Skp2 is phosphorylated on Ser64 and Ser72 in vivo. (A) Rat1 cells transfected with Myc₆-Skp2 were labelled with ³²Pi. Skp2 was immunoprecipitated with anti-Myc and analysed by autoradiography (upper panel) and immunoblotting (lower panel). (B) Phosphopeptide mapping analysis of Myc₆-Skp2 and phosphoamino-acid analysis of the main phosphopeptides. (C) In vivo phosphorylation of the indicated Myc₆-Skp2 constructs expressed in Rat1 cells. (D) Phosphopeptide mapping analysis of Myc₆-Skp2 wild-type, S64A and S72A. Arrows indicate the loss of phosphopeptides. (E) Rat1 cells were transfected with wild-type Skp2-HA and the indicated mutants. Cellular extracts were analysed by immunoblotting with Skp2 phospho-specific antibodies to Ser64 and Ser72. The higher molecular mass species corresponds to the Ser64-phosphorylated form of Skp2. ^* Marks nonspecific bands. (F) Cellular extracts of proliferating HeLa cells were analysed by immunoblotting with Skp2 phospho-specific antibodies. Specificity of antibodies was assessed by depletion of Skp2 using RNA interference. (G) Phylogenetic conservation of Skp2 phosphorylation sites.

Phosphorylation of Skp2 protects it from degradation by APC^Cdh1. (A) Rat1 cells were transfected with various amounts of DNA encoding wild-type Skp2-HA or the indicated mutants. Endogenous Skp2 and ectopic Skp2 were detected by anti-Skp2 immunoblotting. The position of the endogenous protein is indicated. (B) 293 cells were transfected with Skp2-HA. After 48 h, the cells were treated with cycloheximide for the indicated time intervals, in the absence or presence of MG132 (25 μM). Expression of Skp2 and GAPDH was analysed by immunoblotting. (C) HeLa cells were transfected with the indicated Skp2 constructs and synchronized in M phase by mitotic shake-off of cells obtained 8 h after release from a double-thymidine block. The cells were then re-plated and allowed to progress into G1 in the presence of cycloheximide (100 μg/ml). Skp2, cyclin B1, phospho-Ser10 H3 and GAPDH were detected by immunoblotting. The results are representative of four independent experiments. (D) Quantification of Skp2 degradation shown in panel C. (E) NIH 3T3 cells were co-transfected with Skp2-HA and increasing amounts of Flag-Cdh1. Lysates were analysed by immunoblotting with antibodies to HA, Flag and GAPDH. (F) 293 cells were co-transfected with wild-type Skp2-HA or the indicated mutants together with Myc₆-Cdh1. After 48 h, Myc₆-Cdh1 complexes were immunoprecipitated with anti-Myc antibody and analysed by immunoblotting. The ratio of unphosphorylated Ser64 to phospho-Ser64 for wild-type Skp2 in the lysate and the Myc immunoprecipitate was quantified from three independent experiments (lower panel). (G) Control and Skp2^−/− MEFs were co-transfected with Myc₆-EGFP together with low amounts of the indicated Skp2-HA constructs. The cells were then serum starved for 24 h and BrdU was incorporated during the last 10 h. The percentage of Myc-transfected cells that incorporated BrdU was evaluated by immunofluorescence. The results are expressed as mean±SEM of nine experiments (upper panel). Expression of EGFP and Skp2-HA was monitored by immunoblotting (lower panels).

Cdc14B dephosphorylates Skp2 on Ser64. (A) HeLa cells were transfected with Cdh1 siRNA and synchronized in S or M phase (FACS profiles). Cells were allowed to progress into the cell cycle in the presence of cycloheximide and roscovitine (100 μM) for the times indicated. Endogenous expression of Skp2 (upper panel), Cdh1 and GAPDH (lowers panels) was evaluated by immunoblotting. (B) Cell lysates prepared as in panel A were immunodepleted with the phospho-Ser64 antibody and analysed by immunoblotting with anti-Skp2 antibody. Note the enrichment of the faster migrating band of Skp2. In a parallel experiment, endogenous Skp2 was immunoprecipitated and subjected to extensive dephosphorylation using λ phosphatase (left panel). (C) HeLa cells were transfected with the ΔD-box mutant of Skp2. Exponentially proliferating cells, cells synchronized in S phase or cells at M/G1 transition were monitored for Skp2 expression by anti-HA immunoblotting. (D) Rat1 cells were co-transfected with Skp2-HA together with Flag-tagged Cdc14A or Cdc14B or their respective inactive mutants (^*). Lysates were analysed by immunoblotting. (E) 293 cells were transfected with low amounts of Skp2-HA. After 48 h, Skp2-HA was immunoprecipitated with anti-HA and incubated with in vitro-translated Cdc14A or Cdc14B for the indicated times. Skp2 dephosphorylation was analysed by anti-HA immunoblotting. When indicated (4 × ), four times more Cdc14A was used in the assay. (F) Endogenous Skp2 was immunoprecipitated from proliferating HeLa cells and incubated with in vitro-translated Cdc14A or Cdc14B for 2 h. Note that Skp2 dephosphorylation by Cdc14B results in the presence of a faster migrating band.

