Quantitative kinase and phosphatase profiling reveal that CDK1 phosphorylates PP2Ac to promote mitotic entry

Abstract

The reciprocal regulation of phosphoprotein phosphatases (PPPs) by protein kinases is essential to cell cycle progression and control, particularly during mitosis for which the role of kinases has been extensively studied. PPPs perform much of the serine/threonine dephosphorylation in eukaryotic cells and achieve substrate selectivity and specificity through the interaction of distinct regulatory subunits with conserved catalytic subunits in holoenzyme complexes. Using a mass spectrometry-based chemical proteomics approach to enrich, identify, and quantify endogenous PPP holoenzyme complexes combined with kinase profiling, we investigated the phosphorylation-dependent regulation of PPP holoenzymes in mitotic cells. We found that cyclin-dependent kinase 1 (CDK1) phosphorylated a threonine residue on the catalytic subunit of the phosphatase PP2A, which disrupted its holoenzyme formation with the regulatory subunit B55. The consequent decrease in the dephosphorylation of PP2A-B55 substrates promoted mitotic entry. This direct phosphorylation by CDK1 was in addition to a previously reported indirect mechanism, thus adding a layer to the interaction between CDK1 and PP2A in regulating mitotic entry.

Fig. 1.. PIB-TMT reveals quantitative differences between asynchronous and mitotic PPPome.

(A) Asynchronous or mitotic HeLa cell lysates were either treated with free microcystin-LR or left untreated, followed by enrichment of PPPome using PIBs. The PIB eluates from triplicate experiments were digested into peptides, labeled with tandem mass tags (TMT), combined, fractionated and analyzed by LC-MS/MS. (B) Volcano plot of proteins specifically bound to PIBs in asynchronous or mitotic lysates identified using PIB-TMT strategy. The proteins highlighted in red significantly increased and those in blue significantly decreased in their binding to PIBs in mitotic HeLa cells (fold-change mitotic/asynchronous being > or < respectively, P< 0.05, n=3 independent biological replicates). (C) Proteins identified by PIB-TMT strategy as either asynchronous-specific or mitosis-specific PPP binding proteins were validated by immunoblotting. (D) Scatter plot of the log₂ ratio of PIB binding between mitotic and asynchronous cells of proteins specifically bound to PIBs in PIB-TMT experiment compared to global changes in protein abundance to determine correlation between PIB binding and protein expression. n=3 independent biological replicates.

Fig. 2.. PhosPIB approach identifies differential phosphorylation sites on PPP regulatory proteins.

(A) Schematic of process to identify differential phosphorylation sites on PPP regulatory proteins: PIB eluates from asynchronous or mitotic HeLa cells were digested into peptides, TMT-labeled, and enriched for phosphorylated peptides using Fe-NTA columns prior to LC-MS/MS analysis. (B) Volcano plot of phosphorylation sites identified by phosphopeptide enrichment of PIB eluates from asynchronous or mitotic cells. Sites highlighted in blue significantly decreased and those in red significantly increased in mitotic HeLa PIB eluates (P<0.05, n=3 independent biological replicates). (C) Phosphorylation sites identified on PP1, PP2A, PP4 or PP6 catalytic or regulatory subunits that increased or decreased significantly (red or blue, respectively; P<0.05) in mitotic PIB eluates. (D) Enriched phosphorylation site motifs from phosphopeptides that significantly increased (P < 0.05, log₂ fold change of 1.5) in mitotic or asynchronous PIB eluates. (E) Significantly increased or decreased (P < 0.05) phosphorylation sites from mitotic PIB eluates were classified as either proline-directed, basic (basophilic) or acidic (acidophilic) per surrounding amino acids.

Fig. 3.. MIB-TMT approach uncovers the differential kinome between asynchronous and mitotic cells.

(A) Illustration showing the multiplexed inhibitor beads (MIBs), which are sepharose beads covalently attached to linker-adapted small molecule kinase inhibitors. (B) Schematic of process with which MIBs were used to enrich for kinases from asynchronous or mitotically arrested HeLa cells: MIB eluates from three replicates were digested with trypsin, labeled with TMT reagents, combined, fractionated and analyzed by LC-MS/MS. (C) Phylogenetic kinome tree depicting the protein kinase super families of all 280 human kinases identified in the MIB-TMT experiment. (D) Plot showing log₂ fold changes of all kinases that significantly increased or decreased (P < 0.05) in mitotic vs asynchronous MIB eluates. Data are means ± S.E.; n=3 biological replicates.

