Na'}-H+ Exchange and Na' Entry across the Apical Membrane ofNecturus Gallbladder

18 Pages·2003·1.09 MB·English

(cid:9)(cid:9)
Na' -H Exchange and Na' Entry across the
} +
Apical Membrane of Necturus Gallbladder
STEVEN A. WEINMAN and LUIS REUSS
From the Department ofPhysiology andBiophysics, Washington University School of Medicine,
St.Louis, Missouri63110 D
o
w
n
lo
a
d
e
ABSTRACT Therole ofNa''-H+exchange in Na'transportacross theapical d fro
membrane wasevaluated in Necturus gallbladderepithelium by meansofintra- m
h
cellular Na' activity (aNai) and "Na' uptake measurements. Under control ttp
conditions, complete replacement of Na' in the mucosal solution with tetra- ://rup
methylammonium reduced aNai from 14.0 to 6.9 mM in 2 min (P < 0.001). ress.o
MucosaladditionoftheNa'-H' exchange inhibitoramiloride(10-'M)reduced rg
/jg
aNai from 15.0 to 13.3 mM (P< 0.001), whereasbumetanide (10-5 and 10' p/a
M)hadno effect.Na'influxacross theapical membranewasstudiedbytreating rticle
the tissues with ouabain, bathing them in Na-free solutions, and suddenly -pd
replacing the mucosal solution with an Na-containing solution. When the f/83
/1
mucosal solution was replaced with Na-Ringer's, aNai increasedat -I 1 mM/ /57
/1
min. This increase wasinhibitedby 54% byamiloride(10-'M,P<0.001)and 2
4
8
wasunaffectedby bumetanide(10-5 M).Amiloride-inhibitable Na'fluxesacross 50
3
theapical membranewerealsoinducedby theimposition ofpH gradients. Na' /57
.p
influx was also examined in tissues that had not been treated with ouabain. df b
Under control conditions, "Na' influx from the mucosal solution into the y g
u
epithelium waslinear over thefirst60 sandwasinhibitedby 40%by amiloride est o
(10-' M, P < 0.001) and by 19% by bumetanide (10-5 M, P < 0.025). We n 1
2
conclude that Nat-H' exchange is a majorpathwayfor Na' entry in Necturus F
e
b
gallbladder, which accounts for at least half ofapical NO influx both under ru
a
transportingconditions andduring exposure to ouabain. Bumetanide-inhibita- ry 2
0
bleNa'entrymechanisms mayaccountforonlyasmallerfraction ofNa'influx 23
under transporting conditions, and cannot explain influx in ouabain-treated
tissues. Theseresultssupportthehypothesisthat NaCl entry resultsprimarily
from theoperationofparallel Na'-H'andCl--HC09exchangers,andnotfrom
abumetanide-inhibitableNaCl cotransporter.
INTRODUCTION
Leakyepithelia such as gallbladder, renal proximal tubule,andintestine accom-
plish transepithelial transport of NaCl by a neutral process. (For reviews see
Diamond, 1968; Frizzell et al., 1979a; WarnockandEvelofF, 1982.) Numerous
studies have demonstrated an interdependence of net transepithelial Na' and
Addressreprint requests to Dr. Luis Reuss, Dept. of Physiology and Biophysics, Washington
University School ofMedicine,660SouthEuclid Ave., St. Louis, MO 63110.
J.GEN. PHYSIOL. 0TheRockefellerUniversityPress- 0022-1295/84/01/0057/18$1.00 57
Volume83 January 1984 57-74
58 THEJOURNAL OF GENERAL PHYSIOLOGY " VOLUME 83 - 1984
Cl- fluxes (QuayandArmstrong, 1969; Nellanset al., 1973;Frizzell et al., 1975;
Cremaschiand Henin, 1975), and an Na' requirementfor the maintenance of
intracellular chloride activity (aCli) at a level greater than that predicted from
equilibrium distribution (Duffey et al., 1978; Spring and Kimura, 1978; Reuss
and Grady, 1979;Garcia-Diaz andArmstrong, 1980; Oberleithner etal., 1982).
