Expression of polycystin-1 enhances endoplasmic reticulum calcium uptake and decreases capacitative calcium entry in ATP-stimulated MDCK cel
http://www.100md.com
《美国生理学杂志》
Divisions of Renal Medicine and Cardiology, Johns Hopkins Bayview Medical Center, and The Johns Hopkins School of Medicine, Baltimore, Maryland
Department of Biological Research and Technology, Dulbecco Telethon Institute, San Raffaele Scientific Institute, Milan, Italy
ABSTRACT
Autosomal dominant polycystic kidney disease (ADPKD) types 1 and 2 arise as a consequence of mutations in the PKD1 or PKD2 genes, encoding polycystins-1 and -2. Because loss of function of either of the polycystins leads to a very similar phenotype and the two proteins are known to interact, polycystins-1 and -2 are probably active in the same pathway. The way in which loss of either polycystin leads to the development of ADPKD remains to be established, but disturbances of cell calcium regulation are likely to play an important role. Here, we demonstrate that polycystin-1, heterologously expressed in Madin-Darby canine kidney cells, had a pronounced effect on intracellular calcium homeostasis. ATP-induced calcium responses in transfection control cells exhibited a double peak and relatively gradual return to baseline. By contrast, cells expressing heterologous polycystin-1 showed a brief, uniphasic peak and an accelerated rate of decay. Heterologously expressed polycystin-1 accelerated endoplasmic reticulum (ER) calcium reuptake and inhibited capacitative calcium entry; we found no effect of the protein on mitochondrial calcium buffering or plasma membrane calcium extrusion. We therefore propose that polycystin-1 accelerated the decay of the cell calcium response to ATP by upregulation of ER calcium reuptake and consequent minimization of the stimulus for capacitative calcium entry. It is possible that cellular dedifferentiation, fluid secretion, and proliferation might therefore arise in ADPKD as a consequence of disturbances in cytoplasmic and ER calcium homeostasis and aberrant capacitative calcium entry.
autosomal dominant polycystic kidney disease; purinergic; calcium signaling; stable transfection; Madin-Darby canine kidney cells
AUTOSOMAL DOMINANT POLYCYSTIC kidney disease (ADPKD) arises as a consequence of loss of function of either polycystin-1 or polycystin-2, the gene products of the PKD1 and PKD2 genes respectively (10a, 22). ADPKD type 2, arising from mutations in the PKD2 gene, typically progresses somewhat less rapidly than ADPKD type 1, but the latter accounts for the majority of affected individuals. The phenotypes associated with either genotype are otherwise very similar, indicating that the two proteins probably function in the same pathway. The disease, which affects 1 in 1,000 individuals, is characterized by the development of multitudes of renal cysts. As these cysts expand, the resultant compression and distortion of normal renal tissue leads to loss of renal function, which typically progresses to end-stage and dialysis dependency by mid- to late adulthood. Polycystin-1 is a very large transmembrane protein, which appears to be active in a number of diverse pathways. The isolated truncated cytoplasmic COOH terminus of polycystin-1 has shown activity in 1) wnt signal transduction pathways (18); 2) AP-1 transcription factor activation (2); 3) heterotrimeric G protein-coupled signal transduction (25); 4) regulation of calcium channel activity (33, 37); 5) in-gel morphogenesis (24); and 6) modulation of ATP-stimulated cell calcium and chloride secretory responses (16, 34). The full-length protein has been shown to 1) regulate the ion channel properties of polycystin-2 (14), 2) promote tubulogenesis and resistance to apoptosis (5), 3) activate the JAK-STAT pathway (3), and 4) regulate G protein-coupled signal transduction (10). Polycystin-1 has been localized to the plasma membrane (6), at the sites of adherens junctions (17) and focal adhesions (35), and also, more recently, in the cilia (39). Polycystin-2 is predominantly seen in the endoplasmic reticulum (ER) (19) but may also be expressed at the cell membrane (13) and apical cilia (39). The cell calcium response to cilial deflection appears to depend on expression of functional polycystin-1 and -2 (23). The observation that polycystin-2 is a nonspecific cation channel with predominant calcium conductance (19) raised the possibility that ADPKD might develop as a consequence of disturbed intracellular calcium homeostasis (30). However, the effects of the polycystins on cell calcium homeostasis, whether acting independently or as a functional complex, are currently incompletely understood. If dysregulation of cell calcium homeostasis is an important mechanism of disease, in view of the fact that ADPKD types 1 and 2 share an almost identical phenotype, it would be predicted that the two proteins should have concordant effects on some aspect of cell calcium regulation. However, from studies of the effects of the polycystins on cell calcium homeostasis it is not clear that the two proteins always have consistent functions in this respect. Although heterozygous loss of polycystin-2 has been linked to a fall in resting cytoplasmic calcium concentration in vascular smooth muscle cells (27), homozygous loss of polycystin-1 activity seemed not to alter resting calcium concentration in collecting duct cells (23). Overexpression of polycystin-2 appeared to increase the cell calcium response to vasopressin in LLC-PK cells (19); we are not aware of any previous studies of heterologously expressed full-length polycystin-1 on stimulated cell calcium responses, but loss of functional polycystin actually caused an increase in the thrombin-stimulated cell calcium response in collecting duct cells (23). That polycystin-1 and -2 are required for the cell calcium response to cilial deflection is perhaps the most convincing evidence for a functional concordance in the regulation of cell calcium homeostasis (23).
To achieve a more complete understanding of the role of the polycystins in integrated cell calcium homeostasis, it is clear that more work is required to examine the effects of heterologously expressed proteins and loss of polycystin function on resting and stimulated cell calcium responses in renal epithelia and other cell types.
In earlier studies, we found that expression of the isolated cytoplasmic COOH terminus of polycystin-1 resulted in the prolongation of ligand-activated cell calcium responses (16, 34). Others have presented data consistent with these findings (1). We found that this effect resulted in increased ATP-stimulated chloride secretion and proposed that, in promoting a prosecretory phenotype, the truncated polycystin-1 fragment was acting as a dominant negative with respect to endogenous polycystin-1. The possibility that the cytoplasmic COOH-terminal polycystin-1 truncation might act as a dominant negative has been raised by other studies (31, 38). Consequently, we predicted that full-length polycystin-1 should have the opposite effect to that of the truncation, to abbreviate the ATP-stimulated cell calcium response. In the findings of the current study, we show that this is indeed the case. Our experiments indicate that heterologously expressed full-length polycystin-1 abbreviated the cell calcium response to ATP through a combination of acceleration of ER calcium reuptake and attenuation of capacitative calcium entry.
EXPERIMENTAL PROCEDURES
Cell Culture Conditions
Stably transfected Madin-Darby canine kidney (MDCK)-derived cell lines expressing human polycystin-1 were developed as previously described (5). The original parental cells were MDCK type II. It should be noted that heterologous human polycystin-1 was not expressed at high levels in these cells, because the protein was only detectable after enrichment by immunoprecipitation (5). Native expression of polycystin-1 in these cell lines was only detectable by RT-PCR (data not shown). In the experiments described below, we studied two vector control (F6 and F8) and two polycystin-1-expressing cell lines (C8/68 and G7/36) between passages 4 and 16. Cells were propagated in DMEM/F-12 supplemented with 5% fetal calf serum under dual selection with G418 and Zeocin and were passaged weekly at 1 x 106 cells/25-cm2 flask. To prepare cells for calcium imaging, glass coverslips were coated with Matrigel (0.45 mg/ml) for at least 1 h at room temperature. Cells were sparsely seeded onto the coated glass coverslips and grown for 3 days in culture. Small clusters of cells were selected for imaging. These islands of cells were chosen to ensure that the same number and density of cells (25 cells/field) was studied in each experiment. The morphology of these groups was similar in all the cell lines studied.
Fluorescent Cell Calcium Measurements
The coverslips were placed in a perfusion chamber mounted on the stage of a Nikon Diaphot microscope fitted with dual photomultipliers and illuminated with a xenon light source passed through a monochromator and appropriate filters (Photon Technologies International). The cells were excited at 340 and 380 nm, and emission was measured at 510 nm. Cell fluorescence was measured in response to excitation for 100 ms once every second. Shutter opening was controlled using a Spectroscan device driven by Felix software. Imaging was performed with a long-working-distance Nikon Fluor 40 lens. All perfusion was performed at room temperature at 2 ml/min to ensure rapid turnover of the 0.25-ml bath volume.
Bath Calcium-Switch Repeat-Stimulation Studies
Cells were loaded in a 5 μM solution of the cell-permeant acetoxymethyl (AM) ester of the calcium indicator fura 2 (fura 2-AM) at room temperature for 45 min in DMEM/F-12. All fluorescent dyes used in these experiments were made up to a 1 mM stock solution in 10% Pluronic/DMSO. Following incubation, cells were allowed to recover for 30 min in DMEM/F-12. The incubation and recovery media both contained 5 mM probenecid to maximize dye loading through inhibition of dye extrusion. Cells were placed into the perfusion chamber in a zero-calcium EGTA buffer and stimulated with 10 μM ATP for a 30-s pulse in this buffer. Cells were then switched into a calcium-containing buffer for 10 min to refill calcium stores before being stimulated again with a second 30-s pulse of ATP in the presence of bath calcium. This sequence of studies was used to avoid the need for a prolonged zero-calcium washout of calcium-containing buffer.
Measurement of Intracellular Calcium Stores by Ionomycin
The magnitude of compartmentalized calcium content can be inferred from the peak in cytoplasmic calcium concentration in response to the release of this calcium by ionomycin (4). However, we were concerned that the peak might have been rendered very transient as a consequence of concurrent loss of calcium from the cell through the ionomycin-permeabilized plasma membrane. For this reason, we decided to use fura 5F in these experiments, because the rapid response of this indicator made it superior to fura 2 in the capture of short-lived calcium transients. Cells were loaded with 5 μM fura 5F-AM for 45 min at room temperature in the presence of 5 mM probenecid. After being mounted on the microscope stage, cells were perfused with a zero-calcium (200 μM BAPTA) buffer and treated with 10 μM ATP followed by 10 μM ionomycin at 100 or 460 s post-ATP, to release the total sequestered intracellular calcium stores at these time points. Ionomycin was added at 2 ml/min using a gravity-feed system through a separate perfusion pathway to prevent carryover contamination.
Manganese Quench Studies
Because of the duration of these studies and the presence of basal quenching, cells were maximally loaded with 10 μM fura 2-AM for 45 min at room temperature in the presence of 5 mM probenecid. The cells were perfused with nominally calcium-free extracellular buffer containing MnCl2 (which quenches the fura 2 signal on access to the cytoplasm) and treated with 10 μM ATP for 30 s. For these studies, cells were excited at 360 nm, the isosbestic point of fura 2, and the emission was recorded at 510 nm.
ER Calcium Imaging
Cells were loaded with 10 μM mag-fura 2-AM for 1 h 15 min at room temperature in the absence of probenecid. Cells were perfused with zero-calcium EGTA extracellular buffer and then permeabilized with saponin (20 μg/ml). Once the mag-fura 2 escaped from the cytoplasm (defined as a decline in the 340-nm signal to one-third of the starting value), the cells were switched into the intracellular buffer. Transfer to intracellular buffer resulted in an increase in ER calcium signal because of activation of the ER calcium ATPase, previously rendered inactive through ATP depletion in the permeabilization buffer (which contained no ATP). ER calcium release was then induced by treatment with a 30-s pulse of 10 μM inositol1,4,5-triphosphate (IP3), followed by restoration of intracellular buffer to terminate the IP3 response and promote ER calcium reuptake.
