- Email: [email protected]
Characteristics of a rat cortical collecting duct cell line that maintains high transepithelial resistance
Kidney International, Vol. 50 (1996), pp. 36 7—3 76
LABORATORY_INVESTIGATION
Characteristics of a rat cortical collecting duct cell line that maintains high transepithelial resistance MARCEL BLOT-CHABAUD, MONIQUE LAPLACE, FRArrçoIsE CLUZEAUD, CLAUDIA CAPURRO, ROLAND CASSINGENA, ALAIN VANDEWALLE, NICOLE1TE FARMAN, and JEAN PIERRE BON VALET INSERM (1246, Institut Fédératif de Recherches "Cellules Epithéliales'c Faculté de Médecine Xavier Bichat, Parts, and CNRS UPR 42, Lahoratoire de Genetique Moléculaire, Villejuif France
Characteristics of a rat cortical collecting duct cell line that maintains high transepithelial resistance. This study describes the establishment of a rat kidney cortical collecting duct (CCD) clonal cell line (RCCDI cells)
responses have led to the necessity of developing CCD cell lines from a mammalian origin. Vandewalle et al [7] and Prié et a! [8] that maintains high transepithelial resistance and specific hormonal developed distal-like cell lines by SV4O transformation of primary sensitivities. Immortalized cells were obtained by infection of primary cultured CCD cells with the wild-type simian virus 40. Grown on Petri dishes, RCCD1 cells are organized as monolayers of cuboid cells separated by tight junctions and form domes. Grown on permeable filters, confluent 140 G RCCD1 cells exhibit high transepithelial resistance (R,: 2390 cm2), transepithelial potential difference (PD) of l0.5 1.2 mV lumen negative, an associated short-circuit current (Isc) of 4.3 0.5 sA1cm2, and
generated significant Nat, K, H and HC03 gradients, reflecting Na° and H reabsorption and K and HC01 secretion. RCCD1 cells exhibit features of both principal (PC) and intercalated (IC) cells. Consistent with
PC phenotype, about 50% of the cells were positively stained by a PC-specific agglutinin. In situ hybridization studies revealed the presence of a, f3 and y subunit mRNAs of the amiloride-sensitive epithelial Na°
channel and a and 13 subunits of Na-K-ATPase. Moreover, Na -KATPase was immunolocalized at the hasolateral side of the cells. Arginine vasopressin (AVP) induced a significant increase in both cellular cAMP
content and lsc. Amiloride decreased in a dose-dependent manner Isc from untreated and AVP-trcatcd RCCD1 cells. In addition, a bariumsensitive K conductance was evidenced in the apical side of the cells. Consistent with IC phenotype, isoproterenol (ISO) provoked a large
increase in cellular cAMP and stimulated lsc. The effect of ISO on Isc was blocked by 5 >< vi DPC, a chloride channel blocker. Finally, AVP plus
I0
ISO had additive effect on Isc. Taken together, these results provide evidence that the RCCD1 cell line has maintained many of the original properties of rat CCD from which they were derived.
culture of rabbit renal cortical cells and by further selection of distal as opposed to proximal or ascending limb cells. However, grown on porous filters, all these cell lines developed only low transepithelial resistance. Arend et al [9] also established a rabbit cell line that did not result in a tight epithelium. Using another strategy, Stoos et al [10] first established a mouse cortical collecting duct cell line (M1 cells) derived from microdissected CCD of a transgenie mouse carrying the early DNA region of the large T antigen gene. This cell line exhibited a transepithelial resistance ranging between 750 and 950 11 cm2 and vectorialized electrolyte transport properties. However, most of the ion transport studies
on intact CCD have been performed in the rat and most tools, such as antibodies or cDNAs, are available essentially for this species. For these reasons, we sought to establish a rat CCD cell line. This was achieved by transformation of pure rat CCD cells in primary culture with the large T antigen for the simian virus 40 (5V40). This study describes the establishment of a rat clonal CCD cell
line that has both principal and intercalated cell phenotypes, referred to as RCCD1 cells, which exhibit high transepithelial resistance. These cells retain the main features of the parental CCD from which they derive. Such a model of cortical collecting
duct should represent a powerful tool to study ion transport The cortical collecting duct (CCD) plays a major role in the properties and regulatory pathways of this part of the nephron. control of electrolyte reabsorption and secretion in the final Methods portion of the nephron. Ion transport across this epithelium is under the control of several hormones including corticosteroid Primary culture of cortical collecting duct cells
and polypeptide hormones. Since mammalian renal cell lines such CCD cells were isolated using a Percoll gradient technique as MDCK or LLC-PK1 cells express physiological properties of both proximal and distal segments of the nephron [—I, amphib- adapted from Vinay et al [11] and Bello-Reuss and Weher [12]. ian epithelial cell lines like A6 or TBM cells have been used for a Kidneys were removed under sterile conditions from male Spralong time as a model for the mammalian CCD [5, 61. However, gue-Dawley rats (180 to 200 g), decapsulated, and kept at 4°C in species differences between amphibian and mammalian specific a modified Krehs solution containing HBSS medium (Gihco, Les
Ulis, France) with 5 mvi glucose, 2 mtvi glutamine, 10 U/mI
Received for publication October 20, 1995 and in revised form February 20, 1996 Accepted for publication February 20, 1996
© 1996 by the International Society of Nephrology
penicillin-streptomycin and 15 mvi HEPES, pH 7.4. For each cell preparation 12 kidneys were used. Cortices were sliced, rinsed in the modified Krebs solution at 4°C and then incubated in a 1:1
(vol/vol) modified Krebs solution and defined culture medium containing 0.75 mg/mI collagenase (Serva, France) and 2.5 mg/mI bovine serum albumin (BSA, Sigma) for one hour at 37°C under
367
368
Blot-Chabaud et al: Immortalized rat CCD cells
95% air—5% CO2 atmosphere. The defined culture medium (DM) was as follows: HAM's F12, DMEM 1:1; 14 mM NaHCO3; 2 m glutamine; 5 g/m1 transferrin; 5 g/ml insulin; 10 jg/m1 EGF; 5 x 10 M T3; 10 U/mI penicillin-streptomycin; 20 mM HEPES, pH 7.4. Slices were rinsed and centrifuged (1500 rpm for 5 mm) three times. The resulting pellet was then resuspended in 50 mg/ml BSA modified Krebs solution for five minutes at 4°C. After centrifugation (1500 rpm for 5 mm), the pellet was resuspended in a 60% modified Krebs solution 40% Percoll solution and centrifuged at 12,000 g for 30 minutes. This procedure resulted in the separation of tubules in four bands (Fl to F4). The lowest density Fl band was removed, washed three times in modified Krebs solution and passed through a 53 m nylon mesh to remove glomeruli. Small tubule fragments exhibiting the characteristic appearance of collecting ducts (CCD) with minor contamination (<3%) by glomer-
RCCD1 cells grown on glass coverslips and fixed with acetone for 10 minutes at 4°C. Cells were then washed in PBS and sequentially
reach confluence. Primary cultures of CCD obtained by this
extensive washings in PBS. In all cases, specimens were examined and photographed with Zeiss microscope equipped with epifluorescence optics. Cell monolayers labeled with the MCK1 antibody were also examined by confocal laser scanning microscopy using a Leica TCS4D apparatus (Leica, Germany), and quantification of
incubated for 30 minutes at room temperature with Syrian hamster anti-SV4O antibody [71 and fluorescein isothyocyanateconjugated rabbit antiserum to hamster gamma-globulins. Each incubation step was followed by extensive washings in PBS. Immunofluorescence studies were performed on cells grown on glass slides. For staining with Dolichos biflorus agglutinin (DBA; Vector), cells were fixed with periodate-lysine-paraformadehyde
(PLP) for 30 minutes at room temperature, then washed and incubated at room temperature for one hour with biotinylated DBA and then for one hour with avidin D RITC. The expression of the a1 subunit of the Na°,K-ATPase were
examined on cells grown on glass slides covered with collagen. Cells were fixed with methanol for 10 minutes at room temperauli or proximal tubules were obtained. These CCD fragments ture. Cells were incubated with the MCK1 antibody (provided by were then seeded on collagen-coated (type I from rat tail, Institut Dr. K. Sweadner), for two hours, then with a sheep anti-mouse Jacques Boy, France) 25 cm2 flasks and cultured for six to seven biotinylated Ig for one hour and streptavidin FITC for another days in the DM described above in order to allow CCD cells to hour at room temperature. Each incubation step was followed by technique exhibited a high transepithelial resistance (1000 to 1800 fl cm2), morphological appearance and hormonal responses very
similar to that of the intact CCD (data not shown).
SV4O injection of cortical collecting duct cells in primary culture
Confluent cells were trypsinized, seeded (106 cells/dish) on 60 mm diameter Petri dishes (Falcon) and grown in DM for seven
days. Subconfluent cells were rinsed twice with medium and infected with the wild-type SV4O strain LP (100 plaque forming
units/cell). After two hours at 37°C, the cells were washed,
the staining was performed using an image analyzer (Optilab, Macintosh, Graftek, France). In situ hybridization
For in situ hybridization, RCCD1 cells were grown on glass slides covered with collagen. Cells were fixed for 15 minutes in 4%
covered with medium and reincubated at 37°C in defined culture paraformaldehyde with 5 mrvi MgCI2 and then kept at 4°C in 70% medium containing 2% fetal bovine serum (FBS; Gibco). Three ethanol. Prehybridization and post-hybridization steps were as weeks after seeding, foci of transformed cells could be observed. previously described [13, 14]. Cell treatment with proteinase K (20 They were isolated by trypsinization and several colonies were mg/mi) was followed by an acetylation step (0.1 M triethano-
expanded in separate dishes. Four clones were obtained and lamine, 0.025% acetic anhydride). After PBS and 0.9% NaCI subcultured in 2% FBS-supplemented DM. Clones were then rinsing, cells were dehydrated and dried. Hybridization mix was selected for dome formation on non-porous support, development of a transepithelial resistance, and for ultrastructural morphology matching with that of the primary culture. One clone, referred to as RCCD1, was selected. RCCD1 cells were routinely grown and subcultured in 25 cm2 flasks. For experiments, cells were seeded either on petri dishes, Transwell filters or Snapwell filters (Costar)
then spread over the cells and covered with Paraflim. Cells were hybridized for 16 to 18 hours at 50°C. Posthybridization treatment consisted of an initial wash in 5 >< saline sodium citrate (SSC: 150 mM NaC1, 15 m sodium citrate), 10 mi 1,4-dithiothreitol (DTT) at 50°C, a high-stringency wash in 50% formamide, 2 X SSC, 0.1
M DiT at 65°C for 20 minutes, and two 10 minutes washes in
or on glass slides coated with collagen. In all cases the 2% NaCl-hydroxymethyl aminomethane (Tris)-EDTA (0.5 M NaCI, FBS-supplemented DM was changed every two days. Motphological and immunocytochemical studies
Confluent RCCD1 cells grown on collagen-coated petri dishes
were examined and photographed on the stage on an inverted microscope equipped with a phase contrast device (Axiovert 10, Zeiss, Germany). For transmission electron microscopy, cells were grown on collagen-treated transwell filters. Cells were first rinsed with phosphate-buffered saline (PBS) and then fixed for one hour with 2.5% giutaraldehyde in PBS at room temperature. Cells were then washed in PBS, postflxed with 1% osmic acid for 15 minutes, dehydrated in graded series of ethanois and embedded in Epon. Ultrathin sections were performed on transversally oriented confluent monolayers and examined with a Philips EM 410 electron microscope.
Immunolabeling of the large T antigen was performed on
10 tOM Tris HCI, 5 mivi EDTA) at 37°C. Ribonuciease A treatment
(20 mg/mi) was then performed at 37°C for 20 minutes. After rinsing with NaCI-Tris-EDTA and 0.1 X SSC for 15 minutes, sections were dehydrated in graded ethanols (containing ammonium acetate 0.3 M) and dried. Kodak NTB2 film (melted at 42°C)
was applied to the slides, dried, and exposed at —20°C for 3 (Na ,K-ATPase) or 30 days (Na -channel). The film was developed (Kodak D19) and fixed (Kodak fixer). At the end of the autoradiographic process the cells were stained with toluidine blue; specimens were photographed and signal quantification was performed using an image analyzer (Optilab).
