Chloride channels in apical and basolateral membranes of CCD cells (RCCT-28A) in culture.

Chloride channels in apical and basolateral of CCD cells (RCCT-28A) in culture PAUL

DIETL

Department

AND BRUCE

membranes

A. STANTON

of Physiology, Dartmouth Medical School, Hanover, New Hampshire 03756

Dietl, Paul, and Bruce A. Stanton. Chloride channels in apical and basolateral membranes of CCD cells (RCCT28A) in culture. Am. J. Physiol. 263 (Renal Fluid Electrolyte Physiol. 32): F243-F250,1992.-Previously, we found that isoproterenol activates whole cell Cl- conductance by a pathway involving adenosine 3’,5’-cyclic monophosphate and protein kinase A (PKA) in a renal cell line (RCCT-28A) derived from the cortical collecting duct. The goal of the present study was to determine whether PKA activates Cl- channels in the apical and/or basolateral membrane. Using the patch clamp technique we found a 305pS Cl- channel, described previously (22), located exclusively in the apical membrane and an outwardly rectifying Clchannel (13/96 pS) located exclusively in the basolateral membrane. The outward rectifier was highly selective to Cl- versus cations, was inhibited by 4,4’-diisothiocyanostilbene-2,2’-disulfonic acid and 5-nitro-2-(3-phenylpropylamino)-benzoic acid, but was not regulated by cytoplasmic pH or Ca2+. Neither isoproterenol nor PKA activated the 305-pS Cl- channel. In contrast, PKA activated a subset of outwardly rectifying channels in inside-out patches. In another subset of outwardly rectifying channels, formation of the inside-out configuration increased channel activity. These channels, however, were not sensitive to PKA. In conclusion, these experiments show that isoproterenol increases the Cl- conductance of RCCT-28A cells by activating a subset of outwardly rectifying Cl- channels located in the basolateral membrane. intercalated cells; A-intercalated cell; acid clamp; outward rectifier; cystic fibrosis

secretion;

patch

CORTICAL COLLECTING DUCT (CCD) reabsorbs -3% of the NaCl filtered by the glomerulus (30,31). Clreabsorption across the CCD occurs by a paracellular and a cellular pathway (30, 31). As much as 80% of transepithelial Cl- reabsorption may traverse the Cl-permeable paracellular pathway, driven by the lumennegative transepithelial voltage (25). Recent studies indicate that transcellular Cl- reabsorption may be mediated by principal cells (PC) and base-secreting intercalated cells (B-IC). In PC, Cl- may enter the cell via a thiazide-sensitive NaCl cotransporter located in the apical membrane and leave the cell across the basolatera1 membrane through a Cl--conductive pathway (17, l&26,38,39). Reabsorption of Cl- by PC, however, has recently been questioned (12, 27, 28). In B-IC, Cl- enters the cell via a Cl--HCO, exchanger located in the apical membrane and leaves the cell across the basolatera1 membrane through a Cl--conductive pathway (17, 23, 30, 41, 42). A third cell type in the CCD, the acidsecreting IC (A-IC), although not directly responsible for transepithelial Cl- transport, has a Cl-conductive pathway in the basolateral membrane that is important in electrogenic proton secretion (17, 23). Proton secretion by A-IC generates HCO, in the cytoplasm that leaves the cell across the basolateral membrane via the Cl--HCO, exchanger (30). The Cl- entering the cell on this exchanger recycles across the basolateral membrane via a Cl--conductive pathway (23). Accordingly, the

THE

0363-6127/92

$2.00

Copyright

Cl-conductive pathway in the basolateral membrane is crucial for proton secretion. Although Cl- channels have been characterized in the basolateral membrane of PC located in isolated CCD (26) and in the apical membrane of PC in primary culture (8), little is known about the regulation of these channels and about Cl- channels in other cell types in the CCD. Recently, we reported that RCCT-28A cells, a clonal cell line derived from rabbit CCD (1), are primarily conductive to Cl- and are relatively easy to study by the patch clamp technique (33, 34). We characterized a 305-pS Cl- channel located in the apical membrane (22, 33, 34). The channel is inactive in the basal state but is activated by adenosine and cell swelling (33, 36, 37). Furthermore, we found by the whole cell patch clamp technique that the ,&adrenergic agonist isoproterenol activates an outwardly rectifying Cl- conductance by a pathway involving adenosine 3’,5’-cyclic monophosphate (CAMP) and protein kinase A (PKA)(33). However, by use of the whole cell patch clamp technique it was not possible to localize the effect of isoproterenol to Cl- channels located in the apical or basolateral membrane. The goal of the present study, therefore, was to determine whether PKA activates whole cell Cl- conductance in RCCT-28A cells by activating apical and/or basolateral membrane Cl- channels. We report that PKA does not activate the 305~s Cl- channel located in the apical membrane, whereas PKA activates an outwardly rectifying (13/96 pS) Cl- channel located in the basolateral membrane. METHODS Cell culture. As described previously, CCD cells from rabbit kidney were isolated by the immunodissection technique and immortalized by infection with an SV40 adenovirus type 12 hybrid resulting in a continuous cell line designated RCCT-28A (I). RCCT-28A cells were grown on glass cover slips that were coated with Vitrogen plating medium, which contained fibronectin (IO pg/ml; Collaborative Research, Bedford, MA), Vitrogen 100 [I ml/l00 ml Dulbecco’s minimal essential medium (DMEM); Collagen, Palo Alto, CA], and bovine serum albumin (10 pg/ml). Cells were grown in DMEM supplemented with 24 mM HCO, (pH = 7.4,5% C02-balance air at 37”C), 5% heat-inactivated fetal bovine serum, glutamine (2 mM), dexamethasone (1 PM), penicillin (50 U/ml), and streptomycin 50 pg/ml). Subconfluent and confluent monolayers were studied at 23°C. RCCT-28A cells form confluent monolayers, have tight junctions, and produce domes that indicate vectorial salt and water transport (1). In addition, RCCT-28A cells have phenotypic properties of A-IC including N-ethylmaleimide-sensitive electrogenic H+ secretion (3, 4) and Cl--HCO, exchangers in the basolateral membrane (33). All cells contain carbonic anhydrase (33) Analysis of Cl- currents. As described in detail previously (2 1, 22, 33, 34), single-channel Cl- currents were measured with an

0 1992 the American

Physiological

Society

F243

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.

