Thermodynamic analysis of F1-ATPase rotary catalysis using high-speed imaging

Article abstract of DOI:10.1002/pro.2559

F₁-ATPase (F₁) is a rotary motor protein fueled by ATP hydrolysis. Although the mechanism for coupling rotation and catalysis has been well studied, the molecular details of individual reaction steps remain elusive. In this study, we performed high-speed imaging of F₁ rotation at various temperatures using the total internal reflection dark-field (TIRDF) illumination system, which allows resolution of the F₁ catalytic reaction into elementary reaction steps with a high temporal resolution of 72 μs. At a high concentration of ATP, F₁ rotation comprised distinct 80° and 40° substeps. The 80° substep, which exhibited significant temperature dependence, is triggered by the temperature-sensitive reaction, whereas the 40° substep is triggered by ATP hydrolysis and the release of inorganic phosphate (P_i). Then, we conducted Arrhenius analysis of the reaction rates to obtain the thermodynamic parameters for individual reaction steps, that is, ATP binding, ATP hydrolysis, P_i release, and TS reaction. Although all reaction steps exhibited similar activation free energy values, ΔG^? = 53-56 kJ mol^-1, the contributions of the enthalpy (ΔH^?), and entropy (ΔS^?) terms were significantly different; the reaction steps that induce tight subunit packing, for example, ATP binding and TS reaction, showed high positive values of both ΔH^? and ΔS^?. The results may reflect modulation of the excluded volume as a function of subunit packing tightness at individual reaction steps, leading to a gain or loss in water entropy.

Full text of DOI:10.1002/pro.2559

Thermodynamic analysis of F1-ATPase
rotary catalysis using high-speed imaging
Rikiya Watanabe,1,2 Yoshihiro Minagawa,1 and Hiroyuki Noji1*
1Department of Applied Chemistry, School of Engineering, The University of Tokyo, Bunkyo-ku, Tokyo, 113-8656, Japan
2PRESTO, JST, Bunkyo-ku, Tokyo, 113-8656, Japan
Received 28 July 2014; Accepted 23 September 2014
DOI: 10.1002/pro.2559
Published online 27 September 2014 proteinscience.org
Abstract: F1-ATPase (F1) is a rotary motor protein fueled by ATP hydrolysis. Although the mecha-
nism for coupling rotation and catalysis has been well studied, the molecular details of individual
reaction steps remain elusive. In this study, we performed high-speed imaging of F1 rotation at var-
ious temperatures using the total internal reflection dark-field (TIRDF) illumination system, which
allows resolution of the F1 catalytic reaction into elementary reaction steps with a high temporal
resolution of 72 ms. At a high concentration of ATP, F1 rotation comprised distinct 80ꢀ and 40ꢀ sub-
steps. The 80ꢀ substep, which exhibited significant temperature dependence, is triggered by the
temperature-sensitive reaction, whereas the 40ꢀ substep is triggered by ATP hydrolysis and the
release of inorganic phosphate (Pi). Then, we conducted Arrhenius analysis of the reaction rates to
obtain the thermodynamic parameters for individual reaction steps, that is, ATP binding, ATP
hydrolysis, Pi release, and TS reaction. Although all reaction steps exhibited similar activation free
energy values, DG‡ 5 53–56 kJ mol21, the contributions of the enthalpy (DH‡), and entropy (DS‡)
terms were significantly different; the reaction steps that induce tight subunit packing, for example,
ATP binding and TS reaction, showed high positive values of both DH‡ and DS‡. The results may
reflect modulation of the excluded volume as a function of subunit packing tightness at individual
reaction steps, leading to a gain or loss in water entropy.
Keywords: single-molecule biophysics; temperature dependence; molecular motor protein; F1-
ATPase; FoF1-ATP synthase
rotary torque of 40 pN nm for thermophilic Bacillus
PS3 F1 (TF1)11 and 56 pNÁnm for Escherichia coli F1
(EF1).2,12–14 The mechanical work done by one c
rotation, which is estimated to be 2p 3 40 5 ꢁ240
pN nm, is comparable to the energy released from
hydrolysis of ATP molecules. Therefore, F1 is
extremely efficient at converting chemical energy to
mechanical work, and catalysis is tightly coupled
with mechanical work,11,15 which is the prominent
feature of F1 among ATP-driven motor proteins.
Introduction
F1-ATPase (a3b3c), a catalytic subcomplex of FoF1-
ATP synthase, is a rotary motor protein that uses
the energy from ATP hydrolysis to rotate the rotor g
subunit against the surrounding stator of the a3b3
ring.1–4 The three catalytic sites of F1 are located
mainly on the b subunits at each a and b inter-
face.5–7 The rotary motion of F1 can be visualized by
optical microscopy.8–10 Upon ATP hydrolysis, F1
rotates the rotary shaft counterclockwise, generating
The F1 reaction scheme has been established by
3,16
single-molecule studies using TF1
According to
Grant sponsor: The Ministry of Education, Culture, Sports, Sci-
ence and Technology, Japan; Grant numbers: 00343111;
30540108; Grant sponsor: Precursory Research for Embryonic
Science from the Japan Science and Technology Agency.
the current mechanistic understanding (Fig. 1),
hydrolysis or turnover of a single ATP molecule at
each of the three catalytic sites, which are located
120ꢀ apart, is coupled with one revolution of the c
subunit. The rotation step size is 120ꢀ, and each
step is coupled to the turnover of a single ATP
*Correspondence to: Hiroyuki Noji; Department of Applied
Chemistry, School of Engineering, The University of Tokyo,
Tokyo, Japan. E-mail: hnoji@appchem.t.u-tokyo.ac.jp
C
V
Published by Wiley-Blackwell. 2014 The Protein Society
PROTEIN SCIENCE 2014 VOL 23:1773—1779 1773

F1 rotation at various temperatures using the
recently developed total internal reflection dark-field
(TIRDF) illumination system,31 which allows us to
monitor rotation with high spatiotemporal resolution
(ꢁ5 nm and 72 ms). Herein, we present the thermo-
dynamic parameters for the elementary reaction
steps, which represent a major contribution toward
a more detailed understanding of the molecular
mechanism of F1 rotary catalysis.
Results
Temperature-dependence of TF1 rotation
We observed the rotation of TF1 with an 80-nm gold
colloid probe at saturating ATP (200 mM) with a
TIRDF illumination system.31 The viscous friction
exerted by the gold colloid on the g subunit was low
Figure 1. Chemo-mechanical coupling scheme of F1-
ATPase. Red arrows represent the angular positions of the c
subunit. Blue areas represent the catalytic actions of the indi-
vidual catalytic sites. F1 rotates 120ꢀ upon hydrolysis of 1
ATP, comprising the 80ꢀ and 40ꢀ substeps. The 80ꢀ substep
is driven by ATP binding, ADP release, and the temperature-
sensitive (TS) reaction, and the 40ꢀ substep is driven by ATP
hydrolysis and Pi release.
17
enough to allow full-speed rotation of F1 When
observed between 4 and 33ꢀC, F1 showed continuous
counterclockwise rotation [Fig. 2(a)]. Therefore, we
molecule.11 The 120ꢀ step is further divided into 80ꢀ
and 40ꢀ substeps.17,18 The 80ꢀ substep is triggered
by ATP binding and ADP release,19,20 whereas the
40ꢀ substep is triggered by ATP hydrolysis and the
release of inorganic phosphate (Pi).16,18,20,21 The
angular positions of F1 before the 80ꢀ and 40ꢀ sub-
steps are referred to as the ATP binding and cata-
lytic angles, respectively.
The scheme above was obtained at room temper-
ature (ꢁ23ꢀC), whereas the optimum temperature
for TF1 is 75ꢀC,22 which is higher than for other spe-
cies.23,24 Recently, we observed the rotary motion of
TF1 at various temperatures.25,26 At temperatures
below 9ꢀC, we detected the presence of a new reac-
tion intermediate as a clear intervening pause before
the 80ꢀ substep (Fig. 1).25,26 The rate constant for
this step was remarkably sensitive to temperature
and increased by a factor of 6–19 for every 10ꢀC rise
in temperature (Q10 5 6–19),25,26 which is unusually
high compared to conventional Q10 values of ꢁ2.
Moreover, as expected from the high Q10 value,
which reflects a high activation energy barrier, this
reaction step involves a large conformational change
that presumably contributes to torque generation.27
Hereafter, this step is referred to as the
temperature-sensitive (TS) reaction.
Figure 2. Rotation of F1-ATPase at various temperatures. (a)
Time course of F1 rotation in the presence of 200 mM ATP at
11ꢀC (cyan), 16ꢀC (blue), 20ꢀC (green), 25ꢀC (orange), and
33ꢀC (red). The gold colloidal bead (/e 5 80 nm) was used
as the rotational probe. (b) Temperature dependence of the
rotational rate at saturating ATP (200 mM). The rotational rates
determined in the rotational assay using a gold colloidal bead
(red) are plotted together with our previous result using a
magnetic bead (green) and with the rate in bulk solution esti-
mated as onethird of the ATP hydrolysis rate (blue).25 (c)
Arrhenius plots of traces in Figure 2(b).
Except for the TS reaction, the temperature
dependence of the reaction rate has not been exam-
ined at the resolution of elementary reaction steps,
for example, hydrolysis and product release,
although some previous studies attempted to observe
F1 rotation at various temperatures.25,26,28–30 In the
present study, we performed high-speed imaging of
1774 PROTEINSCIENCE.ORG
Thermodynamic Analysis of F1-ATPase Rotation

between the two values around room temperature
[Fig. 2(b,c)].25
Rotational substeps at saturating ATP
concentration
We were able to clearly resolve individual steps in
F1 rotation at 27,000 frames s21 using the 80 nm
bead even at saturating ATP (200 mM). Although we
did not previously detect obvious substeps in the
rotation assay at 200 mM ATP when recording below
8000 frames s21 17
we found molecules in this study
,
that exhibited distinct 80ꢀ and 40ꢀ substeps at room
temperature after careful and thorough observations
[Fig. 3(a,b)], which is similar to observations made
by high-speed imaging of EF1.32 To identify the rate-
limiting reaction of the 80ꢀ substep, we analyzed the
dwell time prior to the 80ꢀ substep at various tem-
peratures. The dwell time distribution exhibited
strong temperature dependence [Fig. 3(c)]. By fitting
the distribution to an exponential function,
y 5 CÁexp(2kÁt), the rate constants at 16, 20, 25, and
33ꢀC were determined to be 287, 1098, 1654, and
5540 s21 [Fig. 3(c)], which are much larger than the
rate constants for binding 200 mM ATP estimated
from the previous study.25 In addition, these fits
gave Q10 5 6.0 [Fig. 4(a)], which is consistent with
previous studies of the TS reaction at low tempera-
tures.25,26 Thus, we confirmed that the rate-limiting
step of the dwell prior to the 80ꢀ substep is the TS
reaction. The rate-limiting reactions of the 40ꢀ sub-
step at saturating ATP were identified as ATP
hydrolysis and Pi release in our previous
study.17,18,20 We analyzed the dwell time before the
40ꢀ substep, which includes the waiting times for
Figure 3. Dwell time analysis of F1-ATPase. (a) Time course
of the rotation of F1 in the presence of 200 mM ATP at 16ꢀC.
The pauses before the 80ꢀ or 40ꢀ substeps are shown as
blue or red, respectively. (b) Histogram of angular positions
from the time course shown in Figure 3(a). (c) Histograms of
the dwell times of the pause before the 80ꢀ substep at 16ꢀC
(blue), 20ꢀC (green), 25ꢀC (orange), and 33ꢀC (red). Curves
were plotted using a first-order reaction scheme,
y 5 CÁexp(2kÁt), where k (16ꢀC) 5 287 s21, k
(20ꢀC) 5 1098 s21, k (25ꢀC) 5 1654 s21, and k
(33ꢀC) 5 5540 s21. (d) Histograms of the dwell times of the
pause before the 40ꢀ substep at 16ꢀC (blue), 20ꢀC (green),
25ꢀC (orange), and 33ꢀC (red). Curves were plotted using a
sequential reaction scheme, y 5 CÁ(exp(2k1Át) 2exp(2k2Át)),
where k1 (16ꢀC) 5 420 s21, k1 (20ꢀC) 5 491 s21, k1
(25ꢀC) 5 710 s21, k1 (33ꢀC) 5 1332 s21, k2 (16ꢀC) 5 2085 s21
k2 (20ꢀC) 5 2500 s21, k2 (25ꢀC) 5 3507 s21, and k2
,
(33ꢀC) 5 5009 s21
.
measured the rotational rate of F1 at various tem-
peratures [Fig. 2(b,c)]. Although the previously
reported rotational rate of F1 using magnetic beads
(/ 5 ꢁ200 nm) was limited by viscous friction from
the probe at room temperature (>9ꢀC),25 the rota-
tional rate measured in this study was not affected
by viscous friction but was determined by the reac-
tion steps comprising ATP hydrolysis, for example,
hydrolysis and product release. In support of this
conclusion, the rotational rate of F1 was close to the
estimated rotational rate, which is onethird of the
ATPase activity in bulk solution because three ATPs
are hydrolyzed for every turn; a previous study
using magnetic beads showed a large difference
Figure 4. Arrhenius plots of reaction rates. (a) Arrhenius plots
of the rates of the temperature-sensitive (TS) reaction (red),
ATP hydrolysis (yellow), and Pi release (cyan) determined
from Figure 3. Solid lines represent linear fits. (b) Magnifica-
tion of Figure 4(a).
Watanabe et al.
PROTEIN SCIENCE VOL 23:1773—1779 1775

hydrolysis, and 53 kJ mol21 for Pi release. Here, we
note that DG‡ determined in this study agrees with
calculations from computational simulations,34,35
supporting the validity of the computational calcula-
tions applied to the F1 catalytic mechanism.
Discussion
In this study, we assessed the temperature-
dependence of the reaction rates of individual steps of
TF1 ATP hydrolysis. For comparison, we reanalyzed
the Arrhenius plot of the bulk phase ATPase rate
shown in Figure 2(c).25 The Arrhenius plot showed a
breakpoint around 9ꢀC [Fig. 2(c)], which suggests a
change in rate-limiting step. A similar phenomenon
was observed for F1 from bovine mitochondria36 and
Escherichia coli.37 From the Arrhenius plot, ATPase
rates below or above 9ꢀC are in good agreement with
the rate constants of the TS reaction or ATP hydroly-
sis, respectively, as determined from single-molecule
assays (Fig. 3). Thus, the comparison shows that the
rate-limiting process of F1 rotary catalysis is tempera-
ture dependent; that is, the TS reaction dominates
the catalytic turnover rate of F1 below 9ꢀC while ATP
hydrolysis dominates above 9ꢀC.
Figure 5. Thermodynamic parameters. Thermodynamic
parameters at 25ꢀC. DG‡ (green), DH‡ (blue), and TDS‡
(yellow) for the temperature-sensitive (TS) reaction, ATP
hydrolysis, and Pi release were calculated from Figure 4(a,b),
while that for ATP binding at 200 mM ATP was calculated
from Watanabe et al.25
both ATP hydrolysis and Pi release. The dwell-time
distribution exhibited a typical convex curve indica-
tive of
a
sequential reaction, y 5 C(exp(2k1Át)
2exp(2k2Át)), which yielded k1 and k2 values for 16,
20, 25, and 33ꢀC of 420 and 2085 s21, 491 and
2500 s21, 710 and 3507 s21, and 1332 and 5009 s21
,
We estimated the rotational rate of TF1 under
physiological conditions, that is, 65ꢀC. According to
the Arrhenius plot [Fig. 4(a)], the rate constant for
ATP hydrolysis at 65ꢀC, where it is the rate-limiting
step for rotation is 8.6 3 103 s21. The rate constant
for rotation of TF1 under physiological conditions is
thus estimated to be (8.6 3 103)/3 5 2.9 3 103 s21
based on the coupling ratio of 3 ATPs per turn. The
physiological rate constant for EF1 rotation is also
estimated to be ꢁ103 s21 from the previous study,29
which shows good agreement with that of TF1.
Accordingly, it is highly probable that the physiologi-
cal rotational rate constant is inherent to F1 regard-
less of species.
respectively [Fig. 3(d)]. Our previous study sug-
gested that the slower step corresponds to ATP
hydrolysis while the faster step represents Pi
release.33 Accordingly, the Q10 factors of ATP hydro-
lysis and Pi release were 1.9 and 1.6, respectively
[Fig. 4(a,b)], which are consistent with the expecta-
tion from previous studies.25,26,28,29
Arrhenius analysis of reaction rates
The Arrhenius plots of the individual reaction rates
are shown in Figure 4(a,b). All Arrhenius plots fit
well to linear functions, which allowed determina-
tion of the transition-state thermodynamic parame-
ters, DG‡, DH‡, and TDS‡, for individual reaction
steps (Fig. 5). The slope of the TS reaction was appa-
rently steeper than the slopes of other elementary
steps, that is, ATP hydrolysis and Pi release, giving
DH‡ values of 121 kJ mol21 for the TS reaction, 51
kJ mol21 for ATP hydrolysis, and 39 kJ mol21 for Pi
release. In our previous study, the DH‡ for ATP bind-
ing at 200 mM ATP was determined to be 107 kJ
mol21, which is almost the same as that of the TS
reaction.25 The y intercept of the TS reaction was
apparently greater than those of the other elemen-
tary steps; at 25ꢀC, TDS‡ values are 67 kJ mol21 for
the TS reaction, 25.4 kJ mol21 for ATP hydrolysis,
and 214 kJ mol21 for Pi release. As was observed
for DH‡, the previously determined TDS‡ of ATP
binding at 200 mM ATP, 54 kJ mol21, was similar to
that of the TS reaction.25 Thus, the activation free
energy, DG‡ 5 DH‡ 2TDS‡, was calculated to be 54
kJ mol21 for the TS reaction, 56 kJ mol21 for ATP
The activation free energy determined in this
study, 53–56 kJ mol21, is almost the same among
elementary reaction steps and shows good agree-
ment with that estimated from the bulk phase
ATPase activity of EF1.29,38 Therefore, the activation
free energy is inherent to F1, regardless of elemen-
tary reaction step or species. The phenomenon that
the activation free energy is equal among elemen-
tary reaction steps is a well-known feature of opti-
mally evolved enzymes.39,40 Recent NMR studies
elucidated that conformational fluctuations of an
enzyme active site occur on the time scale of cata-
lytic turnover (<1 ms),41,42 which suggests that con-
formational fluctuations might constrain the
catalytic turnover rate of an enzyme; alternatively,
there may be no selective pressure to reduce cata-
lytic time scales below those of conformational fluc-
tuation. In fact, the time scales of most steps of the
F1 reaction, that is, ATP hydrolysis, Pi release, and
the TS reaction, are around 1 ms, which coincides
1776 PROTEINSCIENCE.ORG
Thermodynamic Analysis of F1-ATPase Rotation

with that of conformational fluctuations. Thus, F1
catalytic proficiency appears to be optimally evolved,
leading to uniformity of the activation free energies
among the elementary reaction steps.
upon ATP hydrolysis and Pi release at the catalytic
angle.52 To discriminate between these models, we
hope to simultaneously visualize the conformational
change in the b subunit and the rotational motion
with spatiotemporal resolution sufficient to capture
cracking or local unfolding of the b conformation.
Using high-speed imaging of F1 rotation, we
revealed the temperature dependence of the F1 reac-
tion rate at the resolution of elementary reaction
steps. Such information is difficult to obtain from
conventional single-molecule and bulk assays due to
low temporal resolution. Thus, we are convinced
that the high-speed imaging performed in this study
is a powerful tool for understanding the mechanism
of F1 rotary catalysis and holds promise for under-
standing the function of other molecular machines.
The limitation to the current system is the
requirement of a gold colloid rotary probe with diame-
ter above 40 nm for visualization of the rotary motion
of F1 with high temporal resolution, 72 ms. Recently,
we reported that the viscous friction exerted on the
rotary probe affects the reaction rate constants; the
reaction rates of Pi release and the TS reaction are
decelerated as the viscous friction increases.33 To
determine the thermodynamic parameters under
physiological (viscous friction-free) conditions, the
development of a high-speed imaging system with a
smaller rotary probe is highly anticipated.
Although the activation free energy was similar
among the reaction steps, the contributions from the
enthalpy and entropy terms differed significantly
(Fig. 5). The reaction steps that occur at the ATP-
binding angle (Fig. 1), that is, ATP binding and the
TS reaction, exhibited high positive values of DH‡
and TDS‡, while those that occur at the catalytic
angle (Fig. 1), that is, ATP hydrolysis and Pi release,
exhibited small positive DH‡ values and slightly neg-
ative TDS‡ values. When we focus on the interaction
between ATP and the catalytic site, the TDS‡ values
for ATP binding and Pi release determined in this
study seem to conflict with expectations based on
the crystal structure:7 hydrogen-bond formation
upon ATP binding induces
a negative entropy
change, so its disruption upon Pi release necessarily
induces a positive entropy change. Therefore, it is
highly expected that the thermodynamic parameters
determined in this study reflect the overall dynam-
ics at catalytic subunits upon individual reaction
steps. Here, we focus on the entropic contribution
associated with subunit packing. Yoshidome et al.
proposed that tight packing of F1 subunits reduces
the excluded volume and leads to a large gain in
water entropy while decreasing the conformational
entropy of amino acid side chains.43,44 Because the
entropic contribution of water molecules is substan-
tially larger than that of the side chains,43,44 water
entropy is regarded as the key factor in subunit
packing. X-ray crystallography and MD simulation
revealed that the subunit packing is tightened upon
ATP binding and the TS reaction, while it is loos-
ened upon ATP hydrolysis and Pi release, which
results in gain or loss of water entropy, respec-
tively.5,7,43,44 This assessment is consistent with our
experimental results: entropy was gained upon ATP
binding and TS reaction, but entropy was lost upon
ATP hydrolysis and Pi release. Therefore, it is proba-
ble that the TDS‡ measured in this study arises
from changes in water entropy due to subunit pack-
ing. Another possible explanation is that a large-
scale conformational transition involving cata-
strophic events such as cracking or local unfolding,
termed a “proteinquake”,45,46 leads to an increase in
the entropy of the transition state and, accordingly,
lowers the activation barrier height.45,46 Indeed, a
large-scale conformational transition of the b-subu-
nit between the open and closed forms, termed
“hinge-motion”, which is the principal power-stroke
motion of the b-subunit,47–50 occurs upon ATP bind-
ing and the TS reaction at the ATP-binding
angle.5,51 On the other hand, the b subunit exhibits
a small-scale conformational transition, for example,
a loosening of the interface between a–b subunits,
Materials and Methods
Rotation assay
Wild-type F1 from thermophilic Bacillus PS3 (TF1)
was prepared as previously described.16 To visualize
the rotation of F1, the stator region (a3b3 subunits)
was fixed to a glass surface, and a gold colloidal
bead (/ 5 80 nm) was attached to the rotor (c-subu-
nit) as the rotation probe. The rotation assay was
carried out in a 50 mM MOPS-KOH (pH 7.0) buffer
containing 50 mM KCl, 5 mM MgCl2, and 200 mM
ATP. Rotating beads were observed under a custom-
built laser dark-field microscope31 with a 603 objec-
tive lens. The rotation of the bead was recorded at
27,000 frames s21 (FASTCAM 1024PCI-SE, Photo-
ron, Japan). The temperature in the room was con-
trolled with
a room air conditioner and was
monitored using a thermometer located on the sam-
ple stage of the microscope. The precision of the
temperature control was 61ꢀC.
ACKNOWLEDGMENTS
We thank Dr. S. Hayashi and Dr. K. Okazaki for crit-
ical discussion and A. Yukawa for technical support.
REFERENCES
1. Yoshida M, Muneyuki E, Hisabori T (2001) ATP syn-
thase—a marvellous rotary engine of the cell. Nat Rev
Mol Cell Biol 2:669–677.
Watanabe et al.
PROTEIN SCIENCE VOL 23:1773—1779 1777

2. Junge W, Sielaff H, Engelbrecht S (2009) Torque gener-
ation and elastic power transmission in the rotary
FoF1-ATPase. Nature 459:364–370.
20. Adachi K, Oiwa K, Nishizaka T, Furuike S, Noji H,
Itoh H, Yoshida M, Kinosita K, Jr. (2007) Coupling of
rotation and catalysis in F1-ATPase revealed by single-
molecule imaging and manipulation. Cell 130:309–321.
21. Watanabe R, Noji H (2014) Timing of inorganic phos-
phate release modulates the catalytic activity of ATP-
driven rotary motor protein. Nat Commun 5:3486.
22. Yoshida M, Sone N, Hirata H, Kagawa Y (1975) A
highly stable adenosine triphosphatase from a thermo-
phillie bacterium. Purification, properties, and recon-
stitution. J Biol Chem 250:7910–7916.
3. Weber J (2010) Structural biology: toward the ATP syn-
thase mechanism. Nat Chem Biol 6:794–795.
4. Dimroth P, von Ballmoos C, Meier T (2006) Catalytic
and mechanical cycles in F-ATP synthases. Fourth in
the Cycles Review Series. EMBO Rep 7:276–282.
5. Abrahams JP, Leslie AG, Lutter R, Walker JE (1994)
˚
Structure at 2.8 A resolution of F1-ATPase from bovine
heart mitochondria. Nature 370:621–628.
6. Cingolani G, Duncan TM (2011) Structure of the ATP
synthase catalytic complex (F1) from Escherichia coli in
an autoinhibited conformation. Nat Struct Mol Biol 18:
701–707.
23. Wang ZY, Freire E, McCarty RE (1993) Influence of
nucleotide binding site occupancy on the thermal sta-
bility of the F1 portion of the chloroplast ATP synthase.
J Biol Chem 268:20785–20790.
24. Li Z, Neufeld GJ (2001) Isolation and characterization
of mitochondrial F1-ATPase from crayfish (Orconectes
virilis) gills. Comp Biochem Physiol B Biochem Mol
Biol 128:325–338.
7. Kabaleeswaran V, Puri N, Walker JE, Leslie AG,
Mueller DM (2006) Novel features of the rotary cata-
lytic mechanism revealed in the structure of yeast F1
ATPase. EMBO J 25:5433–5442.
25. Watanabe R, Iino R, Shimabukuro K, Yoshida M, Noji
H (2008) Temperature-sensitive reaction intermediate
of F1-ATPase. EMBO Rep 9:84–90.
8. Noji H, Yasuda R, Yoshida M, Kinosita K, Jr. (1997)
Direct observation of the rotation of F1-ATPase. Nature
386:299–302.
26. Enoki S, Watanabe R, Iino R, Noji H (2009) Single-mol-
ecule study on the temperature-sensitive reaction of
9. Sielaff H, Rennekamp H, Wachter A, Xie H, Hilbers F,
Feldbauer K, Dunn SD, Engelbrecht S, Junge W (2008)
Domain compliance and elastic power transmission in
rotary FoF1-ATPase. Proc Natl Acad Sci USA 105:
17760–17765.
