1133(I t]91) 17-24 P~ 1991 Elsevier Science Publishers B.V. All rights reserved (1167-4889/91/$113.511
Biochimica et Biophysica Acta,
17
Regulation of intracellular pH in rabbit cortical connecting tubule and cortical collecting duct Georges Dagher and Claude Sauterey Laboratoire de Physiologic Celhdaire, Coll~gc de France. Paris (France)
(Received 17 July 1991)
Key words: CytosolicpH: Na-H exchange; Rabbit collectingtubule: H secretion; Isoproterenol Proton transport pathways in isolated superfused rabbit cortical connecting (CNT) and collecting tubules (CCD) were determined using the fluorescent pH-seusitive dye BCECF following acid or base load by exposure to NH4CI. Following removal of NH4CI which results in a rapid decline in pH i two mechanisms a p p e a r to be responsible for pH i recovery, a Na-independent NEM-sensitive H etllux with a 5low activity, which was virtually absent in 30% of the segments tested and a seconO ."~pld l'~a-d~pendent I~ c.~. nx. In CCD this latter pathway was shown to have an apparent K , for (Na+)e of 38.2 + 0.4 mM (S.D., n = 7) and was sensitive to EIPA. Similar results were obtained with the CNT. With regard to the H pump in six out of ten CCD isoproterenol (200 nM) resulted in a 2-fold stimulation of H pump activity. These effects of isoprenaline were inhibited both by the non-specific ~l.adrenoceptor antagonist propranolol as well as by the specific b~ antagonist metoprolol. Interestingly, these stimulatory effects of this ~B agonist, which is known to stimulate cAMP formation in rabbit C C v , wcre not reproduced by the addition of exogenous cAMP analogues db cAMP (0.1 raM), ClWl[" cAMP (0.1 raM), 8 Br-cAMP (0.1 mM) or the addition of forskoline (0.3 mM). In conclusion, these" data obtained in isolated rabbit CNT and CCT demonstrate the presence of a n active Na-H exchange which is for the most part responsible for the recovery of PHi. It should be noted also that the contribution of the H pump to pH i regulation appears to be negligible in these segments.
Introduction The distal part of the nephron plays an important role in the regulation of acid secretion. Much of the information concerning the acidification mechanisms comes from studies on isolated perfused cortical collecting duct (CCD) which showed that acidification is an electrogenic and Na-independent process [1,1638]. In contrast, little is known about H transport in the CNT. The regulation of H secretion is not well defined. PCO 2 and transtubular H C O 3 grad:ents have been shown to elicit rapid insertion of cytoplasmic vesicles believed to carry H + pump units. On the other hand, cAMP was found to stimulate net HCO 3 secretion in rabbit CCD [2].
Abbreviations: CNI, cortical connecting tubules; CCD, cortical collecting ducts. Correspondence: G. Dagher, Laboratoire de physiologicceliutaire. Coll6ge de France, 75231 Paris Cedex 05. France.
In the present study, we have examined the mechanisms by which intracellular pH is regulated on the isolated, superfused CNT and C C D segments from rabbit kidney. In both segments, pH recovery after acute acid load is mediated by a NEM-sensitive Na-independent H + efflux and an Na-H exchange. Isoproterenol markedly stimulates the NEM-sensitive H + elflux.
Methods Granular distal connecting tubules (CNT) and cortical collecting tubules (CCD) were obtained from male New-Zealand white rabbits weighting 1-1.5 kg as previously described [3,4]. Single tubules were isolated from collagenase-treated kidney and incubated in a cold Na-Hepes solution (pH 7.4 at 4 ° C ) containing 4 mM essential and non-essential amino acids derived from Eagles minimal essential medium, Cells were loaded with 2',7'-bis(carboxyethyl)-5(6)-carboxyfluorescein (BCECF) by exposing the tubule to 5 p,M
BCECF-AM for 20 rain at room temperature. The tubule was fixed on a glass cover slip by transfering it to a 1 ml Na-Hepes solution containing 1% agarose and then cooled rapidly to 4°C. The cover-slip was glued to a steel chamber mounted on an inverted Leitz Diavert microscope, equipped with a 40-fold magnification fluorescent Nikon immersion objective. The tubule was constantly superfused, and the solutions and chamber were maintained at 37°C. A detailed description of the chamber and the optical system has been published in a preceeding paper from this laboratory by Tanigushi et al. [5]. pH i was determined by alternately exciting the dye at 450 and 500 nm while measuring the fluorescence emission. A 510 nm dichroic mirror was used and a 530 nm band pass filter was placed immediately before the photometer. A neutral density filter was present in the pathway of the light source (a mercury 100 W lamp) to reduce photobleaching of the dye [6]. The emission intensities for each excitation wavelength increase with pH and remain relatively insensitive to probe concentration. The emitted light intensity was quantitated from a rectangular spot (of about 40 ~m in width and 100/tm in length) focussed on a group of random cells. Intercalated cells could not be reliably distinguished from principal cells by light microscopy. Thus, pH i values reflected a weighted average of pH i of both types of cells. After amplification the emission signal from the photomultiplier was digitized and stored in a microcomputer for analysis. Measurements were averaged for 3 s for each excitation wavelength. The background signal was recorded after each experiment. This non cell related intensity never exceeded 2% of the signal in dye-loaded cells and was subtracted from all the measurements of the experiment. Cellular autofluorescence was assessed in tubules not exposed to the fluorescent probe. The signal was always smaller than 1% of that obtained in BCECF-Ioaded cells. No correction for this signal was made.
p H i calibration pH i was calculated using the nigericin technique as described by Thomas et al. [7]. Calibration studies were performed on each tubule used for pH i measurement. At the end of each experiment, cells were exposed to a calibration solution containing: KCI 110 mM, MgCI 2 1 mM. CaCI 2 1 raM, Hepes 20 mM, choline chloride 30 raM, nigericin 10 p.M. Solutions were adjusted to different pH values ranging from 6.0 to 8.0. When transmembrane K concentration is in equilibrium, internal H would tend to equilibrate with external H concentration the ratio of emission intensity assessed at two excitation wavelength 500 nm/450 nm is a linear function of external pH in the range of 6.3 to 7.8.
Solutions Na-Hepes solution contained in mM: NaCI 145, KCI 5, CaCI 2 l, MgCI 2 1, NaH2PO a 0.6, Na2HPO 4 1.4, glucose 20, Hepes 20 (pH 7.4 at 37 ° C). When necessary 20 or 30 mM NH4CI were substituted for NaCI. In Na-free solution, NaCI was replaced with N-methylglucamine (NMG) (pH 7.4 at 37 ° C). All reagents were analytical grade from Merck. Specific inhibitors, /3 agonist, /3 antagonist, dibutyryl cAMP, chlorophenyl thio-cAMP and forskolin were from Sigma. 8-Br-cAMP was from Boehringer. Results One commonly used approach for studying oH regulating mechanisms is to acutely load the cells with acid or base, and monitor the subsequent recovery of cellular pH towards its initial level [8,12]. In the present study, single isolated segments were transiently exposed to 20 mM ammonium chloride, pH i rises rapidly due to NH 3 influx which traps intracellular H +, and then reverses back due to NH4+ entry which dissociates to form NH 3 that leaves the cell and H + which remains trapped inside. NH4CI medium was then removed and replaced with Na+-free solution (NMG medium). This resulted in cellular acidification due to the NH 3 loss. pH i spontaneously recovers from this acid load as a result of one or more acid extrusion mechanisms located in the luminal and basolateral cell membranes. p H i recovery mechanisms from an acid load A typical pH recovery in CCD cells is shown in Fig. l. Cells were transiently exposed to 20 mM NH4CI (a-c) inducing an increase in cell pH (a-b). When NH4CI was removed by substitution with NMG medium, we observed an acidification of 0.7 pH units (c-d) followed by a pH recovery with a slow rate in the absence of external sodium (d-e). The replacement of this medium with Na + solution (e) induces a rapid pH recovery toward its initial value (e-f). The segment was next exposed to 1 mM N-ethylmaleimide (NEM), inducing a decrease in steady state pH i by about 0.2 pH units if-g). The cells were then acid loaded by a transient exposure to NH4C! (g-i). No pH recovery could be observed in Na-free medium (j-k). The introduction of external sodium induces a rapid pH recovery toward its initial value (k-l). The present results suggest that in CCD two pathways mediate cell pH recovery after an acid load. A slow rate NEM-sensitive, Na+-independent pathway which is inhibited by NEM. The second pathway is a highly active Na+-dependent mechanism which in presence of NEM recovers pH~ to its initial level. This pattern of pH recovery from an acid load was also observed in CNT (Fig. lb).
