Molecular Vision 2007; 13:443-456 <http://www.molvis.org/molvis/v13/a47/> Received 5 May 2006 | Accepted 23 February 2007 | Published 27 March 2007

Download Reprint

Ca²⁺ channels in retinal pigment epithelial cells regulate vascular endothelial growth factor secretion rates in health and disease

Rita Rosenthal,¹ Heinrich Heimann,² Hansjürgen Agostini,³ Gottfried Martin,³ Lutz Lothar Hansen,³ Olaf Strauss^1,4
 
 

¹Institut für Klinische Physiologie, Charité-Universitatsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany, ²St. Pauls Eye Unit, Royal Liverpool Hospital, Liverpool, United Kingdom, ³Augenklinik des Universitätsklinikums Freiburg, Albert-Ludwigs-Universität Freiburg, Freiburg, ⁴Experimentelle Ophthalmologie, Klinik und Poliklinik für Augenheilkunde, Universitätsklinikum Hamburg-Eppendorf, Germany

Correspondence to: Olaf Strauss, Experimentelle Ophthalmologie, Klinik und Poliklinik für Augenheilkunde, Universitätsklinikum Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany; Phone: +49 40 42803 9469; FAX: +49 40 42803 5017; email: o.strauss@uke.uni-hamburg.de

Abstract

Purpose: Choroidal neovascularization (CNV) is the most severe complication in age-related macular degeneration. The major angiogenic factor involved is vascular endothelial growth factor (VEGF) secreted by the retinal pigment epithelium (RPE). Since RPE cells express neuroendocrine L-type Ca²⁺ channels we investigated their involvement in VEGF secretion in normal RPE cells and RPE cells from patients with CNV.

Methods: Freshly isolated and cultured RPE cells were studied using the patch-clamp technique and ELISA-based secretion assays.

Results: Both freshly isolated and cultured cells showed whole-cell Ba²⁺ currents with properties of L-type Ca²⁺ currents: high activation threshold, sensitivity to dihydropyridines (10 μM nifedipine) and slow inactivation. VEGF-A secretion was elevated by BayK8644 (10 μM) or basic fibroblast growth factor (bFGF, 10 ng/ml), both of which are able to activate L-type channels. Cells from CNV tissue also showed nifedipine-sensitive Ba²⁺ currents, which displayed a voltage-dependent activation at more negative potentials, faster inactivation and changed regulation by tyrosine kinase pp60^c-src. The CNV RPE cells showed higher VEGF secretion rates which were reduced by nifedipine.

Conclusions: Thus, L-type Ca²⁺ channels in normal RPE cells regulate the secretion of VEGF. RPE cells from eyes with CNV maintain a VEGF secretion regulated by nifedipine-sensitve Ca²⁺ channels which might be of importance for the development of CNV.

Introduction

The retinal pigment epithelium (RPE) is a monolayer of pigmented cells that closely interacts with photoreceptors to maintain their structural integrity and excitability [1-4]. A changed growth factor secretion by the RPE is believed to be involved in the etiology of proliferative eye diseases,such as choroidal neovascularization (CNV) [5-8] in age-related macular degeneration (AMD), which is the most common cause for legal blindness in industrialized countries [9]. Mainly a changed growth factor secretion by the RPE seems to be of importance for initiating CNV: increased secretion of pro-angiogenic factors and decreased secretion of anti-angiogenic factors. CNV, representing the most severe complication in AMD, originates in choroidal blood vessels that grow through Bruch's membrane into the sub RPE as well as the subretinal space. In many studies, but especially in recent clinical trials, vascular endothelial growth factor (VEGF) appears to be the major angiogenic factor in this process.

Several studies using cultured or freshly isolated RPE cells from various species demonstrated the expression of L-type Ca²⁺ channels [10-15]. L-type Ca²⁺ channels represent a group of high-voltage activated Ca²⁺ channels [16,17]. Studies examining the regulation of L-type channels of the RPE suggested that these Ca²⁺ channels provide a Ca²⁺-influx pathway involved in growth factor-dependent intracellular signaling [12,15]. Activation of L-type channels in the RPE by the cytosolic subtype tyrosine kinase pp60^c-src shifts the voltage-dependent activation to a more negative voltage-range, closer to the resting potential of RPE cells [12,18]. This leads to a higher number of active channels and, thus, an increase in intracellular free Ca²⁺. In this way, high-voltage-activated Ca²⁺ channels can contribute to intracellular signaling in epithelial cells. A voltage-dependent activation in a rather negative voltage-range has been discussed as a property of L-type channels composed of Ca_v1.3 α-subunits [19-21]. The expression of Ca_v1.3 subunits has been demonstrated in rat RPE cells [12,15]. Thus, L-type channels in RPE cells can participate in intracellular signaling which could mediate changes in a yet unknown RPE cell function.

L-type channels composed of the Ca_v1.3 α-channel-subunit belong to the neuroendocrine subtype, because these channels are known to regulate the insulin secretion by β-islet cells in the pancreas [16]. The RPE is known to secrete a variety of growth factors like insulin like growth factor-1 (IGF-1), basic fibroblast growth factor (bFGF or FGF2), VEGF or pigment epithelium-derived factor (PEDF) [7,22-30]. In order to show that L-type channels might regulate secretion rates in RPE cells we studied voltage-dependent Ca²⁺ channels and the regulation of VEGF-A secretion by human RPE cells and by RPE cells from surgically excised CNV tissues of AMD patients. We found that VEGF secretion is dependent on the activity of voltage-dependent Ca²⁺ channels, which might provide new targets to prevent CNV by interventions at the source for VEGF.

Methods

Human tissue

For the use of human material, tenets of the Declaration of Helsinki were followed, informed consent was obtained, and Institutional Human Experimentation Committee approval was granted for the studies. CNV membranes were obtained directly after eye surgery of patients with AMD. The mean and the range of the age of the patients are given in Table 1. Patients with severe systemic diseases were not included in the study. The human tissue was obtained about one hour after eye surgery or enucleation and immediately prepared for the experiments The detailed information is listed in Table 1.

Patch-clamp recordings with freshly isolated cells

A portion of the CNV membranes was dissolved into single cell suspension by enzymatic digestion using protease (10%) and trypsin (0.1%). Cells from donor eyes without CNV were placed into single-cell suspension by the same method. Cell suspensions were placed onto a poly-lysine-coated glass cover slip and allowed to settle. The cover slips were placed into the perfusion chamber on the stage of an inverted microscope and constantly superfused. The bath solution contained the following (given in mM): 136.4 NaCl, 1.1 NaH₂PO₄, 4.2 NaHCO₃, 0.9 MgCl₂, 0.95 CaCl₂, 10.0 BaCl₂, 5.8 TEACl, 25.0 HEPES, and 11.1 glucose. Perforated-patch recordings were performed with patch-pipettes of 3-5 mega-Ω resistance, using a pipette solution that contained the following (given in mM): 100.0 CsCl, 10.0 NaCl, 2.0 MgSO₄, 0.5 CaCl₂, 5.5 EGTA, 10.0 HEPES, nystatin (150 mg/ml). For perforated-patch recordings from pigmented cells, we added an antibody against class III β-tubulin (Sigma, Deisenhofen, Germany) to the pipette solution. Currents were measured using an EPC-7 patch-clamp amplifier in conjunction with TIDA hardware and software (HEKA, Lamprecht, Germany). The freshly isolated cells showed a membrane capacitance of 43.3±6.9 pF (n=10), and access resistance of 12.5±2.0 mega-Ω (n=10). Access resistance was compensated to values lower than 10 mega-Ω. Seals were stable for approximately 3-7 min.

For stimulation of Ba²⁺ currents through L-type channels cells were depolarized from a holding potential of -70 mV by 9 voltage-steps of 50 ms duration up to +20 mV (Figure 1A). Measurements of maximal current density and kinetic parameters occurred at +10 mV to ensure data comparison at equal biophysical driving forces.

Patch-clamp recordings from cultured cells

CNV membrane pieces not used for in situ investigations were employed for growing RPE cell cultures, using the method of Schlunck et al. [31]. Cultures from eyes without CNV were established using the method of Aronson [32]. Recordings from cultured cells were performed under the same conditions with the same pipette and bath solutions as used for freshly isolated cells. Purity of cell cultures was confirmed by immunohistochemical staining for class III β-tubulin, vimentin, cytokeratin 18, glial fibrillary acidic protein (GFAP), and endothelial factor 8. Sub confluent cell cultures of both control RPE and CNV RPE from the 5^th to the 12^th passage were used for patch-clamp experiments. Mean membrane capacitance of cultured cells was 128.9±17.8 pF (n=19); access resistance was 16.9±2.1 mega-Ω (n=19).

Intracellular application of pp60^c-src

During the whole cell recording, 30 U/ml pp60^c-src (>90% purity) was applied via the patch-pipette employing a procedure described by Wang and Salter [33]. In addition to pp60^c-src (Biomol, Hamburg, Germany), the pipette solution contained ATP (4x10^-3 M) and the extracellular solution DIDS (5x10^-4 M) to block Cl^- channels possibly activated by the application of ATP [34]. After the patch was opened into the whole-cell configuration, the maximal Ba²⁺ current amplitude was measured every 1 min and plotted over the experimental time normalized to the amplitude measured directly after breaking into the whole-cell configuration. As a control, a series of the same experiments were performed with heat-inactivated pp60^c-src (30 min at 95 °C).

Immunoblotting

Membrane proteins were separated by polyacrylamide gel electrophoresis (8.5% polyacrylamide; Mini-Protean cells Bio-Rad Life Science Group, Hercules, CA; for 1 h at 150 V), and 30 mg of total protein were loaded in each lane. The proteins were blotted to nitrocellulose filter screens (Polyscreen, NEN®, Life Science Products Boston, MA) for 1 h at 100 V. Blots were blocked in phosphate buffered saline (PBS/Tween) with nonfat dry milk (10%) for 2 h and with bovine serum albumin (BSA; 5%) for 4 h at room temperature. They were then probed overnight at 4 °C with antibodies against Ca_v1.2 and Ca_v1.3 channel subunits and with the peroxidase-conjugated secondary antibody for 1 h at room temperature. Next, blots were visualized using a chemiluminescence kit (Amersham Pharmacia Biotech, Braunschweig, Germany) and digitized using the LAS-1000 Image Analyzer (Fujifilm, Berlin, Germany) and the AIDA 2.0 software (Raytest, Berlin, Germany). Specific staining was verified by stripping blots and staining them a second time using the same antibody together with the corresponding blocking peptide (antibody: blocking peptide, 1:1).

Vascular endothelial growth factor secretion by retinal pigment epithelium cells

Confluent cell cultures were used to study VEGF secretion by RPE cells. For this assay 100,000 cells were kept in a chamber of a 12-well plate. Every chamber was equipped with 500 μl test solution (Dulbecco's modified Eagle medium with 0.5% fetal calf serum). Beginning with the last change of test medium, the concentration of VEGF-165 was measured every 4 h in the test medium using a commercially available ELISA kit (Biosource International, Solingen, Germany) according to the manufacturer's instructions. The number of experiments (n) includes the mean values of two repeated measurements of VEGF concentration in the medium in one experiment.

