Functional and morphological differentiation of nonpigmented ciliary body epithelial cells grown on collagen rafts

Summary We have examined the effect of alteration in cell shape on promoting differentiated morphology and physiology in cultured nonpigmented epithelial cells from the ciliary body. We have grown pure populations of nonpigmented cells on collagen gels released from the culture dish to create collagen rafts. Shortly after the gels were detached, the cells shrank in diameter and increased in height while they contracted the gel. Concurrently, the actin cytoskeleton reorganized to the cell cortex as found in vivo. After this differentiated morphology developed, large changes in intracellular Ca²⁺ could be elicited by simultaneous activation of acetylcholine and epinephrine or acetylcholine and somatostatin receptors as seen in intact tissue. Explant cultures of isolated nonpigmented cell layers maintained their actin distribution and also showed synergistic Ca²⁺ increases. Spread cells, grown on rigid substrates, had a disorganized cytoskeleton and rarely showed synergism. These data suggest that the mechanism underlying synergistic Ca²⁺ responses in the ciliary body is functional in nonpigmented cells grown on collagen rafts. In addition, this pathway appears to be sensitive to the disposition of the cell’s cytoarchitecture.     

Adenosine stimulates Cl- channels of nonpigmented ciliary epithelial cells

Ciliaryepithelial cells possess multiple purinergic receptors, and occupancyof A₁ andA₂ adenosine receptors isassociated with opposing effects on intraocular pressure. Aqueousadenosine produced increases in short-circuit current across rabbitciliary epithelium, blocked by removingCl and enhanced by aqueousBa²⁺. Adenosine's actions werefurther studied with nonpigmented ciliary epithelial (NPE) cells fromcontinuous human HCE and ODM lines and freshly dissected bovine cells.With gramicidin present, adenosine (3 µM) triggered isosmoticshrinkage of the human NPE cells, which was inhibited by theCl channel blockers5-nitro-2-(3-phenylpropylamino)benzoate (NPPB) and niflumic acid. At 10 µM, the nonmetabolizable analog 2-chloroadenosine and AMP alsoproduced shrinkage, but not inosine, UTP, or ATP. 2-Chloroadenosine(1 µM) triggered increases of whole cell currents in HCE cells,which were partially reversible,Cl dependent, andreversibly inhibited by NPPB. Adenosine (10 µM) also stimulatedwhole cell currents in bovine NPE cells. We conclude that occupancy ofadenosine receptors stimulatesCl secretion in mammalianNPE cells.

     

Inhibition of NHE-1 Na+/H+ exchanger by natriuretic peptides in ocular nonpigmented ciliary epithelium

The natriuretic peptides (NPs) atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP), and C-type natriuretic peptide (CNP) display hypotensive effects in the mammalian eye by lowering the intraocular pressure (IOP), a function that is mediated by the bilayer ocular ciliary epithelium (CE), in conjunction with the trabecular meshwork. ANP regulates Na⁺/H⁺ exchanger (NHE) activity, and inhibitors of NHE have been shown to lower IOP. We examined whether NPs influence the NHE activity of the CE, which is comprised of pigmented (PE) and nonpigmented (NPE) epithelial cells, by directly recording the rate of intracellular pH (pH_i) recovery from its inner NPE cell layer. NPs inhibited, in a dose-dependent manner (1–100 nM), the rate of pH_i recovery with the order of potency CNP > ANP > BNP, indicative that this inhibition is mediated by the presence of NPR type B receptors. 8-Bromo-cGMP (8-BrcGMP), a nonhydrolyzable analog of cGMP, mimicked NPs in inhibiting the rate of Na⁺-dependent pH_i recovery. In contrast, ethylisopropyl amiloride (EIPA, 100 nM) or amiloride (10 µM) completely abolished the pH_i recovery by NHE. 18-Glycyrrhetinic acid (18-GA), a gap junction blocker, attenuated the inhibitory effect of CNP on the rate of pH_i recovery, suggesting that NHE activity in both cell layers of the CE is coregulated. This interpretation was supported, in part, by the coexpression of NHE-1 isoform mRNA in both NPE and PE cells. The mechanism by which the inhibitory effect of NPs on NHE-1 activity might influence the net solute movement or fluid transport by the bilayer CE remains to be determined. Na⁺/H⁺ exchanger type 1; intracellular pH; aqueous humor     

