Understanding Mitosis: Part 2 Lecture Notes
Key words and terms
Experiments that reveal properties of a complex process, like mitosis, are ways to gain a deeper understanding of the machinery that underlies that process, i.e., how it works. Observations and experiments on a range of different organisms can be informative about the features of mitosis that are basic, but they can also be confusing as a result of natural variability among organisms.
A review-overview of mitosis is given: prophase, prometaphase, metaphase, anaphase, telophase, and what happens in each.
A review of spindle structure is given with an emphasis on spindle symmetry and overall organization. The rest of the lecture will be described below by posing and answering the major questions that are addressed.
Question 1: How do microtubules growing from the spindle’s two centrosomes (the spindle poles) interact to form the spindle’s 2-fold symmetric structure?
The protein Ase1 is localized where microtubules (MTs) that grow from the two poles interdigitate. When Ase1 is removed from a cell (e.g., fission yeast) by gene deletion, the spindles that form have a strong tendency to break into two parts.
The kinesin-like motor enzyme kinesin-5 is also important for the formation of a bipolar (2-fold symmetric) spindle. Its inactivation by mutation or by drugs that block its enzyme activity rather specifically leads to the formation of monopolar spindles in which the two centrosomes have not separated. The localization of Kinesin-5 is more uniform than that of Ase1, but the enzyme is certainly found in the zone of MT overlap, near the spindle midplane. This motor is a homotetrameric assembly of four kinesin-like heavy chains, arranged in a 2-fold symmetric structure. When working as a motor, it walks towards the plus ends of MTs (which are located distal to the spindle poles), so kinesin-5 is likely to contribute to the forces that push spindle poles apart.
Since Ase1 is not a motor but does bind MTs, we can imagine that its presence as a cross-linker between interdigitating MTs impedes the MT sliding that is favored by kinesin-5. This kind of antagonism between molecular functions will come up again.
Chromosomes themselves are not essential for the formation of a 2-fold symmetric spindle, because they can be removed by micromanipulation, and the spindle still develops into a 2-fold symmetric structure, so long as the chromosomes are not removed too early in prometaphase.
Question 2: What defines the length of a bipolar spindle?
In the group of unicellular algae called “Diatoms”, the spindle is seen by polarization microscopy as a shaft that runs from one spindle pole to the other (though electron microscopy shows that this shaft is actually formed from two sets of interdigitating MTs that overlap near the spindle midplane). When a microbeam of ultraviolet light is used to irradiate one side of this midregion, the remaining spindle collapses by bending with the inside of the bend facing the site of irradiation. This result implies that the poles of the spindle are being pulled inward at metaphase, presumably by the reaction to forces that are pulling the chromosomes toward the spindle poles (recall these forces from the Introductory lecture). Thus, we must think of a metaphase spindle as a mechanical system in which kinesin-5 is forcing the sliding apart of the two interdigitating families of MTs that grew from each spindle pole, Ase1 is resisting this sliding and holding the two MT families together in a bundle, the chromosomes are being pulled toward the spindle poles, and the poles are being pulled in toward the spindle midplane (or “equator”, as it is sometimes called).
But there is another motor in the mix: a kinesin-14 is localized in the region of spindle MT overlap, and this kind of motor is minus end-directed. This means that when it cross-links the interdigitating MTs, its motor action will tend to make them INCREASE the extent of their overlap and pull the poles in toward the spindle midplane. Thus, the mechanical equilibrium in the spindle includes multiple, antagonistic actions, much as you might use antagonistic muscles to control a fine motor operation.
Question 3: Are spindle microtubules dynamic, and if so what are their sites of polymerization and depolymerization?
Evidence that is mentioned but not presented shows that spindle MTs are “labile” in the sense that they turn over rapidly. A more complicated and interesting form of MT dynamics is revealed by “speckle imaging”, the use of small amounts of fluorescent tubulin to label MTs. When only a little fluorescent tubulin is added to a cell, the MTs that form have randomly distributed brighter and dimmer spots along their lengths. When these MTs are imaged in with a very sensitive video camera, they are seen to migrate toward both of the spindle poles; all of the spindle MTs seem to be moving away from the spindle midplane, a motion that has been called “flux”. Yet the spindle at metaphase is not getting longer, suggesting that the MTs are depolymerizing at the poles. There is yet another kinesin-like protein, a kinesin-13, that has been found at the poles, and this motor promotes MT depolymerization. Experiments favor the model that flux includes pushing of MTs from the midplane by the action of kinesin-5 and depolymerization of MTs at the poles by kinesin-13. Thus, the dynamics of spindle MTs are complex and include not only the polymerization and depolymerization of tubulin but also the action of multiple motor enzymes.
Question 4: How might MT flux be related to chromosome motion in mitosis? Do the data support the hypothesis that this is a fundamental and widely used mechanism for chromosome motion?
