Protein Tyrosine Kinase and Protein Phosphatase ... - CiteSeerX

Protein Tyrosine Kinase and Protein Phosphatase Signaling Pathways Regulate Volume-Sensitive Chloride Currents in a Nonpigmented Ciliary Epithelial Cell Line Chanjuan Shi,1,2 Steven Barnes,1,3,4 Miguel Coca-Prados,5 and Melanie E. M. Kelly1,2,4 PURPOSE. To investigate whether signaling pathways that incorporate protein tyrosine kinases and phosphatases regulate PKC-sensitive, volume-sensitive Cl⫺ currents (ICl,vol) in cultured rabbit nonpigmented ciliary epithelial cells. METHODS. Activation of ICl,vol in response to hyposmotic stimulation was recorded with whole-cell patch-clamp techniques in the presence of pharmacologic agents that activate or block kinases and phosphatases. RESULTS. ICl,vol in rabbit nonpigmented ciliary epithelial cells was identified as a PKC-sensitive, volume-sensitive Cl⫺ current, because current was downregulated during cell swelling by phorbol-12-dibutyrate, a PKC activator, and the PKC inhibitors, calphostin and chelerythrine, enhanced the current. Activation of c-Src tyrosine kinases, with an Src activator peptide (EPQ(pY)EEIPI), increased ICl,vol after hyposmotic stimulation, whereas the protein tyrosine kinase inhibitor, genistein, but not its inactive analogue daidzein, inhibited the current. The phosphatidylinositol-3-kinase (PI3K) inhibitor, wortmannin, inhibited ICl,vol. Wortmannin did not further inhibit ICl,vol in cells pretreated with the protein tyrosine kinase inhibitor, genistein, but blocked enhancement of ICl,vol by PKC inhibitors. The serine-threonine protein phosphatase (PP) inhibitor, okadaic acid, blocked activation of ICl,vol, whereas insulin, which activates PI3K and PP-1, enhanced the current. The insulin-enhanced current was also blocked by okadaic acid. ICl,vol was not activated under isosmotic conditions by the simultaneous inhibition of PKC with calphostin and activation of PP-1 by insulin. CONCLUSIONS. These data show that PKC-sensitive Cl⫺ currents activated in response to cell swelling in nonpigmented ciliary epithelial cells are modulated by protein tyrosine kinase, PI3K, and PP signaling pathways. Activation of PP and PKC may involve the upstream intermediaries Src tyrosine kinase and PI3K. (Invest Ophthalmol Vis Sci. 2002;43:1525–1532)

T

he ciliary body epithelium (CE) is responsible for aqueous humor secretion in the eye and is composed of two different cell layers: a pigmented ciliary epithelial (PCE) cell layer and nonpigmented ciliary epithelial (NPCE) cell layer. The basal membrane of PCE cells faces the stroma of the ciliary

From the 1Laboratory for Retina and Optic Nerve Research, and Departments of 2Pharmacology, 3Physiology and Biophysics, and 4Ophthalmology, Dalhousie University, Halifax, Nova Scotia, Canada; and the 5Department of Ophthalmology, Yale Medical School, New Haven, Connecticut. Supported by a Canadian Institutes of Health Research (CIHR) grant (MEMK) and a Killam studentship (CS). Submitted for publication July 24, 2001; revised December 17, 2001; accepted January 2, 2002. Commercial relationships policy: N. The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact. Corresponding author: Melanie E. M. Kelly, Department of Pharmacology, Dalhousie University, Halifax, Nova Scotia, B3H 4H7, Canada; [email protected]. Investigative Ophthalmology & Visual Science, May 2002, Vol. 43, No. 5 Copyright © Association for Research in Vision and Ophthalmology

body, whereas the basolateral surface of NPCE cells faces the aqueous humor.1 The apical membranes of these two cell layers are connected by gap junctions allowing PCE and NPCE cells to function as a syncytium in the production of aqueous humor.1,2 In this coupled cell model of aqueous humor formation, solutes and water are taken up by PCE cells from the stromal interstitium, diffuse into NPCE cells through the apical gap junctions, and are secreted at the basolateral surface of NPCE cells into the posterior chamber as aqueous humor.1,3 Although the rate and quantity of aqueous humor are subjected to both endocrine and paracrine modulation, it is likely that cell volume regulatory mechanisms in both ciliary epithelial cell types contribute to aqueous humor production.2,3 For example, Na⫹, K⫹, and Cl⫺ enter PCE cells mainly through Na⫹-K⫹-2Cl⫺ symport and diffuse from PCE to NPCE cells through gap junctions. The overload of NPCE cells with solute and water from PCE cells leads to cell volume expansion and can be mimicked in in vitro tissue preparations by using hyposmotic solution.1,2,4 Cell swelling then activates volumesensitive K and Cl channels in NPCE cells, causing efflux of K⫹ and Cl⫺ into the aqueous humor with water passively following, resulting in cell shrinkage. Cl channels in NPCE cells are thought to be a rate-limiting step in aqueous humor production as well as in volume regulation in these cells.2 Although the molecular identity of volume-sensitive Cl channels remains unresolved,5 previous studies have suggested several candidates for the volume-sensitive channel regulator/Cl channels in NPCE cells.6 –12 These include the multidrug resistant gene product (MDR), the channel regulator, ICln protein (pICln), and ClC-3, a member of the closely related ClC family which also includes ClC-4, ClC-5, and hypotonically-activated ClC-2.13 Furthermore, in addition to PKCsensitive ICl,vol described in human NPCE cells,6 a volumesensitive Cl⫺ current that showed little to no inactivation at positive potentials and that was blocked by extracellular 5 to 10 mM adenosine triphosphate (ATP), has also been reported in native bovine NPCE cells.12 Single-channel recordings from these cells indicated that several distinct volume-sensitive channels are activated during cell swelling and may contribute to the net volume-sensitive Cl⫺ current.9 Cell swelling activates a variety of second-messenger molecules, which may link volume changes to Cl channel activation.13,14 Downregulation by PKC-mediated phosphorylation is a property of volume-sensitive Cl⫺ current.15,16 Regulation of ion transporters and channels by phosphorylation and dephosphorylation during cell swelling has now been demonstrated in several studies.13,17 Although the signaling pathways may vary among different cell types, small guanosine triphosphate (GTP)-binding proteins,18,19 protein tyrosine kinases (PTK),20 –23 phosphatidylinositol-3-kinase (PI3K),17,23,24 and serine-threonine protein phosphatases (PP),16 have been individually implicated in volume regulation in response to cell swelling. This study examined the potential PTK-coupled phosphorylation and dephosphorylation pathway(s) that contribute to regulation of PKC-sensitive ICl,vol in mammalian NPCE cells. 1525

1526

Shi et al.

METHODS Materials Phorbol-12-dibutyrate (PDBu), chelerythrine chloride, calphostin, daidzein, genistein, wortmannin, okadaic acid, okadaic acid methyl ester, and insulin were purchased from Calbiochem (San Diego, CA). These agents, with the exception of insulin were prepared as stocks in dimethyl sulfoxide (DMSO). PDBu, chelerythrine chloride, daidzein, genistein, wortmannin, and insulin were further diluted in extracellular recording solution and bath superfused at the concentrations cited in results. The final DMSO concentration for these agents was 0.01% or less. Calphostin, okadaic acid, and okadaic acid methyl ester were dissolved in the intracellular recording solution at the concentrations cited in results. Src activator and control peptides were generous gifts from Dr. Michael Salter, University of Toronto (Toronto, Canada), and were diluted and applied in the intracellular solution through electrode dialysis.

Cell Culture We used NPCE cells from a cell line derived from simian virus (SV)40 –transformed rabbit NPCE cells, which have properties similar to those of native cultured rabbit NPCE cells.25 Our previous studies26 have shown that the electrophysiological properties of the SV40-transformed NPCE cells resemble those of freshly isolated rabbit NPCE cells.27 SV40-transformed NPCE cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Canadian Life Technologies, Burlington, Ontario, Canada) plus 10% newborn calf serum (NCS) and 1% gentamicin in an atmosphere of 5% CO2-95% O2 at 37°C. Before electrophysiological experiments, cells were seeded onto 12-mm glass coverslips at a density of 106 cells/mL and incubated for another 24 to 48 hours at 37°C in an atmosphere of 5% CO2-95%O2.

Whole-Cell Current Recordings We used tight-seal patch-clamp recording methods to measure wholecell currents from cultured NPCE cells. Coverslips with attached cells were placed in a 1-mL (total volume) recording chamber and positioned on a stage of a Nikon inverted microscope. Cells were continually superfused with external solution at rate of 1 to 2 mL/min. The regular isosmotic external solution contained (in millimolar): Trizma HCl (Sigma Chemical Co., St. Louis, MO), 70; CaCl2, 1.5; MgCl2, 0.8; HEPES, 10; tetraethylammonium (TEA)-Cl, 5; BaCl2, 5; glucose, 10; and sucrose, 105. The osmolarity of the external solution was 295 mOsmol/L. The hyposmotic external solution was made by omitting sucrose, which reduced the osmolarity to 191 mOsmol/L but maintained ionic strength. The pH of external solutions was adjusted to 7.4 with CsOH. The intracellular electrode-filling solution for whole-cell recording was composed of (in millimolar): Trizma HCl, 60; Trizma base, 60; aspartic acid, 60; HEPES, 15; CaCl2, 0.4; MgCl2, 1; EGTA, 1; adenosine triphosphate (ATP)-Mg, 2; and guanosine triphosphate (GTP)-Na2, 0.1. The pH of the internal solution was adjusted to 7.2, with CsOH and the osmolarity was 264 mOsmol/L. Normally the osmolarity of intracellular electrode solutions was at least 20 mOsmol/L hyposmotic to the external solution to prevent transient cell swelling during early periods of whole-cell recording. We used a software program and an amplifier (Axopatch-1D; Axon Instruments, Union City, CA) to generate voltage commands and to record membrane currents. Patch electrodes, with an external diameter of 1.5 mm and an internal diameter of 1.1 mm (Sutter Instruments, Novato, CA), were pulled from borosilicate glass using a two-stage vertical microelectrode puller (model pp83; Narishige, Tokyo, Japan). Electrodes were coated with beeswax to reduce capacitance and had resistances of 3 to 5 M⍀ when filled with internal solutions. A sealed electrode-salt bridge combination (Dri-ref-2; World Precision Instruments, Sarasota, FL) was used as the reference electrode. Before the seal was formed, offset potentials were nulled using the amplifier circuitry. Current-voltage relationships were corrected by subtracting the liquid junction potential, which was measured or calculated by

IOVS, May 2002, Vol. 43, No. 5 computer (JPCal software, ver. 2.00; P. H. Barry, Sydney, Australia) and was found to be approximately 2 mV. Series resistance and cell capacitance were obtained directly from the amplifier’s transient cancellation settings. For all recordings shown, the series resistance was less than 15 M⍀ and was compensated (80%). Experiments were conducted at room temperature (23–25°C).

Statistics Data are expressed as the mean ⫾ SEM. Differences between groups of data were compared using Student’s unpaired t-test or the MannWhitney test. Differences between groups were considered significant when P ⬍ 0.05.

RESULTS Activation of a Clⴚ Conductance by Cell Swelling We have previously identified ICl,vol in SV40-transformed rabbit NPCE cells using solutions designed to isolate Cl⫺ currents.26 Figure 1 shows a typical example of hyposmotic activation of ICl,vol. The voltage-clamp protocol used is shown in Figure 1A (left). Cells were held at ⫺62 mV and the membrane potential stepped from ⫺62 to ⫹98 mV in increments of 20 mV. Figure 1A shows representative current traces recorded in isosmotic solution (Iso), hyposmotic solution (Hypo), and recovery in isosmotic solution again (Rec). In isosmotic solution with regular Na⫹- and K⫹-free solutions, the whole-cell current was very slight at both hyperpolarized and depolarized potentials. Superfusion of cells with hyposmotic solution for approximately 30 minutes activated a large current. Recovery of cells in isosmotic solution for 10 minutes produced a return of the whole-cell current to basal levels. Figure 1B shows the current– voltage (I-V) relationship curves for the cell shown in Figure 1A. The current recorded in the hyposmotic solution shows slight outward rectification and reverses at a membrane potential of ⫺12 mV, which is close to the calculated equilibrium potential for Cl⫺ ions (ECl ⫽ ⫺10 mV).

Effect of Alterations in PKC Activity on ICl,vol PKC is involved in the modulation of ICl,vol.15–16,28,29 In human NPCE cells, PKC has been implicated in inhibition of ICl,vol.6 To confirm PKC regulation of ICl,vol in rabbit NPCE cells, we used a PKC activator, PDBu. Whole-cell currents were recorded in isosmotic and hyposmotic solution in the absence or presence of 100 nM PDBu. Cells were preteated with 100 nM PDBu for 15 minutes before recording. Figure 2A shows representative current traces of ICl,vol recorded in hyposmotic solution in the absence (Hypo) and presence (Hypo⫹PDBu) of 100 nM PDBu. ICl,vol was inhibited in the presence of 100 nM PDBu. Figure 2B shows the mean I-V relationships for currents recorded from PDBu-treated (n ⫽ 6) and control cells (n ⫽ 5). ICl,vol, in this and subsequent figures, is represented as the difference current obtained by subtracting the current recorded in isosmotic solution from the maximum current recorded in hyposmotic solution (usually after 20 minutes of hyposmotic exposure). ICl,vol was decreased by incubation of cells with 100 nM PDBu from 41.95 ⫾ 8.55 pA/pF to 2.27 ⫾ 1.93 pA/pF at ⫹58 mV (P ⬍ 0.01) and from ⫺14.38 ⫾ 2.69 pA/pF to ⫺1.48 ⫾ 0.63 pA/pF at ⫺62 mV (P ⬍ 0.05). The inhibitory actions of PKC activation on ICl,vol in the rabbit NPCE cell line were also supported by experiments examining the effect of two specific PKC inhibitors, chelerythrine chloride and calphostin, on ICl,vol (Fig. 3). Cells were pretreated with 1 ␮M chelerythrine chloride for 30 minutes and superfused with chelerythrine throughout the recordings. The effect of chelerythrine on ICl,vol is shown in Figure 3A. At ⫹58 and ⫺62 mV, ICl,vol was increased by 168% and 123% in 1 ␮M chelerythrine-treated cells (n ⫽ 6) compared with control

IOVS, May 2002, Vol. 43, No. 5

Volume-Sensitive Chloride Currents in Ciliary Epithelial Cells

1527

FIGURE 1. Activation of ICl,vol by cell swelling. (A) Whole-cell currents recorded in isosmotic solution (left), after 25 minutes in hyposmotic solution (middle), and after 10 minutes in isosmotic solution again (right). Top left: voltage protocol. (B) I-V relationship for whole-cell currents recorded from the cell shown in (A). Current was measured 100 ms after the initiation of the voltage step in isosmotic solution (Iso), hyposmotic solution (Hypo), and in isosmotic solution again (Rec).

(n ⫽ 5) cells (P ⬍ 0.01 at ⫹58 mV, P ⬍ 0.05 at ⫺62 mV). Similar findings were obtained with another PKC inhibitor, calphostin (Fig. 3B). In comparison with the control group (n ⫽ 5), inclusion of 0.25 ␮M calphostin in internal solution (n ⫽ 6) significantly increased ICl,vol by 92% and 147% at ⫹58 and ⫺62 mV, respectively (P ⬍ 0.05). Consistent with the absence of effect of PDBu on basal Cl⫺ current, neither calphostin nor

FIGURE 2. PKC activation inhibited ICl,vol. (A) Representative current traces recorded from an NPCE cell after 30 minutes of hyposmotic stimulation (left). Right: currents recorded in another cell after hyposmotic stimulation in the presence of 100 nM of the PKC activator, PDBu. Top right: Voltage protocol. (B) I-V relationship for ICl,vol recorded from five cells in the absence and presence of 100 nM PDBu. Whole-cell currents were normalized to cell capacitance in this and subsequent figures.

chelerythrine significantly enhanced basal Cl⫺ currents in isosmotic solution (Fig. 3C).

Tyrosine Kinase Activation in ICl,vol Regulation PTKs have been reported to be involved in the activation of ICl,vol in some cell types.20 –23 We examined the effects of the tyrosine kinase inhibitor genistein on ICl,vol. The inactive analogue, daidzein, was also used to rule out the nonspecific effects of genistein. Cells were incubated with 25 ␮M genistein or 25 ␮M daidzein for 30 minutes, and both drugs were included in the superfusate during the experiments. Figure 4A shows mean I-V relationships recorded in control, genistein-, and daidzein- treated groups of cells (n ⫽ 10, 10, and 4 respectively). ICl,vol was largely suppressed by 25 ␮M genistein. The mean current amplitude at ⫺62 and ⫹58 mV decreased from ⫺10.05 ⫾ 3.8 and 28.64 ⫾ 7.16 pA/pF in the control group to ⫺1.26 ⫾ 1.60 and 6.68 ⫾ 2.30 pA/pF in the genisteintreated groups, respectively. ICl,vol was not suppressed by daidzein, indicating a selective inhibitory effect of genistein on ICl,vol and a role for tyrosine kinase–signaling pathway(s) in the regulation of volume-sensitive Cl channels in NPCE cells. Among the families of PTK molecules, the c-Src family is one of the largest and has been reported to be involved in regulation of both ligand-gated and voltage-dependent ion channel phosphorylation.30,31 To investigate whether a c-Src PTK is involved in regulation of ICl,vol during cell swelling in rabbit NPCE cells, we used an Src activator peptide, EPQ(pY)EEIPI which was delivered through intracellular dialysis. The phosphorylated Src activator peptide can bind to the Src SH2 domain and disrupt the interaction between the tyr527 and c-Src kinase domains and therefore enhance c-Src kinase activity.32 The nonphosphorylated form of the Src peptide (EPQYEEIPI) binds to the Src SH2 domain weakly and cannot activate c-Src. We used the unphosphorylated peptide to rule out the nonspecific action of the Src activator peptide. The effects of Src control and activator peptides on ICl,vol are shown in Figure 4B. Inclusion of the Src activator peptide (Src-AP, 1 mM) in the electrode (n ⫽ 6) significantly enhanced ICl,vol during cell

1528

Shi et al.

