Full text of DOI:10.1110/ps.9.11.2246

Protein Science ~2000!, 9:2246–2250. Cambridge University Press. Printed in the USA.
Copyright © 2000 The Protein Society
Enhanced internal dynamics of a membrane transport
protein during substrate translocation
KLAUS DÖRING,1,6 THOMAS SURREY,2,6 SYLVIA GRÜNEWALD, EDGAR JOHN,4
3
and FRITZ JÄHNIG5
Max-Planck-Institute for Biology, Department of Membrane Biochemistry, Corrensstrasse 38, 72076 Tübingen, Germany
~Received June 6, 2000; Final Revision August 9, 2000; Accepted September 1, 2000!
Abstract
Conformational changes are essential for the activity of many proteins. If, or how fast, internal fluctuations are related
to slow conformational changes that mediate protein function is not understood. In this study, we measure internal
fluctuations of the transport protein lactose permease in the presence and absence of substrate by tryptophan fluores-
cence spectroscopy. We demonstrate that nanosecond fluctuations of a-helices are enhanced when the enzyme transports
substrate. This correlates with previously published kinetic data from transport measurements showing that millisecond
conformational transitions of the substrate-loaded carrier are faster than those in the absence of substrate. These findings
corroborate the hypothesis of the hierarchical model of protein dynamics that predicts that slow conformational
transitions are based on fast, thermally activated internal motions.
Keywords: helix fluctuations; lactose permease; nanosecond motions; protein dynamics; time-resolved tryptophan
fluorescence anisotropy
Protein motions perform most intricate tasks as is formidably ex-
emplified by the action of transport proteins. Embedded in mem-
branes, they catalyze the specific passage of substrate molecules
from one side of the membrane to the other ~Henderson, 1993;
Kaback et al., 1994!. They do so by switching between two main
conformations exposing the substrate binding site alternatively to
either side of the membrane. One essential and remarkable char-
acteristic of these motions is that they occur in ways always en-
suring a tight seal against other molecules and solutes—no matter
if the transporter is empty or ligand-“filled.” Although a wealth of
kinetic data is available describing the rates of transport, we are far
from understanding on a molecular-mechanical level this dynamic
nature of transport proteins that lies at the heart of their biological
activity. This is not only due to a deficit in structural information
on the atomic level, but stems mainly from the difficulty to mea-
sure the internal protein motions that would then enable us to
correlate them with the kinetic characteristics of protein functioning.
LP from the inner membrane of Escherichia coli is a paradigm
for transport proteins ~Kaback et al., 1994!. It is a member of the
family of 12-transmembrane helix transporters and mediates cou-
pled transport of galactosides and protons across a membrane ~Hen-
derson, 1993; Kaback, 1997!. Typical translocation rates kcat of LP
~Garcia et al., 1983; Wright, 1986! and of other transporters ~Loo
Ϫ1
et al., 1998! are in the range of 10–100 s ; substrate binding can
be considerably faster ~Garcia et al., 1983; Wright, 1986; Wright
et al., 1994!. Spectroscopic and biochemical analysis of a series of
single amino acid mutants indicates that there is considerable con-
formational flexibility in LP ~Jung et al., 1994; Wu & Kaback,
1997; Weinglass & Kaback, 1999!. Especially, helix motions that
occur in the nanosecond time range were suggested to be linked to
the much slower large-scale conformational rearrangements even-
tually leading to transport ~Dornmair & Jähnig, 1989!. To examine
the validity of this hypothesis, we directly measured here these fast
motions for liposome-reconstituted purified LP under various trans-
port conditions. This could be achieved by utilizing LP’s six tryp-
tophan residues, which are all predicted to reside on membrane-
spanning a-helices ~Kaback et al., 1994!, as intrinsic fluorescent
probes for time-resolved fluorescence depolarization experiments.
This approach allowed us to measure internal orientational fluctu-
ations of not artificially labeled, fully functional LP both in the
absence and presence of substrate, to our best knowledge the first
measurement of this kind for a transport protein.
