282 Chapter 13: Intracellular Membrane Traffic

C. Ricin is transported from the ER to the Golgi apparatus to the lysosome,

where it is released into the cytosol.
D. Ricin, by mimicking an unfolded protein, is tagged for transport across
the ER membrane into the cytosol.

Passage 2 (Questions 13–108 to 13–111)

The molecular machinery responsible for vesicle fusion was discovered via bio-
chemical analysis of vesicle transport. These experiments used a virus that hijacks
the host-cell machinery to insert a viral coat protein into the ER membrane. From
there, the viral protein is transported through the Golgi apparatus to the cell sur-
face, where it is packaged into new virus particles that bud off the cell surface. In
the mature virus, the viral protein is exposed to the outside of the virus and forms
part of the viral coat. As the viral protein moves through the Golgi apparatus, it is
modified with N-acetylglucosamine (GlcNAc).
The investigators used the virus to infect a mutant cell line that is incapable of
modifying the viral protein with GlcNAc. The Golgi apparatus was then isolated
from the infected cells. This “donor” Golgi apparatus was mixed with “accep-
tor” Golgi apparatus isolated from uninfected wild-type cells. The investigators
hypothesized that transport of vesicles from the donor to acceptor Golgi appara-
tuses would move the viral protein into the wild-type acceptor Golgi apparatus,
leading to addition of GlcNAc. Thus, vesicle traffic between Golgi stacks could be
detected simply by assaying for the GlcNAc modification on the viral protein.
In addition to purifying donor and acceptor Golgi apparatuses, the investiga-
tors made a preparation of soluble cytosolic proteins by breaking open cells and
removing membrane-bound organelles and other large particles by centrifuga-
tion. When the donor and acceptor Golgi apparatuses were mixed in the absence
of cytosol, no transport of the viral protein could be detected. However, when ATP
and soluble cytosolic proteins were included, the investigators detected trans-
port of viral protein between the Golgi stacks. To identify proteins required for
vesicle transport, the investigators treated the soluble cytoplasmic extract with
N-ethylmaleimide (NEM), which covalently modifies lysine residues, inacti-
vating some proteins. This process inactivated the activity in the extract, which
allowed the investigators to purify a protein from untreated extracts that could be
added back to NEM-treated extract to rescue vesicle transport. This protein was
called NEM-sensitive factor (NSF). NSF was found to be a soluble protein that can
hydrolyze ATP.
NSF was found to bind to the membranes of the Golgi apparatus in the presence
of soluble proteins called “soluble NSF attachment proteins” (SNAPs). Complexes
of NSF and SNAP formed only in the absence of ATP; addition of ATP caused rapid
release of NSF. If Golgi membranes with bound NSF were solubilized by addition
of nonionic detergent, NSF was found to be associated with a large protein com-
plex. To identify proteins that bind to NSF, the investigators attached an antibody
that recognizes NSF to beads. The beads were then used to bind detergent-solu-
bilized NSF—with its associated complex of proteins—from bovine brain extracts.
After washing with buffer to remove unbound proteins, ATP was added to release
the proteins that bind to NSF and SNAP. These proteins were called SNAP recep-
tors (SNAREs). Two of the proteins the investigators identified—syntaxin and
synaptobrevin—had already been found in other studies of neuronal cell exo-
cytosis. SNAREs were hypothesized to be the minimal machinery necessary for
membrane fusion events in the secretory pathway. To test this, a v-SNARE and a
t-SNARE were expressed in bacteria, purified, and reconstituted into lipid vesi-
cles. When vesicles bearing the v-SNARE were mixed with those containing the
t-SNARE, the vesicles fused. NSF and ATP were not required for fusion.
13–108 The investigators hypothesized that the viral protein was transported
between the Golgi stacks inside vesicles. An alternative hypothesis, how-
ever, was that the viral protein was released from one Golgi apparatus
and taken up by the other, without being packaged into vesicles. Which
MCAT STYLE 283

of the following experiments would best distinguish between these two

hypotheses?
A. Add a protease to the system and determine whether the viral protein is
degraded.
B. Determine whether transport still occurs when clathrin is removed from
the extract.
C. Test for association of GlcNAc-modified viral protein with membranes by
centrifugation.
D. Test whether transport between Golgi stacks is blocked by addition of a
detergent.
13–109 To learn more about the function of NSF, the investigators used a cyto-
plasmic extract that lacked NSF activity. To the extract, they added donor
Golgi apparatus, purified NSF, and ATP. They then added NEM to inac-
tivate NSF, followed by acceptor membranes. No vesicle transport was
detected in this situation. What does this experiment tell you about the
function of NSF?
A. NSF is required for packaging viral protein into donor vesicles.
B. NSF is required for formation of donor vesicles from donor Golgi.
C. NSF is required for fusion of donor vesicles to the acceptor Golgi.
D. NSF is required for release of viral protein into the acceptor Golgi.
13–110 Which of the following statements about synaptobrevin and syntaxin are
consistent with what we now know about the key functions of SNAREs?
I. Botulinum toxin proteolytically cleaves synaptobrevin, leading to muscle
paralysis.
II. Synaptobrevin induces NSF to hydrolyze ATP to provide the energy for
membrane fusion.
III. Synaptobrevin is on the vesicle membrane; syntaxin is on the plasma
membrane.
A. I
B. I and II
C. I and III
D. II and III
13–111 Which one of the following statements best explains why NSF and ATP
were required for vesicle fusion in the original reconstituted system for
transport of a viral protein between Golgi stacks, but were not required
for fusion of lipid vesicles containing a v-SNARE and a t-SNARE?
A. Fusion of lipid vesicles requires no extra energy in the presence of high
concentrations of purified SNAREs; however, energy from ATP hydrolysis
is required to fuse vesicles with natural membranes.
B. In the fusion of natural membranes, NSF and ATP are required to remove
a cytosolic inhibitor that binds to SNARE proteins; the fusion of lipid ves-
icles occurs without ATP because the inhibitor is absent.
C. NSF uses ATP hydrolysis to move proteins out of the way, so that natural
vesicles can dock on their target membrane, but ATP hydrolysis is not
needed to dock lipid vesicles containing purified SNAREs.
D. NSF uses ATP to pull apart tight complexes between v-SNAREs and
t-SNAREs to prime them for fusion, a process that is not required when
the SNAREs are already segregated into different lipid vesicles.
High-Current Copper-Brush
Commutated Dynamo.
According to Wikipedia, this large,
belt-driven, high-current dynamo
produced 310 amperes at 7 volts.
Mitochondria are often referred to as the
“power plants” of the cell. The pictured
machine converted mechanical energy
into electrical energy; mitochondria
convert electrical energy into chemical
energy.
Chapter 14 285

CHAPTER

Energy Conversion:
Mitochondria and Chloroplasts 14
THE MITOCHONDRION IN THIS CHAPTER

TERMS TO LEARN THE MITOCHONDRION

chemiosmotic coupling mitochondria
cristae outer mitochondrial membrane THE PROTON PUMPS OF THE
electrochemical gradient oxidative phosphorylation ELECTRON-TRANSPORT CHAIN
inner mitochondrial membrane proton-motive force
intermembrane space respiratory chain ATP PRODUCTION IN
mitochondrial matrix MITOCHONDRIA

CHLOROPLASTS AND
DEFINITIONS
PHOTOSYNTHESIS
Match each definition below with its term from the list above.
THE GENETIC SYSTEMS
14–1 The subcompartment formed between the inner and outer mitochon- OF MITOCHONDRIA AND
drial membranes.
CHLOROPLASTS
14–2 The mitochondrial electron-transport chain, which generates the proton
gradient across the inner mitochondrial membrane that powers ATP syn-
thase, producing most of the cell’s ATP.
14–3 Mechanism by which a pH gradient across a membrane is used to drive
an energy-requiring process, such as ATP production or the rotation of
bacterial flagella.
14–4 Process in bacteria and mitochondria in which ATP formation is driven
by the transfer of electrons from food molecules to molecular oxygen,
with the intermediate generation of a proton gradient across a mem-
brane.
14–5 The result of a combined pH gradient and membrane potential.
14–6 A sievelike membrane surrounding mitochondria that is permeable to all
molecules of 5000 daltons or less.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
14–7 The intermembrane space is chemically equivalent to the cytosol with
respect to small molecules due to the many specialized transport pro-
teins in the mitochondrial outer membrane.
14–8 The most important contribution of the citric acid cycle to energy metab-
olism is the extraction of high-energy electrons during the oxidation of
acetyl CoA to CO2.
14–9 Each respiratory enzyme complex in the electron-transport chain has a
greater affinity for electrons than its predecessors, so that electrons pass
286 Chapter 14: Energy Conversion: Mitochondria and Chloroplasts

sequentially from one complex to another until they are finally trans-
ferred to oxygen, which has the greatest electron affinity of all.

THOUGHT PROBLEMS
14–10 Mitochondria in liver cells appear to move freely in the cytosol, whereas
those in cardiac muscle are immobilized at positions between adjacent
myofibrils. Do you suppose these differences are a trivial consequence
of cell architecture or do they reflect some underlying functional advan-
tage? Explain your answer.
14–11 Electron micrographs show that mitochondria in heart muscle have a
much higher density of cristae than mitochondria in skin cells. Why do
you suppose this should be?
14–12 In the 1860s, Louis Pasteur noticed that when he added O2 to a culture of
yeast growing anaerobically on glucose, the rate of glucose consumption
declined dramatically. Explain the basis for this result, which is known as
the Pasteur effect.
14–13 The citric acid cycle generates NADH and FADH2, which are then used
in the process of oxidative phosphorylation to make ATP. If the citric acid
cycle, which does not use oxygen, and oxidative phosphorylation are
separate processes, as they are, then why is it that the citric acid cycle
stops almost immediately upon removal of O2?
14–14 When dinitrophenol (DNP) is added to mitochondria, the inner mem-
brane becomes permeable to protons. When the drug valinomycin is
added to mitochondria, the inner membrane becomes permeable to K+.
A. How will the electrochemical gradient change in response to DNP?
B. How will it change in response to valinomycin?
14–15 Several coupled transport processes that occur across the inner mito-
chondrial membrane are illustrated in Figure 14–1. For each, decide
whether transport is with the electrochemical gradient, against it, or
unaffected by it. For those transport processes that are affected by the
gradient, identify which component of the gradient (the difference in pH
or the membrane potential) affects transport. MATRIX
PYR–
I
CALCULATIONS
H+
14–16 Heart muscle gets most of the ATP needed to power its continual con-
tractions through oxidative phosphorylation. When oxidizing glucose to ORN+
CO2, heart muscle consumes O2 at a rate of 10 μmol/min per g of tissue, II
in order to replace the ATP used in contraction and give a steady-state CTR
ATP concentration of 5 μmol/g of tissue. At this rate, how many sec-
onds would it take the heart to consume an amount of ATP equal to its HPO42–
steady-state levels? (Complete oxidation of one molecule of glucose to III
CO2 yields 30 ATP, 26 of which are derived by oxidative phosphorylation 2H+
using the 12 pairs of electrons captured in the electron carriers NADH
and FADH2.) CIT3– + H+
IV
MAL3–

Asp–
V
Glu–
Figure 14–1 Five coupled transport processes that occur across the
inner mitochondrial membrane (Problem 14–15). PYR is pyruvate; ORN is
ornithine; CTR is citrulline; CIT is citrate; MAL is malate; Asp is aspartic inner
acid; and Glu is glutamic acid. membrane
THE PROTON PUMPS OF THE ELECTRON-TRANSPORT CHAIN 287

THE PROTON PUMPS OF THE ELECTRON-

TRANSPORT CHAIN
TERMS TO LEARN
cytochrome quinone (Q)
cytochrome c oxidase complex redox pairs
cytochrome c reductase redox potential
iron–sulfur center redox reaction
NADH dehydrogenase complex

DEFINITIONS
Match each definition below with its term from the list above.
14–17 An electron-driven proton pump in the respiratory chain that accepts
electrons from cytochrome c and generates water using molecular oxy-
gen as an electron acceptor.
14–18 Colored, heme-containing protein that transfers electrons during cellu-
lar respiration.
14–19 Electron-transporting group consisting of either two or four iron atoms
bound to an equal number of sulfur atoms.
14–20 A reaction in which one component becomes oxidized and the other
reduced.
14–21 The affinity of a redox pair for electrons, generally measured as the volt-
age difference between an equimolar mixture of the pair and a standard
reference.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
14–22 Most cytochromes have a higher redox potential (higher affinity for elec-
trons) than iron–sulfur centers, which is why the cytochromes tend to
serve as electron carriers near the O2 end of the respiratory chain.
14–23 The three respiratory enzyme complexes in the mitochondrial inner
membrane tend to associate with each other in ways that facilitate the
correct transfer of electrons between appropriate complexes.

THOUGHT PROBLEMS
14–24 Both H+ and Ca2+ are ions that move through the cytosol. Why is the
movement of H+ ions so much faster than that of Ca2+ ions? How do you
suppose the speed of these two ions would be affected by freezing the
solution? Would you expect them to move faster or slower? Explain your
answer.
14–25 Distinguish between a hydrogen atom, a proton, a hydride ion, and a
hydrogen molecule.
14–26 The half reactions for some of the carriers in the respiratory chain are
given in Table 14–1. From their E0 values, what would you guess is their
order in the chain? What would you need to know before you were more
certain of their order?
14–27 The cytochrome c oxidase complex is strongly inhibited by cyanide,
which binds to the Fe3+ form of cytochrome a3. Cyanide kills at very low
concentration because of its effects on the cytochrome c oxidase com-
plex. Carbon monoxide (CO) also inhibits the cytochrome c oxidase
288 Chapter 14: Energy Conversion: Mitochondria and Chloroplasts

TABLE 14–1 Standard redox potentials for electron carriers in the

respiratory chain (Problem 14–26).
Half reaction E0 (V)

ubiquinone + 2 H+ + 2 e– ubiquinol 0.045

cytochrome b (Fe3+) + e– cytochrome b (Fe2+) 0.077

cytochrome c1 (Fe3+) + e– cytochrome c1 (Fe2+) 0.22

cytochrome c (Fe3+) + e– cytochrome c (Fe2+) 0.25

cytochrome a (Fe3+) + e– cytochrome a (Fe2+) 0.29

cytochrome a3 (Fe3+) + e– cytochrome a3 (Fe2+) 0.55

complex by binding to cytochrome a3, but it binds to the Fe2+ form. CO

kills only at much higher doses than cyanide, not because of its binding
to the cytochrome c oxidase complex, but because it binds to the heme
group of hemoglobin, which also carries an Fe2+. In a sense, by mopping
up CO, hemoglobin protects cytochrome c oxidase from being inhibited
by CO at low concentrations. One treatment for cyanide poisoning—if
administered quickly enough—is to give sodium nitrite, which oxidizes
Fe2+ to Fe3+. How do you suppose sodium nitrite protects against the
effects of cyanide?
14–28 The two different diffusible electron carriers, ubiquinone and cytochrome
c, shuttle electrons between the three protein complexes of the electron-
transport chain. In principle, could the same diffusible carrier be used
for both steps? If not, why not? If it could, what characteristics would it
need to possess and what would be the disadvantages of such a situa-
tion?
14–29 If you were to impose an artificially large electrochemical gradient across
the mitochondrial inner membrane, would you expect electrons to flow
up the respiratory chain, in the reverse of their normal direction? Why or
why not?
14–30 Some bacteria have become specialized to live in an alkaline environ-
ment at pH 10. They maintain their internal environment at pH 7. Why
is it that they cannot exploit the pH difference across their membrane to
get ATP for free using a standard ATP synthase? Can you suggest an engi-
neering modification to ATP synthase that would allow it to generate ATP
from proton flow in such an environment?

CALCULATIONS
14–31 One of the problems in understanding redox reactions is coming to grips
with the language. Consider the reduction of pyruvate by NADH:
pyruvate + NADH + H+ lactate + NAD+
In redox reactions, oxidation and reduction necessarily occur together;
however, it is convenient to list the two halves of a redox reaction sepa-
rately. By convention, each half reaction is written as a reduction: oxidant
+ e– reductant. For the reduction of pyruvate by NADH the half reac-
tions are
pyruvate + 2 H+ + 2e– lactate E0 = –0.19 V
NAD+ + H+ + 2e– NADH E0 = –0.32 V
THE PROTON PUMPS OF THE ELECTRON-TRANSPORT CHAIN 289

where E0 is the standard redox potential and refers to a reaction occur-

ring under standard conditions (25°C or 298 K, all concentrations at 1
M, and pH 7). To obtain the overall equation for reduction of pyruvate
by NADH, it is necessary to reverse the NAD+/NADH half reaction and
change the sign of E0 :
pyruvate + 2 H+ + 2e– lactate E0 = –0.19 V
NADH NAD+ + H+ + 2e– E0 = +0.32 V
Summing these two half reactions and their E0 values gives the overall
equation and its E0 value:
pyruvate + NADH + H+ lactate + NAD+ E0 = +0.13 V
When a redox reaction takes place under nonstandard conditions, the
tendency to donate electrons ( E ) is equal to E0 modified by a con-
centration term:

2.3 RT [lactate][NAD+]
∆E = ∆E0′ – log
nF [pyruvate][NADH]

where R = 8.3 × 10–3 kJ/K mole, T = temperature in kelvins, n = the num-

ber of electrons transferred, and F = 96 kJ/V mole.
G is related to E by the equation
G = –nF E
Since the signs of G and E are opposite, a favorable redox reaction has
a positive E and a negative G.
A. Calculate G for reduction of pyruvate to lactate at 37°C with all reactants
and products at a concentration of 1 M.
B. Calculate G for the reaction at 37°C under conditions where the con-
centrations of pyruvate and lactate are equal and the concentrations of
NAD+ and NADH are equal.
C. What would the concentration term need to be for this reaction to have a
G of zero at 37°C?
D. Under normal conditions in vascular smooth muscle (at 37°C), the con-
centration ratio of NAD+ to NADH is 1000, the concentration of lactate is
0.77 μmol/g, and the concentration of pyruvate is 0.15 μmol/g. What is
G for reduction of pyruvate to lactate under these conditions?
14–32 Thiobacillus ferrooxidans, a bacterium that lives on slag heaps at pH
2, is used by the mining industry to recover copper and uranium from
low-grade ore by an acid leaching process. The bacteria oxidize Fe2+ to
produce Fe3+, which in turn oxidizes (and solubilizes) these minor com-
ponents of the ore. It is remarkable that the bacterium can live in such
an environment. It does so by exploiting the pH difference between the
environment and its cytoplasm (pH 6.5) to drive synthesis of ATP and
NADPH, which it can then use to fix CO2 and nitrogen. In order to keep
its cytoplasmic pH constant, T. ferrooxidans uses electrons from Fe2+ to
reduce O2 to water, thereby removing the protons:
4 Fe2+ + O2 + 4 H+ 4 Fe3+ + 2 H2O
What are the energetics of these processes? Is the flow of electrons from
Fe2+ to O2 energetically favorable? How difficult is it to reduce NADP+
using electrons from Fe2+? These are key questions for understanding
how T. ferrooxidans manages to thrive in such an unlikely niche.
A. What are E and G for the reduction of O2 by Fe2+, assuming that the
reaction occurs under standard conditions? The half reactions are
Fe3+ + e– Fe2+ E0 = 0.77 V
O2 + 4 H+ + 4e– 2 H2O E0 = 0.82 V
290 Chapter 14: Energy Conversion: Mitochondria and Chloroplasts

B. Write a balanced equation for the reduction of NADP+ + H+ by Fe2+. What

is G for this reaction under standard conditions? The half reaction for
NADP+ is
NADP+ + H+ + 2e– NADPH E0 = –0.32 V
What is G for the reduction of NADP+ + H+ by Fe2+ if the concentrations
of Fe3+ and Fe2+ are equal, the concentration of NADPH is 10-fold greater
than that of NADP+, and the temperature is 310 K? (Note: adjusting the
number of atoms and electrons in order to balance the chemical equa-
tion does not affect E0 or E0 values. It does, however, affect E by its
influence on the concentration term; each concentration term must be
raised to an exponent equal to the number of atoms or molecules used in
the balanced equation.)
14–33 What is the standard free-energy change ( G°) associated with trans-
fer of electrons from NADH to O2, according to the balanced equation
below?
O2 + 2 NADH + 2 H+ 2 H2O + 2 NAD+
The half reactions are
O2 + 4 H+ + 4e– 2 H2O E0 = 0.82V
NAD+ + H+ + 2e– NADH E0 = –0.32 V
A common way of writing this equation is
½ O2 + NADH + H+ H2O + NAD+
What is G° for this equation? Do the two calculations give the same
answer? Explain why they do or don’t.

DATA HANDLING
14–34 In 1925, David Keilin used a simple spectroscope to observe the charac-
teristic absorption bands of the cytochromes that make up the electron-
transport chain in mitochondria. A spectroscope passes a very bright
light through the sample of interest and then through a prism to display
the spectrum from red to blue. If molecules in the sample absorb light of
particular wavelengths, dark bands will interrupt the colors of the rain-
bow. Keilin found that tissues from a wide variety of animals all showed
the pattern in Figure 14–2. (This pattern had actually been observed
several decades before by an Irish physician named MacMunn, but he
thought all the bands were due to a single pigment. His work was all but
forgotten by the 1920s.)
The different heat stabilities of the individual absorption bands and
their different intensities in different tissues led Keilin to conclude that
the absorption pattern was due to three components, which he labeled
cytochromes a, b, and c (Figure 14–2). His key discovery was that the
absorption bands disappeared when oxygen was introduced (Figure
14–3A) and then reappeared when the samples became anaerobic

cytochrome
absorption bands
c b a

400 500 600 700

Figure 14–2 Cytochrome absorption bands
wavelength (nm) (Problem 14–34).
THE PROTON PUMPS OF THE ELECTRON-TRANSPORT CHAIN 291

(A) AEROBIC Figure 14–3 Cytochrome absorption

bands under a variety of experimental
conditions (Problem 14–34).
(B) ANAEROBIC

(C) AEROBIC PLUS CYANIDE

(D) AEROBIC PLUS URETHANE

(E) CYTOCHROME c PLUS OXYGEN

400 500 600 700

wavelength (nm)

(Figure 14–3B). He later confessed, “This visual perception of an intra-

cellular respiratory process was one of the most impressive spectacles I
have witnessed in the course of my work.”
Keilin subsequently discovered that cyanide prevented the bands
from disappearing when oxygen was introduced (Figure 14–3C). When
urethane (an inhibitor of electron transport that is no longer used) was
added, bands a and c disappeared in the presence of oxygen, but band
Figure p14.09/14.03
b remained (Figure 14–3D). Finally, using cytochrome c extracted from
dried yeast, he showed that the band due to cytochrome c remained
when oxygen was present Problem p14.59/14.34
(Figure 14–3E).
A. Is it the reduced (electron-rich) or the oxidized (electron-poor) forms of
the cytochromes that give rise to the bands that Keilin observed?
B. From Keilin’s observations, deduce the order in which the three
cytochromes carry electrons from intracellular substrates to oxygen.
C. One of Keilin’s early observations was that the presence of excess glu-
cose prevented the disappearance of the absorption bands when oxygen
was added. How do you think that rapid glucose oxidation to CO2 might
explain this observation?
14–35 If isolated mitochondria are incubated with a source of electrons such
as succinate, but without oxygen, electrons enter the respiratory chain,
reducing each of the electron carriers almost completely. When oxygen is
then introduced, the carriers become oxidized at different rates (Figure 100
14–4). How does this result allow you to order the electron carriers in the
reduced cytochrome (percent)

respiratory chain? What is their order? add b

O2
14–36 Inhibitors have provided extremely useful tools for analyzing mitochon- c1
drial function. Figure 14–5 shows three distinct patterns of oxygen elec- 50
trode traces obtained using a variety of inhibitors. In all experiments, c
add
mitochondria were added to a phosphate-buffered solution containing succinate
succinate as the sole source of electrons for the respiratory chain. After (a + a3)
a short interval, ADP was added followed by an inhibitor, as indicated
in Figure 14–5. The rates of oxygen consumption at various times during 0
time
the experiment are shown by downward-sloping lines, with faster rates of
consumption shown by steeper lines. Figure 14–4 Rapid spectrophotometric
A. Based on the descriptions of the inhibitors in Table 14–2, assign each analysis of the rates of oxidation of
electron carriers in the respiratory chain
inhibitor to one of the oxygen traces in Figure 14–5. All these inhibitors (Problem 14–35). Cytochromes a and a3
stop ATP synthesis. cannot be distinguished and thus are
B. Using the same experimental protocol indicated in Figure 14–5, sketch listed as cytochrome (a + a3).
292 Chapter 14: Energy Conversion: Mitochondria and Chloroplasts

(A) mitochondria
TABLE 14–2 Effects of a variety of inhibitors of mitochondrial function
ADP
(Problem 14–36).
Inhibitor Function inhibitor

1. FCCP Makes membranes permeable to protons

(B) mitochondria
2. Malonate Prevents oxidation of succinate
ADP
3. Cyanide Inhibits the cytochrome c oxidase complex
inhibitor
4. Atractyloside Inhibits the ADP/ATP transporter
5. Oligomycin Inhibits ATP synthase
6. Butylmalonate Blocks mitochondrial uptake of succinate
(C) mitochondria
ADP
the oxygen traces that you would expect for the sequential addition of the
pairs of inhibitors in the list below. inhibitor

1. FCCP followed by cyanide

2. FCCP followed by oligomycin
3. Oligomycin followed by FCCP

MEDICAL LINKS
14–37 Normally, the flow of electrons to O2 is tightly linked to the production
of ATP via the electrochemical gradient. If ATP synthase is inhibited, for
example, electrons do not flow down the electron-transport chain and
respiration ceases. Since the 1940s, several substances—such as 2,4-dini- Figure 14–5 Oxygen traces showing three
trophenol—have been known to uncouple electron flow from ATP syn- patterns of inhibitor effects on oxygen
thesis. Dinitrophenol was once prescribed as a diet drug to aid in weight consumption by mitochondria (Problem
loss. How would an uncoupler of oxidative phosphorylation promote 14–36).
weight loss? Why do you suppose dinitrophenol is no longer prescribed?

ATP PRODUCTION IN MITOCHONDRIA

TERM TO LEARN
ATP synthase

DEFINITIONS
Match the definition below with its term from the list above.
Figure p14.11/14.05
14–38 Enzyme in the inner membrane of a mitochondrion that catalyzes the
formation of ATP from ADP and inorganic phosphate. Problem p14.61/14.36
TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
14–39 An average person contains 50 kg of ATP, which is required to meet their
daily energy needs.
14–40 The number of c subunits in the rotor ring of ATP synthase defines how
many protons need to pass through the turbine to make each molecule of
ATP.

THOUGHT PROBLEMS
14–41 The respiratory chain is relatively inaccessible to experimental manipu-
lation in intact mitochondria. After disrupting mitochondria with ultra-
sound, however, it is possible to isolate functional submitochondrial par-
ticles, which consist of broken cristae that have resealed inside-out into
small closed vesicles. In these vesicles, the components that originally
ATP PRODUCTION IN MITOCHONDRIA 293

faced the matrix are now exposed to the surrounding medium. How do
you suppose such an arrangement might aid in the study of electron
+ detergent
transport and ATP synthesis?
14–42 As electrons move down the respiratory chain, protons are pumped
across the inner membrane. Are those protons confined to the inter- bacteriorhodopsin ATP synthase
membrane space? Why or why not?
ADD PHOSPHOLIPIDS,
14–43 You have reconstituted into the membranes of the same vesicles purified
REMOVE DETERGENT
bacteriorhodopsin, which is a light-driven H+ pump from a photosyn-
thetic bacterium, and purified ATP synthase from ox heart mitochondria.
Assume that all molecules of bacteriorhodopsin and ATP synthase are
oriented as shown in Figure 14–6, so that protons are pumped into the
vesicle and ATP synthesis occurs on the outer surface. LIGHT
H+
A. If you add ADP and phosphate to the external medium and shine light
into the suspension of vesicles, would you expect ATP to be generated?
Why or why not?
B. If you prepared the vesicles without being careful to remove all the deter-
gent, which makes the bilayer leaky to protons, would you expect ATP to
be synthesized?
C. If the ATP synthase molecules were randomly oriented so that about
half faced the outside of the vesicle and half faced the inside, would you
expect ATP to be synthesized? If the bacteriorhodopsin molecules were sealed
randomly oriented, would you expect ATP to be synthesized? Explain vesicle
your answers.
D. You tell a friend over dinner about your new experiments. He questions
Figure 14–6 Reconstitution of
the validity of an approach that utilizes components from so widely bacteriorhodopsin and ATP synthase
divergent, unrelated organisms. As he so succinctly puts it, “Why would into lipid vesicles (Problem 14–43).
anybody want to mix vanilla pudding with brake fluid?” Defend your
approach against his criticism.
14–44 An elongated arm—the stator—links the catalytic head group (the 3 3
complex) of the ATP synthase to the membrane-embedded rotor com-
ponent. Attached to the rotor is a stalk (the axle-like subunit) that turns
inside the head group to force the conformational changes that lead to
ATP synthesis. If the stator were missing, would ATP be synthesized in
response to the proton flow? Why or why not?

CALCULATIONS Figure p14-01/14.-06

14–45 In actively respiring liver mitochondria, the pH in the matrix is about half Problem p14-21/14.43
a pH unit higher than it is in the cytosol. Assuming that the cytosol is at
pH 7 and the matrix is a sphere with a diameter of 1 μm [V = (4/3) r3],
calculate the total number of protons in the matrix of a respiring liver
mitochondrion. If the matrix began at pH 7 (equal to that in the cytosol),
how many protons would have to be pumped out to establish a matrix pH
of 7.5 (a difference of 0.5 pH unit)?
14–46 The relationship of free-energy change ( G) to the concentrations of
reactants and products is important because it predicts the direction
of spontaneous chemical reactions. Familiarity with this relationship is
essential for understanding energy conversions in cells. Consider, for
example, the hydrolysis of ATP to ADP and inorganic phosphate (Pi):
ATP + H2O ADP + Pi
The free-energy change due to ATP hydrolysis is
[ADP][Pi]
∆G = ∆G °+ RT ln
[ATP]
[ADP][Pi]
= ∆G °+ 2.3 RT log
[ATP]
294 Chapter 14: Energy Conversion: Mitochondria and Chloroplasts

where the concentrations are expressed as molarities (by convention,

the concentration of water is not included in the expression). R is the gas
constant (8.3 × 10–3 kJ/K mole), T is temperature (assume 37°C, which
is 310 K), and G° is the standard free-energy change (–30.5 kJ/mole for
ATP hydrolysis to ADP and Pi).
A. Calculate G for ATP hydrolysis when the concentrations of ATP, ADP,
and Pi are all equal to 1 M. What is G when the concentrations of ATP,
ADP, and Pi are all equal to 1 mM?
B. In a resting muscle, the concentrations of ATP, ADP, and Pi are approxi-
mately 5 mM, 1 mM, and 10 mM, respectively. What is G for ATP hydrol-
ysis in resting muscle?
C. What will G equal when the hydrolysis reaction reaches equilibrium? At
[Pi] = 10 mM, what will be the ratio of [ATP] to [ADP] at equilibrium?
D. Show that, at constant [Pi], G decreases by 5.9 kJ/mole for every 10-fold
increase in the ratio of [ATP] to [ADP], regardless of the value of G°. (For
example, G decreases by 11.8 kJ/mole for a 100-fold increase, by 17.7
kJ/mole for a 1000-fold increase, and so on.)

DATA HANDLING
14–47 ATP synthase is the world’s smallest rotary motor. Passage of H+ ions
through the membrane-embedded portion of ATP synthase (the Fo com-
ponent) causes rotation of the single, central, axle-like subunit inside
the head group. The tripartite head is composed of the three dimers,
the subunit of which is responsible for synthesis of ATP. The rotation
of the subunit induces conformational changes in the dimers that
allow ADP and Pi to be converted into ATP. A variety of indirect evidence
had suggested rotary catalysis by ATP synthase, but seeing is believing.
To demonstrate rotary motion, a modified form of the 3 3 complex
was used. The subunits were modified so they could be firmly anchored
to a solid support and the subunit was modified (on the end that nor-
mally inserts into the Fo component in the inner membrane) so that a
fluorescently tagged, readily visible filament of actin could be attached
(Figure 14–7A). This arrangement allows rotations of the subunit to
be visualized as revolutions of the long actin filament. In these experi-
ments, ATP synthase was studied in the reverse of its normal mechanism
by allowing it to hydrolyze ATP. At low ATP concentrations, the actin fila-
ment was observed to revolve in steps of 120° and then pause for variable
lengths of time, as shown in Figure 14–7B.
A. Why does the actin filament revolve in steps with pauses in between?
What does this rotation correspond to in terms of the structure of the
3 3 complex?
B. In its normal mode of operation inside the cell, how many ATP molecules
do you suppose would be synthesized for each complete 360° rotation of
the subunit? Explain your answer.

