Interaction of chloride and bicarbonate transport across the basolateral membrane of rabbit proximal straight tubule Evidence for sodium coupled chloride/bicarbonate exchange

Interaction of chloride and bicarbonate

transport across the basolateral membrane of

rabbit proximal straight tubule. Evidence for

sodium coupled chloride/bicarbonate

exchange.

S Sasaki, N Yoshiyama

J Clin Invest. 1988;81(4):1004-1011. https://doi.org/10.1172/JCI113410.

The existence of chloride/bicarbonate exchange across the basolateral membrane and its physiologic significance were examined in rabbit proximal tubules. S2 segments of the proximal straight tubule were perfused in vitro and changes in intracellular pH (pHi) and chloride activity (aCli) were monitored by double-barreled microelectrodes. Total peritubular chloride replacement with gluconate increased pHi by 0.8, and this change was inhibited by a pretreatment with an anion transport inhibitor, SITS. Peritubular bicarbonate reduction increased aCli, and most of this increase was lost when ambient sodium was totally removed. The reduction rates of pHi induced by a peritubular bicarbonate reduction or sodium removal were attenuated by 20% by withdrawal of ambient chloride. SITS application to the bath in the control condition quickly increased pHi, but did not change aCli. However, the aCli slightly decreased in response to SITS when the basolateral bicarbonate efflux was increased by reducing peritubular bicarbonate concentration. It is concluded that sodium coupled chloride/bicarbonate exchange is present in parallel with sodium-bicarbonate cotransport in the basolateral membrane of the rabbit proximal tubule, and it contributes to the basolateral bicarbonate and chloride transport.

Research Article

Find the latest version:

Interaction

of Chloride and Bicarbonate

Transport

across

the Basolateral

Membrane of Rabbit Proximal Straight Tubule

Evidence forSodiumCoupled Chloride/Bicarbonate Exchange

Sei Sasakiand NaokiYoshiyama

Department ofInternal Medicine, TokyoMedicaland DentalUniversity, Tokyo 113, Japan

Abstract

The existence of chloride/bicarbonate exchangeacrossthe

ba-solateral membrane and its physiologic significance were

ex-amined in rabbit proximal tubules. S2segmentsofthe proxi-mal straight tubule wereperfused in vitro and changes in in-tracellular pH (pHi) and chloride activity(aa0)weremonitored

by double-barreled microelectrodes. Total peritubular chloride replacement with gluconate increased pHi by 0.8, and this changewasinhibitedbyapretreatment withananiontransport

inhibitor, SITS. Peritubular bicarbonate reduction increased

aci and mostofthis increasewas lost when ambientsodium

wastotally removed. The reductionratesof pHi induced bya

peritubular bicarbonate reductionorsodiumremovalwere

at-tenuated by 20% by withdrawal of ambient chloride. SITS

ap-plicationtothe bath inthecontrol condition quickly increased pHi, but did not change a01. However, the a0i slightly de-creasedinresponsetoSITS when the basolateral bicarbonate effluxwasincreasedbyreducing peritubular bicarbonate

con-centration. It is concluded that sodium coupled chloride/bicar-bonate exchange ispresentinparallel with sodium-bicarbonate

cotransport inthe basolateral membrane of the rabbit proxi-mal tubule, and it contributes to the basolateral bicarbonate andchloridetransport.

Introduction

There is growing evidence thatmostofbicarbonatetransport acrossthe basolateral membrane (BLM)' ofthe renal proximal tubule cells is mediated by rheogenic Na-HCO3 cotransport

(1-8).However, it hasnotbeen well examined whether there is

a chloride dependent bicarbonate flux across BLM ofthe

proximal tubule. It isinterestingthatachloride coupled

Na-HCO3 cotransport mechanism (Na-HCO3/Cl exchange) has been identifiedinthe snailneuron(9), squidaxon(10),

bama-cle musbama-cle fiber (I 1), and cultured fibroblast (12). Gugginoet Addressreprintrequests toDr.Sasaki,Second Department of Internal

Medicine, Tokyo Medical and Dental University, Bunkyo-Ku,

Yu-shima,Tokyo 113, Japan.

Receivedfor publication1September 1987 andinrevisedform 23

November 1987.

1.Abbreviations usedinthispaper:aa0,intracellular chloride activity; BLM, basolateral membrane; pHi, intracellularpH; PST, proximal straight tubules; RpHi, pHireduction rates; SITS, 4-acetamido-4'-isothiocyanostilbene-2,2'-disulfonic acid; Vbl, basolateral membrane

potential.

al. demonstrated the existence ofan Na-HCO3/Cl exchange mechanism in BLM of Necturus proximal tubules, and he concluded that most ofthe basolateral chloride transport is mediated by this mechanism (13). In the previous study (8),we

observed that the intracellular pH (pHi) of rabbit proximal straight tubules (PST) increased when chloride of the basolat-eral fluidwastotally removed. This observationwasconsistent withachloride/bicarbonate exchange mechanism. If this kind ofexchange exists, it is conceivable that the exchange plays some role inboth bicarbonate and chloridetransport across

BLM of the proximal tubule.

The aim of this study was to define whether chloride

cou-pled bicarbonate transport exists in BLM ofS2 segment of rabbit PST.We monitored pHi and intracellular chloride ac-tivity(acil)of PST cells when compositions of ambientfluid were rapidlychanged. Our data demonstrated the existence of asodium coupled chloride/bicarbonate exchange mechanism

(Na-HCO3/Cl exchange)acrossBLM.

Methods

Isolated segmentsof rabbit PST were dissected and perfused as pre-viouslydescribed (7, 8, 14). Briefly, the proximal portions of PST (S2 segment)weredissected in cooled (4°C)Asolution (TableI)and were thentransferredtothe bath.Toachievearapid bathfluid exchange, bath volumewasreducedto0.1 ml, and bathfluidwascontinuously changedat5-10ml/min. This resulted inacompletebath exchange within 5s.The bathfluidwaspreheated to 38°C. Thecomposition of artificial solutions used in this study is shown in Table I. These solu-tionswerebubbledwith 5%C02/95%02gas, and their osmolarities wereadjusted to 290 mosmol/kg H20 by adding principal salts or water.

