Université Pierre et Marie Curie, Institut National
de la Santé et de la Recherche Médicale Unité
356, Institut Fédératif de Recherche 58, Hôpital
Européen Georges Pompidou, Assistance Publique-Hôpitaux
de Paris, 75015 Paris, France
medullary thick ascending limb; bicarbonate; in vitro
microperfusion; intracellular pH
 |
INTRODUCTION |
TUBULAR REABSORPTION OF
FILTERED bicarbonate (HCO
) is a major function
of the kidney. Under normal conditions, all the filtered
HCO is actively reabsorbed by the renal tubule,
thereby preventing renal HCO loss and metabolic
acidosis. The main site of renal HCO reabsorption is
the proximal tubule, where the mechanisms of apical and basolateral HCO transport are reasonably well known (for review
see Ref. 1). The second most important tubular segment for
HCO reclamation is the thick ascending limb of the
loop of Henle (TALH), which reabsorbs ~15% of the filtered
HCO (13). In addition, in the medulla,
the reabsorbed HCO is delivered to the
interstitium, characterized by a low blood flow; therefore, the
reabsorption of HCO by the medullary TALH (MTALH) is
a critical determinant of interstitial pH that controls secretion of
protons by the neighboring medullary collecting duct (15).
The apical step of HCO entrance into the MTALH cell
relies on proton secretion. Two different mechanisms of proton
secretion have been described in these cells: an electroneutral
Na+/H+ exchange and an electrogenic,
ATP-dependent proton pump (11). However, only the former
seems to be physiologically significant under the various conditions
tested. HCO reabsorption is almost completely
inhibited in the absence of luminal Na+ and/or in the
presence of luminal amiloride (16). Finally, the
expression and activity of the apical Na+/H+
exchanger is enhanced during metabolic acidosis and depressed during
Cl
-depleted metabolic alkalosis (26, 27),
conditions under which MTALH HCO reabsorption is
increased or decreased, respectively (12, 14).
In contrast to the apical mechanisms involved in proton secretion, the
mechanisms involved in HCO exit from the MTALH cells
under physiological conditions are unsettled. However, several
HCO exit mechanisms have been successively
suggested. Previous experiments examining HCO exit
mechanisms have been mainly conducted with MTALH suspensions.
4,4'-Diisothiocyanostilbene-2,2'-disulfonic acid (DIDS)-sensitive
Na+-(HCO)n
and Ba2+-insensitive electroneutral
K+-HCO cotransports have been described in mouse and rat MTALH suspensions, respectively, and no evidence for
the presence of Cl/HCO exchange has
been found in mouse and rat MTALH suspensions (23, 28).
However, these data have recently been challenged by the results of a
study using purified basolateral membrane vesicles from rat MTALH cells
(29). A weak Na+-(HCO)n
cotransport activity and no electroneutral
K+-HCO cotransport activity were
observed in this preparation. In contrast, an electroneutral
Cl/HCO exchange activity was apparent under isosmotic conditions and in the absence of arginine vasopressin in basolateral membrane vesicles from rat MTALH cells
(10). In addition, two distinct isoforms of
Cl/HCO exchangers were reported in
this preparation (10). These results are in keeping with
those recently obtained by Sun (37) in microperfused mouse
MTALH, demonstrating a basolateral
Cl/HCO exchange activity under
baseline conditions.
Because a growing body of evidence indicates that tubule suspensions do
not appear to be suitable for identification of membrane transporters
(10, 29, 32) and because a transporter itself may be
altered and/or a regulatory factor may be lost during the preparation
of membrane vesicles, the present study used microperfused MTALH to
identify the mechanisms physiologically involved in
HCO transport across the basolateral membrane of
intact rat MTALH cells. Here, we demonstrate the presence and the
functionality, under isosmotic conditions, of an electroneutral
Cl/HCO exchange, a DIDS- and
ethylisopropylamiloride (EIPA)-insensitive
Na+-HCO cotransport, and an
electroneutral Ba2+-sensitive
K+-HCO cotransport.
| |
MATERIALS AND METHODS |
Tubule isolation and perfusion.
Pathogen-free male Sprague-Dawley rats (60-75 g body wt; Iffa
Credo) had free access to standard rat chow and distilled water until
the time of experiments. Animals were injected intraperitoneally with 2 mg of furosemide 10 min before they were anesthetized with pentobarbital sodium (50 mg/kg body wt). Both kidneys were cooled in
situ with control bath solution for 1 min and then removed and cut into
thin coronal slices for tubule dissection. MTALH were dissected from
the inner stripe of the outer medulla at 4°C in the control solution
of the respective experiment. The isolated tubule was transferred to
the bath chamber on the stage of an inverted microscope (Axiovert 100, Carl Zeiss), mounted on concentric pipettes, and perfused in vitro. To
prevent motion of the tubule during intracellular pH (pHi)
measurement, the average tubule length exposed to bath fluid was
limited to 300-350 µm.
Solution composition.
