As our experimental techniques get more powerful, we’re constantly having to update our models of what living cells are like. This new paper is a case in point. We know about the cell cycle, the phases of cellular life and the way that cell division is its own carefully managed process that occurs in stages. We can see the reorganization of the cellular contents (and of course the genetic material), even though we certainly don’t understand everything that’s going on. And there are a number of biochemical signals that have been identified with specific timing – for example, the powerful signaling protein c-Myc is normally present in cells at very low levels under very tight regulatory control, but its levels go through a brief, strong spike at the very beginning of the cell cycle, a sort of kick-start. (Similarly, if a cell runs into some sort of defects that cause c-Myc levels to be raised all the time, that turns out to be a very common feature of many cancer types).
But there are still many details that we haven’t been able to work out. We knew, for example, that protein synthesis ramps up during cell division, as well it might when you’re going to end up with two separate cells at the end of the process. I think, though, that many people either picture this as happening at the beginning of the process or (more likely) that it occurs more or less smoothly throughout. Indeed, there is experimental evidence for that latter view. But this new paper shows that there are actually two distinct waves of protein synthesis in yeast cells as they bud, one at the G1 stage and the second at the S/G2/M stage. That was determined by expression of a fluorescent marker protein in single cells, and separately by stopping protein synthesis at various points (with the well-known tool cycloheximide) and measuring NAD(P)H levels in those single cells as well as a measure of metabolic activity. Meanwhile, there is only one wave of biosynthesis for lipids and polysaccharides, also in S/G2/M. That was determined similarly in single cells, using the lipid synthesis inhibitor cerulenin followed by NAD(P)H measurement for the former and an inducible degrader system to take out the key Ugp1 enzyme that produces UDP-glucose, precursor to a number of important polysaccharides. These details seem to have been obscured by noisier data in previous work, although no doubt some of the authors of those papers will be disputing that interpretation. The paper under discussion has a good summary of the experimental results that have been used to suggest a constant or exponential rate of protein biosynthesis and some others that have pointed towards this oscillation behavior.
The authors found that these metabolic oscillations can’t be explained by respiration (they didn’t go away when varying the oxygen levels), and they can’t be explained by changes in carbohydrate storage (deletion of glucose biosynthetic genes didn’t remove the oscillations, either). Changing up the nutrient mixes for the cells also showed that the biosynthesis waves occur under all sorts of feeding (minimal medium, then supplemented individually with fatty acids, with amino acids, with nucleobases, etc.) The total time needed to make it through the cell cycle varied a great deal with those changes, but the biosynthesis oscillations still occurred. They took place, though, with various phase shifts, suggesting that this “temporal segregation” in metabolism is the fundamental process under all these conditions. That is, the need for this phased time course is what drives the demand for metabolic precursor molecules and for cellular energy in general, rather than the other way around.
So this means that we’re going to have to think about cellular growth and division in an even more fine-grained manner. The biosynthesis pathways that are needed are not operating at the same time, but are instead involved in a dance routine that has specific moves at specific times. It looks like the ATP of the cell gets earmarked for extra protein synthesis separately from extra lipid and polysaccharide synthesis, and there must be a good reason for that (and for the dual waves of protein synthesis) that we don’t quite understand yet.
We’ll also want to dig into the signaling mechanisms for these waves: how does this timing work on a molecular level? Is there some signaling peptide produced in that first protein biosynthesis surge that sets off the pathways that leads to lipid and polysaccharide synthesis (and which isn’t produced in the second wave)? Or is there some separate timing regulator system that’s operating off to the side and spitting out crucial signaling molecules? Maybe it’s just run by straight negative and positive feedback loops based on the concentrations of the new biosynthesized molecules – we just don’t know. But without understanding these things, we won’t have a good understanding of how normal cells grow and divide, and we won’t be able to understand the crucial differences in the uncontrolled growth and division phenotype that we know as cancer.