F1-ATPase with
a
hybrid F1 carrying
a
single
b(E190D). J Biol Chem 284:23169–23176.
27. Watanabe R, Noji H (2014) Characterization of the
temperature-sensitive reaction of F1-ATPase by using
single-molecule manipulation. Sci Rep 4:4962.
10. Spetzler D, Ishmukhametov R, Hornung T, Day LJ,
Martin J, Frasch WD (2009) Single molecule measure-
ments of F1-ATPase reveal an interdependence
between the power stroke and the dwell duration. Bio-
chemistry 48:7979–7985.
28. Furuike S, Adachi K, Sakaki N, Shimo-Kon R, Itoh H,
Muneyuki E, Yoshida M, Kinosita K, Jr. (2008) Tem-
perature dependence of the rotation and hydrolysis
activities of F1-ATPase. Biophys J 95:761–770.
11. Yasuda R, Noji H, Kinosita K, Jr., Yoshida M (1998) F1-
ATPase is a highly efficient molecular motor that rotates
with discrete 120 degree steps. Cell 93:1117–1124.
12. Bilyard T, Nakanishi-Matsui M, Steel BC, Pilizota T,
Nord AL, Hosokawa H, Futai M, Berry RM (2013)
High-resolution single-molecule characterization of the
enzymatic states in Escherichia coli F1-ATPase. Philos
Trans R Soc Lond B Biol Sci 368:20120023.
29. Sekiya M, Nakamoto RK, Al-Shawi MK, Nakanishi-
Matsui M, Futai M (2009) Temperature dependence of
single molecule rotation of the Escherichia coli ATP
synthase F1 sector reveals the importance of gamma-
beta subunit interactions in the catalytic dwell. J Biol
Chem 284:22401–22410.
30. Sekiya M, Hosokawa H, Nakanishi-Matsui M, Al-
Shawi MK, Nakamoto RK, Futai M (2010) Single mole-
cule behavior of inhibited and active states of Esche-
richia coli ATP synthase F1 rotation. J Biol Chem 285:
42058–42067.
13. Spetzler D, York J, Daniel D, Fromme R, Lowry D,
Frasch W (2006) Microsecond time scale rotation meas-
urements of single F1-ATPase molecules. Biochemistry
45:3117–3124.
31. Ueno H, Nishikawa S, Iino R, Tabata KV, Sakakihara
S, Yanagida T, Noji H (2010) Simple dark-field micros-
copy with nanometer spatial precision and microsecond
temporal resolution. Biophys J 98:2014–2023.
14. Hornung T, Ishmukhametov R, Spetzler D, Martin J,
Frasch WD (2008) Determination of torque generation
from the power stroke of Escherichia coli F1-ATPase.
Biochim Biophys Acta 1777:579–582.
32. Martin JL, Ishmukhametov R, Hornung T, Ahmad Z,
Frasch WD (2014) Anatomy of F1-ATPase powered
rotation. Proc Natl Acad Sci USA 111:3715–3720.
33. Watanabe R, Hayashi K, Ueno H, Noji H (2013) Catal-
ysis-enhancement via rotary fluctuation of F1-ATPase.
Biophys J 105:2385–2391.
15. Toyabe S, Okamoto T, Watanabe-Nakayama T,
Taketani H, Kudo S, Muneyuki E (2010) Nonequili-
brium energetics of a single F1-ATPase molecule. Phys
Rev Lett 104.
16. Watanabe R, Iino R, Noji H (2010) Phosphate release
in F1-ATPase catalytic cycle follows ADP release. Nat
Chem Biol 6:814–820.
34. Hayashi S, Ueno H, Shaikh AR, Umemura M, Kamiya
M, Ito Y, Ikeguchi M, Komoriya Y, Iino R, Noji H
(2012) Molecular mechanism of ATP hydrolysis in F1-
ATPase revealed by molecular simulations and single-
molecule observations. J Am Chem Soc 134:8447–8454.
35. Mukherjee S, Warshel A (2011) Electrostatic origin of
the mechanochemical rotary mechanism and the cata-
lytic dwell of F1-ATPase. Proc Natl Acad Sci USA 108:
20550–20555.
17. Yasuda R, Noji H, Yoshida M, Kinosita K, Jr., Itoh H
(2001) Resolution of distinct rotational substeps by sub-
millisecond kinetic analysis of F1-ATPase. Nature 410:
898–904.
18. Shimabukuro K, Yasuda R, Muneyuki E, Hara KY,
Kinosita K, Jr., Yoshida M (2003) Catalysis and rota-
tion of F1 motor: cleavage of ATP at the catalytic site
occurs in 1 ms before 40 degree substep rotation. Proc
Natl Acad Sci USA 100:14731–14736.