Theseresults have beeninterpretedto implythe existence of direct coupling of
Na'and CI- fluxes through a ternary-complex NaCl cotransporter at theapical
membrane (Frizzell et al., 1979x). However, the possibility of neutral, coupled
NaCl transport resulting from the simultaneous operation of Na'-H' and Cl--
OH-exchangers hasbeen suggested(Turnberget al., 1970;LiedtkeandHopfer,
D
1977)and has recently been the subject of considerable investigation (Petersen o
w
n
et al., 1981; Warnock and Yee, 1981a; Liedtke and Hopfer, 1982x, b). Distin- lo
a
d
guishing between these two mechanisms is difficult because many of their ed
predictions arethe same. from
The conclusion that coupled NaCl entry results from asingle cotransporter is http
based in part on the observation that the process is inhibited by furosemide or ://ru
p
bdruumgestahnaivdeeb(eEevneldoefmfoentstarl.a,te1d97t8o;iEnrhiibcistonmaannyd cSopruipnlge,d1i9o8n2txr)a.nsAplotrhtopurgohcetshseess,e ress.org
theirspecificity has not been established. /jgp
/a
The case for parallel ion exchangers is also inconclusive and is based on the rticle
demonstrationofboth Na'-H' andCI--OH- exchange processesin apical mem- -p
d
brane vesicles from leaky epithelia (tourer et al., 1976; Kinsella and Aronson, f/83
/1
1980; Warnock and Yee, 1981x; Liedtke and Hopfer, 1982x, b). In addition, /57
/1
attempts to demonstrate the existenceofa ternary-complex NaCl cotransporter 24
8
have been unsuccessful (Liedtke and Hopfer, 1982x). 50
3
Our previous studies (Weinman and Reuss, 1982x) used extracellular and /57
.p
intracellularpHmeasurements todemonstratetheexistenceofNa'-H'exchange df b
at the apical membrane ofNecturus gallbladder. This Na'-H' exchange process y gu
e
occurs continuously under control conditions, but its magnitude could only be st o
n
estimatedcrudelyfromtheextracellularpH measurements. In thepresentstudy, 1
2
F
we evaluate the role of Na'''-H+ exchange in transapical Na' transport by e
b
measurements of intracellular sodium activity (aNai) and unidirectional 22Na+ rua
uptake in Necturus gallbladder. The results demonstrate that in ouabain-treated ry 20
2
3
tissues, at least 50% of apical Na' entry results from amiloride-inhibitable Na+-
H+exchange,andthat undercontrolconditions, 40%oftheunidirectional 22Na+
uptake is amilorideinhibitable. Apical membrane Na'-H' exchange is therefore
a major pathway for transepithelial Na+ transport. Preliminary results of these
studieshave been reported (Weinman andReuss, 1982b, 1983).
MATERIALS AND METHODS
Mudpuppies(Necturusmaculosus) were purchased from Nasco Biologicals (Ft. Atkinson,
WI), kept in aquaria at -10°C, and fedlive fish. Gallbladders were removed, mounted
mucosal side up,andcontinuously perfused on both sidesin a modified Ussing chamber
as previously described (Reuss and Finn, 1975, 1977). Na-Ringer's solution had the
followingcomposition (mM): 109.2 NaCl, 2.5 KCI, 1.0 CaC12, 1.0 Hepes. TMA-Ringer's
contained 109.2 mM tetramethylammonium (TMA) Cl instead ofNaCl.Thesesolutions
(cid:9)
WEINMAN AND REuss Na+-H+ExchangeacrossGallbladderApicalMembrane 59
were titrated with KOH and equilibrated with room air to have a final pH of 7.7.
HC09.Ringer'swasmade byreplacing 10mM NaCl with 10mM NaHCO, andomitting
the Hepes. This solution wasequilibrated with 1% COQ-99% airand had a final pH of
7.6.AmiloridewasagenerousgiftofMerck,Sharp&Dohme, WestPoint,PA;bumetanide
wasagenerous gift ofHoffman-La Roche, Somerville,NJ. Astocksolution (10' M) was
prepared in Na-Ringer'stitratedto pH 9.2'NaCland['HJmannitol werepurchasedfrom
NewEngland Nuclear, Boston,MA.