A3 adenosine receptors regulate Cl- channels of nonpigmented ciliary epithelial cells

Adenosinestimulates Cl channels ofthe nonpigmented (NPE) cells of the ciliary epithelium. We sought toidentify the specific adenosine receptors mediating this action.Cl channel activity inimmortalized human (HCE) NPE cells was determined by monitoring cellvolume in isotonic suspensions with the cationic ionophore gramicidinpresent. The A₃-selective agonistN⁶-(3-iodobenzyl)-adenosine-5'-N-methyluronamide(IB-MECA) triggered shrinkage (apparentK_d = 55 ± 10 nM). A₃-selective antagonists blocked IB-MECA-triggered shrinkage, andA₃-antagonists (MRS-1097, MRS-1191, and MRS-1523) also abolished shrinkage produced by 10 µMadenosine when all four known receptor subtypes are occupied. TheA₁-selective agonistN⁶-cyclopentyladenosineexerted a small effect at 100 nM but not at higher or lowerconcentrations. The A_2A agonistCGS-21680 triggered shrinkage only at high concentration (3 µM), aneffect blocked by MRS-1191. IB-MECA increased intracellularCa²⁺ in HCE cells and alsostimulated short-circuit current across rabbit ciliary epithelium.A₃ message was detected in bothHCE cells and rabbit ciliary processes using RT-PCR. We conclude that human HCE cells and rabbit ciliary processes possessA₃ receptors and that adenosinecan activate Cl channels inNPE cells by stimulating these A₃ receptors.     

Multiple receptor activation elicits synergistic IP formation in nonpigmented ciliary body epithelial cells

We have examined theinteraction between muscarinic and ₂-adrenergic receptoractivation on inositol phosphate (IP) formation in the nonpigmentedcells of the ciliary body epithelium (NPE cells) of the rabbit. We havecompared these changes with those previously observed in theintracellular free Ca²⁺ concentration. Whereas muscarinicreceptor activation causes an increase in intracellularCa²⁺ and IP formation, activation of₂-receptors does not significantly increase eitherintracellular Ca²⁺ or IPs over basal levels. However,simultaneous activation of muscarinic and ₂-adrenergicreceptors with the specific agonists carbachol and UK-14304 producesmassive Ca²⁺ increases and results in a synergisticincrease in IP formation. This synergistic IP formation is inhibited byboth muscarinic and ₂-adrenergic receptor antagonists aswell as by pertussis toxin and an inhibitor of phospholipase C. IPformation is predominantly independent of intracellularCa²⁺, because it is decreased but not prevented by blockingthe entry of Ca²⁺ with LaCl₃ or chelatingintracellular Ca²⁺ with1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid. Thus synergistic IP formation underlies, at least in part, the synergistic increase in intracellular Ca²⁺ resulting fromsimultaneous activation of muscarinic and ₂-adrenergic receptors.

     

Calcitonin gene-related peptide stimulates intracellular cAMP via a protein kinase C-controlled mechanism in human ocular ciliary epithelial cells.

