Pflügers Arch – Eur J Physiol (1999) 437:716–723

© Springer-Verlag 1999

O R I G I N A L A RT I C L E

E. Desfleurs · M. Wittner · S. Pajaud · R. Nitschke R.M. Rajerison · A. Di Stefano

The Ca2+-sensing receptor in the rabbit cortical thick ascending limb (CTAL) is functionally not coupled to phospholipase C

Received: 28 September 1998 / Received after revision: 3 November 1998 / Accepted: 9 December 1998

Abstract The recently cloned rabbit kidney Ca2+-sensing receptor (RabCaR) was functionally characterized in microperfused rabbit cortical thick ascending limb (CTAL) segments. Reverse transcriptase polymerase chain reaction (RT-PCR) confirmed that this nephron segment contains mRNAs coding for the RabCaR. Elevation of the extracellular Ca2+ concentration ([Ca2+]e) from 1 to 5 mmol l–1 induced an increase in the fluorescence emission ratio (R), thus reflecting an increase in intracellular Ca2+ activity ([Ca2+]i). This increase was inhibited by verapamil, nifedipine and SKF 96365, and potentiated by a previous application of Bay K 8644. Neither verapamil nor Bay K 8644 modified the resting [Ca2+]i. This suggests that the basolateral Ca2+ influx induced by a high [Ca2+]e occurs via verapamil- and dihydropyridine-sensitive Ca2+ channels, which are not open under resting conditions. In contrast to that evoked by antidiuretic hormone (ADH), the [Ca2+]i increase induced by a high [Ca2+]e did not result from an accumulation of inositol phosphates. Neomycin, Gd3+, Mg2+, commonly used agonists of the Ca2+-sensing receptor, did not increase the [Ca2+]i. In the presence of verapamil, ADH still produced a transient [Ca2+]i increase that was not observed in the presence of an increased [Ca2+]e. These results suggest that the RabCaR in rabbit CTAL cells is not functionally coupled to phospholipase C. In conclusion, the high [Ca2+]e-induced [Ca2+]i increase involves verapamil- and dihydropyridine-sensitive Ca2+ channels and is independent of phosphoinositide metaboE. Desfleurs · M. Wittner · A. Di Stefano (✉) Département de Biologie Cellulaire et Moléculaire, CEA Saclay, URA CNRS 1859, F-91191 Gif-sur-Yvette, France e-mail: [email protected] Tel.: +33-1-69087841 or +33-1-69087462 Fax: +33-1-69083570 R. Nitschke Physiologisches Institut, Albert-Ludwigs-Universität, Hermann-Herder-Str.7, D-79104 Freiburg i. Br., Germany R.M. Rajerison INSERM, Unité 367, 17, rue du Fer-à-Moulin, F-75 005 Paris, France

lism. Whether these channels are activated by the RabCaR remains to be elucidated. Key words Calcium-sensing receptor · CTAL · Cytosolic calcium · Kidney · Phospholipase C · V1a receptor · RabCaR

Introduction Different types of cells are able to sense changes in the extracellular Ca2+ concentration ([Ca2+]e) and to respond by varying their intracellular Ca2+ activity ([Ca2+]i) [2, 5, 21, 26]. The mechanisms involved in the high-[Ca2+]einduced [Ca2+]i increase are well described for cells of the parathyroid gland, from which a Ca2+-sensing receptor (BoPCaR) has been cloned [2, 3]. Such a receptor has also been functionally described in studies of the rat and mouse cortical thick ascending limb (CTAL) [21, 22]. In these cells, activation of the G-protein-coupled Ca2+sensing receptor (CaR) leads to a peak–plateau increase in the [Ca2+]i, which results from a release of Ca2+ from intracellular inositol 1,4,5-trisphosphate- (IP3-) sensitive stores, and an influx of extracellular Ca2+ through basolateral Ca2+ channels. The mechanisms involved in the [Ca2+]i increase are not always well documented in studies of other tissues or species where this receptor has been cloned or functionally described. This is the case for C cells of the rat thyroid gland [13], where the [Ca2+]i increase occurs in part through dihydropyridine-sensitive Ca2+ channels [25], and for the human and rabbit kidney [1, 4], where these mechanisms have not yet been studied extensively. Cloning of the rabbit CaR (RabCaR) reveals that this receptor shares a high degree of homology (>90% amino acid identity) with the bovine parathyroid (BoPCAR) and human and rat kidney (RaKCaR) Ca2+-sensing receptor [4]. In fact, when assayed functionally in transiently transfected HEK 293 cells, the RabCaR shows apparent affinities for Ca2+, Mg2+ and Gd3+ that are undistinguishable from those observed for the human CaR. Therefore, it

717

was concluded that the hypercalcemia normally present in New Zealand white rabbits and associated with an elevated set point for Ca2+-regulated release of parathyroid hormone (PTH) cannot be explained by different intrinsic functional properties of the RabCaR. However, to date, the transduction mechanisms for this receptor have not been studied. The aim of the present study was therefore to characterize functionally the recently cloned RabCaR in rabbit CTAL cells, as immunological studies reveal that this is where it is expressed. First, we investigated the effect of increasing the [Ca2+]e on the [Ca2+]i in isolated perfused rabbit CTAL segments, and then we searched for the mechanisms involved in the observed [Ca2+]i increase. We tested the effects of commonly used CaR agonists on the [Ca2+]i response and inositol phosphate production to study the transduction pathways involved in the high[Ca2+]e-induced [Ca2+]i increase in this nephron segment. Finally, the responses were compared with those obtained with ADH, a hormone known to activate phospholipase C and to increase the [Ca2+]i via activation of its V1a receptor in the rabbit CTAL [20].

