ABSTRACT The dependency relationships between various events in mitosis in the fission yeast Schizo-eaccharomycee pombe have been investigated using a combination of approaches. The events concerned are those controlled by specific genes, a step or steps in mitosis sensitive to benomyl, and protein synthesis. The anti-microtubule agent benomyl was shown to inhibit mitosis specifically, while not significantly affecting the accumulation of RNA or protein. DNA synthesis was not directly affected, though accumulation ceased after mitosis was inhibited. The inhibition of mitosis was readily reversible and was followed by a synchronous cell division. Reciprocal shift experiments were carried out using benomyl in conjunction with temperature-sensitive mutants defective in mitosis. Consideration of these experiments and the transition points of the mutants allowed the mutant and the steps controlled by the respective genes to be ordered functionally. Group (i) consisted of cdc6 23, which showed an early transition point: the benomyl-sensitive step was dependent on completion of the cdc6 step. Group (ii) consisted of cdc2·33 and cdc2·33 K3, which had transition points close to the time of mitosis itself. The benomyl-sensitive step was dependent on the cdc2 and cdc27 steps, but the relationship between these gene-controlled steps was uncertain. Group (ii) processes may be dependent on those in group (i), or they may act independently of one another on separate pathways. Group (iii) mutants consist of 1·7 cdc13·117 and cdc25·22. These have transition points close to mitosis, and act interdependently with benomyl. It is likely, though not directly proven, that all group (iii) processes are dependent on those in group (ii). The strikingly different terminal phenotype of cdc13·17 compared with other mutants of mitosis allowed further analysis of the group (iii) mutants according to the terminal phenotypes of double mutants of cd13·117 with other mutants. Double mutants with cdc1·7, cdc2·33, 25·22 and cdc27·1<3 all showed a terminal phenotype unlike cdc13·117, and similar to the other single mutant parent. It is therefore likely that the cdc13 step is dependent on cdci, 2, 25 and 27; that is, later in mitosis, consistent with its unusual terminal phenotype. Two mutants defective in stages of the cell cycle other than mitosis were subjected to reciprocal shifts with benomyl. The process controlled by cdc15·140, an event in septum formation, was dependent on the benomyl-sensitive step. The other mutant, cdc10·129, with a defect very early in the cycle, has been proposed to arrest at a stage analogous to ‘start’ in budding yeast. The completion of the cdcio function was shown to be dependent on the occurrence of mitosis in the previous cycle. Cells arrested early in mitosis by cdc 2·33 were able to undergo mitosis when returned to the permissive temperature in the presence of cycloheximide. This supports previous observations that protein synthesis is not required for mitosis once the process has started. Extrapolation of this result to the normal cycle suggests that sequential switching on of cdc genes is not the basis of the dependency relations determined in this study, and that periodic synthesis of proteins coded by cdc genes may not be an important part of the mechanism that controls the sequence of events in mitosis.
Read full abstract