Na
, NH4
I
~
,
Na
7.8 (A)
r~
i
1
NH4
Ntv~
Nn
I
NH4
l ~
I
~-~ 76
b
f
b
74
,
7.2 7.0 68
d
6.6 6.4 u
•
M~.ures
0'
'
Minutes
Fig. I. pH i recoveryfrom an NH~ pulse in CCD and CNT. (A) CCD bathed in CO, free Na-l-lepes solutions At (a) the bathing solution was switched for one containing 20 mM NHaCI in replacement of NaCI inducing an increase in pH i (b-e). NH4CI solution was replaced by N-methylglucamlnesolution (NMG) at (c) Cell pH decreases rapidly (c-d). and then rec(wers at a rate of 0.044 pH units/rain (d-e). The replacement of NMG solution by NaCI at (e) induces a rapid recoveryof pH, (e-f) with a rate of 0.344 pH units/rain. The presence of I mM N-ethylmaleimide(f) induces a decrease of steady state pH by 0.2 units. Cells were then tee×posed to ;0 mM NH4CI (g) which was switched to N-methylglueamine solution at (i). In presence of NEM, pH i recovery from the induced acid load Q-k) had a rate of 0.003 pH/min. Replacement of NMG solution by NaCI at (k) induces a rapid rise in pH i with a rate of 0.318 pH/min. (B) CNT was bathed it. CO2 free Na-Hepes solution. At (a) the bathing solution was switched for one containing 20 mM NH4CI in replacement for NaCI. At (c) NH~CIsolution was replaced by NMG solution, pH i recovery from the induced acid load (e-d) has a rate of 0.037 pH units/rain (d-e). The replacement of NMG solution by NaCI at (e) induces a rapid alkalinization with a rate of 0.36 oH units/min (e-O. Fig, 2 shows another example of p H recovery from an acid load in CCD cells. Cells were first exposed to NH4CI which was replaced with N M G m e d i u m inducing an acid load of about 0.5 p H units• Stlrprisingly, in this segment no p H recovery could be observed in Na-free m e d i a ( d - e ) . T h e introduction of external sodium induces a rapid p H recovery toward its initial value (e-f). Cells were then treated with E I P A (50 # M ) inducing a decrease in steady state p H by 0.2-0.3 units (Fig. 2, f-g), T h e segment was exposed again to N~ I u-~ I ~
ra
i
I Nm e~o^
•
7.S
b
f
Na
h
7.2
s"
d
-
k
i
S.4
0
4
MINUTES
Fig. 2. Effect of EIPA on pH recoveryfrom a NH4CI pulse in CCD cells. At (a) external solution was switched from Na-Hepes to 20 mM NH4CI. At (c) external NH4 was removed and replaced with Nmethylglucamine(NMG) solution. Na-Hepes solution was reintroduced at (e) causing pH i to recover with an initial rate of 0.45 pH units/rain. At (0 60 g,M EIPA was applied, and cells were reexposed to 20 mM NH4CIpulse. Na-Hepes solution was introduced at (i) and no pH recoverycould be observed(k-I).
NH4CI pulse. Replacing this m e d i u m with Na medium induces an acid load which was not followed by a p H recovery in the presence of EIPA. This suggests that the Na-dependent recovery phase is mediated by the N a - H exchanger. A similar observation was made in five different CCD segments. Interestingly the acid load induced by replacement of NHgCI medium with N M G ( c - d ) was of similar magnitude to that obtained with Na m e d i u m ( i - k ) with EIPA, This indicates that the decrease in pH is determined by N H 4 removal ~s it is expected, rather than by Na removal. We have examined the activity of the Na-independent and the Na-dcpendent mechanisms in 30 CCD and 8 CNT. The recovery rate of the Na-independent pathway was variable ranging from 0.007 to 0.076 p H units m i n - t (mean + S.D.: 0.046 + 0.06) in CCD, and from 0.003 to 0.029 p H units rain-1 in C N T (mean + S.D: 0.018 + 0.009). Interestingly, this mechanism was virtually absent in about 60% of the CCD segments studied (for example, Figs 2 ( d - e ) and 5). T h e initial rate of recovery of the Na-dependent pathway in 30 CCD had a mean + S.D. of 0.31 =]: 0.18 p H uni:~ rain - ~ and of 0.32 + 0.13 o H units r a i n - : in 8 CNT. Cell p H modtdation during exposure to N H 4 CI The recovery phase after NH3 induced alkalinization Fig. 1A ~b-c), can be fitted with a single exponential and has been ascribed in several cell types to passive NH~- entry down its electrochemical gradient. However, as shown in Figs. 1 and 2 after initial alkalin-
20
s6
'
o
o,c,2
'
is
Fig. 3. Effect ouabain and BaCIz on recovery rate after NH 4 pulse in CCD. CCD cells were bathed in a Na-Hepes solution which was subtituted at (a) for one co;llaining 2O mM NH4CI. The acidification phase (a-b) fitted a single exponential with time constant of 1.16 rain- i and an initial rate of 0.85 oH units/min. At (b) cells were bathed back in Na-Hepes solution 1 mM ouabain and I mM naCI 2 were introduced as indicated and cells were reexposed twice to 20 mM NH4CI pulse. The acidification phase could be fitted with a single exponential with an initial rate of 0.38 pH units/rain rain le-f) and 0.12 pH units/min (i-j). On the other hand. in Na-Hepes medium oH recovery could be fitted with a single exnonential. The calculated initial velocities+ S.D. were of 0.96+ 0.07 (c-d), fi.25+ 0.07 (g-h), 0.31+0.03 pH units/rain (k-I). One of four experiments givingsimilar results.
ization pH~ decreases to levels below initial value ( b - c ) . This was observed in 22 out of 30 C C D a n d in 3 out of 8 CNT. Interestingly, in the presence of N E M (Fig. IA) or E I P A (Fig. 2) this d r o p below initial value was not observed, suggesting the participation of o t h e r pathways transporting N H ~ into the cells such as the N a - H exchanger. The decrease below initial values occurs with a simultaneous decrease in 20 m M external Na (replaced by NH4CI) a n d N H 4 entry. This would p r o d u c e a decrease in the outside to inside N a gradient, thus inducing an increase in H influx coupled to Na efflux. This is in accord with the absence of a decrease below initial level in presence of E I P A a n d the effect of o u a b a i n as discussed below. Fig. 3 represents a peculiar p H recovery p a t t e r n observed in only 8 out of 36 C C D segments. Addition of NH4CI resulted in an immediate a n d rapid acidification (Fig. 3, a - b ) . Substitution of Na m e d i u m to this m e d i u m induces an acid load followed by a rapid raise in p H to its initial level. P r e t r e a t m e n t with 1 m M o u a b a i n r e d u c e d by 5 0 % the rate of acidification after NH4CI addition which is still below initial level (e-f). The addition of o u a b a i n could increase cellular Na. This would induce a more p r o n o u n c e d decrease in outside to inside Na gradient in NH4CI medium, a n d consequently to a f u r t h e r increase in N a efflux coupled to H influx. This a p p e a r s in concert with the observation of 75% reduction in the N a - d e p e n d e n t recovery from the acid load in presence of o u a b a i n ( g - h ) , strongly suggesting a reduction in t r a n s m e m b r a n e N a concentration gradient following the increase in intracellular Na. O n the o t h e r hand, one c a n n o t exclude
partial contribution of the Na p u m p to N H 4 influx. The addition of BaCl 2 in presence of oubain induced a further decrease in the rate of acidification after NH4CI addition (i-j vs. e - O . U n d e r these conditions steady state p H fails to d r o p below initial p H levels. Removal of NH4CI induced an acid load which was followed by a recovery in Na medium.
Kinetics of Na-H exchange as a function of external sodium T h e above results clearly d e m o n s t r a t e that in C N T a n d C C D , pH~ recovery from N i l 4 induced acid load is much m o r e active in presence of Na than in its absence. In o r d e r to quantify the Na d e p e n d e n c y of this rapid phase, we p e r f o r m e d a series of experiments in which we assessed H efflux at increasing external concentrations of Na as previously described by Chaillet et al. [9]. C C D cells were p r e t r e a t e d with N-ethylmaleimide (1 mM) to inhibit H + efflux in Na-free m e d i a a n d were then acid loaded by exposure to NH4CI. As shown in Fig. 4A, H + efflux was absent in N M G m e d i u m ( b - c ) . T h e addition of 4.5 ~.'tM external Na induced a n increase in pHi. External Na was then removed inducing a slow decree.se in pH~ to a stable value. Addition of 10 m M external N a induces a substantially higher p H i recovery rate which was partially reversed u p o n removal of t=xternal Na. T h e addition of 40 m M N a results in a m o r e r a p i d pH~ recovery which reverses in N M G medium, pH~ recovery rate was even higher at 145 m M N a a n d could be fitted to a single exponential giving a n initial slope of 0.61 p H units m i n - t . F o r e a c h tubule, the initial recovery rate assessed at increasing external c o n c e n t r a tions of N a were normalized to the initial recovery rate at 145 m M Na. The results are shown in Fig. 4B. A K m value for external Na of 38.2 + 10.4 m M was o b t a i n e d by fitting the results to a Miehaelis-Menten equation using a n o n linear analysis p r o g r a m (Enzfitter from Elsevier-Biosoft, Cambridge, U.K). A H a n e s - W o o l f plot of the d a t a is shown in the inset. In this kinetic analysis we have assumed that intracellular N a was uniformely low as the tubule was exposed to Na-free m e d i u m before the addition of increasing external c o n c e n t r a tions of Na.
Effect of isoproterenol C C D cells were acid loaded by exposure to NH4CI prepulse a n d N a - i n d e p e n d e n t a n d N a - d e p e n d e n t H efflux mechanisms were assessed before a n d a f t e r exposure of the tubule to isoproterenol ( 2 ' 10 -7 M)-II B M X (10 - s M) for 3 to 5 min. As seen in Fig. 5, H efflux in Na-free m e d i u m was virtually absent. This has b e e n observed in a large n u m b e r of segments as discussed above. Addition of isoproterenol did not modify steady state p H for the
Fig. 6 s u m m a r i z e s t h e e f f e c t o f i s o p r o t e r e n o l on N a - i n d e p e n d e n t , H efflux in I0 C C D s e g m e n t s , in 6 o f t h e m , a 1 to 6-fold stimulation o f N E M sensitive H + efflux w a s o b s e r v e d in p r e s e n c e o f /3 agonist. In 4 o t h e r s e g m e n t s no significant effect could be o b s e r v e d . M e a n H efflux w a s of 0.077_+0.105 u n d e r control c o n d i t i o n s a n d o f 0.118 _+ 0.112 a f t e r i s o p r o t e r e n o l addition ( P < 0.05 by p a i r e d t-test; a n d P < 0.006 by Wilcox0n s i g n e d R a n k test.) In contrast, p H recovery
following 10 min. Successive N H 4 C I i n d u c e d acid load r e v e a l e d a significant s t i m u l a t i o n o f the N a ~ndcpend e n t H efflux 20 to 30 m i n a f t e r e x p o s u r e to agonist. F u r t h e r , this s t i m u l a t o r y e f f e c t s e e m s to be m o r e potent 50 to 60 rain a f t e r i s o p r o t e r e n o l addition. O n t h e o t h e r h a n d , H efflux r a t e s in N a - f r e e m e d i a w e r e highly r e p r o d u c i b l e in control c o n d i t i o n s as recovery r a t e s a f t e r two s u b s e q u e n t acid loads s h o w e d a maxim a l v a r i a t i o n o f 2 0 % ( m e a n : 12%, n = 7).
I NH4 ] Na
I0
14~10
I~0
I
0
,I 4°1
0
1145
7.9 7.7 7.5 7.3 7.1 6.9 6,7 6.5
I
I
0
I0 minutes
I B) 1.0 o.8
< a:
0.6
0.4
i,o
0.2
I
'
:o
I
I
I
I
40
80
120
160
Na ( m M )
Fig. 4. External Na-dependence of pHi recovery from an NH~ induced acid load in CCD. (A) pH recovery at different concentration of e:~,.rnai Na. Cells were pretreatcd with 1 mM NEM. At (a) Na-Hepcs solution was switched for an isoosmotic 20 mM NH4CI solution. NMG solutioti was introduced at (b), (d), (f) and (h). The solutions applied at (c), (e), (g), (i) contained (in raM): 4.5. 10, 40 and 145 NaCI in replacement o[ an equivalent amount of N-methylglucamine (NMG). Recovery rates from the acid load were of 0.074 (c-d), 0.167 (e-0,0.520 (g-h) and 0.613 pH units/rain (i-i). (B) Normalized acid extrusion as a function of external [Na]. Data were derived from seven experiments similar to that in A the mean initial recovery rate + S.D. at 145 mM Ha was of 0.35 + O.14 pH units/rain. Fluxes obtained at 4.5, tO and 40 mM Na were divided by flux calculated for same pH i at 145 mM [Na] o. phi values at (c), (c), (g). (i) had mean values___S.D. of: 6,76± 0.21, 6.87+0.21.6.88±0.26, 6.81 d:0.23, respectively. A Hanes-Woolf plot of these data is shown in the inset.
7.8
NH4
7.6
4,
7.4 7.2
B
:~ o. 7.0 6.8 6.6 6.4 6.2
I
1
I
o
4 minutes
Fig. 5. Effect of isoproterenol and IBMX in CCD on pH recoveryafter NH4CIpulse. The figure represents pH i recoveryin Na free media of a single tubule from an acid load, before (A) and after exposure to ,8 agonist (B. C). Acid loading was achieved by exposing cells 1o NH4CI then replaced by NMG and NaCI solutions (conditions as in Fig. IB). This three steps protocol was then applied 20 and 50 rain after the addition of isoprolerenol (2-10-7 M)+ IBMX (10-5 M). NH4CI loads were similar with or without agonist. The arrows indicate the replacement of NH4CI solutk,n by NMG. H emux in Na-free media had a recoveryrate of 0.008 pH unit rain- ! in control conditions (A) and increases to 0.021 in (B) and 0.043 in (C), respectively30 and 60 rain after exposure to isoproterenol. in the presence of external Na was not modified by isoproterenol (n = 6, results not shown). It should be noted that isoproterenol effect on H + efflux vanished with time and was virtually absent when assessed in CCD 4 to 6 h after animal killing. On the other hand, IBMX alone did not stimulate H + ¢fflux rate in Na-free media (n = 3) and isoproterenol effect was not observed in presence of either propanolol ( 3 . 1 0 -6 M) (n = 3) or metoprolol (3" 10 -7 M) a /~t antagonist (n = 4).
T h e effect of cAMP analogues was tested in 20 CCD. Segments were pretreated with IBMX (10 -5 M) for 5 rain before their exposure to either dibutyryi cAMP (10 -4 M, n = 4), chlorophenyl thio-cAMP (10 -4 M, n = 8), 8-Br-cAMP (10 -4 M, n = 5) or forskoline ( 3 - 1 0 -6 M, n = 3) in presence of phosphodiesterase inhibitor for 5 min. No modification of H efflux in N-methylglucamine or Na media could be observed (results not shown).
Discussion 8OO
60o
--
fP
400 •
200'
0
~
-
Control
n=4) Isoprolerenol
.30'
60'
Fig. 6. Effect of isoproterenol on Na-independent H + efflux in CCD. 10 experiments similar to that described in Fig. 5 were conducted. H effluxes rates in Na-free media, obtained 30 rain and 60 rain after isoproterenol addition, are expressed as per cent stimulation compared to control values. (n = 4) represents the segments in which no effect of agonist could be observed.
