Activation of the sigma receptor 1 suppresses NMDA responses in rat retinal ganglion cells

Neuroscience 177 (2011) 12–22

ACTIVATION OF THE SIGMA RECEPTOR 1 SUPPRESSES NMDA RESPONSES IN RAT RETINAL GANGLION CELLS The sigma receptor (␴R), once thought as a subtype of the opioid receptor (Martin et al., 1976), is now considered as a distinct pharmacological entity and represents unique nonopiate, nonphencyclidine binding sites in mammalian nervous systems (see Hayashi and Su, 2005 for review). By radioligand assay, it is suggested that there are at least three types of ␴Rs, denoted as ␴R1, ␴R2, ␴R3 (Quirion et al., 1992; Myers et al., 1994). Among them only ␴R1 has been cloned, which is a 29-kDa single polypeptide composed of 223 amino acids (Hanner et al., 1996; Kekuda et al., 1996) and has two transmembrane segments with the NH2 and COOH termini on the cytoplasmic side of the membrane (Aydar et al., 2002). ␴R1 is widely expressed in the CNS and may be involved in several physiological and pathological processes, such as neuronal firing, neurotransmitter release, learning and memory, neuroprotection, antipsychosis and drug abuse etc (see Maurice and Su, 2009 for review). These ␴R1-mediated effects may be largely a consequence of ␴R-induced modulation of kinds of voltage- (Aydar et al., 2002; Cheng et al., 2008; Tchedre et al., 2008) and ligand-gated ion channels (Hayashi et al., 1995; Yamamoto et al., 1995; Mtchedlishvili and Kapur, 2003; Martina et al., 2007). ␴R1 mRNA and protein are expressed in both rat and mouse neural retinas (Ola et al., 2001; Liu et al., 2010). In rat retina, using double-labeling immunohistochemistry, ␴R1 is shown to be mainly expressed in ganglion cells (GCs), GABAergic amacrine cells and a population of glycinergic amacrine cells (Liu et al., 2010). There are also several lines of evidence demonstrating that ␴R1 ligands may protect retinal neurons from damage in both in vitro and in vivo models (Martin et al., 2004; Dun et al., 2007; Smith et al., 2008; Tchedre and Yorio, 2008). However, little is known about possible involvement of ␴R1 in retinal processing and mechanisms underlying ␴R1-mediated neuroprotection. GCs, output neurons of the retina, convey visual information from bipolar cells and amacrine cells to the visual cortex via ON and OFF parallel pathways (Masland, 2001), and progressive death of these cells that occurs in ophthalmic neurodegenerative diseases may lead to vision loss (Osborne et al., 1999). The N-methyl-D-aspartate (NMDA) receptor, an ionotropic glutamate receptor subtype, is abundantly expressed on GCs. This receptor, along with the AMPA receptor, is important for signal transfer from bipolar cells to GCs (Yang, 2004; Shen et al., 2006). In addition, the NMDA receptor is known to be primarily involved in glutamate-induced excitotoxicity. That is, excessive stimulation of NMDA receptors causes intracellular Ca2⫹ concentration ([Ca2⫹]i) overload, thus result-

X.-J. ZHANG, L.-L. LIU, S.-X. JIANG, Y.-M. ZHONG* AND X.-L. YANG* Institute of Neurobiology, Institutes of Brain Science and State Key Laboratory of Medical Neurobiology, Fudan University, 138 Yixueyuan Road, Shanghai 200032, PR China

Abstract—The sigma receptor 1 (␴R1) has been shown to modulate the activity of several voltage- and ligand-gated channels. Using patch-clamp techniques in rat retinal slice preparations, we demonstrated that activation of ␴R1 by SKF10047 (SKF) or PRE-084 suppressed N-methyl-D-aspartate (NMDA) receptor-mediated current responses from both ON and OFF type ganglion cells (GCs), dose-dependently, and the effect could be blocked by the ␴R1 antagonist BD1047 or the ␴R antagonist haloperidol. The suppression by SKF of NMDA currents was abolished with pre-incubation of the G protein inhibitor GDP-␤-S or the Gi/o activator mastoparan. We further explored the intracellular signaling pathway responsible for the SKF-induced suppression of NMDA responses. Application of either cAMP/the PKA inhibitor RpcAMP or cGMP/the PKG inhibitor KT5823 did not change the SKF-induced effect, suggesting the involvement of neither cAMP/PKA nor cGMP/PKG pathway. In contrast, suppression of NMDA responses by SKF was abolished by internal infusion of the phosphatidylinostiol-specific phospholipase C (PLC) inhibitor U73122, but not by the phosphatidylcholinePLC inhibitor D609. SKF-induced suppression of NMDA responses was dependent on intracellular Ca2ⴙ concentration ([Ca2ⴙ]i), as evidenced by the fact that the effect was abolished when [Ca2ⴙ]i was buffered with 10 mM BAPTA. The SKF effect was blocked by xestospongin-C/heparin, IP3 receptor antagonists, but unchanged by ryanodine/caffeine, ryanodine receptor modulators. Furthermore, application of protein kinase C inhibitors Bis IV and Gö6976 eliminated the SKF effect. These results suggest that the suppression of NMDA responses of rat retinal GCs caused by the activation of ␴R1 may be mediated by a distinct [Ca2ⴙ]i-dependent PLC-PKC pathway. This effect of SKF could help ameliorate malfunction of GCs caused by excessive stimulation of NMDA receptors under pathological conditions. © 2011 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: ␴R1, NMDA receptor, neurotransmission, retina, ganglion cells, G protein.

*Corresponding author. Tel: ⫹8621-6422-1975 or ⫹8621-5423-7736; fax: ⫹8621-5423-7647. E-mail address: [email protected] (X.-L. Yang) or ymzhong@ fudan.edu.cn (Y.-M. Zhong). Abbreviations: [Ca2⫹]i, intracellular Ca2⫹ concentration; DAG, diacylglycerol; DMSO, dimethyl sulfoxide; GC, ganglion cell; IP3, inositol1,4,5-triphosphate; IPL, inner plexiform layer; mGluR, metabotropic glutamate receptor; NMDA, N-methyl-D-aspartate; PI, phosphatidylinositol; PKC, protein kinase C; PLC, phospholipase C; SKF, SKF10047; Xe-C, xestospongin-C; ␴R, sigma receptor.

0306-4522/11 $ - see front matter © 2011 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2010.12.064

12

X.-J. Zhang et al. / Neuroscience 177 (2011) 12–22

ing in degeneration of central neurons (Sucher et al., 1997; Smith, 2002; Hynd et al., 2004; Shen et al., 2006). In the present work, we show by patch-clamp techniques that ␴R1 agonists SKF10047 (SKF) and PRE-084 suppress NMDA receptor-mediated current responses (NMDA currents) in rat GCs, an effect that is associated with G-proteins. We further demonstrate that a change in protein kinase C (PKC) activity due to altered calcium release from inositol-1,4,5-triphosphate (IP3)-sensitive intracellular calcium stores induced through the phosphatidylinositol-phospholipase C (PI-PLC) pathway may be responsible for the suppression of NMDA responses.

EXPERIMENTAL PROCEDURES Retinal slice preparation Retinas were prepared from Sprague–Dawley rats ranging in age from P14 to P20, with the day of birth denoted as P0. All efforts were made to minimize the number of animals used and their pain and discomfort in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the guidelines of Fudan University on the ethical use of animals. Retinal slices were prepared following the procedures described previously (Chen and Yang, 2007), with minor modifications. Briefly, following deep anesthesia with 25 mg/ml urethane, the eyes were enucleated, and the retinas were removed. The isolated retinas were cut into 200-␮m-thick slices in Ringer’s using a manual cutter (ST-20, Narishige, Tokyo, Japan). The slices were transferred into a recording chamber with the cut side up and held mechanically in place by a grid of parallel nylon strings glued onto a U-shape frame of platinum wire. They were then viewed through a fixedstage upright microscope (BX51WI, Olympus, Tokyo, Japan) equipped with a 60⫻ water-immersion ceramic objective and DIC optics. Unless described otherwise, retinal slices were perfused continuously with oxygenated and carbogen-bubbled Ringer’s, which contained (in mM) NaCl 125, KCl 2.5, CaCl2 2, MgCl2 1, NaH2PO4 1.25, NaHCO3 25, and glucose 15. While recording NMDA-induced currents, the extracellular solution was supplemented with NBQX (10 ␮M), bicuculline (10 ␮M), strychnine (10 ␮M) and TTX (0.5 ␮M) to block AMPA, GABAA, glycine receptors and voltagegated sodium channels, respectively. Meanwhile, 1 mM MgCl2 was removed and 5 ␮M glycine was added in the solution. All experiments were performed at room temperature (25 °C).

Whole-cell patch clamp recording Whole-cell membrane currents of GCs were recorded with pipettes of 8 –10 M⍀ resistance in voltage-clamp modes filled with a solution containing (in mM) CsCH3SO3 120, TEA-Cl 10, CaCl2 0.1, MgCl2 4, EGTA 1, HEPES 10, ATP 3, GTP 0.5 and creatine 12, pH 7.2 adjusted with CsOH. Lucifer Yellow (0.1%) was dialyzed into neurons after membrane rupture by including it in the pipette. Pipettes were mounted on a motor-driven micromanipulator (MP-285, Sutter, Novato, CA, USA), and connected to an EPC10 patch clamp amplifier (HEKA, Lambrecht, Germany). Lucifer Yellow-filled GCs were visualized using a mercury light source and a FITC filter set, and the images were taken by a cool CCD camera (Photometrics, CoolSNAP ES, Tucson, AZ, USA) or by Leica SP2 confocal laser scanning microscope (Mannheim, Germany). Data were acquired at a sampling rate of 5 kHz, filtered at 2 kHz and then stored for further analysis. Drug-containing extracellular Ringer’s were either locally applied through a puff pipette (tip diameter ⬃2 ␮m), using a pressure micro-injector (PMI-100, DAGAN, Minneapolis, MN, USA), which applied a pressure of 35 kPa (5 p.s.i.) to the top of the pipette, or administrated

13

in bath medium through another inlet by gravity, depending on the purpose of an experiment. In addition, some drugs (cAMP, cGMP, KT5823, BAPTA, heparin, xestospongin-C and ryanodine) were dialyzed into neurons by including them in the pipette. In the experiments in which these drugs were included in the internal solution, the tip of the recording pipette was first filled with control solution, and then the shank was filled with the drug-containing solution. This therefore caused a 1⬃2 min delay for the drug entering into the cell.

Chemicals All chemicals were obtained from Sigma (Sigma-Aldrich, St. Louis, MO, USA) except D-AP5, SKF, BD1047, PB28 and ryanodine, which were purchased from Tocris Bioscience, UK. The drugs were stored in frozen stock solution and dissolved in the extracellular or intracellular solution before an experiment. All the lipophilic drugs (such as NBQX, haloperidol, ryanodine, Gö6976) were first dissolved in DMSO and then added to Ringer’s. The final concentration of DMSO was less than 0.1% with no effect on the currents of GCs obtained by whole-cell recording.

Statistical analysis All data were presented as mean⫾SEM; P values indicate paired (or unpaired) Student’s t-test, as appropriate (and noted), and values of P⬍0.05 were considered statistically significant.

RESULTS Characterization of NMDA receptor-mediated currents in GCs We first characterized NMDA currents of rat GCs. GCs were distinguished from displaced amacrine cells in the ganglion cell layer based on soma diameters and firing characteristics as previously described (Chen and Yang, 2007). To block non-NMDA receptor-mediated synaptic transmission, all experiments of NMDA-induced currents were performed in the presence of the antagonists for blocking AMPA receptor, GABAergic and glycinergic inputs (see Experimental procedures). Fig. 1A shows the currents of an ON type GC induced by local puff of 1 mM NMDA to the dendrites of the cell in normal Ringer’s containing 1 mM Mg2⫹ (left panel). When the cell was clamped at ⫺70 mV, only a small inward current (86 pA) was recorded (lower trace), but a relatively larger outward current (240 pA) was induced at ⫹40 mV (upper trace). When extracellular Mg2⫹ was removed, however, a robust inward current (707 pA) was induced at ⫺70 mV (right panel in Fig. 1A), whereas the current amplitude almost remained unchanged (237 pA) at ⫹40 mV, as compared with that in normal Ringer’s. This was expected because NMDA currents are voltage-dependently blocked by Mg2⫹ (Mayer et al., 1984). In all experiments to be described below, Mg2⫹ was removed from the extracellular solutions. The NMDA current was greatly reduced by co-application of 50 ␮M D-AP5, a specific NMDA receptor antagonist, with an average reduction of 88.82⫾2.21% (n⫽6, P⬍0.01) (Fig. 1B, C). A complete recovery of the current response to the control level was commonly obtained after washout with Ringer’s for 8 min (102.6⫾4.86% of control, n⫽6).

14

X.-J. Zhang et al. / Neuroscience 177 (2011) 12–22

Fig. 1. Characterization of NMDA receptor-mediated currents recorded in rat GCs. (A) Representative current responses elicited by local puff (300 ms) of 1 mM NMDA at intervals of 2 min to the dendrites of a GC in retinal slice preparation. The currents were recorded in the presence of 1 mM Mg2⫹ (left panel) and in the absence of Mg2⫹ (Mg2⫹-free) (right panel) when the cell was clamped at ⫺70 mV and ⫹40 mV, respectively. (B) NMDA-elicited whole-cell currents of a GC clamped at ⫺70 mV in Mg2⫹-free Ringer’s (control), in the presence of 50 ␮M D-AP5, and after wash out. The current was almost completely suppressed by D-AP5 in a reversible manner. (C) Bar chart, showing a comparison of relative amplitudes of NMDA (1 mM)-induced currents obtained in Mg2⫹-free Ringer’s (control), in the presence of 50 ␮M D-AP5 and after washout. Error bars represent SEM. ** P⬍0.01.

Activation of ␴R1 suppresses NMDA currents of both ON and OFF types of GCs We then tested effects of ␴R1 agonists on NMDA currents recorded from ON and OFF subtypes of rat GCs. First, GC subtypes were identified according to well-established morphological and physiological criteria (Pang et al., 2003; Margolis and Detwiler, 2007). Morphologically, ON (Fig. 2A) and OFF (Fig. 2D) type GCs were characterized by their dendrites terminating in proximal and distal parts of the inner plexiform layer (IPL), respectively. Further identification was provided by checking responses of the GCs to a 500 ms negative current step in the current-clamp mode (Margolis and Detwiler, 2007). Negative current injection led to rebound burst firing in the OFF GC (Fig. 2E), but not in the ON GC (Fig. 2B). Effects of SKF, a prototypical ␴R1 agonist, on NMDA currents recorded from ON and OFF GCs are shown in Fig. 2C, F, respectively. The inward current response of the ON GC to 1 mM NMDA at ⫺70 mV was significantly suppressed from 554 to 304 pA with incubation of 1 ␮M SKF for 8 min (Fig. 2C). SKF of 1 ␮M caused a comparable reduction of the NMDA current of the OFF GC from 740 to 510 pA (Fig. 2F). Since there was no significant difference in amplitude reduction extent induced by 1 ␮M SKF between ON and OFF GCs, the data presented in this section were pooled from both types. Fig. 2G shows how SKF of increasing concentrations suppressed the NMDA currents of GCs. For these experiments, data were pooled only

from those cells whose current responses to 1 mM NMDA, repetitively applied at intervals of 2 min, changed in peak amplitudes by less than 5% during a period of 8 min. For each cell, data obtained at these concentrations of SKF were all normalized to the current of that cell in normal Ringer’s. As shown in Fig. 2G, the average peaks of the NMDA responses were respectively suppressed to 67.19⫾1.73%, 50.22⫾3.89% and 28.64⫾4.14% of control with incubation of SKF of 0.1 ␮M (n⫽7), 1 ␮M (n⫽15) and 10 ␮M (n⫽7) (unpaired t-test, in all cases, P⬍0.01). SKFinduced suppression of NMDA currents was not observed when ␴R1 was blocked by BD1047, a ␴R1 antagonist, or haloperidol, a ␴R antagonist. Bath application of either 10 ␮M haloperidol or 10 ␮M BD1047 for 8 min did not change the NMDA currents of GCs (97.76⫾2.62% of control for haloperidol, n⫽6, P⬎0.05; 97.75⫾1.47% of control for BD1047, n⫽5, P⬎0.05). Fig. 2H shows the average changes in NMDA currents caused by 1 ␮M SKF, obtained in normal Ringer’s, in the presence of 10 ␮M BD1047 and of 10 ␮M haloperidol, as a function of time. In normal Ringer’s, perfusion of 1 ␮M SKF produced a progressive reduction of the NMDA currents and a substantial reduction (50.22⫾3.89% of control, n⫽15, P⬍0.01) was observed in 8 min. By contrast, with pre-incubation of BD1047 or haloperidol, 1 ␮M SKF failed to suppress the NMDA currents (100.96⫾2.64% of control, n⫽6, P⬎0.05 for BD1047; 91.94⫾3.06% of control, n⫽7, P⬎0.05 for haloperidol). Application of PRE-084, another ␴R1 agonist, caused a similar suppression of NMDA currents of GCs. Following perfusion of 10 ␮M PRE-084 for 8 min, the NMDA currents were reduced to 67.88⫾3.17% of control (n⫽7, P⬍0.01), which could be fully reversed by co-application of 10 ␮M haloperidol (95.15⫾3.77% of control, Fig. 2I). No effect of PB28, a high-affinity ␴R2 agonist (Cassano et al., 2006), on NMDA currents was observed. As shown in Fig. 2J, bath perfusion of 30 ␮M PB28 for 8 min did not change the NMDA current of the GC. Similar results were obtained in another six GC, and the relative response amplitude was 100.18⫾3.48% of control (P⬎0.05), suggesting that activation of ␴R2 did not modulate NMDA currents of GCs. Involvement of G protein in SKF-induced suppression of NMDA currents It is well documented that ␴R1 activation is associated with both pertussis toxin-sensitive Gi/o and cholera toxin-sensitive Gs proteins in a number of systems (Soriani et al., 1998; Morin-Surun et al., 1999; Soriani et al., 1999; Meyer et al., 2002). However, several studies reported that the transduction mechanism induced by ␴R1 activation is independent of G-proteins (Lupardus et al., 2000; Tchedre et al., 2008). We used patch pipettes filled with the non-hydrolyzable GDP analogue GDP-␤-S to test possible involvement of G proteins. Internal infusion of 3 mM GDP-␤-S resulted in a progressive reduction of NMDA currents in the first 8 min after membrane rupture (81.41⫾4.47% of control, n⫽8, P⬍0.01), and the currents then reached a steady level (Fig. 3A).