(A) direction of rotation (B) Figure 14–7 Experimental set-up for

5 observing rotation of the subunit
of ATP synthase (Problem 14–47).
4 (A) The immobilized 3 3 complex.
actin filament The subunits are anchored to
revolutions

γ 3 a solid support and a fluorescent

inner membrane actin filament is attached to the
matrix 2 subunit. (B) Stepwise revolution of
β β
1
the actin filament. The indicated
trace is a typical example from one
α

0 experiment. The inset shows the

0 20 40 60 80 100 positions in the revolution at which
time (seconds) the actin filament pauses.
ATP PRODUCTION IN MITOCHONDRIA 295

(A) (B)

i
P

AT
y
AT

pt
P

P
AD

em
ADP + Pi ATP ADP + Pi ATP

ATP 3H+ empty ATP 3H+ ADP Pi

3H+ ADP + Pi 3H+ ADP + Pi

AD
AD

i
P
y
em

em
pt
P

P
P
AT

AT

P
P
em

AD
pt

pt

Pi
Pi
y

y
3H+ 3H+ empty
ADP Pi ATP ATP

ADP + Pi ATP ADP + Pi ATP

14–48 The three dimers in each ATP synthase normally exist in three dif- Figure 14–8 The two possible
ferent conformations: one empty, one with ADP and Pi bound, and one arrangements of conformations of the
three dimers in ATP synthase, along
with ATP bound. The conformational changes are driven sequentially by
with the linked conformational changes
rotation of the subunit, which in turn is driven by the flow of protons driven by proton flow (Problem 14–48).
through the ATP synthase. As viewed from the inner membrane, look- One dimer is colored red to emphasize
ing at the underside of the 3 3 complex, these sites could be arranged that its position remains fixed as it
in either of two ways around the central subunit (Figure 14–8). The changes conformation in response to
proton flow. The perspective illustrated is
sequential, linked conformational changes driven by proton flow are from the inner membrane, looking at the
also shown for the two arrangements in the figure. In Problem 14–47, the underside of the tripartite 3 3 complex.
revolutions of the attached actin filaments during ATP hydrolysis were
shown to be counterclockwise when viewed from the same perspective
(see Figure 14–7). Which of the two arrangements of conformations of
dimers shown in Figure 14–8 is correct? Explain your answer.
14–49 A manuscript has been submitted to a prestigious scientific journal. In
it the authors describe an experiment using an immobilized 3 3 com-
plex with an attached actin filament like that shown in Figure 14–7A. The
authors show that they can mechanically rotate the subunit by applying
Figure ofp14.05/14.08
force to the actin filament. Moreover, in the presence ADP and phos-
phate, each 120° clockwise rotation of the subunit is accompanied by
the synthesis of one molecule of ATP. Is this result at all reasonable?
Problem What
p14-32/14.48
would such an observation imply about the mechanism of ATP synthase?
Should this manuscript be considered for publication in one of the best
journals?
14–50 The ADP/ATP transporter in the mitochondrial inner membrane can
exchange ATP for ATP, ADP for ADP, and ATP for ADP. Even though mito-
chondria can transport both ADP and ATP, there is a strong bias in favor
of exchange of external ADP for internal ATP in actively respiring mito-
chondria. You suspect that this bias is due to the conversion of ADP into
ATP inside the mitochondrion. ATP synthesis would continually reduce
the internal concentration of ADP and thereby create a favorable concen-
tration gradient for import of ADP. The same process would increase the
internal concentration of ATP, thereby creating favorable conditions for
export of ATP.
To test your hypothesis, you conduct experiments on isolated mito-
chondria. In the absence of substrate (when the mitochondria are not
respiring and the membrane is uncharged), you find that ADP and ATP
are taken up at the same rate. When you add substrate, the mitochon-
dria begin to respire, and ADP enters mitochondria at a much faster rate
than ATP. As you expected, when you add dinitrophenol, which collapses
the pH gradient, along with the substrate, ADP and ATP again enter at
296 Chapter 14: Energy Conversion: Mitochondria and Chloroplasts

ATP NH2
TABLE 14–3 Entry of ADP and ATP into isolated mitochondria (Problem
N
14–50). N
O O O N N
Experiment Substrate Inhibitor Relative rates –
of entry O P O P O P O CH2
O
O– O– O–
1 Absent None ADP = ATP H H
H H
2 Present None ADP > ATP OH OH

3 Present Dinitrophenol ADP = ATP

ADP NH2
4 Present Oligomycin ADP > ATP
N N
In all cases, the initial rates of entry of ATP and ADP were measured.
O O N N
–
O P O P O CH2
O
O– O–
H H
the same rate. When you add an inhibitor of ATP synthase (oligomycin) H H

along with the substrate, ADP is taken up much faster than ATP. Your OH OH
results are summarized in Table 14–3. You are puzzled by the results
Figure 14–9 Structures of ATP and ADP
with oligomycin, since your hypothesis predicted that the rates of uptake (Problem 14–50).
would be equal.
When you show the results to your advisor, she compliments you
on your fine experiments and agrees that they disprove the hypothesis.
She suggests that you examine the structures of ATP and ADP (Figure
14–9) if you wish to understand the behavior of the ADP/ATP transporter.
What is the correct explanation for the biased exchange by the ADP/ATP
transporter under some of the experimental conditions and an unbiased
exchange under others?

CHLOROPLASTS AND PHOTOSYNTHESIS Figure p14.06/14.09

TERMS TO LEARN
antenna complex photochemical reaction center Problem p14.34/14.50
carbon fixation photosynthetic electron transfer
carbon-fixation reactions photosystem
charge separation stroma
chlorophyll thylakoid membrane
chloroplast

DEFINITIONS
Match each definition below with its term from the list above.
14–51 Light-driven reactions in photosynthesis in which electrons move along
the electron-transport chain in the thylakoid membrane.
14–52 Part of a photosystem that captures light energy and channels it into the
photochemical reaction center.
14–53 Process by which green plants incorporate carbon atoms from atmos-
pheric carbon dioxide into sugars.
14–54 The part of a photosystem that converts light energy into chemical energy.
14–55 Organelle in green algae and plants that contains chlorophyll and carries
out photosynthesis.
14–56 Light-absorbing green pigment that plays a central role in photosynthe-
sis.
14–57 The large space that surrounds the inner chloroplast membrane.
CHLOROPLASTS AND PHOTOSYNTHESIS 297

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
14–58 In a general way, one might view the chloroplast as a greatly enlarged
mitochondrion in which the cristae have been pinched off to form a
series of interconnected submitochondrial particles in the matrix space.
14–59 When a molecule of chlorophyll in an antenna complex absorbs a pho-
ton, the excited electron is rapidly transferred from one chlorophyll mol-
ecule to another until it reaches the photochemical reaction center.

THOUGHT PROBLEMS
14–60 Both mitochondria and chloroplasts use electron transport to pump pro-
tons, creating an electrochemical proton gradient, which drives ATP syn-
thesis. Are protons pumped across the same (analogous) membranes in
the two organelles? Is ATP synthesized in the analogous compartments?
Explain your answers.
14–61 A suspension of the green alga Chlamydomonas is actively carrying out
photosynthesis in the presence of light and CO2. If you turned off the
light, how would you expect the amounts of ribulose 1,5-bisphosphate
and 3-phosphoglycerate to change over the next minute? How about if
you left the light on but removed the CO2?
14–62 Why are plants green?
14–63 Treatment of chloroplasts with the herbicide DCMU stops O2 evolution
and photophosphorylation. If an artificial electron acceptor is added that
accepts electrons from plastoquinone (Q), oxygen evolution is restored
but not photophosphorylation. Propose a site at which DCMU acts in the
flow of electrons through photosystems II and I (Figure 14–10). Explain
your reasoning. Why is DCMU an herbicide?
14–64 In chloroplasts, protons are pumped out of the stroma across the thyla-
koid membrane, whereas in mitochondria, they are pumped out of the
matrix across the crista membrane. Explain how this arrangement allows
chloroplasts to generate a larger proton gradient across the thylakoid photosystem I
membrane than mitochondria can generate across the inner membrane.
photosystem II
14–65 Unlike mitochondria, chloroplasts do not have a transporter that allows
light NADPH
them to export ATP to the cytosol. How, then, does the rest of the cell get
separates
the ATP it needs to survive? charges
Q
light
separates cytochromes
CALCULATIONS charges
pC +

14–66 How much energy is available in visible light? How much energy does
sunlight deliver to Earth? How efficient are plants at converting light H 2O +
energy into chemical energy? The answers to these questions provide an
important backdrop to the subject of photosynthesis. electron flow
Each quantum or photon of light has energy hv, where h is Planck’s
constant (6.6 × 10–37 kJ sec/photon) and v is the frequency in sec–1. Figure 14–10 Flow of electrons
through photosystems II and I during
The frequency of light is equal to c/ , where c is the speed of light photosynthesis in chloroplasts (Problem
(3.0 × 1017 nm/sec) and is the wavelength in nm. Thus, the energy (E) 14–63). Electrons from photosystem II
of a photon is flow to plastoquinone (Q), then to the
cytochrome b6-f complex (cytochromes),
E = hv = hc/ and then to plastocyanin (pC), after which
they enter photosystem I. The protons
A. Calculate the energy of a mole of photons (6 × 1023 photons/mole) at pumped by the cytochrome b6-f complex
400 nm (violet light), at 680 nm (red light), and at 800 nm (near-infrared generate an electrochemical gradient,
light). which is used to drive ATP synthesis.

Figure p14.14/14.10
298 Chapter 14: Energy Conversion: Mitochondria and Chloroplasts

B. Bright sunlight strikes Earth at the rate of about 1.3 kJ/sec per square
meter. Assuming for the sake of calculation that sunlight consists of mon-
ochromatic light of wavelength 680 nm, how many seconds would it take
for a mole of photons to strike a square meter?
C. Assuming that it takes eight photons to fix one molecule of CO2 as car-
bohydrate under optimal conditions (8–10 photons is the currently
accepted value), calculate how long it would take a tomato plant with a
leaf area of 1 square meter to make a mole of glucose from CO2. Assume
that photons strike the leaf at the rate calculated above and, furthermore,
that all the photons are absorbed and used to fix CO2.
D. If it takes 468 kJ/mole to fix a mole of CO2 into carbohydrate, what is the
efficiency of conversion of light energy into chemical energy after pho-
ton capture? Assume again that eight photons of red light (680 nm) are
required to fix one molecule of CO2.
14–67 What fraction of the free energy of light at 700 nm is captured when a
chlorophyll molecule (P700) at the photochemical reaction center in
photosystem I absorbs a photon? The equation for calculating the free
energy available in one photon of light is given in Problem 14–66. If one
assumes standard conditions, the captured free energy ( G = –nF E0 )
can be calculated from the standard redox potential for P700* (excited)
P700 (ground state), which can be gotten from the half reactions:
P700+ + e– P700 E0 = 0.4 V
P700+ + e– P700* E0 = –1.2 V
14–68 The balanced equation for production of NADPH by the Z scheme of
photophosphorylation is
2 H2O + 2 NADP+ 2 NADPH + 2 H+ + O2
How many photons must be absorbed to generate two NADPH and a
molecule of O2? (Assume one photon excites one electron.)
14–69 T. ferrooxidans, the slag-heap bacterium that lives at pH 2, fixes CO2 like pho-
tosynthetic organisms but uses the abundant Fe2+ in its environment as a
source of electrons instead of H2O. T. ferrooxidans oxidizes Fe2+ to Fe3+ to
reduce NADP+ to NADPH, a very unfavorable reaction with a E of about
–1.1 V. It does so by coupling production of NADPH to the energy of the nat-
ural proton gradient across its membrane, which has a free-energy change
( G) of –26.8 kJ/mole H+. What is the smallest number of protons to the
nearest integer that would be required to drive the reduction of NADP+ by
Fe2+? How do you suppose proton flow is mechanistically coupled to the
reduction of NADP+?

DATA HANDLING
14–70 Careful experiments comparing absorption and action spectra of plants
ultimately led to the notion that two photosystems cooperate in chloro-
plasts. The absorption spectrum is the amount of light captured by pho-
tosynthetic pigments at different wavelengths. The action spectrum is the
rate of photosynthesis (for example, O2 evolution or CO2 fixation) result-
ing from the capture of photons.
T.W. Engelmann, who used simple equipment and an ingenious
experimental design, probably made the first measurement of an action
spectrum in 1882. He placed a filamentous green alga into a test tube
along with a suspension of oxygen-seeking bacteria. He allowed the bac-
teria to use up the available oxygen and then illuminated the alga with
light that had been passed through a prism to form a spectrum. After a
short time he observed the results shown in Figure 14–11. Sketch the
CHLOROPLASTS AND PHOTOSYNTHESIS 299

Figure 14–11 Experiment to measure the

action spectrum of a filamentous green
alga (Problem 14–70). Bacteria, which are
indicated by the brown rectangles, were
distributed evenly throughout the test
tube at the beginning of the experiment.

400 500 600 700 wavelength (nm)

ultraviolet blue green yellow red infrared

action spectrum (O2 evolved at different wavelengths of light) for this

alga and explain how this experiment works.
14–71 The most compelling early evidence for the Z scheme of photosynthesis
came from measuring the oxidation states of the cytochromes in algae
under different regimes of illumination (Figure 14–12). Illumination
with light at 680 nm caused oxidation of cytochromes (indicated by the
upward trace in Figure 14–12A). Additional illumination with light at
562 nm caused reduction of the cytochromes (indicated by the down-
ward trace in Figure 14–12A). When the lights were then turned off, both
Figure
effects were reversed (Figure p14.16/14.11
14–12A). In the presence of the herbicide
DCMU (see Problem 14–63), no reduction with 562-nm light occurred
(Figure 14–12B). Problem p14.92/14.70
A. In these algae, which wavelength stimulates photosystem I and which
stimulates photosystem II?
B. How do these results support the Z scheme for photosynthesis; that is,
how do they support the idea that there are two photosystems that are
linked by cytochromes?
C. On which side of the cytochromes does DCMU block electron trans-
port—on the side nearer photosystem I or the side nearer photosystem
II?
14–72 Photosystem II accepts electrons from water, generating O2, and donates
them via the electron-transport chain to photosystem I. Each photon
absorbed by photosystem II transfers only a single electron, and yet four

(A) 680 562 562 680

on on off off
absorbance at 420 nm

(B) 680 562 562 680

on on off off
absorbance at 420 nm

more oxidized
Figure 14–12 Oxidation state of
DCMU cytochromes after illumination of algae
with light of different wavelengths
more reduced (Problem 14–71). (A) In the absence of
DCMU. (B) In the presence of DCMU. An
upward trace indicates oxidation of the
0 5 10 15 20 cytochromes; a downward trace indicates
time (seconds) reduction of the cytochromes.
300 Chapter 14: Energy Conversion: Mitochondria and Chloroplasts

Figure 14–13 Oxygen evolution by spinach chloroplasts in response to 80

saturating flashes of light (Problem 14–72). The chloroplasts were placed
in the dark for 40 minutes prior to the experiment to allow them to come

oxygen produced
60
to the same “ground” state. Oxygen production is expressed in arbitrary
units.
40

electrons must be removed from water to generate a molecule of O2. 20

Thus, four photons are required to produce a molecule of O2:
0
2 H2O + 4 hv 4e– + 4 H+ + O2 0 2 4 6 8 10 12 14
flash number
How do four photons cooperate in the production of O2? Is it necessary
that four photons arrive at a single reaction center simultaneously? Can
four activated reaction centers cooperate to produce a molecule of O2?
Or is there some sort of “gear wheel” that collects the four electrons from
H2O and transfers them one at a time to a reaction center?
To investigate this problem, you expose dark-adapted spinach chlo-
roplasts to a series of brief saturating flashes of light (2 μsec) separated
by short periods of darkness (0.3 sec) and measure the production of O2
that results from each flash. Under this lighting regime, most photosys-
tems capture a photon during each flash. As shown in Figure 14–13, O2
is produced with a distinct periodicity: the first burst of O2 occurs on the
third flash, and subsequent peaks occur every fourth flash thereafter. If
you first inhibit 97% of the photosystem II reaction centers with DCMU
and then repeat the experiment, you observe the same periodicity of O2
production, but the peaks are only 3% of the uninhibited values.
A. How do these results distinguish among the three possibilities posed at Figure p14.20/14.13
the outset (simultaneous action, cooperation among reaction centers,
and a gear wheel)? Problem p14.95/14.72
B. Why do you think it is that the first burst of O2 occurs after the third flash,
whereas additional peaks occur at four-flash intervals? (Consider what
this observation implies about the dark-adapted state of the chloro-
plasts.)
C. Can you suggest a reason why the periodicity in O2 production becomes
less pronounced with increasing flash number?
14–73 In an insightful experiment performed in the 1960s, chloroplasts were
first soaked in an acidic solution at pH 4, so that the stroma and thylakoid
space became acidified (Figure 14–14). They were then transferred to a
basic solution (pH 8). This rapidly increased the pH of the stroma to 8,
while the thylakoid space temporarily remained at pH 4. A burst of ATP
synthesis was observed, and the pH difference between the thylakoid
space and the stroma quickly disappeared.
A. Explain why these conditions lead to ATP synthesis.
B. Is light needed for the experiment to work? Why or why not?
C. What would happen if the solutions were switched so that the first incu-
bation was in the pH 8 solution and the second one was in the pH 4 solu-
tion? Explain your answer.
D. Does this experiment support the chemiosmotic model, or raise ques-
tions about it?

INCUBATE CHANGE
CHLOROPLAST EXTERNAL pH
FOR SEVERAL AND ADD ADP
HOURS AND Pi

Figure 14–14 Soaking chloroplasts in

pH 4 pH 4 pH 8 acidic and basic solutions (Problem
14–73). Pink areas are at pH 4.
THE GENETIC SYSTEMS OF MITOCHONDRIA AND CHLOROPLASTS 301

THE GENETIC SYSTEMS OF MITOCHONDRIA AND

CHLOROPLASTS
TERM TO LEARN
maternal inheritance

DEFINITIONS
Match the definition below with its term from the list above.
14–74 Pattern of mitochondrial inheritance in higher animals that arises
because the egg cells always contribute much more cytoplasm to the
zygote than does the sperm.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
14–75 The mitochondrial genetic code differs slightly from the nuclear code,
but it is identical in mitochondria from all species that have been exam-
ined.
14–76 The presence of introns in organellar genes is not surprising since similar
introns have been found in related genes from bacteria whose ancestors
are thought to have given rise to mitochondria and chloroplasts.
14–77 Mutations that are inherited according to Mendelian rules affect nuclear
genes; mutations whose inheritance violates Mendelian rules are likely
to affect organelle genes.

THOUGHT PROBLEMS
14–78 You have discovered a remarkable, one-celled protozoan that lives in an
anaerobic environment. It has caught your attention because it has abso-
lutely no mitochondria, an exceedingly rare condition among eukary-
otes. If you could show that this organism derives from an ancient line-
age that split off from the rest of eukaryotes before mitochondria were
acquired, it would truly be a momentous discovery. You sequence the
organism’s genome so you can make detailed comparisons. It is clear
from sequence comparisons that your organism does indeed derive from
an ancient lineage. But here and there, scattered around the genome, are
bits of DNA that in aggregate resemble the bacterial genome from which
mitochondria evolved. Propose a plausible evolutionary history for your
organism.
14–79 At the cellular level, evolutionary theories are particularly difficult to
test since fossil evidence is lacking. The possible evolutionary origins
of mitochondria and chloroplasts must be sought in living organisms.
Fortunately, living forms resembling the ancestral types thought to
have established an endosymbiotic relationship that led to the origin
of mitochondria and chloroplasts can be found today. For example,
the plasma membrane of the free-living aerobic bacterium Paracoccus
denitrificans contains a respiratory chain that is nearly identical to the
respiratory chain of mammalian mitochondria—both in the types of res-
piratory components present and in its sensitivity to respiratory inhibi-
tors. Indeed, no significant feature of the mammalian respiratory chain
is absent from Paracoccus. Paracoccus effectively assembles in a single
organism all those features of the mitochondrial inner membrane that
are otherwise distributed at random among other aerobic bacteria.
302 Chapter 14: Energy Conversion: Mitochondria and Chloroplasts

(A)

(B)

(C)

Figure 14–15 A variegated leaf of Aucuba japonica with green and

yellow patches (Problem 14–80).
(D)

Imagine that you are a protoeukaryotic cell looking out for your evo-
lutionary future. You have been observing proto-Paracoccus and are
amazed at its incredibly efficient use of oxygen in generating ATP. With
such a source of energy, your horizons would be unlimited. You plot to
hijack a proto-Paracoccus and make it work for you and your descend- Figure 14–16 Hypothetical pedigrees
ants. You plan to take it into your cytoplasm, feed it any nutrients it representing four patterns of inheritance
(Problem 14–81). Males are shown as
needs, and harvest the ATP. Accordingly, one dark night, you trap a wan- squares; females as circles. Affected
Figure
dering proto-Paracoccus, p14.23/14.15
surround it with your plasma membrane, and individuals are shown as red symbols;
imprison it in a new cytoplasmic compartment. To your relief, the proto- unaffected individuals are shown as white
Paracoccus seems toProblem
enjoy its new environment. After a day of waiting,
p14.109/14.80 symbols.
however, you feel as sluggish as ever. What has gone wrong with your
scheme?
14–80 Examine the variegated leaf shown in Figure 14–15. Yellow patches
surrounded by green are common, but there are no green patches sur-
rounded by yellow. Propose an explanation for this phenomenon.
14–81 The pedigrees in Figure 14–16 show one example each of the following
types of mutation: mitochondrial mutation, autosomal recessive muta-
tion, autosomal dominant mutation, and X-linked recessive mutation. In
each family, the parents have had nine children. Assign each pedigree to
one type of mutation. Explain the basis for your assignments.

DATA HANDLING
14–82 It is well accepted that transfer of DNA from organellar genomes to
nuclear genomes is common during evolution. Do transfers between
i
in

h
ac
ch

rn

in

organellar genomes also occur? One experiment to search for genetic

c

pe
zu

co

sp

transfers between organellar genomes used a defined restriction frag- Figure

m c mp14.24/14.16
c m c m c
ment from spinach chloroplasts, which carried information for the gene origin
for the large subunit of ribulose bisphosphate carboxylase. This gene has Problem p14.110/14.81
no known mitochondrial counterpart. Thus, if a portion of the chloro-
plast DNA in the restriction fragment were transferred to the mitochon-
drial genome, it would show up as a hybridizing band at a novel position.
Mitochondrial and chloroplast DNAs were prepared from zucchini, corn,
spinach, and pea. All these DNAs were digested with the same restriction
nuclease, and the resulting fragments were separated by electrophoresis.
The fragments were then transferred to a filter and hybridized to a radio-
active preparation of the spinach fragment. A schematic representation Figure 14–17 Patterns of hybridization
of the autoradiograph is shown in Figure 14–17. of a probe from spinach chloroplast
A. It is very difficult to prepare mitochondrial DNA that is not contaminated DNA to mitochondrial and chloroplast
to some extent with chloroplast DNA. How do these experiments control DNAs from zucchini, corn, spinach, and
for contamination of the mitochondrial DNA preparation by chloroplast pea (Problem 14–82). Lanes labeled m
contain mitochondrial DNA; lanes labeled
DNA? c contain chloroplast DNA. Restriction
B. Which of these plant mitochondrial DNAs appear to have acquired chlo- fragments to which the probe hybridized
roplast DNA? are shown as dark bands.
THE GENETIC SYSTEMS OF MITOCHONDRIA AND CHLOROPLASTS 303

tRNA genes Figure 14–18 Transcription map of human

F V L IM W D K GR HSL T
mitochondrial DNA (Problem 14–83).
mitochondrial DNA Individual tRNA genes are indicated by
red circles; the amino acids they carry are
13 16 7
shown in the one-letter code. The three
12S 16S mRNAs whose detailed sequences are
mRNAs shown in Figure 14–19 are indicated by
ribosomal
RNA number.

14–83 The majority of mRNAs, tRNAs, and rRNAs in human mitochondria are
transcribed from one strand of the genome. These RNAs are all present
initially on one very long transcript, which is 93% of the length of the DNA
strand. During mitochondrial protein synthesis, these RNAs function as
separate, independent species of RNA. The relationship of the individ-
ual RNAs to the primary transcript and many of the special features of
the mitochondrial genetic system have been revealed by comparing the
sequences of the RNAs with the nucleotide sequence of the genome. An
overview of the transcription map is shown in Figure 14–18.
Three segments of the nucleotide sequence of the human mitochon-
drial genome are shown in Figure 14–19 along with the three mRNAs
that are generated from those regions. The nucleotides that encode tRNA
species are highlighted; the amino acids encoded by the mRNAs are indi-
cated below the center base of the codon.
A. In terms of codon usage and mRNA structure, in what two ways does ini-
tiation of protein synthesis in mitochondria differ from initiation in the
cytoplasm?
B. In what two ways are the termination codons for protein synthesis in
mitochondria unusual? (The termination codons are shown in Figure
14–19 as asterisks.)
C. Does the arrangement of tRNA and mRNA sequences in the genome
suggest a possible mechanism for processing the primary transcript into
individual RNA species?
14–84 A friend of yours has been studying a pair of mutants in the fungus Neu-
rospora, which she has whimsically named poky and puny. Both mutants
grow at about the same rate, but much more slowly than wild type. Your
friend has been unable to find any supplement that improves their
growth rates. Her biochemical analysis shows that each mutant displays
a different abnormal pattern of cytochrome absorption. To characterize
the mutants genetically, she crossed them to wild type and to each other
and tested the growth rates of the progeny.
Figure She has come to you because
p14.28/14.18
she is puzzled by the results.
She explains that haploid
Problemnucleip14.114/14.83
from the two parents fuse during a
Neurospora mating and then divide meiotically to produce four haploid
spores, which can be readily tested for their growth rates. The parents
Figure 14–19 Arrangements of tRNA
tRNAL tRNAI
and mRNA sequences at three places
TTCTTAACAACATACCCAT.........CTCAAACCTAAGAAATATG DNA on the human mitochondrial genome
ACAUACCCAU.........CUCAAACCUAAAAAAAAAA mRNA 13 (Problem 14–83). Highlighted sequences
M P E T * protein indicate tRNA genes. The sequences
of the mRNAs are shown in blue below
tRNAD tRNAK the corresponding genes. The middle
portions of the mRNAs and their genes
TATATCTTAATGGCACATG.........CTCTAGAGCCCACTGTAAA DNA are indicated by dots. The 5 ends of the
AUGGCACAUG.........CUCUAGAGCCAAAAAAAAA mRNA 16 sequences are shown at the left. The
M A H S * protein 5 ends of the mRNAs are unmodified
and the 3 ends have poly-A tails. The
tRNAR tRNAH encoded protein sequences are indicated
in green below the mRNAs, with the letter
ATTTACCAAATGCCCCTCA.........TTTTCCTCTTGTAAATATA DNA
for the amino acid immediately under
AUGCCCCUCA.........UUUUCCUCUUAAAAAAAAA mRNA 7 the center nucleotide of the codon. An
M P L F S S * protein asterisk (*) indicates a termination codon.
304 Chapter 14: Energy Conversion: Mitochondria and Chloroplasts

TABLE 14–4 Genetic analysis of Neurospora mutants (Problem 14–84).

Spore counts
Protoperithecial Fertilizing
Cross parent parent Fast growth Slow growth
1 poky wild 0 1749
2 wild poky 1334 0
3 puny wild 850 799
4 wild puny 793 801
5 poky puny 0 1831
6 puny poky 754 710
7 wild wild 1515 0
8 poky poky 0 1389
9 puny puny 0 1588

contribute unequally to the diploid: one parent (the protoperithecial

parent) donates a nucleus and the cytoplasm; the other (the fertilizing
parent) contributes little more than a nucleus—much like egg and sperm
in higher organisms. As shown in Table 14–4, the “order” of the crosses
sometimes makes a difference: this is a result she has not seen before.
Can you help your friend understand these results?

MCAT STYLE
Passage 1 (Questions 14–85 to 14–87)
Scientists discovered the mechanism for ATP production via glycolysis decades
before they understood the mechanism that generates ATP via oxidative phos-
phorylation. In glycolysis, ATP production is directly linked to the chemical reac-
tions that break glucose down into two molecules of pyruvate. Early studies of oxi-
dative phosphorylation suggested that transport of high-energy electrons down a
cascade of electron acceptors generated the energy for ATP production. By anal-
ogy with glycolysis, it was thought that production of ATP would be directly linked
to high-energy chemical intermediates produced during electron transport.
Despite much effort, such compounds were never found. However, a number of
experiments suggested an alternative hypothesis, in which the energy captured
during electron transport was used to pump protons (H+) across the membrane,
generating a gradient of protons that was subsequently used to drive ATP synthe-
sis. The proposed indirect linkage between electron transport and ATP produc-
tion was known as the chemiosmotic hypothesis for oxidative phosphorylation.
This hypothesis proved to be correct.
14–85 Electron transport is carried out by a series of large multiprotein com-
plexes. Early work found that these complexes were in some way asso-
ciated with the inner mitochondrial membrane, although their func-
tion and the nature of their association with the membrane were poorly
understood. Which one of the following observations regarding the
electron-transport complexes would have been most consistent with the
chemiosmotic hypothesis?
A. Efficient electron transport can be detected in purified preparations of
mitochondrial membranes.
B. Proteins in each electron-transport complex are exposed on both sides of
the inner mitochondrial membrane.
MCAT STYLE 305

C. The electron-transport complexes must be embedded in the membrane

to accept electrons.
D. The proteins in each electron-transport complex are exposed only to the
matrix side of the membrane.
14–86 In one experiment, mitochondrial membranes were mechanically bro-
ken into pieces by subjecting them to high-frequency sound waves.
Which one of the following observations would have been most consist-
ent with the chemiosmotic hypothesis?
A. Adding a source of electrons to the fragmented membranes yielded ATP.
B. Electron-transport complexes were associated with the fragmented
membranes.
C. Fragmented membranes could transport electrons, but could not gener-
ate ATP.
D. Fragmented membranes produced protons in response to electron
transport.
14–87 Which of the following experiments would have provided the clearest
proof of the chemiosmotic hypothesis?
A. A decrease in pH inside the mitochondrial matrix could generate ATP in
the complete absence of electron transport.
B. ATP production in intact mitochondria requires both the entire electron-
transport chain and the ATP synthase complex.
C. Bacteriorhodopsin, which transports protons across membranes in
response to light, could replace the electron-transport chain.
D. Reconstitution of the electron-transport chain in membranes was suffi-
cient to transport protons across the membrane.
Our Colleague, the Late Julian Lewis,
Semaphores the Letter H.
You may like to consider the similarities
and differences between this human
mode of communication and the signaling
networks used by cells.
Chapter 15 307

CHAPTER

Cell Signaling 15
PRINCIPLES OF CELL SIGNALING IN THIS CHAPTER

TERMS TO LEARN PRINCIPLES OF CELL

adaptation (desensitization) monomeric GTPase SIGNALING
adaptor neurotransmitter
contact-dependent signaling paracrine signaling SIGNALING THROUGH
endocrine cell phosphorylation G-PROTEIN-COUPLED
extracellular signal molecule primary cilium RECEPTORS
GTPase-activating protein (GAP) protein kinase
GTP-binding protein protein phosphatase SIGNALING THROUGH ENZYME-
guanine nucleotide exchange receptor COUPLED RECEPTORS
factor (GEF) scaffold protein
hormone second messenger ALTERNATIVE SIGNALING
interaction domain serine/threonine kinase ROUTES IN GENE REGULATION
ion-channel-coupled receptor synaptic signaling
kinase cascade tyrosine kinase SIGNALING IN PLANTS
local mediator

DEFINITIONS
Match each definition below with its term from the list above.
15–1 Protein that binds to a GTP-binding protein and activates it by stimulat-
ing release of tightly bound GDP, thereby allowing it to bind GTP.
15–2 General term for a protein that binds a specific extracellular molecule
(ligand) and initiates a response in the cell.
15–3 Alteration of sensitivity following repeated stimulation, reducing a cell’s
response to that level of stimulus.
15–4 Compact protein module that binds to a particular structural motif in
another protein (or lipid) molecule with which the signaling protein
interacts.
15–5 Short-range cell–cell communication via secreted local mediators that
act on adjacent cells.
15–6 A signal relay chain involving multiple protein kinases, each of which is
activated by phosphorylation and then phosphorylates the next protein
kinase in the sequence.
15–7 Small molecule that is formed in the cytosol, or released into it, in
response to an extracellular signal and that helps to relay the signal to the
interior of the cell.
15–8 Specialized animal cell that secretes a hormone into the blood.
308 Chapter 15: Cell Signaling

15–9 Molecule from outside the cell that communicates the behavior or actions
of other cells in the environment and elicits an appropriate response.
15–10 Enzyme that transfers the terminal phosphate group of ATP to a specific
amino acid of a target protein.
15–11 Small signal molecule secreted by the presynaptic nerve cell at a chemi-
cal synapse to relay the signal to the postsynaptic cell.
15–12 Protein that binds to a GTP-binding protein and inactivates it by stimu-
lating its GTPase activity so that its bound GTP is hydrolyzed to GDP.
15–13 Protein that organizes groups of interacting intracellular signaling pro-
teins into signaling complexes.
15–14 Cell–cell communication in which the signal molecule remains bound to
the signaling cell and only influences cells that physically touch it.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
15–15 There is no fundamental distinction between signaling molecules that
bind to cell-surface receptors and those that bind to intracellular recep-
tors.
15–16 All second messengers are water-soluble and diffuse freely through the
cytosol.