ElectricalPotential Measurements
Transepithelial (V), apical membrane (V.), and basolateral membrane (V.) potentials
were measured as described previously (Reuss and Finn, 1975, 1977). The serosal D
o
referencewas an Ag-AgCl electrode connected to the solution via an Na-Ringer'sagar wn
bridge.Themucosalsolution potentialwasmeasured withacalomelelectrodeconnected loa
d
e
stiodteh;eVm.ucaonsdalV,,sowleurtieonrebfyerarefdlotwointgh,esraetsupreacttievdeKbCalthbirnigdgseol.uVtionwsa.sTrraenfseerprietdhteolitahlecsuerrroesnatl d from
h
pulses were passed via two Ag-AgCl electrodes connected to the respective solutions by ttp
agarbridges. ://ru
p
Micropipettes were pulled from 1-mm-OD inner fiberglass capillaries (Hilgenberg, re
Malsfeld, Federal Republic of Germany) on a horizontal electrode puller (Narishige, ss.org
Japan).Thepipettes were filled with either 3 Mor0.5 MKClandhadresistancesof15- /jg
p
a4n0 MM1S2winhveenrtfeidllemdiwcirtohsc3opMe K(NCilkoann,dIinmcm.,erGsaerddeinnNCai-tyR,inNgYer)'sa.nCdelilsmpwaelreedowbistehrveeidthweirtah /article
-p
motorizedremote controlmicromanipulator (Stoelting, Chicago, IL)ora 3-Dhydraulic df/8
3
microdrive (Narishige). Impalements with conventional microelectrodes were accepted /1
/5
when (a)thepotential change upon impalement wasabrupt and monotonic, (b)thecell 7/1
2
potential was stable for at least 2 min, and (c) simultaneous impalement with a second, 48
5
usuallyNa-selective, microelectroderevealed that bothimpaledcellshadthesame appar- 03
/5
entratioofmembrane resistances(see Results). 7.p
d
f b
Intracellular Na'ActivityMeasurements y g
u
e
IntracellularNa-selectivemicrcelectrodeswereconstructedasdescribedpreviously(Reuss st o
n
et al., 1983). Micropipettes were pulled similarly to those used for conventional micro- 1
2
electrodes; when filled with 3 M KCl and immersed in Na-Ringer's, they had resistances Fe
b
of10-30MR.Pipetteswere driedat 160°Candexposedto vaporofhexamethyldisilazane rua
(Sigma Chemical Co., St. Louis, MO) for -1 h. After .tooling, the tips were filled with ry 2
0
Na' resin (0.2 gel). An inner pipette filled with Na-Ringer's was then inserted into the 23
back of the electrode to make contact with the resin within 100 um ofthe tip. The
electrodewas backfilled with Na-Ringer's and an Ag-AgCl wire was inserted and sealed
in placewith wax.
TheNa'resin used wasthat describedby Steineret a.'. (1979),andconsistedofa 10%
wt/wtsolutionofNa-ligandI(Fluky Chemical Co., Hauppauge, NY)in o-n-octyloxynitro-
benzene(Alfa Products, Danvers, MA) with 0.5% Na-tetraphenylborate (Fluky Chemical
Co.) added. Potential measurements with these electrodes were made with an FD223
ultra-high impedanceelectrometer (W-P Instruments, Inc., New Haven, CT).
Slopeandselectivity(Na-'overK+)fortheseelectrodes were ^-52-55and30-100 mV/
decade, respectively. It was notedthat theslopes of theelectrodes in pure KClandNaCl
solutionswere frequently different. Electrodes were thereforecalibrated in six solutions
containing 120mM KClandvariousconcentrations of NaCl rangingfrom 2to 50 mM,
in amethod similar ot'.-ti cscribed byArmstrongandGarcia-Diaz(1980). Intracellular
(cid:9)
60 THE JOURNAL OF GENERAL PHYSIOLOGY " VOLUME 83 - 1984
Na' activity was measured by impaling two cells simultaneously withaKCI-filledmicro-
electrode and an Na-selective microelectrode, respectively. The potential measured by
the conventional electrode (V.), that measured by the Na-selective electrode (VN.), and
thedifference(VN.-V.) weredisplayed ondigitalpanelmeters andastripchartrecorder
(Gould, Inc.,Cleveland,OH). Intracellular Na'activity was determined from VNa - V~
byinterpolationontothe calibration curveforthatparticularelectrode.
