Calcium Handling by the Mammalian Kidney

In mammals, overall Ca²⁺ homeostasis is maintained by the concerted action of the intestine, bone and kidney. Given a daily dietary calcium intake of 1000mg, net intestinal absorption amounts to about 200mg, with the remaining 800mg being excreted in the faeces. To maintain a net balance, the kidney must excrete approximately 200mg, amounting to less than 2% of the filtered calcium load (Costanzo and Windhager, 1992). Thus, it is evident that the kidney is a crucial organ for the regulation of the calcium balance. As a result, a considerable amount of renal research has focused on the handling of calcium. Our present knowledge of renal calcium handling is mainly based on micropuncture and in vitro microperfusion techniques with which the segmental localisation of Ca²⁺ transport, its hormonal dependence and the relationship between Na²⁺ and Ca²⁺ reabsorption have been investigated. Although these techniques have been extremely useful, their limitations precluded further advances. The development of new techniques, such as single-cell fluorescent video microscopy, molecular biological approaches, patch-clamp and cell culture applications, has greatly furthered our insight into the mechanisms and regulation of renal Ca²⁺ transport. This review focuses on the regulation of active transcellular Ca²⁺ reabsorption in the distal nephron. Within this framework, the general handling of Ca²⁺ by the mammalian kidney will be briefly summarized and subsequently the physiological relevance of recent studies on isolated perfused tubules and isolated and cultured renal cells will be discussed in detail.

Following its filtration across the glomerular capillaries, Ca²⁺ can be reabsorbed along the nephron by two potential pathways (Fig. 1). One is a paracellular and passive movement in response to an electrical or chemical gradient and the other is a transcellular active transport route, which consists of passive influx across the luminal membrane, diffusion through the cytosol and active extrusion across the opposing basolateral membrane (van Os, 1987). Although Ca²⁺ influx proceeds via a steep electrochemical gradient, the molecular nature of this Ca²⁺ entry pathway is still undetermined. Conceivably, a carrier-mediated process or Ca²⁺ channels may be present to allow permeation of Ca²⁺ through the bilayer forming the plasma membrane. Once in the cell, Ca²⁺ diffuses to the opposing basolateral membrane, while intracellular Ca²⁺ concentration ([Ca²⁺ ]_i) is maintained within a narrow range to allow normal cell functioning. This is achieved by controlling the compartmentalisation of Ca²⁺ in intracellular organelles and by the formation of complexes with vitamin-D₃-dependent Ca²⁺ -binding proteins, such as calbindin-D_28K. Theoretical analysis and experimental evidence have confirmed the need for facilitated diffusion of Ca²⁺ through the cytosol.

Presumably, calbindin-D_28K increases the diffusional flux of Ca²⁺ by increasing brush-border entry of Ca²⁺, through decreasing [Ca²⁺ ]_i near the brush border, through increasing cytosolic flow by acting as a diffusional carrier and through increasing efflux by feeding Ca²⁺ to the basolateral extrusion mechanisms (Feher et al. 1992). In this scheme, the diffusion of Ca²⁺ through the cytosol is the rate-limiting step in transcellular Ca²⁺ movement. In the basolateral membrane, two Ca²⁺ transporters, Ca²⁺ -ATPase and a Na²⁺ /Ca²⁺ exchange mechanism, have been shown to be involved in the efflux of Ca²⁺ from the cell.

Micropuncture studies have shown that approximately 55% of the filtered load of Ca²⁺ is reabsorbed along the proximal convoluted tubule and an additional 10% is reabsorbed in the proximal straight tubule (Suki, 1979; Suki and Rouse, 1992; Costanzo and Windhager, 1992). Passive driving forces are the major determinants of Ca²⁺ transport in these nephron segments. Hence, Ca²⁺ reabsorption is a consequence of Na²⁺ and fluid reabsorption and is therefore obligatorily coupled to Na²⁺ transport (Suki and Rouse, 1992). Controversy exists as to whether some residual Ca²⁺ transport occurs via an active, transcellular pathway (Ng et al. 1984; Suki and Rouse, 1992). Ca²⁺ reabsorption in the loop of Henle has been attributed to the thick ascending limb, since the thin limbs have been shown to be relatively impermeable to Ca²⁺ (Rouse et al. 1980). It is generally accepted that about 20% of filtered Ca²⁺ is reabsorbed in the thick ascending loop of Henle (Suki, 1979; Costanzo and Windhager, 1992). The lumen-positive transepithelial potential difference generated by Cl²⁺ reabsorption provides a driving force for passive Ca²⁺ reabsorption through the paracellular pathway. Other studies suggested that there may be an additional, possibly active, Ca²⁺ transport mechanism (Imai, 1978; Suki et al. 1980; Friedman, 1988). In the medullary segment, calcitriol enhances Ca²⁺ reabsorption, while in the cortical segment parathyroid hormone (PTH) exerts a stimulatory effect. In the distal convoluted and connecting tubule, Ca²⁺ reabsorption amounts to about 10% of the filtered load (Costanzo and Windhager, 1992). Both segments are a major regulatory site for Ca²⁺ excretion in the urine. Transport appears to be active, as it can proceed against an electrochemical gradient. Here, hormones such as PTH and calcitriol stimulate Ca²⁺ reabsorption (Costanzo and Windhager, 1992). In the distal convoluted tubule, Ca²⁺ reabsorption parallels Na⁺ reabsorption, but thiazide diuretics dissociate Na⁺ and Ca ²⁺ reabsorption by inhibiting Na⁺ transport and stimulating Ca²⁺ reabsorption (Costanzo and Windhager, 1978). Finally, the contribution of the collecting duct to overall Ca²⁺ reabsorption is small. Although little is known about Ca²⁺ transport in this segment, it is probably active (Sharegi and Stoner, 1981; Bourdeau and Hellstrom-Stein, 1982). The segmental Ca²⁺ reabsorption is summarized in Fig. 2.

The distal tubule is the main site of active transcellular Ca²⁺ transport and determines the fine tuning of Ca²⁺ excretion to the urine. Hitherto, it has been difficult to identify the segments or cell types responsible. This is due both to the heterogeneous nature of this part of the nephron, in which numerous cell types can be recognized, and to considerable species differences. Identification of the cell types responsible has been greatly expedited by immunohistological localisation of the calcium-binding protein, calbindin-D_28K which, as mentioned above, facilitates transcellular Ca²⁺ movement. In mammals, such as rat, rabbit, pig, monkey and human, calbindin-D_28K is present in all the cells of the distal convoluted tubule and in most cells of the connecting tubule (Taylor et al. 1982; Schreiner et al. 1983; Bindels et al. 1991b). In rabbit, cells of the cortical collecting tubule also contain calbindin-D_28K, although less frequently (Taylor et al. 1982; R. J. M. Bindels, unpublished observation); in the rat, the presence of calbindin-D_28K is restricted to the transition between the connecting tubule and the cortical collecting duct (Taylor et al. 1982; Bindels et al. 1991b) (Fig. 3). Apparently, the relative contribution of the individual distal segments to active Ca²⁺ reabsorption differs significantly between rat and rabbit. In rabbit, the connecting tubule seems to be the most important site of hormone-regulated active Ca²⁺ reabsorption, whereas in the rat the distal convoluted tubule is more dominant (Morel, 1981; Imai, 1989; Bourdeau and Lau, 1990). Furthermore, the distal nephron of the rat lacks clear-cut segmentation with sharp transitions (Morel, 1981). Conceivably, hormone responsiveness may be determined by this heterogeneity in cell and site.

Recently, several groups have presented detailed information on the mechanism and regulation of transcellular Ca²⁺ transport in the distal nephron. The involvement of all distal segments in calcium reabsorption is now well documented, although the cellular mechanisms remain to be clarified. Integration of this information into a single model of cellular Ca²⁺ transport is not possible, mainly because of the complexity of the various transport steps and incomplete information on the individual segments. Therefore, the characteristics of active Ca²⁺ transport in the distal segments, such as the thick ascending limb of Henle’s loop, the distal convoluted tubule, the connecting tubule and the cortical collecting duct, will be discussed individually. Finally, at the end of each paragraph a cellular model is presented summarizing the various Ca²⁺ transporters that appear to participate in the transcellular movement of Ca²⁺ in that particular segment.

A considerable fraction of filtered Ca²⁺ is reabsorbed in the thick ascending limb of Henle’s loop. Several groups have investigated the mechanisms of Ca²⁺ transport using micropuncture techniques and microperfused tubules in both rabbit and mouse, but the results are equivocal. Passive reabsorption is suggested by the transepithelial voltage-dependence and the furosemide-sensitivity of Ca²⁺ fluxes (Bourdeau and Burg, 1979; Wittner et al. 1993). Hormonal stimulation of Ca²⁺ transport occurs through hormone-mediated transepithelial voltage alterations (Wittner et al. 1993). Other studies suggest that there is an additional active Ca²⁺ reabsorption, which can be stimulated by PTH and calcitonin (Costanzo and Windhager, 1978; Suki et al. 1980; Friedman, 1988). It is conceivable that under resting conditions passive Ca²⁺ reabsorption prevails and that after hormonal stimulation an additional active cellular pathway is activated (Friedman and Gesek, 1993). This controversy has not been resolved, but the conspicuous absence of calbindin-D_28K from this nephron segment strongly suggests an insignificant contribution of active, transcellular Ca²⁺ transport to the overall Ca²⁺ reabsorption.

Examination of the cellular mechanism of Ca²⁺ transport in the cortical thick ascending limb of Henle’s loop has not provided additional evidence for the presence of active Ca²⁺ transport (Fig. 4). In cultured mouse cortical thick ascending limb of Henle’s loop and distal convoluted cells, Bacskai and Friedman (1990) demonstrated a PTH-activated dihydropyridine-sensitive Ca²⁺ entry pathway. Subsequent studies by the same authors, performed with immortalized cells of similar origin, extended their previous results. PTH primarily increased Cl⁻conductance and the subsequent membrane hyperpolarisation enhanced Ca²⁺ entry through putative Ca²⁺ channels (Gesek and Friedman, 1992). The physiological significance of these data with respect to Ca²⁺ reabsorption is not clear for two reasons. First, a mixture of distal nephron segments, i.e. the thick ascending limb of Henle’s loop and the distal convoluted tubules, was originally used and it is, therefore, difficult to assign these findings to a specific segment. Second, localisation of the Ca²⁺ entry pathway or correlation with transcellular Ca²⁺ fluxes was not possible, since the observations were not made on polarized confluent monolayers. Their findings could also be interpreted as a basolateral Ca²⁺ entry playing a role in signal transduction of PTH. In this respect, it is interesting that a pH-sensitive Ca²⁺ influx pathway in the basolateral membrane of the rabbit cortical thick ascending limb of Henle’s loop has recently been demonstrated (Hanaoka et al. 1993).

There is also no indication of what kind of basolateral Ca²⁺ extrusion mechanism is present in this segment. No functional evidence could be obtained for the presence of a basolateral Na⁺/Ca²⁺ exchanger in rabbit thick ascending limb of Henle’s loop (Hanoaka et al. 1993). Furthermore, immunohistological and molecular biological studies failed to demonstrate the presence of a Na⁺/Ca²⁺ exchanger in the thick ascending limb of Henle’s loop (Reilly et al. 1992; Reilly and Shugrue, 1992; Yu et al. 1992b). Finally, only a low activity of a Ca²⁺/Mg²⁺-ATPase was detected in this segment, albeit sufficient activity to account for active Ca²⁺ transport (Doucet and Katz, 1982).

In the distal convoluted tubule, Ca²⁺ reabsorption proceeds against both chemical and electrical gradients and is, therefore, by definition active in nature. The paracellular permeability for Ca²⁺ is very low (Costanzo and Windhager, 1992). Depending on the species, hormones such as PTH and calcitonin stimulate Ca²⁺ transport (Shareghi and Stoner, 1981; Imai, 1989; Shimizu et al. 1990). The presence of vitamin-D₃-dependent calbindin-D_28K suggests that 1,25-dihydroxyvitamin D₃ (1,25[OH]₂D₃) exerts a stimulatory action in this segment (Taylor et al. 1982; Bindels et al. 1991b), but at present there is no experimental evidence for such action. In general, only a few investigators have addressed the role of the distal convoluted tubule in the regulation of Ca²⁺ reabsorption and this reflects the technical difficulties involved in the identification, micropuncture and microperfusion of this nephron segment (Suki, 1979). Consequently, our knowledge of the mechanism of cellular Ca²⁺ transport in this segment is still incomplete (Fig. 5).

Recently, Poncet et al. (1992) characterised a Ca²⁺-permeable channel in the apical membrane of primary cultures of the rabbit distal convoluted tubule. This is the first study using patch-clamp analysis to identify a Ca²⁺ channel in the distal nephron. The apical Ca²⁺ channel had a conductance of approximately 8pS and was blocked by La³⁺ and by high concentrations of verapamil and nifedipine. This channel could well be the apical entry pathway in transepithelial calcium transport. Interestingly, Yu et al. (1992a) recently identified an mRNA encoding partly for funnel web spider toxin-sensitive P-type Ca²⁺ channel that was exclusively expressed in the distal convoluted tubule of the rat.

The presence of both a Ca²⁺-ATPase and a Na⁺/Ca²⁺ exchange contributing to Ca²⁺ efflux across the basolateral membrane has been well documented. Epitopes for a plasmalemma Ca²⁺-ATPase were exclusively present in the basolateral membrane of this segment in human and rat, as demonstrated with immunocytochemistry using a monoclonal antibody against the human erythrocyte Ca²⁺/Mg²⁺-ATPase (Borke et al. 1987, 1989). The immunoreactivity co-localised perfectly with the presence of calbindin-D_28K in the distal nephron of the rat, which suggests that, in addition to the distal convoluted tubule, the connecting tubule also expresses a Ca²⁺-ATPase immunologically similar to that of human erythrocyte membranes. This idea is confirmed by the presence of numerous intercalated cells in the immunopositive segment. Another approach was used by Magocsi et al. (1992), who studied the localisation of mRNAs coding for isoenzymes of rat plasma membrane Ca²⁺-ATPase using polymerase chain reaction (PCR) analysis. mRNA for one isoenzyme, rPMCA2, was detected in microdissected proximal tubules, cortical thick ascending limb of Henle’s loop, distal convoluted tubules and perhaps also in cortical collecting ducts. The remaining tubule segments were not examined. The presence of this isoenzyme in most cortical tubules suggests a house-keeping function for rPMCA2 rather than a specific role in active transcellular Ca²⁺ transport. The molecular cloning of the cardiac Na⁺/Ca²⁺ exchanger enabled other investigators to identify a similar antiporter in the mammalian kidney (Reilly and Shugrue, 1992; Yu et al. 1992b). PCR analysis performed on microdissected rat tubules revealed that the Na⁺/Ca²⁺ exchanger is primarily expressed in the distal convoluted tubule (Yu et al. 1992b). Confusingly, in rabbit kidney, specific staining for immunoreactive Na⁺/Ca²⁺ exchanger was found only along the basolateral membrane of the connecting tubules compared with the distal convoluted tubules in the rat (Reilly et al. 1992). Presumably, expression of the Na⁺/Ca²⁺ exchanger in the distal nephron varies between different species. Finally, the relative contributions of the Na⁺/Ca²⁺ exchanger and Ca²⁺-ATPase to active Ca²⁺ reabsorption remain to be determined.

This tubule segment, which connects the distal convoluted tubule with the initial portion of the cortical collecting duct, is a heterogeneous epithelium consisting of at least two different cell types: connecting tubule cells and intercalated cells (Madsen et al. 1988). Only the former contain the vitamin-D₃-dependent calbindin-D_28K (Bindels et al. 1991b). Ca²⁺ reabsorption in this segment has been extensively studied by a diversity of technical approaches applied to both isolated perfused tubules and cultured cells. As in the distal convoluted tubule, Ca²⁺ reabsorption is active because it proceeds against a steep electrochemical gradient. In isolated perfused connecting tubules, it has been demonstrated that PTH stimulates Ca²⁺ reabsorption through the activation of protein kinase A (Shareghi and Stoner, 1981; Imai, 1989; Shimizu et al. 1990). These findings have been confirmed in primary cultures of the cortical collecting system (Bindels et al. 1991a), as shown in Fig. 6. This culture system has been established in our laboratory from a mixture of cells from connecting tubules and cortical collecting ducts isolated by immunodissection from rabbit kidney. When cultured on permeable filters, the cells exhibited several functions of the original epithelium (Bindels et al. 1991a). Net basal Ca²⁺ reabsorption across the monolayers ranged between 15 and 20pmolmin⁻¹ mm⁻². This value agrees with published values obtained in isolated perfused tubules (Costanzo and Windhager, 1978; Shimizu et al. 1990). In addition to protein kinase A, activation of protein kinase C plays a regulatory role in transcellular Ca²⁺ reabsorption. The phorbol ester 12-o-tetradecanoylphorbol 13-acetate (TPA) inhibited Ca²⁺ reabsorption and increased membrane-associated protein kinase C activity in primary cultures of rabbit cortical collecting system (Bindels et al. 1993) (Fig. 7). Interestingly, several kidney hormones, such as PTH, bradykinin, calcitriol and vasopressin, have been shown to activate protein-kinase-C-dependent signalling pathways (Shayman and Morrison, 1985; Hruska et al. 1987; Burnatowska-Hledin and Spielman, 1989). Therefore, further studies are needed to delineate the importance of this regulatory pathway. The final target for protein kinase C action has not yet been identified, but it could well be the apical Ca²⁺ channel, the basolateral Ca²⁺-ATPase or the Na⁺/Ca²⁺ exchanger. The phorbol ester treatment also abolished the transepithelial potential and doubled the transepithelial resistance because it also inhibited transepithelial Na⁺ transport. Therefore, possible effects of protein kinase C activation on junctional permeability can be excluded. In the same primary culture system, we have demonstrated for the first time that 1,25[OH]₂D₃ stimulates transcellular Ca²⁺ reabsorption and concomitantly increases cytosolic calbindin-D_28K content (Bindels et al. 1991a) (Fig. 8). Maximal stimulation occurred between 48 and 72h after exposure to calcitriol, suggesting that the mechanism of stimulation is analogous to that of classical steroid hormones, which implies that 1,25[OH]₂D₃ binds first to a specific cytosolic receptor protein and then moves to the nucleus, where transcriptional events are initiated, leading to de novo synthesis of calbindin-D_28K. Calcium-binding proteins such as calbindin-D_28K have been shown to accelerate diffusion of Ca²⁺ through the cytosolic compartment (Feher, 1983; Feher et al. 1992). In addition, another genomic response of 1,25[OH]₂D₃ was recently demonstrated in the intestine (Wasserman et al. 1992; Cai et al. 1993), where it was shown that 1,25[OH]₂D₃ increased the net synthesis of the basolateral Ca²⁺ pump through an increase in Ca²⁺ pump gene expression (Cai et al. 1993). This intriguing finding has not yet been observed in the kidney.

At the cellular level, there are indications that several Ca²⁺ transporters participate in the transcellular movement of Ca²⁺, as summarized in Fig. 9. Until now the molecular nature of the luminal Ca²⁺ entry pathway in the connecting tubule has been unknown. However, some properties of Ca²⁺ influx have been revealed. In isolated perfused rabbit connecting tubules, an apical Ca²⁺ influx has been suggested by the dependency of [Ca²⁺]_i on luminal Ca²⁺ (Bourdeau and Lau, 1992). In the same preparation, PTH stimulates Ca²⁺ influx across the luminal membrane, probably via activation of a cyclic-AMP-dependent protein kinase (Bourdeau and Lau, 1989). Preliminary studies performed on primary cultures of rabbit cortical collecting system demonstrated that Ca²⁺ influx was strongly inhibited by luminal acidification (Bindels et al. 1992b). The mechanism of inhibition resided at the external surface of the apical membrane. Our results are, therefore, at variance with the operation of an apical Ca²⁺/H⁺ exchanger, as suggested by Bourdeau and Lau (1992). These authors noticed that acute changes in luminal and cellular pH slightly influenced [Ca²⁺]_i in isolated perfused connecting tubules. Their findings are also consistent with an effect of pH on other Ca²⁺ channels.