The composition of the various solutions is given in Table
1. In nominally
Na+-free solutions, Na+ was isosmotically
replaced with N-methyl-D-glucamine
(NMDG+). Cl-free solutions contained
gluconate as a replacement for Cl. We compensated for
Ca2+ chelated by Cl substitutes by increasing
total Ca2+ concentration from 2 to 7.5 mM in
Cl-free solutions. The solutions containing
HCO were continuously gassed with 95%
O2-5% CO2 at 37°C, and the
CO2/HCO-free solutions were
continuously gassed with 100% O2 passed through a 3 N KOH
CO2 trap. Before each experiment, Na+,
K+, and Cl concentrations, osmolality, and pH
were measured in bulk solutions.
Alanine and HEPES were obtained from Research Organics
(Cleveland, OH), DIDS from Acros Organics, and
2',7'-bis(2-carboxyethyl)-5(6)-carboxyfluorescein (BCECF)
acetoxymethyl ester from Molecular Probes (Eugene, OR). 5-Nitro-2-(3-phenylpropylamino)benzoate (NPPB),
[(dihydroindenyl)oxy]alkanoic acid (DIOA), and all other chemicals
were obtained from Sigma (Saint-Quentin Fallavier, France).
pHi measurement.
MTALH cells were loaded with the fluorescent probe BCECF, prepared as a
20 mM stock in DMSO, by exposing the cells for ~20 min at room
temperature to the control bath solution containing 20 µM BCECF.
Loading was continued until the fluorescence intensity at 440-nm
excitation was at least one order of magnitude higher than background
fluorescence. The loading solution was then washed out by initiation of
bath flow, and the tubule was equilibrated with dye-free control bath
solution for 5-10 min. The bath solution was delivered at 20 ml/min and warmed to 37 ± 0.5°C by a water jacket immediately
upstream to the chamber. The perfusion rate was adjusted by hydrostatic
pressure to ~20 nl/min to prevent axial changes in composition of the
luminal fluid.
Intracellular dye was excited alternately at 490 and 440 nm
with a 75-W xenon lamp and a computer-controlled chopper assembly. Emitted light was collected through a dichroic mirror, passed through a
530-nm filter, and focused onto a charge-coupled device camera (model
ICCD 2525F, Videoscope International) connected to a computer. The
measured light intensities were digitized with eight-bit precision (256 gray level scale) for further analysis. For each tubule, a region of
interest was drawn, and the mean gray level for each excitation
wavelength was calculated with Starwise Fluo software (Imstar, Paris,
France). Background fluorescence was subtracted from fluorescence
intensity at each excitation wavelength to obtain intensities of
intracellular fluorescence. The ratio of fluorescence at 490 nm to that
at 440 nm was used as an indicator of pHi.
Intracellular dye was calibrated at the end of each experiment using
the high-K+-nigericin technique. Tubules were perfused and
bathed with a HEPES-buffered, 95 mM K+ solution containing
10 µM nigericin, a K+/H+ exchanger. Four
different calibration solutions, titrated to pH 6.5, 6.9, 7.3, or
7.5, were used.
Determination of intrinsic buffering capacity.
The intrinsic buffering capacity (
i) of MTALH
cells was determined using a method similar to that used by Watts and
Good (40), except a weak acid, instead of a weak base, was
used to change pHi and Ba2+ was omitted from
the solutions. To exclude HCO/CO2 as a
buffering component and block Na+-dependent pHi
regulatory mechanisms, Na+-free, HEPES-buffered solutions
containing 1 mM amiloride were used in the luminal perfusate and
basolateral bath. Addition of 20 mM acetate to the bath induced a
decrease in pHi. The dissociation constant of acetic acid
(pKa; 4.74) was used to calculate the intracellular acetate concentration when cell acidification reached a
plateau. The i was calculated as the ratio of the change
in intracellular acetate concentration to the change in
pHi; i ranged from 32.8 to 48.5 mM/pH unit
over the range of initial pHi observed in the experiments.
Calculation of base flux values.
Base flux (Jbase) values (in
pmol · min1 · mm1) were
calculated using the following equation:
Jbase = dpHi/dt × total × V, where dpHi/dt is the initial rate
of change in pHi (in pH units/min), total is
the buffering capacity (in mM · pH
unit1 · l1), and V is the cell
volume (in nl) per 1 mm of tubule length.
Values for dpHi/dt were determined by a
computer-assisted fitting of the pHi vs. time data to a
linear regression line over the first 15-30 s. In the absence of
CO2/HCO, total is
equal to i; in the presence of
CO2/HCO, total is
equal to i + 2.3 × [HCO]i, where
[HCO]i is intracellular
HCO concentration, calculated as follows:
[HCO]i = 0.03PCO2 × 
with the assumption that intracellular PCO2 equals extracellular
PCO2 (37 mmHg). V was calculated by determining
the volume of the tubule [(radius of tubule)2 × 
× length] and subtracting the volume of the lumen [(radius of
lumen)2 × × length] and was equal to 0.31 ± 0.01 nl/mm. A negative value of Jbase
indicates a net base efflux, and a positive value indicates a
net base influx.
