One of the fastest moving areas of modern cell biology is the study of what molecules are involved in regulating the rate at which cells divide. The process seems to so fundamental to eukaryotic organisms that the organisms that evolved earliest and are relatively simple, such as yeast, and some of the most recently evolved organisms that seem to us pretty complex, such as--well, us--use essentially the same molecules for the same purpose. This suggests that the mechanisms for regulation of cell division evolved early and have been maintained relatively unchanged for hundreds of millions of years, though additional controls on the process have evolved in multicellular organisms. This is how similar these proteins are: One can create yeast mutants that are defective in their ability to divide, and then rescue these mutants by giving them human genes, which apparently direct synthesis of a human protein that can substitute for the yeast protein in regulating the yeast cell cycle. We’ll spend some time discussing these proteins, and how they are studied, a little later. The proteins are called cyclins--a group of proteins whose levels in the cell go up and down dramatically as the cell moves into different stages of the cells cycle and another group of proteins with which they interact--the cyclin-dependent protein kinases (cdks).
First, though, I want to remind you about how the cell cycle has been divided up into four distinct stages. This was initially done by cytologists--people who studied the behavior of cells with light microscopes. Early on, they realized that there was a very dramatic group of events that happened whenever a eukaryotic cell divided in two--which they called mitosis. During mitosis, as you know, the nuclear membrane dissolves, the chromosomes which are normally invisible, condense and become observable in light microscopes, and they line up along an array of microtubules called the mitotic spindle, and two (more or less) identical chromatids align themselves along each strand of the spindle and them move apart. Once all the pairs of chromosomes have separated and cytoplasmic organelles like mitochondria, chloroplasts, peroxisomes, Golgi apparatus, ER, etc. have also been divvied up into the two sides of the cell, a new membrane forms down the middle of the old cell, pinching it in half. (The process differs slightly for plants and animals.)
Well, this event is hard to miss even if you’re not too swift, and it marked the defining event of the cell cycle to cytologists. They noticed that often they could repeatedly watch living cells undergo mitosis with a microscope and that the time between one mitosis and the next was pretty constant for a given type of cell. This suggested that cells divided on the basis of some internal clock mechanism (although this view is no longer so widely held). Because they couldn’t see much happening in the cell between periods of mitosis, they assumed that the cell was quiescent during that interval.
This view prevailed from the late 1800s until the 1950s when people realized that the genetic information in a cell was contained in its DNA. Because the genetic information--i.e., the chromosomes--was divided up at each mitosis, it was clear that the amount of DNA in the cells was doubled some time between the end of one mitosis and the beginning of the next (Why could they conclude that?) So scientists asked when the DNA was made. Was it at the end of one mitosis, at the start of the next, or what? To answer this question they fed cells labeled thymidine at various times after mitosis, and asked when the cells incorporated the labeled thymidine into DNA as they synthesized it (which is, conveniently, all that the cell uses thymidine for). What they found was that the DNA was made somewhere in the middle of the period between one mitosis and the next, not right after or right before mitosis. So they defined this middle period as the time of DNA synthesis, which they called S phase (synthesis phase). They had already called mitosis M phase of the cell cycle. So these two phases conveniently divided up the period between divisions. Since the cell waited a while after mitosis before it began to synthesize DNA to prepare for the next mitosis, it seemed reasonable to guess that something had to happen to the cell during that period to tell it that it was okay to begin DNA synthesis--because once DNA synthesis begins the cell is pretty much committed to going on through mitosis. What told the cell that it was time to begin DNA synthesis? How about some guesses? Could it be some external signal, such as what other nearby cells are doing, or the presence of a mitogen? Could it be an internal signal like how big the cell is or how many mitochondria they are? Whatever it was, the signal ended the first, apparently quiescent period, (which was called G1 or gap 1), and caused the cell to begin S phase.
Apparently finishing up DNA synthesis is not enough to make most cells ready to divide because there is another period when nothing obvious happens, between S phase and M phase; this second quiet period is called G2 or gap 2. During this time the cell, although committed irreversibly to going through mitosis, has to do some housekeeping or make something essential for mitosis that is not available in sufficient amounts at the end of DNA synthesis. (See Fig. 13-1 in Lodish et al.)
The time that eukaryotic cells take to divide can range from several hours to years or longer. Even for cells in the same organism, the length of the cell cycle varies greatly. In animals some cells like skin cells can divide about once a day, while others like neurons, have a cell cycle that is longer than the life of the organism. That is, it’s essentially infinite.
