Using our discussion of actin and myosin interactions in muscle cells as background, we now examine the function of actin-myosin structures in nonmuscle cells. At first, scientists thought that most cell movements were caused by a contractile mechanism similar to the sliding of actin and myosin filaments in muscle cells. This idea was based on several properties of at least some nonmuscle cells: the ability of cytosolic extracts to undergo contractile-like movements, the presence of actin and myosin II, and the existence of structures similar to muscle sarcomeres both in their organization and in their having ends anchored to the plasma membrane by proteins also found at the muscle Z disk.
Later biochemical studies led to the extraction of myosin I, which does not form thick filaments, from nonmuscle cells. As discussed earlier in the Chapter, 13 different myosins have been identified to date, but researchers have studied only 3 — myosin I, myosin II, and myosin V — in most detail. In this section, we discuss some of the functions of these three myosins in various nonmuscle cells.
Nonmuscle cells contain prominent bundles of actin and myosin II filaments in the cellular region that contacts the substratum or another cell. When isolated from cells, these bundles contract upon addition of ATP. The contractile bundles of nonmuscle cells differ in two ways from the noncontractile bundles of actin described earlier in this chapter (see Figure 18-5a). First, contractile bundles always are located adjacent to the plasma membrane like a sheet or belt, whereas noncontractile actin bundles form the core of membrane projections (microvilli and filopodia). Second, interspersed among the actin filaments of a contractile bundle is myosin II, which is responsible for the contractility of the bundle. The noncontractile actin bundles in microvilli and filopodia sometimes are associated with myosin I, but myosin II usually is not a major component of these bundles. Despite their ability to contract, contractile bundles probably function primarily in cell adhesion rather than cell movement.
In epithelial tissue, a belt of actin and myosin filaments rings the inner surface of the cell adjacent to the adherens junctions, where cell-cell contacts are maintained. The circumferential belt is attached by linker proteins to cell-adhesion molecules in the plasma membrane (Chapter 22).
; see Figure
15-23). A circumferential belt resembles a primitive sarcomere in its
organization and contains many proteins found in stress fibers and smooth
muscle, including vinculin, tropomyosin, and α-actinin. As a complex
with the adherens junction, the circumferential belt functions as a tension
cable that can internally brace the cell and thereby control its shape.
Contraction of the circumferential belt in epithelial cells surrounding a wound
seals the gap in the sheet of cells and thus aids in wound healing.
As noted early in this chapter, long bundles of actin microfilaments, called stress fibers, lie along the ventral surfaces of cells cultured on artificial (glass or plastic) surfaces (see Figure 18-1b). Fluorescent-antibody techniques reveal that myosin and α-actinin are distributed in alternating patches in stress fibers, much like the pattern of alternating thick filaments and Z bands in muscle sarcomeres. Stress fibers also contain tropomyosin, caldesmon, and the regulatory protein myosin LC kinase, which are found in smooth muscle. The ends of stress fibers terminate at focal adhesions, special structures that attach a cell to the underlying substratum. When stress fibers are separated from focal adhesions by cutting their ends with a laser beam, they contract on addition of ATP, thus demonstrating their contractility.
Although stress fibers are contractile and found in motile cells, several observations suggest that they function in cell adhesion rather than in movement. For one, migrating fibroblasts have few stress fibers during the time of rapid movement; however, once the cells stop migrating, the stress fibers increase in number. Also, when a cultured fibroblast is removed from its substratum, the cell becomes spherical and the stress fibers disappear. If the cell is returned to its substratum, stress fibers reappear within a few hours. In fact, stress fibers may be an artifact caused by culturing cells on glass or plastic surfaces, as they are rarely seen in cells in tissues. Apparently, the adhesion of cells to a substratum induces stresses on the cytoskeleton that cause the random assortment of actin and myosin filaments to align into stress fibers.
In Chapter 22, which covers the integration of cells into tissues, we describe the complex structure of adherens junctions and focal adhesions and how they are attached to contractile bundles.
