NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Lodish H, Berk A, Zipursky SL, et al. Molecular Cell Biology. 4th edition. New York: W. H. Freeman; 2000.

  • By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.
Cover of Molecular Cell Biology

Molecular Cell Biology. 4th edition.

Show details

Section 18.2The Dynamics of Actin Assembly

Thus far we have treated the actin cytoskeleton as if it were an unchanging structure consisting of bundles and networks of filaments. The microfilaments in a cell, however, are constantly shrinking or growing in length, and bundles and meshworks of microfilaments are continuously forming and dissolving. These changes in the organization of actin filaments cause equally large changes in the shape of a cell. In this section, we discuss the mechanism of actin polymerization and the regulation of this process, which is largely responsible for the dynamic nature of the cytoskeleton.

Actin Polymerization in Vitro Proceeds in Three Steps

As we mentioned earlier, addition of salts to a solution of G-actin induces polymerization, creating F-actin filaments. The polymerization process can be monitored by viscometry, sedimentation, and fluorescence spectroscopy. When actin filaments become long enough to become entangled, the viscosity of the solution increases, which is measured as a decrease in its flow rate in a viscometer. The basis of the sedimentation assay is the ability of ultracentrifugation (100,000g for 30 minutes) to pellet F-actin but not G-actin. The third assay makes use of G-actin covalently labeled with a fluorescent dye; the fluorescence spectrum of the modified G-actin monomer changes when it is polymerized into F-actin. These assays are useful in kinetic studies of actin polymerization and during purification of actin-binding proteins, which cross-link or depolymerize actin filaments.

The polymerization of actin filaments proceeds in three sequential phases (Figure 18-11a). The first phase is marked by a lag period in which G-actin aggregates into short, unstable oligomers. Once the oligomer reaches a certain length (three or four subunits) it can act as a stable seed, or nucleus, which in the second phase rapidly elongates into a filament by the addition of actin monomers to both of its ends. As F-actin filaments grow, the concentration of G-actin monomers decreases until it is in equilibrium with the filament. This third phase is called steady state because G-actin monomers exchange with subunits at the filament ends but there is no net change in the total mass of filaments. The kinetic curves shown in Figure 18-11b show that the lag period can be eliminated by addition of a small number of F-actin nuclei to the solution of G-actin.

Figure 18-11. The three phases of G-actin polymerization in vitro.

Figure 18-11

The three phases of G-actin polymerization in vitro. (a) During the initial nucleation phase, ATP – G-actin monomers (pink) slowly form stable complexes of actin (purple). These nuclei are more rapidly elongated in the second phase (more...)

Once the steady-state phase is reached, the equilibrium concentration of the pool of unassembled subunits is called the critical concentration (Cc). This parameter is a measure of the ability of a solution of G-actin to polymerize. Under typical in vitro conditions, the Cc of G-actin is 0.1 μM. Above this value, a solution of G-actin will polymerize; below this value, a solution of F-actin will depolymerize (Figure 18-12).

Figure 18-12. The critical concentration (Cc ) is the concentration of G-actin monomers in equilibrium with actin filaments.

Figure 18-12

The critical concentration (Cc ) is the concentration of G-actin monomers in equilibrium with actin filaments. At monomer concentrations below the Cc, no polymerization occurs. At monomer concentrations above the Cc, filaments assemble until the monomer (more...)

After ATP – G-actin monomers are incorporated into a filament, the bound ATP is slowly hydrolyzed to ADP. As a result of this hydrolysis, most of the filament consists of ADP – F-actin, but ATP – F-actin is found at the ends. However, ATP hydrolysis is not essential for polymerization to occur, as evidenced by the ability of G-actin containing ADP or a nonhydrolyzable ATP analog to polymerize into filaments.

Actin Filaments Grow Faster at One End Than at the Other

We saw earlier that myosin decoration experiments reveal an inherent structural polarity of F-actin (see Figure 18-3). This polarity also is reflected in different rates of monomer addition to the two ends. One end of the filament, the (+) end, elongates five to ten times faster than does the opposite, or (−), end. The unequal growth rates can be demonstrated by a simple experiment using myosin-decorated actin filaments to nucleate polymerization of G-actin. Electron microscopy of the elongated filaments reveals bare sections at both ends, corresponding to the added undecorated G-actin. The newly polymerized (undecorated) actin is five to ten times longer at the (+) end than at the (−) end of the filaments (Figure 18-13a).

Figure 18-13. Experimental demonstration of unequal growth rates at the two ends of an actin filament.

Figure 18-13

Experimental demonstration of unequal growth rates at the two ends of an actin filament. (a) When short myosin-decorated filaments are the nuclei for actin polymerization, the resulting elongated filaments have a much longer undecorated (+) end than (−) (more...)

The difference in elongation rates at the opposite ends of an actin filament is caused by a difference in Cc values at the two ends. This difference can be measured by blocking one or the other end with proteins that “cap” the ends of actin filaments. If the (+) end of an actin filament is capped, it can elongate only from its (−) end; conversely, elongation occurs only at the (+) end when the (−) end of a filament is blocked (Figure 18-13b). Polymerization assays of such capped filaments have shown that the Cc is much lower for G-actin addition at the (+) end (Cc+ = 0.1 μM) than for addition at the (−) end (Cc = 0.8 μM).

As a result of the difference in the Cc values for the (+) and (−) ends of a filament, we can predict the following: at G-actin concentrations below Cc+, no filament growth occurs; at G-actin concentrations between Cc+ and Cc, growth occurs only from the (+) end; and at G-actin concentrations above Cc, growth occurs at both ends, although it is faster at the (+) end than at the (−) end. Once the steady-state phase is reached at G-actin concentrations intermediate between the Cc values for the (+) and (−) ends, subunits continue to be added at the (+) end and lost from the (−) end (Figure 18-13c). The length of the filament remains constant, with the newly added subunits traveling through the filament, as if on a treadmill, until they reach the (−) end, where they dissociate. In the lamellipodia of cells, actin filaments probably turn over by a treadmilling type of mechanism. Subunits released from one end of the filament are rapidly recruited to assemble at the leading edge of the cell.

The Cc for assembly of actin filaments depends on whether the monomers are bound to ATP or ADP. When ADP-actin monomers are incorporated into actin filaments, the Cc at the (+) end becomes equal to that at the (−) end despite the inherent structural polarity of actin filaments.

Toxins Disrupt the Actin Monomer-Polymer Equilibrium

The equilibrium between actin monomers and filaments is easily perturbed by toxins. Two unrelated toxins, cytochalasin D and latrunculin, have two complementary effects. Cytochalasin D, a fungal alkaloid, depolymerizes actin filaments by binding to the (+) end of F-actin, where it blocks further addition of subunits. In contrast, latrunculin, a toxin secreted by sponges, binds G-actin and inhibits it from adding to a filament end. Exposure to either toxin shifts the monomer-polymer equilibrium in the direction of dissociation. When these toxins are added to live cells, the actin cytoskeleton disappears and cell movements like locomo-tion and cytokinesis are inhibited. These observations were among the first that implicated actin filaments in cell motility.

A third toxin, phalloidin, has the opposite effect on actin: It poisons a cell by preventing actin filaments from depolymerizing. Isolated from Amanita phalloides (the “angel of death” mushroom), phalloidin binds at the interface between subunits in F-actin and locks adjacent subunits together. Even when actin is diluted below its critical concentration, phalloidin-stabilized filaments will not depolymerize. Fluorescent-labeled phalloidin, which binds only to F-actin, is commonly used to stain actin filaments for light microscopy (see Figure 18-1b).

Actin Polymerization Is Regulated by Proteins That Bind G-Actin

In the perfect world of a test tube, experimenters can start the polymerization process by adding salts to G-actin or can depolymerize F-actin by simply diluting the filaments. Cells, however, must maintain a nearly constant cytosolic ionic concentration and thus employ a different mechanism for controlling actin polymerization. The cellular regulatory mechanism involves several actin-binding proteins that either promote or inhibit actin polymerization. Here we discuss two such proteins that have been isolated and characterized.

Inhibition of Actin Assembly by Thymosin β4

Calculations based on the Cc of G-actin (0.1 μM), a typical cytosolic total actin concentration (0.5 mM), and the ionic conditions of the cell indicate that nearly all cellular actin should exist as filaments; there should be very little G-actin. Actual measurements, however, show that as much as 40 percent of actin in an animal cell is unpolymerized. What keeps the cellular concentration of G-actin above its Cc? The most likely explanation is that cytosolic proteins sequester actin, holding it in a form that is unable to polymerize.

Because of its abundance in the cytosol and ability to bind ATP – G-actin (but not F-actin), thymosin β4 is considered to be the main actin-sequestering protein in cells. A small protein (5000 MW), thymosin binds ATP – G-actin in a 1:1 complex; in this complex, G-actin cannot polymerize. In platelets, the concentration of thymosin β4 is 0.55 mM, approximately twice the concentration of unpolymerized actin (0.25 mM). At these concentrations, approximately 70 percent of the monomeric actin in a platelet should be sequestered by thymosin β4.

Thymosin β4 (Tβ4) functions like a buffer for monomeric actin as represented in the following reaction:

Image ch18e1.jpg

In a simple equilibrium, an increase in the cytosolic concentration of thymosin β4 would increase the concentration of sequestered actin subunits and correspondingly decrease F-actin, since actin filaments are in equilibrium with actin monomers. This effect of thymosin β4 on the cellular F-actin level has been experimentally demonstrated in live cells.

Promotion of Actin Assembly by Profilin

Another cytosolic protein, profilin (15,000 MW), also binds ATP-actin monomers in a stable 1:1 complex. At most, profilin can buffer 20 percent of the unpolymerized actin in cells, a level too low for it to act as an effective sequestering protein. Rather than sequestering actin monomers, the main function of profilin probably is to promote assembly of actin filaments in cells. It appears to do so by several mechanisms.

First, as a complex with G-actin, profilin is postulated to assist in addition of monomers to the (+) end of an actin filament. This hypothesis is consistent with the threedimensional structure of the profilin-actin complex in which profilin is bound to the (+) end of an actin monomer, leaving the ATP-binding end, the (−) end, free to associate with the (+) end of a filament (Figure 18-14). After the complex binds transiently to the filament, the profilin dissociates from actin.

Figure 18-14. The three-dimensional structure of the profilin-actin complex.

Figure 18-14

The three-dimensional structure of the profilin-actin complex. Profilin (green) binds to the edge of subdomains I and III of G-actin (white) at the end opposite to the ATP-binding cleft. Profilin is unique among actin-binding proteins because it permits (more...)

Second, profilin also interacts with membrane components involved in cell-cell signaling, suggesting that it may be particularly important in controlling actin assembly at the plasma membrane. Profilin binds to the membrane phospholipid phosphoinositol 4,5-bisphosphate (PIP2); this interaction prevents binding of profilin to G-actin. (As we will see in Chapter 20, hydrolysis of PIP2 in response to extracellular signals triggers intracellular signaling pathways.) In addition, profilin binds to proline-rich sequences that are commonly found in membrane-associated signaling proteins such as Vasp and Mena. This interaction, which does not inhibit profilin binding to G-actin, localizes profilin-actin complexes to the membrane. Figure 18-15 depicts how these properties of profilin could play a central role in stimulating actin polymerization in response to cell-cell signals.

Figure 18-15. Model of the complementary roles of profilin and thymosin β4 in regulating polymerization of G-actin.

Figure 18-15

Model of the complementary roles of profilin and thymosin β4 in regulating polymerization of G-actin. (a) At the cell membrane, profilin is bound to PIP2, a membrane lipid, while most of the G-actin is complexed with thymosin β4 and thus (more...)

Finally, profilin also promotes assembly of actin filaments by acting as a nucleotide-exchange factor. Profilin is the only actin-binding protein that allows the exchange of ATP for ADP. When G-actin is complexed with other proteins, ATP or ADP is trapped in the ATP-binding cleft of actin. However, because profilin binds to G-actin opposite to the ATP-binding cleft, it can recharge ADP-actin monomers released from a filament, thereby replenishing the pool of ATP-actin (see Figure 18-15d).

Some Proteins Control the Lengths of Actin Filaments by Severing Them

A second group of proteins, which bind to actin filaments, control the length of actin filaments by breaking them into shorter fragments (Table 18-2). A valuable clue that led to the discovery of these severing proteins came from studies of amebas. Viscosity measurements and light-microscope observations demonstrated that during ameboid movement the cytosol flows forward in the center of the cell and then turns into a gel when it reaches the front end of the cell. As we discuss later, this “sol-to-gel” transformation depends on the assembly of new actin filaments in the front part of a moving ameba and the disassembly of old actin filaments in the rear part. Because the actin concentration in a cell favors the formation of filaments, the breakdown of existing actin filaments and filament networks requires the assistance of severing proteins such as gelsolin and cofilin.

Table 18-2. Some Cytosolic Proteins That Control Actin Polymerization.

Table 18-2

Some Cytosolic Proteins That Control Actin Polymerization.

Severing proteins can break the filaments in a network into shorter fragments. Currently, researchers believe that a severing protein changes the conformation of the subunit to which it binds, thereby causing strain on and subsequent breakage of the intersubunit bonds. In support of this hypothesis are electron micrographs showing that an actin filament with bound cofilin is severely twisted. After a severing protein breaks a filament at one site, it remains bound at the (+) end of one of the resulting fragments, where it prevents the addition or exchange of actin subunits, an activity called capping. The (−) ends of fragments remain uncapped and are rapidly shortened. Thus severing promotes turnover of actin filaments by creating new (−) ends and causes disintegration of an actin network, although many filaments remain cross-linked (Figure 18-16). The turnover of actin filaments promoted by severing proteins is necessary not only for cell locomotion but also for cytokinesis.

Figure 18-16. Action of gelsolin in severing actin filaments.

Figure 18-16

Action of gelsolin in severing actin filaments. At cytosolic Ca2+ levels below 10−6 M, gelsolin does not bind to actin filaments. At higher Ca2+ concentrations, binding of gelsolin to filaments causes a distortion that disrupts the noncovalent (more...)

The capping and severing proteins are regulated by several signaling pathways. For example, both cofilin and gelsolin bind PIP2 in a way that inhibits their binding to actin filaments and thus their severing activity. Hydrolysis of PIP2 releases these proteins and induces rapid severing of filaments. The reversible phosphorylation and dephosphorylation of cofilin also regulates its activity, and the severing activity of gelsolin is activated by an increase in cytosolic Ca2+ to about 10−6 M. The counteracting influence of different signaling molecules, Ca2+, and PIP2 permits the reciprocal regulation of these proteins. At the end of this chapter, we discuss how extracellular signals coordinate the activities of different actin-binding proteins, including severing proteins, during cell migration.

Actin Filaments Are Stabilized by Actin-Capping Proteins

Another group of proteins can cap the ends of actin filaments but, unlike severing proteins, cannot break filaments to create new ends. One such protein, CapZ, binds the (+) ends of actin filaments independently of Ca2+ and prevents the addition or loss of actin subunits from the (+) end. Capping by this protein is inhibited by PIP2, suggesting that its activity is regulated by the same signaling pathways that control cofilin and profilin. Tropomodulin, which is unrelated to CapZ in sequence, caps the (−) ends of actin filaments. Its capping activity is enhanced in the presence of tropomyosin, which suggests that the two proteins function as a complex to stabilize a filament. An actin filament that is capped at both ends is effectively stabilized, undergoing neither addition nor loss of subunits. Such capped actin filaments are needed in places where the organization of the cytoskeleton is unchanging, as in a muscle sarcomere or at the erythrocyte membrane.

Many Movements Are Driven by Actin Polymerization

By manipulating actin polymerization and depolymerization, the cell can create forces that produce several types of movement. A classic example is the growth of a finger of membrane from the cell surface during the acrosome reaction of an echinoderm sperm cell (Figure 18-17). More recent examples suggest that actin polymerization stimulated by profilin is a common mechanism for generating the force for movement of intracellular pathogens and for cell locomotion.

Figure 18-17. The acrosome reaction in echinoderm sperm.

Figure 18-17

The acrosome reaction in echinoderm sperm. In an unactivated sperm, a membrane-bounded acrosomal vesicle (A) lies within an indentation of the nucleus (N). Adjacent to the vesicle is the periacrosomal region (P), which contains unpolymerized profilin-actin (more...)

Intracellular Bacterial and Viral Movements

Most infections are spread by bacteria or viruses that are liberated when an infected cell lyses. However, some bacteria and viruses escape from a cell on the end of a polymerizing actin filament. Examples include Listeria monocytogenes, a bacterium that can be transmitted from a pregnant woman to the fetus, and vaccinia, a virus related to the smallpox virus. When such organisms infect mammalian cells, they move through the cytosol at rates approaching 11 μm/min. Fluorescent-microscopy experiments showed that a meshwork of short actin filaments follows a moving bacterium or virus like the plume of a rocket exhaust (Figure 18-18). These observations suggested that actin generates the force necessary for movement. The first hints about how actin mediates bacterial movement were provided by a microinjection experiment in which fluorescent-labeled G-actin was injected into Listeria-infected cells. In the microscope, the labeled monomers could be seen incorporating into the tail-like meshwork at the end nearest the bacterium, with a simultaneous loss of actin throughout the tail. This result showed that actin polymerizes into filaments at the base of the bacterium and suggested that as the tail-like meshwork assembles, it pushes the bacterium ahead. Studies with mutant bacteria indicate that interaction of cellular profilin with a bacterial membrane protein promotes actin polymerization at the end of the tail nearest the bacterium.

Figure 18-18. Fluorescent micrograph of Listeria in infected fibroblasts.

Figure 18-18

Fluorescent micrograph of Listeria in infected fibroblasts. Bacteria (red) are stained with an antibody specific for a bacterial membrane protein that binds cellular profilin and is essential for infectivity and motility. Behind each bacterium is a “tail” (more...)

Actin Polymerization at the Leading Edge of Moving Cells

Understanding of the role of actin polymerization and profilin in the acrosome reaction and intracellular bacterial movement played a key role in elucidating how cells crawl forward. Cell movement is led by changes in the position of the plasma membrane at the front of the cell (leading edge); video microscopy reveals that a major feature of this movement is the polymerization of actin at the membrane (Figure 18-19). Profilin is thought to play a central role because it is located at the leading edge where polymerization occurs. In addition, actin filaments at the leading edge are rapidly cross-linked into bundles and networks in the projecting filopodia and lamellipodia. As we discuss in later sections, these structures then form stable contacts with the underlying surface and prevent the membrane from retracting.

Figure 18-19. Assembly of actin filaments at the leading edge of migrating cells.

Figure 18-19

Assembly of actin filaments at the leading edge of migrating cells. (a) As shown in this diagram of a fibroblast, profilin is located at the leading edge of the cell, the site where actin filaments are assembled. (b) In a microinjection experiment, the (more...)

SUMMARY

  •  Within cells, the actin cytoskeleton is dynamic, with filaments able to grow and shrink rapidly.
  •  Polymerization of G-actin in vitro is marked by a lag period during which nucleation occurs. Eventually, a polymerization reaction reaches a steady state in which the rates of addition and loss of subunits are equal (see Figure 18-11).
  •  The concentration of actin monomers in equilibrium with actin filaments is the critical concentration (Cc ). At a G-actin concentration above Cc, there is net growth of filaments; at concentrations below Cc, there is net depolymerization of filaments.
  •  Actin filaments grow considerably faster at their (+) end than at their (−) end, and the Cc for monomer addition to the (+) end is lower than for addition at the (−) end.
  •  The assembly, length, and stability of actin filaments are controlled by specialized actin-binding proteins. These proteins are in turn regulated by various mechanisms.
  •  The complementary actions of thymosin β4 and profilin are critical to regulating the actin cytoskeleton near the cell membrane (see Figure 18-15).
  •  The regulated polymerization of actin can generate membrane projections and underlies the movement of certain bacteria and viruses within cells.
Image permission
Image permission

By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.

Copyright © 2000, W. H. Freeman and Company.
Bookshelf ID: NBK21594