If you measure the length of different phases of all these different cells from the same organism, what kinds of variation would you expect to find? That is, how different is the length of mitosis for a muscle cell or a liver cell or a skin cell? Not very different at all; in warm blooded animals it takes 20 or 30 minutes. How different is the length of S phase? Not very; it takes an hour or two. No surprise here; all the cells have the same amount of DNA to duplicate, so it should take the same amount of time. How about the length of G2? Again very little variation, just a couple of hours or so. Whatever the cell has to do to get ready for mitosis after DNA synthesis is complete apparently doesn’t vary a whole lot between cells. What does vary? The length of G1. In fact, some cells that seem to be idling in G1 for very long times--months or even years, are said to be stuck in a special quiescent state called G0 (gee zero); they’re thought to require a jump start to get back into the cell cycle. For instance, after immature lymphocytes form in vertebrates, which mostly happens around the time of birth, they often don’t divide for many years, unless they encounter a molecule to binds to particular receptor proteins on their cell surface. Once this binding event occurs, along with the interaction among several different kinds of lymphocytes, the lymphocyte that was in G0 suddenly gets back on the fast track and zooms along through the cell cycle at a breakneck speed for a while, sometimes dividing in less than a day for several weeks, causing an enormous increase in the number of these lymphocytes.
From these kinds of observations, where would you suspect is the crucial control point at which the cell makes a decision about whether to remain quiescent or to proceed through another round of cell division? G1--at least that’s what people thought in the 1980s when they started tackling this question in earnest. An how would you try to study this? What strategy could you use?
People have used two strategies that have turned out to be informative and that actually have converged to complement each other very nicely. One type of approach was biochemical, using egg cells or oocytes from frogs (especially Xenopus laevis). The other was a genetic approach using genetically tractable eukaryotes like single celled yeast that could be mutated to interfere with cell division and used to identify the proteins that are affected by these mutations. The first system was used because the start of cell division could be controlled very precisely; that is, egg cells aren't dividing but can be stimulated to divide on cue. Then you can see what happens in the cell as a result of fertilization by sperm, and maybe discover what events in the cells were essential for it to enter the cell cycle.. It’s easy to cause fertilization in a test tube, and thus it’s easy to move a cell at will from G0 to G1 and then to S phase. This approach has another advantage, namely that oocytes are really big cells, so they have a very large cytoplasm. Since the molecules that regulate the cell cycle are likely to be in the cytoplasm, this provides a rich source of material for biochemistry to purify the components involved in regulation. And since the oocytes can all be induced to enter the cell cycle simultaneously, it’s relatively easy to identify changes in the control molecules--e.g., changes in concentration of particular proteins, addition of phosphate groups, etc.
Before an oocyte can be fertilized it must move from G2, where it has been arrested, partway through meiosis to M phase of meiosis II--i.e., nearly to the completion of meiosis (Fig. 13-5a, Lodish et al.). Fertilization with sperm completes the process and generates a new diploid cell, the fertilized egg that can rapidly divide to generate a new frog. The maturation of oocyte to egg is stimulated by reproductive hormones in the frog and of course involves movement of the cell from G2 into meiosis. Thus treatment with the hormone must activate some factor required for the cell to enter meiosis or else it must remove a block to the entry into meiosis. So people treated oocytes with hormone in vitro and looked for things that changed, and one thing that changed was the level of a protein that they called maturation promoting factor, MPF. It was assayed as follows. Cytoplasm from cells that had been induced to enter M phase was injected back into oocytes that had been arrested in G2 phase (Fig. 13-5b). This caused the oocytes to enter M phase. Thus something was present in the cytoplasm of M phase cells that was lacking in G2 phase cells, and that something was enough to cause the shift from one phase of the cell cycle to the next. (Similar experiments were done with animal cells in culture, in which resting and dividing cells were fused together; the nuclei of the resting cell was stimulated to begin mitosis, presumably as a result of something in the cytoplasm of the dividing cell). These observations provided an assay that could be used to purify MPF--what is it in the cytoplasm of the egg or dividing cell that when added to a resting cell causes it to divide? It turned out to be a protein consisting of two different subunits--so MPF is a dimer. It was subsequently learned that proteins that are very similar to MPF are required for all cells to enter meiosis or mitosis, so maturation promoting factor is now called M-phase promoting factor, or mitosis-promoting factor, so it could keep the same initials.