IOVS, May 2002, Vol. 43, No. 5 tyrosine kinases.34,35 To investigate whether PI3K is a downstream mediator of PTK involved in the stimulation of ICl,vol in rabbit NPCE cells, a specific PI3K inhibitor, wortmannin, was used in the absence and presence of the PTK inhibitor genistein. Cells were superfused during the experiments with100 nM wortmannin or pretreated with genistein (50 ␮M) and superfused during the experiment with wortmannin (100 nM). Figure 5A shows that the majority of ICl,vol was suppressed by wortmannin (n ⫽ 6) compared with the control group (n ⫽ 8). At ⫺62 and ⫹58 mV, ICl,vol for the wortmannintreated group decreased from ⫺11.61 ⫾ 1.34 to ⫺2.77 ⫾ 1.02 pA/pF and from 33.64 ⫾ 4.25 to 5.74 ⫾ 2.34 pA/pF (P ⬍ 0.01), respectively. In the presence of genistein, application of wortmannin did not produce any further significant inhibition of ICl,vol (n ⫽ 4, P ⬎ 0.05; Fig. 5B). In some cells, PI3K signaling pathways activate PKC.36 To determine whether PI3K could contribute to activation of PKC after cell swelling in NPCE cells, we activated ICl,vol and exposed cells to the PI3K inhibitor, wortmannin (100 nM), followed by the PKC inhibitor chelerythrine (1 ␮M). If PKC is activated downstream of PI3K after cell swelling, then we would expect wortmannin to block the enhancement of ICl,vol by the PKC inhibitor. Figure 6A shows ICl,vol measured at ⫺62 and ⫹58 mV after application of wortmannin. As described earlier, wortmannin blocked ICl,vol, and this inhibition was not decreased after application of chelerythrine. Figure 6B shows the mean data obtained from three cells, in which ICl,vol was decreased at ⫺62 and ⫹58 mV by 57% and 52% in the presence of wortmannin alone and 42% and 51% in the presence of wortmannin and chelerythrine.

FIGURE 3. PKC inhibitors enhanced ICl,vol. (A) Mean ICl,vol recorded in chelerythrine (CHT)-treated (n ⫽ 6) and control cells (n ⫽ 5) at ⫺62 and ⫹58 mV. (B) ICl,vol recorded in the presence (n ⫽ 6) or absence (n ⫽ 6) of 0.25 ␮M calphostin (Calph) in the internal recording solution at ⫺62 and ⫹58 mV. (C) Cl⫺ current recorded in control (n ⫽ 7) or CHT-pretreated (n ⫽ 9) cells in isosmotic solution at ⫺62 and ⫹58 mV (*P ⬍ 0.05; **P ⬍ 0.01).

swelling, compared with the Src control peptide (Src-CP, 1 mM; n ⫽ 11; ). ICl,vol was increased by105% and 97% at ⫺62 and ⫹58 mV, respectively (P ⬍ 0.05). We tested whether the Src-enhanced ICl,vol, recorded in the presence of Src-AP, was sensitive to PKC inhibition by treating cells with 100 nM PDBu. The Src-enhanced ICl,vol recorded from PDBu-treated cells (n ⫽ 11) was decreased by 68% and 64% at ⫺62 and ⫹58 mV, respectively (P ⬍ 0.05, data not shown). Therefore, these data suggest that tyrosine kinase–signaling pathways, including Src PTK, participate in the activation of PKC-sensitive, volumesensitive Cl channels during cell swelling in rabbit NPCE cells.

PI3K in the Regulation of ICl,vol Cell swelling can enhance PI3K activity.17,24,33 In addition, some experiments have demonstrated that PI3K is one of the potential downstream effectors of receptor or nonreceptor

FIGURE 4. PTK pathways activated ICl,vol. (A) Mean ICl,vol recorded in the absence (n ⫽ 10) or presence of genistein (GST, n ⫽ 10) or daidzein (n ⫽ 10). (B) ICl,vol recorded using internal recording solution containing 1 mM Src control peptide (CP; n ⫽ 10) or 1 mM Src activator peptide (AP; n ⫽ 6) at ⫺62 and ⫹58 mV (*P ⬍ 0.05).



1529

strated that insulin can activate PP-1 through a PTK-PI3K pathway.37 Before electrophysiological experiments, NPCE cells were pretreated with 100 nM insulin for 5 minutes. Insulin was also included in the superfusate during the recordings. Figure 7B shows Cl⫺ current traces recorded from a representative NPCE cell (34 pF) in hyposmotic solution without insulin (control, right panel) and current traces from another cell (32 pF) treated with insulin (100 nM) and then stimulated with hyposmotic solution (insulin, left). In cells treated with insulin, ICl,vol was increased. Figure 7C shows that in hyposmotic solution, maximum ICl,vol recorded at ⫺62 and ⫹58 mV, was increased by 136% and 149% in insulin-treated cells compared with control cells (P ⬍ 0.01, n ⫽ 10). Figure 7D demonstrates that ICl,vol recorded in the presence of 100 nM insulin, was also inhibited by 1 ␮M okadaic acid (68% at ⫺62 mV and 69% at ⫹58 mV, n ⫽ 6, P ⬍ 0.01), further confirming that PP activation contributes to ICl,vol activation in rabbit NPCE cells. Insulin, in the absence or presence of the PKC inhibitor calphostin, did not enhance ICl,vol under isosmotic conditions (data not shown), suggesting that, in the absence of cell swelling, PP activation alone or coupled to a decrease in PKC inhibition is not sufficient to increase basal volume-sensitive Cl channel activity.