Linearized rat al- and f31-ATPase cDNAs (al- and 1-cDNA provided by Dr. J. Lingrel) subcloned into Bluescribe or Bluescript were used to synthetize antisense or sense cRNA probes. The 3'-noncoding sequence (that presents little homology with other isoforms) of a1-ATPase and 1-ATPase cDNAs was used.
369
Blot-Chabaud et al: immortalized rat CCD cells
W2. sflSt sa r— aan
C flwS-
- w. .. —
pI-
E
-Ii'
—
-,- 7I -* Ii —'-1
-
a
4'
jf;y -
I -
- - -ç
—
t.
aI
— ---t
Fig. 1. Morphological features of RCCDJ cultured cells. Confluent RCCD1 exhibited 100% nuclear large T antigen positivity (A). When grown on petri
dishes, cells formed layers of epithelioid-shaped cells and formed domes (B). When grown on filters, RCCD1 cells grew as monolayer of cuboid cells (D and E). Note that some cells exhibited numerous apical microprojections and presented differences in cytoplasmic appearance reminiscent of dark and light cells. Closely apposed cells were separated by tight junctions (arrow, E'). RCCD1 cells grown on petri dishes were labeled with DBA as described in the Methods section (C). Note that 50% of RCCD1 cells exhibited positive staining. Bars: A, 10 sm; B, 50 jsm; C, 10 m; D, E and E', 1 pm.
Part of the 3' untranslated regions of a-, J3- and y-cDNA subunits of the rat epithelial sodium channel (rENaC; provided by Dr. B.
Rossier) subcloned into the Bluescript vector were used. After
cAMP assay
Confluent RCCD1 cells (passages 15 to 30) grown on collagencoated 12-well trays were incubated overnight in minimum meSP6, or T3 polymerases. 35S-labeled uridine 5'-triphosphate dium (DM devoided of FBS, hormones and growth factors). Cells (1,000 Ci/mmol) was from Amersham, and the other reagents were then first incubated in minimum medium containing 0.1 mM (adenosine, guanosine, cytosine 5'-triphosphate, ribonucleasin, 3-isobutyl 1-methyl xanthine (IBMX) for 20 minutes at 37°C. DTT, and RNA polymerases) were from Promega. The hybrid- Thereafter, cells were incubated in 2 ml of the same minimum ization mix was 50% formamide (Fluka), 1 ms'i DIT (Boehr- medium supplemented or not with AVP or ISO (Sigma) for seven inger), 2 x SSC, 10% dextran sulfate (Pharmacia), I mg/mi minutes at 37°C. The amounts of hormones are given in the text salmon sperm DNA (Sigma), and the 355-iabeled cRNA. Imme- and the Figure legends. The reaction was stopped by a rapid diately before hybridization, this mix was denatured at 80°C for removal of the medium, immediately followed by the addition of five minutes. 1 ml ice-cold 95% ethanol—5% formic acid solution. After 30 linearization, 35S-iabeied RNA probes were synthetized using T7,
370
Blot-Chabaud et al. Immortalized rat CCD cells
6
170
* ÷
5
25
***
5
E
20
165
Fig. 2. Distribution of Na , K, H4 and HC04 concentrations in the apical and basolateral media bathing RCCDJ cells. The concentrations of Na K, H4 and HC03 were determined in the apical (A) and hasolateral (B) media of RCCD1 cells cultured for 48 hours on Transwell filters with a 2% FBS-supplemented DM. Na and H concentrations were higher in the B medium than in the A medium, whereas K4 and HCO3concentrations were higher in the A medium than in the B one. °P < oo5, ***p < 0.001 hasolateral versus apical medium values.
minutes of incubation at 4°C, the supernatant was evaporated and
vials were frozen at —80°C. Levels of cellular cAMP were determined in the deproteinated supernatants using a radioimmunoassay kit (Institut Pasteur, Lyon, France). All experiments were carried out in duplicate or triplicate. The number of cells per well was determined in parallel.
Table 1. Expression of mRNA encoding for a, f3, y subunits of the subunits of Na,K4-ATPase in RCCD1 cells
Na channel and Na4 channel
a = 14.28 f3 = 1.11
y=
1.02
(1.51 (3)
1.14 (3) 0.30 (3)
Na,K4-ATPase a4 = 11.00 2.50 (4) = 15.25
3.15 (9)
Values are expressed in arbitrary units of specific labeling measured by
Electrophysiological studies
differences between results obtained with antisense and sense probes. Results are the mean SE of N) experiments.
The development of confluence of RCCD1 cells was monitored by measuring the transepithelial voltage VT (mV) and transepithelial resistance RT (f1 cm2) accross the Transwell or Snapwell Statistical analysis filters using a WPI epithelial volt-ohm meter (World Precision Results are expressed as mean values SE. Statistical analysis Instruments, New Haven, CT, USA) connected to sterile elec- was performed using Student's t-test for paired or unpaired data. trodes. For the measurement of short circuit current (Isc), V. and RT Results RCCD1 cells (passages 15 to 30) were cultured on collagen-coated
Snapwell filters. Five to eight days after seeding, filters were incubated overnight in minimum medium and then mounted into a voltage clamp system (Costar; WPI). This voltage-current clamp was used to measure the open-circuit VT, RT and lsc (i.tAIcm2).
Establishment of the RCCDJ cell line
The cloned RCCD1 cells exhibited stable nuclear large T
antigen expression. Results from indirect immunofluorescence using a specific anti-large T antigen antibody indicated that all nuclei were positively labeled (Fig. IA). When grown on petri lsc was measured by clamping VT to 0 mV for one second (clamping V.r to 0 mV for longer times, that is, 5 to 10 minutes, dishes, RCCD1 cells had epithelioid shape and formed domes resulted in similar results) and RT was determined from the (Fig. 1B), which was considered to be features of transporting current deflection in response to a —51+5 mV on V offset of VT epithelia in culture [15]. When grown on filters, the cells formed lasting one second. The polarity of V. measurements was given monolayers of closely apposed cuboid cells (Fig. I D, E) separated using the apical side as the ground. Cells were bathed on each side by junctional complexes (Fig. 1E'). Results from electron microswith 8 ml minimum medium thermostated at 37°C that was copy showed that the cell layers consisted of at least two different cell types showing structural features of principal (with light circulated by a gaslift (95% 02—5% CO2 mixture). appearance) and intercalated (with dark appearance) cells (Fig. 1 D, E) [16, 171. To better evidence these different cells types, Determination of ion concentrations immunocytochemical experiments were performed with DBA, which binds essentially to principal cells of the rat CCD {18J. As The concentrations of Ht HC03, Na4 and K4 were deter- shown in Figure 1C, about 50% of the cells were labeled with mined in apical and basolateral media of RCCD1 cells cultured DBA, suggesting that RCCD1 cells were composed of about 50% for 48 hours on Transwell filters with a 2% FBS-supplemented PC. When grown on filters with 2% FBS-supplemented DM for DM. pH and HCO3 concentration were determined using a five to seven days followed by withdrawal of serum and growth blood gas analyzer (Radiometer). Na and K4 concentrations factors during 12 to 15 hours, RCCD1 cells exhibited baseline were determined by flame photometry. values for VT, RT and Isc of —10.5 1.2 mV lumen negative, 2390
371
Blot-Chabaud et al: Immortalized rat CCD cells
9 •ak•••• •"T•. -It
- :.j,• :
•I
-
D
0 0
C.) 4—
C.)
0,
E
100
100
80
80
0 C.) o
60
.4-
60
40
(1)
40
20
20
II
0 10
io— 106 iü— io—
0 108 io— 106
Amiloride, M Fig.
io
io—4
Ouabain, M
3. Presence of amiloride-sensitive Na channels and NatK-ATPase in RCCDJ cells. Detection of the mRNA of the a-subunit of the
amiloride-sensitive sodium channel by in situ hybridization in RCCD cells (A and B). Illustrations of labeling using antisense and sense probes. A clear signal was observed with antisense probe (A) whereas only background signal could be observed with sense probe (B). Immunocytochemical detection of the cs-subunit of Na',KtATPase in RCCD1 cells (C) and reconstructed view of a transversal section analyzed by confocal laser microscopy (C'). Note that Na,K-ATPase was essentially located in the basolateral domain of RCCDJ cells. Dose-dependent effects of amiloride and ouabain on short circuit current in RCCD1 cells (D and E). Increased concentrations of amiloride were successively added to the apical or basal sides of the cells (D). Apical addition () induced a large fall in Isc, whereas no effect was observed with the basolateral treatment (U). Increasing concentrations of ouahain in the basal bath also induced a large fall in lsc (E). Bars: A, B, C,C', 10 m. Values are the mean SE from four separate experiments.
160 cz
0 () 'i5
140 [ cm2 and 4.3
Ami 105M
120
0.5 pA/cm2, respectively. Accordingly, all
experiments were performed in these conditions in order to evaluate hormonal modulation of electrolyte transport. Finally, RCCD1 cells exhibited a prolonged lifespan since to date they have been subcultured up to the 45th passage.
BaCI2 5•10-3M
80
d 40
RCCD1 cells exhibit polarized ion transport processes
C),
0
To examine whether ion transport could occur in RCCD1 cells, the ionic concentrations from apical and basolateral media were determined for RCCD1 cells cultured during 48 hours on TransTime, minutes well filters. Results are given in Figure 2. In the apical medium, Fig. 4. Effect of amilonde and barium on Isc in RCCDJ cells. Confluent concentrations of K and HCO3 were significantly higher than RCCD1 cells grown on Snapwell filters were used for short-circuit in the basolateral medium, whereas concentrations of Na and experiments. Control value of Isc was 3.2 0.6 AJcm2. The addition of H were significantly lower. The occurrence of ion concentration iO El amiloride on the apical side induced a rapid and large decrease in Isc. The addition of 5 X iO El BaCl2 apically resulted in a further gradients strongly suggests that RCCD1 cells reabsorb Na and
0
20
40
60
80
100
120
decrease in Isc. Values are means SE from 3 separate experiments.
H and secrete K and HCO3.
372 I)
0 0
Blot-Chabaud et al: Immortalized rat CCD cells
25
A
200
20 15
.
E N S
10
0 0)
E
5
0
0
180
140
Fig. 5. Effects of arginine vasopressin (A VP) on CAMP and short circuit current ([sc) in RCCDJ cells. A. cAMP was measured on confluent RCCD1 cells grown on petri dishes and incubated without or with various
120
concentrations of AVP in the presence of iO M IBMX. Values are the mean SE from 3 to
160
06 6Q 000
0
AVP, M
160
00 .4-.
A
0
00
0
Ami 1 0-5M
+AVP 10-6M
U)
subunits of the Na channel or Na,K-ATPase. As shown in
80
of the probes was assessed by using sense probes. Figure 3
0 B
illustrates the results provided by in situ hybridization using the Na channel a subunit antisense (Fig. 3A) and sense (Fig. 3B)
20 40 60 80
120
1O6M
40 0
20
C
40 60 80 +DPC 5•10-3M
100
probes. Results from immunofluorescence studies using the MCK1 antibody against Na,KtATPase and confocal laser microscopy analysis also showed that > 95% of the staining was observed in the basolateral membrane domain of RCCD1 cells (Fig. 3C and 3C'). Consistent with these results Isc was sensitive to amiloride and ouabain. Figure 3D illustrates the dose-dependent effect on Isc of amiloride added either to the apical side or
4,+Ami
80
120
100 120 140 160
105M
0
basal side of RCCD1 cells. When added to the basal side, amiloride did not modify Isc. By contrast, apical addition of 120
+AVP 10-6M
100 80
0 (I)
ments were performed to detect mRNA encoding for the different
Table 1, a, /3 and y mRNAs encoding for the subunits of the Na channel were expressed in RCCD1 cells, a1 and 13) mRNA isoforms of Na,K-ATPase were also expressed. The specificity
160
0
from 4 separate experiments.
120
40 -
.
side of the cells. Values are the mean SE
Concentration, M
0
0U)
0N
0)
6 determinations. B. Isc was measured on RCCD1 cells grown on Snapwell filters. Cells were exposed to increasing concentrations of AVP (D) or dDAVP () added to the basal
concentrations of amiloride resulted in a progressive decrease in Isc with half maximal inhibition occuring at .10_6 M (Fig. 3D). Additional experiments performed in a medium devoid of Na confirmed that a large part of Isc was dependent on Na entry (not shown). Addition of ouabain to the basal side of the cells resulted in a decrease of Isc with half maximal inhibition of Isc occurring at more than i0— M (Fig. 3E).
