Chapter SEVENTEEN
17
Cytoskeleton
The ability of eucaryotic cells to adopt a variety of shapes, organize
the many components in their interior, interact mechanically with the
environment, and carry out coordinated movements depends on the
cytoskeleton—an intricate network of protein filaments that extends
throughout the cytoplasm (Figure 17–1). This filamentous architecture
helps to support the large volume of cytoplasm in a eucaryotic cell, a
function that is particularly important in animal cells, which have no cell
walls. Although some cytoskeletal components are present in bacteria,
the cytoskeleton is most prominent in the large and structurally complex
eucaryotic cell.
IntermedIate FIlaments
mIcrotubules
actIn FIlaments
muscle contractIon
Unlike our own bony skeleton, however, the cytoskeleton is a highly
dynamic structure that is continuously reorganized as a cell changes
shape, divides, and responds to its environment. The cytoskeleton is not
only the “bones” of a cell but its “muscles” too, and it is directly responsible for large-scale movements such as the crawling of cells along a
surface, contraction of muscle cells, and the changes in cell shape that
take place as an embryo develops. Without the cytoskeleton, wounds
would never heal, muscles would be useless, and sperm would never
reach the egg.
The eucaryotic cell, like any factory making a complex product, has a
highly organized interior in which specialized machines are concentrated
in different areas but linked by transport systems (discussed in Chapter
15). The cytoskeleton controls the location of the organelles that conduct
these specialized functions, in addition to providing the machinery for
the transport between them. It is also responsible for the segregation of
chromosomes into daughter cells and the pinching apart of cells at cell
division, as we discuss in Chapter 18.
10 mm
Figure 17–1 The cytoskeleton gives a cell
its shape and allows the cell to organize
its internal components. an animal cell
in culture has been labeled to show two
of its major cytoskeletal systems, the
microtubules (green) and the actin filaments
(red). the DNa in the nucleus is labeled in
blue. (Courtesy of albert tousson.)
572
Chapter 17
Cytoskeleton
The cytoskeleton is built on a framework of three types of protein filaments: intermediate filaments, microtubules, and actin filaments. As shown
in Panel 17–1 (p. 573), each type of filament has distinct mechanical
properties and is formed from a different protein subunit. A family of
fibrous proteins form the intermediate filaments; tubulin is the subunit
in microtubules; and actin is the subunit in actin filaments. In each case,
thousands of subunits assemble into a fine thread of protein that sometimes extends across the entire cell.
Figure 17–2 Intermediate filaments form a
strong, durable network in the cytoplasm
of the cell. (a) Immunofluorescence
micrograph of a sheet of epithelial cells in
culture stained to show the lacelike network
of intermediate keratin filaments (green)
that surround the nuclei and extend through
the cytoplasm of the cells. the filaments
in each cell are indirectly connected to
those of neighboring cells through the
desmosomes (discussed in Chapter 20),
establishing a continuous mechanical link
from cell to cell throughout the tissue. this
mechanical link strengthens the epithelium,
allowing its cells to form a continuous
sheet lining the tissue cavity. a second
protein (blue) has been stained to show
the locations of the cell boundaries.
(B) Drawing from an electron micrograph
of a section of epidermis showing the
bundles of intermediate filaments that
traverse the cytoplasm and are inserted
at desmosomes. (a, courtesy of Kathleen
Green and evangeline amargo; B, from
r.V. Krstić, Ultrastructure of the Mammalian
Cell: an atlas. Berlin: Springer, 1979. With
permission from Springer-Verlag.)
In this chapter we consider the structure and function of the three types
of protein filament networks in turn. We begin with the intermediate filaments that provide cells with mechanical strength. We then see how cell
appendages built from microtubules propel motile cells like protozoa and
sperm, and how the actin cytoskeleton provides the motive force for a
crawling fibroblast. Finally, we discuss how the cytoskeleton drives one
of the most obvious and best-studied forms of cell movement, the contraction of muscle.
IntermedIate FIlaments
Intermediate filaments have great tensile strength, and their main function is to enable cells to withstand the mechanical stress that occurs when
cells are stretched. The filaments are called “intermediate” because, in
the smooth muscle cells where they were first discovered, their diameter (about 10 nm) is between that of the thin actin-containing filaments
and the thicker myosin filaments. Intermediate filaments are the toughest
and most durable of the three types of cytoskeletal filaments: when cells
are treated with concentrated salt solutions and nonionic detergents, the
intermediate filaments survive while most of the rest of the cytoskeleton
is destroyed.
Intermediate filaments are found in the cytoplasm of most animal cells.
They typically form a network throughout the cytoplasm, surrounding
the nucleus and extending out to the cell periphery. There they are often
anchored to the plasma membrane at cell–cell junctions such as desmosomes (discussed in Chapter 20), where the external face of the membrane
is connected to that of another cell (Figure 17–2). Intermediate filaments
intermediate
filaments
desmosome
connecting
two cells
(A)
10 mm
(B)
5 mm
paNel 17–1
The three major types of protein filaments
INTERMEDIATE FILAMENTS
ii
iii
i
100 nm
iv
25 nm
Intermediate filaments are ropelike fibers with a diameter of around
10 nm; they are made of intermediate filament proteins, which constitute a
large and heterogeneous family. One type of intermediate filament forms a
meshwork called the nuclear lamina just beneath the inner nuclear
membrane. Other types extend across the cytoplasm, giving cells mechanical
strength. In an epithelial tissue, they span the cytoplasm from one cell–cell
junction to another, thereby strengthening the entire epithelium.
Micrographs courtesy of Roy Quinlan (i); Nancy L. Kedersha (ii); Mary Osborn (iii); Ueli Aebi (iv).
MICROTUBULES
ii
i
100 nm
iii
iv
25 nm
Microtubules are long, hollow cylinders made of the protein tubulin.
With an outer diameter of 25 nm, they are much more rigid than actin
filaments (below). Microtubules are long and straight and typically have
one end attached to a single microtubule-organizing center (MTOC)
called a centrosome.
Micrographs courtesy of Richard Wade (i); D.T. Woodrum and R.W. Linck (ii); David Shima (iii); Arshad Desai (iv).
ACTIN FILAMENTS
ii
iii
i
100 nm
iv
25 nm
Actin filaments (also known as microfilaments) are two-stranded helical
polymers of the protein actin. They appear as flexible structures, with a
diameter of 5–9 nm, and they are organized into a variety of linear bundles,
two-dimensional networks, and three-dimensional gels. Although actin
filaments are dispersed throughout the cell, they are most highly
concentrated in the cortex, just beneath the plasma membrane.
Micrographs courtesy of Roger Craig (i and iv); P.T. Matsudaira and D.R. Burgess, Cold Spring Harb. Symp. Quant. Biol. 46:845–854, 1982. With permission from Cold Spring Harbor
Laboratory Press (ii); Keith Burridge (iii).
573
574
Chapter 17
Cytoskeleton
are also found within the nucleus; a mesh of intermediate filaments, the
nuclear lamina, underlies and strengthens the nuclear envelope in all
eucaryotic cells. In this section, we see how the structure and assembly
of intermediate filaments makes them particularly suited to strengthening cells and protecting them from mechanical stress.
Intermediate Filaments are strong and ropelike
Figure 17–3 Intermediate filaments are
like ropes made of long, twisted strands
of protein. the intermediate filament
protein monomer shown in (a) consists of
a central rod domain with globular regions
at either end. pairs of monomers associate
to form a dimer (B), and two dimers then
line up to form a staggered tetramer (C).
tetramers can pack together end-to-end
as shown in (D) and assemble into a helical
array containing eight strands of tetramers
(shown here spread out flat for clarity) that
twist together to form the final ropelike
intermediate filament (e). (F) electron
micrograph of the final 10-nm filament.
(F, courtesy of roy Quinlan.)
Intermediate filaments are like ropes with many long strands twisted
together to provide tensile strength (movie 17.1). The strands of this
rope—the subunits of intermediate filaments—are elongated fibrous
proteins, each composed of an N-terminal globular head, a C-terminal
globular tail, and a central elongated rod domain (Figure 17–3a). The
rod domain consists of an extended a-helical region that enables pairs of
intermediate filament proteins to form stable dimers by wrapping around
each other in a coiled-coil configuration (Figure 17–3B), as described in
Chapter 4. Two of these coiled-coil dimers then associate by noncovalent bonding to form a tetramer (Figure 17–3C), and the tetramers then
bind to one another end-to-end and side-by-side, and also by noncovalent bonding, to generate the final ropelike intermediate filament (Figure
17–3D–F).
COOH
NH2
(A)
a-helical region in monomer
NH2
COOH
(B)
coiled-coil dimer
NH2
COOH
48 nm
NH2
COOH
COOH
NH2
(C)
(F)
0.1 mm
NH2
(D)
COOH
COOH
staggered tetramer of two coiled-coil dimers
NH2
two tetramers packed together
(E)
eight tetramers twisted into a ropelike filament
10 nm
Intermediate Filaments
575
stretching a sheet
of cells without
intermediate filaments
stretching a sheet
of cells with
intermediate filaments
CELLS REMAIN INTACT AND TOGETHER
CELLS RUPTURE
The central rod domains of different intermediate filament proteins are all
similar in size and amino acid sequence, so that when they pack together
ECB3 e17.05/17.04
they always form filaments of similar diameter and internal structure. By
contrast, the globular head and tail regions, which are exposed on the
surface of the filament, allow it to interact with other components of the
cytoplasm. The globular domains vary greatly in both size and amino
acid sequence from one intermediate filament protein to another.
Intermediate Filaments strengthen cells against
mechanical stress
Intermediate filaments are particularly prominent in the cytoplasm of cells
that are subject to mechanical stress. They are present in large numbers,
for example, along the length of nerve cell axons, providing essential
internal reinforcement to these extremely long and fine cell extensions.
