Cytoskeleton II: Microtubules and associated proteins

Cytoskeleton II:
Microtubules The second known cytoskeletal element is the microtubule, a hollow proteinaceous tube about 25 nm in diameter that looks a bit like a straw--long, thin and hollow. These are formed by the polymerization of a molecule that consists of two subunits, called alpha and beta tubulin. These two proteins bind to each other, and then these dimers (called "heterodimers" because the two subunits are different from each other) can interact one with another until they stick end to end and side to side in a helical array that forms the wall of the tubule (Figure 19-2. Lodish et al.). A number of microtubule associated proteins--also called MAPs--have been found that promote polymerization or depolymerization of the microtubule; MAPS are especially important in establishing the stable arrangements of microtubules that are found in nerve cells, where microtubules form the central core of axons and dendrites. (Interestingly one of the MAPs, called Tau protein, is a major component of the "fibrillary tangles" that are a characteristic lesion in the brain of people who have Alzheimers disease (p. 67-68 and Fig. 3-19 in Lodish et al.). The role of Tau in Alzheimers, if any, is under intense scrutiny).
The two ends of a microtubule elongate at different rates and the rate of elongation of the microtubule depends on the rate of growth at one end being greater than the rate of dissociation at the other. Otherwise the length decreases. Microtubules, like actin filaments, have positive (+) and negative (-) ends that grow at different rates, and they grow faster from the positive end. However, the process of elongation is somewhat more complex for microtubules than for actin filaments, because of the three dimensional natures of the microtubules. Thus, not only do the tubulin dimers form and add end to end, but they must bind side to side in sheets, and curl up to form the tube-like structure that gives microtubules their name (Fig. 19-11).
Assembly of tubulin dimers into microtubules is regulated not only by MAPs but by GTP which binds to the beta subunit (GTP also binds to the alpha subunit, but this GTP is less important in controlling the stability of microtubules). When GTP is bound to the beta subunit, the dimer associates strongly with other dimers; the microtubule will be stable and may even elongate. However, tubulin is a GTPase that eventually breaks down GTP to GDP, and when it does, the tubulin dimers associate with each other much more weakly. At this point, they will in fact dissociate unless they are held in place by stable cap structures of GTP-tubulin or by MAPs (Fig. 19-15). And indeed microtubules have been observed to undergo dramatic changes of length, rapid growth followed by rapid disassembly, in a cyclic fashion whose controlling factors are not completely understood (Fig. 19-13).
All the microtubules in animal cells seem to grow from the same source, a structure called the centrosome (central body). Microtubules have a directionality or polarity, like actin filaments, because the two tubulins that make them up are not identical and the two ends grow at different rates, called plus and minus as I indicated above. In all cells studied to date, the end of the microtubules closer to the centrosome is the minus end, the end facing away is the plus end. (Fig. 19-6). As shown in Fig. 19-6, the centrosome, or similar structures in non-animal cells, is the "nucleating" region for microtubule growth, the so-called "microtubule organizing center". That is, the centrosome seems to help stabilize small microtubules at low concentrations of tubulin by binding to the (-) end of the microtubule (and thus preventing dissociation of the subunits); this allows the microtubule to grow (i.e., elongate) in the (+) direction. This arrangement of the microtubules also provides a directionality to the whole cell, because microtubules form in a spoke-like array from the centrosome outward (and, as their name implies, centrosomes are centrally located). Thus the microtubules establish a kind of "toward the center" and a "toward the periphery" directionality in eukaryotic cells, that the cells take advantage of when they want to move materials in one direction or the other.
Microtubules, like actin filaments, are essential for movement in cells, though of course different organelles or molecules use the two for movement. There are at least two classes of motor proteins that can drag themselves along microtubules. One is called the dyneins, which are required for movement in the + to - direction, and the other is kinesins, which are required for movement in the - to + direction (Fig. 19-26). Both of these can attach on one part of the molecule to some organelle like mitochondria or vesicles--they can even attach to stryofoam beads. On another part of the molecule they have a structure that binds microtubules and that pulls the motor protein along by shape changes, involving hydrolysis of ATP. Dyneins were originally discovered because of their role in themovement of cilia, beating "arms" on the surface of many eukaryotic cells, including those lining your airways to the lungs; these help clear debris and mucus out of your lungs (unless you smoke, which tends to paralyze the cilia).
The kinesins were discovered from a series of very clever experiments conducted in the mid-1980s using nerve cells. Nerve cells have a stable cytoskeleton that forms the core of their axons, and it's been known for over 50 years that "stuff" (vesicles, mitochondria, etc.) moves up and down the axons, a process called "anterograde" (forward) and "retrograde" (backward) axonal transport. . The microtubules in the axon are organized like others in the cell--that is, the minus end is toward the cell center (i.e., nucleus) and the plus end is toward the cell periphery (i.e., the synapse). It was known that dynein carried material along microtubules toward the minus end, so it could be responsible for retrograde transport, but what carried materials in the forward direction? To find out, scientists squeezed the cytosol out of axons (this stuff is called the axoplasm) onto a microscope slide, and they showed that if they added some ATP back, they could watch movement of materials back and forth in the retrograde and anterograde direction. They also showed that adding pure mitochondria and microtubules together with ATP did not cause movement, but if they mixed some axoplasm with the microtubules, they could get the mitochondria moving around. This provided them with an assay to find out what component(s) of axoplasm are necessary for the reconstitution of the movement on a microscope slide. In this way they purified the first kinesin, and identified the anterograde motor protein. If you want to see some cool movies of this process of movement of material in axons, check out the web sites of Drs. Lochner and Scalletar who study the process of movement of material in nerve cells (at http://www.lclark.edu/~bethe or http://www.lclark.edu/~lochner).
Just as there are numerous types of myosin, there are numerous kinds of kinesin, and all share a generally similar structure--a dimer with head, tail, and neck, a bit like myosin (Fig. 19-23). It's believed that the head binds to microtubules, the tail binds to vesicles, mitochondria, even styrofoam beads, and by a sort of "walking" process, according to the current best studies, they can move materials along the microtubules with them (Fig. 12-24).
Dyneins are big, hunking proteins (often of 1 million or more MW) that serve as minus-end-directed motor proteins. They also have two head groups that bind to microtubules (like kinesins), but they apparently don't bind directly to the organelles that they move. Rather, they use a bunch of accessory proteins that form a link between dynein and the organelle (Fig. 19-25).
The current idea of organelle movement is summarized in Fig. 19-26, which shows the minus end directed movements of organelles like lysosomes or endosomes or Golgi-to-ER vesicles, as well as the plus end directed movement of mitochondria, ER to Golgi directed vesicles, etc. As a quick look at Dr. Scalletar and Lochner's movies will convince you, the same organelle can often be seen to move backward and forward in a very short span of time. This strongly suggests that both kinds of motor proteins--kinesins and dyneins--bind to the same organelle at the same time or in very quick succession. What, if anything, regulates the directionality of movement, and switch between directions, is not yet understood.
Finally, it seems likely that vesicles and organelles can be moved by different motor proteins in different parts of the cell. That is, sometimes a microtubule-based motor might be involved in moving a vesicle, while at other times, an actin-filament based motor might be involved. This is most strikingly illustrated in the axons of nerve cells where the microtubules end just as the thin part of the axon widens out into a nerve terminal; the nerve terminal is filled with actin filaments but not microtubules. But it's also known that synaptic vesicles, the vesicular structures that carry neurotransmitter molecules, go all the way to the plasma membrane of the nerve terminal. It's therefore likely that the synaptic vesicles are carried down the axon by kinesin moving along microtubules, but that when they reach the nerve terminal, the vesicle dissociates from the kinesin/microtubule complex, and is bound by myosin, which carries it along actin filaments out to the plasma membrane.
In short, then, microtubules are, like actin filaments, dynamic structures that can form and disassemble, and that serve as "railroad tracks" for movement of organelles powered by motor proteins. Also like actin filaments, they are sometimes stabilized by accessory proteins like MAPs that allow them to form long-lasting structures that give shape to the cells, such as the microtubular core of an axon.