DISCUSSION In this study, we examined regulation by phosphorylation of volume-sensitive Cl channels in NPCE cells, showing that the activation of ICl,vol is modulated by PKC-mediated phosphorylation, which exerts an inhibitory modulation, and by PPmediated dephosphorylation, which promotes channel activaFIGURE 5. Inhibition of PI3K reduced ICl,vol. (A) Mean ICl,vol recorded in the absence (n ⫽ 8) or presence (n ⫽ 6) of 100 nM wortmannin (Wort) at ⫺62 to ⫹58 mV. (B) ICl,vol recorded in absence (n ⫽ 4) or presence (n ⫽ 3) of 50 ␮M genistein (GST) or 50 ␮M genistein ⫹100 nM wortmannin ( n ⫽ 4) at ⫺62 to ⫹58 mV.

Our results with PTK inhibitors, Src-AP and PI3K inhibition are consistent with a c-Src PTK/PI3K pathway’s contributing to activation of ICl,vol in NPCE cells. Our data also suggest that PI3K signaling pathways may contribute to PKC activation in NPCE cells after cell swelling.

Serine-Threonine PPs and ICl,vol Regulation Dephosphorylation involving serine-threonine PPs may regulate activity of volume-sensitive Cl channels during cell swelling.16 In addition, PTK activation has been reported to activate serine-threonine PP-1, probably through PI3K after hyposmotic stimulation.17 We investigated the role of PP in the activation of volume-sensitive Cl channels in NPCE cells after cell swelling using the PP-1/PP-2A inhibitor okadaic acid. An inactive analogue, okadaic acid methyl ester, was also used to exclude any nonspecific effects of okadaic acid. During whole-cell current recordings, 1 ␮M okadaic acid or 1 ␮M okadaic acid methyl ester was included in internal solution. Figure 7A shows mean ICl,vol recorded in control cells (n ⫽ 8), okadaic acid (n ⫽ 8), and okadaic acid methyl ester (n ⫽ 8) treated cells. At ⫺62 and ⫹58 mV, ICl,vol was inhibited by 60% and 56%, respectively, in the okadaic acid–treated group compared with the okadaic acid methyl ester group and 59% and 56% compared with control groups (P ⬍ 0.05). Okadaic acid methyl ester had no significant effect on the ICl,vol in comparison with the control group. These data indicate that PP activation, after swelling, leads to increased volume-sensitive Cl channel activity in rabbit NPCE cells. Further confirmation that PP is probably involved in dephosphorylation and activation of volume-sensitive Cl channels in NPCE cells was obtained using insulin. Studies have demon-

FIGURE 6. Inhibition of PI3K blocked actions of PKC inhibitors in ICl,vol. (A) Representative currents recorded every 5 minutes at ⫺62 and ⫹58 mV in isosmotic solution (0 minutes), hyposmotic solution (0 –25 minutes), hyposmotic solution ⫹100 nM wortmannin (Wort; 25– 45 minutes), hyposmotic solution ⫹100 nM wortmannin ⫹1 ␮M chelerythrine (45– 65 minutes). (B) Mean ICl,vol recorded in the absence and presence of 100 nM wortmannin or 100 nM wortmannin ⫹1 ␮M chelerythrine (CHT) at ⫺62 (n ⫽ 3; **P ⬍ 0.01) and ⫹58 mV (n ⫽ 3; *P ⬍ 0.05).

1530

Shi et al.


FIGURE 7. Activation of PP-1 enhanced ICl,vol. (A) ICl,vol measured in the absence (n ⫽ 8) or presence of 1 ␮M of the PP inhibitor, okadaic acid (OA; n ⫽ 8) or okadaic acid methyl ester (OAM; n ⫽ 8) at ⫺62 and ⫹58 mV. (B) Representative current traces recorded from an NPCE cell after 25 minutes of hyposmotic stimulation (Control, left). Right: currents recorded in another cell 15 minutes after hyposmotic stimulation in the presence of 100 nM insulin. (B, top left) Voltage protocol. (C) ICl,vol measured in control cells (Con; n ⫽ 9) or in cells pretreated with 100 nM insulin (Ins; n ⫽ 10) at ⫺62 and ⫹58 mV. (D) The insulin-enhanced ICl,vol in the absence (n ⫽ 9) or presence of 1 ␮M okadaic acid (OA; n ⫽ 6). (*P ⬍ 0.05; **P ⬍ 0.01).

tion. Our results provide new evidence that identifies Src tyrosine kinase and PI3K as upstream intermediaries in a signaling pathway linking PP and PKC in the regulation of ICl,vol in NPCE cells. A summary scheme of our results is shown in Figure 8 and is discussed in the following paragraphs. Under isotonic conditions, it has been suggested that the balance between basal PKC and phosphatase activity may keep

FIGURE 8. Proposed signaling pathways regulating ICl,vol in NPCE cells. Cell swelling, which may involve multiple volume sensors (e.g., perturbations in membrane-cytoskeleton tension, changes in macromolecular crowding, and activation of mechanosensitive channels and proteins), initiates increases in PTK (including Src kinase) activity, probably upstream of PI3K, which is activated in turn. PI3K activates protein phosphatases such as PP-1 to produce an increase in Cl channel activity through dephosphorylation. In contrast, PKC-mediated phosphorylation provides inhibitory regulation of ICl,vol through actions either directly on channel proteins or through signaling pathways. Because inhibition of PI3K occluded subsequent channel modulation through PKC, an interrelationship between these enzymes may be a prerequisite for channel modulation, and it is possible that PI3K also activates PKC to provide for fine tuning of the Cl channel activation. Activation of PI3K by insulin increased channel activity and supports the involvement of PP-1.

