater then 4N is shown. (D) Cells were synchronized by a double thymidine block (lane 1 and 5), released for 8 h (lane 2 and 6), further incubated for 4 h in 50 μM roscovitine (lane 3 and 7), or Nocodazole (lane 4 and 8). Extracts of these cells were analyzed by immunoblotting with two different Cdt1 antibodies and antibodies raised against Mcm3 and 4.

Analysis of DNA endoreplication and Mcm binding to chromatin after Cdk1 and Cdk1/2 inhibition in DT40 and HeLa cells. (A) Inhibition of Cdk1 does not trigger endoreplication in cdk1as/cdk2^−/− cells. The cells were treated with 1 μM 1NM-PP1 for 6 h to allow accumulation in G2 phase, followed by addition of 10 μM 1NM-PP1 for an additional hour (1–10 μM block). The inhibitor was washed out in excess medium and the cells were further incubated without the inhibitor for 14 h. At 2, 6, and 14 h after wash out, cell cycle analysis was performed. (B) Evaluation of origin licensing by analyzing chromatin binding of Mcm2-4 in cdk1as/cdk2^−/− cells. Asynchronous populations (lanes 1 and 4), and cells synchronized in G2 by a 12-h incubation in 1 μM 1NM-PP1, and then further treated with 10 μM 1NM-PP1 (lanes 2 and 5), or with 10 μM 1NM-PP1, and 50 μM Roscovitine (lanes 3 and 6) are shown. The indicated samples were subjected to chromatin fractionation and immunoblot analysis of the soluble (left) and chromatin (right) fraction. (C) Cell cycle analysis of HeLa cells released from a double thymidine block. At 7 h after release, 100 μM Roscovitine or DMSO was added to the cells, which were incubated for additional 2 h (9 and 9+Ros). Progression into the next G1 phase was inhibited by simultaneous addition of 100 ng/ml Nocodazole. (D) Lack of origin licensing in HeLa cells following Roscovitine treatment in G2. Cells were prepared as in C. Samples were analyzed by SDS-PAGE and immunoblotting with the indicated antibodies. Note the absence of Cdt1 and Mcms in the chromatin fraction after Cdk1 inhibition in lane 9.

Inhibition of Cdk1 in mitosis causes endoreplication depending on proteolysis. (A) Experimental design. After a 7-h exposure to 10 μM 1NM-PP1 we released the cells from a G2 arrest into medium containing Nocodazole. After a 1-h incubation in Nocodazole, Cdk1 was again inhibited by the addition of 10 μM 1NM-PP1. (B) Endoreplication after inhibition of Cdk1 in mitosis. cdk1as cells arrested in prometaphase as described in A were incubated with 10 μM 1NM-PP1 and pulse labeled with BrdU at the indicated time points. Dot blot analysis is displayed as in Fig. 1 E. (C) Analysis of Histone H3 phosphorylation in samples treated as described in A. Cells were fixed at 10 min after the second Cdk1 inhibition or without inhibition, and stained with both PI (X-axis, linear scale) and anti phospho-Ser10 Histone H3 antibodies (Y-axis, log scale). (D) Analysis of Cdc27 phosphorylation by immunoblotting in cells synchronized in G2 phase or mitosis as described in A. 15 min after release from the 1NM-PP1 block into mitosis (lane M) Cdc27 shifts upwards due to hyper-phosphorylation by Cdk1. It remains phosphorylated after 1 h in Nocodazole (lane 1, Noc 1h). Following Cdk1 inhibition by addition of 10 μM 1NM-PP1, Cdc27 is rapidly de-phosphorylated and shifts back to its G2 form within minutes (see lane 10′). (E) Immunoblot analysis of cyclin A and Histone H2A levels in Nocodazole arrested cdk1as cells at the indicated times after the second Cdk1 block. (F) Endoreplication initiated by the second Cdk1 inhibition is blocked by MG132. The same experiment as in B in the presence of the proteasome inhibitor MG132 is shown. (F) MG132 blocks origin licensing induced by the second Cdk1 inhibition. Chromatin fractionation of cdk1as cells arrested in mitosis as shown in A, before (lane 1 and 4) and 2 h after 1NM-PP1 addition (lane 2 and 5). To inhibit the proteasome, cells had been incubated for 1 h with MG132 before the 1NM-PP1 inhibitor was added (lane 3 and 6). Mcm and Histone H2A levels were measured by SDS PAGE and immunoblotting.

Inhibition of Cdk1 by chemical genetics in DT40 cells results in a G2 arrest. (A) Histone H1 kinase assays with 9E10 immunoprecipitates from CDC2 gene-disrupted DT40 cells, reconstituted with myc-tagged Xenopus cdk1WT or cdk1as cDNAs. Each immunoprecipitate was split and washed with buffer containing equal amounts of either DMSO or 10 μM 1NM-PP1. Samples from the kinase assay were taken at the indicated incubation times and analyzed by SDS-PAGE and autoradiography to detect ³²P-labeled Histone H1. For a comparison of Cdk1 levels in the immunoprecipitates, see Fig. S1 G. (B) Quantification of the Histone H1 kinase activity of cdk1WT and cdk1as immunoprecipitates in the presence (+) or absence of (−) 10 μM 1NM-PP1. Shown are values relative to the activity of cdk1WT kinase, 16 min after initiation of the reaction. The mean values and SD of four independent experiments are shown. (C) Proliferation kinetics of Cdk1-deficient DT40 cells reconstituted with either cdk1WT or cdk1as cDNAs in the presence (+) or absence (−) of 10 μM 1NM-PP1. (D) Cell cycle profile of cdk1WT and cdk1as cells. Cells were synchronized in G1 phase by elutriation at time zero and cultured for 10 h in medium containing 10 μM 1NM-PP1. Samples were taken every 2 h and analyzed by PI staining and FACS. (E) Cell cycle analysis of BrdU pulse-labeled cdk1as cells incubated with 10 μM 1NM-PP1 for 0, 12, 24, and 48 h. The X-axis of the dot blot shows the content of DNA on a linear scale; the Y-axis shows the BrdU uptake on a log scale. (F) Cyclin stability in 1NM-PP1 arrested cdk1as cells. The cells arrested in G2 phase by a 12-h incubation with 10 μM 1NM-PP1. 100 μg/ml CHX were added and whole-cell extracts were analyzed by SDS-PAGE and immunoblotting at the indicated times. To rule out an interference of CHX with proteolysis cdk1as cells were released from a 4-h Nocodazole arrest in the absence or presence of 100 μg/ml CHX. Cyclin B or Cdk1 amounts were measured by immunoblotting at the indicated times after the release. (G) Localization of cyclin B2 in the cytoplasm of cdk1as cells treated with 10 μM 1NM-PP1. Samples were taken at the indicated times, fixed, and analyzed by immunofluorescence (bar, 10 μm), using the anti–chicken B2 antibody (red color), and DAPI staining of DNA (blue color).