Measurement of transepithelial voltage.
The transepithelial voltage (Vte) was measured
with a DP-301 differential electrometer (Warner) by use of a Ag-AgCl
electrode connected to the perfusion pipette via a 0.15 M NaCl or
sodium gluconate agar bridge, as appropriate. A 0.15 M NaCl or sodium gluconate agar bridge also connected the peritubular bath to an Ag-AgCl electrode.
Statistics and data analysis.
Values are means ± SE. Statistical significance was assessed by
paired or impaired Student's t-test, as appropriate.
P < 0.05 was considered significant.
| |
RESULTS |
A functional electroneutral Cl/HCO
exchange is present in the basolateral membrane of MTALH cells.
Because a Cl/HCO exchanger
electroneutrally exchanges Cl that enters the cell for
HCO that leaves the cell, the presence of a
Cl/HCO exchange activity in the
basolateral membrane of the MTALH cells is expected to yield the
following results (31, 37). 1) The isosmotic
removal of bath Cl should induce a reversible
intracellular alkalinization. 2) This effect should be
blunted in the absence of CO2/HCO buffer, as well as in the basolateral presence of an anion-exchange inhibitor such as DIDS. 3) The exchange should operate
independently of the presence of Na+ and changes in the
membrane voltage.
As shown in Fig. 1A and Table
2, in the bilateral presence of
Cl and in the presence of
CO2/HCO (extracellular pH 7.40),
pHi was stable at ~7.08 (segment a-b, Fig.
1A). The isosmotic removal of basolateral Cl
(solution A to solution B) induced a prompt and
significant intracellular alkalinization (segment b-c, Fig.
1A; Table 2). Conversely, readdition of bath
Cl (solution B to solution A)
elicited a rapid fall in pHi, which returned to baseline
values (segment c-d, Fig. 1A; Table 2).
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 1.
A: effect of basolateral Cl
removal on intracellular pH (pHi). Cells were initially
perfused and bathed with a standard solution buffered with
CO2/HCO (external pH 7.40) and
containing 121 mM Cl (segment a-b). Removing
Cl from the bath (replaced with gluconate) caused a large
and fast pHi increase (segment b-c). Returning
Cl to the bath elicited a prompt decrease of
pHi toward control values (segment c-d;
n = 5 tubules). B: effect of basolateral
Cl removal on pHi in the presence of 0.15 mM
4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS) in the bath.
Cells were initially perfused and bathed with a standard solution
buffered with CO2/HCO (external
pH 7.40) also containing 0.15 mM DIDS in the bath. In the bilateral
presence of 121 mM Cl, pHi values were
steadily increasing (segment a-b): the feature was clearly
different from that observed in the absence of basolateral DIDS.
Removing Cl from the bath (replaced with gluconate) in
the presence of DIDS caused small changes in pHi
(segment b-c; n = 6 tubules). C:
effect of basolateral Cl removal in the bilateral absence
of CO2/HCO and the presence of
0.1 mM acetazolamide to inhibit endogenous HCO
production on pHi. In the bilateral presence of 144 mM
Cl, pHi values were nearly constant
(segment a-b). Removing peritubular Cl
(segment b-c) and then returning bath Cl
(segment c-d) caused only small changes in pHi
(n = 6 tubules). D: effect of peritubular
Cl addition in the presence of external
CO2/HCO, voltage clamp, and
Na+ depletion on pHi. Tubules were initially
perfused and bathed in Na+-free solutions containing 120 mM
K+ and 10 µM valinomycin. Voltage clamp resulted in
intracellular alkalinization (segment a-b). Isosmotic
addition of Cl to the bath (Cl replacing
gluconate) elicited a decrease in pHi (segment
b-c) that was reversed by removal of peritubular Cl
(segment c-d).
|
|
In the presence of 0.15 mM bath DIDS and in the basolateral presence of
Cl, pHi was not stable but steadily
increasing (segment a-b, Fig. 1B), at variance
with data obtained in the absence of peritubular DIDS (Fig.
1A). This steadily increasing pHi was probably
accounted for by a DIDS-induced inhibition of basolateral
HCO exit pathway(s) under basal conditions.
Furthermore, the effect of basolateral Cl removal
(solution A to solution B) was significantly
blunted when tubules were studied in the presence of 0.15 mM
basolateral DIDS: the initial rate of alkalinization was reduced by
63% (P < 0.01), and the rise in pHi after
removal of basolateral Cl was diminished by 46%
(P < 0.05; segment b-c, Fig. 1B;
Table 2). The isosmotic readdition of bath Cl induced
only a transient and nonsignificant decrease in pHi
(segment c-d, Fig. 1B).
When studied in the nominal absence of
CO2/HCO (solution C,
external pH 7.40) and in the bilateral presence of 0.1 mM acetazolamide
to inhibit endogenous HCO production,
pHi measured in the bilateral presence of Cl
was significantly higher than pHi measured in the presence
of CO2/HCO (segment
a-b, Fig. 1C; Table 2). In addition, the basolateral
replacement of Cl with gluconate (solution C
to solution D) did not induce any significant change in
pHi (segment b-c, Fig. 1C; Table 2),
indicating that the increase in pHi elicited by basolateral
Cl removal was dependent on the presence of
HCO.
Finally, we studied the effect of basolateral Cl addition
in the bilateral absence of Na+ (isosmotically replaced
with NMDG+) together with a voltage clamp obtained with 10 µM valinomycin in high (120 mM)-K+-containing Ringer
solution (solution F); this maneuver has previously been
shown to efficiently depolarize the plasma membrane when used with
intact tubules (4). First, pHi values were
higher (~7.45; segment a-b, Fig. 1D) under
these conditions (absence of extracellular Na+ and
Cl, high extracellular K+, and valinomycin)
than under control conditions (cf. segment a-b, Fig. 1,
A and D). A high pHi has previously
been noted by others in cortical or medullary TALH, studied under
similar conditions (24, 37). This probably indicates that,
in these cells, a base exit mechanism is inhibited by the absence of
external Cl, the high external K+
concentration, and/or the membrane depolarization induced by valinomycin and K+. The addition of basolateral
Cl (solution F to solution E)
induced a prompt and significant intracellular acidification
(segment b-c, Fig. 1D). The subsequent removal of bath Cl elicited another increase in pHi to
initial values (segment c-d, Fig. 1D), consistent
with the presence of an electroneutral and Na+-independent
Cl/HCO exchange (Fig. 1D,
Table 2).
Taken together, these data demonstrate the presence of an
electroneutral Cl/HCO exchange
activity under basal conditions in the basolateral membrane of the rat
MTALH cells.
Because of the presence of an electroneutral
Cl/HCO exchange activity in the
basolateral membrane of MTALH cells, all the following experiments have
been carried out in the bilateral absence of Cl.
A Cl-independent HCO exit pathway
is present in MTALH cells.
Two Cl-independent HCO efflux
mechanisms have been described in MTALH suspensions: an
Na+-(HCO)n cotransport in
mouse MTALH cells (which is DIDS sensitive), but not in rat MTALH cells
(23), and a DIDS-sensitive and
Ba2+-insensitive K+-HCO
cotransport in rat MTALH (28). Moreover, in a recent study
using purified basolateral membrane vesicles from rat MTALH, besides
the presence of a robust Cl/HCO
exchange and a weak Cl-independent
Na+-HCO cotransport, no conclusive
evidence for a K+-HCO cotransport could
be obtained. To clarify this issue, we wondered whether a basolateral
Cl-independent HCO efflux was present in the rat microperfused MTALH.
A Cl-independent HCO exit pathway in
the basolateral membrane is expected to affect pHi in the following way. 1) In the bilateral absence of
Cl, a decrease in bath pH and HCO
concentration at constant PCO2 should increase
HCO efflux and, therefore, induce a fall in
pHi. 2) The outwardly directed base efflux
induced by a given decrease in bath pH should be greater in the
presence of CO2/HCO than in its
absence. 3) Because many HCO transport
mechanisms previously described in the kidney are stilbene sensitive,
DIDS should reduce the effect of bath HCO decrease
on pHi. 4) The inhibitory effect of DIDS should
be absent in a nominally
CO2/HCO-free Ringer solution.
As shown in Fig. 2, when tubules were
studied in a Cl-free Ringer solution (Cl
replaced with gluconate), a decrease in bath HCO concentration from 23 mM (external pH 7.40) to 3 mM (external pH 6.60)
at constant PCO2 (solution B to
solution G) elicited a prompt intracellular acidification
(segment b-c, Fig. 2) with an initial rate of change in
pHi of 
0.22 ± 0.04 pH unit/min. The initial base
efflux induced by the decrement in peritubular pH and
HCO concentration was calculated to be 4.2 ± 0.8 pmol · min1 · mm1 (see
MATERIALS AND METHODS). In paired experiments conducted in
the same tubules in the presence of 0.15 mM peritubular DIDS, the
magnitude and the initial rate of decrease in pHi after the decrement in bath HCO concentration were
significantly reduced (segment e-f, Fig. 2). The initial rate of base efflux in the presence of peritubular DIDS was reduced and
calculated to be 2.1 ± 0.4 pmol · min1 · mm1
(P < 0.05 vs. without DIDS). Also, before the decrease
in peritubular pH and HCO concentration,
pHi values were rather stable in the absence of DIDS
(segment a-b, Fig. 2) but steadily increased in the presence
of 0.15 mM bath DIDS (segment d-e, Fig. 2); this indicates
that, in the absence of external Cl, peritubular DIDS
inhibited a base efflux pathway under basal conditions.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 2.
