| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Groupe d'étude des protéines membranaires, Université de Montréal, Montreal, Quebec, Canada
Correspondence: Address reprint requests to J.-Y. Lapointe, Groupe d'étude des protéines membranaires (GÉPROM), Université de Montréal, C.P. 6128, succ. Centre-ville, Montréal, Québec H3C 3J7, Canada. Tel.: 514-343-7046; Fax: 514-343-7146; E-mail: jean-yves.lapointe{at}umontreal.ca.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
-methyl-glucose (MG, a nonhydrolyzable glucose analog) concentrations. In sharp contrast, two SGLT1 mutants (C255A and C511A) that lack a recently identified disulfide bridge express the pre-steady-state currents in the presence of MG. The dose-dependent effects of MG on pre-steady-state currents were studied for wild-type (wt) SGLT1 and for the two mutants. Increases in MG concentration reduced the total transferred charge (partially for the mutants, totally for wt SGLT1), shifted the transferred charge versus membrane potential (Q-V) curve toward positive potentials, and significantly modified the time constants of the pre-steady-state currents. A five-state kinetic model is proposed to quantitatively explain the effect of MG on pre-steady-state currents. This analysis reveals that the reorientation of free transporter is the slowest step for wt SGLT1 either in the presence or in the absence of MG. In contrast, the conformational change of the fully loaded mutant transporters constitutes their rate-limiting step in the presence of substrate and explains the persistence of pre-steady-state currents in this situation. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
), expression in Xenopus oocytes enabled the measurement of pre-steady-state currents, i.e., transient currents observed in the absence of substrate, which were suggestive of gating currents observed in voltage-dependent channels. As these currents were Na+ dependent and were absent in the presence of the specific inhibitor phlorizin (Pz) or in the presence of glucose, they were considered to represent charge displacements occurring during the voltage-dependent reorientation of the Na+-binding site and upon Na+ binding (2).
Pre-steady-state currents have been extremely useful for devising a credible transport mechanism, with quantitative estimation of the rate constants linking the different conformational states. The original model (2) proposed in 1992 has been challenged both theoretically and experimentally ((3–8); for review see Wright and Turk (9)), and new steps have been proposed. Recently Loo et al., using fluorescently labeled mutants (10), have reported extremely slow conformational changes (time constants on the order of 100 ms) which have yet to be quantitatively explained by any proposed kinetic model. In particular, this observation is incompatible with a rate-limiting step of 50 s–1, which was proposed for the translocation of the fully loaded transporter in the human isoform of SGLT1 (hSGLT1) (2,11).
One characteristic observed for nearly all cotransporters studied (Na+/glucose cotransporters SGLT1 and SGLT2, Na+/myo-inositol cotransporters SMIT1 and SMIT2, Na+/monocarboxylate cotransporter SMCT1, Na+/Pi cotransporter NaPiII, Na+/I– symporter NIS, gamma-aminobutyric acid transporters GAT1 and GAT3, H+/hexose cotransporter STP1, and Cl–-dependant K+/amino acids transporter KAAT1) is that addition of a saturating concentration of substrate leads to the total inhibition of pre-steady-state currents (11–25). Surprisingly, with the exception of GAT1, no quantitative explanation has been proposed to explain this phenomenon.
Recently, we identified a disulfide bridge between C255 and C511 in hSGLT1 (26). An interesting feature of mutants C255A and C511A, which we have not previously published, is that they express pre-steady-state currents in the presence of a saturating MG concentration, in contrast to what is observed with wild-type (wt) SGLT1. This phenomenon has prompted us to examine the dose-dependent effects of MG on the pre-steady-state currents for the two mutants as well as for wt SGLT1 and to propose a quantitative explanation using a kinetic model displaying different rate-limiting steps for the wt SGLT1 and the mutant cotransporters.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
). One day after defolliculation, oocytes were injected with 46 nl of water containing mRNA (0.1 µg/µl and 0.25 µg/µl for wt SGLT1 and mutants, respectively) to obtain maximal protein expression. Oocytes were maintained in Barth's solution (in mM: 90 NaCl, 3 KCl, 0.82 MgSO4, 0.41 CaCl2, 0.33 Ca(NO3)2, 5 Hepes, pH 7.6) supplemented with 5% horse serum, 2.5 mM Na+ pyruvate, 100 units/ml penicillin, and 0.1 mg/ml streptomycin for 4–7 days before electrophysiological experimentation.
Molecular biology
The constructions prepared for obtaining the mutants C255A and C511A have been described elsewhere (26).