F244

CHLORIDE

CHANNELS

IN CCD

t

open

10 PAI lmin PKA

4

ALKALINE PHOSPHATASE

PKA

Fig. 1. Representative current record of a 305~s Cl- channel in inside-out patches of apical membrane (14-min record). Addition of protein kinase A (PKA; 75 nM) and alkaline phosphatase (2 U/ml) to bath solution (i.e., cytoplasmic surface of membrane) is indicated by arrows. Large deflections in the current record are artifacts caused by addition of PKA and alkaline phosphatase to bath. Downward deflections indicate channel openings. Holding voltage = -40 mV. ATP (1 mM) was present in bath solution throughout experiment.

Axopatch-ID amplifier using a IO-GQ CV-I probe (Axon Instrument, Foster City, CA). Currents were filtered at 1 kHz using a four-pole Bessel filter (model LPF-100; Warner Instruments, Hamden, CT), digitized at 5 kHz using pCLAMP software (version 5.5; Axon Instruments), and stored on the hard disk of an Everex 286 computer. Currents were also recorded continuously on a strip-chart recorder (model 2600s; Gould, Cleveland, OH; full scale frequency response, 125 Hz). As described in detail previously, single-channel amplitudes (i) were measured by constructing amplitude histograms of single-channel currents. Channels were considered open when the current was larger than i/2. Data were recorded on the hard drive of the computer for a minimum of three 10-speriods(with a delay of 30-45 s between each period) during control and experimental periods. Each experimental maneuverwas reversible. The probability of a single channel being open (PO) was defined as the total time the channel was in the open state divided by the total time of data collection. Membrane patches containing multiple patches were rarely encountered; in these casesthe P, was calculated as described(21). Gigaohm sealsof the apical membrane were obtained by touching the apical cell membranewith a patch electrode and applying suction (22). To obtain gigaohmsealsof the basolatera1 membrane, we perfused monolayers with a Ca2+- and Mg2+-free solution (Dulbecco’sphosphate-bufferedsaline;GIBCO) for 15-20 min prior to an experiment. This disrupted tight junctions and causedcellsto partially detach from the Vitrogencoated glasscover slip. Partially detached cells were spherical, and the tight junctions separating the apical and basolateral membraneswere clearly discernable as a constriction around the apical region of the cell (cells viewed with Hoffman modulation contrast optics at x800). In somecases,the basolateral membranewas exposedfurther by using a secondpatch electrode to lift the cell off the cover slip. As discussedin detail later (see RESULTS and DISCUSSION), using this procedure we observeddifferential expressionof Cl- channelsin the apical and basolateralmembranedomains; 305-pS Cl- channels were expressedonly in the apical membrane, whereas 13/96-pS Clchannelswere expressedonly in the basolateralmembrane. Current-voltage (I-V) relations of single Cl; channelswere determined by stepping the command voltage from zero in 20-mV increments (between *80 mV) for IO s using pCLAMP 5.5 routines. Voltages are referenced to a grounded bath solution. The I- Vplots of Cl- channelslocated in the basolateralmembrane outwardly rectified. The I- V data fit a second-orderpolynomial; the slope conductancesat -80, 0, and +80 mV were calculated from the first derivative of the polynomial. Statistics. Data are meanst SE. Statistical significance was determined by use of the Student’s t test for paired observa-

tions. Differences between means are considered significant if P < 0.05.

SoZutions.Patch pipettes were filled with a solution containing (in mM) 140KCl, 1 MgC12,IO N-2-hydroxyethylpiperazineN’-2-ethanesulfonic acid (HEPES, pH 7.4), and 1 CaCl,. The bath solution of cell-attached and inside-out patches contained (in mM) 140 NaCl, 1 MgC12, IO HEPES (pH 7.4), 2 ethylene glycol-bis(P-aminoethyl ether)-N,N,N’,N’-tetraacetic acid, and 1 ATP, as well as 200 nM Ca2+. In bath ion-substitution experiments, Cl- was replacedwith gluconate. Chemicals. 5-Nitro-2-(3-phenylpropylamino)-benzoic acid (NPPB) was a generousgift from Dr. Rainer Greger. All other chemicals, including the catalytic subunit of PKA, were purchasedfrom Sigma Chemical (St. Louis, MO). RESULTS

First, we present data on the effects of isoproterenol and PKA on the 305pS Cl- channel in the apical membrane, and then we present data characterizing a l3/ 96-pS Cl- channel in the basolateral membrane and the effect of PKA on the channel. Apical membrane. The only channel observed in the apical membrane was a 305~s Cl- channel. As described previously (22, 33, 34), the channel has a linear I-V relation, the Cl--to-Na+ permeability ratio (&/&) was 91, and the PolIp HC03 was 2:l. The channel was inhibited by the Cl--channel blockers 4,4’-diisothiocyanostilbene-2,2’-disulfonic acid (DIDS, lo-* M), diphenylamine carboxylic acid (lo-* M), and NPPB ( 10d6 M). This channel is usually inactive in cell-attached patches but is activated by either formation of inside-out patches (i.e., excision activation) or by application of a depolarizing voltage (>30 mV) to inside-out patches (22, 34). Isoproterenol (2 x 10e6 M for 4 min)l did not activate quiescent 305~s Cl- channels in cell-attached patches (n = 6). The presence of channels in cell-attached patches was confirmed by forming inside-out patches and depolarizing the membrane (greater than +30 mV) to activate quiescent channels. In all six experiments in which isoproterenol failed to activate quiescent channels, 3O5-pS Cl- channels were activated by excision or depolarization. To determine whether PKA stimulates excision-activated or voltage-activated 305-pS Cl- channels, we added l Previously, whole cell Cl-

we showed conductance

that after

isoproterenol 2 min (10).