Immunohistochemistry

Freshly isolated and cultured cells were stained with anti-class III β-tubulin antibody (Sigma). The following procedure was used for freshly isolated cells. After the perforated-patch recording of a pigmented cell the patch was opened into a classical whole-cell configuration, and negative voltage was applied to facilitate diffusion of the antibody from the pipette into the cell. The patch-pipette was removed from the cell, and the cell stayed attached to the ground and could be subjected to immuno-staining. In the first step, the cells underwent methanol-based fixation followed by permeabilization with Triton X-100 (0.5% in PBS). Then cells were washed several times and incubated with the secondary fluorescence conjugated antibody. Cultured cells underwent methanol-based fixation and permeabilization before they were incubated with the first antibody for 1 h. After being placed through a series of washes cells were next incubated with the secondary fluorescence conjugated antibody. Fluorescence staining was visualized by analysis (Soft Imaging System, Muenster, Germany).

Confluent cultures, which had been subjected to VEGF-A secretion assay, were tested for their homogeneity by means of immunohistochemistry methods as previously described in reference [31]. In brief, cultures were rinsed with PBS and fixed with acetone/PBS 1:1 followed with iced acetone. Cultures were incubated with antibodies against cytokeratin 18 (Sigma), vimentin, glial fibrillary acidic protein (GFAP), endothelial factor 8 (all from DAKO, Hamburg, Germany). Cell nuclei were counter-stained with hematoxylin. Primary antibodies were stained with alkaline phosphatase- or peroxidase-labeled secondary antibodies (Sigma) with fast red as substrate.

Statistics

Numbers of experiments (n) with freshly isolated cells correspond to numbers of individual AMD patients or eyes without CNV. Data from cultured cells from CNV membranes were obtained from three different cultures from three different patients. All data are stated as mean±SEM. Data were tested for statistical significance by ANOVA (Sigma Plot, Systat Software, Inc., Point Richmond, CA). Significance was considered at p-values lower than 0.05.

Results

Voltage-dependent Ca²⁺ channels in freshly isolated cells

Cells were identified as RPE cells by subsequent immunohistochemical staining and fluorescence microscopy (Figure 1G, Figure 2). For this purpose, the solution of the recording pipette contained antibodies against class III β-tubulin, a marker for RPE cells in both healthy and several forms of disease [35]. Once the membrane currents were recorded, the antibody was intracellularly applied via dialysis by the solution in the patch-pipette.

Freshly isolated RPE cells showed voltage-dependent inward currents in the absence of potassium. With Ba²⁺ as charge carrier, these cells responded to depolarization from a holding potential of -70 mV (Figure 1B) to values more positive than -30 mV with fast activating and inactivating inward currents (Figure 1B). The currents showed a maximal current density at +10 mV (Figure 1C) with 1.9±0.6 pApF^-1 with a range from 1.4 to 2.8 pApF^-1 (n=3; Figure 1F), a potential of the half maximal activation of -9.7±1.9 mV (n=3; Figure 3B), a time to peak of 9.7±2.8 ms (n=3), and an inactivation time constant of 53.9±10.3 ms (n=3).

In six out of 13 (46%) investigated CNV membranes, RPE cells showed voltage-dependent Ba²⁺ currents (Figure 1D-G). The mean current density of the maximal amplitude at +10 mV was 7.0±3.4 pApF^-1 (n=6; current density ranged at +10 mV from 3.5 pApF^-1 to 24.4 pApF^-1), which was reached after 2.8±0.4 ms (n=6). This was faster than in cells from control eyes (p=0.024). The potential of the half maximal activation was -15.4±1.3 mV (n=6) which was significantly more negative than in cells from control eyes (p=0.0254). The currents inactivated with a time constant of 12.6±2.3 ms (n=6). Thus, inactivation was faster than in cells from control eyes (p=0.00094).

The activation threshold at about -30 mV indicates that the Ba²⁺ currents were in freshly isolated control cells and in cells from CNV tissue currents through high-voltage-activated (HVA) Ca²⁺ channels. Dihydropyridine nifedipine, a blocker of an HVA channel subtype, the L-type Ca²⁺channels, inhibited maximal current amplitudes to 39.4±6.3% (n=3) of control in freshly isolated RPE cells from eyes without CNV (Figure 3A,B,E). Application of nifedipine (10^-5 M, Figure 3F,D,E) reduced maximal current amplitudes to 20.7±7.5% (n=2) of the control values in freshly isolated cells from CNV tissues. The voltage-dependence of Ba²⁺ currents was not changed in the presence of nifedipine in cells from CNV tissues as well as in cells from eyes without CNV (Figure 3B,D).

Ba²⁺ currents in cultured cells

To further examine the cellular physiological characteristics of RPE cells, we developed cultures from CNV tissues by following the methods of Schlunck et al. [31]. In order to test these cultures on their origin and homogeneity, cultures were labeled with antibodies against vimentin, cytoskeratin 18, endothelial factor 8, and glial acidic fibrillary protein (GFAP). Cultures were homogeneously stained with antibodies against vimentin or cytokeratin 18 (Figure 2). No staining was observed using antibodies against GFAP or endothelial factor 8. Identity of cells used for patch-clamp recordings was confirmed by staining against class III β-tubulin (Figure 1G).

RPE cells maintain the activity of voltage-dependent Ca²⁺ channels in culture (Figure 4 and Figure 5). The maximal current density at +10 mV in cultured RPE cells from CNV membranes was 3.9±0.8 pApF^-1 (n=9) and in cultured RPE cells from eyes without CNV (2.0±0.4 pApF^-1, n=10; p=0.049). Cells from CNV membranes had a time to peak of with 3.3±0.45 ms (n=9), which was faster than in cells from eyes without CNV (11.9±1.4 ms, n=10; p=0.00004). Western blot analysis of membrane proteins revealed the expression of Ca_v1.3 subunits in cultured RPE cells from eyes without CNV and in cells from CNV membranes. A weak expression of Ca_v1.2 subunits could only be detected in RPE cells from eyes without CNV (Figure 4G).

The currents in cultured cells were sensitive to the dihydropyridine nifedipine, too (Figure 5). Application of nifedipine reduced the maximal current amplitude measured at +10 mV to 53.1±8.7% (n=4) in cells from CNV tissues and to 60.1±4.8% (n=4) in cells from eyes without CNV. Furthermore, as in freshly isolated cells, the voltage-dependence of Ba²⁺ currents was not influenced by nifedipine in both cells from CNV tissues and cells from control eyes.

Vascular endothelial growth factor secretion

Cultured RPE cells release VEGF into the medium, a process that could be stimulated by a test medium containing 0.5% fetal calf serum. Under these conditions, we found that VEGF concentration increased with time in the test medium in both RPE cells from eyes without CNV and in RPE cells from CNV membranes. However, the VEGF concentration increased significantly faster and reached higher concentrations in cultures containing cells from CNV membranes when compared to cultures of cells from eyes without CNV (Figure 6A). This resulted in a higher VEGF secretion rate of 12.8 ± 1.4 pg/h (versus no CNV: 5.4±0.9 pg/h; n=4; p=0.0475). Inhibition of Ca²⁺ channels by nifedipine (10^-5 M) reduced VEGF secretion by both cell populations (Figure 6B,C). To confirm the conclusion that activation of voltage-dependent Ca²⁺ channels leads to secretion of VEGF, we used the L-type channel Ca²⁺ channel modulator BayK8644 (10^-5 M; Figure 6B,C) as (-) enantiomer, which activates L-type channels and as the (+) enantiomer, which inhibits L-type channels. The (-) BayK8644 increased the VEGF secretion rate by RPE cells from eyes without CNV (Figure 6B), whereas it only slightly increased the VEGF-A secretion rate in cells from CNV membranes (Figure 6C). The latter effect was not statistically significant. The (+) BayK8644 reduced the VEGF-A secretion rate in both RPE cells from CNV membranes (Figure 6C) as well as in cells from eyes without CNV (Figure 6B). To achieve a more realistic stimulation of RPE cells, we used a test medium that contained bFGF (10 ng/ml). In both, cells from CNV tissue and cells from eyes without CNV, bFGF increased the VEGF secretion rate. Further inhibition of voltage-dependent Ca²⁺ channels by nifedipine reduced the bFGF-induced increase in the VEGF secretion rate (Figure 6D) to control values. In addition, bFGF increased L-type channel activity in RPE cells from CNV membranes as well as in cells from eyes without CNV (Figure 7D).

Regulation of voltage-dependent Ca²⁺ channels

Since the voltage-dependent Ca²⁺ channels mediate growth factor-dependent secretion of VEGF, we studied the regulation of these ion channels to find a possible link to how growth factors stimulate VEGF secretion by stimulation of Ca²⁺ channels. Intracellular application of the cytosolic subtype of tyrosine kinase pp60^c-src (30 U/ml) during the whole-cell configuration via the patch-pipette increased the maximal current amplitude of L-type channels in RPE (n=4; Figure 7A). Application of heat-inactivated pp60^c-src did not change the L-type channel activity (n=4). Furthermore, intracellular application of pp60^c-src accelerated time-dependent inactivation of L-type channel currents (reduction of inactivation time constant to 22±11% of control measured directly after breaking into the whole-cell configuration; n=7; p=0.00036 paired test) Thus, pp60^c-src subtype of tyrosine kinase is an activator of L-type channel activity and modulator of L-type channel kinetics in human RPE cells. Intracellular application of the active pp60^c-src to RPE cells from CNV membranes did not increase the maximal current amplitude (n=3; Figure 7B) or change the time-dependent inactivation (inactivation time constant 67±21% of control; n=4; not significant with paired test). However, inhibition of pp60^c-src by overnight incubation with herbimycin A (10^-5 M) or by 1 h incubation with SU6656 (10^-6 M, according to [36,37]) reduced the current density to the same base level in both groups of cells. (Figure 7C). Thus, L-type channels in cells from CNV membranes are activated by src-subtype tyrosine kinase, but these channels cannot be further stimulated by pp60^c-src. However, as we demonstrated bFGF can further stimulate L-type channel activity in cells from CNV membranes (Figure 7D).

Discussion

The RPE is known to secrete a variety of growth factors [5-7,38]. In this study, we provide insight into the mechanisms that regulate their secretion rate. We found the rate of VEGF secretion by human RPE cells is dependent on the activation of L-type Ca²⁺ channels. Furthermore, RPE cells from CNV tissue also maintained expression of voltage-dependent Ca²⁺ channels and secreted the angiogenic factor VEGF in response to activation of these channels. Our observations might be of importance to explore new strategies to prevent or reduce CNV.