In addition, apical addition of 5 X i0 M barium (BaCl2)
60 40 0
20
40
60
80
100
120
Time, minutes
significantly decreased the amiloride-insensitive Isc by - 36%, in addition to the effect of amiloride (Fig. 4). These results suggest K channels are present in the apical membrane of the cells, although one cannot totally exclude an indirect effect due to the leak of BaCI2 through the junctionnal pathways leading to the blockade of basolaterally located K channels.
Fig. 6. Influence of amiloride on Isc before and after treatment with atginine vasopressin (A VP) in RCCDJ cells. A. The control value of Isc was 3.4 + 0.3 pA/cm2. The addition of 10—6 M AVP to the basolateral side of RCCD1
cells induced a rapid increase in Isc that was sustained for at least 80 minutes. The addition of io M amiloride on the apical side of the cells induced a rapid decrease in Isc. B. When amiloride was added before AVP
administration, only a very small increase in Isc was observed after the addition of AVP on the basal side of the cells. C. The AVP-induced increase in Isc was not modified when DPC 5 X iO M was added before AVP. Values are the mean SE from 4 separate experiments.
RCCDJ cells are sensitive to arginine vasopressin
Results from cAMP measurements indicated that AVP increased the cellular cAMP content of RCCD1 cells (Fig. 5A). In
basal conditions, the cellular cAMP content measured in the presence of IBMX, art inhibitor of phosphodiesterase, was 2.51
0.80 pmol/7 minutes/106 cells. AVP induced a dose-dependent increase in cAMP with half maximal stimulation at .10_8 M (Fig. 5A).
Results from short circuit current experiments showed that, In CCD, the reabsorption of sodium implies Na entry via after a control period, 10—6 M AVP added to the basolateral side amiloride-sensitive apical sodium channels and Na extrusion of RCCD1 cells significantly increased Isc after less than five through basolateral Na,K-ATPase [19]. Therefore, experi- minutes, and then remained constant for up to 80 minutes
373
Blot-Chabaud et al: Immortalized rat CCD cells
A
B
175 150
160
S
125
140
E
100
0 E 0.
75
130 120
C.)
N S
150
50
Fig. 7. Effects of isoproterenol (ISO) on cAMP and short circuit current (Isc) in RCCDJ cells. cAMP assay and Isc experiments were performed on RCCD1 cells grown as described in Figure 4. A. cAMP was measured in the
110 100
25 C.)
ci - 0)
absence of isoproterenol or in the presence of various concentrations of the hormone. Values are the mean SE from 3 to 5 determinations. B. The addition of increasing concentrations of ISO to the basal side of cells induced an important and sustained increase in Isc. Values
C
C?
Li)
are the mean SE from 3 separate
Isoprotereflol, M
Isoproterenol, M
0.3; +AVP, 4.4 0.2 JLA/cm2; N = 8, P < 0.001; Fig. 6A). The addition of i0— M amiloride on the apical side
experiments.
(control, 3.3
resulted in a rapid fall in Isc (Fig. 6A). When RCCD1 cells were first incubated with i0— M amiloride, AVP induced only a small rise in Isc as compared to its effects in the absence of amiloride (Fig. 6B). Finally, when RCCD1 were first incubated with 5 x
iO M DPC, the addition of AVP resulted in a rapid and large increase in Isc (Fig. 6C). Both AVP and dDAVP, a V2 agonist [201, stimulated Isc with maximal effect at iO M (Fig. 5B). The effect of dDAVP on Isc was greater than that obtained with AVP but the difference did not reach statistical significance. For both hormones, the concentration required to obtain half maximum stimulation of Isc was — 4 >< 10_li
0
200 160
-Iso 10-6 M
A
4,+DPC 5•10 M
120
80
40 Cl)
0
0
20
60
40
80
100
120
B
M.
In additional experiments, the effect of prostaglandin E2 (PGE2) was tested. After treatment of RCCD1 cells with 10 M AVP, Isc increased from 4.5 1.4 to 5.9 1.5 aA/cm2. A further addition of 10-8 M PGE2 to the basal side of the cells significantly decreased Isc to 4.7 1.5 tAJcm2 (NS vs. control; P < 0.01 vs.
AVP treatment).
I
2E
o U)
40 0
RCCDJ cells are sensitive to isoproterenol
Results from immunocytochemistry showed that about 50% of RCCD1 had positive staining for DBA (Fig. 1C). It suggests that the DBA-negative cells correspond to IC. Several studies previously demonstrated that IC are mainly sensitive to f3-adrenergic agents [21]. To test whether the RCCD1 cell line was sensitive to
adrenergic agents, the effects of isoproterenol, a /3-adrenergic agonist acting on IC from CCD [21], was tested on cAMP content
and short circuit current (Fig. 7). ISO strongly stimulated the production of cAMP from RCCDI cells (Fig. 7A). As shown in Figure 8A, 10 M ISO added to the basal side of the cells also rapidly increased Isc. Thereafter, the ISO—elicited Isc remained in plateau (control, 5.0 0.5; + ISO, 7.4 1.1 rAJcm2; N = 6, P <
0.01). The further addition of 5 X i0 M DPC, a blocker of Cl channels [22], on the apical side of the cells induced a rapid fall of Isc (Fig. 8A). DPC (5 >< io M) added alone to the apical side of the cells also decreased Isc (Fig. 8B). In this case, the subsequent
addition of 10_6 M ISO induced only a weak increase in Isc. Finally, when amiloride was first added to the apical side of RCCD1 cells, the addition of isoproterenol led to a large increase in Isc (Fig. 8C). As already observed for AVP, ISO significantly
0
20
40
20
40
60
80
100
120
C
. 0 0
°0
240 160 80
C.)
U)
0
0
60
80
100
120
Time, minutes Fig. 8. Effects of DPC on Isc in RCCD1 cells before and after treatment with
isoproterenol (ISO) in RCCDJ cells. The influence of DPC on Isc was tested before and after addition of 106 M ISO to the basal side of RCCD1 cells.
A. The control value of lsc was 3.8 + 1.1 pAJcm2. The addition of IO M DPC to the apical side of the cells blunted the stimulated effect of ISO. B. When DPC was added first, almost no effect of isoprotenerol could be observed. C. The ISO-induced increase in Isc was not prevented when amiloride iO M was added before ISO. Values are the mean SE from 3 separate experiments.
374
Blot-Chabaud et at: Immortalized rat CCD cells
increased Isc in a dose-dependent manner (Fig. 7B). The concentration of isoproterenol required to obtain a half maximal stimulation of Isc was — 10 M.
200
The effects of AVP and ISO on Isc were additive, since the addition of i0 M ISO after treatment by 10_6 M AVP led to a
120
160
further increase in Isc (Fig. 9). Thereafter, successive additions of
10 NI amiloride, 5 X 10 M DPC, and 5 X i0 M BaCI2 to the apical side of the cells led to an almost complete suppression of Isc (Fig. 9). Discussion
This study describes the functional properties of a new rat cortical collecting duct clonal cell line established by infection of a primary culture of rat kidney cortical collecting ducts with the
wild-type SV4O. To establish the RCCD1 cell line, we used a primary culture of rat CCD obtained using a Percoll gradient technique modified from Vinay et a! [11] and Bello-Reuss et a! [12]. This technique allows an enriched preparation of CCD (> 97%) to be obtained for cell culture. Infection of these cells with the wild-type SV4O permitted us to develop a clonal cell population with the same morphological and functional features as those of primary cultured cells used for viral infection (data not shown).
The strategy of SV4O infection of primary cultured cells has already been used to establish several cell lines in various organs [23, 24], including the kidney [7—9, 25]. These studies have shown
that the presence of the SV4O large T antigen coding sequences was essential to the transformation process. In RCCD1 cells, the viral large T antigen is expressed since 100% of the cells were positively stained (Fig. 1A). Successive passages of RCCD1 cells
resulted in the maintenance of the expression of the large T antigen. RCCD1 cells have maintained many of the electrophysiological and functional properties of the parental CCD [26, 27]. In addition, contrasting with previously established CCD cell lines either derived from rabbit or mouse kidneys [7—10], RCCD1 cells
+lso 106M +DPC 5•1 -3 M +BaCL2 5•1 o— M
80 (2 (I)
40 0 0
20
40
60
80 100 120 140
Time, minutes Fig. 9. Arginine vasopressin (A VP) and isoproterenol (ISO) have additive effects on Isc in RCCDJ cells. Confluent RCCD1 cells were monitored for Isc as described in the legend of Figures 6 and 8. The control value of Isc was 4.4 + 0.8 jsAjcm2. In addition to the effect of i0 M AVP, 106 M ISO
induced a further increase in Isc. Isc could be reduced by successive addition of amiloride (Ami), diphenylamine 2-carboxylate (DPC) and BaCl2. Note that in the presence of these three inhibitors Isc was almost totally blocked. Results are the mean SE from 3 separate experiments.
a-, f3- and y-subunits of the epithelial sodium channel [31] as well as mRNA of a1- and /31-isoforms of the a and 13 subunits of the
Na,K4-ATPase [321. Whereas a1- and /31-subunits mRNA of Na,K4-ATPase are expressed at approximatively the same level, it appears that 13- and y-subunits mRNAs of the Na channel may be 10 times less abundant than a-subunit mRNA. Direct comparison of the absolute level of expression between subunit mRNAs is difficult. However, in the cortical collecting duct, it was estimated that all three subunit mRNAs were at comparable levels, using the same in situ hybridization technique and the same probes [31]. The low signals obtained with /3 and y probes, as compared to a probe for the sodium channel may thus be related to culture conditions and/or to the immortalization process. The expression of Na',K-ATPase is also demonstrated at the protein
develop very high transepithelial resistance (ranging between 1800 and 4000 (1 cm2) that can be maintained after long term level by immunocytochemical detection with an antibody against passages.
the a1-subunit of the Na,K-ATPase [33]. In addition, the
Results from ion concentrations measurements and from mor- presence of an apical K conductance sensitive to barium [28] is phological and immunocytochemical studies indicate that RCCD1 evidenced by Isc experiments. The RCCD1 cell line retain responsiveness to AVP, another cells are polarized. RCCD1 cells create concentration gradients for Na4, K, HCO3 and H4 (Fig. 2). This indicates that cells are property of the PC of the CCD [261. This is evidenced by a rapid able to reabsorb Na4 and H and secrete K4 and HC03. In increase in both cAMP content and Isc following AVP treatment addition, they exhibit typical junctional complexes and basolater- [27, 34]. AVP induces a sustained increase in Isc [35], dependent ally-located Na pumps. Additional evidence of the polarization on a rapid increase in the apical sodium entry [36] and with an state of the cells is provided by the results from short circuit apparent K112 at 4 x 10 M AVP. The dose-dependence curve current experiments: RCCD1 cells exhibit amiloride-sensitive for the V7 agonist of AVP, dDAVP, shows that the effect of Na channels [19] and K conductanccs blocked by barium [28] dDAVP on lsc is somewhat higher than that observed with AVP. on their apical surface. The observed half maximum increases in Finally, the addition of I 0 M exogenous PGE2 inhibits the lsc obtained for various concentrations of ouahain and amiloride AVP-induced increase in lsc. One should notice that higher indicate that Na4 channels from RCCD1 cells have high affinity concentrations of AVP are necessary to induce increase in cellular for amiloride, whereas Na,K4-ATPase has low affinity for cAMP than those required to stimulate Isc. Such differences, also ouabain. These results are in complete agreement with previous observed with isoproterenol, have already been reported in other types of cultured epithelial cells [24]. Since the culture conditions studies obtained in microdissected rat CCD [29, 30]. Ultrastructural studies, DBA binding studies and Isc experi- influence the differentiation state of cultured epithelial cells [7],
ments support the notion that the RCCDJ cell line express such dissociated dose-dependent effects in AVP and ISO refeatures of at least two different cells types that may correspond to sponses may he due to the fact that Isc was measured on cells the principal and intercalated cells constituting the CCD [16—18, grown on filters, whereas the cAMP content was determined on 21]. Principal cells (PC) of RCCD1 retain most of the properties cells grown on petri dishes. While the RCCD1 cell line appears to possess features of PC of the PC of CCD in vivo. Indeed RCCD1 cells express mRNAs of
Blot-Chabaud et al: Immortalized rat CCD cells
from intact CCD, it fails to respond to aldosterone by an increase
in Isc (data not shown). This constrasts with the maintained sensitivities to AVP. One cannot exclude that the SV4O-transformation process may have deleterious consequences on mineralo-
corticoid receptor expression. In fact, no evidence of a highly conserved response to aldosterone in SV4O-transformed distal or CCD cell lines has been reported [7—10, 37]. Alternatively, the
culture conditions may be not optimum for the expression of mineralocorticoid receptors and/or the response to this hormone
375
3. ROY C, PRESTON AS, HANDLER JS: Insulin and serum increase the number of receptors for vasopressin in a kidney-derived line of cells grown in a defined medium. Proc NatI Acad Sci USA 77:5979—5983, 1980 4. GSTRAUNTHALER G, PFALLER W, KOTANKO, P: Biochemical charac-
terization of renal epithelial cell cultures (LLC-PKI and MDCK). Am J Physiol 248:F536—F544, 1985 5. HANDLER JS, STEELE RE, SM-JIB MK, WADE JB, PRESTON AS, LAWSON NL, JOHNSON JP: Toad urinary bladder epithelial cells in
culture: Maintenance of epithelial structure, sodium transport, and response to hormones. Proc NatlAcad Sci USA 76:4151—4155, 1979
[381.