Calcitonin gene-related peptides I and II (CGRP I and II) were found to stimulate cAMP levels by approximately 4-6 fold in human nonpigmented ciliary epithelial cells with half-maximal effective concentrations of 20 x 10(-10) and 3 x 10(-10) M, respectively. Prior exposure of cells to 6 x 10(-7) M phorbol 12-myristate, 13-acetate for 15 min resulted in a 40-50% inhibition of CGRP II-dependent cAMP stimulation. Phorbol didecanoate and dioctanoylglycerol also effectively inhibited, whereas 4 alpha phorbol didecanoate, an ineffective activator of protein kinase C, had no effect. Staurosporine, a protein kinase C inhibitor, blocked the inhibition of cAMP formation by phorbol esters. cAMP stimulation by forskolin or cholera toxin was not inhibited by phorbol esters, suggesting that neither a Gs protein nor adenylyl cyclase is the site of inhibition by protein kinase C. These data therefore suggest that CGRP receptors are required for inhibition of adenylate cyclase by protein kinase C.     

Swelling activates chloride current and increases internal calcium in nonpigmented epithelial cells from the rabbit ciliary body

Membrane current and [Ca]_i in rabbit nonpigmented ciliary body epithelial cells (NPE cells) were monitored with combined patch-clamp and fura-2 measurements during cell swelling induced by anisosmotic conditions. In the presence of K-channel blockers, cell swelling produced an increase in membrane current, accompanied by an increase in [Ca]_i. Structural changes in the cell, associated with membrane deformation, may be the cause of the increase in [Ca]_i during swelling. The conductance activated by swelling was permeable to CI: it was dependent on the CI concentration gradient across the cell membrane, and it was blocked by the CI-channel blockers DIDS, SITS, NPPB, and DIOA. Although swelling increased both CI current and [Ca]_i, there was no evidence that Ca was involved in the regulation of the CI conductance. Cell swelling activated the current even when [Ca]_i was strongly buffered at an elevated level (500 nM) or at a low level (～0) with internal Ca-BAPTA/Cs-BAPTA mixtures. In addition, CI conductance was unaffected when [Ca]_i was increased with a Ca ionophore. There was also no evidence that cAMP participates in the regulation of the CI conductance: swelling activation of the current occurred in the presence of cAMP inhibitor (R_p-cAMP-S) and cAMP mimic (S_p-cAMP-S). The data suggest independent involvement of CI conductance and internal Ca in the regulation of cell volume in NPE cells. © 1995 Wiley-Liss, Inc.     

Potassium transport in nonpigmented epithelial cells of ocular ciliary body: inhibition of a Na+, K+, Cl- cotransporter by protein kinase C.

The mechanisms by which 86Rb+ (used as a tracer for K+) enters human nonpigmented ciliary epithelial cells were investigated. Ouabain-inhibitable bumetanide-insensitive 86Rb+ transport accounted for approximately 70-80% of total, whereas bumetanide-inhibitable ouabain-insensitive uptake accounted for 15-25% of total. K+ channel blockers such as BaCl2 reduced uptake by approximately 5%. Bumetanide inhibited 86Rb+ uptake with an IC50 of 0.5 microM, while furosemide inhibited with an IC50 of about 20 microM. Bumetanide-inhibitable 86Rb+ uptake was reduced in Na(+)-free or Cl(-)-free media, suggesting that Na+ and Cl- were required for optimal uptake via this mechanism. These characteristics are consistent with a Na+, K+, Cl- cotransporter in NPE cells. Treatment of NPE cells for 15 min with phorbol 12-myristate, 13-acetate (PMA), an activator of protein kinase C, caused a 50-70% decrease in 86Rb+ uptake via the Na+, K+, Cl- cotransporter. Other 86Rb+ uptake mechanisms were not affected. 86Rb+ uptake via the Na+, K+, Cl- cotransporter could be inhibited by other phorbol esters and by dioctanoylglycerol, an analog of diacylglycerol, but not by 4 alpha phorbol didecanoate, an ineffective activator of protein kinase C. Staurosporine, a protein kinase C inhibitor, blocked phorbol ester inhibition of 86Rb+ uptake. These data suggest that a Na+, K+, Cl- cotransporter in NPE cells is inhibited by activation of protein kinase C.     