Freshly isolated human RPE cells showed dihydropyridine-sensitive voltage-dependent Ba²⁺ inward currents in response to depolarization to potentials more positive than -30 mV from a holding potential of -70 mV. Thus, fresh human RPE cells exhibited currents with properties of L-type Ca²⁺ channels. We used a mixture of charge carrier ions of 1 mM Ca²⁺ and 10 mM Ba²⁺ to enhance seal formation and stability of the perforated-patch configuration. This would lead to an anomalous mole fraction effect [39-41]. This effect arises from the higher affinity of Ca²⁺ compared to Ba²⁺ at the selectivity filter of the Ca²⁺ channel pore and leads to changes in the current properties such as smaller amplitudes or even occasional shifts in the voltage-dependence of the currents [39,41]. However, the presence of L-type channels is in accordance with many studies using freshly isolated or cultured RPE cells from various species [10-15,18,42,43]. The L-type channel currents in freshly isolated human RPE cells exhibited fast time-dependent activation, rather negative potential of half-maximal activation, and were reduced to only 40% of the control values in the presence of 10^-5 M nifedipine, a rather large concentration. With the low dihydropyridine sensitivity, voltage-dependence, and kinetic behavior, the currents in freshly isolated human RPE cells share characteristics with currents exhibited by L-type Ca²⁺ channel α-subunit Ca_v1.3 investigated in a heterologous expression system [19-21]. The dihydropyridine block was incomplete in which the currents differ from those in heterologous expression system. However, characterization of Ca_v1.3 channels in native cells revealed that these channels were also only incompletely blocked by dihydropyridines at large concentrations [44]. Thus, the currents in freshly isolated RPE cells from eyes without CNV display characteristics of Ca_v1.3 channels by means of voltage-dependence, dihydropyridine sensitivity, and activation kinetics. The application of nifedipine did not change the voltage-dependent activation of the channels. Thus, the L-type channel blocker did not unmask another Ca²⁺ channel with a different voltage-dependence contributing to the whole-cell Ba²⁺ currents. The expression of the Ca_v1.3 subunit was confirmed by Western blot analysis of membrane proteins. The predominant expression of Ca_v1.3 subunits in the RPE was also described in rat and mouse cells [12,15,42,43]. However, the L-type channel currents in human RPE cells showed a fast inactivation with a time constant of 50 ms. Ca_v1.3 subunits in a heterologous expression system showed almost no inactivation [19,20] except in one study [21]. The reason for this difference remains to be elucidated. It might be due to the absence of an anomalous mole fraction effect in other studies that did not use extracellular solutions containing both Ba²⁺ and Ca²⁺, to different phosphorylation in native cells, or different composition of accessory subunits. Another possibility might be the specific splice variant expressed in RPE cells [42]. This splice variant has not been so far described to be present in other tissues, and its characteristics are unknown.

The western blot analysis analysis also revealed a weak expression of Ca_v1.2 (channel α1C) subunits. The time- and voltage-dependent activation indicates that this subtype of L-type channel might only weakly contribute to the whole-cell Ba²⁺ currents in RPE cells, but it cannot be excluded.

The expression of Ca_v1.3 (or α1D subunits) known as neuroendocrine L-type channels [16] in RPE cells implies a role in the control of the secretory activity. Since expression of L-type Ca²⁺ channels was maintained in cultured cells, we analyzed effects of L-type channel modulation on VEGF secretion by RPE cells in vitro. Direct modulation of L-type channels by application of the dihydropyridine compound BayK8644 substantially changed the VEGF secretion rate. Application of the Ca²⁺ channel activating enantiomer of BayK8644 increased the VEGF secretion rate. The same was observed by stimulation with low concentration of fetal calf serum. Furthermore, application of bFGF stimulated both L-type channel currents and VEGF secretion rate. The growth-factor-stimulated increase in VEGF secretion was reduced by L-type channel inhibition using nifedipine or the Ca²⁺ channel blocking enantiomer of BayK8644. Thus, either direct or growth-factor-dependent stimulation of L-type channels resulted in an increase of the VEGF secretion rate. The underlying growth-factor-dependent increase in L-type channel activity might be achieved in two ways. First, the bFGF receptor FGFR-2 can stimulate L-type channels by direct interaction with the Ca_v1.3 subunit [15]. Second, the Ca_v1.3 subunit in rat RPE was found to directly interact with the cytosolic tyrosine kinase pp60^c-src [12]. In human cells pp60^c-src inhibition results in a decrease of L-type channel activity, whereas intracellular application of pp60^c-src leads to an increase of L-type channel activity. Thus, pp60^c-src constitutively maintains a basic level of L-type channel activity, which can be further increased. Since stimulation by several growth factors results in an increase of pp60^c-src activity, the growth-factor-dependent stimulation of VEGF secretion might result from pp60^c-src-dependent activation of L-type channels.

