Upregulation of P2X₇ Receptor Currents in Müller Glial Cells during Proliferative Vitreoretinopathy

Abstract

purpose. Müller glial cells from the human retina express purinergic P2X₇ receptors. Because extracellular adenosine triphosphate (ATP) is assumed to be a mediator of the induction or maintenance of gliosis, this study was undertaken to determine whether the expression of these receptors is different in human Müller cells obtained from retinas of healthy donors and of patients with choroidal melanoma and proliferative vitreoretinopathy (PVR).

methods. Human Müller cells were enzymatically isolated from donor retinas, and whole-cell patch-clamp recordings were made to characterize the density of the P2X₇ currents and the activation of currents through Ca²⁺-activated K⁺ channels of big conductance (I _BK) that reflects the increase of the intracellular Ca²⁺ concentration.

results. Stimulation by external ATP or by benzoylbenzoyl ATP (BzATP) evoked both release of Ca²⁺ from thapsigargin-sensitive intracellular stores and opening of Ca²⁺-permeable P2X₇ channels. These responses caused transient and sustained increases in I _BK. In Müller cells from patients with PVR, the mean density of the BzATP-evoked cation currents was significantly greater compared with cells from healthy donors. As a consequence, such cells displayed an enlarged I _BK during application of purinergic agonists. ATP and BzATP increased the DNA synthesis rate of cultured cells. This effect could be reversed by blocking the I _BK.

conclusions. The increased density of P2X₇ receptor channels may permit a higher level of entry of extracellular Ca²⁺ into cells from patients with PVR. Enhanced Ca²⁺ entry and the subsequent stronger activation of I _BK may contribute to the induction or maintenance of proliferative activity in gliotic Müller cells during PVR.

Müller (radial glial) cells are the main type of macroglial cells within the vertebrate retina. Although the plasma membrane permeability of Müller cells is predominated by inwardly rectifying K⁺ currents (I _Kir), ¹ the cells also express various distinct types of depolarization-activated ion channels—among them, Ca²⁺-activated K⁺ channels of big conductance (BK). ^{2 3} The activity of BK channels has been implicated in the regulation of cultured Müller cell proliferation, ^{2 4} and has been found to be elevated in human Müller cells from patients with proliferative vitreoretinopathy (PVR) compared with cells from control retinas. ⁵  

Müller glial cells express a diversity of receptors for neurotransmitters and other biologically active substances ⁶ that may modulate the membrane conductances of the cells. ⁷ Adenosine triphosphate (ATP) is an important transmitter in the retina, ⁸ being crucially involved in early retinal development ^{9 10} and in the neuronal information processing of the mature retina. ¹¹ Müller cells may express different types of purinergic P2 receptors. In isolated salamander Müller cells and in rat Müller cells in situ, activation of P2Y receptors by extracellular ATP stimulates the release of Ca²⁺ from internal stores. ^{12 13} Activation of P2 receptors inhibits the uptake of γ-aminobutyric acid (GABA) by rat Müller cells. ¹⁴ Recently, the presence of ionotropic P2X receptors was described in Müller cells freshly isolated from the human retina. ¹⁵ In human Müller cells, extracellular ATP and 2′-/3′-O-(4-benzoylbenzoyl)-ATP (BzATP), a more specific agonist of P2X₇ receptors, ¹⁶ open nonselective cation channels that may be permeable for Ca²⁺ ions. ¹⁵ The noninactivating current kinetics, as well as single-cell reverse transcription–polymerase chain reaction (RT-PCR) and immunocytochemical evidence, indicate the expression of P2X₇ receptors in these cells. ¹⁵  

Because the activation of purinergic receptors is thought to be a mediator of the induction of reactive gliosis, ^{17 18 19} we wanted to investigate whether the expression of P2 receptors by human Müller cells is altered under pathologic conditions. For this purpose, the density of BzATP-evoked currents in cells from patients with PVR and those with choroidal melanoma was compared with that in cells from healthy donors. Alterations of cell membrane conductances during activation of P2 receptors were investigated electrophysiologically in two ways: Either the P2X₇ receptor-mediated cation conductance or the stimulation of the Ca²⁺-activated K⁺ currents were recorded. An amplitude increase of the Ca²⁺-activated K⁺ currents reflects the increase of the intracellular Ca²⁺ concentration induced by activation of the purinergic receptors. 