Cytoplasmic p H is known to be strictly regulated in eukaryotic cells by ion transport pathways and a high buffering capacity in the cytosol. O n e commonly used technique to characterize the different mechanisms involved in cytoplasmic p H regulation, is to induce an acid or alkaline load to the cell and then study the recovery of p H towards initial values. T h e cytosolic acidification can be achieved by the NH4CI prepulse technique [8]. In CCD, NH4CI addition induces a rapid alkalinization which reverses slowly. In a variety of ceils, this recovery phase has been ascribed to N H ~ entry down its electrochemical gradient [8,12-14]. In CCD, there is a considerable participation of N H 4 passive diffusion. H o w e v e r other pathways seem to be also involved as p H decreases below control value. T h e inhibition of the recovery rate by E I P A suggests the participation of the N a - H exchanger to this acidification. Thus, H or more likely NH4CI could be exchanged for internal Na. This is in accord with previous reports implicating this pathway in NH4 secretion [15]. T h e contribution of
23 other pathways such as the K channel or the N a / K pump cannot be excluded. The effect of NEM on the acidification rate is not clear, and deserves further investigation. The distal tubule plays an important role in regulating ac!d secretion [16,17]. Except for the CCD, pathways of acid extrusion in other segment of this part of the nephron are largely uncharacterized. The present study shows that CNT and CCD have at least two similar H transport pathways that regulate cellular pH after acute acid load. The first pathway is a Na-independent slow rate mechanism which is inhibited 'by N-ethylmaleimide (an H+-ATPase inhibitor). In more than half of the CCD segments studied the activity of this pathway was negligeable.The present results do not distinguish the presence of this transporter on the apical or basolateral membrane. The second pathway is a substantially active Na-dependent H efflux which rapidly recovers pH to its normal value. This phase is mediated by Na-H exchanger as it is dependent on external Na with a K m of 38.2 _+ 10.4 mM in CCD, and could be completely inhibited by E.I.P.A, an amiloride derivative. These results are in accord with previous observations by Chaillet et al. [9] obtained in microperfused rabbit CCD. These authors demonstrated the presence of a Na-independent H extrusion mechanism on the apical membrane and a Na-H exchanger on the basolateral membrane with a K m for external Na of 27 mM, slightly lower than the value reported here. A CI/base exchanger has been previously described in intercalated cells. Its contribution to the observed pH regulation remains to be determined. CCD and CNT segments are characterized by at least two cell types: (a) an intercalated carbonic anhydrase rich cell present in both segments and representing 30% of total cell number. (b) An abundant principal cell in CCD and connecting cell in CNT segment and responsible for Na absorption and K secretion. Indirect evidence suggests that transport pathways involved in H secretion are confined to intercalated cells [18,23]. The large variability observed in Na-independent H efflux rate could be due to variations in the number of intercalated cells present in the illuminating spot of light. Alternatively, the loss of in vivo regulatory factors could be responsible for this phenomenon. A similar variability in Na-independent H efflux rate was also reported in microperfused rabbit CCD by ChaiUet et al. [9]. On the other hand, control studies in rabbit CCD have consistently shown a spontaneous decrement in K secretion [19,20] and HCO 3 secretion [2] with time. The reason for this is yet unclear, a decrease in hormonal stimulation, for instance catecbolamine or mineralocorticoids, could be a possible explanation, as these hormones have been shown to modulate either K [19,20], IqCO 3 [2] or H transport (this study).
The regulation of acid secretion in the distal tubule is yet unclear. Physiological and morphological studies have identified two subtypes in intercalated cells. 'A' cells have an H pump on the apical membrane and an electroneutral CI/HCO 3 exchange on the basolateral membrane, in 'B" cells these two transporters would have a switched position the apical membrane containing the CI/HCO 3 exchange and the basolateral membrane the H pump [17,18]. Chabard~s et al. demonstrated that isoproterenol stimulates adenylate cyclase activity in rabbit CNT and CCD and a number of evidence suggests that intercalated cells would be responsible for the reported/3 agonist effect [21]. Several studies have reported that isoproterenol induces a transepithelial depolarization in microperfused rabbit CCD [24]. This effect was found to be associated with a significant decrease in K secretion [19] and to be inhibited by propanolol and mimicked by exogenous cAMP. On the other hand an increase in luminal net HCO3 secretion by cAMP was also reported in rabbit CCD [2]. The present results shows that in CCD cells isoproterenol stimulates Na-independent H efflux mediated by the NEM sensitive H pump. This effect does not occur in presence of propranolol, or metoprolol a /31 antagonist, thus suggesting it could be mediated by a /3t receptor. Surprisingly, the isoproterenol effect on NEM sensitive H + efflux does not seem to be directly mediated by cAMP as it does not appear immediately after the addition of /3 agonist. Furthermore, exogenous cAMP analogues do not modify H pump activi~.y. Although the interpretation of these results is hampered by segment heterogeneity, one cannot exclude other molecular mechanisms which remain to be elucidated. This study has identified the presence of a NEM sensitive H pump and a Na-H exchange in CNT and CCD cells. Isoproterenoi stimulates H pump activity without altering that of the Na-H exchange. The physiological role of this activation is not totally clear. It could participate in the homeostatic regulation of K excretion and net HCO 3 secretion, this latter being implicated in the correction of metabolic alkalosis. Further studies are required to define the role of eatecholamine regulation in H + secretion.
Acknowledgements The authors are grateful to Prof. F. Morel for his constant advice and support during this study, and to Drs. A. Doucet and D. Chabard~s for their helpful comments on an earlier oraft of this 15aper. We are indepted to Sylvie Siaume-Perez for her skilfull contribution in preparing the kidneys.
24 References 1 Koeppen, B.M, and Helman S.I. (1982) Am. J. Physiol. 242, F521-F523. 2 Schuster, V.L. (1985) J. Cliu. Invest. 75, 2056-2064. 3 Morel, F., Chabardes, D. and Imbert-Teboul, M. (1977) Methods Pharmacol. 4. 297-334. 4 Sudo, J. and Morel, F. (1984) Am. J. Physiol. 246, C407-C414. 5 Taniguehi, S., Marchetti, J. and Morel, F. (1989) Pfliigers Arch. 414. 125-133. 6 Weiner, I.D. and Hammi, L.L (1989) Am. J. Physiol. 256, F957F964. 7 Thomas, J.A., Bushbaum. R.N., Fimniak, A. and Racker, S. (1979) Biochem. J. 18, 2210-2218. 8 Roos, A. and Boron, W.F. (1981) Physiol. Rev. 61,296-434. 9 Chaillet, J.R., Lopes, A.G. and Boron, W.F. (1985) J. Gen. Physiol. 86. 795-812. 10 Boyarsky, G., Ganz, M.B., Sterzel, R. and Boron W.F. (1988) Am. J. Physiol. 255, C844-C856. 11 Madshus, I.H., (1988) Biochem. J. 250, 1-8. 12 Boron, W.F. (1983)J. Memhr. Biol. 72, 1-16. 13 Aickin, C.C. and Thomas, R.C. (1975) J. Physiol. (Lond.) 252, 803-815.
14 Thomas, R.C. (19741 J. Physiol. (Lond) 238, 159-18/I. 15 Knepper, M.A.. Packer, R. and Good, D.W. (1989) Physiol. Rev. 69, 179-249. 16 Breyer, M.D. and Jacohson, H.R. (1987) Am. J. Nephtol. 7, 150-161. 17 Steinmetz. R.P. (1986) Am. J. Physiol. 251, F173-FI87. 18 Schwartz, '3.J. and AI-Awqati, Q. 0985) J. Clin. Invest. 75, 1638-1644. 19 Kimmel, P.L and Goldfarb. S. (1984) Am. J. Physiol, 246, F804F810. 20 Schwartz, G.J. and Burg, M.B. (1978) Am. J. Physiol. 235. F576F585. 21 Chabard~s, D., Imbert-Teboul, M., Mont~gut, M., Clique, A and Morel, F. (1975) Pfliigers Arch. 361, 9-15. 22 Knepper, M., Garcia-Austt, J. and Burg, H.B. (1'~84) Proc. Int. Cong. Nephrol. 9, 383a. 23 Hoth/Jfer, H., Sehulte, B.A., Pasternack, G., Siegel, G. and Spicer, S. (1987) Lab. Invest. 57, 150-156. 24 lino, Y., Troy, J.L. and Brenner, B.M. (1081) J. Membr. Biol. 61-73.