Reprint requests to: Klaus Döring, e-mail: Klaus.Doering@tecan.com or
Thomas Surrey, e-mail: surrey@embl-heidelberg.de.
1
Present address: Tecan Austria Ges.m.b.H, Untersbergstrasse 1A, 5082
Grödig, Austria.
2Present address: European Molecular Biology Laboratory, Meyerhof-
strasse 1, 69117 Heidelberg, Germany.
3
Present address: BASF-LYNX Bioscience AG, Im Neuenheimer Feld
5
15, 69120 Heidelberg, Germany.
Present address: Novartis Pharma AG, K-127.3.32, 4002 Basel,
4
Switzerland.
5
Fritz Jähnig died on June 18, 1996.
6
Both authors contributed equally to this work.
Abbreviations: LP, lactose permease; TDG, thiodigalactoside.
2
246

Internal dynamics of lactose permease
2247
Results and discussion
cited electronic state of the fluorophores!. Similar to other pro-
teins, LP’s tryptophan fluorescence intensity has—in addition to
shorter lifetimes ~Chen et al., 1991! ~t Ϫ t !—a small amplitude,
We studied orientational fluctuations of LP by time-resolved flu-
orescence depolarization experiments utilizing its six tryptophan
residues as intrinsic fluorescent probes. This yields the time courses
of the fluorescence intensity and anisotropy ~Holzwarth, 1995!
1
4
5
long lifetime component ~t ! of about 30 ns ~Döring et al., 1995,
1997! ~Table 1!. This long lifetime enabled us to observe the
anisotropy decay on a timescale of several tens of nanoseconds. In
this regime, motions of side chains and secondary structural ele-
ments are expected to occur ~Döring et al., 1997!.
~
see Materials and methods; Fig. 1A!. The molecular orientational
fluctuations are reflected by the anisotropy decay. The time win-
dow in which the anisotropy can be detected depends on the range
of the intensity decay ~also containing information about the ex-
Two types of internal fluctuations were indeed detected for LP
as reflected by two anisotropy relaxation processes ~Table 1; Fig. 1!.
External motions like rotational diffusion of the entire protein
within the viscous membrane or tumbling of the membrane vesicle
are too slow to be detectable in a nanosecond time window ~Dorn-
mair et al., 1985! making membrane proteins ideal candidates for
the study of internal nanosecond fluctuations. As established in
previous studies ~Beecham & Brand, 1985; Döring et al., 1995,
1997!, the faster of the two measured relaxation processes is caused
by wobbling motions of the tryptophan side chains relative to the
backbone and the slower relaxation process by protein domains,
which in the present case may be either fluctuations of individual
a-helices or groups of a-helices ~Dornmair & Jähnig, 1989!.
We investigated how these internal fluctuations of a transport
protein are affected by substrate transport. Adding transportable
galactosides to LP at saturating levels resulted in faster decays of
the anisotropy indicating enhanced fluctuations. This substrate ef-
Table 1. Parameters from the fits to the decays of the
tryptophan intensity and anisotropy of LP: intensity lifetimes t
intensity amplitudes a , anisotropy relaxation times f
anisotropy amplitudes b , and angles between excitation
and emission dipoles ue,i (see Materials and methods)a
i
,
i
i
,
i
No substrate
Lactose
TDG
Sucrose
t1
t2
t3
t4
t5
~ns!