Electrophysiology
The saline solution normally used in our electrophysiological experiments is composed of (in mM): 90 NaCl, 3 KCl, 0.82 MgCl2, 0.74 CaCl2, and 10 Hepes and the pH was adjusted to 7.6 with NaOH. Two-electrode voltage-clamp experiments were performed using an Oocyte Clamp OC-725 (Warner Instruments, Hamden, CT) and a data acquisition system (Digidata 1322A and Clampex 8.2, Axon Instruments, Union City, CA). Current and voltage microelectrodes were filled with 1 M KCl and had a resistance of 1–2 M
. The bath current electrode was an Ag-AgCl pellet, and the reference electrode was a 1 M KCl agar bridge. The oocytes were clamped to a membrane potential (Vm) of –50 mV, and three repetitions of Vm steps between +70 and –170 mV (by increments of 20 mV, 300 ms duration, no series resistance compensation used) were applied with an interval of 1.7 s between each step. Ninety-five percent of the command voltage step was reached in 3–4 ms. Data were obtained with a sampling frequency of 10 kHz, without filtering, and the three repetitions were averaged for each experiment.
Data analysis
Pre-steady-state current analysis was performed as described previously (26). Briefly, the transferred charge was obtained at each membrane potential (Q-V curve) by subtracting the integrated baseline-corrected currents in Pz solution (200 µM) from similar currents in saline solution (either in the presence or absence of MG). Thus, the transferred charge calculated corresponds to the total charge in one experimental condition minus the total charge in the presence of Pz. The total charge in the presence of Pz was found to be linear with voltage as expected if it was mainly due to the presence of the oocyte capacitive current. The baseline correction was obtained from the mean current measured between 50 and 80 ms after the initiation of a voltage pulse. A simple Boltzmann equation was fitted to the Q-V curve to estimate V1/2 (the voltage at which half of the charge is transferred), Qmax (the amplitude of the total charge transferred), and z (the valence of the transferred charge) (26). The time constants (
slow) were evaluated by fitting a double exponential on the Itransit (Isaline – IPz) with the Clampfit 8.2 program (Axon Instruments). Only the slow time constant (slow; 2–10 ms), which has the dominant amplitude, was considered.
Statistics
Experiments were performed on at least six oocytes obtained from a minimum of two different donors. Data are reported as means ± SE and are compared using unpaired Student's t-test; statistical significance was set at P < 0.05. Errors bars were omitted when smaller than the symbol size.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
MG for C255A and C511A-methyl-glucose (MG) concentration (26), particularly at depolarizing Vm levels. Figs. 1 A and 2 A show the Pz-sensitive currents with different voltage steps for the two mutant proteins using several MG concentrations (0, 1, 5, and 10 mM MG). As the capacitive currents are eliminated by subtraction of currents measured in the presence of Pz, the transient currents at each voltage step directly represent SGLT1-specific pre-steady-state currents. The integrals of the transient currents measured at different Vm are used to produce transferred charge versus Vm (Q-V) curves. The Q-V curves for the two mutants are shown in Figs. 1 B and 2 B. As the cotransporter conformation at very positive Vm levels is predicted to be independent of the presence of extracellular MG (the binding site being predicted to face inside in all cases (2,4,10)), each Q-V curve was shifted vertically to have Q = 0 at +50 mV under all conditions. This allows direct comparison of the transferred charge at different Vm; it is clear that they decrease as the MG concentration increases. It is also clear from Figs. 1 B and 2 B that the voltage range over which charge can be transferred is reduced in amplitude and displaced toward more positive potentials when the MG concentration is increased. The measured charge, at saturating MG concentration, has reached its plateau level at –50 mV and remains basically constant as Vm becomes more negative. However, a saturating MG concentration does not totally abolish the transferred charge. A Boltzmann relation can be fitted to the Q-V curves, which yields a value for the parameter V1/2, the Vm at which half of the mobile charge is equally balanced between the inward and outward facing positions. For both mutants, an increase in MG concentration shifts the V1/2 toward more positive values. For mutant C255A, the measured V1/2 averaged –35 ± 5 mV at 0 mM and 5 ± 5 mV at 10 mM MG (n = 6). For mutant C511A, the corresponding values were –28 ± 2 mV at 0 mM and 13 ± 3 mV at 10 mM MG (n = 6).
|
|
MG also affected the slow for pre-steady-state currents as depicted in Figs. 1 C and 2 C. They were obtained by fitting double exponentials to the Pz-sensitive currents. As reported previously (26), in the absence of MG, slow for the mutant proteins reached a plateau of 4–5 ms at hyperpolarized Vm, which is about half of the value measured for the wt SGLT1. An increase in MG concentration clearly produced an increase of slow at positive Vm. For Vm more negative than –50 mV, addition of MG accelerated the transient currents, as shown in Fig. 2 C for 1 mM MG. The currents in the presence of higher concentrations can no longer be fitted accurately at these voltages (see Fig. 2 C for 10 mM MG and Fig. 1 C for 5 and 10 mM MG). Consequently, at 1 mM MG, the slow versus Vm (slow-V) curve has a bell shape centered around –50 mV for both mutants. At 10 mM MG, slow starts at very low values for negative Vm and reaches a value of 6–7 ms for both mutants at depolarizing potentials.
