Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 74 (14), 3870-9

Loss of Cdk2 and Cyclin A2 Impairs Cell Proliferation and Tumorigenesis

Affiliations

Loss of Cdk2 and Cyclin A2 Impairs Cell Proliferation and Tumorigenesis

Lakshmi Gopinathan et al. Cancer Res.

Abstract

Cell-cycle inhibition has yet to offer a generally effective approach to cancer treatment, but a full evaluation of different combinations of cell-cycle inhibitors has not been evaluated. Cyclin A2, a core component of the cell cycle, is often aberrantly expressed in cancer where it may impact cell proliferation. In this study, we investigated the role of cyclin A2 in tumorigenesis using a conditional genetic knockout mouse model. Cyclin A2 deletion in oncogene-transformed mouse embryonic fibroblasts (MEF) suppressed tumor formation in immunocompromised mice. These findings were confirmed in mice with cyclin A2-deficient hepatocytes, where a delay in liver tumor formation was observed. Because cyclin A2 acts in complex with Cdk2 in the cell cycle, we explored a hypothesized role for Cdk2 dysregulation in this effect through conditional deletions of both genes. In oncogene-transformed MEFs lacking both genes, tumor formation was strongly suppressed in a manner associated with decreased proliferation, premature senescence, and error-prone recovery from serum deprivation after immortalization. Whereas loss of cyclin A2 led to a compensatory increase in Cdk1 activity, this did not occur with loss of both Cdk2 and cyclin A2. Our work offers a rationale to explore combinations of Cdk1 and Cdk2 inhibitors as a general approach in cancer therapy.

Figures

Fig. 1
Fig. 1. Delayed liver tumorigenesis in the absence of cyclin A2
Liver tumors were induced by hydrodynamic tail vein injection of Ras/shRNA p53 in wild type (cyclin A2WT) or liver-specific cyclin A2 knockout (cyclin A2Liv−/−) mice. Tumor formation (assessed over 8 months) is delayed in cyclin A2Liv−/− mice. 4-6 mice were assessed per time point. Wild type mice with tumors were euthanized within 4 months (A, C). “X” indicates that mice with tumors do not survive for 6 months (C). Cyclin A2Liv−/− livers display a decreased Ki67 proliferative index (BC). *p < 0.05, Student’s t-test.
Fig. 2
Fig. 2. Transformed Cdk2nullcyclin A2null MEFs exhibit decreased tumorigenic potential in nude mice
Cyclin A2flox and Cdk2nullcyclin A2flox MEFs were oncogenically transformed with Ras/p53DN and treated with 4-OHT to induce cyclin A2 knockout. Transformed MEFs were assessed for colony formation (A) or injected subcutaneously into nude mice (B-C). Cdk2nullcyclin A2null MEFs are resistant to transformation (A) and display decreased tumorigenic potential (B-C). Mice in (B) were photographed 4 weeks after MEF allografts. Data is representative of two independent MEF clones and 6 nude mice were used per genotype.
Fig. 3
Fig. 3. Impaired proliferation of primary liver tumor cells upon inhibition of cyclin A2 and Cdk2
Liver tumors were induced by tail vein injection of Ras/shRNA p53 in cyclin A2flox/floxRosa26-CreERT2Tg/Tg mice. Tumors were isolated followed by dissociation and culturing of tumor cells (A-C). Cyclin A2 knockout in primary liver tumor cells was achieved by addition of 4-OHT, whereas Cdk2 was silenced by retroviral shRNA transduction (A). Cyclin A2nullshCdk2 liver tumor cells exhibit decreased proliferation rates as determined by alamarBlue proliferation assay (B) and resistance to colony formation (C). Data in (A-C) is representative of two tumor cell lines established from two mice. NPIU: normalized phosphoimager units.
Fig. 4
Fig. 4. Concomitant loss of Cdk2 and cyclin A2 results in impaired proliferation and premature senescence
Primary cyclin A2flox and Cdk2nullcyclin A2flox MEFs were treated with 4-OHT to induce cyclin A2 knockout and their proliferative potential was determined by alamarBlue proliferation assays (A). Cdk2nullcyclin A2null MEFs display impaired proliferation rates when compared to other genotypes. Loss of cyclin A2 and Cdk2 resulted in an increased number of prematurely senescent cells in early passages, as detected by β-galactosidase staining (B). Analysis of cell cycle profile by BrdU labeling and propidium iodide (PI) staining revealed increased population of cells in S phase, and cells with 4n and >4n DNA content in Cdk2nullcyclin A2null cells (C). Data is representative of three independent MEF lines. Cyclin A2flox and Cdk2nullcyclin A2flox MEFs were treated with 4-OHT to induce cyclin A2 knockout and were cultured over several passages using a 3T3 assay at 21% (left panel) and 3% (right panel) oxygen to determine long-term propagative potential (D). Of 7 clones tested, only one Cdk2nullcyclin A2null clone survived the 3T3 assay at 21% oxygen. MEFs were collected at various passages during the course of the 3T3 assay at 21% oxygen conditions and the absence of cyclin A2 protein in cyclin A2null and DKO MEFs was confirmed by Western blotting of protein extracts (E).
Figure 5
Figure 5. Activity of Cdks in primary MEFs
Primary MEFs were synchronized at G0/G1 by serum starvation for 72 hours and simultaneously treated with 4-OHT to induce cyclin A2 knockout. MEFs were released into S phase by serum addition and were collected at different time points. Protein extracts were subjected to immunoprecipitation with the indicated antibodies followed by in vitro kinase assays using radiolabeled ATP and histone H1 as substrates. Data is representative of three independent kinase assays performed with three different MEF clones. (A). Quantitative analysis of Cdk1-associated activity is shown as an average of phosphoimager units for three MEF clones at 24, 30, 48, and 72 hours following serum starvation. The average value obtained for cyclin A2flox MEFs was normalized to one and values for the other genotypes were calculated comparatively (B). NPIU: normalized phosphoimager units. Protein extracts were prepared from cells and subjected to co-immunoprecipitation with antibodies against cyclin B1, followed by SDS-PAGE and Western blot with antibodies against Cdk1 (C). Increased Cdk1/cyclin B1 complexes are detected in cyclin A2null MEFs but not in the other genotypes. Synchronized primary MEFs were treated with different concentrations of Cdk1 inhibitor RO3306 and their proliferative potential was determined by alamarBlue proliferation assays. Data is representative of three independent proliferation assays performed with three different MEF clones. (D). Loss of cyclin A2 renders MEFs more sensitive to Cdk1 inhibition. *p < 0.05, Student’s t-test; proliferative rate (slope of graphical trendline) of cyclin A2null MEFs is significantly lower than cyclin A2flox MEFs at 5μM RO3306.

Similar articles

See all similar articles

Cited by 29 PubMed Central articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback