Basal Calcium Entry in Retinal Pigment Epithelial Cells Is Mediated by TRPC Channels

Abstract

purpose. Ca²⁺ is a major regulator of cell function. In the retinal pigment epithelium (RPE), intracellular free Ca²⁺ concentration ([Ca²⁺]_i) is essential for the maintenance of normal retinal function. Therefore, accurate control of [Ca²⁺]_i is vital in these cells. Because Ca²⁺ is permanently extruded from the cytosol, RPE cells need a basal Ca²⁺ entry pathway that counteracts this Ca²⁺ efflux. The purpose of this study was to identify the molecular basis of basal Ca²⁺ entry into the RPE.

methods. [Ca²⁺]_i was measured using Fura-2–loaded ARPE-19 cells. The expression pattern of TRPC channels was investigated by RT-PCR with RNA extracted from ARPE-19 cells and freshly isolated RPE cells from human donor eyes.

results. In most cells, basal [Ca²⁺]_i is highly controlled by cell membranes that are only slightly permeable to Ca²⁺ and by the activity of Ca²⁺ pumps and transporters. The authors show here that RPE cells have a basal Ca²⁺ conductance that is dose dependently blocked by La³⁺. Basal [Ca²⁺]_i was also strongly reduced by the TRP channel blockers Gd³⁺, Ni²⁺, 2-APB, and SKF96365 and was insensitive to blockers of other Ca²⁺ channels. In confirmation of this pharmacologic profile, RPE cells expressed TRPC1 and TRPC4 channels, as shown by RT-PCR experiments.

conclusions. Ca²⁺ is needed for several permanently occurring regulatory processes in RPE cells. The Ca²⁺ influx pathway identified in this study is essential to define a resting basal [Ca²⁺]_i. This resting [Ca²⁺]_i may contribute, for example, to basal cytokine secretion essential for the maintenance of normal retinal function.

Many cellular processes are controlled by the intracellular free Ca²⁺ concentration ([Ca²⁺]_i). ^{1 2} These include excitation, secretion, cell differentiation, gene expression, endocytosis, and apoptosis. Altered Ca²⁺ homeostasis may lead to physical impairment, as seen in genetic diseases associated with Ca²⁺ transporters and channels. ³ Generally, two pathways for the elevation of [Ca²⁺]_i exist: Ca²⁺ may enter through plasma membrane Ca²⁺ channels or Ca²⁺ may be released from intracellular Ca²⁺ stores by the activation of ryanodine or inositol trisphosphate receptors. For each case, the cells use a variety of Ca²⁺-transporting proteins. ² The Ca²⁺ extrusion is mediated by Ca²⁺ exchangers and pumps. In resting cells in which the [Ca²⁺]_i is approximately 100 nM, the Ca²⁺ influx, efflux, and intracellular Ca²⁺ buffering are in equilibrium. 

The retinal pigment epithelium (RPE) is located between the neural retina and the choroidal vasculature. ^{4 5} With its tight junctions, it forms part of the blood-retina barrier and is important for the maintenance of retinal function. Its pigment absorbs excess light, it reisomerizes all-trans retinal to 11-cis retinal and delivers it back to the photoreceptors, it controls the transport of metabolites, nutrients, ions, and water between the subretinal space and the choroidal vessels, and it phagocytes shed outer segments of photoreceptors. As a target and source of a variety of growth factors, the RPE maintains retinal integrity and is part of the immune privilege of the subretinal space. ^{4 5 6} Many of these tasks are influenced by changes in the [Ca²⁺]_i. ⁷ Therefore, accurate control of [Ca²⁺]_i in the RPE is vital. 

In the RPE, it has been shown that Ca²⁺ influx through the voltage-operated L-type Ca²⁺ channel α1D plays a role in the control of growth factor secretion. ⁸ Additionally, a specific block of L-type channels led to a reduced light peak in electroretinograms in mice and rats, indicating that these channels are involved in light-induced responses of the RPE. ^{9 10} Furthermore, purinergic stimulation of RPE cells leads to an increase in [Ca²⁺]_i, which seems to be at least partially mediated by ionotropic purinergic receptors (P2X). ¹¹ This ATP-induced [Ca²⁺]_i increase leads to increased transepithelial Cl⁻ and water transport. 

Both L-type and purinergic receptor channels open only in response to specific stimulation, such as depolarization (α1D) or binding of ATP (P2X). Ca²⁺ is permanently extruded from RPE cells through constitutively active plasma-membrane Ca²⁺ ATPases and Na⁺/Ca²⁺ exchangers. ^{12 13} Therefore, unstimulated RPE cells seem to have a Ca²⁺ leak pathway for Ca²⁺ influx that stabilizes the basal [Ca²⁺]_i of 100 nM. In another study we have already demonstrated that this Ca²⁺ influx is not driven by a reverse mode of the Na⁺/Ca²⁺ exchanger. ¹⁴ Here we show in addition that it is not driven by the already identified voltage-operated Ca²⁺ channels or the ionotropic purinergic receptor-operated channels. Instead, we show by pharmacologic reduction of the basal [Ca²⁺]_i that this background Ca²⁺ influx is carried out by a member of the transient receptor potential (TRP) channels. 

Materials and Methods

Cell Culture

The human retinal pigment epithelial cell line ARPE-19 was cultured in Dulbecco modified eagle medium/F-12 nutrient mixture (D-MEM/F-12) containing 10% fetal bovine serum, insulin-transferrin-sodium (Roche, Basel, Switzerland), nonessential amino acids, and penicillin/streptomycin at 37°C in a humidified ambient atmosphere containing 5% CO₂. They were passaged twice per week. For Fura-2 measurements, they were seeded onto coverslips and cultured to confluence. 

For primary cultures of human RPE cells, the anterior part of human donor eyes, including vitreous and retina, were removed. The RPE and choroidea were carefully separated from the sclera, washed with PBS, and incubated overnight with collagenase IA/IV (0.5 mg/mL each) in serum-free culture medium. The dissociated cells were collected by centrifugation (50g, 5 minutes) and cultured on coverslips in the same medium as the ARPE-19 cells. These still pigmented RPE cells were used for Ca²⁺ measurements after they reached confluence. Human material was used in accordance with the tenets of the Declaration of Helsinki. 

Measurement of Intracellular Free Ca²⁺ Concentrations

ARPE-19 cells grown on coverslips to confluence were washed with Ringer solution (130 mM NaCl, 5 mM KCl, 2 mM MgCl₂, 2 mM CaCl₂, 5 mM glucose, 10 mM HEPES, pH 7.3, with NaOH) and loaded with Fura-2 AM ester (Fluka, Buchs, Switzerland) for 40 minutes in the dark at room temperature in Ringer solution containing 10 μM Fura-2 AM. Cells were washed and incubated for at least 30 minutes with Ringer solution. The coverslips were placed into a bath chamber perfused constantly with Ringer solution and mounted onto an inverted microscope (Axiovert 35; Carl Zeiss, Oberkochen, Germany) equipped with a 40× objective (Fluar; Carl Zeiss). To exclude the possible stimulation of mechanosensitive channels by the perfusion system, we stopped the perfusion and found no change in [Ca²⁺]_i. We performed ratiometric measurement Fura-2 fluorescence at 5-second intervals using a high-speed polychromator system (VisiChrome; Visitron Systems, Puchheim, Germany) altering the wavelength of excitation light between 340 and 380 nm. Emitted light was filtered with a 510-nm filter and was detected by a cooled charged-coupled device camera (CoolSNAP; Roper Scientific GmbH, Ottobrunn, Germany). Data were collected (MetaFluor software; Universal Imaging) and analyzed (MetaAnalysis software; Universal Imaging). Intracellular free Ca²⁺ ([Ca²⁺]_i) was calculated from the Fura-2 fluorescence ratio (F₃₄₀/F₃₈₀). ¹⁵ Mean fluorescence intensities after excitation with 340 nm were between 153 and 197 arbitrary units, indicating equal loading of the cells throughout all experiments. 

RNA Isolation and RT-PCR

Human RPE was obtained from organ donors within 24 hours of death. After the cornea was removed for transplantation, the eyes were subjected for RPE preparation. The anterior parts of the eyes, including the vitreous and the retina, were removed. The posterior part was rinsed with ice-cold PBS (without Ca²⁺ and Mg²⁺) to wash away residual material from the neural retina. With the use of fine forceps, the RPE was gently brushed away. RPE cells were collected and lysed in lysis buffer (RNeasy Mini Kit; Qiagen, Valencia, CA). Total RNA from ARPE-19 cells was prepared from confluent cultures grown in a 25-cm² culture flask. RNA was isolated (RNeasy Mini Kit; Qiagen) according to manufacturer’s instructions. RNA (1 μg) was reverse transcribed at 37°C for 1 hour in the following reaction mixture: 1 μg oligo dT primer (Invitrogen, Carlsbad, CA), 1 mM of each dNTP, 20 U RNAguard (Amersham Biosciences, Freiburg, Germany), and 20 U M-MLV reverse transcriptase (Invitrogen). For control PCR reactions, human total brain RNA (Stratagene, La Jolla, CA) was reverse transcribed under the same conditions. PCR experiments were performed with 1 μL cDNA in 50-μL PCR reaction mixtures with Taq DNA polymerase (Stratagene) and 1.5 pmol of sense and antisense oligonucleotides specific to the various TRPC channel subunits (Table 1) . The identity of the amplification product was confirmed by sequencing. 

Data Analysis

Results were presented as mean ± SEM. Statistical significance was tested using one-way analysis of variance (P < 0.05; statistical significance; P < 0.01, strong statistical significance; P < 0.001, very strong statistical significance). 

Results

Blocking of the plasma membrane Ca²⁺ ATPase in RPE cells by the application of 2 mM orthovanadate led to a pronounced increase in [Ca²⁺]_i. The [Ca²⁺]_i increased to 216.78% ± 10.46% of the basal value (n = 4; Fig. 1 ). Given that plasma membrane Ca²⁺ ATPases are responsible for Ca²⁺ efflux to maintain the low [Ca²⁺]_i, this increase must have been caused by a permanent Ca²⁺ influx through a sustained membrane conductance for Ca²⁺. 

It has already been shown that RPE cells have some types of Ca²⁺-conducting ion channels, such as voltage-operated L-type Ca²⁺ channels and purinergic receptors. Therefore, we first tested the influence of blockers of these currents on the resting [Ca²⁺]_i. As shown in Figure 2 , neither 10 μM nifedipine (a blocker of L-type Ca²⁺ channels, n = 9) nor 50 μM PPADS (a blocker of P2X channels, n = 7) had any effect on [Ca²⁺]_i, suggesting that there should be background Ca²⁺ channels distinctive from voltage-operated Ca²⁺ channels and P2X channels expressed in these cells. 

Because lanthanides are known to inhibit Ca²⁺ influx, we tested the effect of La³⁺ and Gd³⁺ on basal [Ca²⁺]_i in RPE cells. Both ions reduced [Ca²⁺]_i significantly; 100 μM La³⁺ led to a reduction to 25.26% ± 7.21% (n = 10), and 100 μM Gd³⁺ to 14.57 ± 2.67% (n = 5) of the resting [Ca²⁺]_i (Fig. 3) . Although La³⁺ is an unspecific blocker of different Ca²⁺ channels, the sensitivity of the individual Ca²⁺ channel types to La³⁺ is very different. Therefore, we investigated the concentration dependence of the La³⁺ effect on the basal [Ca²⁺]_i. At the lowest concentration of 0.2 μM, La³⁺ already reduced basal [Ca²⁺]_i to 39.7% ± 11.72% (n = 7; Fig. 4 ). Ni²⁺ is also known to inhibit Ca²⁺ influx through different Ca²⁺ channels. Application of 2 mM Ni²⁺ reduced the [Ca²⁺]_i to 33.08% ± 8.95% (n = 6). 

Lanthanides and Ni²⁺ are relatively unspecific inhibitors that affect various ion channels and transporters, but at low concentrations they are known to be efficient blockers of some members of the TRP channel family. In addition, we used blockers influencing a more narrow number of channels, including TRPC channels. The application of 2-APB and SKF 96365 to RPE cells resulted in a decrease of [Ca²⁺]_i, comparable to that seen with La³⁺ or Gd³⁺. 2-APB (75 μM, n = 8) and SKF 96365 (50 μM, n = 8) led to a decrease of [Ca²⁺]_i, to 23.55% ± 5.83% and 63.78% ± 8.85%, respectively (Fig. 5) . This reduction of basal [Ca²⁺]_i was qualitatively reproduced with primary cultures of RPE cells from human donor eyes (Fig. 6) . The basal [Ca²⁺]_i was reduced in these cells by the application of 75 μM 2-APB and 50 μM SKF 96365 to 55.7% ± 9.38% and 67.36% ± 4.63%, respectively. 

The pharmacologic profile of blockage of basal Ca²⁺ entry observed here suggested that it was driven by members of the TRPC subfamily. RT-PCR with RNA from the RPE cell line ARPE-19 and from freshly isolated RPE cells showed that TRPC1 and TRPC4 were expressed in the RPE. TRPC7 was also expressed in freshly isolated cells (Fig. 7) . 

Discussion

In this study, we showed for the first time that RPE cells have large basal membrane permeability for Ca²⁺, as indicated by an increase in [Ca²⁺]_i after inhibition of the plasma membrane Ca²⁺ ATPase. This basal Ca²⁺ conductance was inhibited by blockers with the same pharmacologic profiles as TRPC channels. We were able to exclude the contribution of voltage-operated Ca²⁺ channels and purinergic receptors to basal Ca²⁺ entry. By RT-PCR we confirmed the expression of TRPC channels in the RPE. 

The RPE is a target for and a source of various cytokines whose intracellular signaling cascades are coupled to [Ca²⁺]_i. Furthermore, [Ca²⁺]_i changes are involved in many other RPE functions, such as photoreceptor outer segment phagocytosis, transcellular fluid and ion transport, cell differentiation, and the control of gene expression. ⁵ Nevertheless, until now, only voltage-operated L-type Ca²⁺ channel α1D expression has been shown in RPE cells. These channels are involved in growth factor signaling in the RPE. ^{8 16} In addition to these studies about voltage-operated channels, only one other study based on pharmacologic evidence of purinergic receptors suggests that ionotropic purinergic receptors (P2X) are involved in the regulation of fluid and ion transport across the RPE. ¹¹ In addition, it has been reported that NMDA receptors are expressed in the RPE, ^{17 18 19 20 21} though it seems unlikely that these channels may contribute to resting [Ca²⁺]_i because they need glutamate for stimulation. 

Thus, considering the multitude of different [Ca²⁺]_i-regulated RPE cell functions, additional Ca²⁺ channels are likely to be expressed in the RPE. TRP channels are good candidates because they are coupled to a variety of intracellular signaling pathways. RPE cells use energy to continuously extrude Ca²⁺ effectively from the intracellular space through a plasma-membrane Ca²⁺ ATPase and a Na⁺/Ca²⁺ exchanger. Because some of the Ca²⁺-induced processes, such as basal VEGF secretion, must persist without stimulation, RPE cells need an equally effective Ca²⁺ entry pathway. To our knowledge, such a leakage Ca²⁺ pathway is only described for vascular smooth muscle cells, ^{22 23 24} osteoblastlike cells, ²⁵ and chromaffin cells. ²⁶  

Our findings that La³⁺, Gd³⁺, Ni²⁺, 2-APB, and SKF 96365 inhibited this nonstimulated basal Ca²⁺ entry indicated that TRP channels are the molecular correlate of this Ca²⁺ leak. Although none of these blockers is specific for a particular class of TRP channels, conclusions concerning the molecular identity of the channels can be drawn from the combination of their effects. TRPV channels are either insensitive to or activated by 2-APB. ²⁷ Accordingly, these channels can be excluded from their possible role in basal Ca²⁺ entry in RPE cells. TRPM channels can be excluded because of their insensitivity to Ni²⁺. Thus far, it has been found that TRPM channels are highly permeable to Ni²⁺. ^{28 29} Therefore, we concentrated on the TRPC channels. 

Our RT-PCR experiments revealed that TRPC1 and TRPC4 are expressed in ARPE-19 and cultured RPE cells from human donor eyes. In the latter, we also found TRPC7 to be expressed. La³⁺ and Gd³⁺ block a variety of TRP channels. ³⁰ They also inhibit a variety of other ion channels and transporters. ^{31 32 33 34 35} All non-TRP channels, however, have a much lower La³⁺ sensitivity than the leak Ca²⁺ channel investigated in this study. Voltage-operated Ca²⁺ channels had an IC₅₀ of 1.1 mM. ³² Most TRP channels have much higher La³⁺ sensitivity than voltage-operated Ca²⁺ channels. Nevertheless, their sensitivity is much higher (IC₅₀, 3–100 μM) than the IC₅₀ for La³⁺ block of basal Ca²⁺ entry in RPE cells. ^{36 37 38 39}  

Some study results conflict with ours. In these studies, TRPC4-mediated currents were potentiated rather than blocked, as in our study, by La³⁺ in micromolar concentrations. ^{40 41 42} In other studies, TRPC4 channels were blocked by La³⁺. ^{43 44 45} Comparative studies of wild-type and TRPC4 knock-out mice revealed very high La³⁺ sensitivity of these native TRPC4 channels in the nanomolar range. ^{43 46} This discrepancy might be explained by the fact that the potentiation of these currents could only be observed in the heterologous expression system, whereas studies indicating high La³⁺ sensitivity of TRPC4 were conducted with endogenously expressed channels in native cells. 

TRPC4 has been shown to coassemble with TRPC1 to form heteromultimeric channels. ⁴⁷ These heteromultimeric channels might be dominated by the high La³⁺ sensitivity of native TRPC4 channels. 

In another single study on TRP channels in the RPE, it was found that only TRPC1 is expressed in ARPE-19 cells. ⁴⁸ TRPC4 is known to be differentially spliced. ^{41 49 50} Because we amplified different parts of the gene, our results may be attributed to a splice variant that Bollimuntha et al. ⁴⁸ could not detect with their oligonucleotides. In addition, we confirmed the expression of TRPC1 and TRPC4 in native RPE cells. We also found TRPC7 to be expressed in native RPE cells. It has been thought that heteromultimers could only be formed among TRPC1, TRPC4, and TRPC5 and among TRPC3, TRPC6, and TRPC7. ⁴⁷ Nevertheless, some studies show that heteromultimers can also be formed between these groups. ^{51 52} Hence, the additional expressed TRPC7 might be a part of a heteromultimeric channel carrying the basal influx into native RPE cells. 

Ca²⁺ is one of the fundamental intracellular signaling molecules. TRPC channels mediating the leak Ca²⁺ entry to the RPE may be involved in basal cellular processes controlled by [Ca²⁺]_i, such as basal secretion of cytokines. Additionally, TRPC channels are coupled to intracellular signaling pathways that are activated by G protein–coupled receptors. On stimulation, these channels may contribute not only to the adjustment of resting [Ca²⁺]_i but also to that of elevated [Ca²⁺]_i. ⁵³ Because TRPC channels are nonselective cation channels, the TRPC channels identified in this study may also contribute to the resting membrane potential in RPE cells. Human RPE cells have a resting membrane potential of approximately −45 mV. ^{54 55} Given that they have high K⁺ permeability, ⁵⁴ the observed resting potential requires an additional Na⁺-permeable current that shifts the resting potential to values positive to the equilibrium potential of K⁺. TRPC channels are known to set the membrane potential and the basal Ca²⁺ entry ⁵⁶ in other cell types. They may also do so in the RPE. 

 Supported by the Deutsche Forschungsgemeinschaft DFG Grants STR 480/8-2 and STR 480/9-1.

 Submitted for publication April 5, 2007; revised July 9 and August 28, 2007; accepted October 19, 2007.

 Disclosure: S. Wimmers, None; O. Strauss, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Oligonucleotides Used to Amplify Transcripts of TRPC1–7

Gene	Accession No.	Sequence	Length (bp)	Annealing Temperature (°C)
TRPC1	NM_003304	Forward: 5′-TGCTTACCAAACTGCTGGTG-3′	243	59
		Reverse: 5′-AACTGTTTTGCCGTTTGACC-3′
TRPC3	NM_003305	Forward: 5′-GACTTTGGGATGCTGTCCAT-3′	264	59
		Reverse: 5′-ATTTGCAGGGAGGATGTACG-3′
TRPC4	NM_016179	Forward: 5′-TACTCCCTTCAATGTCATCC-3′	183	54
		Reverse: 5′-TTCACCAGGTTCCTCATAAC-3′
TRPC5	NM_012471	Forward: 5′-CAACTGTCGTGGAATGGATG-3′	244	59
		Reverse: 5′-AGTGCTTCCGCAATCAGAGT-3′
TRPC6	NM_004621	Forward: 5′-GGCTCTCATTTACTGGTTTG-3′	171	54
		Reverse: 5′-GTGCTGGTTTCATTAGGAAG-3′
TRPC7	NM_020389	Forward: 5′-GCCCACCAGATACCAGAAAA-3′	241	56
		Reverse: 5′-CACCCTCAGGTGGTCTTTGT-3′

Figure 1.

 

Inhibition of plasma membrane ATPases by 2 mM Na₃VO₄ led to increased [Ca²⁺]_i in RPE cells. (a) [Ca²⁺]_i changes after application of 2 mM Na₃VO₄ calculated from Fura-2 measurements. (b) Mean [Ca²⁺]_i before (control) and maximal [Ca²⁺]_i during application of Na₃VO₄ (n = 4).

Figure 2.

 

Basal [Ca²⁺]_i is independent from the activity of voltage-operated Ca²⁺ channels and ionotropic purinergic receptors. (a) [Ca²⁺]_i changes during application of nifedipine (10 μM) or PPADS (50 μM) calculated from Fura-2 measurements. (b) Mean [Ca²⁺]_i before (black bars) and during (gray bars) application of the drugs (5 μM nifedipine, n = 9; 50 μM PPADS, n = 7) did not differ from each other.

Figure 3.

 

Basal [Ca²⁺]_i is significantly reduced by the application of different nonspecific ion channel blocking ions. (a) [Ca²⁺]_i changes in the presence of La³⁺ (100 μM), Gd³⁺ (100 μM), or Ni²⁺ (2 mM) calculated from Fura-2 measurements. (b) Mean [Ca²⁺]_i before (black bars) and during (gray bars) application of La³⁺ (n = 10), Gd³⁺ (n = 5), and Ni²⁺ (n = 6). All three ions reduced the [Ca²⁺]_i significantly. Data were normalized to the mean values before application of the respective drug. *P < 0.05. ***P < 0.001.

Figure 4.

 

Concentration-response curve for the influence of La³⁺ on basal [Ca²⁺]_i. (a) Stepwise reduction of [Ca²⁺]_i after application of increasing concentrations of La³⁺ calculated from Fura-2 measurements. (b) Influence of increasing concentrations La³⁺ on [Ca²⁺]_i. Compared are the steady state values with the different La³⁺ concentrations. Mean ± SEM [Ca²⁺]_i normalized to the concentration before La³⁺ application (n = 2–15). Note that the x-axis is interrupted between 15 and 45 μM La³⁺.

Figure 5.

 

Influence of 2-APB and SKF 96365 on basal [Ca²⁺]_i in RPE cells. (a) [Ca²⁺]_i changes after application of 2-APB (75 μM) or SKF 96365 (50 μM) calculated from Fura-2 measurements. (b) [Ca²⁺]_i before (black bars) and during (gray bars) application of the TRP channel blockers 2-APB (n = 8) and SKF 96365 (n = 8). Both molecules reduced [Ca²⁺]_i significantly. *P < 0.05. ***P < 0.001.

Figure 6.

 

Influence of 2-APB and SKF96365 on basal [Ca²⁺]_i in primary cultures of RPE cells. (a) [Ca²⁺]_i changes after the addition of SKF96365 (50 μM) calculated from Fura-2 measurements. (b) Mean [Ca²⁺]_i before (black bars) and during (gray bars) application of SKF96365 (n = 5) and 2-APB (n = 5). Both molecules reduced [Ca²⁺]_i significantly in all five independent experiments. *P < 0.05.

Figure 7.

 

Expression profile of TRPC channels in ARPE-19 cells compared with the expression pattern in freshly isolated RPE cells from human donor eyes.

The authors thank Stefanie Schlichting for excellent technical assistance.