THOUGHT PROBLEMS
15–17 Compare and contrast signaling by neurons to signaling by endocrine
cells. What are the relative advantages of these two mechanisms for cel-
lular communication?
15–18 Cells communicate in ways that resemble human communication.
Decide which of the following forms of human communication are anal-
ogous to autocrine, paracrine, endocrine, and synaptic signaling by cells.
A. A telephone conversation
B. Talking to people at a cocktail party
C. A radio announcement
D. Talking to yourself
15–19 How is it that different cells can respond in different ways to exactly the
same signaling molecule even when they have identical receptors?
15–20 Working out the order in which the individual components in a signaling
pathway act is an essential step in defining the pathway. Imagine that
two protein kinases, PK1 and PK2, act sequentially in a kinase cascade.
When either kinase is completely inactivated, cells do not respond to
the normal extracellular signal. By contrast, cells containing a mutant
form of PK1 that is permanently active respond even in the absence of
an extracellular signal. Doubly mutant cells that contain inactivated PK2
and permanently active PK1 respond in the absence of a signal.
In the normal kinase cascade, does PK1 activate PK2 or does PK2 acti-
vate PK1? What outcome would you have predicted for a doubly mutant
cell line with an activating mutation in PK2 and an inactivating mutation
in PK1? Explain your reasoning.
15–21 Why do you suppose that phosphorylation/dephosphorylation, as
opposed to allosteric binding of small molecules, for example, has
evolved to play such a prominent role in switching proteins on and off in
signaling pathways?
PRINCIPLES OF CELL SIGNALING 309

15–22 The two main classes of molecular switches involve changes in phospho-
rylation state or changes in guanine nucleotide binding. Comment on
the following statement. “In the regulation of molecular switches, pro- CYTOSOL
tein kinases and guanine nucleotide exchange factors (GEFs) always turn
proteins on, whereas protein phosphatases and GTPase-activating pro- 1
1
teins (GAPs) always turn proteins off.”
2 2
15–23 Consider a signaling pathway that proceeds through three protein
kinases that are sequentially activated by phosphorylation. In one case,
the kinases are held in a signaling complex by a scaffold protein; in the 3 3
other, the kinases are freely diffusing (Figure 15–1). Discuss the proper-
ties of these two types of organization in terms of signal amplification,
speed, and potential for cross-talk between signaling pathways.
Figure 15–1 A protein kinase cascade
15–24 Proteins in signaling pathways use a variety of binding domains to organized by a scaffolding protein or
composed of freely diffusing components
assemble into signaling complexes. Match the following domains with
(Problem 15–23).
their binding targets. (A binding target can be used more than once.)
A. PH domain 1. phosphorylated tyrosines
B. PTB domain 2. proline-rich sequences Problems p15.02/15.01
C. SH2 domain 3. phosphorylated inositol phospholipids
D. SH3 domain
15–25 Describe three ways in which a gradual increase in an extracellular signal
can be sharpened by the target cell to produce an abrupt or nearly all-or-
none response.

CALCULATIONS
15–26 Suppose that the circulating concentration of hormone is 10–10 M and
the Kd for binding to its receptor is 10–8 M. What fraction of the recep-
tors will have hormone bound? If a meaningful physiological response
occurs when 50% of the receptors have bound a hormone molecule, how
much will the concentration of hormone have to rise to elicit a response?
Recall that the fraction of receptors (R) bound to hormone (H) to form a
receptor–hormone complex (R–H) is [R–H]/([R] + [R–H]) = [R–H]/[R]TOT
= [H]/([H] + Kd).
15–27 Radioimmunoassay (RIA) is a powerful tool for quantifying virtually
any substance of biological interest because it is sensitive, accurate, and
fast. RIA technology arose from studies on adult-onset diabetes. Some
patients had antibodies with high affinity for insulin, and RIA was devel-
oped as a method to distinguish free insulin from antibody-bound insu-
lin.
How can high-affinity antibodies be exploited to measure low con-
centrations of insulin? When a small amount of insulin-specific antise-
rum is mixed with an equally small amount of very highly radioactive
insulin, some binds and some remains free according to the equilibrium.
Insulin (I) + Antibody (A) Insulin–Antibody (I–A) Complex
When increasing amounts of unlabeled insulin are added to a fixed
amount of labeled insulin and anti-insulin antibody, the ratio of bound
to free radioactive insulin decreases as expected from the equilibrium
expression. If the concentration of the unlabeled insulin is known, then
the resulting curve serves as a calibration against which other unknown
samples can be compared (Figure 15–2).
You have three samples of insulin whose concentrations are unknown.
When mixed with the same amount of radioactive insulin and anti-insu-
lin antibody used in Figure 15–2, the three samples gave the following
ratios of bound to free insulin:
310 Chapter 15: Cell Signaling

1.4 Figure 15–2 Calibration curve

for radioimmunoassay of insulin
1.2 (Problem 15–27).
1.0

bound/free (ratio)
0.8

0.6

0.4

0.2

0.0
0 5 10 15
unlabeled insulin (pg/mL)

Sample 1 0.67
Sample 2 0.31
Sample 3 0.46
A. What is the concentration of insulin in each of these unknown samples?
B. What portion of the standard curve is the most accurate,
Problems and why?
p15.03/15.02
C. If the antibodies were raised against pig insulin, which is similar but not
identical to human insulin, would the assay still be valid for measuring
human insulin concentrations?
15–28 Two intracellular molecules, A and B, are normally synthesized at a con-
stant rate of 1000 molecules per second per cell. Each molecule of A sur-
vives an average of 100 seconds, while each molecule of B survives an
average of 10 seconds.
A. How many molecules of A and B will a cell contain?
B. If the rates of synthesis of both A and B were suddenly increased 10-fold
to 10,000 molecules per second—without any change in their average
lifetimes—how many molecules of A and B would be present after 1 sec-
ond?
C. Which molecule would be preferred for rapid signaling? Explain your
answer.

DATA HANDLING

15–29 The cellular slime mold Dictyostelium discoideum is a eukaryote that

lives on the forest floor as independent motile cells called amoebae,
which feed on bacteria and yeast. When their food supply is exhausted,
the amoebae stop dividing and gather together to form tiny, multicellu-
lar, wormlike structures, which crawl about as slugs. How do individual
amoebae know when to stop dividing and how to find their way into a
common aggregate? A set of classic experiments investigated this phe-
nomenon more than half a century ago.
Amoebae aggregate when placed on a glass coverslip under water,
provided that simple salts are present. The center of the aggregation pat-
tern can be removed with a pipette and placed in a field of fresh amoebae,
which immediately start streaming toward it. Thus, the center is emitting
some sort of attractive signal. Four experiments were designed to deter-
mine the nature of the signal. In each, an existing center of aggregation
was used as the source of the signal and previously unexposed amoebae
served as the target cells. The arrangements of aggregation centers and
test amoebae at the beginning and end of the experiments are shown in
Figure 15–3.
Do these results show that Dictyostelium discoideum aggregates
through the action of a secreted chemical signal? Explain your reasoning.
15–30 The nicotinic acetylcholine receptor is a neurotransmitter-dependent ion
channel, which is composed of four types of subunit. Phosphorylation of
PRINCIPLES OF CELL SIGNALING 311

INITIAL ARRANGEMENT FINAL ARRANGEMENT

actively signaling
(A) aggregation center
glass lower center forms
coverslip at random location
layer of amoebae aggregation center
placed at edge
(B)
glass lower amoebae stream
coverslip around the edge

(C) lower center forms

semipermeable exactly below upper
membrane center

(D) (top view)

gentle amoebae downstream

stream of of center stream to
medium join it; upstream
across amoebae ignore the
coverslip center

the receptor by protein kinase A attaches one phosphate to the subu- Figure 15–3 Four experiments to study
nit and one phosphate to the subunit. Fully phosphorylated receptors the nature of the attractive signal
generated by aggregation centers
desensitize much more rapidly than unmodified receptors. To study this (Problem 15–29).
process in detail, you phosphorylate two preparations of receptor to dif-
Problems
ferent extents (0.8 mole phosphate/mole p15.05/15.03
receptor and 1.2 mole phos-
phate/mole receptor) and measure desensitization over several seconds
(Figure 15–4). Both preparations behave as if they contain a mixture of
receptors; one form that is rapidly desensitized (the initial steep portion
of the curves) and another form that is desensitized at the same rate as
the untreated receptor.
A. Assuming that the and subunits are independently phosphorylated
at equal rates, calculate the percentage of receptors that carry zero, one,
and two phosphates per receptor at the two extents of phosphorylation.
B. Do these data suggest that desensitization requires one phosphate or two
phosphates per receptor? If you decide that desensitization requires only
one phosphate, indicate whether the phosphate has to be on one specific
subunit or can be on either of the subunits.

MEDICAL LINKS
15–31 Surgeons use succinylcholine, which is an acetylcholine analog, as a
muscle relaxant. Care must be taken because some individuals recover
abnormally slowly from this paralysis, with life-threatening conse-
quences. Such individuals are deficient in an enzyme called pseudo-
cholinesterase, which is normally present in the blood, where it slowly
inactivates succinylcholine by hydrolysis to succinate and choline.
100
If succinylcholine is an analog of acetylcholine, why do you think it
activity remaining (%)

causes muscles to relax and not contract as they do in the presence of 36

acetylcholine?
18
10

Figure 15–4 Desensitization rates for untreated acetylcholine receptor

and two preparations of phosphorylated receptor (Problem 15–30). Red
squares represent untreated receptors; blue squares represent receptors
with 0.8 mole phosphate/mole receptor; and brown triangles represent 1
receptors with 1.2 mole phosphate/mole receptor. Arrows indicate 0 2 4 6
the fractions of the phosphorylated preparations that behaved like the preincubation time
untreated receptor. (seconds)
312 Chapter 15: Cell Signaling

SIGNALING THROUGH G-PROTEIN-COUPLED

RECEPTORS
TERMS TO LEARN
adenylyl cyclase GPCR kinase (GRK)
arrestin inhibitory G protein (Gi)
Ca2+/calmodulin-dependent inositol phospholipid signaling pathway
kinase (CaM-kinase) inositol 1,4,5-trisphosphate (IP3)
calmodulin IP3 receptor
CaM-kinase II nitric oxide (NO)
cone photoreceptor NO synthase
CRE-binding (CREB) protein olfactory receptor
cyclic AMP (cAMP) phospholipase C- (PLC )
cyclic-AMP-dependent protein phosphatidylinositol 4,5-bisphosphate
kinase (PKA) (PI(4,5)P2)
cyclic AMP phosphodiesterase protein kinase C (PKC)
cyclic GMP regulator of G protein signaling (RGS)
cyclic GMP phosphodiesterase rhodopsin
diacylglycerol rod photoreceptor (rod)
Gq ryanodine receptor
G-protein-coupled receptor stimulatory G protein (Gs)
(GPCR) trimeric GTP-binding protein (G protein)

DEFINITIONS
Match each definition below with its term from the list above.
15–32 G protein that activates adenylyl cyclase and thereby increases cyclic
AMP concentration.
15–33 Protein composed of three subunits, one of which is activated by the
binding of GTP.
15–34 Ubiquitous calcium-binding protein whose interactions with other pro-
teins are governed by changes in intracellular Ca2+ concentration.
15–35 Enzyme that hydrolyzes cyclic AMP to adenosine 5 -monophosphate (5 -
AMP).
15–36 Cell-surface receptor that associates with an intracellular G protein upon
activation by an extracellular ligand.
15–37 Enzyme that participates in desensitization of GPCRs by phosphorylat-
ing them after they have been activated by ligand binding.
15–38 Ca2+-release channel in the ER membrane that is activated by Ca2+ bind-
ing in the absence of IP3.
15–39 Enzyme bound to the cytoplasmic surface of the plasma membrane that
converts membrane PI(4,5)P2 to diacylglycerol and IP3.
15–40 Protein that is an -subunit-specific GTPase-activating protein (GAP).
15–41 Second messenger that is released from a phospholipid in the plasma
membrane and diffuses to the ER, where it opens Ca2+-release channels.
15–42 Enzyme that phosphorylates target proteins in response to a rise in intra-
cellular cyclic AMP.
15–43 A Ca2+-dependent protein kinase that is activated by diacylglycerol.
15–44 Light-sensitive GPCR in rod photoreceptor cells of the retina.
15–45 Protein kinase whose activity is regulated by the binding of Ca2+-activated
SIGNALING THROUGH G-PROTEIN-COUPLED RECEPTORS 313

calmodulin, and which indirectly mediates the effects of Ca2+ by phos-

phorylation of other proteins.
15–46 Protein that binds to the cyclic AMP response elements found in the reg-
ulatory region of many genes activated by cyclic AMP.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
15–47 Different isoforms of protein kinase A in different cell types explain why
the effects of cyclic AMP vary depending on the target cell.
15–48 The activity of any protein regulated by phosphorylation depends on the
balance at any instant between the activities of the kinases that phospho-
rylate it and the phosphatases that dephosphorylate it.
15–49 Most intracellular signaling pathways provide multiple opportunities for
amplifying a response to an extracellular signal.

THOUGHT PROBLEMS
15–50 GPCRs activate G proteins by reducing the strength of GDP binding,
allowing GDP to dissociate and GTP, which is present at much higher
concentrations, to bind. How do you suppose the activity of a G protein
would be affected by a mutation that caused its affinity for GDP to be
reduced without significantly changing its affinity for GTP?
15–51 When adrenaline (epinephrine) binds to adrenergic receptors on the
surface of a muscle cell, it activates a G protein, initiating a signaling
pathway that results in breakdown of muscle glycogen. How would you
expect glycogen breakdown to be affected if muscle cells were injected
with a nonhydrolyzable analog of GTP, which can’t be converted to GDP?
Consider what would happen in the absence of adrenaline and after a
brief exposure to it.
15–52 Should RGS (regulator of G protein signaling) proteins be classified as
GEFs (guanine nucleotide exchange factors) or GAPs (GTPase-activating
proteins)? Explain what role this activity plays in modulating G-protein-
mediated responses in animals and yeasts.
15–53 What is “cyclic” about cyclic AMP?
15–54 Explain why cyclic AMP must be broken down rapidly in a cell to allow
rapid signaling.
15–55 You are trying to purify adenylyl cyclase from brain. The assay is based
on the conversion of -32P-ATP to cAMP. You can easily detect activity
in crude brain homogenates stimulated by isoproterenol, which binds to
-adrenergic receptors, but the enzyme loses activity when low-molec-
ular-weight cofactors are removed by dialysis. What single molecule do
you think you could add back to the dialyzed homogenate to restore
activity?
15–56 Propose specific types of mutations in the gene for the regulatory subunit
of cyclic-AMP-dependent protein kinase (PKA) that could lead to either a
permanently active PKA, or to a permanently inactive PKA.
15–57 Why do you suppose cells use Ca2+ (intracellular concentration 10–7 M)
for signaling rather than the more abundant Na+ (intracellular concen-
tration 10–3 M)?
15–58 EGTA chelates Ca2+ with high affinity and specificity. How would micro-
injection of EGTA affect glucagon-triggered breakdown of glycogen in
314 Chapter 15: Cell Signaling

cAMP Ca2+ P Figure 15–5 Integration of cyclic-AMP-

dependent and Ca2+-dependent signaling
pathways by phosphorylase kinase in
Ca2+ liver and muscle cells (Problem 15–59).
P G1P is glucose 1-phosphate, the product
Ca2+
of phosphorylase kinase, which uses
phosphate to cleave glucose units from
phosphorylase glycogen.
kinase

glycogen
phosphorylase

GLYCOGEN G1P

liver? How would it affect vasopressin-triggered breakdown of glycogen

in liver?
15–59 Phosphorylase kinase integrates signals from the cyclic-AMP-dependent
and Ca2+-dependent signaling pathways that control glycogen break-
down in liver and muscle
Problems (Figure 15–5). Phosphorylase kinase is
cells p15.15/15.05
composed of four subunits. One is the protein kinase that catalyzes the
addition of phosphate to glycogen phosphorylase to activate it for gly-
cogen breakdown. The other three subunits are regulatory proteins that
control the activity of the catalytic subunit. Two contain sites for phos-
phorylation by PKA, which is activated by cyclic AMP. The remaining
subunit is calmodulin, which binds Ca2+ when the cytosolic Ca2+ concen-
tration rises. The regulatory subunits control the equilibrium between
the active and inactive conformations of the catalytic subunit, with each
phosphate and Ca2+ nudging the equilibrium toward the active forma-
tion. How does this arrangement allow phosphorylase kinase to serve its
role as an integrator protein for the multiple pathways that stimulate gly-
cogen breakdown?
15–60 CaM-kinase II is a remarkable molecular memory device. How does
CaM-kinase II “remember” its exposure to Ca2+/calmodulin and why
does it eventually “forget”?
15–61 The outer segments of rod photoreceptor cells can be broken off, isolated,
and used to study the effects of small molecules on visual transduction
because the broken end of each segment remains unsealed. How would
you expect the visual response to be affected by the following additions?
A. An inhibitor of cyclic GMP phosphodiesterase.
B. A nonhydrolyzable analog of GTP.
C. An inhibitor of rhodopsin-specific kinase.
15–62 A rise in the cyclic GMP levels in smooth muscle cells causes relaxation
of the blood vessels in the penis, resulting in an erection. Explain how the
natural signal molecule NO and the drug Viagra® produce an increase in
cyclic GMP.
15–63 In muscle cells, adrenaline binds to the -adrenergic receptor to initi-
ate a signaling cascade that leads to the breakdown of glycogen (Figure
15–6). At what points in this pathway is the signal amplified?
15–64 A critical feature of all signaling cascades is that they must be turned off
rapidly when the extracellular signal is removed. Examine the signaling
cascade in Figure 15–6. Describe how each component of this signaling
pathway is returned to its inactive state when adrenaline is removed.
SIGNALING THROUGH G-PROTEIN-COUPLED RECEPTORS 315

adrenaline adenylyl Figure 15–6 Signaling cascade for

cyclase activation of glycogen breakdown by
adrenaline in muscle cells (Problem 15–63).
G1P is glucose 1-phosphate; cAMP bound
to the regulatory subunits of PKA is shown
α α CYTOSOL as red balls.
β
GDP γ GTP
ATP

dissociated
cAMP R-subunit
of PKA

inactive PKA

active
PKA P

phosphorylase
kinase

glycogen
phosphorylase

GLYCOGEN G1P

CALCULATIONS
15–65 In a classic paper, the number of -adrenergic receptors on the mem-
branes of frog erythrocytes
Problems was determined by using a competitive inhib-
p15.16/15.06
itor of adrenaline, alprenolol, which binds to the receptors 500 times
more tightly than adrenaline. These receptors normally bind adrenaline
and stimulate adenylyl cyclase activity. Labeled alprenolol was mixed
with erythrocyte membranes, left for 10 minutes at 37°C, and then the
membranes were pelleted by centrifugation and the radioactivity in the
pellet was measured. The experiment was done in two ways. The binding
of increasing amounts of 3H-alprenolol to a fixed amount of erythrocyte
membranes was measured to determine total binding. The experiment
was repeated in the presence of a vast excess of unlabeled alprenolol to
measure nonspecific binding. The results are shown in Figure 15–7.
A. On Figure 15–7, sketch the Problem
curve for specific
15-86 binding of alprenolol to
-adrenergic receptors. Has alprenolol binding to the receptors reached
saturation?
B. Assuming that one molecule of alprenolol binds per receptor, calculate
the number of -adrenergic receptors on the membrane of a frog eryth- 40
rocyte. The specific activity of the labeled alprenolol is 1 ×1013 cpm/
bound

mmol, and there are 8 ×108 frog erythrocytes per milligram of membrane total
(cpm/mg x 103)

30
protein.
3H-alprenolol

20
15–66 In visual transduction, one activated rhodopsin molecule leads to the
nonspecific
hydrolysis of 5 ×105 cyclic GMP molecules per second. One stage in this 10
enormous signal amplification is achieved by cyclic GMP phosphodies-
terase, which hydrolyzes 1000 molecules of cyclic GMP per second. The 0
0 2 4 6 8 10
additional factor of 500 could arise because one activated rhodopsin acti- 3H-alprenolol (nM)
vates 500 transducin (Gt) molecules, or because one activated transducin
activates 500 cyclic GMP phosphodiesterases, or through a combination Figure 15–7 Binding of 3H-alprenolol to frog
of both effects. One experiment to address this question measured the erythrocyte membranes (Problem 15–65).
316 Chapter 15: Cell Signaling

12 Figure 15–8 Binding of GppNp to rod-cell

membranes as a function of the fraction

(mmol/mole of rhodopsin)
10 of activated rhodopsin (Problem 15–66).
Background binding of GppNp to rod-

GppNp bound
8 cell membranes in the dark has been
subtracted from the values shown.
6
5.5 mmol/mole
4

2 1.1 × 10–3 %
activated
0
10-5 10-4 10-3 10-2 10-1
activated rhodopsin (%)

amount of GppNp (a nonhydrolyzable analog of GTP) that is bound by

transducin in the presence of different amounts of activated rhodopsin.
As indicated in Figure 15–8, 5.5 mmol of GppNp were bound per mole of
total rhodopsin when 0.0011% of the rhodopsin was activated.
A. Assuming that each transducin molecule binds one molecule of GppNp,
calculate the number of transducin molecules that are activated by each
Problems p15.18/15.08
activated rhodopsin molecule. Which mechanism of amplification does
this measurement support?
B. Binding studies have shown that transducin-GDP has a high affinity for
activated rhodopsin and that transducin-GTP has a low affinity; con-
versely, transducin-GTP has a high affinity and transducin-GDP has a
low affinity for cyclic GMP phosphodiesterase. Are these affinities con-
sistent with the mechanism of amplification you deduced from the above
experiment? Explain your reasoning.

DATA HANDLING
15–67 The mating behavior of yeast depends on signaling peptides termed
pheromones that bind to pheromone GPCRs (Figure 15–9). When the
-factor pheromone binds to a wild-type yeast cell, it blocks cell-cycle
progression, arresting proliferation until a mating partner is found. Yeast
mutants with defects in one or more of the components of the G protein
have characteristic phenotypes in the absence and in the presence of the
-factor pheromone (Table 15–1). Strains with defects in any of these
genes cannot undergo the mating response and are therefore termed
sterile.
A. Based on genetic analysis of the yeast mutants, decide which component
of the G protein normally transmits the mating signal to the downstream
effector molecules.
B. Predict the proliferation and mating phenotypes in the absence and
presence of the -factor pheromone of strains with the following mutant
G protein subunits:
α-factor

GTP EXTRACELLULAR

α GDP α
α-factor β β
receptor γ GTP γ
GDP

Figure 15–9 -Factor pheromone signaling via -factor GPCR and G protein (Problem 15–67).
SIGNALING THROUGH G-PROTEIN-COUPLED RECEPTORS 317

TABLE 15–1 Mating phenotypes of various mutant and nonmutant strains of

yeast (Problem 15–67).
Mutation Phenotype

Minus α factor Plus α factor

None (wild type) Normal proliferation Arrested proliferation, mating
response

subunit deleted Arrested proliferation Arrested proliferation, sterile

subunit deleted Normal proliferation Normal proliferation, sterile

subunit deleted Normal proliferation Normal proliferation, sterile

and deleted Normal proliferation Normal proliferation, sterile

and deleted Normal proliferation Normal proliferation, sterile

and deleted Normal proliferation Normal proliferation, sterile

1. An subunit that can bind GTP but cannot hydrolyze it.

2. An subunit with an altered N-terminus to which the fatty acid myris-
toylate cannot be added, thereby preventing its localization to the plasma
membrane.
3. An subunit that cannot bind to the activated pheromone receptor.
15–68 A particularly graphic illustration of the subtle, yet important, role of
cyclic AMP in the whole organism comes from studies of the fruit fly Dro-
sophila melanogaster. In search of the gene for cyclic AMP phosphodi-
esterase, one laboratory measured enzyme levels in flies with chromo-
somal duplications or deletions and found consistent alterations in flies
with mutations involving bands 3D3 and 3D4 on the X chromosome.
Duplications in this region have about 1.5 times the normal activity of
the enzyme; deletions have about half the normal activity.
An independent laboratory at the same institution was led to the
same chromosomal region through work on behavioral mutants of fruit
flies. The researchers had developed a learning test in which flies were
presented with two metallic grids, one of which was electrified. If the
electrified grid was painted with a strong-smelling chemical, normal flies
quickly learned to avoid it, even when it was no longer electrified. The
mutant flies, on the other hand, never learned to avoid the smelly grid;
they were aptly called Dunce mutants. The Dunce mutation was mapped
genetically to bands 3D3 and 3D4.
Is the learning defect really due to lack of cyclic AMP phosphodieste-
rase or are the responsible genes simply closely linked? Further experi-
ments showed that the level of cyclic AMP in Dunce flies was 1.6 times normal
phosphodiesterase

higher than in normal flies. Furthermore, sucrose-gradient analysis of

homogenates of Dunce and normal flies revealed two cyclic AMP phos-
activity

phodiesterase activities, one of which was missing in Dunce flies (Figure

15–10).
A. Why do Dunce flies have higher levels of cyclic AMP than normal flies?
B. Explain why homozygous (both chromosomes affected) duplications of Dunce

the nonmutant Dunce gene cause cyclic AMP phosphodiesterase levels top bottom
to be elevated 1.5-fold and why homozygous deletions of the gene reduce
Figure 15–10 Sucrose-gradient analysis
enzyme activity to half the normal value. of cyclic AMP phosphodiesterase activity
C. What would you predict would be the effect of caffeine, a phosphodieste- in homogenates of normal and Dunce
rase inhibitor, on the learning performance of normal flies? flies (Problem 15–68).

Problems p15.21/15.10
318 Chapter 15: Cell Signaling

(A) (B) (C) (D) Figure 15–11 Experimental set-up and

typical results of patch-clamp analysis of
K+-channel activation by acetylcholine
+ + + +
(Problem 15–69). The buffer is a salts
solution that does not contain nucleotides
intact buffer buffer buffer or Ca2+. In all these experiments,
cell + GTP acetylcholine is present inside the pipet,
as indicated by the plus sign. The current
through the membrane is measured in
picoamps (pA). In (C), the GTP is added
1 pA to the buffer.

200 seconds

15–69 Acetylcholine acts on muscarinic GPCRs in the heart to open K+ chan-

nels, thereby slowing the heart rate. This process can be directly studied
using the inside-out membrane patch-clamp technique. The external
surface of the membrane is in contact with the solution in the bore of
the pipet, and the cytoplasmic surface faces outward and can be exposed
readily to a variety of solutions (Figure 15–11). Receptors, G proteins,
and K+ channels remain associated with the membrane patch.
When acetylcholine is added to a pipet with a whole cell attached, K+
channels open as indicated by the flowp15.23/15.11
Problems of current (Figure 15–11A). Under
similar circumstances with a patch of membrane inserted into a buffered
salts solution, no current flows (Figure 15–11B). When GTP is added to
the buffer, current resumes (Figure 15–11C). Subsequent removal of GTP
stops the current (Figure 15–11D). The results of several similar experi-
ments to test the effects of different combinations of components are
summarized in Table 15–2.
A. Why do you think it is that G activated the channel when the complete
G protein did not? Is the active component of the G protein in this system
the same as the one that activates adenylyl cyclase in other cells?
B. Addition of GppNp (a nonhydrolyzable analog of GTP) causes the K+
channel to open in the absence of acetylcholine (Table 15–2, line 4). The
flow of current, however, rose very slowly and reached its maximum only
after a minute (compare with the immediate rise in Figure 15–11A and
C). How do you suppose GppNp causes the channels to open slowly in
the absence of acetylcholine?
C. To the extent that these experiments allow, draw a scheme for the activa-
tion of K+ channels in heart cells in response to acetylcholine.

TABLE 15–2 Responses of K+ channel to various experimental manipulations (Problem 15–69).

Additions
Acetylcholine Small molecules G-protein components K+ channel
1 + none none closed
2 + GTP none open
3 – GTP none closed
4 – GppNp none open
5 – none G protein closed
6 – none G closed

7 – none G open

8 – none boiled G protein closed

SIGNALING THROUGH G-PROTEIN-COUPLED RECEPTORS 319

MEDICAL LINKS
15–70 During a marathon, runners draw heavily on their internal reserves of
glycogen (carbohydrate) and triglycerides (fat) to fuel muscle contrac-
tion. Initially, energy is derived mostly from carbohydrates, with increas-
ing amounts of fat being used as the race progresses. If runners use up
their muscle glycogen reserves before they finish the race, they hit what
is known as “the wall,” a point of diminished performance that arises
because fatty acids from triglyceride breakdown cannot be delivered to
the muscles quickly enough to sustain maximum effort. One trick that
marathon runners use to avoid the wall is to drink a cup of strong black
coffee an hour or so before the race begins. Coffee contains caffeine,
which is an inhibitor of cyclic AMP phosphodiesterase. How do you sup-
pose inhibition of this enzyme helps them avoid the wall?
15–71 Patients with Oguchi’s disease have an inherited form of night blindness.
After a flash of bright light, these individuals recover their night vision
(become dark adapted) very slowly. Night vision depends almost entirely
on the visual responses of rod photoreceptor cells. What aspect of the
visual response in these patients’ rod cells do you suppose is defective?
What genes, when defective, might give rise to Oguchi’s disease?
15–72 The primary role of platelets is to control blood clotting. When they
encounter the exposed basement membrane (collagen fibers) of a dam-
aged blood vessel or a newly forming fibrin clot, they change their shape
from round to spiky and stick to the damaged area. At the same time, they
begin to secrete serotonin and ATP, which accelerate similar changes
in newly arriving platelets, leading to the rapid formation of a clot. The
platelet response is regulated by protein phosphorylation. Significantly,
platelets contain high levels of two protein kinases: PKC, which initiates
serotonin release, and myosin light-chain kinase, which mediates the
change in shape.
When platelets are stimulated with thrombin, the light chain of
myosin and an unknown protein of 40,000 daltons are phosphorylated.
When platelets are treated with a calcium ionophore, which increases
membrane permeability to Ca2+, only the myosin light chain is phos-
phorylated; when they are treated with diacylglycerol, only the 40-kD
protein is phosphorylated. Experiments using a range of concentra-
tions of diacylglycerol in the presence or absence of calcium ionophore
show that the extent of phosphorylation of the 40-kD protein depends
only on the concentration of diacylglycerol (Figure 15–12A). Serotonin
release, however, depends on diacylglycerol and the calcium ionophore
(Figure 15–12B).
A. Based on these experimental observations, describe the normal sequence
of molecular events that leads to phosphorylation of the myosin light

(A) PHOSPHORYLATION (B) SEROTONIN RELEASE

100 100
40-kD protein phosphorylation

+ ionophore
serotonin release
(% maximum)

(% maximum)

50 50 Figure 15–12 Treatment of platelets with

calcium ionophore and diacylglycerol
– ionophore
(Problem 15–72). (A) Effects on
phosphorylation of the 40-kD protein.
(B) Effects on serotonin release. Red
0
circles indicate the presence of calcium
0
0 10 20 0 10 20 ionophore and blue circles indicate its
diacylglycerol (µg/mL) diacylglycerol (µg/mL) absence.
320 Chapter 15: Cell Signaling

chain and the 40-kD protein. Indicate how the calcium ionophore and
diacylglycerol treatments interact with the normal sequence of events.
B. Why do you think serotonin release requires both calcium ionophore
and diacylglycerol?

SIGNALING THROUGH ENZYME-COUPLED

RECEPTORS
TERMS TO LEARN
Akt Ras
Cdc42 Ras-GAP
cytokine receptor Ras-GEF
cytoplasmic tyrosine kinase Ras–MAP-kinase signaling pathway
enzyme-coupled receptor Ras superfamily
ephrins receptor serine/threonine kinase
focal adhesion kinase (FAK) receptor tyrosine kinase (RTK)
JAK–STAT signaling pathway Rheb
Janus kinase (JAK) Rho
MAP kinase module Rho family
mTOR SH2 domain
phosphoinositide Smad family
phosphoinositide 3-kinase Src family
(PI 3-kinase) STATs
phospholipase C- (PLC ) TOR
PI-3-kinase–Akt pathway transforming growth factor- (TGF )
pleckstrin homology (PH) domain superfamily
protein tyrosine phosphatase tyrosine-kinase-associated receptor
Rac

DEFINITIONS
Match each definition below with its term from the list above.
15–73 The largest class of cell-surface-bound extracellular signal proteins.
15–74 Large family of structurally related, secreted, dimeric proteins that act as
hormones and local mediators to control a wide range of biological func-
tions in all animals.
15–75 Cell-surface receptor that when activated by ligand binding adds phos-
phates from ATP to tyrosine side chains in its own cytoplasmic domain.
15–76 The founding member of a superfamily of monomeric GTPases that help
to relay signals from cell-surface receptors to the nucleus.
15–77 A group of monomeric GTPases that regulate both the actin and microtu-
bule cytoskeletons.
15–78 Cytoplasmic tyrosine kinase present at cell–matrix junctions in associa-
tion with the cytoplasmic tails of integrins.
15–79 A kinase that is involved in intracellular signaling pathways activated by
cell-surface receptors and that phosphorylates inositol phospholipids at
the 3 position of the inositol ring.
15–80 Cell-surface receptor in which the cytoplasmic domain either has enzy-
matic activity itself or is associated with an intracellular enzyme.
15–81 Cell-surface receptor that activates a tyrosine kinase that is noncova-
lently bound to the receptor.
15–82 A three-component signaling module used in various signaling pathways
in eukaryotic cells.
SIGNALING THROUGH ENZYME-COUPLED RECEPTORS 321

15–83 One of several intracellular signaling pathways that leads from cell-sur-
face receptors to the nucleus, it is distinguished by providing one of the
more direct routes.
15–84 Protein domain found in intracellular signaling proteins by which they
bind to inositol phospholipids phosphorylated by PI 3-kinase.
15–85 A protein domain that is homologous to a region in Src, is present in
many proteins, and binds to a short amino acid sequence containing a
phosphotyrosine.
15–86 A crucial signaling protein in the PI-3-kinase–Akt signaling pathway, so
named because it is the target of rapamycin.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
15–87 Binding of extracellular ligands to receptor tyrosine kinases (RTKs) acti-
vates the intracellular catalytic domain by propagating a conformational
change across the lipid bilayer through a single transmembrane helix.
15–88 PI 3-kinase phosphorylates the inositol head groups of phospholipids at
the 3 position of the ring so that they can be cleaved by phospholipase C
to produce IP3.
15–89 Protein tyrosine phosphatases display exquisite specificity for their sub-
strates, unlike most serine/threonine protein phosphatases, which have
rather broad specificity.

