(Received for publication, June 26, 1996, and in revised form, January 15, 1997)
From the Renal Division, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, and Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, Massachusetts 02115 and the ¶ Yale University, Department of Cellular and Molecular Physiology, School of Medicine, New Haven, Connecticut 06510
Ion-coupled solute transporters exhibit
pre-steady-tate currents that resemble those of
voltage-dependent ion channels. These currents were assumed
to be mostly due to binding and dissociation of the coupling ion near
the extracellular transporter surface. Little attention was given to
analogous events that may occur at the intracellular surface. To
address this issue, we performed voltage clamp studies of
Xenopus oocytes expressing the intestinal H+-coupled peptide cotransporter PepT1 and recorded the
dependence of transient charge movements in the absence of peptide
substrate on changing intra- (pHi) and extracellular pH
(pHo). Rapid steps in membrane potential induced transient
charge movements that showed a marked dependence on pHi and
pHo. At a pHo of 7.0 and a holding potential
(Vh) of 50 mV, the charge movements were mostly inwardly
directed, whereas reduction of pHo to below 7.0 resulted in
outwardly directed charge movements. When pHi was reduced,
inwardly directed charge movements were observed. The data on the
voltage dependence of the transient charge movements were fitted by the
Boltzmann equation, yielding an apparent valence of 0.65 ± 0.03 (n = 7). The midpoint voltage (V0.5) of
the charge distribution shifted linearly as a function of
pHi and pHo. Our results indicate that, as a
first approximation, the magnitude and polarity of the transient charge
movements depend upon the prevailing H+ electrochemical
gradient. We propose that PepT1 has a single proton binding site that
is symmetrically accessible from both sides of the membrane and that
decreasing the H+ chemical potential (
µH)
or increasing the membrane potential (Vm)
shifts this binding site from an outwardly to an inwardly facing
occluded state. This concept constitutes an important extension of
previous kinetic models of ion-coupled solute transporters by including
a more detailed description of intracellular events.
Several ion-coupled solute transporters have been shown to
exhibit pre-steady-state currents in response to step changes in membrane potential. These currents were found to be analogous to gating
currents of voltage-dependent ion channels (1-5; 12). Based on Xenopus oocyte expression studies of
Na+/glucose cotransporters, Wright and coworkers (3, 22)
proposed that they are due to binding and dissociation of
Na+ near the extracellular transporter surface in
conjunction with a slow voltage-sensitive conformational change that
leads to translocation of the charged empty carrier within the plane of
the membrane before binding or after dissociation of Na+
(see Fig. 6A). The investigators reported that analogous
events occur in the human H+-coupled oligopeptide
cotransporter PepT1 and that transient charge movements can be
attributed to the reorientation of the empty carrier within the
membrane in response to H+ binding/dissociation (10). In
contrast, Lester and colleagues (6) described the kinetics and
transient charge movements of the GABA1
transporter GAT1, the 5-HT transporter, and the Na+-glucose
transporter SGLT1 by a model which treated the transporters as a single
file of binding sites with ends open to the external and internal
solutions. Unlike previous reports (see Refs. 1-5, 10, and 17), this
model does not consider conformational changes of the empty transport
protein in response to voltage jumps. Thus, although the findings on
the dynamic aspects of transporters have generated considerable
interest and have stimulated numerous studies, the interpretations of
the pre-steady-state currents vary, and the molecular events underlying
these events are poorly understood. In addition, the effect of changes
of the intracellular concentrations of the coupling ions
(i.e. Na+ or H+) on the
pre-steady-state kinetics have not yet been reported for an ion-coupled
cotransporter.
The intestinal brush border oligopeptide cotransporter PepT1 is the prototype of a new family of mammalian H+-coupled transporters. This transporter was previously cloned in our laboratory using an expression cloning approach. PepT1 mediates electrogenic transport of di-, tri-, and tetrapeptides into epithelial cells and has a coupling ratio of 1:1 for neutral peptides (7, 23; and for review, see Refs. 13 and 16).