Assessment of the mitochondrial contribution to ATP-stimulated calcium responses. To determine the contribution made by mitochondrial calcium uptake to the clearance of calcium from the cytoplasm following ER calcium release, ATP responses were measured after treatment with FCCP. FCCP is a mitochondrial membrane-potential uncoupler, which abolishes the driving force for calcium uptake into this organelle (12). In these experiments, coverslips were incubated in 2 μM FCCP for 1 min before a 30-s stimulation with 10 μM ATP. These experiments were performed in the absence of bath calcium (EGTA buffer) to demonstrate calcium release from mitochondria on exposure to FCCP and to preclude potentially confounding effects of calcium entry.
Assessment of the contribution made by plasma membrane calcium transport. We determined that plasma membrane sodium/calcium exchange did not contribute to calcium extrusion because substitution of bath sodium with choline had no effect on the width of the ATP response (data not shown). For a variety of technical reasons, we were unable to establish a protocol to effectively inhibit the other principal extrusion pathway, the plasma membrane calcium ATPase. We were unable to use lanthanum or calmidazolium because they either abolished or markedly inhibited the ATP response in all cell lines (data not shown). Neither vanadate nor alkalinization of the bath appeared to have any effect on plasma membrane calcium ATPase because neither broadened the cell calcium responses (data not shown). As an alternative approach, we used the C18 conjugate of fura 2 as a "near-membrane" calcium probe in the external plasma membrane leaflet to monitor the appearance rate of calcium in the immediate extracellular compartment (9). Cells were loaded with 10 μM C18 fura 2 for 45 min at room temperature. We confirmed the extracellular location of the indicator by demonstrating complete quenching with 5 mM nickel, which cannot cross the cell membrane and therefore had no effect on the fluorescent signal of fura 2-AM-loaded cells. C18 fura 2 calibration curves were constructed in bath calcium concentrations, ranging from 17 to 600 nM, using a concentrate calibration kit from Molecular Probes.
Buffers and Reagents
The buffers used were composed as follows (in mM): extracellular calcium buffer (140 NaCl, 5 KCl, 2 CaCl2, 1.8 MgCl2, 10 HEPES, 5 probenecid, pH 7.4 with NaOH); 0-calcium EGTA extracellular buffer (140 NaCl, 5 KCl, 1.8 MgCl2, 10 HEPES, 0.1 EGTA, 5 probenecid, pH 7.4 with NaOH); 0-calcium BAPTA extracellular buffer (140 NaCl, 5 KCl, 10 HEPES, 0.2 BAPTA, 5 probenecid, pH 7.4 with NaOH); nominally calcium-free manganese buffer (140 NaCl, 5 KCl, 1.8 MgCl2, 10 HEPES, 0.1 MnCl2, 5 probenecid, pH 7.4 with NaOH); and intracellular buffer (19 NaCl, 125 KCl, 10 HEPES, 1 EGTA, 0.33 CaCl2, 1.4 MgCl2, 3 ATP, pH 7.2 with KOH). The BAPTA and fluorescent dyes were obtained from Molecular Probes, and all other chemicals were from Sigma. All the fluorescent dyes were made up to 1 mM stock solutions in DMSO to aid solubilization. Stock solutions of FCCP (Sigma) were made up to 2 mM in DMSO.
Data Analysis
To present actual cell calcium data, response traces for individual experiments were superimposed (based on the initial upstroke) and plotted as the mean flanked by the SE range using Graph Pad software (Prism). Because the responses were not perfectly synchronous, statistical analysis required manual examination from individual experiments to determine the peak amplitude, time for decay from peak to 80% of peak (tpeak-80%), and the time taken for decay from 80% of peak to 20% of peak (t80–20%). These numerical descriptors were applied as a simple way of quantifying the kinetics of the calcium responses and have no a priori physiological significance. In the manganese quench experiments, the effect of ATP on the rate of manganese entry was determined as the slope for the 1-min interval immediately preceding ATP stimulation subtracted from the slope for the 1-min interval starting 2 min after stimulation. In similar fashion, the rate of decay of the Ca2+ transient induced by FCCP was also expressed as the slope. Comparisons between cell line groups were made by t-test, with significance accepted at the level of P < 0.05.
RESULTS
In a series of concentration-response experiments (Fig. 1), it was determined that 10 μM ATP produced an 80%-of-maximal ATP-stimulated cell calcium response in both control and polycystin-1 expressing cell lines. Because responses in the two groups of cells were very similar, it is unlikely that expression of polycystin-1 was having a dominant effect on the subtype profile or pharmacodynamics of the purinergic receptors expressed by these cells. This dose was used in all subsequent experiments. In the compiling of the numerical values reported in the text, the cell lines were combined into control and polycystin-expressing groups, and data were reported as the mean (SE) for each of the two groups. In the figures, these results are depicted as composite curves for individual cell lines, with the number of experiments for each cell line stated in the captions. In all cases, experiments were performed over multiple days spanning several weeks, with equally matched numbers of cell lines studied on each day.
Effect of Polycystin-1 on Cell Calcium Responses to Stimulation with ATP in the Absence and Presence of Bath Calcium
Store-operated or capacitative calcium entry is thought to make a significant contribution to the latter part of the calcium response (26). Therefore, to examine the contribution made by capacitative calcium entry, ATP responses were studied twice in each preparation, first in the absence of bath calcium, and then in the presence of bath calcium. Following ATP treatment in EGTA buffer, cells were allowed to recover for 10 min in calcium containing buffer before repeat exposure to ATP. As shown in Fig. 1, in the presence of bath calcium, the ATP-stimulated cell calcium responses of the control cell lines F6 and F8 exhibited a broader peak and more gradual decay than the polycystin-1-expressing cell lines C8/68 and G7/36 [tpeak-80% 29.2 (SE 2.8) vs. 10.6 s (1.2 SE), P = 4.44 x 10–6; t80–20% 37.4 (SE 4.8) vs. 22.4 s (SE 1.9), P = 0.001]. The basal fluorescence ratio, given in arbitrary units, was slightly lower in polycystin-1-expressing cells in zero-calcium [control 0.83 (SE 0.02) polycystin-expressing 0.76 (SE 0.02), P = 0.014] but was not significantly different between cell lines in 2 mM calcium buffer [control 0.86 (SE 0.02), polycystin-expressing 0.81 (SE 0.02), P = 0.09]. The peak amplitude of the ATP response was no different between cell lines, irrespective of bath calcium. Manipulation of bath calcium had no effect on the decay rate of the ATP response in either group of cell lines, [t80–20% 0-calcium vs. calcium-containing buffer; control, 36.8 (SE 3.2) vs. 37.4 s (SE 4.8), P = 0.88; polycystin-expressing cells 26.2 (SE 1.8) vs. 22.4 s (SE 1.9), P = 0.26]. However, the peaks of the control cell responses were greatly expanded by replacement of bath calcium, exhibiting a second and often third spike, [tpeak-80% 15.4 (SE 1.8) in zero-calcium vs. 29.2 s (SE 2.8) plus calcium, P = 6.11 x 10–5]. By contrast, the peak calcium response to ATP was always uniphasic in the polycystin-1-expressing cells, being unaffected by bath calcium concentration [tpeak-80% 10.5 (SE 1.0) in zero-calcium vs. 10.6 s (SE 1.2) in plus calcium, P = 0.86]. Because of the influence of bath calcium on the control cells, the difference between the two groups of cell lines in the breadth of the peak in 2 mM calcium buffer [tpeak-80% control 29.2 (SE 2.8), polycystin-expressing 10.6 s (SE 1.2), P = 4.44 x 10–6] was less evident in the zero-calcium bath [tpeak-80% control 15.4 (SE 1.8) vs. polycystin-expressing cells 10.5 s (SE 1.0), P = 0.027].
It therefore appeared that polycystin-1 was influencing cell calcium responses through combined effects on calcium entry (suppression of bath calcium-dependent multiple peaking) and intracellular calcium turnover (bath calcium-independent acceleration of the decay of the calcium response).
Effect of Polycystin-1 on Ligand-Activated Calcium Entry
To assess calcium entry using an independent methodology, the manganese quench approach was used to examine activation of calcium entry pathways. In these experiments, cells were loaded with fura 2-AM and then perfused in a nominally calcium-free bath containing 0.1 mM manganese. Manganese enters the cell via store-operated or capacitative calcium channels and quenches fura 2 fluorescence when excited at the isosbestic wavelength of 360 nm. Therefore, the rate of manganese quench of fura 2 fluorescence provides an index of the degree of activation of calcium entry channels (21). As can be seen in Fig. 2, treatment with ATP resulted in a steepening of the quench slope in the control but much less so in the polycystin-1-expressing cell lines [slope; control –887 (SE 137) vs. polycystin-expressing –143 (SE 67) fluorescent units/s, P = 3.86 x 10–5]. Basal quench slopes were no different between cell lines [control –1,208 (SE 102), polycystin-expressing –1,083 fluorescent units/s (SE 88), P = 0.37]. The manganese quench and zero-calcium experiments therefore indicated that polycystin-1 inhibited ATP-stimulated calcium entry, providing a plausible explanation for the narrowed peaks in the calcium response in the polycystin-1-expressing cell lines C8/68 and G7/36. However, the effect of polycystin-1 on calcium entry could not have accounted for the differences between cell lines over the later phase of the response, which was not sensitive to changes in bath calcium.
As depicted diagrammatically in Fig. 3, there were several alternative explanations for the effect of polycystin-1 on the decay phase of the calcium response. The heterologously expressed protein might have promoted more rapid clearing of cytoplasmic calcium following the calcium release phase through 1) enhanced reuptake into some intracellular compartment (most probably the ER); 2) increased buffering within the cell, (possibly by mitochondria); or 3) increased calcium extrusion across the cell membrane. The next series of experiments was designed to examine each of these possible explanations for the effect of polycystin-1.
Use of Ionomycin to Assess the Effect of Polycystin-1 on the Rate of Store Repletion Following a Calcium Release Stimulus
We reasoned that, if polycystin-1 accelerated calcium reuptake into intracellular stores following a calcium release response, then the total calcium content within these stores should recover more rapidly after stimulation with ATP (Fig. 3). Therefore, if total store calcium could be assayed at a defined time soon after ATP exposure, the levels should be greater in cells expressing polycystin-1. Ionomycin can be used to provide an indication of total intracellular calcium stores, because at high concentrations, this agent will render not only the plasma membrane but also intracellular membranes permeable to calcium (4). If applied to fura 2-loaded cells in a zero-calcium bath, ionomycin-induced cell calcium responses can only arise as a consequence of calcium release from within the cell and therefore provide an indication of the total amount of calcium sequestered within membrane-delimited intracellular compartments. The results obtained using this approach are described in Fig. 4. In these experiments, BAPTA was used as the bath calcium buffer because of its greater affinity and more rapid binding of calcium. Fura 5F-AM was chosen as the calcium indicator because of its capacity to report high-amplitude or fleeting calcium transients more effectively than fura 2, as discussed in EXPERIMENTAL PROCEDURES. Under basal, nonstimulated conditions, ionomycin-induced cell calcium responses in cells perfused with a BAPTA-buffered solution were the same in control and polycystin-1-expressing lines [ionomycin-induced peak fluorescence ratio; controls 0.27 (SE 0.03) vs. polycystin-expressing cell lines 0.27 (SE 0.03), P = 0.95]. This indicated that, in unstimulated cells, the total compartmentalized intracellular calcium content was not influenced by polycystin-1. The response to ionomycin was considerably attenuated in all cell lines when exposed to the ionophore 100 s after the ATP-induced cell calcium response. This was expected because, in the absence of bath calcium, ATP stimulation would have caused depletion of intracellular calcium stores that could not have been replenished by the usual capacitative entry of calcium from outside the cell. Residual ionomycin peaks were significantly greater in cells expressing polycystin-1 [controls 0.053 (SE 0.007); polycystin-1-expressing cells 0.108 (SE 0.008), P = 3.4 x 10–5], indicating that calcium stores were replenished more rapidly in these cells. In these experiments, the ATP peaks preceding exposure to ionomycin were greater in the polycystin-1-expressing cell lines [0.36 (SE 0.02) vs. 0.47 (SE 0.03), P = 0.007], indicating that these cells did not simply release less calcium in response to ATP. In a separate series of experiments, cells were subjected to essentially the same protocol, except that ionomycin was applied 460 s following exposure to ATP, i.e., 3 min later than in the original ionomycin experiments. Following this additional 3-min period, the control cells, but not the polycystin-expressing cells, showed an increase in the ionomycin-releasable pool compared with the 100-s ionomycin response [controls from 0.053 (SE 0.007) to 0.074 (SE 0.005), P = 0.018; polycystin expressors from 0.108 (SE 0.008) to 0.109 (SE 0.01), P = 0.92]. This is consistent with the notion that calcium reuptake rates were increased by polycystin-1 but that the somewhat slower calcium uptake rate of control cells was still sufficient to accumulate an equivalent calcium store from available intracellular sources, given a little more time. This would be expected in view of the lack of difference between cell lines in the basal ionomycin responses.
We are proposing that the ionomycin release peak is related to the rate of calcium reuptake following ATP-stimulated release and that increased reuptake explains the decrease in t80–20% in polycystin-1-expressing cells. Therefore, it would be predicted that there should be an inverse relationship between the magnitude of the ionomycin peak and the t80–20%. This is shown to be the case in Fig. 5.
Direct Measurement of the Effect of Polycystin-1 on the Rate of Recovery of ER Calcium Concentration Following a Calcium Release Stimulus
The ionomycin approach indicated the total intracellular calcium content of compartmentalized calcium. To complement and corroborate the ionomycin experiments, we performed direct measurement of compartmentalized calcium concentration using the low-affinity calcium indicator mag-fura 2-AM. Because cells had to be permeabilized to minimize signal contamination from cytoplasmic dye (therefore mandating the use of an ATP-containing intracellular buffer), it was not possible to examine receptor-mediated ATP responses. For this reason, we used exogenous IP3 as the calcium release stimulus, thereby examining the influence of polycystin-1 on ER calcium responses independently of possible effects at the receptor level. As shown in Fig. 6, administration of IP3 resulted in a prompt fall in compartmentalized calcium concentration, which recovered on washout. It is most likely that we were measuring calcium within the ER, because this is the principal IP3-responsive compartment. The tpeak–80% and t80–20% values were again used to describe two components of the ER response, duration of peak depletion and rate of refilling, respectively. Expression of polycystin-1 was associated with a shortened period of IP3-induced maximal compartmental depletion [tpeak–80%; control 49.9 (SE 1.7) vs. polycystin- expressing 37 s (SE 2), P = 3.6 x 10–5] and an accelerated rate of refilling [t80–20% control 38.7 (SE 2.8) vs. polycystin-expressing 29.6 s (SE 1.4), P = 0.016]. As can be seen from the composite response curves in Fig. 6, the fluorescent signal returned to the prestimulus baseline by 100 s in the polycystin-1-expressing cells, whereas the control cells remained significantly below baseline at the 100-s time point and beyond, up to the termination of the experiment. In keeping with this observation, the 100-s-post-IP3 minus the pre-IP3 fluorescence values were significantly different from an hypothesized mean of 0 for control but not polycystin-1-expressing cells [control –0.031 (SE 0.007), P = 0.0007, polycystin-1-expressing 0 (SE 0.006), P = 0.97]. The pre-IP3 baseline fluorescent signal was not significantly different between groups [control 0.64 (SE 0.04) vs. polycystin-1-expressing 0.59 (SE 0.04), P = 0.4].
Examination of the Role of Mitochondrial Buffering of Calcium in the Effect of Polycystin-1
Because mitochondria can accumulate calcium during peaks in cytoplasmic calcium concentration (11), it was possible that polycystin-1 might have accelerated the decay of the cell calcium response through upregulation of mitochondrial calcium buffering. As seen in Fig. 7, the effect of polycystin-1 to accelerate the decay of the cytoplasmic calcium response to ATP persisted after pretreatment with FCCP to abolish mitochondrial calcium uptake. A small calcium transient was seen after treatment with FCCP, probably due to release from the depolarized mitochondrial compartment. There were no differences between cell lines in the magnitude of the FCCP-associated calcium transient. If anything, the rate of return of the FCCP transient to baseline was more gradual in the polycystin-1-expressing cells.
Evaluation of Effect of Polycystin-1 on Calcium Export from the Cell Following Treatment with ATP
As described in Fig. 3, polycystin-1 might have accelerated the decay of the ATP-stimulated calcium response through enhanced calcium extrusion across the plasma membrane. We used the near-membrane calcium indicator fura C18 to measure the accumulation of calcium in the immediate extracellular domain. As seen in Fig. 8, the extracellular near-membrane calcium transient appeared shortly after the spike in the intracellular response and peaked during the period of the steepest slope of the decay phase of the intracellular response. If polycystin-1 were accelerating the decay phase through increased calcium extrusion, this would have been expected to appear as an increase in the magnitude of the near-membrane transient. If anything, the near-membrane transient was a little greater in the control cells. Based on the calibration curves, the control cell's near-membrane calcium was 500 vs. 150 nM in the polycystin-1 cells in the 25 μM BAPTA-buffered solution used in the ATP stimulation experiments. As solutions were switched at the beginning of the calibration experiments, it was clear that this BAPTA-buffered solution actually had a much greater calcium concentration than that of the lowest calcium (17 nM) standard. In the calibration curve, the difference between cell lines in the near-membrane signal was not so marked, perhaps because of initial washout in the lowest calcium standard.
DISCUSSION
Polycystin-1 and Ligand-Activated Cell Calcium Responses
In the experiments reported in this paper, we used a variety of approaches to examine each of the possible explanations (shown diagrammatically in Fig. 3) for the more rapid decay of the ATP-stimulated calcium response in cells expressing heterologous polycystin-1. Altered calcium entry did not entirely explain the difference in calcium response kinetics, because the decay times of control and polycystin-1 cell lines remained different in the zero-calcium bath. There was no evidence for decreased ER calcium release in the polycystin-1-expressing cells because neither peak ATP-stimulated calcium responses nor total basal intracellular calcium stores were diminished in these cells relative to control. Similarly, increased extrusion of calcium could not have explained the effect of polycystin-1 to shorten the t80–20%, because the post-ATP residual calcium stores were actually elevated in polycystin-1-expressing cells; if polycystin-1 were increasing calcium extrusion, residual cell calcium stores should have been diminished relative to control cells following treatment with ATP. This conclusion is strengthened by the near-membrane data. The FCCP experiments effectively ruled out a role for the mitochondria in the polycystin-1 effect and suggested that polycystin-1-induced activation of calcium clearance from the cytoplasm might only be operative after ligand-activated ER calcium release. Thus the increased ionomycin-releasable calcium stores following ATP stimulation in polycystin-1-expressing cells most probably indicated more rapid reuptake of calcium following calcium release. In keeping with this conclusion, the mag-fura 2 experiments provided direct evidence for an effect of polycystin-1 to increase the activity of calcium reuptake pathways and confirmed that the effect of polycystin-1 was on the kinetics of ER calcium turnover rather than the total capacity of this structure.
We propose that the effect of polycystin-1 to accelerate ER calcium reuptake also accounted for the reduced calcium entry, because capacitative calcium entry appears to be activated in a graded fashion by the degree of ER store depletion (15).
Polycystin-1 and Basal Cytoplasmic Calcium Concentration
In our experiments, heterologous expression of polycystin-1 did not significantly alter resting cytoplasmic calcium concentrations in MDCK cells in calcium-containing buffer. The slight lowering of basal cytoplasmic calcium relative to control cells in the zero-calcium bath raised the possibility that polycystin-1 might have activated ER calcium retention in response to depletion of ER stores by incubation in a zero-calcium bath. A similar lack of effect of polycystin-1 on resting cell calcium concentration was evident in studies of the loss of polycystin-1 activity in embryonic renal tubular epithelial cells (23). By contrast, Qian et al. (27) reported that cytoplasmic calcium concentration was diminished in vascular smooth muscle cells from heterozygous polycystin-2 knockout mice. However, because partial loss of polycystin-2 is not associated with a renal phenotype, this effect on resting cell calcium concentration might be restricted to vascular smooth muscle cells and reflect an activity of polycystin-2 independent of the functions of polycystin-1.
Potential Significance
Based on our data, loss of polycystin-1 would be predicted to lead to prolonged ligand-activated elevation of cell calcium with reciprocally prolonged depletion of ER calcium concentration. In the following sections, we discuss how these effects might influence processes considered necessary for ADPKD pathogenesis.
Fluid secretion. Increased cytoplasmic calcium is a potent stimulus for chloride secretion and inhibition of sodium reabsorption (8). Therefore, one of the functions of polycystin-1 might be to attenuate the net secretory response to ATP. Because the effect of polycystin-1 was evident in cells stimulated with IP3, it is likely that polycystin-1 would similarly attenuate chloride secretion in response to other hormones acting through phospholipase C. If so, then loss of polycystin-1 in the ADPKD cyst epithelium would lead to a prosecretory state through the prolongation of cytoplasmic calcium responses and chloride channel activation following exposure to a number of naturally occurring ligands.
Proliferation. In a recent series of studies, Yamaguchi et al. (38) demonstrated that blockade of plasma membrane calcium channels, or chelation of extracellular calcium, reversed the cAMP growth phenotype to an ADPKD type pattern. Although a decline in cytoplasmic calcium was evident in response to either manipulation, these authors showed that the thapsigargin-releasable calcium store was also depleted. Because ER store depletion has been linked to ERK activation (20), ER calcium depletion might have been a stimulus for the reversal of the cAMP response in the studies of Yamaguchi et al. (38). We have shown that polycystin-1 enhances calcium uptake into the ER and would therefore act to minimize ER calcium depletion in response to a variety of release stimuli. Because, in vivo, renal epithelial cells are continuously exposed to endocrine, paracrine, and autocrine stimulation, the calcium concentration of the ER of cells lacking polycystin-1 might be slower to recover and therefore remain depleted to a greater extent than neighboring unaffected cells. This could promote proliferation through ERK activation in cyst epithelia. Our use of ATP as the calcium release stimulus is pathophysiologically relevant, because all the components for purinergic signal transduction are resident in the cystic environment (29).
Apoptosis. Our observations raise a possible explanation for the increased apoptosis seen in polycystic epithelia in ADPKD (36). As discussed above, calcium released by the ER can be cleared from the cytoplasm by buffering by intracellular organelles such as the mitochondrion (mitochondrial calcium uptake is closely coupled to ER calcium release) (28). There is evidence that mitochondrial calcium loading is an important mechanism for apoptotic cell death (32). The prolonged elevation of cytoplasmic calcium with every calcium release stimulus in ADPKD cyst epithelia could conceivably result in increased apoptosis through progressive accumulation of calcium within mitochondria.
In conclusion, we suggest that it might be important to consider the interplay between ER and cytoplasmic calcium in future studies of the cell biological consequences of loss of polycystin function in ADPKD.
GRANTS
This work was supported by National Institutes of Health (NIH) award KO8-DK-066323–01 and by grants from Satellite Research, the PKD Foundation, and the National Kidney Foundation of Maryland. G. Germino is the Blum Scholar of Johns Hopkins University School of Medicine and is funded by the NIH and the American Heart Association. R. Zeigelstein received support from NIH Grant RO1-HL-63720.
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
Aguiari G, Campanella M, Manzati E, Pinton P, Banzi M, Moretti S, Piva R, Rizzuto R, and del Senno L. Expression of polycystin-1 C-terminal fragment enhances the ATP-induced Ca2+ release in human kidney cells. Biochem Biophys Res Commun 301: 657–664, 2003.
Arnould T, Kim E, Tsiokas L, Jochimsen F, Gruning W, Chang JD, and Walz G. The polycystic kidney disease 1 gene product mediates protein kinase C -dependent and c-Jun n-terminal kinase-dependent activation of the transcription factor AP-1. J Biol Chem 273: 6013–6018, 1998.
Bhunia AK, Piontek K, Boletta A, Liu L, Qian F, Xu PN, Germino FJ, and Germino GG. PKD1 induces p21 (WAF1) and regulation of the cell cycle via direct activation of the JAK-STAT signaling pathway in a process requiring PKD2. Cell 109: 157–168, 2002.
Bird G, Obie J, and Putney J Jr. Functional homogeneity of the non-mitochondrial Ca2+ pool in intact mouse lacrimal acinar cells. J Biol Chem 267: 18382–18386, 1992.
Boletta A, Qian F, Onuchic LF, Bhunia AK, Phakdeekitcharoen B, Hanaoka K, Guggino W, Monaco L, and Germino G. Polycystin-1, the gene product of PKD1, induces resistance to apoptosis and spontaneous tubulogenesis in MDCK Cells. Mol Cell 6: 1267–1273, 2000.
Boletta A, Qian F, Onuchic LF, Bragonzi A, Cortese M, Deen PM, Courtoy PJ, Soria MR, Devuyst O, Monaco L, and Germino GG. Biochemical characterization of bona fide polycystin-1 in vitro and in vivo. Am J Kidney Dis 38: 1421–1429, 2001.
Cuffe JE, Bielfeld-Ackermann A, Thomas J, Leipziger J, and Korbmacher C. ATP stimulates Cl– secretion and reduces amiloride-sensitive Na+ absorption in M-1 mouse cortical collecting duct cells. J Physiol (Lond) 524: 77–90, 2000.
De Luisi A and Hofer AM. Evidence that Ca2+ cycling by the plasma membrane Ca2+-ATPase increases the 'excitability' of the extracellular Ca2+-sensing receptor. J Cell Sci 116: 1527–1538, 2003.
Delmas P, Nomura H, Li X, Lakkis M, Luo Y, Segal Y, Fernandez-Fernandez JM, Harris P, Frischauf AM, Brown DA, and Zhou J. Constitutive activation of G-proteins by polycystin-1 is antagonized by polycystin-2. J Biol Chem 277: 11276–11283, 2002.
The European Polycystic Kidney Disease Consortium. The polycystic kidney disease 1 gene encodes a 14 kb transcript and lies within a duplicated region on chromosome 16. Cell 77: 881–894, 1994.
Fowler MR and Hunter M. Mitochondrial Ca2+ transport in frog early distal tubule. Exp Physiol 90: 195–201, 2004.
Fulceri R, Bellomo G, Mirabelli F, Gamberucci A, and Benedetti A. Measurement of mitochondrial and non-mitochondrial Ca2+ in isolated intact hepatocytes: a critical re-evaluation of the use of mitochondrial inhibitors. Cell Calcium 12: 431–439, 1991.
Gonzalez-Perrett S, Kim K, Ibarra C, Damiano AE, Zotta E, Batelli M, Harris PC, Reisin IL, Arnaout MA, and Cantiello HF. Polycystin-2, the protein mutated in autosomal dominant polycystic kidney disease (ADPKD), is a Ca2+-permeable nonselective cation channel. PNAS 98: 1182–1187, 2001.
Hanaoka K, Qian F, Boletta A, Bhunia AK, Piontek K, Tsiokas L, Sukhatme VP, Guggino WB, and Germino GG. Co-assembly of polycystin-1 and -2 produces unique cation-permeable currents. Nature 408: 990–994, 2000.
Hofer AM, Fasolato C, and Pozzan T. Capacitative Ca2+ entry is closely linked to the filling state of internal Ca2+ stores: a study using simultaneous measurements of ICRAC and intraluminal [Ca2+]. J Cell Biol 140: 325–334, 1998.
Hooper K, Unwin R, and Sutters M. The isolated C-terminus of polycystin-1 promotes increased ATP-stimulated chloride secretion in a collecting duct cell line. Clin Sci (Colch) 104: 217–221, 2003.
Huan Y and van Adelsberg J. Polycystin-1, the PKD1 gene product, is in a complex containing E-cadherin and the catenins. J Clin Invest 104: 1459–1468, 1999.
Kim E, Arnould T, Sellin LK, Benzing T, Fan MJ, Gruning W, Sokol SY, Drummond I, and Walz G. The polycystic kidney disease 1 gene product modulates Wnt signaling. J Biol Chem 274: 4947–4953, 1999.
Koulen P, Cai Y, Geng L, Maeda Y, Nishimura S, Witzgall R, Ehrlich BE, and Somlo S. Polycystin-2 is an intracellular calcium release channel. Nat Cell Biol 4: 191–197, 2002.
Maloney JA, Tsygankova OM, Yang L, Li Q, Szot A, Baysal K, and Williamson JR. Activation of ERK by Ca2+ store depletion in rat liver epithelial cells. Am J Physiol Cell Physiol 276: C221–C230, 1999.
Missiaen L, Declerck I, Droogmans G, Plessers L, De Smedt H, Raeymaekers L, and Casteels R. Agonist-dependent Ca2+ and Mn2+ entry dependent on state of filling of Ca2+ stores in aortic smooth muscle cells of the rat. J Physiol 427: 171–186, 1990.
Mochizuki T, Wu G, Hayashi T, Xenophontos S, Veldhuisen B, Saris JJ, Reynolds DM, Cai Y, Gabow PA, Pierides A, Kimberling WJ, Breuning MH, Deltas CC, Peters DJ, and Somlo S. PKD2, a gene for polycystic kidney disease that encodes an integral membrane protein. Science 272: 1339–1342, 1996.
Nauli S, Alenghat F, YL, Williams E, Vassilev P, Li X, Elia A, Lu W, Brown E, Quinn S, Ingber D, and Zhou J. Polycystins 1 and 2 mediate mechanosensation in the primary cilium of kidney cells. Nat Genet 33: 129–137, 2003.
Nickel C, Benzing T, Sellin L, Gerke P, Karihaloo A, Liu ZX, Cantley LG, and Walz G. The polycystin-1 C-terminal fragment triggers branching morphogenesis and migration of tubular kidney epithelial cells. J Clin Invest 109: 481–489, 2002.
Parnell SC, Magenheimer BS, Maser RL, Rankin CA, Smine A, Okamoto T, and Calvet JP. The polycystic kidney disease-1 protein, polycystin-1, binds and activates heterotrimeric G-proteins in vitro. Biochem Biophys Res Commun 251: 625–631, 1998.
Putney JW Jr, Broad LM, Braun FJ, Lievremont JP, and Bird GSJ. Mechanisms of capacitative calcium entry. J Cell Sci 114: 2223–2229, 2001.
Qian Q, Hunter LW, Li M, Marin-Padilla M, Prakash YS, Somlo S, Harris PC, Torres VE, and Sieck GC. Pkd2 haplo insufficiency alters intracellular calcium regulation in vascular smooth muscle cells. Hum Mol Genet 12: 1875–1880, 2003.
Rizzuto R, Brini M, Murgia M, and Pozzan T. Microdomains with high Ca2+ close to IP3-sensitive channels that are sensed by neighboring mitochondria. Science 262: 744–747, 1993.
Schwiebert EM, Wallace DP, Braunstein GM, King SR, Peti-Peterdi J, Hanaoka K, Guggino WB, Guay-Woodford LM, Bell PD, Sullivan LP, Grantham JJ, and Taylor AL. Autocrine extracellular purinergic signaling in epithelial cells derived from polycystic kidneys. Am J Physiol Renal Physiol 282: F763–F775, 2002.
Somlo S and Ehrlich B. Human disease: calcium signaling in polycystic kidney disease. Curr Biol 11: R356–360, 2001.
Sutters M, Yamaguchi T, Maser RL, Magenheimer BS, St. John PL, Abrahamson DR, Grantham JJ, and Calvet JP. Polycystin-1 transforms the cAMP growth-responsive phenotype of M-1 cells. Kidney Int 60: 484–494, 2001.
Szabadkai G and Rizzuto R. Participation of endoplasmic reticulum and mitochondrial calcium handling in apoptosis: more than just neighborhood FEBS Lett 567: 111–115, 2004.
Vandorpe DH, Chernova MN, Jiang L, Sellin LK, Wilhelm S, Stuart-Tilley AK, Walz G, and Alper SL. The cytoplasmic C-terminal fragment of polycystin-1 regulates a Ca2+-permeable cation channel. J Biol Chem 276: 4093–4101, 2001.
Wildman SS, Hooper KM, Turner CM, Sham JSK, Lakatta EG, King BF, Unwin RJ, and Sutters M. The isolated polycystin-1 cytoplasmic COOH terminus prolongs ATP-stimulated Cl– conductance through increased Ca2+ entry. Am J Physiol Renal Physiol 285: F1168–F1178, 2003.
Wilson PD, Geng L, Li X, and Burrow CR. The PKD1 gene product, "polycystin-1," is a tyrosine-phosphorylated protein that colocalizes with 21-integrin in focal clusters in adherent renal epithelia. Lab Invest 79: 1311–1323, 1999.
Woo D. Apoptosis and loss of renal tissue in polycystic kidney diseases. N Engl J Med 333: 18–25, 1995.
Xu GM, Gonzalez-Perrett S, Essafi M, Timpanaro GA, Montalbetti N, Arnaout MA, and Cantiello HF. Polycystin-1 activates and stabilizes the polycystin-2 channel. J Biol Chem 278: 1457–1462, 2003.
Yamaguchi T, Wallace DP, Magenheimer BS, Hempson SJ, Grantham JJ, and Calvet JP. Calcium restriction allows cAMP activation of the B-Raf/ERK pathway, switching cells to a cAMP-dependent growth-stimulated phenotype. J Biol Chem 279: 40419–40430, 2004.
Yoder BK, Hou X, and Guay-Woodford LM. The polycystic kidney disease proteins, polycystin-1, polycystin-2, polaris, and cystin, are co-localized in renal cilia. J Am Soc Nephrol 13: 2508–2516, 2002.(K. M. Hooper, A. Boletta,)
Department of Biological Research and Technology, Dulbecco Telethon Institute, San Raffaele Scientific Institute, Milan, Italy
ABSTRACT
Autosomal dominant polycystic kidney disease (ADPKD) types 1 and 2 arise as a consequence of mutations in the PKD1 or PKD2 genes, encoding polycystins-1 and -2. Because loss of function of either of the polycystins leads to a very similar phenotype and the two proteins are known to interact, polycystins-1 and -2 are probably active in the same pathway. The way in which loss of either polycystin leads to the development of ADPKD remains to be established, but disturbances of cell calcium regulation are likely to play an important role. Here, we demonstrate that polycystin-1, heterologously expressed in Madin-Darby canine kidney cells, had a pronounced effect on intracellular calcium homeostasis. ATP-induced calcium responses in transfection control cells exhibited a double peak and relatively gradual return to baseline. By contrast, cells expressing heterologous polycystin-1 showed a brief, uniphasic peak and an accelerated rate of decay. Heterologously expressed polycystin-1 accelerated endoplasmic reticulum (ER) calcium reuptake and inhibited capacitative calcium entry; we found no effect of the protein on mitochondrial calcium buffering or plasma membrane calcium extrusion. We therefore propose that polycystin-1 accelerated the decay of the cell calcium response to ATP by upregulation of ER calcium reuptake and consequent minimization of the stimulus for capacitative calcium entry. It is possible that cellular dedifferentiation, fluid secretion, and proliferation might therefore arise in ADPKD as a consequence of disturbances in cytoplasmic and ER calcium homeostasis and aberrant capacitative calcium entry.
autosomal dominant polycystic kidney disease; purinergic; calcium signaling; stable transfection; Madin-Darby canine kidney cells
AUTOSOMAL DOMINANT POLYCYSTIC kidney disease (ADPKD) arises as a consequence of loss of function of either polycystin-1 or polycystin-2, the gene products of the PKD1 and PKD2 genes respectively (10a, 22). ADPKD type 2, arising from mutations in the PKD2 gene, typically progresses somewhat less rapidly than ADPKD type 1, but the latter accounts for the majority of affected individuals. The phenotypes associated with either genotype are otherwise very similar, indicating that the two proteins probably function in the same pathway. The disease, which affects 1 in 1,000 individuals, is characterized by the development of multitudes of renal cysts. As these cysts expand, the resultant compression and distortion of normal renal tissue leads to loss of renal function, which typically progresses to end-stage and dialysis dependency by mid- to late adulthood. Polycystin-1 is a very large transmembrane protein, which appears to be active in a number of diverse pathways. The isolated truncated cytoplasmic COOH terminus of polycystin-1 has shown activity in 1) wnt signal transduction pathways (18); 2) AP-1 transcription factor activation (2); 3) heterotrimeric G protein-coupled signal transduction (25); 4) regulation of calcium channel activity (33, 37); 5) in-gel morphogenesis (24); and 6) modulation of ATP-stimulated cell calcium and chloride secretory responses (16, 34). The full-length protein has been shown to 1) regulate the ion channel properties of polycystin-2 (14), 2) promote tubulogenesis and resistance to apoptosis (5), 3) activate the JAK-STAT pathway (3), and 4) regulate G protein-coupled signal transduction (10). Polycystin-1 has been localized to the plasma membrane (6), at the sites of adherens junctions (17) and focal adhesions (35), and also, more recently, in the cilia (39). Polycystin-2 is predominantly seen in the endoplasmic reticulum (ER) (19) but may also be expressed at the cell membrane (13) and apical cilia (39). The cell calcium response to cilial deflection appears to depend on expression of functional polycystin-1 and -2 (23). The observation that polycystin-2 is a nonspecific cation channel with predominant calcium conductance (19) raised the possibility that ADPKD might develop as a consequence of disturbed intracellular calcium homeostasis (30). However, the effects of the polycystins on cell calcium homeostasis, whether acting independently or as a functional complex, are currently incompletely understood. If dysregulation of cell calcium homeostasis is an important mechanism of disease, in view of the fact that ADPKD types 1 and 2 share an almost identical phenotype, it would be predicted that the two proteins should have concordant effects on some aspect of cell calcium regulation. However, from studies of the effects of the polycystins on cell calcium homeostasis it is not clear that the two proteins always have consistent functions in this respect. Although heterozygous loss of polycystin-2 has been linked to a fall in resting cytoplasmic calcium concentration in vascular smooth muscle cells (27), homozygous loss of polycystin-1 activity seemed not to alter resting calcium concentration in collecting duct cells (23). Overexpression of polycystin-2 appeared to increase the cell calcium response to vasopressin in LLC-PK cells (19); we are not aware of any previous studies of heterologously expressed full-length polycystin-1 on stimulated cell calcium responses, but loss of functional polycystin actually caused an increase in the thrombin-stimulated cell calcium response in collecting duct cells (23). That polycystin-1 and -2 are required for the cell calcium response to cilial deflection is perhaps the most convincing evidence for a functional concordance in the regulation of cell calcium homeostasis (23).
To achieve a more complete understanding of the role of the polycystins in integrated cell calcium homeostasis, it is clear that more work is required to examine the effects of heterologously expressed proteins and loss of polycystin function on resting and stimulated cell calcium responses in renal epithelia and other cell types.
In earlier studies, we found that expression of the isolated cytoplasmic COOH terminus of polycystin-1 resulted in the prolongation of ligand-activated cell calcium responses (16, 34). Others have presented data consistent with these findings (1). We found that this effect resulted in increased ATP-stimulated chloride secretion and proposed that, in promoting a prosecretory phenotype, the truncated polycystin-1 fragment was acting as a dominant negative with respect to endogenous polycystin-1. The possibility that the cytoplasmic COOH-terminal polycystin-1 truncation might act as a dominant negative has been raised by other studies (31, 38). Consequently, we predicted that full-length polycystin-1 should have the opposite effect to that of the truncation, to abbreviate the ATP-stimulated cell calcium response. In the findings of the current study, we show that this is indeed the case. Our experiments indicate that heterologously expressed full-length polycystin-1 abbreviated the cell calcium response to ATP through a combination of acceleration of ER calcium reuptake and attenuation of capacitative calcium entry.
EXPERIMENTAL PROCEDURES
Cell Culture Conditions
Stably transfected Madin-Darby canine kidney (MDCK)-derived cell lines expressing human polycystin-1 were developed as previously described (5). The original parental cells were MDCK type II. It should be noted that heterologous human polycystin-1 was not expressed at high levels in these cells, because the protein was only detectable after enrichment by immunoprecipitation (5). Native expression of polycystin-1 in these cell lines was only detectable by RT-PCR (data not shown). In the experiments described below, we studied two vector control (F6 and F8) and two polycystin-1-expressing cell lines (C8/68 and G7/36) between passages 4 and 16. Cells were propagated in DMEM/F-12 supplemented with 5% fetal calf serum under dual selection with G418 and Zeocin and were passaged weekly at 1 x 106 cells/25-cm2 flask. To prepare cells for calcium imaging, glass coverslips were coated with Matrigel (0.45 mg/ml) for at least 1 h at room temperature. Cells were sparsely seeded onto the coated glass coverslips and grown for 3 days in culture. Small clusters of cells were selected for imaging. These islands of cells were chosen to ensure that the same number and density of cells (25 cells/field) was studied in each experiment. The morphology of these groups was similar in all the cell lines studied.
Fluorescent Cell Calcium Measurements
The coverslips were placed in a perfusion chamber mounted on the stage of a Nikon Diaphot microscope fitted with dual photomultipliers and illuminated with a xenon light source passed through a monochromator and appropriate filters (Photon Technologies International). The cells were excited at 340 and 380 nm, and emission was measured at 510 nm. Cell fluorescence was measured in response to excitation for 100 ms once every second. Shutter opening was controlled using a Spectroscan device driven by Felix software. Imaging was performed with a long-working-distance Nikon Fluor 40 lens. All perfusion was performed at room temperature at 2 ml/min to ensure rapid turnover of the 0.25-ml bath volume.
Bath Calcium-Switch Repeat-Stimulation Studies
Cells were loaded in a 5 μM solution of the cell-permeant acetoxymethyl (AM) ester of the calcium indicator fura 2 (fura 2-AM) at room temperature for 45 min in DMEM/F-12. All fluorescent dyes used in these experiments were made up to a 1 mM stock solution in 10% Pluronic/DMSO. Following incubation, cells were allowed to recover for 30 min in DMEM/F-12. The incubation and recovery media both contained 5 mM probenecid to maximize dye loading through inhibition of dye extrusion. Cells were placed into the perfusion chamber in a zero-calcium EGTA buffer and stimulated with 10 μM ATP for a 30-s pulse in this buffer. Cells were then switched into a calcium-containing buffer for 10 min to refill calcium stores before being stimulated again with a second 30-s pulse of ATP in the presence of bath calcium. This sequence of studies was used to avoid the need for a prolonged zero-calcium washout of calcium-containing buffer.
Measurement of Intracellular Calcium Stores by Ionomycin
The magnitude of compartmentalized calcium content can be inferred from the peak in cytoplasmic calcium concentration in response to the release of this calcium by ionomycin (4). However, we were concerned that the peak might have been rendered very transient as a consequence of concurrent loss of calcium from the cell through the ionomycin-permeabilized plasma membrane. For this reason, we decided to use fura 5F in these experiments, because the rapid response of this indicator made it superior to fura 2 in the capture of short-lived calcium transients. Cells were loaded with 5 μM fura 5F-AM for 45 min at room temperature in the presence of 5 mM probenecid. After being mounted on the microscope stage, cells were perfused with a zero-calcium (200 μM BAPTA) buffer and treated with 10 μM ATP followed by 10 μM ionomycin at 100 or 460 s post-ATP, to release the total sequestered intracellular calcium stores at these time points. Ionomycin was added at 2 ml/min using a gravity-feed system through a separate perfusion pathway to prevent carryover contamination.
Manganese Quench Studies
Because of the duration of these studies and the presence of basal quenching, cells were maximally loaded with 10 μM fura 2-AM for 45 min at room temperature in the presence of 5 mM probenecid. The cells were perfused with nominally calcium-free extracellular buffer containing MnCl2 (which quenches the fura 2 signal on access to the cytoplasm) and treated with 10 μM ATP for 30 s. For these studies, cells were excited at 360 nm, the isosbestic point of fura 2, and the emission was recorded at 510 nm.
ER Calcium Imaging
Cells were loaded with 10 μM mag-fura 2-AM for 1 h 15 min at room temperature in the absence of probenecid. Cells were perfused with zero-calcium EGTA extracellular buffer and then permeabilized with saponin (20 μg/ml). Once the mag-fura 2 escaped from the cytoplasm (defined as a decline in the 340-nm signal to one-third of the starting value), the cells were switched into the intracellular buffer. Transfer to intracellular buffer resulted in an increase in ER calcium signal because of activation of the ER calcium ATPase, previously rendered inactive through ATP depletion in the permeabilization buffer (which contained no ATP). ER calcium release was then induced by treatment with a 30-s pulse of 10 μM inositol1,4,5-triphosphate (IP3), followed by restoration of intracellular buffer to terminate the IP3 response and promote ER calcium reuptake.
Assessment of the mitochondrial contribution to ATP-stimulated calcium responses. To determine the contribution made by mitochondrial calcium uptake to the clearance of calcium from the cytoplasm following ER calcium release, ATP responses were measured after treatment with FCCP. FCCP is a mitochondrial membrane-potential uncoupler, which abolishes the driving force for calcium uptake into this organelle (12). In these experiments, coverslips were incubated in 2 μM FCCP for 1 min before a 30-s stimulation with 10 μM ATP. These experiments were performed in the absence of bath calcium (EGTA buffer) to demonstrate calcium release from mitochondria on exposure to FCCP and to preclude potentially confounding effects of calcium entry.
Assessment of the contribution made by plasma membrane calcium transport. We determined that plasma membrane sodium/calcium exchange did not contribute to calcium extrusion because substitution of bath sodium with choline had no effect on the width of the ATP response (data not shown). For a variety of technical reasons, we were unable to establish a protocol to effectively inhibit the other principal extrusion pathway, the plasma membrane calcium ATPase. We were unable to use lanthanum or calmidazolium because they either abolished or markedly inhibited the ATP response in all cell lines (data not shown). Neither vanadate nor alkalinization of the bath appeared to have any effect on plasma membrane calcium ATPase because neither broadened the cell calcium responses (data not shown). As an alternative approach, we used the C18 conjugate of fura 2 as a "near-membrane" calcium probe in the external plasma membrane leaflet to monitor the appearance rate of calcium in the immediate extracellular compartment (9). Cells were loaded with 10 μM C18 fura 2 for 45 min at room temperature. We confirmed the extracellular location of the indicator by demonstrating complete quenching with 5 mM nickel, which cannot cross the cell membrane and therefore had no effect on the fluorescent signal of fura 2-AM-loaded cells. C18 fura 2 calibration curves were constructed in bath calcium concentrations, ranging from 17 to 600 nM, using a concentrate calibration kit from Molecular Probes.
Buffers and Reagents
The buffers used were composed as follows (in mM): extracellular calcium buffer (140 NaCl, 5 KCl, 2 CaCl2, 1.8 MgCl2, 10 HEPES, 5 probenecid, pH 7.4 with NaOH); 0-calcium EGTA extracellular buffer (140 NaCl, 5 KCl, 1.8 MgCl2, 10 HEPES, 0.1 EGTA, 5 probenecid, pH 7.4 with NaOH); 0-calcium BAPTA extracellular buffer (140 NaCl, 5 KCl, 10 HEPES, 0.2 BAPTA, 5 probenecid, pH 7.4 with NaOH); nominally calcium-free manganese buffer (140 NaCl, 5 KCl, 1.8 MgCl2, 10 HEPES, 0.1 MnCl2, 5 probenecid, pH 7.4 with NaOH); and intracellular buffer (19 NaCl, 125 KCl, 10 HEPES, 1 EGTA, 0.33 CaCl2, 1.4 MgCl2, 3 ATP, pH 7.2 with KOH). The BAPTA and fluorescent dyes were obtained from Molecular Probes, and all other chemicals were from Sigma. All the fluorescent dyes were made up to 1 mM stock solutions in DMSO to aid solubilization. Stock solutions of FCCP (Sigma) were made up to 2 mM in DMSO.
Data Analysis
To present actual cell calcium data, response traces for individual experiments were superimposed (based on the initial upstroke) and plotted as the mean flanked by the SE range using Graph Pad software (Prism). Because the responses were not perfectly synchronous, statistical analysis required manual examination from individual experiments to determine the peak amplitude, time for decay from peak to 80% of peak (tpeak-80%), and the time taken for decay from 80% of peak to 20% of peak (t80–20%). These numerical descriptors were applied as a simple way of quantifying the kinetics of the calcium responses and have no a priori physiological significance. In the manganese quench experiments, the effect of ATP on the rate of manganese entry was determined as the slope for the 1-min interval immediately preceding ATP stimulation subtracted from the slope for the 1-min interval starting 2 min after stimulation. In similar fashion, the rate of decay of the Ca2+ transient induced by FCCP was also expressed as the slope. Comparisons between cell line groups were made by t-test, with significance accepted at the level of P < 0.05.
RESULTS
In a series of concentration-response experiments (Fig. 1), it was determined that 10 μM ATP produced an 80%-of-maximal ATP-stimulated cell calcium response in both control and polycystin-1 expressing cell lines. Because responses in the two groups of cells were very similar, it is unlikely that expression of polycystin-1 was having a dominant effect on the subtype profile or pharmacodynamics of the purinergic receptors expressed by these cells. This dose was used in all subsequent experiments. In the compiling of the numerical values reported in the text, the cell lines were combined into control and polycystin-expressing groups, and data were reported as the mean (SE) for each of the two groups. In the figures, these results are depicted as composite curves for individual cell lines, with the number of experiments for each cell line stated in the captions. In all cases, experiments were performed over multiple days spanning several weeks, with equally matched numbers of cell lines studied on each day.
Effect of Polycystin-1 on Cell Calcium Responses to Stimulation with ATP in the Absence and Presence of Bath Calcium
Store-operated or capacitative calcium entry is thought to make a significant contribution to the latter part of the calcium response (26). Therefore, to examine the contribution made by capacitative calcium entry, ATP responses were studied twice in each preparation, first in the absence of bath calcium, and then in the presence of bath calcium. Following ATP treatment in EGTA buffer, cells were allowed to recover for 10 min in calcium containing buffer before repeat exposure to ATP. As shown in Fig. 1, in the presence of bath calcium, the ATP-stimulated cell calcium responses of the control cell lines F6 and F8 exhibited a broader peak and more gradual decay than the polycystin-1-expressing cell lines C8/68 and G7/36 [tpeak-80% 29.2 (SE 2.8) vs. 10.6 s (1.2 SE), P = 4.44 x 10–6; t80–20% 37.4 (SE 4.8) vs. 22.4 s (SE 1.9), P = 0.001]. The basal fluorescence ratio, given in arbitrary units, was slightly lower in polycystin-1-expressing cells in zero-calcium [control 0.83 (SE 0.02) polycystin-expressing 0.76 (SE 0.02), P = 0.014] but was not significantly different between cell lines in 2 mM calcium buffer [control 0.86 (SE 0.02), polycystin-expressing 0.81 (SE 0.02), P = 0.09]. The peak amplitude of the ATP response was no different between cell lines, irrespective of bath calcium. Manipulation of bath calcium had no effect on the decay rate of the ATP response in either group of cell lines, [t80–20% 0-calcium vs. calcium-containing buffer; control, 36.8 (SE 3.2) vs. 37.4 s (SE 4.8), P = 0.88; polycystin-expressing cells 26.2 (SE 1.8) vs. 22.4 s (SE 1.9), P = 0.26]. However, the peaks of the control cell responses were greatly expanded by replacement of bath calcium, exhibiting a second and often third spike, [tpeak-80% 15.4 (SE 1.8) in zero-calcium vs. 29.2 s (SE 2.8) plus calcium, P = 6.11 x 10–5]. By contrast, the peak calcium response to ATP was always uniphasic in the polycystin-1-expressing cells, being unaffected by bath calcium concentration [tpeak-80% 10.5 (SE 1.0) in zero-calcium vs. 10.6 s (SE 1.2) in plus calcium, P = 0.86]. Because of the influence of bath calcium on the control cells, the difference between the two groups of cell lines in the breadth of the peak in 2 mM calcium buffer [tpeak-80% control 29.2 (SE 2.8), polycystin-expressing 10.6 s (SE 1.2), P = 4.44 x 10–6] was less evident in the zero-calcium bath [tpeak-80% control 15.4 (SE 1.8) vs. polycystin-expressing cells 10.5 s (SE 1.0), P = 0.027].
It therefore appeared that polycystin-1 was influencing cell calcium responses through combined effects on calcium entry (suppression of bath calcium-dependent multiple peaking) and intracellular calcium turnover (bath calcium-independent acceleration of the decay of the calcium response).
Effect of Polycystin-1 on Ligand-Activated Calcium Entry
To assess calcium entry using an independent methodology, the manganese quench approach was used to examine activation of calcium entry pathways. In these experiments, cells were loaded with fura 2-AM and then perfused in a nominally calcium-free bath containing 0.1 mM manganese. Manganese enters the cell via store-operated or capacitative calcium channels and quenches fura 2 fluorescence when excited at the isosbestic wavelength of 360 nm. Therefore, the rate of manganese quench of fura 2 fluorescence provides an index of the degree of activation of calcium entry channels (21). As can be seen in Fig. 2, treatment with ATP resulted in a steepening of the quench slope in the control but much less so in the polycystin-1-expressing cell lines [slope; control –887 (SE 137) vs. polycystin-expressing –143 (SE 67) fluorescent units/s, P = 3.86 x 10–5]. Basal quench slopes were no different between cell lines [control –1,208 (SE 102), polycystin-expressing –1,083 fluorescent units/s (SE 88), P = 0.37]. The manganese quench and zero-calcium experiments therefore indicated that polycystin-1 inhibited ATP-stimulated calcium entry, providing a plausible explanation for the narrowed peaks in the calcium response in the polycystin-1-expressing cell lines C8/68 and G7/36. However, the effect of polycystin-1 on calcium entry could not have accounted for the differences between cell lines over the later phase of the response, which was not sensitive to changes in bath calcium.
As depicted diagrammatically in Fig. 3, there were several alternative explanations for the effect of polycystin-1 on the decay phase of the calcium response. The heterologously expressed protein might have promoted more rapid clearing of cytoplasmic calcium following the calcium release phase through 1) enhanced reuptake into some intracellular compartment (most probably the ER); 2) increased buffering within the cell, (possibly by mitochondria); or 3) increased calcium extrusion across the cell membrane. The next series of experiments was designed to examine each of these possible explanations for the effect of polycystin-1.
Use of Ionomycin to Assess the Effect of Polycystin-1 on the Rate of Store Repletion Following a Calcium Release Stimulus
We reasoned that, if polycystin-1 accelerated calcium reuptake into intracellular stores following a calcium release response, then the total calcium content within these stores should recover more rapidly after stimulation with ATP (Fig. 3). Therefore, if total store calcium could be assayed at a defined time soon after ATP exposure, the levels should be greater in cells expressing polycystin-1. Ionomycin can be used to provide an indication of total intracellular calcium stores, because at high concentrations, this agent will render not only the plasma membrane but also intracellular membranes permeable to calcium (4). If applied to fura 2-loaded cells in a zero-calcium bath, ionomycin-induced cell calcium responses can only arise as a consequence of calcium release from within the cell and therefore provide an indication of the total amount of calcium sequestered within membrane-delimited intracellular compartments. The results obtained using this approach are described in Fig. 4. In these experiments, BAPTA was used as the bath calcium buffer because of its greater affinity and more rapid binding of calcium. Fura 5F-AM was chosen as the calcium indicator because of its capacity to report high-amplitude or fleeting calcium transients more effectively than fura 2, as discussed in EXPERIMENTAL PROCEDURES. Under basal, nonstimulated conditions, ionomycin-induced cell calcium responses in cells perfused with a BAPTA-buffered solution were the same in control and polycystin-1-expressing lines [ionomycin-induced peak fluorescence ratio; controls 0.27 (SE 0.03) vs. polycystin-expressing cell lines 0.27 (SE 0.03), P = 0.95]. This indicated that, in unstimulated cells, the total compartmentalized intracellular calcium content was not influenced by polycystin-1. The response to ionomycin was considerably attenuated in all cell lines when exposed to the ionophore 100 s after the ATP-induced cell calcium response. This was expected because, in the absence of bath calcium, ATP stimulation would have caused depletion of intracellular calcium stores that could not have been replenished by the usual capacitative entry of calcium from outside the cell. Residual ionomycin peaks were significantly greater in cells expressing polycystin-1 [controls 0.053 (SE 0.007); polycystin-1-expressing cells 0.108 (SE 0.008), P = 3.4 x 10–5], indicating that calcium stores were replenished more rapidly in these cells. In these experiments, the ATP peaks preceding exposure to ionomycin were greater in the polycystin-1-expressing cell lines [0.36 (SE 0.02) vs. 0.47 (SE 0.03), P = 0.007], indicating that these cells did not simply release less calcium in response to ATP. In a separate series of experiments, cells were subjected to essentially the same protocol, except that ionomycin was applied 460 s following exposure to ATP, i.e., 3 min later than in the original ionomycin experiments. Following this additional 3-min period, the control cells, but not the polycystin-expressing cells, showed an increase in the ionomycin-releasable pool compared with the 100-s ionomycin response [controls from 0.053 (SE 0.007) to 0.074 (SE 0.005), P = 0.018; polycystin expressors from 0.108 (SE 0.008) to 0.109 (SE 0.01), P = 0.92]. This is consistent with the notion that calcium reuptake rates were increased by polycystin-1 but that the somewhat slower calcium uptake rate of control cells was still sufficient to accumulate an equivalent calcium store from available intracellular sources, given a little more time. This would be expected in view of the lack of difference between cell lines in the basal ionomycin responses.
We are proposing that the ionomycin release peak is related to the rate of calcium reuptake following ATP-stimulated release and that increased reuptake explains the decrease in t80–20% in polycystin-1-expressing cells. Therefore, it would be predicted that there should be an inverse relationship between the magnitude of the ionomycin peak and the t80–20%. This is shown to be the case in Fig. 5.
Direct Measurement of the Effect of Polycystin-1 on the Rate of Recovery of ER Calcium Concentration Following a Calcium Release Stimulus
The ionomycin approach indicated the total intracellular calcium content of compartmentalized calcium. To complement and corroborate the ionomycin experiments, we performed direct measurement of compartmentalized calcium concentration using the low-affinity calcium indicator mag-fura 2-AM. Because cells had to be permeabilized to minimize signal contamination from cytoplasmic dye (therefore mandating the use of an ATP-containing intracellular buffer), it was not possible to examine receptor-mediated ATP responses. For this reason, we used exogenous IP3 as the calcium release stimulus, thereby examining the influence of polycystin-1 on ER calcium responses independently of possible effects at the receptor level. As shown in Fig. 6, administration of IP3 resulted in a prompt fall in compartmentalized calcium concentration, which recovered on washout. It is most likely that we were measuring calcium within the ER, because this is the principal IP3-responsive compartment. The tpeak–80% and t80–20% values were again used to describe two components of the ER response, duration of peak depletion and rate of refilling, respectively. Expression of polycystin-1 was associated with a shortened period of IP3-induced maximal compartmental depletion [tpeak–80%; control 49.9 (SE 1.7) vs. polycystin- expressing 37 s (SE 2), P = 3.6 x 10–5] and an accelerated rate of refilling [t80–20% control 38.7 (SE 2.8) vs. polycystin-expressing 29.6 s (SE 1.4), P = 0.016]. As can be seen from the composite response curves in Fig. 6, the fluorescent signal returned to the prestimulus baseline by 100 s in the polycystin-1-expressing cells, whereas the control cells remained significantly below baseline at the 100-s time point and beyond, up to the termination of the experiment. In keeping with this observation, the 100-s-post-IP3 minus the pre-IP3 fluorescence values were significantly different from an hypothesized mean of 0 for control but not polycystin-1-expressing cells [control –0.031 (SE 0.007), P = 0.0007, polycystin-1-expressing 0 (SE 0.006), P = 0.97]. The pre-IP3 baseline fluorescent signal was not significantly different between groups [control 0.64 (SE 0.04) vs. polycystin-1-expressing 0.59 (SE 0.04), P = 0.4].
Examination of the Role of Mitochondrial Buffering of Calcium in the Effect of Polycystin-1
Because mitochondria can accumulate calcium during peaks in cytoplasmic calcium concentration (11), it was possible that polycystin-1 might have accelerated the decay of the cell calcium response through upregulation of mitochondrial calcium buffering. As seen in Fig. 7, the effect of polycystin-1 to accelerate the decay of the cytoplasmic calcium response to ATP persisted after pretreatment with FCCP to abolish mitochondrial calcium uptake. A small calcium transient was seen after treatment with FCCP, probably due to release from the depolarized mitochondrial compartment. There were no differences between cell lines in the magnitude of the FCCP-associated calcium transient. If anything, the rate of return of the FCCP transient to baseline was more gradual in the polycystin-1-expressing cells.
Evaluation of Effect of Polycystin-1 on Calcium Export from the Cell Following Treatment with ATP
As described in Fig. 3, polycystin-1 might have accelerated the decay of the ATP-stimulated calcium response through enhanced calcium extrusion across the plasma membrane. We used the near-membrane calcium indicator fura C18 to measure the accumulation of calcium in the immediate extracellular domain. As seen in Fig. 8, the extracellular near-membrane calcium transient appeared shortly after the spike in the intracellular response and peaked during the period of the steepest slope of the decay phase of the intracellular response. If polycystin-1 were accelerating the decay phase through increased calcium extrusion, this would have been expected to appear as an increase in the magnitude of the near-membrane transient. If anything, the near-membrane transient was a little greater in the control cells. Based on the calibration curves, the control cell's near-membrane calcium was 500 vs. 150 nM in the polycystin-1 cells in the 25 μM BAPTA-buffered solution used in the ATP stimulation experiments. As solutions were switched at the beginning of the calibration experiments, it was clear that this BAPTA-buffered solution actually had a much greater calcium concentration than that of the lowest calcium (17 nM) standard. In the calibration curve, the difference between cell lines in the near-membrane signal was not so marked, perhaps because of initial washout in the lowest calcium standard.
DISCUSSION
Polycystin-1 and Ligand-Activated Cell Calcium Responses
In the experiments reported in this paper, we used a variety of approaches to examine each of the possible explanations (shown diagrammatically in Fig. 3) for the more rapid decay of the ATP-stimulated calcium response in cells expressing heterologous polycystin-1. Altered calcium entry did not entirely explain the difference in calcium response kinetics, because the decay times of control and polycystin-1 cell lines remained different in the zero-calcium bath. There was no evidence for decreased ER calcium release in the polycystin-1-expressing cells because neither peak ATP-stimulated calcium responses nor total basal intracellular calcium stores were diminished in these cells relative to control. Similarly, increased extrusion of calcium could not have explained the effect of polycystin-1 to shorten the t80–20%, because the post-ATP residual calcium stores were actually elevated in polycystin-1-expressing cells; if polycystin-1 were increasing calcium extrusion, residual cell calcium stores should have been diminished relative to control cells following treatment with ATP. This conclusion is strengthened by the near-membrane data. The FCCP experiments effectively ruled out a role for the mitochondria in the polycystin-1 effect and suggested that polycystin-1-induced activation of calcium clearance from the cytoplasm might only be operative after ligand-activated ER calcium release. Thus the increased ionomycin-releasable calcium stores following ATP stimulation in polycystin-1-expressing cells most probably indicated more rapid reuptake of calcium following calcium release. In keeping with this conclusion, the mag-fura 2 experiments provided direct evidence for an effect of polycystin-1 to increase the activity of calcium reuptake pathways and confirmed that the effect of polycystin-1 was on the kinetics of ER calcium turnover rather than the total capacity of this structure.
We propose that the effect of polycystin-1 to accelerate ER calcium reuptake also accounted for the reduced calcium entry, because capacitative calcium entry appears to be activated in a graded fashion by the degree of ER store depletion (15).
Polycystin-1 and Basal Cytoplasmic Calcium Concentration
In our experiments, heterologous expression of polycystin-1 did not significantly alter resting cytoplasmic calcium concentrations in MDCK cells in calcium-containing buffer. The slight lowering of basal cytoplasmic calcium relative to control cells in the zero-calcium bath raised the possibility that polycystin-1 might have activated ER calcium retention in response to depletion of ER stores by incubation in a zero-calcium bath. A similar lack of effect of polycystin-1 on resting cell calcium concentration was evident in studies of the loss of polycystin-1 activity in embryonic renal tubular epithelial cells (23). By contrast, Qian et al. (27) reported that cytoplasmic calcium concentration was diminished in vascular smooth muscle cells from heterozygous polycystin-2 knockout mice. However, because partial loss of polycystin-2 is not associated with a renal phenotype, this effect on resting cell calcium concentration might be restricted to vascular smooth muscle cells and reflect an activity of polycystin-2 independent of the functions of polycystin-1.
Potential Significance
Based on our data, loss of polycystin-1 would be predicted to lead to prolonged ligand-activated elevation of cell calcium with reciprocally prolonged depletion of ER calcium concentration. In the following sections, we discuss how these effects might influence processes considered necessary for ADPKD pathogenesis.
Fluid secretion. Increased cytoplasmic calcium is a potent stimulus for chloride secretion and inhibition of sodium reabsorption (8). Therefore, one of the functions of polycystin-1 might be to attenuate the net secretory response to ATP. Because the effect of polycystin-1 was evident in cells stimulated with IP3, it is likely that polycystin-1 would similarly attenuate chloride secretion in response to other hormones acting through phospholipase C. If so, then loss of polycystin-1 in the ADPKD cyst epithelium would lead to a prosecretory state through the prolongation of cytoplasmic calcium responses and chloride channel activation following exposure to a number of naturally occurring ligands.
Proliferation. In a recent series of studies, Yamaguchi et al. (38) demonstrated that blockade of plasma membrane calcium channels, or chelation of extracellular calcium, reversed the cAMP growth phenotype to an ADPKD type pattern. Although a decline in cytoplasmic calcium was evident in response to either manipulation, these authors showed that the thapsigargin-releasable calcium store was also depleted. Because ER store depletion has been linked to ERK activation (20), ER calcium depletion might have been a stimulus for the reversal of the cAMP response in the studies of Yamaguchi et al. (38). We have shown that polycystin-1 enhances calcium uptake into the ER and would therefore act to minimize ER calcium depletion in response to a variety of release stimuli. Because, in vivo, renal epithelial cells are continuously exposed to endocrine, paracrine, and autocrine stimulation, the calcium concentration of the ER of cells lacking polycystin-1 might be slower to recover and therefore remain depleted to a greater extent than neighboring unaffected cells. This could promote proliferation through ERK activation in cyst epithelia. Our use of ATP as the calcium release stimulus is pathophysiologically relevant, because all the components for purinergic signal transduction are resident in the cystic environment (29).
Apoptosis. Our observations raise a possible explanation for the increased apoptosis seen in polycystic epithelia in ADPKD (36). As discussed above, calcium released by the ER can be cleared from the cytoplasm by buffering by intracellular organelles such as the mitochondrion (mitochondrial calcium uptake is closely coupled to ER calcium release) (28). There is evidence that mitochondrial calcium loading is an important mechanism for apoptotic cell death (32). The prolonged elevation of cytoplasmic calcium with every calcium release stimulus in ADPKD cyst epithelia could conceivably result in increased apoptosis through progressive accumulation of calcium within mitochondria.
In conclusion, we suggest that it might be important to consider the interplay between ER and cytoplasmic calcium in future studies of the cell biological consequences of loss of polycystin function in ADPKD.
GRANTS
This work was supported by National Institutes of Health (NIH) award KO8-DK-066323–01 and by grants from Satellite Research, the PKD Foundation, and the National Kidney Foundation of Maryland. G. Germino is the Blum Scholar of Johns Hopkins University School of Medicine and is funded by the NIH and the American Heart Association. R. Zeigelstein received support from NIH Grant RO1-HL-63720.
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
Aguiari G, Campanella M, Manzati E, Pinton P, Banzi M, Moretti S, Piva R, Rizzuto R, and del Senno L. Expression of polycystin-1 C-terminal fragment enhances the ATP-induced Ca2+ release in human kidney cells. Biochem Biophys Res Commun 301: 657–664, 2003.
Arnould T, Kim E, Tsiokas L, Jochimsen F, Gruning W, Chang JD, and Walz G. The polycystic kidney disease 1 gene product mediates protein kinase C -dependent and c-Jun n-terminal kinase-dependent activation of the transcription factor AP-1. J Biol Chem 273: 6013–6018, 1998.
Bhunia AK, Piontek K, Boletta A, Liu L, Qian F, Xu PN, Germino FJ, and Germino GG. PKD1 induces p21 (WAF1) and regulation of the cell cycle via direct activation of the JAK-STAT signaling pathway in a process requiring PKD2. Cell 109: 157–168, 2002.
Bird G, Obie J, and Putney J Jr. Functional homogeneity of the non-mitochondrial Ca2+ pool in intact mouse lacrimal acinar cells. J Biol Chem 267: 18382–18386, 1992.
Boletta A, Qian F, Onuchic LF, Bhunia AK, Phakdeekitcharoen B, Hanaoka K, Guggino W, Monaco L, and Germino G. Polycystin-1, the gene product of PKD1, induces resistance to apoptosis and spontaneous tubulogenesis in MDCK Cells. Mol Cell 6: 1267–1273, 2000.
Boletta A, Qian F, Onuchic LF, Bragonzi A, Cortese M, Deen PM, Courtoy PJ, Soria MR, Devuyst O, Monaco L, and Germino GG. Biochemical characterization of bona fide polycystin-1 in vitro and in vivo. Am J Kidney Dis 38: 1421–1429, 2001.
Cuffe JE, Bielfeld-Ackermann A, Thomas J, Leipziger J, and Korbmacher C. ATP stimulates Cl– secretion and reduces amiloride-sensitive Na+ absorption in M-1 mouse cortical collecting duct cells. J Physiol (Lond) 524: 77–90, 2000.
De Luisi A and Hofer AM. Evidence that Ca2+ cycling by the plasma membrane Ca2+-ATPase increases the 'excitability' of the extracellular Ca2+-sensing receptor. J Cell Sci 116: 1527–1538, 2003.
Delmas P, Nomura H, Li X, Lakkis M, Luo Y, Segal Y, Fernandez-Fernandez JM, Harris P, Frischauf AM, Brown DA, and Zhou J. Constitutive activation of G-proteins by polycystin-1 is antagonized by polycystin-2. J Biol Chem 277: 11276–11283, 2002.
The European Polycystic Kidney Disease Consortium. The polycystic kidney disease 1 gene encodes a 14 kb transcript and lies within a duplicated region on chromosome 16. Cell 77: 881–894, 1994.
Fowler MR and Hunter M. Mitochondrial Ca2+ transport in frog early distal tubule. Exp Physiol 90: 195–201, 2004.
Fulceri R, Bellomo G, Mirabelli F, Gamberucci A, and Benedetti A. Measurement of mitochondrial and non-mitochondrial Ca2+ in isolated intact hepatocytes: a critical re-evaluation of the use of mitochondrial inhibitors. Cell Calcium 12: 431–439, 1991.
Gonzalez-Perrett S, Kim K, Ibarra C, Damiano AE, Zotta E, Batelli M, Harris PC, Reisin IL, Arnaout MA, and Cantiello HF. Polycystin-2, the protein mutated in autosomal dominant polycystic kidney disease (ADPKD), is a Ca2+-permeable nonselective cation channel. PNAS 98: 1182–1187, 2001.
Hanaoka K, Qian F, Boletta A, Bhunia AK, Piontek K, Tsiokas L, Sukhatme VP, Guggino WB, and Germino GG. Co-assembly of polycystin-1 and -2 produces unique cation-permeable currents. Nature 408: 990–994, 2000.
Hofer AM, Fasolato C, and Pozzan T. Capacitative Ca2+ entry is closely linked to the filling state of internal Ca2+ stores: a study using simultaneous measurements of ICRAC and intraluminal [Ca2+]. J Cell Biol 140: 325–334, 1998.
Hooper K, Unwin R, and Sutters M. The isolated C-terminus of polycystin-1 promotes increased ATP-stimulated chloride secretion in a collecting duct cell line. Clin Sci (Colch) 104: 217–221, 2003.
Huan Y and van Adelsberg J. Polycystin-1, the PKD1 gene product, is in a complex containing E-cadherin and the catenins. J Clin Invest 104: 1459–1468, 1999.
Kim E, Arnould T, Sellin LK, Benzing T, Fan MJ, Gruning W, Sokol SY, Drummond I, and Walz G. The polycystic kidney disease 1 gene product modulates Wnt signaling. J Biol Chem 274: 4947–4953, 1999.
Koulen P, Cai Y, Geng L, Maeda Y, Nishimura S, Witzgall R, Ehrlich BE, and Somlo S. Polycystin-2 is an intracellular calcium release channel. Nat Cell Biol 4: 191–197, 2002.
Maloney JA, Tsygankova OM, Yang L, Li Q, Szot A, Baysal K, and Williamson JR. Activation of ERK by Ca2+ store depletion in rat liver epithelial cells. Am J Physiol Cell Physiol 276: C221–C230, 1999.
Missiaen L, Declerck I, Droogmans G, Plessers L, De Smedt H, Raeymaekers L, and Casteels R. Agonist-dependent Ca2+ and Mn2+ entry dependent on state of filling of Ca2+ stores in aortic smooth muscle cells of the rat. J Physiol 427: 171–186, 1990.
Mochizuki T, Wu G, Hayashi T, Xenophontos S, Veldhuisen B, Saris JJ, Reynolds DM, Cai Y, Gabow PA, Pierides A, Kimberling WJ, Breuning MH, Deltas CC, Peters DJ, and Somlo S. PKD2, a gene for polycystic kidney disease that encodes an integral membrane protein. Science 272: 1339–1342, 1996.
Nauli S, Alenghat F, YL, Williams E, Vassilev P, Li X, Elia A, Lu W, Brown E, Quinn S, Ingber D, and Zhou J. Polycystins 1 and 2 mediate mechanosensation in the primary cilium of kidney cells. Nat Genet 33: 129–137, 2003.
Nickel C, Benzing T, Sellin L, Gerke P, Karihaloo A, Liu ZX, Cantley LG, and Walz G. The polycystin-1 C-terminal fragment triggers branching morphogenesis and migration of tubular kidney epithelial cells. J Clin Invest 109: 481–489, 2002.
Parnell SC, Magenheimer BS, Maser RL, Rankin CA, Smine A, Okamoto T, and Calvet JP. The polycystic kidney disease-1 protein, polycystin-1, binds and activates heterotrimeric G-proteins in vitro. Biochem Biophys Res Commun 251: 625–631, 1998.
Putney JW Jr, Broad LM, Braun FJ, Lievremont JP, and Bird GSJ. Mechanisms of capacitative calcium entry. J Cell Sci 114: 2223–2229, 2001.
Qian Q, Hunter LW, Li M, Marin-Padilla M, Prakash YS, Somlo S, Harris PC, Torres VE, and Sieck GC. Pkd2 haplo insufficiency alters intracellular calcium regulation in vascular smooth muscle cells. Hum Mol Genet 12: 1875–1880, 2003.
Rizzuto R, Brini M, Murgia M, and Pozzan T. Microdomains with high Ca2+ close to IP3-sensitive channels that are sensed by neighboring mitochondria. Science 262: 744–747, 1993.
Schwiebert EM, Wallace DP, Braunstein GM, King SR, Peti-Peterdi J, Hanaoka K, Guggino WB, Guay-Woodford LM, Bell PD, Sullivan LP, Grantham JJ, and Taylor AL. Autocrine extracellular purinergic signaling in epithelial cells derived from polycystic kidneys. Am J Physiol Renal Physiol 282: F763–F775, 2002.
Somlo S and Ehrlich B. Human disease: calcium signaling in polycystic kidney disease. Curr Biol 11: R356–360, 2001.
Sutters M, Yamaguchi T, Maser RL, Magenheimer BS, St. John PL, Abrahamson DR, Grantham JJ, and Calvet JP. Polycystin-1 transforms the cAMP growth-responsive phenotype of M-1 cells. Kidney Int 60: 484–494, 2001.
Szabadkai G and Rizzuto R. Participation of endoplasmic reticulum and mitochondrial calcium handling in apoptosis: more than just neighborhood FEBS Lett 567: 111–115, 2004.
Vandorpe DH, Chernova MN, Jiang L, Sellin LK, Wilhelm S, Stuart-Tilley AK, Walz G, and Alper SL. The cytoplasmic C-terminal fragment of polycystin-1 regulates a Ca2+-permeable cation channel. J Biol Chem 276: 4093–4101, 2001.
Wildman SS, Hooper KM, Turner CM, Sham JSK, Lakatta EG, King BF, Unwin RJ, and Sutters M. The isolated polycystin-1 cytoplasmic COOH terminus prolongs ATP-stimulated Cl– conductance through increased Ca2+ entry. Am J Physiol Renal Physiol 285: F1168–F1178, 2003.
Wilson PD, Geng L, Li X, and Burrow CR. The PKD1 gene product, "polycystin-1," is a tyrosine-phosphorylated protein that colocalizes with 21-integrin in focal clusters in adherent renal epithelia. Lab Invest 79: 1311–1323, 1999.
Woo D. Apoptosis and loss of renal tissue in polycystic kidney diseases. N Engl J Med 333: 18–25, 1995.
Xu GM, Gonzalez-Perrett S, Essafi M, Timpanaro GA, Montalbetti N, Arnaout MA, and Cantiello HF. Polycystin-1 activates and stabilizes the polycystin-2 channel. J Biol Chem 278: 1457–1462, 2003.
Yamaguchi T, Wallace DP, Magenheimer BS, Hempson SJ, Grantham JJ, and Calvet JP. Calcium restriction allows cAMP activation of the B-Raf/ERK pathway, switching cells to a cAMP-dependent growth-stimulated phenotype. J Biol Chem 279: 40419–40430, 2004.
Yoder BK, Hou X, and Guay-Woodford LM. The polycystic kidney disease proteins, polycystin-1, polycystin-2, polaris, and cystin, are co-localized in renal cilia. J Am Soc Nephrol 13: 2508–2516, 2002.(K. M. Hooper, A. Boletta,)