Báo cáo khoa học: NMR investigations of subunit c of the ATP synthase from Propionigenium modestum in chloroform/methanol/water (4 : 4 : 1) pot

NMR investigations of subunit c of the ATP synthase from
Propionigenium modestum
in chloroform/methanol/water (4 : 4 : 1)
Ulrich Matthey
1
, Daniel Braun
2
and Peter Dimroth
1
1
Institut fu
¨
r Mikrobiologie, and
2
Institut fu
¨
r Molekularbiologie und Biophysik, Eidgeno
¨
ssische Technische Hochschule, Zu
¨
rich,
Switzerland
The subunit c from the A TP synthase of Propionigenium
modestum was studied by NMR in chloroform/methanol/
water (4 : 4 : 1). In this s olvent, subunit c consists of two
helical segments, comprised of residues L5 to I26 and G 29 to
N82, respectively. On comparing the secondary structure of
subunit c from P. modestum in the organic solvent mixture
with that in dodecylsulfate micelles several deviations
became apparent: in the organic solvent, the interruption of
the a helical structure within the conserved GXGXGXGX

motif was shortened from five to two residues, the p rominent
interruption of the a helical structure in the cystoplasmic
loop region was not apparent, and neither was there a break
in the a helix after the sodium ion-binding Glu65 residue.
The folding of subunit c of P. modestum in the organic
solvent a lso d eviated from that o f Escherichia coli in the same
environment, the most important difference being that sub-
unit c of P. modestum did not adopt a stable hairpin struc-
ture like subunit c of E. coli.
Keywords: ATP synthase; stable isotope labeling, NMR
spectroscopy; Propionigenium modestum; subunit c.
F
1
F
0
ATP synthases catalyse the formation of ATP from
ADP and inorganic phosphate that is driven by an
electrochemical gradient of protons or in some cases Na
+
ions. Similar e nzymes are found in chloroplast, mitochon-
dria and bacteria. They consist of a cytoplasmic F
1
part with
the subunit composition a
3
b
3
cde and a membrane intrinsic
F
0

moiety, which in bacteria has the subunit composition
ab
2
c
x
. The mechanism for ATP synthesis, proposed to
involve binding changes of the three catalytic binding sites
on the b sunbunits [1], was in remarkable agreement with
the atomic resolution X-ray structure of F
1
[2]. Based on
these data rotation of subunit c within the cylinder made of
alternating a and b subunits was suggested and confirmed
[3–5]. More recent structural data have shown that the c and
e subunits forming the central stalk are permanently fixed to
the ring of c subunits [6], and consequently, all three
subunits were demonstrated to rotate as a unit [7–9].
A high-resolution structure of the ion-translocating F
0
part remains to b e determined. Electron and atomic f orce
microscopy of F
0
indicated that subunits a and b a re
attached to the periphery of an oligomeric ring of c subunits
[10,11]. The subunits a and c a re directly involved in ion
translocation [12–15], whereas subunit b is presumed to
form a peripheral stalk, which connects the F
0
part to F
1

via
associationwiththed subunit [16–18]. Based on structu ral
data, the number of c subunits forming the ring is c
10
for the
yeast ATP synthase [6], c
14
for t he chloroplast enzyme [19],
and c
11
for the Ilyobacter tar taricus ATP s ynthase [20].
According to cross-linking studies, c
10
appears to be the
preferred stoichiometry for the ATP synthase of Escherichia
coli [21].
The NMR structure of the monomeric E. coli subunit c
was determined in chloroform/methanol/water (4 : 4 : 1),
in which the protein folds like a hairpin [22,23]. Two
extended a helices are connected by a hydrophilic loop, and
the proton-binding residue D61 is located in the centre of
the C-terminal helix. The monomeric P. modestum sub-
unit c was studied by NMR in SDS micelles [ 24]. In this
biphasic system the protein consists of four a helical
segments, t hat are connected by short linker peptides with
nonregular secondary structures. The Na
+
-binding residues
Q32, E65 and S66 [12] are lo cated in the I–II and III–IV
helix connections. No lo ng-range NOEs could be identified

that would indicate the presence of a three-dimensional fold
with close packing of the helices.
In order to enable a direct comparison of the P. mode-
stum subunit c with its E. coli homologue, we now prepared
an NMR sample in chloroform/methanol/water (4 : 4 : 1).
The key questions to be investigated were how the
secondary structure in the organic solvent mixture compares
with that in dod ecylsulfate mic elles, and w hether the
P. modestum subunit c forms similar interhelix contacts as
the E. coli protein in the same solvent.
EXPERIMENTAL PROCEDURES
Overproduction and purification of subunit c
Unifo rml y
13
C,
15
N-labelled subunit c was overproduced
in E. coli PEF42(DE3)pT7c on Martek 9-CN medium
as described previously [24]. Subunit c was purified by
chloroform/methanol extraction and anion-exch ange
Correspondence to U. Matthey, Institut fu
¨
r Mikrobiologie,
Eidgeno
¨
ssische Technische Hochschule Zu
¨
rich, ETH-Zentrum, CH
8092 Zu
¨

rich, Switzerland.
Fax: + 41 1632 13 78, Tel.: + 41 1632 55 23,
E-mail:
Enzymes:H
+
-transporting ATP synthase (EC 3.6.1.34);
Na
+
-transporting ATP synthase (EC 3.6.1.37).
Note: web page available at h ttp://www.micro.biol.ethz.ch
(Received 3 December 2001, r evised 18 February 2001, accepted 20
February 2001)
Eur. J. Biochem. 269, 1942–1946 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.02851.x
chromatography [25]. Aliquots of 4 mL were applied to a
Sephadex LH20 column (160 mm · 20 mm) and eluted
with chloroform/methanol (2 : 1). Fractions of 3 mL were
collected and the protein was monitored by A
280
.
NMR sample preparation
Subunit c was transferred into C
2
HCl
3
/C
2
H
3
OH/H
2

O
(4 : 4 : 1), applying the same conditions as for the E. co li
protein (M. Girvin, Biochemistry Department, Albert
Einstein College of Medicine, New York, USA, personal
communication). Ten milliliters of C
2
HCl
3
were added to
11 mL of subunit c [1 mgÆmL
)1
in chloroform/methanol
(2 : 1)] and the mixture was concentrated to less than
0.1 m L by a gentle stream of Argon with periodical swirling
of the s olution to k eep the s olvent composition uniform.
After addition of 0.5 mL C
2
HCl
3
/C
2
H
3
OH (2 : 1), the
sample was incubated for 15 min at room temperature, and
then 0.4 mL C
2
HCl
3
was added. The sample was concen-

trated to 0.05–0.1 mL and the same c ycle of solvent
addition and subsequent volume reduction was repeated.
The sample was then brought to complete dryness with a
stream of Argon. Whenever the sample became cloudy
during the preparation, C
2
HCl
3
wasaddedtoredissolvethe
material. The dried sample was covered with 0.6 mL of
C
2
HCl
3
/C
2
H
3
OH/H
2
O containing 25 m
M
D
11
-Tris/HCl at
pH 7.5, and incubated for 1 h at 30 °C with periodical,
gentle swirling. After adjusting the pH to 7.5, the sample
was centrifuged and transferred to an NMR tube, w hich
was flame-sealed. The final concentration of the protein was
2m

M
.
NMR spectroscopy
Spectra were recor ded at 27 °C o n B ruker D RX500,
DRX600, DRX750 and DRX800 spectrometers. For the
resonance assignments and the collection of conformational
constraints, the following experiments were recorded: 3D
CBCA(CO)NH [26], 3D HNCACB [27], 3D
15
N-resolved
[
1
H,
1
H]-TOCSY (mixing time s
m
¼ 60 ms) [28], 3D
13
C-resolved HCCH-TOCSY (s
m
¼ 14 ms) [29], 3D
15
N-resolved [
1
H,
1
H]-NOESY (s
m
¼ 60 ms) [ 30], 3D
13

C-
resolved [
1
H,
1
H]-NOESY (s
m
¼ 60 ms) [31], and 3D
13
C-
resolved [
1
H,
1
H]-NOESY (s
m
¼ 150 ms). Spectra were
processed and analysed with the programs
PROSA
[32] and
XEASY
[33]. Chemical shifts were calibrated with sodium
3-(trimethylsilyl)propane-1-sulfonate.
RESULTS
Stability of
P. modestum
subunit c in
chloroform/methanol/water (4 : 4 : 1)
Samples of 2 m
M

unlabelled subunit c of Propionigenium
modestum in C
2
HCl
3
/C
2
H
3
OH/H
2
O (4:4:1), 25m
M
D11-Tris/HCl were prepared to test the stability of the
protein. Two-dimensional homonuclear NMR spectra
were recorded to monitor structural changes over 4 weeks.
During this period, no spectral differences were observed
with samples at pH 5.8, pH 7.0 and pH 7.5 that were kept
at 20 °C. These observations indicated that P. modestum
subunit c was stable in the solvent mixture at the pH values
indicated.
Resonance assignment
Sequence-specific backbone assignments f or subunit c were
obtained from 3D HNCA, 3D CBCA(CO)NH and
HNCACB experiments using a 2-m
M
15
N/
13
C-labelled

sample in chloroform/methanol/water (4 : 4 : 1). With the
exception of the N-terminal dipeptide se gment H-Met-Asp-
all amide protons and nitrogen resonances could be
assigned. Proton resonances of aliphatic side chains were
achieved by 3D
15
N-resolved [
1
H,
1
H]-TOCSY and 3D
15
N-
resolved [
1
H,
1
H]-NOESY. Furthermore, homonuclear 2D
[
1
H,
1
H]-TOCSY and 2D [
1
H,
1
H]-NOESY spectra were
used to assign the aromatic spin systems. While all aromatic
1
H s ide chain resonances of Y34, Y70 and Y80 could be

determined, only one aromatic side chain resonance was
found for F84 due to signal overlap.
The
13
C chemical shifts were obtained by 3D
13
C-resolved
HCCH-COSY, 3D
13
C-resolved HCCH-TOCSY and 3D
13
C-resolved [
1
H,
1
H]-NOESY experiments. However,
chemical shift dispersion of the methyl groups was limited,
complete assignment of all CH
n
groups was received except
those of the N-terminal methionine.
Conformational constraints
Overall 3035 NOESY cross-peaks were assigned. We found
515 intraresidual, 283 sequential and 331 medium-range
NOEs. Observed d
ad
NOEs showed that the peptide bonds
G27–P28, Q46–P47 a nd N82 –P83 are all in trans-confor-
mation. No long-range NOEs could be identified in the
spectra recorded with a mixing time of 60 ms that would

indicate the occurrence of close interhelix contacts. There-
fore, additional 2D [
1
H,
1
H]-NOESY and 3D
13
C-resolved
[
1
H,
1
H]-NOESY spectra with 150 ms mixing time were
recorded on a Bruker DRX800 spectrometer and analysed.
Possible long-range NOEs were collected using the chemical
shift comparison function of CANDID. Evaluation of these
signals did not indicate the presence of any long-range
NOEs. In particular, cross-peaks of aromatic protons could
be completely assigned as short-range and medium-range
NOEs (Fig. 1 ).
The secondary structure characteristic connectivities were
collected from NOESY spectra recorded with 60 ms.
Surprisingly, the proposed hydrophilic loop between Q46
and D52 showed significant a helix con nectivities (Fig. 2).
In contrast to previous NMR studies of P. modestum
subunit c in SDS micelles [24], subunit c in chloroform/
methanol/water (4 : 4 : 1) exhibited no ahelix interruption
at t he C-terminal part (T67, G68) of the protein. Similar to
the SDS sample, subunit c in chloroform/methanol/water
(4 : 4 : 1) contains a nonhelical linker peptide near P28,

which is shorter t han in the SDS structure. Overall, typical
a helix connectivities were f ound from L5 to I26, from G29
to Q46 and from P47 to N82. A s indicated by t he
d
aN
(i,i + 2) connectivities the two helices possibly end with
3
10
helix turns comprising residues I23 to I26 and N82 to
L87, respectively.
Chemical shift deviations
The deviations from the random coil chemical shifts were
calculated as the difference b etween the measured chemical
shifts and the corresponding random coil values in aqueous
Ó FEBS 2002 NMR studies of P. modestum subunit c (Eur. J. Biochem. 269) 1943
solution [34]. Continuous
13
C
a
downfield chemical shift
deviations were found from V4 to G23, V30 to A44 and
I53 to Y80 (Fig. 2). Significantly smaller deviations were
observed for the peptide segments G25 to G29, R45 to
D52 and A81 to G89. The shape of the chemical shift
deviations did not change when calculated with corre-
sponding random coil values in chloroform/methanol
(1 : 1), whic h w ere referenced with tetramethylsilane
(H. K essler, Institute fu
¨
rOrganischeChemieandBio-

chemie, Tu M u
¨
nchen, Garching, Germany, personal
communication). However, the chemical shift deviations
were about 1.9 p.p.m. higher, which can most probably
be attributed to the usage of different refer ence stan-
dards.
Secondary structure and global fold
As indicated by helix-characteristic NOE connectivities [35],
subunit c in chloroform/methanol/water (4 : 4 : 1) consists
of two helices. Helix I comprises residues L5 to I 26, and
helix II comprises the segment G 29 to N82 with a short
interruption between Q46 and P47. In particular no signal
intensities were obtained for the a helix characteristic
connectivities d
aN
(23,26), d
ab
(23,27), d
aN
(24,27), d
ad
(24,28),
d
aN
(25,29), d
ab
(25,28), d
ad
(25,28) and d

aN
(26,30). The
interruption between helix I and helix II is further confirmed
by the small
13
C
a
downfield chemical shift deviation of the
segment A24 to G29.
The proposed hydrophilic loop Q46 to D52, is a helical
according to the medium-range NOEs. However, the
13
C
a
chemical shift deviation of this peptide segment is signifi-
cantly smaller than that of the other helical segments.
Therefore, it is likely that the a helix conformation of this
segment is poorly populated.
DISCUSSION
Structural studies of membrane proteins by NMR in
solution require the usage of either organic solvents or
detergents. T he s tructure o f the E. coli s ubunit c was
studied in organic s olvent by NMR [23]. Addition of water
to E. coli subunit c in chloroform:methanol (1 : 1) was
found to stabilize interhelix contacts in a concentration
depending manner (M. Girvin, Biochemistry Department,
Albert Einstein College of Medicine, New York, USA,
personal communication). In chloroform/methanol/water
(4 : 4 : 1) the protein consisted of two elongated helices,
which w ere connected by a h ydrophilic loop. Two different

structures at pH 5 and pH 8 were found and a conforma-
tional change of subunit c during ion translocation was
proposed [22].
The secondary structure of subunit c of P. modestum in
SDS micelles [24] deviates significantly from that of E. coli
in chloroform/methanol/water (4 : 4 : 1): P. modestum sub-
unit c folds into four helices, that are connected by small
linker peptides with nonregular secondary structure. The
Na
+
-binding ligands (Q32, E65, S66) [12] are located in the
peptides connecting helices I and II, and III and IV,
respectively.
As described a bove, the two NMR s tructures were
determined in different solvent system s and for t he c
subunits from two different b acteria. It w as therefore
intriguing to investigate whether the structure of the c
subunit is dependent on the solvent conditions. For this
purpose, we determined the structure of subunit c from
P. modestum in chloroform/methanol/water (4 : 4 : 1).
Under these conditions, the protein f olds significantly dif-
ferent from subunit c in dodecylsulfate micelles (Fig. 3B,B¢)
and it also folds significantly different than subunit c of
E. coli in the organic solvent mixtures ( Fig. 3A). The
secondary structure of P. modestum subunit c in the organic
solvent is mainly helical with only a short linker peptide
consisting of residues G27 and P 28, but without interrup-
tion of the helical folding in t he hydrophilic loop region as
indicated b y h elix-characteristic NOEs. Subunit c of
P. modestum therefore does not fold into a s table helical

hairpin in chloroform/methanol/water (4 : 4 : 1). The hair-
pin structure of subunit c is indicated, however, by a wealth
of experiments [12,36–40] and was unequivocally demon-
strated by the F
1
c crystal structure from yeast mitochondria
[6]. Moreover, the inconsistency between chemical shift
deviations and NOE connectivities at the peptide seg-
ment R45 to D52 implies structural polymorphism. The
P. modestum subunit c showed structural variety in differ-
ent solvent systems. It is likely that small changes i n the
organic solvent composition can cause structural polymor-
phism. Further, homologue proteins (P. modestum and
E. coli subunit c) can form different structures in the same
solvent, depending on their individ ual sequences.
Fig. 1. Part of a contour plot of a two-dimensional [
1
H,
1
H]-NOESY
spectrum showing the methyl-aromatic proton cross-peaks o f subunit c
in chloroform/methanol/water (4 : 4 : 1). The spectrum was reco rded
with 150 ms mixing times on a Bruker DRX800 spectrometer.
Assignments of the signals are indicated.
1944 U. Matthey et al. (Eur. J. Biochem. 269) Ó FEBS 2002
We hypothesize that monomeric subunit c may be
considerably prone to structural polymorphism. Regardless
of whether the protein is dissolved in organic solven ts or
embedded into detergent micelles, these environments are
not a good mimic of the natural situation where the protein

forms strong protein/protein contacts to assemble into rings
of 10, 11 or 14 c subunits. Th ese rings can be extremely s table,
and resist boiling in SDS for 5 min in the case of the
undecameric c ring of P. modestum [41]. This stability of the
ring makes structural flexibility rather unlikely, indicating
that the structural polymorphism of monomeric subunit c is
due to the interaction of the monomer with the artificial
environment lacking the stabilizing protein/protein contacts
within the ring. The undecameric c ring of the ATP synthase
from Ilyobacter tartaricus, a close relative to P. modestum,
has recently been crystallized in two dimensions and
subjected to structure determination by cryo-transmission
electron microscopy [20]. All c subunits of the ring show th e
hairpin-like folding and the structure of all monomeric units
appears to be the same. Hence, the structural flexibility
observed for the subunit c monomer is apparently lost upon
assembly into the ring, probably b ecause r ings of the stability
observed require defined structures of the monomeric units
in order to gen erate strong protein/protein interactions.
ACKNOWLEDGEMENTS
We thank Georg Kaim and Mark E. Girvin for critical reading of the
manuscript. We are grateful to Torsten H errmann for his support i n
CANDID, Reto H orst for introduction into the DRX800 spectrometer
and Mark Girvin for providing the detailed protocol of the NMR sample
preparation. We express a special thank to Kurt W u
¨
thrich for his
support and the provision of NMR equipment.
REFERENCES
1. Boyer, P.D. (1993) The binding change mechanism for ATP