In the present study, we have analyzed the pre-steady-state currents displayed by rabbit PepT1 and have determined their dependence on intra- and extracellular pH. Our data provide new insights into the H+-coupling mechanism of PepT1 and show a striking symmetry of the pre-steady-state characteristics obtained in response to either intra- or extracellular acidification.
The rabbit intestinal oligopeptide transporter PepT1, was expressed in Xenopus laevis oocytes by microinjection of transporter cRNA as described previously (7). For intracellular acidification using the human high affinity glutamate transporter EAAC1 (9), PepT1 cRNA was coinjected into oocytes with EAAC1 cRNA at a ratio of 1:1 (w/w). Control oocytes were injected with 50 nl of deionized water. Experiments were performed 3 to 7 days after injection.
Voltage Jump ExperimentsTransporter currents were measured
by using the two-electrode voltage clamp technique (GenClamp 500, Axon
Instruments, Foster City, CA) as described previously (7, 9). The
resistance of the microelectrodes was 0.3-0.6 M. The standard bath
solution contained 100 mM NaCl, 2 mM KCl, 1 mM MgCl2, 1 mM CaCl2, 3 mM Hepes, and 3 mM Mes. The pH was adjusted to
the value of interest with Tris buffer. To lower the intracellular pH
(pHi) of the oocytes, PepT1 cRNA-injected oocytes were
incubated either in standard bath solution (see above) in the presence
of either 2 mM Gly-Leu or 2 mM Trp-Gly or in
acetate buffer (30 mM sodium acetate, 58 mM
NaCl, 5 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, 5 mM Hepes, 5 mM] Mes, pHo 6.0). Alternatively, oocytes
expressing PepT1 and EAAC1 were incubated in standard bath solution in
the presence of 2 mM L-glutamate. Intracellular
pHi changes were monitored using pH-sensitive
microelectrodes filled with a hydrogen-selective ionophore (Fluka
Chemie AG, ionophore I, coctail B) as described previously (7, 9). To
measure charge movements, oocytes were clamped at a holding potential
(Vh) of
50 mV, and step changes in membrane
potential to transient voltages (Vt) were
applied for 250 ms followed by a 1-s interpulse interval using the
pCLAMP software package (Version 6.0, Axon Instruments, CA). Analog
signals were averaged from three sweeps, low pass filtered at 5 kHz,
digitized, and saved for further processing.
To analyze the pre-steady-state currents, the current traces were
fitted to I(t) = I1
exp(t/
1) + I2
exp(
t/
2) + Iss;
where I1 is a capacitive current with time
constant
1 associated with the oocyte plasma membrane
(
1 is also observed in non-injected control oocytes),
I2 is a transient current associated with PepT1 expression with time constant
2, and
Iss is the steady-state current. The parameters
1,
2, I1,
I2, and Iss were allowed
to vary in all fits. The fractional error using a Simplex algorithm was
set to 0.0001 (pClamp software package, Version 6.0, Axon Instruments).
The transient charge movements Q were obtained from the time
integrals of Itransient(t) = I2 exp(
t/
2) during
the ON and OFF responses for all depolarizing and hyperpolarizing potentials and fitted by the Boltzmann equation, Q = Qmax/(1 + exp((V
V0.5)zF/RT) + Qhyp; where Qmax
represents the total charge movement, z represents the
effective valence, V0.5 represents the midpoint
of the charge distribution, and Qhyp represents
the charge movement for extreme hyperpolarizing potentials (for
reviews, see Refs. 1-6 and 17). R, T, and
F are the usual thermodynamic constants. In our
experiments RT/F was 25.3 mV.
Pre-steady-state currents of rabbit PepT1
expressed in Xenopus oocytes were recorded as a function of
pHo (Fig. 1A). These experiments
were performed in the absence of peptides to suppress the "normal"
transport cycle of PepT1 (step 3 in Fig. 6B) and to isolate
the partial reactions that account for the pre-steady-state currents.