Biochimica et Biophysica Acta,
17
Regulation of intracellular pH in rabbit cortical connecting tubule and cortical collecting duct Georges Dagher and Claude Sauterey Laboratoire de Physiologic Celhdaire, Coll~gc de France. Paris (France)
(Received 17 July 1991)
Key words: CytosolicpH: Na-H exchange; Rabbit collectingtubule: H secretion; Isoproterenol Proton transport pathways in isolated superfused rabbit cortical connecting (CNT) and collecting tubules (CCD) were determined using the fluorescent pH-seusitive dye BCECF following acid or base load by exposure to NH4CI. Following removal of NH4CI which results in a rapid decline in pH i two mechanisms a p p e a r to be responsible for pH i recovery, a Na-independent NEM-sensitive H etllux with a 5low activity, which was virtually absent in 30% of the segments tested and a seconO ."~pld l'~a-d~pendent I~ c.~. nx. In CCD this latter pathway was shown to have an apparent K , for (Na+)e of 38.2 + 0.4 mM (S.D., n = 7) and was sensitive to EIPA. Similar results were obtained with the CNT. With regard to the H pump in six out of ten CCD isoproterenol (200 nM) resulted in a 2-fold stimulation of H pump activity. These effects of isoprenaline were inhibited both by the non-specific ~l.adrenoceptor antagonist propranolol as well as by the specific b~ antagonist metoprolol. Interestingly, these stimulatory effects of this ~B agonist, which is known to stimulate cAMP formation in rabbit C C v , wcre not reproduced by the addition of exogenous cAMP analogues db cAMP (0.1 raM), ClWl[" cAMP (0.1 raM), 8 Br-cAMP (0.1 mM) or the addition of forskoline (0.3 mM). In conclusion, these" data obtained in isolated rabbit CNT and CCT demonstrate the presence of a n active Na-H exchange which is for the most part responsible for the recovery of PHi. It should be noted also that the contribution of the H pump to pH i regulation appears to be negligible in these segments.
Introduction The distal part of the nephron plays an important role in the regulation of acid secretion. Much of the information concerning the acidification mechanisms comes from studies on isolated perfused cortical collecting duct (CCD) which showed that acidification is an electrogenic and Na-independent process [1,1638]. In contrast, little is known about H transport in the CNT. The regulation of H secretion is not well defined. PCO 2 and transtubular H C O 3 grad:ents have been shown to elicit rapid insertion of cytoplasmic vesicles believed to carry H + pump units. On the other hand, cAMP was found to stimulate net HCO 3 secretion in rabbit CCD [2].
Abbreviations: CNI, cortical connecting tubules; CCD, cortical collecting ducts. Correspondence: G. Dagher, Laboratoire de physiologicceliutaire. Coll6ge de France, 75231 Paris Cedex 05. France.
In the present study, we have examined the mechanisms by which intracellular pH is regulated on the isolated, superfused CNT and C C D segments from rabbit kidney. In both segments, pH recovery after acute acid load is mediated by a NEM-sensitive Na-independent H + efflux and an Na-H exchange. Isoproterenol markedly stimulates the NEM-sensitive H + elflux.
Methods Granular distal connecting tubules (CNT) and cortical collecting tubules (CCD) were obtained from male New-Zealand white rabbits weighting 1-1.5 kg as previously described [3,4]. Single tubules were isolated from collagenase-treated kidney and incubated in a cold Na-Hepes solution (pH 7.4 at 4 ° C ) containing 4 mM essential and non-essential amino acids derived from Eagles minimal essential medium, Cells were loaded with 2',7'-bis(carboxyethyl)-5(6)-carboxyfluorescein (BCECF) by exposing the tubule to 5 p,M
BCECF-AM for 20 rain at room temperature. The tubule was fixed on a glass cover slip by transfering it to a 1 ml Na-Hepes solution containing 1% agarose and then cooled rapidly to 4°C. The cover-slip was glued to a steel chamber mounted on an inverted Leitz Diavert microscope, equipped with a 40-fold magnification fluorescent Nikon immersion objective. The tubule was constantly superfused, and the solutions and chamber were maintained at 37°C. A detailed description of the chamber and the optical system has been published in a preceeding paper from this laboratory by Tanigushi et al. [5]. pH i was determined by alternately exciting the dye at 450 and 500 nm while measuring the fluorescence emission. A 510 nm dichroic mirror was used and a 530 nm band pass filter was placed immediately before the photometer. A neutral density filter was present in the pathway of the light source (a mercury 100 W lamp) to reduce photobleaching of the dye [6]. The emission intensities for each excitation wavelength increase with pH and remain relatively insensitive to probe concentration. The emitted light intensity was quantitated from a rectangular spot (of about 40 ~m in width and 100/tm in length) focussed on a group of random cells. Intercalated cells could not be reliably distinguished from principal cells by light microscopy. Thus, pH i values reflected a weighted average of pH i of both types of cells. After amplification the emission signal from the photomultiplier was digitized and stored in a microcomputer for analysis. Measurements were averaged for 3 s for each excitation wavelength. The background signal was recorded after each experiment. This non cell related intensity never exceeded 2% of the signal in dye-loaded cells and was subtracted from all the measurements of the experiment. Cellular autofluorescence was assessed in tubules not exposed to the fluorescent probe. The signal was always smaller than 1% of that obtained in BCECF-Ioaded cells. No correction for this signal was made.
p H i calibration pH i was calculated using the nigericin technique as described by Thomas et al. [7]. Calibration studies were performed on each tubule used for pH i measurement. At the end of each experiment, cells were exposed to a calibration solution containing: KCI 110 mM, MgCI 2 1 mM. CaCI 2 1 raM, Hepes 20 mM, choline chloride 30 raM, nigericin 10 p.M. Solutions were adjusted to different pH values ranging from 6.0 to 8.0. When transmembrane K concentration is in equilibrium, internal H would tend to equilibrate with external H concentration the ratio of emission intensity assessed at two excitation wavelength 500 nm/450 nm is a linear function of external pH in the range of 6.3 to 7.8.
Solutions Na-Hepes solution contained in mM: NaCI 145, KCI 5, CaCI 2 l, MgCI 2 1, NaH2PO a 0.6, Na2HPO 4 1.4, glucose 20, Hepes 20 (pH 7.4 at 37 ° C). When necessary 20 or 30 mM NH4CI were substituted for NaCI. In Na-free solution, NaCI was replaced with N-methylglucamine (NMG) (pH 7.4 at 37 ° C). All reagents were analytical grade from Merck. Specific inhibitors, /3 agonist, /3 antagonist, dibutyryl cAMP, chlorophenyl thio-cAMP and forskolin were from Sigma. 8-Br-cAMP was from Boehringer. Results One commonly used approach for studying oH regulating mechanisms is to acutely load the cells with acid or base, and monitor the subsequent recovery of cellular pH towards its initial level [8,12]. In the present study, single isolated segments were transiently exposed to 20 mM ammonium chloride, pH i rises rapidly due to NH 3 influx which traps intracellular H +, and then reverses back due to NH4+ entry which dissociates to form NH 3 that leaves the cell and H + which remains trapped inside. NH4CI medium was then removed and replaced with Na+-free solution (NMG medium). This resulted in cellular acidification due to the NH 3 loss. pH i spontaneously recovers from this acid load as a result of one or more acid extrusion mechanisms located in the luminal and basolateral cell membranes. p H i recovery mechanisms from an acid load A typical pH recovery in CCD cells is shown in Fig. l. Cells were transiently exposed to 20 mM NH4CI (a-c) inducing an increase in cell pH (a-b). When NH4CI was removed by substitution with NMG medium, we observed an acidification of 0.7 pH units (c-d) followed by a pH recovery with a slow rate in the absence of external sodium (d-e). The replacement of this medium with Na + solution (e) induces a rapid pH recovery toward its initial value (e-f). The segment was next exposed to 1 mM N-ethylmaleimide (NEM), inducing a decrease in steady state pH i by about 0.2 pH units if-g). The cells were then acid loaded by a transient exposure to NH4C! (g-i). No pH recovery could be observed in Na-free medium (j-k). The introduction of external sodium induces a rapid pH recovery toward its initial value (k-l). The present results suggest that in CCD two pathways mediate cell pH recovery after an acid load. A slow rate NEM-sensitive, Na+-independent pathway which is inhibited by NEM. The second pathway is a highly active Na+-dependent mechanism which in presence of NEM recovers pH~ to its initial level. This pattern of pH recovery from an acid load was also observed in CNT (Fig. lb).
