Mitosis and cytokinesis are highly coordinated in eukaryotic cells. But procyclic-form Trypanosoma brucei under G1 or mitotic arrest is still capable of dividing, resulting in anucleate daughter cells (zoids). Okadaic acid (OKA), an inhibitor of protein phosphatases PP1 and PP2A, is known to inhibit kinetoplast replication and cell division yielding multinucleate cells with single kinetoplasts. However, when OKA was applied to cells arrested in G1 or G2/M phase via RNAi knockdown of specific cdc2-related kinases (CRKs), DNA synthesis and nuclear division were resumed without kinetoplast replication or cell division, resulting in multinucleate cells as in the wild type. Cells arrested in G2/M via depleting the mitotic cyclin CycB2 or an aurora B kinase homologue TbAUK1 were, however, not released by OKA treatment. The phenomenon is thus similar to the OKA activation of Cdc2 in Xenopus oocyte by inhibiting PP2A [Maton, et al., Differential regulation of Cdc2 and Aurora-A in Xenopus oocytes: a crucial role of phosphatase 2A. J. Cell Sci. 118 (2005) 2485-2494]. A simultaneous knockdown of the seven PP1s or the PP2A catalytic subunit in T. brucei by RNA interference did not, however, result in multinucleate cells. This could be explained by assuming a negative regulation, either directly or indirectly, of CRK by an OKA-sensitive phosphatase, which could be a PP2A as in the Xenopus oocyte and a positive regulation of kinetoplast replication by an OKA-susceptible protein(s). Test of a PP2A-specific inhibitor, fostriecin, on cells arrested in G2/M via CRK depletion or a knockdown of the PP2A catalytic subunit from the CRK-depleted cells both showed a partial lift of the G2/M block without forming multinucleate cells. These observations support the abovementioned assumption and suggest the presence of a novel OKA-sensitive protein(s) regulating kinetoplast replication that still remains to be identified.