the majority of volume-sensitive chloride channels in a phosphorylated, closed state, with a few dephosphorylated channels generating the basal current.8,15,16 Cell swelling, after hyposmotic challenge, is associated with an increase in ICl,vol as increasing dephosphorylation allows more channels to activate.13,15,16 The increasing dephosphorylation of the channels has been hypothesized to be due either to decreased PKC activity, which may reflect dilution of cytosolic contents, or other alterations in kinase and/or PP activity. In rabbit NPCE cells, we showed that PKC inhibitors together with PP activation (which should result in increased activation of ICl,vol),16 had no effect on the basal Cl⫺ conductance recorded under isosmotic conditions. This could suggest that under isosmotic conditions, very few volume-sensitive Cl channels are active and that cell swelling appears to be a prerequisite for modulation of ICl,vol by phosphorylation. These findings are in agreement with the lack of basal activation of ICl,vol in native bovine NPCE cells12,38 and several other cell types, including chick and canine cardiac myocytes.11,13,19 However, in a human NPCE cell line, even under isosmotic conditions, a PKC inhibitor, staurosporine, was shown to upregulate ICl,vol,6 and in NIH/3T3 cells, expressed ICl,vol was basally active in absence of cell swelling.15 Differences in basal PKC activity have been suggested to account for cell– cell differences in the phosphorylation and activation of ICl,vol under isotonic conditions.16 Furthermore, although PKC inhibits IClC-3 in NIH/3T3 cells and ICl,vol in native guinea pig and rabbit atrial myocytes,15,16 studies of ICl,vol in several other cell types, such as human cervical cancer cells39 and canine atrial cells,29 have also shown that ICl,vol may be stimulated by PKC. Activation of specific PKC isoforms or kinase–phosphatase signaling pathways targeted by PKC in individual cell types may contribute to differences in phosphorylation and activation of volume-sensitive Cl channels between cells. Alternatively, contributions of several distinct Cl channel types to ICl,vol may also give rise to variability in the response to PKC. In rabbit NPCE cells, our electrophysiological data support the involvement of PKC in the inhibitory modulation of ICl,vol. Although it is possible that decreases in basal PKC activity may contribute to Cl channel activation after cell swelling, it is also possible that activation of PKC, together with other downstream mediators, after cell swelling, allows for reciprocal inhibition of ICl,vol through phosphorylation of the channel



protein directly and/or through other signaling intermediaries. Our data showing that PI3K inhibitors block enhancement of ICl,vol by PKC inhibitors suggest that PKC could be activated downstream of PI3K after cell swelling. Other kinases could also contribute to channel phosphorylation, but their identity remains elusive at present. A volume-regulated serine-threonine protein kinase (h-sgk) has been described and is upregulated by cell shrinkage and inhibited by cell swelling.40 In addition to decreased kinase activity, it has also been suggested that swelling-induced activation of ICl,vol may occur as a result of the increased activities of signaling pathways involving tyrosine kinases and serine-threonine PP.16 Reports in various cell types have shown that swelling-induced activation of ICl,vol requires tyrosine phosphorylation.20 –23 Our study implicates the c-Src family of tyrosine kinase in the activation of ICl,vol during NPCE cell swelling. Although the mechanism(s) underlying the involvement of tyrosine kinase pathways by cell swelling are speculative, it has been suggested that cell swelling can affect the cytoskeleton through the activation of signaling pathways involving various small G proteins including Rho, Rac, and CDC42,9,19 which in turn can lead to the activation of cytosolic tyrosine kinases such as c-Src.34 One signaling molecule activated downstream of c-Src tyrosine kinase is PI3K.34 Enhanced PI3K activity has been demonstrated during cell swelling and a PI3K inhibitor, wortmannin, blocks ICl,vol in several different cell types.23,24 In cholangiocytes, intracellular application of the lipid products of PI3K gave rise to a Cl⫺ current analogous to those observed during cell swelling.24 In rabbit NPCE cells, our experiments using wortmannin showed that PI3K contributed to the activation of ICl,vol and further supported the involvement of PTKPI3K pathways in the activation of volume-sensitive Cl channels after cell swelling. Our data in rabbit NPCE cells suggest that PTK and PI3K signaling pathways may lead to the activation of volume-sensitive Cl channels through stimulation of PP activity. These conclusions are based on results indicating that signaling molecules, such as c-Src and PI3K, are involved in the activation of ICl,vol, and that exposure of NPCE cells to the PP inhibitor, okadaic acid, inhibits ICl,vol. Activation of the insulin signaling cascade, has been demonstrated to stimulate tyrosine kinase activity, phosphorylation of PI3K, and activation of PP-1.37 In rabbit NPCE cells, insulin significantly enhanced ICl,vol. The insulin-enhanced ICl,vol was inhibited by okadaic acid, further confirming PP involvement in the regulation of ICl,vol. Insulin has also been demonstrated to stimulate transepithelial ion transport in other epithelia, such as glandular endometrial epithelial cells, an action blocked by PI3K and PP inhibitors.41 Although further investigations are still required to clarify insulin’s actions on transepithelial ion transport in the ciliary epithelium, our demonstration that insulin increases ICl,vol suggests that increases in plasma insulin levels could result in acute enhancement of transepithelial ion transport and increase aqueous humor secretion. Some reports have suggested that PI3K may activate PP-1 through PKC or protein kinase B (PKB)/Akt.17,37 Although PKB activation of PP-1 was not addressed in this study, it is unlikely that PKC activation of PP can account for the increase in ICl,vol in NPCE cells, because PKC activators abolish ICl,vol activation and PKC inhibitors are successful in enhancing Cl⫺ current after cell swelling. Thus, although our data suggest that PKC activation may occur downstream of c-Src and PI3K in rabbit NPCE cells, they do not support PKC activation of PP-1. Taken together, our results demonstrate that ICl,vol in rabbit NPCE cells was subject to inhibition through PKC-dependent phosphorylation and was activated through stimulation of signaling pathway(s) involving c-Src tyrosine kinase, PI3K, and PP-1. Manipulation of either PKC or PP signaling pathways in the absence of cell swelling was not sufficient to alter basal