Na
, NH4
I
~
,
Na
7.8 (A)
r~
i
1
NH4
Ntv~
Nn
I
NH4
l ~
I
~-~ 76
b
f
b
74
,
7.2 7.0 68
d
6.6 6.4 u
•
M~.ures
0'
'
Minutes
Fig. I. pH i recoveryfrom an NH~ pulse in CCD and CNT. (A) CCD bathed in CO, free Na-l-lepes solutions At (a) the bathing solution was switched for one containing 20 mM NHaCI in replacement of NaCI inducing an increase in pH i (b-e). NH4CI solution was replaced by N-methylglucamlnesolution (NMG) at (c) Cell pH decreases rapidly (c-d). and then rec(wers at a rate of 0.044 pH units/rain (d-e). The replacement of NMG solution by NaCI at (e) induces a rapid recoveryof pH, (e-f) with a rate of 0.344 pH units/rain. The presence of I mM N-ethylmaleimide(f) induces a decrease of steady state pH by 0.2 units. Cells were then tee×posed to ;0 mM NH4CI (g) which was switched to N-methylglueamine solution at (i). In presence of NEM, pH i recovery from the induced acid load Q-k) had a rate of 0.003 pH/min. Replacement of NMG solution by NaCI at (k) induces a rapid rise in pH i with a rate of 0.318 pH/min. (B) CNT was bathed it. CO2 free Na-Hepes solution. At (a) the bathing solution was switched for one containing 20 mM NH4CI in replacement for NaCI. At (c) NH~CIsolution was replaced by NMG solution, pH i recovery from the induced acid load (e-d) has a rate of 0.037 pH units/rain (d-e). The replacement of NMG solution by NaCI at (e) induces a rapid alkalinization with a rate of 0.36 oH units/min (e-O. Fig, 2 shows another example of p H recovery from an acid load in CCD cells. Cells were first exposed to NH4CI which was replaced with N M G m e d i u m inducing an acid load of about 0.5 p H units• Stlrprisingly, in this segment no p H recovery could be observed in Na-free m e d i a ( d - e ) . T h e introduction of external sodium induces a rapid p H recovery toward its initial value (e-f). Cells were then treated with E I P A (50 # M ) inducing a decrease in steady state p H by 0.2-0.3 units (Fig. 2, f-g), T h e segment was exposed again to N~ I u-~ I ~
ra
i
I Nm e~o^
•
7.S
b
f
Na
h
7.2
s"
d
-
k
i
S.4
0
4
MINUTES
Fig. 2. Effect of EIPA on pH recoveryfrom a NH4CI pulse in CCD cells. At (a) external solution was switched from Na-Hepes to 20 mM NH4CI. At (c) external NH4 was removed and replaced with Nmethylglucamine(NMG) solution. Na-Hepes solution was reintroduced at (e) causing pH i to recover with an initial rate of 0.45 pH units/rain. At (0 60 g,M EIPA was applied, and cells were reexposed to 20 mM NH4CIpulse. Na-Hepes solution was introduced at (i) and no pH recoverycould be observed(k-I).
NH4CI pulse. Replacing this m e d i u m with Na medium induces an acid load which was not followed by a p H recovery in the presence of EIPA. This suggests that the Na-dependent recovery phase is mediated by the N a - H exchanger. A similar observation was made in five different CCD segments. Interestingly the acid load induced by replacement of NHgCI medium with N M G ( c - d ) was of similar magnitude to that obtained with Na m e d i u m ( i - k ) with EIPA, This indicates that the decrease in pH is determined by N H 4 removal ~s it is expected, rather than by Na removal. We have examined the activity of the Na-independent and the Na-dcpendent mechanisms in 30 CCD and 8 CNT. The recovery rate of the Na-independent pathway was variable ranging from 0.007 to 0.076 p H units m i n - t (mean + S.D.: 0.046 + 0.06) in CCD, and from 0.003 to 0.029 p H units rain-1 in C N T (mean + S.D: 0.018 + 0.009). Interestingly, this mechanism was virtually absent in about 60% of the CCD segments studied (for example, Figs 2 ( d - e ) and 5). T h e initial rate of recovery of the Na-dependent pathway in 30 CCD had a mean + S.D. of 0.31 =]: 0.18 p H uni:~ rain - ~ and of 0.32 + 0.13 o H units r a i n - : in 8 CNT. Cell p H modtdation during exposure to N H 4 CI The recovery phase after NH3 induced alkalinization Fig. 1A ~b-c), can be fitted with a single exponential and has been ascribed in several cell types to passive NH~- entry down its electrochemical gradient. However, as shown in Figs. 1 and 2 after initial alkalin-
20
s6
'
o
o,c,2
'
is
Fig. 3. Effect ouabain and BaCIz on recovery rate after NH 4 pulse in CCD. CCD cells were bathed in a Na-Hepes solution which was subtituted at (a) for one co;llaining 2O mM NH4CI. The acidification phase (a-b) fitted a single exponential with time constant of 1.16 rain- i and an initial rate of 0.85 oH units/min. At (b) cells were bathed back in Na-Hepes solution 1 mM ouabain and I mM naCI 2 were introduced as indicated and cells were reexposed twice to 20 mM NH4CI pulse. The acidification phase could be fitted with a single exponential with an initial rate of 0.38 pH units/rain rain le-f) and 0.12 pH units/min (i-j). On the other hand. in Na-Hepes medium oH recovery could be fitted with a single exnonential. The calculated initial velocities+ S.D. were of 0.96+ 0.07 (c-d), fi.25+ 0.07 (g-h), 0.31+0.03 pH units/rain (k-I). One of four experiments givingsimilar results.
ization pH~ decreases to levels below initial value ( b - c ) . This was observed in 22 out of 30 C C D a n d in 3 out of 8 CNT. Interestingly, in the presence of N E M (Fig. IA) or E I P A (Fig. 2) this d r o p below initial value was not observed, suggesting the participation of o t h e r pathways transporting N H ~ into the cells such as the N a - H exchanger. The decrease below initial values occurs with a simultaneous decrease in 20 m M external Na (replaced by NH4CI) a n d N H 4 entry. This would p r o d u c e a decrease in the outside to inside N a gradient, thus inducing an increase in H influx coupled to Na efflux. This is in accord with the absence of a decrease below initial level in presence of E I P A a n d the effect of o u a b a i n as discussed below. Fig. 3 represents a peculiar p H recovery p a t t e r n observed in only 8 out of 36 C C D segments. Addition of NH4CI resulted in an immediate a n d rapid acidification (Fig. 3, a - b ) . Substitution of Na m e d i u m to this m e d i u m induces an acid load followed by a rapid raise in p H to its initial level. P r e t r e a t m e n t with 1 m M o u a b a i n r e d u c e d by 5 0 % the rate of acidification after NH4CI addition which is still below initial level (e-f). The addition of o u a b a i n could increase cellular Na. This would induce a more p r o n o u n c e d decrease in outside to inside Na gradient in NH4CI medium, a n d consequently to a f u r t h e r increase in N a efflux coupled to H influx. This a p p e a r s in concert with the observation of 75% reduction in the N a - d e p e n d e n t recovery from the acid load in presence of o u a b a i n ( g - h ) , strongly suggesting a reduction in t r a n s m e m b r a n e N a concentration gradient following the increase in intracellular Na. O n the o t h e r hand, one c a n n o t exclude
partial contribution of the Na p u m p to N H 4 influx. The addition of BaCl 2 in presence of oubain induced a further decrease in the rate of acidification after NH4CI addition (i-j vs. e - O . U n d e r these conditions steady state p H fails to d r o p below initial p H levels. Removal of NH4CI induced an acid load which was followed by a recovery in Na medium.