Oocytes were voltage clamped at a holding potential (Vh) of 50 mV, and pre-steady-state currents
were induced by a series of test voltage pulses
(Vt) (ON response). At pHo 7.0, stepping the oocyte membrane potential from Vh
to hyperpolarizing potentials (Vt =
150,
110, and
70 mV) resulted in large inwardly directed transient
currents, whereas for depolarizing pulses (Vt =
30, +10, and +50 mV) only small outwardly directed currents were
observed. Virtually the opposite observations were made when pHo was less than 6.0 where we measured relatively large
outwardly directed pre-steady-state currents in response to
depolarizing pulses. Stepping the membrane potential back to
Vh (OFF response) produced opposite transient
currents that were identical in magnitude to those obtained for the ON
response (see below).
Addition of substrate completely abolished the transient currents of PepT1 for pHo between pH 5.5 and 7.5 and produced large inwardly directed steady-state currents that increased with hyperpolarization of the membrane. Fig. 1B shows a typical current recording in the presence of saturating Gly-Leu concentration (2 mM) in response to the voltage steps at pHo 7.0. Water-injected control oocytes (Fig. 1C) exhibited capacitive currents that were symmetrical with respect to hyper- and depolarization but did not reveal significant transient currents when compared with those depicted in Fig. 1A. Consistent with our previous current recordings for Gly-Sar (7), the steady-state I/V relationship for Gly-Leu taken at pHo 7.0 was membrane potential-dependent (Fig. 1D).
Relaxation time constants of pre-steady-state currents were obtained by
fitting the pre-steady-state currents to a steady-state component and
two exponential functions. Fig. 1E shows the transient currents displayed in Fig. 1A in response to depolarizing
potentials at pHo 6.0 on a semilogarithmic scale and
indicates that the data can be adequately described by two
exponentials. The exponential functions represented the contributions
of a transient capacitive component associated with the oocyte plasma
membrane and a transient component associated with expression of PepT1.
The time constant 1, which reflects the oocyte membrane
capacitance, was insensitive to voltage and external pHo
(Fig. 2A) but varied with the size of the
oocytes. For example, relatively small PepT1 cRNA or water-injected control oocytes gave
1 values of 0.72 ± 0.06 ms
(mean ± S.D.; n = 90, number of oocytes). By
contrast, the largest oocytes gave
1 values of 1.35 ± 0.09 ms (n = 72) (data not shown).
For the ON response, the relationship between 2 and
Vm (Vt) was bell-shaped,
and the maximum of
2 (
2max) increased
with decreasing pHo (Fig. 2A), consistent with
data of Mackenzie et al. (10) on human PepT1. At
Vm = +50 mV,
2 increased from 5 to 15 ms (n = 5) when pHo was reduced from
pH 7.5 to pH 6.0. The Vm at which
2 was maximal moved in the depolarizing direction with
decreasing pHo, consistent with the view that reducing
pHo promotes H+ binding to the extracellular
transporter surface, resulting in a shift toward the "outwardly
occluded state" (Fig. 6B). Thus, after reduction of
pHo, a more depolarizing potential is required to maintain
the transporter at
2max condition. At hyperpolarizing potentials (e.g.
120 mV),
2max was much
smaller, and the
2 values were less dependent on
pHo (Fig. 2A). However, as will be shown later,
2max at negative potentials goes back up again after
intracellular acidification (Fig. 3B).
Analysis of the transient currents during the OFF response exhibited
relaxation constants 2 of 10-15 ms that were relatively insensitive to Vt and pHo (data not
shown). These data are consistent with the findings on human PepT1
(10), Na+/glucose cotransporters (3, 22), and the GABA
transporter, GAT1 (4). The Vt insensitivity
during the OFF response is analogous to the Vh
insensitivity of the transient currents during the ON response.
To further study the pHo-dependence of the pre-steady-state behavior of PepT1, we calculated the charge transfer Q for the ON and OFF responses from the time integrals of the transient currents. The charge movements were virtually identical for ON and OFF responses at all values of Vt and pHo (Fig. 2B), indicating strict charge conservation, a property that is in agreement with studies of Na+-coupled transporters (1, 3-5, 17). When plotted against Vt, Q saturated with both hyper- and depolarization of the membrane (data not shown).