Cdc14B regulates S-phase entry in an Skp2-dependent manner. (A) HeLa cells were treated with siRNAs targeting Cdc14B and/or Skp2. Mitotic cells were collected and released into the cell cycle for the indicated times. CDC14B mRNA levels were measured by real-time Q-PCR. Expression of the indicated endogenous proteins was analysed by immunoblotting. (B) Mitotic HeLa cells treated as in panel A were re-plated on coverslips and allowed to progress into the cell cycle in the presence of BrdU. The percentage of cells that incorporated BrdU was evaluated by fluorescence microscopy. The results are expressed as mean±SEM of six experiments. (C) Model of the regulation of Skp2 expression by phosphorylation during cell cycle progression. During mitotic exit, Cdc14B actively dephosphorylates Skp2 on Ser64, thereby allowing its efficient association with APC^Cdh1 and subsequent degradation. Cdc14B might also stimulate APC^Cdh1 activity by promoting Cdh1 dephosphorylation. During G1 progression, induction of CDK2/cyclin E activity stabilizes Skp2 expression by phosphorylating Ser64, even in the presence of active APC^Cdh1. In S phase, CDK2/cyclin A further stabilizes Skp2 by inactivating Cdh1 by phosphorylation. Ser72 phosphorylation plays a significant but minor role in controlling Skp2 stability.

Phosphorylation of Skp2 on Ser64/Ser72 does not affect SCF^Skp2 activity. (A) 293 cells were co-transfected with epitope-tagged Cul1, Skp1, Rbx1, Cks1 and the indicated Myc₆-Skp2 constructs. SCF complexes were immunoprecipitated with anti-Myc antibody and analysed by immunoblotting. (B) Immunoprecipitated SCF^Skp2/Cks1 complexes obtained as in panel A were used in p27 in vitro ubiquitination assays. In vitro-translated ³⁵S-labelled wild-type (WT) p27 or T187A (TA) mutant was phosphorylated by recombinant CDK2/cyclin A before the reaction. Methyl ubiquitin was added to the assay. The reaction products were analysed by autoradiography. (C) HeLa cells were co-transfected with p27-Myc together with HA-tagged EGFP or Skp2 constructs. Expression of ectopic p27 and Skp2 was detected by immunoblotting.

Ser64 phosphorylation catalysed by CDK2 stabilizes Skp2 expression in G1 phase. (A) Expression of Skp2 occurs before other APC^Cdh1 substrates in G1 phase. Rat1 cells were made quiescent by contact inhibition and released into G1 by re-plating at low density for the times indicated. The expression of Skp2, Cdc20, cyclin B1, cyclin A, p27 and GAPDH was analysed by immunoblotting. ^* Marks nonspecific bands. (B) Rat1 cells expressing doxycycline-inducible HA-tagged Skp2 constructs were serum starved for 24 h and then stimulated with 10% serum and doxycycline (1 μg/ml) for the indicated times. HA-Skp2 and GAPDH were detected by immunoblotting. (C) Rat1 cells were co-transfected with Skp2-HA together with wild-type or dominant-negative forms (^*) of HA-CDK2 or HA-CDK4. Lysates were analysed by anti-HA immunoblotting. (D) NIH 3T3 cells were co-transfected with high amounts of Skp2-HA together with the indicated Myc₆-tagged cyclins. Lysates were analysed by anti-HA, anti-phospho-Ser64 and anti-Myc immunoblotting. (E) Contact-inhibited Rat1 cells were released into G1 in the presence of roscovitine (25 μM) or vehicle for the indicated times. Expression of endogenous Skp2 was analysed by immunoblotting. (F) Q-PCR analysis of Skp2 mRNA in Rat1 cells treated as in panel E. (G) Quiescent Rat1 cells expressing doxycycline-inducible Skp2-HA were stimulated with serum and doxycycline, in the presence or absence of roscovitine. Expression of Skp2 was analysed by immunoblotting. Rat1 (H) and NIH 3T3 (I) cells were transfected with the indicated constructs. Lysates from exponentially proliferating cells were analysed by immunoblotting.

The phosphatase activity of Cdc14B stimulates Skp2 degradation by APC^Cdh1. (A–C) NIH 3T3 cells were transfected with the indicated constructs and lysates were analysed by immunoblotting. (D) NIH 3T3 cells were co-transfected with Skp2-HA and Flag-Cdh1 with or without Myc₆-Cdc14B. After 48 h, the half-life of Skp2 was evaluated by cycloheximide-chase analysis. (E) HeLa cells were co-transfected with Skp2-HA together with Myc₆-Cdc14B (wild-type or inactive mutant) and synchronized in M phase. Cells were released into G1 in the presence of cycloheximide. Expression of Skp2, Cdc14B and GAPDH was analysed by immunoblotting. Mitotic exit was monitored by phospho-Ser10 histone H3 immunoreactivity. (F) HeLa cells were transfected with control siRNA or siRNAs targeting Cdc14B. Mitotic cells were released into G1 phase for the indicated times. CDC14B mRNA levels were measured by real-time Q-PCR (upper panel). Expression of endogenous Skp2, cyclin B1 and GAPDH was analysed by immunoblotting. (G) HeLa cells were transfected with either wild-type Skp2-HA or AA mutant and treated with control or Cdc14B siRNA. Mitotic cells were collected and allowed to enter G1 for 2 h. Lysates were analysed by immunoblotting.