0.1
1.1
3.6
6.0
33
0.4
1.4
3.3
5.8
19
0.1
1.7
4.1
6.4
0.6
2.6
4.8
6.4
33
33
a1
a2
a3
a4
a5
0.36
0.13
0.25
0.27
0.19
0.15
0.32
0.33
0.50
0.13
0.22
0.14
0.20
0.20
0.39
0.22
0.0005
0.0012
0.0005
0.0010
f1
f2
(ns)
3.0
14.2
2.6
9.6
2.5
12.9
3.1
14.7
Fig. 1. A: Time course of the measured intensity and anisotropy of the
tryptophan fluorescence of LP in the absence of substrate. The experimen-
tal data ~dots! are presented together with fits ~lines! of multi-exponential
decays convoluted with the apparatus response function ~the “shape” of the
exciting light pulse!. B: The relaxation times fi and amplitudes bi of the
anisotropy decays in the absence and presence of sugars as obtained from
the fits ~Table 1! are replotted as “motional anisotropy decays.” Rm~t! ϭ
b1
b2
b3
0.066
0.11
0.23
0.053
0.14
0.21
0.053
0.14
0.21
0.070
0.10
0.23
ue1
ue2
(deg)
38
47
38
49
37
46
38
48
͚
jϭ1–2 bj exp~t0fj! ϩ b3. In this form, the anisotropy decays reflect only
molecular motions and are free from influences of the apparatus response
function and certain spectroscopic properties of the tryptophan fluores-
cence on the measured anisotropy decay ~Döring et al., 1995, 1997; see
aThe anisotropy relaxation times fi and the anisotropy amplitudes bi ~in
bold! reflect internal orientational fluctuations of LP and are replotted as
motional anisotropy decays Rm in Figure 1B. The substrate concentrations
are 200 mM lactose and 5 mM TDG, as a control 200 mM sucrose was
used ~dissociation constants for lactose and TDG are 15 mM and 70 mM,
respectively; sucrose does not bind to LP!. The standard deviations are
0.1 ns for t1–t4, 2 ns for t5, 0.01 for a1–a4, 0.0002 for a5, 0.1 ns for f1,
0.5 ns for f2, 0.01 for b1 and b2, and 0.58 for ue1 and ue2.
_
_ _ _
;
also Materials and methods!. No substrate,
; 200 mM lactose,
_
mM TDG, {{{; 200 mM sucrose
_
_
5
. The dissociation constants for
lactose and TDG are 15 mM and 70 mM, respectively, sucrose does not
bind to LP. Inset: logarithmic motional anisotropy log@~Rm~t! Ϫ b3!0
~
b1 ϩ b2!#.

2248
K. Döring et al.
fect on the relaxation times f
i
and amplitudes b
i
of the anisotropy
We can now ask whether there is a correlation between nano-
second internal motions and biochemically measured millisecond
conformational changes. Kinetic analysis of transport measure-
ments reveals that substrate-loaded LP performs these millisecond
conformational transitions 5–30 times faster than empty LP in the
absence of a membrane potential ~Garcia et al., 1983; Wright,
1986; Kiefer, 1992! ~Fig. 3!; a similar behavior was found for
other 12-transmembrane helix transporters ~Wright et al., 1994!.
Additionally, LP translocates lactose two to six times faster than
the artificial substrate TDG ~Garcia et al., 1983; Wright, 1986;
Kiefer, 1992!. This can be compared with the nanosecond fluctu-
ations of LP’s a-helices measured in this paper. They are slowest
for “empty” LP, faster for TDG-loaded LP, and fastest during
lactose translocation. While the timescales of these two types of
protein motions are about six orders of magnitude apart, substrates
affect these motions in the same order on both timescales. This
remarkable correlation suggests that conformational changes lead-
ing to substrate translocation are causally related to the much faster
nanosecond fluctuations.
decays is illustrated in Figure 1B. To extract the characteristics of
these underlying molecular wobbling motions, we converted ~van
der Meer et al., 1984! these relaxation times f and amplitudes b
i
i
to rotational diffusion constants D4i and root-mean-square ~RMS!