(a) In cells containing a normal cytoskeleton, the membrane resists deformation when a bead-topped poker is pushed against it. In a mutant cell lacking myosin II, the membrane is easily deformed when the same force is applied with a poker. (b) Integral membrane proteins that are attached to the underlying cytoskeleton aggregate when bound by specific antibodies or lectins, forming a patch or “cap” on the cell surface. Actin and myosin filaments also collect beneath the membrane even though they are not bound by the antibodies or lecithin. In cells lacking functional myosin II, membrane proteins are unable to cap, suggesting that movement of membrane proteins depends on cortical myosin.
). Second, capping
experiments outlined in Fig 18-36b
indicate that cortical myosin II is responsible for the movements of some
cell-surface proteins. Although cell-surface proteins are normally immobile in
the membrane, they will cluster, or “cap,” at one region in
the membrane in the presence of antibodies or lectins that bind them. Capping is
inhibited in cells lacking myosin II, suggesting that myosin II in the cortex
provides the force that aggregates the membrane proteins. In the last section of
this chapter, we discuss how myosin-dependent tension in the cortex may
contribute to cell locomotion.
(a) As the spindle poles and aster microtubules separate during mitosis, myosin (blue) and actin (red) in the cortex of the cell assemble into an equatorial contractile ring around the cell. As the nuclei in the daughter cells start to re-form, the contractile ring constricts, causing the membrane to form a cleavage furrow. In the last step of cell division, cytokinesis, the cell is pinched into two parts. (b) During cell division of a Dictyostelium ameba, myosin II (red) is concentrated in the cleavage furrow, while myosin I (green) is localized at the poles of the cell. [Part (b) Courtesy of Y. Fukui.]
). There they align
into a contractile ring, which is similar to a stress fiber or
circumferential belt and encircles the cell. As division of the cytoplasm
(cytokinesis) proceeds, the diameter of the contractile ring decreases, so that
the cell is pinched into two parts by a deepening cleavage furrow. Dividing
cells stained with antibodies against myosin I and myosin II show that myosin II
is localized to the contractile ring, while myosin I is at the cell poles (Figure 18-37b). This localization
indicates that myosin II but not myosin I is involved in cytokinesis.
The activity of myosin II was inhibited either by deleting its gene or by microinjecting anti-myosin II antibodies into a cell. A cell that lacked myosin II was able to replicate its DNA and nucleus, but it failed to divide; this defect caused the cell to become large and multinucleate. In comparison, an untreated cell during the same period continued to divide and formed a multicellular ball of cells in which each cell contained a single nucleus.
). In one type of experiment, anti-myosin II
antibodies were microinjected into one blastomere of a sea urchin embryo at the
two-cell stage. In other experiments, expression of myosin II was inhibited in
the Dictyostelium ameba by genetic deletion of the myosin gene
or by antisense inhibition of myosin mRNA expression. The results were identical
in all cases: cells lacking myosin II became multinucleated because cytokinesis
but not karyokinesis (chromosome separation) was inhibited. Without myosin II,
the cells failed to assemble a contractile ring, although other events in the
cell cycle proceeded.
Among the many movements exhibited by cells, vesicle translocation has been one of the most fascinating to cell biologists. In early studies of the cytoplasm, researchers found that certain particles, now known to be membrane-bounded vesicles, moved in straight lines within the cytosol, sometimes stopping and then resuming movement, at times after changing direction. This type of behavior could not be caused by diffusion because the movement was clearly not random. Therefore, researchers reasoned, there must be tracks, most likely actin filaments or microtubules, along which the particles travel, as well as some type of motor to power the movement.
Unlike cytokinesis, which involves myosin II, other motile processes, such as vesicle transport and membrane movements at the leading edge, possibly involve other myosins and, as discussed in the next chapter, some microtubule motor proteins also. Here we present evidence that some myosins, including myosins I and V, can move along an actin filament while carrying a membrane vesicle as cargo. In these processes, the interaction of actin and myosin differs from that in the sarcomere. We defer discussion of the role of myosin I in directing the movement of the leading edge of cells until the last section of this chapter.