Basolateral membranepotential (Vbl), pHi, andaaiweremeasured by double-barreled micro pH and Cl- electrodes. The method of mak-ing double-barreled ion-selective microelectrodes was described else-where(7,8). Double-barreledborosilicate glass tubing of unequal

di-ameter(fiber-containing; Hilgenberg,FRG) was pulledon ahorizontal microelectrode puller. The larger barrel (1.5mm OD, 0.87mmID) was usedfor the ion-selective electrode, and the smaller barrel (0.75

mmOD, 0.35mmID)was usedforthereference electrode. The inside of the large barrelwasmadehydrophobicby silane vapor, and thetip portionwasbackfilledwith either H+ligand (15),orCl-ligand

(Corn-ing 477913;CorningGlassWorks,Corning,NY). Thereference barrel of the pH electrode wasfilled with 0.5M KCI and that of the Cl-electrode with0.5 MK2SO4+ 10mMKCI solution. Ag-AgClwires wereinserted intothebarrelsandthe electrode was mountedon a

micromanipulator(Leitz, FRG).

The calibration method and general characteristics of pH elec-trodes used inthis studywerethesame asthosereportedpreviously(7). General characteristics of the Cl- electrode were recently reported from this laboratory (16).Theelectrical resistance of theCl--selective

barrel ranged from2 to 5 X 1010 ohm, and the response time(95%

voltagechange)waswithin2s.Theslopeof the response was 55-60 mV/10-fold change in Cl- activity. The selectivity coefficients for other anions(Ka,anion)thatweredetermined in the previous study(16) were,

J.Clin.Invest.

©TheAmerican Society for ClinicalInvestigation,Inc.

0021-9738/88/04/1004/08 $2.00

Table ILCompositionofArtificialSolutions(mM)

A B C D E F G H I J KL

NaCi 125 125 125 145 145

NaHCO3 25 25 5 25 5 25 5 5

Nagluconate 125

Naisethionate 125 145

Choline Cl 125 125

CholineHCO3 25 5

Cholinegluconate 20

NMDG Cl 125

NMDGHCO3 25 25

NMDGgluconate 20 125

MgS04 1 1 1 1 1 1 1I 1 1

K2HP04 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5

CaC12 1.5 1.5 1.5 1.5 1.5 1.5

Cagluconate 6.7 6.7

Caacetate 1.5 1.5 1.5 1.5

Glucose 2 2 2 2 2 2 2 2 2 2 2 2

Alanine 2 2 2 2 2 2 2 2 2 2 2 2

Allsolutionswerebubbledwith 5%

_C02J95%

02gas. NMDG,N-methyl-D-glucammonium.

KI,C.HCO,:

0.1I1,

Kau~:0.03, Ki,i0:0.14. Theaa'was calcu-latedas

adl

=_(Clb+

_Kcu,HCO3

XHC03b)X IO0cv (1)

whereClb andHCO3bareC1- andHCO5activities of the bathfluid,

respectively. VCI is the differential output between the voltages re-cordedbytheCl- selective and reference barrels and S is theslopeof thevoltageresponseof Cl- electrode. It ispossiblethatunmeasurable intracellularanions interact with the Cl- electroderesulting inan overestimation of

_aad

values. Inourprevious study (16),wemeasured

aa'

of PST in the total absence of ambientCl1 (gluconate replacement),

andfound 4.2 mM of unmeasurable anions. We didnotsubstract this valuefrom the measured

_aai'

values in thisstudy.Thus, all

_aai

values

reportedinthispaperarenotthecorrected values. Theelectrical

po-tentials(referenceandion-selectiveelectrodes)weremeasuredwithan electrometer(FD223; W-P Instruments, New Haven, CT) and re-cordedsimultaneouslyon atwo-penchart recorder(R-20,RikaDenki,

Tokyo, Japan). Acommonbath reference electrodewas a3-MKCI

flowingelectrode inadirectcontactwith the exit of bathsolutionto

makea_{liquidjunction potentialnegligibly}small.

Thedataareexpressedasmeans±SE. Forintracellular measure-ments,nequalsthe numberof cells.Onlyoneintracellular

measure-mentwasperformedin each tubule. Student'sttestwasusedto deter-mine statisticalsignificance.

EIc-)

Vbl(unV)

I-1 minl--4

pHi

Figure1.Effect of total bathchloridereplacement

withgluconateonVbl and

pHi.Achloride freeperiod

is denotedbyCl(-).

Results

Effect ofbath CL- elimination on pHi. We have previously reportedthat total bathCl- replacementwith isethionate

de-polarized Vblby4.5 mVandalkalinizedpHi by0.06(8). It is

possiblethatthis effect is notdueto Cl-removal, but dueto

the isethionate addition. To deny this possibility, we

per-formed studies in which bath Cl- wasreplaced with another

anion, gluconate. Intheexperimentshown inFig. 1,total bath Cl- replacement withgluconate (solutionsA and B)

depolar-ized Vblby9.0 mV and alkalinizedpHi by Asummaryof

eightsuchstudies isgivenin TableIILThese resultsare

essen-tiallythe same asthose in thepreviously reportedisethionate

study,indicatingthat theeffect of bath Cl- removal is dueto

the Cl-elimination rather thantothe addition ofsubstituting

anions.