Effect on pHi of lowering bath pH and
HCO in the nominal absence of Cl by
decreasing basolateral HCO concentration at fixed
PCO2 in the absence and then in the presence of
0.15 mM DIDS in the bath. Tubules were initially perfused and bathed
with a Cl-free solution (Cl replaced with
gluconate) buffered with CO2/HCO
(23 mM HCO, external pH 7.40; segment
a-b). Lowering basolateral HCO concentration
from 23 to 3 mM at constant PCO2 caused a
larger pHi decrease in the absence (segment b-c)
than in the presence of 0.15 mM DIDS in the bath (segment
e-f). Both effects on pHi were reversed by returning
basolateral pH and HCO concentration to normal
(segments c-d and f-g). Reported values of
pHi were measured at the end of each experimental period
(n = 5 tubules); dpHi/dt,
rate of change of pHi. *P < 0.05, **P < 0.02, and ***P < 0.01 vs. the
same period without DIDS.
|
|
In separate experiments, the same maneuvers were performed in the
nominal absence of external
CO2/HCO and in the bilateral
presence of 0.1 mM acetazolamide (Fig.
3). Under these conditions, a decrease in
basolateral pH from 7.4 to 6.6 (solution D to solution
H) elicited a significant intracellular acidification
(segment b-c, Fig. 3) that appeared to be completely DIDS
resistant (segment e-f compared with segment b-c,
Fig. 3); the decrement in pHi and the rate of change in
pHi were the same in the absence and presence of 0.15 mM
DIDS. The average base efflux elicited by the decrease in bath pH was
calculated to be 2.6 ± 0.3 pmol · min1 · mm1
(P < 0.05 vs. Jbase measured in
the presence of CO2/HCO).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 3.
Effect on pHi of lowering basolateral pH in
the absence of CO2/HCO in
Cl-free solutions. The study was performed in the absence
and then in the presence of 0.15 mM DIDS in the bath. Tubules were
perfused and bathed with a
CO2/HCO- and
Cl-free solution (HEPES-Tris solution, external pH 7.40;
segment a-b). Lowering basolateral pH from 7.40 to 6.60 in
the absence of bath DIDS caused a pHi decrease
(segment b-c) that was unchanged in the presence of 0.15 mM
DIDS in the bath (segment e-f). Both effects on
pHi were reversed by returning basolateral pH to normal
(segments c-d and f-g). Reported values of
pHi were measured at the end of each experimental period.
|
|
In summary, in the bilateral absence of Cl, a decrease in
bath pH and HCO concentration caused a reversible
intracellular acidification and a net base efflux; both values were
significantly decreased in the presence of bath DIDS or in the absence
of external CO2/HCO. Taken
together, these data demonstrate a Cl-independent,
DIDS-sensitive HCO efflux pathway in the basolateral
membrane of MTALH cells.
DIDS-insensitive, EIPA-insensitive
Na+-HCO
cotransport
activity is present in the basolateral membrane of MTALH cells.
In the presence of an Na+-HCO
cotransport in the basolateral membrane of MTALH cells, the following changes in pHi should be observed. The isosmotic removal of
bath Na+ (Na+ replaced with NMDG+)
induces an outwardly directed Na+ efflux that drives
HCO out of the cell, resulting in a decrease in
pHi; however, the analysis is complicated by the presence
of Na+/H+ exchange activity in this basolateral
membrane that acidifies the cell as a consequence of peritubular
Na+ removal. Nevertheless, if an
Na+-HCO cotransport is present in
addition to an Na+/H+ exchange, removal of
peritubular Na+ should induce a higher base efflux in
the presence of CO2/HCO than in
its absence (HEPES-Tris condition).
The results are shown in Fig. 4. In the
presence of external CO2/HCO,
pHi remained stable in the peritubular presence of 140 mM
Na+ (segment a-b, Fig. 4A). Replacing
peritubular Na+ with NMDG+ (solution
B to solution I) elicited a prompt intracellular
acidification (from 6.92 ± 0.04 to 6.41 ± 0.05 pH units;
segment b-c, Fig. 4A), indicating a net base
efflux. The initial base efflux was calculated to be 6.32 ± 0.56 pmol · min1 · mm1;
finally, returning bath Na+ concentration to the control
value induced a cellular realkalinization (segment c-d, Fig.
4A). When the same maneuvers were performed in paired
experiments in the peritubular presence of 30 µM EIPA, the changes in
pHi induced by the removal/readdition of bath
Na+ were significantly decreased, but not suppressed,
indicating an EIPA-sensitive component in the Na+-dependent
net base efflux (Fig. 4A).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 4.
Effect on pHi of basolateral Na+
removal (replaced by
N-methyl-D-glucamine+) in
Cl-free solutions in the absence and then in the presence
of 30 µM ethylisopropylamiloride (EIPA) in the bath in the presence
(A) and absence (B) of
CO2/HCO. A: tubules
were initially perfused with a Cl- and
Na+-free solution containing 23 mM HCO
(1 mM amiloride was also added in the luminal fluid) and bathed with a
Cl-free solution containing 23 mM HCO
and 140 mM Na+ (external pH 7.40; segment
a-b). In the absence of EIPA, basolateral Na+ removal
induced a reversible decrease in pHi (segment
b-c) that was partially blunted in the presence of 30 µM EIPA in
the bath (segment f-g). Both effects on pHi were
reversed by returning basolateral Na+ concentration to
normal (segments c-d and g-h; n = 7 tubules). B: tubules were initially perfused with a
Cl- and Na+-free solution (also containing 1 mM amiloride) and bathed with a Cl-free solution
containing 140 mM Na+ in the bilateral absence of
CO2/HCO and in the presence of
0.1 mM acetazolamide (external pH 7.40; segment a-b). In the
absence of EIPA, basolateral Na+ removal induced a
reversible pHi decrease (segment b-c) that was
dramatically reduced in the presence of 30 µM EIPA in the bath
(segment f-g). Both effects on pHi were reversed
by returning basolateral Na+ concentration to normal
(segments c-d and g-h; n = 6 tubules). Reported values of pHi were measured at the end
of each experimental period. **P < 0.01 vs. the same
period in the absence of EIPA; ***P < 0.001 vs. the same
period in the absence of EIPA.
|
|
The same experimental maneuvers were then performed in the complete
absence of external CO2/HCO (Fig.
4B). Removal of peritubular Na+ (solution
D to solution J) induced a significant intracellular acidification (segment b-c, Fig. 4B) that was
reversed by returning peritubular Na+ concentration to its
initial value (segment c-d, Fig. 4B). The initial
rates of change in pHi after peritubular Na+
removal were almost identical in the presence and absence of external
CO2/HCO; however, because the
intracellular buffering power was significantly higher in the presence
than in the absence of CO2/HCO (58.0 ± 0.55 vs. 34.9 ± 3.61 mmol · pH
unit1 · l1, P < 0.001), the Na+-driven base efflux was accordingly lower
under CO2/HCO-free conditions
(3.42 ± 0.38 pmol · min1 · mm1,
P < 0.01). Finally, the effect of peritubular EIPA was
studied in CO2/HCO-free
solutions. The magnitude and the rate of changes in pHi
were significantly decreased in the presence of 30 µM peritubular
EIPA (segment a-b-c-d compared with segment
e-f-g-h, Fig. 4B).
Taken together, these results indicated the presence in the basolateral
membrane of an EIPA-sensitive Na+/H+ exchange
activity and an additional Na+-HCO
cotransport activity, the latter being demonstrated by the higher
Na+-driven base efflux measured in the presence than in the
absence of external CO2/HCO. In
this regard, it is noteworthy that the sum of the EIPA-resistant
Na+-driven base efflux measured in
CO2/HCO-containing solutions
(2.92 ± 0.21 pmol · min1 · mm1) and the
Na+-driven base efflux in
CO2/HCO-free solutions
(3.42 ± 0.38 pmol · min1 · mm1) was
identical to the Na+-driven base efflux measured in
CO2/HCO-containing solutions
(6.32 ± 0.56 pmol · min1 · mm1). In
addition, because the Na+-driven base efflux was always
higher in CO2/HCO-containing than
in CO2/HCO-free solutions,
irrespective of the absence or presence of EIPA, the
Na+-HCO cotransport activity was EIPA resistant, whereas the Na+/H+ exchange activity
was EIPA sensitive.
In a second set of experiments, we studied the sensitivity to DIDS of
the Na+-dependent HCO cotransport across the basolateral membrane of the rat MTALH cells (Fig.
5). Under control conditions,
pHi was stable in the presence of 140 mM bath Na+ (segment a-b, Fig. 5). Removing bath
Na+ (solution B to solution I)
induced a fall in pHi (from 6.98 ± 0.04 to 6.63 ± 0.04 pH units; segment b-c, Fig. 5) that was
reversed after bath Na+ concentration was returned to the
control value (segment c-d, Fig. 5). Paired experiments have
been conducted in the presence of 0.15 mM (n = 3) or
0.5 mM (n = 3) DIDS in the bath. First, in the
bilateral presence of Na+, the presence of DIDS in the bath
led the pHi to steadily increase with time, at
variance with our observation in the absence of DIDS (segment
d-e compared with segment a-b, Fig. 5). As previously noted, this favors a Cl-independent, DIDS-sensitive base
exit mechanism in the basolateral membrane of the MTALH cells under
basal conditions. Second, removing peritubular Na+ in the
presence of DIDS elicited a fall in pHi (segment
e-f, Fig. 5) that was indistinguishable from the cell
acidification observed in the absence of DIDS (segment b-c,
Fig. 5). Indeed, the magnitude and the rate of cell acidification were
the same in the presence and absence of DIDS. Third, the magnitudes of pHi increase after readdition of Na+ to the
bath were the same in the presence and absence of DIDS (segment
f-g compared with segment c-d, Fig. 5). Finally, the rates of change in pHi after bath Na+
readdition in the presence and absence of bath DIDS were not readily
comparable, because pHi measured at the end of the 0 Na+ period significantly differed when DIDS was present and
when it was absent (point f compared with point
c, Fig. 5). Also, the rates of changes in pHi were not
readily comparable between Fig. 4 and 5, because luminal
Na+ was present in experiments in Fig. 5 and absent in
experiments in Fig. 4.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 5.