The use oftwosimultaneous impalements rather thanseparate single impalements at
differenttimesallowed ustohaveacontinuousrecordofaNai.Inaddition,itallowed the
use of more stringent criteria to validate the impalements. These criteria have been
describedpreviously(ReussandWeinman, 1979;Weinmanand Reuss, 1982a)andinvolve
acomparisonofthevoltagedeflectionsmeasuredbyeachelectrode whenthecellpotential
D
was changed byeithertransepithelialcurrent pulsesormucosal solution ioniccomposition o
w
changes. Anexample ofthesecriteriaispresented in Fig. 1 (seeResults). nlo
a
d
e
d
Intracellular Na'ActivityMeasurements inOuabain-treated Tissues fro
m
Insomeexperiments, gallbladdersweretreated withouabain (10-'M)ontheserosalside http
for atleast45 minandaNai was measured during exposuretoTMA-Ringer'soneither ://ru
themucosal sideonlyoronbothsides.The influx ofNa'acrosstheapical membrane was pre
assessedbysuddenlyreplacingthemucosal TMA-Ringer'swithNa-Ringer'sandrecording ss.o
theresultingchanges inaNai. Insometissues theNa' influx rate,whenthe tissuewas re- rg/jg
p
exposed to control Na-Ringer's,either declinedorincreased as afunctionoftime. For /a
thisreason, influxratesinthepresenceofamilorideorbumetanidewerealwayscompared rticle
withcontrol onesmeasured both before andafterthe exposure tothe drug. Intwoout -pd
f/8
ofninetissues, thecontrol influx ratechanged by afactorof>4frombeginning toend 3
/1
ofouabain exposure. Resultsfrom thesetissues werediscarded. /57
/1
2
4
8
Unidirectional "Na'Influx 50
3
The fluxof"Na' from the mucosal solutionintothe epithelium (J.)was measured in /57.p
anapparatussimilar tothatdescribedbyothers (Schultzetal., 1967;Frizzell etal., 1975). df b
Four gallbladdersweremounted on aplexiglassbase, serosal side down,restingon filter y g
u
paperwetted with Na-Ringer's. The mucosal surface was then isolated by placement of est o
an upper chamberwith cylindrical wellson top ofthe tissues. The sealswere made by n 1
2
rubber 0-rings and the exposed mucosal surfaces each had an area of0.42 cms. The F
e
mucosal solution (200 Al) was gently stirred bybubblingwithair. The bathing solution bru
a
couldberemoved bysuctionand addedbypipettingdirectlyintothetop ofthechamber. ry 2
Gallbladders mounted inthischamberwereexposed onthe mucosal sideto Na-Ringer's 02
3
for 20-30 min. After this preincubation, the mucosal solution was removed by suction
andreplacedbyNa-Ringer's containing12NaCl(-15ACi/ml)and['H]mannitol (-10ACi/
In]).Afteratimedinterval(15-90 s), the tracersolutionwas removedbysuction and the
tissue was rinsed by washing for2s with 5 ml ofisotonic sucrose solution at-.5°C The
gallbladder was thenpunched out,blotted gently, andtransferred toa scintillation vial.
Gallbladdersweredigested byincubationat65°Cfor 1 hwith 200 Al HC103 (60%)and
20AlH202(30%)(Mahin andLofberg, 1966).Afterdigestionandcooling, 10 mlBudget
solve(Research Products International, MountProspect, IL)wasadded,andsampleswere
counted inaTriCarbscintillation counter(PackardInstrumentCo.,Inc.,DownersGrove,
IL)for'Hand22Na.'H counts werecorrected forthefraction ofthe 22Naspectrumthat
counted inthe 'H window (2.10% ofthe total22Nacpm).The 'Hcounting efficiency in
the"Nawindowwas zero.4sNa'influxwascorrectedfortheYsNa'presentinextracellular
fluid contamination asestimated by thedistribution of['I-I]mannitol.This was generally
<5% ofthe totals2Nacounts. Uptake was shown to be linear for 60s (see Results), and
(cid:9)
WEINMAN AND REUSS Na'-H+ExchangeacrossGallbladderApicalMembrane 61
uptake rates were determined from 45-s exposures. When the effects ofamiloride or
bumetanideonf.weretested, the tissues werepreincubated as described and exposed
todrug-containingsolutionfor1minpriortoexposuretothedrug-containingradioactive
tracersolution.
Statistics
Results are presented as means t SE. When specified, comparisons were made by
conventionalpaired dataanalysis.Otherwise,comparisons weremadebyStudent's ttest;
avalueofP< 0.05was consideredsignificant.
RESULTS D
o
w
n
Validation ofImpalements lo
a
d
e
Intracellular Na' activity (aNa;) was determined by simultaneous impalement d fro
with two electrodes, a KCl-filled microelectrode and a liquid-membrane Na- m h
selective microelectrode. The criteria used to validate impalements were those ttp://ru
described previously (Reuss and Weinman, 1979; Weinman and Reuss, 1982a). p
re
The most rigorous testsconsisted ofdemonstratingequal changesofmembrane ss.o
potentialsofthe two impaled cells bothduringtransepithelialcurrent pulsesand rg/jg
p
during ionic substitutions ofthe mucosal solution. Since mucosal Na' removal /a
leadstoanimmediate change in aNa;,however, thedepolarization produced by rticle
-p
K' for Na' substitution could not be used for validation. As demonstrated in df/8
3
Fig. 1, mucosal Na'' replacement with TMA' leads to a fall ofaNa;and a new /1
/5
steady stateis reachedin ^-2 min. IfatthistimeTMA'' isreplacedwithK+,very 7/1
2
4
large changes in apical and basolateral membrane potentials occur that are 8
5
0
identical in both impaled cells. This observation makes it veryunlikely thatthe 3/5
7
recorded valuesofaNa;are inerror because ofnonspecific impalement damage .pd
ofthe apical membrane. All impalements in thisstudywere validated bypassage f by g
u
oftransepithelialcurrent pulses (Weinmanand Reuss, 1982a)and onlyoccasion- e
ally was the selectivity criterion applied. st on
1
Approximately one-third ofthe simultaneous conventional and Na-selective 2 F
e
impalements that satisfied the criteria of abruptness and stability produced bru
a
different basolateral membranevoltage deflections across the two impaledcells. ry 2
0
Theseimpalements were discarded. 2
3
IntracellularNa''Activity:EffectsofNa'Removal, Amiloride, andBumetanide
When Necturus gallbladder was bathed on both sides with Na-Ringer's (1 mM
Hepes), aNa; was 14.1 ± 1.9 mM (n = 52 tissues). When tissues werebathed on
bothsides with 10mM HCO3-Ringer's, aNaiwas 15.6± 1.2mM (n = 10 tissues).
Inordertoevaluate the mechanismsresponsible for the maintenanceofsteady
stateaNa;, the effects ofmucosal Na'removal or mucosaladdition ofamiloride
(10-s M) or bumetanide (10-5 M) were studied. As illustrated in Fig. 2 and
summarized in TableI, Na'replacement withTMA+reduced aNa; by -50% in
2 min. Amiloride, after 2 miff, caused a 12% reduction, which was statistically
significant, and bumetanide caused no measurable change. Similar results were
obtained with amiloride and bumetanide in 10 mM HCO3-Ringer's (Table I).
The initial rates ofchange ofaNa;in Na-Ringer's, calculatedoverthe first 20s,
(cid:9)
62 THEJOURNAL OF GENERALPHYSIOLOGY " VOLUME 83 - 1984
were 10.8 ± 1.6 mM/min after Na' removal (n = 20) and 2.2 ± 0.4 mM/min
afteramiloride addition (n = 20).
Changes in aNaj do not provide quantitative information about the inhibition
of the Na' influx step by these experimental perturbations. In the case of
completeNa' substitution withTMA+, allapical Na' entryhas certainly stopped,
but the decline ofaNajmaybepartlydueto Na' exit across the apical membrane.