Methods

Human tissue was used in accordance with applicable laws and with the Declaration of Helsinki, after approvement by the ethics committee of the Leipzig University Medical School. Eyes obtained at autopsy from organ donors with no reported history of eye disease (referred to in the text as healthy donors) were supplied within 12 and 24 hours after death. Retinal tissue from patients with PVR was obtained from vitreoretinal surgery 1 to 3 hours after the tissue was removed. Retinal tissue from patients with choroidal melanoma was obtained 1 to 3 hours after enucleation. Müller cells were isolated from retinal areas far away from the regions where the melanoma cells were located. Müller cells were isolated using papain- and DNase I–containing solutions, as described previously. ^{3 20} The cell suspensions were stored at 4°C (up to 10 hours) before use. 

Electrophysiological Recordings

Records were made in the whole-cell or in the excised patch configuration of the patch-clamp technique. ²¹ To create outside–out patches, the whole-cell configuration was established, and thereafter the pipette was drawn back to excise a membrane patch. Voltage-clamp records were performed at room temperature (22°C–25°C) using an amplifier (EPC 7; List, Darmstadt, Germany) and a computer program (Tida ver. 5.72, Heka Elektronik, Lambrecht, Germany). The signals were low-pass filtered at 4 kHz (three-pole Bessel filter) at a sampling rate of 15 kHz. The series resistance (10–18 MΩ) was compensated by 30% to 50%. Patch pipettes were pulled from thick-walled borosilicate glass (WPI, Sarasota, FL) and had resistances between 3 and 5 MΩ when K⁺-containing bath and pipette solutions were used. To investigate ATP-evoked responses, the whole-cell currents were elicited by a standard step protocol (holding potential, V _h −80 mV; depolarizing and hyperpolarizing voltage steps of 250-msec duration with an increment of 20 mV) or by continuous recording at a V _h of −60 mV with voltage steps of 50 msec duration to +120 mV and to −100 mV at a frequency of 2.5 Hz. The traces were not leak subtracted. Data were not corrected for liquid junction potentials, because these did not exceed 3 mV. The membrane capacitance of the cells was measured by the integral of the uncompensated capacitive artifact evoked by a hyperpolarizing voltage step from −80 to −90 mV when Ba²⁺ ions (1 mM) were present in the bath solution to block the K⁺ conductance. For recording of the capacitive artifact, the sampling rate was 30 kHz, and the frequencies above 10 kHz were cut off. 

Solutions

The recording chamber was continuously perfused with bath solution. Test substances were added by fast (<15 seconds) changes of the perfusate. For recording the effects of purinergic agonists on the whole-cell currents and on the BK channel activity in excised membrane patches, a low-divalent cation bath solution was used composed of (mM) 110 NaCl, 3 KCl, 0.5 CaCl₂, 10 HEPES, and 11 glucose with pH adjusted to 7.4 with Tris. The pipette solution was made of (mM) 10 NaCl, 130 KCl, 3 MgCl₂, 0.1 EGTA, and 10 HEPES with pH adjusted to 7.2 with Tris. When the BzATP-induced cation currents were recorded in K⁺-free conditions, the bath solution consisted of (mM) 116 NaCl, 1 Na₂HPO₄, 25 NaHCO₃, 11 glucose, 10 HEPES (pH 7.4), and was gassed with 95% O₂-5% CO₂. The pipette solution contained (mM) 10 NaCl, 130 CsCl, 1 CaCl₂, 2 MgCl₂, 10 EGTA, 10 HEPES (pH 7.1). Iberiotoxin was obtained from Alomone Laboratories (Jerusalem, Israel) and papain from Boehringer–Mannheim (Mannheim, Germany). All other substances were from Sigma (Deisenhofen, Germany). 

Cell Culture

Primary cultures of Müller cells were obtained from retinas of healthy donors. The excised retinas were dispersed in Ca²⁺, Mg²⁺-free phosphate buffer supplemented with nagarse (1 mg/ml) for 30 minutes at 37°C. After they were washed 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 cultured at 37°C in a gas mixture of 95% air-5% CO₂. The minimum essential medium was supplemented with 10% fetal calf serum. The medium was exchanged twice a week. After 3 weeks in culture, the test substances were added to the culture medium 16 hours before the cultures were fixed. During this latter period, substances were tested in serum-free medium. Lipophilic substances were dissolved in dimethyl sulfoxide (DMSO). Vehicle alone did not affect the DNA synthesis rate. 

Determination of the DNA Synthesis Rate

The DNA synthesis rate was determined by measuring the bromodeoxyuridine (BrdU) incorporation. BrdU (10 μM) was added 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 acridine orange or Hoechst 33258. In the peripheral (i.e., nonconfluent) regions of the cultures, six distinct areas of each coverslip (each approximately 60,000μ m², 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. The experiments involved four independent cultures. The ratio of BrdU immunoreactive versus total cell nuclei was taken as marker for the DNA synthesis rate. 

