Self-organization of the Fission Yeast Actin Cytoskeleton

The actin cytoskeleton is a large and dynamic network that mediates many important cellular processes, including polarity, endocytosis, motility, and cytokinesis. In an extremely crowded cytoplasm, cells must simultaneously assemble multiple distinct filamentous actin (F-actin) networks with the proper architecture and dynamics at the correct time and place. How a cell initiates and maintains these distinct F-actin structures to drive essential cellular functions is a fundamental question in cell biology. In cells, actin assembly factors initiate F-actin assembly from monomeric globular actin subunits (G-actin). The organization of these filaments into higher-order networks with specific architectures and dynamics is regulated by the coordinated efforts of actin-binding proteins (ABPs) that nucleate, crosslink, sever, stabilize, and cap actin filaments. Therefore, the accurate localization of particular ABPs to specific F-actin networks is paramount, as the identity of each F-actin network is defined by the subgroup of ABPs with which it associates. However, the mechanistic principles guiding the sorting of ABPs to distinct F-actin networks remain largely unclear. Previous work from our lab revealed that competitive and cooperative interactions between ABPs drive their sorting to specific F-actin networks in fission yeast. In particular, fimbrin, a crosslinking protein found in actin patches, competes with tropomyosin, a side-binding protein that localizes to the contractile ring, for binding F-actin both in vitro and fission yeast cells (Skau and Kovar, 2010; Christensen et al., 2017). Here, we investigated other potential competitive interactions between ABPs by conducting a survey in fission yeast. We discovered a competitive interaction between fimbrin and α-actinin, a bundling protein found in the contractile ring. Fimbrin and α-actinin compete for binding F-actin in vitro, in actin patches, and in the contractile ring in fission yeast cells. Additionally, tropomyosin synergizes with α-actinin to inhibit fimbrin's association with F-actin. While competitive and cooperative interactions between ABPs certainly tune their sorting to distinct F-actin networks, how ABP sorting is initially established is currently unknown. Our longstanding hypothesis is that actin assembly factors build actin filaments with specific characteristics that are preferred by certain ABPs, thereby recruiting them to particular F-actin networks and triggering the cascade of ABP sorting. We used fluorescence microscopy in fission yeast and in vitro reconstitution to investigate how actin assembly factors Arp2/3 complex and formin Cdc12 influence the localization of ABPs fimbrin and tropomyosin. We show that in fission yeast, fimbrin is preferentially recruited to actin patches (assembled by Arp2/3 complex), while tropomyosin prefers to localize to the contractile ring (assembled by formin Cdc12). Interestingly, we also see that actin filaments assembled by different actin assembly factors are not sufficient to induce ABP sorting in vitro. Instead, F-actin with distinct architectures and competition between ABPs are both necessary to recapitulate fimbrin's preferred association with Arp2/3 complex-mediated actin filaments and tropomyosin's localization to formin Cdc12-generated F-actin. In addition to driving ABP sorting, formins generate straight actin filaments that regulate fundamental processes such as cell division and polarity. While formins assemble F-actin via a well-conserved molecular mechanism, diverse cell types require many distinct formin isoforms to carry out specific cellular functions. Therefore, we sought to understand whether a specific formin is tailored for a particular cellular role by its actin assembly properties. We engineered formin chimera fission yeast strains in which the actin assembly domains of the contractile ring formin Cdc12 were replaced by the actin assembly domains of functionally diverse formins. These formin chimeras produce viable cells, but exhibit a range of cytokinesis defects and delays in contractile ring assembly. We determined that formin chimeras with nucleation efficiencies similar to Cdc12 are generally less defective in cytokinesis than those that are poor F-actin nucleators. Additionally, we used a computational model to isolate specific formin actin assembly properties and found that the probability and timing of contractile ring formation are both critically impacted by variations in formin nucleation efficiency