maximally

increased


CHLORIDE

CHANNELS

PKA to the bath solution of inside-out patches. As shown in the representative experiment illustrated in Fig. 1, addition of PKA [ 125 U/ml (75 nM) with 1 mM ATP] to the cytoplasmic solution of inside-out patches decreased the P, from 0.60 t 0.05 to 0.19 t 0.06 (n = 7;P < 0.01). Washing PKA from the bath and simultaneously adding alkaline phosphatase to the bath solution increased the P, to control values (Fig. 1; P, increased to 0.57; n = 3 of 3 experiments). ATP (1 mM), a cofactor required for PKA phosphorylation of its substrate, had no effect on the P, of active channels (n = 12). These experiments show that PKA inhibits excision- and voltage-activated 305-pS Cl- channels in the apical membrane by a-phosphorylation-dependent mechanism. Because isoprotereno1 had no effect on quiescent channels and because PKA inhibited active channels, we conclude that these compounds do not stimulate whole cell Cl- conductance by activating the apical 305-pS Cl- channel. &sol&e& membrane. Our procedure for exposing the basolateral membrane to the patch pipette enabled us to obtain gigaohm seals. Active, anion-permeable channels were occasionally observed in cell-attached patches of the basolateral membrane. Unfortunately, it was not possible to maintain cell-attached patches for a sufficient period (i.e., several minutes) to characterize the channels completely in this configuration or to examine the effects of isoproterenol or CAMP on the P,. We do not know why the cell-attached configuration could not be maintained on basolateral membranes. Instability of the cell-attached configuration may be related to the Ca2+- and Mg2+-free solution or to the continuously flowing bath solution, which often dislodged cells from the cover slip and patch electrode. Spontaneous excision of cell-attached patches often resulted in the formation of inside-out patches con-

F245

IN CCD

taining active Cl- channels whose current signature appeared similar to anion channels active in cell-attached patches. In some inside-out patches the P, was low (CO. 5), whereas in others formation of the inside-out configuration was accompanied by a dramatic increase in channel activity (P, increased from -0 to >0.5). Figure 2 illustrates a family of Cl- channel current traces recorded in the inside-out configuration. Figure 3 is an I-V plot of a typical channel studied in solutions asymmetric with respect to cations (i.e., 140 mM KC1 in the pipette and 140 mM NaCl in the bath). The I- V plot outwardly rectified; the slope conductance was 96 pS at +80 mV, 55 pS at 0 mV, and 13 pS at -80 mV (n = 5). The reversal potential was -2.2 t 3.6 mV (n = 5). This reversal potential indicates the channel either cannot discriminate between Na+ and K+ (i.e., nonselective to cations) or is selective to Cl-. To discriminate between these possibilities, we conducted ion-substitution experiments. As illustrated in Fig. 3, a reduction of the NaCl concentration in the bath from 140 to 40 mM (substitution of 100 mM NaCl with 200 mM mannitol to maintain osmolality) changed the reversal potential from -3.2 t 3.1 to -33.5 t 3.7 mV (n = 5), as predicted by the Nernst equation [i.e., -30.3 mV (measured) vs. -31.6 mV (predicted)] for a highly selective Cl- channel. From the reversal potential and the Goldman-Hodgkin-Katz equation we calculate a pcl/& of ~64. To characterize further the permeability properties of the Cl- channel, we replaced 100 mM NaCl with 100 mM sodium gluconate in the bath solution. The reversal potential changed from -1.5 mV $- 1.3 to -26.9 t 1.4 mV (n = 5), a value consistent with a Cl--selective channel. From the reversal potential and the Goldman-Hodgkin-Katz equation we calculate a &/Pgluconate of 11.4. Thus the channel is a

+90

mV

-C

+60 mV -

C

+30 mV -C

Fig. 2. Representative current records of a Cl- channel in an inside-out patch of basolateral membrane. Vm, holding voltage of each current record. C, closed state.

0 mV -C

-30 mV

-C

-60 mV


F246

CHLORIDE 8TI(~A)

CHANNELS

T

-100

-4 t -b

0

Symmetrical

l

Low

Bath

ClCl-

Fig. 3. Current (I)-to-voltage (V) plot of outwardly rectifying (13/96 pS) Cl- channel in inside-out patches of basolateral membrane. Open symbols, data from channels studied in asymmetric solutions (i.e., 140 mM KC1 pipette and 140 mM NaCl in bath; n = 5). Closed symbols, data from same channels studied in presence of 140 mM KC1 in pipette solution and 40 mM NaCl and 200 mM mannitol in bath solution (i.e., a reduction of bath NaCl of 100 mM; n = 5).