Fig. 2C shows the voltage- and
pHo-dependence of the transient charge movements. The data
were fitted by a single Boltzmann distribution and represent the
average of five oocytes. To correct for different levels of PepT1
expression in the oocyte plasma membrane, the data were normalized
against the maximal charge movement Qmax. The
slope z, which provides a measure of the equivalent charge
moving within the membrane electric field in response to voltage jumps,
shifted from 0.62 to 0.53 when changing pHo from 5.5 to
7.5. The midpoint potential (V0.5) of the
Boltzmann equation shifted from approximately +60 to 63 mV when
changing pHo from 5.5 to 7.5 and was linearly dependent on
pHo (Fig. 2D). V0.5
changed by
61 mV (r2 = 0.99, n = 5) per pHo unit. The slope of this relationship, however,
varied between different batches of oocytes. In another set of
PepT1-injected oocytes V0.5 changed by
38 mV
per pHo unit (r2 = 0.92, n = 7; Fig. 5). This variation may be
related to different intracellular ion concentrations in different
batches of oocytes. Comparison between the V0.5
values of rabbit and human PepT1 (10) revealed marked differences at
similar pHo values. This may be due to differences in
pHi (see below) or species variations as exhibited by
rabbit and human SGLT1 (19) or wild-type and mutant SGLT1 (20). Taken
together, our data suggest that V0.5 roughly follows the Nernst potential for protons, which is
58 mV per pH unit.
Thus, as a first approximation, we hypothesize that
V0.5 depends on the electrochemical potential
for protons.
As expected, Qmax increased with the expression
levels of PepT1 (Fig. 2E) and with the oocyte size. The
latter became clear when we plotted Qmax
versus 1. The
1 value,
which is proportional to the membrane capacitance and thus to the
oocyte surface area, was linearly dependent on
Qmax (data not shown). Three to five days after
injection of PepT1 cRNA, analysis of transient currents of 10 oocytes
revealed a median value for the total transient charge movement of
48 ± 7.0 nCi. This value is 4 to 10-fold higher than that
determined for human and rat SGLT1, human EAAT2, and human PepT1 (3, 5,
10) but is comparable with previous estimates for rabbit SGLT1 (19, 20)
and the rat GABA transporter GAT1 (4). As Fig. 2E shows, one
oocyte displayed an exceptionally high Qmax
value of
140 nC. The corresponding
1 value of that oocyte was also high (1.37 ms), indicating that the high charge movement was due to a larger number of transporters expressed over a
large oocyte surface area. Given that Qmax is
proportional to the carrier density N in the oocyte membrane
according to Qmax = N z e, where
z is the effective valence of the moving charge (see below)
and e is the elementary charge, we estimate an average expression level of 4.6 × 1011 PepT1 transporters.
This level is on the same order of magnitude as that of other solute
transporters (rabbit SGLT1, rat GAT1, and human PepT1;
~1011 transporters) (10, 19, 20). It is worth noting that
this high estimate was the result of both relatively high charge
movements and small z values (see below).
The turnover rate of PepT1 is the slope of the Qmax
versus Imax plot
(Imax was obtained at 70 mV at saturating
peptide concentration (2 mM Gly-Leu, Fig. 2E))
and was 6.6 s
1 (r2 = 0.83) based
on data from 12 oocytes. The turnover rate is thus similar to that of
the GABA transporter GAT1 (4), the glutamate transporter EAAT2 (5), and
other Na+-coupled solute transporters (3, 10). The turnover
rate of human PepT1 (10) was about 10-15-fold greater than the 6.6 s
1 value we obtained for the rabbit cotransporter,
probably because the human data were obtained at a
Vh of
150 mV (10).