Materials and methods Measurements of [Ca2+]i Experiments were carried out on isolated perfused CTAL of the rabbit nephron as usual in our laboratory [24]. CTAL segments were dissected at 4°C in a standard solution from kidney slices of female New Zealand rabbits (body weight: 700–800 g). Dissected tubules were loaded with the acetoxymethyl ester of the Ca2+-sensitive dye fura-2 (fura-2/AM, 2×10–6 mol/l) in the presence of pluronic F-127 (0.1 g/l) at room temperature in the dark for about 45 min. Each fura-2-loaded tubule was then transferred to a perfusion chamber mounted on the stage of an inverted microscope (Axiovert 10, Zeiss, France) and perfused by gravity. Loading of the tubules was stopped by washing them with the standard solution at 37°C at a flow rate of approximately 15 ml/min. After an equilibration period of 5–10 min, fura-2 fluorescence was measured photometrically, as originally described by R. Nitschke, U. Fröbe and R. Greger (Physiologisches Institut Freiburg, Germany) [20]. Calibration of the fura-2 fluorescence signal was carried out with the so-called “without tubule” calibration procedure, using a fura-2 concentration in the perfusion chamber of 2×10–6 mol/l. Solutions containing 5 mmol/l EGTA and zero Ca2+ for the determination of Rmin and 2.0 mmol/l for the determination of Rmax were used. λ, at 380 nm excitation, is the ratio of the fura-2 fluorescence signal at zero Ca2+ over the fura-2 fluorescence signal at a saturating Ca2+ concentration. The estimated Rmin value was 0.39±0.04, Rmax was 8.15±0.8 and λ was 11.7±1.5 (n=10). Measurement of phosphoinositidase C activity All measurements were carried out as described in detail elsewhere [18]. The following modifications were made: tubules were dissected without collagenase and labelled with the standard solution containing [myo-3H]inositol for 2 h at 37°C instead of at 30°C. Results are expressed as the ratio between the measured radioactivity of the inositol phosphates and the total radioactivity measured from all compounds with the labelled [myo-3H]inositol.

RT-PCR of the

Ca2+-sensing

receptor

CTAL segments were isolated from rabbit kidney treated with collagenase as described elsewhere [11]. RNAs were extracted from the microdissected segments (total tubular length 35 mm, corresponding to about 60 tubular segments), using a micromethod [11] adapted from the guanidinium thiocyanate-phenol/chloroform method developed by Chomczynski and Sacchi [6]. The primers used for RT-PCR were selected from the sequence of RabCaR cDNAs [4] using Oligo Primer Analysis Software (Medprobe, Oslo, Norway). A 457-bp cDNA fragment was obtained by RT-PCR from 1 mm total RNA extracted from the rabbit CTAL. The sense (5’–GAAAGCATCCCAGGCAATCTCTAC–3’) and antisense (5’–GTAACCAGCGAAAACCACGAAA–3’) primers corresponded to positions 1754–1777 and 2189–2210, respectively, from the beginning of the published sequence [4]. Reverse transcription was carried out at 41°C for 45 min. Then amplification was performed over 31 cycles (95°C for 30 s, 56°C for 30 s, 72°C for 1 min), followed by an additional cycle with an elongation time of 10 min. The presence of possible contaminants was checked by control RT-PCR reactions, carried out using samples from which RNA was excluded (blank) and by omitting reverse transcriptase from the reverse transcription mixture (RT–). Solutions and chemicals Fluorescence measurements Lumen (in mmol/l): 145 NaCl, 1.6 K2HPO4, 0.4 KH2PO4, 1.0 CaCl2, 1.0 MgCl2. Bath (in mmol/l): 145 NaCl, 1.6 K2HPO4, 0.4 KH2PO4, 1.0 CaCl2, 1.0 MgCl2, 5 D-glucose. The pH of the solutions was adjusted to 7.4. Pluronic F-127, fura-2 penta-potassium salt and fura-2/AM were obtained from Calbiochem (La Jolla, Calif., USA), [myo3H]inositol (370–740 GBq/mmol, or 10–20 Ci/mmol) and [α32P]dCTP were obtained from Amersham International (Buckinghamshire, UK). Ribonuclease inhibitor (rRNasin) was purchased from Promega, Madison, Wis., USA; 2’-deoxynucleotide 5’-triphosphate from Pharmacia, Orsay, France; Moloney Murine Leukemia Virus reverse transcriptase from Life Technologies, USA and Taq Polymerase from Eurobio, Les Ulis, France. All other chemicals were of the highest grade of purity available and obtained from Sigma (St-Quentin Fallavier, France). Statistical analysis The data are mean values ±SEM (n), where n refers to the number of tubules, except for measurements of inositol phosphates where n refers to the number of animals. The Anova test and Bonferroni’s multiple comparison test were used to compare mean values within each experimental series. Student’s unpaired t-test was used to compare mean values of two experimental series. A P value of <0.05 was accepted as indicating statistical significance.

Results Identification of a RabCaR mRNA fragment in the rabbit CTAL To confirm the presence of RabCaR transcripts in the rabbit CTAL, RT-PCR was performed on isolated microdissected tubules and entire rabbit kidney. As shown in Fig. 1, a single PCR product of 457 bp was consistently found to be abundantly expressed in cDNA from CTAL

718

Fig. 1 Expression of rabbit kidney Ca2+-sensing receptor (RabCaR) mRNA in isolated rabbit cortical thick ascending limb (CTAL) segments and kidney preparation. RNA corresponding to 1 mm CTAL were submitted to reverse transcriptase polymerase chain reaction (RT-PCR, 32 cycles in the presence of [α32P]dCTP). For CTAL segments, the reaction was also carried out in the absence of reverse transcriptase (RT–). (Bl Blank – reaction performed without RNA.) The products of the RT-PCR were analysed on a 2% agarose gel and detected by autoradiography. The size of the band indicated on the figure was evaluated by comparison with the molecular weight marker pBR322 MspI digest stained with ethidium bromide on the agarose gel

Fig. 2A, B Two [Ca2+]i recordings showing the effect of an increase in the peritubular Ca2+ concentration from 1 to 5 mmol l–1 on the 340/380 excitation fluorescence emission ratio (R) of two isolated perfused rabbit CTAL segments. A The increase in the [Ca2+]i exhibited only a plateau; B the plateau was preceded by a transient [Ca2+]i increase

and rabbit kidney, suggesting the presence of mRNA coding for the RabCaR in each of the samples. The band was absent when reverse transcriptase was omitted. This observation was made twice. Effect of changes in the peritubular Ca2+ concentration ([Ca2+]e) on the [Ca2+]i of non-stimulated tubules CTAL segments, perfused with 1 mmol l–1 Ca2+ in the lumen and bath, presented a fluorescence ratio of 0.69±0.01 (n=227), which corresponds to a [Ca2+]i of 105±1 nmol l–1. Rabbit CTAL cells are sensitive to changes in [Ca2+]e. Increasing the [Ca2+]e from 1 to 5 mmol l–1 induced within about 30 s significant increases in R (plateau phase) from 0.69±0.01 to 0.88±0.02 (n=227). In 199 CTAL segments the [Ca2+]i increased to a variable degree, whereas 28 tubules did not respond. These non-responsive tubules had a resting [Ca2+]i similar to that of the other tubules and also a similar transepithelial voltage (data not shown). Furthermore, it is interesting to note that the tubules obtained from each animal displayed similar variations in [Ca2+]i in response to high [Ca2+]e. Therefore, the observed variability of the [Ca2+]i increase in response to a high [Ca2+]e cannot be attributed to tubule variability within the same animal. Among tubules that did respond to changes in [Ca2+]e, two types of kinetics of the [Ca2+]i increase were recorded. The first was characterized by a stable plateau (Fig. 2A) and the second by a biphasic response, consisting of an initial rapid [Ca2+]i increase which then relaxed to a lower stable plateau value (Fig. 2B). Of the 199 tubules responding to 5 mmol l–1 Ca2+, 153 CTALs showed the monophasic response and 46 tubules the biphasic one. The [Ca2+]i response of the CTAL cells to a high [Ca2+]e showed no desensitization, because several successive [Ca2+]i responses could be obtained upon repetitive elevation of [Ca2+]e. Figure 3 illustrates the dependence of the [Ca2+]i increase, measured at the plateau phase, on

Fig. 3 The concentration dependency of the increase in the fluorescence ratio 340/380 (R) in response to raised peritubular Ca2+ concentrations. Values correspond to R measured at the maximum of the response to the increases in peritubular Ca2+ concentrations from 1 to 2, 3, 4 and 5 mmol l–1. Measurements were performed in a paired fashion on the same tubule. The columns represent means ±SEM of data from 14 tubules. *Significantly different from the pre- and post-experimental control period (C1, C2)

the [Ca2+]e level, which was increased progressively from 1 to 2, 3, 4 and 5 mmol l–1. A significant increase in R was observed with 2 mmol l–1 peritubular Ca2+ and the maximal effect was reached at 4 mmol l–1 Ca2+. On the other hand, reducing [Ca2+]e to 10–8 mol l–1 resulted in a significant decrease in R from 0.67±0.04 to 0.49±0.02 (n=21). When Ca2+ was added again, R returned to the control value. No change in [Ca2+]i was seen when the luminal Ca2+ concentration was modified (data not shown). Effect of ADH on the [Ca2+]i of rabbit CTAL cells Figure 4A shows an individual recording of the effect of 10–8 mol l–1 ADH in the bath solution on the R value of an isolated perfused CTAL segment. In the presence of ADH, R rapidly increased within 30 s to a maximum (peak) and then decreased to a plateau despite the contin-

719

Fig. 4A, B Examples of two [Ca2+]i recordings showing the effect of antidiuretic hormone (ADH, 10–8 mol l–1, bath) on the 340/380 excitation fluorescence emission ratio (R) of an isolated perfused rabbit CTAL segment in the presence of 1 mmol l–1 peritubular Ca2+ (A) and in the presence of 5 mmol l–1 peritubular Ca2+ (B). The peritubular Ca2+ concentration was increased from 1 to 5 mmol l–1 at the same time as ADH was added to the bath solution. Note that the plateau phase of the [Ca2+]i response to ADH was greater with 5 than with 1 mmol l–1 Ca2+, but that it was not different from that obtained with 5 mmol l–1 Ca2+ alone

Fig. 5 Effect of 5 mmol l–1 Ca2+ or 10–8 mol l–1 ADH on inositol phosphate (IP) production in rabbit CTAL cells. Data are means ±SEM of IP levels expressed as percentages of total radioactivity found in all inositol-containing pools. IP levels were measured in a paired manner under control conditions with 1 mmol l–1 Ca2+ (C) and under experimental conditions (5 mmol l–1 Ca2+, 10–8 mol l–1 ADH). (n Number of experiments.) In each experiment, five measurements were performed for each condition. *Significantly different from the control value (C)

ued presence of the hormone. Of the 16 tubules responding to ADH, the average value of R transiently increased after ADH application from a basal value of 0.63±0.03 to a peak value of 0.96±0.05 and then fell to 0.75±0.04 (plateau phase). After the hormone was removed, R returned to 0.63±0.03 in the post-experimental control period. As mentioned for the experiments in which the [Ca2+]e was increased from 1 to 5 mmol l–1, not all animals responded to ADH. Out of 21 animals tested, 5 animals did not respond to ADH. The tubules of the responding animals as well as those of the non-responding ones displayed comparable basal [Ca2+]i and transepithelial voltage values (data not shown).

crease in R from 0.66±0.05 to 0.86±0.06 (n=16); with 5 mmol l–1 Ca2+ alone the increase in R was from 0.68±0.06 to 0.81±0.06 (n=16).