They are also abundant in muscle cells and in epithelial cells such as
those of the skin. In all these cells, intermediate filaments, by stretching and distributing the effect of locally applied forces, keep cells and
their membranes from breaking in response to mechanical shear (Figure
17–4). A similar principle is used to manufacture composite materials
such as fiberglass or reinforced concrete, in which tension-bearing linear
elements such as carbon fibers (in fiberglass) or steel bars (in concrete)
are embedded in a space-filling matrix to give the material strength.
Figure 17–4 Intermediate filaments
strengthen animal cells. If a sheet of
epithelial cells is stretched by external
forces (due to the growth or movements of
the surrounding tissues, for example), then
the network of intermediate filaments and
desmosomal junctions that extends through
the sheet develops tension and limits the
extent of stretching. If the junctions alone
were present, then the same forces would
cause a major deformation of the cells,
even to the extent of causing their plasma
membranes to rupture.
Intermediate filaments can be grouped into four classes: (1) keratin filaments in epithelial cells; (2) vimentin and vimentin-related filaments in
connective-tissue cells, muscle cells, and supporting cells of the nervous
system (glial cells); (3) neurofilaments in nerve cells; and (4) nuclear lamins, which strengthen the nuclear membrane of all animal cells (Figure
17–5). The first three filament types are found in the cytoplasm, the fourth
INTERMEDIATE FILAMENTS
CYTOPLASMIC
keratins
in epithelia
vimentin and
vimentin-related
in connective
tissue, muscle
cells, and
glial cells
NUCLEAR
neurofilaments
nuclear lamins
in nerve cells
in all
animal cells
Figure 17–5 Intermediate filaments can be
divided into several different categories.
576
Chapter 17
Cytoskeleton
Figure 17–6 Plectin aids in the
bundling of intermediate filaments
and links these filaments to other
cytoskeletal protein networks. plectin
(green) links intermediate filaments
(blue) to other intermediate filaments,
to microtubules (red), and to actin
filaments (not shown). In this electron
micrograph, the yellow dots are gold
particles linked to antibodies that
recognize plectin. the actin filament
network has been removed to reveal
these protein linkages. (From
t.M. Svitkina and G.G. Borisy,
J. Cell Biol. 135:991–1007, 1996. With
permission from the rockefeller
University press.)
0.5 mm
in the cell nucleus. Filaments of each class are formed by polymerization
of their corresponding protein subunits.
The keratins are the most diverse class of intermediate filament. Every
kind of epithelium in the vertebrate body—whether in the tongue, the
cornea, or the lining of the gut—has its own distinctive mixture of keratin
proteins. Specialized keratins also occur in hair, feathers, and claws. In
each case, the keratin filaments are formed from a mixture of different
keratin subunits. Keratin filaments typically span the interiors of epithelial cells from one side of the cell to the other, and filaments in adjacent epithelial cells are indirectlyECB3
connected
through cell–cell junctions
e17.07/17.06
called desmosomes (see Panel 17–1, p. 573). The ends of the keratin filaments are anchored to the desmosomes, and they associate laterally
with other cell components through their globular head and tail domains,
which project from the surface of the assembled filament. This cabling
of high tensile strength, formed by the filaments throughout the epithelial sheet, distributes the stress that occurs when the skin is stretched.
The importance of this function is illustrated by the rare human genetic
disease epidermolysis bullosa simplex, in which mutations in the keratin
genes interfere with the formation of keratin filaments in the epidermis.
As a result, the skin is highly vulnerable to mechanical injury, and even a
gentle pressure can rupture its cells, causing the skin to blister.
Question 17–1
Which of the following types of cells
would you expect to contain a high
density of intermediate filaments
in their cytoplasm? explain your
answers.
a. Amoeba proteus (a free-living
amoeba)
b. skin epithelial cell
c. smooth
mooth muscle cell in the
digestive tract
d. Escherichia coli
e. nerve
erve cell in the spinal cord
F. sperm cell
G. Plant cell
?
Many of the intermediate filaments are further stabilized and reinforced
by accessory proteins, such as plectin, that cross-link the filament bundles
into strong arrays. In addition to holding together bundles of intermediate
filaments (particularly vimentin), these proteins link intermediate filaments to microtubules, to actin filaments, and to adhesive structures in
the desmosomes (Figure 17–6). Mutations in the gene for plectin cause a
devastating human disease that combines features of epidermolysis bullosa simplex (caused by disruption of skin keratin), muscular dystrophy
(caused by disruption of intermediate filaments in muscle), and neurodegeneration (caused by disruption of neurofilaments). Mice lacking a
functional plectin gene die within a few days of birth, with blistered skin
and abnormal skeletal and heart muscles. Thus, although plectin may not
be necessary for the initial formation of intermediate filaments, its crosslinking action is required to provide cells with the strength they need to
withstand the mechanical stresses inherent to vertebrate life.
the nuclear envelope Is supported by a meshwork of
Intermediate Filaments
Whereas cytoplasmic intermediate filaments form ropelike structures,
the intermediate filaments lining and strengthening the inside surface of
the inner nuclear membrane are organized as a two-dimensional mesh
(Figure 17–7). The intermediate filaments within this tough nuclear lamina are constructed from a class of intermediate filament proteins called
Microtubules
577
CYTOSOL
nuclear envelope
nuclear
pore
nuclear
lamina
NUCLEUS
chromatin
(A)
lamins (not to be confused with laminin, which is an extracellular matrix
protein). In contrast to the very stable cytoplasmic intermediate filaments
found in many cells, the intermediate filaments of the nuclear lamina
disassemble and re-form at each cell division, when the nuclear envelope breaks down during mitosis and then re-forms in each daughter cell
(discussed in Chapter 18).
Disassembly and reassembly of the nuclear lamina are controlled by
ECB3
E17.08/17.07
the phosphorylation and dephosphorylation
(discussed
in Chapter 4) of
the lamins by protein kinases. When the lamins are phosphorylated, the
consequent conformational change weakens the binding between the
tetramers and causes the filament to fall apart. Dephosphorylation at the
end of mitosis causes the lamins to reassemble (see Figure 18–31).
Defects in a particular nuclear lamin are associated with certain types of
progeria—rare disorders that cause affected individuals to appear to age
prematurely. Children with progeria have wrinkled skin, lose their teeth
and hair, and often develop severe cardiovascular disease by the time
they reach their teens. Although researchers do not yet know how loss of
the nuclear lamins leads to these symptoms, some have suggested that
the resulting nuclear instability could lead to impaired cell division or a
diminished capacity for tissue repair.
mIcrotubules
Microtubules have a crucial organizing role in all eucaryotic cells. They
are long and relatively stiff hollow tubes of protein that can rapidly disassemble in one location and reassemble in another. In a typical animal
cell, microtubules grow out from a small structure near the center of the
cell called the centrosome (Figure 17–8a). Extending out toward the cell
periphery, they create a system of tracks within the cell, along which vesicles, organelles, and other cell components are moved. These and other
systems of cytoplasmic microtubules are the part of the cytoskeleton
mainly responsible for anchoring membrane-enclosed organelles within
the cell and for guiding intracellular transport.
When a cell enters mitosis, the cytoplasmic microtubules disassemble
and then reassemble into an intricate structure called the mitotic spindle.
As described in Chapter 18, the mitotic spindle provides the machinery
that will segregate the chromosomes equally into the two daughter cells
just before a cell divides (Figure 17–8B). Microtubules can also form permanent structures, as exemplified by the rhythmically beating hairlike
structures called cilia and flagella (Figure 17–8C). These extend from the
Figure 17–8 Microtubules usually grow out of an organizing
structure. Unlike intermediate filaments, microtubules (dark green)
extend from an organizing center such as (a) a centrosome, (B) a
spindle pole, or (C) the basal body of a cilium.
(B)
1 mm
Figure 17–7 Intermediate filaments support
and strengthen the nuclear envelope.
(a) Schematic cross section through the
nuclear envelope. the intermediate filaments
of the nuclear lamina line the inner face of the
nuclear envelope and are thought to provide
attachment sites for the DNa-containing
chromatin. (B) electron micrograph of a
portion of the nuclear lamina from a frog egg,
or oocyte. the lamina is formed from a square
lattice of intermediate filaments composed of
lamins. (Nuclear laminae from other cell types
are not always as regularly organized as the
one shown here.) (B, courtesy of Ueli aebi.)
(A)
INTERPHASE CELL
centrosome
(B)
DIVIDING CELL
spindle poles
of mitotic spindle
(C)
CILIATED CELL
cilium
basal body
578
Chapter 17
Cytoskeleton
surface of many eucaryotic cells, which use them either as a means of
propulsion or to sweep fluid over the cell surface. The core of a eucaryotic
cilium or flagellum consists of a highly organized and stable bundle of
microtubules. (Bacterial flagella have an entirely different structure and
act as propulsive structures by a different mechanism.)
In this section, we first look at the structure and assembly of microtubules and then discuss their role in organizing the cytoplasm. Their
organizing function depends on the association of microtubules with
accessory proteins, especially the motor proteins that propel organelles
along cytoskeletal tracks. Finally, we discuss the structure and function of
cilia and flagella, in which microtubules are permanently associated with
motor proteins that power ciliary beating.
microtubules are Hollow tubes with structurally distinct
ends
Microtubules are built from subunits—molecules of tubulin—each of
which is itself a dimer composed of two very similar globular proteins
called a-tubulin and b-tubulin, bound tightly together by noncovalent
bonding. The tubulin dimers stack together, again by noncovalent bonding, to form the wall of the hollow cylindrical microtubule. This tubelike
structure is made of 13 parallel protofilaments, each a linear chain of tubulin dimers with a- and b-tubulin alternating along its length (Figure 17–9).