Then,weexamined whether this bath Cl- removal related cellalkalinization is inhibitable with the aniontransport

inhib-itor, 4-acetamido-4-isothiocyanostilbene-2,2'-disulfonic acid

(SITS).Fig. 2showsonesuchexperiment. SITS(0.5 mM)was

addedtothe bathbeforeaCl- removalstudy,and>4minwas

dCl-)

-70r

Vbl(.V)-60 J

-501

IL-in

nu-pHi

allowedas anequilibration period.Theadditionof SITStothe

bathcausedahyperpolarization of Vblaspreviously reported by Biagi (2) and an alkalinization ofpHi. The bath Cl-

re-movalperformed in this condition causedaninitial rapid de-polarization of Vbl by 4 mV, whichwasfollowedbyagradual

depolarization.The initial Vbldepolarizationmaybedueto a

circularcurrentflowgenerated byadiffusion potentialatthe tightjunction as considered in our previous study (8). The mechanism of the later depolarization isnotclearatpresent.

Bath Cl- removal didnot cause anychange in pHi,

demon-strating that the cell alkalinization induced by bath Cl-

re-movalwasprevented with SITSpretreatment. A summaryof five studies isalsogiven inTableII.Theinitial JVbl

depolariza-tionwassmaller in SITS-treatedgroupthan innormalgroup (1.6 vs. 8.1 mV;Table II).The reasonfor thisdifference isnot

clear,but itmaybe duetoalterations of membrane resistance

ofthe luminal and basolateral membranes, and the shunt pathway,orboth.

Effect ofreducingbath HCOyon

acd.

Theresultsdescribed above indicate the existence ofanexchange mechanism of Cl-and HCO- (or related base) at BLM of S2 cells. If this

ex-changerexists,it isanticipatedthatanalteration ofbath

HCO3

shouldchangead,.Toexaminethisprediction, adiwas moni-tored and bath HCO- wasreducedfrom 25to5mM(solutions

AandC). Bath

HCO5

wasreplacedby gluconatetomaintain bath Cl- concentration constant. As shown in Fig. 3, bath

HCO reduction caused alarge and sharp depolarization of Vbland anincrease in aad. This Vbldepolarizationcould be

duetotherheogenicNa-HCO3cotransport. Assummarizedin Table III, the resultsof eight studiesshowed that Vbl instanta-neously depolarized by 34 mVand that adi increased from 25.5to47.6 mM(the maximum peak value). These changes

wereallsignificant. This increase in

aai

isconsistentwith the presenceofCl/HCO3exchange.

Wenextexaminedwhether this

aad

increase inresponse to

bathHCO- reductioncanbeobserved in thenominalabsence

of ambient Na+(solutionsDandE). Na+ of luminal and bath

fluidswassubstitutedwith choline. In alow

HCO5

solution, 20 mM of HCO- was replaced with gluconate to keep

Cl-constant. A representative experimentis shown in Fig. 4. In

the absence ofNa+, bath HCO- reduction from 25to 5 mM caused asmalldepolarization ofVbl andagradual but small

increasein

aai.

Asummaryof five studies (TableIII) wasthat

acdi

increased from 30.5to34.2mM(peak values,P <0.02).

Table I.EffectofBathChloride Removal

Vbl pHi

mV

Control (n=8) -46.6±2.4 7.36±0.01

Cl(-) -38.5±3.1* 7.44±0.02*

SITS (n=5) -72.6±0.9 7.45±0.04

CI(-) -71.0±1.4 7.45±0.04

P<0.05.

Cl(-)denotes thestudies where total bath Cl wasreplaced with

glu-conate.

VblandpHi values of Cl(-) are the valueswhen

Vbl

initially

depo-larized,andpHireached the maximumafterbathC-removal,

re-spectively.

IBOO3

5

-60.

-40-Vbl(iuV)

-20-40.

axcl(MM) 30 20L

|- 2 min .-j

Figure3. Effectof bath bicarbonate reduction from 25 to 5 mMon Vblandadi.Bathbicarbonate wassubstitutedwith gluconate to keep

chlorideconstant.

This

adi

increasewasclearly smaller than thatobservedin the

presenceof Na+ (TableIII, andcompareFigs.3 and4).

Reduction rates ofpHi in the presence and absence ofCLt.

To confirm thatCl- interacts with basolateral

HCO5

trans-port, thereduction ofpHi induced byeitherbath

HCO3

re-ductionorbath Na+ removalwerecompared in thepresence

and absence of ambientCl-.Asshown inFig. 5, bath

HCO3

reduction from25 to 5 mM (at constant pco2) wasrepeated

twice; first, in the presence ofCl- and second, in the total absenceofCl-(solutions F-H, andI). Bath HCO- reduction caused both a large and sharp depolarization of Vbl and a

reduction inpHiin thepresenceand absence of ambientCl-.

However, an inspection of Fig. 5 shows that pHi reduction induced by lowering bath HCO5 was slightly smaller in the absenceofCl-than in thepresence ofCl-. To quantify this effect, the initial pHi reduction rate (RpHi; pH/min) was

measured.RpHiwasmeasuredas(pHi°-pHi'°)X

_{(l/6 min)-',}

wherepHi° andpHi'°werethe pHi valuesatthestartandlOs

afterthebath HCO- reduction, respectively. BecauseourpH electrodespossessedsometime delay inresponse,RpHiwould beslightly underestimated. Thisunderestimation didnot

af-fect our interpretation ofthe data because the studywas a

paired study, and thesame electrodewas usedthrough each

pair.The dataaresummarizedinFig.6.Removalof ambient

C1- slowed RpHiby 26% from 1.80±0.13to 1.33±0.06

pH/

TableIII.EffectofBathBicarbonate Reductionon

acd

Vbl ad

mV mm

Control (n=8) -46.5±2.6 25.5±2.3 HCO- 5 mM -12.6±4.8* 47.6±4.1*

Na+free (n=₅₎ -36.8±3.4 30.5±3.9

HCO- 5mM -34.8±3.4 34.2±4.0*

*P

<0.05.

HCO5

5 mM: bath HCO- wasreducedto5mMbyreplacingwith 20mMgluconatetokeepbathCL-constant.

