80K-H as a New Ca2+ Sensor Regulating the Activity of the Epithelial Ca2+ Channel Transient Receptor Potential Cation Channel V5 (TRPV5)*

The epithelial Ca2+ channel transient receptor potential cation channel V5 (TRPV5) constitutes the apical Ca2+ entry pathway in the process of active Ca2+ reabsorption. Ca2+ influx through TRPV5 is tightly controlled by modulators of Ca2+ homeostasis, including 1,25-dihydroxyvitamin D3 and dietary Ca2+. However, little is known about intracellular proteins that interact with TRPV5 and directly regulate the activation of this channel. By the use of cDNA microarrays, the present study identified 80K-H as the first protein involved in the Ca2+-dependent control of the epithelial Ca2+ channel TRPV5. 80K-H was initially identified as a protein kinase C substrate, but its biological function remains to be established. We demonstrated a specific interaction between 80K-H and TRPV5, co-localization of both proteins in the kidney, and similar transcriptional regulation by 1,25-dihydroxyvitamin D3 and dietary Ca2+. Furthermore, 80K-H directly bound Ca2+, and inactivation of its two EF-hand structures totally abolished Ca2+ binding. Electrophysiological studies using 80K-H mutants showed that three domains of 80K-H (the two EF-hand structures, the highly acidic glutamic stretch, and the His-Asp-Glu-Leu sequence) are critical determinants for TRPV5 activity. Importantly, inactivation of the EF-hand pair reduced the TRPV5-mediated Ca2+ current and increased the TRPV5 sensitivity to intracellular Ca2+, accelerating the feedback inhibition of the channel. None of the 80K-H mutants altered the TRPV5 plasma membrane localization nor the association of 80K-H with TRPV5, suggesting that 80K-H has a direct effect on TRPV5 activity. In conclusion, we report a novel function for 80K-H as a Ca2+ sensor controlling TRPV5 channel activity.

The maintenance of the body Ca 2ϩ balance is of crucial importance for many vital physiological functions, including neuronal excitability, muscle contraction, and bone formation.
The recent identification of two novel members of the transient receptor potential (TRP) 1 superfamily, TRPV5 and TRPV6, has provided new insight into the molecular mechanisms controlling the extracellular Ca 2ϩ balance (1)(2)(3). TRPV5 and TRPV6 are Ca 2ϩ -selective channels involved in transcellular Ca 2ϩ transport across vitamin D 3 -sensitive epithelia, such as kidney, intestine, and placenta (4,5). Furthermore, these two channels possess unique functional characteristics among the highly heterogeneous group of TRP channels (6). They exhibit a constitutively activated Ca 2ϩ permeability, a high Ca 2ϩ selectivity, and a Ca 2ϩ -dependent feedback mechanism regulating channel activity (7,8).
TRPV5 activity is controlled by various regulatory mechanisms ranging from long-term transcriptional modulations to short-term direct physical interactions with intracellular factors. TRPV5 is regulated at the transcriptional level by 1,25dihydroxyvitamin D 3 (1,25-(OH) 2 D 3 ), dietary Ca 2ϩ , and 17␤estradiol (5,9,10). Previous studies have shown that these hormones and dietary Ca 2ϩ normalize the hypocalcemia in 25-hydroxyvitamin D 3 -1␣-hydroxylase knock-out mice, which is accompanied by an up-regulation of TRPV5 mRNA expression levels. Furthermore, TRPV5 modulation occurs at the post-translational level by controlling translocation of the channel from intracellular pools to the plasma membrane. Recently, the S100A10-annexin 2 complex was demonstrated to modulate the functional plasma membrane distribution of TRPV5 by direct association to the carboxyl (C)-terminal tail of the channel. Down-regulation of annexin 2 inhibited TRPV5mediated currents in TRPV5-transfected cells (11). Finally, the terminal tails of the channel contain potential regulatory motifs, such as protein kinase C (PKC) phosphorylation sites, PDZ motifs, and ankyrin repeats (4). This suggests a regulatory role for the intracellular tails of TRPV5 in the (in)activation or the trafficking of the channel. In line with this observation, Niemeyer et al. (13) demonstrated that TRPV6 is competitively regulated by PKC and calmodulin.
