|
|
||||||||
1School of Biomedical Sciences, Worsley Building, University of Leeds, Leeds, West Yorkshire LS2 9NQ; and 2Department of Biomedical Science, The University of Sheffield, Western Bank, Sheffield S10 2TN, United Kingdom
Submitted 18 July 2003 ; accepted in final form 9 February 2004
| ABSTRACT |
|---|
|
|
|---|

m, the complex 1 inhibitor rotenone and the protonophore FCCP. Ionomycin was able to mobilize store Ca2+ following exposure to IP3. These results suggest that the endoplasmic reticulum is a dominant Ca2+ store in the frog EDT. A second pool, sensitive to ionomycin but not IP3, may overlap with the IP3-sensitve pool. The data also rule out any contribution by mitochondria to EDT Ca2+ cycling. pump-leak coupling; permeability
NaCl reabsorption is energized by the basolateral Na+-K+-ATPase, which maintains a low intracellular Na+ concentration ([Na+]) and establishes a favorable chemical gradient for Na+ uptake from the luminal fluid. The movement of Na+ across the apical membrane is coupled to that of Cl and K+ via an electroneutral cotransporter. Working in concert, these two mechanisms complete the transcellular transport of Na+. The basolateral passage of Cl is coupled to that of K+ via either cotransport or parallel movement through channels. The K+ that exits the cell via the basolateral membrane is immediately available for reuptake via the Na+-K+-ATPase. In this manner, K+ recycles across the basolateral membrane and acts as a substrate supporting Na+ efflux from the cell. Peculiar to this nephron segment, K+ also recycles across the apical membrane. After its absorption from the lumen, K+ reenters the tubule fluid via an apical channel, where it is once again taken up by the electroneutral NKCC1 cotransporter (17). This apical recycling is mandatory for the continued reabsorption of NaCl, because the luminal delivery via the glomerular filtrate of K+ to the diluting segment is not of itself sufficient to sustain NaCl reabsorption; mutations of the apical K+ channels in humans lead to the salt-wasting symptoms of Bartter's syndrome (29). Thus the uptake of NaCl is determined by the availability of luminal K+ (16) and is obligatorily coupled to its movement on the cotransporter. Furthermore, the secretory flux of K+ via the apical channels is the rate-limiting step in NaCl absorption and is the principal regulator of salt absorption in this segment (16, 17).
Previous work demonstrated an indirect link between the activity of the apical K+ channels and cytosolic Ca2+ concentration ([Ca2+]); the activity of the apical K+ channels is directly regulated by the intracellular pH, where raising pH increases channel activity (23). Intracellular pH is itself principally determined by the activity of the Na+/H+ exchangers located on the basolateral membrane (8). Inhibition of the apical cotransporters by the loop-acting diuretic furosemide results in a rise in cytosolic [Ca2+] and is associated with increased activity of the apical K+ channels. Furosemide inhibits Cl uptake on the apical cotransporter and results in a rapid fall in intracellular [Cl]. This fall in intracellular [Cl] results in Ca2+ release from intracellular stores, thereby elevating cytosolic [Ca2+] (7). The increase in intracellular [Ca2+] activates the basolateral Na+/H+ exchangers, via calmodulin, and the consequent intracellular alkalinization directly upregulates apical K+ channel activity, increasing the availability of luminal K+ (9). This complex series of regulatory steps can be seen as a mechanism to maintain Na+ absorption in the face of reduced Na+ delivery to the diluting segment. Therefore, a rise in cytosolic Ca2+ is central to membrane cross talk, pump-leak coupling, and, therefore, the regulation of salt absorption in this nephron region.
The principal question remaining from the experiments described above is the source of intracellular Ca2+ involved in this feedback response. Previous work suggested that this was the endoplasmic reticulum (ER), because depletion of the intracellular Ca2+ stores with the SERCA-specific inhibitor 2,5-di(tert-butyl) hydroquinone (TBQ) ablated the furosemide-induced rise in intracellular Ca2+ (9). The purpose of the experiments described in this paper was to seek functional evidence for the identity of intracellular Ca2+ storage pools in the renal diluting segment, which may be involved in pump-leak coupling. In this manuscript, we describe experiments designed to elucidate functional calcium storage pools within the amphibian diluting segment. To our knowledge, this is the first time that such experiments have been conducted in native renal tissue. The results show that the predominant calcium pool is the ER and further show modulation of ER calcium content with physiological regulators. These studies provide strong corroborative evidence in support of our earlier findings concerning the ER as the source of calcium in pump-leak coupling in the amphibian EDT.