Data Analysis

The steady state whole-cell currents were measured at the end of 250-msec voltage steps. To determine disease-related changes of currents, 6 to 13 cells per donor were recorded; in most of the further statistical analysis, only the mean values of the cells from each donor were used. Statistical analysis (Mann–Whitney test, two-tailed; nonparametric regression analysis) and curve fits were made by computer (Prism; GraphPad, San Diego, CA). Data are expressed as means ± SD (electrophysiological data) or as means ± SEM (proliferation experiment). 

Results

Effect of ATP on the Whole-Cell Currents

Freshly isolated human Müller cells from patients with various eye diseases do not have the I _Kir normally expressed by human Müller cells from healthy retinas. ^{5 20} Therefore, in these cells, hyperpolarizing voltage steps evoked only a small“ leak” current, whereas depolarizing voltage steps activated delayed rectifying K⁺ currents and Ca²⁺-activated K⁺ currents (Fig. 1A ). Extracellular application of Na-ATP (1 mM) reversibly increased the amplitudes of both the inward and the outward currents. In particular, the currents at strongly depolarized potentials were elevated by ATP (the uppermost noisy current traces in Fig. 1A ). A similar strong activation of outwardly directed K⁺ currents was observed when cells from healthy donors were exposed to BzATP (50 μM; Fig. 1B ). Figure 1C illustrates the mean steady state currents of four cells from patients with PVR before (control) and during external exposure of Na-ATP (1 mM), and after washout of the drug. ATP increased the inwardly directed currents at negative membrane potentials and the outwardly directed current at positive potentials. During ATP exposure, the Müller cells depolarized, as indicated by the shift of the zero current potential of the whole-cell currents by 17.0 ± 5.3 mV toward more positive voltages (P < 0.01; inset in Fig. 1C ). 

To isolate the currents that were evoked by ATP-induced Ca²⁺ entry from the extracellular space, the whole-cell currents were recorded in two different bath solutions: one containing 0.5 mM Ca²⁺ and the other nominally Ca²⁺- and Mg²⁺-free and containing 1 mM EGTA. Figure 1D illustrates the Na-ATP (1 mM)–evoked currents that were recorded under the two conditions in cells from patients with PVR (the ATP-evoked currents were calculated by subtraction of the control currents from the currents recorded during exposure of ATP). The inwardly directed current (downward) was only slightly modulated by extracellular Ca²⁺. The density of the ATP-induced inward current was 5.32 ± 1.12 pA/pF with 0.5 mM Ca²⁺ in the bath solution (measured at the voltage step to −160 mV; n = 6), whereas in the Ca²⁺-free solution, it was 4.78 ± 1.53 pA/pF (n = 4; not significant). Thus, the inwardly directed current represents mainly a nonselective cation conductance, as also indicated by its reversal potential near 0 mV (Fig. 1D) . In contrast, the outwardly directed currents (upward) were strongly depressed after omitting the Ca²⁺ ions from the bath solution. In the Ca²⁺-containing bath solution, the current density of the ATP-sensitive currents was 11.3 ± 7.9 pA/pF (measured at the voltage step to +140 mV). In Ca²⁺-free conditions, the current density was reduced to 0.2 ± 1.5 pA/pF (P < 0.05). No significant inactivation of the ATP-induced cation conductance could be observed up to 5 minutes after beginning of the drug exposure. 

Similar results were obtained with external application of BzATP. As illustrated in Figure 1E , BzATP (50 μM) reversibly increased both inward and outward currents of Müller cells from patients with PVR, whereas the outward currents at positive potentials were much more increased than the inward currents at negative potentials. Moreover, BzATP depolarized the cells, as indicated by the positive shift of the zero current potential of the whole-cell currents (by 9.6 mV; inset in Fig. 1E ). The BzATP-induced outward currents displayed significantly different amplitudes when the drug was tested in Ca²⁺-containing or in Ca²⁺-free bath solution. In Ca²⁺-containing bath solution, the BzATP-induced current had a mean density of 17.3 ± 10.1 pA/pF (n = 7; measured at the voltage step to +140 mV), whereas in Ca²⁺-free solution the current density was only 3.3 ± 4.2 pA/pF (n = 8, P < 0.01). It is concluded that a large portion of the BzATP-induced outward current was evoked by Ca²⁺ entry from the extracellular space and may represent the activation of Ca²⁺-activated K⁺ currents. A similar increase of depolarization-evoked, Ca²⁺-activated K⁺ currents was observed in cells from healthy donors (Fig. 3) . 