Effect of elevated [Ca2+]e on the [Ca2+]i of CTAL segments stimulated by ADH In the rabbit CTAL, the plateau phase of the ADHevoked response is sensitive to a decrease in the [Ca2+]e from 1 to nominally 0 mmol l–1 [20]. The effect of increasing [Ca2+]e from 1 to 5 mmol l–1 on the plateau phase of the ADH-evoked response was therefore investigated. Figure 4B depicts an individual tracing showing that, when compared to the ADH-evoked response in the presence of 1 mmol l–1 Ca2+ (Fig. 4A), 5 mmol l–1 peritubular Ca2+ resulted in an increased plateau phase, but that it did not influence the magnitude of the peak of the ADH-evoked response. Thus, in the presence of 5 mmol l–1 Ca2+, the plateau phase of the [Ca2+]i response to ADH was significantly larger, with an increase in R of 0.22±0.03 (n=14), when compared to 0.12±0.02 (n=16) in the presence of 1 mmol l–1 Ca2+. However, no difference could be observed between the plateau phase obtained with 5 mmol l–1 Ca2+ in the presence or in the absence of ADH (Fig. 4B): addition to the bath solution of 10–8 mol l–1 ADH and 5 mmol l–1 Ca2+ induced an in-

Characterization of the [Ca2+]i increase induced by high [Ca2+]e Search for a phospholipase-C-coupled CaR Inositol phosphate formation in rabbit CTAL cells was measured after application of 10–8 mol l–1 ADH or 5 mmol l–1 Ca2+ (Fig. 5). As expected from the peak–plateau kinetics of the [Ca2+]i response to ADH, the hormone increased inositol phosphate synthesis. By contrast, 5 mmol l–1 Ca2+ had no effect on inositol phosphate synthesis, suggesting that the [Ca2+]i increase observed after increases in [Ca2+]e is not mediated by increased inositol phosphate formation via a G-protein-coupled sensing mechanism. ADH experiments and those with 5 mmol l–1 Ca2+ were performed on tubules dissected from the kidney of the same animal. Table 1 Effect of commonly used agonists of the Ca2+-sensing receptor (CaR) on the 340/380-nm excitation fluorescence emission ratio (R) in rabbit CTAL cells. Data are means ±SEM of the 340/380 nm excitation fluorescence emission ratio (R). R was measured under the following conditions, each performed on the same tubule: (1) control (C1) with 1 mmol l–1 Ca2+ in the perfusate and bath solution, (2) experimental conditions (E) after addition of the agent to the bath solution, (3) after return to control conditions (C2). (n Number of tubules) Compound

C1

E

C2

n

Mg2+ (5×10–3 mol l–1) 0.73±0.05 0.71±0.05 0.73±0.04 6 Neomycin (10–3 mol l–1) 0.78±0.04 0.72±0.03* 0.76±0.04 23 0.64±0.04 0.64±0.03 0.65±0.04 3 Gd3+ (5×10–4 mol l–1) *Significantly different from the pre-experimental control period (C1)

720

Fig. 6 Frequency distribution of the changes in the 340/380 excitation fluorescence emission ratio (R) in response to 5 mmol l–1 peritubular Ca2+ and 10–4 mol l–1 verapamil in the bath solution. The abscissa represents the difference between R measured in the presence of 5 mmol l–1 Ca2+ (R5Ca) and in the presence of 5 mmol l–1 Ca2+ plus verapamil (R5Ca+Vera). In five experiments, verapamil had no effect on the increase in R induced by 5 mmol l–1 Ca2+. In the other 22 experiments, the drug reduced the increase in R to varying degrees

To pharmacologically characterize the rabbit CaR, the effects of commonly used agonists of the mouse and rat CaR were tested. The results are summarized in Table 1. Addition of 5×10–3 mol l–1 Mg2+ or 5×10–4 mol l–1 Gd3+ to the bath solution induced no change in the resting [Ca2+]i. A significant decrease in R was observed in the presence of 10–3 mol l–1 neomycin. The pharmacological approach First, the effects of verapamil and nifedipine, two known inhibitors of voltage-operated Ca2+ channels, on the [Ca2+]i increase induced by [Ca2+]e variations were studied. The effects of a rather high verapamil concentration (10–4 mol l–1) on the [Ca2+]e-mediated [Ca2+ ]i increase could be tested, since this concentration does not affect the resting [Ca2+ ]i (see below). Although 10–4 mol l–1 verapamil depolarizes the basolateral membrane voltage by some 7 mV [10], this depolarizing effect is not caused by inhibition of the basolateral K+ or Cl– conductances, but probably by inhibition of the basolateral KCl cotransporter [10]. The effects of verapamil and nifedipine were tested on tubules that responded to increases in the [Ca2+]e. Figure 6 shows the frequency distribution of the inhibitory effect of 10–4 mol l–1 peritubular verapamil on the [Ca2+]i increase induced by elevating the [Ca2+]e from 1 to 5 mmol l–1. In a few tubules (18%) the [Ca2+]i response observed after [Ca2+]e elevation was insensitive to verapamil (n=5), whereas in the other tubules the drug inhibited the high-[Ca2+]-mediated increase in R to a variable degree (n=22). In these 22 tubules, verapamil led to a significant decrease in R from 1.12±0.06 to 0.81±0.04. Under post-experimental control conditions, R returned to a value of 0.69±0.03. When its effects on the resting [Ca2+]i were tested (R=0.58±0.03, n=10), verapamil did not modify R (R=0.55±0.03, n=10). A marked inhibitory effect of verapamil on the plateau

