Restoration of an absent G₁ arrest and protection from apoptosis in embryonic stem cells after ionizing radiation

Abstract

Somatic cells have genomic requirements that are very different from those of germ cells or embryonic stem (ES) cells. Somatic cells have restricted patterns of gene expression characteristic of their specific differentiated lineages. ES cells retain the potential to produce any cell type in the body. The consequences of mutation in a somatic cell are limited to that particular cell lineage and may result in somatic diseases, e.g., cancer, but will not be passed on to the progeny. In contrast, mutation in ES cells potentially can compromise multiple cell lineages and affect the well-being of subsequent generations. Thus, ES cells have mechanisms, beyond those used by somatic cells, that protect the integrity of their genomes. One mechanism involves suppression of mutation and mitotic recombination. The spontaneous mutation frequency at the endogenous mouse adenine phophoribosyltransferase (Aprt) reporter gene in somatic cells in vivo is very high, approaching 10 ⁽1, 2). In contrast, the mutation frequency in mouse ES cells is suppressed by about two orders of magnitude (3). A complementary mechanism to maintain genomic integrity is facilitated cell death that rids cells with a mutational burden from the population. Consistent with this proposition, mouse ES cells lack a G₁ checkpoint (4, 5) and are hypersensitive to DNA damage (6, 7).

Two known pathways, both of which involve inhibition of Cdk2 activity, govern the G₁/S checkpoint in somatic cells (8, 9). A rapid but transient response functions through Cdc25A degradation, whereas a p53/p21-mediated pathway supports a sustained but delayed response (10). In ES cells, p53 protein appears not to be fully functional because it is predominantly cytoplasmic and inefficiently translocated to the nucleus after DNA damage (4). Also, the Cdk inhibitors p21 and p27 are undetectable (11). In somatic cells, after DNA damage by ionizing radiation (IR), the ATM-Chk2-Cdc25A checkpoint pathway is initiated by autophosphorylation and activation of the phosphatidylinositol 3-kinase, ATM (12, 13). Activated ATM, in turn, phosphorylates Chk2, a serine/threonine kinase that phosphorylates several effector proteins such as BRCA1 and members of the Cdc25 family of phosphatases (14-16). In unperturbed cycling somatic cells, the Cdc25A phosphatase facilitates the transition from G₁ to S phase by removing the inhibitory phosphates on the Cdk2/cyclin E complex (17, 18). After ionizing irradiation, the activated Chk2 phosphorylates Cdc25A on serine 123, causing its ubiquitin-mediated degradation (19). Thus, in the absence of functional Cdc25A, Cdk2 is not dephosphorylated and the cells arrest in G₁. A second, but slower, response involves activation of p53 and induction of the Cdk inhibitor, p21.

This report establishes that the absence of a G₁ checkpoint after IR exposure in ES cells is due to an altered intracellular localization of Chk2 kinase. In somatic cells, Chk2 is predominantly diffuse within the nucleus. Here we show that in ES cells, Chk2 localizes to centrosomes and is unavailable to phosphorylate some of its substrates. Furthermore, the data demonstrate that a G₁ arrest can be restored by ectopic expression of Chk2 and that the restoration of a G₁ arrest protects cells from apoptosis. We argue that natural absence of a G₁ checkpoint in ES cells is beneficial to survival of the ES cell population and the species by facilitating removal of ES cells with damaged DNA and thus maintaining a pristine cell population. Conversely, acquisition of a G₁ checkpoint by ES cells and protection from apoptosis would be disadvantageous.

Acknowledgments

Notes

This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: ES, embryonic stem; IR, ionizing radiation; MEF, mouse embryonic fibroblast.