Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Sep 18;10(1):4252.
doi: 10.1038/s41467-019-12084-x.

A Mitotic CDK5-PP4 Phospho-Signaling Cascade Primes 53BP1 for DNA Repair in G1

Affiliations
Free PMC article

A Mitotic CDK5-PP4 Phospho-Signaling Cascade Primes 53BP1 for DNA Repair in G1

Xiao-Feng Zheng et al. Nat Commun. .
Free PMC article

Abstract

Mitotic cells attenuate the DNA damage response (DDR) by phosphorylating 53BP1, a critical DDR mediator, to prevent its localization to damaged chromatin. Timely dephosphorylation of 53BP1 is critical for genome integrity, as premature recruitment of 53BP1 to DNA lesions impairs mitotic fidelity. Protein phosphatase 4 (PP4) dephosphorylates 53BP1 in late mitosis to allow its recruitment to DNA lesions in G1. How cells appropriately dephosphorylate 53BP1, thereby restoring DDR, is unclear. Here, we elucidate the underlying mechanism of kinetic control of 53BP1 dephosphorylation in mitosis. We demonstrate that CDK5, a kinase primarily functional in post-mitotic neurons, is active in late mitotic phases in non-neuronal cells and directly phosphorylates PP4R3β, the PP4 regulatory subunit that recognizes 53BP1. Specific inhibition of CDK5 in mitosis abrogates PP4R3β phosphorylation and abolishes its recognition and dephosphorylation of 53BP1, ultimately preventing the localization of 53BP1 to damaged chromatin. Our results establish CDK5 as a regulator of 53BP1 recruitment.

Conflict of interest statement

J.A.M. serves on the Scientific Advisory Board of 908 Devices. P.S. has been a consultant at Novartis, Genovis, Guidepoint, The Planning Shop, ORIC Pharmaceuticals and Exo Therapeutics; his laboratory receives research funding from Novartis. The remaining authors declare no competing interests.