6. PERKIN FM, HANDLER JS: Transport properties of toad kidney
specific marker for rat IC, equivalent to peanut lectin in the rabbit
7. VANDEWALLE A, LELONGT B, GENITEAU-LEGENDRE M, BAU000IN B, ANTOINE M, ESTRADE S, CHATELET F, VERROUST P, CASSINGENA R,
The RCCD1 cell line also exhibits features of IC. Since no
[39], is available, DBA was used in this study. Half of RCCD1 cells
were not stained by DBA, a result compatible with those from cAMP assays and short circuit current experiments. This is also in agreement with ultrastructural studies which indicated the presence of both dark (IC) and light (PC) cells [16]. In addition, Isc experiments clearly indicate that AVP and ISO have cumulative effects and modulate different transport pathways; this argues for
epithelia in culture. Am J Physiol 241:C154—C159, 1981
RoNco P: Maintenance of proximal and distal cell functions in SV4O-transformed tubular cell lines derived from rabbit kidney cortex. J Cell Physiol 141:203—221, 1989 8. PRIE D, RoNco PM, BAUDOUIN B, GENITEAU-LEGENDRE M, AroINE M, PIEDAGNEL R, ESTRADE S, LELONGT B, VERROUST P, CASSINGEENA
R, VANDEWALLE, A: Activation of the simian virus 40 (SV4O) genome
abrogates sensitivity to AVP in a rabbit collecting tubule cell line by repressing membrane expression of AVP receptors. J Cell Biol 113:
the presence of both cell types in the RCCD1 cell, although we 951—962, 1991 cannot totally exclude that RCCD1 cells would be constituted of 9. AREND LI, HANDLER JS, RrnM JS, GUSOVSKY F, SPIELMAN WS: Adenosine-sensitive phosphoinositide turnover in a newly established only one cell type with hybrid characteristics of both PC and IC. renal cell line. Am J Physiol 256:F1067—F1074, 1989 ISO exerts a rapid increase in Isc with an apparent K112 at BA, NARAY-FEJES-TOTH A, CARRETERO OA, ITO S, FEJE5M, which is mediated by an increase in apical Cl conductance 10. STOOS Tom G: Characterization of a mouse cortical collecting duct cell line. (Fig. 8). This observation is in close agreement with previous Kidney mt 39:1168—1175, 1991 studies [27, 40] and suggests that IC from RCCD1 cells mainly 11. VINAY P, Gouooux A, LEMIEUX G: Isolation of a pure suspension of rat proximal tubules. Am J Physiol 241:F403—F411, 1981 correspond to 13-IC [21]. This statement is also supported by the
10
fact that RCCD1 have been shown to secrete HC03 and reabsorb H (Fig. 2), a characteristic of J3-IC. Fejes-Toth and Naray-Fejes-Toth [41] showed that a conversion of 13-IC to cs-IC
and PC may occur in CCD cultured cells. Although one cannot exclude that such a phenomenon also occurs in RCCD1 cells, results from Isc experiments strongly suggest that this cell line still possesses /3-IC. In conclusion, this new cell line appears to be the first rat CCD
cell line with high transepithelial resistance and properties of vectorial transport close to that of intact rat CCD. Therefore, it could represent an attractive model to study ion transport and its control by polypeptide hormones in cortical collecting duct cells.
12. BELLO-REUSS E, WEBER MR: Electrophysiological studies of primary
cultures of rabbit distal tubule cells. Am J Physiol 252:F899—F909, 1987 13. Cox KH, DELEON DV, ANGERER LM, ANGERER RC: Detection of mRNAs in urchin embryos by in situ hybridization using asymmetric RNA probes. Dev Biol 101:485—502, 1984 14. FARMAN N, CORTHESY-THEULAZ I, BONVALET JP, ROSSIER BC: Localization of cs-isoforms of Na-K-ATPase in rat kidney by in situ hybridization. Am J Physiol 260:C468—C474, 1991 15. CEREIJIDO ME, R0BBINS 5, DOLEN WJ, ROTUNNO CA, SABATINI DD:
Polarized monolayers formed by epithelial cells on a permeable and translucent support. J Cell Biol 77:853—880, 1978 16. STANTON BA, BIEMESDERFER D, WADE JB, GIEBIscH U: Structural
and functional study of the rat distal nephron: Effects of potassium adaptation and depletion. Kidney mt 19:36—48, 1988 17. H0LTHOFER H, SCHULTE BA, PASTERNACK G, SIEGEL GJ, SPtCER SS:
Acknowledgments This work was supported by INSERM U246. We are grateful to Dr.
Sweadner (Harvard Medical School) who kindly provided the antiNa,K-ATPase antibody. The rat a1- and f31-ATPase cDNAs were a generous gift from Dr. Lingrel (University of Cincinnati). The rat a, /3, and 'y sodium channel cDNAs were kindly provided by Dr. Rossier (Institut de Pharmacologie, Lausanne). We thank the laboratory of blood gas analysis
(X. Bichat Hospital) for pH and HC01 concentration determinations and Mrs. M. Ferrera for determination of Na and K concentrations. We thank T. Carison for editing the manuscript. We also thank Mrs. V. Levêque and C. Tritscher for secretarial assitance. Reprint requests to Marcel Blot-C'hahaud, Ph.D., INSERM U246, Faculté de Médecine Xavier Bichat, BP 416, 75870 Paris Ceder 18, France.
References 1. GAUSCH CR, HARD WL, SMITH TF: Characterization of an established line of canine kidney cells (MDCK). Proc Soc Exp Biol Med 122:931— 935, 1966 2. Hvu. RN, CHERRY R, WEAVER OW: The origin and characteristics of a pig kidney cell strain LLCPK. In Vitro 12:670—677, 1976
Three distinct cell populations in rat kidney collecting duct. Am J Physiol 253:C323—C328, 1987 18, HOLTI-IOFER H, SCHULTE BA, SPICER SS: Expression of binding Sites
for dolichos biflorus agglutinin at the apical aspect of collecting duct cells in rat kidney. Cell Tissue Res 249:481—485, 1987 19. STOKES JB: Ion transport by the cortical and outer medullary collecting tubule. Kidney mt 22:473—484, 1982 20. BREYER MD, ANDO Y: Hormonal signaling and regulation of salt and water transport in collecting duct. Annu Rev Physiol 56:711—739, 1994 21. SCHUSTER VL: Organization of collecting duct intercalated cells. Kidney mt 38:668—672, 1990 22. STuns MJ, HENKE DC, BOUcHER RC: Diphenylamine-2-carboxylate (DPC) inhibits C1 conductance and cyclooxygenase of canine tracheal epithelium. Pfidgers Arch 415:611—616, 1990 23. ISOM HC, TEUETHIA MJ, TAYLOR JM: Transformation of isolated rat hepatocytes with simian virus 40. J Cell Biol 85:651—659, 1980 24. HERMAN P CASSINGENA R, FRIEDLANDER G, SOLaR P, GRODET A,
TRAN BA HUY P, AMIEL C: Middle ear cell line that maintains vectorial electrolyte transport. J Cell Physiol 154:615—622, 1993 25. Scorr DM, MACDONALD C, BREZSKI H, KINNE R: Maintenance of expression of differentiated function of kidney cells following transformation by SV 40 early region DNA. Exp Cell Res 162:391—398, 1986 26. BURG MB, lsCsoN L, GRANTHAM I, ORLOFF J: Electrical properties
376
Blot-Chabaud et al: Immortalized rat CCD cells
of isolated perfused rabbit renal tubules. Am J Physiol 215:788—794, 1968
27. MOREL F: Sites of hormone action in the mammalian nephron. Am J Physiol 240:F159—F164, 1981
28. GIEBISCH G: Renal potassium transport, in Membrane Transport in Biology, edited by GIEBISCH 0, TOSTESON D, USSING HH, Berlin, Springer Verlag, 1978, IV. A, p 215
cultures of rabbit cortical collecting duct cells.AmfPhysiol 262:F454— F461, 1992 35. CHEN L, WILLIAMS SK, SCHAFER JA: Differences in synergistic actions
of vasopressin and deoxycorticosterone in rat and rabbit CCD. Am J Physiol 259:F147—F156, 1990 36. SCHLATrER E, SCHAFER JA: Electrophysiological studies in principal
cells of rat cortical collecting tubules. ADH increases the apical membrane Na- conductance. Pflugers Arch 409:81—92, 1987
29. SARIBAN-SOHRABY S, BEN05 DJ: The amiloride sensitive Na channel. Am J Physiol 250:C175—C190, 1986 30. SWEADNER KJ: Isozymes of the Na-K-ATPase. Biochim Biophys
37. RAFESTIN-OBLIN ME, FARMAN N, CASSINGENA R, RONCO P, VANDE-
Acta 998:185-220, 1989 31. Duc C, FARMAN N, CANESSA CM, BONVALET JP, ROSSIER BC: Cell-specific expression of epithelial sodium channel a, /3, -y subunits
1993 38. HUSTED RF, MATSUSHITA K, STOKES JB: Induction of resistance to
in aldosterone-responsive epithelia from the rat: Localization by in situ hybridization and immunocytochemistry. J Cell Biol 127:1907— 1921, 1994 32. FARMAN N, COUTRY N, LOGVINENKO N, BLOT-CHABAUD M, BouR-
BouzE R, BONVALET JP: Adrenalectomy reduces a1 and not /3
Na-K-ATPase mRNA expression in rat distal nephron. Am J Physiol 263:C810—C817, 1992 33. FELSENFELD DP, SWEADNER KJ: Fine specificity mapping and topog-
raphy of an isozyme-specific epitope of the rat Na,K-ATPase catalytic subunit. J Biol Chem 263:10932—10942, 1988 34. CANE55A CM, SCHAFER JA: AVP stimulates Na transport in primary
WALLE A: Mineralocorticoid receptors in SV-40 transformed tubule cell lines derived from rabbit kidney. J Ster Bioch Mol Biol 44:45—52,
mineralocorticoid hormone in cultured inner medullaxy collecting duct cells by TGF-/31. Am J Physiol 36:F767—F775, 1994 39. L HIR M, KAISSLING B, KOEPPEN BM, WADE JB: Binding of peanut lectin to specific epithelial cell types in kidney. Am J Physiol 242:C1 17— C120, 1982 40. LIGHT DB, SCHWIEBERT EM, FEJES-TOTH G, NARAY-FEJES-TOTH A, KARLSON KH, MCCANN FV, STANTON BA: Chloride channels in the
apical membrane of cortical collecting duct cells. Am J Physiol 258:F273—F280, 1990 41. FEJES-TOTH G, NARAY-FEJES-TOTH A: Differentiation of renal /3-in-
tercalated cells to a-intercalated and principal cells in culture. Proc NatI Acad Sci USA 89:5487—5491, 1992
LABORATORY_INVESTIGATION
Characteristics of a rat cortical collecting duct cell line that maintains high transepithelial resistance MARCEL BLOT-CHABAUD, MONIQUE LAPLACE, FRArrçoIsE CLUZEAUD, CLAUDIA CAPURRO, ROLAND CASSINGENA, ALAIN VANDEWALLE, NICOLE1TE FARMAN, and JEAN PIERRE BON VALET INSERM (1246, Institut Fédératif de Recherches "Cellules Epithéliales'c Faculté de Médecine Xavier Bichat, Parts, and CNRS UPR 42, Lahoratoire de Genetique Moléculaire, Villejuif France
Characteristics of a rat cortical collecting duct cell line that maintains high transepithelial resistance. This study describes the establishment of a rat kidney cortical collecting duct (CCD) clonal cell line (RCCDI cells)
responses have led to the necessity of developing CCD cell lines from a mammalian origin. Vandewalle et al [7] and Prié et a! [8] that maintains high transepithelial resistance and specific hormonal developed distal-like cell lines by SV4O transformation of primary sensitivities. Immortalized cells were obtained by infection of primary cultured CCD cells with the wild-type simian virus 40. Grown on Petri dishes, RCCD1 cells are organized as monolayers of cuboid cells separated by tight junctions and form domes. Grown on permeable filters, confluent 140 G RCCD1 cells exhibit high transepithelial resistance (R,: 2390 cm2), transepithelial potential difference (PD) of l0.5 1.2 mV lumen negative, an associated short-circuit current (Isc) of 4.3 0.5 sA1cm2, and
generated significant Nat, K, H and HC03 gradients, reflecting Na° and H reabsorption and K and HC01 secretion. RCCD1 cells exhibit features of both principal (PC) and intercalated (IC) cells. Consistent with
PC phenotype, about 50% of the cells were positively stained by a PC-specific agglutinin. In situ hybridization studies revealed the presence of a, f3 and y subunit mRNAs of the amiloride-sensitive epithelial Na°
channel and a and 13 subunits of Na-K-ATPase. Moreover, Na -KATPase was immunolocalized at the hasolateral side of the cells. Arginine vasopressin (AVP) induced a significant increase in both cellular cAMP
content and lsc. Amiloride decreased in a dose-dependent manner Isc from untreated and AVP-trcatcd RCCD1 cells. In addition, a bariumsensitive K conductance was evidenced in the apical side of the cells. Consistent with IC phenotype, isoproterenol (ISO) provoked a large
increase in cellular cAMP and stimulated lsc. The effect of ISO on Isc was blocked by 5 >< vi DPC, a chloride channel blocker. Finally, AVP plus
I0
ISO had additive effect on Isc. Taken together, these results provide evidence that the RCCD1 cell line has maintained many of the original properties of rat CCD from which they were derived.
