| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institute for Biophysical Chemistry and Centre for Biomolecular Magnetic Resonance, J. W. Goethe University, Frankfurt, Germany
3Correspondence: Intitut für Biophysikalische Chemie, J. W. Goethe Universität, Max von Laue Str. 9, 60438 Frankfurt, Germany. E-mail: glaubitz{at}em.uni-frankfurt.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
pH, which results in enhanced ethidium fluorescence intensity, which is abolished when pH or 
is collapsed or when the essential residue Glu-13 in TBsmr is exchanged with Ala. This observation shows the formation of a pH-dependent, transient substrate-protein complex between binding and release of ethidium. We have further characterized this state by determining the Kd, by inhibiting ethidium transport through titration with nonfluorescent substrate and by fluorescence anisotropy measurements. Our findings support a model with a single occluded intermediate state in which the substrate is highly immobile.—Basting, D., Lorch, M., Lehner, I., Glaubitz, C. Transport cycle intermediate in small multidrug resistance protein is revealed by substrate fluorescence.
Key Words: bacteriorhodopsin • EmrE • ethidium bromide • TBsmr
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
. Because of their small size, members of the SMR protein family offer a convenient paradigm for studying the mechanism of substrate-proton antiport in secondary multidrug transporters (2)
.
EmrE is the most studied SMR protein attracting significant interest and recently stirred controversial discussions with respect to its three-dimensional structure and oligomeric arrangement (3
, 4)
. EmrE transports a diverse array of aromatic, positively charged substrates in a proton/drug antiport fashion (5)
. Other SMR proteins have overlapping, but significantly different, substrate specificities with measured affinities in the nanomolar to millimolar range (6
, 7)
. All SMR proteins are of similar size (
11–12 kDa), four-transmembrane helix topology, and have a highly conserved key residue Glu-14 (6)
. For EmrE, the four-helix topology predicted by hydropathy analysis has been confirmed with Fourier transform infrared (FTIR) and solution state NMR analysis (8
, 9)
. On the basis of cryoelectron microscopy data of two-dimensional crystals (10)
, a structural model of a dimer has been deposited in the protein database, and an alternate access model for the transport mechanism has been proposed (11
, 12)
. In this model, substrate translocation is initiated by substrate binding, which is followed by release of the substrate, triggered by proton binding and proton antiport. Although the transfer of substrate from a binding to a release site must involve conformational changes within EmrE, no statement could be made about potential intermediate states. To address this open mechanistic question, we have selected the typical SMR protein TBsmr from Mycobacterium tuberculosis which expressed well in our hands. TBsmr (gene accession number RV 3065, previously also named mmr, ref. 13
) has 70% similarity and 43% identity to EmrE. Both proteins impart resistance to TPP+, acriflavine, ethidium bromide (EtBr), benzalkonium, and methyl viologen, whereas TBsmr also confers resistance to safranin O, pyronin Y, and erythromycin (13)
. It has been shown to cause uptake of methyl viologen into proteoliposomes (14)
. Because of its intrinsic fluorescence, we have selected EtBr for the following study to probe substrate transport across the membrane.
One requirement for in vitro transport studies is the generation of a stable and reproducible pH gradient. Here, TBsmr has been co-reconstituted with the light-driven proton pump bacteriorhodopsin (bR) into unilamellar vesicles. On illumination bR generates a stable pH gradient, which TBsmr requires for substrate transport. Co-reconstitution of pH-dependent proteins with bR has been extensively used in functional assays for ATPase activity (15)
, sodium channels from eel (16)
, and glutamate transporter from rat brain vesicles (17)
. The transport process must involve steps for substrate binding, translocation, and release. During transport, a substrate such as EtBr could experience a significant change of environment, which would influence its fluorescence properties. Therefore, EtBr itself could report on transient substrate-transporter complexes, which could form during the transport cycle. We have measured the fluorescence quantum yield of EtBr, as a proton gradient is formed by bR. As the same light source (545 nm) can be used to activate bR and excite EtBr fluorescence, the experiment can be carried out using a standard fluorescence spectrometer. The basic layout of the experiment is depicted in Fig. 1
.