Fig. 7A–D [Ca2+]i recordings showing the effect of verapamil (verap, 10–4 mol l–1, bath) on the [Ca2+]i response to 10–8 mol l–1 ADH and 5 mmol l–1 peritubular Ca2+. A The peritubular Ca2+ concentration was 1 mmol l–1; and B it was 5 mmol l–1. In both cases, verapamil was added to the bath solution after the stable plateau phase of the [Ca2+]i response to ADH was obtained. Note that verapamil almost completely inhibited the ADH-induced Ca2+ influx observed in the presence of either 1 or 5 mmol l–1 peritubular Ca2+. C ADH and verapamil were added at the same time to the bath solution and only a transient increase in the [Ca2+]i (peak) was observed. D Ca2+ (5 mmol l–1) was tested alone first, and a [Ca2+]i increase characterized by plateau kinetics was observed. After a control period, verapamil and 5 mmol l–1 Ca2+ were administered simultaneously. The plateau phase of the [Ca2+]i response was very low and no peak was observed

phase of the [Ca2+]i response was also observed when ADH (10–8 mol l–1) was used to increase the [Ca2+]i, whether the bath solution contained 1 or 5 mmol l–1 Ca2+, as shown in Fig. 7A ([Ca2+]e: 1 mmol l–1) and Fig. 7B ([Ca2+]e: 5 mmol l–1). Under both conditions, verapamil almost completely abolished the plateau phase of the [Ca2+]i response to ADH: in the presence of 1 mmol l–1 Ca2+ and ADH, addition of verapamil resulted in a decrease in R from 0.71±0.02 to 0.60±0.01 (post-experimental R value: 0.61±0.02, n=5) and in the presence of ADH and 5 mmol l–1 Ca2+, verapamil decreased R from 0.86±0.02 to 0.68±0.02 (post-experimental R value: 0.63±0.03, n=6). When verapamil and ADH were added simultaneously to the bath solution, the transient increase in the [Ca2+]i (peak) was still observed (Fig. 7C), whereas when verapamil and 5 mmol l–1 Ca2+ were added simultaneously to the bath solution, no transient [Ca2+]i increase was observed (Fig. 7D). These two observations confirm that IP3-sensitive Ca2+ pools are not involved in the high-[Ca2+]e-mediated [Ca2+]i increase.

721 Table 2 Effect of different agents known to change the basolateral membrane potential in rabbit CTAL cells on the 340 nm/380-nm excitation fluorescence emission ratio (R). Data are means ±SEM of the 340/380 excitation fluorescence emission ratio (R). R was measured under the following conditions, each performed on the same tubule: (1) control (C1) with 1 mmol l–1 Ca2+ in the perfusate and bath solution, (2) experimental conditions (E) after addition of the agent to the bath solution, (3) after return to control conditions (C2). [n Number of tubules, NPPB 5-nitro-2-(3-phenylpropylamino)-benzoic acid]

Fig. 8 Example of a [Ca2+]i recording showing the effect of the Ca2+ channel opener Bay K 8644 (10–6 mol l–1, bath) on the [Ca2+]i increase induced by 5 mmol l–1 Ca2+. Bay K 8644 was applied in the presence of 1 mmol l–1 Ca2+. Three minutes after Bay K 8644 withdrawal, the effect of an increase in bath Ca2+ concentration from 1 to 5 mmol l–1 on the [Ca2+]i was tested. Note that, after application of Bay K 8644, the amplitude of the [Ca2+]i response to 5 mmol l–1 peritubular Ca2+ was markedly enhanced

The effects of nifedipine and Bay K 8644 were studied to examine whether dihydropyridine-sensitive Ca2+ channels might account for the [Ca2+]i increase observed in response to increases in [Ca2+]e. When nifedipine was added to the bath at a concentration of 10–5 mol l–1, it inhibited the [Ca2+]i response induced by 5 mmol l–1 Ca2+. R was significantly reduced from 1.09±0.14 to 0.94±0.13 and returned to a control value of 0.73±0.08 (n=5) under post-experimental conditions. The effect of nifedipine, however, was smaller than that of verapamil. Figure 8 shows that peritubular application of Bay K 8644 (10–6 mol l–1), L-type Ca2+ channel opener, only very slightly modified the resting [Ca2+]i. The tiny [Ca2+]i increase visible in Fig. 8 may be attributed to the fluorescence of the substance. Results obtained from eight tubules, however, showed that the effect on the resting [Ca2+]i was not significant (P<0.05). R was 0.46±0.03 under control conditions and 0.48±0.03 in the presence of Bay K 8644. Interestingly, the exposure to Bay K 8644 resulted in an enhanced responsiveness of the tubules to increased peritubular Ca2+ concentrations. Indeed, the increase in R by 5 mmol l–1 Ca2+ was 0.11±0.02 before and 0.28±0.05 (n=8) after exposure to Bay K 8644. The effect of Bay K 8644 was tested under two experimental conditions. In one experimental series, the substance was applied simultaneously with 5 mmol l–1 Ca2+, in the other series (see Fig. 8) it was applied in the presence of 1 mmol l–1 Ca2+ and withdrawn 2 or 3 min before application of 5 mmol l–1 Ca2+. The activation of the Ca2+ channel by Bay K 8644 was found, for unknown reasons, to be more efficient in the second experimental series. However, we may anticipate that Bay K 8644 competes with Ca2+ ions for its binding site on the Ca2+ channel, so that in the presence of 5 mmol l–1 Ca2+ Bay K 8644 does not efficiently activate the channel. The observation that the [Ca2+]i response to 5 mmol l–1 Ca2+ was markedly enhanced after withdrawal of Bay K 8644 suggests that the

