Involvement of Calcium-Activated Potassium Channels in the Regulation of DNA Synthesis in Cultured Müller Glial Cells

Abstract

purpose. To determine the involvement of Ca²⁺-activated K⁺ channels of big conductance (BK) and of Ca²⁺ channels in the regulation of DNA synthesis in cultured guinea pig Müller cells. DNA synthesis was stimulated by elevated extracellular potassium, by serum, or by epidermal growth factor.

methods. Dissociated retinas from guinea pigs were cultured for 8 days. Just before confluence was achieved, the cultures were treated with the test substances in serum-free or serum-containing media. The rates of DNA synthesis were assessed by a quantitative bromodeoxyuridine immunoassay. The intracellular Ca²⁺ concentration was measured by the fura-2 fluorescence technique.

results. Blocking the BK channels with tetraethylammonium or by iberiotoxin had no effect at normal extracellular K⁺ (5.8 mM) but decreased the rate of DNA synthesis at higher extracellular K⁺ (10 or 25 mM). Epidermal growth factor-induced DNA synthesis was decreased by block of BK channels or by application of the Ca²⁺ channel blockers nimodipine and flunarizine. Application of epidermal growth factor elevated the intracellular Ca²⁺ concentration of cultured Müller cells. This elevation was diminished by co-application of iberiotoxin or of flunarizine.

conclusions. The activity of BK channels is necessary for elevated DNA synthesis in Müller cells when their membranes are depolarized and/or when the Ca²⁺ influx into Müller cells is increased by growth factors. BK channels may contribute to the maintenance of DNA synthesis by increasing mitogen-induced increase in intracellular Ca²⁺ concentration.

During proliferative vitreoretinopathy (PVR), periretinal membranes are generated by various types of cells, among others by retinal glial (Müller) cells, which become proliferative and migrate out of the neural retina. ^{1 2 3} The transformation of Müller cells from a differentiated into a proliferative state was found to be accompanied by changes in the ion channel activity in their membranes. In Müller cells from patients with PVR, the inwardly rectifying K⁺ currents are downregulated. This is accompanied by a positive shift of the mean resting membrane potential. ^{4 5} On the contrary, Ca²⁺-activated K⁺ channels of big conductance (BK) show an elevated activity in Müller cells from patients with PVR compared with cells from healthy human donors. ⁵ It has been suggested that both membrane depolarization and elevated intracellular Ca²⁺ may cause this stimulation of the BK channel activity. ⁵  

The BK channels have been implicated in the regulation of the proliferation rate of cultured Müller cells. ⁶ Generally, the mechanisms of K⁺ channel-mediated regulation of cell proliferation have been discussed in recent years. It has been observed that various K⁺ and Cl⁻ channel blockers inhibit cell proliferation: This has been ascribed to a modification of calcium signaling, ^{7 8} to a drug-induced alteration of the intracellular pH, ⁹ and to a reduced ability of cell volume regulation during exposure to the blockers. ^{10 11} In cultured human Müller cells, a block of L-type Ca²⁺ channels was found to decrease the mitotic response to growth factors ^{12 13} ; furthermore, growth factors increased the amplitude of L-type Ca²⁺ currents ¹² and enhanced the activity of other calcium-permeable ion channels. ¹⁴ These observations support the idea that the increase in BK channel activity in response to mitogenic factors ⁶ results from an increase in Ca²⁺ entry into Müller cells through Ca²⁺ channels, and that enhanced BK channel activity may be necessary for the maintenance of sufficient Ca²⁺ entry. The present study was performed to determine whether Müller cell proliferation is dependent on extracellular K⁺ concentration, and whether BK channel-mediated effects on Müller cell proliferation is mediated by modulation of Ca²⁺ entry through the Müller cell membrane. 

Methods

Materials

Iberiotoxin and apamin were obtained from Alomone Laboratories (Jerusalem, Israel); nimodipine, flunarizine, and human recombinant epidermal growth factor (EGF) from Calbiochem (Bad Soden, Germany); nagarse (subtilisin, EC 3.4.21.14) from Serva (Heidelberg, Germany); and Hoechst 33258 and fura-2/AM from Molecular Probes (Eugene, OR). All other substances were obtained from Sigma (Deisenhofen, Germany). 