The aim of the present study was to identify novel regulators of Ca 2ϩ reabsorption via TRPV5. To this end, cDNA microarray experiments were performed to identify candidates that exhibit a similar transcriptional regulation to well known proteins involved in Ca 2ϩ transport. Analysis of the data pointed to a PKC substrate, 80K-H, of which the biological function remains unclear. The putative role of this protein in active Ca 2ϩ reabsorption was investigated by immunohistochemical, biochemical, and electrophysiological analysis of 80K-H interacting with TRPV5.

EXPERIMENTAL PROCEDURES
DNA Constructs and cRNA Synthesis-TRPV5 constructs were generated as described previously (11). The coding sequence of 80K-H wild-type protein was amplified from total mouse kidney cDNA material, cloned into the pCINeo/IRES-GFP vector (14) as an EcoRI-NheI fragment, and subsequently subcloned into the pGEX6p-2 vector (Amersham Pharmacia Biotech AB, Uppsala, Sweden) and the pT7Ts vector (15). Using in vitro mutagenesis (QuickChange site-directed mutagenesis kit, Stratagene, La Jolla, CA), three mutants of 80K-H were generated in the pCINeo/IRES GFP vector: one mutant was generated by alanine substitution of the first two aspartates of each EF-hand structure (80K-H EF ), the second mutant was generated by deletion of a stretch containing 21 glutamates (80K-H ⌬Glu ) and the third mutant was generated by deletion of the C-terminal His-Asp-Glu-Leu (HDEL) sequence (80K-H ⌬HDEL ). The obtained mutants were subsequently subcloned into the pT7Ts and pGEX6p-2 vectors. All constructs were verified by sequence analysis. pTLN and pT7Ts constructs were linearized, and cRNA was synthesized in vitro using SP6 and T7 RNA polymerase, respectively (16).
RT-PCR Analysis-Total RNA from mouse brain, eye, salivary tissue, thymus, heart, lung, liver, spleen, pancreas, stomach, duodenum, colon, kidney, prostate, testis, uterus, skin, skeletal muscle, and bone was isolated using TRIzol (GIBCO/BRL, Life Technologies, Breda, The Netherlands). Total RNA (2 g) was subjected to reverse transcription using Moloney murine leukemia virus reverse transcriptase, and a PCR for 80K-H was performed in a species-conserved region using the primers 5Ј-TAC GTC TAC CGG CTT TGC C-3Ј and 5Ј-AGG TAC TCA CAG CGA CTG GG-3Ј.
Glutathione S-transferase (GST)-TRPV5 Fusion Protein and Interaction Assays-pGEX6p-2 constructs were transformed in Escherichia coli BL21 and GST-fusion proteins were expressed and purified according to the manufacturer's protocol (Amersham Pharmacia Biotech AB). [ 35 S]Methionine-labeled 80K-H, 80K-H EF , and 80K-H ⌬Glu were prepared using a reticulocyte lysate and added to GST or GST-TRPV5 fusion proteins immobilized on glutathione-Sepharose 4B beads (Amersham Pharmacia Biotech AB) in PBS containing 1% (v/v) Triton X-100. After 2-h incubation at room temperature, the beads were washed extensively, and bound proteins were eluted with SDS-PAGE loading buffer, separated on 10% (w/v) SDS-PAGE gels, and visualized by autoradiography.

45
Ca 2ϩ -binding Assay-To study the direct Ca 2ϩ -binding properties of 80K-H, GST-fused 80K-H proteins migrated on SDS-12% PAGE gel and were blotted to polyvinylidene difluoride membrane. Blotted proteins were visualized with Ponceau S staining (Sigma). 45 Ca 2ϩ binding was performed as described by Maruyama et al. (20). Briefly, polyvinylidene difluoride membranes were thoroughly washed and incubated for 10 min with 1 Ci/ml 45 CaCl 2 (PerkinElmer Life Sciences). After washing, the membrane was dried and 45 Ca 2ϩ binding was visualized by autoradiography.