In the case ofamiloride, the small magnitude ofthe change may in part reflect
a changing basolateral Na' extrusion rate or cell shrinkage, which would tend
to bluntthe change ofaNai(Springand Ericson, 1982).In thecaseofbumetanide,
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
://ru
p
re
ss.o
rg
/jg
p
/a
rticle
-p
d
f/8
3
/1
/5
7
/1
2
4
8
5
0
3
/5
7
.p
d
f b
y g
u
e
st o
n
1
FIGURE 1. Validation ofaNaj measurements. The records begin with both KCI- 2 F
e
filled and Na-selective microelectrodesincells.Potentialsare inmillivoltsasdefined bru
in Materials and Methods. At the two arrows, transepithelial current pulses (100 ary 2
AA/cm', 10 s duration) were passed. At the beginning andend of the record, the 02
3
mucosal solution was Na-Ringer's. For the periods indicated by the bar, it was
replaced with TMA-Ringer's, K-Ringer's, and TMA-Ringer's. In the lower trace, a
downward change representsadecline in aNa;.
however, the absence ofany effect on aNai would suggestthat therehas been no
inhibition ofNa' influx.Even ifcellvolume decreasedinparallel withareduction
of Na' content, aNaj would be expected to decrease, since Na' constitutes a
small fraction ofthe intracellular cation pool.
Since Larson and Spring(1983) have reported that bumetanidecauses a large,
rapid reduction of aNaj at a higher concentration (10-' M), we also tested this
concentration in fourexperiments. No significant changesin aNaj were observed
(aNai values were 11 .3 ± 1 .2, 12.3 ± 1.1, and 11 .7 ± 1.0 mM before, during [3
min], and after exposure to bumetanide, respectively).
(cid:9)(cid:9)(cid:9)
WEINMAN AND REuss Na+-H+ExchangeacrossGallbladderApicalMembrane 63
Effect ofOuabain on aNaj
Since changes in the rate ofNa' extrusion by the basolateral Na'-K' ATPase
might tend to minimizechanges inaNajproduced byinhibitionofNa'entry, we
chose to study the Na' entry step after inhibiting the basolateral pump with
ouabain. After exposure to ouabain, the cells became somewhat more difficult
to impale, but validsimultaneous impalements couldbeobtained.
The gallbladder was bathed in Na-Ringer's on both sides, and ouabain (10-4
M) was added to the serosal bathing solution. After a 5-10-min lag, aNai
increased roughly linearly atarateof0.8 mM/min(Fig.3). ThisincreaseinaNaj
reduces the driving force for net Na' entry across the apical membrane and D
o
therefore makes it difficult to study this process. Hence, we reduced aNaj in wn
lo
ouabain-treated tissues by complete replacement of Nay with TMA+, either ad
e
bilaterally or on the mucosal side only. When tissues were first exposed to d fro
m
h
-100 ttp
v40 ://rup
-1401 re
ss.o
rg
-a0 /jgp
/a
-80 rticle
-p
d
f/8
3
20 /1/5
110aNai 7/1
2
4
8
5
0
3
TMA Amilwide Bumetanide _ /5
7
1min .p
d
FIGURE 2. Effects ofNaremoval, amiloride,andbumetanideonaNai. Potentials f by g
were measured and transepithelial current pulses were passed as in Fig. 1. All ue
records startwith bothmicroelectrodesincells. Atthe timesindicated bythe bars, st on
mucosal Na-Ringer's was replaced with TMA-Ringer's, Na-Ringer's plus 10' M 12 F
amiloride, or Na-Ringer's plus 10-5 M bumetanide. eb
ru
a
ouabain in Na-containing solutions and then Na' was replaced on the mucosal ry 20
2
3
sideonly, aNaicouldbelowered to -20 mM. If, however, Na' was replaced by
TMA'onboth sides ofthe epithelium, aNaj couldbeconsistently lowered to -5
mM. The need to remove Na' from both sides in order to lower aNaj could
reflect either a component of Na' entry across the basolateral membrane or a
serosa-to-mucosaparacellular Na' leak followed byapical membraneNa'entry.