Compound

C1

E

C2

n

Ba2+ (3 mmol l–1) Ouabain (10–4 mol l–1) KCl (25 mmol l–1) NPPB (10–5 mol l–1)

0.69±0.05 0.73±0.06 0.73±0.03 0.79±0.03

0.80±0.05* 0.66±0.07* 0.60±0.01* 0.79±0.01

0.80±0.05 0.66±0.07 0.72±0.03 0.76±0.03

5 5 6 5

*Significantly different from the pre-experimental control period (C1)

binding of the drug to the Ca2+ channel persists after its removal from the bath solution. The effect of SKF 96365, initially described as an inhibitor of receptor-operated Ca2+ channels [19] and subsequently as an inhibitor of store-operated and voltageoperated Ca2+ channels [7, 12], on the [Ca2+]e-induced [Ca2+]i increase was also tested. Addition of 10–4 mol l–1 SKF 96365 to the bath solution containing 5 mmol l–1 Ca2+ significantly decreased R from 0.83±0.07 to 0.63±0.03 (n=5). This inhibitory effect of SKF 96365 was fully reversible. The search for a voltage-dependent Ca2+ channel To test the hypothesis that external Ca2+ ions enter the rabbit CTAL cell compartment through voltage-sensitive Ca2+ channels in the basolateral membrane, the effects of various agents known to change the membrane potential in the rabbit CTAL [10] were studied. The results are summarized in Table 2. Ba2+ (3 mmol l–1), a blocker of epithelial K+ channels that depolarizes rabbit CTAL cells, induced a significant but very slight and irreversible increase in R. Peritubular application of the (Na+K+)-ATPase inhibitor ouabain (10–4 mol l–1) or elevation of the K+ concentration (25 mmol l–1) in the bath solution, manoeuvres that depolarize the basolateral membrane [14–16], induced a significant decrease in R. On the other hand, addition to the bath solution of 5-nitro-2(3-phenylpropylamino)-benzoic acid (NPPB, 10–5 mol l–1), a potent Cl– channel blocker that hyperpolarizes the basolateral membrane of rabbit and mouse TAL cells [23], did not change R. Considering the contrasting effects on the [Ca2+ ]i of membrane-voltage-depolarizing substances (Ba2+, ouabain, 25 mmol l–1 KCl) and the hyperpolarizing agent (NPPB), it is suggested that there are no voltage-gated Ca2+ channels in the basolateral membrane of rabbit CTAL cells.

722

Discussion The present study was undertaken to characterize functionally in rabbit CTAL cells the recently cloned rabbit Ca2+-sensing receptor (RabCaR) and the cellular components that account for the [Ca2+]i variations following increases in [Ca2+]e. It is shown that the RabCaR is expressed at the mRNA level in the rabbit CTAL, and that, at variance with the mouse and rat TAL [8, 9], this receptor is not functionally coupled to phospholipase C. The high-[Ca2+]e-mediated [Ca2+]i increase due to Ca2+ influx across the basolateral membrane of rabbit CTAL cells occurs via dihydropyridine- and verapamil-sensitive Ca2+ channels, which are not open under resting conditions. An increased [Ca2+]e did not augment the accumulation of IP3 in the rabbit CTAL, whereas in the same experimental series phosphatidylinositol breakdown was stimulated by ADH. The lack of effect of a high [Ca2+]e on IP3- mediated Ca2+ release from intracellular stores was confirmed by the observation that, in the presence of verapamil, the transient [Ca2+]i increase (peak) was still observed in the presence of ADH but not in a high [Ca2+]e. It was surprising that phospholipase C was not activated by the high [Ca2+]e, since the recently cloned RabCaR [4] has been shown to be functionally expressed in the thick ascending limb of Henle’s loop and the inner medullary collecting duct [5]. When expressed in HEK 293 cells [4], the RabCaR increases the [Ca2+]i in response to a high [Ca2+]e, as demonstrated for the human CaR, and presents a similar pharmacology (activation by Gd3+ and Mg2+) as described for the G-protein-coupled bovine (BoPCaR) and rat (RaKCaR) Ca2+-sensing receptor. The presence of mRNA coding for the RabCaR in the rabbit CTAL cells was confirmed in the present study by RT-PCR. However, commonly used agonists of the CaR, i.e. Gd3+, Mg2+ and neomycin [3, 21, 22], did not increase the [Ca2+]i in rabbit CTAL cells, suggesting that phospholipase C is not activated by these compounds. From these results we anticipate that a CaR not coupled to activation of phospholipase C is functionally expressed in the rabbit TAL. A similar observation has been reported by Champigneulle et al. [5] for the rat cortical collecting duct. They showed that a CaR pharmacologically different from the cloned rat CaR (RaKCaR) led to a stable increase in the [Ca2+]i and concluded that a CaR not coupled to phospholipase C is responsible for the [Ca2+]i response observed in the rat cortical collecting duct. How may our observations be reconciled with those obtained in studies of transfected HEK 293 cells? First, in HEK 293 cells, the phosphoinositide metabolism was not measured; therefore, it is uncertain whether the [Ca2+]i increase in these cells in response to a high [Ca2+]e results from phospholipase C activation. Second, the different pharmacological profile of the CaR when expressed in rabbit CTAL cells and when transfected into HEK293 cells suggests that either, at variance with HEK 293 cells, the Ca2+ channels in the rabbit CTAL (opened-up via activation of the CaR by neomycin or