highly selective Cl- channel. To characterize further the outwardly rectifying (13/96 pS) Cl- channel, we examined the effects of the Cl-channel blockers DIDS and NPPB and examined the effects of pH and Ca2+ activity. DIDS (10m4 M) reduced the P, from 0.69 t 0.12 to 0.07 t 0.01 (n = 5). NPPB also inhibited the channel. At a concentration of 2 x 10m5 M, NPPB reduced the P, to 0, whereas at 2 x 10V6 M, NPPB caused a flickering, partial block of the channel (the P, decreased from 0.85 t 0.08 to 0.14 t 0.10; P < 0.05; n = 4). Thus the Cl--channel blockers DIDS and NPPB inhibit the outwardly rectifying (13/96 pS) Cl- channel. The pH of the bath solution did not influence the P, of the outward rectifier. Changing the pH from 7.2 to 7.6 (HEPES buffer) did not alter the P, (P, in pH 7.2 was 0.80 t 0.08 and in pH 7.6 was 0.78 t 0.08; n = 3) Changing the Ca2+ activity of the bath solution also was’without effect on the P, of the channel. An increase in Ca2+ activity from 100 to 1,000 nM did not change the P, (P, in 100 nM was 0.75 t 0.05 and in 1,000 nM was 0.77 t

a F 4

I

PKA

2.5

pA

1

IN CCD

0.05; n = 5). Taken together, these experiments demonstrate that the outwardly rectifying (13/96 pS) Cl- channel is not regulated directly by cytosolic pH or Ca2+ activity. We next examined the effects of PKA on the Cl- channel in inside-out patches. As discussed above, the P, of these Cl- channels often increased, from a value near 0 in the cell-attached configuration to >0.5 with formation of the inside-out configuration (36 of 43 patches with channels). However, in 7 of 43 experiments, the P, did not increase following excision and remained below a value of 0.5. Because the response to PKA was dependent on the F0 in the control condition (i.e., following excision into the control bath solution containing 1 mM ATP), we report the effects of PKA on channels with PO < 0.5 separately from those with P, > 0.5. Figure 4 illustrates a typical experiment on an insideout patch containing a channel with a P, value near 0. Occasional channel openings were observed after formation of the inside-out configuration, but these are not included in this -lo-min current record. PKA (125 U/ml; 75 nM) rapidly and dramatically increased the P,. Addition of alkaline phosphatase to the bath solution returned the P, to near 0. These observations are consistent with the conclusion that PKA activates the channel by a phosphorylation-dependent mechanism. Figure 5 illustrates a typical experiment on a channel with a P, > 0.5. In this - 15-min current record it can be seen that the P, value before PKA treatment was greater than 0.5 and that neither PKA nor alkaline phosphatase changed the PO. Figure 6 summarizes the results of all experiments with PKA. In the subset of channels with a P, value in control ~0.5, PKA increased the P, from 0.17 t 0.08 to 0.40 t 0.15 (P C 0.05; n = 6). In three experiments alkaline phosphatase in the bath solution reversed the increase in P, elicited by PKA (P, declined from 0.43 t 0.03 to 0.26 t 0.06; P < 0.05). In contrast, in the subset of channels with a P, value in control ~0.5, PKA did not change the P, (P, was 0.77 -t 0.06 in control and 0.76 t 0.06; n = 5),

_ .

.

ALKAL

Fig. 4. Representative current record of outwardly rectifying (13/96 pS) Cl- channel [open probability (P,,) < 0.51 in inside-out patches of basolateral membrane (w lo-min record). Addition of PKA (75 nM) and alkaline phosphatase (2 U/ml) to bath solution (i.e., cytoplasmic surface of membrane) is indicated by arrows. Large deflections in current record are artifacts caused by addition of PKA and alkaline phosphatase to bath. In this experiment, single-channel current amplitude declined during continued exposure to PKA. This was not a consistent observation (see Fig. 5) and indicates that the outward rectifier may have substates. Because current amplitudes less than the full open state were infrequent in the channels studied, we could not determine whether the channel has substates. Downward deflections indicate channel openings. Holding voltage = -50 mV. ATP (1 mM) was present in bath solution throughout experiment.