Na'Influx in Ouabain-treated Tissues:Effects ofAmiloride and Bumetanide
Tissues were treated with ouabain for at least 45 min while aNai was kept low
by exposure to TMA-Ringer's on either the mucosal side only or on both sides
of the tissue. After a steady value of aNai was measured by simultaneous
impalements, the mucosal solution was suddenly changed from TMA-Ringer's
to Na-Ringer's containing either no drug (control), amiloride (10-s M), or
bumetanide(10-5 M). The increases inaNajafter 1 min werecomparedfor these
(cid:9)
64 THEJOURNALOF GENERALPHYSIOLOGY " VOLUME 83 - 1984
three conditions. As illustrated in Fig. 4 and summarized in Table II, upon re-
exposure of the mucosal side ofthe tissue to Na-Ringer's, Na' enters the cells
andcausesarise in aNaiof-I1 mM in thefirstminute.This increase is inhibited
by -50% byamiloride; it is unaffected bybumetanide.
Transepithelial and cell membrane potentials before and afterre-exposure to
Na' under these conditions arepresented in Tables IIIand IV. It is interesting
to note that there is no significant difference between control, amiloride, or
bumetanide. This suggests that the mechanism of Na' entry across the apical
membrane is electrically silent (see Discussion).
Effect ofExternalpH on aNai D
o
w
n
Theobservationthat ^-50% ofNa'entry in ouabain-treated gallbladders can be loa
inhibited by I mM amiloride suggests that Na'}-H+ exchange accounts for a ded
substantial fraction ofentry. In principle, however, inhibition of Na' entry by from
h
amiloride does not prove that it results from Na'-H' exchange. As a result of ttp
://ru
p
EffectofMucosalSTolAutBiLonERepIlacementon aNa; ress.org
/jg
Controlsolution Experimental Controlsolution p/a
(before) solution (after) n P rticle
-p
HeTpMesA--bRuifnfgeerre'ds 14.0±1.2 6.9±0.6 13.8±1.2 21 <0.001 df/83/1
Amiloride(10' M) 15.0±1.8 13.3±1.6 15.9±1.8 21 <0.001 /57
Bumetanide(10' M) 14.4±2.7 14.6±2.7 14.7±3.1 13 NS /12
4
8
5
0
HCO3-buffered 3/5
Amiloride(10"'M) 16.1±1.4 15.0±1.2 16.1±1.2 9 <0.025 7.p
d
Bumetanide(10' M) 13.3±1.3 13.6±1.2 14.2±1.5 8 NS f b
y g
ValuesareintracellularNa'activities(inmillimolar)measuredin Na-orHCO3-Ringer'sbefore,2minafter ue
solutionchange,andafterreturningtothecontrolsolution.DrugswereaddedtotheappropriateRinger's st o
n
solutionattheindicatedconcentrations. 1
2
F
e
b
inhibiting Na'-H' exchange, amiloride decreases intracellular pH in Necturus rua
gallbladder (Weinman and Reuss, 1982a). Conceivably, the lower rate of Na+ ry 2
0
2
entry in amiloride-treated tissues could result from a difference in intracellular 3
pH after re-exposureto Na+. Even ifthe only effect ofamiloride were to block
Na+-H+ exchange, the intracellular acidification thus caused might inhibit an-
otherpathwayforNa'entry, such as NaCl cotransport.Thispossibilitywasruled
outby the experiments shown in Fig. 5andTable V. In theseexperiments, the
external Na'concentrationwasadjusted to keep theNa'activity ratioacross the
apical membrane near unity. IfNa'-H' exchange were themechanism by which
Na+ crossesthe membrane, sizablenetfluxes couldthen beinduced by mucosal
solution pH changes.
Tissues were exposed to TMA-Ringer's with ouabain on the serosal side and
TMA-Ringer'scontaining 11 mM Na+ on themucosalside. Asshown in Fig. 5,
changing the pH of the mucosal solution from 7.7 to 6.1 (HCl titration), at
constant mucosal Na'concentration, caused a fall in aNai,consistent with a net
Na+ flux mediated by Na'-H' exchange. When the experiment was repeated
(cid:9)(cid:9)(cid:9)(cid:9)
WEINMAN AND REUSS Na+-H'ExchangeacrossGallbladderApicalMembrane 65
with amiloride added to the pH 6.1 solution, aNai changes were abolished. A
subsequent mucosalacidification withoutamiloride demonstrated that the tissue
retained the ability to respond to mucosal solution pH changes. The results of
similar experiments in five tissues are presented in Table V. They demonstrate
that pH gradients across the apical membrane induce large changes in aNai in
ouabain-treatedNecturus gallbladder. These changes areabolished by amiloride.
