Characterization of single channel currents from primary cilia of renal epithelial cells.

The primary cilium is a ubiquitous, non-motile microtubular organelle lacking the central pair of microtubules found in motile cilia. Primary cilia are surrounded by a membrane, which has a unique complement of membrane proteins, and may thus be functionally different from the plasma membrane. The function of the primary cilium remains largely unknown. However, primary cilia have important sensory transducer properties, including the response of renal epithelial cells to fluid flow or mechanical stimulation. Recently, renal cystic diseases have been associated with dysfunctional ciliary proteins. Although the sensory properties of renal epithelial primary cilia may be associated with functional channel activity in the organelle, information in this regard is still lacking. This may be related to the inherent difficulties in assessing electrical activity in this rather small and narrow organelle. In the present study, we provide the first direct electrophysiological evidence for the presence of single channel currents from isolated primary cilia of LLC-PK1 renal epithelial cells. Several channel phenotypes were observed, and addition of vasopressin increased cation channel activity, which suggests the regulation, by the cAMP pathway of ciliary conductance. Ion channel reconstitution of ciliary versus plasma membranes indicated a much higher channel density in cilia. At least three channel proteins, polycystin-2, TRPC1, and interestingly, the alpha-epithelial sodium channel, were immunodetected in this organelle. Ion channel activity in the primary cilium of renal cells may be an important component of its role as a sensory transducer.

The primary cilium is a ubiquitous, non-motile organelle lacking the central pair of microtubules found in motile cilia (1,2). Recent studies indicate that primary cilia of renal epithelial cells may have important sensory properties (3). Bending of primary cilia either by fluid flow or mechanical stimulation initiates extracellular Ca 2ϩ influx amplified by Ca 2ϩ release from intracellular stores (4). Renal cystic (5)(6)(7)(8)(9) and other diseases (10) have recently been associated with dysfunctional ciliary proteins. Although sensory function by renal epithelial primary cilia may implicate functional channel activity in the organelle, information in this regard is still completely lacking (2,11). Except for nodal cilia, which exhibit an unusual twirling movement (12), primary cilia are thought to be non-motile because of the lack of axonemal dyneins. Primary cilia, like all eukaryotic cilia and flagella, are surrounded by a membrane that is continuous with the plasma membrane with a unique complement of membrane proteins such as receptors (13,14) and channel proteins such as polycystin (PC) 5 -2 (7,15). The function of this organelle is still largely unknown. Several hypotheses have been raised as to the potential role(s) of the primary cilium in cell function, ranging from vestigial appendage with no apparent function, to a potentially relevant role in the cell cycle, as it is derived from the centriole (2). Primary cilia arise from the centrioles that organize the mitotic spindle (16), and their expression is highly cell cycle-dependent. Cilia are most abundant in non-proliferating G 0 cells and can often be found in cells during interphase. Cilia assemble in G 1 until spindle formation is initiated by centriole separation (16). Renal primary cilia extend several micrometers from their apical surface (4,17). A mechanosensory role of renal primary cilia has been recently implicated in cell activation. Bending of primary cilia of cultured kidney cells by either fluid flow or mechanical stimulation has recently been shown to initiate extracellular Ca 2ϩ influx, which is amplified by Ca 2ϩ release from intracellular stores (4,18). Interestingly, the PC1⅐PC2 channel complex is not only present in primary cilia of renal epithelial cells (7, 15) but may be a requirement for Ca 2ϩ signaling and subsequent cell activation (15). It is thus speculated that the Ca 2ϩ influx signals initiated by cilia transduction may be the initial step in cell activation, including cell replication and differentiation. How this long and narrow organelle behaves as a signal transducer is still unknown. Previous studies in sensory cilia, would indicate, however, that channel activity in ciliary membranes may be an important factor in its sensory function (19). In this report we provide the first direct evidence for the presence of various channels phenotypes in nonmotile primary cilia of LLC-PK1 renal epithelial cells. Ciliary channel density is much higher than plasma membrane channel activity, suggesting an electrical phenomenon, where depolarization of this organelle, with respect to the cytoplasm, may contribute to its sensory function.