X.-J. Zhang et al. / Neuroscience 177 (2011) 12–22

15

Fig. 2. Activation of ␴R1 suppresses NMDA currents in both ON and OFF GCs. (A, D) Representative Lucifer Yellow-filled ON GC (A) and OFF GC (D) which possess dendrite arborizations respectively in proximal and distal parts of the IPL. INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. (B, E) Response to a 500 ms negative current step of an ON GC (B) and an OFF GC (E). Negative current injection led to rebound burst firing in the OFF cell but not in the ON GC. Current stimuli are indicated below voltage records. (C, F) SKF (1 ␮M) suppressed NMDA currents in both ON (C) and OFF (F) GCs. (G) SKF suppressed NMDA currents dose-dependently. All data for each cell were normalized to the amplitude of the cell in normal Ringer’s, and then averaged. Cell numbers are marked inside the bars. ** P⬍0.01. (H) Averaged time courses of 1 ␮M SKF induced suppression of NMDA currents in normal Ringer’s (control), in the presence of 10 ␮M BD1047 or 10 ␮M haloperidol. Note that no obvious current reduction was observed in the presence of BD1047/haloperidol. (I) 10 ␮M PRE-084 suppressed NMDA current, and the effect could be reversed by 10 ␮M haloperidol. (J) Application of 30 ␮M PB28 did not change the NMDA current.

Under this condition, addition of 1 ␮M SKF for another 8 min did not further change the currents (78.81⫾6.85% of control), and the current amplitudes remained at the same level even after SKF perfusion was stopped. The current waveforms of a representative GC recorded at times indicated by a, b, c and d in Fig. 3A are illustrated in Fig. 3B. Effects of mastoparan, a peptide activator of Gi and Go (Shpakov and Pertseva, 2006), were also tested. Fig. 3C, D show how internal dialysis of 30 ␮M mastoparan changed the NMDA currents of GCs. Mastoparan increased the current amplitudes progressively and caused an average potentiation of 62.94⫾16.28% (vs. control, n⫽7, P⬍0.01) in 6 min. With mastoparan infusion, 1 ␮M SKF no longer suppressed the currents (160.91⫾16.88% of control) and the current amplitudes were kept at a

similar level (150.81⫾10.26% of control) even after SKF was removed. These results obtained with GDP-␤-S and mastoparan suggest that the SKF effect is associated with G-proteins. cAMP-PKA and cGMP-PKG signaling pathways are not involved in SKF-induced suppression of NMDA currents Activation of ␴R1 may regulate several second messengers, thus modulating functions of a series of cellular substrates by changing their phosphorylation states (Akunne et al., 2001; Kim et al., 2008). To explore intracellular signaling pathways underlying SKF-induced suppression of NMDA currents, cAMP/PKA and cGMP/PKG systems were first examined.

16

X.-J. Zhang et al. / Neuroscience 177 (2011) 12–22

Fig. 3. Involvement of G-protein in SKF-induced suppression of NMDA currents. (A) Changes in NMDA currents caused by 1 ␮M SKF are plotted as a function of time during internal infusion of 3 mM GDP-␤-S. Note that GDP-␤-S by itself suppressed the NMDA currents, but addition of SKF did not cause further reduction. (B) Representative current responses recorded at times indicated by a, b, c and d shown in (A). (C) Averaged time course of the effect of 1 ␮M SKF on NMDA currents when the cells were internally infused with 30 ␮M mastoparan. Mastoparan potentiated the NMDA currents, and SKF did not change the NMDA currents during internal infusion of mastoparan. (D) Representative responses recorded at the time indicated by a, b, c and d shown in (C). ** P⬍0.01.