THOUGHT PROBLEMS
15–90 Antibodies are Y-shaped molecules that carry two identical binding sites.
Imagine that you have obtained an antibody that is specific for the extra-
cellular domain of a receptor tyrosine kinase. If cells were exposed to the
antibody, would you expect the receptor tyrosine kinase to be activated,
inactivated, or unaffected? Explain your reasoning.
15–91 Genes encoding mutant forms of a receptor tyrosine kinase can be intro-
duced into cells that express the normal receptor from their own genes.
If the mutant genes are expressed at considerably higher levels than the
normal genes, what will be the consequences for receptor-mediated sig-
naling of introducing genes for the following mutant receptors?
A. A mutant receptor tyrosine kinase that lacks its extracellular domain.
B. A mutant receptor tyrosine kinase that lacks its intracellular domain.
15–92 The SH3 domain, which comprises about 60 amino acids, recognizes
and binds to structural motifs in other proteins. The motif recognized
by SH3 domains was found by constructing a fusion protein between an
SH3 domain and glutathione-S-transferase (GST). GST fusions allow for
easy purification using a glutathione affinity column, which binds GST
specifically. After tagging the purified GST–SH3 protein with biotin to
make it easy to detect, it was used to screen filters containing E. coli colo-
nies expressing a cDNA library. Two different clones were identified that
bound to the SH3 domain: in both cases, binding was shown to occur at
short proline-rich sequences.
A. Could you use biotin-tagged GST–SH2 proteins in the same way to find
cDNAs for proteins that bind to SH2 domains? Why or why not?
B. Many proteins bind to short strings of amino acids in other proteins. How
do you think these kinds of interactions differ from the kinds of interac-
tions found between the protein subunits of multisubunit enzymes?
322 Chapter 15: Cell Signaling

15–93 The Ras protein functions as a molecular switch that is turned on by a (A) RECEPTORS
guanine nucleotide exchange factor (GEF) that causes it to bind GTP. A 1 2 3
GTPase-activating protein (GAP) turns the switch off by inducing Ras to
hydrolyze its bound GTP to GDP much more rapidly than in the absence
of the GAP. Thus, Ras works like a light switch that one person turns CYTOSOL
on and another turns off. In a cell line that lacks the Ras-specific GAP,
what abnormalities in Ras activity, if any, would you expect to find in the Y Y
absence of extracellular signals, and in their presence? Y Y

15–94 What are the similarities and differences between the reactions that lead Y Y

to the activation of G proteins and those that lead to the activation of Ras?
kinase + – +
15–95 In principle, the activated, GTP-bound form of Ras could be increased P sites + + –
by activating a guanine nucleotide exchange factor (GEF) or by inactivat-
ing a GTPase-activating protein (GAP). Why do you suppose that Ras- (B) PROTEIN GEL
mediated signaling pathways always increase Ras-GTP by activating a 1 2 3 1+2 1+3 2+3
GEF rather than inactivating a GAP?
2
15–96 A single amino acid change in Ras eliminates its ability to hydrolyze GTP, 1
even in the presence of a GTPase-activating protein (GAP). Roughly
30% of human cancers have this change in Ras. You have just identified 3
a small molecule that prevents the dimerization of a receptor tyrosine
kinase that signals via Ras. Would you expect this molecule to be effec-
tive in the treatment of cancers that express this common, mutant form
(C) RADIOACTIVITY
of Ras? Why or why not?
1 2 3 1+2 1+3 2+3

DATA HANDLING 2
1
15–97 What does autophosphorylation mean? When a receptor tyrosine kinase
binds its ligand and forms a dimer, do the individual receptor molecules 3
phosphorylate themselves or does one receptor cross-phosphorylate the
other, and vice versa? To investigate this question, you’ve constructed
genes for three forms of a receptor tyrosine kinase: the normal form with Figure 15–13 Analysis of
an active kinase domain and three sites of phosphorylation; a large form autophosphorylation (Problem 15–97).
(A) Normal and mutant receptor tyrosine
that carries an inactivating point mutation in the kinase domain but kinases. P sites refers to the sites
retains the three phosphorylation sites; and a short version that has an of phosphorylation. (B) Expression
Problems p15.25/15.13
active kinase domain but is lacking the sites of phosphorylation (Figure of receptor tyrosine kinases.
15–13A). You express the genes singly and in combination in a cell line (C) Phosphorylation of receptor
that lacks this receptor tyrosine kinase, and then break open the cells and tyrosine kinases.
add the ligand for the receptor in the presence of radioactive ATP. You
immunoprecipitate the receptors and analyze them for expression levels
by staining for protein (Figure 15–13B) and for phosphorylation by auto-
radiography (Figure 15–13C).
A. What results would you expect on the autoradiograph if individual recep-
tors only phosphorylated themselves?
B. What would you expect if receptors cross-phosphorylated each other?
C. Which model for autophosphorylation do your data support?
15–98 When activated, the platelet-derived growth factor (PDGF) receptor
phosphorylates itself on multiple tyrosines. These phosphorylated tyros-
ines serve as assembly sites for several SH2-domain-containing proteins
that include phospholipase C- (PLC ), a Ras-specific GTPase-activat-
ing protein (GAP), a subunit of phosphoinositide 3-kinase (PI3K), and
a phosphotyrosine phosphatase (PTP) (Figure 15–14). PDGF binding
stimulates several changes in the target cell, one of which is an increase
in DNA synthesis, as measured by incorporation of radioactive thymi-
dine or bromodeoxyuridine into DNA.
To determine which of the bound proteins is responsible for acti-
vation of DNA synthesis, you construct several mutant genes for the
PDGF receptor that retain individual or combinations of tyrosine
SIGNALING THROUGH ENZYME-COUPLED RECEPTORS 323

PDGF Figure 15–14 The signaling complex

assembled on the PDGF receptor
(Problem 15–98). Numbers refer to the
positions of the phosphorylated amino
acids in the sequence of the PDGF
receptor.
P 740 P CYTOPLASM
PI3K PI3K
P 751 P

GAP P 771 P GAP

PTP P 1009 P PTP

1021
P P
PLCγ PLCγ

phosphorylation sites. When expressed in cells that do not make a PDGF

receptor of their own, each of the receptors is phosphorylated at its tyros-
ines upon binding of PDGF. in Figure 15–15, DNA synthesis is
As shownp15.26/15.14
Problems
stimulated to different extents in cells expressing the mutant receptors.
What roles do PI3K, GAP, PTP, and PLC play in the stimulation of
DNA synthesis by PDGF?

100 Figure 15–15 Stimulation of DNA

synthesis by the normal PDGF receptor
DNA synthesis
(% maximum)

and by receptors missing some

phosphorylation sites (Problem 15–98).
50 Stimulation by the normal receptor is set
arbitrarily at 100%. The presence of a
phosphorylation site (P site) is indicated
0 by +; absence of a site by –.
protein P site 1 2 3 4 5 6 7 8 9
PI3K 740, 751 + + – – – + – – –
GAP 771 + – + – – – + – –
PTP 1009 + – – + – – – + –
PLCγ 1021 + – – – + + + + –

15–99 MAP kinase kinase kinase (MAPKKK) activates MAP kinase kinase
(MAPKK) by phosphorylation of two serine side chains. Doubly phos-
phorylated (active) MAPKK, in turn, activates MAP kinase (MAPK) by
the phosphorylation of a threonine and a tyrosine. The doubly phospho-
rylated MAPK then phosphorylates a variety of target proteins to bring
about complex changes in cell behavior. It is possible to write down all of
MAPK
the rate equations for the individual steps in this activation cascade, as 100
MAPKK
well as for the removal of the phosphates (inactivation) by protein phos-
kinase activity

phatase, and to solve them by making reasonable assumptions about the

MAPKKK
concentrations of the proteins. The calculated plot of activation of the 50
kinases versus input stimulus is shown in Figure 15–16. Why is the very
steep response curve for MAPK a good thing for this signaling pathway?
15–100 An explicit assumption in theProblems
analysis in p15.27/15.15
Problem 15–99 is that the 0
0 1 2 3
components of the MAP kinase module operate independently of one
input stimulus
another, so that the dual phosphorylation events that activate MAPKK
and MAPK occur one at a time as molecules collide in solution. How do Figure 15–16 Stimulus–response curves
you suppose the curves in Figure 15–16 would change if a scaffold protein for the components of the MAPK cascade
(Problem 15–99). For ease of comparison,
held the kinases of the MAP kinase cascade together? Most MAP kinase the curves have been normalized so
modules are scaffolded. What is the advantage of linking these kinases that an input stimulus of 1 gives 50%
Problems p15.28/15.16
onto scaffold proteins? activation of the kinases.
324 Chapter 15: Cell Signaling

progesterone (A) POOLED OOCYTES

– +
Mos active (+ P )
Mos mRNA
inactive (– P)
100

MAP kinase
MEK1

% active
50

MAP kinase 0
0.001 0.01 0.1 1 10
progesterone (µM)

Figure 15–17 Progesterone-induced (B) INDIVIDUAL OOCYTES

MAP kinase activation, leading to mature
oocyte maturation (Problem 15–101). oocytes
–+ 0.03 µM progesterone
MEK1 is the frog’s MAP kinase
kinase.
0.1 µM progesterone
1 mm

0.3 µM progesterone

15–101 Activation (“maturation”) of frog oocytes is signaled through a MAP Figure 15–18 Activation of frog oocytes
kinase signaling module. An increase in the hormone progesterone trig- (Problem 15–101). (A) Phosphorylation
of MAP kinase in pooled oocytes.
Problems
gers the module by stimulating thep15.10/15.17
translation of the mRNA for Mos, (B) Phosphorylation of MAP kinase in
which is the frog’s MAP kinase kinase kinase (Figure 15–17). Maturation individual oocytes. MAP kinase was
is easy to score visually by the presence of a white spot in the middle of detected Problems p15.11/15.18
by immunoblotting using a
the brown surface of the oocyte (Figure 15–17). To determine the dose– MAP-kinase-specific antibody. The
first two lanes in each gel contain
response curve for progesterone-induced activation of MAP kinase,
nonphosphorylated, inactive MAP
you place 16 oocytes in each of six plastic dishes and add various con- kinase (–) and phosphorylated, active
centrations of progesterone. After an overnight incubation, you crush MAP kinase (+).
the oocytes, prepare an extract, and determine the state of MAP kinase
phosphorylation (hence, activation) by SDS polyacrylamide-gel electro-
phoresis (Figure 15–18A). This analysis shows a graded response of MAP
kinase to increasing concentrations of progesterone.
Before you crushed the oocytes, you noticed that not all oocytes in
individual dishes had white spots. Had some oocytes undergone partial
activation and not yet reached the white-spot stage? To answer this ques-
tion, you repeat the experiment, but this time you analyze MAP kinase
activation in individual oocytes. You are surprised to find that each
oocyte has either a fully activated or a completely inactive MAP kinase
(Figure 15–18B). How can an all-or-none response in individual oocytes
give rise to a graded response in the population? Akt Akt
construct Akt
T308A K179M
15–102 Akt is a key protein kinase in the signaling pathway that leads to cell IGF1 – + – + – +
growth. Akt is activated by a phosphatidylinositol-dependent pro- anti-Akt
tein kinase (PDK1), which phosphorylates threonine 308. At the same
time, serine 473 is phosphorylated. Your advisor has been unsuccess-
anti-P473
ful in purifying the protein kinase responsible for the phosphorylation
of serine 473, but you think you know what is going on. You construct
genes encoding two mutant forms of Akt: one carries a point mutation anti-P308
in the kinase domain, Akt-K179M, which renders it kinase-dead, and the 1 2 3 4 5 6
other carries a point mutation in the domain required to bind to PDK1
Figure 15–19 Expression levels of
(Akt-T308A), which cannot be activated by PDK1. You transfect each of various forms of Akt and their degree
these constructs, and a construct for wild-type Akt, into cells that do not of phosphorylation in the presence and
express their own Akt. You treat a portion of the cells with an insulin-like absence of IGF1 (Problem 15–102).
growth factor (IGF1), which activates PDK1, and analyze the phosphor- Anti-Akt recognizes all three forms of
Akt regardless of their phosphorylation
ylation state of the various forms of Akt using antibodies specific for Akt
state; anti-P473 specifically recognizes
or for particular phosphorylated amino acids (Figure 15–19). the phosphorylated serine at position
What is the identity of the enzyme that phosphorylates serine 473 on 473; anti-P308 specifically recognizes the
Akt? phosphorylated threonine at position 308.
ALTERNATIVE SIGNALING ROUTES IN GENE REGULATION 325

heptad Figure 15–20 Sequence elements in the

repeats SH3 SH2 P transcription factor that responds to IFN
(Problem 15–103).

15–103 Interferon- (IFN ) is a cytokine produced by activated T lymphocytes. It

binds to surface receptors on macrophages and stimulates their efficient
scavenging of invading viruses and bacteria via a JAK–STAT signaling
pathway. A number of genes are activated in response to IFN binding,
all of which contain a DNA sequence element with partial dyad symme-
try (TTCCXGTAA) that is required for the IFN response.
You have cloned the gene for the STAT transcription factor that is
activated in response to IFN binding. The sequence of the gene indi-
Problems
cates that the protein contains p15.30/15.20
several heptad repeat sequences near its
N-terminus—a common dimerization domain in many transcription
factors—and SH2 and SH3 domains adjacent to a site for tyrosine phos-
phorylation near the C-terminus (Figure 15–20). By making antibodies
to the protein, you show that it is normally located in the cytosol. After
15 minutes exposure to IFN , the protein becomes phosphorylated on a
tyrosine and moves to the nucleus.
Suspecting that tyrosine phosphorylation is the key to the regulation
of this transcription factor, you assay its ability to bind the DNA sequence
element in the presence of high concentrations of free phosphotyrosine
or when mixed with anti-phosphotyrosine antibodies. Both treatments
inhibit binding of the protein to DNA, as does treatment with a protein
phosphatase. Finally, you measure the molecular weight of the cytosolic
and nuclear forms of the protein, which suggest that the cytosolic form is
a monomer and the nuclear form is a dimer.
A. Do you think that phosphorylation of the transcription factor is neces-
sary for the factor to bind to DNA, or do you think phosphorylation is
required to create an acidic activation domain to promote transcription?
B. Bearing in mind that SH2 domains bind phosphotyrosine, how do you
suppose free phosphotyrosine might interfere with the activity of the
transcription factor?
C. How might tyrosine phosphorylation of the protein promote its dimeri-
zation? How do you think dimerization enhances its binding to DNA?

ALTERNATIVE SIGNALING ROUTES IN GENE

REGULATION
TERMS TO LEARN
-catenin I B Patched
circadian clock LDL-receptor-related Smoothened
Cubitus interruptus (Ci) protein (LRP) steroid hormone
Delta NF B proteins Wnt/ -catenin
Dishevelled Notch pathway
Frizzled nuclear receptor Wnt proteins
Hedgehog protein superfamily
iHog

DEFINITIONS
Match each definition below with its term from the list above.
15–104 Receptor protein involved in what may be the most widely used signal-
ing pathway in animal development; its ligands are cell-surface proteins
such as Delta.
326 Chapter 15: Cell Signaling

15–105 A family of secreted signal molecules that act as local mediators and
morphogens during development; they were initially discovered as the
products of the Wingless gene in flies and the Int1 gene in mice.
15–106 A signaling pathway activated by Wnt binding to both the Frizzled recep-
tor and the LRP co-receptor.
15–107 A group of secreted signal molecules that act as local mediators and mor-
phogens during development and whose effects are mediated through
the cell-surface receptor Patched and its binding partner Smoothened.
15–108 A target of Hedgehog signaling, this gene regulatory molecule is a full-
length gene activator in the presence of Hedgehog and a partially prote-
olyzed gene repressor in its absence.
15–109 Latent gene regulatory proteins that are present in most cells in both ani-
mals and plants and are central to many stress, inflammatory, and innate
immune responses.
15–110 Hydrophobic signaling molecule with a characteristic four-ringed struc-
ture derived from cholesterol.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
15–111 Signaling pathways that activate latent gene regulatory proteins depend
on regulated proteolysis to control activity and location.
15–112 Notch is both a cell-surface receptor and a latent gene regulatory protein.

15–113 Because one of the targets of NF B activation is the gene for I B , the
cytoplasmic inhibitor of NF B, a negative feedback loop is established
that limits the duration of the NF B response.

THOUGHT PROBLEMS
15–114 Why do signaling responses that involve changes in proteins already
present in the cell occur in milliseconds to seconds, whereas responses
that require changes in gene expression require minutes to hours?
15–115 Like Notch, the -amyloid precursor protein (APP) is cleaved near its
transmembrane segment to release an extracellular and an intracellu-
lar component. Explain how the fragments of APP relate to the amyloid
plaques that are characteristic of Alzheimer’s disease.
15–116 The Wnt planar polarity signaling pathway normally ensures that each
wing cell in Drosophila has a single hair. Overexpression of the Frizzled
gene from a heat-shock promoter (hs-Fz) causes multiple hairs to grow
from many cells (Figure 15–21A). This phenotype is suppressed if hs-Fz is
combined with a heterozygous deletion (Dsh /+) of the Dishevelled gene
(Figure 15–21B). Do these results allow you to order the action of Friz-
zled and Dishevelled in the signaling pathway? If so, what is the order? (A) (B)
Explain your reasoning.
15–117 There are two common mutational routes to the uncontrolled cell pro-
liferation and invasiveness that characterize cancer cells. The first is to
make a stimulatory gene (a proto-oncogene) hyperactive: this type of
mutation has a dominant effect so that only one of the cell’s two gene hs-Fz / + hs-Fz / +
copies needs to undergo change. The second is to make an inhibitory +/+ Dsh ∆ / +
gene (a tumor suppressor gene) inactive: this type of mutation usually is Figure 15–21 Pattern of hair growth
recessive so that both the cell’s gene copies must be inactivated. on wing cells in genetically different
Mutations of the Apc (adenomatous polyposis coli) gene occur in 80% Drosophila (Problem 15–116).

Problems p15.32/15.21
ALTERNATIVE SIGNALING ROUTES IN GENE REGULATION 327

of human colon cancers. Normal APC increases the affinity of the degra-
dation complex for -catenin, which in excess can enter the nucleus and
promote transcription of key target genes for cell proliferation. Given this
information, which category—oncogene or tumor suppressor—would
you expect the Apc gene to belong to? Why? HO

cholesterol
15–118 Latent gene regulatory proteins are prevented from entering the nucleus
until the cell receives an appropriate signal. List four ways by which cells CH2OH
keep gene regulatory proteins out of the nucleus, and give an example of
C O
a latent gene regulatory protein that is controlled by each mechanism.
HO OH
15–119 The steroid hormones cortisol, estradiol, and testosterone are all derived
from cholesterol by modifications that introduce polar groups such as –
OH and =O (Figure 15–22). If cholesterol itself was not normally found O
in cell membranes, do you suppose it could be used effectively as a hor- cortisol
mone, provided that an appropriate intracellular receptor was available?
15–120 Most people who are completely blind have circadian rhythms that OH
are ‘free-running;’ that is, their circadian rhythms are not synchro-
nized to environmental time cues and they oscillate on a cycle of about
24.5 hours. Why do you suppose the circadian clocks of blind people are
not entrained to the same 24-hour clock as the majority of the popula- O
tion? Can you guess what symptoms might be associated with a free-
testosterone
running circadian clock? Do you suppose that blind people have trouble
sleeping?
OH

DATA HANDLING
15–121 -Catenin can be phosphorylated by glycogen synthase kinase 3 (GSK3)
and it can be degraded in proteasomes. -Catenin could be sensitized for HO

degradation by phosphorylation, it could be protected from degradation estradiol

by phosphorylation, or its phosphorylation status could be irrelevant for
degradation. To distinguish among these possibilities, you generate cell Figure 15–22 Steroid hormones and
lines that express either a mutant GSK3 that cannot carry out phospho- their parent molecule, cholesterol
(Problem 15–119).
rylation, or a mutant -catenin that is missing its site of phosphorylation.
In the presence and absence of the proteasome inhibitor, ALLN, both
cell lines yield -catenin that migrates as a single band, with no slower
migrating bands visible. In contrast, nonmutant -catenin and GSK3 (A) HEDGEHOG PRECURSOR
Problems p15.01/15.22
display several slower migrating bands in the presence of ALLN, but no G257 C258
slower migrating bands in its absence. What is the relationship between 1 471
N- -C
-catenin phosphorylation and its degradation in proteasomes? Explain
your answer.
(B) TIME COURSE
15–122 The Hedgehog gene encodes the Hedgehog precursor protein, which is
471 amino acids long. The precursor protein (Figure 15–23A) is normally
0 0.5 1 2 4 8 16
cleaved between glycine 257 (G257) and cysteine 258 (C258) to gener-
hours
ate a fragment that is active in local and long-range signaling. Cleavage
is essential for signaling. You clone a segment of the Hedgehog gene
(C) CONCENTRATION DEPENDENCE
encoding a portion of the protein that includes the cleavage site and
the entire C-terminus. When you purify this protein and incubate it in
buffer, you observe cleavage over the course of several hours, as shown in 0.05 0.2 0.8 3.2 12.8
Figure 15–23B. If you vary its concentration over a 256-fold range and µM
assay cleavage after 4 hours of incubation, you observe the results shown
Figure 15–23 Mechanism of cleavage of
in Figure 15–23C. the Hedgehog precursor protein (Problem
A. Explain how these data support the idea that the Hedgehog precursor 15–122). (A) Site of cleavage (red arrow)
protein cleaves itself. How do they rule out the possibility that the puri- in the Hedgehog precursor protein.
fied protein is contaminated with a bacterial protease, for example? (B) Time course of cleavage of the
fragment of the precursor protein.
B. Does a molecule of precursor protein cleave itself, or does it cleave
(C) Dependence of cleavage on
another molecule of the precursor; that is, is the reaction intramolecular concentration of the precursor protein
or intermolecular? fragment.

Problems p15.34/15.23
328 Chapter 15: Cell Signaling

(A) CONSTRUCTS (B) FLIES (C) CELLS Figure 15–24 Fate of the fragments of
wt 1–257 (N) Hedgehog after cleavage (Problem 15–
123). (A) Constructs encoding different

e
bl
forms of the Hedgehog precursor protein.

cl N)
va

m
ce m
un 7 (
ea
w or

iu
iu
(B) Results of expression in Drosophila

25

ed
ed

lls
lls
ct
t
ve

1–

ce

m
m
embryos. (C) Results of expression in
G257 C258
insect cells. Hedgehog fragments were
1 471
wild type detected using antibodies specific for the
N-terminal segment.
1 A258
471
uncleavable

1 257
1–257 (N)

1 2 3 4 5 6 7 8

15–123 To find out what happens to the fragments of Hedgehog after cleavage,
you express three versions: wild-type Hedgehog precursor, an uncleav-
able form, and the N-terminal cleavage product (Figure 15–24A). In
fly embryos, the constructs behave as expected: wild-type Hedgehog is
cleaved, the uncleavable version is not, and the N-terminal segment is
expressed (Figure 15–24B). When wild-type Hedgehog and the N-termi-
nal segment are expressed in insect cells, however, the N-terminal seg-
ment from wild-type Hedgehog remains associated with the cells, while
the synthesized N-terminal segment is secreted into the medium (Fig-
ure 15–24C). Can you suggest possible explanations for the difference in
localization of the N-terminal segment?
15–124 If you overexpress various Hedgehog constructs (see Figure 15–24A) in
normal fly embyrosProblems
and examinep15.35/15.24
the pattern of Wnt expression (a well-
characterized target of Hedgehog signaling), you observe a striped pat-
tern of expression in all cases, but some constructs lead to thicker stripes
than normal (Figure 15–25).
A. Which part of the Hedgehog molecule is responsible for signaling?
B. All the cells in the embryo are overexpressing the various Hedgehog
constructs. Why is it, do you suppose, that you observe the same basic
striped pattern of Wnt expression in all of them?
C. Why do you see stripes of Wnt expression even in the absence of Hedge-
hog overexpression?
15–125 Studies with the fruit fly Drosophila provided initial clues to the complex
changes in patterns of gene expression that a simple hormone can trig-
ger. Drosophila larvae molt in response to an increase in the concentra-
tion of the steroid hormone ecdysone. The polytene chromosomes of the
Drosophila salivary glands are an excellent experimental system in which
to study the pattern of gene activity initiated by the hormone because
active genes enlarge into puffs that are visible in the light microscope.
Furthermore, the size of a puff is proportional to the rate at which the
gene is transcribed. Prior to addition of ecdysone, a few puffs—termed
intermolt puffs—are already active. Upon exposure of dissected salivary
glands to ecdysone, these intermolt puffs regress, and two additional sets

forms of Hedgehog
vector wild type uncleavable 1–257 (N) 257–471 (C)

Figure 15–25 Patterns of Wnt

expression in Drosophila embryos that
are overexpressing various Hedgehog
constructs (Problem 15–124). Wnt
expression was detected by in situ
1 2 3 4 5 hybridization.
SIGNALING IN PLANTS 329

of puffs appear. The early puffs arise within a few minutes after addition (A) NORMAL

puff size (% maximum)

of ecdysone; the late puffs arise within 4–10 hours. The concentration of intermolt early late
ecdysone does not change during this time period. The pattern of puff 100
appearance and disappearance is illustrated for a typical puff in each cat-
egory in Figure 15–26A. 50
Two critical experiments helped to define the relationships between
the different classes of puff. In the first, cycloheximide, which blocks pro-
tein synthesis, was added at the same time as ecdysone. As illustrated 0
0 2 4 6 8 10 12
in Figure 15–26B, under these conditions the early puffs did not regress time (hours)
and the late puffs were not induced. In the second experiment, ecdysone add
was washed out after a 2-hour exposure. As illustrated in Figure 15–26C, ecdysone
this treatment caused an immediate regression of the early puffs and a (B) CYCLOHEXIMIDE
premature induction of the late puffs.

puff size (% maximum)

A. Why do you think the early puffs didn’t regress and the late puffs weren’t 100
intermolt early
induced in the presence of cycloheximide? Why do you think the inter-
molt puffs were unaffected?
B. Why do you think the early puffs regressed immediately when ecdysone 50
was removed? Why do you think the late puffs arose prematurely under
these conditions?
0
C. Outline a model for ecdysone-mediated regulation of the puffing pattern. 0 2 4 6 8 10 12
time (hours)
add
SIGNALING IN PLANTS ecdysone

TERMS TO LEARN (C) ECDYSONE PULSE

puff size (% maximum)

auxin leucine-rich repeat (LRR) receptor kinase intermolt late
100
brassinosteroids phototropin
cryptochrome phytochrome early
ethylene plant growth regulator (plant hormone) 50

DEFINITIONS 0
0 2 4 6 8 10 12
Match each definition below with its term from the list above. time (hours)
add remove
15–126 A cytoplasmic serine/threonine kinase in plants that is activated by red ecdysone ecdysone
light and inactivated by far-red light.
15–127 Small gas molecule influential in various aspects of plant development, Figure 15–26 Puffing patterns in
including fruit ripening and leaf abscission. Drosophila salivary gland giant
chromosomes (Problem 15–125).
15–128 General term for a signal molecule that helps coordinate growth and (A) Normal puffing pattern.
(B) Puffing pattern in the presence
development in plants. Problems p15.09/15.26
of cycloheximide. (C) Puffing pattern
after removal of ecdysone.
15–129 Flavoprotein responsive to blue light, found in both plants and animals;
in animals it is involved in circadian rhythms.
15–130 A growth regulator that helps plants grow toward light, grow upward
rather than branch out, and extend their roots downward.
15–131 Common type of receptor serine/threonine kinase in plants, character-
ized by an extracellular portion rich in repeated segments containing a
high proportion of leucine.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
15–132 Even though plants and animals independently evolved multicellularity,
they use virtually all the same signaling proteins and second messengers
for cell–cell communication.
15–133 Remarkably, the auxin efflux transporters in the cap cells of the root
quickly redistribute themselves in response to a change in the direction
330 Chapter 15: Cell Signaling

(A) (B) (C) Figure 15–27 Three possible phylogenetic

plants fungi animals plants fungi animals plants animals fungi
relationships among plants, animals, and
fungi (Problem 15–134).

of the gravity vector, so that they pump auxin toward the side of the root
pointing downward.

THOUGHT PROBLEMS
Problems p15.37/15.27
15–134 The last common ancestor to plants and animals was a unicellular eukar-
yote. Thus, it is thought that multicellularity and the attendant demands
for cell communication arose independently in these two lineages. This
evolutionary viewpoint accounts nicely for the vastly different mecha-
nisms that plants and animals use for cell communication. Fungi use
signaling mechanisms and components that are very similar to those
used in animals. Which of the phylogenetic trees shown in Figure 15–27
does this observation support?
15–135 If signaling arose as a solution to the demands of multicellularity, how
then do you account for the very similar mechanisms of signaling that are
used in animals and the unicellular fungus Saccharomyces cerevisiae?
15–136 How is it that plant growth regulators can be present throughout a plant
and yet have specific effects on particular cells and tissues?

DATA HANDLING
15–137 The ripening of fruit is a complicated process of development, differ-
entiation, and death (except for the seeds, of course). The process is
triggered by minute amounts of ethylene gas. (This was discovered by
accident many years ago; the paraffin stoves used to heat greenhouses
in the olden days gave off enough ethylene to initiate the process.) The
ethylene is normally produced by the fruits themselves in a biochemical
pathway, the rate-limiting step of which is controlled by ACC synthase,
which converts S-adenosylmethionine to a cyclopropane compound
that is the immediate precursor of ethylene. Ethylene initiates a program
of sequential gene expression that includes the production of several
new enzymes, including polygalacturonase, which probably contributes
to softening the cell wall.
Your company, Agribucks, is trying to make mutant tomatoes that
cannot synthesize their own ethylene. Such fruit could be allowed to stay
longer on the vine, developing their flavor while remaining green and
firm. They could be shipped in this robust unripe state and exposed to
ethylene just before arrival at market. This should allow them to be sold
at the peak of perfection, and the procedure involves no artificial addi-
tives of any kind.
You decide to use an antisense approach, which works especially well
in plants. You place an ACC synthase cDNA into a plant expression vector
so that the gene will be transcribed in reverse, introduce it into tomato
cells, and regenerate whole tomato plants. Sure enough, ethylene pro-
duction is inhibited by 99.5% in these transgenic tomato plants, and their
fruit fails to ripen. But when placed in air containing a small amount of
ethylene, they turn into beautiful, tasty, ripe red fruit in about 2 weeks.
A. How do you imagine that transcribing the ACC synthase gene in reverse
blocks the production of ethylene?
B. Will you be a millionaire before you are 30?
SIGNALING IN PLANTS 331

MCAT STYLE
Passage 1 (Questions 15–138 to 15–140)
The Ras GTPase was first discovered as a gene that plays an important role in
transforming normal cells into cancer cells. Although Ras is normally activated by
a receptor tyrosine kinase (RTK), in many kinds of cancer the Ras gene has sus-
tained a mutation that makes the Ras protein hyperactive. This mutant Ras sends
unregulated signals that drive cell proliferation and contribute to tumor forma-
tion. Activated Ras binds and activates a MAP kinase kinase kinase (MAPKKK)
called Raf, which activates a MAP kinase kinase (MAPKK) called Mek, which then
activates a MAP kinase (MAPK) called Erk.
15–138 What kinds of mutations in the Ras gene could lead to hyperactive Ras?
I. Mutations that stimulate Ras to bind the Ras GTPase-activating protein
II. Mutations that decrease the ability of Ras to hydrolyze GTP
III. Mutations that block Ras binding to Ras-GEF
A. I
B. II
C. I and III
D. II and III
15–139 Mutant forms of Raf have also been found to play an important role in
cancer. A mutant called Raf-V600E causes Raf to become hyperactive
independently of signals from Ras. Drugs that inhibit Raf-V600E cause
rapid regression of tumors that express Raf-V600E. It was recently discov-
ered that treatment of cancer cells with these drugs increases Ras activ-
ity. Which of the following hypotheses best explains this observation?
A. Erk normally phosphorylates and inhibits Raf to restrict the duration of
RTK signaling. Inhibition of Erk therefore leads to increased Ras activity.
B. Erk normally phosphorylates the RTK and inhibits its signaling. Inhibi-
tion of Raf-V600E decreases Erk activity, which leads to increased RTK
signaling.
C. Erk normally phosphorylates the RTK and stimulates its signaling. Inhi-
bition of Raf-V600E increases RTK signaling, which leads to increased
Ras activity.
D. Raf normally phosphorylates the RTK and stimulates its signaling. Inhi-
bition of Raf-V600E therefore increases RTK signaling and increases Ras
activity.
15–140 Imagine you are working in a cancer clinic and encounter a patient with
a cancer that has the Raf-V600E mutation. You treat with a Raf inhibi-
tor, but the cancer does not respond. You are working on developing a
new treatment plan. Which of the following drugs would make the most
sense?
A. An inhibitor of Ras-GEF
B. An inhibitor of Erk
C. An inhibitor of Ras
D. An inhibitor of the RTK

Passage 2 (Questions 15–141 to 15–143)

Scaffold proteins are thought to constrain signaling specificity by bringing mul-
tiple kinases into close proximity to ensure that they signal to each other, rather
than to other proteins in the cell. The role of scaffolds in signaling was elucidated
in studies aimed at understanding the specificity of MAP kinase cascades in yeast.
In the MAP kinase cascade that prepares the cell for mating, an extracellular mat-
ing pheromone activates a G protein, which then activates a MAP kinase cascade
that includes Ste11 (MAPKKK), Ste7 (MAPKK), and Fus3 (MAPK) (Figure 15–28).
Activation of this cascade occurs over a time scale of 5–10 minutes. The MAP
kinase cascade that controls the response to starvation is activated by the Ras
332 Chapter 15: Cell Signaling

GTPase and also includes Ste11 and Ste7; however, the pathway works through a mating starvation
MAPK called Kss1, rather than Fus3. This cascade is activated over a time scale of
several hours. How can activation of the same kinases—Ste11 and Ste7—lead to Ste11 MAPKKK Ste11
completely different outputs? The discovery that Ste5 binds to the G protein and

Ste5
to all of the MAP kinases in the mating-response pathway led to the idea that Ste5 Ste7 MAPKK Ste7
acts as a scaffold to sequester the MAP kinases and link their activation to the acti-
vation of the G protein. Recent work suggests that scaffolds may play even more Fus3 MAPK Kss1
complex and interesting roles.
15–141 Which of the following observations would make you question the Figure 15–28 Mating-pheromone
sequestration model for scaffolds? activation of the MAP kinase cascade in
A. Ste7 activates Fus3 and Kss1 in vitro in the absence of Ste5. yeast (Problems 15–141 to 15–143).
B. Ste7 dissociates from Ste5 with a half-life of 5 seconds.
C. The Ste5 scaffold binds to Fus3, but does not bind to Kss1.
D. When activated in the mating pathway, Fus3 inactivates Kss1.
15–142 In one series of experiments, purified proteins were used to measure the
ability of Ste7 to phosphorylate Fus3 and Kss1. Kinase reactions were car-
ried out in the presence or absence of a domain of Ste5 that was found
to play an important role in Fus3 activation. Ste7 robustly phosphoryl-
ated Kss1 by itself, and addition of the Ste5 domain had no effect on the
Km or Kcat of the reaction. In contrast, the Ste5 domain gave a 5000-fold
increase in the rate of phosphorylation of Fus3 by Ste7, with little effect
on the Km. Which one of the following hypotheses could explain these
experimental results?
A. Fus3 induces a conformational change in Ste5 that activates Ste7. Figure 15-301
B. Ste5 alters the conformation of Fus3 to allow phosphorylation by Ste7.
C. The binding of the Ste5 domain to Ste7 activates its kinase activity. Problem 15-??
D. The Ste5 domain positions Ste7 so that it binds more tightly to Fus3.
15–143 In another series of in vitro experiments, activation of Fus3 by Ste7 was
measured in the presence of either full-length Ste5 or the Ste5 domain
that activates Fus3. The rate of activation of Fus3 in the presence of full-
length Ste5 was 10-fold lower than in the presence of the Ste5 domain. In
addition, it was found that cells expressing the Ste5 domain, instead of
full-length Ste5, inappropriately activated Fus3 in response to starvation.
Which of the following hypotheses would explain these observations?
I. Full-length Ste5 inhibits Fus3 activation in the absence of mating phe-
romone, ensuring that starvation signals relayed by Ste7 cannot activate
Fus3.
II. Full-length Ste5 contains a domain that promotes feedback activation of
Ste7 by Fus3.
III. In the absence of mating pheromone, full-length Ste5 is in a conforma-
tion that inhibits its ability to facilitate Fus3 activation. Mating-pherom-
one signaling triggers a conformational change in Ste5 that relieves the
inhibition.
A. I
B. II
C. I and III
D. II and III
Chapter 16 333

CHAPTER

The Cytoskeleton 16
FUNCTION AND ORIGIN OF THE CYTOSKELETON IN THIS CHAPTER

TERMS TO LEARN FUNCTION AND ORIGIN OF

cytoskeleton motor protein protofilament THE CYTOSKELETON

DEFINITIONS ACTIN AND ACTIN-BINDING

PROTEINS
Match each definition below with its term from the list above.
MYOSIN AND ACTIN
16–1 A linear chain of protein subunits joined end to end, which associates
laterally with other such chains to form cytoskeletal components. MICROTUBULES
16–2 System of protein filaments in the cytoplasm of a eukaryotic cell that
INTERMEDIATE FILAMENTS
gives the cell its shape and the capacity for directed movement.
AND SEPTINS
TRUE/FALSE CELL POLARIZATION AND
Decide whether each of these statements is true or false, and then explain why. MIGRATION

16–3 Microtubules determine the shape of the cell’s surface and are necessary
for whole-cell locomotion, and drive the pinching of one cell into two.
16–4 Even though the actin bundles at the cores of stereocilia on the hair cells
of the inner ear maintain their stable organization for the entire lifetime
of the animal, they are continuously remodeled and replaced on average
every 48 hours.
16–5 Because bacteria lack the elaborate networks of intracellular mem-
brane-enclosed organelles typical of eukaryotic cells, they do not require
cytoskeletal filaments.