CHLORIDE

CHANNELS

IN CCD

F247

rectifying (13/96 pS) channel. The 305pS channel was found exclusively in the apical membrane, whereas the 13/96-pS channel was only found in the basolateral memclosed brane. This observation indicates that RCCT-28A cells, open grown on glass cover slips coated with Vitrogen plating medium, are polar with respect to the distribution of Cl+ t ALKALINE PKA channels. In addition to this geographic separation, the PHOSPHATASE 5 PA I channels are differentially regulated by PKA; the kinase inactivates the 305pS channel, whereas it activates the 1 min outwardly rectifying (13/96 pS) channel. Fig. 5. Representative current record of an outwardly rectifying (13/96 The 305-pS Cl- channel2 is inactive in cell-attached pS) Cl- channel (P, > 0.5) in inside-out patches of basolateral mempatches of unstimulated cells. We found that neither isobrane (-20-min record). Addition of PKA (75 nM) and alkaline phosphatase (2 U/ml) to bath solution (i.e., cytoplasmic surface of memproterenol, added to cell-attached patches, nor PKA, brane) is indicated by arrows. Large deflections in current record are added to inside-out patches, activated quiescent chanartifacts caused by addition of PKA and alkaline phosphatase to bath. nels. This confirms our previous observation that CAMP Downward deflections indicate channel openings. Holding voltage = does not activate the channel in cell-attached patches -40 mV. ATP (1 mM) was present in bath solution throughout exper(22). Moreover, PKA reduced the P, of excision- and iment. voltage-activated 305-pS channels. Furthermore, the 1.ooo 1 305-pS Cl- channel has a linear I-V relation, whereas isoproterenol activated outwardly rectifying whole cell k--------A---co.750 1 1 Cl- currents. These observations are consistent with our --I iii conclusion that the 305-pS Cl- channel is not responsible 2 for the isoproterenol-induced increase in whole cell Clg 0.500 CL currents. I El In a previous study we showed that the 305-pS ClCL 00.250 channel is activated by adenosine via a sequential pathI way involving an A1 receptor, G protein, phospholipase C, 1 diacylglycerol, and protein kinase C (33). Because the 0.000 A PKA Alkaline Control channel is inactivated by PKA and activated by protein Phosphatase kinase C, we speculate that these kinases either have Fig. 6. Effects of PKA (75 nM) and alkaline phosphatase (2 U/ml) on P, different substrates or phosphorylate different sites on of outwardly rectifying (13/96 pS) Cl- channel in basolateral memthe 305-pS channel. In a preliminary study we showed brane. Circles, experiments where P, < 0.5 in control (n = 6); triangles, that the 305-pS channel is activated by exposure to a experiments where P, > 0.5 in control (n = 5). Lines connecting means different from control indicate experiments were paired. * Significantly hypotonic medium and that the channel plays an imporand alkaline phosphatase (P < 0.05). tant role in cell volume regulation (36, 37). Because the hypotonic medium increases diacylglycerol production in and, furthermore, alkaline phosphatase had no demonRCCT-28A cells (B. Stanton and P. Friedman, unpubstrable effect on the P, either before or after addition of lished observations) it is possible that cell swelling actiexogenous PKA. Except for the differential response to vates the 305-pS Cl- channel by a pathway involving PKA, we could not discern any other differences between diacylglycerol and protein kinase C. Thus we speculate Cl- channels with low and high P, values. Channels in that the 305-pS Cl- channel in the apical membrane both groups had similar slope conductances and ion se- plays an important role in cell volume regulation. lectivities. These results show that a subset of outwardly The outwardly rectifying (13/96 pS) Cl- channel we rectifying (13/96 pS) Cl- channels in the basolateral describe was found exclusively in the basolateral memmembrane are activated by PKA and likely represent the brane. The channel was highly selective for Cl- over catchannels activated in whole cell patch clamp experiments ions, and the permeability ratio of Cl- to gluconate was by isoproterenol, CAMP, and PKA (33). 11:l. Furthermore, the channel was inhibited by DIDS and NPPB and was insensitive to changes in cytoplasmic DISCUSSION pH and Ca2+ activity within the physiological range. This The major findings of this study are that PKA actichannel is very different from the 46-p& double-barrelled vates, by a phosphorylation-dependent mechanism, out- Cl- channel (23 pS/barrel) responsible for the Cl- conwardly rectifying (13/96 pS) Cl- channels in the basolatductance of the basolateral membrane of PC (26). The era1 membrane and inhibits 305~s Cl- channels in the 2 The 305-pS Cl- channel has a few similarities with the voltageapical membrane of RCCT-28A cells. Accordingly, we anion channel (VDAC) found in the outer membrane of conclude that the stimulation by isoproterenol of the dependent mitochondria, including a high single-channel conductance and voltwhole cell Cl- conductance in this cell line is referable to age-dependent gating (9). In contrast to the 305-pS Cl- channel in activation of the outwardly rectifying Cl- channel. The RCCT-28A cells, however, the VDAC has several unique characteristics including 1) poor selectivity to Cl- vs. K+ at +20 mV (PC1/PK = observation that whole cell Cl- currents (33) and single2:l) (at -20 mV VDAC is more permeable to K+ than to Cl-), 2) a high channel currents through the basolateral Cl- channel outsingle-channel conductance (500 to 1,000 pS), 3) a steep relationship wardly rectify are consistent with this conclusion. between membrane voltage and P,, 4) pH modulation of voltageOur study identifies two types of Cl- channels in RCdependent gating, and 5) complex kinetics of the open and closed states that depend on the length of channel inactivation. CT-28A cells, i.e., a 305~s channel and an outwardly

1

----*


F248

CHLORIDE

CHANNELS

46-pS channel has a linear I-V relation, is moderately selective to Cl- versus cations (i.e., El), and is voltage dependent, i.e., the channel open time decreases with depolarizing voltages. Furthermore, the 46-pS Cl- channel is insensitive to DIDS and NPPB, since these Cl-channel blockers failed to block the Cl- conductance of the basolateral membrane of PC (26). Thus the 46-pS Clchannel in the basolateral membrane of PC has very different characteristics from the outwardly rectifying (13/96 pS) Cl- channel in the basolateral membrane of RCCT-28A cells which have many phenotypic properties of A-IC. The outward rectifier in the basolateral membrane of RCCT-28A cells is very similar to the outward rectifier located in the apical membrane of airway epithelial cells (13, 1920, 29). Similarities include outward rectification of the I-V plot, single-channel conductance (-50 pS at 0 mV), activation by CAMP via PKA (see below), inhibition by DIDS and NPPB, high selectivity to Cl- versus cations, activation by formation of the inside-out configuration (i.e., excision activation) and by depolarization, and insensitivity to cytoplasmic pH and Ca2+ activity (13, 19, 29, 43). It is important to note that PKA activation of the outward rectifier in airway epithelial cells, as in the present study on collecting duct cells in culture, also occurs in a subset of channels (43,45). Furthermore, in cell-attached patches CAMP and epinephrine activate a subset of outward-rectifying Cl- channels in airway epithelial cells (43,45). In contrast, in a previous study we showed that protein kinase C consistently activates the 305-pS Cl- channel in the apical membrane of RCCT28A cells (33). The outward rectifier in RCCT-28A ceils was, like the outward rectifier in airway cells, frequently but not always activated by formation of the inside-out configuration (19, 44). Although the mechanism responsible for excision activation is unknown (19, 44), results from our experiments with PKA provide the basis for a hypothesis. Activation of quiescent channels by excision suggests that an inhibitor present in the cell-attached configuration normally keeps the channel closed. Excision may physically separate the inhibitor from the channel and thereby activate the channel. In 6 of 43 experiments, excision did not lead to activation of the channel (i.e., channels with P, c 0.5), suggesting that the putative inhibitor was present in the inside-out membrane patch. In these experiments, PKA increased the P, (6 of 6 cases), and alkaline phosphatase reversed the increase in P, induced by PKA. These observations indicate that, in these patches, the putative inhibitor may be inactivated by PKA by a phosphorylation-dependent mechanism. The observation that neither PKA nor alkaline phosphatase alters the P, of excision-activated channels (e.g., P, > 0.5) suggests that the inhibitor is not present in these excised patches. One potential candidate for a kinase-sensitive inhibitor of Cl- channels is a heterotrimeric G protein. G proteins inhibit a variety of ion channels (5) including Cl- channels in airway epithelial cells (32). G proteins are closely associated with Na+ and Ca2+ channels in plasm .a membranes (2, 11 35) Fu .rthermore, it has been shown that PKA, either di rectly * or indirectly,