Figure 2A illustrates the time course of the whole-cell current changes induced by BzATP (50 μM) exposure. Examples of records from four cells are shown that were made in Ca²⁺ (0.5 mM)-containing (Fig. 2A , left) or in Ca²⁺-free (Fig. 2A , right) bath solution. In Ca²⁺-containing bath solution, the cells showed a biphasic elevation of the amplitude of the outward currents at +120 mV. Just after beginning of drug exposure, there was a transient elevation of the current amplitude (asterisks) that thereafter switched into a sustained elevation (Fig. 2A , left). This biphasic response mimicked the ATP-induced biphasic increase of the intracellular Ca²⁺ concentration described previously in human Müller cells. ¹⁵ Therefore, it is assumed that the transient response was caused by release of Ca²⁺ ions from intracellular stores (through metabotropic ATP receptors), whereas the sustained response was mainly caused by a Ca²⁺ entry through an ATP-induced cation conductance (i.e., ionotropic ATP receptors). To test this assumption, cells were recorded in Ca²⁺-free extracellular solution (Fig. 2A , right). Indeed, under these conditions the majority of cells responded to BzATP with a large transient elevation of the outward currents (asterisks), whereas the sustained response was greatly reduced from 303.4% ± 124.1% to 124.3% ± 14.2% (as compared with the control currents set as 100%; P < 0.01; Fig. 2B ). The transient elevation of the mean outward current amplitude was virtually independent of extracellular Ca²⁺ (Fig. 2B) which strongly supports the assumption that this response is caused by Ca²⁺ release from intracellular stores. The release of intracellular Ca²⁺ was relatively fast and independent of the activation of the inwardly directed cation currents and could occur before, during, or after the onset of inward currents (Fig. 2A) . By contrast, the inward currents developed very slowly, and full activation of these currents was observed 1 to 2 minutes after the beginning of drug exposure. 

Exposure of the cells to thapsigargin (1 μM, 3 minutes) in Ca²⁺-free bath solution did not alter the sustained BzATP responses at +120 mV (amplitude 116.0% ± 17.2% compared with the value before drug exposure [100%], n = 5, not significant). Thapsigargin itself did not induce a transient activation of the outward currents (not shown). However, pretreatment with thapsigargin resulted in a significantly decreased amplitude of the BzATP-induced initial transient response (decrease to 51.1% ± 21.1% compared with five cells without pretreatment, P < 0.05), which is consistent with the assumption that a large portion of the transient activation of the outward currents by BzATP was caused by release of Ca²⁺ from intracellular stores. The specific types of P2Y receptors involved in the Ca²⁺ release from internal stores remain to be determined in future experiments. 

To determine whether the outward current elevated by BzATP represents a BK-channel–mediated current (I _BK), the effect of iberiotoxin was tested. Figure 3A illustrates an example of current records in one Müller cell from a healthy donor. Extracellular application of BzATP (50 μM) induced a strong increase of the outwardly directed currents that was blocked by simultaneous exposure to iberiotoxin (100 nM). The time course of the drug’s effect is shown in Figure 3B for one cell. Iberiotoxin fully reversed the BzATP-induced increase of the outward currents at +120 mV but had no effect on the inward currents at −100 mV. The voltage-dependence of the whole-cell currents reveals that BzATP induced a negative shift of the activation of I _BK; under control conditions, the activation threshold of the I _BK was at+ 120 mV, and during BzATP exposure, the I _BK was evoked at potentials positive to 0 mV (Fig. 3C) . Iberiotoxin blocked the BzATP-activated I _BK measured at +120 mV (Fig. 3D , left), but did not influence the amplitude of the BzATP-induced inwardly directed currents measured at −100 mV (Fig. 3D , right). These results indicate that activation of P2 receptors enhances the amplitude of I _BK in human Müller cells and that the depressing effect of iberiotoxin on I _BK was not caused by inhibition of the ATP-induced cation conductance. 

In excised outside–out patches, BzATP reversibly increased the activity of iberiotoxin-sensitive (Fig. 4A ) K⁺ channels of large conductance (134.3 ± 16.6 pS). As shown in Figure 4B , exposure to BzATP (50 μM) of the extracellular side of the membrane increased the open probability of BK channels at positive membrane potentials. In three patches, BzATP increased the mean channel-open probability at +80 mV from 0.04 ± 0.05 to 0.14 ± 0.10. After washout of the drug, the value returned to 0.04 ± 0.04. 

