Cdk1 activity acts as a quantitative platform for coordinating cell cycle progression with periodic transcription

Abstract

Cell proliferation is regulated by cyclin-dependent kinases (Cdks) and requires the periodic expression of particular gene clusters in different cell cycle phases. However, the interplay between the networks that generate these transcriptional oscillations and the core cell cycle machinery remains largely unexplored. In this work, we use a synthetic regulable Cdk1 module to demonstrate that periodic expression is governed by quantitative changes in Cdk1 activity, with different clusters directly responding to specific activity levels. We further establish that cell cycle events neither participate in nor interfere with the Cdk1-driven transcriptional program, provided that cells are exposed to the appropriate Cdk1 activities. These findings contrast with current models that propose self-sustained and Cdk1-independent transcriptional oscillations. Our work therefore supports a model in which Cdk1 activity serves as a quantitative platform for coordinating cell cycle transitions with the expression of critical genes to bring about proper cell cycle progression.

Figure 1. Periodic transcription in G1-arrested fission yeast cells.

(a) Schematic representation of the experimental procedure. Inhibitor-sensitive minimal cells were synchronized in G2 by addition of 1 μM 3-MBPP1 for 2 h 40 min. Cultures were washed to allow cells to resume their cycle, progress through the metaphase to anaphase transition (15 min, Release), and finally treated with 10 μM 3-MBPP1 to induce a G1 arrest (DMSO was used as a control). Samples were then taken every 10 min for 120 min and gene expression changes were assessed. (b) Percentage of binucleated cells in a. Both control and inhibitor-treated cells initially undergo mitosis. However, cells subsequently treated with inhibitor at T0 remain blocked in G1 (see c) and do not progress to the next mitosis. n>100 at each time point. (c) DNA content analysis of cells in a. Although control cells progress through S phase (black profile) and cytokinesis, inhibitor-treated cells accumulate in G1 prior to S phase with a 1C DNA content (see Methods section for the interpretation of flow cytometry profiles in fission yeast). (d–f). Changes in gene expression for the indicated genes in a. C and A refer to cycling (DMSO-treated) and arrested (inhibitor-treated) cells, respectively. Fold changes are normalized to actin RNA levels and represented relative to the values at T0 (set to 1). See Supplementary Fig. 1 for additional representative genes.

Figure 2. Periodic transcription in G2-arrested fission yeast cells.

(a) Schematic representation of the experimental procedure. Inhibitor-sensitive minimal cells were synchronized in G2 with 1 μM inhibitor (3-MBPP1) for 2 h 40 min as described. Upon washing off the inhibitor, cells resumed their cycle (Release), progressing through mitosis, G1, S and entering the next G2 after ∼70 min (see b and c). The culture was subsequently treated with 1 μM 3-MBPP1 (T0), blocking the onset of the next mitosis. DMSO-treated cells were used as a control. Samples were then taken every 10 min for 120 min, and gene expression changes were assessed. (b) Percentage of binucleated cells in a. Unlike the control culture, cells treated with inhibitor at T0 do not undergo mitosis. Release is as in a. n>100 at each time point. (c) DNA content analysis of cells in a. At T0, cells have undergone DNA replication (black profile, bottom) and have entered G2. Subsequently, control cells undergo a second round of replication after 40 min (black profile, top), while inhibitor-treated cells are blocked in G2. Note that the reduced synchrony in the second cell cycle makes the next S phase more difficult to monitor by flow cytometry (see DMSO profiles). (d–f) Changes in gene expression for the indicated genes in a. C and A refer to cycling (DMSO-treated) and arrested (inhibitor-treated) cells, respectively. Fold changes are normalized to actin RNA levels and represented relative to the values at T0 (set to 1). (g) Heatmaps reflecting the differences in expression levels (ratios between conditions) for periodic genes between cycling (C) and arrested (A) cells at the indicated time points. Included are ∼450 periodic genes as described in Cyclebase (Supplementary Data 1), sorted by induction amplitude. Coloured boxes on the left show the known targets of PBF, Ace2, MBF and Ams2. Within each cluster, shades of blue indicate three groups according to their transcription amplitude, each containing the same number of genes. Darker blue corresponds to higher fold inductions. See Supplementary Table 1 for the percentages of genes above the cutoffs for the ratios in expression between the cultures.

Figure 3. Coupling of periodic transcription and cell cycle phases.