culture of rabbit renal cortical cells and by further selection of distal as opposed to proximal or ascending limb cells. However, grown on porous filters, all these cell lines developed only low transepithelial resistance. Arend et al [9] also established a rabbit cell line that did not result in a tight epithelium. Using another strategy, Stoos et al [10] first established a mouse cortical collecting duct cell line (M1 cells) derived from microdissected CCD of a transgenie mouse carrying the early DNA region of the large T antigen gene. This cell line exhibited a transepithelial resistance ranging between 750 and 950 11 cm2 and vectorialized electrolyte transport properties. However, most of the ion transport studies
on intact CCD have been performed in the rat and most tools, such as antibodies or cDNAs, are available essentially for this species. For these reasons, we sought to establish a rat CCD cell line. This was achieved by transformation of pure rat CCD cells in primary culture with the large T antigen for the simian virus 40 (5V40). This study describes the establishment of a rat clonal CCD cell
line that has both principal and intercalated cell phenotypes, referred to as RCCD1 cells, which exhibit high transepithelial resistance. These cells retain the main features of the parental CCD from which they derive. Such a model of cortical collecting
duct should represent a powerful tool to study ion transport The cortical collecting duct (CCD) plays a major role in the properties and regulatory pathways of this part of the nephron. control of electrolyte reabsorption and secretion in the final Methods portion of the nephron. Ion transport across this epithelium is under the control of several hormones including corticosteroid Primary culture of cortical collecting duct cells
and polypeptide hormones. Since mammalian renal cell lines such CCD cells were isolated using a Percoll gradient technique as MDCK or LLC-PK1 cells express physiological properties of both proximal and distal segments of the nephron [—I, amphib- adapted from Vinay et al [11] and Bello-Reuss and Weher [12]. ian epithelial cell lines like A6 or TBM cells have been used for a Kidneys were removed under sterile conditions from male Spralong time as a model for the mammalian CCD [5, 61. However, gue-Dawley rats (180 to 200 g), decapsulated, and kept at 4°C in species differences between amphibian and mammalian specific a modified Krehs solution containing HBSS medium (Gihco, Les
Ulis, France) with 5 mvi glucose, 2 mtvi glutamine, 10 U/mI
Received for publication October 20, 1995 and in revised form February 20, 1996 Accepted for publication February 20, 1996
© 1996 by the International Society of Nephrology
penicillin-streptomycin and 15 mvi HEPES, pH 7.4. For each cell preparation 12 kidneys were used. Cortices were sliced, rinsed in the modified Krebs solution at 4°C and then incubated in a 1:1
(vol/vol) modified Krebs solution and defined culture medium containing 0.75 mg/mI collagenase (Serva, France) and 2.5 mg/mI bovine serum albumin (BSA, Sigma) for one hour at 37°C under
367
368
Blot-Chabaud et al: Immortalized rat CCD cells
95% air—5% CO2 atmosphere. The defined culture medium (DM) was as follows: HAM's F12, DMEM 1:1; 14 mM NaHCO3; 2 m glutamine; 5 g/m1 transferrin; 5 g/ml insulin; 10 jg/m1 EGF; 5 x 10 M T3; 10 U/mI penicillin-streptomycin; 20 mM HEPES, pH 7.4. Slices were rinsed and centrifuged (1500 rpm for 5 mm) three times. The resulting pellet was then resuspended in 50 mg/ml BSA modified Krebs solution for five minutes at 4°C. After centrifugation (1500 rpm for 5 mm), the pellet was resuspended in a 60% modified Krebs solution 40% Percoll solution and centrifuged at 12,000 g for 30 minutes. This procedure resulted in the separation of tubules in four bands (Fl to F4). The lowest density Fl band was removed, washed three times in modified Krebs solution and passed through a 53 m nylon mesh to remove glomeruli. Small tubule fragments exhibiting the characteristic appearance of collecting ducts (CCD) with minor contamination (<3%) by glomer-
RCCD1 cells grown on glass coverslips and fixed with acetone for 10 minutes at 4°C. Cells were then washed in PBS and sequentially
reach confluence. Primary cultures of CCD obtained by this
extensive washings in PBS. In all cases, specimens were examined and photographed with Zeiss microscope equipped with epifluorescence optics. Cell monolayers labeled with the MCK1 antibody were also examined by confocal laser scanning microscopy using a Leica TCS4D apparatus (Leica, Germany), and quantification of
incubated for 30 minutes at room temperature with Syrian hamster anti-SV4O antibody [71 and fluorescein isothyocyanateconjugated rabbit antiserum to hamster gamma-globulins. Each incubation step was followed by extensive washings in PBS. Immunofluorescence studies were performed on cells grown on glass slides. For staining with Dolichos biflorus agglutinin (DBA; Vector), cells were fixed with periodate-lysine-paraformadehyde
(PLP) for 30 minutes at room temperature, then washed and incubated at room temperature for one hour with biotinylated DBA and then for one hour with avidin D RITC. The expression of the a1 subunit of the Na°,K-ATPase were
examined on cells grown on glass slides covered with collagen. Cells were fixed with methanol for 10 minutes at room temperauli or proximal tubules were obtained. These CCD fragments ture. Cells were incubated with the MCK1 antibody (provided by were then seeded on collagen-coated (type I from rat tail, Institut Dr. K. Sweadner), for two hours, then with a sheep anti-mouse Jacques Boy, France) 25 cm2 flasks and cultured for six to seven biotinylated Ig for one hour and streptavidin FITC for another days in the DM described above in order to allow CCD cells to hour at room temperature. Each incubation step was followed by technique exhibited a high transepithelial resistance (1000 to 1800 fl cm2), morphological appearance and hormonal responses very
similar to that of the intact CCD (data not shown).
SV4O injection of cortical collecting duct cells in primary culture
Confluent cells were trypsinized, seeded (106 cells/dish) on 60 mm diameter Petri dishes (Falcon) and grown in DM for seven
days. Subconfluent cells were rinsed twice with medium and infected with the wild-type SV4O strain LP (100 plaque forming
units/cell). After two hours at 37°C, the cells were washed,
the staining was performed using an image analyzer (Optilab, Macintosh, Graftek, France). In situ hybridization
For in situ hybridization, RCCD1 cells were grown on glass slides covered with collagen. Cells were fixed for 15 minutes in 4%
covered with medium and reincubated at 37°C in defined culture paraformaldehyde with 5 mrvi MgCI2 and then kept at 4°C in 70% medium containing 2% fetal bovine serum (FBS; Gibco). Three ethanol. Prehybridization and post-hybridization steps were as weeks after seeding, foci of transformed cells could be observed. previously described [13, 14]. Cell treatment with proteinase K (20 They were isolated by trypsinization and several colonies were mg/mi) was followed by an acetylation step (0.1 M triethano-
expanded in separate dishes. Four clones were obtained and lamine, 0.025% acetic anhydride). After PBS and 0.9% NaCI subcultured in 2% FBS-supplemented DM. Clones were then rinsing, cells were dehydrated and dried. Hybridization mix was selected for dome formation on non-porous support, development of a transepithelial resistance, and for ultrastructural morphology matching with that of the primary culture. One clone, referred to as RCCD1, was selected. RCCD1 cells were routinely grown and subcultured in 25 cm2 flasks. For experiments, cells were seeded either on petri dishes, Transwell filters or Snapwell filters (Costar)
then spread over the cells and covered with Paraflim. Cells were hybridized for 16 to 18 hours at 50°C. Posthybridization treatment consisted of an initial wash in 5 >< saline sodium citrate (SSC: 150 mM NaC1, 15 m sodium citrate), 10 mi 1,4-dithiothreitol (DTT) at 50°C, a high-stringency wash in 50% formamide, 2 X SSC, 0.1
M DiT at 65°C for 20 minutes, and two 10 minutes washes in
or on glass slides coated with collagen. In all cases the 2% NaCl-hydroxymethyl aminomethane (Tris)-EDTA (0.5 M NaCI, FBS-supplemented DM was changed every two days. Motphological and immunocytochemical studies
Confluent RCCD1 cells grown on collagen-coated petri dishes
were examined and photographed on the stage on an inverted microscope equipped with a phase contrast device (Axiovert 10, Zeiss, Germany). For transmission electron microscopy, cells were grown on collagen-treated transwell filters. Cells were first rinsed with phosphate-buffered saline (PBS) and then fixed for one hour with 2.5% giutaraldehyde in PBS at room temperature. Cells were then washed in PBS, postflxed with 1% osmic acid for 15 minutes, dehydrated in graded series of ethanois and embedded in Epon. Ultrathin sections were performed on transversally oriented confluent monolayers and examined with a Philips EM 410 electron microscope.
Immunolabeling of the large T antigen was performed on
10 tOM Tris HCI, 5 mivi EDTA) at 37°C. Ribonuciease A treatment
(20 mg/mi) was then performed at 37°C for 20 minutes. After rinsing with NaCI-Tris-EDTA and 0.1 X SSC for 15 minutes, sections were dehydrated in graded ethanols (containing ammonium acetate 0.3 M) and dried. Kodak NTB2 film (melted at 42°C)
was applied to the slides, dried, and exposed at —20°C for 3 (Na ,K-ATPase) or 30 days (Na -channel). The film was developed (Kodak D19) and fixed (Kodak fixer). At the end of the autoradiographic process the cells were stained with toluidine blue; specimens were photographed and signal quantification was performed using an image analyzer (Optilab).