2Na
Unidirectiona12 +Influx
The studies on ouabain-treated Necturus gallbladder have shown bumetanide-
insensitive, amiloride-inhibitable Na'-H' exchange to be a major mechanism of
D
Na' flux across theapical membrane. However, it is possible that themagnitude o
w
n
of Na'-H' exchange in tissues exposed to ouabain is different from that under loa
d
control conditions. It was thus necessary to establish whether Na'-H' exchange ed
is also a sizable component of Na' entry in tissues in which the Na' pump is from
h
ttp
240 ://rup
re
ss.o
rg
/jg
p
aZa 30 /article-pd
f/8
3
/1
J20 /5
7
J /12
J 4
8
5
0
3
/5
7
.p
d
L-L__L_1- 11 1 f b
10 20 30 40 y g
u
TIME IN OUABAIN, min est o
FIGURE 3. IntracellularNa'activity as afunction oftime ofexposure to ouabain. n 1
2
aNai was determined from continuous intracellular recordings after ouabain (10-4 F
e
M) wasaddedto theserosalsolution. For each pointn ranges from 8to 13 tissues. bru
a
ry 2
0
operative.This wasaccomplished by measurements of22Na'influx from mucosal 23
solution into theepithelium with methodssimilarto thoseused by others (Schultz
et al., 1967; Frizzell et al., 1975).
Thetime course of22Na' entry into theepithelium is shown in Fig. 6. During
the first 60 s, influx is linear and can therefore be presumed to represent the
entry process without any significant contribution of 22Na' backflux (Schultz et
al., 1967).
Theeffectsamilorideand bumetanide onJmearesummarized in Table VI.Jme
was reduced by 40% by the presence of amiloride (10-s M) in the mucosal
solution. This result is statistically significant (P <0.001)and is in good quanti-
tative agreement with the 54% reduction of Na' entry estimated from aNai
measurements in ouabain-treated tissues (Table I1). Bumetanide (10-5 M) re-
ducedJrr,e by 19%. This result is also statistically significant (P < 0.05) and may
reflect a smallercomponent of Na'influx that is bumetanide inhibitable.
(cid:9)(cid:9)
66 THEJOURNAL OF GENERAL PHYSIOLOGY " VOLUME 83 " 1984
DISCUSSION
The importance of the resultspresented in this paper lies in how they relate to
the mechanism of neutral NaCl entry in leaky epithelia. Na' and Cl- entry are
linked in many systems and this observation has been interpreted to result from
either a direct NaCI cotransporter (Frizzell et al., 1979a), or parallel Na'-H',
Cl--OH- exchangers (Turnberget al., 1970; Liedtkeand Hopfer, 1977).
The proposal of an apical membrane NaCl cotransporter is based on demon-
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
://ru
p
re
ss.o
rg
/jg
p
/a
rticle
-p
d
f/8
3
/1
/5
7
/1
2
4
8
5
0
3
/5
7
.p
d
f b
y g
u
e
st o
n
1
2
F
FIGURE 4. Changes in aNa;upon sudden exposure to mucosal Na' in aouabain- eb
treated gallbladder. Theseare five consecutive records obtained in asingle tissue. rua
Theinterval between traces is 1-4 min. The left-hand scale is VNa - V«, the right- ry 20
2
hand oneis aNa;. Thetissue wasperfused on both sideswith TMA-Ringer'sand, at 3
the timesindicatedby the bars, themucosal solution was replaced with either Na-
Ringer's without drugs (A, C, E), or with bumetanide (B), or amiloride (D). The
deflections in thelatter halfoftraces B, C, andD were produced by transepithelial
currentpulses.
strationsofinterdependenceof Na'and Cl-fuxes andtheeffectsoffurosemide
or bumetanide on cell volume changesand Na-dependent Cl- fluxes or intracel-
lular Cl- activity (Quay and Armstrong, 1969; Frizzell et al., 1975, 1979b;
Cremaschi and Hénin, 1975; Eveloff et al., 1978; Oberleithner et al., 1982;
Ericsonand Spring, 1982a). However, theseargumentsareinconclusive because
flux interdependence can also be explained by parallel exchangers, and because
the implied specificity of the loop diuretics is uncertain, particularly since