IN CCD

phosphorylates G proteins and prevents G protein activation of their effecters (6, 7, 14,40). In the present case, phosphorylation may prevent the inhibitory action of the G protein, thereby stimulating the outwardly rectifying Cl- channel. Although our hypothesis states that there is one class of 13/96-pS Cl- channels (i.e., PKA activated and PKA insensitive), we cannot exclude the possibility that there are actually two classes of outwardly rectifying Cl- that have similar conductances and ion selectivities and yet differ with respect to the sensitivity to activation by excision, voltage, and PKA. Additional experiments are required to test our hypothesis and to elucidate the mechanism of excision activation and the variable response to PKA. The observation that PKA stimulates an outwardly rectifying (13/96 pS) Cl- channel in the basolateral membrane of RCCT-28A cells, a cell line with many phenotypic properties of A-IC, is relevant to A-IC function. Koeppen and co-workers (15, 24) have shown that isoproterenol, CAMP, and PKA increase an outwardly rectifying Cl- conductance in primary cultures of rabbit outer medullary collecting duct cells isolated from the inner stripe (OMCD;,). CAMP increases electrogenic proton secretion by the OMCDi,, a segment composed of acid-secreting cells (16). The CAMP-induced increase in proton secretion by OMCDi, may occur by several mechanisms including 1) direct stimulation of the H+-ATPase in the apical membrane, 2) stimulation of the Cl-HCO, exchanger in the basolateral membrane, and 3) stimulation of the Cl- conductance of the basolateral membrane, an effect that would facilitate Cl- exit from the cell and reduce cell Cl- activity. This, in turn, would stimulate the Cl--HCO, exchanger in the basolateral membrane and reduce intracellular pH, an effect that would increase electrogenic proton secretion. Our finding that isoproterenol, CAMP, and PKA increase an outwardly rectifying Cl- conductance in RCCT-28A cells suggests that CAMP may increase proton secretion in part by stimulating outwardly rectifying Cl- channels in the basolateral membrane. Additional experiments are required to determine whether isoproterenol and CAMP also increase proton secretion and Cl--HCO, exchange in RCCT-28A cells. In summary, we report that RCCT-28A cells, a cell line derived from rabbit CCD with many phenotypic properties of A-IC, selectively expresses 305-pS Cl- channels in the apical membrane and outwardly rectifying (13/96 pS) Cl- channels in the basolateral membrane. Furthermore, isoproterenol via CAMP and PKA stimulate the whole cell Cl- conductance by selectively activating the outward rectifier located in the basolateral membrane. Activation of Cl- channels by PKA may play an important role in proton secretion by this cell line. We thank Kathy Karlson and Sabine Diet1 for able technical assistance, Dr. Wenhui Wang and Erik Schwiebert for valuable discussions, and Drs. Neil Kizer and Bruce Koeppen for comments on the manuscript. Erik Schwiebert also performed several experiments with PKA. This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-34533 and was conducted during the tenure of B. A. Stanton as an American Heart Association Established Investigator. P. Diet1 was supported by a fellowship from


CHLORIDE the Max Kade Foundation. Present address of P. Dietl: Physiologisches Innsbruck, 6020 Innsbruck, Austria. Address reprint requests to B. A. Stanton. Received

14 January

1992; accepted

in final

form

Institut,

Universitat

4 March

1992.