Linearized rat al- and f31-ATPase cDNAs (al- and 1-cDNA provided by Dr. J. Lingrel) subcloned into Bluescribe or Bluescript were used to synthetize antisense or sense cRNA probes. The 3'-noncoding sequence (that presents little homology with other isoforms) of a1-ATPase and 1-ATPase cDNAs was used.
369
Blot-Chabaud et al: immortalized rat CCD cells
W2. sflSt sa r— aan
C flwS-
- w. .. —
pI-
E
-Ii'
—
-,- 7I -* Ii —'-1
-
a
4'
jf;y -
I -
- - -ç
—
t.
aI
— ---t
Fig. 1. Morphological features of RCCDJ cultured cells. Confluent RCCD1 exhibited 100% nuclear large T antigen positivity (A). When grown on petri
dishes, cells formed layers of epithelioid-shaped cells and formed domes (B). When grown on filters, RCCD1 cells grew as monolayer of cuboid cells (D and E). Note that some cells exhibited numerous apical microprojections and presented differences in cytoplasmic appearance reminiscent of dark and light cells. Closely apposed cells were separated by tight junctions (arrow, E'). RCCD1 cells grown on petri dishes were labeled with DBA as described in the Methods section (C). Note that 50% of RCCD1 cells exhibited positive staining. Bars: A, 10 sm; B, 50 jsm; C, 10 m; D, E and E', 1 pm.
Part of the 3' untranslated regions of a-, J3- and y-cDNA subunits of the rat epithelial sodium channel (rENaC; provided by Dr. B.
Rossier) subcloned into the Bluescript vector were used. After
cAMP assay
Confluent RCCD1 cells (passages 15 to 30) grown on collagencoated 12-well trays were incubated overnight in minimum meSP6, or T3 polymerases. 35S-labeled uridine 5'-triphosphate dium (DM devoided of FBS, hormones and growth factors). Cells (1,000 Ci/mmol) was from Amersham, and the other reagents were then first incubated in minimum medium containing 0.1 mM (adenosine, guanosine, cytosine 5'-triphosphate, ribonucleasin, 3-isobutyl 1-methyl xanthine (IBMX) for 20 minutes at 37°C. DTT, and RNA polymerases) were from Promega. The hybrid- Thereafter, cells were incubated in 2 ml of the same minimum ization mix was 50% formamide (Fluka), 1 ms'i DIT (Boehr- medium supplemented or not with AVP or ISO (Sigma) for seven inger), 2 x SSC, 10% dextran sulfate (Pharmacia), I mg/mi minutes at 37°C. The amounts of hormones are given in the text salmon sperm DNA (Sigma), and the 355-iabeled cRNA. Imme- and the Figure legends. The reaction was stopped by a rapid diately before hybridization, this mix was denatured at 80°C for removal of the medium, immediately followed by the addition of five minutes. 1 ml ice-cold 95% ethanol—5% formic acid solution. After 30 linearization, 35S-iabeied RNA probes were synthetized using T7,
370
Blot-Chabaud et al. Immortalized rat CCD cells
6
170
* ÷
5
25
***
5
E
20
165
Fig. 2. Distribution of Na , K, H4 and HC04 concentrations in the apical and basolateral media bathing RCCDJ cells. The concentrations of Na K, H4 and HC03 were determined in the apical (A) and hasolateral (B) media of RCCD1 cells cultured for 48 hours on Transwell filters with a 2% FBS-supplemented DM. Na and H concentrations were higher in the B medium than in the A medium, whereas K4 and HCO3concentrations were higher in the A medium than in the B one. °P < oo5, ***p < 0.001 hasolateral versus apical medium values.
minutes of incubation at 4°C, the supernatant was evaporated and
vials were frozen at —80°C. Levels of cellular cAMP were determined in the deproteinated supernatants using a radioimmunoassay kit (Institut Pasteur, Lyon, France). All experiments were carried out in duplicate or triplicate. The number of cells per well was determined in parallel.
Table 1. Expression of mRNA encoding for a, f3, y subunits of the subunits of Na,K4-ATPase in RCCD1 cells
Na channel and Na4 channel
a = 14.28 f3 = 1.11
y=
1.02
(1.51 (3)
1.14 (3) 0.30 (3)
Na,K4-ATPase a4 = 11.00 2.50 (4) = 15.25
3.15 (9)
Values are expressed in arbitrary units of specific labeling measured by
Electrophysiological studies
differences between results obtained with antisense and sense probes. Results are the mean SE of N) experiments.
The development of confluence of RCCD1 cells was monitored by measuring the transepithelial voltage VT (mV) and transepithelial resistance RT (f1 cm2) accross the Transwell or Snapwell Statistical analysis filters using a WPI epithelial volt-ohm meter (World Precision Results are expressed as mean values SE. Statistical analysis Instruments, New Haven, CT, USA) connected to sterile elec- was performed using Student's t-test for paired or unpaired data. trodes. For the measurement of short circuit current (Isc), V. and RT Results RCCD1 cells (passages 15 to 30) were cultured on collagen-coated
Snapwell filters. Five to eight days after seeding, filters were incubated overnight in minimum medium and then mounted into a voltage clamp system (Costar; WPI). This voltage-current clamp was used to measure the open-circuit VT, RT and lsc (i.tAIcm2).
Establishment of the RCCDJ cell line
The cloned RCCD1 cells exhibited stable nuclear large T
antigen expression. Results from indirect immunofluorescence using a specific anti-large T antigen antibody indicated that all nuclei were positively labeled (Fig. IA). When grown on petri lsc was measured by clamping VT to 0 mV for one second (clamping V.r to 0 mV for longer times, that is, 5 to 10 minutes, dishes, RCCD1 cells had epithelioid shape and formed domes resulted in similar results) and RT was determined from the (Fig. 1B), which was considered to be features of transporting current deflection in response to a —51+5 mV on V offset of VT epithelia in culture [15]. When grown on filters, the cells formed lasting one second. The polarity of V. measurements was given monolayers of closely apposed cuboid cells (Fig. I D, E) separated using the apical side as the ground. Cells were bathed on each side by junctional complexes (Fig. 1E'). Results from electron microswith 8 ml minimum medium thermostated at 37°C that was copy showed that the cell layers consisted of at least two different cell types showing structural features of principal (with light circulated by a gaslift (95% 02—5% CO2 mixture). appearance) and intercalated (with dark appearance) cells (Fig. 1 D, E) [16, 171. To better evidence these different cells types, Determination of ion concentrations immunocytochemical experiments were performed with DBA, which binds essentially to principal cells of the rat CCD {18J. As The concentrations of Ht HC03, Na4 and K4 were deter- shown in Figure 1C, about 50% of the cells were labeled with mined in apical and basolateral media of RCCD1 cells cultured DBA, suggesting that RCCD1 cells were composed of about 50% for 48 hours on Transwell filters with a 2% FBS-supplemented PC. When grown on filters with 2% FBS-supplemented DM for DM. pH and HCO3 concentration were determined using a five to seven days followed by withdrawal of serum and growth blood gas analyzer (Radiometer). Na and K4 concentrations factors during 12 to 15 hours, RCCD1 cells exhibited baseline were determined by flame photometry. values for VT, RT and Isc of —10.5 1.2 mV lumen negative, 2390
371
Blot-Chabaud et al: Immortalized rat CCD cells
9 •ak•••• •"T•. -It
- :.j,• :
•I
-
D
0 0
C.) 4—
C.)
0,
E
100
100
80
80
0 C.) o
60
.4-
60
40
(1)
40
20
20
II
0 10
io— 106 iü— io—
0 108 io— 106
Amiloride, M Fig.
io
io—4
Ouabain, M
3. Presence of amiloride-sensitive Na channels and NatK-ATPase in RCCDJ cells. Detection of the mRNA of the a-subunit of the
amiloride-sensitive sodium channel by in situ hybridization in RCCD cells (A and B). Illustrations of labeling using antisense and sense probes. A clear signal was observed with antisense probe (A) whereas only background signal could be observed with sense probe (B). Immunocytochemical detection of the cs-subunit of Na',KtATPase in RCCD1 cells (C) and reconstructed view of a transversal section analyzed by confocal laser microscopy (C'). Note that Na,K-ATPase was essentially located in the basolateral domain of RCCDJ cells. Dose-dependent effects of amiloride and ouabain on short circuit current in RCCD1 cells (D and E). Increased concentrations of amiloride were successively added to the apical or basal sides of the cells (D). Apical addition () induced a large fall in Isc, whereas no effect was observed with the basolateral treatment (U). Increasing concentrations of ouahain in the basal bath also induced a large fall in lsc (E). Bars: A, B, C,C', 10 m. Values are the mean SE from four separate experiments.
160 cz
0 () 'i5
140 [ cm2 and 4.3
Ami 105M
120
0.5 pA/cm2, respectively. Accordingly, all
experiments were performed in these conditions in order to evaluate hormonal modulation of electrolyte transport. Finally, RCCD1 cells exhibited a prolonged lifespan since to date they have been subcultured up to the 45th passage.
BaCI2 5•10-3M
80
d 40
RCCD1 cells exhibit polarized ion transport processes
C),
0
To examine whether ion transport could occur in RCCD1 cells, the ionic concentrations from apical and basolateral media were determined for RCCD1 cells cultured during 48 hours on TransTime, minutes well filters. Results are given in Figure 2. In the apical medium, Fig. 4. Effect of amilonde and barium on Isc in RCCDJ cells. Confluent concentrations of K and HCO3 were significantly higher than RCCD1 cells grown on Snapwell filters were used for short-circuit in the basolateral medium, whereas concentrations of Na and experiments. Control value of Isc was 3.2 0.6 AJcm2. The addition of H were significantly lower. The occurrence of ion concentration iO El amiloride on the apical side induced a rapid and large decrease in Isc. The addition of 5 X iO El BaCl2 apically resulted in a further gradients strongly suggests that RCCD1 cells reabsorb Na and
0
20
40
60
80
100
120
decrease in Isc. Values are means SE from 3 separate experiments.
H and secrete K and HCO3.
372 I)
0 0
Blot-Chabaud et al: Immortalized rat CCD cells
25
A
200
20 15
.
E N S
10
0 0)
E
5
0
0
180
140
Fig. 5. Effects of arginine vasopressin (A VP) on CAMP and short circuit current ([sc) in RCCDJ cells. A. cAMP was measured on confluent RCCD1 cells grown on petri dishes and incubated without or with various
120
concentrations of AVP in the presence of iO M IBMX. Values are the mean SE from 3 to
160
06 6Q 000
0
AVP, M
160
00 .4-.
A
0
00
0
Ami 1 0-5M
+AVP 10-6M
U)
subunits of the Na channel or Na,K-ATPase. As shown in
80
of the probes was assessed by using sense probes. Figure 3
0 B
illustrates the results provided by in situ hybridization using the Na channel a subunit antisense (Fig. 3A) and sense (Fig. 3B)
20 40 60 80
120
1O6M
40 0
20
C
40 60 80 +DPC 5•10-3M
100
probes. Results from immunofluorescence studies using the MCK1 antibody against Na,KtATPase and confocal laser microscopy analysis also showed that > 95% of the staining was observed in the basolateral membrane domain of RCCD1 cells (Fig. 3C and 3C'). Consistent with these results Isc was sensitive to amiloride and ouabain. Figure 3D illustrates the dose-dependent effect on Isc of amiloride added either to the apical side or
4,+Ami
80
120
100 120 140 160
105M
0
basal side of RCCD1 cells. When added to the basal side, amiloride did not modify Isc. By contrast, apical addition of 120
+AVP 10-6M
100 80
0 (I)
ments were performed to detect mRNA encoding for the different
Table 1, a, /3 and y mRNAs encoding for the subunits of the Na channel were expressed in RCCD1 cells, a1 and 13) mRNA isoforms of Na,K-ATPase were also expressed. The specificity
160
0
from 4 separate experiments.
120
40 -
.
side of the cells. Values are the mean SE
Concentration, M
0
0U)
0N
0)
6 determinations. B. Isc was measured on RCCD1 cells grown on Snapwell filters. Cells were exposed to increasing concentrations of AVP (D) or dDAVP () added to the basal
concentrations of amiloride resulted in a progressive decrease in Isc with half maximal inhibition occuring at .10_6 M (Fig. 3D). Additional experiments performed in a medium devoid of Na confirmed that a large part of Isc was dependent on Na entry (not shown). Addition of ouabain to the basal side of the cells resulted in a decrease of Isc with half maximal inhibition of Isc occurring at more than i0— M (Fig. 3E).