PepT1-injected oocytes were voltage clamped at
50 mV, and pre-steady-state currents were recorded before and after
intracellular acidification. Intracellular acidification was achieved
using three different protocols: 1) a 60-min pre-incubation of
PepT1-injected oocytes with 2 mM Gly-Leu (pHo
5.5), 2) a 10-min incubation of the oocytes in acetate buffer, and 3) a
10-min incubation with glutamate using oocytes expressing both PepT1
and EAAC1. Pre-steady-state currents were recorded at an extracellular
pHo of 7.0. At this pHo, the inwardly and
outwardly directed pre-steady-state currents were symmetrical upon
depolarization or hyperpolarization of the membrane so that changes in
their direction could be easily detected by loss of symmetry of the
current traces.
Typical recordings before and after intracellular acidification are
shown in Fig. 3A. Here, intracellular acidification was achieved by PepT1-mediated H+-coupled uptake of Gly-Leu
under voltage clamp conditions. The decrease of pHi was
monitored simultaneously on the same batch of PepT1-injected oocytes
using pH-sensitive microelectrodes (Fig. 3D). Because
intracellular acidification in the latter experiments was measured
under non-clamping conditions, and because application of Gly-Leu
elicited a membrane depolarization (data not shown) and the Gly-Leu
elicited current is voltage-dependent (Fig. 1D), the
reduction of pHi under clamping condition at the resting
membrane potential (50 mV) would be somewhat greater.
Our data show that the current relaxations become slower after
intracellular acidification with Gly-Leu (pHo = 7.0), with 2 increasing from 10 to 28 ms (n = 4;
Fig. 3B). This increase in
2 was partially
reversed when the oocytes were incubated for an additional 10 min in
pHo 8.0 bath solution at
50 mV, to allow pHi
to recover partially (Fig. 3B). Recovery of
2
was further enhanced if oocytes were incubated for 1 h in
pHo 8.0 bath solution at +10 mV (data not shown).
Measurement of pHi using pH-sensitive microelectrodes
showed that the recovery of
2 is associated with a
partial recovery of pHi (Fig. 3D). The increase
in
2 is consistent with our model that reducing
pHi increases H+ binding to the intracellular
transporter surface and thereby shifts the carrier toward an inwardly
occluded state (Fig. 6B). In further agreement, the maximum
of
2 shifted to more negative potentials when
pHi was decreased because a more negative
Vm is required to dissociate H+.
To rule out the possibility that the changes in the relaxation kinetics of PepT1 are due to accumulation of dipetides in the oocytes, control experiments were performed with Trp-Gly (2 mM). We recently showed that Trp-Gly taken up by PepT1 in oocytes is rapidly hydrolyzed into Trp and Gly (14). Incubation with Trp-Gly had the same effect on PepT1 pre-steady-state kinetics as Gly-Leu (data not shown).
To further test whether the changes in the relaxation kinetics are due
to intracellular acidification, we incubated oocytes in 30 mM acetate buffer to reduce pHi. Under
non-clamping conditions, application of 30 mM acetate
reduced the intracellular pHi by ~0.2 pH units (Fig.
3E) without changing the resting membrane potential, which
was around 50 mV (data not shown). The effect of acetate on the
relaxation characteristics (Fig. 3C) and on the recovery of
pHi after a 20 min incubation in pH 8.0 bath solution at
50 mV was similar to that obtained by incubation with peptide, except
that the acetate procedure was accompanied by more pronounced recovery
in
2 values (Fig. 3C).
The third line of evidence in support of our hypothesis that the
relaxation kinetics of PepT1 depend on pHi was derived from co-expression studies of PepT1 and the neuronal and epithelial high
affinity glutamate transporter EAAC1 (9, 13) (Fig. 4A). Because EAAC1-mediated uptake of L-glutamate is coupled to
the cotransport of one H+ (or alternatively to the
countertransport of one OH) uptake of
L-glutamate causes intracellular acidification. Previously, we demonstrated that EAAC1-mediated uptake of glutamate decreased pHi from ~7.6 to 7.5 (9). Because EAAC1 per se
does not produce large pre-steady-state currents in response to voltage
jumps (9), it can be assumed that co-expression of EAAC1 with PepT1
does not interfere with the pre-steady-state kinetics of PepT1. Thus, any observed changes of the pre-steady-state currents of oocytes expressing both EAAC1 and PepT1 can be mainly attributed to PepT1 expression. Transient currents displayed by these oocytes were analyzed
before and after a 10-min incubation with L-glutamate (2 mM, Vh =
100 mV). The
pre-steady-state currents recorded at pHo 7.0 in the
absence of peptide and L-glutamate were both inwardly and
outwardly directed, similar to the symmetrical currents shown in Fig.
3A, top. Pre-incubation with
L-glutamate resulted in predominantly inwardly directed
pre-steady-state currents at Vh =
50 mV,
analogous to the data shown in Fig. 3A, bottom.
Analysis of the pre-steady-state currents revealed that the dependence
of
2 on membrane potential (Fig. 4B) was
consistent with those shown in Fig. 3, B and C. Analysis of charge movements at pHo 7.0 (Fig.
5) revealed that V0.5 shifted by
17.6 mV in response to EAAC1-mediated intracellular acidification
(from
30.1 ± 6.9 mV (n = 3) to
47.7 ± 1.6 mV (n = 3)), which is in line with our hypothesis
that V0.5 roughly follows the electrochemical
potential for protons.
When we
analyzed the pHo dependence of the transient charge
movements before and after intracellular acidification, either by
incubation of oocytes in Gly-Leu or acetate, we found nearly the same
linear dependence of V0.5 on pHo
(Fig. 5). The apparent valence z, moving in response to the
voltage steps, was 0.65 ± 0.03 (S.E., n = 7) and
did not change significantly with reduction of pHi. The
slopes of the pHo dependence of V0.5
for Gly-Leu and acetate-treated oocytes were 49 mV
(r2 = 0.99, n = 4) and
60 mV
(r2 = 0.99, n = 3) per
pHo unit, respectively. The line representing the linear
relationship between V0.5 and pHo shifted
toward more negative potentials in response to intracellular
acidification. This suggests a qualitative equivalence between raising
pHo and lowering pHi, in agreement with the
hypothesis that V0.5 is proportional to
µH, and thus to the pHi
pHo
difference. The observed effects of pHi and pHo
also are consistent with the
2/voltage relationships (Figs. 2A and 3, B and C), where a
decrease of the inwardly directed H+ gradient through
manipulation of either pHi or pHo results in a
shift of the
2 maximum toward more negative potentials.
Thus, our data indicate that there is a fixed relationship between
µH, Vm, and
2.
Several studies on Na+- and H+-coupled
solute transporters indicated that varying the external ion
concentration affects their pre-steady-state kinetics, shifting the
voltage dependence of the transient currents (1-5, 10). Our data show
that pHo affects the pre-steady-state kinetics of rabbit
PepT1 in a similar way, with the charge movements dependent on
pHo. The shift of the maximal time constant
2max toward more negative potentials in
response to increasing pHo is in agreement with studies of
Hilgemann and coworkers (2, 11) addressing the charge movements of the Na+,K+-pump. These investigators showed that
the Vm dependence of the maximal capacitance, which is proportional to
2max, shifts toward more negative potentials when the
extracellular Na+ concentration is lowered.
A new aspect of our study is the determination of the effect of changes
of the intracellular pH on the pre-steady-state kinetics. To our
knowledge, the effect of intracellular coupling ion-concentration on
the pre-steady-state kinetics has not yet been determined for an
ion-coupled solute transporter. In the present investigation, we show
that the PepT1-elicited pre-steady-state currents depend on both
pHo and pHi. The effect of increasing
pHo on the voltage dependence of the transient charge
movement was equivalent to that of decreasing pHi. We found
that either increasing pHo or decreasing pHi
shifts the 2max of the pre-steady-state currents toward
more negative potentials. Our data, therefore, indicate that an
important determinant of the polarity of the pre-steady-state kinetics
of PepT1 is the electrochemical potential for protons across the oocyte
plasma membrane.