1bl

andaaivaluesarethe values when Vblinitiallydepolarized,and

5F1

₃

Vbl(mV)

9-1 min

-40

-a 1(hM) 30'

201

Figure 4.Effect of bath bicarbonate reduction in the absence of so-dium.Experimentwasperformedin the total absence of ambient so-dium(choline substitution),andbicarbonatewassubstitutedwith gluconatetokeep chlorideconstant.

I Bath C1-)

-0 LuenCl(-)

-60 |H0

Vbl(=V)

-20[

-10- >1 min z

pHi 7.2z

7.0L

min (n= 6,P<0.01). Theseexperimentswereperformed in

theorderwhere the controlperiod always precededthe Cl- free

period.It waspossiblethat theattenuation of theRpHiwould be dueto atime-dependent processrather than to aspecific Cl- effect. To deny this possibility, we performed the time

control study where bath HCO5 reduction was performed

twiceinthepresenceof ambient Cl-. Fivesuch studies showed that RpHi did not change in the first and second trials (1.97±0.30vs. 1.96±0.27pH/min).

The above result indicated that Cl- removal affects the HCO- flux, which is induced by bath HCO5 reduction. To determine whether this Cl--sensitive HCO- flux is mediated byaNa+-dependent process,the RpHiinduced by bath Na+ removal was also compared in the presence and absence of ambient Cl- (solutionsA, B, J, and K).A representative

ex-perimentis shown inFig.7.Theresultwasalmostidenticalto

that observed when bath HCO5 was reduced (Fig. 5). The removalof ambientCl- slowedRpHiby 19% from 1.78±0.30

to 1.44±0.20 pH/min (P<0.05, n = 6). Asummaryof the

data (Fig. 6)demonstrates that ambientCl- removal slowed

RpHi to a comparable degree when either bath

HCO3

was reduced or Na+ was removed indicating that Cl- sensitive HCO5 flux is probably coupled to a Na-HCO3 cotransport

(Na-HCO3/Cl exchange).2

Effect ofSITSapplicationtothe bathonpHi and

aci.

The results thus farindicatetheexistence of Na-HCO3/Clexchange inBLM(seeDiscussion).Inordertodetermine the rolesofthis exchanger,SITSwasrapidly appliedtothe bath. SITShas been shown to inhibit both the Na-HCO3 cotransport and

Na-HCO3/Clexchange inBLM ofrenalproximal tubules(1-13, 17). Ourpredictionwasthat ifNa-HCO3/Cl exchange

contrib-utes tothe basolateral HCO- andCl- fluxes in the steady-state

condition,asuddenstopoftheexchangewill alterpHiandacl.

Fig. 8Ashowsanexperiment where the effectonpHi of0.5

mMSITSappliedtothebath wasexamined.SITSapplication induced anabrupt increase in Vbl by 3 mV, which was

fol-2. We interpretedtheeffect of Cl- removal on RpHi as indicating evidencefor directcouplingof Cl- and Na-HCO3 cotransport.

How-ever,wehaveto notethat other secondary effects of Cl- removal such

asalterations in cell functions and luminal H+ secretionratesmight induce thesameresult. Forexample, if Cl- removal reducesmetabolic

acidproduction(byinhibitingNa+ transport), thenRpHi induced by HCO-orNaeremoval may besmallin Cl-free condition becausecell

acid load is less.Atpresentwe cannotneglectthesepossibilities.

Figure 5. Effect oflowering bath bicarbonate from 25to5mMin the presence and absence ofambient chloride. Bath bicarbonate

re-ductionwasperformedtwice. After thefirst bicarbonate reduction, bothluminal and bathfluidswerechanged tochloride-free solution, and the secondbicarbonatereductionwasrepeated.

lowed by a gradual hyperpolarization. pHi also quickly in-creasedby 0.9in response to SITSapplication,and it further

increased gradually in the late phase.Asummaryofsevensuch experiments is givenin Table IV. The above results indicate

that S2 cellsin the control condition continuouslytransport

HCO-across BLMfrom the cell totheperitubular fluid via SITS inhibitable transport mechanisms, presumablythe

Na-HCO3cotransport. Theabrupt Vbl hyperpolarization is also

consistentwith theNa-HCO3cotransport mechanism because thistransporterhas been showntoberheogenic (1-8).

Fig. 8Bisanexampleofthestudy where the effect of SITS

on

aai

was examined. As clearly shown in the figure, SITS

applicationinducedaquick hyperpolarizationofVbl(sameas

inFig.8 A), but didnotchange

ada

appreciably.A summaryof thisprotocolis shown in Table IV. This resultmay indicate

that basolateral Cl- flux mediated by the Na-HCO3/Cl

ex-change isnotsignificant(seeDiscussion).

Wefurtherexaminedthepossible contributionofa baso-lateralNa-HCO3/Cl exchange in the condition where

transcel-2.0f

RpHi (pH/min)

1.5

1.

K

HCO3 reduction

.0

Control C1(-)

Figure6. Asummary ofpHireductionrates(RpHi)inducedbybath bicarbonate reduction(25to5mM)orbath sodiumremovalinthe

presence and absence of chloride.

-I Bath Cl(-)

I

Lumen C1 -)

-60r

-40 r

Vbl(mV)

_20

-O

_t

7.4

pHi 7.2 7.0

6.8

II mini

Figure 7. Effect of the total removal of bath sodium in the presence andabsence ofambientchloride. Bath sodium removal was per-formedtwice; first,inthe presence ofchloride,andsecond, in the total absenceofchloride (lumen and bath).

lularCl-transport is stimulated. For thispurpose, ahigh C1-, low HCO-, and formate containing solution (solution L

+formate 1 mM)wasusedas aluminal perfusate. Recently Schild et al. (18) clearly demonstrated that the addition of formate to a high Cl-, low HCO5 luminal fluid increased

transcellularCl- transport in the rabbitS2 segment. As

sum-marized in Table IV, SITSapplicationtothetubule thatwas

perfused with the formate containing luminal perfusate did

notchange

ac,'

aswell.