Electrophysiology-The full-length cDNAs encoding TRPV5 and 80K-H mutants were transiently transfected in HEK293, and patchclamp experiments were performed in the whole-cell configuration, as described previously (11). Monovalent cation currents were measured in nominal divalent-free (DVF) solutions, in which Ca 2ϩ and Mg 2ϩ were omitted from Krebs solution. For measuring the Ca 2ϩ currents, Krebs was used with 10 mM CaCl 2 instead of 1 mM CaCl 2 . The internal (pipette) solution contained 10 mM BAPTA, 20 mM CsCl, 100 mM Csaspartate, 1 mM MgCl 2 , 4 mM Na 2 ATP, 10 mM HEPES-CsOH, pH 7.2. The Ca 2ϩ concentration in the pipette was adjusted to 10 nM, 20 nM, 50 nM, 100 nM, 250 nM, or 1 M by adding 0.747, 1.391, 2.880, 4.457, 6.682, or 8.916 mM CaCl 2 , respectively, in the presence of 10 mM BAPTA, as calculated by the CaBuf program. 2 All experiments were performed at room temperature (20 -22°C). Reported current densities were calculated from the current at Ϫ80 mV during the ramp protocol for the Na ϩ currents or at the peak amplitude in the step protocol, respectively. The software package ASCD 3 was used for analysis of the whole-cell data. All current amplitudes in the dose-response curves were normalized to Ϫ80 mV with 10 mM BAPTA in the pipette solution and fitted with a modified Hill function (using Origin version 7.0 software, OriginLab Corporation, Northampton, MA).
Statistical Analysis-In all experiments, the data are expressed as mean Ϯ S.E. Overall statistical significance was determined by analysis of variance. In case of significance (p Ͻ 0.05), individual groups were compared by Student's t test.

Identification of 80K-H as a Gene Regulated by 1,25-(OH) 2 D 3 and Dietary
Ca 2ϩ -Mice, in which the renal D 3 -1␣-hydroxylase enzyme catalyzing the 1,25-(OH) 2 D 3 synthesis was inactivated, exhibited undetectable levels of 1,25-(OH) 2 D 3 , resulting in severe hypocalcemia (21). We performed a microarray analysis on kidney cDNA of these 1,25-(OH) 2 D 3 -deficient mice, which were rescued by 1,25-(OH) 2 D 3 supplementation or by dietary Ca 2ϩ (9,22). Analysis of the gene expression data provided information about RNA transcripts important for the maintenance of the Ca 2ϩ balance and pointed to the 80K-H protein, which is a PKC substrate containing two putative Ca 2ϩ -binding sites. 80K-H was significantly down-regulated at the mRNA level in the absence of 1,25-(OH) 2 D 3 (79% compared with control values in the presence of 1,25-(OH) 2 D 3 ) and up-regulated by dietary Ca 2ϩ supplementation (189% compared with the 1,25-(OH) 2 D 3deficient mice). The regulation of 80K-H at the transcription level was confirmed by quantitative real-time PCR analysis (from 100% to 157 Ϯ 12% and 155 Ϯ 7%, respectively; p Ͻ 0.05).
Localization of 80K-H in Ca 2ϩ -transporting Epithelia-Tissue distribution of 80K-H was first investigated by reverse transcription PCR on mouse tissues using primers located in a species-conserved region within the 80K-H C-terminal tail. The expected 80K-H amplicon of 242 bp was detected in all tissues, including Ca 2ϩ -transporting epithelia such as kidney and duodenum (Fig. 1A). Localization of 80K-H was further studied by immunohistochemistry of mouse kidney, where immunopositive staining for 80K-H and TRPV5 was present along the apical membrane of distal convoluted and connecting tubules (Fig. 1B).
