In vitro phosphorylation of p21^Cip1. (A) myc-tagged p21^Cip1 (p21 wt or the nonphosphorylatable T145A construct), active Akt (T308D/S473D), or inactive Akt (K179 M) was overexpressed in COS-7 cells and immunoprecipitated with an anti-myc antibody. The immunoprecipitates were combined, and kinase activity towards p21^Cip1 was assayed as described in Materials and Methods. Phosphorylated proteins were resolved by SDS-PAGE. The lower panels demonstrate the equal expression of the proteins by Western blot analysis (WB) with anti-myc antibodies. (B) Comparison of Akt and PKA in vitro kinase activities towards p21 wt and Thr 145 constructs. (C) Lack of an effect of SGK in phosphorylating p21^Cip1 in vitro. As a positive control, histone phosphorylation by SGK is shown. (D) Lack of p21Cip1 in vitro phosphorylation activity by p70 S6 kinase. Histone phosphorylation is shown as a positive control. Immunoblot analysis demonstrates p21^Cip1 expression (top) and p70 S6 kinase immunopurification (IP) (right). Representative autoradiographs out of at least three different experiments are shown (A through D).

Regulation of PCNA binding by Akt-dependent phosphorylation of p21^Cip1. (A) Coimmunoprecipitation of p21^Cip1-PCNA complexes. myc-tagged p21^Cip1 constructs were expressed in COS-7 cells and immunoprecipitated (IP) with an anti-myc antibody followed by Western blotting (WB) against PCNA (n = 4 different experiments). (B) Endogenous PCNA was immunoprecipitated from COS-7 cells overexpressing the various myc-tagged p21^Cip1 constructs followed by Western blot analysis using an anti-myc antibody (n = 3). (C) COS-7 cells were cotransfected with myc-tagged p21^Cip1 and vector, active Akt (T308D/S473D), or kinase-inactive Akt (K179 M). Then, PCNA was detected by immunoblot analysis of anti-myc immunoprecipitates (n = 3). (D) COS-7 cells were transfected with p21^Cip1 constructs mutated at Thr 145 and/or Ser 146. Again, endogenous PCNA was detected from anti-myc immunoprecipitates. Right lane, anti-myc antibody without cell lysate. Expression of p21^Cip1 constructs and endogenous PCNA is shown in the lower panels (A through D). Similar results were obtained using HUVEC.

Regulation of Cdk function by p21^Cip1 phosphorylation at Thr 145. (A) Effect of p21^Cip1 phosphorylation at Thr 145 on cyclin D-Cdk4 assembly. Endogenous cyclin D was detected in Cdk4 immunoprecipitates (IP) from HUVEC overexpressing the various Thr 145 constructs of p21^Cip1 (n = 4). The lower panel shows endogenous cyclin D expression in HUVEC transfected with vector or p21^Cip1 constructs. (B) Densitometric analysis of cyclin D expression and binding to Cdk4 in p21^Cip1-transfected HUVEC. Data are mean ± SEM, n = 4, *P < 0.05. (C and D) Cdk2 kinase activity. Immunoprecipitated Cdk2 from HUVEC overexpressing p21^Cip1 constructs was incubated with histone H1 as an in vitro substrate. A representative autoradiograph is shown in panel C. The quantitative analysis is shown in panel D (n = 4, *, P < 0.05 versus the wild type). WB, Western blotting.

Regulation of endothelial cell proliferation by the PI3K/Akt pathway. (A and B) Serum-stimulated HUVEC were incubated with the PI3K inhibitor Ly294002 (10 μM) or starved for 24 h in serum-free medium plus 1% bovine serum albumin. Cell cycle phases were detected by FACS (data are mean ± SEM; n = 4). (C) HUVEC were incubated with Ly294002 (10 μM) for 24 h, and cell proliferation was assessed by measuring BrdU incorporation via enzyme-linked immunosorbent assay (data are mean ± SEM; n = 4). (D) HUVEC were cotransfected with GFP (1 μg) and the respective pcDNA3.1 constructs (2 μg) and incubated for 20 h. During the last 12 h, cells were starved in FCS-free medium plus 1% bovine serum albumin. GFP-positive cells were isolated, and proliferating endothelial cells were identified by immunostaining against the proliferative marker protein Ki67. Nuclei were counterstained with DAPI. Ki67-positive cells were counted, and values were calculated as the number of Ki67-positive cells/the number of DAPI-stained nuclei × 100 (data are mean ± SEM; n = 3).

Proposed Akt effector pathways regulating cell cycle progression. The scheme illustrates the data reported in this study in the context of previous publications demonstrating p27 and cyclin D regulation in response to Akt (15, 24, 33, 34). FKH, Forkhead transcription factors; GSK-3, glycogen synthase kinase 3.

Akt phosphorylation site of p21^Cip1: sequence homology and localization. Cy1 (amino acids [aa] 17 to 24) and Cy2 (aa 152 to 158), cyclin-binding motifs (6, 7); Cdk, Cdk-binding motif (7); PCNA binding, PCNA binding site (aa 144 to 151) (23, 46).