Each protofilament has a structural polarity, with a-tubulin exposed at
one end and b-tubulin at the other, and this polarity—the directional
arrow embodied in the structure—is the same for all the protofilaments,
giving a structural polarity to the microtubule as a whole. One end of the
microtubule, thought to be the b-tubulin end, is called its plus end, and
the other, the a-tubulin end, its minus end.
tubulin heterodimer
(= microtubule subunit)
(B)
Figure 17–9 Microtubules are hollow
tubes of tubulin. (a) One tubulin molecule
(an ab dimer) and one protofilament are
shown schematically, together with their
location in the microtubule wall. Note that
the tubulin molecules are all arranged in the
protofilaments with the same orientation, so
that the microtubule has a definite structural
polarity. (B and C) Schematic diagrams
of a microtubule, showing how tubulin
molecules pack together in the microtubule
wall. at the top, the 13 molecules are shown
in cross section. Below this, a side view
of a short section of a microtubule shows
how the tubulin molecules are aligned in
linear protofilaments. (D) Cross section of
a microtubule with its ring of 13 distinct
subunits, each of which corresponds to
a separate tubulin dimer. (e) Microtubule
viewed lengthwise in an electron
microscope. (D, courtesy of richard linck;
e, courtesy of richard Wade.)
(D)
10 nm
lumen
protofilament
plus
end
50 nm
minus
end
(A)
(C)
microtubule
(E)
50 nm
Microtubules
579
In vitro, in a concentrated solution of pure tubulin, tubulin dimers will add
to either end of a growing microtubule, although they add more rapidly to
the plus end than the minus end (which is why the ends were originally
named this way). The polarity of the microtubule—the fact that its structure has a definite direction, with the two ends being chemically different
and behaving differently—is crucial, both for the assembly of microtubules and for their role once they are formed. If they had no polarity,
they could not serve their function in defining a direction for intracellular
transport, for example.
the centrosome Is the major microtubule-organizing
center in animal cells
Microtubules in cells are formed by outgrowth from specialized organizing centers, which control the number of microtubules formed, their
location and their orientation in the cytoplasm. In animal cells, for example, the centrosome, which is typically close to the cell nucleus when the
cell is not in mitosis, organizes the array of microtubules that radiates
outward from it through the cytoplasm (see Figure 17–8A). Centrosomes
contain hundreds of ring-shaped structures formed from another type of
tubulin, g-tubulin, and each g-tubulin ring serves as the starting point, or
nucleation site, for the growth of one microtubule (Figure 17–10a). The
ab-tubulin dimers add to the g-tubulin ring in a specific orientation, with
the result that the minus end of each microtubule is embedded in the
centrosome and growth occurs only at the plus end—that is, at the outward-facing end (Figure 17–10B).
In addition to its g-tubulin rings, the centrosome in most animal cells also
contains a pair of centrioles, curious structures each made of a cylindrical
array of short microtubules. The centrioles have no role in the nucleation
of microtubules in the centrosome (the g-tubulin rings alone are sufficient), and their function remains something of a mystery, especially as
plant cells lack them. Centrioles are, however, similar, if not identical, to
the basal bodies that form the organizing centers for the microtubules in
cilia and flagella (see Figure 17–8C), as discussed later in this chapter.
Microtubules need nucleating sites such as those provided by the g-tubulin rings in the centrosome because it is much harder to start a new
microtubule from scratch, by first assembling a ring of ab-tubulin dimers,
than to add such dimers to a preexisting microtubule structure. Purified
free ab-tubulin can polymerize spontaneously in vitro when at a high
concentration, but in the living cell, the concentration of free ab-tubulin
Question 17–2
Figure 17–10 Tubulin polymerizes from
nucleation sites on a centrosome.
(a) Schematic drawing showing that a
centrosome consists of an amorphous
matrix of protein containing the g-tubulin
rings that nucleate microtubule growth.
In animal cells, the centrosome contains
a pair of centrioles, each made up of a
cylindrical array of short microtubules.
(B) a centrosome with attached
microtubules. the minus end of each
microtubule is embedded in the
centrosome, having grown from a
nucleating ring, whereas the plus end of
each microtubule is free in the cytoplasm.
(C) a reconstructed image shows a dense
thicket of microtubules emanating from the
centrosome of a C. elegans cell. (C, from
e.t. O’toole et al., J. Cell Biol. 163:451–456,
2003. With permission from the rockefeller
University press.)
+
+
+
nucleating sites
(g-tubulin ring complexes)
+
+
+
+
+
+
centrosome
matrix
+
+
+
+
+
+
+
+
pair of
centrioles
+
+
+
+
+
+
(A)
(B)
+
+ microtubules growing from
g-tubulin ring complexes
of the centrosome
?
Why do you suppose it is much
easier to add tubulin to existing
microtubules than to start a new
microtubule from scratch? explain
explain
how g-tubulin
-tubulin in the centrosome
helps to overcome this hurdle.
(C)
580
Chapter 17
Cytoskeleton
is too low to drive the difficult first step of assembling the initial ring of a
new microtubule. By providing organizing centers containing nucleation
sites, and keeping the concentration of free ab-tubulin dimers low, cells
can thus control where microtubules form.
Growing microtubules show dynamic Instability
Figure 17–11 Each microtubule filament
grows and shrinks independently of
its neighbors. the array of microtubules
anchored in a centrosome is continually
changing as new microtubules grow
(red arrows) and old microtubules shrink
(blue arrows).
ECB3 E17.12/17.11
tubulin molecule
with bound GTP
GTP tubulin molecules
add to end of microtubule
addition proceeds faster
than GTP hydrolysis
GTP cap
(A)
GROWING MICROTUBULE
protofilaments containing GDP
tubulin peel away from the
microtubule wall
GDP tubulin is released
to the cytosol
tubulin molecule
with bound GDP
(B)
SHRINKING MICROTUBULE
Once a microtubule has been nucleated, its plus end typically grows outward from the organizing center by the addition of ab-tubulin subunits
for many minutes. Then, without warning, the microtubule suddenly
undergoes a transition that causes it to shrink rapidly inward by losing
subunits from its free end (movie 17.2). It may shrink partially and then,
no less suddenly, start growing again, or it may disappear completely, to
be replaced by a new microtubule from the same g-tubulin ring (Figure
17–11).
This remarkable behavior, known as dynamic instability, stems from the
intrinsic capacity of tubulin molecules to hydrolyze GTP. Each free tubulin
dimer contains one tightly bound GTP molecule that is hydrolyzed to GDP
(still tightly bound) shortly after the subunit is added to a growing microtubule. The GTP-associated tubulin molecules pack efficiently together
in the wall of the microtubule, whereas tubulin molecules carrying GDP
have a different conformation and bind less strongly to each other.
When polymerization is proceeding rapidly, tubulin molecules add to the
end of the microtubule faster than the GTP they carry is hydrolyzed. The
end of a growing microtubule is therefore composed entirely of GTPtubulin subunits, forming what is known as a GTP cap. In this situation,
the growing microtubule will continue to grow (Figure 17–12a). Because
of the randomness of chemical processes, however, it will occasionally
happen that tubulin at the free end of the microtubule hydrolyzes its GTP
before the next tubulin has been added, so that the free ends of protofilaments are now composed of GDP-tubulin subunits. This change tips
the balance in favor of disassembly (Figure 17–12B). Because the rest of
the microtubule is composed of GDP-tubulin, once depolymerization has
started, it will tend to continue, often at a catastrophic rate; the microtubule starts to shrink rapidly and continuously, and may even disappear.
The GDP-containing tubulin molecules that are freed as the microtubule
depolymerizes join the unpolymerized tubulin molecules already in the
cytosol. In a typical fibroblast, for example, at any one time about half of
the tubulin in the cell is in microtubules, while the remainder is free in
the cytosol, forming a pool of subunits available for microtubule growth.
This situation is quite unlike the arrangement with the more stable intermediate filaments, where the subunits are typically almost completely
in the fully assembled form. The tubulin molecules joining the pool then
exchange their bound GDP for GTP, thereby becoming competent again
to add to another microtubule that is in a growth phase.
Figure 17–12 GTP hydrolysis controls the growth of microtubules.
(a) tubulin dimers carrying Gtp (red) bind more tightly to one
another than do tubulin dimers carrying GDp (dark green). therefore,
microtubules that have freshly added tubulin dimers at their end with
Gtp bound tend to keep growing. (B) From time to time, however,
especially when microtubule growth is slow, the subunits in this Gtp
cap will hydrolyze their Gtp to GDp before fresh subunits loaded
with Gtp have time to bind. the Gtp cap is thereby lost; the GDpcarrying subunits are less tightly bound in the polymer and are readily
released from the free end, so that the microtubule begins to shrink
continuously (movie 17.3 and movie 17.4).
Microtubules
581
microtubules are maintained by a balance of assembly
and disassembly
The relative instability of microtubules allows them to undergo rapid
remodeling, and this is crucial for microtubule function. In a normal cell,
the centrosome (or other organizing center) is continually shooting out
new microtubules in an exploratory fashion in different directions and
retracting them. A microtubule growing out from the centrosome can,
however, be prevented from disassembling if its plus end is somehow
permanently stabilized by attachment to another molecule or cell structure so as to prevent tubulin depolymerization. If stabilized by attachment
to a structure in a more distant region of the cell, the microtubule will
establish a relatively stable link between that structure and the centrosome (Figure 17–13). The centrosome can be compared to a fisherman
casting a line: if there is no bite at the end of the line, the line is quickly
withdrawn and a new cast is made; but if a fish bites, the line remains in
place, tethering the fish to the fisherman. This simple strategy of random
exploration and selective stabilization enables the centrosome and other
nucleating centers to set up a highly organized system of microtubules
linking selected parts of the cell. It is also used to position organelles
relative to one another.