Fig. 4A shows the effect of internal infusion with 3 mM cAMP on NMDA currents of GCs. In the first 8 min after membrane rupture, the NMDA currents increased in size progressively with an average increase of 85.40⫾17.57% (vs. control, n⫽7, P⬍0.01), and during the infusion of 3 mM cAMP, application of 1 ␮M SKF caused a significant reduction of the NMDA current amplitudes, and in another

10 min they even dropped to a level lower than the control level. Compared with the amplitudes obtained after an 8 min-cAMP infusion, an average reduction of the amplitudes was 42.30⫾4.69% (n⫽7, P⬍0.01). When SKF was removed, the currents gradually returned to the level obtained before the application of SKF. Moreover, pre-incubation with 50 ␮M Rp-cAMP, a specific PKA inhibitor, did not

Fig. 4. No involvement of cAMP-PKA and cGMP-PKG signaling pathways in the suppression by SKF of NMDA currents. (A, C) Plots of average peak current amplitudes as a function of time, showing that SKF (1 ␮M) persisted to suppress the NMDA currents during internal application of either cAMP (3 mM) (A) or cGMP (4 mM) (C). (B, D) Plots of average peak current amplitudes as a function of time, showing the time courses of the effect of SKF on NMDA currents in the presence of 50 ␮M Rp-cAMP (B) or 30 ␮M KT5823 (D). Note that SKF produced similar suppression effects, just as observed in normal Ringer’s.

X.-J. Zhang et al. / Neuroscience 177 (2011) 12–22

much change the current amplitudes, but addition of 1 ␮M SKF significantly suppressed the NMDA currents (61.14⫾ 2.97% vs. the amplitudes obtained just after a 6 min RpcAMP perfusion, Fig. 4B). The effect was also reversible. The results obtained with cGMP and KT-5823, a specific inhibitor of PKG, were rather similar. As shown in Fig. 4C, NMDA currents were gradually potentiated with internal infusion of 4 mM cGMP and then reached a steady level in 8 min (144.15⫾8.06% of control, n⫽8, P⬍0.01). In the presence of 4 mM cGMP, application of 1 ␮M SKF significantly reduced the NMDA current amplitudes. Compared with the amplitudes obtained after an 8 min-cGMP infusion, an average reduction of the amplitudes was 42.31⫾3.04% (n⫽8, P⬍0.01). Again, like Rp-cAMP, internal dialysis of 30 ␮M KT-5823 hardly changed the NMDA currents, and in the presence of KT5823 1 ␮M SKF persisted to suppress the currents (58.78⫾4.02% of the amplitudes obtained in a 6-min KT5823 perfusion) (n⫽6, P⬍0.01, Fig. 4D). SKF-induced suppression of NMDA currents is mediated via PI-PLC, but not PC-PLC pathway Following activation of ␴R1, PLC/PKC dependent pathways, which are either phosphatidylcholine (PC)- or PIPLC specific (Zawalich and Zawalich, 1996), could come into play (Morin-Surun et al., 1999; Cheng et al., 2008). Fig. 5 shows the results obtained with internal dialysis of D609, a PC-PLC inhibitor, and U73122, a PI-PLC inhibitor.

Fig. 5. PI-PLC, but not PC-PLC, is involved in modulation by SKF of NMDA currents. (A) Average peak amplitudes of NMDA currents are plotted as a function of time, showing that SKF-induced suppression of NMDA currents remained in the presence of 50 ␮M D609. (B) Plot of average peak NMDA current amplitudes as a function of time, showing that effective inhibition of endogenous PI-PLC by U73122 eliminated SKF-induced suppression of NMDA currents.

17

Dialysis of 50 ␮M D609 into GCs for 6 min through recording pipettes did not change the NMDA currents (100.37⫾3.10% of control, n⫽7, P⬎0.05, Fig. 5A) and addition of 1 ␮M SKF for 8 min persisted to suppress the currents significantly (54.88⫾4.56% of control, P⬍0.01). In contrast, dialysis of 10 ␮M U73122 suppressed the NMDA currents, and the current amplitudes reached a steady level in 6 min after membrane rupture (61.59⫾5.06% of control, n⫽6, P⬍0.01). Co-application of 1 ␮M SKF did not cause a further reduction of the current amplitudes (60.45⫾4.78% of control). These results suggest the involvement of the PI-PLC pathway, but not the PC-PLC one. Effects of Ca2ⴙ on SKF-induced suppression of NMDA currents Ca2⫹ is considered to be a mediator between PI-PLC and PKC (Monnet, 2005). We further examined possible roles of extracellular Ca2⫹ influx across the plasma membrane through voltage-gated calcium channels and/or calcium release from intracellular stores may play in the SKF effect observed. Whilst there is a recent work showing that ␴R1 modulates L-type calcium channels by a direct interaction in rat GCs (Tchedre et al., 2008), as shown in Fig. 6A, NMDA currents of rat GCs were unaffected (103.3⫾5.20% of control, n⫽7, P⬎0.05) by external application of 10 ␮M nimodipine, a specific L-type calcium channel blocker, under our experimental conditions. In the presence of 10 ␮M nimodipine, application of 1 ␮M SKF reversibly suppressed the currents (54.65⫾5.45% of control, P⬍0.01). However, the SKF effect critically depended on [Ca2⫹]i, and internal infusion of Ca2⫹-free solution containing 10 mM BAPTA, which led to a dramatic reduction of [Ca2⫹]i of GCs (Bers et al., 1994), actually abolished the SKF effect: bath application of 1 ␮M SKF for 8 min no longer suppressed the NMDA currents (97.75⫾2.22% of control, n⫽7, P⬎0.05, Fig. 6B). This result implied the importance of calcium release from intracellular stores for the SKF effect. The release could be mediated by IP3- and/or ryanodine-sensitive pathways (Dropic et al., 2005). Involvement of calcium release through the ryanodine-sensitive pathway seemed to be negligible. Fig. 6C shows the result of the experiments with internal infusion of 50 ␮M ryanodine, which depletes ryanodine-sensitive calcium sites (Buck et al., 1992). Ryanodine of 50 ␮M by itself did not change the NMDA currents (108.4⫾4.10% of control, n⫽6, P⬎0.05). And bath application of 1 ␮M SKF for 8 min caused a significant reduction of the currents (58.13⫾ 4.71% of control, P⬍0.05), just as observed in normal Ringer’s. Experiments using 1 mM caffeine, a ryanodine receptor agonist, yielded a similar result (data not shown). The results obtained when IP3 receptors were blocked by xestospongin-C (Xe-C), however, were totally different. Internal infusion of 20 ␮M Xe-C resulted in a progressive suppression of NMDA currents in the first 6 min, and the current amplitudes then tended to be unchanged (57.64⫾1.99% of control, n⫽5, P⬍0.01, Fig. 6D). Bath application of 1 ␮M SKF under this condition failed to further suppress the NMDA currents (52.06⫾2.07% of