Fluid transport by human nonpigmented ciliary epithelial layers in culture: a homeostatic role for aquaporin-1

We report for thefirst time that cultured nonpigmented human ciliary epithelial (NPE)cell layers transport fluid. Cells were grown to confluence onpermeable membrane inserts, and fluid transport across the resultingcell layers was determined by volume clamp at 37°C. These cell layerstranslocated fluid from the apical to the basal side at a steady rateof 3.6 µl · h¹ · cm²(n = 4) for 8 h. This fluid movement wasindependent of hydrostatic pressure and was completely inhibited by 1 mM ouabain, suggesting it arose from fluid transport. Mercuricchloride, a nonspecific but potent blocker ofHg²⁺-sensitive aquaporins, and aquaporin-1 antisenseoligonucleotides both partially inhibited fluid transport across thecell layers, which suggests that water channels have a role in NPE cellhomeostasis. In addition, these results suggest that of the two ciliaryepithelial layers in tandem, the NPE layer by itself can transportfluid. This cultured layer, therefore, constitutes an interesting model that may be useful for physiological and pharmacologicalcharacterization of ciliary epithelial fluid secretion.

     

Molecular evidence that human ocular ciliary epithelium expresses components involved in phototransduction

Here we report the expression, in the human ocular ciliary epithelium and in a human nonpigmented (NPE) ciliary epithelial cell line, of genes usually restricted to cone and rod photoreceptor cells of the retina. By RT-PCR and DNA sequencing we identified the expression of rhodopsin and components linked to its deactivation, including rhodopsin kinase, recoverin, and visual arrestin. We also detected the expression of transducin (T-alpha), phosphodiesterase (PDE-alpha), and cGMP-gated channel alpha-subunits. Cultured NPE cells responded to treatment with phorbol ester by enhancing the expression of rhodopsin mRNA three- to fourfold. Indirect immunofluorescence of the intact ciliary epithelium with monoclonal antibodies (MAbs) against rhodopsin, rhodopsin kinase, and visual arrestin revealed labeling preferentially restricted to the NPE cells. Furthermore, Western blot analysis of whole lysates from the pars plicata region of the human ciliary epithelium with MAbs demonstrated immunochemical cross-reactivity with proteins of molecular mass similar to rhodopsin (36 kDa), rhodopsin kinase (64 to 66 kDa), and arrestin (48-52 kDa) from the human retina. These results provide the first molecular evidence that components of a non-visual phototransduction pathway are expressed in the human ocular NPE ciliary epithelium, which may be linked to circadian entrainment tasks.     