Drugs that prevent the polymerization or depolymerization of tubulin can
have a rapid and profound effect on the organization of the cytoskeleton—
and the behavior of the cell. Consider the mitotic spindle, the microtubule
framework that guides the chromosomes during mitosis (see Figure
17–8B). If a cell in mitosis is exposed to the drug colchicine, which binds
tightly to free tubulin and prevents its polymerization into microtubules,
the mitotic spindle rapidly disappears and the cell stalls in the middle of
mitosis, unable to partition its chromosomes into two groups. This shows
that the mitotic spindle is normally maintained by a continuous balanced
addition and loss of tubulin subunits: when tubulin addition is blocked by
colchicine, tubulin loss continues until the spindle disappears.
The drug taxol has the opposite action at the molecular level. It binds
tightly to microtubules and prevents them from losing subunits. Because
new subunits can still be added, the microtubules can grow but cannot
shrink. However, despite the differences in molecular detail, taxol has
the same overall effect on the cell as colchicine: it also arrests dividing
cells in mitosis. We learn from this that for the spindle to function, microtubules must be able not only to assemble but also to disassemble. The
behavior of the spindle is discussed in more detail in Chapter 18, when
we consider mitosis.
The inactivation or destruction of the mitotic spindle eventually kills
dividing cells. Cancer cells, which are dividing with less control than
most other cells of the body, can sometimes be killed preferentially by
microtubule-stabilizing and microtubule-destabilizing antimitotic drugs.
Thus, drugs that interfere with microtubule polymerization or depolym-
nucleus
(A)
centrosome
microtubule
(B)
capping
protein
(C)
unstable
microtubules
(D)
stable
microtubules
Figure 17–13 The selective stabilization of
microtubules can polarize a cell.
a newly formed microtubule will persist
only if both its ends are protected from
depolymerization. In cells, the minus ends of
microtubules are generally protected by the
organizing centers from which the filaments
grow. the plus ends are initially free but can
be stabilized by other proteins. here, for
example, a nonpolarized cell is depicted
in (a) with new microtubules growing
from and shrinking back to a centrosome
in many directions randomly. Some of
these microtubules happen by chance to
encounter proteins (capping proteins) in a
specific region of the cell cortex that can
bind to and stabilize the free plus ends of
microtubules (B). this selective stabilization
will lead to a rapid reorientation of the
microtubule arrays (C) and convert the cell
to a strongly polarized form (D).
582
Chapter 17
Cytoskeleton
TablE 17–1 DruGS ThaT affECT fIlaMENTS aND MICroTubulES
action
microtubule-specific drugs
taxol
binds and stabilizes microtubules
Colchicine, colcemid
binds subunits and prevents polymerization
Vinblastine, vincristine
binds subunits and prevents polymerization
actin-specific drugs
phalloidin
binds and stabilizes filaments
Cytochalasin
caps filament plus ends
latrunculin
binds subunits and prevents polymerization
erization, including colchicine, taxol, vincristine, and vinblastine, are
used in the clinical treatment of cancer. As we discuss shortly, there also
exist compounds that stabilize and destabilize actin filaments. Together,
these drugs, listed in table 17–1, allow biologists to study the function of
the cytoskeleton.
microtubules organize the Interior of the cell
Cells are able to modify the dynamic instability of their microtubules for
particular purposes. As cells enter mitosis, for example, microtubules
become initially more dynamic, switching between growing and shrinking much more frequently than cytoplasmic microtubules normally do.
This enables them to disassemble rapidly and then reassemble into the
mitotic spindle. On the other hand, when a cell has differentiated into a
specialized cell type and taken on a definite fixed structure, the dynamic
instability of its microtubules is often suppressed by proteins that bind to
the ends of microtubules or along their length and stabilize them against
disassembly. The stabilized microtubules then serve to maintain the
organization of the cell.
Figure 17–14 Microtubules transport
cargo along a nerve cell axon. In nerve
cells, all the microtubules in the axon
point in the same direction, with their
plus ends toward the axon terminal. the
oriented microtubules serve as tracks
for the directional transport of materials
synthesized in the cell body but required
at the axon terminal (such as membrane
proteins required for growth). For an axon
passing from your spinal cord to a muscle in
your shoulder, say, the journey takes about
two days. In addition to this outward traffic
of material (red circles) driven by one set of
motor proteins, there is inward traffic (blue
circles) in the reverse direction driven by
another set of motor proteins. the inward
traffic carries materials ingested by the tip of
the axon or produced by the breakdown of
proteins and other molecules back toward
the cell body.
Most differentiated animal cells are polarized; that is, one end of the cell
is structurally or functionally different from the other. Nerve cells, for
example, put out an axon from one end of the cell and dendrites from the
other; cells specialized for secretion have their Golgi apparatus positioned
toward the site of secretion, and so on. The cell’s polarity is a reflection of
the polarized systems of microtubules in its interior, which help to position organelles in their required location within the cell and to guide the
streams of traffic moving between one part of the cell and another. In the
nerve cell, for example, all the microtubules in the axon point in the same
direction, with their plus ends toward the axon terminal (Figure 17–14).
Along these oriented tracks the cell is able to send cargoes of materials,
such as membrane vesicles and proteins for secretion, that are made in
the cell body but required far away at the end of the axon.
cell body
axon
terminal
microtubule
axon
+
+
–
–
inward
transport
outward
transport
Microtubules
Some of these materials move at speeds in excess of 10 cm a day, which
means that they may still take a week or more to travel to the end of a long
axon in larger animals. But movement along microtubules is immeasurably faster and more efficient than free diffusion. A protein molecule
traveling by free diffusion would take years to reach the end of a long
axon—if it arrived at all (see Question 17–12).
But the microtubules in living cells do not act alone. Their activity, like
those of other cytoskeletal filaments, depends on a large variety of accessory proteins that bind to them. Some microtubule-associated proteins
stabilize microtubules against disassembly, for example, while others
link microtubules to other cell components, including the other types
of cytoskeletal filaments. Yet other microtubule-associated proteins are
motor proteins that carry organelles, vesicles and other cellular materials
along the microtubules. Because the components of the cytoskeleton can
interact with each other, their functions can be coordinated.
motor Proteins drive Intracellular transport
If a living cell is observed in a light microscope, its cytoplasm is seen
to be in continual motion (Figure 17–15). Mitochondria and the smaller
membrane-enclosed organelles and vesicles move in small, jerky steps—
that is, they move for a short period, stop, and then start again. This
saltatory movement is much more sustained and directional than the continual, small Brownian movements caused by random thermal motions.
Both microtubules and actin filaments are involved in saltatory and other
directed intracellular movements in eucaryotic cells. In both cases the
movements are generated by motor proteins, which use the energy
derived from repeated cycles of ATP hydrolysis to travel steadily along the
actin filament or the microtubule in a single direction (see Figure 4–42).
At the same time, these motor proteins also attach to other cell components and thus transport this cargo along the filaments. Dozens of motor
proteins have been identified. They differ in the type of filament they bind
to, the direction in which they move along the filament, and the cargo
they carry.
583
Question 17–3
dynamic instability causes
microtubules either to grow or to
shrink rapidly. consider an individual
microtubule that is in its shrinking
phase.
a. What must happen at the end
of the microtubule in order for it to
stop shrinking and to start growing?
b. How would a change in the
tubulin concentration affect this
switch?
c. What would happen if only Gd
GdP,
G
dP,
P,
but no GtP,
P, were present in the
solution?
d. What would happen if the
GtP
solution contained an analog of Gt
G
tP
P
that cannot be hydrolyzed?
?
The motor proteins that move along cytoplasmic microtubules, such as
those in the axon of a nerve cell, belong to two families: the kinesins
generally move toward the plus end of a microtubule (away from the centrosome; outward from the cell body in Figure 17–14), while the dyneins
move toward the minus end (toward the centrosome; inward in Figure
17–14). These kinesins and dyneins are both dimers with two globular
ATP-binding heads and a single tail (Figure 17–16a). The heads interact
5 mm
Figure 17–15 organelles move along microtubules at different speeds. In this series of video-enhanced images of a flattened area of
an invertebrate nerve cell, numerous membrane vesicles and mitochondria are present, many of which can be seen to move. the white
circle provides a fixed frame of reference. these images were recorded at intervals of 400 milliseconds. (Courtesy of p. Forscher.)
584
Chapter 17
Cytoskeleton
Figure 17–16 Motor proteins move along
microtubules using their globular heads.