The expression of P2X₇ receptors was electrophysiologically investigated in freshly isolated Müller cells from various species (human [n = 318 cells], pig[ n = 14], rat [n = 8], mouse [n = 9], guinea pig [n = 6], and rabbit [n = 18]), by using extracellular application of BzATP (50 μM). It was found, however, that BzATP induced a cation conductance only in the case of human Müller cells, whereas metabotropic P2Y receptors seem to be present also in Müller cells from other species (data not shown). Therefore, the expression of P2X₇ receptors by Müller cells may be a specific phenomenon of the human retina. 

Taken together, the results indicate that extracellular ATP may have three effects on human Müller cells: (1) Through activation of P2X₇ receptors, ATP evokes the opening of a nonselective cation conductance which mediates a Ca²⁺ entry from the extracellular space and which depolarizes the cells; (2) Ca²⁺ ions are released from intracellular stores, probably through an activation of P2Y receptors; and (3) the depolarization and the elevation of the intracellular Ca²⁺ concentration together increase I _BK. 

Disease-Related Changes of BzATP-Evoked Currents

To determine whether P2 receptors of the Müller cells are implicated in transdifferentiation processes accompanying gliosis in cases of human retinal disease, two kinds of retinal diseases were investigated: choroidal melanoma and PVR. Müller cell gliosis is characterized by different features, including upregulation of the immunoreactivity of intermediate filaments and by cell hypertrophy. In the case of PVR, Müller cells may also become proliferative, whereas in other types of gliosis (e.g., during choroidal melanoma), Müller cells do not proliferate. Müller cells from patients with PVR displayed a significantly greater cell membrane capacitance (81.5 ± 22.4 pF, n = 19 patients) compared with cells from healthy retinas (54.3 ± 10.9 pF, n = 13 donors, P < 0.001) indicating hypertrophy of Müller cells in PVR retinas. Similarly, Müller cells from three patients with melanoma displayed hypertrophy (mean membrane capacitance, 93.0 ± 12.3 pF, n = 3). 

Whole-cell currents were recorded before and during exposure to BzATP (50 μM) in K⁺-containing or K⁺-free solutions. Figure 5A shows the mean steady state current density–voltage relations of BzATP-evoked currents measured in K⁺-containing solutions after the transient stimulations of I _BK had ceased (in most cells, 2 to 3 minutes after the beginning of drug exposure; currents were calculated by subtraction of the control currents from the drug-induced currents). Mean curves from three donor groups are shown: healthy donors, patients with PVR, and patients with melanoma. The BzATP-evoked currents consist of two components: At negative membrane potentials, the inwardly (downwardly) directed currents reflect the activation of the BzATP-evoked cation conductance, and, at positive potentials, the outwardly directed currents reflect the amplitude increase of the sustained I _BK. The density of the inwardly directed cation currents evoked by BzATP was significantly greater in cells from patients with PVR (2.9 ± 1.1 pA/pF, measured between −100 and −140 mV) than in cells from healthy donors (1.1 ± 0.6 pA/pF, P < 0.001). The elevated amplitude of the inwardly directed currents was accompanied by a stronger stimulation of I _BK at positive membrane potentials. However, in cells from the patients with melanoma, both the BzATP-evoked inward currents (0.53 ± 0.25 pA/pF) and the BzATP-stimulated I _BK were in the same range as found in cells from healthy donors. 

To rule out the possibility that the disease-related alterations of the BzATP-induced cation conductance were artificially caused by alterations of K⁺ conductances (e.g., by Ca²⁺ _i-induced decrease of I _Kir ⁷ ), the BzATP-evoked cation conductance was measured under K⁺-free conditions. In fact, the density of the BzATP-evoked inward currents at a holding potential of −80 mV was significantly greater in cells from patients with PVR (3.6 ± 3.0 pA/pF, n = 58 cells) than in cells from healthy donors (2.1 ± 1.4 pA/pF, n = 70 cells, P < 0.001). 

Figure 5B shows mean values of the amplitudes of the maximally BzATP-stimulated I _BK from the three donor groups. Although the amplitudes of the transient I _BK were similar in all groups, the amplitude of the sustained I _BK was significantly greater in cells from patients with PVR compared with healthy donors. This indicates a specific upregulation of P2X₇ receptor-mediated Ca²⁺ entry, whereas the P2Y receptor-mediated release of intracellular Ca²⁺ was not altered in pathologic conditions. 