| METHODS |
|---|
|
|
|---|
Frogs (Rana temporaria) of either sex were kept in tap water at 4°C. Animals were stunned by concussion, and the brain and spinal cord were destroyed by pithing in accordance with the UK legislature. The kidneys were removed, cut into 1-mm sections, and stored in ice-cold Leiboitz incubation medium (pH 7.4, 204 mosmol/kgH2O). Individual EDT segments were then manually dissected in ice-cold amphibian Ringer solution (for composition see below) and incubated for an hour in amphibian Ringer solution containing 14 µM mag-fura 2-AM at room temperature. Tubules were transferred to a perfusion chamber mounted on the stage of a Nikon Diaphot inverted microscope and held at each end with a pair of microperfusion pipettes (9). Tubule segments were unperfused. Figure 1A illustrates a mag-fura 2-loaded EDT segment before permeabilization. Permeabilization of the basolateral membrane with the detergent saponin (15) (25 ng/ml) selectively disrupts the basolateral cell membrane and permits washout of the probe in the cytosol. Other detergents used in similar studies are also accepted to preferentially permeabilize the cell membrane, which means that internal membranes are left intact (5, 15, 19, 33). This permeabilization procedure was monitored by following the fluorescence signal online and terminated by washing off the saponin when the 350-nm fluorescence signal had fallen to
30% of its maximum, indicating loss of cytosolic probe into the bath solution. By the end of the permeabilization period, the signal was
1520% of the unpermeabilized value and was
50100 times background (Fig. 1B). In this way, the mag-fura present in the cytosol was removed and access to the intracellular Ca2+ stores could be gained via the bath solution; 0.5 mM Mn2+ was added to the bath solution during permeabilization to quench any cytosolic probe that was not washed out by the permeabilization process, such that changes in fluorescence from dye trapped within intracellular compartments could be followed.
|
Amphibian Ringer solution was used as dissection medium and for incubation of tubule segments in fluorescent probe and had the following composition (in mM): 97 NaCl, 3 KCl, 2 CaCl2, 1 MgCl2, and 10 HEPES, titrated to pH 7.4 with NaOH. The "intracellular" solution had the following composition (in mM): 88 K-gluconate, 12 NaCl, 2 MgSO4, 1 free [Mg2+], 4 Ca(NO3)2, 200 nM free [Ca2+], 10 EGTA, and 10 HEPES, titrated to pH 7 with KOH/glucuronic acid lactone as appropriate.
Where required, additions to the intracellular solution were made from the following stock solutions (unless otherwise stated, all chemicals were obtained from Sigma, Poole, UK, and dissolved in water): 1 mM or 1 M K-ATP, 0.5 M K-ADP (FLUKA), 1 M KH2PO4, 0.5 M succinic acid, 1 M Ca(NO3)2, 50 mg/ml saponin, 100 mM MnCl2, 10 mM rotenone in DMSO, 10 mM FCCP in DMSO, 10 mM TBQ in DMSO, 10 mM IP3, 100 mg/ml heparin, 10 mM ionomycin in DMSO, and 5 mg/ml oligomycin in DMSO (mix of A, B, and C).
Fluorescence Microscopy
The compartmentalization within intracellular organelles of the low-affinity calcium-sensitive probe mag-fura 2 (Molecular Probes, Leiden, The Netherlands) has been exploited here to monitor changes in intrastore Ca2+ following permeabilization of the basolateral membrane. Although the fluorescent probe chosen is also sensitive to Mg2+, it is of much lower affinity than that for Ca2+ and has been shown to be insensitive to Mg2+ in physiological type intracellular buffers (20, 33). The basic recording system (Newcastle Photometrics) has been described in detail previously (9). Briefly, mag-fura 2 was alternately excited at 350 and 380 nm for 500 ms, and emitted light was collected at 520 nm by a photomultiplier tube and digitized before being displayed and stored on an IBM-compatible computer.
Data Analysis and Statistics
Data are presented as continuous experimental recordings with time on the x-axis and 350/380 ratio on the y-axis, or as mean steady-state ratio values ± SE. No attempt has been made to calibrate the intrastore signals because a mixed population of Ca2+ stores invalidates such an approach (21). On the other hand, it was necessary to ascertain that the probe was sensitive to changes in Ca2+ following permeabilization, which was achieved by exposure to a Ca2+ ionophore and different levels of Ca2+: after incubation with 5 µM ionomycin, the 350/380 fluorescence ratio was 0.26 ± 0.003 in Ca2+-free and 1.07 ± 0.01 (n = 3) in a 5 mM Ca2+ solution, respectively. All starting ratios were above this minimum and all peak fluorescence ratios fell below the maximum. Additionally, changes in fluorescence while adding intracellular regulators occurred rapidly, suggesting that the probe was effectively reporting changes in Ca2+, even with very low levels of substrates (e.g., see ATP and TBQ dose reponses in RESULTS), without any appreciable delay other than that caused by the lag in solution exchange.
