Pom1 is not the size ruler.

Cell size is under genetic control, and size at division may be mutationally changed so that cells divide at a larger or smaller size without any alteration in the generation time. Larger and smaller cells thus double their size during the same time period, illustrating that larger cells grow faster than smaller cells. However, the increase of cellular growth rate with cell size in certain genetic backgrounds could amplify size heterogeneity of the population in the absence of additional regulation. Therefore, growing cells possess a homeostatic size control mechanism, which shortens the cycle time of larger cells, while it extends the cycle time for smaller ones (Fig. 1). This homeostasis works through the size control mechanisms that regulate the length of G1 or G2 phase, thereby controlling the timing of entry into DNA replication or mitosis, respectively (G1/S and G2/M size controls). The molecular mechanisms underlying these size controls are still obscure.

Figure 1. Mechanism of size homeostasis. The average mass doubling time (red curve) is independent of birth size, if larger cells can double their size during the same time as small ones. The size control mechanism modulates the time between 2 successive divisions (cycle time, green curve). Cells, born at the size where the 2 curves intersect (dashed line), double their size exactly until cell division. Cells born smaller than this size extend by more than their birth size, because the cycle time is longer than the time required to double their size. Therefore their size at birth will be larger in the subsequent cycle, i.e., these cells are returning toward the birth size that allows exact doubling during the cycle. Exactly the opposite scenario happens in large cells, which move to smaller sizes during successive cycles.

A few years ago, 2 papers published in Nature demonstrated an intracellular concentration gradient of Pom1 kinase in rod-shaped fission yeast cells., The level of the mitotic inhibitory Pom1 kinase was found to be highest at cell tips and minimal around the nucleus in the middle of the cell. As cells grow, the Pom1 concentration at the cell center decreases, thus relieving its mitotic inhibitory effect. These observations lead to the proposal that the Pom1 gradient generated by a reaction–diffusion mechanism could serve as the basis of mitotic size control in fission yeast.- A straightforward prediction of this hypothesis, that balanced growth and division should be compromised in Pom1-deleted cells, has been tested now by Wood and Nurse and published in the October 1, 2013 issue of Cell Cycle. By using live-cell imaging, the cycle time vs. birth size relationship (for example the green curve on Fig. 1) of Pom1-deleted cells was determined. This analysis showed no significant difference compared with wild-type cells, suggesting that Pom1 and its intracellular gradient cannot play a central role in cell size control. The existence of size control in Pom1-deleted cells has been also confirmed by analysis of cells returning to the normal cell size after a reversible cell cycle block. In this situation, cells are born at a larger size than normal, and their subsequent cycle time is shorter than the time required to double their size (see Fig. 1). As a consequence, cells become progressively smaller in each cycle, until they recover their normal size. The rate of reversion to normal cell size was indistinguishable in pom1Δ and wild-type cells, confirming the existence of size control in the mutant cells.

Back-up mechanisms are common in biology, and size control is no exception. Fission yeast cells also have a G1/S size control, which provides size homeostasis in cells with short G2 phase, e.g., wee1 mutants. However, the kinetics of cell division after nutritional-shift experiments with Pom1-deleted cells is consistent with these cells still having a G2/M size control rather than depending on the G1/S size control.

Pom1 regulates Cdk1 inhibitory phosphorylation,, which is normally essential for cell viability of fission yeast. However, cells with non-phosphorylable Cdk1 are viable if the oscillation of Cdk1 activity is driven only by a single Cdk1- mitotic B-type cyclin (Cdc13) fusion protein. Despite having a more variable size at cell division, these cells still have a homeostatic size control mechanism., These observations suggest that inhibitory Cdk1 phosphorylation regulated by Pom1 is dispensable for fission yeast size control.

It is very likely that cells with non-phosphorylable Cdk1 regulate their cell size by G1/S control, because they have an extended G1 phase. The length of G1 phase in fission yeast is controlled by the stoichiometric Cdk1 inhibitor, Rum1. Rum1-deleted cells, after inactivation of the major Cdk1-inhibiting kinase (Wee1), become progressively smaller at each cell division. The lack of size homeostasis in the absence of any G1- and G2-specific Cdk1 inhibitors suggest that balanced growth and division is a systems-level property of the Cdk1 activity control system.