THOUGHT PROBLEMS
16–6 In general terms, what are the cellular functions of intermediate fila-
ments, microtubules, and actin filaments?
16–7 If each type of cytoskeletal filament is made up of subunits that are held
together by weak noncovalent bonds, how is it possible for a human
being to lift heavy objects?
16–8 List differences between bacteria and animal cells that could have
depended on the appearance during evolution of some or all of the com-
ponents of the present eukaryotic cytoskeleton. Why do you suppose
a cytoskeleton might have been crucial for each of these differences to
evolve?
16–9 The amino acid sequences of actins and tubulins from all eukaryotes are
remarkably well conserved, yet the large numbers of proteins that inter-
act with these filaments are no more conserved than most other proteins
334 Chapter 16: The Cytoskeleton

in different species. How can it be that the filament proteins themselves

are highly conserved, while the proteins that interact with them are not?

CALCULATIONS
16–10 The average time it takes particles to diffuse a distance of x cm is
t = x2/2D
where t is the time in seconds and D is the diffusion coefficient, which is
a constant that depends on the size and shape of the particle.
A. How long would it take for a small molecule, a protein molecule, and a
membrane-enclosed vesicle to diffuse across a cell 10 μm in diameter? A
typical diffusion coefficient for a small molecule is 5 × 10–6 cm2/sec, for
a protein molecule 5 × 10–7 cm2/sec, and for a membrane vesicle 5 × 10–8
cm2/sec.
B. Why do you suppose a cell relies on the strategy of polymerizing and
depolymerizing cytoskeletal filaments, rather than on diffusion of the
filaments themselves, to accomplish its cytoskeletal rearrangements?

DATA HANDLING
16–11 One of the most striking examples of a purely actin-based cellular move-
ment is the extension of the acrosomal process of a sea cucumber sperm.
The sperm contains a store of unpolymerized actin in its head. When a
sperm makes contact with a sea cucumber egg, the actin polymerizes
rapidly to form a long spearlike extension. The tip of the acrosomal proc-
ess penetrates the egg, and it is probably used to pull the sperm inside.
Are actin monomers added to the base or to the tip of the acrosomal
bundle of actin filaments during extension of the acrosomal process? If
the supply of monomers to the site of assembly depends on diffusion, it
should be possible to distinguish between these alternatives by meas-
uring the length of the acrosomal process with increasing time. If actin
monomers are added to the base of the process, which is inside the head,
the rate of growth should be linear because the distance between the site
of assembly and the pool of monomers does not change with time. On the
other hand, if the subunits are added to the tip, the rate of growth should
decline progressively as the acrosomal process gets longer because the
monomers must diffuse all the way down the shaft of the process. In this
case, the rate of extension should be proportional to the square root of
time. Plots of the length of the acrosomal process versus time and the
square root of time are shown in Figure 16–1.
A. Are the ascending portions of the plots in Figure 16–1 more consistent
with the addition of actin monomers to the base or to the tip of the acro-
somal process?
B. Why do you suppose the process grows so slowly at the beginning and at
the end of the acrosomal reaction?

80

60
length ( µm)

40

20

0 Figure 16–1 Plots of the length of the

0 2 4 6 8 10 0 1 2 3 4 acrosome versus time and the square root
time (seconds) square root of time ( seconds0.5 ) of time (Problem 16–11).
ACTIN AND ACTIN-BINDING PROTEINS 335

ACTIN AND ACTIN-BINDING PROTEINS 4

TERMS TO LEARN C

fluorescence intensity
3
Arp2/3 complex formin
cell cortex treadmilling B
2
A
DEFINITIONS
1
Match each definition below with its term from the list above.
0
16–12 The process by which a polymeric protein filament is maintained at con- 0 30 60 90
stant length by addition of protein subunits at one end and loss of subu- time (seconds)
nits at the other.
16–13 Specialized layer of cytoplasm on the inner face of the plasma mem- Figure 16–2 Formation of actin filaments
over time, starting with purified actin
brane, rich in actin filaments. monomers that are labeled with a
16–14 Protein assembly that nucleates actin filament growth from the minus Problems
fluorescent p16.01/16.02
probe (Problem 16–17).
Upon polymerization, the fluorescence
end, allowing rapid growth at the plus end and forming a treelike web of of the probe increases, which allows
filaments. polymerization to be measured. The
intensity of fluorescence at zero seconds
is due to the background fluorescence of
TRUE/FALSE the actin monomers. The three phases of
polymerization are indicated as A, B, and
Decide whether each of these statements is true or false, and then explain why. C. Fluorescence intensity is measured in
arbitrary units.
16–15 In the treelike web of actin filaments that form the cell cortex, an Arp2/3
complex anchors each actin filament branch to the side of another actin
filament.
16–16 All the proteins that bind to the ends of actin filaments cap the ends to
prevent further polymerization.

THOUGHT PROBLEMS
16–17 A typical time course of polymerization of actin filaments from actin sub-
units is shown in Figure 16–2.
A. Explain the properties of actin polymerization that account for each of
the three phases of the polymerization curve.
B. How would the curve change if you doubled the concentration of actin?
Would the concentration of free actin at equilibrium be higher or lower
than in the original experiment, or would it be the same in both?
16–18 Figure 16–3 shows the equilibrium distribution of actin in free subunits
(monomers) and in filaments, as a function of actin concentration. Indi-
cate the critical concentration of actin on this diagram.
16–19 Imagine that the polymer in Figure 16–4A can add subunits at either
end, just like actin filaments (and microtubules). Imagine also three
hypothetical types of free subunit, as shown in Figure 16–4B. Each type

(A) POLYMERIZATION
filament
mass

monomer

or

actin concentration

(B) SUBUNIT CONFORMATIONS Figure 16–4 Polymerization of a polymer

Figure 16–3 Mass of actin monomers and (Problem 16–19). (A) Addition of a subunit
filaments as a function of actin concentration to a polymer. (B) Conformations of three
(Problem 16–18). 1 2 3 hypothetical subunits.
336 Chapter 16: The Cytoskeleton

of subunit can add to the polymer and, once added, it adopts the confor-
mation of the other subunits in the polymer (Figure 16–4A). For each of
these subunits, decide which end of the polymer, if either, will grow at the
faster rate when the concentration of that subunit is higher than the criti-
cal concentration required for polymerization. Explain your reasoning.
For any of the subunits, will there be a concentration at which one end
will preferentially grow while the other shrinks? Why or why not?
16–20 Some actin-binding proteins significantly increase the rate at which the
formation of actin filaments is initiated in the cytosol. How might such
proteins do this? What must they not do when binding the actin mono-
mers? Figure 16–5 Myosin-decorated actin
filament after a few minutes in a solution
16–21 The concentration of actin in cells is 50–100 times greater than the critical with excess actin monomers (Problem
concentration observed for pure actin in a test tube. How is this possible? Problems
16–23). p16.06/16.05
The shorter, thicker segment is
What prevents the actin subunits in cells from polymerizing into fila- the myosin-decorated actin filament.
ments? Why is it advantageous to the cell to maintain such a large pool of
actin subunits?
16–22 Cofilin preferentially binds to older actin filaments and promotes their
disassembly. How does cofilin distinguish old filaments from new ones?

DATA HANDLING
16–23 If you add short actin filaments marked by bound myosin heads (myosin-
decorated filaments) to a solution with an excess of actin monomers, wait
for a few minutes, and then examine the filaments by electron micros-
copy, you see the picture shown in Figure 16–5.
A. Which is the plus end of the myosin-decorated filament and which is the
minus end? Which is the “barbed” end and which is the “pointed” end?
How can you tell?
B. If you diluted the mixture so that the actin concentration was below the (A) MEASUREMENTS
critical concentration, which end would depolymerize more rapidly? 300
(molecules per second)

C. When the actin filament depolymerizes, why are subunits removed

exclusively from the ends and not from the middle of the filament? 200 plus end
growth rate

16–24 The growth rates at the plus and minus ends of actin filaments as a
function of actin concentration are shown in Figure 16–6A and, on an 100

expanded scale, in Figure 16–6B. minus end

A. The data in Figure 16–6A were gathered by measuring initial growth rates 0
at each actin concentration. Similar data gathered for any Michaelis– 0 5 10 15 20
Menten enzyme would generate a hyperbolic plot, instead of the linear actin (μM)

plots shown here. Why does the growth rate of actin filaments continue
to increase linearly with increasing actin concentration, whereas an (B) EXPANDED SCALE
enzyme-catalyzed reaction reaches a plateau with increasing substrate
15
concentration?
B. Figure 16–6B shows the filament growth rates at low actin concentration
on an expanded scale. Imagine that you could add actin filaments to a plus end
(molecules per second)

solution of actin subunits at the concentrations indicated as A, B, C, D, 10

growth rate

and E. For each of these concentrations, decide whether the added actin
filament would grow or shrink at its plus and minus ends. What is the
critical concentration for the plus end? What is the critical concentration 5
for the minus end? Would treadmilling occur at any of these concentra- minus end
tions? AB C D

E
Figure 16–6 Growth rates at the plus and minus ends of actin filaments
as a function of actin concentration (Problem 16–24). (A) Measurements of
–5
growth rates over a broad range of actin concentrations. (B) Growth rates 0 0.5 1.0
at low actin concentrations, shown on an expanded scale. actin (μM)
ACTIN AND ACTIN-BINDING PROTEINS 337

Figure 16–7 The kinetics of actin

ATP hydrolyzed (μmol)

150
30 polymerization and ATP hydrolysis 0.8 – cytochalasin B

(arbitrary units)
light scattering (Problem 16–25).

light scattering
100 0.6
20

viscosity
+ cytochalasin B

10 ATP hydrolysis 50 0.4

0 0 0.2
0 100 200 300
time (seconds) 0.0
0 5 10 15 20
time (minutes)
16–25 Your ultimate goal is to understand human consciousness, but your advi-
sor wants you to understand some basic facts about actin assembly first.
He tellsProblems
you that ATPp16.09/16.07
binds to actin monomers and is required for assem- Figure 16–8 Increase in the viscosity
bly. But, ATP hydrolysis is not necessary for polymerization since ADP of actin solutions in the presence and
can, under certain circumstances, substitute for the ATP requirement. absence of cytochalasin B (Problem
16–26).
Problems p16.12/16.08
ADP filaments, however, are much less stable than ATP filaments, sup-
porting your secret suspicion that the free energy of ATP hydrolysis really
is used to drive actin assembly.
Your advisor suggests that you make careful measurements of the
quantitative relationship between the number of ATP molecules hydro-
lyzed and the number of actin monomers linked into polymer. The
experiments are straightforward. To measure ATP hydrolysis, you add
32P-ATP to a solution of polymerizing actin, take samples at intervals,
and determine how much radioactive phosphate has been produced. To
assay polymerization, you measure the increase in light scattering that
is caused by formation of the actin filaments. Your results are shown in
Figure 16–7. Your light-scattering measurements indicate that 20 μmoles
of actin monomers were polymerized. Since the number of polymerized
actin monomers matches exactly the number of ATP molecules hydro-
lyzed, you conclude that one ATP is hydrolyzed as each new monomer is
added to an actin filament.
When you show your advisor the data and tell him your conclusions, – cytochalasin B
he smiles and very gently tells you to look more closely at the graph. He
minus plus
says your data prove that actin can polymerize without ATP hydrolysis. end end
A. What does your advisor see in the data that you have overlooked?
B. What do your data imply about the distribution of ATP and ADP in
polymerizing actin filaments?
16–26 Cytochalasin B strongly inhibits certain forms of cell motility, such
as cytokinesis and the ruffling of growth cones, and it dramatically
decreases the viscosity of gels formed with mixtures of actin and a wide
variety of actin-binding proteins. These observations suggest that cyto- + cytochalasin B
chalasin B interferes with the assembly of actin filaments. In the classic
experiment that defined its mechanism, short lengths of actin filaments
were decorated with myosin heads and then mixed with actin subunits
in the presence or absence of cytochalasin B. Assembly of actin filaments
was measured by assaying the viscosity of the solution (Figure 16–8) and
by examining samples by electron microscopy (Figure 16–9).
A. Suggest a plausible mechanism to explain how cytochalasin B inhibits
actin filament assembly. Account for the appearance of the filaments
in the electron micrographs and the viscosity measurements (both the
altered rate and extent). Figure 16–9 Appearance of typical actin
B. The normal growth characteristics of an actin filament and the actin- filaments formed in the presence and
binding properties of cytochalasin B argue that actin monomers undergo absence of cytochalasin B (Problem
16–26). The decorated actin filaments
a conformational change upon addition to an actin filament. How so? present before the addition of actin
monomers are shown at the top of each
16–27 Phalloidin, which is a toxic peptide from the mushroom Amanita phal- set of three. Filaments present after
loides, binds to actin filaments. Phalloidin tagged with a fluorescent Problems
increasing timesp16.13/16.09
of incubation with actin
probe is commonly used to stain actin filament assemblies in cells monomers (red circles) are shown below.
338 Chapter 16: The Cytoskeleton

(A) CELLS (B) LOW (C) HIGH Figure 16–10 Binding of phalloidin to
actin filaments (Problem 16–27).
(A) The actin cytoskeleton stained with
fluorescent phalloidin. (B) An actin
filament bound by gold-tagged phalloidin
at low contrast. (C) The same actin
filament as in (B), but at high contrast.
Bright bands mark the positions of six
11 nm gold particles.

20 μm

(Figure 16–10A). If phalloidin is attached to a gold particle instead, its

binding to actin filaments can be examined at high resolution by scan-
ning transmission electron microscopy. Figure 16–10B shows a micro-
Problems p16.14/16.10
graph of an actin filament with bound phalloidin and Figure 16–10C
shows the same picture with the contrast adjusted so that only the points
of highest intensity (the gold particles) are visible. Does phalloidin bind
to every actin subunit? How can you tell?
16–28 Isolated bundles of actin filaments from the acrosomal processes of
Limulus polyphemus (horseshoe crab) sperm have readily distinguish-
able plus ends (tapered) and minus ends (blunt). Assembly at the ends
of such bundles was used to determine the mechanism of action of phal-
loidin, which has a marked effect on actin assembly. When phalloidin is (A) ASSEMBLY
mixed with actin in a molar ratio of at least 1:1, the growth rate increases 45
at both ends, as shown for minus ends in Figure 16–11A. Because growth
(molecules per second)

rate = kon[actin]initial – koff, these plots have the form y = mx + b, so that the 30
+ phalloidin
slope of the line equals kon and the y intercept equals –koff.
growth rate

15
A. By analyzing the on and off rates, decide how phalloidin increases the – phalloidin
growth rate of actin filaments. Explain your reasoning. 0
B. In Figure 16–11B, actin filaments grown in the presence or absence of
phalloidin were diluted in the absence of actin monomers and their dis- –15
assembly was assayed. Do these results confirm or contradict your con-
0 1 2 3 4
clusions from part A? Explain your answer.
actin (µM)
C. What is the critical concentration for actin assembly at the minus end
in the absence of phalloidin? What is the critical concentration for actin
(B) DISASSEMBLY
assembly at the minus end in the presence of phalloidin?
D. Propose a molecular mechanism for the effects of phalloidin on actin
assembly.
percentage of length

+ phalloidin
before dilution

100
16–29 Swinholide A is a member of a class of lipophilic compounds termed
macrolides, which include a number of useful antibiotics such as
erythromycin, that are synthesized by Actinomycetes. Swinholide A is 50
– phalloidin
a “twin” molecule, composed of two identical halves (Figure 16–12A).
When added to cells growing in culture, swinholide A disrupts the actin
cytoskeleton. Your advisor has shown conclusively that swinholide A 0
0 10 20
binds a pair of actin monomers. She suspects that swinholide A causes time after dilution (minutes)
actin filaments to depolymerize by sequestering actin subunits in a non-
functional dimeric form and thus accelerating depolymerization through Figure 16–11 Effects of phalloidin on
mass-action effects. She wants you to test this hypothesis. actin filaments (Problem 16–28).
You prepare actin filaments tagged with a probe that fluoresces (A) Growth rates at the minus ends of
acrosomal bundles in the presence and
intensely in the filament but much less so in the free subunits (or swin- absence of phalloidin. (B) Disassembly
holide-bound subunits). This allows you to follow depolymerization read- of actin filaments upon dilution in the
ily and rapidly as a loss of fluorescence. Just as your advisor predicted, presence and absence of phalloidin.

Problems p16.15/16.11
ACTIN AND ACTIN-BINDING PROTEINS 339

(A) SWINHOLIDE A STRUCTURE (B) DISASSEMBLY ASSAY (C) SWINHOLIDE A TITRATION

depolymerization rate (nm/sec)

0.16

fluorescence (arbitrary units)

5
0
0.12 4

152
3
0.08
280
480 2
0.04 1000
1

0.00 0
0 10 20 30 40 50 0 50 100 150
time (seconds) swinholide A (nM)

Figure 16–12 Effects of swinholide A on actin filaments (Problem 16–29). (A) Structure of swinholide A. The identical halves of
swinholide A are arranged head to tail, so that if the molecule were rotated 180° about the indicated axis (circle with an X in it),
it would superimpose on itself. For this reason it is said to have a twofold axis of symmetry. (B) Time course of actin filament
depolymerization in the presence and absence of swinholide A. Numbers indicate the concentration of swinholide A (nM) used in
each depolymerization assay. (C) Initial rates of depolymerization as a function of swinholide A concentration.

Problems p16.16/16.12

depolymerization increases in the presence of increasing concentrations (A) NO ADDED PROTEIN

of swinholide A (Figure 16–12B). But you notice two features of these
curves that suggest to you that swinholide A may actually sever actin fila- filament
ments. One of these features is illustrated in Figure 16–12C, which shows

mass
a nonlinear dependence of the initial rate of depolymerization on the monomer
concentration of swinholide A. A simple mass-action effect—the seques-
tering of actin monomers by binding to swinholide A—predicts a linear
dependence; however, increasing increments in swinholide A concen-
tration have a progressively greater effect on depolymerization. 0 0.2 0.4 0.6 0.8
actin concentration ( μM)
A. In Figure 16–12B, why does fluorescence reach a plateau value (at about
0.03) instead of decreasing to zero?
(B) PROTEIN 1
B. The other odd feature you noticed about depolymerization in the pres-
ence of swinholide A (Figure 16–12B) is that the lines have a “hump” in
them in the first few seconds (when their fluorescence is still above 0.12).
filament
Why does this hump suggest that swinholide A severs actin filaments?
mass

C. Assuming that swinholide A does sever actin filaments, is one molecule

enough, or are multiple molecules needed? How do you know? monomer

16–30 Accessory proteins that regulate the nucleation of actin filaments pro-
mote binding of the Arp2/3 complex to actin filaments so that most new 0 0.2 0.4 0.6 0.8
filaments form as branches from existing ones. These proteins could actin concentration (μ M)
stimulate Arp2/3 binding to the sides of existing filaments or to the plus
end of a growing filament in a way that does not interfere with growth. (C) PROTEIN 2
Both possibilities would yield the final characteristic branched network
of filaments. To distinguish between these alternatives, you mix the regu- monomer
latory proteins with the Arp2/3 complex and actin subunits in the pres-
ence of actin filaments that are capped at their plus ends. After a short
mass

incubation you examine the resulting structures by electron microscopy. filament

How will this experiment distinguish between these alternatives? What
structures would you expect to see according to each model for nuclea-
tion by the Arp2/3 complex? 0 0.2 0.4 0.6 0.8
actin concentration (μ M)
16–31 You have two proteins that you suspect cap the ends of actin filaments. To Figure 16–13 Effects of two proteins
determine whether they do and, if so, which protein caps which end, you on actin polymerization (Problem 16–31).
measure filament formation as a function of actin concentration in the (A) Polymerization of pure actin.
absence of either protein, in the presence of protein 1, and in the pres- (B) Actin polymerization in the presence
of protein 1. (C) Actin polymerization in
ence of protein 2 (Figure 16–13). Which protein caps the plus end and the presence of protein 2. The mass of
which caps the minus end? How can you tell? Give examples of proteins actin, as monomers or filaments, was
in the cell that you would expect to behave like protein 1 and protein 2. determined at equilibrium.

Problems p16.25/16.13
340 Chapter 16: The Cytoskeleton

(A) TIME-LAPSE MOVIE (B) EM Figure 16–14 Movement of a bacterium

through the cytosol on a comet tail of
actin filaments (Problem 16–32). (A) Time-
lapse movie. (B) Electron micrograph. The
bacterium is 2 μm in length.
10 μm

0 10 20 30 40 50 60 70 2 μm
time (seconds)

MEDICAL LINKS
16–32 The intracellular pathogenic bacterium Listeria monocytogenes propels
Problems
itself through the cytosol p16.23/16.14
on a comet tail of actin filaments (Figure 16–14).
Remarkably, only a single bacterial protein, the transmembrane protein
ActA, is required for this motility. ActA is distributed unequally on the
surface of the bacterium, with maximum concentrations at the pole in
contact with the actin tail. The effects of ActA on actin polymerization
in the presence and absence of the Arp2/3 complex (ARP) are shown in
Figure 16–15A. The first few seconds of the reactions are shown on an
expanded scale in Figure 16–15B. Polymerization of actin was followed
using pyrene-actin, which exhibits much higher fluorescence intensity
when actin is polymerized.
A. What are the effects of ActA and the Arp2/3 complex, separately and
together, on the rate of nucleation of actin filaments? Explain your
answer.
B. How do you suppose that the polymerization of actin by ActA and the
Arp2/3 complex propels the bacterium across the cell? In the comet
tail of actin filaments, which ends—plus or minus—do you suppose are
pointed at the bacterium?

(A) KINETICS OF ACTIN ASSEMBLY (B) EXPANDED SCALE Figure 16–15 Effects of ActA and
10,000 the Arp2/3 complex (ARP) on actin
actin + ARP polymerization (Problem 16–32).
(A) Kinetics of polymerization of actin
7500
(arbitrary units)

in the presence of ActA and the Arp2/3

fluorescence

actin + ARP + ActA

actin + ARP + ActA complex. (B) Kinetics of polymerization
5000 on an expanded scale. In all cases, actin
actin alone,
actin + ActA
was present at 2 μM, and ActA and the
2500 actin alone,
Arp2/3 complex were present at 30 nM.
actin + ActA,
actin + ARP
0
0 500 1000 1500 2000 10 40 70 100
time (seconds) time (seconds)

MYOSIN AND ACTIN

TERMS TO LEARN
myofibril Problems p16.24/16.15
stress fiber
myosin

DEFINITIONS
Match each definition below with its term from the list above.
16–33 The motor protein in muscle that generates the force for muscle contrac-
tion.
MYOSIN AND ACTIN 341

16–34 Long, highly organized bundle of actin, myosin, and other proteins in
the cytoplasm of muscle cells that contracts by a sliding-filament mecha-
nism.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
16–35 Myosin II molecules have two motor domains and a rodlike tail that
allows them to assemble into bipolar filaments, which are crucial for the
efficient sliding of oppositely oriented actin filaments past each other.
16–36 Motor neurons trigger action potentials in muscle cell membranes that
open voltage-sensitive Ca2+ channels in T tubules, allowing extracellular
Ca2+ to enter the cytosol, bind to troponin C, and initiate rapid muscle
contraction.
16–37 When activated by Ca2+ binding, troponin C causes troponin I to release
its hold on actin, thereby allowing the tropomyosin molecules to shift
their positions slightly so that the myosin heads can bind to the actin fila-
ments.

THOUGHT PROBLEMS
16–38 Living systems continually transform chemical free energy into motion.
Muscle contraction, ciliary movement, cytoplasmic streaming, cell divi-
sion, and active transport are examples of the ability of cells to transduce
chemical free energy into mechanical work. In all these instances, a pro-
tein motor harnesses the free energy released in a chemical reaction to
drive an attached molecule (the ligand) in a particular direction. Analysis
of free-energy transduction in favorable biological systems suggests that
a set of general principles governs the process in cells.
1. A cycle of reactions is used to convert chemical free energy into
mechanical work.
2. At some point in the cycle a ligand binds very tightly to the protein
motor.
3. At some point in the cycle the motor undergoes a major conforma-
tional change that alters the physical position of the ligand.
4. At some point in the cycle the affinity for the ligand markedly
decreases, allowing the ligand to detach from the motor.
These principles are illustrated by the two cycles for free-energy
transduction shown in Figure 16–16: (1) the sliding of actin and myosin
filaments against each other and (2) the active transport of Ca2+ from
(A) SLIDING FILAMENT (B) ACTIVE TRANSPORT

actin
Ca2+
Pi ATP ADP
INSIDE Pi
ATP ADP ADP
Ca Ca
myosin
Pi ADP

ATP Pi
ATP Pi Pi
Ca
OUTSIDE
Ca2+

Figure 16–16 Transduction of chemical free energy into mechanical work (Problem 16–38). (A) Sliding of actin filaments relative to
myosin filaments. (B) Active transport of Ca2+ from the inside to the outside of the cell. In both cycles, arrows are drawn in only one
direction to emphasize their normal operation. The phosphorylation and dephosphorylation steps in the active transport cycle are
catalyzed by enzymes that are not shown in the diagram.
342 Chapter 16: The Cytoskeleton

inside the cell, where its concentration is low, to the cell exterior, where
its concentration is high. An examination of these cycles underscores the
principles of free-energy transduction.
A. What is the source of chemical free energy that powers these cycles, and
what is the mechanical work that each cycle accomplishes?
B. What is the ligand that is bound tightly and then released in each of
the cycles? Indicate the points in each cycle where the ligand is bound
tightly.
C. Identify the conformational changes in the protein motor that constitute
the “power stroke” and “return stroke” of each cycle.
16–39 Which one of the following changes takes place when a skeletal muscle
contracts?
A. Z discs move farther apart.
B. Actin filaments contract.
C. Myosin filaments contract.
D. Sarcomeres become shorter.
16–40 Two electron micrographs of striated muscle in longitudinal section are
shown in Figure 16–17. The sarcomeres in these micrographs are in two
different states of contraction.

(A)

(B) 1 μm

Figure 16–17 Two electron micrographs

of striated muscle in longitudinal section
(Problem 16–40). The micrograph in
(B) is a much lighter exposure than the
one in (A). At the same exposure, the
entire space between the thin dark lines
in (B) would be as dark as the fat dark
1 μm band in (A).
MYOSIN AND ACTIN 343

CAGED ATP NH2 Figure 16–18 Caged ATP (Problem 16–43).

N N

CH3 O O O N N
C O P O P O P O H 2C
O
H O– O– O– H H
NO2 H

OH OH
CH3
C LASER
O LIGHT
NH2
NO2
N N

O O O N N
–O P O P O P O H 2C
O
O– O– O– H H
H

ATP OH OH

A. Using the micrograph in Figure 16–17A, identify the locations of the fol-
lowing:
1. Dark band
2. Light band
3. Z disc
4. Myosin II filaments
5. Actin filaments (show plus and minus ends)
6. -Actinin
7. Nebulin
8. Titin
B. Locate the same features on the micrograph in Figure 16–17B. Be careful!
16–41 Problems
Troponin molecules are evenly spaced p16.35/16.18
along an actin filament with one
troponin bound at every seventh actin molecule. How do you suppose
troponin molecules can be positioned this regularly?
16–42 What two major roles does ATP hydrolysis play in muscle contraction?

DATA HANDLING
16–43 As a laboratory exercise, you and your classmates are carrying out experi-
ments on isolated muscle fibers using “caged” ATP (Figure 16–18). Since
caged ATP does not bind to muscle components, it can be added to a
muscle fiber without stimulating activity. Then, at some later time it can
be split by a flash of laser light to release ATP instantly throughout the add caged laser
muscle fiber. ATP flash
To begin the experiment, you treat an isolated, striated muscle fiber
with glycerol to make it permeable to nucleotides. You then suspend
tension

it in a buffer containing ATP in an apparatus that allows you to meas- remove

ure any tension generated by fiber contraction. As illustrated in Figure ATP

16–19, you measure the tension generated after several experimental

manipulations: removal of ATP by dilution, addition of caged ATP, and
activation of caged ATP by laser light. You are somewhat embarrassed
because your results are very different from everyone else’s. In checking 0 2 4 6 8 10

over your experimental protocol, you realize that you forgot to add Ca2+ time (minutes)
to your buffers. The teaching assistant in charge of your section tells you Figure 16–19 Tension in a striated muscle
that your experiment is actually a good control for the class, but you will fiber as a result of various experimental
have to answer the following questions to get full credit. manipulations (Problem 16–43).

Problems p16.36/16.19
344 Chapter 16: The Cytoskeleton

Figure 16–20 Tension as a function

2.0 2.2 of sarcomere length during isometric

tension (% of maximum)
1.6 contraction (Problem 16–44).
100
II
75 III

50
1.3 I
IV 3.6
25

0
1 2 3 4
sarcomere length (μm)

A. Why did the ATP in the suspension buffer not cause the muscle fiber to
contract?
B. Why did the subsequent removal of ATP generate tension? Why did ten-
sion develop so gradually? (If our muscles normally took a full minute to
contract, we would move very slowly.)Problems p16.37/16.20
C. Why did laser illumination of a fiber containing caged ATP lead to relaxa-
tion?
16–44 Detailed measurements of sarcomere length and tension during iso-
metric contraction in striated muscle provided crucial early support for
the sliding-filament model of muscle contraction. Based on your under-
standing of the sliding-filament model and the structure of a sarcomere,
propose a molecular explanation for the relationship of tension to sar-
comere length in the portions of Figure 16–20 marked I, II, III, and IV. (In
this muscle, the length of the myosin filament is 1.6 μm, and the lengths
of the actin thin filaments that project from the Z discs are 1.0 μm.)