Functional coupling in bovine ciliary epithelial cells is modulated by carbachol

The functional coupling of the ciliaryepithelium was studied in isolated pairs (couplets) of pigmentedciliary epithelial (PCE) and nonpigmented ciliary epithelial (NPCE)cells using the whole cell patch clamp and the fluorescent dye luciferyellow. One cell of the pair (usually the NPCE cell of a NPCE-PCE cell couplet) was accessed with a 2-5 M electrode, containing1-2 mM lucifer yellow, in the whole cell configuration of thepatch clamp. After voltage-clamp experiments were completed, cells were viewed under a fluorescent microscope to confirm that the cells werecoupled. The electrical coupling of the cells was also studied bycalculating the capacitance (using the time-domain technique), assuminga "supercell" model for coupled cells. The mean capacitance ofcoupled pairs was 79.8 ± 4.3 (SE) pF(n = 47) compared with single cellcapacitances of 36.8 ± 3.4 pF (n = 10) for PCE cells and 38.1 ± 3.1 pF(n = 15) for NPCE cells. Octanol,carbachol (CCh), and raised extracellularCa²⁺ concentration([Ca²⁺]_o)all caused uncoupling in pairs (couplets) of coupled NPCE and PCEcells. At room temperature (22-24°C), the capacitance of thecouplets decreased from 70.5 ± 8.0 to 48.0 ± 5.2 pF(n = 5) when exposed to octanol (1 mM), from 73.8 ± 9.2 to 43.2 ± 9.5 pF(n = 4) when exposed to CCh (100 µM), and from 80.5 ± 6.7 to 49.9 ± 7.8 pF(n = 4) when exposed to 10 mM[Ca²⁺]_o.The response to CCh was dose dependent; at higher temperatures of34-37°C, 10 µM CCh caused a 38% reduction in capacitance,from 53.7 ± 9.7 to 33.5 ± 3.3 pF(n = 7) with a half-time of 249 s, and100 µM CCh caused a 49% reduction in capacitance, from 51.3 ± 5.6 to 26.0 ± 2.4 pF (n = 7) witha half-time of 124 s. After pairs uncoupled and the uncoupling agentwas washed out, the cell pairs often exhibited an increase incapacitance that we interpreted as "recoupling" or a reopening ofthe gap junctional communication pathway; the half-time for thisprocess was 729 s after uncoupling with 100 µM CCh and 211 s afteruncoupling with 10 µM CCh. This interpretation was confirmedoptically by the spread of lucifer yellow into both cells of anuncoupled pair with a time course corresponding to the increase inelectrical coupling. The controllable coupling of ciliary epithelialcells extends the idea of a functional syncytium involved in activetransport. PCE cells take up solute and water from the blood, whichthen cross to NPCE cells via gap junctions and from there are secretedinto the posterior chamber of the eye. Modulation of the couplingbetween NPCE and PCE cells may provide a mechanism to controlsecretion.

     

Similarity of A(3)-adenosine and swelling-activated Cl(-) channels in nonpigmented ciliary epithelial cells

Chloride release from nonpigmented ciliary epithelial (NPE)cells is a final step in forming aqueous humor, and adenosine stimulates Cl transport by these cells. Whole cell patchclamping of cultured human NPE cells indicated that theA₃-selective agonist1-deoxy-1-(6-[([3-iodophenyl]methyl)amino]-9H-purin-9-yl)-N-methyl--D-ribofuranuronamide (IB-MECA) stimulated currents (I_IB-MECA) by~90% at +80 mV. Partial replacement of external Clwith aspartate reduced outward currents and shifted the reversal potential (V_rev) from 23 ± 2 mV to0.0 ± 0.7 mV. Nitrate substitution had little effect. Perfusionwith the Cl channel blockers5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB) and niflumic acidinhibited the currents. Partial Cl replacement withaspartate and NO₃, and perfusion with NPPB, hadsimilar effects on the swelling-activated whole cell currents(I_Swell). Partial cyclamate substitution for external Cl inhibited inward and outward currents of bothI_IB-MECA and I_Swell. Bothsets of currents also showed outward rectification and inactivation atlarge depolarizing potentials. The results are consistent with theconcept that A₃-subtype adenosine agonists and swellingactivate a common population of Cl channels.

     

C-type natriuretic peptide receptor expression in pancreatic alpha cells

Atrial natriuretic peptide (ANP), brain type natriuretic peptide (BNP) and C-type natriuretic peptide (CNP) comprise a family of natriuretic peptides that mediate their biological effects through three natriuretic peptide receptor subtypes, NPR-A (ANP, BNP), NPR-B (CNP) and NPR-C (ANP, BNP, CNP). Several reports have provided evidence for the expression of ANP and specific binding sites for ANP in the pancreas. The purpose of this study was to identify the ANP receptor subtype and to localize its expression to a specific cell type in the human pancreas. NPR-C immunoreactivity, but neither ANP nor NPR-A, was detected in human islets by immunofluorescent staining. No immunostaining was observed in the exocrine pancreas or ductal structures. Double-staining revealed that NPR-C was expressed mainly in the glucagon-containing alpha cells. NPR-C mRNA and protein were detected in isolated human islets by RT-PCR and Western blot analysis, respectively. NPR-C expression was also detected by immunofluorescent staining in glucagonoma but not in insulinoma. ANP, as well as BNP and CNP, stimulated glucagon secretion from perifused human islets (1,111 ± 55% vs. basal [7.3 fmol/min]; P < 0.001). This response was mimicked by cANP(4–23), a selective agonist of NPR-C. In conclusion, the NPR-C receptor is expressed in normal and neoplastic human alpha cells. These findings suggest a role for natriuretic peptides in the regulation of glucagon secretion from human alpha cells.     