From our studies, the involvement of Ca²⁺ channels resembling the voltage-dependent types seems unlikely, since administration of Ca²⁺ entry blockers, such as methoxyverapamil, diltiazem, flunarizine and felodipine, had no effect on Ca²⁺ transport (Bindels et al. 1992a,b). Taken together, these data are consistent with a pH-sensitive Ca²⁺ channel regulated by a cyclic-AMP-dependent kinase, but they do not prove the existence of such a channel. Therefore, more work is needed to define the nature of this putative channel.

Our observation that the calcitriol-induced increase in transcellular Ca²⁺ transport is accompanied by a substantial increase in cytosolic calbindin-D_28K confirms the idea that calbindin-D_28K facilitates diffusion of Ca²⁺ through the cytosol (Bindels et al. 1991a). These findings are in line with studies performed in the mammalian intestine, where 1,25[OH]₂D₃ also stimulates Ca²⁺ absorption and increases the cellular content of calbindin-D_9K (Gross and Kumar, 1990).

Finally, Ca²⁺ must be extruded at the basolateral membrane against an electrochemical gradient. Two transport mechanisms exist in this membrane. One is a high-affinity Ca²⁺-ATPase, whose activity is highest in the connecting tubule (Doucet and Katz, 1982). The presence of the Na⁺/Ca²⁺ antiporter exclusively in the basolateral membrane of the rabbit connecting tubule has been demonstrated by cytochemistry (Reilly et al. 1992). Several investigators evaluated the role of this extrusion mechanism in the regulation of [Ca²⁺]_i by measuring [Ca²⁺]_i during selective manipulation of extracellular [Na⁺] (Taniguchi et al. 1989; Bourdeau and Lau, 1990). For instance, it has been shown that, in isolated perfused tubules, [Ca²⁺]_i rose upon removal of Na⁺ from the bathing medium and that this increase was virtually abolished in the absence of luminal Na⁺ or basolateral Ca²⁺ (Bourdeau and Lau, 1990). In addition to involvement in regulating [Ca²⁺]_i, this antiporter may participate in transcellular Ca²⁺ transport. Shimizu et al. (1990) demonstrated a decrease in Ca²⁺ reabsorption on removal of bath Na⁺ or following treatment with ouabain. In primary cultures of the cortical collecting system, a dose-dependent relationship between basolateral [Na⁺] and transcellular Ca²⁺ transport was found and a Hill coefficient of around 2 pointed to at least two binding sites for Na⁺ on the Na⁺/Ca²⁺ exchanger (Bindels et al. 1992a). Ouabain inhibited Ca²⁺ transport to the same extent as did Na⁺-free solutions. Depolarisation of the basolateral membrane, by raising basolateral [K⁺], also decreased transcellular Ca²⁺ transport, which suggests that the Ca²⁺ extrusion is electrogenic. Bourdeau and Lau (1990) calculated that the exchanger is quiescent under control physiological conditions because the net driving force for Na⁺ and Ca²⁺ movements are balanced. Consequently, Na⁺/Ca²⁺ exchange does not contribute to Ca²⁺ extrusion under physiological conditions. This is not in agreement with our observations, since reduction in basolateral [Na⁺] from 150 to 100mmol l⁻¹ did not reduce Ca²⁺ transport across the cultured collecting tubules. However, lowering [Na⁺] to 50mmol l⁻¹ significantly reduced Ca²⁺ transport. Taken together, these results suggest that connecting tubules exhibit a basolateral Na⁺/Ca²⁺ exchanger which contributes to transcellular Ca²⁺ transport and is involved in regulating [Ca²⁺]_i. In primary cultures of the rabbit cortical collecting system, a residual Ca²⁺ absorption of approximately 40% was always observed, regardless of the mode of inhibition. The partial inhibition caused by removal of basolateral Na⁺ and treatment with ouabain suggests that a Ca²⁺ extrusion mechanism other than Na⁺/Ca²⁺ exchange participates in Ca²⁺ reabsorption. This residual Ca²⁺ absorption probably reflects the activity of a basolateral Ca²⁺-ATPase. Unfortunately, no specific inhibitors are available to evaluate the role of this transporter directly.