Effect on pHi of basolateral Na+
removal (replaced by
N-methyl-D-glucamine+) in
Cl-free solutions in the absence and then in the presence
of 0.15 mM (n = 3 tubules) or 0.5 mM (n = 3 tubules) DIDS in the bath. Tubules were initially perfused and
bathed with a Cl-free solution containing 23 mM
HCO and 140 mM Na+ (external pH 7.40;
segment a-b). In the absence of DIDS, basolateral
Na+ removal induced a reversible pHi decrease
(segment b-c) that was unchanged in the presence of 0.15 or
0.5 mM DIDS in the bath (segment e-f). Both effects on
pHi were reversed by returning basolateral Na+
concentration to normal (segments c-d and f-g).
Reported values of pHi were measured at the end of each
experimental period (n = 6 tubules).
**P < 0.001 vs. the same period in the absence of
DIDS.
|
|
These results indicated that the EIPA-resistant
Na+-HCO cotransport activity in the
basolateral membrane of MTALH cells was also DIDS resistant.
A Ba2+-sensitive electroneutral
K+-HCO cotransport is
present in the basolateral membrane of MTALH cells.
The presence of a K+-HCO cotransport is
expected to affect pHi in the following way. Increasing peritubular K+ concentration reduces the outwardly directed
K+ efflux; therefore, if K+ and
HCO efflux are electroneutrally coupled, increasing
bath K+ concentration should reduce HCO
efflux and elicit a rise in pHi. However, because of the
presence of basolateral K+ channel(s), increasing
peritubular K+ concentration may induce an acute
depolarization of the basolateral membrane, thereby affecting any
electrogenic HCO exit mechanism located in the same
membrane. That is, an increase in pHi elicited by a rise in
peritubular K+ concentration does not allow a distinction
between an electrogenic HCO transport mechanism and
an electroneutral K+-HCO cotransport.
However, in the case of the thick ascending limb (TAL), the
mechanism through which an increase in peritubular K+
concentration induces a basolateral membrane depolarization is not
quite so simple. Greger and Schlatter (19) proposed that a
high peritubular K+ concentration may directly inhibit an
electroneutral K+-Cl cotransport, leading to
a rise in cytosolic Cl activity and a secondary
basolateral membrane depolarization. In addition, in their experiments,
peritubular Ba2+ was able to abolish the effect of
peritubular K+ through a direct inhibition of the
K+-Cl cotransport activity. More recently, Di
Stefano et al. (9) confirmed that peritubular
Ba2+ does not act on a basolateral K+ channel
but on the electroneutral K+-Cl cotransport.
Therefore, to precisely determine the nature of a putative
K+-dependent HCO transport, we tested the effect of the changes in the peritubular concentrations of K+ and Ba2+, alone or in combination, on
pHi. In addition, we measured the acute effect of
peritubular Ba2+ addition on Vte,
taken as an index of basolateral membrane voltage. We reasoned that, if
K+ and Ba2+ are able to inhibit the same
K+-coupled HCO transport, both would be
expected to induce a rise in pHi; in addition, the effect
of the increase in peritubular K+ concentration on
pHi would be attenuated in the presence of Ba2+
(absence of an additive effect); finally, if K+ and
Ba2+ inhibit an electroneutral HCO
transport, the addition of peritubular Ba2+ would not
change Vte.
In the bilateral presence of
CO2/HCO, increasing bath
K+ concentration from 4 to 40 mM (solution K to
solution L) provoked a sharp and reversible intracellular
alkalinization (from 6.99 ± 0.03 to 7.28 ± 0.05 pH units;
segment b-c, Fig. 6). This
alkalinization rapidly reversed (segment c-d, Fig. 6) when
bath K+ concentration was returned to 4 mM (solution
L to solution K).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 6.
Effect on pHi of basolateral K+
concentration increase in the absence and then in the presence of 2 mM
Ba2+ in the bath. Tubules were initially perfused and
bathed with a Cl-free solution containing 4 mM
K+ and 23 mM HCO (segment
a-b). Basolateral increase in K+ concentration from 4 to 40 mM caused a large and rapid rise in pHi
(segment b-c). When the same maneuver was performed in the
presence of 2 mM Ba2+ in the bath, the rate of change in
pHi was dramatically inhibited (segment f-g).
Both effects on pHi were reversed by returning
basolateral K+ concentration to normal
(segments c-d and g-h). Reported values of
pHi were measured at the end of each experimental period
(n = 7 tubules). *P < 0.05, **P < 0.01, and ***P < 0.001 vs. the
same period without Ba2+.
|
|
When bath K+ concentration was changed in paired
experiments on the same tubules in the presence of 2 mM
Ba2+ in the bath (solutions M and N),
the magnitude and rate of change in pHi were clearly
inhibited (Fig. 6, right). The rate of change in
pHi was inhibited by 55 ± 7% when bath
K+ was changed from 4 to 40 mM (segment f-g
compared with segment c-d, Fig. 6) and by 75 ± 6%
when bath K+ was decreased from 40 to 4 mM
(segment g-h compared with segment c-d,
Fig. 6).