Figures

Fig. 1
Fig. 1
PP4R3β S840 phosphorylation is required for 53BP1 recruitment to damaged chromatin. a MS/MS spectrum of the PP4R3β phospho-peptide harboring S840. Phosphorylated residue is highlighted in green. b Co-immunoprecipitation of 53BP1 or PP4C with PP4R3β S840 phospho-variants. HeLa cells transiently transfected with the indicated FLAG-53BP1 and Myc-PP4R3β constructs were left in an asynchronous state (Asyn) or synchronized to prometaphase (M) by treatment with nocodazole (Supplementary Fig. 1f–h). Whole cell extracts were immunoprecipitated (IP) with indicated agarose beads and analyzed by immunoblot using indicated antibodies. EV: empty vector; S/D: S840D; S/F: S840F; S/A: S840A. c Immunoblots showing the kinetics of 53BP1 phosphorylation throughout mitosis in cells expressing PP4R3β S840 phospho-variants . HeLa cells were transfected with the indicated siRNAs and complemented with the indicated siRNA-resistant Myc-PP4R3β S840 phospho-variant constructs. Transfected cells were released from prometaphase arrest and harvested at indicated time points for immunoblot analysis using indicated antibodies. Phospho-histone H3 at Ser10 (H3pS10) indicates mitotic state. d Irradiation-induced 53BP1 foci formation in cells expressing PP4R3β S840 phospho-variants. Mitotic HeLa cells were complemented with Myc-PP4R3β S840 phospho-variant constructs as described in c, collected from prometaphase arrest and seeded on poly-D-lysine-coated coverslips. Four hours after release, cells were irradiated with 5 Gy. Two hours post-irradiation, cells were fixed and analyzed by immunofluorescence with indicated antibodies. γ-H2AX staining marks sites of DNA damage. e Quantifications of 53BP1 foci in d, pooled from triplicate repeat experiments, expressed as mean ± s.d. Total number of cells is indicated in parenthesis. P-values, Mann–Whitney U Test. f Radiosensitivity of cells expressing PP4R3β S840 phospho-variants. HeLa cells were complemented with siRNA-resistant Myc-PP4R3β S840 phospho-variant constructs as described in c. Viability was evaluated by clonogenic survival. Immunoblots confirm siRNA efficiency and expression of siRNA-resistant constructs. Data are expressed as mean ± s.d; n = 3. g Response of BRCA1-null UWB1.289 cells expressing PP4R3β S840 phospho-variants to indicated concentration of clinical-grade PARP inhibitor Olaparib. Cells were complemented as described in c. Viability was assessed by CellTiter-Glo colorimetric assay. Immunoblots confirm siRNA efficiency and expression of siRNA-resistant constructs. Data are expressed as mean ± s.d; n = 3
Fig. 2
Fig. 2
CDK5 is active in mitosis. a Phosphorylation of 53BP1 T1609/S1618 at distinct stages of mitosis and in G1. RPE1 cells were fixed at 0, 15, 30, 60, 90, and 120 min after release from G2/M arrest and stained with phospho-specific antibody against T1609/S1618 (pT1609/pS1618). Cells at indicated stages of mitosis were selected based on chromatin morphology, as indicated by DAPI staining. b Fluorescence intensity of phosphorylated 53BP1 over area of interest was quantified. Staining intensities from number of events per stage (indicated in the parenthesis) were expressed as mean ± s.d. A.U: arbitrary unit. c Phosphorylation of PP4R3β S840 at distinct stages of mitosis and in G1. RPE1 cells were synchronized and fixed at time points after release as described in a and stained with a phospho-specific antibody against S840 (pS840). d Fluorescence intensity of phosphorylated PP4R3β over area of interest was quantified as described in b. e Distribution of HeLa cells across distinct stages of mitosis at indicated time points after release from prometaphase arrest. HeLa cells were arrested at G2/M by RO-3306 and released into nocodazole to enrich for cells in prometaphase for 1 h. Prometaphase-arrested cells were collected by shake-off and released to progress through mitosis into G1 and fixed at indicated time points. Cells corresponding to stages of mitosis were tallied based on chromatin morphology, as indicated by DAPI staining. f Kinase activity levels of CDK5 in mitosis and G1. HeLa cells were synchronized and released as described in e. CDK5, immunoprecipitated from cells collected at indicated time points after release, was incubated with substrate peptide and [32P]-ATP. Radioactivity of labeled substrate peptide was measured in a liquid scintillation counter. Data are expressed as mean ± s.d; n = 3. Representative immunoblots show indicated proteins present in the cell lysate for kinase assay. g Immunoblots showing the kinetics of 53BP1 T1609/S1618 phosphorylation and PP4R3β S840 phosphorylation throughout mitosis. HeLa cells were collected at indicated time points after release from synchronization as described in e and were analyzed with indicated antibodies
Fig. 3
Fig. 3
PP4R3β is a bona fide substrate of CDK5. a Reconstituted CDK5/p25 complex was incubated with purified wild-type or S840A PP4R3β. 32P-ATP signal indicates CDK5/p25-mediated phosphorylation. Ponceau-stained blots show input proteins. b HeLa cells stably expressing FLAG-HA empty vector (EV) or FLAG-HA PP4R3β were arrested in prometaphase, lysed, and immunoprecipitated for FLAG-HA PP4R3β. Elution from FLAG immunoprecipitation was analyzed by immunoblot with indicated antibodies. c RPE1 cells released from RO-3306-induced G2/M arrest were immediately treated with DMSO or 10 µM of CDK5 inhibitor compound 20-223. One hour after release from RO-3306, prometaphase cells were collected by shake-off and seeded on poly-D-lysine-coated coverslips. Three hours after release, cells were irradiated with 5 Gy IR. Two hours post-irradiation, cells were fixed and stained with indicated antibodies to detect irradiation-induced 53BP1 foci. γ-H2AX staining marks sites of DNA damage. d Quantifications of 53BP1 foci in c, pooled from triplicate repeat experiments, expressed as mean ± s.d. Total number of cells is indicated in parenthesis. P-values, Mann–Whitney U Test. e Effect of CDK5 inhibition on radiosensitivity. Hela cells were treated with DMSO or 250 nM of 20-223 for 48 h, followed by irradiation at the indicated doses. Cell viability was evaluated by clonogenic survival. Data are expressed as mean ± s.d; n = 3. Immunoblots indicate abrogation of PP4R3β S840 phosphorylation from treatment with 20-223. f Response of BRCA1-null UWB1.289 cells treated with DMSO or 250 nM of 20-223 to indicated concentration of clinical-grade PARP inhibitor Olaparib. Cell viability was assessed by CellTiter-Glo assay. Data are expressed as mean ± s.d; n = 3. Immunoblots indicate abrogation of PP4R3β S840 phosphorylation from treatment with 20-223. g Melanoma A375 cells stably expressing CDK5 analog-sensitive variant (A375-AS) were arrested in prometaphase. Mitotic cells were released into media containing either DMSO or 10 µM of non-hydrolyzable ATP analog, 1NMPP1, to specifically inhibit CDK5. Cells were harvested at indicated time points for immunoblot analysis with indicated antibodies. h Quantification of immunoblot intensity in g, from duplicate repeat experiments, by ImageJ
Fig. 4
Fig. 4
CDK5 regulates dephosphorylation and the recruitment of 53BP1 to DNA damage. a Endogenous 53BP1 was immunoprecipitated from DMSO-treated or 1NMPP1-treated A375-AS cells and collected at indicated time points after release from arrest at prometaphase. Levels of co-immunoprecipitated endogenous PP4R3β were analyzed by immunoblot. b Endogenous 53BP1 was immunoprecipitated from DMSO-treated or 1NMPP1-treated A375-AS cells, collected at indicated time points, after release from nocodazole-induced arrest at prometaphase. Levels of 53BP1 T1609/s1618 phosphorylation from immunoprecipitated endogenous 53BP1 were analyzed by immunoblot. c Mitotic A375-AS cells, collected by shake-off from arrest at prometaphase, were immediately treated with DMSO or 10 µM of 1NMPP1 and seeded on poly-D-lysine-coated coverslips. 90 min after release, treated cells were irradiated with 5 Gy. 240 min after release, cells were fixed and stained with indicated antibodies to detect irradiation-induced 53BP1 foci. γ-H2AX was stained to mark sites of DNA damage. d Quantifications of 53BP1 foci in c, pooled from triplicate repeat experiments, expressed as mean ± s.d. Total number of cells is indicated in parenthesis. P-values, Mann–Whitney U Test. e A375-AS cells were exposed to 0.2 µM Aphidicolin and DMSO or 10 µM 1NMPP1 for 12 h. Cells were fixed and stained for 53BP1 and Cyclin-A, a negative marker for G1 cells. f Quantification of percentage of cells containing 53BP1-containing nuclear bodies in e out of total cells, indicated in parenthesis, pooled from triplicate repeat experiments. P-values, Mann–Whitney U Test
Fig. 5
Fig. 5
A CDK5-PP4 signaling axis drives 53BP1 recruitment to damaged chromatin. a A375-AS cells were depleted of endogenous PP4R3β by siRNA knockdown and transfected with indicated siRNA-resistant Myc-PP4R3β S840 phospho-variant constructs. Transfected cells were synchronized to prometaphase. Mitotic cells collected by shake-off were immediately treated with DMSO or 10 µM of 1NMPP1 and seeded on poly-D-lysine–coated coverslips. Ninety minutes after release, treated cells were irradiated with 5 Gy. 240 min after release, cells were fixed and stained with indicated antibodies to detect irradiation-induced 53BP1 foci. γ-H2AX was stained to mark sites of DNA damage. b Quantifications of 53BP1 foci in a, pooled from triplicate repeat experiments, expressed as mean ± s.d. Total number of cells is indicated in parenthesis. P-values, Mann–Whitney U Test. c A375-AS cells were depleted of endogenous PP4R3β by siRNA knockdown and transfected with indicated siRNA-resistant Myc-PP4R3β S840 phospho-variant constructs. The transfected cells were exposed to 0.2 µM Aphidicolin and DMSO or 10 µM 1NMPP1 for 12 h. Cells were fixed and stained for 53BP1 and Cyclin-A, a negative marker for G1 cells. d Quantification of percent cells containing 53BP1-containing nuclear bodies in c out of total cells (indicated in parenthesis) pooled from triplicate repeat experiments. P-values, Mann–Whitney U Test. e Model depicting CDK5 as a regulator of 53BP1 recruitment to damaged chromatin