MICROTUBULES
TERMS TO LEARN
axoneme -tubulin ring complex ( -TuRC)
centriole kinesin
centrosome kinesin-1
cilium microtubule-associated protein (MAP)
dynamic instability microtubule-organizing center (MTOC)
dynein tubulin
flagellum

DEFINITIONS
Match each definition below with its term from the list above.
16–45 The property of sudden conversion from growth to shrinkage, and vice
versa, in a protein filament such as a microtubule or an actin filament.
16–46 Centrally located organelle of animal cells that is the primary microtu-
bule-organizing center and acts as the spindle pole during mitosis.
16–47 Protein assembly containing a special form of tubulin, along with other
proteins, that is an efficient nucleator of microtubule growth.
16–48 Short cylindrical array of microtubules, a pair of which are embedded in
the major microtubule-organizing center of an animal cell.
16–49 A member of the family of motor proteins that move along microtubules
by walking toward the minus end.
16–50 A motor protein that moves along microtubules by walking toward the
plus end.
MICROTUBULES 345

16–51 Bundle of microtubules and associated proteins that forms the core of a
cilium or flagellum in a eukaryotic cell and is responsible for their move-
ments.
16–52 Long, hairlike protrusion from the surface of a eukaryotic cell whose
undulations drive the cell through a liquid medium. seam
+ end
TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why. 8 nm

16–53 The structural polarity of all microtubules is such that -tubulin is

exposed at one end and -tubulin is exposed at the opposite end.
16–54 The role of GTP hydrolysis in tubulin polymerization is similar to the
role of ATP hydrolysis in actin polymerization: both serve to weaken the
bonds in the polymer and thereby promote depolymerization. β
α
16–55 All microtubule-organizing centers contain centrioles that help nucleate – end
microtubule polymerization.
tubulin
protofilament microtubule
16–56 In most animal cells, minus-end directed microtubule motors deliver
their cargo to the periphery of the cell, whereas plus-end directed micro- Figure 16–21 Structure of a
tubule motors deliver their cargo to the interior of the cell. 13-protofilament microtubule, showing
the seam between the first and thirteenth
protofilaments (Problem 16–58).
THOUGHT PROBLEMS
16–57 Why do you suppose it is much easier to add tubulin to existing microtu-
bules than to start a new microtubule from scratch?
16–58 In a 13-filament microtubule, the majority of lateral interactions are
between like subunits, with -tubulin binding to -tubulin, and -tubulin
binding to -tubulin. Between the first and thirteenth protofilaments,
however, there is a seam at which -tubulin interacts with -tubulin
(Figure 16–21). Are these heterotypic interactions ( with ) likely to be Problems p16.03/16.21
stronger than, weaker than, or the same strength as homotypic interac-
tions ( with , or with )? Explain your reasoning.
16–59 The microtubules in Figure 16–22A were obtained from a population
that was growing rapidly, whereas the one in Figure 16–22B came from
microtubules undergoing catastrophic shrinkage. Comment on any dif-
ferences between the two images and suggest likely explanations for
those you observe.
16–60 Dynamic instability causes microtubules either to grow or to shrink rap-
idly. Consider an individual microtubule that is in its shrinking phase.
(A) GROWING (B) SHRINKING
A. What must happen at the end of the microtubule in order for it to stop
shrinking and start growing?
B. How would an increase in the tubulin concentration affect this switch
from shrinking to growing?
C. What would happen if GDP, but no GTP, were present in the solution?
D. What would happen if the solution contained an analog of GTP that
could not be hydrolyzed?
16–61 The -tubulin subunit of an -tubulin dimer retains its bound GTP for
a short time after it has been added to a microtubule, yielding a GTP
cap whose size depends on the relative rates of polymerization and GTP
hydrolysis. A simple notion about microtubule growth dynamics is that
the ends with GTP caps grow, whereas ends without GTP caps shrink. To
Figure 16–22 Electron microscopic
test this idea, you allow microtubules to form under conditions where
analysis of microtubule dynamics
you can watch individual microtubules. You then sever one microtubule (Problem 16–59). (A) Rapidly growing
in the middle using a laser beam. Would you expect the newly exposed microtubules. (B) Catastrophically
plus and minus ends to grow or to shrink? Explain your answer. Problems
shrinking p16.05/16.22
microtubule.
346 Chapter 16: The Cytoskeleton

LINEAR GROWTH LATERAL ASSOCIATION Figure 16–23 Model for microtubule

nucleation by pure -tubulin dimers
(Problem 16–63).

16–62 The drugs Taxol®, extracted from the bark of yew trees, and colchicine, an
alkaloid from autumn crocus, have opposite effects. Taxol binds tightly to
microtubules and stabilizes them. When added to cells, it causes much
of the free tubulin to assemble into microtubules. In contrast, colchicine
prevents microtubule formation. Taxol and colchicine are equally toxic
to dividing cells, and both are used as anticancer drugs. Based on your
Problems
knowledge of microtubule p16.17/16.23
dynamics, explain why these drugs are toxic
to dividing cells despite their opposite modes of action.
16–63 A solution of pure -tubulin dimers is thought to nucleate microtu-
bules by forming a linear protofilament about seven dimers in length.
At that point, the probabilities that the next -dimer will bind laterally
or to the end of the protofilament are about equal. The critical event for
microtubule formation is thought to be the first lateral association (Fig-
ure 16–23). How does lateral association promote the subsequent rapid
formation of a microtubule?
16–64 How does a centrosome “know” when it has found the center of the cell?
16–65 How are -TuRC and the Arp2/3 complex similar, and how are they dif-
ferent?
16–66 When cells enter mitosis, their existing array of cytoplasmic microtu-
bules has to be rapidly broken down and replaced with the mitotic spin-
dle, which pulls the chromosomes into the daughter cells. The enzyme
katanin, named after Japanese samurai swords, is activated during the
onset of mitosis and cleaves microtubules into short pieces. What do you
suppose is the fate of the microtubule fragments created by katanin?
8
16–67 Kinesin-1 motors are highly processive, moving long distances on micro-
7
tubule tracks without dissociating. By contrast, myosin II motors in skel- 1
etal muscle do not move processively; they take only one or a few steps
before letting go. How are these different degrees of processivity adapted
to the biological functions of kinesin-1 and myosin II? 6

16–68 An electron micrograph of a cross section through a flagellum is shown in

Figure 16–24.
A. Assign the following components to the indicated positions on the figure. 2

A microtubule 3
B microtubule
Outer dynein arm 5 4
Inner dynein arm
Inner sheath 100 nm
Nexin
Figure 16–24 Electron micrograph of
Radial spoke a cross section through a flagellum of
Singlet microtubule Chlamydomonas reinhardtii (Problem
B. Which of the above structures are composed of tubulin? 16–68).

Problems p16.32/16.24
MICROTUBULES 347

16–69 The sliding-microtubule mechanism for ciliary bending is undoubtedly hooks

correct. The consequences of sliding are straightforward when a pair of
outer doublets is considered in isolation. The dynein arms are arranged
so that, when activated, they push their neighboring outer doublet out-
ward toward the tip of the cilium. If the pair of outer doublets is linked
together by nexin molecules, they will bend so that the one that has
been pushed toward the tip will define the inside of the curve (see Figure
16–28A). It is confusing, however, to think about sliding in the circular microtubules
array of outer doublets in the axoneme. If all the dynein arms in a circular
array were equally active, there could be no significant relative motion. Figure 16–25 Tubulin-decorated
(The situation is equivalent to a circle of strongmen, each trying to lift his microtubules in a cross section through
neighbor off the ground; if they all succeeded, the group would levitate.) a nerve axon (Problem 16–70). The hooks
represent the tubulin decoration.
Devise a pattern of dynein activity (consistent with axoneme struc-
ture and the directional pushing of dynein) that could account for bend- Problems p16.34/16.25
ing of the axoneme in one direction. How would this pattern change for
bending in the opposite direction?
16–70 In addition to conducting impulses in both directions, nerve axons carry
vesicles to and from the cell body along microtubule tracks. Do outbound
vesicles move along microtubules that are oriented in one direction and
incoming vesicles move along oppositely oriented microtubules? Or are
microtubules all oriented in the same direction, with different motor
proteins providing the directionality?
To distinguish between these possibilities, you prepare a cross sec-
tion through a nerve axon and decorate the microtubules with tubulin,
which binds to the tubulin subunits of the microtubule to form hooks.
The decorated microtubules are illustrated in Figure 16–25. Do all the
microtubules run in the same direction or not? How can you tell?

CALCULATIONS
16–71 At 1.4 mg/mL pure tubulin, microtubules grow at a rate of about 2 μm/
min. At this growth rate, how many -tubulin dimers (8 nm in length) (A) NO CENTROSOMES
40
are added to the ends of a microtubule each second?
microtubule mass

30
16–72 The function of microtubules depends on their specific spatial organi-
zation within the cell. How are specific arrangements created, and what 20

determines the formation and disappearance of individual microtu- 10

bules?
0
To address these questions, investigators have studied the in 0 10 20 30 40 50
vitro assembly of -tubulin dimers into microtubules. Below 15 μM
-tubulin, no microtubules are formed; above 15 μM, microtubules (B) WITH CENTROSOMES
form readily (Figure 16–26A). If centrosomes are added to the solution of 60
microtubule number

tubulin, microtubules begin to form at less than 5 μM (Figure 16–26B).

(Different assays were used in the two experiments—total weight of 40
microtubules in Figure 16–26A and the average number of microtubules
per centrosome in Figure 16–26B—but the lowering of the critical con- 20
centration for microtubule assembly in the presence of centrosomes is
0
independent of the method of assay.) 0 10 20 30 40 50
A. Why do you think that the concentration at which microtubules begin to αβ -tubulin (µM)
form (the critical concentration) is different in the two experiments?
Figure 16–26 Analysis of microtubule
B. Why do you think that the plot in Figure 16–26A increases linearly with
assembly (Problem 16–72). (A) Mass of
increasing tubulin concentration above 15 μM, whereas the plot in Fig- microtubules assembled in the absence
ure 16–26B reaches a plateau at about 25 μM? of centrosomes as a function of tubulin
C. The concentration of -tubulin dimers (the subunits for assembly) in concentration. Tubulin assembly into
a typical cell is 1 mg/mL and the molecular weight of a tubulin dimer is microtubules
Problemswas measured by the
p16.18/16.26
110,000. What is the molar concentration of tubulin dimers in cells? How increase in solution turbidity. (B) Average
number of microtubules per centrosome
does the cellular concentration compare with the critical concentrations as a function of tubulin concentration.
in the two experiments in Figure 16–26? What are the implications for the Concentrations refer to -tubulin dimers,
assembly of microtubules in cells? which are the subunits of assembly.
348 Chapter 16: The Cytoskeleton

(A) POLYMERIZATION KINETICS (B) CRITICAL CONCENTRATION Figure 16–27 Effects of -tubulin on
0.3 microtubule polymerization (Problem 16–
73). (A) Kinetics of polymerization in the

polymerized tubulin
+ γ-tubulin

tubulin assembly
10 presence and absence of -tubulin.
0.2 – γ-tubulin (B) The critical concentration for
+ γ-tubulin

(µM)
microtubule assembly in the presence of
5 0.6 nM -tubulin and in its absence.
0.1
– γ-tubulin
0 0
0 30 60 0 5 10
time (minutes) αβ-tubulin (µM)

16–73 The -tubulin ring complex ( -TuRC), which nucleates microtubule

assembly in cells, includes -tubulin and several accessory proteins.
To get at its mechanism of nucleation, you have prepared monomeric
-tubulin by in vitroProblems p16.19/16.27
translation and purification. You measure the effect
of adding monomeric -tubulin to a solution of -tubulin dimers, as
shown in Figure 16–27.
A. In the presence of monomeric -tubulin, the lag time for assembly of
microtubules decreases, and assembly occurs more rapidly (Figure
16–27A). How would you account for these two effects of -tubulin? (A)
B. The critical concentration of -tubulin needed for the assembly of outside outer doublet
microtubules is reduced from about 3.2 μM in the absence of -tubulin
to about 1.7 μM in its presence (Figure 16–27B). How do you suppose
-tubulin lowers the critical concentration? How does this account for
the greater extent of polymerization in Figure 16–27A? (Think about the r r + 180
end—plus or minus—at which polymerization occurs in the presence of
-tubulin.)
16–74 Using the equation for diffusion given in Problem 16–10, calculate the base of
axoneme
average time it would take for a vesicle to diffuse to the end of an axon 10
cm in length. The diffusion coefficient of a typical vesicle is 5 × 10–8 cm2/
sec.
16–75 A mitochondrion 1 μm long can travel the 1 meter length of the axon inside
from the spinal cord to the big toe in a day. The Olympic men’s freestyle outer
swimming record for 200 meters is 1.75 minutes. In terms of body lengths doublet

per day, who is moving faster: the mitochondrion or the Olympic record
holder? (Assume that the swimmer is 2 meters tall.) (B)

16–76 During the flagellar beat cycle in Chlamydomonas, the bent segment of
the flagellum extends roughly through half the circumference of a circle
(Figure 16–28). r r + 30

A. How much sliding of microtubule doublets against one another is

required to account for the observed bending of the flagellum into a
semicircle? Calculate how much farther the doublet on the inside of the
semicircle protrudes beyond the doublet on the outside of the semicircle unstretched
at the tip of the flagellum (Figure 16–28A). The width of a flagellum is nexin
180 nm.
B. The elastic protein molecule (nexin) that links adjacent outer doublets
must stretch to accommodate the bending of a flagellum into a semicir-
cle. If the length of an unstretched nexin molecule at the base of a flagel-
stretched
lum is 30 nm, what is the length of a stretched molecule at the tip of a nexin
flagellum (Figure 16–28B)? Adjacent doublets are 30 nm apart.
Figure 16–28 Flagella bent into half-
DATA HANDLING circles (Problem 16–76).
(A) Representation showing the
“inside” and “outside” doublets, which
16–77 The orientation of the -tubulin dimer in a microtubule was deter-
are 180 nm apart. (B) Representation
mined in several ways. GTP-coated fluorescent beads, for example, were showing adjacent doublets, which are
found to bind exclusively at the plus ends of microtubules. By contrast, 30 nm apart, and the nexin molecules
gold beads coated with antibodies specific for a peptide of -tubulin that link them.
MICROTUBULES 349

(A) PLUS ENDS Figure 16–29 Analysis of growth kinetics

8 of individual microtubules (Problem
16–78). (A) Changes in length at the
plus ends. Results from two individual
a microtubules are indicated by different

length ( µm)
shades of red. (B) Changes in length
at the minus ends. Results from two
4 b individual microtubules are indicated by
different shades of blue.

0
0 5 10 15 20
time (minutes)
(B) MINUS ENDS

4
length ( µm)

0
0 5 10 15 20
time (minutes)

bound exclusively at the minus end. How do these observations define

the orientation of the -tubulin dimer in the microtubule? Which tubu-
lin subunit, or , is at which end? Explain your reasoning.
16–78 The complex kinetics of microtubule assembly make it hard to predict the
behavior of individual microtubules. Some microtubules in a population
can grow, even as the majority shrink to nothing. One simple hypothesis
proposed to explain this behavior is that a growing end is protected from
disassembly by a GTP cap and that a faster-growing end has a longer
GTP cap. Real-time video observations
Problems of changes in length with time are
p16.07/16.29
shown for two individual microtubules in Figure 16–29. Measurements
of their rates of growth and shrinkage show that the plus end of each
microtubule grows three times faster than the minus end, and shrinks at
half the rate.
A. Are changes in length at the two ends of a microtubule dependent or
independent of one another? How can you tell?
B. What does the GTP-cap hypothesis predict about the rate of switching
between growing and shrinking states at the fast-growing end relative to
the slow-growing end? Does the outcome of this experiment support the
GTP-cap hypothesis?
C. What do you suppose would happen if centrosomes were used to nucleate
growth? What would happen if microtubule-associated proteins (MAPs)
were included?
16–79 Comparisons of microtubule behavior between species point to differ-
ences that raise questions about the biological importance of dynamic
instability. Notothenioid fish, for example, which live in the Southern
Ocean at a constant temperature of –1.8°C, have remarkably stable micro-
tubules compared with warm-blooded vertebrates such as the cow. This
is an essential modification for notothenioid fish because normal micro-
tubules disassemble completely into -tubulin dimers at 0°C. Meas-
urements on individual microtubules in solutions of pure tubulin show
that notothenioid fish microtubules grow at a much slower rate, shrink
at a much slower rate, and only rarely switch from growth to shrinkage
(catastrophe) or from shrinkage to growth (rescue) (Table 16–1).
A. The amino acid sequences of the - and -tubulin subunits from noto-
thenioid fish differ from those of the cow at positions and in ways that
might reasonably be expected to stabilize the microtubule, in accord with
350 Chapter 16: The Cytoskeleton

TABLE 16–1 Properties of individual microtubules in notothenioid fish and the

domestic cow (Problem 16–79).
Microtubules Growth Shrinkage Catastrophe Rescue
rate rate frequency frequency
(μm/min) (μm/min) (min–1) (min–1)
Notothenioid fish 0.27 0.8 0.008 <0.0004
Domestic cow 2.18 61.2 0.52 3.1
Multiple individual microtubules were observed by video microscopy near the body temperature
for each species: 5°C for fish and 37°C for cow. Average growth rates were calculated for
growing microtubules, and average shrinkage rates were calculated for shrinking microtubules.
Changes from growth to shrinkage (catastrophe) and from shrinkage to growth (rescue) were
averaged over the observation period and expressed as frequency of events per minute.

the data in Table 16–1. Would you expect these changes to strengthen
the interactions between the - and -tubulin subunits in the -dimer,
between adjacent dimers in the protofilament, or between tubulin subu-
nits in adjacent protofilaments? Explain your reasoning.
B. Dynamic instability is thought to play a fundamental role in the rapid
microtubule rearrangements that occur in cells. How do you suppose
cells in these notothenioid fishes manage to alter their microtubule
architecture quickly enough to accomplish essential cell functions? Or
do you suppose that these cells exist with a stable microtubule cytoskel-
eton that only slowly rearranges itself?
16–80 A standard purification scheme for tubulin is to prepare a cell extract,
chill it to 0°C, spin it at high speed, and save the supernatant. The super-
natant is then warmed to 37°C and incubated in the presence of GTP. The
mixture is then spun at high speed and the pellet is saved and redissolved.
Then the cycle is repeated: chill the dissolved pellet, spin at high speed,
save the supernatant, incubate with GTP at 37°C, spin at high speed, save
the pellet. After a few cycles one obtains a pure preparation of tubulin.
Explain how this procedure yields pure tubulin.
16–81 In addition to centrosomes, flagellar axonemes also can serve as nucle-
ation sites for microtubule assembly. The following experiment was
designed to determine whether these two structures nucleate microtu-
bule growth by binding to the plus end or to the minus end of the nascent
microtubule. Flagellar axonemes were included as a control since their
plus and minus ends can be distinguished. Centrosomes and flagellar
axonemes were incubated briefly in unlabeled tubulin to nucleate micro-
tubule growth. A high concentration of biotin-labeled tubulin was then
added and the incubation was continued for 10 minutes. At that point the
preparations were fixed and the biotin-labeled segments were visualized
by adding fluorescein-labeled antibodies specific for biotin. The lengths
of the biotin-labeled segments were measured and plotted as shown in
Figure 16–30.

(A) AXONEMES (B) CENTROSOMES

60 80
minus end
60
microtubule

40
number

plus end 40 Figure 16–30 Length distributions of

20 microtubules (Problem 16–81).
20 (A) Nucleation by flagellar axonemes,
whose plus and minus ends can
0 0
0 5 10 15 0 5 10 15 be distinguished. (B) Nucleation by
length (µm) length (µm) centrosomes.
MICROTUBULES 351

(A) MICROGRAPHS (B) PLUS ENDS (C) MINUS ENDS Figure 16–31 Effects of -TuRC on
microtubule assembly (Problem 16–82).
60 60 (A) Example of microtubules grown in
– γ-TuRC – γ-TuRC – γ-TuRC the presence and absence of -TuRC.

% total microtubules
50 50
Scale bar is 10 μm. (B) Distribution of the
40 40 lengths of bright segments at microtubule
30 30
plus ends in the presence and absence
of -TuRC. (C) Distributions of the lengths
20 20 of bright segments at microtubule minus
10 10 ends in the presence and absence of
-TuRC. In (B) and (C), only microtubules
0 0 with a defined dim segment and one
0 3 6 9 12 15 18 21 0 3 6 9 12 15 18 21
or two bright terminal segments were
60 60 counted.
+ γ-TuRC + γ-TuRC + γ-TuRC
% total microtubules

50 50

40 40
30 30
20 20
10 10
0 0
0 3 6 9 12 15 18 21 0 3 6 9 12 15 18 21
microtubule length (µm) microtubule length (µm)

A. Which end of a newly assembled microtubule is attached to the plus end

of the flagellar axoneme?
B. Which end Problems p16.22/16.31
of a microtubule assembled on a flagellar axoneme grows
faster?
C. Which end of an assembled microtubule is attached to a centrosome?
Explain your reasoning.
16–82 In the paper that defined the -tubulin ring complex ( -TuRC), the
authors purified the complex from Xenopus oocytes and showed that it
dramatically stimulated the nucleation of microtubules. To determine
whether nucleation occurred at plus ends or at minus ends, they polym-
erized microtubules in two steps. In the first step, microtubules nucle-
ated with or without -TuRC were allowed to form in the presence of a
small amount of -tubulin containing a low proportion of rhodamine-
labeled -tubulin, which makes the microtubules fluoresce dimly. In
the second step, these microtubules were allowed to extend at both ends
in the presence of a higher proportion of rhodamine-tagged tubulin to
label the ends brightly. The longer bright segment identifies the plus end,
and the shorter segment the minus end (Figure 16–31A). Measurements
of the lengths of a large number of bright segments in individual micro-
tubules yielded the data in Figure 16–31B and C. At which end of the
microtubule is -TuRC when it nucleates? Explain your reasoning.
16–83 A useful technique for studying a microtubule motor is to attach the
motor proteins by their tails to a glass coverslip (the tails stick avidly to a
clean glass surface) and then to allow microtubules to settle onto them.
In the light microscope, the microtubules can be seen to move over the
surface of the coverslip as the heads of the motors propel them (Figure
Figure 16–32 Movement of microtubules
16–32).
on a bed of microtubule motor molecules
(Problem 16–83). Red arrows mark the
movement of a microtubule with a gold
bead attached via antibodies to the
minus end of a microtubule; white arrows
mark the movement of a microtubule
without an attached bead. Pictures
5 µM were taken using video-enhanced
differential-interference-contrast (VE-DIC)
0 sec 24 sec 48 sec 72 sec microscopy.
352 Chapter 16: The Cytoskeleton

Figure 16–33 Microtubules with -tubulin

(microtubules/sec/mm2 )
antibody-coated gold beads attached to 100
one end (Problem 16–84). At the vertical

landing rate
line a section of each microtubule has
10 2 heads
been removed so that the two ends can
be displayed side by side.
1 head
1

0.1
A. Since the motor proteins attach in random orientations to the coverslip, 1 10 100 1000

how can they generate

Problems coordinated movement of individual microtu-
p16.27/16.33 molecules/µm2
bules, rather than engaging in a tug-of-war?
Figure 16–34 Landing rates—binding and
B. In which direction will microtubules crawl on a bed of dynein motor
moving—of microtubules as a function
molecules (that is, will they move plus-end first or minus-end first)? of motor protein density (Problem 16–85).
C. In the experiment shown in Figure 16–32, some of the microtubules Problems
Results p16.29/16.34
with wild-type kinesin are shown
were marked by gold beads that were bound by minus-end-specific anti- as green squares, while those with
bodies. Is the motor protein on the coverslip a plus-end or minus-end recombinant kinesin are shown as brown
circles.
directed motor? How can you tell?
16–84 In Problem 16–77, the orientation of the -tubulin dimer in the micro-
tubule was determined by showing that -tubulin antibody-coated gold
beads bound to the minus end. The electron micrographs, however, just
showed microtubules with beads at one end (Figure 16–33). How do you
suppose the investigators knew which end was which? Design an experi-
ment to determine the orientation of microtubules labeled at one end
with a gold bead.
16–85 Kinesin carries vesicles for long distances along microtubule tracks in
the cell. Are the two motor domains of a kinesin molecule essential to
accomplish this task, or could a one-headed motor protein function just
as well? Using recombinant DNA techniques, a version of kinesin was
prepared that was identical to normal kinesin except that one motor
domain was absent. Wild-type kinesin with two motor domains and
recombinant kinesin with one were attached to coverslips at a variety
of densities and the rate at which microtubules were bound and moved
(collectively, termed the landing rate) was measured (Figure 16–34).
A. Why do you suppose that the curves at low motor densities are so differ-
ent?
B. What do these experiments say about the design of the kinesin motor: are
two heads required for vesicle transport, or is only one needed? Explain
your reasoning.
16–86 The movements of single motor-protein molecules can be analyzed
directly. Using polarized laser light, it is possible to create interference
patterns that exert a centrally directed force, ranging from zero at the
center to a few piconewtons at the periphery (about 200 nm from the
center). Individual molecules that enter the interference pattern are rap-
idly pushed to the center, allowing them to be captured and moved at the
experimenter’s discretion.
Using such “optical tweezers,” single kinesin molecules can be posi-
tioned on a microtubule that is fixed to a coverslip. Although a single
kinesin molecule cannot be seen optically, it can be tagged with a silica
bead and tracked indirectly by following the bead (Figure 16–35A). In
the absence of ATP, the kinesin molecule remains at the center of the
interference pattern, but with ATP it moves toward the plus end of the
microtubule. As kinesin moves along the microtubule, it encounters the
force of the interference pattern, which simulates the load kinesin carries
during its actual function in the cell. Moreover, the pressure against the
silica bead counters the effects of Brownian (thermal) motion, so that the
position of the bead more accurately reflects the position of the kinesin
molecule on the microtubule.
MICROTUBULES 353

(A) EXPERIMENTAL SET-UP (B) POSITION OF KINESIN

silica
bead
80
trace 1 trace 2

distance (nm)
60

40

kinesin
20

microtubule 0 2 4 6 8 0 1 2 3 4 5
time (seconds) time (seconds)

Traces of the movements of two kinesin molecules along a microtu- Figure 16–35 Movement of kinesin
bule are shown in Figure 16–35B. along a microtubule (Problem 16–86).
(A) Experimental set-up, with kinesin
A. As shown in Figure 16–35B, all movement of kinesin is in one direction linked to a silica bead, moving along
(toward the plus end of theProblems p16.30/16.35
microtubule). What supplies the free energy a microtubule. (B) Position of kinesin
needed to ensure a unidirectional movement along the microtubule? (as visualized by the position of the
B. What is the average rate of movement of each kinesin along the microtu- silica bead) relative to the center of the
interference pattern, as a function of
bule?
time of movement along the microtubule.
C. What is the length of each step that a kinesin takes as it moves along a The jagged nature of the trace results
microtubule? from Brownian motion of the bead. The
D. From other studies it is known that kinesin has two globular domains that movements of two different kinesin
can each bind to -tubulin, and that kinesin moves along a single proto- molecules are shown.
filament in a microtubule. In each protofilament, the -tubulin subunit
repeats at 8-nm intervals. Given the step length and the interval between
-tubulin subunits, how do you suppose a kinesin molecule moves along
a microtubule?
E. Is there anything in the data in Figure 16–35B that tells you how many
ATP molecules are hydrolyzed per step?

MEDICAL LINKS
16–87 Mice that are homozygous for a knockout of the gene for the kinesin
motor protein KIF1B die at birth. Heterozygous knockouts survive, but
suffer from a progressive muscle weakness similar to human neuropa-
thies. Humans with Charcot–Marie–Tooth disease type 2A have a muta-
tion in one copy of the gene for KIF1B that prevents the protein from
binding to ATP. The heterozygous mice and the human patients have very
similar progressive neuropathies. How do you suppose that the loss of
one copy of a gene for a kinesin motor can have such profound effects on
nerve function?

INTERMEDIATE FILAMENTS AND SEPTINS

TERMS TO LEARN
keratin septum
neurofilament

DEFINITIONS
Match each definition below with its term from the list above.
16–88 Member of a family of intermediate filaments that are found in high con-
centrations along the axons of vertebrate neurons.
16–89 The most diverse intermediate filament family, with about 20 types found
in human epithelial cells and 10 more specific to hair and nails.
354 Chapter 16: The Cytoskeleton

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
16–90 Like actin filaments and microtubules, cytoplasmic intermediate fila-
ments are found in all eukaryotes.
16–91 The cytoplasmic cytoskeleton and the nuclear lamina are connected by
proteins that extend from one side to the other of the nuclear pore.

THOUGHT PROBLEMS
16–92 Why is it that intermediate filaments have identical ends and lack polar-
ity, whereas actin filaments and microtubules have two distinct ends
with a defined polarity?
16–93 Which of the following types of cell would you expect to contain a high
density of cytoplasmic intermediate filaments? Explain your answers.
A. Amoeba proteus (a free-living amoeba).
B. Human skin epithelial cell.
C. Smooth muscle cell in the digestive tract of a vertebrate.
D. Nerve cell in the spinal cord of a mouse.
E. Human sperm cell.
F. Plant cell.
16–94 Disulfide bonds do not form in the cytosol of eukaryotic cells (see Prob-
lem 3–37). Yet keratin intermediate filaments in the skin are cross-linked
by disulfide bonds. How can that be?
16–95 Although knockouts of genes for some intermediate filaments have
detectable phenotypes in mice, gene knockouts for vimentin or another
vimentin family member, glial fibrillary acid protein (GFAP), appear nor-
mal. Mice with knockouts for both vimentin and GFAP, however, exhibit
impaired function of their astrocytes, which are accessory cells in the
central nervous system. Why do you suppose that individual knockouts
for vimentin and GFAP are normal, while the combined knockout has a
demonstrable deficiency?
16–96 There are no known motor proteins that move on intermediate filaments.
Suggest an explanation for this observation.
16–97 Compare the structure of intermediate filaments with that of the myo-
sin II filaments in skeletal muscle cells. What are the major similarities?
What are the major differences? How do the differences in structure
relate to their function?

DATA HANDLING
16–98 The intermediate filament networks in cells must be dealt with in some
way when a cell divides. Figure 16–36 shows the vimentin networks
[tagged with green fluorescent protein (vimentin–GFP)] in kidney cells
from baby hamster (BHK-21) and rat kangaroo (PtK2) that are undergo-
ing division. By examining these photographs, decide how each of these
cell types handles its vimentin network.
16–99 Although the mechanism of disassembly of most intermediate filaments
is unclear, it is well defined for their ancestors, the nuclear lamins. The
nuclear envelope is strengthened by a fibrous meshwork of lamins (the
nuclear lamina), which supports the membrane on the nuclear side.
When cells enter mitosis, the nuclear envelope breaks down and the
nuclear lamina disassembles. Assembly and disassembly of the nuclear
INTERMEDIATE FILAMENTS AND SEPTINS 355

(A) BHK-21 CELLS Figure 16–36 Vimentin networks during

cell division (Problem 16–98). (A) BHK-
21 cells. The three images from left to
right correspond to prometaphase and
anaphase of mitosis and to the daughter
cells. Note that the first two images
are magnified relative to the third.
(B) PtK2 cells. The images from left
to right correspond to prometaphase,
(B) PtK2 CELLS
telophase, and late cytokinesis. Note
that in late cytokinesis, the cells are still
connected by a bridge of cytoplasm.
The scale bar in each picture is 5 μm.
Bright areas represent vimentin-GFP
fluorescence.

Problems p16.10/16.36
lamina may be controlled by reversible phosphorylation of lamins A, B,
and C, since the lamins from cells that are in mitosis carry significantly
more phosphate than do the lamins from cells that are in interphase.
To investigate the role of phosphorylation, you label cells with
35S-methionine and purify lamins A, B, and C from mitotic cells and from

interphase cells. You then analyze the purified lamins from each source,
along with a mixture of them, by two-dimensional gel electrophoresis
(Figure 16–37A). You also treat the samples with alkaline phosphatase,
which removes phosphates from proteins, and analyze them in the same
way (Figure 16–37B).
A. Why does treatment with alkaline phosphatase reduce the number of
lamin spots to three, regardless of the number seen in the absence of
phosphatase treatment?
B. How many phosphate groups are attached to lamins A, B, and C during
interphase? How many are attached during mitosis? How can you tell?
C. Why was 35S-methionine rather than 32P-phosphate used to label lamins
in experiments designed to measure phosphorylation differences? How
would the autoradiograms have differed if 32P-phosphate had been used
instead?
D. Do you think these results prove that lamin disassembly during mitosis is
caused by their reversible phosphorylation? Why or why not?

(A) MINUS ALKALINE PHOSPHATASE

interphase mitosis mixture

kd
separation by molecular mass

A A A
B B B 79 Figure 16–37 Two-dimensional separation
C C C of nuclear lamins from cells in interphase
65 and mitosis (Problem 16–99). (A) No
treatment with alkaline phosphatase.
(B) Treatment with alkaline phosphatase.
(B) PLUS ALKALINE PHOSPHATASE Letters identify the positions of lamins
A, B, and C. The purified lamins from
A A A
B B B 79 interphase and mitotic cells were added
together to create the mixture. Acidic
C C C proteins are more negatively charged;
65
basic proteins are more positively
charged. Each rectangle represents a
acidic basic
distinct two-dimensional separation of
separation by charge nuclear lamins.
356 Chapter 16: The Cytoskeleton

CELL POLARIZATION AND MIGRATION

TERMS TO LEARN
blebbing invadopodium Rho family protein
chemotaxis lamellipodium WASp protein
filopodium

DEFINITIONS
Match each definition below with its term from the list above.
16–100 A group of closely related monomeric GTPases that includes Cdc42, Rac,
and Rho.
16–101 Flattened, two-dimensional protrusion of membrane, supported by a
meshwork of actin filaments, that is extended from the leading edge of
crawling epithelial cells, fibroblasts, and some neurons.
16–102 Essentially a one-dimensional structure that protrudes from a cell and (A) QUIESCENT
contains a core of long, bundled actin filaments.
16–103 A distinct form of membrane protrusion that is often observed when cells
are cultured on a pliable extracellular matrix substrate.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
16–104 When fragments of the lamellipodium of cultured keratocytes are sliced
(B) STRESS FIBERS
off, they continue to crawl normally, looking like tiny keratocytes.
16–105 Neutrophils move toward a source of bacterial infection by chemotaxis,
using receptors on their surface to respond to a gradient of N-formylated
peptides derived from bacterial proteins.

