Asymmetric meiotic divisions in mammalian oocytes rely on the eccentric positioning of the spindle and the remodeling of the overlying cortex, resulting in the formation of small polar bodies. The mechanism of this cortical polarization, exemplified by the formation of a thick F-actin cap, is poorly understood. Cdc42 is a major player in cell polarization in many systems; however, the spatio-temporal dynamics of Cdc42 activation during oocyte meiosis, and its contribution to mammalian oocyte polarization, have remained elusive. In this study, we investigated Cdc42 activation (Cdc42–GTP), dynamics and role during mouse oocyte meiotic divisions. We show that Cdc42–GTP accumulates in restricted cortical regions overlying meiotic chromosomes or chromatids, in a Ran–GTP-dependent manner. This polarized activation of Cdc42 is required for the recruitment of N-WASP and the formation of F-actin-rich protrusions during polar body formation. Cdc42 inhibition in MII oocytes resulted in the release of N-WASP into the cytosol, a loss of the polarized F-actin cap, and a failure to protrude the second polar body. Cdc42 inhibition also resulted in central spindle defects in activated MII oocytes. In contrast, emission of the first polar body during oocyte maturation could occur in the absence of a functional Cdc42/N-WASP pathway. Therefore, Cdc42 is a new protagonist in chromatin-induced cortical polarization in mammalian oocytes, with an essential role in meiosis II completion, through the recruitment and activation of N-WASP, downstream of the chromatin-centered Ran–GTP gradient.

Tissue morphogenesis relies on the production of active cellular forces. Understanding how such forces are mechanically converted into cell shape changes is essential to our understanding of morphogenesis. Here, we use myosin II pulsatile activity during Drosophila embryogenesis to study how transient forces generate irreversible cell shape changes. Analyzing the dynamics of junction shortening and elongation resulting from myosin II pulses, we find that long pulses yield less reversible deformations, typically a signature of dissipative mechanics. This is consistent with a simple viscoelastic description, which we use to model individual shortening and elongation events. The model predicts that dissipation typically occurs on the minute timescale, a timescale commensurate with that of force generation by myosin II pulses. We test this estimate by applying time-controlled forces on junctions with optical tweezers. Finally, we show that actin turnover participates in dissipation, as reducing it pharmacologically increases the reversibility of contractile events. Our results argue that active junctional deformation is stabilized by actin-dependent dissipation. Hence, tissue morphogenesis requires coordination between force generation and dissipation.