(a) Kinesins and cytoplasmic dyneins are
microtubule motor proteins that generally
move in opposite directions along a
microtubule. each of these proteins (drawn
here to scale) is a dimer composed of
two identical molecules. each protein
has two globular heads that interact with
microtubules at one end and a single tail at
the other. (B) Schematic diagram of a motor
protein showing atp-dependent “walking”
along a filament (see also Figure 4–42).
tail
dynein
globular
head
kinesin
microtubule
plus end
minus end
10 nm
(A)
(B)
with microtubules in a stereospecific manner, so that the motor protein
will attach to a microtubule in only one direction. The tail of a motor
protein generally binds stably to some cell component, such as a vesicle
or an organelle, and thereby determines the type of cargo that the motor
protein can transport (Figure 17–17). The globular heads of kinesin and
dynein are enzymes with ATP-hydrolyzing (ATPase) activity. This reaction provides the energy for a cycle of conformational changes in the
head that enable it to move along the microtubule by a cycle of binding,
release, and rebinding
to the microtubule (see Figure 17–16B and Figure
ECB3 e17.17/17.16
4–42). For a discussion of the discovery and study of motor proteins, see
How We Know, pp. 586–588.
organelles move along microtubules
Microtubules and motor proteins play an important part in positioning
membrane-enclosed organelles within a eucaryotic cell. In most animal cells, for example, the tubules of the endoplasmic reticulum reach
almost to the edge of the cell (movie 17.5). The Golgi apparatus, in contrast, is located in the interior of the cell near the centrosome (Figure
17–18). Both the endoplasmic reticulum and the Golgi apparatus depend
on microtubules for their alignment and positioning. The membranes of
the endoplasmic reticulum extend out from their points of connection
with the nuclear envelope (see Figure 1–22), aligning with microtubules
that extend from the centrosome out to the plasma membrane. As the
cell develops and the endoplasmic reticulum grows, kinesins attached to
the outside of the endoplasmic reticulum membrane (via receptor proteins) pull it outward along microtubules, stretching it like a net. Dyneins,
Figure 17–17 Different motor proteins
transport cargo along microtubules. Most
kinesins move toward the plus end of a
microtubule, whereas dyneins move toward
the minus end (movie 17.6). Both types of
microtubule motor proteins exist in many
forms, each of which is thought to transport
a different cargo. the tail of the motor
protein determines what cargo the protein
transports.
cargo
tail
motor head
minus
end
KINESINS
plus
end
microtubule
motor head
DYNEINS
tail
cargo
Microtubules
(A)
(B)
10 mm
(C)
Figure 17–18 Microtubules help to arrange the organelles in a eucaryotic cell. (a) Schematic
diagram of a cell showing the typical arrangement of microtubules (dark green), endoplasmic
reticulum (blue), and Golgi apparatus (yellow). the nucleus is shown in brown, and the centrosome
in light green. (B) Cell stained with antibodies to endoplasmic reticulum (upper panel) and to
microtubules (lower panel). Motor proteins pull the endoplasmic reticulum out along microtubules.
(C) Cell stained with antibodies to the Golgi apparatus (upper panel) and to microtubules (lower
panel). In this case, motor proteins move the Golgi apparatus inward to its position near the
centrosome. (B, courtesy of Mark terasaki, lan Bo Chen, and Keigi Fujiwara; C, courtesy of Viki allan
and thomas Kreis.)
similarly attached to the Golgi membranes, pull the Golgi apparatus the
other way along microtubules, inward toward the cell center. In this way
the regional differences in internal membranes, on which the successful
function of the cell depends, are created and maintained.
ECB3 E17.23/17.22
When cells are treated with a drug such as colchicine
that causes microtubules to disassemble, both of these organelles change their location
dramatically. The endoplasmic reticulum, which has connections to the
nuclear envelope, collapses to the center of the cell, while the Golgi
apparatus, which is not attached to any other organelle, fragments into
small vesicles, which disperse throughout the cytoplasm. When the drug
is removed, the organelles return to their original positions, dragged by
motor proteins moving along the re-formed microtubules.
cilia and Flagella contain stable microtubules moved by
dynein
Earlier in this chapter we mentioned that many microtubules in cells are
stabilized through their association with other proteins, and therefore no
longer show dynamic instability. Stable microtubules are employed by
cells as stiff supports on which to construct a variety of polarized structures, including the remarkable cilia and flagella that allow eucaryotic
cells to move water over their surface. Cilia are hairlike structures about
0.25 mm in diameter, covered by plasma membrane, that extend from
the surface of many kinds of eucaryotic cells (see Figure 17–8C). A single
cilium contains a core of stable microtubules, arranged in a bundle, that
grow from a basal body in the cytoplasm; the basal body serves as the
organizing center for the cilium.
585
586
hOW We KNOW:
PurSuING MoTor ProTEINS
The movement of organelles throughout the cell cytoplasm has been observed, measured, and speculated
about since the middle of the nineteenth century. But
it was not until the mid-1980s that biologists identified
the molecules that drive this movement of organelles
and vesicles from one part of the cell to another.
Why the lag between observation and understanding?
The problem was in the proteins—or, more precisely,
in the difficulty of studying them in isolation outside the
cell. To investigate the activity of an enzyme, for example, biochemists first purify the polypeptide: they break
open cells or tissues and separate the protein of interest
from other molecular components (see Panels 4–4 and
4–5, pp. 164–167). They can then study the protein on
its own, in vitro, controlling its exposure to substrates,
inhibitors, ATP, and so on. Unfortunately, this approach
did not seem to work for studies of the motile machinery
that underlies intracellular transport. It is not possible
to break open a cell and pull out a fully active transport
system, free of extraneous material, that continues to
carry mitochondria and vesicles from place to place.
The techniques needed to move the research forward
came from two different sources. First, advances in
microscopy allowed biologists to see that an operational transport system (with extraneous material still
attached) could be squeezed from the right kind of living cell. At the same time, biochemists realized that
they could assemble a working transport system from
scratch—using purified cables, motors, and cargo—outside the cell. The breakthrough started with a squid.
Teeming cytoplasm
As we saw in Chapter 12, neuroscientists interested in
the electrical properties of nerve cell membranes have
long studied the giant axon from squid (see How We
Know, pp. 412–413). Because of its large size, researchers found that they could squeeze the cytoplasm from
the axon like toothpaste, and then study how ions move
back and forth through various channels in the empty,
tubelike membrane. The physiologists simply discarded
the cytoplasmic jelly, as it appeared to be inert (and thus
uninteresting) when examined under a standard light
microscope.
Then along came video-enhanced microscopy. This
type of microscopy, developed by Shinya Inoué, Robert
Allen, and others, allows one to detect structures that
are smaller than the resolution power of standard light
microscopes, about 0.2 mm, or 200 nm (see Panel 1–1,
pp. 8–9). Sample images are captured by a video camera
and then enhanced by computer processing to reduce
the background and heighten contrast. When researchers in the early 1980s applied this new technique to
preparations of squid axon cytoplasm (axoplasm), they
~1 mm
squid giant
axon
extruded axoplasm
5 mm
Figure 17–19 Video-enhanced microscopy of cytoplasm
squeezed from a squid giant axon reveals the motion of
organelles. In this micrograph numerous cytoskeletal filaments
are visible, along with transported particles including a
mitochondrion (red arrow) and smaller vesicles (blue arrows).
(From r.D. Vale et al., Cell 40:449–454, 1985. With permission
from elsevier.)
ECB3 e17.19/17.18
observed, for the first time, the motion of vesicles and
membrane-enclosed organelles along cytoskeletal
filaments.
Under the video microscope, extruded axoplasm is
seen to be teeming with tiny particles—from vesicles
30–50 nm in diameter to mitochondria some 5000 nm
long, all moving to and fro along cytoskeletal filaments
at speeds of up to 5 mm per second. If the axoplasm is
spread thinly enough, individual filaments can be seen
(Figure 17–19).
The movement continues for hours, allowing researchers to manipulate the preparation and study the effects.
Ray Lasek and Scott Brady discovered, for example, that
organelle movement requires ATP. Substitution of ATP
analogs, such as AMP-PNP, which bind to the enzyme
active site but cannot be hydrolyzed (and thus provide
no energy), inhibit the translocation.
pursuing Motor proteins
587
Figure 17–20 a motor protein
causes microtubule gliding.
In an active in vitro motility
assay, purified kinesin is mixed
with microtubules in a buffer
containing atp. When a drop of
the mixture is placed on a glass
slide and examined by videoenhanced microscopy, individual
microtubules can be seen gliding
over the slide driven by kinesin
molecules. Images recorded at
1 second intervals. (Courtesy of
Nick Carter and rob Cross.)
1 mm
Snaking tubes
More work was needed to identify the individual components that drive the transport system in squid axons.
What are the filaments made of? What are the molecular
machines that shuttle the vesicles and organelles along
these filaments? Identifying the cables was relatively
easy. Studies using antibodies to a-tubulin revealed
that the filaments are microtubules. But what about the
motor proteins? To find these, Ron Vale, Thomas Reese,
and Michael Sheetz set up a system in which they could
fish for proteins that power organelle movement.
Their strategy was simple yet elegant: add together
purified cables and purified cargo and then look for
molecules that induce motion. They took purified microtubules from squid optic lobe, added organelles isolated
from squid axons, and showed that movement could be
triggered by the addition of an extract from squid axon
cytoplasm. In this preparation, the researchers could
watch organelles travel along the microtubules, and
watch microtubules glide snakelike over the surface of
a glass coverslip (see Question 17–18). Their challenge
was to isolate the protein responsible for movement in
this reconstituted system.
To do that, Vale and his colleagues took advantage of
Lasek and Brady’s earlier work with the ATP analog
AMP-PNP. Although this analog inhibits the movement
of vesicles along microtubules, it still allows these
components to attach to the microtubule filaments. So
the researchers incubated the cytoplasm extract with
microtubules in the presence of AMP-PNP; they then
pulled out the microtubules with what they hoped were
the motor proteins still attached. Vale and his team then
added ATP to release the attached proteins, and they
found a 110-kilodalton polypeptide that could bind to,
and initiate movement of, microtubules in vitro (Figure
17–20). They dubbed the molecule kinesin (from the
Greek kinein, “to move”).
Such in vitro motility assays have been instrumental in the study of motor proteins and their activities.
Subsequent studies showed that kinesin moves along
microtubules from the minus end to the plus end,
and also identified many other kinesin-related motor
proteins.
lights, camera,
action
ECB3 e17.20/17.19
Combining these in vitro assays with ever more refined
microscopic techniques, researchers can now monitor the movement of individual motor proteins along
single microtubules, even in living cells. In an assay
developed by Steven Block and his colleagues in 1990,
microscopic silica beads coated with low concentrations of kinesin (so that only one molecule of kinesin is
present on each bead) can be monitored as they make
their way down a microtubule (Figure 17–21). Other
observations of single kinesin molecules are made possible by coupling the motor protein with a fluorescent
marker protein such as GFP.