Gd3+) are also inhibited by these compounds, or that the coupling of the RabCaR to a specific transduction pathway in the native tissue is different from that in transfected cells. Therefore, it is attractive to postulate that the CaR is coupled to different transduction mechanisms, depending on the cell or the species considered. Alternatively, sequence differences within several regions of extra- and intracellular domains of the RabCaR, which are conserved in the other mammalian homologous of the receptor studied so far, could affect the coupling of the CaR to its intracellular transduction pathway. Since the [Ca2+]e-induced [Ca2+]i increase in the rabbit CTAL cannot be explained simply by the classic way, i.e. activation of a CaR coupled to phospholipase C, we investigated the mechanisms responsible for the [Ca2+]einduced [Ca2+]i increase. The involvement of a Na+/Ca2+ exchanger was not studied since its presence in the rabbit CTAL was excluded by studies of Hanaoka et al. [17]. The presence of voltage-gated Ca2+ channels also seemed very unlikely, since substances known to influence the membrane potential in the rabbit CTAL [14–16, 23] elicit only very limited effects on the [Ca2+]i. Ca2+ channels in the basolateral membrane of rabbit CTAL cells are also involved in the ADH-induced [Ca2+]i response. The peak–plateau kinetics of that response ([20] and present study) are thought to be due to IP3 formation, following phospholipase C activation by the V1a receptor. Indeed, our results confirm that ADH increases phosphoinositide metabolism. The Ca2+ channels involved in the plateau phase of the [Ca2+]i response have been found in the present study to be sensitive to verapamil. Whether these channels are also responsible for the high-[Ca2+]e-induced Ca2+ influx remains to be established. However, there is evidence that the same type of Ca2+ channel is responsible for the high-[Ca2+]eand the ADH-stimulated Ca2+ entry in the rabbit CTAL, since: (1) the plateau phase induced by 5 mmol l–1 peritubular Ca2+ alone was the same as that induced by the simultaneous presence of 5 mmol l–1 Ca2+ and ADH; (2) the plateau phases of both responses were inhibited by verapamil; and (3) the animals that responded to 5 mmol l–1 peritubular Ca2+ generally also responded to ADH. Can the high-[Ca2+]e-induced [Ca2+]i increase be explained simply by an increased driving force for Ca2+ influx when the bath Ca2+ concentration is elevated from 1 to 5 mmol l–1? From the calculated electrochemical driving forces for Ca2+ influx this hypothesis seems rather unlikely. In fact, if basal [Ca2+]i is estimated to be 100 nmol l–1 and the basolateral membrane potential –80 mV, then the electrochemical driving force for Ca2+ influx would amount to 200 mV at a bath concentration of 1 mmol l–1 Ca2+, and to 221 mV at a bath Ca2+ concentration of 5 mmol l–1. This represents a rather small change (about 10%) in the electrochemical driving forces. Furthermore, Ba2+, a high KCl concentration (25 mmol l–1), ouabain and NPPB, which lead, via large changes of the membrane voltage, to larger modifications of the electrochemical driving forces for Ca2+ influx, had only very limited effects on Ca2+ movement through the basolateral cell membrane. The experiment with the Ca2+ channel

723

opener Bay K 8644 also argues against simple Ca2+ influx being evoked by changes in the electrochemical driving force. In the presence of 1 mmol l–1 Ca2+, Bay K 8644 led to only a very small [Ca2+]i increase, indicating that even when the Ca2+ channel is activated by Bay K 8644, 1 mmol/l extracellular Ca2+ is not sufficient to account for a larger [Ca2+]i increase. The presence of a membrane Ca2+ channel gated by extracellular Ca2+ is supported by our finding that Bay K 8644 evoked a considerable [Ca2+]i increase only in the presence of a high [Ca2+]e. Therefore, these last observations favour the presence of an extracellular Ca2+ sensor, which accounts for the high-[Ca2+]e-induced [Ca2+]i increase. The localization of the recently cloned RabCaR in the TAL, together with our finding that the dose/response curve for the high-[Ca2+]e-induced [Ca2+]i increase in the rabbit CTAL is very similar to those reported for the CaR of the bovine parathyroid gland (BoPCaR) and the mouse TAL (RaKCaR) [3, 21], suggests that the Ca2+-sensing mechanism in the rabbit TAL could be the RabCaR. In conclusion, our study demonstrates that in the rabbit CTAL the recently cloned and expressed RabCaR is not coupled to phosphoinositide metabolism. Therefore, the [Ca2+]i increase induced by high [Ca2+]e cannot be explained by phospholipase C activation, as was shown for the rat and mouse CTAL, but it involves a dihydropyridine- and verapamil-sensitive Ca2+ channel. Whether the recently cloned RabCaR is involved needs further investigation.

Ca2+-sensing

9.

10.

11.

12. 13.

14. 15.

16. 17.

Acknowledgements This study was supported by the Commission des Communautés européennes, grant number ERBCHRXCT 940595.

18.

References

19.