amplitudes qRMS,i ~Fig. 2!. We found that substrate addition mainly
affects LP’s a-helix fluctuations ~relaxation process 2! while the
fluctuations of the tryptophan side chains themselves ~relaxation
process 1! are hardly changed ~Fig. 2!. The diffusion coefficient
D42 of the helix fluctuations is increased by 80% upon lactose
Ϫ1
addition ~from 3.8 to 6.9 ms ! and by 40% upon addition of the
Ϫ1
artificial substrate thiodigalactoside ~TDG! ~to 5.1 ms !; hence,
the fluctuations become faster. The effect on the RMS amplitudes
qRMS,2 of these orientational motions is less dramatic: they in-
crease by 10% ~from 20 to 228! indicating a slight increase of the
angle covered by the fluctuating helices. The extent of these sub-
strate effects is remarkable because the detected motions are an
average measured over all tryptophans present in LP. In summary,
a-helix dynamics are increased globally in the transport protein
when it translocates substrate across the membrane. Adding su-
crose, which is not bound by LP, does not change the internal
fluctuations significantly.
This result agrees well with the concept of protein energy land-
scapes that has been developed on the basis of low temperature
studies of myoglobin ~Ansari et al., 1985!. Briefly, it states that
folded proteins can adopt a large number of conformational sub-
states with slightly different potential energies ~Frauenfelder et al.,
Fig. 3. Minimal kinetic model for the transport cycle of LP ~adopted from
Patlak, 1957; LP’s two substrates, one sugar molecule and one proton, are
represented as a single filled black circle!. The transport cycle consists of
four steps: substrate binding, a major conformational transition ~with rate
kc! exposing the substrate to the other side of the membrane, substrate
dissociation, and finally a second major conformational change ~with rate
k0! reorienting the binding site back to the initial side of the membrane.
Substrate binding and dissociation are fast as compared to the two confor-
mational transitions ~Garcia et al., 1983; Wright, 1986!. The slowest step in
this cycle is transposition of the empty binding site. In the absence of a
membrane potential, k0 is about 5–10 times slower than kc for TDG and
about 15–30 times slower than kc for lactose ~Garcia et al., 1983; Wright,
1986!. A recent reinvestigation of these rates for LP reconstituted in lipid
vesicles ~under conditions very similar to our fluorescence anisotropy mea-
surements! agrees well with these ratios, the values obtained at pH 7.3 were
Ϫ1
Ϫ1
Ϫ1
k0 ϭ 1.6 s , kc,TDG ϭ 6.0 s , and kc,lactose ϭ 23.8 s ~Kiefer, 1992!. The
fluorescence measurements of the present study were either performed in
the total absence of substrate or in the presence of equal substrate concen-
trations on both sides of the membrane. Under these conditions, LP per-
formed conformational transitions exclusively either with the rate k0 or kc,
respectively.
Fig. 2. Diffusion coefficients D4,1 and D4,2 and RMS amplitudes qRMS,1
and qRMS,2 of the two measured relaxation processes of LP. These relax-
ation processes 1 and 2 are interpreted to arise from tryptophan side-chain
fluctuations and a-helix fluctuations, respectively.

Internal dynamics of lactose permease
997!. A hierarchy of internal fluctuations covering a broad range
2249
1
labeled lipid, and LP activity was measured in a binding assay.
Mock reconstitutions for fluorescence background measurements
were performed in the same way, however, in the absence of
protein.
of timescales corresponds to transitions between all these sub-
states. To which extent which fast internal fluctuations are coupled
to major conformational changes involved in protein function on
slower timescales has since then been an open question ~Frauen-
felder & Leeson, 1998!. We have presented here one of the rare
studies addressing this question by directly measuring fast internal
protein dynamics under close-to-native conditions of protein func-
tioning: galactoside transport at ambient temperatures.
We conclude that substrate binding appears to alter the energy
landscape of LP in a way that helix fluctuations can sample the
accessible conformational space more efficiently. On the level of
helix fluctuations, the protein becomes more dynamic. The fluctu-
ations of the tryptophan side chains themselves are hardly altered
Time resolved fluorescence anisotropy (for a review see,
for example, Holzwarth, 1995)
The fluorescence measurements were performed at room temper-
ature as described ~Döring et al., 1995, 1997!. Tryptophan fluo-
rophores of liposome-reconstituted LP are repeatedly excited with
polarized picosecond light pulses. As light source, a mode-locked
Nd-YAG-laser and a cavity-dumped dye laser were used with a
pulse repetition rate of 4.1 MHz. The excitation wavelength was
300 nm. Each pulse selectively excites a population of fluoro-
phores with their absorption dipole moments mainly parallel to the
polarization of the exciting light. The emitted light is detected at
350 nm under two orthogonal polarizations—one parallel and the
other perpendicular to the polarization of the excitation light. The
time courses of these two fluorescence intensity decays are mon-
itored by time-correlated single photon counting ~Holzwarth, 1995!.
The total intensity is s~t! ϭ i ~t! ϩ 2Gi ~t! and the anisotropy is
~
see also Ansari et al., 1985!, most probably because the trypto-
phans are not close to the binding site ~Weitzman et al., 1995!. On
the other hand, a galactoside binding protein behaves completely
different from a galactoside transport protein. For maltose-binding
protein, we observed the opposite dynamic behavior in response to
substrate binding ~Döring et al., 1999!. When it closes its binding
cleft around the ligand, its domain motions are strongly restricted.
Hence, on the level of secondary structure elements or protein
domains the protein becomes less dynamic upon ligand binding.
Since the energy landscape determines the internal protein dynam-
ics, it is this energy landscape that appears to be specifically de-
signed for distinct protein functions.
The observed correlation between helix fluctuations and slow
conformational changes might therefore be a consequence of the
typical character of the energy landscape of a transport protein.
Although conformational changes are often described by models
implying an energy barrier along a low-dimensional reaction co-
ordinate, similar to a chemical reaction ~Kramers, 1940; Patlak,
5
4
defined as r~t! ϭ @i ~t! Ϫ Gi ~t!#0s~t!, with G representing a
5
4
scaling factor determined experimentally by calibration.
Data analysis
The experimental data were analyzed as described ~Döring et al.,
1997!. Briefly, a multi-exponential decay S~t! ϭ ͚
ai
iϭ1Ϫ5
exp~Ϫt0t ! was assumed for the total intensity. For the anisotropy,
i
a
modified multi-exponential decay R~t! ϭ f ~t!$͚jϭ1–2 bj
exp~t0f ! ϩ b % with f ~t! ϭ $͚ P ~cos u !a exp~Ϫt0t ! ϩ
j
3
iϭ1–4
2
e1
i
i
1957!, in the high dimensionality of phase space, a protein might
2 e2 5 5
P ~cos u !a exp~Ϫt0t !%0S~t! was used to account for the more
be able to circumvent high energy barriers on a long journey
through a labyrinth of pathways. Extensive sampling of the con-
formational space by thermally activated fluctuations will eventu-
ally lead to the passage through these paths corresponding to the
large conformational change. The overall rate of the slow confor-
mational transition would then depend strongly on how efficiently
the diffusive motions explore the labyrinth. Such an interpretation
of protein functioning is supported by our finding that fast internal
helix fluctuations and slow conformational transitions in lactose
permease are coupled. To which extent energy barriers have still to
be overcome and therefore contribute to the rate of conformational
changes is an intriguing question for future studies.
complex spectroscopy caused by a long lifetime component of
tryptophan ~Döring et al., 1995!. The angles ue,i between the ab-
sorption and emission dipole moments had to be explicitly taken
into account, since the longest lifetime component t originates
5
from a different dipole moment. From these equations, expressions
for the parallel and perpendicular components of the fluorescence
intensity were derived and convoluted with the apparatus response
function ~determined by use of a strictly mono-exponential fluo-
rescence standard! and fitted to the measured intensities i ~t! and
5
i4~t!. The obtained parameters are presented in Table 1. The in-
tensity data i ~t! and i ~t! were background-corrected by subtract-
5
4
ing mock-sample data from the actual protein sample data ~Döring
et al., 1997!. To achieve a sufficient signal-to-noise ratio even at
5
0 ns after the excitation pulse, extensive data sampling over 10 h
Materials and methods
was required, demanding an extremely high light-source and overall-
setup performance.
Purification and reconstitution
LP was purified from strain T206 and reconstituted in lipid vesi-
cles essentially as described ~Dornmair, 1988!. However, an ad-
ditional purification step using S Sepharose Fast Flow ~Pharmacia,
Uppsala, Sweden! in 20 mM MES, 0.1 mM Na-EDTA, 1 mM
DTT, 0.01% dodecyl maltoside, pH 6.2 was performed, followed
by adjusting the dodecyl maltoside concentration to 0.1% and
reconstituting the protein in vesicles of a mixture of 1-palmitoyl-
Molecular motions
Depolarization, as seen in the anisotropy decay, is caused by ro-
tational motions of the excited fluorophores. The orientational fluc-
tuations in a protein may be described as ~angularly restricted!
diffusional motions. Their velocity and amplitude can be expressed
by rotational diffusion coefficients D and RMS amplitudes qRMS.
4
2-oleoylphosphatidyl-ethanolamine and 1-palmitoyl-2-oleoylphos-
Their values can be calculated from the anisotropy decay: From
phatidyl-glycerol at a molar ratio of 401 as described ~Dornmair &
Jähnig, 1989!. The lipid0protein molar ratio was 1,000. The lipid
concentration was determined using trace amounts of radioactively
the amplitudes b of the anisotropy decay, the orientational order
j
2
parameters ^P & can be derived according to ͗P & ϭ ~b ϩ b !0
2
i
2
3
2 1
2
2 2
~b1 ϩ b2 ϩ b ! and ͗P & ϭ b 0~b ϩ b !. Since ^P & ϭ
3
3
2
3
2 i

2250
K. Döring et al.
2
^
~3 cos q
axis relative to the mean axis resulting from relaxation process i,
in the limit of high order the RMS amplitude of q is obtained as
i
Ϫ 1!02& with q
i
denoting the angle of the fluorophore
Svart P, eds. 1997. Landscape paradigms in physics and biology—Concepts
structures and dynamics. Amsterdam: North-Holland Publishing Co.
Frauenfelder H, Leeson DT. 1998. The energy landscape in non-biological and
biological molecules. Nat Struct Biol 5:757–759.
Garcia ML, Viitanen P, Foster DL, Kaback HR. 1983. Mechanism of lactose
translocation in proteoliposomes reconstituted with lac carrier protein pu-
rified from Escherichia coli. 1. Effect of pH and imposed membrane po-
tential on efflux, exchange, and counterflow. Biochemistry 22:2524–2531.
Henderson PJF. 1993. The 12-transmembrane helix transporters. Curr Opin Cell
Biol 5:708–721.
Holzwarth AR. 1995. Time-resolved fluorescence spectroscopy. Methods Enzy-
mol 246:334–378.
Jung H, Jung K, Kaback HR. 1994. A conformational change in the lactose
permease of Escherichia coli is induced by ligand binding or membrane
potential. Protein Sci 3:1052–1057.
i
102
q
~
RMS,i ϭ @~203!~1 Ϫ ^P
van der Meer et al., 1984! by D4,i ϭ ~1 Ϫ ^P
2
&
i
!# . The diffusion coefficient is given
!06f . Flexible
2
&
i
i
molecules display a rapid anisotropy decay, i.e., short anisotropy
relaxation times and0or high amplitudes for the internal relaxation
processes, indicative of large rotational diffusion coefficients and
large RMS amplitudes. Because the external rotational mobility of
membrane proteins is hindered and too slow to be detected in a
fluorescence depolarization experiment, they show a finite residual
anisotropy ~i.e., an additional amplitude in the anisotropy decay!.
Kaback HR. 1997. A molecular mechanism for energy coupling in a membrane
transport protein, the lactose permease of Escherichia coli. Proc Natl Acad
Sci USA 94:5539–5543.