Finally,weexamined thepossibilitythat theNa-HCO3/Cl

exchangecontributesas aHCO- exitmechanismwhen

baso-lateralHCO- transportis stimulated.Tostimulate basolateral HCO-transport,low HCO- solution(solutionL)wasusedas a

A

bath fluid. Asshown inFig.9,anapplicationofSITSin this

condition causedasharp and largehyperpolarization ofVbl anddecreased

ad,

slightly.A summaryof thedata(Table IV)

shows that

ac'

decreasedsignificantly by3mMinresponseto

SITSapplication. VblhyperpolarizationinducedbySITS ad-ditionwaslargeinthisprotocol comparedwith thatobserved

whenthe bathfluidwasthecontrolsolution(15.2vs.3.3mV,

TableIV). The difference is in accord with ourview that this

Vbl hyperpolarization is dueto the rheogenic Na-HCO3 co-transport. Interestingly, effects ofSITS on Vbl and

_aad

were

reversible, indicatingthat SITS effects may not be due to a

covalentbinding ofSITStothetransporter. Theobserved

re-ductioninac, inresponse toSITSaddition indicatesthatthe

Na-HCO3/Cl exchanger works as a HCO- exit mechanism

whenbasolateral HCO-effluxis stimulated.

Discussion

Mechanisms of the interaction of

CV

and

HCOY

transport.

Recently, HCO- transport across BLM inrenal proximal

tu-bule cells has been showntobemediatedbyaNa-HCO3 co-transport in amphibian (1) and mammals (2-8, 19). This

transport is knowntoberheogenicdueto astoichiometryof

HCO3/Na > 1, thus, this transport is sensitive to voltage changes. In both previous (8) and this study (Fig. 1), we

showed that total elimination ofbathchloride alkalinizedthe cell, indicating the presence of counter-transport ofchloride and bicarbonate. Several mechanisms are considered to ac-count for this exchange. Fig. 10 shows five possible mecha-nisms(mechanisms1-5).Mechanism l isadirect inhibition of theNa-HCO3cotransportbysubstitutinganionsfor chloride.

An inhibition of Na-HCO3 cotransport will increase pHi.

Mechanism2predicts that bath chloride removal depolarizes

Vbl by acircularcurrentflow generatedby adiffusion

poten-tial acrossthetight junction,and this Vbldepolarization in-hibits the Na-HCO3cotransportsince thistransportis voltage

sensitive. Mechanism 3istheexistence ofachloride

conduc-tance in parallel with the Na-HCO3 cotransport in BLM. If there is a chloride conductance in BLM,a bath chloride

re-moval depolarizes Vbl, and this depolarization inhibits Na-HCO3cotransport. Mechanism 4 is theparallel existence of the Na-HCO3 cotransport and Cl/HCO3 exchange. Mecha-nism 5 is the parallel existence of the Na-HCO3cotransport

andNa-HCO3/Clexchange. The aim of this studywas to

dif-B

SITS 2mM

-70

-601-Vbl(mV)

-50_

-40 I-l mini

7.6

pHi 7 4

7.2.

-70 _

-60 . Vbl(mV)

-50 'J

-40

30. a

3.

(MM) 20

Figure 8. Effect of SITSapplication

tothe bath.(A)Anapplicationof SITS(0.5mM)tothe bath causeda Vblhyperpolarizationandan

in-creasein

_ad.

(B)Anapplicationof SITS(2 mM)tothebath increased

Table IV.Effect ofSITS on pHi and

aci

Vbl pHi aa

mV mM

Control (n=7) -53.7±3.7 7.29±0.03 SITS(0.5 mM) -56.l±3.8* 7.38±0.03*

Control (n=8) -52.3±3.6 23.7±2.1 SITS(0.5or2mM)** -56.3±3.3* 24.8±2.2

Lumenhigh Cl- (n=4)t -49.8±3.3 30.0±1.2 SITS(I mM) -56.5±1.7* 29.9±1.2

Bath low HCO- (n=5)§ -21.0±2.9 44.5±4.8

SITS (1 mM) -36.2±4.7* 41.6±4.6*

*P<0.05.

Values in SITSperiod areasfollows; Vbl value is when itinitially hyperpolarized after bath SITSapplication,andpHianda01values

arethevaluesat30safterSITSapplication.

**0.5mMSITSwasappliedin four tubules and in the other four tu-bules 2mMSITSwasexamined. The effectswereessentiallythe same, and the resultsweregathered.

Luminal perfusatewas alowHCO-,high Cl-and I mMformate

containingfluid.

§ Bath fluidwas alowHCO5andhighCl- solution.

ferentiate these mechanisms, and to define the role of this exchange.

Mechanism 1 can notexplain the resultsthat

aad

increased

when bathbicarbonatewas reduced(Fig. 3 and Table II). This

result is in favorof the existence of chloride/bicarbonate

ex-change. Mechanism2alsodoesnotexplainsomeofourdata. The decreasedRpHi by chloride removalisnotconsistent with this mechanism. This mechanism predicts that RpHi is the

same in the presence and absence of ambient chloride. The

result thatadi increased in response tobath bicarbonate

re-ductioncan notbeexplainedbymechanism2 aswell. Mechanism 3,onthe otherhand,canexplain allourdata. However, we have not been able toobtain thesupportive

evi-denceforasignificantbasolateralchloride conductance.In our recent study (20), Vbl depolarization or hyperpolarization

-20 Vbl(mV)

0L

I SITS I |

~-1

min

-60r

a (mM)

ci

40

Figure9.Effect of SITSapplicationonacliwhen bathbicarbonate

was5 mM. A lowbicarbonate solutionwasusedas abathfluid and

SITS(I mM)wasappliedtothe bath.

Lumen Cell Peritubular

Na+

1) C m anions

Na+

2) [

nNHC03-3) I nHCO3

ci-Na+

4) ciH-c

I

~~HC03-Na+ 5)

~~~Na

,

.nHC03-

+-

nHC03-Figure 10. Possible

mecha-nisms thatexplain the chlo-ride andbicarbonate inter-actionacrossthe basolat-eral membrane(see thetext

for thedetail).

causedbyacurrentinjectioninto the tubular lumenthrough

the perfusion pipette changed neither

aad

or

aad

changes in

response to bath chloride removal. This result demonstrated thatmostof basolateral chloridetransportin rabbit S2segment

is electroneutral. Inthepresentstudy(Fig. 1) Vbl depolarized

when bath chloride was removed, but the magnitude ofthe

depolarizationwassmall. This small Vbldepolarizationcould beeasily explained byacircularcurrentflow generated by the chloridediffusion potentialacrossthetight junctionas consid-eredpreviously (8). Thus,asmall Vbldepolarizationis in favor of the absence of basolateral chloride conductance, although there is a possibility that Na-HCO3/Cl exchange is

electro-genic,andthiselectrogenicityobscuresachloride-induced

de-polarization.Our data donotnecessarily neglect the existence of small chloride conductance. As a summary, we conclude that thereisnot asignificant chloride conductanceinthe baso-lateral membrane.

Bothmechanisms4and 5predicttheexistenceofexchange

of chloride and bicarbonate in BLM: oneisa pureCl/HCO3 exchange(mechanism 4), and the otherisa sodiumcoupled

Cl/HCO3 exchange, possibly Na-HCO3/Cl exchange (mecha-nism 5). Thechloride/bicarbonate exchange would explainall ourdata. Todifferentiate whether the exchange is coupledto

sodium,theresponseof

_aad

tobathbicarbonate reductionwas

examined in the absence of sodium(Fig.4and TableIII). The result clearly demonstrated that the most ofCl/HCO3

ex-change is dependent on sodium (mechanism 5). The small

increase in

a0i

in response to bath bicarbonate reductionin

sodium freecondition (Fig. 4) may indicate theexistence of sodium independent Cl/HCO3 exchange (mechanism 4),

al-though thepossibility existsthat a small amountof residual intracellular sodium drives the Na-HCO3/Cl exchange. The result that totalchlorideremoval reduced the RpHi induced by bath sodiumremovalisalsoagainst mechanism4.Ifapure

Cl/HCO3 exchange exists in parallelto theNa-HCO3 cotrans-port, the Cl/HCO3 exchange will compensate pHi changes causedbyabathsodiumwithdrawal. Thus,totalchloride re-moval is expected to exaggerate RpHi, which isinduced by

bathsodiumremoval(contrasttoourobservation).As a sum-maryof theabovediscussion, weconclude that Na-HCO3/Cl

exchange is a major mechanism forthe interaction between

chlorideandbicarbonate inBLMof rabbitPST.

Chloride/bicarbonate exchangeinBLM. We conclude that

a sodium coupled Cl/HCO3 exchange mechanism exists in

BLMof rabbit PST (S2segment). This kind oftransport sys-temshas beenexaminedin other cells. Anelectroneutral

Na-HCO3/Cl-(H) exchange mechanism has been demonstrated in invertebrate nonepithelial cells such asthe snail neuron (9), barnacle muscle(11), and squid giantaxon(10). This mecha-nismhas beenidentifiedas anacid extrusion mechanismwhen

thecellswereacid loaded (Na-HCO3 influx and Cl-(H) efflux).

An apparentlysimilartransport system was alsoidentified in theculturedfibroblasts of hamster ( 12).Intheepithelial cells, Gugginoetal.recentlyidentifiedtheNa-HC03/Clexchangein

BLMofNecturusproximal tubule (13). They concluded that

this mechanism is the major route for chloride movement across BLMintheirpreparation. Alpern reported the possible existence of sodium dependent Cl/HCO3 exchange

mecha-nism inBLMofratPCT(17). Vesicle studies inthe ratkidney (21), andnotin the rabbit kidney(22) have presentedevidence for Na-HCO3/Cl exchange. The demonstration of Na-HCO3/ Clexchange in thepresentstudy strengthens the viewthatthis kind oftransportmechanism prevails inmanykinds ofcells.

Althoughweidentified the Na-HCO3/Cl exchange

mecha-nism inBLMofrabbit PST,wehavetoadmit thatwedidnot

clarify the characteristics of this exchanger. Wecould not

de-termineeven theelectrogenicityof thissystem.Thisdifficulty

was clearly due tothe coexistence of the Na-HCO3 cotrans-porter,which is electrogenic, andpossesses ahighertransport

activity.Anychange that is dueto theNa-HCO3/Clexchange could be easily masked by a dominant Na-HCO3

cotrans-porter. Furtherstudiesareneededtoidentifythe

stoichiome-try,andthetruesubstrate ofthis exchanger.

We observed a small increase in

_aod

in response to bath

bicarbonate reduction in theabsenceof ambientsodium(Fig.

4) which is consistent with asodium independent Cl/HCO3 exchange. The sodiumindepondent Cl/HCO3 exchange

mech-anismhas been shown in otherepithelial preparations(19, 22, 23). In oursodium free condition thepossibilityexiststhat a

small amount of sodium remains in the cell, and that this

sodiumdrivestheNa-HCO3/Cl exchange.At anyrate wecan

safelyconclude that the contribution of sodium independent

Cl/HCO3 exchange isverysmall in the rabbitS2segment.

The roleof

Na-HCO3/Cl

exchange. If we assume that the

stoichiometry of theNa-HCO3/Cl exchange of rabbit PST is similartothat of cells frominvertebrates, i.e., Na-2HC03/C1

(electroneutral), wecantell thedirectionof theexchanger by

calculatingthedriving force.The energyforthe electroneutral

Na-2HC03/Clexchangeacross BLM can begivenas

AG RT (aNa)(aHcO3i)2(aco)

F F (aNaoXaHCO3o)(a) (

where AGisthefree energychange, and F, R, and T are their usualmeanings.

axi

and

ax0

aretheionic activity ofX in the

celland bath, respectively. A negative valueof G means the

influxof sodium and bicarbonate,andeffiluxofchloride out of the cell. Puttingthe intracellular ionic activity values in the controlcondition determined intheprevious(8)and present

studiestoEq.2(aNai47mM,aHCO3 1 1mM,

adi

21mM),AG/F

of-1 1 mVisobtained. Thus, ifthisstoichiometry isthe case,

thistransporteris expectedtoworkas achlorideextrudingand

Na-HCO3 loading mechanismin thecontrol condition. The latter component mayeasilybecompensated bythedominant Na-HCO3cotransporter.

Todetermine thevalidity of this speculation, we applied

SITStothe bath andimmediatechangesinpHiand

acdi

were

monitored. Since we demonstrated the parallel existence of

two SITS sensitive transporters, namely, Na-HCO3 cotrans-port and Na-HCO3/Cl exchange, if the sum ofbicarbonate fluxes mediated bythese transportersis bicarbonateeffluxout

ofthe cell, SITSapplicationwillalkalinize the cell.Similarly,if

chlorideflux mediated by theNa-HCO3/Clexchange isa

chlo-rideeffilux, SITSapplication is expectedto increase

aai.

Our results(Fig. 8) clearly showed that addition of SITS increased

pHi,but didnotchange

aad.

This fact demonstrated that the

netflux of bicarbonateacross BLMis bicarbonateeffiluxthatis mainly mediated by the Na-HCO3 cotransport, and implied thatthe basolateral chloride fluxmediatedby the Na-HCO3/Cl

exchangeisverysmall. Theapplicationof SITS didnotchange

aai

evenin thecondition where the transcellular chloride flux

wasaugmented byperfusingtubular lumen withahigh chlo-ride and formate containing solution, indicatingthat the

Na-HCO3/Clexchangemightnotbe themainroutefor basolateral chloride exit. Other mechanisms besides the Na-HCO3/Cl

ex-change, suchas K-Clsymport (20, 24, 25) maycontributeto

basolateralchloridetransport.However, the lack of shortterm

effect of SITSon

aad

doesnotnecessarily indicatetheabsence of basolateral chloride transport. Our present interpretation

dependson theassumptionthattheapicalchlorideentry

con-tinuesafter basolateraltransportis blocked. If thisassumption

isnottrue,thenit isimpossibletospeculate transcellular chlo-ridefluxbymonitoring the-changes in

adi.

Itmightbepossible

that chloride transport across the luminal and basolateral

membrane isfunctionally coupled, thus,preventingan acute

change in

ad'.

Such functional coupling of solute transport across the two membranes has been considered for sodium (26, 27). Further study isrequiredtoexamine thispossibility. Our results indicate that Na-HCO3/Cl exchange

contrib-utes tothebasolateral bicarbonateeffluxwhenthe basolateral bicarbonatetransportis stimulated.Thiswasshownbythe fact that the RpHi induced by bath bicarbonate reduction was

slowedbyremovalof chloride(Fig. 5),and that-

_acd

decreased

whenSITSwasappliedtothe lowbicarbonate bath solution

(Fig.9, and TableIV).

Speculation regarding the role of the Na-HCO3/Cl

ex-change in physiologic conditions may be interesting. As dis-cussedabove,thisexchangeworksas abicarbonate exit mech-anism as proximal bicarbonate

reabsorption

is stimulated. This stimulation ofthe basolateral bicarbonateeffluxincreases

adi

which in turn will inhibit theuptakeofchlorideacrossthe

luminal membrane. The net effect is that a stimulation of

bicarbonate reabsorption inhibits transcellularchloride

reab-sorption. Thus, Na-HCO3/Clexchangemay worktokeepthe

totaltransportratesof anions(chloride+ bicarbonate)across

BLM constant in the proximal tubule. For example, hyper-capnia has been known toinduce chloruresis (28).

Micro-puncture studies in therat observed thatacute

hypercapnia

stimulates bicarbonate reabsorption and inhibits chloride reabsorptionin PCT(29, 30). Coganconcluded thatthis inhi-bition of chloride absorption occurred on transcellular chlo-rideabsorption (30).Theexistence ofNa-HCO3/Cl exchange

may fit with this interaction of chloride and bicarbonate. A

mechanisms not known) causes an increase in bicarbonate

efflux across BLM. This increased efflux may behandled in partby the Na-HCO3/Cl exchange, and this causes anincrease in ad. An increase in adi will inhibit the luminal chloride uptake mechanism leading to the inhibition of the transcellu-lar chloridereabsorption. Thus, it is conceivable that the Na-HCO3/Cl exchange may play a role when basolateral bicar-bonate and chloride fluxes are altered in several physiologic conditions.

Acknowledgments

We aregrateful to Dr. M. Imai, and Dr. K. Yoshitomi for critically reading themanuscript. We thank Drs. T. Shiigai, Y. Iino, and K. Tomita for encouragement of this work.

This studywassupportedbyaGrant-in-Aid fromMinistry of Edu-cation, Science, and Culture, Japan.

References

1. Boron, W. F., and E. L. Boulpaep. 1983. Intracellular pH regula-tion in the renalproximaltubule ofthe salamander. BasolateralHCO3