Figures 5C and 5D show the voltages at which the sustained I _BK was half-maximally activated, and the densities of the maximal I _BK, respectively, for cells from diseased and healthy retinas. Both parameters were significantly correlated with the density of the BzATP-evoked cation conductance. The higher the density of the BzATP-evoked inward currents, the more shifted the voltage at which I _BK was half-maximally activated toward more negative membrane potentials, and the higher the maximal density of stimulated I _BK. Probably, this was mediated by the intracellular concentration of Ca²⁺. We did not find any correlation of the density of the ATP-gated cation currents with the age of the donors (not shown). 

Modulation of the DNA Synthesis Rate by Extracellular ATP

Both extracellular Na-ATP (500 μM) and BzATP (20 μM) increased the DNA synthesis rate of cultured human Müller cells (Fig. 6A ). Simultaneous exposure to iberiotoxin (70 nM) decreased the effects of the purinergic agonists indicating an involvement of BK channel activity in mediating the proliferation-stimulating effect of extracellular ATP. 

Discussion

I _BK of Human Müller Cells

In the whole-cell records (e.g., Figs. 1 and 3 ), I _BK activated at very positive membrane potentials (positive to +100 mV; Fig. 1C ). Application of extracellular ATP induced a shift of the I _BK activation threshold toward more positive potentials (Figs. 1C 3C) . The I _BK activation at positive membrane potentials was previously described using whole-cell records in Müller cells of several mammalian species ^{22 23} including humans. ⁵ However, this activation at positive potentials does not reflect the real activation state of BK channels in Müller cells, because in cell-attached records, single BK channel activity can be observed at the native resting membrane potential and at slightly depolarized potentials. ^{5 22 23 24} In Müller cells from patients with PVR, the single BK channel activity at the resting membrane potential was even found to be significantly increased if compared with cells from healthy donors. ⁵ Very probably, the strong positive shift of the I _BK activation in whole-cell records was mainly caused by the fact that the artificial pipette solution (replacing the normal intracellular milieu in the whole-cell configuration) did not contain essential cytoplasmic components that may have coactivated BK channels, such as activators of protein kinase A. ^{3 24}  

Purinergic Receptors of Müller Cells

In human Müller cells, both ATP and BzATP activate a nonselective, noninactivating cation current, ¹⁵ consistent with the involvement of P2X₇ receptors. The nonselective cation channels allow for an entry of Ca²⁺ ions from the extracellular space ¹⁵ that causes a sustained increase of I _BK (Fig. 1D) . ATP and BzATP caused also transient elevations of I _BK that persisted in Ca²⁺-free extracellular solution (Fig. 2A) , indicating that they induced a release of Ca²⁺ from thapsigargin-sensitive intracellular stores. Although BzATP is assumed to preferentially activate ionotropic P2X₇ receptors the present results indicate that BzATP also stimulates internal Ca²⁺ release, as previously described for other cell types. ²⁵ The mechanism of the BzATP-induced release of internal Ca²⁺ remains to be identified. This may be a secondary step after activation of P2X₇ receptors or a direct (additional) activation of P2Y receptors by BzATP, as previously shown for P2Y₂ receptors. ²⁶ The presence of P2Y receptors in human Müller cells was also indicated by preliminary experiments that showed that, in addition to ATP, extracellular uridine triphosphate (UTP) and guanosine triphosphate (GTP) evoke a transient activation of I _BK (not shown). 

Involvement of Purinergic Receptors in Müller Cell Gliosis

The involvement of purinergic receptors in induction or maintenance of gliosis in vivo has been previously discussed. ^{17 19} In the brain, activation of P2 receptors may induce astrogliosis, leading to hypertrophy and proliferation of astrocytes. ^{18 27} In the current study, we showed for the first time that gliosis in vivo may be connected with an upregulation of a distinct type of purinergic receptor. Although the currents through P2X₇ receptor channels were upregulated in Müller cells from patients with PVR compared with healthy donors, the release of intracellular Ca²⁺ that is probably mediated by P2Y receptors was unchanged in these cells. However, further investigations are necessary to provide evidence for a causal relationship between the expression of P2X₇ receptor-mediated currents and the induction or maintenance of Müller cell gliosis. 