Interaction of 80K-H with TRPV5-To examine the association of 80K-H with TRPV5, immunoprecipitation experiments were performed using the X. laevis oocyte expression system. Membrane extracts of X. laevis heterologously expressing TRPV5 and 80K-H were first immunoprecipitated with anti-80K-H antibody, and the precipitated proteins were analyzed by immunoblotting. The two specific bands of HA-tagged TRPV5, corresponding to the core and the non-glycosylated proteins (23), were detected in the precipitated complex ( Fig.  2A). Likewise, this interaction was confirmed in the reverse manner by immunoprecipitation of TRPV5 from membrane extracts using the anti-HA antibody and subsequent immunoblotting with the anti-80K-H antibody (Fig. 2B). Ca 2ϩ -depend-ence of 80K-H association with TRPV5 was investigated by GST pull-down assays in the presence (1 mM CaCl 2 ) and absence (5 mM EDTA) of Ca 2ϩ . 80K-H protein bound specifically to TRPV5 in both conditions because no interaction was observed with GST alone (Fig. 2C). Immunoblot analysis of mouse kidney lysates did not result in a specific immunopositive signal (data not shown), which is probably due to the relatively low abundance of these channels (24). Endogenous immunoprecipitation of the two proteins was, therefore, not possible. Mapping of the 80K-H-binding Site in TRPV5-To determine the 80K-H binding site in TRPV5, a series of deletion mutants between positions 598 and 680 of the TRPV5 C-terminal tail was constructed. Truncated forms of TRPV5 were expressed as GST-fusion proteins and tested for their interaction with in vitro translated 80K-H protein using pull-down experiments (Fig. 2D). The interaction between the two proteins was abolished when TRPV5 was truncated at position 603, whereas truncations at positions 680 up to 608 did not disturb the interaction with 80K-H. The sequence of TRPV5 protein corresponding to the binding site of 80K-H was restricted to the short region between the positions 598 and 608.
Ca 2ϩ -binding Capacity of 80K-H-80K-H contains two putative EF-hand structures and a highly negatively charged glu-tamate stretch that might be a Ca 2ϩ -interacting region (25). Ca 2ϩ -binding properties of 80K-H were investigated by performing a 45 Ca 2ϩ overlay assay, using 80K-H wild type and mutants of the two putative Ca 2ϩ -related regions (Fig. 3A). The first mutant was generated by substitution of the critical aspartates for alanine residues that impair Ca 2ϩ coordination in the two EF-hands (80K-H EF mutant). The second mutant was generated by excision of the glutamate stretch (80K-H ⌬Glu mutant). Pull-down experiments demonstrated that mutation of 80K-H, in the regions described above, did not disturb binding to the TRPV5 C-terminal tail (Fig. 3B). Subsequently, GSTfused proteins of 80K-H were separated by SDS-PAGE followed by blotting. An equal expression level of 80K-H proteins was verified by Ponceau S staining. After incubation with 45 CaCl 2 , the capacity of the blotted proteins to bind 45 Ca 2ϩ was determined (Fig. 4). The amino acid substitutions in the two EFhands of 80K-H abolished the Ca 2ϩ binding, whereas deletion of the glutamate stretch did not interfere with Ca 2ϩ binding. GST protein alone was used as a negative control.
Cell-surface-associated Expression of 80K-H and TRPV5-TRPV5 plasma membrane expression was examined in HEK293 cells transfected with TRPV5 and 80K-H with a biotinylation method using Sulfo-NHS-SS-Biotin (Pierce). TRPV5 was expressed at the plasma membrane of HEK293 cells when transfected alone and together with 80K-H. Co-expression of TRPV5 and 80K-H mutants did not significantly affect the amount of TRPV5 expression on the plasma membrane fraction. Non-transfected cells were used as a negative control (Fig.  5A). In the same experiment, the association of 80K-H with TRPV5 channels located in the plasma membrane was studied. Interestingly, 80K-H and mutants were detected in the bio- tinylated membrane fraction when co-expressed with TRPV5 (Fig. 5B). The endogenously and heterologously expressed 80K-H was not detected at the plasma membrane fraction in the absence of TRPV5, despite their abundant expression in the cell lysates (Fig. 5B, right panel). Thus, it could be assessed that the expression of 80K-H at the plasma membrane depends on the presence of TRPV5.