In addition, apical addition of 5 X i0 M barium (BaCl2)
60 40 0
20
40
60
80
100
120
Time, minutes
significantly decreased the amiloride-insensitive Isc by - 36%, in addition to the effect of amiloride (Fig. 4). These results suggest K channels are present in the apical membrane of the cells, although one cannot totally exclude an indirect effect due to the leak of BaCI2 through the junctionnal pathways leading to the blockade of basolaterally located K channels.
Fig. 6. Influence of amiloride on Isc before and after treatment with atginine vasopressin (A VP) in RCCDJ cells. A. The control value of Isc was 3.4 + 0.3 pA/cm2. The addition of 10—6 M AVP to the basolateral side of RCCD1
cells induced a rapid increase in Isc that was sustained for at least 80 minutes. The addition of io M amiloride on the apical side of the cells induced a rapid decrease in Isc. B. When amiloride was added before AVP
administration, only a very small increase in Isc was observed after the addition of AVP on the basal side of the cells. C. The AVP-induced increase in Isc was not modified when DPC 5 X iO M was added before AVP. Values are the mean SE from 4 separate experiments.
RCCDJ cells are sensitive to arginine vasopressin
Results from cAMP measurements indicated that AVP increased the cellular cAMP content of RCCD1 cells (Fig. 5A). In
basal conditions, the cellular cAMP content measured in the presence of IBMX, art inhibitor of phosphodiesterase, was 2.51
0.80 pmol/7 minutes/106 cells. AVP induced a dose-dependent increase in cAMP with half maximal stimulation at .10_8 M (Fig. 5A).
Results from short circuit current experiments showed that, In CCD, the reabsorption of sodium implies Na entry via after a control period, 10—6 M AVP added to the basolateral side amiloride-sensitive apical sodium channels and Na extrusion of RCCD1 cells significantly increased Isc after less than five through basolateral Na,K-ATPase [19]. Therefore, experi- minutes, and then remained constant for up to 80 minutes
373
Blot-Chabaud et al: Immortalized rat CCD cells
A
B
175 150
160
S
125
140
E
100
0 E 0.
75
130 120
C.)
N S
150
50
Fig. 7. Effects of isoproterenol (ISO) on cAMP and short circuit current (Isc) in RCCDJ cells. cAMP assay and Isc experiments were performed on RCCD1 cells grown as described in Figure 4. A. cAMP was measured in the
110 100
25 C.)
ci - 0)
absence of isoproterenol or in the presence of various concentrations of the hormone. Values are the mean SE from 3 to 5 determinations. B. The addition of increasing concentrations of ISO to the basal side of cells induced an important and sustained increase in Isc. Values
C
C?
Li)
are the mean SE from 3 separate
Isoprotereflol, M
Isoproterenol, M
0.3; +AVP, 4.4 0.2 JLA/cm2; N = 8, P < 0.001; Fig. 6A). The addition of i0— M amiloride on the apical side
experiments.
(control, 3.3
resulted in a rapid fall in Isc (Fig. 6A). When RCCD1 cells were first incubated with i0— M amiloride, AVP induced only a small rise in Isc as compared to its effects in the absence of amiloride (Fig. 6B). Finally, when RCCD1 were first incubated with 5 x
iO M DPC, the addition of AVP resulted in a rapid and large increase in Isc (Fig. 6C). Both AVP and dDAVP, a V2 agonist [201, stimulated Isc with maximal effect at iO M (Fig. 5B). The effect of dDAVP on Isc was greater than that obtained with AVP but the difference did not reach statistical significance. For both hormones, the concentration required to obtain half maximum stimulation of Isc was — 4 >< 10_li
0
200 160
-Iso 10-6 M
A
4,+DPC 5•10 M
120
80
40 Cl)
0
0
20
60
40
80
100
120
B
M.
In additional experiments, the effect of prostaglandin E2 (PGE2) was tested. After treatment of RCCD1 cells with 10 M AVP, Isc increased from 4.5 1.4 to 5.9 1.5 aA/cm2. A further addition of 10-8 M PGE2 to the basal side of the cells significantly decreased Isc to 4.7 1.5 tAJcm2 (NS vs. control; P < 0.01 vs.
AVP treatment).
I
2E
o U)
40 0
RCCDJ cells are sensitive to isoproterenol
Results from immunocytochemistry showed that about 50% of RCCD1 had positive staining for DBA (Fig. 1C). It suggests that the DBA-negative cells correspond to IC. Several studies previously demonstrated that IC are mainly sensitive to f3-adrenergic agents [21]. To test whether the RCCD1 cell line was sensitive to
adrenergic agents, the effects of isoproterenol, a /3-adrenergic agonist acting on IC from CCD [21], was tested on cAMP content
and short circuit current (Fig. 7). ISO strongly stimulated the production of cAMP from RCCDI cells (Fig. 7A). As shown in Figure 8A, 10 M ISO added to the basal side of the cells also rapidly increased Isc. Thereafter, the ISO—elicited Isc remained in plateau (control, 5.0 0.5; + ISO, 7.4 1.1 rAJcm2; N = 6, P <
0.01). The further addition of 5 X i0 M DPC, a blocker of Cl channels [22], on the apical side of the cells induced a rapid fall of Isc (Fig. 8A). DPC (5 >< io M) added alone to the apical side of the cells also decreased Isc (Fig. 8B). In this case, the subsequent
addition of 10_6 M ISO induced only a weak increase in Isc. Finally, when amiloride was first added to the apical side of RCCD1 cells, the addition of isoproterenol led to a large increase in Isc (Fig. 8C). As already observed for AVP, ISO significantly
0
20
40
20
40
60
80
100
120
C
. 0 0
°0
240 160 80
C.)
U)
0
0
60
80
100
120
Time, minutes Fig. 8. Effects of DPC on Isc in RCCD1 cells before and after treatment with
isoproterenol (ISO) in RCCDJ cells. The influence of DPC on Isc was tested before and after addition of 106 M ISO to the basal side of RCCD1 cells.
A. The control value of lsc was 3.8 + 1.1 pAJcm2. The addition of IO M DPC to the apical side of the cells blunted the stimulated effect of ISO. B. When DPC was added first, almost no effect of isoprotenerol could be observed. C. The ISO-induced increase in Isc was not prevented when amiloride iO M was added before ISO. Values are the mean SE from 3 separate experiments.
374
Blot-Chabaud et at: Immortalized rat CCD cells
increased Isc in a dose-dependent manner (Fig. 7B). The concentration of isoproterenol required to obtain a half maximal stimulation of Isc was — 10 M.
200
The effects of AVP and ISO on Isc were additive, since the addition of i0 M ISO after treatment by 10_6 M AVP led to a
120
160
further increase in Isc (Fig. 9). Thereafter, successive additions of
10 NI amiloride, 5 X 10 M DPC, and 5 X i0 M BaCI2 to the apical side of the cells led to an almost complete suppression of Isc (Fig. 9). Discussion
This study describes the functional properties of a new rat cortical collecting duct clonal cell line established by infection of a primary culture of rat kidney cortical collecting ducts with the
wild-type SV4O. To establish the RCCD1 cell line, we used a primary culture of rat CCD obtained using a Percoll gradient technique modified from Vinay et a! [11] and Bello-Reuss et a! [12]. This technique allows an enriched preparation of CCD (> 97%) to be obtained for cell culture. Infection of these cells with the wild-type SV4O permitted us to develop a clonal cell population with the same morphological and functional features as those of primary cultured cells used for viral infection (data not shown).
The strategy of SV4O infection of primary cultured cells has already been used to establish several cell lines in various organs [23, 24], including the kidney [7—9, 25]. These studies have shown
that the presence of the SV4O large T antigen coding sequences was essential to the transformation process. In RCCD1 cells, the viral large T antigen is expressed since 100% of the cells were positively stained (Fig. 1A). Successive passages of RCCD1 cells
resulted in the maintenance of the expression of the large T antigen. RCCD1 cells have maintained many of the electrophysiological and functional properties of the parental CCD [26, 27]. In addition, contrasting with previously established CCD cell lines either derived from rabbit or mouse kidneys [7—10], RCCD1 cells
+lso 106M +DPC 5•1 -3 M +BaCL2 5•1 o— M
80 (2 (I)
40 0 0
20
40
60
80 100 120 140
Time, minutes Fig. 9. Arginine vasopressin (A VP) and isoproterenol (ISO) have additive effects on Isc in RCCDJ cells. Confluent RCCD1 cells were monitored for Isc as described in the legend of Figures 6 and 8. The control value of Isc was 4.4 + 0.8 jsAjcm2. In addition to the effect of i0 M AVP, 106 M ISO
induced a further increase in Isc. Isc could be reduced by successive addition of amiloride (Ami), diphenylamine 2-carboxylate (DPC) and BaCl2. Note that in the presence of these three inhibitors Isc was almost totally blocked. Results are the mean SE from 3 separate experiments.
a-, f3- and y-subunits of the epithelial sodium channel [31] as well as mRNA of a1- and /31-isoforms of the a and 13 subunits of the
Na,K4-ATPase [321. Whereas a1- and /31-subunits mRNA of Na,K4-ATPase are expressed at approximatively the same level, it appears that 13- and y-subunits mRNAs of the Na channel may be 10 times less abundant than a-subunit mRNA. Direct comparison of the absolute level of expression between subunit mRNAs is difficult. However, in the cortical collecting duct, it was estimated that all three subunit mRNAs were at comparable levels, using the same in situ hybridization technique and the same probes [31]. The low signals obtained with /3 and y probes, as compared to a probe for the sodium channel may thus be related to culture conditions and/or to the immortalization process. The expression of Na',K-ATPase is also demonstrated at the protein
develop very high transepithelial resistance (ranging between 1800 and 4000 (1 cm2) that can be maintained after long term level by immunocytochemical detection with an antibody against passages.
the a1-subunit of the Na,K-ATPase [33]. In addition, the
Results from ion concentrations measurements and from mor- presence of an apical K conductance sensitive to barium [28] is phological and immunocytochemical studies indicate that RCCD1 evidenced by Isc experiments. The RCCD1 cell line retain responsiveness to AVP, another cells are polarized. RCCD1 cells create concentration gradients for Na4, K, HCO3 and H4 (Fig. 2). This indicates that cells are property of the PC of the CCD [261. This is evidenced by a rapid able to reabsorb Na4 and H and secrete K4 and HC03. In increase in both cAMP content and Isc following AVP treatment addition, they exhibit typical junctional complexes and basolater- [27, 34]. AVP induces a sustained increase in Isc [35], dependent ally-located Na pumps. Additional evidence of the polarization on a rapid increase in the apical sodium entry [36] and with an state of the cells is provided by the results from short circuit apparent K112 at 4 x 10 M AVP. The dose-dependence curve current experiments: RCCD1 cells exhibit amiloride-sensitive for the V7 agonist of AVP, dDAVP, shows that the effect of Na channels [19] and K conductanccs blocked by barium [28] dDAVP on lsc is somewhat higher than that observed with AVP. on their apical surface. The observed half maximum increases in Finally, the addition of I 0 M exogenous PGE2 inhibits the lsc obtained for various concentrations of ouahain and amiloride AVP-induced increase in lsc. One should notice that higher indicate that Na4 channels from RCCD1 cells have high affinity concentrations of AVP are necessary to induce increase in cellular for amiloride, whereas Na,K4-ATPase has low affinity for cAMP than those required to stimulate Isc. Such differences, also ouabain. These results are in complete agreement with previous observed with isoproterenol, have already been reported in other types of cultured epithelial cells [24]. Since the culture conditions studies obtained in microdissected rat CCD [29, 30]. Ultrastructural studies, DBA binding studies and Isc experi- influence the differentiation state of cultured epithelial cells [7],
ments support the notion that the RCCDJ cell line express such dissociated dose-dependent effects in AVP and ISO refeatures of at least two different cells types that may correspond to sponses may he due to the fact that Isc was measured on cells the principal and intercalated cells constituting the CCD [16—18, grown on filters, whereas the cAMP content was determined on 21]. Principal cells (PC) of RCCD1 retain most of the properties cells grown on petri dishes. While the RCCD1 cell line appears to possess features of PC of the PC of CCD in vivo. Indeed RCCD1 cells express mRNAs of
Blot-Chabaud et al: Immortalized rat CCD cells
from intact CCD, it fails to respond to aldosterone by an increase
in Isc (data not shown). This constrasts with the maintained sensitivities to AVP. One cannot exclude that the SV4O-transformation process may have deleterious consequences on mineralo-
corticoid receptor expression. In fact, no evidence of a highly conserved response to aldosterone in SV4O-transformed distal or CCD cell lines has been reported [7—10, 37]. Alternatively, the
culture conditions may be not optimum for the expression of mineralocorticoid receptors and/or the response to this hormone
375
3. ROY C, PRESTON AS, HANDLER JS: Insulin and serum increase the number of receptors for vasopressin in a kidney-derived line of cells grown in a defined medium. Proc NatI Acad Sci USA 77:5979—5983, 1980 4. GSTRAUNTHALER G, PFALLER W, KOTANKO, P: Biochemical charac-
terization of renal epithelial cell cultures (LLC-PKI and MDCK). Am J Physiol 248:F536—F544, 1985 5. HANDLER JS, STEELE RE, SM-JIB MK, WADE JB, PRESTON AS, LAWSON NL, JOHNSON JP: Toad urinary bladder epithelial cells in
culture: Maintenance of epithelial structure, sodium transport, and response to hormones. Proc NatlAcad Sci USA 76:4151—4155, 1979
[381.