Patch-clamp recordings with freshly isolated cells from CNV revealed that RPE cells can maintain the expression of voltage-dependent Ca²⁺ channels in disease. In a number of investigated membranes, none of the RPE cells displayed any voltage-dependent current. These cells showed a linear current/voltage relationship with maximal current amplitudes in the range of +/- 20 pA. This corresponds with observations in histological investigations of CNV membranes. In these studies, RPE cells in CNV membranes appeared as atrophic cells, a consequence of a disease that starts in the RPE [9]. However, we found that other CNV membranes contained RPE cells appearing to be functional active cells. These cells expressed voltage-dependent Ca²⁺ channels. The currents activate at relatively positive potentials, showed fast activation kinetics and were inhibited by the dihydropyridine nifedipine. With these characteristics, the currents were comparable with those in cells from eyes without CNV, mostly L-type channel currents. However, currents in cells from CNV membranes showed unusual fast inactivation kinetics, and the potential of half-maximal activation was shifted to more negative potentials. Thus, it could be that in these cells another voltage-dependent Ca²⁺ channel contributes to the observed Ba²⁺ currents. Since application of nifedipine resulted in significant current inhibition without shifts in the voltage-dependent activation, the presence of nifedipine did not unmask the contribution by another Ca²⁺ channel with different voltage-dependence to the whole-cell currents. Furthermore, we could also detect the presence of Ca_v1.3 subunits in RPE cells from CNV tissues. Thus, it is likely that these L-type channels contribute, at least in part to the Ba²⁺ currents in these cells. In all cells from CNV membranes, the voltage-dependent activation was shifted to more negative values, and the currents reached their maximum more rapidly. Thus, in RPE cells from CNV membranes, smaller depolarizing shifts in membrane potential led to faster activation of L-type channels, which, in turn, generated larger Ca²⁺ influxes into the cell. These differences could be again due to an anomalous mole fraction effect on different Ca²⁺ channel subtypes [39,41]. Another reason for this might be a higher pp60^c-src-dependent stimulation of L-type channels. CNV tissue represents a growth-factor-rich environment [5,6,8,27,45] that should increase the pp60^c-src activity in general. Furthermore, intracellular application of pp60^c-src to RPE cells from control eyes resulted in faster inactivation of L-type channel currents and has been found to shift the potential of the half-maximal activation to more negative values in rat RPE cells [12]. Since intracellular application of pp60^c-src did not further increase L-type channel activity in RPE cells from CNV membranes, it appears that L-type channels in these cells are maximally stimulated by this tyrosine kinase. This is supported by the observation that inhibition of pp60^c-src in cells from CNV membranes reduced the L-type channel activity to the same level as in cells from control eyes.

RPE cells from CNV membranes not only maintained the expression of nifedipine-sensitive voltage-dependent Ca²⁺ channels, but the VEGF secretion rate is also dependent on the activity of these channels. Application of nifedipine or the Ca²⁺ channel inhibiting (+) enantiomer of BayK8644 significantly reduced the VEGF-A secretion to the same level. The Ca²⁺ channel activating (-) enantiomer of BaK8644 slightly, but not significantly, increased VEGF-A secretion by RPE cells from CNV membranes. This is not unlike experiments where intracellular application of src-kinase was used to stimulate Ca²⁺ channel activities in CNV RPE cells. Here, also, no further stimulatory effect was observed. Several angiogenic and growth factors are present in CNV membranes. Vitronectin [46], a major component of drusen [47,48], VEGF [49] via KDR/flk-1 receptor expressed by RPE cells [50,51], or IGF-1 [25-27,52,53] act via activation of the src-subtype tyrosine kinase, and many are produced by the RPE [5-7,22-27] and bFGF [54]. We propose that the secretion of these factors is regulated by L-type channels at the beginning of the disease. In the progression of the disease growth of CNV tissue is promoted by VEGF-A secretion under control of nifedipine-sensitive voltage-dependent Ca²⁺ channels. The CNV membranes, which were found to contain RPE cells with active Ca²⁺ channels, might reflect a state of active CNV development. Ca²⁺ channel activity and VEGF secretion rate appears to be higher in cells from CNV tissues compared to cells from eyes without CNV. Immunohistological analysis of RPE cell cultures from CNV tissues revealed that these cultures contained homogeneous RPE cells. RT-PCR analysis revealed the expression of the RPE marker RPE65 in these cultures [31]. The presence of endothelial cells could be excluded by absence of endothelial factor 8 staining, and the presence of glia cells could by excluded by absence of GFAP staining. It is unlikely that cultured photoreceptors or neurons were present.

It is intriguing to argue that this might be a pathophysiological effect. However, this could also be an effect resulting from a different reaction to the experimental procedures. In contrast, taking the different Ca²⁺ channel properties into account it might be that RPE cells in the proliferating CNV tissue underwent a transdifferentiation into a different phenotype, which is characterized by predominant growth factor secretion. This phenotype is maintained in culture as can be seen by the conserved Ca²⁺ channel characteristics. As already stated, these characteristics could be due to the expression of additional Ca²⁺ channel subtypes.

In summary, we showed for the first time that regulation of the angiogenic factor secretion rate by human RPE cells involves the activation of voltage-dependent L-type Ca²⁺ channels. Since a changed growth factor secretion by the RPE is thought to play a role in the induction of various eye diseases involving cell proliferation, voltage-dependent Ca²⁺ channels or their regulatory proteins could be included as targets in therapeutic strategies acting directly at the secretion site of angiogenic factors.

Acknowledgements

The authors thank Sarah Coupland for important helpful discussions and Marianne Boxberger for excellent technical assistance. This work was supported by the grants of the Deutsche Forschungsgemeinschaft DFG STR480/8-1 (O.S.) and STR 480/8-2 (O.S.).

References

1. Bok D. The retinal pigment epithelium: a versatile partner in vision. J Cell Sci Suppl 1993; 17:189-95.

2. Steinberg RH. Interactions between the retinal pigment epithelium and the neural retina. Doc Ophthalmol 1985; 60:327-46.

3. Thompson DA, Gal A. Vitamin A metabolism in the retinal pigment epithelium: genes, mutations, and diseases. Prog Retin Eye Res 2003; 22:683-703.

4. Strauss O. The retinal pigment epithelium in visual function. Physiol Rev 2005; 85:845-81.

5. Amin R, Puklin JE, Frank RN. Growth factor localization in choroidal neovascular membranes of age-related macular degeneration. Invest Ophthalmol Vis Sci 1994; 35:3178-88.

6. Frank RN. Growth factors in age-related macular degeneration: pathogenic and therapeutic implications. Ophthalmic Res 1997; 29:341-53.

7. Witmer AN, Vrensen GF, Van Noorden CJ, Schlingemann RO. Vascular endothelial growth factors and angiogenesis in eye disease. Prog Retin Eye Res 2003; 22:1-29.