The contribution of this terminal segment to overall Ca²⁺ reabsorption is small and the nature of the transport is controversial. Some studies reported an insignificant net Ca²⁺ flux that was voltage-dependent (Shareghi and Stoner, 1981; Bourdeau and Hellstrom-Stein, 1982). At variance with these results are those of Shimizu et al. (1990), who found a net Ca²⁺ reabsorption which, owing to the lumen-negative voltage in the cortical collecting duct, must proceed via a transcellular pathway. In line with this active component is the presence of calbindin-D_28K in the cortical portion of this segment in the rabbit. The cellular mechanisms of Ca²⁺ reabsorption await further investigations.

Renal cells involved in transcellular Ca²⁺ transport are continuously challenged by a substantial Ca²⁺ traffic through the cytosol while simultaneously maintaining low levels of [Ca²⁺]_i. Calcium-binding protein, which is exclusively present in these renal cells, could fulfil this paradoxical function. In cultured cells of the rabbit cortical collecting system, the average calbindin-D_28K concentration amounts to at least 70 μmol l⁻¹ (Bindels et al. 1991a). According to Feher (1983), the calbindin-D_28K concentration in the cytosol should reach values of 50 μmol l⁻¹ or higher to facilitate cytosolic Ca²⁺ diffusion significantly. Consequently, calbindin-D_28K concentration in the cultured cells is high enough to play a significant role in cytosolic Ca²⁺ movement. The cellular calbindin-D_28K content is regulated by 1,25[OH]₂D₃ (Kumar, 1990). It is, therefore, conceivable that the stimulation of Ca²⁺ reabsorption by this hormone, as demonstrated in the cultured cortical collecting system cells, is due to the increased calbindin-D_28K content. Calbindin-D_28K also acts as a Ca²⁺ buffer: during changes in transcellular Ca²⁺ transport rates, [Ca²⁺]_i remains buffered between 100 and 200nmoll⁻¹. Stimulated levels of Ca²⁺ transport were never accompanied by an elevation in [Ca²⁺]_i (R. J. M. Bindels, unpublished observation). However, in the same cell preparation, oscillations in [Ca²⁺]_i could be provoked irrespective of the calbindin-D_28K concentration (Koster et al. 1993). Similar Ca²⁺ oscillations were observed in intercalated cells, but here calbindin-D_28K is totally absent (Fig. 10). These findings suggest that, at physiological [Ca²⁺]_i, calbindin-D_28K is saturated with Ca²⁺, which is in line with an affinity of calbindin-D_28K for Ca²⁺ in the nanomolar range, as suggested by Feher (1983). Consequently, Ca²⁺ signalling regulating transport of other ions can occur during and independently of transcellular Ca²⁺ movement. For example, Na⁺ reabsorption and K⁺ secretion in the distal nephron are partly regulated by the activity of [Ca²⁺]_i-dependent Na⁺ and K⁺ channels, respectively (Hébert et al. 1991; Schlatter et al. 1993). Besides these initial data, further experiments are needed to unravel the complex relationship between transcellular Ca²⁺ transport and [Ca²⁺]_i homeostasis.

The author thanks Dr C. H. van Os for helpful criticism of the manuscript and for stimulating discussions, and expresses his gratitude to Drs J. Dempster, A. Hartog, H. Koster, P. Ramakers and P. Willems for stimulating support. Work in the author’s laboratory has been partly funded by the Dutch Kidney Foundation (grant nos 87.0716 and 91.1112) and by the Netherlands Organisation for Scientific Research (NWO grant no. 900.541.122).