Aurora kinase A (AURKA), which is a centrosome-localized serine/threonine kinase crucial for cell cycle control, is critically involved in centrosome maturation and spindle assembly in somatic cells. Active T288 phosphorylated AURKA localizes to the centrosome in the late G(2) and also spreads to the minus ends of mitotic spindle microtubules. AURKA activates centrosomal CDC25B and recruits cyclin B1 to centrosomes. We report here functions for AURKA in meiotic maturation of mouse oocytes, which is a model system to study the G(2) to M transition. Whereas AURKA is present throughout the entire GV-stage oocyte with a clear accumulation on microtubule organizing centers (MTOC), active AURKA becomes entirely localized to MTOCs shortly before germinal vesicle breakdown. In contrast to somatic cells in which active AURKA is present at the centrosomes and minus ends of microtubules, active AURKA is mainly located on MTOCs at metaphase I (MI) in oocytes. Inhibitor studies using Roscovitine (CDK1 inhibitor), LY-294002 (PI3K inhibitor) and SH-6 (PKB inhibitor) reveal that activation of AURKA localized on MTOCs is independent on PI3K-PKB and CDK1 signaling pathways and MOTC amplification is observed in roscovitine- and SH-6-treated oocytes that fail to undergo nuclear envelope breakdown. Moreover, microinjection of Aurka mRNA into GV-stage oocytes cultured in 3-isobutyl-1-methyl xanthine (IBMX)-containing medium to prevent maturation also results in MOTC amplification in the absence of CDK1 activation. Overexpression of AURKA also leads to formation of an abnormal MI spindle, whereas RNAi-mediated reduction of AURKA interferes with resumption of meiosis and spindle assembly. Results of these experiments indicate that AURKA is a critical MTOC-associated component involved in resumption of meiosis, MTOC multiplication, proper spindle formation and the metaphase I-metaphase II transition.

CDK1 is a pivotal regulator of resumption of meiosis and meiotic maturation of oocytes. CDC25A/B/C are dual-specificity phosphatases and activate cyclin-dependent kinases (CDKs). Although CDC25C is not essential for either mitotic or meiotic cell cycle regulation, CDC25B is essential for CDK1 activation during resumption of meiosis. Cdc25a -/- mice are embryonic lethal and therefore a role for CDC25A in meiosis is unknown. We report that activation of CDK1 results in a maturation-associated decrease in the amount of CDC25A protein, but not Cdc25a mRNA, such that little CDC25A is present by metaphase I. In addition, expression of exogenous CDC25A overcomes cAMP-mediated maintenance of meiotic arrest. Microinjection of Gfp-Cdc25a and Gpf-Cdc25b mRNAs constructs reveals that CDC25A is exclusively localized to the nucleus prior to nuclear envelope breakdown (NEBD). In contrast, CDC25B localizes to cytoplasm in GV-intact oocytes and translocates to the nucleus shortly before NEBD. Over-expressing GFP-CDC25A, which compensates for the normal maturation-associated decrease in CDC25A, blocks meiotic maturation at MI. This MI block is characterized by defects in chromosome congression and spindle formation and a transient reduction in both CDK1 and MAPK activities. Lastly, RNAi-mediated reduction of CDC25A results in fewer oocytes resuming meiosis and reaching MII. These data demonstrate that CDC25A behaves differently during female meiosis than during mitosis, and moreover, that CDC25A has a function in resumption of meiosis, MI spindle formation and the MI-MII transition. Thus, both CDC25A and CDC25B are critical for meiotic maturation of oocytes.

Aurora kinase A (AURKA) is an important mitotic kinase involved in the G2/M transition, centrosome maturation and separation, and spindle formation in somatic cells. We used transgenic models that specifically overexpress in mouse oocytes either wild-type (WT-AURKA) or a catalytically inactive (kinase-dead) (KD-AURKA) AURKA to gain new insights regarding the role of AURKA during oocyte maturation. AURKA activation occurs shortly after hCG administration that initiates maturation in vivo. Although AURKA activity is increased in WT-AURKA oocytes, resumption of meiosis is not observed in the absence of hCG administration. Control oocytes contain one to three microtubule organizing centers (MTOCs; centrosome equivalent) at prophase I. At the time of germinal vesicle breakdown (GVBD), the first visible marker of resumption of meiosis, the MTOC number increases. In WT-AURKA oocytes, the increase in MTOC number occurs prematurely but transiently without GVBD, whereas the increase in MTOC number does not occur in control and KD-AURKA oocytes. AURKA activation is biphasic with the initial activation not requiring CDC25B-CDK1 activity, whereas full activation, which is essential for the increase in MTOCs number, depends on CDK1 activity. AURKA activity also influences spindle length and regulates, independent of its protein kinase activity, the amount of MTOC associated with gamma-tubulin. Both WT-AURKA and KD-AURKA transgenic mice have normal fertility during first 6 mo of life. These results suggest that although AURKA is not a trigger kinase for G2/M transition in mouse oocytes, it regulates MTOC number and spindle length, and, independent of its protein kinase activity, gamma-tubulin recruitment to MTOCs.

The multiprotein complexes known as condensins (I and II) are major players in chromosome dynamics in mitotic and meiotic cells. Here, we report for the first time the detection of different condensin subunits from both complexes in mammalian oocytes. Using immunoblotting analysis we examined expression levels of condensin subunits during meiotic maturation of porcine oocytes. The expression of the core subunit structural maintenance of chromosomes 2 (SMC2), identical in both condensin complexes, did not change significantly during maturation. Similarly, there was no significant change in the expression of the chromosome associated protein (CAP)-H and CAP-H2 subunits, components of condensin I and II, respectively. Conversely, the expression profiles of CAP-G, CAP-D2 (condensin I) and CAP-D3 (condensin II) were more interesting. At least two isoforms of the CAP-D2 subunit were detected, along with three isoforms of the CAP-D3 and CAP-G subunits. We suggest that this diverse migration of subunit isoforms is due to post-translational modification. Earlier, it was reported that non-SMC proteins are phosphorylated by cyclin-dependent kinase 1. In the present study, we analysed the phosphorylation status of the three subunits in oocyte extracts using alkaline phosphatase treatment and we found that at least the fastest migrating form of CAP-D3 was likely to be phosphorylated in maturing porcine oocytes. In addition, the localisation of CAP-H and CAP-H2 subunits was examined using immunofluorescence staining with specific antibodies, as well as following microinjection of their enhanced green fluorescent protein-tagged mRNA into germinal vesicle-stage oocytes. CAP-H was found in the cytoplasm, whereas CAP-H2 was localised within the nucleus.