THOUGHT PROBLEMS
16–106 Distinguish among the three processes—protrusion, attachment, and
traction—that make up the crawling movements of cells. (C) LAMELLIPODIA

16–107 Actin filaments are said to “push” on the cell membrane to cause it to
form a protrusion. But there are problems with a pushing mechanism at
both ends of the filaments. When a plus end reaches the membrane and
abuts it, how are new subunits added to extend the filament (allowing it
to push)? And how is the minus end of the filament anchored so that the
filament isn’t simply pushed back into the cell’s interior? What do you
suppose might be the answers to these questions?
16–108 The characteristic actin staining in a quiescent cell is shown in Figure (D) FILOPODIA
16–38A. When such cells are injected with a constitutively activated form
of Rac, Rho, or Cdc42 monomeric GTPases, they dramatically alter their
actin cytoskeletons. Which GTPase is associated with formation of stress
fibers (Figure 16–38B), lamellipodia (Figure 16–38C), and filopodia
(Figure 16–38D)?
16–109 How is the unidirectional motion of a lamellipodium maintained?

DATA HANDLING
Figure 16–38 Actin cytoskeleton in
16–110 Kinesin motors transport oligomers of neurofilament proteins down different cells (Problem 16–108).
axonal microtubules to sites where they are used in the construction (A) Quiescent cell. (B) Cell with prominent
or repair of neurofilaments. Classic studies that followed pulse-labeled stress fibers. (C) Cell with multiple
lamellipodia. (D) Cell with many long
neurofilament proteins during axonal transport agree that the peak of filopodia. Cells in B, C, and D were
radioactivity broadens markedly during transport. More recent stud- injected with an activated form of a
ies demonstrated that unphosphorylated neurofilament proteins bind monomeric GTPase.
Problems p16.33/16.38
CELL POLARIZATION AND MIGRATION 357

(A) MICROSCOPIC DATA (B) MEASUREMENTS Figure 16–39 Regulation of filopodial

6 extension and retraction (Problem
polymerization 16–111). (A) Time-lapse observations
filopodial 4
tip of a single filopodium. The yellow line

distance (µm)
2 tip identifies the position of the tip of the
filopodium; the blue line marks the
fluorescent 0 position of the fluorescent segment.
flow
mark -2 The black scale bar at the right is 5 μm.
(B) Summary of the data. For the data
-4 summary, the positions of the tip and
0 1 2 3 4 the fluorescent segment were arbitrarily
time (minutes) set at zero at zero minutes, as was
the difference between the tip and the
fluorescent mark. The position of the
strongly to kinesin motors and weakly to existing neurofilaments. By filopodium is labeled “tip”; the position
contrast, the phosphorylated
Problems forms bind weakly to kinesin motors and
p16.43/16.39 of the fluorescent mark is labeled “flow”
strongly to neurofilaments. for retrograde flow; and the difference
A. How might the phosphorylation-dependence of oligomer binding to is labeled “polymerization” for actin
polymerization.
kinesin motors and neurofilaments account for the broadening of the
transport wave?
B. If you could track the movement of single oligomers down an axon, how
would you expect them to move? Consider an oligomer at the leading
edge of the transport wave and one at the trailing edge.
16–111 Nerve growth cones navigate along stereotyped pathways during devel-
opment by continually extending and retracting slender filopodia to
sense directional cues in the environment. The bundled actin cytoskel-
eton in such a filopodium grows at the tip by actin polymerization, and is
pulled back into the cell over time, a phenomenon known as retrograde
flow. In principle, extension and retraction of filopodia could be con-
trolled by regulating the rate of actin polymerization or the rate of retro-
grade flow. The experiments below were carried out to determine which
rate—polymerization or retrograde flow—is regulated.
Actin monomers tagged with caged rhodamine were injected into
cells and allowed to incorporate into actin filaments. The rhodamine
in a narrow segment of the actin bundles near the tip of a single filo-
podium was uncaged by brief irradiation, yielding a fluorescent mark
that allowed retrograde flow to be observed directly over time (Figure
16–39A). Extension and retraction of the tip of the filopodium was fol-
lowed microscopically (Figure 16–39A). Actin polymerization was taken
as the distance between the fluorescent mark and the tip of the filopo-
dium (Figure 16–39A). A summary of the data for this single filopodium
is shown in Figure 16–39B. Are extension and contraction of this filopo-
dium regulated by the rate of actin polymerization or by the rate of retro-
grade flow? Explain your reasoning.
16–112 Activation of Cdc42, a monomeric GTPase, triggers actin polymerization
and bundling to form either filopodia or shorter cell protrusions called
microspikes. These effects of Cdc42 could be mediated by N-WASp,
which is a multifunctional protein. As shown in Figure 16–40A, N-WASp
Figure 16–40 The role of N-WASp in
Cdc42-triggered actin polymerization
(A) DOMAINS OF N-WASp (B) ACTIN POLYMERIZATION and bundling (Problem 16–112).
(A) Domain structure of N-WASp and
fluorescence (arbitrary units)

pleckstrin GTPase verprolin 2.4 untreated two mutants defective in individual

in

ic
fil

id

homology binding homology domains. H208D carries a mutation in the

co

ac

G domain so that it cannot bind Cdc42.

N-WASp PH G V C A
2.0 cof is deleted for the cofilin domain.
(B) Actin polymerization in untreated
H208D PH G V C A and N-WASp-depleted egg extracts. In
N-WASp
1.6 depleted the polymerization assays, the extracts
were supplemented with pyrene-actin,
∆cof PH G V Δ A which fluoresces much more highly in
1.2 the filament than it does on the subunit.
0 200 400 600 Cdc42 with bound GTP S was present in
time (seconds) both reactions.
358 Chapter 16: The Cytoskeleton

(A) (B)
(fluorescence, arbitrary units) 10.0 10.0

+ VCA + ARP + VCA

7.5 + N-WASp + ARP 7.5 + N-WASp + Cdc42 + PIP2
actin polymerization

+ N-WASp + Cdc42
+ N-WASp + PIP2
+ N-WASp
5.0 5.0
+ N-WASp
+ ARP
actin alone actin + ARP
2.5 2.5

0 0
0 500 1000 1500 0 500 1000 1500
time (seconds) time (seconds)

contains a pleckstrin homology (PH) domain, which binds to PIP2 (phos- Figure 16–41 Polymerization of
phatidylinositol bisphosphate, PI(4,5)P2); a Cdc42 GTPase-binding fluorescent actin in the presence of
various purified components (Problem
domain (G); a verprolin homology domain (V), which binds to actin; a 16–113). (A) Mixtures of actin, N-WASp,
cofilin homology domain (C), which can bind to actin filaments; and a the Arp2/3 complex (ARP), and the
C-terminal acidic domain (A), which binds the Arp2/3 complex. C-terminal segment of N-WASp (VCA).
In Xenopus egg extracts, a convenient source of cytoskeletal com- (B) Mixtures of actin, N-WASp, ARP,
Cdc42-GTP S, and PIP2-containing
ponents, the addition of Cdc42 charged with GTP S, a nonhydrolyz-
vesicles. Vesicles without PIP2 do not
able analog of GTP, stimulates actin polymerization (Figure 16–40B). stimulate in any combination with the
If the extract is depleted of N-WASp using N-WASp-specific antibodies, other components.
no actin polymerization is observed when Cdc42-GTP S is added (see
Figure 16–40B). Actin polymerization can be restored by the addition of
purified N-WASp, but not by the addition of either of two mutant forms
of N-WASp: one (H208D) that cannot bind to Cdc42, and a second ( cof )
that eliminates the function of the cofilin domain (see Figure 16–40A).
Do these experiments support a role for N-WASp in the rearrange-
ment of actin filaments in response to Cdc42 activation? Explain your
reasoning. Include a discussion ofProblems
why the twop16.46/16.41
mutant forms of N-WASp
do not restore actin polymerization.
16–113 To determine the mechanism by which N-WASp mediates activation by
Cdc42, actin polymerization was measured in the presence of purified
components. In the presence of the Arp2/3 complex (ARP), N-WASp
stimulates actin polymerization substantially over ARP or N-WASp alone,
but not nearly so dramatically as the C-terminal segment of N-WASp
that contains just the verprolin (V), cofilin (C), and acidic (A) domains
(Figure 16–41A). To account for the difference between N-WASp and its
C-terminal VCA segment, N-WASp and ARP were mixed with combina-
tions of Cdc42-GTP S and vesicles containing PIP2 [PI(4,5)P2], as shown
in Figure 16–41B.
A. What is required for full-length N-WASp to stimulate actin polymeriza-
tion as efficiently as its C-terminal VCA segment? Explain your reason-
ing.
B. Based on these results, propose a model for the activation of N-WASp and
its stimulation of actin polymerization.

MCAT STYLE
Passage 1 (Questions 16–114 to 16–116)
Studying the motility of the pathogenic bacteria Listeria and Shigella, which cause
food poisoning and dysentery, respectively, significantly advanced our under-
standing of actin polymerization. These bacteria escape immune surveillance by
entering one cell’s cytoplasm and then spreading from cell to cell without expos-
ing themselves to the outside environment. They harness actin polymerization
CELL POLARIZATION AND MIGRATION 359

to push themselves against the plasma membrane, generating a membrane pro-

trusion that is engulfed by a neighboring cell, giving them direct access to that
cell’s cytoplasm. The comet tail of actin filaments in the wake of a moving Listeria
bacterium initially suggested that motility was actin based. Biochemical studies
identified host-cell surface proteins required for polymerization of actin, which
then led to the discovery of the Arp2/3 complex and the full in vitro reconstitution
of bacterial motility.
16–114 Early analysis in Listeria pointed to the ActA protein, an integral mem-
brane protein expressed on the cell surface, as the sole Listeria protein
required for motility. Which combination of the following observations
provides evidence that ActA is necessary and sufficient for motility?
I. ActA binds to the Arp2/3 complex.
II. E. coli expressing ActA can move in host-cell cytoplasm.
III. Listeria lacking the ActA gene fail to move in host-cell cytoplasm.
A. I and II
B. I and III
C. II and III
D. I, II, and III
16–115 The mechanism of bacterial motility was initially mysterious and contro-
versial: Was actin polymerization sufficient or did motility require actin-
based motor proteins, as well? Which one of the following observations
suggests that actin polymerization by itself generates the crucial force
that drives motility?
A. Bacteria that lack ActA do not show motility.
B. Drugs that depolymerize actin block motility.
C. Inhibiting myosins does not inhibit motility.
D. The Arp2/3 complex is required for motility.
16–116 Efficient movement of both Shigella and Listeria in a reconstituted sys-
tem requires the protein cofilin. Which one of the following actions
describes the correct role for cofilin in bacterial motility?
A. Blocks hydrolysis of ATP by actin to preserve the ATP cap.
B. Bundles actin filaments to maintain a tight comet tail of actin.
C. Depolymerizes actin filaments to allow subunits to recycle.
D. Promotes actin polymerization to propel bacteria forward.

Passage 2 (Questions 16–117 to 16–119)

Early analyses of microtubule dynamics were carried out before development
of methods for imaging microtubule growth and shrinkage in real time by video
microscopy. These early studies used electron microscopy to measure the num-
ber and length of microtubules at fixed time points after experimental manipula-
tions. One important experiment compared microtubule growth in the presence
and absence of centrosomes. In the absence of centrosomes, microtubule growth
was only observed at tubulin concentrations above 14 μM. In the presence of cen-
trosomes, microtubule growth was observed at tubulin concentrations of 4 μM
and above.
16–117 Which of the following hypotheses best accounts for how centrosomes
lower the concentration of tubulin required to support microtubule
growth?
A. Centrosomes decrease the frequency of changes from normal growth to
rapid shrinkage—catastrophes—at microtubule plus ends.
B. Centrosomes increase the rate of microtubule growth at both ends, mak-
ing it easier to grow large numbers of longer microtubules.
C. Centrosomes lower the critical concentration for growth at both ends,
allowing microtubules to grow at lower tubulin concentration.
D. Centrosomes nucleate microtubules by capping the minus end, allowing
growth at the lower critical concentration typical of plus ends.
360 Chapter 16: The Cytoskeleton

(A) (B) Figure 16–42 Dilution of centrosome-

50 nucleated microtubule arrays, pre-formed
at 25 μM tubulin, into solutions with
15 µM 15
tubulin at 15 μM (blue), 7.5 μM (red), or
microtubules per centrosome

40 5 μM (green) (Problem 16–118).

microtubule length (µm)

(A) Number of microtubules per
7.5 µM
centrosome. (B) Average length of
30 10
microtubules attached to a centrosome.

20
5
10 5 µM

0
0 1 4 8 0 1 4 8
minutes minutes

16–118 In another experiment, microtubules were grown from centrosomes at a

tubulin concentration of 25 μM. The centrosomes were then diluted into
lower concentrations of tubulin, and the number and length of microtu-
bules were plotted as a function of time (Figure 16–42). This experiment
revealed that when diluted to 7.5 μM tubulin, the number of microtu-
bules attached to each centrosome decreased with time, and the average
length of the remaining microtubules increased. At neither 15 μM nor
5 μM did both these changes occur. Which interpretation best accounts
for the decrease in microtubule number and the increase in length upon
dilution to 7.5 μM tubulin?
A. After dilution, some microtubules continue to hydrolyze GTP, which
allows them to continue to grow from their plus ends.
B. After dilution, the tubulin concentration is below the critical concentra-
tion for plus-end growth, so some microtubules shrink.
C. At lower tubulin levels, microtubules with a GTP cap continue to grow,
but those with a GDP cap rapidly depolymerize.
D. Dilution releases some microtubules from the centrosome, allowing
Figure
them to rapidly depolymerize from 16-301
their minus ends.
16–119 A key experiment analyzed microtubule growth in the absence of cen-
“MCAT”
trosomes at 20 μM tubulin. Since Problemare
microtubules 16large structures that
scatter light, their polymerization can be assayed by monitoring the tur-
bidity of the solution. During polymerization, the reaction mixture was
pulled repeatedly through a narrow-gauge syringe to generate shear
forces. This procedure caused a rapid and large decrease in turbidity that
lasted for only a short time before turbidity again increased. Which one
of the following statements provides the most reasonable explanation for
this observation?
A. Shear forces increased the rate of GTP hydrolysis at the plus and minus
ends.
B. Shear forces increased the rate of microtubule depolymerization at both
ends.
C. Shear forces induced frequent catastrophes at both ends of the microtu-
bule.
D. Shear forces induced microtubule breakage, exposing ends with bound
GDP.
Chapter 17 361

CHAPTER

The Cell Cycle 17

OVERVIEW OF THE CELL CYCLE IN THIS CHAPTER

TERMS TO LEARN OVERVIEW OF THE CELL CYCLE

cell cycle restriction point
G1 phase sister chromatid THE CELL-CYCLE CONTROL
G2 phase Start SYSTEM
interphase
S PHASE
DEFINITIONS MITOSIS
Match the definition below with its term from the list above. CYTOKINESIS
17–1 The long period of the cell cycle between one mitosis and the next.
MEIOSIS
17–2 The orderly sequence of events by which a cell duplicates its contents
and divides into two. CONTROL OF CELL DIVISION
AND CELL GROWTH
17–3 The checkpoint in the cell cycle that governs the cell’s commitment to
enter S phase.
17–4 The phase of the eukaryotic cell cycle between the end of cytokinesis and
the start of DNA synthesis.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
17–5 Since there are about 1013 cells in an adult human, and about 1010 cells
die and are replaced each day, we become new people every three years.
17–6 Although the lengths of all phases of the cell cycle are variable to some
extent, by far the greatest variation occurs in the duration of G1.

THOUGHT PROBLEMS
17–7 If the most basic function of the cell cycle is to duplicate accurately the
DNA in the chromosomes and then distribute the copies precisely to the
daughter cells, why are there gaps between S phase and M phase?
17–8 Many cell-cycle genes from human cells function when expressed in
yeast cells. Why do you suppose that is considered remarkable? After
all, many human genes encoding enzymes for metabolic reactions also
function in yeast, and no one thinks that is remarkable.
17–9 The budding yeast Saccharomyces cerevisiae and the fission yeast
Schizosaccharomyces pombe provide facile genetic systems for studying
a wide range of eukaryotic cell biological processes. If cell-cycle progres-
sion is essential for cell viability, as it is in these yeasts, how is it possible
to isolate cells that are defective in cell-cycle genes?
362 Chapter 17: The Cell Cycle

17–10 For many experiments, it is desirable to have a population of cells that

are traversing the cell cycle synchronously. One of the first, and still often
used, methods for synchronizing cells is the so-called double thymidine
block. When high concentrations of thymidine are added to the culture
fluid, cells in S phase stop DNA synthesis, though other cells are not
affected. The excess thymidine blocks the enzyme ribonucleotide reduct-
ase, which is responsible for converting ribonucleotides into deoxyribo-
nucleotides. When this enzyme is inhibited, the supply of deoxyribonu-
cleotides falls and DNA synthesis stops. When the excess thymidine is
removed by changing the medium, the supply of deoxyribonucleotides
rises and DNA synthesis resumes normally.
For a cell line with a 22-hour cell cycle divided so that M phase = 0.5
hour, G1 phase = 10.5 hours, S phase = 7 hours, and G2 phase = 4 hours, a
typical protocol for synchronization by a double thymidine block would
be as follows:
1. At 0 hours (t = 0 hours) add excess thymidine.
2. After 18 hours (t = 18 hours) remove excess thymidine.
3. After an additional 10 hours (t = 28 hours) add excess thymidine.
4. After an additional 16 hours (t = 44 hours) remove excess thymidine.
A. At what point in the cell cycle is the cell population when the second thy-
midine block is removed?
B. Explain how the times of addition and removal of excess thymidine syn-
chronize the cell population.

CALCULATIONS
17–11 The fraction of cells in a population that are undergoing mitosis (the
mitotic index) is a convenient way to estimate the length of the cell cycle.
You have decided to measure the cell cycle in the liver of the adult mouse
by measuring the mitotic index. Accordingly, you have prepared liver
slices and stained them to make cells in mitosis easy to recognize. After 3
days of counting, you have found only 3 mitoses in 25,000 cells. Assum-
ing that M phase lasts 30 minutes, calculate the average length of the cell
cycle in the liver of an adult mouse.
17–12 The overall length of the cell cycle can be measured from the doubling
time for a population of exponentially proliferating cells. The doubling
time of a population of mouse L cells was determined by counting the
number of cells in samples of culture fluid at various times (Figure 17–1).
What is the overall length of the cell cycle in mouse L cells?

DATA HANDLING
17–13 You have isolated a temperature-sensitive mutant of budding yeast. It
proliferates well at 25°C, but at 35°C all the cells develop a large bud and
then halt their progression through the cell cycle. The characteristic mor- 107

phology of the cells at the time they stop cycling is known as the land-
mark morphology.
number of cells

It is very difficult to obtain synchronous cultures of this yeast, but you

would like to know exactly where in the cell cycle the temperature-sensi- 106
tive gene product must function—its execution point, in the terminology
of the field—in order for the cell to complete the cycle. A clever friend,
who has a good microscope with a heated stage and a video camera, sug-
gests that you take movies of a field of cells as they experience the tem-
perature increase, and follow the morphology of the cells as they stop 105
0 20 40 60 80 100
cycling. Since the cells do not move much, it is relatively simple to study time (hours)
individual cells. To make sense of what you see, you arrange a circle of
pictures of cells at the start of the experiment in order of the size of their Figure 17–1 Increase in the number of
daughter buds. You then find the corresponding pictures of those same mouse L cells with time (Problem 17–12).
OVERVIEW OF THE CELL CYCLE 363

100

labeled mitotic cells (%)

75

50

increasing 25
bud
size at time of
0
temperature 0 5 10 15 20 25 30
shift
time (hours)

Figure 17–3 Percentage of mitotic cells

that were labeled as a function of time
after a brief incubation with radioactive
thymidine (Problem 17–14).
Problems p17.03/17.03

Figure 17–2 Time-lapse photography of a temperature-sensitive mutant of yeast

(Problem 17–13). Cells on the inner ring are arranged in order of their bud size,
which corresponds to their position in the cell cycle. After 6 hours at 37°C, they
have given rise to the cells shown on the outer ring. No further growth or division
occurs.
Problems p17.02/17.02

cells 6 hours later, when growth and division have completely stopped.
The results with your mutant are shown in Figure 17–2.
A. Indicate on the diagram in Figure 17–2 where the execution point for
your mutant lies.
B. Does the execution point correspond to the time at which the cell cycle is
arrested in your mutant? How can you tell?
17–14 Cells that grow and divide in medium containing radioactive thymidine
incorporate the thymidine into their DNA during S phase. Consider a
simple experiment in which cells were labeled by a brief (30 minute)
exposure to radioactive thymidine. The medium was then replaced
with one containing unlabeled thymidine, and the cells were allowed to
grow and divide for some additional time. At different time points after
replacement of the medium, cells were examined in a microscope. Cells
in mitosis were identified by their condensed chromosomes. The fraction
of mitotic cells that had radioactive DNA was determined by autoradi-
ography and plotted as a function of time after the thymidine labeling
(Figure 17–3).
A. Would all the cells in the population be expected to contain radioactive
DNA after the labeling procedure?
B. Initially, there were no mitotic cells that contained radioactive DNA (Fig-
ure 17–3). Why is this?
C. Explain the rise and fall of the curve in Figure 17–3.
number of cells

D. Given that mitosis lasts 30 minutes, estimate the lengths of the G1, S, and
G2 phases from these data. (Hint: use the points where the curves corre-
spond to 50% labeled mitoses to estimate the lengths of phases in the cell
cycle.)
17–15 Hoechst 33342 is a membrane-permeant dye that fluoresces when it
binds to DNA. When a population of cells is incubated briefly with
Hoechst dye and then sorted in a flow cytometer, which measures the 0 1 2
fluorescence of each cell, the cells display various levels of fluorescence relative fluorescence per cell
as shown in Figure 17–4. Figure 17–4 Analysis of Hoechst 33342
A. Which cells in Figure 17–4 are in the G1, S, G2, and M phases of the cell fluorescence in a population of cells sorted
cycle? Explain the basis for your answer. in a flow cytometer (Problem 17–15).

Problems p17.05/17.04
364 Chapter 17: The Cell Cycle

B. Sketch the sorting distributions you would expect for cells that were
treated with inhibitors that block the cell cycle in the G1, S, or M phase.
Explain your reasoning.

THE CELL-CYCLE CONTROL SYSTEM

TERMS TO LEARN
anaphase-promoting cyclin M-Cdk
complex or cyclosome cyclin–Cdk complex M-cyclin
(APC/C) cyclin-dependent metaphase-to-
Cdc20 kinase (Cdk) anaphase transition
Cdc25 G1-Cdk S-Cdk
Cdh1 G1-cyclin SCF
Cdk-activating kinase (CAK) G1/S-Cdk S-cyclin
Cdk inhibitor protein (CKI) G1/S-cyclin Wee1
cell-cycle control system G2/M transition

DEFINITIONS
Match the definition below with its term from the list above.
17–16 A member of the family of protein kinases that have to be complexed with
a cyclin protein in order to act.
17–17 The ubiquitin ligase that promotes the destruction of a specific set of
proteins, thereby promoting the separation of sister chromatids and the
completion of M phase.
17–18 The cyclin–Cdk complex responsible for stimulating entry into mitosis at
the G2/M checkpoint.
17–19 One of a family of proteins that rise and fall in concentration in step with
the eukaryotic cell cycle, thereby regulating the activity of the crucial pro-
tein kinases that control progression through the cell cycle.
17–20 The final major checkpoint in the cell cycle, where the control system
stimulates sister-chromatid separation, leading to the completion of
mitosis and cytokinesis.
17–21 A timing mechanism that triggers events of the cell cycle in a set sequence,
using feedback from the processes it controls to ensure that one stage is
complete before the next one begins.
17–22 General term for one of the several protein assemblies that form peri-
odically during the cell cycle as the level of cyclin increases, and partially
activate the cyclin-dependent kinase component.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
17–23 The regulation of cyclin–Cdk complexes depends entirely on phosphor-
ylation and dephosphorylation.
17–24 In order for proliferating cells to maintain a relatively constant size, the
length of the cell cycle must match the time it takes for the cell to double
in size.

THOUGHT PROBLEMS
17–25 Vertebrate cells use several different Cdks to manage the various transi-
tions in the cell cycle, yet budding yeast is able to get by with a single Cdk.
How do budding yeast cells manage that neat trick?
THE CELL-CYCLE CONTROL SYSTEM 365

meiotic Figure 17–5 Progesterone- and MPF-induced

germinal metaphase maturation of oocytes (Problem 17–26).
vesicle spindle
progesterone
oocyte egg

50 nL
1000 nL transferred

fresh oocyte egg

DATA HANDLING
17–26 Frog oocytes mature Problems p17.06/17.05
into eggs when incubated with progesterone. Egg
maturation is characterized by disappearance of the nucleus (termed
germinal vesicle breakdown) and formation of a meiotic spindle. The
requirement for progesterone can be bypassed by microinjecting 50 nL of
egg cytoplasm directly into a fresh oocyte (1000 nL), which then matures
normally (Figure 17–5). Progesterone-independent maturation is trig-
gered by maturation-promoting factor (MPF) activity in the egg cyto-
plasm—later called mitosis-promoting factor and shown to be M-Cdk.
At early times after progesterone treatment, inhibition of protein syn-
thesis by cycloheximide blocks egg maturation. However, a few hours
before oocytes become eggs—a time that corresponds to the appearance
of MPF activity—progesterone-induced maturation can no longer be
blocked by cycloheximide.
Is synthesis of MPF itself the cycloheximide-sensitive event? To
test this possibility, you transfer MPF serially from egg to oocyte to test
whether its activity diminishes with dilution. You first microinject 50 nL
of cytoplasm from an activated egg into an immature oocyte as shown
in Figure 17–5; when the oocyte matures into an egg, you transfer 50 nL
of its cytoplasm into another immature oocyte; and so on. Surprisingly,
you find that you can continue this process for at least 10 transfers, even
when the recipient oocytes are bathed in cycloheximide! Moreover, the
apparent MPF activity in the last egg is equal to that in the first egg.
A. What dilution factor is achieved by 10 serial transfers of 50 nL into 1000
nL? Do you consider it likely that a molecule might have an undimin-
ished biological effect over this concentration range?
B. How do you suppose MPF activity can be absent from immature oocytes
yet appear in activated eggs, even when protein synthesis has been
blocked by cycloheximide?
C. Propose a means by which MPF activity might be maintained through
repeated serial transfers.
17–27 You have isolated two temperature-sensitive strains of yeast (which
you’ve named giant and tiny) that show very different responses to
elevated temperature. At high temperature, giant cells grow until they
become enormous, but no longer divide. By contrast, tiny cells have a
short cell cycle and divide when they are much smaller than usual. You
are amazed to discover that these strains arose by different mutations in
the same gene. Based on your understanding of cell cycle regulation by
Cdk1, Wee1, and Cdc25, propose an explanation for how two different
mutations in one of these genes might have given rise to the giant and
tiny strains.
366 Chapter 17: The Cell Cycle

S PHASE
TERMS TO LEARN
Cdc6 geminin
Cdt1 origin recognition complex (ORC)
cohesin prereplicative complex (pre-RC)

DEFINITIONS
Match the definition below with its term from the list above.
17–28 Complex of proteins that holds sister chromatids together along their
length until they separate at mitosis.
17–29 Large protein complex that is bound throughout the cell cycle at origins
of replication in eukaryotic chromosomes.
17–30 Protein that binds to and inhibits a key component of the prereplicative
complex.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
17–31 Initiation of DNA synthesis is permitted only at origins of replication that
contain a prereplicative complex.
17–32 While other proteins come and go during the cell cycle, the proteins of
the origin recognition complex remain bound to the DNA throughout.

THOUGHT PROBLEMS
17–33 In budding yeast, a prereplicative complex, consisting of ORC, Cdc6, and
Mcm proteins, is established at origins of replication during the G1 phase.
S-Cdk then triggers origin firing and helps to prevent re-replication. But
not all yeast origins begin replication at the same time: some fire early in
S phase, while others fire late. How is it possible for S-Cdk to trigger origin
firing at a variety of different times, yet also prevent re-replication at all
origins? Propose a scheme that could account for this behavior of S-Cdk.
17–34 The yeast cohesin subunit Scc1, which is essential for sister-chromatid
cohesion, can be artificially regulated for expression at any point in the
cell cycle. If expression is turned on at the beginning of S phase, all the
cells divide satisfactorily and survive. By contrast, if Scc1 expression is
turned on only after S phase is completed, the cells fail to divide and
they die, even though Scc1 accumulates in the nucleus and interacts effi-
ciently with chromosomes. Why do you suppose that cohesin must be
present during S phase for cells to divide normally?

DATA HANDLING
17–35 Early clues about the regulation of S phase came from studies in which
human cells at various cell-cycle stages were fused to form single cells
with two nuclei. Figure 17–6 shows the outcome of pairwise fusions
between G1, S, and G2 cells. Given what we now know about the roles of
cyclin–Cdk complexes in progression of the cell cycle, how would you
interpret the outcomes of each of these experiments? Do these experi-
ments suggest that there may be a block to re-replication in the cell cycle?
17–36 Using a clever genetic screen, you have identified a temperature-sen-
sitive (ts) mutant in a yeast gene (Scc1) that appears to be required for
S PHASE 367

(A) (B) (C)

G1 S S G2 G1 G2

G1-phase nucleus G2-phase nucleus G2-phase nucleus

immediately enters S phase; stays in G2 phase; stays in G2 phase;
S-phase nucleus continues S-phase nucleus continues G1-phase nucleus enters
DNA replication DNA replication S-phase according
to its own timetable

Figure 17–6 Results of cell-fusion experiments using mammalian cells at different stages of the cell
cycle (Problem 17–35). (A) Fusion of S and G1 cells. (B) Fusion of S and G2 cells. (C) Fusion of G1
and G2 cells.

sister-chromatid cohesion. To Problems p17.11/17.06

assay directly for sister-chromatid cohe-
sion, you insert a tandem array of 336 short DNA sequences, to which
a bacterial protein can bind tightly, adjacent to the centromere of chro-
mosome V. You then express a fusion of the bacterial protein with green
fluorescent protein (GFP) in the same cells. When the GFP fusion protein
binds to its recognition sequences, it creates a bright dot of green fluo-
rescence on the chromosome. To test for the effects of mutant Scc1 on
sister-chromatid cohesion, you isolate unbudded cells from wild-type
and Scc1ts cells that were grown at 25°C, and grow them at 37°C for vari-
ous times. Representative examples of small-budded cells in S phase and
large-budded cells that have passed the metaphase-to-anaphase transi-
tion are shown for both strains in Figure 17–7.
A. Do sister chromatids in wild-type cells adhere to each other normally
during S phase, and separate normally during mitosis? How can you tell?
B. Do sister chromatids in Scc1ts cells adhere normally in S phase, and sepa-
rate normally during mitosis? How can you tell?
C. In the large-budded cells from the Scc1ts strain, why do both sister chro-
matids remain in one cell?

wild type Scc1ts

cells fluorescence cells fluorescence

small-
budded
cells

Figure 17–7 Small- and large-budded

large- cells from wild-type and Scc1ts cells
budded grown at 37°C (Problem 17–36). For each
cells strain, a matched set of pictures shows
the appearance of the cells and the
corresponding sites of fluorescence.
368 Chapter 17: The Cell Cycle

MITOSIS
TERMS TO LEARN
anaphase A interpolar microtubule securin
anaphase B kinetochore separase
astral microtubule kinetochore microtubule spindle assembly
bi-orientation metaphase plate checkpoint
centrosome microtubule flux telophase
condensin mitotic spindle

DEFINITIONS
Match the definition below with its term from the list above.
17–37 Movement of tubulin subunits toward the spindle poles as a result of
addition of new subunits at the plus ends of microtubules and their dis-
assembly at minus ends.
17–38 Stage of mitosis in which the spindle poles move apart.
17–39 Mechanism ensuring that cells do not enter anaphase until all chromo-
somes are correctly bi-oriented on the mitotic spindle.
17–40 Centrally located organelle of animal cells that after duplication organ-
izes each spindle pole during mitosis.
17–41 Imaginary plane midway between the spindle poles in which chromo-
somes are positioned at metaphase.
17–42 Microtubules that overlap in the spindle midzone and interact via their
plus ends, generating an antiparallel array.
17–43 Final stage of mitosis in which the two sets of separated chromosomes
decondense and become enclosed by nuclear envelopes.
17–44 Complex of proteins that uses the energy of ATP hydrolysis to promote
the compaction and resolution of sister chromatids.
17–45 Protease whose activation at the end of metaphase results in the cleavage
of cohesin and the separation of sister chromatids.
17–46 Microtubule that radiates outward from the spindle pole and contacts
the cell cortex, helping to position the spindle in the cell.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
17–47 After the nuclear envelope breaks down, microtubules gain access to the
chromosomes and, every so often, a randomly probing microtubule con-
nects with a kinetochore and captures the chromosome.
17–48 Chromosomes are positioned on the metaphase plate by equal and
opposite forces that pull them toward the two poles of the spindle.
17–49 Once formed, kinetochore microtubules depolymerize at their plus ends
(the ends attached to the kinetochores) throughout mitosis.
17–50 The six stages of M phase—prophase, prometaphase, metaphase, ana-
phase, telophase, and cytokinesis—occur in strict sequential order.

THOUGHT PROBLEMS
17–51 A living cell from the lung epithelium of a newt is shown at differ-
ent stages in M phase in Figure 17–8. Order these light micrographs
MITOSIS 369

(A) (B) (C) Figure 17–8 Light micrographs of a

single cell at different stages of M phase
(Problem 17–51). (Courtesy of Conly L.
Rieder.)