Statistical analysis was carried out in Excel (Microsoft) using paired or unpaired t-tests as appropriate. Where appropriate, ANOVA analysis was peformed with Minitab (Mintab, State College). Significance was assumed at the 5% level. Sigma Stat (Jandel Scientific) was used for nonlinear curve fitting in the determination of Kd values.
| RESULTS |
|---|
|
|
|---|
ATP and Ca2+. Addition of ATP to the bath solution in the presence of 200 nM Ca2+ promoted a dose-dependent increase in the 350/380 ratio, consistent with the movement of Ca2+ into the internal store (Fig. 2). The mean data from these experiments are summarized in the inset and are described well by saturation kinetics with a Kd of 2.6 ± 0.6 µM (n = 5, P < 0.05). The degree of store filling also depended on the Ca2+ concentration of the bath fluid: in the presence of a saturating concentration of ATP (0.1 mM), incremental increases in the 350/380 ratio were seen on the addition of Ca2+ to the bathing solution over the range of 10 nM to 10 µM (data not shown) and reflect the ability of the store to accumulate Ca2+ over a wide concentration range. These results suggest that Ca2+ is moved against its concentration gradient at the expense of ATP and that the transport mechanisms operate over a wide range of intracellular Ca2+ concentrations.
|
|
Passive Ca2+ leak. In the steady state, the store Ca2+ content reflects equal rates of influx and efflux, a change in either of these processes will alter the store Ca2+ equilibrium. In an attempt to determine the pathway of the passive leak, the effect of heparin, an IP3 receptor antagonist (14), on unstimulated store Ca2+ release was examined. Permeabilized tubules were exposed to 0.1 mM ATP to fill the internal store, and ATP was subsequently removed from the bath solution to reveal the passive loss of store Ca2+ (Fig. 4A). Calcium leak from the store followed an exponential time course that was unaffected by the addition of the IP3 receptor blocker heparin (100 µg/ml, a concentration that has been previously determined to cause complete reversal of IP3-mediated release, n = 6; Fig. 4B); time constant for 350/380 ratio decline following removal of ATP: control 1,750 ± 690 s, heparin 1,633 ± 529 s, n = 5, P > 0.05 (Fig. 4A). Loss of Ca2+ from the store was also apparent when Ca2+ was removed from the bath (data not shown). It appears therefore that the presence of both ATP and Ca2+ is required to maintain store Ca2+ load.
|
IP3-sensitive Ca2+ efflux. Store filling was promoted by the addition of 2 mM ATP to the bath, a concentration known to maximize the open probability of the IP3 receptor (3), before the addition of IP3 to the bath. The efflux of Ca2+ was clearly dependent on the concentration of IP3 (Fig. 5), with an apparent Kd of 1 µM (see inset) and showing saturation at 3 µM. Because virtually all Ca2+ was released with 3 µM IP3, this concentration was used in all subsequent experiments. Furthermore, the IP3 receptor antagonist heparin (100 µg/ml) abolished the effect of IP3 on store calcium (see Fig. 4B as previously discussed, n = 6). IP3 therefore appears to be an important regulator of store Ca2+ content.
|
Permeabilized tubules were exposed to ionomycin in the absence of ATP, i.e., without store loading. There was no effect on the 350/380 ratio following this intervention (350/380 ratio: control 0.27 ± 0.003; ionomycin 0.27 ± 0.003, n = 7, P > 0.05). As previously shown, after Ca2+ loading with ATP, IP3 (3 µM) promoted a rapid fall in store Ca2+ that represented 58 ± 5% of the total release. However, addition of ionomycin promoted a further reduction in the 350/380 ratio, consistent with a release from an IP3-insensitive store that was 42 ± 5% of the total release (n = 6; Fig. 6A). It is possible that these data represent Ca2+ release from a single compartment that occurs with two different time courses. However, we excluded this time-dependent effect on Ca2+ within the store because Fig. 6, A and B, shows that store Ca2+ decreases only after the addition of ionomycin, despite differences in the time course of initial application of the compound. Furthermore, Fig. 5 demonstrates that steady-state levels of store Ca2+ are only altered by additions of IP3 and not by any effect of time (total exposure time to IP3 >5 min).