6. PERKIN FM, HANDLER JS: Transport properties of toad kidney
specific marker for rat IC, equivalent to peanut lectin in the rabbit
7. VANDEWALLE A, LELONGT B, GENITEAU-LEGENDRE M, BAU000IN B, ANTOINE M, ESTRADE S, CHATELET F, VERROUST P, CASSINGENA R,
The RCCD1 cell line also exhibits features of IC. Since no
[39], is available, DBA was used in this study. Half of RCCD1 cells
were not stained by DBA, a result compatible with those from cAMP assays and short circuit current experiments. This is also in agreement with ultrastructural studies which indicated the presence of both dark (IC) and light (PC) cells [16]. In addition, Isc experiments clearly indicate that AVP and ISO have cumulative effects and modulate different transport pathways; this argues for
epithelia in culture. Am J Physiol 241:C154—C159, 1981
RoNco P: Maintenance of proximal and distal cell functions in SV4O-transformed tubular cell lines derived from rabbit kidney cortex. J Cell Physiol 141:203—221, 1989 8. PRIE D, RoNco PM, BAUDOUIN B, GENITEAU-LEGENDRE M, AroINE M, PIEDAGNEL R, ESTRADE S, LELONGT B, VERROUST P, CASSINGEENA
R, VANDEWALLE, A: Activation of the simian virus 40 (SV4O) genome
abrogates sensitivity to AVP in a rabbit collecting tubule cell line by repressing membrane expression of AVP receptors. J Cell Biol 113:
the presence of both cell types in the RCCD1 cell, although we 951—962, 1991 cannot totally exclude that RCCD1 cells would be constituted of 9. AREND LI, HANDLER JS, RrnM JS, GUSOVSKY F, SPIELMAN WS: Adenosine-sensitive phosphoinositide turnover in a newly established only one cell type with hybrid characteristics of both PC and IC. renal cell line. Am J Physiol 256:F1067—F1074, 1989 ISO exerts a rapid increase in Isc with an apparent K112 at BA, NARAY-FEJES-TOTH A, CARRETERO OA, ITO S, FEJE5M, which is mediated by an increase in apical Cl conductance 10. STOOS Tom G: Characterization of a mouse cortical collecting duct cell line. (Fig. 8). This observation is in close agreement with previous Kidney mt 39:1168—1175, 1991 studies [27, 40] and suggests that IC from RCCD1 cells mainly 11. VINAY P, Gouooux A, LEMIEUX G: Isolation of a pure suspension of rat proximal tubules. Am J Physiol 241:F403—F411, 1981 correspond to 13-IC [21]. This statement is also supported by the
10
fact that RCCD1 have been shown to secrete HC03 and reabsorb H (Fig. 2), a characteristic of J3-IC. Fejes-Toth and Naray-Fejes-Toth [41] showed that a conversion of 13-IC to cs-IC
and PC may occur in CCD cultured cells. Although one cannot exclude that such a phenomenon also occurs in RCCD1 cells, results from Isc experiments strongly suggest that this cell line still possesses /3-IC. In conclusion, this new cell line appears to be the first rat CCD
cell line with high transepithelial resistance and properties of vectorial transport close to that of intact rat CCD. Therefore, it could represent an attractive model to study ion transport and its control by polypeptide hormones in cortical collecting duct cells.
12. BELLO-REUSS E, WEBER MR: Electrophysiological studies of primary
cultures of rabbit distal tubule cells. Am J Physiol 252:F899—F909, 1987 13. Cox KH, DELEON DV, ANGERER LM, ANGERER RC: Detection of mRNAs in urchin embryos by in situ hybridization using asymmetric RNA probes. Dev Biol 101:485—502, 1984 14. FARMAN N, CORTHESY-THEULAZ I, BONVALET JP, ROSSIER BC: Localization of cs-isoforms of Na-K-ATPase in rat kidney by in situ hybridization. Am J Physiol 260:C468—C474, 1991 15. CEREIJIDO ME, R0BBINS 5, DOLEN WJ, ROTUNNO CA, SABATINI DD:
Polarized monolayers formed by epithelial cells on a permeable and translucent support. J Cell Biol 77:853—880, 1978 16. STANTON BA, BIEMESDERFER D, WADE JB, GIEBIscH U: Structural
and functional study of the rat distal nephron: Effects of potassium adaptation and depletion. Kidney mt 19:36—48, 1988 17. H0LTHOFER H, SCHULTE BA, PASTERNACK G, SIEGEL GJ, SPtCER SS:
Acknowledgments This work was supported by INSERM U246. We are grateful to Dr.
Sweadner (Harvard Medical School) who kindly provided the antiNa,K-ATPase antibody. The rat a1- and f31-ATPase cDNAs were a generous gift from Dr. Lingrel (University of Cincinnati). The rat a, /3, and 'y sodium channel cDNAs were kindly provided by Dr. Rossier (Institut de Pharmacologie, Lausanne). We thank the laboratory of blood gas analysis
(X. Bichat Hospital) for pH and HC01 concentration determinations and Mrs. M. Ferrera for determination of Na and K concentrations. We thank T. Carison for editing the manuscript. We also thank Mrs. V. Levêque and C. Tritscher for secretarial assitance. Reprint requests to Marcel Blot-C'hahaud, Ph.D., INSERM U246, Faculté de Médecine Xavier Bichat, BP 416, 75870 Paris Ceder 18, France.
References 1. GAUSCH CR, HARD WL, SMITH TF: Characterization of an established line of canine kidney cells (MDCK). Proc Soc Exp Biol Med 122:931— 935, 1966 2. Hvu. RN, CHERRY R, WEAVER OW: The origin and characteristics of a pig kidney cell strain LLCPK. In Vitro 12:670—677, 1976
Three distinct cell populations in rat kidney collecting duct. Am J Physiol 253:C323—C328, 1987 18, HOLTI-IOFER H, SCHULTE BA, SPICER SS: Expression of binding Sites
for dolichos biflorus agglutinin at the apical aspect of collecting duct cells in rat kidney. Cell Tissue Res 249:481—485, 1987 19. STOKES JB: Ion transport by the cortical and outer medullary collecting tubule. Kidney mt 22:473—484, 1982 20. BREYER MD, ANDO Y: Hormonal signaling and regulation of salt and water transport in collecting duct. Annu Rev Physiol 56:711—739, 1994 21. SCHUSTER VL: Organization of collecting duct intercalated cells. Kidney mt 38:668—672, 1990 22. STuns MJ, HENKE DC, BOUcHER RC: Diphenylamine-2-carboxylate (DPC) inhibits C1 conductance and cyclooxygenase of canine tracheal epithelium. Pfidgers Arch 415:611—616, 1990 23. ISOM HC, TEUETHIA MJ, TAYLOR JM: Transformation of isolated rat hepatocytes with simian virus 40. J Cell Biol 85:651—659, 1980 24. HERMAN P CASSINGENA R, FRIEDLANDER G, SOLaR P, GRODET A,
TRAN BA HUY P, AMIEL C: Middle ear cell line that maintains vectorial electrolyte transport. J Cell Physiol 154:615—622, 1993 25. Scorr DM, MACDONALD C, BREZSKI H, KINNE R: Maintenance of expression of differentiated function of kidney cells following transformation by SV 40 early region DNA. Exp Cell Res 162:391—398, 1986 26. BURG MB, lsCsoN L, GRANTHAM I, ORLOFF J: Electrical properties
376
Blot-Chabaud et al: Immortalized rat CCD cells
of isolated perfused rabbit renal tubules. Am J Physiol 215:788—794, 1968
27. MOREL F: Sites of hormone action in the mammalian nephron. Am J Physiol 240:F159—F164, 1981
28. GIEBISCH G: Renal potassium transport, in Membrane Transport in Biology, edited by GIEBISCH 0, TOSTESON D, USSING HH, Berlin, Springer Verlag, 1978, IV. A, p 215
cultures of rabbit cortical collecting duct cells.AmfPhysiol 262:F454— F461, 1992 35. CHEN L, WILLIAMS SK, SCHAFER JA: Differences in synergistic actions
of vasopressin and deoxycorticosterone in rat and rabbit CCD. Am J Physiol 259:F147—F156, 1990 36. SCHLATrER E, SCHAFER JA: Electrophysiological studies in principal
cells of rat cortical collecting tubules. ADH increases the apical membrane Na- conductance. Pflugers Arch 409:81—92, 1987
29. SARIBAN-SOHRABY S, BEN05 DJ: The amiloride sensitive Na channel. Am J Physiol 250:C175—C190, 1986 30. SWEADNER KJ: Isozymes of the Na-K-ATPase. Biochim Biophys
37. RAFESTIN-OBLIN ME, FARMAN N, CASSINGENA R, RONCO P, VANDE-
Acta 998:185-220, 1989 31. Duc C, FARMAN N, CANESSA CM, BONVALET JP, ROSSIER BC: Cell-specific expression of epithelial sodium channel a, /3, -y subunits
1993 38. HUSTED RF, MATSUSHITA K, STOKES JB: Induction of resistance to
in aldosterone-responsive epithelia from the rat: Localization by in situ hybridization and immunocytochemistry. J Cell Biol 127:1907— 1921, 1994 32. FARMAN N, COUTRY N, LOGVINENKO N, BLOT-CHABAUD M, BouR-
BouzE R, BONVALET JP: Adrenalectomy reduces a1 and not /3
Na-K-ATPase mRNA expression in rat distal nephron. Am J Physiol 263:C810—C817, 1992 33. FELSENFELD DP, SWEADNER KJ: Fine specificity mapping and topog-
raphy of an isozyme-specific epitope of the rat Na,K-ATPase catalytic subunit. J Biol Chem 263:10932—10942, 1988 34. CANE55A CM, SCHAFER JA: AVP stimulates Na transport in primary
WALLE A: Mineralocorticoid receptors in SV-40 transformed tubule cell lines derived from rabbit kidney. J Ster Bioch Mol Biol 44:45—52,
mineralocorticoid hormone in cultured inner medullaxy collecting duct cells by TGF-/31. Am J Physiol 36:F767—F775, 1994 39. L HIR M, KAISSLING B, KOEPPEN BM, WADE JB: Binding of peanut lectin to specific epithelial cell types in kidney. Am J Physiol 242:C1 17— C120, 1982 40. LIGHT DB, SCHWIEBERT EM, FEJES-TOTH G, NARAY-FEJES-TOTH A, KARLSON KH, MCCANN FV, STANTON BA: Chloride channels in the
apical membrane of cortical collecting duct cells. Am J Physiol 258:F273—F280, 1990 41. FEJES-TOTH G, NARAY-FEJES-TOTH A: Differentiation of renal /3-in-
tercalated cells to a-intercalated and principal cells in culture. Proc NatI Acad Sci USA 89:5487—5491, 1992