Abstract
The epithelial cells of Reissner's membrane (RM) are capable of transporting Na+ out of endolymph via epithelial Na+ channel (ENaC). However, much remains to be known as to mechanism of regulation of Na+ absorption in RM. We investigated P2Y signaling as a possible regulatory mechanism of ENaC in gerbil RM using voltage-sensitive vibrating probe technique and immunohistochemistry. Results showed that UTP induced partial inhibition of the amiloride-sensitive short-circuit current but did not change short-circuit current when applied in the presence of amiloride. The inhibitory effect of UTP was not completely reversible in minutes. The response to UTP was inhibited by reactive blue-2 and 2′,3′-O-(4-benzoylbenzoyl)adenosine 5′-triphosphate but not by suramin or pyridoxalphosphate-6-azophenyl-2′, 4′-disulfonic acid, which indicates this P2Y receptor as the P2Y4 subtype. The phospholipase C (PLC) inhibitors 1-[6[[(17β)-3-methoxyestra-1,3,5(10)-trien-17-yl]amino]hexyl]-1H-pyrrole-2,5-dione and 1-O-octadecyl-2-O-methyl-rac-glycero-3-phosphocholine markedly inhibited the effect of UTP on ENaC. In contrast, neither modulation of protein kinase C nor application of 2-aminoehoxydiphenyl borate affected P2Y4-mediated inhibition of ENaC. Immunoreactive staining for P2Y4 was observed in the RM, apical membrane of stria vascularis, spiral ligament, and organ of Corti, including outer hair cell, inner hair cell, outer pillar cell, Deiters' cell, and Hensen cell. These results suggest that the physiological role of P2Y4 receptor in RM is likely to regulate Na+ homeostasis in the endolymph. The acute inhibition of ENaC activity by activation of P2Y4 receptor is possibly mediated by decrease of phosphatidylinositol 4,5-biphosphate in the plasma membrane through PLC activation.
Introduction
The unique ion composition of cochlear endolymph (high-K+, low-Na+) is essential for the function of the sensory hair cells. K+ transport mechanisms in the cochlea have been mainly elucidated (Wangemann, 2002; Zdebik et al., 2009). The process involved in Na+ transport in the cochlea also has been described recently. Epithelial cells of the Reissner's membrane (RM) (Lee and Marcus, 2003; C. H. Kim et al., 2009b; S. H. Kim et al., 2009) and outer sulcus cells (Marcus and Chiba, 1999; Lee et al., 2001) were reported to contribute to endolymphatic homeostasis by absorbing Na+.
The RM consists of two cell layers (tight epithelia, which face endolymph, and mesothelia, which face perilymph) that are separated by a basement membrane and a thin layer of intercellular substance. Recently, there are increasing evidences that RM contributes endolymphatic Na+ homeostasis via apical epithelial Na+ channel (ENaC). The amiloride-sensitive Na+ channel-like immunoreactivity was detected in the luminal surface of the epithelial cells of gerbil RM but not of the mesothelial cells (Mizuta et al., 1995), and all three subunits of ENaC were immunolocalized in rat RM (Zhong and Liu, 2004). ENaC mRNA was localized by in situ hybridization in the epithelial cells of rat RM (Couloigner et al., 2001). Transcripts for three subunits of ENaC were present in mouse RM and glucocorticoid upregulated transcriptions for Na+ transport gene expression (S. H. Kim et al., 2009). The electrogenic transepithelial Na+ transport was demonstrated in freshly dissected gerbil RM using vibrating probe technique (Lee and Marcus, 2003). It was also demonstrated that transepithelial Na+ current in RM increases during early development (C. H. Kim et al., 2009b). The α-subunit of Na+, K+-ATPase was found to be expressed at the basolateral membrane of the epithelial cells of RM (Iwano et al., 1989). Several types of ion channels, including stretch-activated nonselective cationic channel in the epithelial cell membrane of RM, were identified by patch-clamp studies (Yeh et al., 1997, 1998).
It is known that the activity of ENaC is regulated by hormones such as glucocorticoid, aldosterone, and vasopressin (Garty and Palmer, 1997) and local paracrine factors such as ATP (Kunzelmann et al., 2005). Recently, it has been reported in many other epithelia that extracellular ATP decreases ENaC activity through P2Y purinergic signaling by depletion of phosphatidylinositol 4,5-bisphosphate [PI(4,5)P2] in the plasma membrane (Inglis et al., 1999; Cuffe et al., 2000; Thomas et al., 2001; Lehrmann et al., 2002; Leipziger, 2003). Purinergic receptors appear to form the basis of paracrine and autocrine communication systems in the cochlea (Housley, 1998; Lee and Marcus, 2008), whereas the regulatory mechanism of Na+ transport through purinergic activation in the inner ear has not been identified yet.
In this study, we investigated the expression of P2Y receptor and its regulation mechanism of ENaC in the epithelial cells of gerbil RM using vibrating probe technique and immunohistochemistry. Our results demonstrate that P2Y4 receptor is expressed in RM and plays a role in the inhibitory regulation of ENaC activity presumably through depletion of PI(4,5)P2 in the plasma membrane.
Part of this work has been published previously in abstract form (C. H. Kim et al., 2009a).
Materials and Methods
Tissue preparation.
Gerbils (3–4 weeks old) were anesthetized with sodium pentobarbital (50–100 mg/kg, i.p.) and killed to remove temporal bones. The methods used for dissecting RM have been described previously (Lee and Marcus, 2003). The stria vascularis was removed from the lateral wall of the apical cochlear turn, and the attached portion of RM was folded over the suprastrial portion of the spiral ligament. The tissue was mounted in a perfusion chamber on the stage of an inverted microscope (Olympus IX70) and continuously perfused at 37°C at an exchange rate of three times per minute. All procedures conformed to protocols approved by the Institutional Animal Care and Use Committee of Seoul National University.
Voltage-sensitive vibrating probe.
The vibrating probe technique was used to measure transepithelial currents under short-circuit conditions attributable to the small size of RM epithelium. The diameter of the vibrating probe tip was ∼20 μm, and it allowed the detection of voltages in the low nanovolt range; the vibration between two positions within the line of current flow yields voltages that correspond to current flow through the resistive physiological saline (Marcus, 1996). The vibrating probe technique used was identical to a previously described method (Marcus and Shipley, 1994; Marcus, 1996). Briefly, the short-circuit current (Isc) was monitored by vibrating a platinum–iridium wire microelectrode insulated with parylene-C (Micro Electrodes) and that had been coated with platinum-black on its exposed tip. The vibration was ∼20 μm along both horizontal (x) and vertical (z) axes. The x-axis was perpendicular to the face of the epithelium, and the probe was positioned at 30 μm from the apical surface of the epithelium using computer-controlled, stepper-motor manipulators (Applicable Electronics) and specialized probe software (ASET version 2.0; Science Wares). The bath references were 26 gauge platinum-black electrodes. Calibration was performed in physiologic saline (see below) using a glass microelectrode (tip <1 μm outer diameter) filled with 3 m KCl as a point source of current. The frequencies of vibration used were in the range 200–400 Hz and were well separated for the two orthogonal directions. Signals from the oscillators driving the probe were also fed to a dual-channel phase-sensitive detector. Asymmetry of probe design yielded different resonant frequencies for the two directions of vibration. X and Z detector signals were connected to a 16 bit analog-to-digital converter (CIO-DAS1602/16; ComputerBoards) in a Pentium IV computer. The sampling interval was 0.6 s, which was the minimum interval allowed by this software. The electrode was positioned where Isc showed a maximum X value and minimum Z value; data are expressed as X value and plotted using Origin software, version 6.1 (OriginLab Software). The output from the vibrating probe depended not only on the specific short-circuit current of the epithelium but also on the position of the probe from the surface of the tissue and on the precise geometry of each tissue sample. The current density reported here refers to the flux at the probe position and represents only a fraction of the current crossing the epithelium. No changes in the relative position of the probe were observed as a result of swelling or shrinking of tissue during experimental treatments.
Immunohistochemistry.
Gerbils at the age of 21 d were transcardially perfused with PBS and then with 4% paraformaldehyde in PBS. The cochleas were dissected out and postfixed, by immersion, with a fresh solution of 4% formaldehyde in PBS for 1 h. After postfixation, the tissues were washed with PBS and transferred to decalcifying solution (0.12 m EDTA, pH 7.2) for 48 h. The EDTA solution was changed after 24 h. The tissues were embedded in Tissue-Tek (Sakura Finetek USA), cryosectioned to a 10 μm thickness using a cryostat (−23°C chamber), and mounted on ProbeOn Plus charged glass slides. The tissue sections were warmed for 15 min at 37°C and then rehydrated with PBS for 10 min. The tissues were then permeabilized, and the nonspecific antigenic sites were blocked using the blocking solution (0.3% Triton X-100 and 10% normal donkey serum in PBS). The sections were incubated overnight at 4°C with the primary antibody. The primary antibody was rabbit anti-P2Y4 (Alomone Labs) diluted at 1:200. The specificity of immunohistochemical stain was controlled by preincubation of antiserum with peptide antigen (the ratio of concentration of peptide to antibody was 1:1) and omitting the primary antibody. The sections were extensively washed with PBS and then incubated in the dark for 1 h at room temperature with the secondary antibody (anti-rabbit Alexa Fluor 488; Invitrogen) diluted in 0.01% Triton X-100 and 1% normal donkey serum in PBS to a final dilution of 1:1000. Finally, the sections were extensively washed with PBS and overlaid with 20 ml of Gel/Mount and a cover glass. The sections were observed with a confocal microscope (LSM510 META; Carl Zeiss).
Solutions and chemicals.
The perfusate used as control solution was a perilymph-like physiologic saline of pH 7.4 containing the following (in mm): 150 NaCl, 3.6 KCl, 1 MgCl2, 0.7 CaCl2, 5 glucose, and 10 HEPES. Reactive blue-2 (RB-2) (R-115; Sigma), 2′,3′-O-(4-benzoylbenzoyl)adenosine 5′-triphosphate (BzATP) (B-6396; Sigma), suramin (S-2671; Sigma), UTP (U-4630; Sigma), and pyridoxalphosphate-6-azophenyl-2′,4′-disulfonic acid (PPADS) (P-178; Sigma) were directly dissolved in the control solution just before use. Amiloride (A-7410; Sigma), U-73122 (1-[6[[(17β)-3-methoxyestra-1,3,5(10)-trien-17-yl]amino]hexyl]-1H-pyrrole-2,5-dione)(U-6756; Sigma), U-73343 (1-[6-((17β-3-methoxyestra-1,3,5(10)-trien-17-yl)amino) hexyl]-2,5-pyrrolidine-dione) (U-6881; Sigma), 1-O-octadecyl-2-O-methyl-rac-glycero-3-phosphocholine (edelfosine; O-9262; Sigma), GF 109203X (2-[1-(3-dimethylaminopropyl)-1H-indol-3-yl]-3-(1H-indol-3-yl) maleimide) (catalog #0741; Tocris Bioscience), phorbol myristate acetate (PMA) (P8139; Sigma), and 2-aminoehoxydiphenyl borate (2-APB) (catalog #1224; Tocris Bioscience) were predissolved in dimethylsulfoxide (DMSO) and then diluted to 0.1% DMSO in the control solution before application. DMSO at this concentration had no effect on Isc.
Data presentation and statistics.
The baseline Isc values in the control solution were obtained by averaging the data for 9 s just before solution change. Each drug was applied for 2–3 min to see the effect on Isc. For the analysis of the effect of each drug, the data were averaged for 9 s after reaching steady state. Increases or decreases in Isc were considered significant at the p < 0.05 level. Statistical comparisons between two means were obtained with t test (Mann–Whitney test, if n < 5) or paired t test (Wilcoxon's signed rank test, if n < 5). The data shown were expressed as mean values ± SEM (n = number of tissues) of the Isc.
Results
Effect of UTP on Na+ absorption in the epithelial cells of RM
We measured baseline Isc of RM in the control solution and the change of Isc after application of amiloride (10 μm) or UTP (100 μm). Negative baseline Isc values were observed in the control solution before application of UTP or amiloride (Fig. 1, Table 1). The application of UTP led to a partial inhibition of Isc (42.9 ± 2.6%, n = 7), and subsequent addition of amiloride inhibited the remaining Isc completely (Fig. 1A, Table 1). Interestingly, application of amiloride showed positive Isc values (Fig. 1, Table 1), which was also observed in previous work (Lee and Marcus, 2003), although the cause remains unknown. In contrast, the application of UTP after pretreatment of amiloride, which abolished most of baseline Isc, showed no change of Isc (Fig. 1B, Table 1). These results indicate that UTP inhibits amiloride-sensitive Na+ absorption partially in the epithelial cells of RM.
Dose–response relationship of UTP
UTP inhibited amiloride-sensitive Na+ absorption in a concentration-dependent manner (from 0.1 μm to 3 mm). The initial Isc was −16.97 ± 0.79 μA/cm2 (n = 10), which was reduced by applying UTP to −16.33 ± 0.71 (n = 10, 0.1 μm), −14.93 ± 0.67 (n = 10, 1 μm), −12.03 ± 0.69 (n = 10, 10 μm), −9.31 ± 0.75 (n = 10, 30 μm), −7.55 ± 0.96 (n = 10, 100 μm), −5.94 ± 0.89 (n = 10, 300 μm), −4.45 ± 0.40 (n = 5, 1 mm), and −4.32 ± 0.39 (n = 5, 3 mm). The data were fitted to the following Hill equation (Fig. 2): where y is the predicted inhibition by UTP, Vmax is a maximum inhibition, [UTP] is UTP concentration, IC50 is the UTP concentration of the half-maximal inhibitory effect, and h is the Hill coefficient (slope factor) of sigmoidicity. Best-fit estimates yielded maximum inhibition 74.7 ± 0.9%, IC50 of 17.2 ± 1.1 μm, and Hill coefficient of 1.1 ± 0.1.
Characterization of UTP-responsive P2Y receptor
We introduced the antagonists of purinergic receptors to characterize the subtype of UTP-responsive P2Y receptor and compared the effects of UTP on Na+ absorption in the absence and presence of each antagonist (Fig. 3, Table 2). We have consistently observed that the effect of UTP on Na+ absorption was not completely reversible in minutes. The recovery rate of Isc was 75.5 ± 2.4% at 3 min after UTP washout (n = 56) (Tables 2, 3). Therefore, we conducted control experiments (time-controlled) using the protocol consisting of two consecutive applications of UTP (100 μm) with 6 min interval (Fig. 3A). The magnitude of inhibition of Isc by the second application of UTP was partially reduced (18.4 ± 6.0%, n = 6) compared with that by the first application of UTP in the control experiments (Fig. 3A, Table 2). This ratio of the magnitude of inhibition obtained in the control experiments was statistically compared with that obtained in the each experimental protocol using antagonists of purinergic receptors (Fig. 3B–E) and phospholipase C (PLC) signaling modulators (Fig. 4) to determine the effect of each drug on Na+ absorption.
The application of RB-2 (100 μm) for 3 min resulted in no significant change in Isc (n = 4) (Fig. 3B, Table 2). The magnitude of inhibition of Isc by UTP in the presence of RB-2 was reduced by 97.3 ± 1.6% (n = 4) (Fig. 3B, Table 2) compared with that observed in the absence of RB-2. This value (97.3% reduction) was significantly different from the value (18.4% reduction) observed in the control experiments. UTP response was significantly reduced by the application of 100 μm BzATP (n = 4, 90.2% reduction) (Fig. 3C). In contrast, neither 100 μm suramin (Fig. 3D) nor 100 μm PPADS (Fig. 3E) had any significant effect on the action of UTP. The magnitude of inhibition of Isc by UTP in the presence of suramin or PPADS was reduced by 11.3 ± 10.9% (n = 7) (Fig. 3C, Table 2) or 20.4 ± 16.0% (n = 6) (Fig. 3D, Table 2) compared with that observed in the absence of each antagonist. These values were not significantly different from the data (18.4% reduction) observed in the control experiments. These pharmacological results indicate that UTP-responsive P2Y receptor is P2Y4 subtype (see Discussion).
Mechanism of P2Y4-mediated regulation of ENaC
We investigated the role of PLC signaling cascade in the regulation of ENaC activity by UTP. We perfused 100 μm UTP for 3 min. After washout of UTP for 3 min, drugs (U-73122, U-73343, GF 109203X, PMA, and 2-APB) were applied for 6 min. UTP was added during the last 3 min of perfusion of drugs. Perfusion of U-73122 (10 μm) had no significant effect on Isc, but the magnitude of inhibition of Isc by UTP in the presence of U-73122 was reduced by 97.6 ± 0.7% (n = 8) (Fig. 4A, Table 3) compared with that observed in the absence of U-73122. This value was significantly different from the value (18.4% reduction) observed in the control experiments (Table 2). UTP response was significantly reduced by the application of 10 μm edelfosine (n = 5, 98.4% reduction) (Fig. 4C, Table 3). U-73343, GF 109203X, PMA, and 2-APB showed no effect on Isc and did not affect the effect of UTP on Na+ absorption (Fig. 4B,D,E,F, Table 3). The sensitivities of purinergic antagonists and modulators of P2Y signaling to UTP response were summarized in Figure 5.
Dose–response relationship of RB-2 and U-73122
Dose–response relationship of RB-2 on P2Y4 receptor was obtained while perfusing 100 μm UTP. For the control experiment, 100 μm UTP was applied for 16 min to investigate the change of Isc during the prolonged perfusion of UTP before assessing the dose–response relationship of RB-2. The initial Isc was −17.89 ± 1.20 μA/cm2 and changed to −9.96 ± 0.19 μA/cm2 at 2 min after UTP application and to −10.05 ± 0.36 μA/cm2 at the end of UTP application (n = 4) (Fig. 6A). There was no significant Isc change during the prolonged perfusion of UTP after reaching the plateau.
RB-2 inhibited UTP response in a concentration-dependent manner (n = 4 at each concentration from 0.1 to 300 μm) (Fig. 6B). The initial Isc was −18.34 ± 1.93 μA/cm2 and changed to −7.18 ± 0.94 μA/cm2 at 2 min after application of 100 μm UTP. Application of RB-2 increased Isc to −7.55 ± 1.10 (0.1 μm), −9.92 ± 0.77 (1 μm), −11.99 ± 0.87 (3 μm), −14.12 ± 0.69 (10 μm), −15.97 ± 1.02 (30 μm), −17.95 ± 1.73 (100 μm), and −17.99 ± 1.74 μA/cm2 (300 μm). The data were fitted to the Hill equation (Fig. 6C). The estimated IC50 value of RB-2 was 4.9 ± 1.0 μm.
The dose–response relationship of U-73122 was obtained in the presence of 100 μm UTP. The control experiments for the dose–response relationship of U-73122 were same to those used in the dose–response relationship of RB-2. U-73122 inhibited UTP response in a concentration-dependent manner (n = 4 at each concentration from 0.01 to 30 μm) (Fig. 6D). The initial Isc was −19.02 ± 0.27 μA/cm2 and changed to −9.47 ± 0.59 μA/cm2 at 2 min after application of 100 μm UTP. Application of U-73122 increased Isc to −9.75 ± 0.68 (0.01 μm), −12.06 ± 0.35 (0.1 μm), −13.80 ± 0.31 (0.3 μm), −15.85 ± 0.22 (1 μm), −17.40 ± 0.37 (3 μm), −18.21 ± 0.56 (10 μm), and −18.32 ± 0.61 μA/cm2 (30 μm). The data were fitted to the Hill equation (Fig. 6E). The IC50 value of U-73122 was 3.1 ± 0.2 μm.
Immunohistochemistry of P2Y4 receptor
Immunoreactive staining for P2Y4 was observed in the epithelial cells of RM, apical membrane of the stria vascularis, spiral ligament (Fig. 7A,D,E), and organ of Corti (Fig. 7A,G,H). In the organ of Corti, moderate staining was observed in the outer hair cells, inner hair cells, outer pillar cells, and Hensen's cells. The greatest intensity of staining was observed in Deiters' cells and head of outer pillar cells (Fig. 7G,H). Anti-P2Y4 antibody preabsorbed with antigenic peptide was negative for staining (Fig. 7B), and omitting the primary antibody from the procedure also showed negative results (Fig. 7C).
Discussion
Inhibition of Na+ absorption by activation of P2Y4 receptor
In this study, the presence of metabotropic P2Y receptor, which regulates the activity of amiloride-sensitive Na+ channel, was demonstrated in the epithelial cells of RM, and pharmacologic study pointed to P2Y subtype as P2Y4 receptor. UTP is not a specific agonist of P2Y4 but an agonist of several subtypes of P2Y receptors, such as rat P2Y2, P2Y4, and P2Y6 (von Kügelgen, 2006). P2X2 purinergic receptors studied by another group (King et al., 1998) would not be activated by the agonist UTP. Suramin and PPADS are also not specific antagonists of P2X and/or P2Y receptor subtype. Suramin is a potent antagonist of rat P2Y2 (Wildman et al., 2003) but not of P2Y4 (Bogdanov et al., 1998; Wildman et al., 2003). PPADS is an antagonist of rat P2X1, P2X2, P2X3, P2X5, P2X7, and P2Y1 (Ralevic and Burnstock, 1998) and of human and mouse P2Y6 (Robaye et al., 1997; Housley et al., 2002). RB-2 is a potent antagonist of rat P2Y4 receptor, which was expressed in Xenopus oocytes (Wildman et al., 2003). BzATP is a potent agonist for rat P2Y2 and an antagonist for rat P2Y4 receptor, which was expressed in Xenopus oocytes (Wildman et al., 2003). These criteria applied to the present results point to the P2Y purinergic receptor in the epithelial cells of RM as the P2Y4 subtype.
Our results agree with those reported in other epithelial cell types in several aspects. First, the activation of P2Y receptors by 100 μm UTP resulted in partial inhibition of amiloride-sensitive Na+ transport (42.9%) (Fig. 1A). This partial inhibition at this concentration has been reported in other epithelia, i.e., 76% in human bronchial epithelia (Devor and Pilewski, 1999), 40% in normal human airways (Inglis et al., 1999), and 49.1% in mouse distal nephron (Lehrmann et al., 2002). Second, the concentration dependence of the inhibitory effect of UTP on amiloride-sensitive Na+ current had an IC50 value of 17.2 ± 1.1 μm (Fig. 2). IC50 values for concentration–response of UTP or ATP in other epithelia such as distal colon and cortical collecting duct of kidney were in the range of 0.6–30 μm (Cuffe et al., 2000; Yamamoto and Suzuki, 2002; Matos et al., 2007). Third, the inhibitory effect of UTP was not completely reversible after washout (75.5 ± 2.4% recovered). Although slow perfusion rate may be one of the reasons of the limited reversibility, this phenomenon has also been reported in other studies (Marcus et al., 2005; Matos et al., 2007). ATP- or UTP-induced inhibition of the amiloride-sensitive Na+ current was not completely reversible in other organs, i.e., ∼85 or 89% recovered in mouse cortical collecting duct cells (Cuffe et al., 2000; Lehrmann et al., 2002). Interestingly, it was reported that it took 60 min for complete washout in human cystic fibrosis airways (Mall et al., 2000).
Mechanism of P2Y4-mediated inhibition of amiloride-sensitive Na+ absorption
Involvement of PLC activation was demonstrated by the markedly reduced effect of UTP in the presence of U-73122, a PLC inhibitor, but not in the presence of U-73343, an inactive analog. Another PLC inhibitor, edelfosine, also significantly inhibited UTP response. Furthermore, we investigated whether protein kinase C (PKC) modulators affect the effect of UTP on amiloride-sensitive Na+ channel. GF 109203X, a PKC inhibitor, and PMA, a PKC activator, induced no significant effect on UTP response; therefore, it is not likely that PKC signaling branch is involved in the acute regulation of ENaC activity. This finding was consistent with another previous report (Pochynyuk et al., 2008). It has been reported that activation of PLC signaling inhibits ENaC activity either by activating PKC or depletion of the inner membrane PI(4,5)P2 content (Kunzelmann et al., 2005). The activation of PKC is known to evoke later response and to maintain long-term (up to 48 h) downregulation of ENaC by decreasing plasma membrane levels of the channel (Stockand et al., 2000; Booth and Stockand, 2003). The level of γ-ENaC decreased by PKC with a time constant of 3.7 h, and the level of β-ENaC decreased in 13.9 h. The fact that treatment of tissues with 2-APB, a cell-permeable antagonist of inositol trisphosphate (IP3) receptor and store-operated channels (Bootman et al., 2002), did not affect UTP effect on ENaC activity suggests that the involvement of the second branch of the PLC pathway (IP3 production) in response to UTP is negligible. It has been shown that 2-APB also inhibits gap junctions composed of connexin 26 or connexin 32 and modifies the transient receptor potential channel activity (Xu et al., 2005; Tao and Harris, 2007). Anionic PI(4,5)P2 is located in the inner layer of plasma membrane and interacts with ion channels or transporters (Hilgemann et al., 2001). Recently, evidences that dynamic interaction between PI(4,5)P2 and β-subunit of ENaC enhances open probability of ENaC and depletion of PI(4,5)P2 in plasma membrane by activation of metabotropic purinergic receptor decreases ENaC activity, have been reported (Kunzelmann et al., 2005; Ma et al., 2007; Pochynyuk et al., 2008). Together, UTP-mediated inhibition of ENaC activity in the epithelial cells of RM might be caused by decreased open probability of ENaC attributable to depletion of PI(4,5)P2 concentration in the plasma membrane.
Immunolocalization of P2Y4 receptors
The electrophysiological evidence of the presence of P2Y4 receptors in RM was further supported by immunohistochemical staining with which P2Y4 receptors were localized in RM. However, subcellular expression pattern could not be discerned within the extremely thin epithelia of RM by immunohistochemistry, and the technique for measuring Isc allowed no means to control the sidedness of agonist application. We could get additional findings as below. In the stria vascularis, as reported previously (Sage and Marcus, 2002), immunoreactivity for P2Y4 receptor was shown at the apical membrane of the strial marginal cells in which P2Y4 is known to inhibit K+ secretion via KCNQ1/KCNE1 K+ channel (Marcus et al., 1998; Loussouarn et al., 2003).
Interestingly, P2Y4 receptors were identified in many cellular types of the organ of Corti. The finding that Deiters' cells had the most intense immunostaining for P2Y4 was consistent with the previous report in guinea pigs (Parker et al., 2003). P2Y4 was also localized in the sensory hair cells. The inner hair cells showed moderate immunostaining in the lateral wall of cell body, and the outer hair cells showed heterogeneous distribution of P2Y4 receptors. In outer hair cells, immunoreactivity was detected more intensely at the apex around the cuticular plate that is facing endolymphatic space than at the lateral wall of cell body. This finding was inconsistent with that P2Y4 receptors were expressed only at the basal pole of the isolated outer hair cells in guinea pigs (Szücs et al., 2004). To our knowledge, the functional roles of P2Y receptors in these cells are uncertain yet.
Physiological significance of P2Y4 receptors in epithelial cells of RM
ATP is released from most cell types and functions as a natural agonist. UTP has been reported to be released from cultured cells of different tissues and suggested that it may act as a natural agonist (Lazarowski et al., 1997; Homolya et al., 2000). In the cochlea, it has been considered that nucleotide such as ATP reduces the sensitivity of sound transduction, especially under the condition of noise exposure (Munoz et al., 2001; Housley et al., 2002). When noise exposure elevates ATP levels in the scala media, there may be the increase of parasensory K+ extrusion from endolymph via ionotropic P2X receptor bounding the endolymphatic duct (Housley et al., 1998) and also the decrease of K+ transport into scala media from the stria vascularis by activation of metabotropic P2Y4 receptor (Marcus and Scofield, 2001). In view of Na+, P2Y4 receptor that is expressed in RM mediates to inhibit Na+ absorption from the endolymph, and consequently Na+ concentration in the scala media can increase. However, because the natural agonist ATP and, possibly UTP, can act on purinergic receptors widely distributed in the epithelial cells of the endolymphatic duct, the total net movement of endolymphatic Na+ would be regulated by balance of P2X and P2Y purinergic signaling.
In conclusion, P2Y4 receptor was expressed in the epithelial cells of RM. It is likely that the physiological role of P2Y4 receptor is to regulate Na+ extrusion from the endolymph to the perilymph, probably in response to noise exposure. The acute inhibition of ENaC activity by activation of P2Y4 receptor is possibly mediated by decrease of PI(4,5)P2 in the plasma membrane through PLC activation.
Footnotes
-
This work was supported by Korea Research Foundation Grant KRF-2008-313-E00383 funded by the Korean Government.
- Correspondence should be addressed to Dr. Jun Ho Lee, Department of Otorhinolaryngology, Seoul National University College of Medicine, Seoul National University Hospital, 28 Yeongon-dong, Chongro-gu, Seoul 110-744, Korea. junlee{at}snu.ac.kr