Similar articles

See all similar articles

Cited by 2 articles

References

    1. Giunta S, Belotserkovskaya R, Jackson SP. DNA damage signaling in response to double-strand breaks during mitosis. J. Cell. Biol. 2010;190:197–207. doi: 10.1083/jcb.200911156. - DOI - PMC - PubMed
    1. Lee DH, et al. Dephosphorylation enables the recruitment of 53BP1 to double-strand DNA breaks. Mol. Cell. 2014;54:512–525. doi: 10.1016/j.molcel.2014.03.020. - DOI - PMC - PubMed
    1. Orthwein A, et al. Mitosis inhibits DNA double-strand break repair to guard against telomere fusions. Science. 2014;344:189–193. doi: 10.1126/science.1248024. - DOI - PubMed
    1. Cundell MJ, et al. A PP2A-B55 recognition signal controls substrate dephosphorylation kinetics during mitotic exit. J. Cell. Biol. 2016;214:539–554. doi: 10.1083/jcb.201606033. - DOI - PMC - PubMed
    1. Zheng XF, Kalev P, Chowdhury D. Emerging role of protein phosphatases changes the landscape of phospho-signaling in DNA damage response. DNA Repair. 2015;32:58–65. doi: 10.1016/j.dnarep.2015.04.014. - DOI - PubMed

Publication types

Feedback