|
In general, the total fluorescence intensity Ftotal contains contributions from free EtBr in solution within and outside the liposomes (Finfree, Foutfree), from EtBr nonspecifically bound to the liposomes (Fliposomebound), and EtBr bound to TBsmr (Fbound):
 | (1) |
pH = 0 and the same EtBr concentration inside and outside. On illumination, a pH gradient is generated, and TBsmr starts to transport EtBr into the liposomes. In the following, we will show that this transport process causes an increase in total fluorescence from its equilibrium value F0 by F:
 | (2) |
F contains mainly fluorescence contributions of EtBr bound to a TBsmr transport cycle intermediate (F=Ftrans). Our experiments demonstrate for the first time that SMR proteins form an intermediate substrate-protein complex during the transport cycle. This substrate-protein complex is further characterized by a Kd and by fluorescence anisotropy. Furthermore, we show that co-reconstitution of secondary transporters with bacteriorhodopsin provides a convenient platform for transport and binding experiments.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
derivative with a myc- and his-tag (14)
. The mutant TBsmr E13A was cloned using the Quikchange® Site-Directed Mutagenesis Kit (Stratagene, Amsterdam, Netherlands). A forward primer 5'-GCGATCTTCGCGGCAGTGGTGGCAACC-3' and a reverse primer 5'-GGTTGCCACCACTGCCGCGAAGATCGC-3') was used in a two-stage PCR reaction (19)
.
Expression and purification
Overexpression and purification of TBsmr were adapted from protocols described (14
, 20)
. E. coli T7 Express were grown at 37°C in LB media. Seven hundred fifty milliliters of LB containing 100 µg/ml ampicillin in a 2-L flask was inoculated with 15 ml of overnight culture. Expression of TBsmr was induced with 0.5 mM isopropyl β-D-thiogalactoside at a cell density of A600 = 1. Two hours later, the cells were harvested by centrifugation (3800 g for 10 min at 4°C).
Cells were resuspended in 5 ml lysis buffer/g wet cells (250 mM sucrose, 150 mM NaCl, 10 mM Tris-Cl pH 7.5, 2.5 mM MgSO4, DNase I) and broken twice in a cell disrupter at 1.5 kbar. Cell debris was removed by centrifugation at 4550 g for 10 min. After centrifuging the supernatant at 223,000 g for 30 min, we collected the membrane fraction, froze it in liquid nitrogen, and stored it in fractions at –80°C until use. A fraction of the membrane pellet was solubilized for 40 min at 4°C in 15 mM Tris-Cl pH 7.5, 300 mM NaCl, 20 mM imidazole, 1% (w/v) DDM and complete protease inhibitor (Roche Diagnostics, Mannheim, Germany). After centrifugation at 184,000 g for 50 min, the supernatant was mixed with Ni-NTA resin (Qiagen, Hilden, Germany) for 80 min at 4°C. The resin was washed with 15 mM Tris-Cl pH 7.5, 300 mM NaCl, 20 mM imidazole and 0.08% (w/v) DDM until the A280 was below 0.05. The protein was eluted with the same buffer containing 300 mM imidazole and either reconstituted directly or stored at –80°C in 0.2-mg fractions (according to A280).
Bacteriorhodopsin
Bacteriorhodopsin in purple membrane patches was prepared from Halobacterium salinarium strain JW5, as described previously (21)
. Purple membrane was used directly after density gradient purification with three additional washing steps to remove the sucrose (22)
.
Reconstitution of TBsmr
10 mg of E. coli total lipids (Avanti Polar Lipids, Alabaster, AL, USA) were solubilized in 1 ml of 150 mM KCl, 5 mM K-EDTA pH 7.3 and 20 mg/ml n-dodecyl-β-D-maltoside (DDM; Glycon, Luckenwalde, Germany). To 0.752 ml of the resulting suspension 0.1 mg of TBsmr, in elution buffer, was added. Elution buffer was used to produce a final sample volume of 2.4 ml. Excess detergent was removed with 250 mg of degassed SM-2 biobeads (Bio-Rad, Munich, Germany) in an overnight incubation at room temperature. The biobeads were then removed.
Fusion of bR with TBsmr-containing liposomes
Bacteriorhodopsin (0.41 mg) was added to 1 ml of the TBsmr-proteoliposomes. The mixture was freeze-thawed 3 times, which results in an 90% inside-out orientation of bR in the vesicles (16)
. If necessary, the sample was stored at –80°C after the last freezing step. Proteoliposomes were centrifuged at 15,000 g, 30 min, 15°C, and the resulting pellet was resuspended in 1 ml of 150 mM KCl, 5 mM K-EDTA at pH 7.3. The sample was sonicated in a water bath for 3 min to form unilamellar vesicles. The sample typically contained final concentrations of 11 µM TBsmr, 15.4 µM bR, and 4.19 mM lipids.
Fluorescence measurements
All fluorescence measurements were carried out with a Jasco FP 6500 fluorescence spectrometer (Jasco, Groβ-Umstadt, Germany). A 1- x 1-cm cuvette was kept at 25°C by a thermostatic circulating water bath. EtBr measurements were carried out with excitation and emission wavelengths of 545 and 610 nm, respectively. Emission slits were kept constant at 10 nm; excitation slits between 2.5 and 20 nm were used.
In the cuvette, 200 µl of the bR/TBsmr proteoliposomes was diluted into 2 ml of 150 mM KCl, 5 mM K-EDTA at pH 7.3, and EtBr. The sample was incubated in the dark for 5 min. Illumination at 545 nm excites EtBr and enables bacteriorhodopsin to pump protons. The change in EtBr fluorescence was measured over 200 s. Illumination was switched off and the system left to equilibrate for 5 min. EtBr control experiments without protein
(Fig. 3)
were measured in 150 mM KCl and 5 mM K-EDTA at pH 7.3.
|
|
Steady-state anisotropy was measured using Melles Griot film polarizer (Bensheim, Germany) on the excitation and emission beam. Correct alignment of the polarizer was confirmed with diluted glycogen, which gave an anisotropy of r > 0.97. The fluorescence intensities were measured parallel (F||) and perpendicular (F
) to the direction of the vertically polarized excitation light, where Fwas corrected for the instrumental bias by a G-factor of 2.89 at 610 nm. The total fluorescence anisotropy r is defined by
 | (3A) |
f x r + f0 x r0, where f0 and f are the fractional fluorescence of EtBr before and during the generation of pH (f = F/F, f0 = F0/F), and r0 and r are the corresponding anisotropies. To determine r, the net emission intensities of the intermediate F were used (Eq. 3)
. F was obtained by subtracting the parallel and perpendicular equilibrium fluorescence intensities (F||0 and F0) from the corresponding fluorescence F, which gives
 | (3B) |
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
. When the formation of pH is initiated by sample illumination at 545 nm, a time-dependent increase of the EtBr fluorescence at 610 nm is detected. The fluorescence time course Ftotal(t) starts at equilibrium F0 where [EtBr]in = [EtBr]out and saturates after 100 s at F0 + Fmax. Representative time traces are shown in Fig. 2
a and are described by
 | (4) |
pH generation by bR, EtBr transport into the liposomes, and EtBr and proton leakage out of the liposomes. A detailed kinetic analysis is beyond the scope of this article. In the following, we only consider the steady state reached after 100 s and use Fmax as parameter for the change in EtBr fluorescence intensity.
The TBsmr/bR/lipid molar ratio was screened with respect to largest pH and Fmax. Best results were obtained for 1:1.4:380 (Supplemental Fig. 1). A pH of 0.2 agrees well with published data (16)
and can be kept constant for some hours (Supplemental Figs. 2 and 3). It was found that doubling the amount of TBsmr from 0.5 to 1 µM also causes a doubling of Fmax, as shown by time traces [2] and [1] in Fig. 2a
, respectively. Abolishing pH by adding the ionophore nigericin as well as eliminating the proton motive force by adding carbonyl cyanide 3-chlorophenylhydrazone (CCCP) caused a total collapse of F (Fig. 2a
). This observation supports the notion that the detected fluorescence increase is dependent on a pH gradient. To exclude unspecific binding to bR or the liposomes, the experiment was repeated with TBsmr E13A, in which the essential residue E13 (in EmrE E14) had been replaced with alanine. No change in EtBr fluorescence could be detected (Fig. 2a
) with this inactive mutant. From these observations, we can conclude that the observed fluorescence change F is caused by TBsmr and depends on the pH gradient as well as on the TBsmr concentration.
Detection of ethidium bromide transport into the liposomes
Our data suggest pH-dependent binding of EtBr to TBsmr. If TBsmr pumps EtBr into the liposomes, the substrate concentration inside should increase while decreasing outside. But the fluorescence of free EtBr inside and outside is indistinguishable and its contribution to Ftotal does not change (Finfree + Foutfree = const.). To demonstrate that transport actually takes place in our experimental setup, we have added the quencher KI to the outside of the liposomes. EtBr is now transported from a quenched environment outside the liposome into a less quenching environment inside and Finfree + Foutfree is no longer constant. If transport takes place, an increase of Finfree proportional to [EtBr]in is expected. Indeed, the time trace 1 shown in Fig. 2b
is the sum of two components. The first component stems from binding to TBsmr, while the second, linearly increasing component arises from the accumulation of EtBr inside the liposomes. Without quencher, or with quencher inside and outside, only the first component as in Fig. 2a
is observed and saturates after 100 s (trace 2 in Fig. 2b
). With quencher, Ftotal is generally smaller because the contributions from Foutfree are reduced. Our data show that EtBr is transported by TBsmr in a pH-dependent fashion while F arises from the formation of pH-dependent protein substrate complex formed during transport.
Other factors that could influence F (pH, EtBr concentration, lipids)
Several control experiments were carried out to exclude other factors, which could also potentially contribute to F.
As a pH gradient is created by bR, the inside of the liposome is acidified, which could affect the fluorescence of EtBr inside. However, a pH titration of TBsmr in solution revealed no change in fluorescence intensity between pH 5 and pH 7 (Supplemental Fig. 4), which is in agreement with its low pKa values of 0.2 and 2 (23)
.
EtBr is hydrophobic (predicted LogP of 0.3, www.molinspiration.com) and partitions into the inner and outer membrane leaflet. This more hydrophobic environment could cause an EtBr fluorescence change but when E. coli lipids were titrated to a given amount of EtBr in solution (Fig. 3
a), no fluorescence increase was observed.
When increasing the EtBr concentration in solution, a small fluorescence decay due to the formation of nonfluorescent dimers (24)
was seen (Fig. 3b
). This effect was also observed in the presence of liposomes (Fig. 3c
).
These controls show that neither pH nor binding to lipids nor concentration effects influence the EtBr fluorescence intensity in a way relevant for the interpretation of our experimental data in Fig. 2
.
Competitive inhibition with TPP+
To compare our results to established radioactive methyl viologen transport assays, we have selected another substrate, TPP+, for competitive inhibition of EtBr transport. TPP+ does not show an intrinsic fluorescence and is highly compatible with our experimental setup. The pH gradient generated by bR is stable over long periods of time, and measurements can be repeated multiple times. Therefore, only a single TBsmr sample per titration curve is required. Because F stems from an EtBr-TBsmr complex formed during the transport process, Fmax can serve as a measure of EtBr transport activity. In Fig. 4
a, (Fmax) is plotted vs. [TPP+]. A sigmoidal model was fitted to the data giving for TPP+ an IC50 of 10 ± 2 µM, which corresponds to a Ki of 1.5 µM.
|
Fmax depends on [EtBr] and is described by a single site-binding model
If the fluorescence changes observed for EtBr arise from substrate binding to TBsmr during the transport cycle, then it must be possible to determine a Kd for this substrate protein complex. Figure 4b
shows Fmax plotted against the EtBr concentration. Fmax saturates at 10 µM EtBr. Neither altering the incident light intensity, nor increasing the number of transporters in the vesicles (by changing the TBsmr/lipid ratio) nor altering pH (by changing the bR/lipid ratio) had an effect on the hyperbolic shape of these binding curves. The data fitted well to a single-site binding equation yielding a Kd of 2.6 ± 0.2 µM (Fig. 4b
).
Anisotropy measurement
To further characterize binding of EtBr to TBsmr during transport, we have performed "steady-state" fluorescence anisotropy measurements. In analogy to Eq. 2
, the observed total anisotropy rtotal consists of contributions from the equilibrium state at the beginning of the experiment and of contributions from EtBr bound to TBsmr in the presence of pH. The anisotropy is given as
 | (5) |
f stand for the fractional fluorescence intensities and r0, r for the corresponding anisotropies without and with pH. Eq. 5
is solved for r, which corresponds to the anisotropy of EtBr bound to TBsmr during the transport cycle:
 | (6) |
a shows the increase of rtotal with time and follows in principle the time course of Ftotal in Fig. 2a
. The total anisotropy starts from its equilibrium value (pH=0), increases when pH is activated, and saturates above 100 s when a steady state is reached. The more EtBr-TBsmr complexes form, the higher will be rtotal. We have determined the actual anisotropy of EtBr bound to TBsmr by calculating rtrans (Eq. 6)
, which is shown in Fig. 5b
. As rtrans does not depend on the number of complexes formed, it remains constant over time with a relatively high value of 0.33 ± 0.04. The large error of the data points below 25 s is caused by the small signal-to-noise ratio of their fluorescence intensities. They have been ignored for calculating the average value of rtrans.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
F (Fig. 2a
, traces 1 and 2). This fluorescence increase disappears when adding nigericin (Fig. 2a
, trace 3) or CCCP (Fig. 2a
, trace 5), which collapses transmembrane pH gradients or the proton motive force. These controls prove that F is pH dependent. Another control experiment has been carried out to exclude pH-dependent unspecific binding: A single, highly conserved Glu residue in the first helix is essential for transport in all SMR proteins (14)
. Therefore, we have introduced an E13A mutation in TBsmr and found F to be 0, which proves that F is caused by TBsmr (Fig. 2a
, trace 4). The observation, that the maximum value Fmax doubles when the number of transport proteins within the liposome is doubled, led us to the conclusion that F arises from the pH-dependent formation of a substrate-protein complex (Fig. 2a
).
During the TBsmr transport cycle, several events such as EtBr binding, translocation, release, and accumulation in the membrane and the lumen of the liposomes take place. All of them could involve alterations of the substrate environment, which potentially contribute to changes in total fluorescence intensity F. In analogy to Eq. 1
, F can be separated in potential contributions from EtBr free in solution, bound to TBsmr and nonspecifically bound to the liposomes, which could all change when pH is generated:
 | (7) |
Contributions from Fliposomebound
The fluorescence properties of EtBr are not altered when bound to lipids, as demonstrated in Fig. 3a
, and therefore, EtBr/liposome binding makes no contribution to F and Fliposomebound. EtBr (LogP=0.3) probably behaves like doxorubicin (LogP=0.6), another multidrug transporter substrate, which has a similar hydrophobicity and accumulates in the interface region of the lipids near the charged head groups, as shown by solid-state NMR spectroscopy (25)
. The interface region is much less hydrophobic than the acyl chains (26)
, which explains why no fluorescence change is detected on EtBr binding.
Contributions from Ffree
At the beginning of the experiment without a proton gradient, the EtBr concentration on the inside and outside is equal. During illumination, the liposomes are continuously acidified on the inside, which could create a pH-dependent change of EtBr fluorescence on the inside. We have experimentally verified that EtBr fluorescence is independent of pH in the range of pH 5–7 (Supplemental Fig. 4). This is consistent with measurements reported by others, where it was shown that EtBr fluorescence only changes near its pKa of 0.2 and 2 (23)
. So our experimental pH differences between the inside and outside of the proteoliposomes have no significant effect onto the fluorescence of free EtBr (Ffree).
The increasing concentration of EtBr within the lumen could change its fluorescence. However, it will eventually cause a slight fluorescence quenching due to the formation of nonfluorescent dimers (24)
, as shown in Fig. 3b, c
. Therefore, EtBr concentration effects can be ruled out as a source of the observed fluorescence increases.
Contributions from Fbound
Our data clearly show that F arises from the pH-dependent formation of an EtBr-TBsmr complex. Such a complex could be either the formation of an intermediate state, a TBsmr conformation with an increased binding capacity, or simply arise from binding of EtBr to inversely oriented TBsmr inside the vesicles. In the first case, F is caused by a change in environment of EtBr, while in the second and third case, it would be simply the result of a greater capacity to bind substrate.
An increase in binding capacity does not explain our data for two reasons. 1) We have shown that EtBr is actually transported, which requires more than just a binding step (Fig. 2b
). 2) Our data in Fig. 4b
are well described with a single-site binding model, while an increase in binding capacity would be equivalent with multiple binding sites. This conclusion is consistent with results for LacY, in which no change in binding capacity was found in the presence of an H+ electrochemical gradient (27)
.
In our experiment, TBsmr is most likely oriented both ways within the membrane. When EtBr is transported into the vesicles, it could bind to inversely oriented transporters and cause a fluorescence increase Fmax. However, our experiment starts from an equilibrium situation with [EtBr]in = [EtBr]out. When increasing this initial substrate concentration, Fmax should become smaller, as more and more binding sites would be already occupied before pH is activated and transport into the liposomes takes place. However, Fig. 4b
shows that actually the opposite effect is observed.
Since these simple binding models can be ruled out and a conformational change is needed to switch the binding pocket into a release site, an intermediate state is the simplest explanation to explain F:
 | (8) |
and by chemical cross-linking (29)
that the equilibrium state binding pocket of EmrE is water accessible. Therefore, the proposed intermediate can be explained by a temporarily occluded state, in which EtBr is not accessible on either side to water molecules, which strongly quench its fluorescence (30)
. A model summarizing our findings is shown in Fig. 6
: To perform substrate transport, a pH-dependent conformational change takes place, in which EtBr shows enhanced fluorescence because it is shielded from water molecules.
|
Besides energized uptake, transport can take place by substrate exchange. Because the experiment starts with equal concentrations of EtBr inside the liposomes and in the bulk solution, there should be a continuous exchange of substrate. As there is no indication of a second substrate pathway (31)
, the exchange should also proceed via the proposed occluded state. Therefore, the fluorescence increase could only be observed if the active transport would be faster than substrate exchange. Indeed, this has been shown for EmrE, in which energized transport of methyl viologen is up to 15 times faster than substrate exchange (32)
.
Characterization of an intermediate substrate-protein complex
We have carried out several experiments to further strengthen our conclusions and to characterize the TBsmr transport cycle intermediate state. In analogy to methyl viologen transport assays, it is possible to inhibit the transport by TBsmr by another substrate such as TPP+ (Fig. 4a
). The resulting IC50 (10 µM) and Ki (1.5 µM) seem rather high compared to a Ki of 30 nM and IC50 of 35 nM known for EmrE but are consistent with previous results of TBsmr in a methyl viologen transport assays (IC50=4.5 µM, Ki=4.3 µM) (14)
. For this comparison, the Ki values were calculated by us from the data reported by Schuldiner and coworkers (14
, 20)
.
The observed hyperbolic increase in Fmax with increasing EtBr concentration can be analyzed in terms of a single-site binding equation (Fig. 4b
). This suggests that F arises from a single intermediate state and not from multiple binding events. The determined Kd of 2.6 µM corresponds to a steady state between EtBr free and bound to the intermediate state.
The formation of this intermediate complex is also supported by fluorescence anisotropy measurements. Although the total anisotropy of free and bound ethidium increases slightly (Fig. 5a
), the fluorescence anisotropy of the substrate-protein complex does not change during the experiment, supporting a single state with immobile substrate (Fig. 5b
). The high anisotropy rtrans = 0.33 ± 0.04 of ethidium in the complex is consistent with a tightly bound, rigid substrate. We compare EtBr intercalated in DNA, which has been well studied: At a high viscosity of 
= 16 mPa · s, EtBr has a similarly high anisotropy of 0.3. Depending on the DNA length, the anisotropy decreases to 0.2 (97 bp) and 0.1 (32 bp) at a viscosity of = 1 mPa · s, which is attributed to internal motions for the larger DNA and helix tumbling for the shorter DNA (33)
. Considering the size of the protein in comparison to the DNA, we attribute the decrease of the anisotropy compared to a theoretical vale of r = 0.4 to wobbling of the drug in the binding site and to internal motions of the protein and the whole liposome.
Relevance for transport models
We have presented experimental data on changes of the intrinsic fluorescence properties of ethidium bromide while being transported through the membrane by TBsmr. The combination of bacteriorhodopsin to create a stable pH gradient over a long period of time with EtBr is especially convenient, as only a standard fluorescence spectrometer is needed. Experiments can be repeated many times on the same sample, enabling time-efficient titration studies as shown in Fig. 4
. It is a useful complement to radioactive measurements for nonfluorescent substrates.
Our data suggest that TBsmr and also EmrE are undergoing conformational changes in the presence of a pH gradient, as monitored indirectly by changes of the EtBr environment. In a preliminary report of this experimental setup (34)
, EmrE did behave similar to TBsmr with an intermediate state Kd of 4.2 µM. These findings show the existence of a transport cycle intermediate, which only forms in the presence of pH.
The occluded substrate-protein complex found here represents an intermediate state in the alternating access model proposed for EmrE (12)
in which the protein switches between conformations with alternate access to the inside or the outside. It is similar to the alternating access model proposed for another secondary transporter, LacY (35)
, where a release of water molecules and a closing of the binding pocket has been observed in MD simulations (36)
. A transport intermediate state between open conformations accessible to either side has also been found for the Na+-K+-ATPase, in which potassium ions temporarily become occluded (37)
.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The application of NMR, FTIR, or EPR methods with sample illumination to SMR proteins co-reconstituted with bacteriorhodopsin will allow steady-state investigations of this transport cycle intermediate in more detail. The kinetics of its formation could be followed by time-resolved fluorescence spectroscopy, in which a pH gradient is generated using caged protons (38)
.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Current address: Department of Chemistry, University of Hull, Hull HU6 7RX, UK.
Received for publication June 9, 2007. Accepted for publication August 16, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  I. Lehner, D. Basting, B. Meyer, W. Haase, T. Manolikas, C. Kaiser, M. Karas, and C. Glaubitz The Key Residue for Substrate Transport (Glu14) in the EmrE Dimer Is Asymmetric J. Biol. Chem., February 8, 2008; 283(6): 3281 - 3288. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