Kinetics of Na-H exchange as a function of external sodium T h e above results clearly d e m o n s t r a t e that in C N T a n d C C D , pH~ recovery from N i l 4 induced acid load is much m o r e active in presence of Na than in its absence. In o r d e r to quantify the Na d e p e n d e n c y of this rapid phase, we p e r f o r m e d a series of experiments in which we assessed H efflux at increasing external concentrations of Na as previously described by Chaillet et al. [9]. C C D cells were p r e t r e a t e d with N-ethylmaleimide (1 mM) to inhibit H + efflux in Na-free m e d i a a n d were then acid loaded by exposure to NH4CI. As shown in Fig. 4A, H + efflux was absent in N M G m e d i u m ( b - c ) . T h e addition of 4.5 ~.'tM external Na induced a n increase in pHi. External Na was then removed inducing a slow decree.se in pH~ to a stable value. Addition of 10 m M external N a induces a substantially higher p H i recovery rate which was partially reversed u p o n removal of t=xternal Na. T h e addition of 40 m M N a results in a m o r e r a p i d pH~ recovery which reverses in N M G medium, pH~ recovery rate was even higher at 145 m M N a a n d could be fitted to a single exponential giving a n initial slope of 0.61 p H units m i n - t . F o r e a c h tubule, the initial recovery rate assessed at increasing external c o n c e n t r a tions of N a were normalized to the initial recovery rate at 145 m M Na. The results are shown in Fig. 4B. A K m value for external Na of 38.2 + 10.4 m M was o b t a i n e d by fitting the results to a Miehaelis-Menten equation using a n o n linear analysis p r o g r a m (Enzfitter from Elsevier-Biosoft, Cambridge, U.K). A H a n e s - W o o l f plot of the d a t a is shown in the inset. In this kinetic analysis we have assumed that intracellular N a was uniformely low as the tubule was exposed to Na-free m e d i u m before the addition of increasing external c o n c e n t r a tions of Na.
Effect of isoproterenol C C D cells were acid loaded by exposure to NH4CI prepulse a n d N a - i n d e p e n d e n t a n d N a - d e p e n d e n t H efflux mechanisms were assessed before a n d a f t e r exposure of the tubule to isoproterenol ( 2 ' 10 -7 M)-II B M X (10 - s M) for 3 to 5 min. As seen in Fig. 5, H efflux in Na-free m e d i u m was virtually absent. This has b e e n observed in a large n u m b e r of segments as discussed above. Addition of isoproterenol did not modify steady state p H for the
Fig. 6 s u m m a r i z e s t h e e f f e c t o f i s o p r o t e r e n o l on N a - i n d e p e n d e n t , H efflux in I0 C C D s e g m e n t s , in 6 o f t h e m , a 1 to 6-fold stimulation o f N E M sensitive H + efflux w a s o b s e r v e d in p r e s e n c e o f /3 agonist. In 4 o t h e r s e g m e n t s no significant effect could be o b s e r v e d . M e a n H efflux w a s of 0.077_+0.105 u n d e r control c o n d i t i o n s a n d o f 0.118 _+ 0.112 a f t e r i s o p r o t e r e n o l addition ( P < 0.05 by p a i r e d t-test; a n d P < 0.006 by Wilcox0n s i g n e d R a n k test.) In contrast, p H recovery
following 10 min. Successive N H 4 C I i n d u c e d acid load r e v e a l e d a significant s t i m u l a t i o n o f the N a ~ndcpend e n t H efflux 20 to 30 m i n a f t e r e x p o s u r e to agonist. F u r t h e r , this s t i m u l a t o r y e f f e c t s e e m s to be m o r e potent 50 to 60 rain a f t e r i s o p r o t e r e n o l addition. O n t h e o t h e r h a n d , H efflux r a t e s in N a - f r e e m e d i a w e r e highly r e p r o d u c i b l e in control c o n d i t i o n s as recovery r a t e s a f t e r two s u b s e q u e n t acid loads s h o w e d a maxim a l v a r i a t i o n o f 2 0 % ( m e a n : 12%, n = 7).
I NH4 ] Na
I0
14~10
I~0
I
0
,I 4°1
0
1145
7.9 7.7 7.5 7.3 7.1 6.9 6,7 6.5
I
I
0
I0 minutes
I B) 1.0 o.8
< a:
0.6
0.4
i,o
0.2
I
'
:o
I
I
I
I
40
80
120
160
Na ( m M )
Fig. 4. External Na-dependence of pHi recovery from an NH~ induced acid load in CCD. (A) pH recovery at different concentration of e:~,.rnai Na. Cells were pretreatcd with 1 mM NEM. At (a) Na-Hepcs solution was switched for an isoosmotic 20 mM NH4CI solution. NMG solutioti was introduced at (b), (d), (f) and (h). The solutions applied at (c), (e), (g), (i) contained (in raM): 4.5. 10, 40 and 145 NaCI in replacement o[ an equivalent amount of N-methylglucamine (NMG). Recovery rates from the acid load were of 0.074 (c-d), 0.167 (e-0,0.520 (g-h) and 0.613 pH units/rain (i-i). (B) Normalized acid extrusion as a function of external [Na]. Data were derived from seven experiments similar to that in A the mean initial recovery rate + S.D. at 145 mM Ha was of 0.35 + O.14 pH units/rain. Fluxes obtained at 4.5, tO and 40 mM Na were divided by flux calculated for same pH i at 145 mM [Na] o. phi values at (c), (c), (g). (i) had mean values___S.D. of: 6,76± 0.21, 6.87+0.21.6.88±0.26, 6.81 d:0.23, respectively. A Hanes-Woolf plot of these data is shown in the inset.
7.8
NH4
7.6
4,
7.4 7.2
B
:~ o. 7.0 6.8 6.6 6.4 6.2
I
1
I
o
4 minutes
Fig. 5. Effect of isoproterenol and IBMX in CCD on pH recoveryafter NH4CIpulse. The figure represents pH i recoveryin Na free media of a single tubule from an acid load, before (A) and after exposure to ,8 agonist (B. C). Acid loading was achieved by exposing cells 1o NH4CI then replaced by NMG and NaCI solutions (conditions as in Fig. IB). This three steps protocol was then applied 20 and 50 rain after the addition of isoprolerenol (2-10-7 M)+ IBMX (10-5 M). NH4CI loads were similar with or without agonist. The arrows indicate the replacement of NH4CI solutk,n by NMG. H emux in Na-free media had a recoveryrate of 0.008 pH unit rain- ! in control conditions (A) and increases to 0.021 in (B) and 0.043 in (C), respectively30 and 60 rain after exposure to isoproterenol. in the presence of external Na was not modified by isoproterenol (n = 6, results not shown). It should be noted that isoproterenol effect on H + efflux vanished with time and was virtually absent when assessed in CCD 4 to 6 h after animal killing. On the other hand, IBMX alone did not stimulate H + ¢fflux rate in Na-free media (n = 3) and isoproterenol effect was not observed in presence of either propanolol ( 3 . 1 0 -6 M) (n = 3) or metoprolol (3" 10 -7 M) a /~t antagonist (n = 4).
T h e effect of cAMP analogues was tested in 20 CCD. Segments were pretreated with IBMX (10 -5 M) for 5 rain before their exposure to either dibutyryi cAMP (10 -4 M, n = 4), chlorophenyl thio-cAMP (10 -4 M, n = 8), 8-Br-cAMP (10 -4 M, n = 5) or forskoline ( 3 - 1 0 -6 M, n = 3) in presence of phosphodiesterase inhibitor for 5 min. No modification of H efflux in N-methylglucamine or Na media could be observed (results not shown).
Discussion 8OO
60o
--
fP
400 •
200'
0
~
-
Control
n=4) Isoprolerenol
.30'
60'
Fig. 6. Effect of isoproterenol on Na-independent H + efflux in CCD. 10 experiments similar to that described in Fig. 5 were conducted. H effluxes rates in Na-free media, obtained 30 rain and 60 rain after isoproterenol addition, are expressed as per cent stimulation compared to control values. (n = 4) represents the segments in which no effect of agonist could be observed.