If MPF is the stuff that makes cell cycles go (or at least part of the cell cycle), why do cells sometimes arrest before they enter mitosis? Guesses? Well when people started to measure the amount of MPF in a cell as a function of time in the cell cycle, they found that the levels of the substance varied a lot during the course of a cell cycle (Fig. 13-6). It was very low just after meiosis, or mitosis, then rose continuously during interphase, reaching a peak just as the cell started to enter the next round of meiosis or mitosis, then fell precipitously again.
Using the Xenopus oocyte system it was possible to work out many of the details of the involvement of MPF on entry of cells into and exit from mitosis. The basic expermental protocol was worked out, largely by Tim Mitchison and Andrew Murray, in Marc Kirschner's lab at UCSF in the 1980s, as shown in Fig. 13-7. First they showed that extracts of frog oocytes could be induced to go through cycles of "cell division" by the addition of sperm chromatin. That is, it was possible to mimic the cell cycle in vitro, which allowed them to manipulate the system to see what its necessary parts were. The reason that this works at all is that the first few divisions of a fertilized egg cell do not require RNA synthesis. That is, a mature egg contains a large number of inactive mRNA molecules and is chock full of protein and nutrients. When the egg is fertilized, the mRNAs are translated into protein, and the newly synthesized proteins are sufficient to promote several rounds of cell division. Eventually, the one cell dividing many times, reduces the average cell size, and concentration of necessary nutrients, so that the cells must begin to synthesize new RNA and acquire new nutrients, make new proteins, etc. Thus, an egg cell extract in vitro has all it needs to undergo 3 or 4 division cycles before it runs out of steam, and that provided the handle that Kirschner et al. needed to analyze the role of MPF in the process.
What they did is summarized in Fig. 13-7. First they showed that the egg extract could in fact go through mitosis when stimulated with sperm chromatin. That is, the extracts went through all the stages of mitosis, from prophase through metaphase, anaphase and telophase, then back into interphase, just as fertilized egg cells do (see figs. 19-34 and 35, and the accompanying videos). In addition they showed that the levels of MPF rose and fell during the cell cycle, increasing during interphase and then suddenly declining during mitosis. (The assay for MPF is to add it to unfertilzed eggs and show that they then complete meiosis--see Fig. 13-5b again). So Fig. 13-7a shows the control experiment that establishes the ability of the extract system to faithfully mimic the behavior of fertilized egg cells. It also shows that a particular protein, cyclin B, changes its level in cells exactly in parallel with MPF levels. In Fig. 13-7b, they treated the extracts with low levels of RNase to destroy mRNA, which prevents the cycling seen in untreated extracts; this confirms that new protein synthesis is necessary for the cell cycle. They then added back to the RNase-treated extracts the mRNA encoding cyclin B, and that alone restored the ability of the extract to "divide" (Fig. 13-7c). This indicates what? That cyclin B is the magic stuff necessary to induce mitosis in these egg extracts. Finally, they produced a non-degradable form of cyclin B added mRNA for it to the Rnase-treated extracts. As shown in Fig. 13-7d, these extracts began mitosis, and progressed through metaphase, but got stuck in late anaphase, so the chromosomes never decondensed and no nuclei formed; levels of cyclin B and MPF remained high. What does this show? That destruction of cyclin B is necessary to end mitosis. That is, an increase of cyclin B levels is required for cells to enter mitosis, but it must be destroyed before they can leave mitosis. Cyclin B is thus some sort of mitotic signal that tells a cell it's "mitosis time", and if it doesn't go away, the mitosis goes on and on.
Why do cyclin B levels and MPF levels parallel each other so closely? Because, of course, cyclin B is part of MPF. MPF is a dimer of two proteins--cyclin B is one, and the other is a protein kinase, called a cyclin-dependent kinase (cdk). That is, the kinase is around all the time, but it's only active as MPF when it's bound by the activating subunit, which is cyclin B. And cyclin B is only around in appreciable quantities at the end of G2 phase and early M phase of the cell cycle.
Why does cyclin B suddenly go away at the end of mitosis? The decrease in cyclin at the end of mitosis turns out to be caused by rapid destruction of the protein. It is known that a marker protein, called ubiquitin, is attached to cyclin at the end of mitosis by an enzyme called the "anaphase-promoting complex (APC)". You may recall that attachment of ubiquitin to a protein, targets it to a protein degrading structure called a proteosome that degrades the cyclin (Fig. 13-8). Ironically what turns on APC, which is normally inactive, is none other than MPF. The rise of cyclin B creates MPF which activates APC, which in turn causes the destruction of cyclin B (Fig. 13-9), which is required to move the cell out of anaphase and through the end of mitosis.
No comments:
Post a Comment