Studies with amebas provided the initial clues that myosin I participates in vesicle transport. Indeed the first myosin I molecule to be identified and characterized was from these organisms. The cDNA sequences of three myosin I genes have now been identified in Acanthameba, a common soil ameba. Using antibodies specific for each myosin I isoform, researchers found that they are localized to different membrane structures in the cell. For example, myosin IA is associated with small cytoplasmic vesicles. Myosin IC, by contrast, is found at the plasma membrane and at the contractile vacuole, a vesicle that regulates the osmolarity of the cytosol by fusing with the plasma membrane. The introduction of antibodies against myosin IC into a living ameba prevents transport of the vacuole to the membrane; as a result, the vacuole expands uncontrollably, eventually bursting the cell.
Myosin I is also implicated in vesicle transport in vertebrate cells. For example, in intestinal epithelial cells, myosin I co-purifies with vesicles derived from Golgi membranes. The presence of this motor on Golgi membranes suggests myosin I moves membrane vesicles between membrane compartments in the cytoplasm. In addition myosin I serves as a membrane-microfilament linkage in microvilli (see Figure 18-10), another example of a membrane-associated function.
In these cells, which are found in nervous tissue, myosin V is concentrated in the Golgi stacks (bright central region) and at the tips of membrane processes that extend from the cell. This distribution indicates that myosin V is associated with membranes and suggests that the protein is involved in membrane transport from the Golgi to the cell periphery. [From E. M. Espreafico et al., 1992, J. Cell Biol. 119:1541; courtesy of R. E. Cheney and M. Mooseker.]
). This association with membranes is
consistent with the effects of myosin V mutations in mice. Such mutations
are implicated in defects in synaptic transmission and eventually cause
death from seizures.
(a) The center of a Nitella cell is filled with a single large water-filled vacuole, which is surrounded by a layer of moving cytoplasm (indicated by blue arrows). A nonmoving layer of cortical cytoplasm filled with chloroplasts lies just under the plasma membrane (enlarged lower figure). On the inner side of this layer are bundles of stationary actin filaments (red), all oriented with the same polarity. A myosinlike motor protein (blue dots) carries portions of the endoplasmic reticulum (ER) along the actin filaments. The movement of the ER network propels the entire viscous cytoplasm, including organelles that are enmeshed in the ER network. (b) An electron micrograph of the cortical cytoplasm shows a large vesicle connected to an underlying bundle of actin filaments. This vesicle, which is part of the endoplasmic reticulum (ER) network, contacts the stationary actin filaments and moves along them by a myosinlike motor. [Part (b) from B. Kachar.]
). The rapid flow of
cytosol, sustained over millimeter-long distances, is a principal mechanism
for distributing cellular metabolites, especially in large cells such as
plant cells and amebas. This type of movement probably represents an
exaggerated version of the smaller-scale movements exhibited during the
transport of membrane vesicles.
). The bulk cytosol is propelled by
myosin attached to parts of the ER lying along the stationary actin
filaments. Although the Nitella myosin has not been
isolated, it must be one of the fastest known, because the flow rate of the
cytosol in Nitella is at least 15 times faster than the
rate produced by any other myosin.In nonmuscle cells, interactions of actin filaments with various myosins are important in such cellular functions as support, cytokinesis, and transport.
Actin filaments and myosin II form contractile bundles that have a primitive sarcomerelike organization and function in cell adhesion. Common examples are the circumferential belt present in epithelial cells and stress fibers found along the ventral surface of cells cultured on plastic or glass surfaces. The latter rarely occur in cells in tissues and may be an artifact.
Interaction of myosin II with cortical actin filaments helps stiffen the plasma membrane, reducing the likelihood of surface deformation.
).
).