Effect of Mutations in 80K-H on TRPV5 Channel Activity-The effect of endogenous 80K-H WT and mutants on TRPV5 activity was determined by whole-cell patch-clamp analysis in transiently transfected HEK293 cells (Fig. 6). In line with previous results, TRPV5 conducted Na ϩ and Ca 2ϩ currents, including large monovalent currents, strong inward rectification, and high Ca 2ϩ selectivity over Na ϩ (7). Cells expressing either TRPV5 alone or in combination with 80K-H WT displayed similar current densities and electrophysiological properties: Na ϩ current at Ϫ80 mV 441 Ϯ 149 pA/pF (n ϭ 4) versus 562 Ϯ 169 pA/pF (n ϭ 6), and Ca 2ϩ peak current 337 Ϯ 109 pA/pF (n ϭ 4), compared with 299 Ϯ 162 pA/pF (n ϭ 4) for cells expressing TRPV5 alone or TRPV5 together with 80K-H WT , respectively. Thus, expression of 80K-H WT did not significantly alter TRPV5 channel activity (p Ͼ 0.1). Co-expression of TRPV5 with the 80K-H EF , 80K-H ⌬Glu , and 80K-H ⌬HDEL mutants had no significant effect upon the shape or amplitude of the current in a nominally DVF solution (p Ͼ 0.8, p Ͼ 0.8, and p Ͼ 0.1, respectively, compared with 80K-H WT ) (Fig. 6, A and  C). In the presence of extracellular Ca 2ϩ , TRPV5 became Ca 2ϩselective, and the inward Ca 2ϩ currents exhibited pronounced inactivation during hyperpolarizing steps (Fig. 6B). In contrast to endogenous 80K-H WT , the 80K-H mutants had profound effects upon the TRPV5 currents measured in the presence of 10 mM extracellular Ca 2ϩ (Fig. 6B). Both 80K-H EF and 80K-H ⌬Glu reduced the peak inward current by ϳ60% (p Ͻ 0.05 for both) (Fig. 6, B and D). These mutants had no effect on the extent of Ca 2ϩ current inactivation, which was quantified as the ratio between the peak current and the current at the end of the 3-s voltage step (Fig. 6E). 80K-H ⌬HDEL did not affect the peak inward current (p Ͼ 0.9), but significantly delayed the Ca 2ϩ current inactivation (p Ͻ 0.5) (Fig. 6, B, D, and E).
Modulation of TRPV5 Sensitivity for Intracellular Ca 2ϩ by 80K-H EF -Currents through TRPV5 are inhibited by increasing the intracellular Ca 2ϩ concentration ([Ca 2ϩ ] i ) (8). A 40% reduction was observed in TRPV5 expressing cells dialyzed with a solution containing 50 nM free Ca 2ϩ when compared with an intracellular solution containing 10 mM BAPTA and no added Ca 2ϩ (Fig. 7A). In contrast, the currents of cells coexpressing TRPV5 and 80K-H EF in nominally DVF solution were nearly blocked with 50 nM Ca 2ϩ in the pipette, indicative of a higher sensitivity to intracellular Ca 2ϩ than cells expressing TRPV5 solely (Fig. 7B). Fig. 7C displays normalized inward Na ϩ currents as a function of [Ca 2ϩ ] i for cells expressing TRPV5 in the absence and presence of 80K-H EF . Ca 2ϩ concentrations for half-maximal inhibition (IC 50 ) of inward current were 72 Ϯ 12 nM and 12 Ϯ 2 nM for TRPV5 alone and in combination with 80K-H EF , respectively. All other 80K-H mutants had no significant effect on the Ca 2ϩ sensitivity of TRPV5 (data not shown).