18

X.-J. Zhang et al. / Neuroscience 177 (2011) 12–22

Fig. 6. Roles of Ca2⫹ and intracellular calcium stores in SKF-induced suppression of NMDA currents. (A) Changes in NMDA currents caused by SKF are plotted as a function of time during the extracellular application of 10 ␮M nimodipine. Note that SKF (1 ␮M) still suppressed the NMDA currents in such experiments. (B) Bar chart showing that SKF (1 ␮M) did not suppress NMDA currents in Ca2⫹-free intracellular solution (0 mM Ca2⫹ with 10 mM BAPTA addition). (C) Internal dialysis of ryanodine (50 ␮M) did not block SKF-induced suppression of NMDA currents. (D) Internal infusion of 20 ␮M xestospongin-C (Xe-C) reduced the NMDA currents and in the presence of Xe-C, 1 ␮M SKF failed to suppress the currents.

control). Similar results were obtained with internal infusion of heparin (5 mg/ml), a well-characterized competitive inhibitor of IP3 binding to its receptor (data not shown). Therefore, the IP3-stimulated calcium release may be crucial for the SKF-induced suppression of NMDA currents. Role of PKC activity Changes in [Ca2⫹]i are known to modulate the activity of PKC (Ducibella and Fissore, 2008). Two inhibitors of PKC, bisindolylmaleimide IV (Bis IV) and Gö6976, were used to test whether inhibition of PKC may affect NMDA currents of rat GCs. Fig. 7A shows that internal application of 10 ␮M Bis IV by itself caused a rapid decrease of the current amplitudes in 2 min (64.20⫾3.44% of control, n⫽6, P⬍0.01). When the NMDA currents became unchanged, 1 ␮M SKF was added, which did not further suppress the current amplitudes (65.48⫾6.13% of control). Similar results were obtained using Gö6976, an inhibitor of classic PKC (cPKC) specific for ␣ and ␤1 isozymes (Fig. 7B). Compared with the effect of Bis IV, it appeared to take longer time (10 min) for Gö6976 to reduce the NMDA currents to a steady level (55.43⫾5.75% of control, n⫽8, P⬍0.01).

DISCUSSION In the present work we found that ␴R1 agonists (SKF and PRE-084) suppressed NMDA currents of both ON and OFF types of rat GCs. Such suppression was blocked by

BD1047/haloperidol, suggesting that the effect was mediated by ␴R1. Indeed, immunohistochemical analysis demonstrates the expression of ␴R1 in rat GCs (Liu et al., 2010). Even though we used 5 ␮M glycine in the bath solution to saturate the coagonist site of NMDA receptors, it would still be possible that activation of ␴R1 could reduce the release of D-serine, thus causing a reduction of NMDA currents (Kalbaugh et al., 2009). To rule out this possibility, we tested effects of D-serine on NMDA currents of rat GCs and found that application of 100 ␮M D-serine did not affect the NMDA currents of GCs in normal glycine (5 ␮M)containing Ringer’s (98.61⫾1.62% of control, n⫽8, P⬎0.05). Furthermore, 100 ␮M D-serine did not further change the NMDA currents when the receptors were suppressed by 1 ␮M SKF in the presence of 5 ␮M glycine (94.49⫾3.29% vs. the amplitudes obtained after an 8 min SKF perfusion, n⫽6, P⬎0.05). Modulation by ␴R1 activation of NMDA receptors In the hippocampus, it has been reported that activation of ␴R1 enhances NMDA currents (Monnet et al., 1990; Bergeron et al., 1996; Martina et al., 2007), which may be mediated by preventing small conductance Ca2⫹-activated K⫹ (SK) channels to open (Martina et al., 2007) and/or through G-proteins (Bergeron et al., 1996). In our experiments, no K⫹ was present in the intracellular solution, suggesting that SK channels could not be involved in the SKF-induced suppression of NMDA currents. On the other hand, there are several studies showing that ␴R1 ligands

X.-J. Zhang et al. / Neuroscience 177 (2011) 12–22

19

Excessive stimulation of NMDA receptors is known to be largely responsible for glutamate-induced excitotoxicity in retinal GCs (Sucher et al., 1997; Smith, 2002; Shen et al., 2006). ␴R1 agonists were found to antagonize the excitotoxicity induced by glutamate/NMDA, thus preventing GC apoptosis, and the effects were blocked by coapplication of ␴R1 antagonists in culture (Senda et al., 1998; Martin et al., 2004; Dun et al., 2007; Tchedre and Yorio, 2008) and in vivo (Smith et al., 2008). The effects of ␴R1 agonists may be related to regulating [Ca2⫹]i and preventing activation of pro-apoptotic genes (Tchedre and Yorio, 2008). The ␴R1-activated suppression of NMDA currents of GCs observed in this work is well consistent with the above observations, since the suppression of NMDA currents would reduce direct calcium influx via NMDA receptors. Signaling pathway involved in ␴R1-activated suppression of NMDA receptors

Fig. 7. SKF-induced suppression of NMDA currents is PKC-dependent. (A) Internal infusion of Bis IV reduced the NMDA currents, and the currents reached a steady level in 6 min. Under such condition, SKF (1 ␮M) failed to suppress the NMDA currents. (B) Similar result was obtained with internal infusion of Gö6976 (2 ␮M).