CHANNELS

REFERENCES 1. Arend, L. J., J. S. Handler, J. S. Rhim, F. Gusovsky, and W. S. Spielman. Adenosine-sensitive phosphoinositide turnover in a newly established renal cell line. Am. J. Physiol. 256 (Renal Fluid Electrolyte Physiol. 25): F1067-F1074, 1989. 2. Ausiello, D. A., J. L. Stow, H. F. Cantiello, J. B. Almeida, and D. J. Benos. Purified epithelial Na+ channel complex contains the pertussis toxin-sensitive GaiM3 protein. J. Biol. Chem. 267: 4759-4765, 1992. E., and L. Cabell-Kluch. Membrane voltages 3. Bello-Reuss, and polarity of acid and lactate secretion by a peanut-lectin positive cell line (RCCT-28A) (Abstract). Kidney Int. 37: 533, 1990. 4. Bello-Reuss, E., P. Tyler, and L. Cabell-Kluch. Acidification mechanisms by the peanut-lectin (+) cell line RCCT-28A (P+) (Abstract). J. Am. Sot. Nephrol. 1: 646, 1990. 5. Brown, A. M., and L. Birnbaumer. Ionic channels and their regulation by G protein subunits. Annu. Reu. Physiol. 52: 197-213, 1990. 6. Bushfield, M., B. E. Lavan, and M. D. Houslay. Okadaic acid identifies a phosphorylation/dephosphorylation cycle controlling the inhibitory guanidine-nucleotide-binding regulatory protein Gi2. Biochem. J. 274: 317-321, 1991. 7. Bushfield, M., G. J. Murphy, B. E. Lavan, P. J. Parker, V. J. Hruby, G. Milligan, and M. D. Houslay. Hormonal regulation of Gi2 a-subunit phosphorylation in intact hepatocytes. Biochem. J. 268: 449-457, 1990. 8. Christine, C. W., F. H. Laskowski, A. H. Gitter, K. W. Beyenbach, P. Gross, and E. Fromter. Anion channels in the apical membrane of collecting duct principal cells in culture. Cell. Physiol. Biochem. 1: 76-88, 1991. 9. Colombini, M. Voltage gating in the mitochondrial channels, VDAC. J. Membr. Biol. 111: 103-111, 1989. 10. Dietl, P., N. Kizer, and B. A. Stanton. Conductive properties of a rabbit cortical collecting duct cell line: regulation by isoproterenol. Am. J. Physiol. 262 (Renal Fluid Electrolyte Physiol. 31): F578-F582, 1992. 11. Hamilton, S. L., J. Codina, M. J. Hawkes, A. Yatani, T. Sawada, F. Strickland, S. C. Froehner, A. M. Spiegel, L. Toro, E. Stefani, L. Birnbaumer, and A. M. Brown. Evidence for direct interaction of GSCY. with the Ca2+ channel of skeletal muscle. J. Biol. Chem. 266: 19528-19535, 1991. 12. Hawk, C. T., and J. A. Schafer. Effects of AVP and deoxycorticosterone on Na+ and water transport in the Dahl salt-sensitive rat CCD. Am. J. Physiol. 260 (Renal Fluid Electrolyte Physiol. 29): F471-F478, 1991. 13. Hwang, T.-C., L. Lu, P. L. Zeitlin, D. C. Gruenert, R. Huganir, and W. B. Guggino. Cl- channels in CF: lack of activation by protein kinase C and CAMP-dependent protein kinase. Science Wash. DC 244: 1351-1353, 1989. 14. Imaizumi, T., Y. Wantanabe, and H. Yoshida. Phosphorylation of Gi protein by cyclic AMP-dependent protein kinase inhibits its dissociation into a-subunits and &-subunits by Mg2+ and GTP+. Eur. J. PharmacoZ. 207: 189-194, 1991. 15. Koeppen, B., C. Pappas, T. Manger, and A. Oyler. Cellular mechanisms of Cl transport in outer medullary collecting duct. Kidney Int. 40, Suppl.: S131-S135, 1991. 16. Koeppen, B. M. Conductive properties of the rabbit outer medullary collecting duct: inner stripe. Am. J. Physiol. 248 (Renal Fluid EZectroZyte Physiol. 17): F500-F506, 1985. 17. Koeppen, B. M. Electrophysiological identification of principal and intercalated cells in the rabbit outer medullary collecting duct. Pfluegers Arch. 409: 138-141, 1987. 18. Koeppen, B. M., B. A. Biagi, and G. H. Giebisch. Intracellular microelectrode characterization of the rabbit cortical collecting duct. Am. J. Physiol. 244 (Renal Fluid Electrolyte Physiol. 13): F35-F47, 1983. 19. Kunzelmann, K., H. Pavenstadt, and R. Greger. Properties and regulation of chloride channels in cystic fibrosis and normal

IN CCD

F249

airway cells. Pfluegers Arch. 415: 172-182, 1989. 20. Li, M., J. D. McCann, C. M. Liedtke, A. C. Nairn, P. Greengard, and M. J. Welsh. Cyclic AMP-dependent protein kinase opens chloride channels in normal but not cystic fibrosis airway epithelium. Nature Lond. 331: 358-360, 1988. 21. Light, D. B., F. V. McCann, T. M. Keller, and B. A. Stanton. Amiloride-sensitive cation channel in apical membrane of inner medullary collecting duct. Am. J. Physiol. 255 (Renal Fluid Electrolyte Physiol. 24): F278-F286, 1988. 22. Light, D. B., E. M. Schwiebert, G. Fejes-Toth, A. NarayFejes-Toth, K. H. Karlson, F. V. McCann, and B. A. Stanton. Chloride channels in the apical membrane of cortical collecting duct cells. Am. J. Physiol. 258 (Renal Fluid Electrolyte Physiol. 27): F273-F280, 1990. 23. Muto, S., K. Yasoshima, K. Yoshitomi, M. Imai, and Y. Asano. Electrophysiological identification of alpha and beta-intercalated cells and their distribution along the rabbit distal nephron segments. J. CZin. Inuest. 86: 1829-1839, 1990. 24. Pappas, C. A., and B. M. Koeppen. P-Adrenergic regulation of Cl- currents in primary cultures of rabbit outer medullary collecting duct cells (OMCDi) (Abstract). J. Am. Sot. Nephrol. 2: 747, 1991. 25. Sansom, S., E. J. Weinman, and R. G. O’Neil. Microelectrode assessment of chloride-conductive properties of cortical collecting duct. Am. J. Physiol. 247 (Renal Fluid Electrolyte Physiol. 16): F291-F302, 1984. 26. Sansom, S. C., B.-Q. La, and S. L. Carosi. Double-barreled chloride channels of collecting duct basolateral membrane. Am. J. Physiol. 259 (Renal Fluid Electrolyte Physiol. 28): F46-F52, 1990. 27. Schlatter, E., R. Greger, and J. A. Schafer. Principal cells of cortical collecting ducts of the rat are not a route of transepithelial Cl- transport. Pfluegers Arch. 417: 317-323, 1990. 28. Schlatter, E., and J. A. Schafer. Electrophysiological studies in principal cells of rat cortical collecting tubules. ADH increases the apical membrane Na+-conductance. Pfluegers Arch. 409: Bl92, 1987. 29. Schoumacher, R. A., R. L. Shoemaker, D. R. Halm, E. A. Tallant, R. W. Wallace, and R. A. Frizzell. Phosphorylation fails to activate chloride channels from cystic fibrosis airway cells. Nature Lond. 330: 752-754, 1987. 30. Schuster, V. L. Bicarbonate reabsorption and secretion in the cortical and outer medullary collecting tubule. Semin. Nephrol. 10: 139-147, 1990. 31. Schuster, V. L., and J. B. Stokes. Chloride transport by the cortical and outer medullary collecting duct. Am. J. Physiol. 253 (Renal Fluid Electrolyte Physiol. 22): F203-F212, 1987. 32. Schwiebert, E. M., D. C. Gruenert, and B. A. Stanton. G protein regulation of chloride channels and whole cell chloride conductance in tracheal epithelial cells (Abstract). Pediatr. Pulmon. 6, Suppl.: 259, 1991. 33. Schwiebert, E. M., K. H. Karlson, P. A. Friedman, P. Dietl, W. S. Spielman, and B. A. Stanton. Adenosine regulates a chloride channel via protein kinase C and a G-protein in a rabbit cortical collecting duct cell line. J. Clin. Inuest. 89: 834-841, 1992. 34. Schwiebert, E. M., D. B. Light, G. Fejes-Toth, A. NarayFejes-Toth, and B. A. Stanton. A GTP-binding protein activates chloride channels in a renal epithelium. J. Biol. Chem. 265: 7725-7728, 1990. 35. Smith, P. R., G. Saccomani, E.-H. Joe, K. J. Angelides, and D. J. Benos. Amiloride-sensitive sodium channel is linked to the cytoskeleton in renal epithelial cells. Proc. Natl. Acad. Sci. USA 88: 6971-6975, 1991. 36. Stanton, B. A., S. Dietl, and E. Schwiebert. Cell volume regulation in the cortical collecting duct: stretch activated Cl channels (Abstract). J. Am. Sot. Nephrol. 1: 492, 1990. 37. Stanton, B. A., J. A. Mills, and E. M. Schwiebert. Role of the cytoskeleton in regulatory volume decrease in cortical collecting duct cells in culture (Abstract). J. Am. Sot. Nephrol. 2: 751, 1991. Y., and M. A. Knepper. Thiazide-sensitive NaCl ab38. Terada, sorption in rat cortical collecting duct. Am. J. Physiol. 259 (Renal Fluid Electrolyte Physiol. 28): F519-F528, 1990. 39. Tomita, K., J. J. Pisano, M. B. Burg, and M. A. Knepper. Effects of vasopressin and bradykinin on anion transport by rat