The effect of addition of 2 mM peritubular Ba2+ on
pHi (solution K to solution S) is
shown in Fig. 7. Peritubular
Ba2+ acutely alkalinizes the cell (from 7.13 ± 0.03 to 7.21 ± 0.04 pH units, P < 0.05).
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 7.
Effect on pHi of 2 mM Ba2+
addition in the bath in
CO2/HCO-containing solutions in
the nominal absence of Cl. Tubules were initially
perfused and bathed with a Cl-free solution containing 4 mM K+ and 23 mM HCO. Peritubular
addition of 2 mM Ba2+ induced a significant cellular
alkalinization. Reported values of pHi were measured at the
end of each experimental period (n = 6 tubules).
|
|
The effect of 2 mM peritubular Ba2+ on
Vte is presented in Fig.
8. In the presence of Cl,
the peritubular addition of 2 mM Ba2+ (solution
A to solution Q) induced a quick and significant
increase in Vte (from 9.6 ± 1.3 to
15.0 ± 2.3 mV, P < 0.05). In contrast, Ba2+ was unable to induce a change in
Vte in the bilateral absence of Cl
(from 2.0 ± 0.73 to 2.3 ± 0.73 mV; solutions B
and R).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 8.
Effect on transepithelial voltage of 2 mM
Ba2+ addition in the bath in
CO2/HCO-containing solutions in
the presence ( ) and nominal absence of Cl
(replaced with gluconate;  ). Ba2+ induced
an increase in Vte in the presence but not in
the absence of Cl.
|
|
Finally, when the effect of the same changes in bath K+
concentration was studied in
CO2/HCO-free solutions and in the
absence of Ba2+ (solutions O and P),
the changes in pHi were blunted compared with those
observed in the presence of
CO2/HCO (segment
a-b-c-d, Fig. 6 compared with Fig.
9). The rate and the magnitude of
cellular alkalinization were inhibited by 55 and 59%, respectively, in
the absence of CO2/HCO in the
solutions compared with values obtained in
CO2/HCO-buffered solutions.
Furthermore, the basolateral presence of 2 mM Ba2+
(solutions S and T) did not influence the
magnitude or the rate of change in pHi when it was studied
in the absence of CO2/HCO (segment a-b-c-d compared with segment e-f-g-h,
Fig. 9). Thus increasing peritubular K+ concentration in
HEPES-buffered solutions inhibits a Ba2+-insensitive proton
influx (or OH efflux).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 9.
Effect on pHi of basolateral K+
concentration increase in the absence of
CO2/HCO. The study was performed
in the absence and then in the presence of 2 mM peritubular
Ba2+. Tubules were initially perfused and bathed with a
CO2/HCO-free solution containing
4 mM K+ (HEPES-Tris solution; segment a-b).
Basolateral K+ concentration increase from 4 to 40 mM
caused a small pHi increase (segment b-c) that
was totally unchanged in the presence of 2 mM peritubular
Ba2+ (segment f-g). Both effects on
pHi were reversed by returning basolateral K+
concentration to normal (segments c-d and g-h).
|
|
Taken together, these results are consistent with the presence of a
Ba2+-sensitive, K+-coupled
HCO exit pathway in the basolateral membrane of rat
MTALH cells that accounts for the major component of
Cl-independent base efflux. Moreover, this pathway
appears to be electroneutral, because addition of Ba2+
inhibited the K+-coupled HCO exit but
did not change Vte. Our results are also
consistent with the presence of a Ba2+-resistant
K+/H+ exchange (or
K+-OH cotransport) in the same membrane that
accounts for a minor proportion of the Cl-independent
basolateral base efflux, in agreement with previous studies from our
laboratory using basolateral membrane vesicles from rat MTALH
(29).
Pharmacological characterization of the electroneutral
K+-coupled HCO exit
pathway.
The Ba2+ sensitivity of the K+-coupled
HCO exit is reminiscent of that described for the
electroneutral K+-Cl cotransport (3, 9,
19, 30).
To address this issue, we tested the effect of several drugs previously
reported to be efficient inhibitors of the
K+-Cl cotransport on the activity of the
K+-coupled HCO exit pathway. We compared the initial rates of change in pHi induced by increasing
peritubular K+ concentration from 4 to 40 mM in the absence
and then in the peritubular presence of one of the following drugs:
furosemide (2 mM), DIOA (0.1 mM), and NPPB (0.1 mM). Finally, because
the Cl-independent HCO efflux pathway
was sensitive to DIDS (Fig. 2), we also tested the effect of this latter drug on the K+-induced cellular alkalinization. As
shown in Fig. 10 and Table 3, bath furos
-, Na+-,
and K+-coupled base transport mechanisms in rat
MTALH
TOP
ABSTRACT
INTRODUCTION