The stability of the preparation as a function of time was tested in each experiment by comparing pre-steady-state currents in the absence of MG measured before and after having presented the different MG concentrations. In all cases, the Q-V and slow-V curves were found identical. In addition, the effects of MG on the pre-steady-state currents were found to be independent on the order of the MG concentrations applied (increasing or decreasing concentrations).
Pre-steady-state currents in the presence of MG for wt SGLT1
Although the inhibitory effect of MG on wt SGLT1 pre-steady-state currents has long been known (11,25), to our knowledge it has never been studied quantitatively in a dose-dependent manner nor has it been explained with the use of a kinetic model. Given our findings with the two SGLT1 mutant proteins, we sought to characterize this effect in detail on wt SGLT1 and to compare it with that observed for the mutants. Fig. 3 A shows Pz-sensitive pre-steady-state currents recorded from wt SGLT1 in the absence or in the presence of different MG concentrations (0.5 mM, 1 mM, and 5 mM). In agreement with previous reports, addition of extracellular MG leads to a progressive decrease in the pre-steady-state currents and to the appearance of steady-state inward Na+/glucose flux, which have been described in detail (11,25). The Q-V curves obtained with different MG concentrations is illustrated in Fig. 3 B. It is obvious that a saturating concentration of MG (5 mM) completely abolished charge transfer (n = 9), at least within the time resolution provided by the two-electrode voltage-clamp technique. In the presence of 5 mM MG, the amplitude of the transferred charge that can be detected is approximately equal to that measured from a noninjected oocyte (<1 nC). The time constants of these pre-steady-state currents are shown in Fig. 3 C. Increased MG concentrations produce a clear acceleration of the transient currents at negative Vm, whereas the value of the time constant remains the same at positive Vm. In the presence of 5 mM MG, the transient currents are difficult to fit to a double exponential equation because slow shows only modest voltage dependence and reaches a value of 
2.5 ms (at +70 mV) with very small amplitude (n = 7). Both parameters are close to the limits inherent to the two-electrode voltage-clamp technique.
|
MG on the V1/2 and on the normalized amplitude of the total transferred charge for wt SGLT1 versus the two mutants. In Fig. 4 A, the shift in V1/2 produced by the addition of MG to the wt SGLT1 is compared with the shifts mentioned above for the two mutants. For wt SGLT1, V1/2 goes rapidly from –60 ± 5 mV at 0 mM to 12 ± 12 mV at 1 mM MG (n = 9) and, at MG concentrations higher than 1 mM, the amplitude of the Q-V curve is reduced to such an extent that the fitting of a Boltzmann curve is not reliable. In all cases, the V1/2 is progressively shifted toward positive Vm levels as the external [MG] increases, but this shift is less marked for the mutants than for wt SGLT1.
|
MG, the decreases in transferred charge caused by MG were normalized to the total transferred charge in the absence of substrate (i.e., 
). A simple Michaelis-Menten equation was fitted to taken at –170 mV as a function of MG concentration after setting Q = 0 for Vm = +50 mV (as done in Figs. 1 B, 2 B, and 3 B). It is clear that for the wt SGLT1, a high MG concentration inhibits 100% of the transferred charge, whereas only partial inhibition (65–75% inhibition) was observed for the mutant proteins. It was found that the MG-sensitive charge is consistent with MG affinity constants 
of 0.48 ± 0.05 mM for wt SGLT1 (see Fig. 4 B), 5 ± 2 mM for mutant C255A, and 2 ± 1 mM for mutant C511A whereas their 
constants, using MG-sensitive cotransport current (steady-state values, Iss(MG)), were 0.97 ± 0.1 mM for wt SGLT1 (26,27), 1.6 ± 0.2 mM (C255A), and 1.6 ± 0.2 mM (C511A) at –170 mV (26).
Kinetic model for pre-steady-state currents
A scheme of the simple five-state kinetic model used is presented in Fig. 5. The substrate-binding and -debinding steps (k45 and k54) are voltage independent (2,28) as are the lumped reactions k41 and k51, which represent the conformational changes of the Na+-loaded transporter (involved in the leak current) and the Na+- and MG-loaded transporter, respectively, with their associated intracellular release steps. It was found that the extracellular Na+-binding reaction could be assumed to be a fast reaction at equilibrium without losing any fitting performance. The voltage-dependent reactions are expressed as follows (for i = 1, 2, 3; j = i + 1):
 | (1) |
i represents the asymmetry of the energy barrier, Vm is the membrane potential, and F, R, and T have their usual meanings (see also Table 1).