DISCUSSION
The present study demonstrates a new function for the 80K-H protein as a Ca 2ϩ -dependent regulator of TRPV5 channel activity. This conclusion is based upon the following observations: (i) 80K-H and TRPV5 genes are similarly controlled by common regulators of Ca 2ϩ homeostasis; (ii) 80K-H and TRPV5 co-localize in the distal part of the nephron; (iii) 80K-H forms a heteromeric complex with TRPV5 along the plasma membrane and a distinct region of the TRPV5 C-terminal tail is important for this binding; and (iv) the two EF-hand structures of 80K-H control TRPV5 activity in a Ca 2ϩ -dependent way. is poorly understood, and there are no indications for a role in Ca 2ϩ homeostasis or ion channel regulation. However, several studies propose a role in cell activation pathways through receptors located at the plasma membrane. 80K-H was initially identified as a PKC substrate and was subsequently associated with intracellular signaling (26). Tyrosine residues of 80K-H are phosphorylated upon ligand binding to the fibroblast growth factor receptor 3 and activate the mitogen-activated protein kinase pathway (27,28). 80K-H is also implicated in signaling and auto-regulation of the advanced glycosylation end products receptor (29), which participates in endocytic and cell-activation pathways and is linked to nephropathy in type 1 diabetic patients (30). In addition, 80K-H homologous proteins have been identified, such as the vacuolar system-associated protein 60 which plays a role in membrane-trafficking events (31) and the non-catalytic ␤ subunit of the endoplasmic reticulum (ER) enzyme glucosidase-II GII␤ involved in the protein folding via deglycosylation (18,32).

80K-H Gene Expression Is Regulated by 1,25-(OH) 2 D 3 and
Dietary Ca 2ϩ -80K-H was identified using cDNA microarray analysis as a transcript down-regulated in 1,25-(OH) 2 D 3 -deficient mice (22). Normalization of the severe hypocalcemia in these mice by 1,25-(OH) 2 D 3 or dietary Ca 2ϩ supplementation was accompanied by an increase of the 80K-H mRNA expression. These two key regulators of Ca 2ϩ (re)absorption are important for the transcriptional modulation of renal proteins participating in the mechanism of transcellular Ca 2ϩ transport, including the epithelial Ca 2ϩ channel TRPV5, calbindin-D 28K , Na ϩ /Ca 2ϩ exchanger NCX1, and the Ca 2ϩ -ATPase PMCA1b (9).
80K-H Interacts with TRPV5-80K-H was ubiquitously expressed in mouse tissues and displayed consistent co-localization with TRPV5 along the apical membrane of distal convoluted and connecting tubules (17,33). Co-localization of 80K-H and TRPV5 in the same cellular compartments substantiates the physiological relevance of the demonstrated interaction between these two proteins. Indeed, a short peptide sequence between amino acids 598 and 608 of the TRPV5 C-terminal tail is necessary for 80K-H binding. This particular region of TRPV5 is highly conserved among all identified species (4). This domain could play an important role in the regulation of the channel via protein-protein interaction, because the S100A10-annexin 2 complex binds here (11). Biotinylation assays on cells co-expressing 80K-H and TRPV5 confirmed cellsurface expression of both proteins. This observation, together with the binding data, demonstrates the participation of 80K-H in a heteromeric complex with TRPV5 that is located at the plasma membrane. Importantly, 80K-H was not expressed at the plasma membrane in the absence of TRPV5, indicating a unique role of this protein in TRPV5-expressing tissues. Likewise, a tissue-dependent activity could explain the implication of 80K-H in diverse biological processes, ranging from vesicular transport and ER quality control to receptor signaling and channel regulation (18,29,32).