(D) (E) (F)

into the correct sequence, and identify the stage in M phase that each
represents.
17–52 It is remarkable that the concentration
Problems p17.14/17.08of M-cyclin in the cleaving clam
egg rises very slowly and steadily throughout the cell cycle, whereas
M-Cdk activity increases suddenly at mitosis (Figure 17–9). How is
the activity of M-Cdk so sharply regulated in the presence of a gradual
increase in M-cyclin?
Figure 17–9 The rise and fall of M-Cdk
M-Cdk activity
activity and M-cyclin concentration during
relative level

the cell cycle in a cleaving clam egg

concentration (Problem 17–52).
of cyclin B

mitosis interphase mitosis interphase

17–53 Describe the three main classes of spindle microtubule in animal cells
and their functions during mitosis.
17–54 When kinesin-5 motor proteins, which contain two plus-end directed
motor domains, are incubated with microtubules, they will organize the
microtubules into an astral array. How do you suppose such an array is
generated? Will the plus ends of the microtubules be located in the center
of the array or at the periphery? Or will some p17.15/17.09
Problems plus ends be at the center
and others at the periphery?
17–55 Examine the schematic representation of centrosome duplication in
Figure 17–10. By analogy with DNA replication, would you classify cen-
trosome duplication as conservative or semiconservative? Explain your
reasoning.

G2
G1

S Figure 17–10 The centrosome duplication

cycle (Problem 17–55). The individual
centrioles that make up the centrosome
are shown as cylinders.
370 Chapter 17: The Cell Cycle

17–56 How many kinetochores are there in a human cell at mitosis?

17–57 Both sister chromatids of a chromosome occasionally end up in one
daughter cell. Suggest some possible causes for such an event. What
would be the consequences for the daughter cells if this event occurred
in mitosis?
17–58 How can there be a constant poleward flux of tubulin subunits in the
absence of any visible change in the appearance of the spindle?
17–59 Nocodazole reversibly inhibits microtubule polymerization, which is
essential for formation of the mitotic spindle. By treating a population of
mammalian cells with nocodazole for a time and then washing it out of
the medium, it is possible to synchronize the cell population. In the pres-
ence of nocodazole, where in the cell cycle would you expect the cells to
accumulate? What mechanism do you suppose is responsible for stop-
ping cell-cycle progression in the presence of nocodazole?
17–60 Budding yeast cells that are deficient for Mad2, a component of the spin-
dle assembly checkpoint, are killed by treatment with benomyl, which
causes microtubules to depolymerize. In the absence of benomyl, how-
ever, the cells are perfectly viable. Explain why Mad2-deficient cells live
in the absence of benomyl but die in its presence.
17–61 If a fine glass needle is used to manipulate a chromosome inside a liv-
ing cell during early M phase, it is possible to trick the kinetochores on
the two sister chromatids into attaching to the same spindle pole. This
arrangement is normally unstable and is rapidly converted to the stand-
ard arrangement with sister chromatids attached to opposite poles. The
abnormal attachment can be stabilized if the needle is used to gently pull
the chromosome so that the microtubules that attach it to the same pole
are under tension. What does this suggest to you about the mechanism
by which kinetochores normally become attached and stay attached to
microtubules from opposite spindle poles during M phase? Explain your
answer.
17–62 Discuss the following analogy: “Chromosomes are pulled to the spindle
pole like fish on a line.”
17–63 Order the following events in animal cell division.
A. Alignment of chromosomes at the spindle equator.
B. Attachment of microtubules to chromosomes.
C. Breakdown of nuclear envelope.
D. Condensation of chromosomes.
E. Decondensation of chromosomes.
F. Duplication of centrosome.
G. Elongation of the spindle.
H. Pinching of cell in two.
I. Re-formation of nuclear envelope.
J. Separation of centrosomes.
K. Separation of sister chromatids.

CALCULATIONS
17–64 High doses of caffeine (Figure 17–11) interfere with the DNA damage
response in mammalian cells. Why then do you suppose the Surgeon
General has not yet issued an appropriate warning to heavy coffee and
cola drinkers? A typical cup of coffee (150 mL) contains 100 mg of caf-
feine (196 g/mole). How many cups of coffee would you have to drink
to reach the dose (10 mM) required to interfere with the DNA damage
response? (A typical adult contains about 40 liters of water.)
MITOSIS 371

O CH3 Figure 17–11 Structure of caffeine

H3C (Problem 17–64).
N
N

O N
N

CH3

17–65 How much DNA does a single microtubule carry in mitosis? From the
Problems
information in Table p17.17/17.11
17–1, calculate the average length of chromosomes
in each organism in base pairs (bp) and in millimeters (1 bp = 0.34 nm),
and then calculate how much DNA (in base pairs) each microtubule car-
ries on average in mitosis. Do microtubules carry about the same amount
of DNA or does it vary widely in different organisms?

TABLE 17–1 DNA content, haploid number of chromosomes, and microtubules

per chromosome in a variety of organisms (Problem 17–65).
Type of Species DNA Number of Microtubules/
organism content chromosomes chromosome
(bp)
Yeast S. cerevisiae 1.4 × 107 16 1
Yeast S. pombe 1.4 × 107 3 3
Protozoan Chlamydomonas 1.1 × 108 19 1
Fly Drosophila 1.7 × 108 4 10
Human Homo sapiens 3.2 × 109 23 25
Plant Haemanthus 1.1 × 1011 18 120

DATA HANDLING
17–66 The activities of Wee1 kinase and Cdc25 phosphatase determine the
state of phosphorylation of tyrosine 15 in the Cdk1 component of M-Cdk.
When tyrosine 15 is phosphorylated, M-Cdk is inactive; when tyrosine 15
is not phosphorylated, M-Cdk is active (Figure 17–12). Just as the activity
of M-Cdk itself is controlled by phosphorylation, so too are the activities
of Wee1 kinase and Cdc25 phosphatase.
The regulation of these various activities can be studied in extracts
of frog oocytes. In such extracts, Wee1 kinase is active and Cdc25 phos-
phatase is inactive. As a result, M-Cdk is inactive because its Cdk1 com-
ponent is phosphorylated on tyrosine 15. M-Cdk in these extracts can be
rapidly activated by addition of okadaic acid, which is a specific inhibitor
of serine/threonine protein phosphatases. Using antibodies specific for
Cdk1, Wee1, and Cdc25, it is possible to examine their phosphorylation

M-Cdk
active

Wee1 Wee1 Cdc25 Cdc25

kinase kinase phosphatase phosphatase
inactive active active inactive
P Figure 17–12 Control of M-Cdk activity by
M-Cdk Wee1 tyrosine kinase and Cdc25 tyrosine
inactive phosphatase (Problem 17–66).
372 Chapter 17: The Cell Cycle

okadaic Figure 17–13 Effects of okadaic acid

– + – + – +
acid on the phosphorylation states of Cdk1,
kd Wee1, and Cdc25 (Problem 17–66).
Molecular mass markers are shown in
kilodaltons on the left.
107
70

34

Cdk1 Wee1 Cdc25

states by changes in mobility upon gel electrophoresis (Figure 17–13).

(Phosphorylated proteins generally migrate more slowly than their non-
phosphorylated counterparts.)
A. Based on the results with okadaic acid,
Problems decide whether the active forms
p17.19/17.13
of Wee1 kinase and Cdc25 phosphatase are phosphorylated or nonphos-
phorylated. In Figure 17–12, indicate the phosphorylated forms of Wee1
and Cdc25. Also, label the arrows connecting their active and inactive
forms to show which transitions are controlled by protein kinases and
which are controlled by protein phosphatases.
B. Are the protein kinases and phosphatases that control Wee1 and Cdc25
specific for serine/threonine side chains or for tyrosine side chains? How
do you know?
C. How does addition of okadaic acid cause an increase in phosphorylation
of Wee1 and Cdc25, but a decrease in phosphorylation of Cdk1?
D. If you assume that Cdc25 and Wee1 are targets for phosphorylation by
active M-Cdk, can you explain how the appearance of a small amount of
active M-Cdk would lead to its rapid and complete activation?
17–67 Cohesins and condensins are very similar in structure yet carry out quite
different biochemical tasks: cohesion of sister chromatids and conden-
sation of chromosomes, respectively. You are skeptical that such similar
molecules can perform such distinct functions, and set out to determine
if they are truly different using purified components. You incubate pure
cohesin or condensin with nicked circular DNA and ATP. You then add
topoisomerase II to link duplexes that have been juxtaposed. (Topoi-
somerase II binds to one duplex and breaks both strands, attaching itself
covalently to the ends and holding them together. The topoisomerase
II complex can gate the passage of a second duplex through the break
and then reseal the original duplex. By linking—or unlinking—duplexes,
topoisomerase II can alter their topology in informative ways.)
As shown in Figure 17–14, condensin and cohesin yield very differ-
ent results upon incubation with topoisomerase II and analysis by gel

(A) CONDENSIN (B) COHESIN Figure 17–14 Topological analysis of

0 0 the functions of condensin and cohesin
(Problem 17–67). (A) Electrophoretic
analysis of migration of circular
catenanes

DNA incubated with condensin and

topoisomerase II. The knots formed by
incubation are all like the one illustrated.
(B) Electrophoretic analysis of migration
of circular DNA incubated with cohesin
and topoisomerase II. The products
of incubation are catenanes; a dimeric
catenane is shown. In both gels, the
heavy band of material corresponds to
the nicked circles that were added to
knots
the incubation mixture. Orange wedges
indicate increasing concentrations of
1 2 3 4 5 6 7 8 condensin and cohesin.
MITOSIS 373

electrophoresis. Incubation with condensin followed by topoisomerase

generates a particular kind of trefoil knot (Figure 17–14A), whereas incu-
bation with cohesin followed by topoisomerase generates a series of cat-
enanes (circles joined like links of a chain, Figure 17–14B).
A. Do these results support the proposed roles of cohesin and condensin?
How so?
B. Suggest a plausible mechanism by which binding of cohesin might allow
topoisomerase II to link molecules into catenanes.
C. Knots are much more difficult to think about, but often are very revealing
of mechanistic details. See if you can figure out a way to use the binding before dilution
of condensin molecules, coupled with one duplex-crossing event cata-
lyzed by topoisomerase II, to tie a circular molecule into any kind of a
knot.
17–68 A classic paper clearly distinguished the properties of astral microtu-
bules from those of kinetochore microtubules. Centrosomes were used
to initiate microtubule growth, and then chromosomes were added. The
chromosomes bound to the free ends of the microtubules, as illustrated
in Figure 17–15. The complexes were then diluted to very low tubulin
concentration (well below the critical concentration for microtubule after dilution
assembly) and examined again (Figure 17–15). As is evident, only the
Figure 17–15 Arrangements of
kinetochore microtubules were stable to dilution. centrosomes, chromosomes,
A. Why do you think the kinetochore microtubules are stable? kinetochores, and microtubules
B. How would you explain the disappearance of the astral microtubules before and after dilution to low tubulin
after dilution? Do they detach from the centrosome, depolymerize from Problems
concentrationp17.23/17.15
(Problem 17–68).
an end, or disintegrate along their length at random?
C. How would a time course after dilution help to distinguish among these
possible mechanisms for disappearance of the astral microtubules?
17–69 In higher eukaryotes, rare chromosomes containing two centromeres at
different locations are highly unstable: they are literally torn apart at ana-
phase when the chromosomes separate. You wonder whether the same
phenomenon occurs in yeast, whose chromosomes are too small to ana-
lyze microscopically.
You construct a plasmid with two centromeres, as shown in Figure
17–16. Growth of this plasmid in bacteria requires the bacterial origin
of replication (Ori) and a selectable marker (AmpR); its growth in yeast
requires the yeast origin of replication (Ars1) and a selectable marker
(Trp1). You prepare a plasmid stock by growth in E. coli. This dicentric
plasmid transforms yeast with about the same efficiency as a plasmid
that contains a single centromere. Individual colonies, however, were
found to contain plasmids with a single centromere or no centromere,
but never a plasmid with two centromeres. By contrast, colonies that
arose after transformation with a monocentric plasmid invariably con-
tained intact plasmids.
A. Considering their extreme instability in yeasts, why are dicentric plas-
mids stable in bacteria?
B. Why do you suppose the dicentric plasmid is unstable in yeasts?
C. Suggest a mechanism for deletion of one of the centromeres from a Ars1 Cen4 Cen3 Ori
dicentric plasmid grown in yeast. Can this mechanism account for loss of
both centromeric sequences from some of the plasmids?
17–70 Among the variety of microtubule-dependent motors associated with
mitotic spindles are ones that bind to chromosome arms. The role of
one such motor protein, Xkid, during spindle assembly was investigated
by removing the protein (by immunodepletion) from frog egg extracts, Trp1 AmpR
which will form spindles under defined conditions. Extracts that have
Figure 17–16 Structure of a dicentric
Xkid, and immunodepleted extracts that are missing Xkid, both assemble plasmid (Problem 17–69). Cen3 and
normal-looking spindles, as assessed by tubulin staining (Figure 17–17). Cen4 refer to centromeres from yeast
In the presence of Xkid the chromosomes are aligned on the metaphase chromosomes 3 and 4, respectively.
Problems p17.24/17.16
374 Chapter 17: The Cell Cycle

plate (Figure 17–17A), whereas in its absence the chromosomes are dis- (A) + Xkid
persed throughout the spindle (Figure 17–17B).
A. Suggest a mechanism by which Xkid might function to align chromo-
somes on the metaphase plate. Include in your description whether you
think Xkid is a plus-end or a minus-end directed motor, and why.
B. Is Xkid a plausible candidate for the mediator of the polar ejection force
that pushes chromosomes away from the poles?
C. At the transition from metaphase to anaphase, Xkid is normally degraded. (B) – Xkid

What would you expect to happen to the chromosomes if the cell were
unable to degrade Xkid?
17–71 Bipolar spindles assemble in the absence of centrosomes in Sciara when
development occurs parthenogenetically. (Normally, the sperm delivers
a centrosome to the egg along with a haploid genome.) These spindles
look normal except that they lack astral microtubules (Figure 17–18). tubulin DNA

They can also support the rapid, synchronous series of early nuclear divi-
Figure 17–17 Assembly of mitotic
sions that occur in a common cytoplasm in Sciara (similar to the early spindles in frog egg extracts (Problem
nuclear divisions in Drosophila). The products of these early mitotic 17–70). (A) In the presence of Xkid.
events, however, are clearly different in normal embryos and parthe- (B) In the absence of Xkid. The spindle
nogenetic ones. The nuclei in normal embryos are well distributed in microtubules were made visible with
the common cytoplasm (Figure 17–18A), but those in parthenogenetic fluorescent tubulin and the DNA was
embryos are clustered together (Figure 17–18B). Can you suggest a way Problems
visualized p17.29/17.17
with a fluorescent stain.

in which astral microtubules might function to keep nuclei well distrib-

uted in the common cytoplasm?
17–72 At the transition from metaphase to anaphase, M-Cdk is inactivated and
chromosomes begin to separate into sister chromatids. M-Cdk is inacti-
vated by the anaphase-promoting complex (APC/C), which destroys the
cyclin B component of M-Cdk, eliminating its kinase activity. You want to
know how the separation of sister chromatids is related to M-Cdk inacti-
vation. To answer this question, you make cell-free extracts from unferti-
lized frog eggs. When nuclei are added to the extract, they spontaneously
form spindles with condensed chromosomes aligned on the metaphase
plate. Anaphase and the separation of sister chromatids can be triggered
by addition of Ca2+, which activates APC/C and turns off M-Cdk.
To investigate the control of sister-chromatid separation, you make
use of two mutant forms of cyclin B (Figure 17–19). Cyclin B 90 is missing
the destruction box, a sequence of amino acids required for inactivation

(A) NORMAL (B) PARTHENOGENETIC

spindle with asters anastral spindle

embryo after division 1 embryo after division 1

after division 5 after division 5

Figure 17–18 Bipolar spindles and

nuclear divisions in Sciara (Problem
17–71). (A) In normal embryos. (B) In
parthenogenetic embryos.
MITOSIS 375

CYCLIN B Figure 17–19 Cyclin B and two mutants

(Problem 17–72).
destruction
box Cdk1-binding region

CYCLIN B∆90

CYCLIN B13-110

(A) TETRAPOLAR MITOSIS

b aa a
b bb bb
c cc cc
aacd ac d –
by APC/C, but it retains its ability to bind to Cdk1 and make functional b aa a
d b b
M-Cdk. Cyclin B13-110 retains the destruction box, but cannot bind to d c c
Cdk1. When either Problems p17.32/17.19
protein is added in excess to the extract, M-Cdk activ- ddd dd

ity remains high after addition of Ca2+. The two proteins differ, however,
in their effects on chromatid separation. In the presence of cyclin B 90,
sister chromatids separate normally; in the presence of cyclin B13-110, (B) TRIPOLAR MITOSIS: 4 CHROMOSOMES
sister chromatids remain linked.
aaa a
A. Why does M-Cdk remain active in the presence of Ca2+ when cyclin B 90 a bbb b
b – ccc
is added to the extract? d d ddd
bc c aa
B. Why does M-Cdk remain active when cyclin B13-110 is added to the b cd
aa d bb
extract? ccc
dd
C. How is the separation of sister chromatids related to M-Cdk inactivation?
Do sister chromatids separate because a linker protein must be phos-
phorylated by M-Cdk in order for it to hold the chromatids together? Or
do chromatids separate because APC/C degrades the linker protein? (C) TRIPOLAR MITOSIS: 1 CHROMOSOME

17–73 By the turn of the twentieth century, it was clear that chromosomes were
the carriers of hereditary information, but it was not clear whether each a aaa a
chromosome carried the total hereditary information or just a portion. aa
According to the first view, multiple chromosomes were required to aa
raise the total quantity of hereditary material above the threshold value
needed for proper development. According to the second view, multiple
chromosomes were needed so that all portions of the hereditary infor- Figure 17–20 Distributions of
chromosomes among multiple spindles
mation would be represented. This question was answered definitively in (Problem 17–73). (A) Example of a
a classic series of experiments carried out by Theodor Boveri from 1901 random arrangement of three sets
to 1905. of four chromosomes (indicated by
To answer this question, Boveri followed the development of sea Problems
letters) among thep17.34/17.20
four mitotic spindles
urchin eggs that had been fertilized by two sperm, a frequent occurrence in a tetrapolar egg. Note that in this
example, all four cells have at least four
during artificial fertilization. Instead of a normal bipolar mitotic spin- chromosomes. If total number were
dle and division into two cells, these abnormally fertilized eggs form a the critical aspect of chromosomes in
tetrapolar mitotic spindle and then divide into four cells. The three sets of heredity, these cells should develop into
chromosomes—one from the egg and two from the sperm—are distrib- a pluteus. On the other hand, if each of
the four cells has to have at least one
uted randomly among the four spindles as shown for four chromosomes
copy of each different chromosome,
in Figure 17–20A. Sometimes one of the spindle poles fails to form, these cells would not be expected to
resulting in a tripolar mitotic spindle followed by division into three cells form a pluteus because one cell (shaded)
(Figure 17–20B). is missing a chromosome. (B) Example
The species of sea urchin that Boveri studied has a diploid chromo- of a random arrangement of three sets
of four chromosomes among the three
some number of 18, but will develop normally to a free-swimming plu- mitotic spindles in a tripolar egg. Once
teus larva with a haploid number of 9. Boveri reasoned that for a tripolar again, if total number of chromosomes
or tetrapolar egg to develop to a normal pluteus, each cell resulting from were critical, these cells should develop
the initial three-way or four-way division would need to have either 9 into a pluteus; but if chromosome type is
total chromosomes or 9 different chromosomes—depending on which critical, they will not since the shaded
cell is missing one chromosome.
view of chromosome inheritance was correct. Boveri followed the devel- (C) Example of one arrangement—out of
opment of 695 tripolar eggs and found that 58 developed into a normal 10 possible—of three chromosomes on a
pluteus. Among 1170 tetrapolar eggs, none formed a normal pluteus. tripolar spindle.
376 Chapter 17: The Cell Cycle

A. To set up the expectations for this experiment, it is instructive to con-

sider first a hypothetical case in which the egg and two sperm each con-
tribute a single chromosome. For tripolar spindles, there are 10 different
arrangements of three chromosomes on three spindles. Sketch these 10
arrangements. Upon separation of sister chromatids and cell division,
how many of these arrangements would be expected to produce three
cells that each carry at least one chromosome? One arrangement and its
division into three cells is shown in Figure 17–20C. (If you want to try
your hand at tetrapolar spindles, there are 20 arrangements.)
B. If the total number of chromosomes were the critical factor, the number
of tripolar eggs in which each cell gets the minimum number of chromo-
somes would be the same as that calculated in part A, regardless of the
number of chromosomes. By contrast, if the distribution of chromosomes
were the critical factor, the number of tripolar eggs that generate three
cells, each with at least one copy of each different chromosome, would
be expected to decrease with increasing numbers of chromosomes. The
number of plutei should decrease according to the fraction calculated in
part A raised to the power of 9 (the number of different chromosomes).
Which hypothesis—total number of chromosomes or distribution of
chromosomes—do Boveri’s observations support?

CYTOKINESIS
TERMS TO LEARN
cell plate midbody preprophase band
contractile ring phragmoplast syncytium
cytokinesis

DEFINITIONS
Match the definition below with its term from the list above.
17–74 Cytoplasm containing many nuclei enclosed by a single plasma mem-
brane.
17–75 Structure formed at the end of cleavage that can persist for some time as
a tether between the two daughter cells.
17–76 Division of the cytoplasm of a plant or animal cell into two.
17–77 Structure made of microtubules and actin filaments that forms in the
prospective plane of division of a plant cell and guides formation of the
cell plate.
17–78 Circular band containing actin and myosin that forms under the surface
of animal cells undergoing cell division and contracts to pinch the two
daughter cells apart.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
17–79 Cytokinesis follows mitosis as inevitably as night follows day.
17–80 Whether cells divide symmetrically or asymmetrically, the mitotic spin-
dle positions itself centrally in the cytoplasm.

THOUGHT PROBLEMS
17–81 What are the two distinct cytoskeletal machines that are assembled to
carry out the mechanical processes of mitosis and cytokinesis in animal
cells?
CYTOKINESIS 377

17–82 You have obtained an antibody to myosin that prevents the movement
of myosin molecules along actin filaments. If this antibody were injected
into cells, would you expect the movement of chromosomes at anaphase
to be affected? How would you expect antibody injection to affect cytoki-
nesis? Explain your answers.
17–83 If a cell just entering mitosis is treated with nocodazole, which destabi-
lizes microtubules, the nuclear envelope breaks down and chromosomes
condense, but no spindle forms and mitosis arrests. In contrast, if such
a cell is treated with cytochalasin D, which destabilizes actin filaments,
mitosis proceeds normally, but a binucleate cell is generated and pro-
ceeds into G1. Explain the basis for the different outcomes of these treat-
ments with cytoskeleton inhibitors. What do these results tell you about
cell-cycle checkpoints in M phase?

CALCULATIONS
17–84 When cells divide after mitosis, their surface area increases—a natural
consequence of dividing a constant volume into two compartments. The
increase in surface requires an increase in the amount of plasma mem-
brane. One can estimate this increase by making certain assumptions
about the geometry of cell division. Assuming that the parent cell and the
two progeny cells are spherical, one can apply the familiar equations for
the volume [(4/3) r3] and surface area (4 r2) of a sphere.
A. Assuming that the progeny cells are equal in size, calculate the increase
in plasma membrane that accompanies cell division. (Although this
problem can be solved algebraically, you may find it easier to substitute
real numbers. For example, let the volume of the parent cell equal 1.) Do
you think that the magnitude of this increase is likely to cause a problem
for the cell? Explain your answer.
B. During early development, many fertilized eggs undergo several rounds
of cell division without any overall increase in total volume. For example,
Xenopus eggs undergo 12 rounds of division before growth commences
and the total cell volume increases. Assuming once again that all cells
are spherical and equal in size, calculate the increase in plasma mem-
brane that accompanies development of the early embryo, going from
one large cell (the egg) to 4096 small cells (12 divisions).

DATA HANDLING
17–85 Megakaryocytes, which are the precursor cells of blood platelets, undergo
a unique differentiation program, becoming polyploid through repeated
cycles of DNA synthesis without concomitant cell division. Such cells
contain some 4 to 128 times the normal DNA content in a single large
nucleus. Ultimately, mature megakaryocytes begin to bud off platelets as
shown in Figure 17–21. Careful observations of individual precursor cells
that were stimulated to undergo polyploidization show the sequence
of events in Figure 17–22. How do these events differ from the normal
sequence in cell division? At what stage in M phase do these cells deviate
from normal cells? What sorts of molecular differences might you expect
to find among the components involved in M phase in these cells versus
normal cells?

MEDICAL LINKS
17–86 Globoid cell leukodystrophy (GLD, also known as Krabbe’s disease) is a
hereditary metabolic disorder characterized morphologically by distinc-
tive multinucleated globoid cells in the white matter of the brain. Defi- Figure 17–21 A megakaryocyte budding
ciency of an enzyme of sphingolipid catabolism leads to accumulation off platelets (Problem 17–85).

Problems p17.38/17.21
378 Chapter 17: The Cell Cycle

(A) NORMAL CELL CYCLE

G1 phase S phase prophase metaphase anaphase A anaphase B telophase cytokinesis

2N to 4N

(B) CELL CYCLE IN MEGAKARYOCYTE PRECURSOR CELLS

first cycle
Figure 17–22 Schematic diagrams of cell
cycles (Problem 17–85). (A) Cell cycle
in normal cells. (B) Two cell cycles in a
megakaryocyte precursor cell.

G1 G2 metaphase anaphase

second cycle

G1 G2 metaphase anaphase

of psychosine in the brain (Figure 17–23A). Psychosine binds to a G-pro-

tein-coupled receptor that is expressed in only a few cell types. To test
whether there might be a relationship between psychosine, its receptor,
and multinucleate cells, you express the psychosine receptor in cells
Problems
that normally lack it, and measure p17.39/17.22
the effects of psychosine treatment by
FACS (fluorescence-activated cell sorting) analysis (Figure 17–23B).
Do these results support the idea that psychosine acts through its
receptor to inhibit cytokinesis? Explain your reasoning.

(A) STRUCTURE OF PSYCHOSINE

CH2OH NH3
O
HO O
H
OH H
H H
OH
H OH
galactose

(B) FACS ANALYSIS – psychosine + psychosine

cell number

– receptor

Figure 17–23 Analysis of the role

of psychosine in the generation of
multinucleate cells (Problem 17–86).
cell number

(A) Structure of psychosine. (B) Effects

+ receptor
of psychosine on cells that do or do not
express the psychosine receptor.
FACS analysis measures the DNA
101 102 103 104 101 102 103 104 content of individual cells using a
DNA content DNA content fluorescent DNA dye.
MEIOSIS 379

MEIOSIS
TERMS TO LEARN
bivalent nondisjunction
chiasma pairing
meiosis I synaptonemal complex
meiosis II

DEFINITIONS
Match the definition below with its term from the list above.
17–87 The failure of homologs to separate properly.
17–88 The unique form of cell division that segregates homologs.
17–89 An inter-homolog connection that arises from an individual crossover
event between nonsister chromatids.
17–90 A four-chromatid structure that arises from the close juxtaposition of
homologs as prophase progresses.
17–91 The closely packed array of transverse filaments that links the axial cores
of paired homologs.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
17–92 Meiosis segregates the paternal homologs into sperm and the maternal
homologs into eggs.
17–93 During meiosis I, the sister chromatids of each homolog are pulled apart
and separated into different daughter nuclei.
17–94 In most species, paired homologs are locked together by homologous
recombination events that lead to reciprocal DNA exchanges between
nonsister chromatids.

THOUGHT PROBLEMS
17–95 You are studying a mutant of maize called plural abnormalities of meiosis
I (PamI), which is severely compromised in meiosis. When you visualize
paired chromosomes in the pachytene stage of prophase I, you observe
many examples of the kind of defects shown in Figure 17–24 in PamI
homozygous strains. These abnormal structures are rarely if ever present
in wild-type cells that are undergoing meiosis. What is abnormal about
the chromosomes in Figure 17–24?
17–96 Down syndrome (trisomy 21) and Edwards syndrome (trisomy 18) are
the most common autosomal trisomies seen in human infants. Does this
fact mean that these chromosomes are the most difficult to segregate
properly during meiosis?

CALCULATIONS
17–97 The human genome consists of 23 pairs of chromosomes (22 pairs of auto-
somes and one pair of sex chromosomes). During meiosis, the maternal
and paternal sets of homologs pair, and then are separated into gametes,
so that each contains 23 chromosomes. If you assume that the chromo-
somes in the paired homologs are randomly assorted to daughter cells, Figure 17–24 Electron micrograph of
chromosomes in PamI cells undergoing
how many potential combinations of paternal and maternal homologs meiosis (Problem 17–95). A region of
can be generated during meiosis? (For the purposes of this calculation, the PamI nucleus containing abnormal
assume that no recombination occurs.) structures is shown.

Problems p17.201/17.24
380 Chapter 17: The Cell Cycle

Figure 17–25 Shugoshin (Problem 17–98). (A) Expression of various (A)

combinations of cohesins and Sgo1 in S. pombe. Rad21 is the normal
mitotic cohesin. Absence of growth is apparent when Rec8 and Sgo1 are +Rad21 +Rad21
coexpressed. (B) Shugoshin, a guardian spirit of Japanese temples. +Sgo1

+Rec8 +Rec8
+Sgo1
DATA HANDLING
17–98 In mitosis, sister chromatids are held together by cohesins, which are
cleaved at the metaphase-to-anaphase transition, allowing the sister (B)
chromatids to separate at anaphase. The same is true for meiosis, except
that sister chromatids stay together during meiosis I and then separate
during meiosis II. How is cohesion of sisters maintained during meio-
sis I? In the fission yeast S. pombe, a special cohesin subunit called Rec8
is expressed only during meiosis, and found to be concentrated at cen-
tromeres, where meiotic homologs are stuck together until meiosis II.
Rec8 is essential for meiosis and cannot be replaced by the mitotic ver-
sion known as Rec21. If Rec8 is deliberately expressed during mitosis,
when it is not normally present, it does not inhibit mitosis because it is
degraded. Thus, you suspect that another protein stops Rec8 degrada-
tion during meiosis I. To find this putative inhibitor protein, you devise
a clever search strategy. You search for other proteins that when coex-
pressed with Rec8 during mitosis cause cell death. As shown in Figure
17–25A, you find one protein, which you name shugoshin (Sgo1) after
the guardian spirits of Japanese temples (Figure 17–25B), that is lethal
when coexpressed with Rec8 during mitosis.
A. Explain why coexpression of Rec8 and Sgo1 in mitosis is lethal.
B. How might Sgo1 act to prevent sister-chromatid separation during meio-
sis I? Problems p17.203/17.25
C. What do you suppose would happen if Sgo1 were not expressed during
meiosis I?

CONTROL OF CELL DIVISION AND CELL GROWTH

TERMS TO LEARN
ATM mitogen replicative cell senescence
ATR Myc retinoblastoma protein (Rb)
E2F protein p53 telomerase
G0 Ras telomere
growth factor

DEFINITIONS
Match the definition below with its term from the list above.
17–99 A specialized, nondividing state that cells enter by partly disassembling
their cell-cycle control system and exiting from the cell cycle.
17–100 Extracellular substance that stimulates cell growth.

17–101 End of a chromosome, associated with a characteristic DNA sequence

that is replicated in a special way.
17–102 Phenomenon observed in primary cell cultures as they age, in which cell
proliferation slows down and finally halts.
17–103 Extracellular substance that stimulates cell division.

17–104 Gene regulatory factor that is activated by G1-Cdk complexes in animal

cells and binds to specific DNA sequences in the promoters of genes that
encode proteins required for S-phase entry.
CONTROL OF CELL DIVISION AND CELL GROWTH 381

17–105 The protein kinase that is initially activated in response to x-ray-induced

DNA damage and is defective in patients with ataxia telangiectasia.

TRUE/FALSE
Decide whether each of these statements is true or false, and then explain why.
17–106 Serum deprivation causes proliferating cells to stop where they are in the
cell cycle and enter G0.
17–107 Budding yeast and mammalian cells respond to DNA damage in the
same way: they transiently arrest their cell cycles to repair the damage
and if repair cannot be completed, they resume their cycles despite the
damage.
17–108 If we could turn on telomerase activity in all our cells, we could prevent
aging.

THOUGHT PROBLEMS
17–109 How do mitogens, growth factors, and survival factors differ from one
another?
17–110 For each of the following, decide whether such cells exist in humans and,
if they do, give examples.
A. Cells that do not grow and do not divide.
B. Cells that grow, but do not divide.
C. Cells that divide, but do not grow.
D. Cells that grow and divide.
17–111 Why do you suppose cells have evolved a special G0 state to exit the cell
cycle, rather than just stopping in G1 at a G1 checkpoint?
17–112 Platelet-derived growth factor (PDGF) is encoded by a gene that can
cause cancer when expressed inappropriately. Why then do cancers not
arise at wounds when PDGF is released from platelets?
17–113 One important biological effect of a large dose of ionizing radiation is to
halt cell division.
A. How does a large dose of ionizing radiation stop cell division?
B. What happens if a cell has a mutation that prevents it from halting cell
division after being irradiated?
C. What might be the effects of such a mutation if the cell was not irradi-
ated?
D. An adult human who has reached maturity will die within a few days of
receiving a radiation dose large enough to stop cell division. What does
this tell you (other than that one should avoid large doses of radiation)?
17–114 What do you suppose happens in mutant cells with the following defects?
A. Cannot degrade M-phase cyclins.
B. Always express high levels of p21.
C. Cannot phosphorylate Rb.
17–115 Replicative cell senescence occurs at a characteristic number of popu-
lation doublings, typically about 40 for cells taken from normal human
tissue. This observation suggests that in some way individual cells can
“count” the number of times they have divided. How does the structure
of telomeres figure into a cell’s calculations?
17–116 Liver cells proliferate both in patients with alcoholism and in patients
with liver tumors. What do you suppose are the differences in the mecha-
nisms by which cell proliferation is induced in these diseases?