synthase – some probabilities and possibilities. Biochim. Biophys.
Acta. 1140, 215–250.
2. Abrahams, J.P., Leslie, A.G.W., Lutter, R. & Walker, J.E. (1994)
Structure at 2 .8 A
˚
resolution of F
1
-ATPasefrombovineheart
mitochondria. Nature 370, 621–628.
3. Noij, H., Yasuda, R., Yoshida, M. & Kinosita, K. (1997)
Direct ob servatio n o f t he rotation of F
1
-ATPase. Na ture 38 6, 299–
302.
4.Duncan,T.M.,Bulygin,V.V.,Zhou,Y.,Hutcheon,M.L.&
Cross, R.L. (1995) Rotation of subunits during catalysis by
Escherichia coli F
1
-ATPase. Proc.NatlAcad.Sci.USA92, 10964–
10968.
5. Sabbert, D., Engelbrecht, S. & Junge, W. (1996) Intersubunit
rotation in active F-ATPase. Nature 381, 623–625.
Fig. 3. Alignment of c subunits from E. coli and P. modestum and location of the a helical secondary structures in different subunit c preparations. (a)
Amino-acid sequence of E. coli subunit c and location of a helices in the st ructure in ch loroform/meth anol/water (4 : 4 : 1) [23], amino-acid
sequence of P. modestum subunit c with the helix locations in (b) SDS micelles [24], and (b¢) in chloroform/methanol/water (4 : 4 : 1 ). Identical
amino ac ids are denoted by white l etters on dark gre y background and conservative substitutions are indicated by light grey background. The
conserved carboxylate residues involved in ion b inding (E. coli D61, P. modestum E65) are indicated with an asterisk.
Fig. 2. Amino-acid sequence of P. modestum subunit c a nd survey o f s elected N MR data. Sequential NOE connectivities are displayed b y c ontinuous
bold line s extending over the residues that showed these connectivities. Medium-range NOEs are represented by lines connec ting the two in teracting
residues. I n the row Dd(

13
C
a
) t he difference be tween the ob served shifts and the co rrespo nding random coil values [34] is sh own. The sequ ence
locations of the a helical secondary structu res derived from these data are indicated by bars at the bottom.
Ó FEBS 2002 NMR studies of P. modestum subunit c (Eur. J. Biochem. 269) 1945
6. Stock, D., Leslie, A.G.W. & Walker, J.E. (1999) Molecular
architecture of the rotary motor in ATP synthase. Science 28 6 ,
1700–1705.
7. Sambongi, Y., Iko, Y., Tanabe, M., Omote, H., I wamoto-Kihara,
A., Ueda, I., Yanagida, T., Wada, Y. & Futai, M. (1999)
Mechanical rotation of the c subunit oligomer in ATP synthase
(F
0
F
1
): Direct observation. Science 286, 1722–1724.
8. Pa
¨
nke, O., Gumbiowski, K., Junge, W. & Engelbrecht, S. (2000)
F-ATPase: sp ecific observation of the rotating c subunit oligomer
of EF
0
EF
1
. FEBS Lett. 472, 34–38.
9. Tsunoda, S.P., Aggeler, R., Noji, H., Kinosita, K., Jr, Yosh ida,
M. & C apaldi, R.A. ( 2000) Observations of the rotation within the
F
0

F
1
-ATP synthase: Deciding between rotation of the F
0
c subunit
ring and artifact. FEB S Lett. 470, 1–5.
10. Takeyasu, K ., Omote, H., N ettik adan, S., Tokumasu, F.,
Iwamotu-Kihara, A. & Futai, M. (1996) Molecular imaging of
Escherichia coli F
1
F
0
-ATPase in r econstituted membranes using
atomic force microscopy. FEBS Lett. 392, 110–113.
11. Birkenha
¨
ger, R., Hoppert, M., Deckers-Hebestreit, G., Mayer, F.
& Altendorf, K. (1995) The F
0
complex of the Escherichia coli
ATP synthase. Investigation by electron spectroscopic imaging
and immunoelectron microscop y. Eur. J. Biochem. 230, 58–67.
12. Kaim, G., Wehrle, F., Gerike, U. & Dimroth, P. (1 997) Molecular
basis for the coupling ion selectivity of F
1
F
0
ATP synthases:
probing the liganding groups for Na
+