An important question is how
the H+ electrochemical ion gradient affects the direction
and amplitude of the transient charge movements of PepT1. A mechanistic
concept of transient charge movements in ion-coupled transporters was
proposed by Wright and coworkers for the Na+-coupled
glucose transporter SGLT1 (3). This concept was additionally proposed
to account for the pre-steady-state currents observed for human PepT1
(10, 18) (Fig. 6A). The investigators
proposed that the pre-steady-state charge movements recorded in the
absence of peptide substrate were mostly due to reorientation of the
empty carrier in the membrane following binding/dissociation of
H+ (Fig. 6A, see voltage-sensitive
conformational change). They predicted that the empty carrier has a
negative charge (Fig. 6A, ) which moves inwardly within
the membrane following H+ dissociation and that this
movement is associated with a structural change of the empty carrier
from an outwardly to an inwardly facing conformation. The negative
charge is proposed to represent the center of movable, positive, or
negative charges on the carrier. However, the molecular basis of the
charge movements has not yet been elucidated, and it is currently
unknown which amino acid residues in PepT1 or SGLT1 contribute to
charge movement.
Since the above model does not consider the affect of intracellular pH
on the pre-steady-state kinetics of PepT1, we have developed a revised
model in which the empty carrier has a single H+-binding
site that is symmetrically accessible from both sides of the membrane
(Fig. 6B). We hypothesize that H+ binding to the
extra- or intracellular surface is followed by a conformational change
resulting in the formation of outwardly or inwardly facing occluded
states, respectively, and that these protonated states cannot cycle
between the inside or outside in the absence of peptide. The
pre-steady-state currents in response to voltage jumps are predicted to
be due to binding or dissociation of H+ within the membrane
electric field near the extra- or intracellular transporter surface
(steps 2/1 and 4/5) and to reorientation of the empty carrier during
the voltage sensitive conformational change (step 6). Increasing the
H+ chemical gradient (µH) and/or
decreasing Vm (hyperpolarization) are predicted
to shift the carrier toward the outwardly occluded state. Similarly,
decreasing the H+ chemical gradient (
µH)
and/or increasing Vm (depolarization) are
predicted to shift the carrier toward the inwardly occluded state.
Thus, consistent with our findings, the model predicts that the
magnitude and polarity of the transient charge movements depend on the
H+ electrochemical gradient.
According to our model, the 2
values for dissociation/association of H+ at the
extracellular surface depend on the rate constants of steps 2, 1, and
6, and those for dissociation/association at the intracellular surface
on the rate constants of steps 4, 5, and 6. Since the transient
currents occur in the millisecond range (Fig. 2a), whereas binding and
dissociation of ions into or out of an "ion well" near the membrane
surface more likely occur in the microsecond range (2), steps 1 and 5 are expected to contribute less to the
2 values. We
therefore propose that the
2 values are set by the slow
conformational changes (steps 2 and 4) which precede H+
dissociation at the extra or intracellular surface and/or the voltage-sensitive conformational change (step 6).
We propose that dissociation/association of H+ will predominantly occur at the extracellular surface if pHi of oocytes is not reduced by acidification (Figs. 1 and 2), whereas at elevated pHo and reduced pHi (Figs. 3 and 4), dissociation/association is predicted to be increased at the intracellular surface. According to our model, there will be a pHo/pHi combination for any given Vm in the physiological range that yields an equal population of inwardly and outwardly facing states, resulting in binding/dissociation of H+ at both sides of the membrane. If the voltage-sensitive conformational change (step 6) accounts for most of the charge movements, both intra- and extracellular events can be described by a single Boltzmann distribution. However, if steps 2/1 and 4/5 of the transport cycle are rate-limiting, the Q/Vm relationships probably represent the composite of at least two closely related Boltzmann distributions that exist at the extra- and intracellular surface.