EXPERIMENTAL PROCEDURES
Isolation of Cilia from LLC-PK1 Cells-Wild type LLC-PK1 cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, as reported (20). Primary cilia from LLC-PK1 cells were isolated as follows. Confluent monolayers (2-3 weeks) were scraped with Ca 2ϩ -free phosphate-buffered saline and centrifuged for 5 min at 52 ϫ g. The cell pellet was suspended in a high Ca 2ϩ "deciliation" solution containing 112 mM NaCl, 3.4 mM KCl, 10 mM CaCl 2 , 2.4 mM NaHCO 3 , 2 mM HEPES, pH 7.0. Resuspended cells were shaken in this solution for 10 min at 4°C. Ciliary membranes were separated by centrifugation at 7,700 ϫ g for 5 min. The supernatant was loaded on top of a 45% sucrose solution in high Ca 2ϩ saline solution and centrifuged for 1 h at 100,000 ϫ g. The sucrose-supernatant interface band was collected and diluted (1:10) and centrifuged again for 1 h at 100,000 ϫ g. The pellet was resuspended in normal saline solution, pH 7.0, and supplemented with 2.0 mM EGTA and 0.5 mM sucrose. Samples were aliquotted and stored at Ϫ80°C until further use. Plasma membranes from confluent LLC-PK1 cells were obtained as reported (20). A 50-fold higher protein content in the plasma membranes compared with the isolated cilia (1160 Ϯ 190 g/ml versus 15.8 Ϯ 4.64 g/ml, p Ͻ 0.004, n ϭ 3) was observed by the method of Lowry.
Immunochemistry-Confluent cells were fixed with paraformaldehyde (4%) and 2% sucrose, for 10 min. Cells were rinsed three times with phosphate buffer solution and permeabilized with either 0.1% Triton X-100 or 1% Nonidet P-40. Some cells were immunolabeled without permeabilization. Cells were blocked with 1% bovine serum albumin or Image-it (MP) for 30 min prior to exposure to the primary antibody. Cells were immunolabeled with anti-acetylated ␣-tubulin antibody (mAb, Sigma) at concentrations of 2.8 g/ml of stock solution. The polyclonal anti-PC2 antibody (0.2 mg/ml stock) was obtained from PolyFast™ (Zymed Laboratories Inc.). The TRPC1 antibody (1:200 dilution from 0.3 mg/ml stock, Sigma) is from rabbit against a peptide corresponding to amino acids 557-571 of human TRPC1. The anti-rat ␣-epithelial sodium channel antibody was a kind gift from Dr. Tom Kleyman. A goat anti-rabbit secondary Alexa Fluor 594 (Molecular Probes) antibody was used at 2 g/ml.
Electron Microscopy-Isolated cilia were fixed in a solution containing 2% glutaraldehyde and Na ϩ cacodylate. Samples were with the buffer, spun down at 90,000 ϫ g in a TL 100 ultracentrifuge for 1 h at 4°C. The pellet was stained for 2 h at room temperature with 2% OsO 4 . Conversely, aliquots (10 l) were placed on a gold grid, wicked off with Whatman paper (#5), and negatively stained for 10 s with 2% phosphotungstic acid. Samples were observed with a Phillips CM10 electron microscope at 80 kV.
Electrophysiology of Isolated Cilia-Cilium-attached patches were obtained from isolated cilia. Currents and command voltages were obtained with a Dagan 3900 patch clamp amplifier (Dagan Corp.) under voltage clamp configuration with leak subtraction adjustments. Ion channel reconstitution of ciliary membranes was conducted as reported for other membranes (21) with some modifications. Briefly, isolated cilia were mixed and sonicated in the lipid mix (21) prior to reconstitution in a lipid bilayer system. All electrical signals were filtered at 5 kHz with the internal four-pole Bessel filter (Dagan Corp.). Data were first analyzed with PClamp 6.0.3 (Axon Instruments), and the basal line of current records was manually corrected with Clampfit 8.0. The patch pipette was filled with a solution containing 140 mM NaCl, 5.0 mM KCl, 1.0 mM MgCl 2 , 2.5 mM CaCl 2 , and 10 mM HEPES, adjusted to pH 7.4 with N-methylglucamine. The bathing solution contained an identical solution, and whenever indicated it was replaced with a solution containing 145 mM KCl, 1.0 mM MgCl 2 , 2.5 mM CaCl 2 , and 10 mM HEPES, adjusted to pH 7.4 with N-methylglucamine. EGTA was added from a 100 mM stock solution. The patches were intrinsically leaky, which constrains their analyses (22,23). Thus, the following restrictions apply to the data. Current records were only obtained under a given set of solutions, after offsetting tip potential to "zero current" in the absence of leak subtraction. This procedure was repeated each time that a solution was changed. Changes in baseline currents after experimental procedures prevented quantification. Thus, only identification of channel levels was considered relevant.

Identification and Isolation of Primary Cilia from LLC-PK1 Cells-
The expression of primary cilia in LLC-PK1 cells was determined by acetylatedtubulin immunocytochemical labeling, most particularly in fluid-filled domes and spontaneously formed cysts (Fig. 1a). Most cells contained identifiable cilia as a prominent thick bud and/or more elongated structures in all stages of growth. Primary cilia were detected in cells lining the wall of transporting domes (Fig. 1b). A closer look at cellular localization indicated a clear development toward the side of the nucleus (Fig. 1c). Some cilia extended several micrometers from the apical membrane, which had the distinctive fragmented pattern of acetylated tubulin (Fig. 1d). Primary cilia from confluent LLC-PK1 cells were isolated with a technique adapted from previous reports (24,25). Briefly, confluent LLC-PK1 cells were deciliated by a quick exposure to a high Ca 2ϩ solution and concentrated in a sucrose density gradient. Isolated cilia were identified after fixation and immunolabeling with anti-acetylated tubulin antibody (Fig. 2a). Isolated cilia were further visualized and identified at higher resolution (ϫ25,000) by electron microscopy (Fig. 2b) where ciliary stems and tips can be identified as reported (26).
Patch Clamping of Isolated Cilia-To obtain direct electrophysiological information, isolated cilia were identified under oil phase contrast (ϫ100) in saline solution, and after location, further visualized at lower resolution (ϫ40) to place the patch pipette in the proximity of the isolated cilium (Fig. 2, c and d). Tip resistance was followed by applying 1-mV square pulses before and after suction of the patch pipette against the isolated cilium (Fig. 2e). Most frequently, the resistance increased after touching, indicating that vacuum suction was sufficient to produce a "leaky patch." "Leak subtraction" compensation followed with the patch clamp amplifier. Despite the intrinsic "leakiness" of the patches, channel activity was clearly observed under spontaneous conditions (n ϭ 22/22, Fig. 3a). The most frequent ion channel observed (Fig. 3, a-d) had a single channel conductance of 83.6 Ϯ 1.0 pS (n ϭ 3) in symmetrical NaCl. Replacement of the bathing solution with either Na ϩ -aspartate (71.5 Ϯ 1.47 pS, Fig. 3d) or KCl (73.2 Ϯ 1.58 pS, Fig. 3d) did not largely modify the single conductance, reversal potential, or rectification properties, suggesting its activity as a nonselective cation channel. Another non-selective cation channel phenotype with a single channel conductance of 173 Ϯ 4.40 pS (n ϭ 3) was observed in the primary cilium (Fig. 3d, inset). Whether this channel phenotype represents a "dimer" of the 83-pS channel is currently unknown. Further, a small Na ϩ -permeable channel (8.07 Ϯ 0.50 pS, n ϭ 3) was also observed after the addition of AVP (10 M, Fig. 3, e and f). The presence of V2R vasopressin receptors is being currently explored and would suggest that a local second messenger pathway is present in isolated primary cilia (to report elsewhere). Interestingly, addition of cAMP-dependent protein kinase (50 nM) and MgATP (3-6 mM) also increased cation channel activity (Fig. 3g). This suggests that vasopressin acts on a V2R, instead of a V1R-type of response. However, we do not presently know how the enzyme reaches the intraciliary compartment. It is likely, however, that both the leaky and openended nature of the isolated cilium help diffuse the drugs in place. Channel function was inhibited by addition of amiloride (5 M, n ϭ 4, Fig. 3h).
Reconstitution of Isolated Cilia-Ciliary membrane channels were also observed by reconstitution in a lipid bilayer system (Fig. 3, i and j) in the presence of a Na ϩ gradient (150 versus 15 mM, in cis and trans compartment, respectively). Channels were observed in 297 of 303 ciliary membranes. The most frequent cation-selective channel had a single channel conductance of 156 pS, which was inhibited by an anti-PC2 antibody (Fig.  3i). Reconstituted ciliary membranes seldom displayed 75-pS Cl Ϫ -permeable channels (3/303, Fig. 3j), explaining why it was not observed in ciliumattached patches. The data indicate, however, that abundant cation-permeable channel activity is present in the ciliary membranes. A comparison of reconstituted ciliary versus plasma membranes, indicated as much as 400fold higher channel activity in ciliary membranes as averaged mean currents were divided by protein content (Fig. 3k).
Presence of Channel Proteins in the Primary Cilium-To begin an identification of the channel proteins contributing to the electrical activity of the primary cilium in renal epithelial cells immunolocalization was conducted in isolated cilia identified by co-labeling with acetylated tubulin. Several ion channels were identified in primary cilia. The TRP-type channels TRPC1 and polycystin-2 (TRPP2) were immunodetected in the primary cilia of confluent LLC-PK1 cells (Fig. 4) and also isolated cilia (data not shown). Interestingly, the epithelial Na ϩ channel subunit ␣-epithelial sodium channel was also observed lining the entire surface of the primary cilium (Fig. 4b). The presence of several channel species is in agreement with the high electrodifussional permeability observed in the primary cilium, with respect to the plasma membrane.

DISCUSSION
Recent evidence indicates that renal epithelial cells may sense environmental forces by mechanosensory activity of the primary cilium (4,15). Little is known, however, about the molecular mechanisms associated with sensory function of renal primary cilia. Electrical activity in this organelle, mediated by functional ion channels, may be an important contributor to this sensory function. Among the most studied ciliary membranes are those from sensory neurons of the olfactory bulb (27)(28)(29)(30) and sensory cilia of ciliated invertebrates (31,32). The reconstitution of olfactory ciliary membranes in planar lipid bilayers originally conducted by Ehrlich et al. (33) demonstrated the presence of voltage-dependent Ca 2ϩ channels from Paramecium cilia. Reconstitution of ciliary membranes from wild type and mutant Tetrahymena thermophila also showed the presence of other cation channels (31,34). Reconstituted sensory ciliary membranes have also allowed identification of signaling mechanisms involved in ciliary channel regulation. Labarca et al. (35) observed that ion channels from reconstituted ciliary membranes of the Rana catesbeiana olfactory epithelium, are sensitive to nanomolar concentrations of odorant ligands. This may have been one of the first demonstrations of TRP vanilloid receptor-type functional channels in cilia. Cyclic AMP activated the 23-pS cation channels in vertebrate olfactory cilia (24) and modulated a high conductance K ϩ channel, displaying several open substates with conductances of 34, 80, and 130 pS (24), which is highly reminiscent of PC2 function (21). The presence of cAMP-activated 8 pS chloride channels has also been determined by steady-state noise analysis of ligand-induced currents (36). Sensory cilia membranes also contain inositol 3-phosphate-gated channels (37). In that study, two types of non-selective cation channels were observed by current fluctuations in rat olfactory cilia membranes fused onto phospholipid bilayers. Direct electrical information from sensory cilia in situ was obtained by Frings and Lindemann (19), who conducted highly demanding studies to determine the biophysical properties and regulation of ionic conductances in cilia from frog olfactory bulb. The authors managed to pull sensory cilia from olfactory receptor cells into a patch pipette. Despite the fact that the pipette did not form a tight electrical seal with the ciliary membrane, transient record currents driven by action potentials arising from the olfactory neuron were collected. With this method, odorant thresholds in the picomolar range were obtained (19), indicating that ligand affinity is much higher in situ than with reconstituted preparations. This finding also suggests that, however difficult, these experiments may provide more reliable information not available by other methods (28,38). The encompassed evidence suggests an abundance of ion channels in sensory cilia. Several G protein receptors, including those for somatostatin and serotonin have also been found in primary cilia of brain neurons (13,14). Thus sensory function in cilia is intrinsically related to the presence of functional channels in these membranes. To date, no information is available on either membrane receptors and/or ion channel activity in primary cilia from renal epithelial cells. Recent studies indicate, however, that the autosomal dominant polycystic kidney disease gene products PC1 and PC2, a novel TRP channel member (TRPP2), are present in the primary cilia of renal epithelial cells (7,9,15). A functional interaction between the two may seem central to environmental cell signaling events leading to Ca 2ϩ transport regulation and subsequent cell signaling (15).
In the present study we described a method to acquire single channel data from isolated primary cilia and provided the first direct evidence for the presence of functional channels in the membrane coating the primary cilium of renal epithelial cells. At least three cation-selective channel phenotypes were observed, although the ϳ80-pS cation-selective channel may be the most prevalent subconductance state of a large channel as observed FIGURE 2. Experimental approach to acquiring electrical information from primary cilia. a, isolated primary cilia from LLC-PK1 renal epithelial cells were obtained with a cell deciliation procedure as described under "Experimental Procedures." Isolated cilia were visualized by immunocytochemistry at ϫ100 magnification after fixation and labeling with anti-acetylated tubulin antibody. b, electron micrograph of isolated cilia. Arrowheads and arrows indicate cilia and cilia tip structures, respectively, consistent with previous reports (26). Right panels detail cilia and tips. c, the patch pipette was approached under lower magnification (ϫ40) and then switched to higher magnification (d, ϫ60). e, suction of the cilium to the patch pipette increased the tip resistance, which further increased by leak subtraction with the patch clamp amplifier. FIGURE 3. Single channel currents from isolated cilia of LLC-PK1 cells. a, spontaneous single channel currents in a cilium-attached patch. b, the all-point histogram shows the main conductance, and a superimposed substate (asterisk). c, single channel currents at different holding potentials. d, current-to-voltage relationships for single channel currents in the presence of symmetrical NaCl (n ϭ 17), asymmetrical Cl Ϫ /aspartate (inverted triangles, n ϭ 3), and asymmetrical Na ϩ /K ϩ (diamonds, n ϭ 2). A large conductance channel, 173 pS was seldom observed (n ϭ 3, circles, Na ϩ /Na ϩ , and Na ϩ /K ϩ , diamonds, inset). e, vasopressin-induced small single channel currents (asterisks), with an 8-pS single channel conductance in symmetrical NaCl (f). g, channel activity increased after addition of cAMP-dependent protein kinase and MgATP (n ϭ 3). h, cation channel activity was completely inhibited by addition of amiloride (2 M, n ϭ 4). i, reconstituted cation channels in a lipid bilayer system. The high conductance channel showed several substates and was inhibited by an anti-PC2 antibody. j, anion-selective single channel currents were also observed in reconstituted LLC-PK1 ciliary membranes. The single channel conductance of this Cl Ϫ -permeable channel was 75 pS (n ϭ 3). k, comparison of average channel activity (as mean current, pA) between LLC-PK1 plasma (Mbs) and ciliary membranes (Cilia). There is a statistically significant Ͼ400-fold higher channel activity in ciliary membranes when corrected by protein content (p Ͻ 0.001, see "Experimental Procedures").
in both ciliary patches and reconstituted membranes. The present data are in agreement with the presence of TRP-type channel activity in primary cilia from renal epithelial cells. An ability to patch and identify single channel currents in these narrow organelles may likely lie in the fact that the process of isolation swells the cilia to dimensions where the membrane area is large enough to interact with the patch pipette. Cation channel in primary cilia of renal epithelial cells may help maintain this compartment depolarized with respect to the cytoplasm of the cell, as depicted in the hypothetical model of ciliary function (Fig. 4c). The presence of a cAMP response is consistent with the presence of the cAMP-dependent protein kinase-regulated ␣-epithelial sodium channel protein in this organelle. This also is in agreement with data obtained in voltage-clamped sensory cilia (36) and the presence G protein receptors, such as those for somatostatin and serotonin in the cilia of brain neurons (13,14). Thus high intrinsic cation channel activity in non-motile primary cilia of renal epithelial cells may contribute to the sensory properties of this organelle.