Regulation of Kir channels in bovine retinal pigment epithelial cells by ...

3 downloads 0 Views 2MB Size Report
Jul 29, 2009 - phosphatidylinositol 4,5-bisphosphate (PIP2) regeneration by ATP-de- pendent lipid kinases. ... retinal pigment epithelium; Kir7.1; phosphatidylinositol 4-kinases; ... HEPES, 3 EDTA-KOH, and 10 glucose, titrated to pH 7.4 with.
Am J Physiol Cell Physiol 297: C1001–C1011, 2009. First published July 29, 2009; doi:10.1152/ajpcell.00250.2009.

Regulation of Kir channels in bovine retinal pigment epithelial cells by phosphatidylinositol 4,5-bisphosphate Bikash R. Pattnaik1 and Bret A. Hughes1,2 Departments of 1Ophthalmology and Visual Sciences and 2Molecular and Integrative Physiology, University of Michigan, Ann Arbor, Michigan Submitted 9 June 2009; accepted in final form 29 July 2009

Pattnaik BR, Hughes BA. Regulation of Kir channels in bovine retinal pigment epithelial cells by phosphatidylinositol 4,5-bisphosphate. Am J Physiol Cell Physiol 297: C1001–C1011, 2009. First published July 29, 2009; doi:10.1152/ajpcell.00250.2009.—The inwardly rectifying K⫹ (Kir) current in mammalian retinal pigment epithelial (RPE) cells, which is largely mediated by Kir7.1 channels, is stable in cells dialyzed with MgATP but runs down when intracellular ATP is depleted. A potential mechanism for this rundown is a decrease in phosphatidylinositol 4,5-bisphosphate (PIP2) regeneration by ATP-dependent lipid kinases. Here, we used the whole cell voltage-clamp technique to investigate the membrane PIP2 dependence of Kir channels in isolated bovine RPE cells. When RPE cells were dialyzed with ATP-free solution containing PIP2 (25–50 ␮M), rundown persisted but was markedly reduced. Removal of Mg2⫹ from the pipette solution also slowed rundown, indicating that elevated intracellular Mg2⫹ concentration contributes to rundown. Cell dialysis with the PIP2 scavenger neomycin in MgATP solution diminished Kir current in a voltagedependent manner, suggesting that it acted at least in part by blocking the Kir channel. Kir current in MgATP-loaded cells was partially inhibited by bath application of quercetin (100 ␮M), phenylarsine oxide (100 ␮M), or wortmannin (50 ␮M), inhibitors of phosphatidylinositol (PI) kinases, and was completely inhibited by cell dialysis with 2 mM adenosine, a PI4 kinase inhibitor. Both LY-294002 (100 ␮M), an inhibitor of PI3 kinases, and its inactive analog LY-303511 (100 ␮M) rapidly and reversibly inhibited Kir current, suggesting that these compounds act as direct channel blockers. We conclude that the activity of Kir channels in the RPE is critically dependent on the regeneration of membrane PIP2 by PI4 kinases and that this may explain the dependence of these channels on hydrolyzable ATP. retinal pigment epithelium; Kir7.1; phosphatidylinositol 4-kinases; phosphatidylinositol 4,5-bisphosphate THE RETINAL PIGMENT EPITHELIUM (RPE) carries out a host of functions that are critical to the adjacent photoreceptors, including phagocytosis of outer segments, regeneration of photopigment, and supply of nutrients and removal of wastes. In addition, the RPE helps control the ionic composition of fluid in the subretinal space, the extracellular compartment that is bounded by the photoreceptor outer segments and the apical aspects of the RPE and Mu¨ller (radial glial) cells (33). In particular, photoreceptor function critically depends on the maintenance of subretinal K⫹ concentration within narrow limits, which is achieved by K⫹ transport mechanisms in the RPE and Mu¨ller cells. The direction and magnitude of net K⫹ transport across the RPE is determined by the coordinated activity of transporters and channels in its apical and basolateral membranes. At the

Address for reprint requests and other correspondence: B. A. Hughes, W. K. Kellogg Eye Center, Univ. of Michigan, 1000 Wall St., Ann Arbor, MI 48105 (E-mail: [email protected]). http://www.ajpcell.org

apical membrane, operating in parallel with the Na⫹-K⫹-2Cl⫺ cotransporter and Na⫹-K⫹ pump, a large K⫹ conductance provides a major pathway for K⫹ efflux. Composed mainly of weak inwardly rectifying (Kir) channels, the apical K⫹ conductance is well suited to sustain large K⫹ fluxes necessary to buffer the K⫹ concentration outside the photoreceptor outer segments, support Na⫹-K⫹-2Cl⫺ cotransporter and Na⫹-K⫹ pump activities by recycling K⫹, and maintain the resting membrane potential (15, 16, 44). The K⫹ conductance in the apical membrane of the RPE is believed to be composed largely of Kir7.1 channels (18, 31, 44). In a previous whole cell patch-clamp study of bovine RPE cells, we reported that the Kir current is dependent on ATP hydrolysis (15). Several observations, including the requirement of millimolar ATP in the pipette solution for sustained Kir current and the failure of ATP␥S or protein phosphatase inhibitors to slow Kir current rundown, led to the conclusion that this dependency on ATP is not likely related to channel phosphorylation by protein kinases (15). Several recent studies have shown that, in common with many other ion channels (13, 17, 35), Kir7.1 channels expressed in heterologous systems are regulated by membrane phospholipids, principally phosphatidylinositol 4,5-bisphosphate (PIP2) (4, 28). The concentration of PIP2 in membranes is determined by the balance between its synthesis by multiple phosphatidylinositol (PI) kinases and its hydrolysis by lipid phosphatases and lipases. Because the rate of PIP2 synthesis by PI kinases is linked to the intracellular concentration of its substrate MgATP, it seemed possible that the rundown of the RPE Kir channel in the absence of MgATP results from the depletion of membrane PIP2. In the present study, we investigated the dependence of Kir current in the RPE on PIP2 by determining the effect of altering PIP2 levels directly by intracellular dialysis with exogenous PIP2, by disrupting PIP2-protein interactions with neomycin, and indirectly through the inhibition of lipid kinases. The results provide the first evidence that Kir channels in the RPE are modulated by alterations in the level of PIP2 and suggest that the MgATP requirement of these channels for sustained activity is related at least in part to PIP2 regeneration. Some aspects of this study were presented previously in abstract form (25). MATERIALS AND METHODS

Cell isolation. Bovine eyes obtained from local slaughter house were transported to the laboratory in ice-cold HEPES Ringer solution (see solutions), with approval of the University of Michigan Institutional Animal Care and Use Committee. RPE cells were isolated essentially as described previously (15). Briefly, pieces of RPEchoroid were incubated in a Na⫹-, Ca2⫹-, and Mg2⫹-free solution [in mM: 135 N-methyl-D-glucamine (NMDG) chloride, 5 KCl, 10 HEPES, 3 EDTA-KOH, and 10 glucose, titrated to pH 7.4 with NMDG-free base] containing activated papain (25 U/ml, Worthington

0363-6143/09 $8.00 Copyright © 2009 the American Physiological Society

C1001

C1002

MODULATION OF Kir CURRENT IN THE RPE BY PIP2

Biochemical, Lakewood, NJ) for 30 min at 37°C, washed in HEPES Ringer solution containing 1% bovine serum albumin, and then gently triturated. Isolated RPE cells were stored in HEPES Ringer solution containing 0.5 mM taurine and 1 mM reduced glutathione at 4°C and used within 8 h. Solutions. The standard bathing solution (HEPES Ringer solution) consisted of (in mM): 135 NaCl, 5 KCl, 10 HEPES, 10 glucose, 1.8 CaCl2, and 1 MgCl2, and was titrated to pH 7.4 with NaOH. The composition of Cs⫹ Ringer solution was identical except that the concentration of NaCl was reduced to 115 mM and 20 mM CsCl was added. Phenylarsine oxide (PAO), LY-294002, LY-303511, and wortmannin were dissolved in DMSO and added to bathing solutions as concentrated stock solutions. The final concentration of DMSO did not exceed 0.1%. Quercetin was dissolved in a dilute solution of NaOH before being added to bathing solutions, and the final solution was titrated to pH 7.4 with HCl. All bath solutions included 100 ␮M Gd3⫹ to block nonselective cation currents. Pipette solutions used in this study are listed in Table 1. All chemicals and drugs were obtained from Sigma-Aldrich (St. Louis, MO) except for LY-294002 (Cayman Chemicals, Ann Arbor, MI), LY-303511 (Axxora, San Diego, CA), PIP2 (Avanti Polar, Alablaster, AL), and phosphatidylinositol-4,5-bisphosphate C8 (DiC8 PIP2; Cayman Chemicals, Ann Arbor, MI). Lipids were stored as 1 mM stock solutions in water at ⫺80°C in glass vials and sonicated on ice for 30 min in pipette solution before use. A fresh aliquot of lipid stock solution was used the day of the experiment. Free Mg2⫹ and free Ca2⫹ concentrations were calculated using the MaxChelator program at www.stanford.edu/⬃cpatton. Electrophysiology. Cells were plated onto the glass coverslip bottom of a recording chamber and superfused continuously (⬃3–5 chamber volumes/min) with bathing solution at room temperature. Solutions were delivered to the chamber by gravity feed and changed by a fluid distribution system consisting of a slider valve connected to two six-way valves. Membrane currents were recorded in the whole cell recording configuration using an Axopatch 200 patch-clamp amplifier and Digidata 1322A analog-to-digital converter (Molecular Devices, Sunnyvale, CA). Voltage clamping and data acquisition were controlled by pCLAMP 9 software (Molecular Devices). Patch pipettes were pulled from thinwalled glass capillaries (PG-52165, World Precision Instruments, Sarasota, FL) using a puller (P-97, Sutter Instrument, Novato, CA) and fire polished to a resistance of 1.5–3 M⍀. Immediately after gigaseal formation and rupture of the membrane patch by gentle suction, whole cell currents were evoked from a holding potential of ⫺10 mV by 1-s voltage ramps from ⫺160 mV to 40 mV every 10 s to allow the monitoring of time-dependent changes in the current-voltage relationship. Kir current was measured as the

Table 1. Composition of pipette solutions MgATP

K-gluconate KCl KF HEPES-KOH CaCl2 Free Ca2⫹ MgCl2 Free Mg2⫹ EGTA-KOH EDTA-KOH K2-ATP Na3VO4 K4P2O7 Mannitol pH

83 30 10 0.5 1.8 ⫻ 10⫺5 4 0.57 5.5

0 ATP

83 30 10 0.5 2.0 ⫻ 10⫺5 4 3.4 5.5

0 Mg

FVPP

83 30

40 30 4 10

10 0.5 4.1 ⫻ 10⫺6

5.5 5.5

4

7.2

7.2

7.2

Pipette solutions are in mM. AJP-Cell Physiol • VOL

3 10 35 7.2

component of whole cell current that was inhibited by 20 mM extracellular Cs⫹, an approach that has been validated by us previously (15). The time course of Kir current was reconstructed by sampling the current at the reversal potential of the Cs⫹-insensitive current. All voltages were corrected for a liquid junction potential at the pipette tip estimated to be ⫺10 mV (15). Recordings were analyzed off-line using Clampfit 9 in conjunction with Microsoft Excel or Sigmaplot 10 (Systat Software, Chicago, IL). Data are presented as means ⫾ SE. Comparisons between two experimental conditions were evaluated by a Student’s t-test. Comparisons involving multiple conditions were evaluated by ANOVA with a Tukey post hoc test. RESULTS

Evidence for the dependence of Kir channel activity on PIP2. We recorded whole cell currents in solitary bovine RPE cells and isolated Kir current as the Cs⫹-sensitive component. Kir current ran down rapidly when bovine RPE cells were recorded with ATP-free (0 ATP) pipette solution containing 4 mM total Mg (3.4 mM free Mg2⫹; Table 1). Figure 1A shows the time course of outward Kir current in a representative experiment. During the first 40 s after breaking into the cell, there was an increase in Kir current, possibly resulting from the washout of inhibitory factors, such as polyamines, from the cytoplasm. An initial current increase was observed in many cells and with all pipette solutions used in this study (Table 1). After the initial rise, Kir current declined over the next several minutes, with a half-time (t1/2) of ⬃2 min. Figure 1B plots the current-voltage (I-V) relationships of the whole cell current measured when Kir current reached a peak at t ⫽ 40 s, after 5 min of dialysis, and during exposure of the cell to 20 mM extracellular Cs⫹, and shows a dramatic decrease in inwardly rectifying K⫹ current. Comparison of I-V curves of Cs⫹-sensitive current (Fig. 1C) shows that Kir current rundown was nearly complete after 5 min of dialysis and that both inward and outward currents were affected. The Cs⫹-insensitive current (Fig. 1B, trace c), which is likely mediated by Cl⫺ and Na⫹ channels, varied among cells and sometimes exhibited small time-dependent changes that resulted in a small shift in the reversal potential (Vr) of the Cs⫹-sensitive current. Qualitatively similar results were obtained in six other cells dialyzed with 0 ATP solution, with the t1/2 of Kir current rundown averaging 2.3 ⫾ 0.4 min (mean ⫾ SE) and the amplitude of Kir current remaining after 5 min of dialysis averaging 29.8 ⫾ 8.8% of its peak value (Fig. 1G, closed circles). The reversal potential of the rundown current ⫺78.7 ⫾ 1.3 mV (means ⫾ SE, n ⫽ 9) was within a few millivolts of EK (approximately ⫺82 mV), indicating that ATP depletion had minimal effects on other currents. In contrast, when RPE cells were dialyzed with pipette solution containing the same solution plus 4 mM ATP (0.6 mM free Mg2⫹, Table 1), Kir current increased during the first 5 min (Fig. 1, D–F) and then remained stable for 20 min or more (results not shown). In six cells, the amplitude of Kir current after 5 min of dialysis with 4 mM ATP solution averaged 123 ⫾ 1.7% of its value at 1 min. These results are similar to those reported in our previous study (15) in which we showed that hydrolyzable ATP was required for sustained Kir current in bovine RPE cells. Recent studies have demonstrated that, in common with KCNQ and other Kir channels, heterologously expressed Kir7.1 channels are gated by PIP2 (4, 28). To determine

297 • OCTOBER 2009 •

www.ajpcell.org

MODULATION OF Kir CURRENT IN THE RPE BY PIP2

C1003

Fig. 1. Effect of internal dialysis with ATP-free pipette solution. A: time course of changes in currents measured at ⫺26 [inward rectifying K⫹ (Kir) current] and ⫺81 mV (residual current) in a cell dialyzed with ATP-free pipette solution containing 3.4 mM free Mg2⫹. Open box indicates exposure of cell to 20 mM Cs⫹ and filled circles represent time points at which current-voltage (I–V) curves are presented in B and C. B: I–V curves recorded in the same cell as depicted in A at t ⫽ 40 s (a), t ⫽ 5 min (b), and in the presence of Cs⫹ (c). C: I–V curves of Cs⫹-sensitive currents calculated from the data in B. D: time course of changes in currents measured at ⫺27 (Kir current) and ⫺80 mV (residual current) in a cell dialyzed with pipette solution containing MgATP pipette solution. E: I–V curves recorded in the same cell as depicted in D at t ⫽ 1 min (a), t ⫽ 5 min (b), and in the presence of Cs⫹ (c). F: I–V curves of Cs⫹-sensitive currents calculated from the data in E. G: average time courses of normalized Kir current in cells dialyzed with ATP-free pipette solution (n ⫽ 9) or MgATP solution (n ⫽ 6). Symbols and error bars represent means and SE, respectively. Where they are not visible, the error bars are smaller than the size of the symbol.

whether PIP2 contributes to the gating of Kir channels in the RPE, we tested whether exogenous PIP2 could sustain Kir current during ATP depletion. Figure 2, A–C, shows the results of a representative experiment in which an RPE cell was dialyzed with ATP-free pipette solution containing 50 ␮M PIP2. Kir current increased during the first few minutes of recording and then slowly declined with a t1/2 of ⬃5 min. The reversal potential of the rundown current in the presence of PIP2 was close to EK (⫺82.6 ⫾ 1.1 mV, n ⫽ 7), indicating that the main effect on whole cell current was a decrease in K⫹ current. Similar results were obtained in six other cells dialyzed with 25 or 50 ␮M PIP2 (Fig. 2D). The summary data in Fig. 2E show that the percentage of Kir conductance remaining AJP-Cell Physiol • VOL

after 5 min of dialysis was greater with ATP-free solution containing PIP2 (n ⫽ 6) than with ATP-free solution alone (n ⫽ 9; P ⬍ 0.05; ANOVA). We also dialyzed seven cells with ATP-free solution containing 100 or 200 ␮M DiC8 PIP2, a more water-soluble short-chain PIP2 analog that partitions into the membrane less readily. Although the average Kir conductance at 5 min was somewhat larger in cells dialyzed with DiC8 PIP2-containing solution than in control cells, this difference was not statistically significant (Fig. 2E). We considered the possibility that the presence of 3.4 mM free Mg2⫹ in the ATP-free pipette solution (Table 1) might contribute to Kir current rundown. Although intracellular Mg2⫹ acts as a voltage-dependent blocker of some Kir channels (12, 22, 24), the

297 • OCTOBER 2009 •

www.ajpcell.org

MODULATION OF Kir CURRENT IN THE RPE BY PIP2

C

8

C1004

Fig. 2. Effect of internal dialysis with exogenous phosphatidylinositol 4,5-bisphosphate (PIP2). A: time course of changes in currents measured at ⫺22 mV (Kir current) and ⫺82 mV (residual current) in a cell dialyzed with ATP-free pipette solution containing 50 ␮M PIP2. B: I–V curves recorded in the same cell as depicted in A at t ⫽ 1 min (a), t ⫽ 11 min (b), and in the presence of Cs⫹ (c). C: I–V curves of Cs⫹-sensitive currents calculated from the data in B. D: average time course of normalized Kir current in cells dialyzed with ATP-free pipette solution containing 25 or 50 ␮M PIP2 (⫾SE, n ⫽ 7). E: percentage of Kir conductance measured at ⫺110 mV remaining after 5 min of dialysis with ATP-free pipette solution alone (control, n ⫽ 9), ATP-free pipette solution plus 25 or 50 ␮M PIP2 (n ⫽ 6), or ATP-free solution containing 100 or 200 ␮M DiC8 PIP2 (n ⫽ 7). Bars and error bars represent means and SE. *Significant differences from control (P ⬍ 0.05; ANVOA, post hoc).

fact that both inward and outward Kir currents decreased during dialysis of RPE cells with ATP-free solution (Fig. 1B) argues against this being the primary mechanism. Alternatively, elevated free Mg2⫹ could accelerate PIP2 degradation by activating phospholipases and/or lipid phosphatases, or it could disrupt Kir channel-PIP2 interactions by screening the negative charges of PIP2 (7, 13, 34). To test for a contribution of Mg2⫹ in Kir current rundown, we recorded whole cell currents in RPE cells using an ATP- and Mg2⫹-free pipette solution containing 5.5 mM EDTA (0 Mg, Table 1). Figure 3A shows the time course of Kir current in a representative cell dialyzed with 0 Mg pipette solution. After an initial increase, Kir current underwent rundown that was substantially slower than that observed in cells dialyzed with 0 ATP pipette solution containing 3.4 mM free Mg2⫹; after 5 min of dialysis, about 38% of the peak Kir current still remained (Fig. 3, B and C). In a total of 14 cells, Kir current remaining after 5 min of dialysis with 0 Mg solution averaged 51.7 ⫾ 3.9% of its value at 1 min (Fig. 3G), which is significantly greater (P ⬍ 0.05; t-test) than that obtained in cells dialyzed with 0 ATP solution (Fig. 1G; 29.8 ⫾ 8.8%). In other experiments, we explored the effect of intracellular dialysis with FVPP solution, a Mg-free solution containing phosphatase inhibitors (fluoride, vanadate, and pyrophosphate; Table 1) that stabilizes Kir7.1 currents in excised Xenopus oocyte membrane patches (14). In general, we found it difficult to achieve tight-seal whole cell recordings of RPE cells with AJP-Cell Physiol • VOL

FVPP in the pipette but, nevertheless, we were able to record successfully in eight cells. Results in cells dialyzed with FVPP were highly variable; after 10 min of recording, Kir current rundown was complete in two cells, essentially eliminated in two cells, and intermediate in four others. Figure 3, D–F, shows the results obtained in a cell in which the extent of Kir current rundown was intermediate. In eight cells dialyzed with FVPP solution, Kir current remaining 5 min after breaking into the cell averaged 66.5 ⫾ 10.1% of its value at 1 min (Fig. 3G, open circles), which is not significantly different from the results obtained with Mg2⫹-free solution (P ⬎ 0.10). We also tested whether PIP2 might be more effective when Mg2⫹ is omitted from the pipette solution. Figure 3H summarizes results of a series of experiments in which RPE cells were dialyzed with 0 Mg solution alone or with 0 Mg solution containing PIP2 or DiC8 PIP2 and shows that after 5 min of dialysis, Kir current rundown was about the same with or without PIP2 or DiC8 PIP2. The reason for the failure of PIP2 to sustain Kir current in these experiments is not clear, but it is possible that the lack of Mg2⫹ somehow interferes with the diffusion of phosphoinositides to the vicinity of the Kir channels in the RPE apical membrane or impedes their integration into the plasma membrane (37). Neomycin is an aminoglycoside with a net charge of ⫹6 that binds to PIP2 and disrupts electrostatic interactions between channels and PIP2 (7, 13, 17, 34). As shown in Fig. 4A, dialysis of an RPE cell with standard pipette solution containing MgATP plus

297 • OCTOBER 2009 •

www.ajpcell.org

C1005

C

8

MODULATION OF Kir CURRENT IN THE RPE BY PIP2

Fig. 3. Effect of internal dialysis with Mg-free pipette solution. A: time course of changes in currents measured at ⫺14 mV (Kir current) and ⫺83 mV (residual current) in a cell dialyzed with ATP- and Mg-free pipette solution buffered with 5.5 mM EDTA. B: I–V curves recorded in the same cell as depicted in A at t ⫽ 1 min (a), t ⫽ 5 min (b), and in the presence of Cs⫹ (c). C: I–V curves of Cs⫹-sensitive currents calculated from the data in B. D: time course of changes in currents measured at ⫺25 mV (Kir current) and ⫺83 mV (residual current) in a cell dialyzed with ATP- and Mg-free pipette solution containing phosphatase inhibitors (FVPP: fluoride, vanadate, and pyrophosphate). E: I–V curves recorded in the same cell as depicted in D at t ⫽ 1 min (a), t ⫽ 10 min (c), and in the presence of Cs⫹ (b). F: I–V curves of Cs⫹-sensitive currents calculated from the data in E. G: average time courses of normalized Kir current in cells dialyzed with 0 Mg pipette solution (n ⫽ 15) or FVPP solution (n ⫽ 8). Symbols and error bars represent means and SE, respectively. H: percentage of Kir conductance at ⫺110 mV remaining after 5 min dialysis with Mg-free pipette solution alone (control, n ⫽ 23), Mg-free pipette solution plus 50 ␮M PIP2 (n ⫽ 3), or Mg-free solution containing 100 or 200 ␮M DiC8 PIP2 (n ⫽ 8).

50 ␮M neomycin resulted in Kir current rundown. Similar results were observed in four other cells, with the t1/2 for neomycininduced rundown averaging 4.2 ⫾ 1 min. Inspection of I-V curves revealed that the neomycin-inhibited current reversed near EK (⫺76.1 ⫾ 2.5, n ⫽ 5), indicating that the main effect was a decrease in K⫹ conductance. The neomycin-induced inhibition of Kir current was voltage dependent, being greater for outward than for inward Kir currents (Fig. 4, B and C). Internal neomycin also produced changes in the kinetics of Kir current activation. Figure 4, D and E, compares families of whole cell currents recorded AJP-Cell Physiol • VOL

with MgATP solution (left) or with MgATP plus neomycin (right). In control recordings with MgATP solution, the isolated Kir current activated and deactivated rapidly (Fig. 4D, bottom), consistent with previous findings (15). In contrast, with neomycin, outward Kir current was virtually absent and inward current activated with a quasi-instantaneous component followed by a slow relaxation (Fig. 4E, bottom). Figure 4, F and G, summarizes the effects of neomycin on steady-state inward and outward Kir currents measured at the end of 1-s voltage steps. Inward current was inhibited by 80%, whereas outward current was inhibited by

297 • OCTOBER 2009 •

www.ajpcell.org

C1006

MODULATION OF Kir CURRENT IN THE RPE BY PIP2

Fig. 4. Effect of internal dialysis with neomycin. A: time course of changes in currents measured at ⫺29 mV (Kir current) and ⫺80 mV (residual current) in a cell dialyzed with MgATP pipette solution containing 50 ␮M neomycin. B: I–V curves of whole cell currents in the same cell as depicted in A at t ⫽ 1 min (a), t ⫽ 10 min (b), and in the presence of 20 mM Cs⫹ (c). C: I–V curves of Cs⫹-sensitive currents calculated from the data in B. Note that outward Kir current was more strongly inhibited than was inward Kir current after 10 min of dialysis with neomycin. D: family of whole cell currents recorded in a cell dialyzed with MgATP pipette solution. Top: HEPES-buffered Ringer (HR) in bath; middle: 20 mM Cs⫹ Ringer in bath; bottom: Cs⫹-sensitive current obtained by taking the difference between currents recorded in the absence and presence of Cs⫹. The horizontal line to the left of the current families indicates the zero-current level. E: families of whole cell currents recorded in a cell dialyzed with MgATP pipette solution containing 50 ␮M neomycin. Currents are presented at a higher gain than in D because of their smaller amplitude. F: inward and outward Kir currents measured at 0 and ⫺160 mV, respectively, after dialysis of cells for ⬃10 min with MgATP pipette solution (n ⫽ 6) or with MgATP pipette solution containing 50 ␮M neomycin (n ⫽ 5). Bars and error bars represent means and SE. The bar and error bar for the outward current of the neomycin group are barely visible. G: ratios of inward to outward Kir current calculated from the data in F, showing stronger inward rectification for cells dialyzed with MgATP pipette solution containing neomycin (⫹neo; n ⫽ 5) than with MgATP pipette solution alone (n ⫽ 6).

95% (Fig. 4F) and this difference resulted stronger inward rectification compared with control (Fig. 4G). A simple interpretation of these voltage-dependent effects is that neomycin is drawn into the pore at positive potentials, blocking outward K⫹ flux, and is expelled from the channel at negative potentials, allowing inward K⫹ flux (20). Assuming that a 1-s voltage step to ⫺160 mV is AJP-Cell Physiol • VOL

sufficient to fully relieve channel block by neomycin, the inhibition of Kir current that remained (⬃80%) can probably be attributed to the disruption of PIP2-Kir channel interactions. Inhibitors of PIP2 synthesis cause Kir current rundown. Membrane PIP2 is regenerated from PI by phosphatidylinositol (PI)4 and PI5 kinases, which catalyze the synthesis of PIP2

297 • OCTOBER 2009 •

www.ajpcell.org

MODULATION OF Kir CURRENT IN THE RPE BY PIP2

from PI by phosphorylation of the inositol ring at the 4 and 5 positions at the expense of ATP hydrolysis. To determine the dependence of Kir channel function on PI kinase activity, we tested the effects of several lipid kinase inhibitors on cells dialyzed with pipette solution containing 4 mM MgATP, which maintains Kir current (see Fig. 1G). Figure 5 shows the effect of quercetin, a flavonoid that inhibits PI4 kinases by competing with ATP at the ATP-binding site. After breaking into the cell and allowing whole cell currents to stabilize, the RPE cell was briefly exposed to 20 mM Cs⫹ to determine Kir current amplitude. Addition of 100 ␮M quercetin to the bath caused inhibition of Kir current, which decreased to a new steady state after 10 min of exposure (Fig. 5A). Comparison of the Cs⫹-sensitive currents measured before and after 15 min of exposure to quercetin revealed that Kir current had decreased by ⬃55% (Fig. 5C). In a total of five cells, exposure to quercetin for 10 min caused Kir current to decrease to 33 ⫾ 8% of its control value (Fig. 5D). Similar results were obtained with PAO and wortmannin, two other inhibitors of PI4 kinases. In six cells, exposure to 100 ␮M PAO for 10 min caused Kir current to decrease to 65 ⫾ 13% of its control value (Fig. 5D). The application of 500 nM wortmannin had little effect (not shown), but at a concentration of 50 ␮M it caused slow inhibition of Kir current, resulting in a 46 ⫾ 18% decrease from its control value after 10 min of exposure (n ⫽ 4, Fig. 5D). The effects of these agents were not rapidly reversible. Because quercetin, PAO, and wortmannin inhibit PI3 kinases as well as PI4 kinases, we also tested the effect of LY-294002, a specific PI3 kinase inhibitor (40). Figure 6, A–C,

C1007

shows that exposure of an RPE cell to 100 ␮M LY-294002 produced a rapid inhibition of Kir current (13 ⫾ 3% of control, n ⫽ 4, Fig. 6F), an effect that was nearly completely reversed upon washout of the drug (Fig. 6A). LY-294002 also inhibited an outward component of the residual current, which accounts for the negative slope region in the I-V relationship of the Cs⫹-sensitive current (Fig. 6C). Reversible inhibition (22 ⫾ 9% of control, n ⫽ 3) was also produced by 100 ␮M LY303511 (Fig. 6, A, D, E), a structurally related compound that does not inhibit PI3 kinase, suggesting that the rapid inhibitory effect of LY-294002 reflects a direct block of the Kir channel. The fact that the inhibition of Kir current by LY-294002 on Kir current was rapidly and nearly fully reversible after ⬎5 min of exposure suggests that inhibition of PI3 kinase in itself has no effect on Kir channel activity in the RPE. We also tested the effect of intracellular adenosine, an inhibitor of PI4 kinase (6). In cells dialyzed with pipette solution containing MgATP plus 2 mM adenosine, Kir current underwent rundown after an initial increase (Fig. 7, A–C). Similar results were obtained in five other cells and are summarized in Fig. 7D. After 10 min of dialysis with 2 mM adenosine, Kir current of the cells decreased to 23 ⫾ 12% of its peak value, whereas in cells dialyzed with 4 mM MgATP alone, Kir current increased by 23 ⫾ 1.7% (Fig. 1G). PI4 kinase type II is potently inhibited by millimolar adenosine, whereas PI4 kinase type III is less sensitive (3), suggesting that the effects of adenosine on RPE Kir current were likely the result of inhibition of PI4 kinase type II.

Fig. 5. Effects of phosphatidylinositoal (PI) kinase inhibitors on Kir current. A: time course of changes in current measured at ⫺38 mV (Kir current) and ⫺80 mV (residual current) produced by the exposure of a cell recorded with MgATP solution in the pipette to 100 ␮M quercetin in the bath. B: I–V curves recorded in the same cell as depicted in A in the presence of 20 mM Cs⫹ at t ⫽ 1.5 min (a), in HR at t ⫽ 3 min (b), after exposure to 100 ␮M quercetin for 10 min (c), and during a second exposure to 20 mM Cs⫹ (d). C: I–V curves of Cs⫹-sensitive currents calculated from the data in B. D: percent control Kir current remaining after 7–10 min exposure to 100 ␮M quercetin (querc, n ⫽ 5), 100 ␮M PAO (n ⫽ 6), or 50 ␮M wortmannin (wort, n ⫽ 5). AJP-Cell Physiol • VOL

297 • OCTOBER 2009 •

www.ajpcell.org

C1008

MODULATION OF Kir CURRENT IN THE RPE BY PIP2

Fig. 6. Effect of PI3 kinase inhibitor LY-294002. A: time course of changes in currents measured at ⫺43 mV (Kir current) and ⫺80 mV (residual current) in response to bath exposure of a cell to 100 ␮M LY-294002 and 100 ␮M LY-303511, an inactive LY-294002 analog. Pipette contained MgATP solution. B: I–V curves recorded in the same cell as depicted in A in the presence of 20 mM Cs⫹ at t ⫽ 3.3 min (a), in HR at t ⫽ 5 min (b), after exposure to 100 ␮M LY-294002 for 10 min (c), and after washout of LY-294002 (d). C: I–V curves of Cs⫹-sensitive currents calculated from the data in B. D: I–V curves recorded in the same cell as depicted in A in the presence of HR at t ⫽ 21 min (d), after exposure to 100 ␮M LY-301511 for 8 min (e), and after washout of LY-303511 (f). E: I–V curves of Cs⫹-sensitive currents calculated from the data in D. F: percent control Kir current at ⫺160 mV remaining after exposure to 100 ␮M LY-294002 for 5.5–13 min (n ⫽ 4) or 100 ␮M LY-303511 for 3.7– 8 min (n ⫽ 3) and following washout.

Taken together, the results obtained with several structurally unrelated PI kinase inhibitors suggest that PI4 kinase type II, and, to a lesser extent, PI4 kinase type III, play important roles in supporting Kir channel activity in the RPE by maintaining the basal level of PIP2 in the plasma membrane. DISCUSSION

The aim of this study was to determine the basis for the dependence of the Kir conductance in the RPE, which is largely composed of Kir7.1 channels, on intracellular ATP. In a previous study (15), we reported that the depletion of ATP from bovine RPE cells by metabolic inhibition or whole cell dialysis with ATP-free solution resulted in the rapid rundown of Kir current and that the requirement for ATP was specific for hydrolyzable nucleotide triphosphates. The underlying mechanism, however, was not determined. Collectively, the results of the present study suggest that the ATP dependence of the Kir conductance in the RPE stems from the role of ATP as a necessary substrate in the regeneration of PIP2 by PI4 kinases and the requirement of PIP2 for Kir channel activity. To our knowledge, these findings are the first to report a role of PIP2 in the regulation of Kir7.1 channel activity in a native epithelial cell. Dependence of RPE Kir conductance on PIP2. We confirmed our previous finding that intracellular dialysis with ATP-free pipette solution containing millimolar free Mg2⫹ AJP-Cell Physiol • VOL

results in substantial rundown of Kir current (15). Rundown was significantly slowed when the pipette solution contained 25–50 ␮M PIP2, such that more than 50% of Kir current remained after 5 min of recording. The fact that cell dialysis with PIP2 did not prevent rundown in ATP-depleted cells suggests that the rate of degradation of PIP2 by phospholipases and/or lipid phosphatases exceeded the rate at which exogenous PIP2 was incorporated into the plasma membrane. Several factors may have impeded the arrival of PIP2 to the RPE apical membrane, where Kir7.1 channels reside (31, 44), including the diffusion limitations of vesicles (in the case of PIP2) and the incorporation of phospholipid in intracellular membranes. Nevertheless, the ability of exogenous PIP2 to slow the rundown of Kir current in ATP-depleted cells is consistent with the idea that this phospholipid plays an important role in maintaining Kir channel activity in the RPE. This does not exclude the possibility that some other ATP-dependent mechanism also plays a role in maintaining Kir channel activity. In the last decade, it has become recognized that, in addition to being the precursor of the second messengers inositol 1,4,5-trisphosphate (IP3) and diacylglycerol, PIP2 is an important regulator of ion channels (36). Members of the Kir channel family were among the first channels demonstrated to be gated by PIP2 (10, 36). More recently, Kir7.1 channels, which are thought to make a major contribution to the Kir conductance in

297 • OCTOBER 2009 •

www.ajpcell.org

MODULATION OF Kir CURRENT IN THE RPE BY PIP2

C1009

Fig. 7. Effect of internal dialysis with adenosine. A: time course of changes in currents measured at ⫺20 mV (Kir current) and ⫺76 mV (residual current) in a cell dialyzed with MgATP pipette solution containing 2 mM adenosine. B: I–V curves recorded in the same cell as depicted in A at t ⫽ 1 min (a), t ⫽ 12 min (b), and in the presence of Cs⫹ (c). C: I–V curves of Cs⫹-sensitive currents calculated from the data in B. D: average time course of Kir current in cells dialyzed with 4 mM MgATP pipette solution plus 2 mM adenosine (n ⫽ 6).

the RPE (18, 31, 44), were shown to be gated by PIP2 and, to a lesser extent, PI(3,4,5)P3 (28). Thus it seems likely that the dependence of the RPE Kir conductance on PIP2 stems from a direct interaction between Kir7.1 channels and this phosphoinositide. We cannot exclude the possibility, however, that changes in some other PIP2-dependent mechanism, such as vesicle trafficking (9, 32) or cytoskeleton organization (38), contributes to the inhibition of RPE Kir conductance in response to PIP2 depletion. Role of Mg2⫹. We found that Kir current rundown was slowed when the 0 ATP pipette solution contained EDTA and no added Mg2⫹, suggesting that intracellular Mg2⫹ accelerates Kir channel rundown in the absence of ATP. There are several ways in which Mg2⫹ might act. Some ion channels, such as Kir2.1, are directly blocked by intracellular Mg2⫹ in a voltagedependent manner (24). The fact that both inward and outward Kir current declined slowly over several minutes, however, argues against this mechanism. Recent studies on TRPM7 (17) and KCNQ2/KCNQ3 channels (34) indicated that Mg2⫹ inhibits these PIP2-gated channels by electrostatic binding to the negative charges of PIP2. If disruption of PIP2-channel interactions were the predominant mechanism by which Mg2⫹ caused rundown of Kir current, one might have expected the intracellular dialysis of RPE cells with 0 Mg solution to cause an increase in current, as has been observed for other channels (34). Instead, we observed only a transient increase followed by slow rundown. Alternatively, Mg2⫹ may promote rundown by stimulating the degradation of PIP2 by lipid phosphatases (1). Although we found that dialysis of some cells with MgAJP-Cell Physiol • VOL

free pipette solution containing lipid phosphatase inhibitors (FVPP) eliminated rundown, in other cells rundown persisted. Hence, the present results do not allow us to specify the degree to which the stimulation of PIP2 hydrolysis by lipid phosphatases contributes to the Mg2⫹-induced inhibition of Kir current in ATP-depleted RPE cells. Effects of neomycin. Neomycin is a polycationic aminoglycoside that inhibits a variety of channels by disrupting their electrostatic interaction with PIP2 (13, 17, 30, 34). In general, the sensitivity of Kir channels to inhibition by neomycin appears to be correlated with channel’s affinity for PIP2 (27), with IC50 values in the range of 10 –100 ␮M (30). We found that intracellular dialysis of RPE cells with 50 ␮M neomycin resulted in strong inhibition of Kir current. Importantly, we determined that intracellular neomycin inhibited Kir current more strongly at positive than at negative potentials, resulting in stronger inward rectification, and that it slowed Kir current activation and deactivation. These results are consistent with a voltage-dependent block by neomycin, an effect reported previously for channels underlying ATP-induced currents in guinea pig outer hair cells (19), rat skeletal muscle Na⫹ channels (45), ryanodine receptors (43), and inwardly rectifying K⫹ channels in RBL cells (17). In the present study, steady-state Kir currents at hyperpolarized potentials, where the voltage-dependent block by neomycin is relieved, were about fivefold smaller when compared with control, suggesting that neomycin also inhibits Kir channels by disrupting PIP2-Kir channel interactions. Dependence of Kir channel activity on PI4 kinases. The rapid rundown of Kir current in ATP-depleted RPE cells is

297 • OCTOBER 2009 •

www.ajpcell.org

C1010

MODULATION OF Kir CURRENT IN THE RPE BY PIP2

Fig. 8. Schematic diagram depicting the gating of RPE Kir channels by their interaction with membrane PIP2, the degradation of PIP2 to PI by lipid phosphatases (PP-ase), and its regeneration by PI4 and PI5 kinases (PI kinases). Also depicted is the disruption of channel-PIP2 interactions via the screening of PIP2 negative charges by polyvalent cations (⫹⫹).

consistent with the idea that, under basal conditions, plasma membrane PIP2 undergoes significant turnover, with its degradation by lipid phosphatases and phospholipases compensated by its regeneration by lipid kinases (Fig. 8). PI4 kinases catalyze the first reaction in the multistep synthesis of PIP2 from PI. In mammalian cells, four PI4 kinase isoforms have been identified: PI4 kinase type II␣, PI4 kinase type II␤, PI4 kinase type III␣, and PI4 kinase type III␤ (2). Type II PI 4 kinases in mammalian cells are membrane-bound proteins that are present mainly in intracellular membranes but which also account for a significant fraction of the plasma membrane PI4 kinase activity under basal conditions or after platelet-derived growth factor stimulation (2). Type III PI4 kinases, on the other hand, are responsible for production of the plasma membrane PIP2 pools after stimulation of G protein or tyrosine kinase receptors (2). We found that several inhibitors of PI4 kinases caused the rundown of Kir current when applied to RPE cells dialyzed with MgATP pipette solution. Quercetin, which inhibits all PI4 kinases (42) and causes the rapid depletion of PIP2 from the plasma membrane in other cell types (11, 39), inhibited Kir current by 67% when applied at a concentration of 100 ␮M. PAO, a more potent inhibitor of PI4 kinases type III (IC50 ⫽ 1–5 ␮M) than PI4 kinases type II (IC50 ⬎100 ␮M) (2), was somewhat less effective, blocking 35% of Kir current at a concentration of 100 ␮M. Finally, wortmannin, an inhibitor of PI4 kinases type III (IC50 ⫽ 100 nM) but not PI4 kinases type II (2), blocked 52% of Kir current when applied at a concentration of 50 ␮M. Because quercetin, PAO, and wortmannin are also potent inhibitors of PI3 kinases, we tested the effect of the specific PI3 kinase inhibitor LY-294002. Exposure of RPE cells to 100 ␮M LY-294002 produced a rapid (⬍1 min) and reversible inhibition of Kir current. This inhibition was likely due to a direct block of the Kir channels rather than through a PI3 kinasedependent mechanism because the structurally related compound LY-303511, which does not affect PI3 kinases, also blocked Kir currents. Our findings add the Kir channel in the RPE, presumably Kir7.1, to the list of ion channels that are blocked by LY-294002 (5, 41). Thus inhibition of PI3 kinase does not appear to affect Kir channels, at least in the short term. In support for a role of PI4 kinases in the maintenance of Kir channel activity, we found that cell dialysis with 2 mM adenosine caused nearly complete inhibition of Kir current. Adenosine is a more potent inhibitor of PI4 kinase type II (IC50 ⫽ 10 –70 ␮M) than PI4 kinase type III (IC50 ⫽ 1.5 mM) (6) but does not affect PI3 kinases. Together, our results suggest that both type II and type III PI4 kinases contribute to the regeneration of the PIP2 pool that maintains Kir channel activity in the RPE under basal conditions but that PI4 kinase type II may play a more important role. Physiological significance. The efflux of K⫹ through Kir channels in the RPE apical membrane helps minimize the AJP-Cell Physiol • VOL

decrease in subretinal K⫹ concentration at light onset that arises from changes in photoreceptor activity and also serves to support general transport function in the RPE by sustaining Na⫹-K⫹ pump activity through K⫹ recycling. Inhibition of these channels stimulates K⫹ reabsorption by reducing the recycling of K⫹ and allowing more K⫹ that enters the cell via the Na⫹-K⫹ pump to exit through K⫹ channels in the basolateral membrane (23). The findings of our present study suggest that PIP2 is an important regulatory molecule of native Kir7.1 channels in the RPE and that local PIP2 undergoes significant turnover under resting conditions. This implies that reduction of PIP2 in the apical membrane due to either PI kinase inhibition secondary to ATP depletion or stimulation of lipid phosphatases or phospholipases would lead to the stimulation of K⫹ absorption across the RPE. The RPE apical membrane contains several Gq protein-coupled receptors that signal through second messengers (IP3 and diacyglycerol) generated by phospholipase C-mediated PIP2 hydrolysis (8, 21, 23, 26, 29). Depending on the affinity of PIP2-channel interactions, phospholipase C-mediated PIP2 depletion could potentially modulate Kir channel activity in the RPE. In support of this possibility, heterologously expressed Kir7.1 channels were inhibited following activation of M1 muscarinic receptors (4) (Pattnaik and Hughes, unpublished observations), consistent with a relatively weak affinity for PIP2- Kir7.1 channel interaction. In bovine RPE, activation of apical ␣1-adrenergic (29) or P2Y2 receptors (21, 26) decreased the apical membrane K⫹ conductance and stimulated K⫹ reabsorption. From the present results, it seems reasonable to conclude that the decreases in apical K⫹ conductance associated with the activation of ␣1-adrenergic and P2Y2 receptors in the RPE reflects the inhibition of native Kir7.1 channels as a direct consequence of phospholipase C-mediated PIP2 depletion. In summary, we have shown that native Kir channels in the RPE are modulated by conditions that alter the membrane content of PIP2, consistent with a requirement of the underlying Kir7.1 channels for this phospholipid for activity. Our results also indicate that PI4 kinases play an important role in PIP2 regeneration in the RPE under resting conditions. This is a likely explanation for why the Kir conductance in the RPE shuts down under conditions of ATP depletion. ACKNOWLEDGMENTS Present address of B. Pattnaik: Dept. of Pediatrics, University of Wisconsin, 202 S. Park Street, Madison, WI 53715. GRANTS This study was spported by National Institutes of Health Grants EY-08850 and EY-07003 and a Lew Wasserman award from Research to Prevent Blindness to B. A. Hughes.

297 • OCTOBER 2009 •

www.ajpcell.org

MODULATION OF Kir CURRENT IN THE RPE BY PIP2 REFERENCES 1. Alvarez-Leefmans FJ, Giraldez F, Gamino SM. Intracellular free magnesium in excitable cells: its measurement and its biologic significance. Can J Physiol Pharmacol 65: 915–925, 1987. 2. Balla A, Balla T. Phosphatidylinositol 4-kinases: old enzymes with emerging functions. Trends Cell Biol 16: 351–361, 2006. 3. Barylko B, Gerber SH, Binns DD, Grichine N, Khvotchev M, Sudhof TC, Albanesi JP. A novel family of phosphatidylinositol 4-kinases conserved from yeast to humans. J Biol Chem 276: 7705–7708, 2001. 4. Du X, Zhang H, Lopes C, Mirshahi T, Rohacs T, Logothetis DE. Characteristic interactions with phosphatidylinositol 4,5-bisphosphate determine regulation of Kir channels by diverse modulators. J Biol Chem 279: 37271–37281, 2004. 5. El-Kholy W, Macdonald PE, Lin JH, Wang J, Fox JM, Light PE, Wang Q, Tsushima RG, Wheeler MB. The phosphatidylinositol 3-kinase inhibitor LY294002 potently blocks KV currents via a direct mechanism. FASEB J 17: 720 –722, 2003. 6. Endemann G, Dunn SN, Cantley LC. Bovine brain contains two types of phosphatidylinositol kinase. Biochemistry 26: 6845– 6852, 1987. 7. Fan Z, Makielski JC. Anionic phospholipids activate ATP-sensitive potassium channels. J Biol Chem 272: 5388 –5395, 1997. 8. Feldman EL, Randolph AE, Johnston GC, DelMonte MA, Greene DA. Receptor-coupled phosphoinositide hydrolysis in human retinal pigment epithelium. J Neurochem 56: 2094 –2100, 1991. 9. Guo J, Wenk MR, Pellegrini L, Onofri F, Benfenati F, De Camilli P. Phosphatidylinositol 4-kinase type II␣ is responsible for the phosphatidylinositol 4-kinase activity associated with synaptic vesicles. Proc Natl Acad Sci USA 100: 3995– 4000, 2003. 10. Hilgemann DW, Ball R. Regulation of cardiac Na⫹,Ca2⫹ exchange and KATP potassium channels by PIP2. Science 273: 956 –959, 1996. 11. Hirono M, Denis CS, Richardson GP, Gillespie PG. Hair cells require phosphatidylinositol 4,5-bisphosphate for mechanical transduction and adaptation. Neuron 44: 309 –320, 2004. 12. Horie M, Irisawa H, Noma A. Voltage-dependent magnesium block of adenosine-triphosphate-sensitive potassium channel in guinea-pig ventricular cells. J Physiol 387: 251–272, 1987. 13. Huang CL, Feng S, Hilgemann DW. Direct activation of inward rectifier potassium channels by PIP2 and its stabilization by G␤␥. Nature 391: 803– 806, 1998. 14. Hughes BA, Swaminathan A. Modulation of the Kir7.1 potassium channel by extracellular and intracellular pH. Am J Physiol Cell Physiol 294: C423–C431, 2008. 15. Hughes BA, Takahira M. ATP-dependent regulation of inwardly rectifying K⫹ current in bovine retinal pigment epithelial cells. Am J Physiol Cell Physiol 275: C1372–C1383, 1998. 16. Hughes BA, Takahira M. Inwardly rectifying K⫹ currents in isolated human retinal pigment epithelial cells. Invest Ophthalmol Vis Sci 37: 1125–1139, 1996. 17. Kozak JA, Matsushita M, Nairn AC, Cahalan MD. Charge screening by internal pH and polyvalent cations as a mechanism for activation, inhibition, and rundown of TRPM7/MIC channels. J Gen Physiol 126: 499 –514, 2005. 18. Kusaka S, Inanobe A, Fujita A, Makino Y, Tanemoto M, Matsushita K, Tano Y, Kurachi Functional Kir7 Y.1. channels localized at the root of apical processes in rat retinal pigment epithelium. J Physiol 531: 27–36, 2001. 19. Lin X, Hume RI, Nuttall AL. Voltage-dependent block by neomycin of the ATP-induced whole cell current of guinea-pig outer hair cells. J Neurophysiol 70: 1593–1605, 1993. 20. Lopatin AN, Makhina EN, Nichols CG. Potassium channel block by cytoplasmic polyamines as the mechanism of intrinsic rectification. Nature 372: 366 –369, 1994. 21. Maminishkis A, Jalickee S, Blaug SA, Rymer J, Yerxa BR, Peterson WM, Miller SS. The P2Y2 receptor agonist INS37217 stimulates RPE fluid transport in vitro and retinal reattachment in rat. Invest Ophthalmol Vis Sci 43: 3555–3566, 2002. 22. Matsuda H, Saigusa A, Irisawa H. Ohmic conductance through the inwardly rectifying K channel and blocking by internal Mg2⫹. Nature 325: 156 –159, 1987. 23. Miller SS, Edelman JL. Active ion transport pathways in the bovine retinal pigment epithelium. J Physiol 424: 283–300, 1990.

AJP-Cell Physiol • VOL

C1011

24. Murata Y, Fujiwara Y, Kubo Y. Identification of a site involved in the block by extracellular Mg2⫹ and Ba2⫹ as well as permeation of K⫹ in the Kir2.1 K⫹ channel. J Physiol 544: 665– 677, 2002. 25. Pattnaik BR, Hughes BA. Regulation of inwardly rectifying K⫹ (Kir) channels in the retinal pigment epithelium (RPE) by phosphatidylinositol 4,5-bisphosphate (PIP2) (Abstract). Invest Ophthalmol Vis Sci 47: 4722, 2006. 26. Peterson WM, Meggyesy C, Yu K, Miller SS. Extracellular ATP activates calcium signaling, ion, and fluid transport in retinal pigment epithelium. J Neurosci 17: 2324 –2337, 1997. 27. Rohacs T, Lopes C, Mirshahi T, Jin T, Zhang H, Logothetis DE. Assaying phosphatidylinositol bisphosphate regulation of potassium channels. Methods Enzymol 345: 71–92, 2002. 28. Rohacs T, Lopes CM, Jin T, Ramdya PP, Molnar Z, Logothetis DE. Specificity of activation by phosphoinositides determines lipid regulation of Kir channels. Proc Natl Acad Sci USA 100: 745–750, 2003. 29. Rymer J, Miller SS, Edelman JL. Epinephrine-induced increases in [Ca2⫹]in and KCl-coupled fluid absorption in bovine RPE. Invest Ophthalmol Vis Sci 42: 1921–1929, 2001. 30. Schulze D, Krauter T, Fritzenschaft H, Soom M, Baukrowitz T. Phosphatidylinositol 4,5-bisphosphate (PIP2) modulation of ATP and pH sensitivity in Kir channels. A tale of an active and a silent PIP2 site in the N terminus. J Biol Chem 278: 10500 –10505, 2003. 31. Shimura M, Yuan Y, Chang JT, Zhang S, Campochiaro PA, Zack DJ, Hughes BA. Expression and permeation properties of the K⫹ channel Kir7.1 in the retinal pigment epithelium. J Physiol 531: 329 –346, 2001. 32. Sorensen SD, Linseman DA, McEwen EL, Heacock AM, Fisher SK. A role for a wortmannin-sensitive phosphatidylinositol-4-kinase in the endocytosis of muscarinic cholinergic receptors. Mol Pharmacol 53: 827– 836, 1998. 33. Strauss O. The retinal pigment epithelium in visual function. Physiol Rev 85: 845– 881, 2005. 34. Suh BC, Hille B. Electrostatic interaction of internal Mg2⫹ with membrane PIP2 Seen with KCNQ K⫹ channels. J Gen Physiol 130: 241–256, 2007. 35. Suh BC, Hille B. Recovery from muscarinic modulation of M current channels requires phosphatidylinositol 4,5-bisphosphate synthesis. Neuron 35: 507–520, 2002. 36. Suh BC, Hille B. Regulation of ion channels by phosphatidylinositol 4,5-bisphosphate. Curr Opin Neurobiol 15: 370 –378, 2005. 37. Summers SA, Guebert BA, Shanahan MF. Polyphosphoinositide inclusion in artificial lipid bilayer vesicles promotes divalent cation-dependent membrane fusion. Biophys J 71: 3199 –3206, 1996. 38. van Rheenen J, Jalink K. Agonist-induced PIP2 hydrolysis inhibits cortical actin dynamics: regulation at a global but not at a micrometer scale. Mol Biol Cell 13: 3257–3267, 2002. 39. Varnai P, Balla T. Visualization of phosphoinositides that bind pleckstrin homology domains: calcium- and agonist-induced dynamic changes and relationship to myo-[3H]inositol-labeled phosphoinositide pools. J Cell Biol 143: 501–510, 1998. 40. Walker EH, Pacold ME, Perisic O, Stephens L, Hawkins PT, Wymann MP, Williams RL. Structural determinants of phosphoinositide 3-kinase inhibition by wortmannin, LY294002, quercetin, myricetin, and staurosporine. Mol Cell 6: 909 –919, 2000. 41. Welling A, Hofmann F, Wegener JW. Inhibition of L-type Cav1.2 Ca2⫹ channels by 2,(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002) and 2-[1-(3-dimethyl-aminopropyl)-5-methoxyindol-3-yl]-3-(1H-indol-3-yl) maleimide (Go6983). Mol Pharmacol 67: 541–544, 2005. 42. Wiedemann C, Schafer T, Burger MM. Chromaffin granule-associated phosphatidylinositol 4-kinase activity is required for stimulated secretion. EMBO J 15: 2094 –2101, 1996. 43. Xu L, Wang Y, Gillespie D, Meissner G. Two rings of negative charges in the cytosolic vestibule of type-1 ryanodine receptor modulate ion fluxes. Biophys J 90: 443– 453, 2006. 44. Yang D, Pan A, Swaminathan A, Kumar G, Hughes BA. Expression and localization of the inwardly rectifying potassium channel Kir7.1 in native bovine retinal pigment epithelium. Invest Ophthalmol Vis Sci 44: 3178 –3185, 2003. 45. Yeiser AJ, Cox JR, Wright SN. Voltage-dependent inhibition of rat skeletal muscle sodium channels by aminoglycoside antibiotics. Pflu¨gers Arch 448: 204 –213, 2004.

297 • OCTOBER 2009 •

www.ajpcell.org