Relationship ofAn important aspect of our work is that we
have extended the pre-steady-state kinetics described by Wright and
colleagues (10) by including H+-dissociation/binding events
at the intracellular transporter surface. In addition, we show that
intracellular acidification reverses the decrease in
2max after hyperpolarization (Figs. 3B and
4B versus Fig. 2A). This finding is in favor of
the view that an additional step may participate in the overall
pre-steady-state cycle. For example, if steps 4/5 become rate
determining when decreasing pHi, this would explain the
increase in
2max at negative potentials. However, to
fully understand the molecular basis of the pre-steady-state kinetics
of PepT1, further studies addressing the rate constants of individual
steps of the transport cycle and their dependences on the
H+ electrochemical gradient will be required.
PepT1 is predicted to bind H+ first and then peptide in an orderly fashion, consistent with studies of Mackenzie et al. (10) who showed that the apparent maximal transport rate of human PepT1 for Gly-Sar is not significantly attenuated when Ho+ is reduced 10-fold. At lower pHo, we predict that the number of PepT1 molecules that exist in the outwardly facing state is increased and that H+ binding results in a conformational change (Fig. 6B, step 2), leading to the outwardly occluded state and that this state has an increased affinity for peptides. From a physiological perspective, this concept makes sense because the pHo in the unstirred layer near the intestinal brush border membrane is acidic (pH 5.5-6.0), which favors the outwardly occluded state and promotes efficient peptide absorption (21, 24).
The Multi-substrate Single File ModelIn some respects, our kinetic model resembles the multi-substrate single file model recently proposed by Lester and coworkers (6) to account for the pre-steady-state kinetics of ion-coupled transporters. The model treats an ion-coupled transporter as a single file of ion binding sites, with ends open to the external and internal solutions. It predicts that, once an ion is within the transporter, it is allowed to move between sites and to the solutions on either end of the transporter. A state of the transporter is defined by the occupancy of multiple binding sites or wells by ions in the permeation pore. The kinetics of the ion-coupled transporter are then reconstructed by possible transitions between different states in response to voltage jumps. Steps 2, 1, 6, 5, and 4 of our model (Fig. 6B) could be considered as sites in the single-file model. However, PepT1 is predicted to have only a single ion binding site that is either inwardly or outwardly facing, depending on intra- and extracellular pH.
PepT1 as a Potential pH BiosensorSince the magnitude and
polarity of the transient currents depend on the proton chemical
gradient µH, we examined whether this relationship can
be used to estimate changes in pHi by analysis of the
pre-steady-state currents of PepT1. The midpoint potential V0.5 can be determined according to
(pHi
pHo) =
*
V0.5, where
represents a conversion factor
(0.023 pH units/mV) that roughly corresponds to
F/RT.
is given by the slope of the
relationship V0.5 to pHo (Fig. 5).
In support of our finding that V0.5 follows
µH, the slopes for different pHi values
were similar. We estimated the decrease in pHi induced by
EAAC1-mediated H+-dependent glutamate uptake by
coexpression of EAAC1 and PepT1 (Fig. 5) and compared the result with
pHi-measurements obtained using pH microelectrodes. In
oocytes injected with both EAAC1 and PepT1, we observed a
glutamate-induced shift in V0.5 of
17.6 mV at
pH0 = 7.0 (Fig. 5, closed triangles). From the above
equation, the observed shift in V0.5 predicts an
intracellular acidification of 0.4 pH units. Previous measurements
using pH-sensitive microelectrodes gave a shift of ~0.1 pH units (9).
The difference may be related to monitoring pHi close to
the plasma membrane (PepT1) as opposed to measuring it deep in the
cytoplasm (microelectrodes). Thus, our data raise the possibility that
PepT1 can be used as a biosensor to monitor pHi changes in
close vicinity of the plasma membrane.
A challenging question that remains to be addressed is whether the symmetry of binding of the coupling ion binding to either side of PepT1 also pertains to Na+-coupled transporters. Increasing [Na+]i in oocytes expressing SGLT1 would be expected to have a similar effect as intracelluar acidification on PepT1. The effect of changing [Na+]i on the pre-steady-state kinetics of a Na+-coupled solute transporter have not yet been reported.
We thank Sela Mager (California Institute of Technology, Pasadena, CA) and Dr. Bryan Mackenzie for stimulating discussions.