transport. J. Gen.Physiol.81:53-94.

2.Biagi,B. A. 1985.Effects of the anion transport inhibitors, SITS,

on the proximal straight tubule of the rabbit perfused in vitro. J. Membr.Bio.88:25-31.

3.Biagi,B.A., andM.Sohtell. 1986.Electrophysiology of basolat-eralbicarbonate transport in the rabbitproximaltubule.Am.J.

Phys-iol. 250(RenalFluidElectrolyte Physiol. 19):F267-F272.

4. Yoshitomi, K., B.-Ch. Burckhardt, and E. Fromter. 1985. Rheogenic sodium-bicarbonate cotransport in the peritubular cell membraneofratrenalproximaltubule.PfluegersArch. Eur. J.

Phys-iol. 405:360-366.

5. Alpern, R. J. 1985. Mechanism ofbasolateral membrane

H'/OH-/HCO5 transport in theratproximal convoluted tubule. A

sodium-coupled electrogenic process. J. Gen.Physiol. 86:613-636. 6.Jentsch,T.J., B. S.Schill,P.Schwartz,H.Matthes, S. K. Keller, and M. Wiederholt. 1985. Kidney epithelial cellsof monkeyorigin

(BSC-1) express a sodium bicarbonate cotransport. J. Bio. Chem.

260:15554-15560.

7.Sasaki,S., T.Shiigai,and J.Takeuchi. 1985. IntracellularpHin theisolatedperfusedrabbitproximalstraighttubule.Am.J.Physiol.

249(RenalFluid andElectrolyte Physiol. 18):F417-F423.

8. Sasaki, S., T. Shiigai, N. Yoshiyama, and J. Takeuchi. 1987.

Mechanismofbicarbonate exitacrossbasolateral membraneof rabbit proximal straight tubule.Am.J.Physiol.252(Renal FluidElectrolyte Physiol. 21):F11-F18.

9. Thomas, R. C. 1977. The role ofbicarbonate, chloride and sodium ions in theregulationof intracellular pH in snailneurones.J.

Physiol.273:317-338.

10. Boron,W. F., and J.M.Russel. 1983.Stoichiometry and ion

dependenciesof the intracellular-pH-regulating mechanism insquid giantaxons.J. Gen.Physiol. 81:373-399.

11. Boron, W. F., W. C. McCormick, and A. Roos. 1981. pH

regulation inbarnaclemuscle fibers: dependenceonextracellular so-dium andbicarbonate.Am.J.Physiol.240(Cell Physiol.9):C80-C89.

12. L'Allemain, G., S. Paris,and J. Pouyssegur. 1984. Role ofa

Na-dependent Cl/HCO3 exchange in regulation of intracellular pH in fibroblasts. J.Biol. Chem. 259:5809-5815.

13. Guggino, W. B., R. London, E. L. Boulpaep, and G. Giebisch. 1983.Chloride transport across the basolateral cell membrane of the necturusproximaltubule: dependence onbicarbonate and sodium. J. Membr. Bio. 71:227-240.

14.Burg, M. B., J. Grantham, M. Abramow, and J.Orloff. 1966. Preparation and study offragments of single rabbit nephrons. Am. J. Physiol. 210:1293-1298.

15. Ammann, D., F. Lanter, R. A.Steiner, P. Schulthess, Y. Shijo, and W. Simon. 1981. Neutralcarrier based hydrogen ion selective microelectrode for extra- and intracellular studies. Anal. Chem. 53:2267-2269.

16.Ishibashi, K., S. Sasaki, N. Yoshiyama, T. Shiigai, and J. Ta-keuchi. 1987.Intracellular chloride activity of rabbit proximal straight tubules.KidneyInt.31 :437a. (Abstr.)

17. Alpern, R. J., andM. Chambers. 1986. Sodium dependent Cl/HCO3 exchangeonthe basolateral membraneof theratproximal

convoluted tubules.Clin. Res. 34:587a. (Abstr.)

18. Schild, L., G.Giebisch, L. P. Karniski, and P. S. Aronson. 1987.Effect of formate on volume reabsorption in the rabbitproximal

tubule. J.Clin. Invest. 79:32-38.

19. Jentsch,T.J., I.Janiche,D.Sorgenfrei, S.K.Keller, andM.

Wiederholt. 1986. Theregulationof intracellular pH inmonkey kid-neyepithelialcells(BSC- 1). J. Biol. Chem. 261:12120-12127.

20.Sasaki,S.,K.Ishibashi,N.Yoshiyama,and T.Shiigai. 1988. KCI cotransportacrossthe basolateral membrane of rabbit renal prox-imalstraighttubules. J.Clin.Invest.81:194-199.

21. Grassl, S. M., P. D. Holohan, and C. R. Ross. 1987. HCO3 transport inbasolateral membrane vesicles isolated fromratrenal cor-tex.J.Biol. Chem.262:2682-2687.

22. Grassl, S. M., and P. S. Aronson. 1986.Na/HCO3 cotransport in basolateral membranevesicles isolated from rabbit renalcortex.J. Biol.Chem.261:8778-8783.

23. Chaillet, J. R., K. Amsler, and W. F. Boron. 1986. Optical

measurementsof intracellular pH in singleLLC-PK, cells: demonstra-tion ofCl-HCO3exchange. Proc. Natl.Acad.Sci. USA. 83:522-526.

24. Reuss,L.1983.Basolateral KCI co-transport ina NaCI-absorb-ingepithelium. Nature(Lond.).305:723-726.

25.Eveloff,J., andD.G. Warnock. 1987.KCI transport systems in rabbit renal basolateral membrane vesicles.Am.J.Physiol.252(Renal

FluidElectrolytePhysiol. 21):F883-F889.

26. Schultz, S. G. 1981. Homocellular regulatory mechanisms in

sodium-transporting epithelia: avoidance of extinction by "flush-through." Am. J. Physiol. 241 (Renal FluidElectrolyte Physiol.

10):F579-F590.

27.Diamond,J. M. 1982. Transcellular cross-talk between

epithe-lialcellmembranes. Nature(Lond.).300:683-685.

28. Levitin, H., W. Branscome, and F. H. Epstein. 1958. The

pathogenesisofhypochloremiainrespiratoryacidosis.J. Clin. Invest. 37:1667-1675.

29. Warren, Y.,R.G. Luke,M.Kashgarian, and H. Levitin. 1970.

Micropuncturestudies ofchloride and bicarbonateabsorptionin the

proximalrenal tubuleof theratinrespiratoryacidosis and inchloride depletion.Clin. Sci. 38:375-383.

30. Cogan, M. G. 1984. Effect ofacutealterations in PCO2on

proximal HCO3, Cl,andH20 reabsorption.Am.J.Physiol.246(Renal