Figure 1.
Gene expression changes induced by p21 in K562 cells.
A. Expression of p21 in Kp21-4 cells in response to ZnSO4. Cells were treated with 75 µM ZnSO4 for the periods of time indicated. Cell extracts were prepared and the p21 levels analyzed by immunoblot. B. The transcriptome of Kp21-4 cells (Kp21) treated for 12 h with ZnSO4 (to induce p21) were compared to that of cells with induced p27 (Kp27), parental K562 and Kp21-4 treated for 72 h with ZnSO4. The heat map shows the hierarchical clustering with those genes with expression variation ≥2.3-fold between uninduced and p21-induced Kp21-4 cells (P<0.001). The heat map shows 350 regulated genes (360 gene probes). C. Distribution of the regulated genes shown in A according with their cell functions. The ontogeny analysis has been carried out with the dChip program. D. Expression changes in 65 genes related to cell cycle and mitosis according to the ontogenic classification. The heat map was obtained as described in B. The genes further validated by RT-qPCR (Figure 2A) are shown in red.
Figure 2.
p21-mediated the down-regulation of genes involved in cell cycle.
A.The expression of p21 was induced in Kp21-4 cells by 75 µM ZnSO4 and 3, 6 and 12 h later, total RNA was prepared and expression of the indicated genes was determined by RT-qPCR. In some cases an alternative name is given into brackets. Cyc., cyclin. The values are means ±S.E.M. from two independent experiments and two determinations for each RNA. B. Kp21-4 cells were treated for 6 h with 75 µM ZnSO4 to induce p21. Chromatin immunoprecipitation was carried out with anti-histone H3 and anti-acetylated-histone H3 antibodies and (as specificity control) rabbit IgG. The DNA in the immunoprecipitated chromatin was measured by quantitative PCR. The amplicons encompass the transcription start site of the indicated genes. A regulatory sequence of the promoter of rDNA was used as a control. The results are expressed as the ratio of DNA enrichment in chromatin immunoprecipitated with anti-H3 versus acetylated-H3 (“Ac-H3”) and normalized to the values obtained in uninduced cells. The values are the means ±S.E.M of two independent ChIP experiments with two PCR determinations each.
Figure 3.
p21-mediated repression of genes is not dependent on CDK2.
A. K562 cells were transiently transfected with a short-hairpin CDK2 (“shCDK2”) vector and the empty vector (pLKO.1, “Vo”). 24 h after transfection the cells were treated with 75 µM ZnSO4 for 12 h to induce p21 and the silencing of CDK2 was assayed at the protein level by immunoblot (left panel) and at the mRNA level by RT-qPCR (right panel) The expression of p21 and actin was also determined to control the p21 induction by Zn2+ and the protein level, respectively. B. The expression of p21 was induced with 75 µM ZnSO4 in K562 transfected with sh-CDK2 and 12 h later, total RNA was prepared and expression of the indicated genes was determined by RT-qPCR. The values are means ±S.E.M. from two independent experiments and two determinations for each RNA.
Figure 4.
The N-terminal region of p21 is required for its effect as gene repressor.
A. Schematic representation of the p21 constructs used. The proteins carry the green fluorescent protein (GFP) in the N-terminal. The position of the CDK binding domain (CDB) and the PCNA binding domain (PCNA) are indicated. B. K562 cells were transfected with expression vectors for the full-length p21 protein (FL), the p21 amino-terminal region (NT) and the p21 carboxy-terminal region (CT). 24 h after transfection the cells were sorted by flow cytometry and the expression of the indicated genes was analysed by RT-qPCR. The values are means ±S.E. from two transfections and two determinations of each mRNA.
Figure 5.
p21 binds to human CCNE2 promoter.
A. Kp21-4 cells were treated for 12 h with ZnSO4 and the nuclear extracts were fractionated between insoluble (chromatin) fraction and soluble (nucleoplasmic) fraction. The p21 levels were determined by immunoblot. The transcription factor UBF was used as control of chromatin-bound protein. B. Cells were treated for 12 h ZnSO4 to induce p21 (in Kp21-4 cells), p27 (in Kp27-5 cells), parental K562 cells were also treated with ZnSO4 as negative control. Chromatin immunoprecipitation was performed with anti-p21, anti-p27 and anti-CDK2 antibodies as indicated, and (as specificity control) rabbit IgG. The DNA in the immunoprecipitated chromatin was analysed by quantitative PCR. The amplicons encompass the transcription start site of the indicated genes. rRNA promoter sequences were used negative controls. The results are expressed as enrichment of DNA in chromatin immunoprecipitated with anti-p21 (with respect the signal with anti-rabbit IgG) in cells induced with Zn2+, and normalized to the values obtained in uninduced cells. The values are the means ±S.E.M of three PCR determinations, each from two independent ChIP experiments.
Figure 6.
p21 represses the activity of human CCNE2 promoter.
A. K562 cells were transfected with luciferase reporters carrying the promoters cyclin E1 and E2 genes (CCNE1 and CCNE2), along with an expression vector for p21 and the corresponding empty vector, and a beta-gal plasmid for transfection efficiency normalization. 24 h after transfection the luciferase activities were determined. The data are normalized to the activity of cells transfected with the empty vector. Lower panel: immunoblot analysis of the transfected cells to assess the expression of p21. B. Kp21-4 cells were transfected with luciferase reporters carrying the promoters cyclin E1 and E2 genes as in (A) and 24 h later the cells were treated with 75 µM ZnSO4 and after 24 h the luciferase activities were determined. The data are normalized to the activity of cells without ZnSO4. Lower panel: immunoblot analysis of the transfected cells to assess the expression of p21.
Figure 7.
Genes regulated by p21 in human keratinocytes related to cell cycle and cell division.
A. Primary human keratinocytes were infected with recombinant adenoviruses expressing p21 (Ad-p21), p27 (Ad-p27). As a control, cells were also infected with adenovirus expressing GFP (Ad-GFP). The mRNA expression of p21 (upper graph) and p27 (lower graph) was determined by RT-qPCR 24 h after infection. Data are represented relative to the expression in Ad-GFP-infected cell. B. Heat-map showing the 82 genes changed by p21 related to cell cycle and cell division with an expression change > 2.3-fold. The names of the regulated genes are indicated at the right. Green indicates genes down-regulated by p21 and red genes up-regulated.
Figure 8.
p21-mediated down-regulation of genes involved in mitotic control in human keratinocytes.
A. Primary human keratinocytes were infected with adenovirus expressing p21 and ad-GFP as a control. 24 h after infection total RNA was prepared and the mRNA levels of the indicated genes were analysed by RT-qPCR. The values are means ±S.E. from three determinations. B. keratinocytes were transfected with siRNA for p21 and a negative control siRNA. Total RNA was prepared and the mRNA levels of the indicated genes were analysed by RT-qPCR. C. Primary human keratinocytes were transfected with p21 siRNA or control siRNA. 48 h after lipofection total RNA was prepared and the mRNA levels of the indicated genes were analysed by RT-qPCR.
Figure 9.
Putative regulatory sites in genes down-regulated by p21.
Schematic representation of the E2F (red), CDE (blue) and CDE/CHR (green) sites in the p21-regulated genes analysed in Figure 2. The region analysed encompass 2 kb upstream and downstream the transcription start site (marked by an arrow). Exons are represented as grey boxes. E2F sites: TTTC/GC/GC/GC/G. CDE sites: T/GGGCGG. CDE/CHR sites: GCGC/GN2–5 TTA/GAA.