|

m), and hence the driving force for Ca2+ entry, was maintained by operation of the respiratory chain. This was achieved by adding ATP (2 mM) to the bath alongside "respiratory substrates" that included ADP (1 mM), KH2PO4 (1 mM phosphate), and a Krebs or TCA cycle intermediate, succinate (substrates of glycolysis would have been ineffective in this preparation because glycolysis occurs in the cytosol, and we were unable to rely on the integrity of the constituent enzymes in the permeabilized preparation, which may have been washed away or otherwise compromised). We relied on atmospheric gases, dissolved in the bath solution, to supply the preparation with oxygen. Figure 7A shows that such a cocktail of substrates promoted an increase in the 340/380 ratio, consistent with Ca2+ uptake, from 0.34 ± 0.02 to 0.38 ± 0.01 (n = 7). This increase in Ca2+ was unaffected by the addition of the electron transport chain complex 1 inhibitor rotenone (1 µM) (350/380 ratio: substrates 0.38 ± 0.009; rotenone 0.38 ± 0.001, n = 7, P > 0.05) but was reversed by the addition of IP3 (3 µM), suggesting that Ca2+ was entering the ER/IP3-sensitive store. Similarly, no further increase in store Ca2+ was observed when permeabilized tubules were exposed to respiratory substrates after Ca2+ accumulation had been induced with ATP (2 mM) (350/380 ratio: ATP 0.37 ± 0.01; ATP and substrates 0.38 ± 0.01, n = 6, P > 0.05; Fig. 7B), presumably, again, into the ER/IP3-sensitive store. Consistent with this, FCCP (5 µM) was without effect on accumulated Ca2+ (350/380 ratio: ATP 0.37 ± 0.01; FCCP 0.38 ± 0.01; analysis by ANOVA with post hoc Dunnet's comparison), whereas IP3 was able to promote Ca2+ mobilization.
|
| DISCUSSION |
|---|
|
|
|---|
The ER appears to be a major Ca2+ store within the cells of the diluting segment because Ca2+ accumulation is sensitive to the SERCA-specific blocker TBQ; inhibition of SERCA activity with this compound prevents the global, furosemide-induced increase in intracellular Ca2+ (9). Furthermore, the application of TBQ alone, before store loading, never resulted in a decline in the 350/380 ratio (an effect indicative of Ca2+ loss via the leak pathway), which is consistent with the idea that there is little SERCA activity and thus ER loading following time spent in an ATP-free environment during dye loading. The data display an apparent Ki of 2 µM toward TBQ. This is in good agreement with inhibitory assays on rat liver microsomes (26) that demonstrated a Ki of 1 µM and a maximal inhibitory effect at 10 µM. The depletion of accumulated Ca2+ in the absence of ATP (or Ca2+) is also consistent with the notion that the continued presence of both these substrates is required to maintain a constant luminal Ca2+ concentration and thus the integrity of the pump-leak coupling mechanism. The affinity of Ca2+ uptake toward ATP strongly suggests that ATP is unlikely to be rate limiting in maintaining store Ca2+ content, assuming an intracellular ATP concentration in excess of 0.1 µM (1). Furthermore, our calculated Kd for ATP with respect to the uptake process of 2.6 µM is several orders of magnitude lower than other estimates using a similar technique in gastric epithelial cells (
1.5 mM, see Ref. 19). However, both these estimates are considerably higher than values obtained using COS cell microsomes whereby the K1/2 for the formation of a phosphorylated intemediate for SERCA 2b and SERCA3 (the most likely nonmuscle isoforms) are 20 and 50 nM, respectively (24). These differences are probably due to the fact that the fluorescence assay represents a global average of all ATP-dependent uptake processes. Similarly, data where Ca2+ transport varied with bath [Ca2+] could be fitted using a Kd of 230 nm. This value is similar to Ca2+ transport values published for SERCA2b, which elicited a Kd of 270 nM (24).