attenuate NMDA-induced current responses in cultured pyramidal neurons and other central neurons (Fletcher et al., 1995; Hayashi et al., 1995; Yamamoto et al., 1995; Shimazu et al., 2000; Kume et al., 2002). Moreover, investigations, using other experimental approaches, suggest that activation of ␴R1 may lead to an inhibition of NMDA receptors in cultured telencephalic neurons (Yamamoto et al., 1995), cortical neurons (Hayashi et al., 1995) and midbrain dopaminergic neurons (Shimazu et al., 2000). It seems likely that modulation by ␴R1 of NMDA receptors may be largely cell-specific. While NMDA receptors are mainly located extrasynaptically on GCs (Shen et al., 2006; Zhang and Diamond, 2006), these receptors are recently demonstrated to be involved not only in light/electrically evoked excitatory postsynaptic currents (EPSCs), but also in spontaneous EPSCs of OFF type GCs, along with AMPA receptors (Sagdullaev et al., 2006; Zhang and Diamond, 2009). ␴R1activated suppression of NMDA receptors of GCs suggests that ␴R1 may play a neuromodulatory role in retinal information processing. Although the endogenous ligand for ␴R1 is not clearly identified, it is generally accepted that neurosteroids may be the most plausible candidates (Baulieu, 1998; Maurice et al., 2006). In the retina, neurosteroids are synthesized by specific types of cells (retinal pigment epithelium cells, GCs and some amacrine cells), and metabolized via the blood-circulating system (Guarneri et al., 2003).

PKA and PKG signaling pathways have been shown to be involved in ␴R1-mediated effects (Akunne et al., 2001; Kim et al., 2008), but these two pathways were unlikely involved in the SKF-induced suppression of NMDA currents (Fig. 4). Instead, the PI-PLC/IP3 pathway may be responsible for the SKF-induced suppression, because the suppression was abolished once this pathway was blocked (Figs. 5 and 6). This is consistent with the previous studies showing that modulation of the PI-PLC/IP3/Ca2⫹ pathway suppresses NMDA currents in pyramidal neurons of prefrontal cortex and hippocampus (Grishin et al., 2005; Gu et al., 2005). In many cases modulation of ligand-gated receptors by calcium release from intracellular stores is mediated by PKC-dependent phosphorylation of these receptors (Ducibella and Fissore, 2008; Kim et al., 2008; Guzman-Lenis et al., 2009). It also seemed the case for the SKF effect observed because it was abolished when the activity of PKC was inhibited by Bis IV/Gö6976 (Fig. 7). Based on the results described in the present work, the putative signal pathway mediating the suppression of NMDA currents of rat GCs by SKF is summarized in the schematic diagram shown in Fig. 8. Whether ␴R1 is directly coupled to G-proteins is controversial. Many studies show that ␴R1 activation is coupled to G-protein signaling (Morin-Surun et al., 1999; Soriani et al., 1999; Ueda et al., 2001; Meyer et al., 2002). However, the putative two-membrane-spanning structure of the cloned ␴R1 protein may be too small for this receptor to be a classic seven-membrane-spanning G-protein-coupled receptor (Hanner et al., 1996). Indeed, recent studies reported that modulation of K⫹ and Ca2⫹ channels by ␴R1 was G-protein-independent (Lupardus et al., 2000; Tchedre et al., 2008). The result that SKF-induced suppression of NMDA currents was not observed when the G-protein inhibitor was applied (Fig. 3) indeed suggests the involvement of G-protein. It is noteworthy, however, that internal infusion of GDP-␤-S led to a progressive reduction of NMDA currents, thus mimicking the consequence of SKF application. In contrast, internal dialysis of mastoparan

20

X.-J. Zhang et al. / Neuroscience 177 (2011) 12–22

REFERENCES

Fig. 8. A schematic diagram showing the putative signaling pathway mediating suppression by SKF of NMDA responses of rat retinal GCs. Activation of ␴R1 reduces the activity of PI-PLC probably through modulating a Gi/o-coupled metabotropic receptor, thus altering the production of IP3 and the level of [Ca2⫹]i. The reduced activity of PI-PLC/IP3 results in a suppressed activity of NMDA receptors via PKC. PC-PLC, ryanodine receptor mediated intracellular calcium release, PKA and PKG pathways are not involved.

increased NMDA current amplitudes. These results raise an intriguing possibility that ␴R1 activation may lead to an inhibition of the G-proteins. The facts that the inhibition of the PI-PLC/IP3/PKC pathway (by U73122, Xe-C or Bis IV/Gö6976) reduced the currents, just like observed with SKF, and under these conditions the addition of SKF no longer suppressed the currents (Figs.5–7), further strengthen this possibility. It seems hard to understand that SKF would inhibit the G-proteins if the ␴R1 in retinal GCs is directly coupled to G-protein. A possible explanation for this apparent inhibition of G-proteins may be that SKF actually modulates some kind(s) of G-protein-involved receptor system(s), thus resulting in NMDA current suppression. Among others, metabotropic glutamate receptors (mGluRs) may be a possible mediator. There is evidence, showing that ␴R ligands could attenuate mGluR-mediated intracellular calcium dynamics (Klette et al., 1997). In addition, activation of group I mGluR significantly enhances NMDA currents, thus causing excitotoxicity (Skeberdis et al., 2001). Our preliminary experiment showed that, in the presence of 10 ␮M MPMQ, a selective mGluR1 antagonist, or 20 ␮M MTEP, a potent mGlu5 antagonist, the suppression of NMDA currents by 1 ␮M SKF (26.36⫾ 4.22% of control, n⫽7 for MPMQ; 15.52⫾4.61% of control, n⫽9 for MTEP) was much less than that obtained in normal Ringer’s (49.78⫾3.89% of control, unpaired ttest, P⬍0.01). Acknowledgments—This work was supported by grants from the National Program of Basic Research sponsored by the Ministry of Science and Technology of China (2006CB500805, 2007CB512205, 2011CB504602), the National Natural Science Foundation of China (30770698, 30930634, 31070967), and the 211 Project sponsored by the Ministry of Education of China.