Preparation of Cell Cultures

Animal care and handling were performed in accordance with applicable German laws and with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Adult guinea pigs (250–400 g) were deeply anesthetized by urethane (2.0 g/kg intraperitoneally) before decapitation and enucleation of the eyes. The excised retinas were dispersed in Ca²⁺, Mg²⁺-free phosphate buffer supplemented with nagarse (1.0 mg/ml) for 30 minutes at 37°C. After washing in phosphate buffer containing DNase I (200 U/ml), the dissociated cells were seeded on coverslips (100 μl cell suspension per coverslip; the retinal cells from two eyes were distributed on 54 coverslips) and were cultured at 37°C under a gas mixture of 95% air-5% CO₂. The minimal essential medium was supplemented with 10% fetal calf serum (FCS). The medium was exchanged twice a week. The concentrations of FCS (10%) and EGF (100 ng/ml) were chosen to stimulate Müller cell proliferation at maximal rate. ¹⁵ Just before confluence was achieved after 8 days in culture (i.e., when approximately half the surface area of each coverslip was occupied by growing cells), the test substances were added to the culture medium either 24 or 16 hours before the cultures were fixed. During this latter period, substances were tested in medium either without or with 10% FCS. 

Immunohistochemistry and Determination of the DNA Synthesis Rate

Glial fibrillary acidic protein (GFAP) immunoreactivity was revealed with a polyclonal rabbit anti-cow GFAP serum (Dakopatts, Copenhagen, Denmark), diluted 1:500, and cyanogen (Cy)2-tagged secondary antibodies (pig anti-rabbit; Dianova, Hamburg, Germany). Vimentin immunoreactivity was determined using a murine anti-vimentin IgG antibody (Immunotech, Hamburg, Germany) and Cy3-tagged secondary antibodies (goat anti-mouse; Dianova). 

The rate of DNA synthesis was determined by a bromodeoxyuridine (BrdU) immunoassay. BrdU (10 μM) was added either 4 or 16 hours before fixation with 4% paraformaldehyde. BrdU incorporation into nuclei of mitotically active cells was revealed by a murine anti BrdU IgG-antibody (Bu 33; Sigma) and Cy3-tagged secondary antibodies. Counter-labeling of all cell nuclei was performed with either acridine orange or Hoechst 33258. In the peripheral (i.e., nonconfluent) regions of the cultures, seven distinct areas of each coverslip (each approximately 60,000 μm², resulting in a total area of 0.42 mm² per coverslip) were studied by means of a semiautomatic image analysis system (SIS, Soft-Imaging Systems, Münster, Germany). The results from three coverslips per culture were summarized; every experiment involved at least three independent cultures. The ratio of BrdU immunoreactive versus total (Hoechst 33258 labeled) cell nuclei was taken as the labeling index. 

Ca²⁺ Imaging

Cells were cultured for 8 days in medium containing 10% FCS and then for 16 hours in serum-free medium. Cultured cells were loaded with fura-2/AM (10 μM) for 30 minutes at 37°C. Measurements were obtained in room temperature by using a bath solution containing (in millimolar) 129 NaCl, 3 KCl, 1 CaCl₂, 0.2 MgCl₂, 20 glucose, and 10 HEPES (pH 7.4 adjusted with NaCl). A fluorescent measurement system Fucal 5.12B (Till-Photonics, München, Germany) with an inverted microscope was used. Fluorescence was excited at 340 nm (F₃₄₀) and 380 nm (F₃₈₀). Images were recorded every 15 seconds. Test substances were applied by rapidly (<15 seconds) changing the bath solution in the recording chamber. 

Data Analysis

The fluorescence ratio F₃₄₀/F₃₈₀ is presented to describe relative changes in the intracellular Ca²⁺ concentration ([Ca²⁺]_i). An increase in the ratio indicates an increase in[ Ca²⁺]_i. ¹⁶ Statistical analysis (paired Student’s t-test) was made using the Prism program (Graphpad Software, San Diego, CA). Data are expressed as means ± SEM. 