Although gliosis was present in eyes with choroidal melanoma, evidenced by a hypertrophy of the cells, all other investigated membrane conductances, including the density of P2X₇ receptor currents, were in the range of the cells from healthy retinas. By contrast, gliosis accompanying PVR was characterized by hypertrophy, by a strong downregulation of I _Kir and a less negative resting membrane potential, ^{5 20} by an increased expression of voltage-gated Na⁺ channels, ²⁸ and by a decrease of currents through voltage-gated Ca²⁺ channels, ²⁹ whereas the density of P2X₇ receptor currents was upregulated. When data from all donor groups used in the present study were considered, an increase of the density of sustained BzATP-evoked inward currents was correlated with a decrease of the I _Kir density (r =− 0.574, n = 31 donors, P < 0.001), a depolarization of the membrane (r = 0.596, P < 0.001), a decrease of the peak currents through high-voltage–activated Ca²⁺ channels (r = −0.600, P < 0.001), and a higher density of peak Na⁺ currents (r = 0.604, P < 0.001). As a mean, the more membrane features were altered the stronger the upregulation of P2X₇ receptor-mediated currents by human Müller cells. This may implicate a causal relationship between purinoceptor activation and the strength of gliosis. 

I _BK and ATP-Induced Müller Cell Proliferation

A specific upregulation of P2X₇ receptor currents in cells from patients with PVR may indicate that this type of purinergic receptors is involved in processes that are activated when gliotic Müller cells become proliferative. A role of P2X₇ receptors in induction and/or maintenance of proliferation was previously described for lymphocytes. ^{30 31} Moreover, many tumor cell lines are characterized by high P2X₇ receptor expression levels. ³² We found that both ATP and BzATP stimulated the DNA synthesis rate of cultured human Müller cells (Fig. 6A) . Coapplication of iberiotoxin fully reversed the effects of the purinergic agonists. When we excluded an unspecific effect of iberiotoxin on Müller cells, the data indicated that the activation of I _BK may be a step that is necessary for the purinergic induction of Müller cell proliferation. The mechanism of involvement of I _BK in the regulation of the proliferation is unclear. One mechanism may be the regulation of the strength of Ca²⁺ entry into Müller cells by BK channels, ³³ as it was shown for the agonist-induced Ca²⁺ entry in other cell types. ³⁴ Because elevated intracellular Ca²⁺ concentration is necessary for both gliosis and maintenance of proliferative activity, the increased expression of P2X₇ receptors and the subsequent stronger activation of I _BK may promote the induction of both processes in cells from PVR retinas. However, further experiments are necessary to investigate the relationships between purinergic receptor activation, BK current stimulation and induction and/or maintenance of Müller cell proliferation. 

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

 Submitted for publication September 7, 2000; revised November 15, 2000; accepted November 29, 2000.

 Commercial relationships policy: N.

 

Figure 1.

 

Extracellular application of ATP modulated the whole-cell currents in human Müller cells. The records were made under steady state conditions, 2 to 3 minutes after the beginning of drug exposure. In these and all experiments that are presented in the figures, K⁺-containing bath and pipette solutions were used. (A) Example of whole-cell records in a cell from a patient with PVR that had no I _Kir before (control) and during exposure to Na-ATP (1 mM) and after washout of the drug. Depolarizing (up to +140 mV) and hyperpolarizing voltage steps (up to −160 mV) were applied in increments of 20 mV. The holding potential was −80 mV. ATP evoked an increase in the currents, particularly at strongly depolarized potentials (uppermost noisy current traces). (B) Example of records in a cell from a healthy donor before, during, and after exposure to BzATP (50 μM). Depolarizing (up to +160 mV) and hyperpolarizing voltage steps (up to− 160 mV) were applied from a holding potential of −80 mV (increment 20 mV). (C) Mean (± SD) voltage dependence of the steady state currents in four cells from a patient with PVR. The whole-cell currents were recorded before (control) and during external exposure of Na-ATP (1 mM) and after washout of the drug. Inset: A portion of the current–voltage curve showing the agonist-induced shift of the zero current potential. (D) Mean (± SD) voltage dependence of the Na-ATP (1 mM)-induced currents in cells from a patient with PVR. In six cells, the whole-cell currents were recorded in a bath solution containing 0.5 mM Ca²⁺; in another four cells, the currents were recorded in a bath solution that contained EGTA (1 mM) and no added Ca²⁺ (0 Ca; n = 4). The ATP-induced currents were calculated by subtraction of the control currents from the currents recorded during exposure of ATP. (E) Mean (± SD) voltage dependence of the steady state currents in seven cells from a patient with PVR. The whole-cell currents were recorded before (control) and during external exposure of BzATP (50 μM) and after washout of the drug. Inset: As in (C).

Figure 2.