(a) Top panel: schematic representation of the experimental procedure. Cultures of the nda3-km311 cold-sensitive mutant grown at 30 °C were shifted to the restrictive temperature of 18 °C (T1) for 6 h (mitotic block) before being shifted back to the permissive temperature (Release, T2). Samples were taken at the indicated times during the mitotic block and after the release, and changes in expression of representative periodic genes were assessed. Bottom panel: percentage of binucleated and septated cells during the experiment (n=100 at each time point). (b–f) Changes in gene expression in a for the indicated genes. A and R refer to arrested (restrictive temperature) and released (30 °C upshift) cultures, respectively. Fold changes are normalized to actin RNA levels and represented relative to the values at T1 (set to 1).

Figure 4. Periodic transcription and Cdk1 activity dynamics.

(a) Schematic representation of the experimental procedure. Inhibitor-sensitive minimal cells carrying the nda3-km311 cold-sensitive mutation were synchronized in G2 using 1 μM 3-MBPP1 at the permissive temperature (32 °C). Forty minutes before the end of the G2 block, the culture was shifted down to the restrictive temperature of 18 °C for the rest of the experiment to inactivate Nda3. Upon subsequent release from the inhibitor block, cells resumed their cycle but arrested in mitosis as a result of the nda3-km311 mutation at this temperature (M block, also see Supplementary Fig. 2). After 60 min (T0), different concentrations of 3-MBPP1 were added to inhibit Cdk1 activity and samples were taken every 10 min for 90 min to assess changes in gene expression (DMSO was used as a control). (b–e) Changes in gene transcription in a for the indicated genes. Fold changes are normalized to actin RNA levels and represented relative to the values at B (set to 1). Legends in c–e are as in b.

Figure 5. Periodic transcription upon bypass of cell cycle phases.

(a) Schematic representation of the experimental procedure. Inhibitor-sensitive minimal cells were blocked in G2 by addition of 1 μM 3-MBPP1 for 2 h 40 min. The cultures were then treated with 10 μM 3-MBPP1 for 90 min to induce G1-reset without an intervening mitosis. Subsequent release into 1 μM inhibitor at T0 (control cells were maintained in 10 μM 3-MBPP1) induces re-replication without an intervening mitosis. Samples were collected during the reset and after the release at the indicated time points for assessing changes in gene transcription. (b) DNA content analysis of cells in a. Although control cells remain arrested with a 2C DNA content, cells released into 1 μM 3-MBPP1 enter S phase (black profiles), re-replicating their genome to a 4C DNA content. Note the shift of the cytometry profiles during the G1 reset, which results from cell elongation (see Methods). The dashed line indicating 2C DNA content was set based on the T0 profiles. (c) Changes in gene expression for the indicated genes in a. Fold changes are normalized to actin RNA levels and represented relative to the values at B (G2 block, set to 1).

Figure 6. Interplay between cell cycle transitions and periodic transcription.

(a) Schematic representation of the experimental procedure. Inhibitor-sensitive minimal cells were synchronized in G2 by addition of 1 μM 3-MBPP1 for 2 h 40 min, allowed to resume their cell cycle and progress through the metaphase to anaphase transition (mitosis, 15 min), and treated with 10 μM 3-MBPP1 (B0) for 90 min to induce a G1-arrest (G1 block). Cultures were released from the G1 block at T0, inducing overlapping S and M phases (cells maintained in 10 μM 3-MBPP1 were used as a control). Samples were collected at the indicated time points for assessing changes in gene expression. (b) Top panel: percentage of aberrant nuclei for cells in a (cells undergo DNA replication and mitosis simultaneously, giving rise to ‘cut' phenotypes). Bottom panel: corresponding percentages of binucleated cells. n>100 at each time point. (c) DNA content analysis of cells in a. Although the control cells remain blocked in G1 with a 1C DNA content, the released cells undergo rapid S phase (black profiles). (d,e) Changes in gene transcription for the indicated genes during the release in a. R and A refer to released (DMSO-washed) and arrested (maintained in 10 μM inhibitor after T0) cells, respectively. Fold changes are normalized to actin RNA levels and represented relative to the values at T0 (set to 1). (f) Heatmaps reflecting the differences in expression levels, comparing 20 and 40 min after the release with T0. Included are ∼450 periodic genes as determined in Cyclebase (Supplementary Data 2), sorted by transcription amplitude. Coloured boxes on the left show the known targets of PBF, Ace2, MBF and Ams2. Within each cluster, shades of blue indicate three groups according to their transcription amplitude, each containing the same number of genes. Darker blue corresponds to higher fold inductions (also see Fig. 2). See Supplementary Table 2 for the percentages of genes above the ratio cutoff.