Fig. 4.. PP2Ac is phosphorylated at Thr³⁰⁴ during mitosis and regulates PP2A holoenzyme composition.

(A and B) Relative quantification of iBAQ abundances, by LC-MS/MS, of B55 family regulatory proteins in complex with FLAG-tagged WT or mutant [T304D (A) or T304A (B)] PP2Ac purified by FLAG-M2 resin from HEK293T cells. Data are means ± S.E. from n=3 biological replicates. * P < 0.05 by two-tailed Student’s t-test. (C) Immunoblotting for FLAG, B56δ and B55α in FLAG eluates from cells expressing FLAG-tagged PP2Ac WT, T304A or T304D; n=3. (D and E) FLAG-M2 affinity pulldown and LI-COR immunoblotting analysis in asynchronous or mitotically arrested HEK293T cells stably expressing FLAG-tagged WT or T304A PP2Ac to validate the phospho-specific Thr³⁰⁴ PP2Ac antibody. Data are means ± S.E. from n=4 independent blots. * P < 0.05 by two-tailed Student’s t-test. (F) Representative in vitro kinase assay using bacteria-purified GST-PP2Ac and the CDK1-cyclin B complex; n≥2. (G and H) LI-COR Western blotting and analysis of FLAG-PP2Ac eluates from HEK293T cells stably expressing FLAG-tagged PP2Ac were left asynchronous (Async), synchronized in mitosis (Mitotic) and treated with calyculin A (100 nM for 30 min), flavopiridol (10 μM for 15 min) or RO-3306 (10 μM for 15 min). Data are means ± S.E., n=3 independent replicates; * P < 0.05 by two-tailed Student’s t-test. (I) Immunoblotting for pThr³⁰⁴ and total PP2Ac using LI-COR in PIB-mediated PP2Ac pull-downs from G1/S-arrested or mitotically arrested HeLa cells. Data is representative of n=3 experiments.

Fig. 5.. Phosphorylation of PP2Ac regulates mitotic timing and PP2A-B55–dependent dephosphorylation at mitotic exit.

(A) Representative blotting for PP2Ac, Myc, GFP and tubulin in HeLa-Flp-In T-REx cells that were knocked down of endogenous α and β isoforms of PP2Ac and doxycycline-induced to express siRNA-resistant Myc-tagged WT PP2Ac. (B) Live cell imaging of cells expressing doxycycline-inducible Myc-PP2Ac as they go through mitosis. (C) Bar plot of the percentage of cells (depleted endogenous PP2Ac and expressing siRNA-resistant WT or Thr³⁰⁴-mutant PP2Ac) that entered mitosis, as determined by nuclear envelope breakdown and cell rounding. Data are mean ± SD from at least 80 cells per condition in a representative of three independent experiments; ns: non-significant, *P<0.05 by Mann-Whitney U-test. (D) Time from nuclear envelope breakdown to mitotic exit of cells depleted of endogenous PP2Ac and expressing either WT or Thr³⁰⁴-mutant PP2Ac. Each dot represents a single cell; red line indicates the median time. A representative result from three independent experiments is shown; ns: non-significant, *P<0.05 by Mann-Whitney U-test. (E) Nocodozaole-treated (mitotically arrested) HeLa cells that had been depleted of endogenous PP2Ac and doxycycline-induced to express WT, T304A or T304D PP2Ac, were pre-treated with MG132 (30 μM for 30 mins) then split and treated with flavopiridol (20 μM) to induce mitotic exit and collected every 5 mins for immunoblotting analysis as indicated. (F) Immunofluorescent imaging and analysis of TPX2 and tubulin staining in cells expressing doxycycline-induced Myc-PP2Ac T304D and control or PP2Ac shRNA. Data are relative intensity of TPX2 at telophase bridges compared to tubulin intensity, calculated from at least 20 cells imaged per condition over three independent experiments; *P<0.05. Scale bar, 10 μm. (G) Model depicting the role of PP2Ac Thr³⁰⁴ phosphorylation in regulating PP2A holoenzyme composition.