36. Harris DA, Dall-Larsen T, Klungsoyr L (1981) Studies
of the kinetics of the isolated mitochondrial ATPase
using dinitrophenol as a probe. Biochim Biophys Acta
635:412–418.
19. Nishizaka T, Oiwa K, Noji H, Kimura S, Muneyuki E,
Yoshida M, Kinosita K, Jr. (2004) Chemomechanical
coupling in F1-ATPase revealed by simultaneous obser-
vation of nucleotide kinetics and rotation. Nat Struct
Mol Biol 11:142–148.
37. Lowry DS, Frasch WD (2005) Interactions between
beta D372 and gamma subunit N-terminus residues
1778 PROTEINSCIENCE.ORG
Thermodynamic Analysis of F1-ATPase Rotation

gamma K9 and gamma S12 are important to catalytic
activity catalyzed by Escherichia coli F1Fo-ATP syn-
thase. Biochemistry 44:7275–7281.
structure-based molecular dynamics simulations. Proc
Natl Acad Sci USA 103:11844–11849.
46. Miyashita O, Onuchic JN, Wolynes PG (2003) Nonlin-
ear elasticity, proteinquakes, and the energy land-
scapes of functional transitions in proteins. Proc Natl
Acad Sci USA 100:12570–12575.
47. Tanigawara M, Tabata KV, Ito Y, Ito J, Watanabe R,
Ueno H, Ikeguchi M, Noji H (2012) Role of the DEL-
SEED loop in torque transmission of F1-ATPase. Bio-
phys J 103:970–978.
48. Usukura E, Suzuki T, Furuike S, Soga N, Saita E,
Hisabori T, Kinosita K, Yoshida M (2012) Torque gener-
ation and utilization in motor enzyme FoF1-ATP syn-
thase half-torque F1 with short-sized pushrod helix
and reduced ATP synthesis by half-torque FoF1. J Biol
Chem 287:1884–1891.
49. Watanabe R, Okuno D, Sakakihara S, Shimabukuro K,
Iino R, Yoshida M, Noji H (2012) Mechanical modula-
tion of catalytic power on F1-ATPase. Nat Chem Biol 8:
86–92.
50. Arai HC, Yukawa A, Iwatate RJ, Kamiya M, Watanabe
R, Urano Y, Noji H (2014) Torque generation mecha-
nism of F1-ATPase upon NTP binding. Biophys J 107:
156–164.
51. Masaike T, Koyama-Horibe F, Oiwa K, Yoshida M,
Nishizaka T (2008) Cooperative three-step motions in
catalytic subunits of F1-ATPase correlate with 80
degrees and 40 degrees substep rotations. Nat Struct
Mol Biol 15:1326–1333.
52. Okazaki K, Takada S (2011) Structural comparison of
F1-ATPase: interplay among enzyme structures, cataly-
sis, and rotations. Structure 19:588–598.
38. al-Shawi MK, Parsonage D, Senior AE (1990) Thermo-
dynamic analyses of the catalytic pathway of F1-
ATPase from Escherichia coli. Implications regarding
the nature of energy coupling by F1-ATPases. J Biol
Chem 265:4402–4410.
39. Albery WJ, Knowles JR (1976) Evolution of enzyme
function and the development of catalytic efficiency.
Biochemistry 15:5631–5640.
40. Ellington AD, Benner SA (1987) Free-energy differen-
ces between enzyme bound-states. J Theor Biol 127:
491–506.
41. Bhabha G, Lee J, Ekiert DC, Gam J, Wilson IA, Dyson
HJ, Benkovic SJ, Wright PE (2011) A dynamic knock-
out reveals that conformational fluctuations influence
the chemical step of enzyme catalysis. Science 332:
234–238.
42. Henzler-Wildman K, Kern D (2007) Dynamic personal-
ities of proteins. Nature 450:964–972.
43. Yoshidome T, Ito Y, Ikeguchi M, Kinoshita M (2011)
Rotation mechanism of F1-ATPase: crucial importance of
the water entropy effect. J Am Chem Soc 133:4030–4039.
44. Yoshidome T, Ito Y, Matubayasi N, Ikeguchi M,
Kinoshita M (2012) Structural characteristics of yeast
F1-ATPase before and after 16-degree rotation of the
gamma subunit: theoretical analysis focused on the
water-entropy effect. J Chem Phys 137:035102.
45. Okazaki K, Koga N, Takada S, Onuchic JN, Wolynes
PG (2006) Multiple-basin energy landscapes for large-
amplitude conformational motions of proteins:
Watanabe et al.
PROTEIN SCIENCE VOL 23:1773—1779 1779