Release of Ca2+ from the ER appears to be dominated by the IP3 receptor; there was no effect of the endogenous ryanodine receptor regulator cADPR. This is in agreement with our earlier work, in which addition of ryanodine directly to the bathing medium was without effect on store Ca2+ (7). Therefore, it is unlikely that there is a mixed expression of intracellular Ca2+ release channels in the EDT region. In vivo, it is therefore conceivable that release from the ER may well be mediated by IP3. In human fibroblasts, it has been shown that a reduction in extracellular Na+ promotes the formation of inositol polyphosphates and a release of Ca2+ (30). Similarly furosemide, which causes an abrupt fall in intracellular Na+ in the EDT (27), also results in the release of store Ca2+, raising the possibility that this mechanism is mediated by IP3.
Additional Ca2+ release following store loading with ATP was induced by the addition of ionomycin to the bath following store depletion with IP3. There are several possibilities to explain this observation. 1) This represents anatomic homogeneity but functional heterogeneity (i.e., not all the Ca2+ within the store is released) within the IP3-sensitive pool. This may explain the ATP-dependent uptake but the lack of effect of ionomycin in the absence of store loading. An example of such a store would be the nuclear envelope, which is not only continuous with the ER (13) but has been shown to have functional Ca2+ uptake (13) and release pathways (12, 22, 31). 2) It is also conceivable that this result represents a differential expression of IP3 receptors to the ER membrane and as such Ca2+ release in the presence of IP3 occurs from the part of the store with the highest level of receptor expression. We have been unable to image the patterns of release and are therefore unable to resolve this possibility, but it is tempting to speculate that focal Ca2+ release could target the Ca2+ signal to discrete calmodulin-rich areas to accomplish activation of the sodium hydrogen exchanger. 3) This observation may represent the activity of a spatially inhomogenous store such as mitochondria (19) or another endomembranous compartment, e.g., lysosomes (18). Under the current conditions, the fluorescence signal is unaffected by addition of the protonphore FCCP, the actions of which are not confined to mitochondria, but to all compartments across which a proton gradient is maintained (see Fig. 6B and also Ref. 25). This has two implications. It suggests 1) that lysosomes do not make a significant contribution to Ca2+ cycling in the frog EDT and rules out any artefactual, pH-induced, fluorescence changes that result from the activity of proton-transporting ATPases on the lysosomal membrane and 2) that mitochondria do not participate directly in the uptake and release of Ca2+ under these conditions. This may not be too surprising given that the affinity of the Ca2+ uniporter is in excess of 1 µM (2), which is an order of magnitude higher than the Ca2+ levels used in the present study.
The data presented in this investigation rely on the compartmentalization into cellular pools of a low-affinity Ca2+-sensitive probe. In our system, mag-fura 2 appears to respond to large changes in Ca2+ (from 0 to 5 mM), equilibrated across the organelle membranes using ionomycin, which produces concomitantly large changes in the fluorescence ratio that would be well in excess of the levels of Ca2+ anticipated to reside within the stores (4). However, signal responses from this preparation may be limited because of 1) a low pumping activity with low levels of "resting" Ca2+, 2) washout of cellular regulators during permeabilization that modulate the activity of store-filling mechanisms, or 3) that "silent" compartments bias the fluorescence signal in favor of probe trapped in pools where Ca2+ resides below the detection limit of the probe (
5 µM) (21). This occurs because fluorescence ratio increases rely on quenching of fluorescent probe by Ca2+. Therefore, a probe trapped equally in two compartments will return an average fluorescence signal of the two pools; the compartment containing high Ca2+ will have a quenched signal compared with the compartment containing low Ca2+. Thus a proportionately larger part of the signal is composed of fluorescence from the low-Ca2+ compartment. If Ca2+ is not moved into this pool at all during uptake, then changes in the fluorescence signal will arise from movements into and out of just one store, which will suppress the magnitude of the observed changes (21). Due to these complications, the current data remain uncalibrated.
In summary, frog EDT cells maintain a constant internal store Ca2+ content at the expense of ATP. The availability of ATP is unlikely to be rate limiting to Ca2+ accumulation, given the high affinity of the Ca2+ uptake mechanism relative to the expected intracellular concentration of ATP. Ca2+ is released via the IP3 receptor, and not the ryanodine receptor, whereas the sensitivity toward the SERCA-specific inhibitor TBQ and the endogenous releasing agent IP3 strongly implicates the ER as the major site of Ca2+ storage within the EDT.
| GRANTS |
|---|
|
|
|---|
| FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
B. M. Walsh, H. B. Naik, J. M. Dubach, M. Beshire, A. M. Wieland, and D. I. Soybel Thiol-oxidant monochloramine mobilizes intracellular Ca2+ in parietal cells of rabbit gastric glands Am J Physiol Cell Physiol, November 1, 2007; 293(5): C1687 - C1697. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |