Permeation and Gating Properties of the Novel Epithelial Ca2+ Channel*

The recently cloned epithelial Ca2+ channel (ECaC) constitutes the Ca2+influx pathway in 1,25-dihydroxyvitamin D3-responsive epithelia. We have combined patch-clamp analysis and fura-2 fluorescence microscopy to functionally characterize ECaC heterologously expressed in HEK293 cells. The intracellular Ca2+ concentration in ECaC-expressing cells was closely correlated with the applied electrochemical Ca2+ gradient, demonstrating the distinctive Ca2+ permeability and constitutive activation of ECaC. Cells dialyzed with 10 mm1,2-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid displayed large inward currents through ECaC in response to voltage ramps. The corresponding current-voltage relationship showed pronounced inward rectification. Currents evoked by voltage steps to potentials below −40 mV partially inactivated with a biexponential time course. This inactivation was less pronounced if Ba2+or Sr2+ replaced Ca2+ and was absent in Ca2+-free solutions. ECaC showed an anomalous mole fraction behavior. The permeability ratioP Ca:P Na calculated from the reversal potential at 30 mm[Ca2+] o was larger than 100. The divalent cation selectivity profile is Ca2+ > Mn2+ > Ba2+ ∼ Sr2+. Repetitive stimulation of ECaC-expressing cells induced a decay of the current response, which was greatly reduced if Ca2+ was replaced by Ba2+ and was virtually abolished if [Ca2+] o was lowered to 1 nm. In conclusion, ECaC is a Ca2+ selective channel, exhibiting Ca2+-dependent autoregulatory mechanisms, including fast inactivation and slow down-regulation.

the proximal small intestine and the distal part of the nephron. This process of transcellular Ca 2ϩ transport is a three-step operation consisting of passive apical Ca 2ϩ entry followed by cytosolic diffusion facilitated by calbindins and active extrusion across the basolateral membrane by a high affinity Ca 2ϩ -ATPase and/or a Na ϩ -Ca 2ϩ exchanger (2). The apical influx of Ca 2ϩ is the rate-limiting step in this process and, therefore, a prime regulatory target for stimulatory and inhibitory hormones. The epithelial Ca 2ϩ channel (ECaC), recently cloned from rabbit kidney, is the candidate channel for being the gatekeeper of this apical Ca 2ϩ influx mechanism (3).
ECaC is exclusively present in 1,25(OH) 2 D 3 -responsive epithelia, including intestine, kidney, and placenta, and is structurally related to the family of transient receptor potential channels, capsaicin receptors, and the growth factor-regulated channel (4 -6). These Ca 2ϩ -permeable cation channels contain six putative transmembrane domains, including a pore-forming region, but share only 30% homology that is mainly restricted to the pore-forming region and flanking transmembrane segments. ECaC contains putative phosphorylation sites for protein kinase C, cAMP-dependent and cGMP-dependent protein kinase, calcium-calmodulin-dependent protein kinase, and structural domains, such as N-linked glycosylation sites and ankyrin repeats (3).
ECaC-expressing Xenopus laevis oocytes mediate a saturable Ca 2ϩ uptake determined with tracer studies and a hyperpolarization-stimulated Ca 2ϩ influx measured indirectly as Ca 2ϩinduced Cl Ϫ current (3,7). These initial studies indicate that ECaC exhibits a distinct Ca 2ϩ permeability. A similar conclusion was reached for a Ca 2ϩ transporter (calcium transporter 1) cloned from rat intestine, which shares with ECaC a structural similarity of 75% and several basic functional properties (8).
The identification of ECaC offers for the first time a realistic approach to study the functional and regulatory aspects of this Ca 2ϩ influx pathway. Knowledge of ECaC functioning is of vital importance to understand the Ca 2ϩ handling by Ca 2ϩ absorbing epithelia and will, in particular, provide a molecular basis for achieving a better understanding of Ca 2ϩ malabsorption/malreabsorption. The aim of the present study is to functionally characterize ECaC by a combined whole-cell patchclamp analysis and fura-2 fluorescence microscopy using HEK293 cells heterologously expressing ECaC.

Vector Construction for ECaC-Green Fluorescent Protein (GFP) Coexpression-
The entire open reading frame from rbECaC was cloned as a PvuII-BamHI fragment in the pCINeo/IRES-GFP vector (3,9). This bicistronic expression vector pCINeo/IRES-GFP/rbECaC was used to co-express rbECaC and enhanced GFP.
Cell Culture and Transfection-Human embryonic kidney cells, HEK293, were grown in Dulbecco's modified Eagle's medium containing 10% (v/v) human serum, 2 mM L-glutamine, 2 units/ml penicillin, and 2 mg/ml streptomycin at 37°C in a humidity-controlled incubator with 10% CO 2 . HEK293 cells were transiently transfected with the pCINeo/IRES-GFP/rbECaC vector using methods described previously (10). Transfected cells were visually identified in the patch-clamp set up. GFP was excited at a wavelength between 450 and 490 nm, and the emitted light was passed through a 520 nm long-pass filter. The ECaCexpressing cells were identified by their green fluorescence, and GFPnegative cells from the same batch were used as controls.
Electrophysiology-Electrophysiological methods and Ca 2ϩ measurements have been described in detail previously (11). Whole-cell currents were measured with an EPC-9 (HEKA Elektronik, Lambrecht, Germany; sampling rate, 1 ms; 8-pole Bessel filter, 2.9 kHz) or an L/M-EPC-7 (List Elektronics, Darmstadt, Germany) using ruptured patches. Electrode resistances were between 2 and 5 M⍀, and capacitance and access resistances were monitored continuously. The ramp protocol consisted of linear voltage ramps changing from Ϫ100 or Ϫ150 mV to ϩ100 mV within 400 ms, applied every 5 s. The step protocol consisted of a series of 400-ms-long voltage steps applied from a holding potential of ϩ20 mV to voltages between Ϫ100 and ϩ100 mV, with an increment of 25 mV. The current density was calculated from the size of net current at Ϫ80 mV during the ramp protocol. The permeability of Cs ϩ , the only other permeable monovalent cation besides Na ϩ in pipette and bath solution, was measured relative to that of Na ϩ from the reversal potential in the absence of extracellular Ca 2ϩ ( Permeability of the divalent cations Ca 2ϩ , Ba 2ϩ , Sr 2ϩ , and Mg 2ϩ rela-tive to Na ϩ was calculated from the reversal potential measured with 30 mM of the respective cation in the extracellular solution. ] i are the extra-and intracellular concentrations for Na ϩ and Cs ϩ , respectively; and V rev is the reversal potential (10). Ca 2ϩ Measurements-ECaC-expressing cells were loaded with fura-2 via the patch pipette and excited alternately at wavelengths of 360 and 390 nm through a filter wheel rotating at 2 cycles/s. The fluorescence emitted at each excitation wavelength was measured at 510 nm using a photomultiplier. Autofluorescence was subtracted. Apparent free [Ca 2ϩ ] was calculated from the fluorescence ratio R by [Ca 2ϩ ] i ϭ K eff (R Ϫ R 0 )/(R 1 Ϫ R), where K eff is the effective binding constant, R 0 the fluorescence ratio at zero Ca 2ϩ , and R 1 that at high Ca 2ϩ . These calibration constants were determined experimentally for the given set-up and the actual experimental conditions used.
Cells were kept in a nominally Ca 2ϩ -free medium to prevent Ca 2ϩ overload and exposed for a maximum of 5 min to a Krebs solution containing 1.5 mM Ca 2ϩ before sealing the patch pipette to the cell. All experiments were performed at room temperature (20 -22°C).
Statistical Analysis-In all experiments, the data are expressed as the mean Ϯ S.E. Overall statistical significance was determined by analysis of variance. In the case of significance (p Ͻ 0.01), individual groups were compared by Student's t test. Experimental data were fitted to multiple exponentials using the fitting routine of the ASCD program (G. Droogmans).

RESULTS
Evidence that ECaC forms a constitutively active Ca 2ϩ entry pathway was obtained from experiments showing a close correlation between the level of intracellular calcium ([Ca 2ϩ ] i ) and the electrochemical Ca 2ϩ gradient in ECaC-expressing HEK293 cell (Fig. 1). The driving force for Ca 2ϩ entry was therefore modified either by changing extracellular ([Ca 2ϩ ] o , Fig. 1A) or the membrane potential (V M ) (Fig. 1B). In Fig. 1 ] o from 0 to 30 mM in cells clamped at ϩ20 mV markedly increases [Ca 2ϩ ] i in ECaC-expressing cells, whereas the effect was much less pronounced in nontransfected cells (Fig. 1A). In another experiment, we varied the membrane potential at a constant [Ca 2ϩ ] o of 1.5 mM. Depolarization from ϩ20 to ϩ60 mV decreased [Ca 2ϩ ] i , whereas hyperpolarization from ϩ20 to Ϫ100 mV increased [Ca 2ϩ ] i (Fig. 1B). The pooled data from several cells are summarized in the bottom panels of Fig. 1 The characteristics of the Ca 2ϩ entry pathway in ECaCexpressing cells were further investigated in electrophysiological whole-cell experiments in cells dialyzed with 10 mM BAPTA and using the described ramp and step protocols. Under these circumstances, voltage ramps induced large inward currents in ECaC-expressing HEK293 cells with amplitudes that were obviously enhanced by increasing [Ca 2ϩ ] o (Fig. 2B). Nontransfected HEK293 cells exhibited under these experimental conditions only a small background current, which was in contrast with ECaC-expressing cells inhibited by increasing [Ca 2ϩ ] o ( Fig. 2A). Because of the strong inward rectification, it was difficult to obtain reliable values of the reversal potential. Moreover, the reversal potential of ϩ54 Ϯ 4 mV (range, 30 -77; n ϭ 13) at 30 mM [Ca 2ϩ ] o is probably an underestimation due to the presence of background currents. Currents in response to voltage-steps to potentials more negative than Ϫ40 mV showed time-dependent inactivation, whereas currents at more positive potentials did not. Inactivation was manifest at 30 mM [Ca 2ϩ ] o but was also present at 1 mM (Fig. 2, C and D (Fig. 3, A and B) but a single slow component if extracellular Ca 2ϩ was replaced with Ba 2ϩ or Sr 2ϩ (Fig. 3, C and D).
To corroborate our insight in the mechanism of permeation through ECaC, we have measured current densities at Ϫ80 mV as a function of [Ca 2ϩ ] o in ECaC-expressing cells, as shown in  ] o dependence of the ECaC-specific current density, calculated as the difference current in ECaC-expressing and control cells, shows the typical anomalous mole fraction behavior known for highly Ca 2ϩ -selective channels (Fig. 4B). Note also that the current density in control cells in contrast with ECaC-expressing cells diminished with increasing [Ca 2ϩ ] o (Fig. 4A).
We have also compared the current densities at Ϫ80 mV in ECaC-expressing cells at different [Ca 2ϩ ] i , i.e. in cells buffered with 10 mM BAPTA at extremely low values or at 1 M (Fig.  4C). It is obvious that an increase in [Ca 2ϩ ] i down-regulates the current.
Current densities and inactivation pattern of ECaC was different for the various divalents (Fig. 5, A and B, and Fig. 3). The conductance sequence for divalent cations of the ECaC channel derived from the current densities at Ϫ80 mV is Ca 2ϩ Ͼ Ba 2ϩ Ϸ Sr 2ϩ Ͼ Mn 2ϩ (Fig. 5C); the values for Sr 2ϩ and Ba 2ϩ are not statistically different (p Ͼ 0.01). The permeability ratio P Ca :P Na , determined from the reversal potentials at 30 mM [Ca 2ϩ ] o and in Ca 2ϩ -free solution according to Equations 1 and 2, was 107 Ϯ 32 (range, 7-420; n ϭ 13). This value probably underestimates the real ratio because of the mentioned under-estimation of the reversal potential. The permeabilities of Mn 2ϩ , Ba 2ϩ , and Sr 2ϩ relative to that of Na ϩ were 20, 5.6, and 4, respectively. ECaC is thus despite the smaller current densities observed with Mn 2ϩ as the charge carrier, more permeable for Mn 2ϩ than for Ba 2ϩ and Sr 2ϩ . (Fig. 5D). In 1 nM [Ca 2ϩ ] o , we calculated a P Cs :P Na of 0.84, indicating that ECaC is more permeable for Na ϩ than for Cs ϩ .
Another feature of the current through ECaC is its decay during repetitive stimulations (Fig. 6), with voltage ramps from Ϫ100 to ϩ100 mV applied with an interval of 5 s from a holding potential of ϩ20 mV. In the presence of 30 mM [Ca 2ϩ ] o , the current practically disappeared after 8 ramps (Fig. 6A). In contrast, more sweeps were necessary to down-regulate the current when Ba 2ϩ was the charge carrier (Fig. 6B). Decreasing [Ca 2ϩ ] o to 1 nM virtually abolished the rundown phenomenon (Fig. 6C). The time constants of half-maximal decay of the ECaC currents are summarized in Fig. 6D, showing the dramatic lowering of ECaC currents when Ca 2ϩ is the charge carrier. This finding indicates that the decay process may be associated with a Ca 2ϩ -dependent mechanism. DISCUSSION The present study describes functional characteristics of the recently cloned epithelial Ca 2ϩ channel, ECaC, that are consistent with its putative role as apical Ca 2ϩ entry channel that mediates transcellular Ca 2ϩ transport in 1,25(OH) 2 D 3 -respon-  (n ϭ 3). C, current steps and inactivation time constant versus voltage plotted for Ba 2ϩ being the charge carrier through ECaC. Inactivation was delayed compared with Ca 2ϩ (n ϭ 3). D, when Sr 2ϩ was substituted for Ca 2ϩ , currents through ECaC were still inactivated. Decay was faster than in Ba 2ϩ but slower than Ca 2ϩ (n ϭ 7). sive epithelia. These distinctive properties include a constitutively activated Ca 2ϩ permeability at physiological membrane potentials, a high calcium selectivity, Ca 2ϩ -dependent feedback regulation of channel activity, anomalous mole-fraction behavior, and rundown of channel activity.
Transient expression of ECaC dramatically increased the Ca 2ϩ permeability of HEK293 cells, which was reflected by an elevated cytosolic Ca 2ϩ level proportional to the electrochemical driving force for Ca 2ϩ . At physiological membrane potentials, ECaC is apparently constitutively active in the absence of any known stimulus. Extrapolating these findings to the in vivo situation suggests that additional regulatory factors or at least an increased Ca 2ϩ buffer capacity are a necessity for ECaC-expressing cells to maintain constant low Ca 2ϩ levels during transcellular Ca 2ϩ transport. The 1,25(OH) 2 D 3 -dependent calcium-binding proteins, named calbindins, which are abundantly present in the cytosol of ECaC-expressing epithelia, could fulfil such a function (2,12,13). Indeed, calbindins are known to facilitate the cytosolic diffusion of Ca 2ϩ from the apical membrane to the basolateral extrusion mechanisms, while maintaining low cytosolic Ca 2ϩ levels (14).
ECaC-expressing HEK293 cells displayed large inward currents in the presence of BAPTA in the pipette solution. These currents displayed high Ca 2ϩ dependence and very positive reversal potentials. These findings are indicative of a Ca 2ϩselective current. The current-voltage relationship shows prominent inward rectification. Thus, under unstimulated physiological conditions when the membrane potential is typically around Ϫ70 mV in the distal nephron (13), ECaC has a substantial Ca 2ϩ conductance permitting basal Ca 2ϩ influx in the epithelial cell. This influx would be enhanced when the driving force is increased by reduction of [Ca 2ϩ ] i or hyperpolarization of the cells. Indeed, PTH has been shown to hyperpolarize the membrane during its stimulatory action and thereby promotes Ca 2ϩ influx in the renal cell. Likewise, thia-zide diuretics-induced impairment of NaCl reabsorption hyperpolarizes the plasma membrane (15) and thus facilitates additional Ca 2ϩ entry through ECaC. This could explain the calcium-sparing effect of this diuretic.
Detailed knowledge about the mechanism of permeation through ECaC was obtained from the Ca 2ϩ current response to hyperpolarizing voltage steps. These responses are characterized by a rapid but incomplete inactivation, which consists of a fast and a slow component. The fast component was eliminated when Ba 2ϩ or Sr 2ϩ was used as a charge carrier. This latter finding suggests that ECaC activity is controlled by a Ca 2ϩ -dependent feedback mechanism. It is generally known that Ca 2ϩ ions that enter through Ca 2ϩ channels, including voltage-dependent Ca 2ϩ channels and transient receptor potential channels, exert a negative effect on channel activity (16). Recently, several independent studies have reached the conclusion that in the case of voltage-activated Ca 2ϩ channels, direct binding of calmodulin, which is ubiquitously expressed as Ca 2ϩ sensor inside cells, plays a critical role in this Ca 2ϩ -dependent inactivation process. (17)(18)(19)(20). So far, we do not have any indication that a similar mechanism is responsible for the inactivation of ECaC. The IQ motif, identified in the C-tail of these voltageoperated channels is not present in ECaC. It might be possible, however, that a still unidentified calmodulin-binding motif is present in ECaC or that a different feedback mechanism is operative.
An additional feature of the gating mechanism is that the current response slowly disappears during successive voltage ramps, generally referred to as rundown or decay. The rundown was significantly diminished when Ca 2ϩ was replaced by Ba 2ϩ as charge carrier and abrogated when extracellular Ca 2ϩ was lowered to 1 nM, indicating, as suggested for the inactivation process, that a Ca 2ϩ -operated process inhibits ECaC activity during repetitive stimulation. This form of regulation could have a significant impact on the amount of Ca 2ϩ that enters the cell during repetitive activation. As postulated for other ion channels, it is possible that this rundown involves phosphorylation and/or dephosphorylation of the channel or associated proteins (16,21). In this respect, it is tempting to explore the role of a calcium-calmodulin-dependent protein kinase II for which functionally conserved consensus sites are present in ECaC (Ser-142 and Ser-693) (3,8), because it has recently been demonstrated that this kinase is involved in the Ca 2ϩ -dependent regulation of channel activity (22,23). Alternatively, calbindin-D 28K has been implicated in the regulation of the rundown process of N-methyl-D-glucamine receptor channel activity, possibly through buffering local Ca 2ϩ elevations and thereby preventing calcium-induced depolymerization of the actin cytoskeleton (24). Together with the striking co-localization of ECaC and calbindin-D in Ca 2ϩ -transporting cells, a calbindin-mediated rundown process represents a possible mechanism that could adjust the amount of Ca 2ϩ that enters the cell during repetitive activation.
The present data unambiguously demonstrate that ECaC possesses a high selectivity for Ca 2ϩ , illustrated by P Ca :P Na values of more than 100. This implies that under physiological conditions when Ca 2ϩ ions present in the pro-urine or intestinal fluid are largely outnumbered by Na ϩ ions, ECaC can efficiently discriminate between Ca 2ϩ and other cations, thereby specifically controlling the Ca 2ϩ permeability of the apical membrane of Ca 2ϩ -transporting epithelial cells. This restriction toward monovalent cations was eliminated in the absence of extracellular Ca 2ϩ ions, which can be explained by the observed anomalous mole-fraction behavior of ECaC, indicating multiple ion binding sites in ECaC (16).
The cation selectivity of ECaC is distinguishably different  Figs. 3 and 4. B, IV curves obtained from voltage ramps for different divalent cations (ramps from Ϫ100 to ϩ100 mV, holding potential ϩ20 mV). C, current densities for divalent cations. Concentration is always 30 mM. The number of cells is indicated near the error bars. Currents were measured from voltage ramps at Ϫ80 mV. D, permeation of mono-and divalent cations through ECaC. The relative permeation to Na ϩ was calculated according to Equations 1 and 2. The P Cs /P Na permeation ratio was used to calculate permeation of the bivalents by Equation 2. See under "Experimental Procedures" for details.
FIG. 6. Current through ECaC induced by repetitive stimulation exhibits rundown. A, rundown of the current through ECaC in the presence of 30 mM Ca 2ϩ . Voltage ramps were applied every 5 s (ramps of 400 ms, from Ϫ100 to ϩ100 mV; holding potential, ϩ20 mV). B, if Ba 2ϩ (30 mM) is the charge carrier, rundown is strikingly delayed. The interval between the ramps is 5 s (same protocol as in A). C, time course of the current rundown from three single cells. D, pooled data representing the time to half-maximal decay from all cells. Note that in 1 nM [Ca 2ϩ ] o , the estimated time for rundown is longer than 4 min. Current values were normalized to the maximal current value in the specific condition (I norm ). from that observed for homologous Ca 2ϩ channels, including members of the Trp family and the ligand-operated vanilloid receptor and growth factor-regulated channels, and is exemplified by an eminent selectivity for Ca 2ϩ over monovalent cations and a permeability sequence for divalent cations of Ca 2ϩ Ͼ Ba 2ϩ ϳ Sr 2ϩ Ͼ Mn 2ϩ . Furthermore, the current-voltage relationship of vanilloid receptor-like 1, vanilloid receptor 1, and growth factor-regulated channels reveals dominant outward rectification of the corresponding current, allowing us to discriminate these channels from the ECaC current (2)(3)(4)(5)(6)25). ECaC, however, resembles several features typical for storeoperated Ca 2ϩ entry. The Ca 2ϩ -release activated Ca 2ϩ current (I CRAC ) is remarkably selective for Ca 2ϩ , with a divalent cation selectivity profile similar to that of ECaC, and gives rise to prominent inward rectification at negative voltages (26,27). Despite these similarities, it could not be demonstrated that ECaC is activated by store depletion (data not shown), indicating that ECaC does not encode I CRAC . However, it remains intriguing to consider channels homologous to ECaC as candidates for store-operated Ca 2ϩ entry.
A homologue of ECaC was recently cloned from rat duodenum and named calcium transporter 1 (8). The main part of this protein is highly identical (Ͼ90%) to ECaC, whereas the last part of the C-tail is structurally different. Based on its macroscopic kinetic properties, it was suggested that this protein represents a unique transition between a channel and a transporter. The current paper, however, clearly demonstrates that ECaC exhibits defining characteristics typical for a Ca 2ϩ channel, including a positive reversal potential and anomalous mole fraction behavior.
The present study has established the basic kinetic aspects of ECaC, which are important to understanding the control of apical Ca 2ϩ entry in Ca 2ϩ -transporting epithelial cells. It has been shown previously that this process of transcellular Ca 2ϩ transport is under the regulation of multiple signaling pathways involving protein kinase C and cAMP-and cGMP-dependent kinase (13,28,29). Together with the observation that relevant consensus sites for these kinases are present in ECaC (3), these findings warrant further investigations to unravel the hormone-activated mechanisms regulating the kinetic properties of ECaC.