However, in LatB-treated cells, endosomes rapidly fused with each other, resulting in enlarged endosomes with few short tubules. After washout of LatB, Tfncontaining tubular structures immediately QNZ 46 segregated from endosomes and clusters of vacuolar domains dissociated from each other. At 15 min after washout, these clusters were dissociated, and at 60 min after washout, EGF-containing endosomes localized around the perinuclear R 568 hydrochloride region and finally disappeared. These data clearly indicate that disruption of the actin filaments induced aggregation of EEs, resulting in the formation of enlarged EEs. On the other hand, actin polymerization made the vacuolar domains pull apart and severed the tubules containing recycling molecules. We demonstrated that LatB treatment induced abnormal enlargement of EEs, judging from colocalization with EEA1. However, there was a possibility that LatB treatment blocked the transition from EEs to LEs and/or REs because EEs have a mosaic structure. EEs move from the cell periphery to perinuclear region in a microtubule-dependent manner and mature to LEs; this process is accompanied by both recruitment of an LE marker LAMP1 and intraluminar acidification. Therefore, we investigated the effect of actin polymerization on endosomal maturation. In control cells, the EGF signals were colocalized with Lamp1 at 30, 60, and 120 min after internalization. Interestingly, the same results were obtained in LatB-treated cells, indicating that EEs containing EGF were partially converted to LEs. The same results were obtained using lysotracker, an acidic sensor. On the other hand, Rab11, a marker of REs, was not colocalized with EGF, suggesting that transferrin did not reach recycling endosomes. When we analyzed whether early and late endosomes fuse together in a heterotypic manner by localization of these specific markers, they were not co-localized but adjacently localized. These results indicate that the transition from EE to LE did not depend on actin dynamics, although the degradative/ recycling components remain the same organelle. Actin filaments have been reported to be responsible for shortrange movement of peripheral endosomes. In contrast, microtubules are responsible for long-range movements between the perinuclear and peripheral region. Therefore, we compared endosomal motility in the presence of LatB and nocodazole. In control cells, long-range directional movements toward the cell center were observed. In contrast, we hardly detected any endosomal movements in nocodazole treated cells, suggesting that endosomal movements largely depend on microtubules. However, in LatB-treated cells, EGF-containing endosomes moved rapidly in random directions and fused with each other.