Hyposmotic activation of ICl,swell in rabbit nonpigmented ciliary epithelial cells involves increased ClC-3 trafficking to the plasma membrane.

In mammalian nonpigmented ciliary epithelial (NPE) cells, hyposmotic stimulation leading to cell swelling activates an outwardly rectifying Cl(-) conductance (I(Cl,swell)), which, in turn, results in regulatory volume decrease. The aim of this study was to determine whether increased trafficking of intracellular ClC-3 Cl channels to the plasma membrane could contribute to the I(Cl,swell) following hyposmotic stimulation. Our results demonstrate that hyposmotic stimulation reversibly activates an outwardly rectifying Cl(-) current that is inhibited by phorbol-12-dibutyrate and niflumic acid. Transfection with ClC-3 antisense, but not sense, oligonucleotides reduced ClC-3 expression as well as I(Cl,swell). Intracellular dialysis with 2 different ClC-3 antibodies abolished activation of I(Cl,swell). Immunofluorescence microscopy showed that hyposmotic stimulation increased ClC-3 immunoreactivity at the plasma membrane. To determine whether this increased expression of ClC-3 at the plasma membrane could be due to increased vesicular trafficking, we examined membrane dynamics with the fluorescent membrane dye FM1-43. Hyposmotic stimulation rapidly increased the rate of exocytosis, which, along with ICl,swell, was inhibited by the phosphoinositide-3-kinase inhibitor wortmannin and the microtubule disrupting agent, nocodazole. These findings suggest that ClC-3 channels contribute to I(Cl,swell) following hyposmotic stimulation through increased trafficking of channels to the plasma membrane.     

Expression of Na,K-ATPase alpha subunit isoforms in the human ciliary body and cultured ciliary epithelial cells

We have analyzed the expression of Na,K-ATPase alpha subunit isoforms in the transporting ciliary processes of the human eye and in cultured cells derived from non-pigmented (NPE) and pigmented (PE) ciliary epithelium. Northern hybridization analysis shows that the mRNAs encoding all the three distinct forms of Na,K-ATPase alpha subunit [alpha 1, alpha 2, and alpha 3] are expressed in the human ciliary processes in vivo. Immunohistochemical analysis using antibodies specific for each of the three alpha subunit isoforms confirms that these polypeptides are present in the microsomal fraction from the human ciliary processes. The monoclonal antibody McB2, which is specific to the Na,K-ATPase alpha 2 subunit isoform, has been found to decorate specifically the basolateral membrane domains of NPE cells but not of the PE cells, suggesting its expression in vivo only in the ocular NPE ciliary epithelium. However, cultured cells derived from the NPE and PE layers exhibit a different pattern of expression of mRNA and protein for the Na,K-ATPase alpha subunit isoforms when compared to the tissue. Both the NPE and PE cells express alpha 1 and alpha 3 mRNA and polypeptide, whereas alpha 2 mRNA and polypeptide are undetectable in these cells. The established cell lines derived from the NPE layer express comparable levels of the alpha 1 and alpha 3 isoforms of Na,K-ATPase as detected in the primary culture. However, the established NPE cell lines are also distinguishable from the normal PE cells when analyzed by Western blot analysis with A x 2 antibodies. The results presented here clearly show that the NPE and PE cells in the ciliary body have a distinct expression of Na,K-ATPase alpha subunit isoforms as compared to cultured cells.