Acknowledgments
Kaback HR, Frillingos S, Jung H, Jung K, Prive GG, Ujwal ML, Weitzman C,
Wu J, Zen K. 1994. The lactose permease meets Frankenstein. J Exp Biol
196:183–195.
We thank Hans Kiefer for helpful discussions, Peter Overath for continuous
support, and our colleagues for critically reading the manuscript.
Kiefer H. 1992. Kinetic characterization of a cotransporter: Lactose permease
@
Thesis#. Eberhard-Karls-University Tübingen.
Kramers HA. 1940. Brownian motion in a field of force and the diffusion model
of chemical reactions. Physica 7:284–304.
Loo DD, Hirayama BA, Gallardo EM, Lam JT, Turk E, Wright EM. 1998.
Conformational changes couple Naϩ and glucose transport. Proc Nat Acad
Sci USA 95:7789–7794.
Patlak CS. 1957. Contributions to the theory of active transport: II. The gate,
type non-carrier mechanism and generalizations concerning tracer flow,
efficiency, and measurement of energy expenditure. Bull Math Biophys
19:209–235.
References
Ansari A, Berendzen J, Bowne SF, Frauenfelder H, Iben IET, Sauke TB, Shy-
amsunder E, Young RD. 1985. Protein states and protein quakes. Proc Natl
Acad Sci USA 82:5000–5004.
Beecham JM, Brand L. 1985. Time-resolved fluorescence of proteins. Ann Rev
Biochem 54:43–71.
Chen RF, Knutson JR, Ziffer H, Porter D. 1991. Fluorescence of tryptophan
dipeptides: Correlation with the rotamer model. Biochemistry 30:5184–5195.
Döring K, Beck W, Konermann L, Jähnig F. 1997. The use of a long-lifetime
component of tryptophan to detect slow orientational fluctuations of pro-
teins. Biophys J 72:326–334.
Döring K, Konermann L, Surrey T, Jähnig F. 1995. The use of a long-lifetime
component of tryptophan to detect slow orientational fluctuations of pro-
teins. Eur Biophys J 23:423–432.
van der Meer W, Pottel H, Herremann W, Ameloot M, Hendricks H, Schröder
H. 1984. Effect of orientational order on the decay of the fluorescence
anisotropy in membrane suspensions. Biophys J 46:515–523.
Weinglass AB, Kaback HR. 1999. Conformational flexibility at the substrate
binding site in the lactose permease of Escherichia coli. Proc Natl Acad Sci
USA 96:11178–11182.
Weitzman C, Consler TG, Kaback HR. 1995. Fluorescence of native single-Trp
mutants in the lactose permease from Escherichia coli: Structural properties
and evidence for a substrate-induced conformational change. Protein Sci
4:2310–2318.
Döring K, Surrey T, Nollert P, Jähnig F. 1999. Effects of ligand binding on the
internal dynamics of maltose-binding protein. Eur J Biochem 266:477–483.
Dornmair K. 1988. An improved purification of lactose permease. FEBS Lett
Wright EM, Loo DD, Panayotova-Heiermann M, Lostao MP, Hirayama BH,
Mackenzie B, Boorer K, Zampighi G. 1994. “Active” sugar transport in
eukaryotes. J Exp Biol 196:197–212.
2
35:35–39.
Dornmair K, Corin AF, Wright JK, Jähnig F. 1985. The size of the lactose
ϩ
permease derived from rotational diffusion measurements. EMBO J 4:3633–
Wright JK. 1986. The kinetic mechanism of galactoside0H cotransport in
3
638.
Escherichia coli. Biochim Biophys Acta 855:391–416.
Dornmair K, Jähnig F. 1989. Internal dynamics of lactose permease. Proc Natl
Acad Sci USA 86:9827–9831.
Frauenfelder H, Bishop A, Garcia AE, Perelson A, Sherington D, Schuster P,
Wu J, Kaback HR. 1997. Helix proximity and ligand-induced conformational
changes in the lactose permease of Escherichia coli determined by site-
directed chemical crosslinking. J Mol Biol 270:285–293.