and Li
+
in the c subu nit of
the ATP synthase from Propionigenium modestum. Biochemistry
36, 9185–9194.
13. Kaim, G. & Dimroth, P. (1998) A triple mutation in the a subunit
of the Escherichia coli/Propionigenium modestum F
1
F
0
ATPase
hybrid causes a switch from Na
+
stimula tion to N a
+
inhibition.
Biochemistry 37, 4626–4634.
14. Kaim, G. & Dimroth, P. (1995) A double mutation in subunit c of
the Na
+
-specific F
1
F
0
-ATPase of Propionigenium modestum
results in a switch from Na
+
to H
+
-coupled ATP synthesis in the

Escherichia coli host cells. J. Mol. Biol. 253, 726–738.
15. Fillingame, R.H. ( 1990) Molecular mechanism of ATP synthesis
by F
1
F
0
-type H
+
-transporting ATP synthases. I n The Bacteria,
pp. 345–391. Academic Press, San D iego, CA.
16. Lill, H., Hensel, F., Junge, W. & Engelbrecht, S. (1996) Cross-
linking of engineered subunit d to (ab)
3
in chloroplast F-ATPase.
J. Biol. Chem. 271, 32737–32742.
17. Ogilvie, I., A ggeler, R. & Capaldi, R.A. ( 1997) Cros s-linking of the
d subunit to on e of the three a subunits has no effect on func-
tioning, as expected if d is a part of the stator that links the F
1
and
F
0
parts of the Escherichia coli ATP synthase. J. Biol. Chem. 272 ,
16652–16656.
18. Bo
¨
ttcher,B.,Schwarz,L.&Gra
¨
ber, P. (1998) Direct indication for
the existence o f a double stalk in CF

0
F
1
. J. Mo l. Biol. 281,757–
762.
19. Seelert,H.,Poetsch,A.,Dencher,N.A.,Engel,A.,Stahlberg,H.&
Mu
¨
ller, D.J. (2000) Proton-powered turbine of a plant motor.
Nature 405, 418–419.
20. Stahlberg, H., Mu
¨
ller, D., Suda, K., Fotiadis, D., Engel, A.,
Meier, T., Matthey, U. & Dimroth, P. (2001) Bacterial sodium
ATP synthase has an undecameric rotor. EMBO Rep. 2,229–
233.
21. Jiang, W., Hermoli, J. & Fillingame, R.H. (2001) The preferred
stoichiometry of c subunits in the rotary sector of Escherichia
coli ATP synthase is 10. Proc. Natl Acad. Sci. USA 98, 4966–
4971.
22. Rastogi, V.K. & Girvin, M.E. (1999) Structural changes linked to
proton translocation by subunit c of the ATP synthase. Nature
402, 263–268.
23. Girvin, M .E., Rastogi, V.K., Albildgaard, F., Markley, J .L. &
Fillingame, R.H. (1998) Solution structure of the transmembrane
H
+
-transporting subun it c of the F
1
F

0
ATP syn thase. Biochem-
istry 37, 8817–8824.
24. Matthey, U., Kaim, G., Braun, D., Wu
¨
thrich,K.&Dimroth,P.
(1999) NMR studies of subunit c of the ATP synthase from
Propionigeniu m modest um dodecylsulfate micelles. Eur. J. Bio-
chem. 261, 459–467.
25. Matthey, U., Kaim, G. & Dimroth, P. (1997) Subunit c from the
sodium-ion-translo cating F
1
F
0
-ATPase of Propionigenium mod-
estum. Production, purification and properties of the protein in
dodecylsulfate solution. Eur. J. Biochem. 247, 820–825.
26. Grzesiek, S. & Bax, A. (1992) Improved 3D triple-resonance
NMR t echniq ues applied to a 31-kDa protein. J. Magn. Reson. 96 ,
432–440.
27. Wittekind, M. & Mueller, L. (1993) HNCACB, a high-sensitivity
3D NMR experiment to correlate amide-proton and nitrogen
resonances with th e alpha-carbon and beta carb on resonances in
proteins. J. Magn. Reson. 101, 201–205.
28. Marion,D.,Kay,L.E.,Sparks,S.W.,Torchia,D.A.&Bax,A.
(1989) Three-dimensional heteronuclear NMR of
15
N-labeled
proteins. J. Am. Chem. Soc. 111, 1515–1517.
29. Bax, A., Clore, G.M. & Gronenborn, A .M. (1990)

