The actin module of endocytic internalization in Aspergillus nidulans: A critical role of the WISH/DIP/SPIN90 family protein Dip1.

Using fluorescent protein-tagged F-actin reporters we studied the actin cytoskeleton in Aspergillus nidulans. F-actin probes labeled endocytic patches, contractile actin rings and the Spitzenkörper (SPK), but not exocytic cables generated by the SPK-associated formin, illuminated only by tropomyosin. The SPK actin mesh contains tropomyosin and capping protein, but not fimbrin or Arp2/3, showing that it does not involve branched actin. Arp2/3 and fimbrin are recruited to endocytic patches at the end of their lifecycle, staying in them for 12-14 sec, coinciding with the burst of branched actin polymerization that powers vesicle internalization, whereas verprolin stays only during the first half of this actin phase. Hyphal growth requires endocytic recycling, which we exploited to assess the efficiency of endocytosis following genetic interventions. Ablation of SlaBSla2, Arp2/3, cofilin and fimbrin is lethal, whereas that of Srv2, verprolin and capping protein are debilitating, with the lifetime of actin in mutant patches roughly correlating with the extent of growth and endocytic defects. An outstanding problem is the origin of seed filaments required to prime Arp2/3 during endocytosis. Actin patches associate with cortical cables that give rise to long distance-moving "actin worms" that are different from tropomyosin-containing cables emanating from the SPK. Cables and worms are dependent on formin, yet inactivation of formin does not affect the F-actin patch lifecycle, arguing against formin playing an endocytic role. Ablation of the WISH/DIP/SPIN90 protein Dip1 priming Arp2/3 for the synthesis of linear actin delocalizes the endocytic machinery and severely impairs, but does not preclude, endocytosis. This establishes the existence of Dip1-independent mechanism(s) that synthesize seed filaments. Our data negate the possibility that this alternative mechanism results from a priming role of formin that is unmasked in dip1∆ cells, but do not exclude that cofilin-mediated filament severing could produce seed microfilaments for Arp2/3, as suggested for Schizosaccharomyces pombe.