|
|
was previously estimated for rabbit SGLT1 in giant, excised, inside-out patches and was found to vary from 44 to 70 mM (28,29). In agreement with these estimates, we recently found that intracellular Na+ has to be increased by blocking the Na+/K+-ATPase overnight to generate a measurable outward Na+/glucose current upon intracellular glucose injection (30). This is consistent with an intracellular 
much higher than the physiological intracellular Na+ concentration. On the other hand, the estimation of 
for the intracellular site is 35 mM (28,29). Consequently, under physiological conditions, the inverse mode of transport is highly unlikely, which is reflected by very low values of k14 and k15 (see Table 1). Thus, we assumed that the probability of finding the intracellular site loaded with Na+ and MG was negligible, and we reduced the potential seven-state kinetic model into the five-state kinetic model shown in Fig. 5.
The numerical simulations were performed using MATLAB 6.5.0 software (MathWorks, Natick, MA). Transferred charges were calculated as the integral of the pre-steady-state currents as done during the analysis of the experimental data and were also fitted with a simple Boltzmann curve to deduce the V1/2, Qmax, and z parameters. As the analytical expression for the time constant in the presence of substrate cannot be obtained in a five-state model, the numerical values of the time constant were obtained by taking the reciprocal of the eigenvalues of the following matrix, as previously reported (10):
 |
Moreover, the model predicts the relaxation of pre-steady-state currents toward steady-state currents. The current (I) versus Vm curves are sigmoid, as observed for the experimental current versus membrane potential (I-V) curves in the absence and in the presence of MG (not shown). The rate constants of the model were adjusted by trial and error to obtain a satisfactory fit to the measured Q-V curves, the V1/2, and the slows of the pre-steady-state currents. The steady-state parameters (I-V curves, , or ) were not considered as criteria for the adjustment of the model parameters but were found afterwards to be in accordance with the experimental values.
Simulation of pre-steady-state currents in the absence of MG
Table 1 gives the rate constants used by Chen et al. (4) and by Loo et al. (11) as well as the rate constants used by this study to simulate the time constants of the pre-steady-state currents. As shown in Fig. 6 A, our new set of rate constants can reproduce, in a generally satisfying manner, the currents, the Q-V curves, and the slow-V curves for wt SGLT1 exposed to different MG concentrations. In the presence of 90 mM Na+, a k210/k120 ratio of 0.5 is required for the plateau effect seen on the slow-V curve at hyperpolarizing Vm. k23 is a crucial rate constant because it is voltage independent and becomes rate limiting for Vm below –70 mV (at 90 mM Na+). Indeed, the value of 1/k23 closely corresponds to the plateau value reached by slow at very negative Vm. The ratio k23/k32 is also responsible for the voltage dependence of slow observed at depolarizing Vm. At 0 mV, K34 (the ratio k43/k34) is 0.1 M2 for wt SGLT1 and 0.07 M2 for the mutants and is largely responsible for the Na+ affinity measured at low MG concentration. This ratio also has a large influence on the position of the V1/2 of the Q-V curve. The K34 and the k23 values were modified for simulation of transferred charges through the mutant proteins to account for the faster slow at negative Vm and the new V1/2 of –30 mV (instead of –50 mV), which was observed for both mutants (26). These two simple modifications yielded the fit for the Q-V curve of Fig. 6 B (left panel), shown for mutant C511A, and the slow-V curve shown in Fig. 6 B (right panel) at 5 mM MG.
|
MG on the pre-steady-state currentsMG on the pre-steady-state currents, appropriate values for the parameters k45, k54, and k51 have to be determined. Our strategy was to start by establishing parameters that could explain the behavior of the transferred charges of wt SGLT1 in the presence of different MG concentrations. Our first criterion was that the transferred charge had to disappear in the presence of a saturating MG concentration. The kij of the four-state model established in the absence of MG were maintained constant, and we investigated the effects of the three new kij on the simulated Q-V curve. We started with the parameters proposed by Loo et al. (11) (see Table 1) but needed to increase the value of k51 by up to 40-fold over the original value (2000 vs. 50 s–1) to reproduce the charge disappearance observed at high MG concentration. It was also clear that the ratio k45/k54 influenced the V1/2 of the Q-V curve: an increase in this ratio shifts the V1/2 toward more positive Vm. However, the absolute values of k45 and k54 (and not only their ratio) were also important in the global behavior of the Q-V curve as a function of MG concentration. With respect to the values proposed by Loo et al. (11), the values of k45 and k54 had to be increased by factors of 30 and 50, respectively.
For the mutants, we first established the values of the parameters k210, k23, and K340 to account for the faster time constants and the positive shift in V1/2 in the absence of MG. Changes in k210 and k23 were necessary along with more modest changes in the remaining parameters describing the MG-independent steps of the kinetic model. The new values of k45, k54, and k51 found for the wt transporter could not reproduce the observed Q-V and slow-V curves of the mutants. We found that the reactions describing MG binding and debinding had to be reduced by an order of magnitude and the reorientation of the fully loaded carrier (k51) had to be massively decreased from 2000 to 80 s–1. The best parameter set found is presented in Table 1 for wt SGLT1 and the mutants.
Simulations of wt SGLT1
The simulated wt SGLT1 Q-V curves are shown in Fig. 6 A (middle panel), superimposed on the experimental data points (in gray). Although, the shapes of the theoretical and experimental Q-V curves differ slightly, the general decrease due to external MG concentration is well represented. Three parameters were used to measure the accuracy of the model predictions: the V1/2 of the Q-V curves, the normalized transferred charge curves in the presence of different MG concentrations 
, as well as the slow-V curve. Fig. 4 A illustrates the phenomenological parameter V1/2 as a function of MG concentration extracted from the fits of a simple Boltzmann relation to our theoretical Q-V curves (solid line). Although the theoretical Q-V curve differs slightly from a simple Boltzmann relation, the model correctly represents the voltage shift produced by MG addition. The model also correctly predicts the total inhibition of the transferred charge by a saturating MG concentration. We estimated an apparent affinity for MG , using the remaining charge in the presence of MG , to compare it with that obtained experimentally. The parameters given in Table 1 yield a 
value of 0.40 ± 0.07 mM at –170 mV, which is not significantly different from the experimental value reported above. Finally, Fig. 6 A (right panel) shows that the slow-V curve values are close to the experimental values as the model reproduces very well the acceleration of the transient currents at hyperpolarizing Vm and shows the bell-shaped curve peak shifting toward more positive Vm as the MG concentration increases.
Simulations of the mutant SGLT1s
The simulated Q-V curves for the mutant SGLT1s are shown in Fig. 6 B (middle panel), superimposed on the experimental data points for mutant C511A alone (in gray) because mutant C255A produced very similar values. The charge plateau value reached at hyperpolarizing Vm, at 5 and 10 mM MG, is reproduced very well by the modeled Q-V curve. The dotted line on Fig. 4 A illustrates V1/2 as a function of MG concentration for the mutants. It is clear that the estimated V1/2 for the modeled Q-V curves of the mutants closely reproduced the characteristics of both mutants. The model accounts for the partial inhibition of the transferred charge at high MG concentrations. In addition, the was estimated with the remaining charge in the presence of MG and provided the value of 4 ± 2 mM at –170 mV, which is close to the experimental values reported above for C255A and, to a lesser extent, for C511A (5 ± 2 mM and 2 ± 1 mM, respectively). Finally, the theoretical slow values were superimposed on the experimental values for both mutants at 5 mM MG in Fig. 6 B (right panel). The model predicts two exponentials with significant amplitudes with time constants in the range of 2–6 ms in the presence of MG. The first one is almost identical with that observed in the absence of MG. The second one is slower at depolarizing Vm where it reaches a plateau value of 5.5 ms. Experimentally, a single exponential with a time constant in the millisecond range could be detected. Given the limited speed of the voltage pulse, the typical noise level found in our current recording, and given the fact that the two predicted time constants are in the same order of magnitude, it is conceivable that our experimental time constant would correspond to some intermediate value between the predicted ones. Thus it is concluded that the model reproduces fairly well the experimental time constants measured in the presence of MG.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
). The two mutants C255A and C511A were found to display further interesting features which confirmed the importance of this region of the cotransporter. In this study we report that, in contrast with wt SGLT1, these two mutants exhibit pre-steady-state currents in the presence of a saturating MG concentration. By analyzing the dose-dependent effects of MG on the pre-steady-state currents of these mutants as well as for wt SGLT1, we sought to identify a satisfying kinetic explanation for both the partial diminution of mutant pre-steady-state currents by MG and for the complete disappearance of the wt SGLT1 transient currents.
The pre-steady-state currents in the absence of MG have been studied using cut open oocytes exposed to various Na+ concentrations, and a simple four-state kinetic model (4) was found to be consistent with the amplitudes and the time constants (fast (<1 ms) and slow (1–10 ms)) of the experimentally determined pre-steady-state currents as a function of the external Na+ concentration. The presence of these two time constants was more recently confirmed for rabbit SGLT1 (7,8) and for hSGLT1 (10). In this last study, fluorescently labeled cotransporters were also used and a slower time constant of 100 ms was reported in addition to fast and slow. A seven-state model was suggested for the translocation of the free transporter and the binding of two external Na+ ions, but the authors could not find a parameter set that would be in quantitative agreement with their own observations. Considering the time resolution provided by the two-electrode voltage-clamp technique, we decided to use the four-state model proposed by Chen et al. (4) to explain the effects of disrupting the disulfide bridge C255-C511 on the V1/2 of the Q-V and slow-V curves in the absence of MG (26). In the original model, it was assumed that a single Na+ ion was involved in the pre-steady-state currents. To incorporate MG binding, and given that the cotransport stoichiometry is 2 Na+:1 glucose (31), we simply replaced the original rate constant for Na+ binding (k34) by k34/[Na+] to account for both Na+ ions and made it a second order rate constant in M–2s–1. As the extracellular Na+ concentration is constant in this study, further studies will have to test whether the model used is consistent with the effects of changes in external Na+ concentration.
Occupancy probabilities in the presence and absence of MG
In the absence of MG and at –50 mV, the set of rate constants proposed in Table 1 leads to occupancy probabilities (Ci) of 5%, 22%, 43%, and 30% (from i = 1 to 4), indicating that 73% of the Na+-binding sites are exposed outside either in a free or Na+-bound state (4,26). Obviously this situation is highly voltage dependent and, at +70 mV, C1 and C2 now represent 52% and 40% of the cotransporter conformations, respectively. If, from state C4 to state C1, the total number of unitary charges that can move across the entire membrane electrical field is 2 (|z1 + z2 + z3| = 2), the occupancy probabilities at +70 mV indicate that all but 11% of it has already moved into the inward facing configuration. Fig. 7 A presents the occupancy probabilities in the absence of MG for an extreme voltage step from +70 to –150 mV. Upon hyperpolarization, C1 rapidly transforms into C2 and the free binding sites exposed to the extracellular solution (C3) become Na+-bound immediately. The step C1 
C2 is considered to be mainly responsible for the fast component to the transient current. In contrast, the following slow transformation of C2 into C3 (which is in equilibrium with C4(2Na+)) is clearly responsible for the slow component of the observed transient currents. Fig. 7, B and C, presents the changes in occupancy probability for a similar voltage step but in the presence of MG at 1 or 5 mM. At +70 mV, the starting probabilities are independent of the presence of MG in the extracellular solution as are the fast events occurring in the first millisecond after hyperpolarization. At –150 mV, the slowest rate constant in the reactions leading to the inward Na+/glucose current is clearly k23. This is why C2 accumulates transiently (75%) then relaxes to a value consistent with the steady-state cotransport rate allowed by the external MG concentration. This is shown in Fig. 7 D where the probability of finding C2 is plotted as a function of time for different MG concentrations.
|
MG, which is quite similar to the Ci probabilities found for wt SGLT1. At +70 mV, the distribution is slightly different from the wt SGLT1 as C2 now dominates with a probability of 42% with respect to C1 (34%), whereas the outward facing, free binding site (C3) presents a significant probability of 23%. Fig. 8 A presents the changes in Ci as a function of time for an extreme voltage pulse from +70 mV to –150 mV. Once again, in the absence of MG, C2 is transiently accumulated before relaxing to <5% as the Na+-bound form (C4(2Na+)) progressively rises to more than 90%. Fig. 8, B and C, depicts the occupancy probabilities in the presence of 1.5 and 5 mM MG. Under these circumstances, and in marked contrast to wt SGLT1, C2 continues to relax to a low value and it is C5(2Na+S), the fully loaded transporter, that progressively increases and attains up to 48% (this value increases to 58% at 10 mM MG). As illustrated in Fig. 8 D, contrary to what was seen for wt SGLT1, the C2 state increases (55%) and then relaxes to much lower steady-state values of 5% and 21% in the absence or presence of MG, respectively. This simply reflects the fact that, for the mutant proteins, the slowest rate constant in the steps mediating Na+/glucose cotransport at –150 mV is k51. As the steps involved in generating the slow component of pre-steady-state currents are the transition between C2 and C4(2Na+), Figs. 7 D and 8 D illustrate the reason transient currents disappear in the presence of MG for the wt transporter but not for the mutants.
|
MGMG)) and obtain 
or the substrate-dependent charge disappearance to obtain 
. For the wt SGLT1, both experimental (0.97 and 0.48 mM) and theoretical (0.36 and 0.40 mM) approaches show that the and the are close in value. However, the two experimental Km estimates for the mutants are significantly different, particularly for mutant C511A. It is important to specify that these two Km are apparent Km and depend not only on the rate constants k45 and k54 but also on the other rate constants. It seems that the rate-limiting step position is crucial for this discrepancy and that the two methods of obtaining an apparent affinity constant for MG should be considered with caution. The accordance between their values for wt SGLT1 may simply be coincidental.
Role of the disulfide bridge C255-C511 in SGLT1
In a previous study, we have shown that the breakage of a disulfide bridge between C255 and C511 using dithiothreitol or by disruption through specific alanine mutations led to a displacement of the equilibrium position of the "voltage sensor" and to an acceleration of time constant of pre-steady-state current in the absence of MG (26). In this study, we established that the disulfide bridge C255-C511 (in hSGLT1) also plays a major role in facilitating the conformational change of the fully loaded cotransporter. In addition, a minor role was also detected in the MG-binding and -debinding reactions. In the absence of a tridimensional structure, it is impossible to know the exact position of this disulfide bridge in relation to the Na+ or MG-binding sites, but it is certainly important for the mechanical structure involved in those processes.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
MG for the wt SGLT1 but not for the mutant transporters. In wt SGLT1 and in the presence of substrate, the rate-limiting step is from state 2 to state 3. The transferred charges are not observed in this case because, upon hyperpolarization from a very positive to a very negative Vm, the large steady-state current requires a high C2 probability. Under these circumstances, the steady-state transporter distribution is predicted to simply move from state 1 to state 2, which should generate only a very fast transient current ( 
0.5 ms). In contrast, for the mutants C255A and C511A in the presence of MG, the rate-limiting step is from state 5 to state 1. The transporter after having reached a high C2 probability will relax to a much lower level to reach the required C5 probability to account for the steady-state current. As the transporter moves from state 2 to states 3–5 through electrogenic steps, a slow transient current is generated. The behavior of the mutants underscores the role played by the rate-limiting step in the possibility of observing pre-steady-state currents. It also reveals the importance of the disulfide bridge C255-C511 in facilitating the translocation of the fully loaded transporter from the outward facing to the inward facing configuration. |
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
This work was supported by the Canadian Institutes of Health Research (grant No. MOP-10580). D.G.G. is a Natural Sciences and Engineering Research Council of Canada and Fonds de la recherche en santé du Québec postgraduate scholar.
Submitted on June 26, 2006; accepted for publication October 5, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
2. Parent, L., S. Supplisson, D. D. Loo, and E. M. Wright. 1992. Electrogenic properties of the cloned Na+/glucose cotransporter: II. A transport model under nonrapid equilibrium conditions. J. Membr. Biol. 125:63–79.[Medline]
3. Chen, X. Z., M. J. Coady, F. Jalal, B. Wallendorff, and J. Y. Lapointe. 1997. Sodium leak pathway and substrate binding order in the Na+-glucose cotransporter. Biophys. J. 73:2503–2510.
4. Chen, X. Z., M. J. Coady, and J. Y. Lapointe. 1996. Fast voltage clamp discloses a new component of presteady-state currents from the Na(+)-glucose cotransporter. Biophys. J. 71:2544–2552.
5. Falk, S., A. Guay, C. Chenu, S. D. Patil, and A. Berteloot. 1998. Reduction of an eight-state mechanism of cotransport to a six-state model using a new computer program. Biophys. J. 74:816–830.
6. Falk, S., N. Oulianova, and A. Berteloot. 1999. Kinetic mechanisms of inhibitor binding: relevance to the fast-acting slow-binding paradigm. Biophys. J. 77:173–188.
7. Krofchick, D., S. A. Huntley, and M. Silverman. 2004. Transition states of the high-affinity rabbit Na(+)/glucose cotransporter SGLT1 as determined from measurement and analysis of voltage-dependent charge movements. Am. J. Physiol. Cell Physiol. 287:C46–C54.
8. Krofchick, D., and M. Silverman. 2003. Investigating the conformational states of the rabbit Na+/glucose cotransporter. Biophys. J. 84:3690–3702.
9. Wright, E. M., and E. Turk. 2004. The sodium/glucose cotransport family SLC5. Pflugers Arch. 447:510–518.[CrossRef][Medline]
10. Loo, D. D., B. A. Hirayama, A. Cha, F. Bezanilla, and E. M. Wright. 2005. Perturbation analysis of the voltage-sensitive conformational changes of the Na+/glucose cotransporter. J. Gen. Physiol. 125:13–36.[CrossRef][Medline]
11. Loo, D. D., A. Hazama, S. Supplisson, E. Turk, and E. M. Wright. 1993. Relaxation kinetics of the Na+/glucose cotransporter. Proc. Natl. Acad. Sci. USA. 90:5767–5771.
12. Boorer, K. J., D. D. Loo, and E. M. Wright. 1994. Steady-state and presteady-state kinetics of the H+/hexose cotransporter (STP1) from Arabidopsis thaliana expressed in Xenopus oocytes. J. Biol. Chem. 269:20417–20424.
13. Bossi, E., E. Centinaio, M. Castagna, S. Giovannardi, S. Vincenti, V. F. Sacchi, and A. Peres. 1999. Ion binding and permeation through the lepidopteran amino acid transporter KAAT1 expressed in Xenopus oocytes. J. Physiol. 515:729–742.
14. Coady, M. J., M. H. Chang, F. M. Charron, C. Plata, B. Wallendorff, J. F. Sah, S. D. Markowitz, M. F. Romero, and J. Y. Lapointe. 2004. The human tumour suppressor gene SLC5A8 expresses a Na+-monocarboxylate cotransporter. J. Physiol. 557:719–731.
15. Coady, M. J., B. Wallendorff, D. G. Gagnon, and J. Y. Lapointe. 2002. Identification of a novel Na+/myo-inositol cotransporter. J. Biol. Chem. 277:35219–35224.
16. Eskandari, S., D. D. Loo, G. Dai, O. Levy, E. M. Wright, and N. Carrasco. 1997. Thyroid Na+/I- symporter. Mechanism, stoichiometry, and specificity. J. Biol. Chem. 272:27230–27238.
17. Fesce, R., S. Giovannardi, F. Binda, E. Bossi, and A. Peres. 2002. The relation between charge movement and transport-associated currents in the rat GABA cotransporter rGAT1. J. Physiol. 545:739–750.
18. Forster, I., N. Hernando, J. Biber, and H. Murer. 1998. The voltage dependence of a cloned mammalian renal type II Na+/Pi cotransporter (NaPi-2). J. Gen. Physiol. 112:1–18.
19. Hager, K., A. Hazama, H. M. Kwon, D. D. Loo, J. S. Handler, and E. M. Wright. 1995. Kinetics and specificity of the renal Na+/myo-inositol cotransporter expressed in Xenopus oocytes. J. Membr. Biol. 143:103–113.[Medline]
20. Kanai, Y., and M. A. Hediger. 2004. The glutamate/neutral amino acid transporter family SLC1: molecular, physiological and pharmacological aspects. Pflugers Arch. 447:469–479.[CrossRef][Medline]
21. Li, M., R. A. Farley, and H. A. Lester. 2000. An intermediate state of the gamma-aminobutyric acid transporter GAT1 revealed by simultaneous voltage clamp and fluorescence. J. Gen. Physiol. 115:491–508.
22. Sacher, A., N. Nelson, J. T. Ogi, E. M. Wright, D. D. Loo, and S. Eskandari. 2002. Presteady-state and steady-state kinetics and turnover rate of the mouse gamma-aminobutyric acid transporter (mGAT3). J. Membr. Biol. 190:57–73.[CrossRef][Medline]
23. Virkki, L. V., I. C. Forster, J. Biber, and H. Murer. 2005. Substrate interactions in the human type IIa sodium-phosphate cotransporter (NaPi-IIa). Am. J. Physiol. Renal Physiol. 288:F969–F981.
24. Mackenzie, B., D. D. Loo, M. Panayotova-Heiermann, and E. M. Wright. 1996. Biophysical characteristics of the pig kidney Na+/glucose cotransporter SGLT2 reveal a common mechanism for SGLT1 and SGLT2. J. Biol. Chem. 271:32678–32683.
25. Parent, L., S. Supplisson, D. D. Loo, and E. M. Wright. 1992. Electrogenic properties of the cloned Na+/glucose cotransporter: I. Voltage-clamp studies. J. Membr. Biol. 125:49–62.[Medline]
26. Gagnon, D. G., P. Bissonnette, and J. Y. Lapointe. 2006. Identification of a disulfide bridge linking the fourth and the seventh extracellular loops of the Na+/glucose cotransporter. J. Gen. Physiol. 127:145–158.
27. Bissonnette, P., J. Noel, M. J. Coady, and J. Y. Lapointe. 1999. Functional expression of tagged human Na+-glucose cotransporter in Xenopus laevis oocytes. J. Physiol. 520:359–371.
28. Eskandari, S., E. M. Wright, and D. D. Loo. 2005. Kinetics of the reverse mode of the Na+/glucose cotransporter. J. Membr. Biol. 204:23–32.[CrossRef][Medline]
29. Sauer, G. A., G. Nagel, H. Koepsell, E. Bamberg, and K. Hartung. 2000. Voltage and substrate dependence of the inverse transport mode of the rabbit Na(+)/glucose cotransporter (SGLT1). FEBS Lett. 469:98–100.[CrossRef][Medline]
30. Charron, F. M., M. G. Blanchard, and J. Y. Lapointe. 2006. Intracellular hypertonicity is responsible for water flux associated with Na+/glucose cotransport. Biophys. J. 90:3546–3554.
31. Chen, X. Z., M. J. Coady, F. Jackson, A. Berteloot, and J. Y. Lapointe. 1995. Thermodynamic determination of the Na+: glucose coupling ratio for the human SGLT1 cotransporter. Biophys. J. 69:2405–2414.
| ||||||||||||||||||||||||||||||
0.05; **, P 