A plausible model for chromosome motion is that the dynamics of the spindle’s MTs makes a sort of conveyer belt, and anaphase occurs when sister chromatids separate and join the flux of MTs toward the two spindle poles. This may be an explanation for many aspects of mitosis, at least in some cells.
However, biological variability lets us see that this is not the whole story. In fission yeasts there does not appear to be any flux, but there is a motion of the spindle MTs from the midplane toward the poles during anaphase B (spindle elongation), as revealed by experiments in which marks are placed on fluorescent spindle MTs by photobleaching them with a laser microbeam.
Moreover, in some fungi, the motion apart of the spindle poles that occurs during anaphase B is not driven from the zone of MT overlap but is a result of pulling on the spindle poles by astral microtubules that interact with the cell’s cortex. This motion is probably driven by dynein.
Question 5: How do kinetochores form stable attachments with spindle fibers?
To understand chromosome-spindle MT interaction, we must look carefully at the parts of the spindle where this happens and the processes that result from such interactions. Moreover, such work should be done in several organisms. Electron microscopy of both algal and vertebrate cells in prometaphase shows kinetochores interacting with both the walls and the ends of MTs. Descriptions of chromosome attachment by light microscopy of living cells show that a chromosome starts to move poleward as soon as its kinetochore associates with a MT. This is a minus end-directed motion over the surface of a MT; dynein is localized on kinetochores, so it is plausible that initial chromosome-MT interactions are mediate by this motor enzyme, in at least some cells.
The interaction between a chromosome and its spindle fiber has been probed by micromanipulation. A metaphase chromosome that has been detached from the spindle will soon re-associate with the spindle and go to the metaphase plate. If the detached chromosome forms an inappropriate attachment to the spindle, e.g., with both kinetochores associating with MTs from one pole, these connections soon break, and the chromosome reattaches to the spindle, trying again to achieve biorientation. If, however, a maloriented chromosome (one whose kinetochores are both attached to the same pole) is put under tension by the action of a microneedle, it is stable and does not reorient. This suggests that the stability of chromosome-spindle fiber attachment is a result of tension. Dynein may contribute to the generation of this tension, but there are many additional possibilities, as discussed in the next lecture.
The idea that tension enhances stability is important because it allows us to understand how chromosomes that are properly associated with the spindle form stable connections, while those that are improperly associated (e.g., both kinetochores attached on one pole) do not. When sister kinetochores are associated with sister poles, a chromosome is under tension; when sister kinetochores are associate with the same pole, there is little or no tension, so this chromosome will be released and can try again to do the job right. Thus, biologically valuable chromosome attachments to the spindle are “selected”, because the others are unstable and break, allowing the chromosome to try again.
Question 6: How does the spindle generate tension on kinetochores?
Dynein is on kinetochores, at least in early mitosis in many cell types, so it is a candidate for tension development. However, the injection of antibodies to dynein does not block chromosome attachment. One can argue that these reagents are simply not effective enough, but using more rigorous experimental approaches poses problems. Dynein does many things in cells, so a deletion mutation is not a good tool for studying the role of dynein in chromosome attachment to the spindle; too many other processes are affected. No good temperature sensitive alleles of dynein are currently available to investigate this issue, so it is still not well resolved. There are drugs and other experimental treatments that block the action of some kinetochore dynein-associated proteins, but exactly what these reagents do is still open to question.
The whole problem is made more complicated by the fact that there are other motors at kinetochores. Some of these are plus end-directed motors, and it is not clear how these could develop tension, but there are kinesin-13s at the kinetochores of some organisms and kinesin-8s at the kinetochores of others. Both these motors promote MT depolymerization, so this might be a source of tension. Moreover, the only kinetochore proteins so far identified that are essential for chromosome-spindle fiber attachment and also universal among organisms are not motors. They are simply kinetochore and MT-binding proteins. MT depolymerization is therefore a plausible source of tension generation in mitosis, and this is the subject of the next McIntosh lecture. For now, we must leave this issue as unresolved.
Question 7: How does the spindle get the chromosomes to the spindle midplane?
Experiments with laser microsurgery show that pieces of chromosome that lack a kinetochore are pushed away from the pole by the spindle of a vertebrate cell. This suggests that at least some spindles not only pull on kinetochores, they push on chromosome arms. A “bioriented” chromosome, which is attached by its sister kinetochores to opposite poles, will experience two approximately equal and opposite pulls at its kinetochores, so another force, like a push from the poles that is stronger when the chromosome is near a pole, would result in chromosome motion to the spindle midplane. Such a mechanism probably contributes to the formation of a metaphase spindle in at least some cell types. But not all cells have this pushing mechanism, so there may be other factors that contribute to pre-anaphase chromosome motion.
Question 8: How does the spindle pull its chromosomes to the poles in Anaphase A?
Answers to this question are implied in what is said above, but this issue is the subject of the next McIntosh iBio Seminar and will be left until that lecture.