Bruce McEwen Laboratory
Chromosome Motion
The elegant movements of chromosomes during mitosis have long fascinated biologists, even before the genetic significance of this phenomena for genetic segregation was appreciated. Mitosis is a process whereby a complete copy of the genetic information is distributed to each new cell during cell division. The genetic material that is passed from one cell to two daughter cells is contained in the segregating chromosomes.
Figure 1: An electron micrograph of a PtK1 cell in prometaphase. The view is of a serial thin section prepared by conventional specimen preparation techniques.
Mitosis begins with a bipolar spindle forming and DNA condensing into distinct chromosomes. Chromosomes contain duplicate copies of the genes and they bind to the spindle with one copy facing each spindle pole. After attachment, chromosomes move to the center of the spindle. Here they split apart with one of the identical copies migrating to each spindle pole and hence to each of the news cells. In this way, a complete complement of DNA is transmitted to all cells of a multicellular organism, and by a similar process, to succeeding generations of the organism. The beautiful and intricate chromosome movements involved in this process are critically important, with even small errors leading to birth defects or cancer. New developments in light and electron microscopy, and computer image processing, serve to probe the interactions between chromosomes and spindle fibers that produce chromosome motion.
What is the Role of the Kinetochore Fiber During Mitosis?
In mammalian cells, each kinetochore typically binds between 10 and 50 microtubules. The exact number varies between species, but even within one cell, the number can vary over a two-fold range. One cause of this variation is the relatively slow rate that kinetochores bind microtubules.
Figure 2: Time course for kinetochore microtubule acquisition (from McEwen et al., 1997).
Thus even at late metaphase (the stage of mitosis when all the chromosomes are bi-oriented and aligned at the equator), not all of the kinetochores are saturated with microtubules. Another factor is the variation in the size of kinetochore plates. One of the unresolved questions in mitosis research is why do kinetochores bind so many microtubules when one is enough to initate movement? Another is how do kinetochores interact with kMts to produce poleward motion?
Is Poleward Force Production Related to Kinetochore Microtubule Numbers?
One area of active research is to discover the mechanism that switches kinetochores between poleward (P) and away from the pole (AP) motion. Earlier ideas envisioned a tug-of-war between poleward pulling sister kinetochores, with the relative strength of each kinetochore being dependent upon the number and length of microtubules it bound. When we counted the number and length of microtubules bound to congressing kinetochores we found the simple tug-of-war model to be incorrect.
Figure 3: Numbers of bound microtubules and distances from the spindle poles for sister kinetochores on congressing chromosomes (from McEwen et al., 1997).
Note that the leading sister kinetochore always had less than half the number of microtubules as the trailing kinetochore. In one case, the leading kinetochore only had one microtubule while the trailing kinetochore had 26! Thus, number and length of kinetochore bound microtubules is not correlated to the direction of chromosome motion.
Is the Direction of Chromosome Motion Related to the Depth of Microtubule Penetration?
A variety of experimental evidence demonstrates that increasing tension on a kinetochore switches off poleward motion, but the mechanism by which this occurs is unknown. Recently we tested the hypothesis that poleward force is generated on the inner surface of the kinetochore and that high-tension levels switch off poleward motion by pulling microtubules to the outer surface of the kinetochores. This would predict that microtubules penetrated more deeply into the kinetochore during poleward motions. However, our results show that depth of penetration is not related to direction of motion. This also means that growth and shrinkage of kinetochore microtubules is not tightly controlled or coordinated with the direction of motion.
Figure 4: Distribution of the penetration of kinetochore microtubules into the outer plate of kinetochores traveling poleward (P) and away from the pole (AP) in monooriented and congressing chromosomes (from McEwen and Heagle, 1997).
Is the Signal for Anaphase Onset Dependent Upon the Number of Kinetochore Microtubules?
It is known that single unbound kinetochore will delay anaphase onset. Therefore, a full complement of microtubules on each kinetochore may be a signal that the cell is ready to initiate anaphase. We measured the number of microtubules bound to cells treated with taxol, an anti-cancer drug that inhibits anaphase onset.
Figure 5: The number of microtubules bound on kinetochores of anaphase and taxol inhibited PtK1 cells (from McEwen et al., 1997).
Kinetochores in these cells had the same number of microtubules as early anaphase cells, demonstrating that taxol does not induce its inhibition by reducing the number of kinetochore microtubules. Furthermore, some of the kinetochores of late metaphase and early anaphase chromosomes have no more than half their full complement of kinetochore microtubules. From these data we conclude that a full complement of kinetochore microtubules is not necessary for anaphase onset, and an adequate number does not insure anaphase onset.
What is the Relationship Between Kinetochore Size and the Number of Kinetochore Microtubules?
If the number of microtubules bound to a kinetochore is not that important for force generation or the signaling of anaphase onset, why does the cell use so many? To address the question of does the cell need to bind so many, we measured density of bound microtubules as a function of kinetochore size in PtK cells.
Figure 6: Plot of microtubule binding density verses kinetochore size (McEwen, Ding, and Heagle, 1998).
This graph shows that smaller kinetochores pack microtubules more tightly, as if they had to be crowded together to achieve a minimum number. This conclusion is reinforced by the observation that distribution of kinetochore sizes in tissue culture cells, where kinetochore rearrangements occur frequently, is skewed toward the larger end of the scale (Cherry et al., 1987, J. Cell Sci. 92:281-289). The skewed size distribution indicates that it is critical for the kinetochore to be larger than a minimum size, which when combined with our packing data, gives a strong indication that kinetochores must be able to bind more than a minimum number of kinetochore microtubules.
We suggest that capacity to bind more than a minimum number of kinetochore microtubules is required for a chromosome to be stable on the mitotic spindle. Since kinetochores acquire microtubules slowly, relative to the duration of mitosis, kinetochores are frequently less than half saturated upon entry into anaphase. There is ample evidence that microtubule binding is dynamic, with microtubules frequently coming on and off. If a kinetochore can only bind 3 - 4 microtubules, there would be a high probability that in some cell divisions the kinetochore would lose all of its microtubules and fail to segregate properly during anaphase.