8. Campochiaro PA. Retinal and choroidal neovascularization. J Cell Physiol 2000; 184:301-10.

9. Ambati J, Ambati BK, Yoo SH, Ianchulev S, Adamis AP. Age-related macular degeneration: etiology, pathogenesis, and therapeutic strategies. Surv Ophthalmol 2003; 48:257-93.

10. Ueda Y, Steinberg RH. Voltage-operated calcium channels in fresh and cultured rat retinal pigment epithelial cells. Invest Ophthalmol Vis Sci 1993; 34:3408-18.

11. Ueda Y, Steinberg RH. Dihydropyridine-sensitive calcium currents in freshly isolated human and monkey retinal pigment epithelial cells. Invest Ophthalmol Vis Sci 1995; 36:373-80.

12. Strauss O, Buss F, Rosenthal R, Fischer D, Mergler S, Stumpff F, Thieme H. 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; 270:806-10.

13. Strauss O, Mergler S, Wiederholt M. Regulation of L-type calcium channels by protein tyrosine kinase and protein kinase C in cultured rat and human retinal pigment epithelial cells. FASEB J 1997; 11:859-67.

14. Rosenthal R, Strauss O. Ca2+-channels in the RPE. Adv Exp Med Biol 2002; 514:225-35.

15. Rosenthal R, Thieme H, Strauss O. Fibroblast growth factor receptor 2 (FGFR2) in brain neurons and retinal pigment epithelial cells act via stimulation of neuroendocrine L-type channels (Ca(v)1.3). FASEB J 2001; 15:970-7.

16. Catterall WA. Structure and regulation of voltage-gated Ca2+ channels. Annu Rev Cell Dev Biol 2000; 16:521-55.

17. Striessnig J. Pharmacology, structure and function of cardiac L-type Ca(2+) channels. Cell Physiol Biochem 1999; 9:242-69.

18. Mergler S, Strauss O. Stimulation of L-type Ca(2+) channels by increase of intracellular InsP3 in rat retinal pigment epithelial cells. Exp Eye Res 2002; 74:29-40.

19. Bell DC, Butcher AJ, Berrow NS, Page KM, Brust PF, Nesterova A, Stauderman KA, Seabrook GR, Nurnberg B, Dolphin AC. Biophysical properties, pharmacology, and modulation of human, neuronal L-type (alpha(1D), Ca(V)1.3) voltage-dependent calcium currents. J Neurophysiol 2001; 85:816-27.

20. Koschak A, Reimer D, Huber I, Grabner M, Glossmann H, Engel J, Striessnig J. alpha 1D (Cav1.3) subunits can form l-type Ca2+ channels activating at negative voltages. J Biol Chem 2001; 276:22100-6.

21. Scholze A, Plant TD, Dolphin AC, Nurnberg B. Functional expression and characterization of a voltage-gated CaV1.3 (alpha1D) calcium channel subunit from an insulin-secreting cell line. Mol Endocrinol 2001; 15:1211-21.

22. Adamis AP, Shima DT, Yeo KT, Yeo TK, Brown LF, Berse B, D'Amore PA, Folkman J. Synthesis and secretion of vascular permeability factor/vascular endothelial growth factor by human retinal pigment epithelial cells. Biochem Biophys Res Commun 1993; 193:631-8.

23. Tanihara H, Inatani M, Honda Y. Growth factors and their receptors in the retina and pigment epithelium. Prog Retin Eye Res 1997; 16:271-301.

24. Sternfeld MD, Robertson JE, Shipley GD, Tsai J, Rosenbaum JT. Cultured human retinal pigment epithelial cells express basic fibroblast growth factor and its receptor. Curr Eye Res 1989; 8:1029-37.

25. Slomiany MG, Rosenzweig SA. Autocrine effects of IGF-I-induced VEGF and IGFBP-3 secretion in retinal pigment epithelial cell line ARPE-19. Am J Physiol Cell Physiol 2004; 287:C746-53.

26. Slomiany MG, Rosenzweig SA. IGF-1-induced VEGF and IGFBP-3 secretion correlates with increased HIF-1 alpha expression and activity in retinal pigment epithelial cell line D407. Invest Ophthalmol Vis Sci 2004; 45:2838-47.

27. Lambooij AC, van Wely KH, Lindenbergh-Kortleve DJ, Kuijpers RW, Kliffen M, Mooy CM. Insulin-like growth factor-I and its receptor in neovascular age-related macular degeneration. Invest Ophthalmol Vis Sci 2003; 44:2192-8.

28. King GL, Suzuma K. Pigment-epithelium-derived factor--a key coordinator of retinal neuronal and vascular functions. N Engl J Med 2000; 342:349-51.

29. Campochiaro PA, Sugg R, Grotendorst G, Hjelmeland LM. Retinal pigment epithelial cells produce PDGF-like proteins and secrete them into their media. Exp Eye Res 1989; 49:217-27.

30. Dawson DW, Volpert OV, Gillis P, Crawford SE, Xu H, Benedict W, Bouck NP. Pigment epithelium-derived factor: a potent inhibitor of angiogenesis. Science 1999; 285:245-8.

31. Schlunck G, Martin G, Agostini HT, Camatta G, Hansen LL. Cultivation of retinal pigment epithelial cells from human choroidal neovascular membranes in age related macular degeneration. Exp Eye Res 2002; 74:571-6.

32. Aronson JF. Human retinal pigment cell culture. In Vitro 1983; 19:642-50.

33. Wang YT, Salter MW. Regulation of NMDA receptors by tyrosine kinases and phosphatases. Nature 1994; 369:233-5.

34. Strauss O, Steinhausen K, Mergler S, Stumpff F, Wiederholt M. Involvement of protein tyrosine kinase in the InsP3-induced activation of Ca2+-dependent Cl- currents in cultured cells of the rat retinal pigment epithelium. J Membr Biol 1999; 169:141-53.

