Etiology of endometrial cancer (EMC) is not fully understood. Animal models with rapidly and spontaneously developing EMC will help explore mechanisms of cancer initiation and progression. Pten +/À mice are currently being used as a model to study EMC. These females develop atypical endometrial hyperplasia of which f20% progresses to EMC. In addition, tumors develop in other organs, complicating the use of this model to specifically study EMC. Here, we show that conditional deletion of endometrial Pten results in EMC in all female mice as early as age 1 month with myometrial invasion occurring by 3 months. In contrast, conditional deletion of endometrial p53 had no phenotype within this time frame. Whereas mice with endometrial Pten deletion had a life span of f5 months, mice with combined deletion of endometrial Pten and p53 had a shorter life span with an exacerbated disease state. Such rapid development of EMC from homozygous loss of endometrial Pten suggests that this organ is very sensitive to this tumor suppressor gene for tumor development. All lesions at early stages exhibited elevated Cox-2 and phospho-Akt levels, hallmarks of solid tumors. More interestingly, levels of two microRNAs miR-199a * and miR-101a that posttranscriptionally inhibit Cox-2 expression were down-regulated in tumors in parallel with Cox-2 upregulation. This mouse model in which the loxP-Cre system has been used to delete endometrial Pten and/or p53 allows us to study in detail the initiation and progression of EMC. These mouse models have the added advantage because they mimic several features of human EMC.

Male mice lacking cyclin A1 protein are sterile. Their sterility results from an arrest in the meiotic cell cycle of spermatocytes, which we now identify as occurring at late diplotene, immediately before diakinesis. The stage of arrest in cyclin A1-deficient mice is distinct from the arrest seen in spermatocytes that are deficient in its putative catalytic partner Cdk2, which occurs much earlier in pachytene. The arrest in cyclin A1-deficient spermatocytes is also accompanied by an unusual clustering of centromeric heterochromatin. Consistent with a possible defect in the centromeric region, immunofluorescent staining of cyclin A1 protein shows localization in the region of the centromere. Phosphorylation of histone H3 at serine 10 in pericentromeric heterochromatin, which normally occurs in late diplotene, is reduced in spermatocytes from heterozygous Ccna1(+/-) testes and completely absent in spermatocytes with no cyclin A1 protein. Concomitantly, the levels of pericentromeric aurora B kinase, known to phosphorylate histone H3 during meiosis, are partially reduced in spermatocytes from testes of heterozygous mice and further reduced in homozygous null spermatocytes. These data suggest a critical and concentration-dependent function for cyclin A1 in the pericentromeric region in late diplotene of meiosis, perhaps in assembly or function of the passenger protein complex.