1. Aida K, Koishi S, Tawata M, Onaya T (1995) Molecular cloning of a putative Ca2+-sensing receptor cDNA from human kidney. Biochem Biophys Res Commun 214:524–529 2. Brown EM (1991) Extracellular Ca2+-sensing, regulation of parathyroid cell function, and role of Ca2+ and other ions as extracellular (first) messengers. Physiol Rev 71:371–411 3. Brown EM, Gamba G, Riccardi D, Lombardi M, Butters RR, Kifor O, Sun A, Hediger MA, Lytton J, Hebert S (1993) Cloning and characterization of an extracellular Ca2+-sensing receptor from bovine parathyroid. Nature 366:575–580 4. Butters RR, Chattopadhyay N, Nielson P, Smith CP, Mithal A, Kifor O, Quinn S, Goldsmith P, Hurwitz S, Krapacho K, et al (1997) Cloning and characterization of a calcium-sensing receptor from the hypercalcemic New Zealand white rabbit reveals unaltered responsiveness to extracellular calcium. J Bone Miner Res 12:568–579 5. Champigneulle A, Siga A, Vassent G, Imbert-Teboul M (1997) Relationship between extra- and intracellular calcium in distal segments of the renal tubule. Role of the Ca2+ receptor RaKCaR. J Membr Biol 156:117–129 6. Chomczynski P, Sacchi N (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159 7. Clementi E, Meldolesi J (1996) Pharmacological and functional properties of voltage-independent Ca2+ channels. Cell Calcium 19:269–279 8. De Jesus Ferreira MC, Héliès-Toussaint C, Imbert-Teboul M, Bailly C, Verbavatz JM, Bellanger AC, Chabardès D (1998) Co-expression of a Ca2+-inhibitable adenylyl-cyclase and of a

20.

21. 22.

23.

24.

25.

26.

receptor in the cortical thick ascending limb cell of the rat kidney. J Biol Chem 273:15192–15202 Desfleurs E, Wittner M, Simeone S, Pajaud S, Moine G, Di Stefano A (1999) Calcium-sensing receptor: regulation of electrolyte transport in the thick ascending limb. Kidney Blood Press Res 21:401–412 Di Stefano A, Greger R, Desfleurs E, Rouffignac C de, Wittner M (1998) A Ba2+-insensitive K+ conductance in the basolateral membrane of rabbit cortical thick ascending limb cells. Cell Physiol Biochem 8:89–105 Elalouf JM, Buhler JM, Tessiot C, Bellanger AC, Dublineau I, Rouffignac C de (1993) Predominent expression of beta 1-adrenergic receptor in the thick ascending limb of rat kidney. Absolute mRNA quantitation by reverse transcription and polymerase chain reaction. J Clin Invest 91:264–272 Favre CJ, Nüsse O, Lew PD, Krausse KH (1996) Store-operated Ca2+ influx: what is the message from the stores to the membrane. J Lab Clin Med 128:19–26 Grazzini E, Durroux T, Payet MD, Bilodeau L, Gallo-payet N, Guillon G (1996) Membrane-delimited G protein-mediated coupling between V1a vasopressin receptor and dihydropyridine binding sites in rat glomerulosa cells. Mol Pharmacol 50:1273–1283 Greger R (1981) Cation selectivity of the isolated perfused cortical thick ascending limb of Henle’s loop of rabbit kidney. Pflügers Arch 390:30–37 Greger R, Schlatter E (1983) Properties of the basolateral membrane of the cortical thick ascending limb of Henle’s loop of rabbit kidney. A model for secondary active chloride transport. Pflügers Arch 396:325–334 Greger R, Schlatter E (1983) Properties of the lumen membrane of the cortical thick ascending limb of Henle’s loop of rabbit kidney. Pflügers Arch 396:315–324 Hanaoka K, Sakai O, Imai M, Yoshitomi K (1993) Mechanisms of calcium transport across the basolateral membrane of the rabbit cortical thick ascending limb of Henle’s loop. Pflügers Arch 422:339–346 Meneton P, Imbert-Teboul M, Bloch-Faure M, Rajerison RM (1996) Cholinergic agonists increase phosphoinositide metabolism and cell calcium in isolated rat renal proximal tubule. Am J Physiol 271:F382–F390 Merrit JE, Armstrong WP, Benham CD, Hallam TJ, Jacob R, Jaxa-Chamiec A, Leigh BK, McCarthy S, Moores KE, Rink TJ (1990) SKF 96365, a novel inhibitor of receptor-mediated calcium entry. Biochem J 271:515–522 Nitschke R, Fröbe U, Greger R (1991) Antidiuretic hormone acts via V1 receptors on intracellular calcium in the isolated perfused rabbit cortical thick ascending limb. Pflügers Arch 417:622–632 Paulais M, Baudouin-Legros M, Teulon J (1996) Functional evidence for a Ca2+/polyvalent cation sensor in the mouse thick ascending limb. Am J Physiol 271:F1052–F1060 Riccardi D, Park J, Lee WS, Gamba G, Brown EM, Hebert SC (1995) Cloning and functional expression of rat kidney extracellular calcium/polyvalent cation-sensing receptor. Proc Natl Acad Sci USA 92:131–135 Wangemann P, Wittner M, Di Stefano A, Englert HC, Lang HJ, Schlatter E, Greger R (1986) Cl–-channel blockers in the thick ascending limb of the loop of Henle. Structure activity relationship. Pflügers Arch 407 [Suppl. 2]:S128–S141 Wittner M, Di Stefano A, Wangemann P, Nitschke R, Greger R, Bailly C, Amiel C, Roinel N, Rouffignac C de (1988) Differential effects of ADH on sodium, chloride, potassium, calcium and magnesium transport in cortical and medullary thick ascending limbs of mouse nephron. Pflügers Arch 412:516–523 Yamashita N, Hagiwara S (1990) Membrane depolarization and intracellular Ca2+ increase caused by high external Ca2+ in a rat calcitonin-secreting cell line. J Physiol (Lond) 431: 243–267 Zaidi M, Alam ASMT, Huang CLH, Pazianas M, Bax CMR, Bax BE, Moonga BS, Bevis PJR, Shankar VS (1993) Extracellular Ca2+-sensing by the osteoclast. Cell Calcium 14:271–277