35. Vinores SA, Derevjanik NL, Mahlow J, Hackett SF, Haller JA, deJuan E, Frankfurter A, Campochiaro PA. Class III beta-tubulin in human retinal pigment epithelial cells in culture and in epiretinal membranes. Exp Eye Res 1995; 60:385-400.

36. Blake RA, Broome MA, Liu X, Wu J, Gishizky M, Sun L, Courtneidge SA. SU6656, a selective src family kinase inhibitor, used to probe growth factor signaling. Mol Cell Biol 2000; 20:9018-27.

37. Bowman T, Broome MA, Sinibaldi D, Wharton W, Pledger WJ, Sedivy JM, Irby R, Yeatman T, Courtneidge SA, Jove R. Stat3-mediated Myc expression is required for Src transformation and PDGF-induced mitogenesis. Proc Natl Acad Sci U S A 2001; 98:7319-24.

38. Hageman GS, Luthert PJ, Victor Chong NH, Johnson LV, Anderson DH, Mullins RF. An integrated hypothesis that considers drusen as biomarkers of immune-mediated processes at the RPE-Bruch's membrane interface in aging and age-related macular degeneration. Prog Retin Eye Res 2001; 20:705-32.

39. Hess P, Tsien RW. Mechanism of ion permeation through calcium channels. Nature 1984; 309:453-6.

40. Lansman JB, Hess P, Tsien RW. Blockade of current through single calcium channels by Cd2+, Mg2+, and Ca2+. Voltage and concentration dependence of calcium entry into the pore. J Gen Physiol 1986; 88:321-47.

41. McDonald TF, Pelzer S, Trautwein W, Pelzer DJ. Regulation and modulation of calcium channels in cardiac, skeletal, and smooth muscle cells. Physiol Rev 1994; 74:365-507.

42. Rosenthal R, Bakall B, Kinnick T, Peachey N, Wimmers S, Wadelius C, Marmorstein A, Strauss O. Expression of bestrophin-1, the product of the VMD2 gene, modulates voltage-dependent Ca2+ channels in retinal pigment epithelial cells. FASEB J 2006; 20:178-80.

43. Wollmann G, Lenzner S, Berger W, Rosenthal R, Karl MO, Strauss O. Voltage-dependent ion channels in the mouse RPE: comparison with Norrie disease mice. Vision Res 2006; 46:688-98.

44. Michna M, Knirsch M, Hoda JC, Muenkner S, Langer P, Platzer J, Striessnig J, Engel J. Cav1.3 (alpha1D) Ca2+ currents in neonatal outer hair cells of mice. J Physiol 2003; 553:747-58.

45. Lopez PF, Sippy BD, Lambert HM, Thach AB, Hinton DR. Transdifferentiated retinal pigment epithelial cells are immunoreactive for vascular endothelial growth factor in surgically excised age-related macular degeneration-related choroidal neovascular membranes. Invest Ophthalmol Vis Sci 1996; 37:855-68.

46. Felsenfeld DP, Schwartzberg PL, Venegas A, Tse R, Sheetz MP. Selective regulation of integrin--cytoskeleton interactions by the tyrosine kinase Src. Nat Cell Biol 1999; 1:200-6.

47. Hageman GS, Mullins RF, Russell SR, Johnson LV, Anderson DH. Vitronectin is a constituent of ocular drusen and the vitronectin gene is expressed in human retinal pigmented epithelial cells. FASEB J 1999; 13:477-84.

48. Crabb JW, Miyagi M, Gu X, Shadrach K, West KA, Sakaguchi H, Kamei M, Hasan A, Yan L, Rayborn ME, Salomon RG, Hollyfield JG. Drusen proteome analysis: an approach to the etiology of age-related macular degeneration. Proc Natl Acad Sci U S A 2002; 99:14682-7.

49. Chou MT, Wang J, Fujita DJ. Src kinase becomes preferentially associated with the VEGFR, KDR/Flk-1, following VEGF stimulation of vascular endothelial cells. BMC Biochem 2002; 3:32.

50. Ohno-Matsui K, Yoshida T, Uetama T, Mochizuki M, Morita I. Vascular endothelial growth factor upregulates pigment epithelium-derived factor expression via VEGFR-1 in human retinal pigment epithelial cells. Biochem Biophys Res Commun 2003; 303:962-7.

51. Wada M, Ogata N, Otsuji T, Uyama M. Expression of vascular endothelial growth factor and its receptor (KDR/flk-1) mRNA in experimental choroidal neovascularization. Curr Eye Res 1999; 18:203-13.

52. Adams TE, Epa VC, Garrett TP, Ward CW. Structure and function of the type 1 insulin-like growth factor receptor. Cell Mol Life Sci 2000; 57:1050-93.

53. Rosenthal R, Wohlleben H, Malek G, Schlichting L, Thieme H, Bowes Rickman C, Strauss O. Insulin-like growth factor-1 contributes to neovascularization in age-related macular degeneration. Biochem Biophys Res Commun 2004; 323:1203-8.

54. Rosenthal R, Malek G, Salomon N, Peill-Meininghaus M, Coeppicus L, Wohlleben H, Wimmers S, Bowes Rickman C, Strauss O. The fibroblast growth factor receptors, FGFR-1 and FGFR-2, mediate two independent signalling pathways in human retinal pigment epithelial cells. Biochem Biophys Res Commun 2005; 337:241-7.