Results

Müller Cell Cultures

After 8 days in culture, guinea pig Müller cells formed monolayers of flat polygonal cells. In three independent cultures, 96.0% ± 2.0% of the cells expressed immunoreactivity for vimentin, and 98.6% ± 2.0% of the cells expressed immunoreactivity for GFAP. Thus, the majority of the cultured cells were considered to be Müller cells. Typical cultured cells are shown in Figure 1A , which also reveals that among individual cells, the relative dominance of vimentin (red) versus GFAP (green) varied considerably. The moderate density of cells easily allowed for a quantification of all cell nuclei (basic green label in Fig. 1B ) and BrdU-positive nuclei (yellow-orange, due to additional red label in Fig. 1B ). 

Modulation of the Proliferation Rate by[ K⁺]_e

The first purpose of the present study was to determine whether the BK channel-mediated effects on Müller cell proliferation depend on extracellular K⁺ concentration ([K⁺]_e). Cultures were maintained in either serum-free or serum-containing media at different levels of [K⁺]_e. In serum-free media, a basal DNA synthesis was present in cultured guinea pig Müller cells. BrdU incubation for 4 hours resulted in a basic (control) proliferation rate (labeling index) of approximately 0.1 (Fig. 2) . After BrdU incubation for 16 hours, a labeling index of approximately 0.3 was determined (Fig. 3) . Addition of 10% FCS consistently increased the BrdU labeling rate from 0.12 ± 0.02 (serum-free medium) to 0.31 ± 0.03 (serum-supplemented medium containing 5.8 mM K⁺; P < 0.001; Fig. 2A ). In serum-free medium, elevated[ K⁺]_e failed to increase the rate of DNA synthesis (Fig. 2A) . However, when the proliferation was stimulated by serum, elevation of the[ K⁺]_e significantly stimulated the DNA synthesis. Elevating the[ K⁺]_e from 5.8 to 10 mM and to 25 mM increased the labeling index by 18.8% ± 11.0% (P < 0.05) and by 27.8% ± 5.5% (P < 0.001), respectively, when 10% FCS was present in the media (Fig. 2A) . 

Inhibition of Ca²⁺-Activated K⁺ Channels

To determine the involvement of Ca²⁺-activated K⁺ channels in the regulation of the rate of DNA synthesis, the following specific blockers were tested: tetraethylammonium (TEA), which blocks BK channels specifically in Müller cells at a concentration of 1 mM ¹⁷ ; iberiotoxin, a specific blocker of BK channels ¹⁸ ; and apamin, a blocker of SK channels. As shown in Figure 2A , addition of TEA (1 mM) to the culture medium resulted in a depression of the DNA synthesis rate that was dependent on the[ K⁺]_e. The more the cells were depolarized by increasing[ K⁺]_e levels, the more pronounced was the inhibiting effect of TEA. In serum-free media, the DNA synthesis was decreased by 9.6% ± 20.8% at 5.8 mM[ K⁺]_e (not significant) and was significantly blocked by 40.1% ± 33.4% (P < 0.01) and by 62.6% ± 7.6% (P < 0.001) at 10 mM and at 25 mM [K⁺]_e, respectively. Similar results were obtained in serum-containing media (Fig. 2A) . 

Addition of both iberiotoxin (100 nM) and apamin (200 nM) to the culture media did not change the DNA synthesis rate in the presence of 5.8 mM [K⁺]_e (Fig. 2B) . At 25 mM [K⁺]_e, however, iberiotoxin application reduced the labeling index significantly by 36.5% ± 30.4% (P < 0.05), whereas apamin had no effect. Simultaneous application of iberiotoxin and apamin blocked the DNA synthesis by 39.1% ± 30.4% (P < 0.05), which is similar to the blocking effect of iberiotoxin alone (Fig. 2B) . Iberiotoxin (70 nM) also partly blocked the serum-induced DNA synthesis at 25 mM [K⁺]_e (Fig. 2C) but had no effect at 5.8 mM[ K⁺]_e (not shown). Thus, in accordance with the above-mentioned effects of TEA, iberiotoxin decreased the rate of DNA synthesis only when[ K⁺]_e was elevated. 

To determine whether BK channels modulate growth factor–induced proliferation, the effect of epidermal growth factor (EGF) was investigated. EGF was shown to enhance the proliferation rate of cultured Müller cells. ^{15 19 20} In 11 independent cultures, the addition of EGF (100 ng/ml) to the medium increased the BrdU labeling index from 0.27 ± 0.02 to 0.46 ± 0.03 (P < 0.001; Figs. 3A 3B ). The EGF-induced DNA synthesis was fully reversed by blocking the BK channels. Figure 3A shows that both iberiotoxin (at 50 nM) and TEA (at 1 mM) completely blocked the EGF-stimulated DNA synthesis. 

Inhibition of Voltage-Gated Ca²⁺ Channels

BK channels may act as feedback regulators of Ca²⁺ entry into the Müller cells, perhaps through voltage-gated Ca²⁺ channels. Therefore, we investigated whether voltage-gated Ca²⁺ channels may modulate the rate of DNA synthesis in cultured Müller cells. The effects of two Ca²⁺ channel blockers were tested: nimodipine, which blocks both L-type and T-type Ca²⁺ channels, and flunarizine, which preferentially blocks T-type Ca²⁺ channels. ²¹ As illustrated in Figure 3B , neither of the two blockers changed the DNA synthesis rate at control conditions (serum-free, 5.8 mM[ K⁺]_e). However, when the DNA synthesis was stimulated by EGF (100 ng/ml), both blockers fully reversed the growth factor–induced DNA synthesis although at significantly different concentrations. Flunarizine completely blocked the growth factor-induced DNA synthesis at low concentrations (1 μM), whereas higher concentrations of nimodipine were necessary to reverse the EGF-induced DNA synthesis (median inhibitory concentration[ IC₅₀], approximately 13 μM; Fig. 3B ). Simultaneous addition of flunarizine and iberiotoxin to the culture medium blocked the EGF-induced DNA synthesis but did not reduce the DNA synthesis rate below the control level (Fig. 3C) . 

EGF-Induced Increases of [Ca²⁺]_i

In 70.4% of the investigated cultured cells (n = 98), extracellular application of EGF (200 ng/ml) induced a fast increase in [Ca²⁺]_i that lasted at least 15 minutes when EGF was continuously present in the bath solution (not shown). In most of the cells, a fast transient increase was followed by a continuous elevation of the[ Ca²⁺]_i (Fig. 4B ). Coapplication of iberiotoxin (100 nM) significantly decreased the EGF-induced continuous elevation of the[ Ca²⁺]_i (Fig. 4A) . In the mean, the EGF-induced increase in steady state[ Ca²⁺]_i was decreased by 64.2% ± 12.4% when iberiotoxin was added to the bath solution (n = 20). Furthermore, addition of flunarizine (1 μM) to the bath solution fully reversed the EGF-induced steady state increase in [Ca²⁺]_i (Fig. 4B) . The results indicate that the depressive effects of iberiotoxin and flunarizine on EGF-induced DNA synthesis may be mediated by their depressive effects on the EGF-induced increase in[ Ca²⁺]_i. 

Discussion

The main result of the present study is that, under certain conditions, both BK (but not SK) Ca²⁺-activated K⁺ channels and voltage-gated Ca²⁺ channels contributed to the mechanism(s) that maintain an elevated rate of DNA synthesis in cultured Müller cells. Neither BK channels nor voltage-gated Ca²⁺ channels modified the basic DNA synthesis rate in unstimulated cells–that is, under control conditions ([K⁺]_e 5.8 mM; Figs. 2A 2B 3B ). However, the DNA synthesis of cells cultured under serum-free (Fig. 2A , left; Fig. 2B , right) or serum-containing conditions (Fig. 2A , right) required, at least partly, the activity of BK channels when the cells were exposed to increased[ K⁺]_e. The more the Müller cells were depolarized by increasing[ K⁺]_e the more the BK channels contributed to the maintenance of DNA synthesis. This reflects the well-known voltage dependence of BK channel activity. ^{17 18}  

Although EGF has been shown by several research groups ^{15 19 20} to enhance the proliferation rate of cultured Müller cells, the mechanisms of its mitogenic action on these cells have not been studied in detail. In other cell types, EGF has been found to depolarize the membrane by increasing its Ca²⁺ conductance, ²² to hyperpolarize the membrane by stimulating both Ca²⁺-activated K⁺ channels and Na,K-ATPase, ²³ or to cause a transient increase in Ca²⁺ influx followed by delayed hyperpolarization due to enhanced activity of Ca²⁺-activated K⁺ channels. ²⁴ In all these cases, EGF enhanced the rate of cell proliferation. 

In cultured guinea pig Müller cells, the stimulating effects of EGF on DNA synthesis were diminished or even reversed by blocking the BK channels (Fig. 3A) or by blocking the voltage-gated Ca²⁺ channels (Fig. 3B) . Because exposure to TEA (Fig. 2A) , iberiotoxin (Fig. 2B) , and nimodipine or flunarizine (Fig. 3B) caused virtually no changes in the rate of DNA synthesis at 5.8 mM[ K⁺]_e, the inhibiting actions of the tested substances were certainly not due to unspecific effects such as direct toxic lesions or changes of the intracellular pH. ⁹ Although a possible effect on cellular volume regulation ^{10 11} cannot be ruled out, the results of the imaging experiments (Fig. 4) suggest that both BK and voltage-gated Ca²⁺ channels are involved in the maintenance of the steady state increase in[ Ca²⁺]_i induced by EGF. Interaction of EGF with the receptor leads to an increase in[ Ca²⁺]_i, which, in turn, activates BK channels. Openings of BK channels would hyperpolarize the membrane of the cells and therefore increase the driving force for sustained Ca²⁺ entry through Ca²⁺-permeable cation channels. ^{7 25} It remains to be determined whether EGF application leads to a membrane hyperpolarization in cultured Müller cells. 

The intracellular signaling mechanisms of EGF in the activity of BK and voltage-gated Ca²⁺ channels are not yet understood. Intravitreal injection of EGF has been found to induce the expression of the immediate early gene c-fos in Müller cells. ²⁶ In chicken ciliary ganglion neurons, EGF has been shown to increase the Ca²⁺-activated K⁺ currents by stimulating the functional expression of BK channels. ²⁷ Other mechanisms may include BK channel activation by tyrosine phosphorylation, which has been associated with the activation of growth factor receptors; by depolarization of the cell membrane; or by stimulation of the Ca²⁺ entry. For bFGF, an amplitude-enhancing effect on the currents through L-type Ca²⁺ channels was described in cultured human Müller cells. ¹² However, because Müller cells of several mammalian species express not only L-type but also T-type Ca²⁺ channels ²⁸ that are sensitive to both flunarizine and nimodipine, ²¹ it is difficult to conclude from the present results which of the Ca²⁺ channel types is involved in the proliferation-enhancing effects of EGF. The depressive effect of flunarizine may indicate an involvement of T-type channels in the induction and/or maintenance of EGF-induced Müller cell proliferation, as previously described for growth factor-induced proliferation of other cell types. ^{29 30 31} T-type channels have been shown to mediate the sustained increase in intracellular calcium induced by different biologically active substances (e.g., by angiotensin II, ³² by endothelin, ³³ and by platelet-derived growth factor ²⁹ ). Further electrophysiological investigations are necessary to identify the type(s) of Ca²⁺ channels modulated by EGF. 

Conclusions

Both BK channels and voltage-gated Ca²⁺ channels may be important mediators of the effects of mitogenic factors on Müller glial cells. BK channels may be implicated in the maintenance of proliferative activity by increasing the mitogen-induced increase in intracellular Ca²⁺ concentration. This view is supported by the previous observation of elevated BK channel activity in Müller cells from patients with PVR. However, further research is necessary to determine the intracellular pathways causing increased activity of BK and Ca²⁺ channels after growth factor stimulation. This research may lead to new therapeutic concepts for the treatment of PVR. 

 Supported by a grant from the Bundesministerium für Bildung, Forschung und Technologie; by Grant 01KS9504, Project C5, from the Interdisciplinary Center for Clinical Research at the University of Leipzig, and by Grant Re 849/8-1 from the Deutsche Forschungsgemeinschaft.

 Submitted for publication November 5, 1999; revised April 8 and July 20, 2000; accepted August 16, 2000.

 Commercial relationships policy: N.

 

Figure 1.

 

Cultured guinea pig Müller cells. (A) The majority of cells expressed vimentin (red) and GFAP (green) immunoreactivity (yellow-orange double label). Cell nuclei were counterstained with Hoechst 33258 (blue). (B) Proliferation assay with BrdU immunocytochemistry (red) and counterstaining of all cell nuclei with acridine orange (green). Cycling cells were revealed by yellow-orange double label (filled arrows). For reasons of clarity, counts were made in single fluorescence. Faint punctuate red label (e.g., open arrow, B) was not taken as an indication of proliferative activity. Scale bar, 100 μm.

Figure 2.

 

TEA, a specific blocker of BK channels in Müller cells at a concentration of 1 mM and iberiotoxin inhibit the proliferation of cultured Müller cells from guinea pigs in accordance with[ K⁺]_e concentration. (A) Cells were cultured in serum-free (left) or serum-containing media (right), at 5.8, 10, or 25 mM[ K⁺]_e. (B) Effects of the addition of iberiotoxin (100 nM) and apamin (200 nM), specific blockers of BK and SK channels, respectively, on the proliferation rate of cells that were cultured in serum-free media. The culture media contained 5.8 mM (left side) or 25 mM K⁺ (right side). (C) Addition of iberiotoxin (70 nM) to the culture medium decreased the proliferation that was induced by 10% FCS at 25 mM [ K⁺]_e. Mean results of three to four independent cultures. The cells were exposed to the test substances during the last 24 hours of culturing. BrdU was added to the cultures 4 hours before fixation. Significant differences are indicated as *P < 0.05, **P < 0.01, and*** P < 0.001; n.s., not significant.

Figure 3.

 

Blockers of BK channels and of voltage-gated Ca²⁺ channels reverse the proliferation induced by addition of EGF to the culture medium. (A) EGF (100 ng/ml; black bar) significantly enhanced the proliferation rate compared with the control conditions (serum-free,[ K⁺]_e 5.8 mM). Simultaneous application of iberiotoxin or of TEA blocked the EGF-induced proliferation at different concentrations. Mean results of four independent cultures. (B) Under control conditions (serum-free, [K⁺]_e 5.8 mM) neither nimodipine (up to 50 μM) nor flunarizine (1 μM) had any effect on the proliferation rate. After addition of EGF (100 ng/ml; black bars) both Ca²⁺ channel blockers decreased the proliferation rate. Mean results of three to six independent cultures. (C) Flunarizine (1 μM) or iberiotoxin (70 nM) blocked the EGF-stimulated proliferation when they were added alone or simultaneously to the culture medium. Mean results of three independent cultures. Significant differences are indicated as in Figure 3 . Both the test substances and BrdU were added 16 hours before fixation of the cultures.

Figure 4.

 

Blockers of BK channels and of voltage-gated Ca²⁺ channels diminished the Ca²⁺ response to EGF revealed by fura-2/AM imaging. (A) Extracellular application of EGF (200 ng/ml) induced an increase in[ Ca²⁺]_i. Coapplication of iberiotoxin (100 nM) greatly diminished the EGF (200 ng/ml)-induced steady state increase in[ Ca²⁺]_i. Examples of records in two cells. (B) Coapplication of flunarizine (1μ M) fully reversed the EGF (200 ng/ml)-induced steady state increase in [Ca²⁺]_i. Examples of recordings in three cells. The ratio of F₃₄₀ to F₃₈₀ is proportional to the concentration of[ Ca²⁺]_i.

The authors thank Jana Krenzlin for the preparation of the cell cultures.