Identification of novel human Cdt1-binding proteins by a proteomics approach: proteolytic regulation by APC/CCdh1

Abstract

In mammalian cells, Cdt1 activity is strictly controlled by multiple independent mechanisms, implying that it is central to the regulation of DNA replication during the cell cycle. In fact, unscheduled Cdt1 hyperfunction results in rereplication and/or chromosomal damage. Thus, it is important to understand its function and regulations precisely. We sought to comprehensively identify human Cdt1-binding proteins by a combination of Cdt1 affinity chromatography and liquid chromatography and tandem mass spectrometry analysis. Through this approach, we could newly identify 11 proteins, including subunits of anaphase-promoting complex/cyclosome (APC/C), SNF2H and WSTF, topoisomerase I and IIalpha, GRWD1/WDR28, nucleophosmin/nucleoplasmin, and importins. In vivo interactions of Cdt1 with APC/C(Cdh1), SNF2H, topoisomerase I and IIalpha, and GRWD1/WDR28 were confirmed by coimmunoprecipitation assays. A further focus on APC/C(Cdh1) indicated that this ubiquitin ligase controls the levels of Cdt1 during the cell cycle via three destruction boxes in the Cdt1 N-terminus. Notably, elimination of these destruction boxes resulted in induction of strong rereplication and chromosomal damage. Thus, in addition to SCF(Skp2) and cullin4-based ubiquitin ligases, APC/C(Cdh1) is a third ubiquitin ligase that plays a crucial role in proteolytic regulation of Cdt1 in mammalian cells.

Figure 1.

Resolution of Cdt1-binding proteins on CBB-stained SDS-PAGE. Nuclear extracts obtained from HeLa cells were loaded onto GST or GST-Cdt1 columns as described in Materials and Methods. The bound proteins were eluted and then separated in a 4–20% SDS-PAGE gel followed by CBB staining.

Figure 2.

Confirmation of interactions between Cdt1 and proteins identified by the MS analysis. (A) GST-Cdt1 or GST were incubated with HeLa cell nuclear extracts, and bound proteins were analyzed by CBB staining (left panel) or immunoblotting with the indicated antibodies. Fifteen percent of the input was also loaded. (B) Cdt1-PCNA complexes were immunoprecipitated with anti-T7 antibody from 293T cells stably expressing T7-tagged Cdt1, and immunoprecipitates (IP) were subjected to immunoblotting with anti-T7 or anti-PCNA antibodies. Five percent of the input (for T7-Cdt1) or 0.5% of the input (for PCNA) were analyzed to show the precipitation efficiency. (C–E) 293T cells were transfected with the indicated expression vectors, and nuclear extracts were prepared. After immunoprecipitation with anti-Cdt1 antibody or control rabbit IgG, the precipitates were blotted with the indicated antibodies. Ten percent of the input was loaded. (F) 293T cells were transfected with the indicated expression vectors, and the soluble fractions were prepared. After immunoprecipitation with anti-HA antibody, the precipitates were blotted with anti-HA or anti-T7 antibodies. Five percent of the input was loaded. (G) The 293T cells stably expressing T7-tagged Cdt1 were first treated with formaldehyde and then solubilized with SDS and sonication. The lysates were immunoprecipitated with anti-Cdt1 antibodies or control IgG, and the immunoprecipitates (IP) were immunoblotted with anti-Cdt1 or anti-Cdh1 antibodies. Five percent of the input for Cdt1 or 0.5% of the input for Cdh1 were loaded.

Figure 3.

Interaction of Cdt1 with APC/C^Cdh1 requires amino-terminal destruction boxes 1–3. (A) Top, schematic illustrations of the N-terminal amino acids of wild-type human Cdt1 and the mutants used in this study. Bottom, table summarizing the properties of the mutants. PIP-motif, PCNA-interaction protein motif (QxxI/L/VxxFF) required for PCNA binding and subsequent recognition by Cullin4-DDB1 ubiquitin ligase; Cy-motif, RXL-type cyclin-binding motif required for Cdt1 phosphorylation by cyclin A/Cdks and subsequent recognition by SCF^Skp2 ubiquitin ligase; and T29, threonine 29 residue phosphorylated by cyclin A/Cdks. (B) Wild type and the indicated mutant GST-Cdt1 or GST were incubated with HeLa cell nuclear extracts, and bound proteins were analyzed by immunoblotting with the indicated antibodies. PCNA and geminin served as control proteins whose binding to Cdt1 is not affected by the mutations. Twenty five percent of the input was also loaded. (C) 293T cells were transfected with wild-type T7-Cdt1, T7-Cdt1 Cy+D1m, or control vector, cross-linked with formaldehyde, and lysed as described in Figure 2G. The lysates were immunoprecipitated with anti-T7 antibody and analyzed by immunoblotting with the indicated antibodies. Five percent of the input for Cdt1, geminin, and Skp2 or 0.5% of the input for Cdh1 and E2F-1 was analyzed to show the precipitation efficiency.

Figure 4.

Elimination of D-boxes 1–3 stabilizes Cdt1 and confers partial resistance to Cdh1 overexpression in 293T cells. (A) Overexpression of Cdh1 promotes Cdt1 degradation in 293T cells. 293T cells were transfected with wild-type T7-Cdt1 expression vector (3 μg), along with a construct expressing myc-Cdh1 or myc-Cdc20 (3 μg), and whole cell lysates were prepared 48 h after transfection. When necessary, cells were supplemented with 40 μM MG132 for 5 h before harvest. Protein levels of T7-Cdt1, myc-Cdh1, and myc-Cdc20 were measured by immunoblotting with anti-T7 (left panel, top row) or anti-myc (middle row) antibodies. The membranes were also subjected to CBB staining to show equal loading (bottom row). The signal intensities of the bands were quantified, and the mean and SDs from two independent experiments are shown with Cdt1 alone without MG132 set at 100 (right panel). (B) Overexpression of Cdh1 promotes degradation of endogenous Cdt1. 293T cells were transfected with myc-Cdh1 (6 μg) and GFP (0.1 μg), and whole cell lysates were immunoblotted with the indicated antibodies. (C and D) Triple mutations in the D-boxes 1–3 increase the stability of 3HA-Cdt1[2-101] proteins. 293T cells were transfected with 3HA-Cdt1[2-101] expression vectors (3 μg), along with plasmids expressing myc-Cdh1 (3 μg) and GFP (0.1 μg). (C) Left panel, whole cell lysates were prepared 48 h after transfection. Protein levels of 3HA-Cdt1[2-101], myc-Cdh1, and GFP were measured by immunoblotting with anti-HA, anti-myc, or anti-GFP antibodies. The membranes were also subjected to CBB staining to confirm equal loading. The signal intensities of the bands were quantified, and the means and SDs from two independent experiments are shown with 3HA-Cdt1[2-101] D1,2,3m without Cdh1 set at 100 (right panel). (D) Cycloheximide (100 μg/ml) was added to the medium 48 h after transfection, and cells were sequentially harvested at the indicated time points for immunoblot analysis. (E and F) Triple mutations in the D-boxes 1–3 increase the stability of full-length Cdt1 proteins. Similar experiments to C and D were performed with full-length T7-Cdt1 expression vectors. For detection of T7-Cdt1, anti-T7 antibody was used.

Figure 5.

Silencing of Cdh1 stabilizes Cdt1 mutants with impaired binding to Skp2 in HeLa cells. (A) HeLa cells stably expressing the T7-Cdt1 Cy mutant were transfected with siRNA corresponding to a nonrelevant mRNA (control siRNA) or to a portion of Cdh1 mRNA (termed Cdh1-1). Forty-eight hours after transfection, whole cell lysates were subjected to immunoblotting (left). The signal intensities of the bands were quantified, and the mean and SDs from five independent experiments are shown with endogenous Cdt1 or T7-Cdt1 Cy in cells treated with control siRNA set at 100 (right panel). (B) HeLa cells stably expressing T7-Cdt1 T29A, Cy, or Cy+D1m were transfected with the control siRNA, siRNA Cdh1-1, and another siRNA against Cdh1 (termed Cdh1-2). Forty-eight hours after transfection, whole cell lysates were subjected to immunoblotting (top). The signal intensities of the bands were quantified and are shown with endogenous Cdt1 or each of the T7-Cdt1 mutants in cells treated with control siRNA set at 100 (bottom).

Figure 6.

APC/C^Cdh1-dependent ubiquitination of Cdt1 in vitro. (A) APC/C^Cdh1 complexes were purified from HeLa cells synchronized in G1 phase with anti-Cdc27 antibody. Recombinant Cdt1 proteins were incubated with E1, E2 (UbcH10), His-tagged ubiquitin, ATP, and the immunopurified APC/C or the control immunoprecipitate at 30°C for 180 min, separated by SDS-PAGE, and subjected to immunoblotting with anti-Cdt1 (left panel) or anti-His (right panel) antibodies. (B) The experiment was repeated using either the wild-type Cdt1 or the Cdt1 Cy+D1m mutant. The reacted samples were immunoblotted with anti-Cdt1 antibody.

Figure 7.

The Cdt1 mutant with the triple mutations in the D-boxes 1–3 induces the strongest rereplication, chromosomal damage, and checkpoint activation. (A) 293T cells were transiently transfected with T7-Cdt1 expression vectors (3 μg), along with a plasmid expressing GFP (0.1 μg) and an empty vector (3 μg) as a carrier. Whole cell lysates were prepared 48 h after transfection and immunoblotted with the indicated antibodies. The signal intensities of the bands were quantified, and the means and standard deviations from four independent experiments are shown with the Cdt1 wild type set at 100 (right panel). (B–E) HeLa cells were infected with the high-titer retroviruses expressing wild-type, T29A, or Cy+D1m T7-Cdt1 or control retroviruses and selected. (B) At 4 d after infection, cells were stained with DAPI to visualize nuclei. Scale bars, 50 μm. In the right panel, the percentage of cells with large nuclei in which the major axis of nuclei was larger than 26.6 μm, corresponding to the average + 2 SDs for that of cells infected with control viruses, is shown. Two hundred random nuclei were measured for each and the means and SDs from two independent experiments are shown. (C) At 4 d after infection, cells were collected and DNA content was analyzed by flow cytometry. The means and SDs of the percentage of rereplicated cells (the DNA content higher than. 5N) from two independent experiments are shown (right panel). (D) Expression of T7-Cdt1 Cy+D1m induces rereplication without centrosome overduplication. At 4 d after infection, cells were stained with anti-γ-tubulin antibody and the number of γ-tubulin spots was counted at least in 200 cells. Cells infected with the control virus were analyzed randomly, and as to T7-Cdt1 Cy+D1m infection, cells larger than 26.6 μm were analyzed. The typical images of cells with 1 (top), 2 (middle), or >2 (bottom) centrosomes are shown in left panels. Scale bars, 20 μm. The percentages of cells with the indicated centrosome numbers are shown in right. (E) At 6 d after infection, whole cell lysates were immunoblotted with the indicated antibodies, and the signal intensities of the bands were quantified, here shown with the Cdt1 wild type set at 100 (right panel).

Figure 8.

Cdt1 is destabilized in G0 dependent on D-boxes 1–3 and their elimination augments its function to induce chromosomal damage in G0. (A) T98G cells were infected with retroviruses expressing either T29A or Cy+D1m mutant T7-Cdt1. At 3 wk after infection, infected and parental cells were arrested in a quiescent state. Forty-eight hours after serum starvation, whole cell lysates were subjected to immunoblotting with the indicated antibodies (left). The signal intensities of the bands were quantified and shown with the values at 0 h set at 100 (for each of Cdt1 proteins) or with the maximum values set at 100 (for p-ATM and p-Chk2) (right). (B) T98G cells expressing T7-Cdt1 T29A were transfected with a mixture of siRNAs against Cdh1 (Cdh1-1 + Cdh1-2) and, 18 h after transfection, were deprived of serum for the indicated times. Whole cell lysates were then subjected to immunoblotting with the indicated antibodies (left). The signal intensities of the bands were quantified and shown with the values at 0 h set at 100 (right). The means and SDs from two independent experiments are shown.

Figure 9.

A model for roles of APC/C^Cdh1-mediated Cdt1 proteolytic regulation. For details, see the text.