On G0 and cell cycle controls.

The current model of the animal cell division cycle proposes that the G, phase of the division cycle is the location of most, if not all, of the events regulating the growth and division of animal cells., Fantes2 has reviewed the ideas involved in this current model, and in particular the relationship between the Go phase and the G, period of the division cycle. His point of departure was the work of Zetterberg and Larsson: who divided the G, period into two parts, a G,pm (post-mitotic) phase and a G,ps (pre-S) phase. The G,pm period, which in the 3T3 cells studied by Zetterberg and Larsson lasted for 3.5 hours after mitosis, is the only period of the division cycle from which cells can leave the cycle to enter the Go period. As Fantes* points out in his review of this work, the existence of a control point in the GI period fits in with other ideas, such as the unique restriction point of which has become ‘one of the basic tenets of workers in the field’.’ Besides the restriction point, Fantes observes that the idea of a control point in the GI period also receives support from the transition-probability model of Smith and Martin.5 I have argued, and will argue here, that there are no GI control points and there is no G,-specific regulatory function. In bacteria and animal cells the relevant regulatory events occur continuously during the period between the starts of S phases; there are no G,specific controls or control points.6” Here I will apply this alternative view of the division cycle to the results of Zetterberg and La r~son .~ I will show not only that their experiments lead to predictions that have not been supported experimentally, but I will also propose a new principle, the Law of Cell Age Order Invariance, against which the model and results of Zetterberg and Larsson can be tested. What did Zetterberg and Larsson do? The experiment was quite simple. They used a time-lapse video recorder to study cells growing in monolayers. After determining the cycle ages of the growing cells by noting the time since the last mitosis, all of the cells were subjected to varying periods in serumfree medium. The division of the cells was then followed during the treatment and after the serum was replaced. To explain their results with numbers, consider that the interdivision time was 16 h, that the G, period was 7 h, and that the S, G, and M phases accounted for the remaining 9 h. Let us look at the experiment where the cells were starved for 1 h. Zetterberg and Larsson3 observed that only cells that were within 3-5 h of the last mitosis (that is, the youngest cells) were affected in the next division; not only was the division delayed for the time of starvation, but the division was delayed or ‘set back’ an additional 8 h. Those cells that were older than 3-5 h and that were now within 12.5 h of the next mitosis went through the next mitosis on schedule, whether or not the cells were starved for between 1 and 8 h. The cells that were more than 12-5 h from mitosis (i.e. not older than 3.5 h) were ‘set back’ or delayed in their next scheduled mitosis for at least 8 h. The model that Zetterberg and Larsson3 propose to explain their results is that during the first 3.5 h the cells are in a G,pm (post-mitotic) phase; when these cells are starved of serum they can enter the Go phase of the cell cycle. Return to the G, period from the Go period takes 8 h and this accounts for the 8 h ‘setback’ in cell division for the younger cells starved for serum. At 3.5 h after mitosis the cells leave the G, pm phase and enter the G, ps (preS) phase. Cells in this phase go on to divided even if incubated in serum-free medium. Only cells in a particular part of the GI period can enter the Go phase,