In vivo phosphorylation of p21^Cip1. (A) Detection of phosphorylated p21^Cip1 by immunoblot analysis of HUVEC extracts using a phospho-specific antibody against the Akt phosphorylation consensus motif. Right lane, effect of Ly294002 (10 μM) for 1 h before lysis. The lower panel shows total endogenous p21^Cip1 for comparison. (B) In vivo phosphorylation of p21^Cip1 by a serum-induced, PI3K-sensitive mechanism. HUVEC were labeled with ³²P and starved for 1 h in FCS-free medium before the addition of 10% phosphate-free FCS and Ly294002 (10 μM) for 30 min as indicated. Endogenous p21^Cip1 was immunoprecipitated (IP) with anti-p21^Cip1 antibodies. (C). In vivo phosphorylation of p21^Cip1 by Akt. COS-7 cells overexpressing myc-tagged p21^Cip1 constructs and vector (pcDNA3.1) or Akt constructs were labeled with ³²P for 3 h, and p21^Cip1 was immunoprecipitated with anti-myc antibodies. In panels B and C, representative autoradiographs are shown: lower panels, expression of p21^Cip1 as a loading control. (D) Phospho-peptide analysis. (Top) Elution of the synthetic peptide QTSMTDFYHSK (where T is the phospho-acceptor amino acid) corresponding to amino acids 144 to 154 of p21 from the reverse phase column. This peptide contains the putative Akt phosphorylation site (¹⁴⁵Thr) in p21 and would result from tryptic digestion of p21. (Middle) Endogenous p21 was immunoprecipitated from COS cells, and tryptic p21 peptides were separated by reverse-phase chromatography. Note the peak corresponding to the peptide QTSMTDFYHSK. (Bottom) Following serum starvation, COS cells were serum treated in the presence of ³²P with or without Ly294002 (10 μM). Tryptic p21 peptides were seperated as above, and fractions surrounding peptide QTSMTDFYHSK were collected for determination of radioactivity. A representative result is shown. WB, Western blotting.

Effect of p21^Cip1 phosphorylation at Thr 145 on Cdk binding. COS-7 cells were transfected with p21^Cip1 constructs, and endogenous Cdk2 (A) or Cdk4 (B) was detected from anti-myc immunoprecipitates (IP) by immunoblot analysis (n = 4 experiments). The expression of the constructs is shown in the lower panel.

Subcellular distribution of p21^Cip1 in dependency on Thr 145 phosphorylation. (A) Immunostaining of human endothelial cells transfected with p21 wt or the Thr 145 constructs. Immunocytochemistry was performed using anti-myc antibody and counterstained with DAPI (n = 4). Due to the limited transfection efficiency, not all cells reveal FITC staining. (B) Immunoblot analysis of p21^Cip1 protein levels in the nuclear versus cytoplasmic fraction of HUVEC overexpressing p21 wt and Thr 145 constructs. (C) Densitometric analysis of four individual experiments as in panel B. Data are mean ± SEM; *, P < 0.05 versus the wild type and T145A. (D) Relative subcellular distribution of Akt in nucleus and cytosol of HUVEC transfected with p21^Cip1 constructs or vector. A representative result of three individual experiments is shown. In the results shown in panels B and D, immunoblots were reprobed with anti-topoisomerase antibody to verify separation of the nuclear fraction and as a loading control. WB, Western blotting.

Effect of p21^Cip1 Thr 145 phosphorylation on endothelial cell proliferation. (A) HUVEC were cotransfected with GFP (1 μg) and the respective pcDNA3.1 constructs (2 μg) and incubated for 20 h. Then, cells were treated with Ly294002 (10 μM) for 12 h. GFP-positive cells were isolated by FACS, and proliferative endothelial cells were identified by immunostaining against the proliferative marker protein Ki67 (data are mean ± SEM; n = 3 to 4). (B) (Top) Western blot analysis (WB) of pRb phosphorylation in intact cells; the upper band corresponds to hyperphosphorylated pRb and the lower band corresponds to hypophosphorylated pRb in HUVEC overexpressing p21 wt or T145D in the absence or presence of Ly294002. In the second panel from the top, expression of p21^Cip1 constructs is shown; in the third panel, an autoradiograph of in vitro-phosphorylated pRb in a Cdk2 kinase assay is shown. The bottom panel shows a Western blot of Cdk2 immunoprecipitates from the respective lysates. (C) HUVEC were transfected with p21^Cip1 antisense or sense oligonucleotides, and p21^Cip1 protein expression was detected by Western blot analysis using an antibody against endogenous p21^Cip1. A representative blot of three individual experiments is shown. (D) Effect of Akt in p21^Cip1 antisense-transfected HUVEC that were cotransfected with p21^Cip1 antisense or sense oligonucleotides, GFP, and active Akt (T308D/S473D) or vector. Eighteen hours after transfection, the proliferation of GFP-positive cells was assessed by BrdU staining, followed by counterstaining with DAPI. Representative photomicrographs are shown (data are mean ± SEM, n = 3; P < 0.05 versus p21 sense plus vector).