You'll not be surprised to learn that cell division works much like many of the other complex biochemical pathways that I've described previously. Namely it is mediated by a Rube Goldberg procession of events, one reaction cascading into another, to produce a final endpoint or endpoints. Moreover, like many of the processes I've previously described, the individual steps are often controlled by protein kinases, enzymes that add a phosphate group onto a protein that cause it to transition from an inactive to an active state (or less frequently, vice versa). However, here is one major difference between a Rube Goldberg machine and all the pathways previously noted is that most biological processes make use of mechanisms that check to see if a step has been completed before going on to the next. The cell cycle offers no exception: it proceeds by a complex series of steps; it utilizes protein kinases; and there are checkpoints along the way that monitor progression through the cycle. Cancer cells are particularly adept at subverting these last surveillance mechanisms. In addition, all the complex pathways that I've discussed can be slowed or stopped by molecules that inactivate one or more steps. Tumor suppressors are a case in point.
I'll begin with the role of kinases because they are the main players in the cell cycle. There's lots of them and they are involved with virtually all of the steps that occur during cell division. For example, during mitosis the nuclear membrane breaks down just before mitosis, a process that is promoted by particular kinases that have the job of transferring phosphate groups to the nuclear membrane. Kinases also phosphorylate transcription factors that help to turn on genes appropriate to particular steps during the cell cycle. There are as many as a hundred other examples. But what are the regulatory molecules that turn on these kinases? As you might have guessed, the answer is other kinases. Foremost among them are a group of enzymes called cyclin dependent kinases. As their name implies, these proteins are inactive by themselves. Only when bound by a second protein, one devoid of kinase activity, can they transfer phosphate groups on to other molecules. This second protein, first discovered By Tim Hunt and named by him, is called cyclin.
In actuality there are four cyclins, called D, E, A, and B, that change in concentration (or availability) at various stages of the cell cycle. A figure illustrating how they fluctuate is shown below.
It is cyclin D that initiates the cell cycle after a cell has completed division. Its concentration is dependent on the presence of growth factors outside of cells and the subsequent cascades of reactions that follow their binding to their appropriate receptors. Ultimately, cyclin D synthesis is driven by transcription factors that are activated by these cascades. As its concentration increases it binds to two cyclin dependent kinases, cdk4 and cdk6. The pairing of a cyclin with these kinases activates them. They add phosphate groups to a variety of molecules that take the cell from the start of G1 to a point near the end of G1 called "R", the restriction point. If a cell makes it past R, it will continue its way through the cycle, through the remainder of G1 and then through S, G2, and M, without further input from growth factors. In fact, it will be largely unresponsive to growth inhibitors that are only effective before R. If a cell fails to pass the restriction point, it will depart the cell cycle, and will enter a stage called G0 where it may differentiate.
Because the restriction point is so critical, it is thought that most cancers have figured out ways to bypass it. Before describing how the restriction point works, I should make clear that in addition to the various cyclins shown above, there are a variety of cyclin dependent kinases. These associate with the cyclins,become active, and move the cell cycle along (see the figure at the right). The black arcs in the illustration represent the times in which these cyclin/cdk complexes operate during the cell cycle.
The Restriction Point
The protein in charge of the restriction point is the product of the retinoblastoma gene, RB. After a cell leaves mitosis, RB lacks phosphate groups. As it progresses through G1, more and more phosphate groups are added to it, until, at the restriction point, most of the possible sites of phosphorylation are filled. It turns out that RB in its unphosphorylated form acts a cell cycle inhibitor, preventing the cell from leaving G1. When it gets fully phosphorylated, it looses its inhibitory behavior, thereby allowing the cell to pass R and progress through the remainder of the cell cycle. As Weinberg so beautifully puts it, RB serves as a guardian of the R gate, keeping it closed through G1 until it is inactivated by the addition of phosphate groups.
But this explanation begs two further questions. First, what is it that unphosphorylated RB does to prevent cells from progressing through G1? Second, what proteins add the phosphate groups to RB? The answer to the first question is that RB is a "pocket protein". it has the ability to bind other proteins, among them a transcription factor that is important for the G1 to M transition. Because this transcription factor is sequestered, it cannot act. Upon phosphorylation, RB releases its pocketed transcription factor which, in turn turns on the synthesis of cyclin E. As to which protein places phosphate groups on RB, initially, it is the cyclin D/cdk4 and 6 complex that begins the job. As the cell progresses to R, the activity of the newly appearing cyclin E complexed with cdk2 completes the process. Since cyclin D concentration is directly linked to growth factors, the complex series of events from growth factor to initiation of the cell cycle should be apparent. For further clarification, take a look at the figure above.
The RB protein is depicted as a "pocket", holding onto transcription factors that are necessary for the cell to pass through the restriction point. Growth factors, like epidermal growth factor mentioned previously, start several cascades of reactions that eventually result in the appearance of cyclin D. Complexed with cdk4 and 6, it begins to phosphorylate RB, resulting in the release of other transcription factors that increase transcription of the cyclin E gene and production of the cyclin E protein. In turn, cyclin E complexes with cdk2, producing an enzyme that further phosphorylates RB, inactivating it, thereby allowing the cell to pass through R and proceed through the other steps of the cell cycle.
All this is complicated, but I've actually left out several steps and only described the essence of the process. For example, I haven't mentioned the role of protein degradation in controlling cyclin D. And I've left out several key components. However, what should be clear is that the retinoblastoma protein (RB) plays a critical role in blocking the cell cycle until it is signaled to advance by the end products of growth factor stimulation. If something were to happen to prevent retinoblastoma from doing its duty, by a mutation in the retinoblastoma gene for example, the cell cycle would lose one of its main control elements. Cells then could proliferate inappropriately and cancer might result. It is for this reason that Weinberg writes that is may well be that "virtually all human tumors" may bear defects in RB signalling.
By way of emphasis, I want to repeat one point mentioned previously. Recall that an oncogene stimulates carcinogenesis by acquiring mutations that increase its growth promoting activities. RB, as a representative of tumor suppressors, increases growth when it loses activity via mutation. In the next post, I'll write about another important tumor suppressor, one that acts entirely differently than RB, but whose absence seems to be correlated with nearly all cancers.