Such single-molecule studies have revealed that
kinesin moves along microtubules processively—that
is, each molecule takes 100 or so “steps” along the filament before falling off (Figure 17–22). The length of
each step is 8 nm, which corresponds to the spacing of
tubulin dimers along the microtubule. Combining these
observations with assays of ATP hydrolysis, researchers
have confirmed that one molecule of ATP is consumed
per step. Kinesin can move in a processive man-
588
pursuing Motor proteins
silica bead
+
microtubule
kinesin
0.1 mm
(A)
(B)
1 mm
Figure 17–21 Video microscopy can be used to track the movement of a single kinesin molecule. (a) In
this assay, silica beads are coated with kinesin molecules at a concentration such that each bead, on average,
will have only one kinesin molecule attached to it. Kinesin is then allowed to walk along a microtubule, and its
movement is monitored by tracking the movement of the bead. (B) In this series of images, the bead is captured
by a laser-based optical tweezer, placed on a microtubule filament, and then allowed to move. thirty seconds
elapses between each frame. (From S. Block et al., Nature 348:348–352, 1990. With permission from Macmillan
publishers ltd.)
they can watch not only single molecules of kinesin,
ner because it has two heads (see Figure 17–22). The
but also each individual head as it moves, in relation
motor is thought to walk its way toward the plus end
to its partner, along the microtubule. The results will
of the microtubule in a hand-over-hand fashion, each
ECB3Further
e17.21/17.20yield additional insights into the molecular movements
head binding and releasing the filament in turn.
that underlie the organization and activity of eucaryotic
studies are required to refine this model, and researchcells.
ers are now working to improve their methods so that
kinesin
heads
4
3
(A)
10 nm
2
1
2
1
3
(B)
1 mm
(C)
microtubule
16 nm
Figure 17–22 a single molecule of kinesin
moves along a microtubule. (a) electron
micrograph of a single kinesin molecule showing
the two head domains (red arrows). (B) three
frames, separated by intervals of 1 second,
record the movement of an individual kinesinGFp molecule (green) along a microtubule (red)
at a speed of 0.3 mm/sec. (C) Series of molecular
models of the two heads of a kinesin molecule,
showing how they are thought to processively
walk their way along a microtubule in a series of
8-nm steps (movie 17.7). (a, courtesy of John
heuser; B and C, courtesy of ron Vale.)
Microtubules
589
Figure 17–23 hairlike cilia coat the
surface of many eucaryotic cells. Scanning
electron micrograph of the ciliated
epithelium on the surface of the human
respiratory tract. the thick tufts of cilia
on the ciliated cells are interspersed with
the dome-shaped surfaces of nonciliated
epithelial cells. (reproduced from
r.G. Kessel and r.h. Karden, tissues and
Organs. San Francisco: W.h. Freeman
& Co., 1979. With permission from
W.h. Freeman & Co.)
5 mm
Cilia move fluid over the surface of a cell or propel single cells through a
fluid. Some protozoa, for example, use cilia to collect food particles, and
others use them for locomotion. On the epithelial cells lining the human
respiratory tract (Figure 17–23), huge numbers of cilia (more than a billion per square centimeter) sweep layers of mucus containing trapped
dust particles and dead cells up toward the throat, to be swallowed and
eventually eliminated from the body. Similarly, cilia on the cells of the
oviduct wall create a current that helps
move eggs along the oviduct.
ECB3toE17.24/17.23
Each cilium acts as a small oar, moving in a repeated cycle that generates
the current that washes over the cell surface (Figure 17–24).
The flagella (singular flagellum) that propel sperm and many protozoa
are much like cilia in their internal structure but are usually very much
longer. They are designed to move the entire cell, and instead of generating a current, they propagate regular waves along their length that drive
the cell through liquid (Figure 17–25).
The microtubules in cilia and flagella are slightly different from the cytoplasmic microtubules; they are arranged in a curious and distinctive
pattern that was one of the most striking revelations of early electron
microscopy. A cross section through a cilium shows nine doublet microtubules arranged in a ring around a pair of single microtubules (Figure
17–26a). This “9 + 2” array is characteristic of almost all forms of eucaryotic cilia and flagella, from those of protozoa to those found in humans.
The movement of a cilium or a flagellum is produced by the bending of
its core as the microtubules slide against each other. The microtubules
are associated with numerous proteins (Figure 17–26B), which project at
regular positions along the length of the microtubule bundle. Some serve
as cross-links to hold the bundle of microtubules together; others generate the force that causes the cilium to bend.
The most important of these accessory proteins is the motor protein ciliary dynein, which generates the bending motion of the core. It closely
resembles cytoplasmic dynein and functions in much the same way.
Ciliary dynein is attached by its tail to one microtubule, while its heads
interact with an adjacent microtubule to generate a sliding force between
the two filaments. Because of the multiple links that hold the adjacent
microtubule doublets together, what would be a simple parallel sliding
movement between free microtubules is converted to a bending motion
power stroke
Figure 17–24 a cilium beats by
performing a repetitive cycle of
movements consisting of a power stroke
followed by a recovery stroke. In the fast
power stroke, the cilium is fully extended
and fluid is driven over the surface of the
cell; in the slower recovery stroke, the
cilium curls back into position with minimal
disturbance to the surrounding fluid. each
ECB3requires
e7.25/17.24
cycle typically
0.1–0.2 second and
generates a force perpendicular to the axis
of the cilium.
590
Chapter 17
Cytoskeleton
Figures 17–25 flagella propel a cell using a repetitive wavelike
motion. the wavelike motion of a single flagellum on a tunicate sperm
is seen in a series of images captured by stroboscopic illumination at
400 flashes per second. (Courtesy of Charles J. Brokaw.)
in the cilium (Figure 17–27). In humans, hereditary defects in ciliary
dynein cause Kartagener’s syndrome. Men with this disorder are infertile because their sperm are non-motile, and all those affected have an
increased susceptibility to bronchial infections because the cilia that line
their respiratory tract are paralyzed and thus unable to clear bacteria and
debris from the lungs.
actIn FIlaments
Actin filaments are found in all eucaryotic cells and are essential for
many of their movements, especially those involving the cell surface.
Without actin filaments, for example, an animal cell could not crawl
along a surface, engulf a large particle by phagocytosis, or divide in two.
Like microtubules, many actin filaments are unstable, but by associating
with other proteins they can also form stable structures in cells, such as
the contractile apparatus of muscle. Actin filaments interact with a large
number of actin-binding proteins that enable the filaments to serve a variety of functions in cells. Depending on their association with different
proteins, actin filaments can form stiff and relatively permanent structures, such as the microvilli on the brush-border cells lining the intestine
(Figure 17–28a) or small contractile bundles in the cytoplasm that can
contract and act like the “muscles” of a cell (Figure 17–28B); they can also
form temporary structures, such as the dynamic protrusions formed at
the leading edge of a crawling fibroblast (Figure 17–28C) or the contractile ring that pinches the cytoplasm in two when an animal cell divides
(Figure 17–28D). In this section, we see how the arrangements of actin
filaments in a cell depend on the types of actin-binding proteins present.
outer dynein arm
radial spoke
inner sheath
nexin
central singlet
microtubule
ECB3 e17.26/17.25
plasma membrane
inner dynein arm
(A)
100 nm
(B)
A microtubule
B microtubule
outer doublet microtubule
Figure 17–26 Microtubules in a cilium or flagellum are arranged in a “9 + 2” array. (a) electron micrograph of a flagellum of
Chlamydomonas shown in cross section, illustrating the distinctive 9 + 2 arrangement of microtubules. (B) Diagram of the flagellum
in cross section. the nine outer microtubules (each a special paired structure) carry two rows of dynein molecules. the heads of
these dyneins appear in this view like pairs of arms reaching toward the adjacent microtubule. In a living cilium, these dynein heads
periodically make contact with the adjacent microtubule and move along it, thereby producing the force for ciliary beating. Various
other links and projections shown are proteins that serve to hold the bundle of microtubules together and to convert the sliding motion
produced by dyneins into bending, as illustrated in Figure 17–27. (a, courtesy of lewis tilney.)
ECB3 e17.27/17.26
actin Filaments
+
+
+
+
+
+
linking
proteins
+
+ATP
+
bend
591
Figure 17–27 The movement of dynein
causes the flagellum to bend.
(a) If the outer doublet microtubules
and their associated dynein molecules
are freed from other components of a
sperm flagellum and then exposed to
atp, the doublets slide against each other,
telescope-fashion, due to the repetitive
action of their associated dyneins. (B) In an
intact flagellum, however, the doublets are
tied to each other by flexible protein links
so that the action of the system produces
bending rather than sliding.
–
–
–
–
–
–
–
–
(A)
IN ISOLATED DOUBLET
MICROTUBULES: DYNEIN
PRODUCES
MICROTUBULE SLIDING
(B)
IN NORMAL
FLAGELLUM: DYNEIN
CAUSES MICROTUBULE
BENDING
Even though actin filaments and microtubules are formed from unrelated
types of proteins, we shall see that the principles according to which they
assemble and disassemble, control cell structure, and bring about movement are strikingly similar.
actin Filaments are thin and Flexible
Actin filaments appear in electron micrographs as threads about 7 nm
in diameter. Each filament is a twisted chain of identical globular actin
molecules, all of which “point” in the same direction along the axis of the
chain. Like a microtubule, therefore, an actin filament has a structural
ECB3 m16.83/17.27
polarity, with a plus end and a minus end (Figure 17–29).
Question 17–4
dynein arms in a cilium are arranged
so that, when activated, the heads
push their neighboring outer
doublet outward toward the tip of
the cilium. consider a cross section
of a cilium (see Figure 17–26). Why
would no bending motion of the
cilium result if all dynein molecules
were active at the same time?