1531

ICl,vol in NPCE cells under isosmotic conditions. This implies that either the level of phosphorylation is tightly regulated in NPCE cells and may involve multiple signaling pathways, or alterations in the conformation of channels through the swelling-induced stretching of the membrane cytoskeleton are a prerequisite for subsequent regulation of the channel’s phosphorylation. A growing body of evidence now suggests that a number of different Cl channel types and regulators (e.g., P-gp, pICln) may underlie ICl,vol in mammalian NPCE cells.3,6 –12,38,42 In rabbit NPCE cells26 and a human NPCE cell line (ODM-2),6 ICl,vol has similar electrophysiological and pharmacologic properties and is upregulated by PKC inhibitors. In acutely isolated bovine NPCE cells, ICl,vol was reported to be ATP-dependent and inhibited by verapamil, quinidine, tamoxifen, and dideoxyforskolin, properties consistent with the Cl channel associated with P-glycoprotein.38 This identification was further supported by dialysis with antibodies that targeted the ATP-binding domain and prevented channel activation.38 In bovine NPCE cells, two volume-sensitive Cl channels were identified,9 a 7.3-pS channel and a 19-pS, tamoxifen-sensitive channel sharing properties with the volume-sensitive organic osmolyte and an anion-sensitive channel suggested to be pICln.43,44 In addition, evidence consistent with a contribution of the ClClC-3 channel to the volume- and PKC-sensitive Cl⫺ transport was also reported in ODM-2 cells8 and antisense against ClC-3 was found to partially inhibit ICl,vol in bovine12 and rabbit45 NPCE cells.45 Although recent evidence does not support ClC-3 as a volume-sensitive Cl channel,5,46 it is possible that ClC-3 and related family members (e.g., ClC-4, ClC-5) together with the identified channel regulators may contribute to ICl,vol in NPCE cells. Variations in contributions from multiple channel types in cells derived from different species and under different conditions (acutely isolated versus cell lines) may give rise to some of the reported differences in the pharmacology and regulation of ICl,vol in these cells. Chloride ions are a major contributor to fluid transport and aqueous humor formation by the ciliary epithelium. The ciliary transepithelial potential is reduced by Cl channel blockers, and Cl channels are rate limiting in the efflux of ions across the basolateral membrane of the NPCE cells.3,42 The Cl channels that underlie ICl,vol in NPCE cells allow regulatory volume decrease after a salt load and cell swelling and probably contribute to transepithelial solute secretion.1– 4,41,42 Our data demonstrating that ICl,vol may be reciprocally modulated by signaling pathways activating Src-tyrosine kinase, PI3K, and PP, as well as PKC, provide further insight into our understanding of the regulation of NPCE Cl channels and suggest that, in addition to cell swelling, activation of these signaling pathways by various paracrine modulators may contribute to modulation of aqueous humor secretion.

References 1. Edelman JL, Sachs G, Adorante JS. Ion transport asymmetry and functional coupling in bovine pigmented and nonpigmented ciliary epithelial cells. Am J Physiol. 1994;266:C1210 –C1221. 2. Walker VE, Stelling JW, Miley HE, Jacob TJ. Effect of coupling on volume-regulatory response of ciliary epithelial cells suggests mechanism for secretion. Am J Physiol. 1999;276:C1432–C1438. 3. Jacob TJ, Civan MM. Role of ion channels in aqueous humor formation. Am J Physiol. 1996;271:C703–C720. 4. Farahbakhsh NA, Fain GL. Volume regulation of non-pigmented cells from ciliary epithelium. Invest Ophthalmol Vis Sci. 1987;28: 934 –944. 5. Clapham D. How to lose your hippocampus by working on chloride channels. Neuron. 2001;29:1–3. 6. Coca-Prados M, Anguita J, Chalfant ML, Civan MM. PKC-sensitive Cl⫺ channels associated with ciliary epithelial homologue of pIcln. Am J Physiol. 1995;268:C572–C579.

1532

Shi et al.

7. Anguita J, Chalfant ML, Civan MM, Coca-Prados M. Molecular cloning of the human volume-sensitive chloride conductance regulatory protein, pICln, from ocular ciliary epithelium. Biochem Biophys Res Commun. 1995;208:89 –95. 8. Coca-Prados M, Sanchez-Torres J, Peterson-Yantorno K, Civan MM. Association of ClC-3 channel with Cl-transport by human nonpigmented ciliary epithelial cells. J Membr Biol. 1996;150:197–208. 9. Zhang JJ, Jacob TJ. Three different Cl channels in the bovine ciliary epithelium activated by hypotonic stress. J Physiol. 1997;499:379 – 389. 10. Wang L, Chen L, Walker V, Jacob TJ. Antisense to MDR mRNA reduces P-glycoprotein expression, swelling-activated C1- current and volume regulation in bovine ciliary epithelial cells. J Physiol. 1998;511:33– 44. 11. Chen L, Wang L, Jacob TJ. Association of intrinsic pICln with volume-sensitive Cl⫺ current and volume regulation in a native epithelial cell. Am J Physiol. 1999;276:C182–C192. 12. Wang L, Chen L, Jacob TJC. The role of ClC-3 in volume-sensitive chloride currents and volume-regulation in bovine epithelial cells demonstrated by antisense inhibition. J Physiol. 2000;524:63–75. 13. Okada Y. Volume expansion-sensing outward-rectifier Cl channel: fresh start to the molecular identity and volume sensor. Am J Physiol. 1997;273:C755–C789. 14. Hoffmann EK, Dunham PB. Membrane mechanisms and intracellular signalling in cell volume regulation. Int Rev Cytol. 1995;161: 173–249. 15. Duan D, Winter C, Cowley S, Hume JR, Horowitz B. Molecular identification of a volume-regulated chloride channel. Nature. 1997;390:417– 421. 16. Duan D, Cowley S, Horowitz B, Hume JR. A serine residue in ClC-3 links phosphorylation-dephosphorylation to chloride channel regulation by cell volume. J Gen Physiol. 1999;113:57–70. 17. Bewick NL, Fernardes C, Pitt AD, Rasmussen H, Whalley DW. Mechanisms of Na⫹-K⫹ pump regulation in cardiac myocytes during hyposmolar swelling. Am J Physiol. 1999;276:C1091–1099. 18. Tilly BC, Edixhoven MJ, Tertoolen LGJ, et al. Activation of the osmo-sensitive chloride conductance involves P21rho and is accompanied by a transient reorganization of the F-actin cytoskeleton. Mol Biol Cell. 1996;7:1419 –1427. 19. Nilius B, Voets T, Prenen J, et al. Role of Rho and Rho kinase in the activation of volume-regulated anion channels in bovine endothelial cells. J Physiol. 1999;516:67–74. 20. Voets T, Manolopoulos V, Eggermont J, Ellory C, Groogmans G, Nillus B. Regulation of a swelling-activated chloride current in bovine endothelium by protein tyrosine phosphorylation and G proteins. J Physiol. 1998;506:341–352. 21. Crepel V, Panenka W, Kelly MEM, MacVicar BA. Mitogen-activation protein and tryosine kinase in the activation of astrocyte volumesensitive chloride current. J Neurosci. 1998;18:1196 –1205. 22. Thoroed SM, Bryan-Sisneros A, Doroshenko D. Protein phosphotyrosine phosphatase inhibitors suppress regulatory volume decrease and the volume-sensitive Cl- conductance in mouse fibroblasts. Pflugers Arch. 1999;438:133–140. 23. Webster CRL, Blanch CJ, Phillips J, Anwer MS. Cell swellinginduced translocation of rat liver NTPC is mediated via the PI3K signaling pathway. J Biol Chem. 2000;275:29754 –29760. 24. Feranchak AP, Roman RM, Doctor RB, Salter KD, Toker A, Fitz JD. The lipid products of phosphoinositide 3-kinase contribute to regulation of cholangiocyte ATP and chloride transport. J Biol Chem. 1999;274:30979 –30986. 25. Delamere NA, Coca-Prados M, Aggarwal S. Studies on the regulation of ascorbic acid transporter in a cell line derived from rabbit nonpigmented ciliary epithelium. Biochim Biophys Acta. 1993; 149:102–108. 26. Shi C, Ryan SJ, French AS, Coca-Prados M, Kelly MEM. Hyposmotically activated chloride channels in cultured rabbit non-pigmented ciliary epithelial cells. J Physiol. 1999;521:57– 67.

IOVS, May 2002, Vol. 43, No. 5 27. Botchkin LM, Matthews G. Swelling activates chloride current and increases internal calcium in nonpigmented epithelial cells from the rabbit ciliary body. J Cell Physiol. 1995;164:286 –294. 28. Kawasaki M, Uchida S, Monkawa T, et al. Cloning and expression of a protein kinase C-regulated chloride channel abundantly expressed in rat brain neuronal cells. Neuron. 1994;12:597– 604. 29. Du X, Sorota S. Protein kinase C stimulates swelling-induced chloride current in canine atrial cells. Pflugers Arch. 1999;437:227– 234. 30. Yu XM, Salter MW. Src, a molecular switch governing gain control of synaptic transmission mediated by N-methyl-D-aspartate receptors. Proc Natl Acad Sci USA. 1999;96:7697–7704. 31. Strauss O, Buss F, Rosenthal R, et al. Activation of neuroendocrine L-type channels (alpha1D subunits) in retinal pigment epithelial cells and brain neurons by pp60(c-src). Biochem Biophys Res Commun. 2000;270:806 – 810. 32. Liu X, Brodeur SR, Gish G, et al. Regulation of c-Src tyrosine kinase activity by the Src SH2 domain. Oncogene. 1993;8:1119 –1126. 33. Krause U, Rider MH, Hue L. Protein kinase signaling pathway triggered by cell swelling and involved in the activation of glycogen synthesis and acetyl-CoA carboxylase in isolated rat hepatocytes. J Biol Chem. 1996;271:16668 –16673. 34. Lebrun P, Mothe-Satney I, Delahaye L, Obberghent E, Baron V. Insulin receptor substrate-1 as a signaling molecule for focal adhesion kinase pp125FAK and pp60src. J Biol Chem. 1998;273: 32244 –32253. 35. Reddy MA, Prasadarao NV, Wass VA, Kim WK. Phosphatidylinositol 3-kinase activation and interaction with focal adhesion kinase in E. coli K1 invasion of human brain microvascular endothelial cells. J Biol Chem. 2000;275:36769 –36774. 36. Toker A. Protein kinases as mediators of phosphoinositide 3-kinase signaling. Mol Pharm. 2000;57:652– 658. 37. Ragolia L, Cherpalis B, Srinivasan M, Begum N. Role of serine/ threonine protein phosphatases in insulin regulation of Na⫹/K⫹ATPase activity in cultured rat skeletal muscle cells. J Biol Chem. 1997;272:23653–23658. 38. Wu J, Zhang JJ, Koppel H, Jacob TJ. P-glycoprotein regulates a volume-sensitive chloride current in bovine non-pigmented ciliary epithelial cells. J Physiol. 1996;491:743–55. 39. Chou CY, Shen MR, Hsu KS, Huang HY, Lin HC. Involvement of PKC-alpha in regulatory volume decrease responses and activation of volume-sensitive chloride channels in human cervical cancer HT-3 cells. J Physiol. 1998;512:435– 448. 40. Waldegger S, Barth P, Raber G, Lang F. Cloning and characterization of a putative human serine/threonine protein kinase transcriptionally modified during anisotonic and isotonic alterations of cell volume. Proc Natl Acad Sci USA. 1997;94:4440 – 4445. 41. Deachapunya C, Palmer-Densmore M, O’Grady SM. Insulin stimulates transepithelial sodium transport by activation of a protein phosphatase that increases Na-K ATPase activity in endometrial epithelial cells. J Gen Physiol. 1999;114:561–574. 42. Jacob TJC. Chloride channels in the ciliary epithelium. In: Civan MM, ed. The Eye’s Aqueous Humor Current, Topics in Membranes. Vol 45. San Diego: Academic Press; 1998:55– 68. 43. Jackson PS, Morrison R, Strange K. The volume-sensitive organic osmolyte-anion channel VSOAC is regulated by nonhydrolytic ATP binding. Am J Physiol. 1995;267:C1203–C1209. 44. Strange K, Emma F, Jackson PS. Cellular and molecular physiology of volume-activated anion channels. Am J Physiol. 1996;270: C711–C730. 45. Shi C, Kelly MEM. Regulation of a volume-activated chloride channel in rabbit nonpigmented ciliary epithelial cells. Exp Eye Res. 2000;71(suppl 1):S130. 46. Stobrawa SM, Breiderhoff T, Takamori S, et al. Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to a loss of the hippocampus. Neuron. 2001;29:185–196.