F250

CHLORIDE

CHANNELS

cortical collecting duct. Evidence for an electroneutral sodium chloride transport pathway. J. Clin. Inuest. 77: 136-141, 1986. 40. Watanabe, Y., T. Imaizumi, N. Misaki, K. Iwakura, and H. Yoshida. Effects of phosphorylation of inhibitory GTP-binding protein by cyclic AMP-dependent protein kinase on its ADPribosylation by pertussis toxin, islet activating protein. FEBS Lett. 236: 372-374, 1988. 41. Weiner, I. D., and L. L. Hamm. Regulation of intracellular pH in the rabbit cortical collecting tubule. J. Clin. Inuest. 85: 274-281, 1990.

IN CCD

42. Weiner, I. D., and L. L. Hamm. Regulation of Cl-/HCO, exchange in the rabbit cortical collecting tubule. J. CZin. Invest. 87: 1553-1558, 1991. 43. Welsh, M. J. Abnormal regulation of ion channels in cystic fibrosis epithelia. FASEB J. 4: 2718-2725, 1990. 44. Welsh, M. J., M. Li, and J. D. McCann. Activation of normal and cystic fibrosis Cl- channels by voltage, temperature, and trypsin. J. Clin. Invest. 84: 2002-2007, 1989. 45. Widdicombe, J. H., and J. J. Wine. The basic defect in cystic fibrosis. Trends Biochem. Sci. 16: 474-477, 1991.


Cation channels in basolateral membranes of sheep parotid secretory cells.

Chloride channels of biological membranes.

Nucleotide-activated chloride channels in lysosomal membranes.

Distinct mechanisms of zinc uptake at the apical and basolateral membranes of caco-2 cells.

Fatty acids inhibit apical membrane chloride channels in airway epithelia.

Small conductance chloride channels in the apical membrane of thyroid cells.

Turtle colon: keeping track of transporters in the apical and basolateral membranes.

Histamine-induced Cl- secretion in human nasal epithelium: responses of apical and basolateral membranes.

Asymmetrical distribution of G-proteins among the apical and basolateral membranes of rat enterocytes.

Localization of several G-protein subunits to the apical and basolateral membranes of cortical tubular cells from the rat kidney.

Chloride channels in the apical membrane of normal and cystic fibrosis airway and intestinal epithelia.

Cl- channels in basolateral renal medullary membranes, VI. Cl- conductance expression in Xenopus oocytes.

Hypercompliant apical membranes of bladder umbrella cells.

Differential effects of basolateral and apical iron supply on iron transport in Caco-2 cells.

Iterative sorting of apical and basolateral cargo in Madin-Darby canine kidney cells.

Differential effect of basolateral and apical adenosine on AVP-stimulated cAMP formation in primary culture of IMCD.

Microtubule- and motor-dependent fusion in vitro between apical and basolateral endocytic vesicles from MDCK cells.

Apical and basolateral transferrin receptors in polarized BeWo cells recycle through separate endosomes.

Inhibition of apical Na+ channels in rabbit cortical collecting tubules by basolateral prostaglandin E2 is modulated by protein kinase C.

Apical membrane ionic channels in the rabbit cortical thick ascending limb in primary culture.

TRPC1 regulates calcium-activated chloride channels in salivary gland cells.

Calcium and cAMP activate different chloride channels in the apical membrane of normal and cystic fibrosis epithelia.

Effect of nocodazole on vesicular traffic to the apical and basolateral surfaces of polarized MDCK cells.

A quantitative dot-blot immunoassay for integral membrane proteins: preparation of pancreatic plasma membranes containing apical and basolateral domains.