What pattern of dynein activity can
account for the bending of a cilium
in one direction?
?
Actin filaments are thinner, more flexible, and usually shorter than microtubules. There are, however, many more of them, so that the total length
of all the actin filaments in a cell is generally many times greater than
the total length of all of the microtubules. Actin filaments rarely occur in
isolation in the cell; they are generally found in cross-linked bundles and
networks, which are much stronger than the individual filaments.
actin and tubulin Polymerize by similar mechanisms
Actin filaments can grow by the addition of actin monomers at either end,
but the rate of growth is faster at the plus end than at the minus end. A
naked actin filament, like a microtubule without associated proteins, is
inherently unstable, and it can disassemble from both ends. Each free
actin monomer carries a tightly bound nucleoside triphosphate, in this
case ATP, which is hydrolyzed to ADP soon after the incorporation of
(A)
(B)
(C)
(D)
Figure 17–28 actin filaments allow
eucaryotic cells to adopt a variety
of shapes and perform a variety of
functions. Various actin-containing
structures are shown here in red:
(a) microvilli; (B) contractile bundles in the
cytoplasm; (C) sheetlike (lamellipodia) and
fingerlike (filopodia) protrusions from the
leading edge of a moving cell;
(D) contractile ring during cell division.
592
Chapter 17
Cytoskeleton
Figure 17–29 actin filaments
are thin, flexible protein threads.
(a) electron micrographs of negatively
stained actin filaments. (B) arrangement
of actin molecules in an actin filament.
each filament may be thought of as a twostranded helix with a twist repeating every
37 nm. Strong interactions between the two
strands prevent the strands from separating.
(C) the identical subunits of an actin
filament are depicted in different colors to
emphasize the close interaction between
each actin molecule and its four nearest
neighbors. (a, courtesy of roger Craig;
C, from K.C. holmes et al., Nature
347:44–49, 1990. With permission from
Macmillan publishers ltd.)
actin molecule
minus end
minus end
37 nm
plus end
plus end
(A)
50 nm
(B)
(C)
the actin monomer into the filament. As with the GTP bound to tubulin,
hydrolysis of ATP to ADP in an actin filament reduces the strength of
binding between monomers and decreases the stability of the polymer.
Nucleotide hydrolysis thereby promotes depolymerization, helping the
cell to disassemble filaments after they have formed (Figure 17–30).
Question 17–5
the formation of actin filaments in
the cytosol is controlled by actinome actin-binding
binding proteins. some
proteins significantly increase the
rate at which the formation of an
actin filament is initiated. suggest
uggest a
mechanism by which they
might do this.
?
As with microtubules, the ability to assemble and disassemble is required
for many of the functions performed by actin filaments, such as their role
in cell locomotion. Actin filament function can be perturbed experimentally by certain toxins produced by fungi or marine sea sponges. Some,
such as the cytochalasins, prevent
actin polymerization; others, such as
ECB3 E17.30/17.29
phalloidin, stabilize actin filaments against depolymerization (see Table
17–1, p. 582). Addition of these toxins to the medium bathing cells or
tissues, even in low concentrations, instantaneously freezes cell movements such as the crawling motion of a fibroblast. Thus, the function of
actin filaments depends on a dynamic equilibrium between the actin filaments and the pool of actin monomers. Filaments often persist for only a
few minutes after they are formed.
many Proteins bind to actin and modify Its Properties
About 5% of the total protein in a typical animal cell is actin; about half
of this actin is assembled into filaments, and the other half remains as
actin monomers in the cytosol. The concentration of monomer is thereADP
Figure 17–30 aTP hydrolysis decreases
the stability of the actin polymer. actin
monomers in the cytosol carry atp, which
is hydrolyzed to aDp soon after assembly
into a growing filament. the aDp molecules
remain trapped within the actin filament,
unable to exchange with atp until the actin
monomer that carries them dissociates from
the filament.
ATP
actin with
bound ATP
actin with
bound ADP
Pi
actin Filaments
593
actin monomers
monomersequestering
protein
nucleating protein
severing protein
actin filaments
bundling protein
(in filopodia)
motor protein
cross-linking
protein (in cell cortex)
side-binding
protein
capping (end-blocking) protein
fore high—much higher than the concentration required for purified actin
monomers to polymerize in vitro. What, then, keeps the actin monomers
in cells from polymerizing totally into filaments? The answer is that cells
contain small proteins, such as thymosin and profilin, that bind to actin
monomers in the cytosol, preventing them from adding to the ends of
actin filaments. By keeping actin monomers in reserve until they are
required, these proteins play a crucial role in regulating actin polymerization. When actin filaments are needed, other actin-binding proteins
promote their assembly. Proteins called formins
actin-related proteins
ECB3 and
E17.32/17.31
(ARPs) both control actin assembly at the advancing front of a migrating
cell.
There are a great many actin-binding proteins in cells. Most of these bind
to assembled actin filaments rather than to actin monomers and control
the behavior of the intact filaments (Figure 17–31). Actin-bundling proteins, for example, hold actin filaments together in parallel bundles in
microvilli; other cross-linking proteins hold actin filaments together in
a gel-like meshwork within the cell cortex—the layer of cytoplasm just
beneath the plasma membrane; filament-severing proteins, such as gelsolin, fragment actin filaments into shorter lengths and thus can convert
an actin gel to a more fluid state. Actin filaments can also associate with
motor proteins to form contractile bundles, as in muscle cells. And they
often form tracks along which motor proteins transport organelles, a
function that is especially conspicuous in plant cells.
Figure 17–31 actin-binding proteins
control the behavior of actin filaments
in vertebrate cells. actin is shown in red,
and the actin-binding proteins are shown in
green.
594
Chapter 17
Cytoskeleton
In the remainder of this chapter, we consider some characteristic structures that actin filaments can form, and discuss how different types of
actin-binding proteins are involved in their formation. We begin with the
actin-rich cell cortex and its role in cell locomotion, and in the final section we consider the contractile apparatus of muscle cells as an example
of a stable structure based on actin filaments.
an actin-rich cortex underlies the Plasma membrane of
most eucaryotic cells
Although actin is found throughout the cytoplasm of a eucaryotic cell,
in most cells it is highly concentrated in a layer just beneath the plasma
membrane. In this region, called the cell cortex, actin filaments are
linked by actin-binding proteins into a meshwork that supports the outer
surface of the cell and gives it mechanical strength. In red blood cells, as
described in Chapter 11, a simple and regular network of fibrous proteins
attached to the plasma membrane provides it with support necessary to
maintain its simple discoid shape (see Figure 11–31). The cell cortex of
other animal cells, however, is thicker and more complex and supports
a far richer repertoire of shapes and movements. Like the cortex in a red
cell, it contains spectrin and ankyrin; however, it also includes a dense
network of actin filaments that project into the cytoplasm, where they
become cross-linked into a three-dimensional meshwork. This cortical
actin mesh governs the shape and mechanical properties of the plasma
membrane and the cell surface: the rearrangement of actin filaments
within the cortex provides the molecular basis for changes in cell shape
and cell locomotion.
cell crawling depends on actin
Many cells move by crawling over surfaces, rather than by swimming
by means of cilia or flagella. Carnivorous amoebae crawl continually,
in search of food. The advancing tip of a developing axon migrates in
response to growth factors, following a path of substrate-bound and diffusible chemicals to its eventual synaptic target. White blood cells known
as neutrophils migrate through tissues when they ‘smell’ small diffusing molecules released by bacteria, which the neutrophils seek out and
destroy. For these immune hunters, chemotactic molecules binding to
receptors on the cell surface trigger changes in actin filament assembly
that drive the cells toward their prey.
The molecular mechanisms of these and other forms of cell crawling
entail coordinated changes of many molecules in different regions of
the cell, and no single, easily identifiable locomotory organelle, such as
a flagellum, is responsible. In broad terms, however, three interrelated
processes are known to be essential: (1) the cell pushes out protrusions
at its “front,” or leading edge; (2) these protrusions adhere to the surface
over which the cell is crawling; and (3) the rest of the cell drags itself forward by traction on these anchorage points (Figure 17–32).
All three processes involve actin, but in different ways. The first step, the
pushing forward of the cell surface, is driven by actin polymerization. The
leading edge of a crawling fibroblast in culture regularly extends thin,
sheetlike lamellipodia, which contain a dense meshwork of actin filaments, oriented so that most of the filaments have their plus ends close
to the plasma membrane (Figure 17–33). Many cells also extend thin, stiff
protrusions called filopodia, both at the leading edge and elsewhere on
their surface. These are about 0.1 mm wide and 5–10 mm long, and each
contains a loose bundle of 10–20 actin filaments, again oriented with
their plus ends pointing outward. The advancing tip (growth cone) of a
actin Filaments
actin cortex
lamellipodium
substratum
actin polymerization at
plus end protrudes
lamellipodium
cortex under tension
PROTRUSION
movement of unpolymerized actin
myosin-II
ATTACHMENT AND
TRACTION
CONTRACTION
595
Figure 17–32 forces generated in the
actin-rich cortex move a cell forward.
In this proposed mechanism for cell
movement, actin polymerization at the
leading edge of the cell pushes the plasma
membrane forward (protrusion) and forms
new regions of actin cortex, shown here in
red. New points of anchorage are made
between the actin filaments and the surface
on which the cell is crawling (attachment).
Contraction at the rear of the cell then
draws the body of the cell forward (traction).