Cytoplasmic p H is known to be strictly regulated in eukaryotic cells by ion transport pathways and a high buffering capacity in the cytosol. O n e commonly used technique to characterize the different mechanisms involved in cytoplasmic p H regulation, is to induce an acid or alkaline load to the cell and then study the recovery of p H towards initial values. T h e cytosolic acidification can be achieved by the NH4CI prepulse technique [8]. In CCD, NH4CI addition induces a rapid alkalinization which reverses slowly. In a variety of ceils, this recovery phase has been ascribed to N H ~ entry down its electrochemical gradient [8,12-14]. In CCD, there is a considerable participation of N H 4 passive diffusion. H o w e v e r other pathways seem to be also involved as p H decreases below control value. T h e inhibition of the recovery rate by E I P A suggests the participation of the N a - H exchanger to this acidification. Thus, H or more likely NH4CI could be exchanged for internal Na. This is in accord with previous reports implicating this pathway in NH4 secretion [15]. T h e contribution of
23 other pathways such as the K channel or the N a / K pump cannot be excluded. The effect of NEM on the acidification rate is not clear, and deserves further investigation. The distal tubule plays an important role in regulating ac!d secretion [16,17]. Except for the CCD, pathways of acid extrusion in other segment of this part of the nephron are largely uncharacterized. The present study shows that CNT and CCD have at least two similar H transport pathways that regulate cellular pH after acute acid load. The first pathway is a Na-independent slow rate mechanism which is inhibited 'by N-ethylmaleimide (an H+-ATPase inhibitor). In more than half of the CCD segments studied the activity of this pathway was negligeable.The present results do not distinguish the presence of this transporter on the apical or basolateral membrane. The second pathway is a substantially active Na-dependent H efflux which rapidly recovers pH to its normal value. This phase is mediated by Na-H exchanger as it is dependent on external Na with a K m of 38.2 _+ 10.4 mM in CCD, and could be completely inhibited by E.I.P.A, an amiloride derivative. These results are in accord with previous observations by Chaillet et al. [9] obtained in microperfused rabbit CCD. These authors demonstrated the presence of a Na-independent H extrusion mechanism on the apical membrane and a Na-H exchanger on the basolateral membrane with a K m for external Na of 27 mM, slightly lower than the value reported here. A CI/base exchanger has been previously described in intercalated cells. Its contribution to the observed pH regulation remains to be determined. CCD and CNT segments are characterized by at least two cell types: (a) an intercalated carbonic anhydrase rich cell present in both segments and representing 30% of total cell number. (b) An abundant principal cell in CCD and connecting cell in CNT segment and responsible for Na absorption and K secretion. Indirect evidence suggests that transport pathways involved in H secretion are confined to intercalated cells [18,23]. The large variability observed in Na-independent H efflux rate could be due to variations in the number of intercalated cells present in the illuminating spot of light. Alternatively, the loss of in vivo regulatory factors could be responsible for this phenomenon. A similar variability in Na-independent H efflux rate was also reported in microperfused rabbit CCD by ChaiUet et al. [9]. On the other hand, control studies in rabbit CCD have consistently shown a spontaneous decrement in K secretion [19,20] and HCO 3 secretion [2] with time. The reason for this is yet unclear, a decrease in hormonal stimulation, for instance catecbolamine or mineralocorticoids, could be a possible explanation, as these hormones have been shown to modulate either K [19,20], IqCO 3 [2] or H transport (this study).
The regulation of acid secretion in the distal tubule is yet unclear. Physiological and morphological studies have identified two subtypes in intercalated cells. 'A' cells have an H pump on the apical membrane and an electroneutral CI/HCO 3 exchange on the basolateral membrane, in 'B" cells these two transporters would have a switched position the apical membrane containing the CI/HCO 3 exchange and the basolateral membrane the H pump [17,18]. Chabard~s et al. demonstrated that isoproterenol stimulates adenylate cyclase activity in rabbit CNT and CCD and a number of evidence suggests that intercalated cells would be responsible for the reported/3 agonist effect [21]. Several studies have reported that isoproterenol induces a transepithelial depolarization in microperfused rabbit CCD [24]. This effect was found to be associated with a significant decrease in K secretion [19] and to be inhibited by propanolol and mimicked by exogenous cAMP. On the other hand an increase in luminal net HCO3 secretion by cAMP was also reported in rabbit CCD [2]. The present results shows that in CCD cells isoproterenol stimulates Na-independent H efflux mediated by the NEM sensitive H pump. This effect does not occur in presence of propranolol, or metoprolol a /31 antagonist, thus suggesting it could be mediated by a /3t receptor. Surprisingly, the isoproterenol effect on NEM sensitive H + efflux does not seem to be directly mediated by cAMP as it does not appear immediately after the addition of /3 agonist. Furthermore, exogenous cAMP analogues do not modify H pump activi~.y. Although the interpretation of these results is hampered by segment heterogeneity, one cannot exclude other molecular mechanisms which remain to be elucidated. This study has identified the presence of a NEM sensitive H pump and a Na-H exchange in CNT and CCD cells. Isoproterenoi stimulates H pump activity without altering that of the Na-H exchange. The physiological role of this activation is not totally clear. It could participate in the homeostatic regulation of K excretion and net HCO 3 secretion, this latter being implicated in the correction of metabolic alkalosis. Further studies are required to define the role of eatecholamine regulation in H + secretion.
Acknowledgements The authors are grateful to Prof. F. Morel for his constant advice and support during this study, and to Drs. A. Doucet and D. Chabard~s for their helpful comments on an earlier oraft of this 15aper. We are indepted to Sylvie Siaume-Perez for her skilfull contribution in preparing the kidneys.
24 References 1 Koeppen, B.M, and Helman S.I. (1982) Am. J. Physiol. 242, F521-F523. 2 Schuster, V.L. (1985) J. Cliu. Invest. 75, 2056-2064. 3 Morel, F., Chabardes, D. and Imbert-Teboul, M. (1977) Methods Pharmacol. 4. 297-334. 4 Sudo, J. and Morel, F. (1984) Am. J. Physiol. 246, C407-C414. 5 Taniguehi, S., Marchetti, J. and Morel, F. (1989) Pfliigers Arch. 414. 125-133. 6 Weiner, I.D. and Hammi, L.L (1989) Am. J. Physiol. 256, F957F964. 7 Thomas, J.A., Bushbaum. R.N., Fimniak, A. and Racker, S. (1979) Biochem. J. 18, 2210-2218. 8 Roos, A. and Boron, W.F. (1981) Physiol. Rev. 61,296-434. 9 Chaillet, J.R., Lopes, A.G. and Boron, W.F. (1985) J. Gen. Physiol. 86. 795-812. 10 Boyarsky, G., Ganz, M.B., Sterzel, R. and Boron W.F. (1988) Am. J. Physiol. 255, C844-C856. 11 Madshus, I.H., (1988) Biochem. J. 250, 1-8. 12 Boron, W.F. (1983)J. Memhr. Biol. 72, 1-16. 13 Aickin, C.C. and Thomas, R.C. (1975) J. Physiol. (Lond.) 252, 803-815.
14 Thomas, R.C. (19741 J. Physiol. (Lond) 238, 159-18/I. 15 Knepper, M.A.. Packer, R. and Good, D.W. (1989) Physiol. Rev. 69, 179-249. 16 Breyer, M.D. and Jacohson, H.R. (1987) Am. J. Nephtol. 7, 150-161. 17 Steinmetz. R.P. (1986) Am. J. Physiol. 251, F173-FI87. 18 Schwartz, '3.J. and AI-Awqati, Q. 0985) J. Clin. Invest. 75, 1638-1644. 19 Kimmel, P.L and Goldfarb. S. (1984) Am. J. Physiol, 246, F804F810. 20 Schwartz, G.J. and Burg, M.B. (1978) Am. J. Physiol. 235. F576F585. 21 Chabard~s, D., Imbert-Teboul, M., Mont~gut, M., Clique, A and Morel, F. (1975) Pfliigers Arch. 361, 9-15. 22 Knepper, M., Garcia-Austt, J. and Burg, H.B. (1'~84) Proc. Int. Cong. Nephrol. 9, 383a. 23 Hoth/Jfer, H., Sehulte, B.A., Pasternack, G., Siegel, G. and Spicer, S. (1987) Lab. Invest. 57, 150-156. 24 lino, Y., Troy, J.L. and Brenner, B.M. (1081) J. Membr. Biol. 61-73.