1
H-
1
Hcorre-
lation via isotropic mixing of
13
C magnetization, a new three-
dimensional approach f or assigning
1
Hand
13
C spect ra of
13
C
enriched proteins. J. Magn. Reson. 88, 425–431.
30. Fesik, S.W. & Zuiderweg, E.R.P. (1988) Heteronuclear three-
dimensional NMR spectroscopy. A strategy for the simplification
of homonucle ar two-dime nsional NMR spectra. J. Magn. Reson.
78, 588–593.
31. Ikura, M., Kay, L .E., Tschudin, R. & Bax, A. (1990) Three-
dimensional NOESY-HMQC spectroscopy of a
13
C-labeled pro-
tein. J. Magn. Reson. 86, 204–209.
32. Gu
¨
ntert, P., Do
¨
tsch, V., Wider, G. & Wu
¨

thrich, K. (1992) Pro-
cessing of multi-dimensional NMR data with the new software
PROSA. J. Biomol. N MR 2, 619–629.
33. Bartels, C., Xia, T ., Billeter, M., Gu
¨
ntert, P. & W u
¨
thrich, K . (1995)
The program XEASY for com puter-sup ported N MR spectral
analysis of biological macromolecules. J. Biol. NMR 6, 1–10.
34. Richarz, R. & Wu
¨
thrich, K. (1978) Carbon-13 NMR chemical
shifts of the common amino acid residues measured in aqueous
solutions of the linear tetrapeptides H-Gly-Gly-X-L-Ala-OH.
Biopolymers 17, 2133–2141.
35. Wu
¨
thrich, K. (1986) NMR of Proteins and Nucleic Acids. John
Wiley & Sons, New York.
36. Deckers-Hebestreit, G., Schmidt, R., Kiltz, H.H. & Altendorf, K.
(1987) F
0
portion of Escherichia coli ATP synthase: orientation of
subunit c in the membrane. Biochemistry 26, 5486–5492.
37. Zhang, Y. & Fillingame, R.H. (1994) Essential aspartate in sub-
unit c o f F
1
F
0

ATP synthase. Effect o f position 61 substitutions
in helix-2 on function of Asp24 in helix-1. J. Biol. Chem. 269,
5473–5479.
38. Watts, S.D., Zhang, Y., Fillingame, R.H. & Capaldi, R.A. (1995)
The c subunit in the Escherichia coli ATP synthase complex
(ECF
1
F
0
) extends through the stalk and contacts the c subunits of
the F
0
part. FEBS Lett. 368, 235–238.
39. Watts, S.D., Tang, C. & Capaldi, R.A. (1996) The stalk region of
the Escherichia c oli ATP synthase. Tyrosine 205 of the c subu nit is
in the interface between the F
1
and F
0
parts and can interact with
both the e and c oligomer. J. Biol. Chem. 271, 28341–28347.
40. Watts, S.D. & Capaldi, R.A. ( 1997) I nteractions between the F
1
and F
0
parts in the Escherichia coli ATP synthase. Associations
involving t he loop region of C subunits. J. Biol. Chem. 272, 15065–
15068.
41. Laubinger, W. & Dimroth, P. (1988) Characterization of the ATP
synthase of Propionigenium modestum as a primary sodium pump.

Biochemistry 27, 7531–7537.
1946 U. Matthey et al. (Eur. J. Biochem. 269) Ó FEBS 2002