Critical Regions in 80K-H for TRPV5 Activity-80K-H sequence analysis predicts several potential threonine and tyrosine phosphorylation sites, two EF-hand domains, a highly acidic domain, and a potential ER-targeting sequence (18). The overall structure of 80K-H (Fig. 3A) suggests two potential Ca 2ϩ -interacting motifs, including a pair of EF-hand motifs and the acidic stretch of glutamates. In this study, we demonstrated that 80K-H coordinates Ca 2ϩ with the EF-hand pair, whereas the highly negatively charged poly-glutamate stretch does not account for Ca 2ϩ binding. Thus, 80K-H can be distinguished from ER proteins that contain similar acidic stretches with low Ca 2ϩ affinity (34). In this respect, deletion of the 80K-H putative ER-targeting signal (HDEL) did not disturb routing of the 80K-H-TRPV5 complex to plasma membrane. 80K-H ⌬HDEL had no effect upon TRPV5-mediated Ca 2ϩ currents; surprisingly, it increased the fraction of open channels. At this point, it is difficult to explain the action of the HDEL motif as a delay factor in the decay of TRPV5 Ca 2ϩ currents. These data can not exclude the possibility that the HDEL sequence constitutes an ER retention or retrieval signal (35), but they clearly support the hypothesis that motifs like the yeast HDEL or the respective mammalian KDEL are not sufficient for ER localization (36). Mutation of the EF-hand pair and the acidic glutamate region of 80K-H resulted in a dramatic decrease of TRPV5-mediated Ca 2ϩ influx without disturbing the association of 80K-H to TRPV5. Additionally, the plasma membrane localization of 80K-H and TRPV5, together with the fact that the Na ϩ current density was not altered, excludes the possibility of a routing defect. Previous findings have also indicated other EF-hand proteins, such as calmodu- lin, to be involved in the activation and inactivation process of ion channels and most interestingly, TRP channels (13,37). Moreover, the enriched glutamate stretch could provide an acidic milieu essential for phosphorylation of the 80K-H PKC substrate. Evolutionary and functional studies showed that variation in length and acidity of similar domains are related to the degree of phosphorylation needed for receptor desensitization (38,39).
80K-H as Ca 2ϩ Sensor-A major question that emerges from this study is how 80K-H controls the TRPV5-mediated Ca 2ϩ influx. Electrophysiological studies demonstrated that Ca 2ϩ influx through TRPV5 induces a feedback inhibition controlled by the ambient Ca 2ϩ concentration in a micro domain near the inner mouth of the channel (8). 80K-H EF significantly reduced the amount of intracellular Ca 2ϩ needed for the induction of TRPV5 feedback inhibition, but did not delay the recovery of the channel. Therefore, 80K-H could act as a Ca 2ϩ -sensing subunit of TRPV5 and facilitate Ca 2ϩ influx through TRPV5 after binding Ca 2ϩ , situated in the proximity of the inner mouth of the channel. In addition, our findings indicate roles for the glutamate stretch and the HDEL sequence of 80K-H in the inactivation and restoration processes of TRPV5 Ca 2ϩ currents, but these need further investigation. 80K-H could have a multifaceted mechanism regulating TRPV5 Ca 2ϩ -mediated influx. A dual function has also been proposed for the Ca 2ϩ sensor calmodulin, according to which calmodulin mediates both Ca 2ϩdependent facilitation and inactivation of L-type Ca 2ϩ channels (40). Finally, a Ca 2ϩ -sensing role of this PKC substrate can be postulated if we take into account the involvement of 80K-H in signal transduction cascades. In contrast to the Ca 2ϩ buffers parvalbumin, calretinin, and calbindin-D 28K , the Ca 2ϩ sensors calmodulin and troponin C do not only bind Ca 2ϩ next to the inner mouth of channels, but also activate various downstream effectors (12).
In conclusion, our findings describe 80K-H as a new Ca 2ϩ sensor and regulator of the epithelial Ca 2ϩ channel TRPV5. Given the wide tissue distribution and the functional diversity of 80K-H, this novel functional feature of 80K-H could also apply to other Ca 2ϩ -related physiological processes.