 

Time-dependent changes of the membrane conductance of human Müller cells in response to extracellular application of BzATP (50 μM). The records were made in cells from patients with PVR. (A) Examples of records in four cells that showed transient elevations of the outward currents after the beginning of drug exposure ( ). The records were made in bath solutions containing 0.5 mM Ca²⁺ (left) or in Ca²⁺-free bath solutions containing 1 mM EGTA (right). Voltage steps were applied at a frequency of 2.5 Hz from a holding potential of −60 mV (middle traces) to +120 mV (upper traces) and to −100 mV (lower traces; inset). Dashed lines: Zero current levels. The amplitudes were measured at the end of each of the 50-msec voltage steps. Arrows: Beginning of drug exposure; horizontal bars: 10 seconds; vertical bars: 250 pA. (B) Mean amplitudes of the transient and sustained outward currents activated by BzATP. The cells were examined in Ca²⁺-free (n = 6) or in Ca²⁺-containing bath solution (n = 7). The amplitudes of the transient currents and of the sustained currents were measured at the end of 50-msec voltage steps to +120 mV from a holding potential of −60 mV and are calculated as the percentage of the control currents that were measured 20 seconds before drug application (100%). Significant differences of P < 0.05 (•) and of P < 0.01 (••). n.s., not significant.

Figure 3.

 

BzATP increased the amplitude of iberiotoxin-sensitive outward currents in human Müller cells. (A) Example of current records in a cell derived from a healthy human donor. For current activation, the membrane was stepped to increasing depolarizing and hyperpolarizing potentials between −160 and +200 mV (250 msec, 20-mV increments, holding potential −80 mV). Cells were exposed to BzATP (50 μM) in the absence and thereafter in the presence of iberiotoxin (100 nM). (B) Time course of whole-cell current changes in one cell. Bars: Application of BzATP (50 μM) and iberiotoxin (100 nM). Currents were measured at +120 mV (upper trace), at− 60 mV (middle trace), and at −100 mV (lower trace). Dashed line: Zero current level. For voltage step protocol, see inset in Figure 2A . (C) Mean steady state current versus voltage relationships of three cells. Currents were recorded before (control) and during BzATP (50 μM) exposure and during simultaneous exposure to BzATP and iberiotoxin (100 nM). Currents were evoked using the voltage step protocol shown in (A). (D) Mean amplitudes of the sustained outward currents at +120 mV (left) and of the inwardly directed currents at −100 mV (right) in 12 cells from patients with PVR. The currents were measured during BzATP (50 μM) exposure, in the absence and in the presence of iberiotoxin (100 nM). Significant differences of P < 0.05 (•) and of P < 0.001 (•••). The currents were evoked using the continuous step protocol shown in the inset of Figure 2A .

Figure 4.

 

Single BK channels in outside–out membrane patches were activated by BzATP. (A) The activity of large-conductance K⁺ channels in a patch that was excised from the end foot membrane of a cell from a healthy donor was reversibly inhibited by iberiotoxin to the extracellular side of the membrane. Arrows: Closed state level. (B) Exposure of the extracellular side of outside–out membrane patches to BzATP reversibly increased the open probability of BK channels. Example of channel record in a patch that was excised from the soma membrane of a cell from a healthy donor. The open probability–voltage curve was calculated from records in which positive voltage steps were applied from a holding potential of 0 mV.

Figure 5.

 

The BzATP-evoked sustained currents are increased in human Müller cells from PVR retinas compared with those in cells from healthy donors. (A) Mean current density–voltage relations of BzATP (50 μM)-evoked sustained whole-cell currents in human Müller cells from three donor groups. The currents were calculated by subtracting the control currents from the currents that were recorded during agonist application. The currents were evoked by depolarizing and hyperpolarizing voltage steps to the potentials indicated from a holding potential of −80 mV. The steady state currents were measured at the end of the 250-msec voltage steps. (B) Mean amplitudes of the transient and sustained I _BK activated by BzATP (50 μM). Maximal amplitudes were measured at the voltage step from −60 to +120 mV. Donor numbers in parentheses. (A, B) Significant differences between cells from PVR retinas and cells from healthy donors of P < 0.01 (•) and P < 0.001 (••) respectively. n.s., not significant. Dependences of the voltage, at which the sustained I _BK was half-maximally activated (C) and of the density of the maximally activated sustained I _BK (D), from the density of the BzATP-evoked inward currents. The BzATP-evoked inward currents were measured at the voltage step from −100 mV to −140 mV. (C, D) Each symbol represents the mean of all investigated cells from one donor.

Figure 6.

 

Iberiotoxin (70 nM) decreased the Na-ATP (500 μM)- and BzATP (20μ M)-induced DNA synthesis in cultured human Müller cells. Mean DNA synthesis rates of four independent cultures. (•) Significant differences of P < 0.05.