New anchorage points are established at
the front, and old ones are released at the
back as the cell crawls forward. the same
cycle is repeated over and over again,
moving the cell forward in a stepwise
fashion.
focal contacts
(contain integrins)
PROTRUSION
developing nerve cell axon extends even longer filopodia, up to 50 mm
long, which help it to probe its environment and find the correct path to
its target. Both lamellipodia and filopodia are exploratory, motile structures that form and retract with great speed, moving at around 1 mm per
second. Both are thought to be generated by the rapid local growth of
actin filaments, which assemble
close to the plasma membrane and elonECB3 m16.86/17.32
gate by the addition of actin monomers at their plus ends. In this way the
filaments push out the membrane without tearing it.
(C)
(A)
contractile bundle
(B)
lamellipodium
filopodium
5 m
Figure 17–33 actin filaments
allow animal cells to migrate.
(a) Schematic drawing of a fibroblast
showing flattened lamellipodia and fine
filopodia projecting from its surface,
especially in the regions of the leading
edge. (B) Details of the arrangement of actin
filaments in three regions of the fibroblast
are shown, with arrowheads pointing
toward the plus end of the filaments.
(C) Scanning electron micrograph showing
lamellipodia and filopodia at the leading
edge of a human fibroblast migrating in
culture. an arrow shows the direction of cell
movement. (C, courtesy of Julian heath.)
596
Chapter 17
Cytoskeleton
newborn
actin filament
cell leading edge
(A)
Figure 17–34 a web of actin filaments
pushes the leading edge of a
lamellipodium forward. (a) highly motile
keratocytes from frog skin were fixed, dried,
and shadowed with platinum, and examined
in an electron microscope. actin filaments
form a dense network, with the fast-growing
ends of the filaments terminating at the
leading margin of the lamellipodium (top of
figure). (B) Nucleation of new actin filaments
(red) is mediated by arp complexes
(orange) attached to the sides of preexisting
filaments. the resulting branching structure
pushes the plasma membrane forward. the
plus ends of the actin filaments become
protected by capping proteins (blue), while
the minus ends of actin filaments nearer the
center of the cell continually disassemble
through the action of depolymerizing
proteins (not shown). the web of actin as
a whole thereby undergoes a continual
rearward movement due to the assembly of
filaments at the front and their disassembly
at the rear. (a, courtesy of tatyana Svitkina
and Gary Borisy.)
(B)
0.5 mm
actin
monomer
capping
protein
The formation and growth of actin filaments at the leading edge of a
cell are assisted by various actin-binding accessory proteins. One set
of proteins—the actin-related proteins, or ARPs—promotes the formation of a web of branched actin filaments in lamellipodia. These proteins
form complexes that bind to existing actin filaments and nucleate the
formation of new filaments, which grow out at an angle to produce side
branches (Figure 17–34). With the aid of additional actin-binding proteins,
this web undergoes continual assembly at the leading edge and disassembly further back, pushing the lamellipodia forward. The other kind of
cell protrusion, the filopodium, depends on formins, which attach to the
growing ends of actin filaments and promote the addition of new monomers to form straight unbranched filaments (Figure 17–35). Formins are
also used elsewhere to assemble unbranched filaments, as in the cleavage furrow of a dividing animal cell.
When the lamellipodia and filopodia touch down on a favorable patch of
surface, they stick: transmembrane proteins in their plasma membrane,
known as integrins, adhere to molecules in the extracellular matrix that
surrounds cells or on the surface of a neighboring cell over which the
moving cell is crawling. Meanwhile, on the intracellular face of the crawling cell’s membrane, integrins capture actin filaments, thereby creating a
robust anchorage for the system of actin filaments inside the crawling cell
(see Figure 20–14C). To use this anchorage to drag its body forward, the
cell now makes use of internal contractions to exert a pulling force (see
Figure 17–32). These too depend on actin, but in a different way—through
the interaction of actin filaments with motor proteins known as myosins.
plus end
Figure 17–35 formins help drive the
elongation of actin filaments. Formin
dimers (green) attach to the growing end of
an actin filament (red). each formin subunit
binds to one actin monomer. the formin
dimer promotes filament growth by holding
onto one of the two actin subunits exposed
at the plus end and pulling in a new actin
monomer.
ARP
complex
formin
dimer
actin
filament
minus end
actin Filaments
597
It is still not certain how this pulling force is produced: contraction of
bundles of actin filaments in the cytoplasm or contraction of the actin
meshwork in the cell cortex, or both, may be responsible. The general
principles of how myosin motor proteins interact with actin filaments to
cause movement is clear, however, as we now discuss.
actin associates with myosin to Form contractile
structures
All actin-dependent motor proteins belong to the myosin family. They
bind to and hydrolyze ATP, which provides the energy for their movement
along actin filaments from the minus end of the filament toward the plus
end. Myosin, along with actin, was first discovered in skeletal muscle,
and much of what we know about the interaction of these two proteins
was learned from studies of muscle. There are several different types
of myosins in cells, of which the myosin-I and myosin-II subfamilies are
most abundant. Myosin-II is the major myosin found in muscle. Myosin-I
is found in all types of cells, and because it is simpler in structure and
mechanism of action we shall discuss it first.
Myosin-I molecules have only one head domain and a tail (Figure
17–36a). The head domain interacts with actin filaments and has an ATPhydrolyzing motor activity that enables it to move along the filament in
a cycle of binding, detachment, and rebinding (movie 17.9). The tail varies among the different types of myosin-I, and it determines what cell
components will be dragged along by the motor. For example, the tail
may bind to a particular type of membrane vesicle and propel it through
the cell along actin filament tracks (Figure 17–36B), or it may bind to the
plasma membrane and move it relative to cortical actin filaments, thus
pulling the membrane into a different shape (Figure 17–36C).
Question 17–6
suppose that the actin molecules
in a cultured skin cell have been
randomly labeled in such a way
that 1 in 10,000 molecules carries
a fluorescent marker. What would
you expect to see if you examined
the lamellipodium (leading edge)
of this cell through a fluorescence
microscope? assume
ssume that your
microscope is sensitive enough to
detect single fluorescent molecules.
?
extracellular signals control the arrangement of actin
Filaments
We have seen that myosin and other actin-binding proteins can regulate the location, organization, and behavior of actin filaments. But the
activity of these accessory proteins is, in turn, controlled by extracellular
signals, allowing the cell to rearrange its cytoskeleton in response to the
environment.
(A)
myosin-I
70 nm
(B)
+
myosin-l
vesicle
(C)
+
myosin-l
plasma membrane
Figure 17–36 The short tail of a myosin-I
molecule contains sites that bind to
various components of the cell, including
membranes. (a) Myosin-I has a single
globular head and a tail that attaches to
another molecule or organelle in the cell.
this arrangement allows the head domain
to move a vesicle relative to an actin
filament (B), or an actin filament and the
plasma membrane relative to each other
(C). Note that the head group of the myosin
always walks toward the plus end of the
actin filament it contacts.
598
Chapter 17
Cytoskeleton
Figure 17–37 activation of GTP-binding
proteins has a dramatic effect on
the organization of actin filaments in
fibroblasts. In these micrographs, actin
has been stained with fluorescently
labeled phalloidin, a molecule that binds
to actin filaments (see table 17–1,
p. 582). (a) Unstimulated fibroblasts have
actin filaments primarily in the cortex.
(B) Microinjection of an activated form
of rho promotes the rapid assembly of
long, unbranched contractile bundles.
(C) Microinjection of an activated form
of rac, a Gtp-binding protein similar to
rho, causes the formation of an enormous
lamellipodium that extends from the entire
circumference of the cell. (D) Microinjection
of an activated form of Cdc42, another rho
family member, stimulates the protrusion of
many long filopodia at the cell periphery.
(From a. hall, Science 279:509–514, 1998.
With permission from aaaS.)
(A) QUIESCENT CELLS
(B)
Rho ACTIVATION
(C) Rac ACTIVATION
(D) Cdc42 ACTIVATION
20 mm
For the actin cytoskeleton, such structural rearrangements are triggered
by activation of a variety of receptor proteins embedded in the plasma
membrane. All of these signals then seem to converge inside the cell on
ECB3 m16.97/17.37
a group of closely related GTP-binding
proteins called the Rho protein
family. As we saw in Chapter 16, proteins of this kind behave as molecular switches that control cellular processes by cycling between an active,
GTP-bound state and an inactive, GDP-bound state (see Figure 16–14B).
In the case of the cytoskeleton, activation of different members of the Rho
family affects the organization of actin filaments in different ways. For
example, activation of one Rho family member triggers actin polymerization and bundling to form filopodia; activation of another promotes the
formation of sheetlike lamellipodia and membrane ruffles; and activation
of Rho itself drives the bundling of actin filaments with myosin II and the
clustering of integrins that promotes cell crawling (Figure 17–37).
Question 17–7
at the leading edge of a crawling
cell, the plus ends of actin filaments
are located close to the plasma
membrane, and actin monomers
are added at these ends, pushing
the membrane outward to form
lamellipodia or filopodia. What do
you suppose holds the filaments at
their other ends to prevent them
from just being pushed into the
cell’s interior?
?
These dramatic and complex structural changes occur because the GTPbinding proteins, together with the protein kinases and accessory proteins
with which they interact, act like a computational network to control
actin organization and dynamics. This network receives external signals from nutrients, growth factors, and contacts with neighboring cells,
along with ‘inside information’ regarding the cell’s nutritional state, size,
and readiness for division. The Rho network then processes these inputs
and produces signals that shape the actin cytoskeleton—for example, by
activating the formin proteins that promote the formation of filopodia
(see Figure 17–35) or by enhancing the actin-nucleating activities of ARP
complexes at the leading edge of the cell to generate large lamellipodia.
One of the most tightly regulated rearrangements of cytoskeletal elements occurs when actin associates with myosin in muscle fibers in
response to signals from the nervous system. We now discuss how this
molecular interaction generates the rapid, repetitive, forceful movements
characteristic of the contraction of vertebrate muscles.