Akunne HC, Zoski KT, Whetzel SZ, Cordon JJ, Brandon RM, Roman F, Pugsley TA (2001) Neuropharmacological profile of a selective sigma ligand, igmesine: a potential antidepressant. Neuropharmacology 41:138 –149. Aydar E, Palmer CP, Klyachko VA, Jackson MB (2002) The sigma receptor as a ligand-regulated auxiliary potassium channel subunit. Neuron 34:399 – 410. Baulieu EE (1998) Neurosteroids: a novel function of the brain. Psychoneuroendocrinology 23:963–987. Bergeron R, de Montigny C, Debonnel G (1996) Potentiation of neuronal NMDA response induced by dehydroepiandrosterone and its suppression by progesterone: effects mediated via sigma receptors. J Neurosci 16:1193–1202. Bers DM, Patton CW, Nuccitelli R (1994) A practical guide to the preparation of Ca2⫹ buffers. Methods Cell Biol 40:3–29. Buck E, Zimanyi I, Abramson JJ, Pessah IN (1992) Ryanodine stabilizes multiple conformational states of the skeletal muscle calcium release channel. J Biol Chem 267:23560 –23567. Cassano G, Gasparre G, Contino M, Niso M, Berardi F, Perrone R, Colabufo NA (2006) The sigma-2 receptor agonist PB28 inhibits calcium release from the endoplasmic reticulum of SK-N-SH neuroblastoma cells. Cell Calcium 40:23–28. Chen L, Yang XL (2007) Hyperpolarization-activated cation current is involved in modulation of the excitability of rat retinal ganglion cells by dopamine. Neuroscience 150:299 –308. Cheng ZX, Lan DM, Wu PY, Zhu YH, Dong Y, Ma L, Zheng P (2008) Neurosteroid dehydroepiandrosterone sulphate inhibits persistent sodium currents in rat medial prefrontal cortex via activation of sigma-1 receptors. Exp Neurol 210:128 –136. Dropic AJ, Brailoiu E, Cooper RL (2005) Presynaptic mechanism of action induced by 5-HT in nerve terminals: possible involvement of ryanodine and IP3 sensitive Ca2⫹ stores. Comp Biochem Physiol A Mol Integr Physiol 142:355–361. Ducibella T, Fissore R (2008) The roles of Ca2⫹, downstream protein kinases, and oscillatory signaling in regulating fertilization and the activation of development. Dev Biol 315:257–279. Dun Y, Thangaraju M, Prasad P, Ganapathy V, Smith SB (2007) Prevention of excitotoxicity in primary retinal ganglion cells by (⫹)-pentazocine, a sigma receptor-1 specific ligand. Invest Ophthalmol Vis Sci 48:4785– 4794. Fletcher EJ, Church J, Abdel-Hamid K, MacDonald JF (1995) Blockade by sigma site ligands of N-methyl-D-aspartate-evoked responses in rat and mouse cultured hippocampal pyramidal neurones. Br J Pharmacol 116:2791–2800. Grishin AA, Benquet P, Gerber U (2005) Muscarinic receptor stimulation reduces NMDA responses in CA3 hippocampal pyramidal cells via Ca2⫹-dependent activation of tyrosine phosphatase. Neuropharmacology 49:328 –337. Gu Z, Jiang Q, Fu AK, Ip NY, Yan Z (2005) Regulation of NMDA receptors by neuregulin signaling in prefrontal cortex. J Neurosci 25:4974 – 4984. Guarneri P, Cascio C, Russo D, D’Agostino S, Drago G, Galizzi G, De Leo G, Piccoli F, Guarneri M, Guarneri R (2003) Neurosteroids in the retina: neurodegenerative and neuroprotective agents in retinal degeneration. Ann N Y Acad Sci 1007:117–128. Guzman-Lenis MS, Navarro X, Casas C (2009) Selective sigma receptor agonist 2-(4-morpholinethyl)1-phenylcyclohexanecarboxylate (PRE084) promotes neuroprotection and neurite elongation through protein kinase C (PKC) signaling on motoneurons. Neuroscience 162:31–38. Hanner M, Moebius FF, Flandorfer A, Knaus HG, Striessnig J, Kempner E, Glossmann H (1996) Purification, molecular cloning, and expression of the mammalian sigma1-binding site. Proc Natl Acad Sci U S A 93:8072– 8077. Hayashi T, Kagaya A, Takebayashi M, Shimizu M, Uchitomi Y, Motohashi N, Yamawaki S (1995) Modulation by sigma ligands of

X.-J. Zhang et al. / Neuroscience 177 (2011) 12–22 intracellular free Ca2⫹ mobilization by N-methyl-D-aspartate in primary culture of rat frontal cortical neurons. J Pharmacol Exp Ther 275:207–214. Hayashi T, Su T (2005) The sigma receptor: evolution of the concept in neuropsychopharmacology. Curr Neuropharmacol 3:267–280. Hynd MR, Scott HL, Dodd PR (2004) Glutamate-mediated excitotoxicity and neurodegeneration in Alzheimer’s disease. Neurochem Int 45:583–595. Kalbaugh TL, Zhang J, Diamond JS (2009) Coagonist release modulates NMDA receptor subtype contributions at synaptic inputs to retinal ganglion cells. J Neurosci 29:1469 –1479. Kekuda R, Prasad PD, Fei YJ, Leibach FH, Ganapathy V (1996) Cloning and functional expression of the human type 1 sigma receptor (hSigmaR1). Biochem Biophys Res Commun 229:553–558. Kim HW, Roh DH, Yoon SY, Seo HS, Kwon YB, Han HJ, Kim KW, Beitz AJ, Lee JH (2008) Activation of the spinal sigma-1 receptor enhances NMDA-induced pain via PKC- and PKA-dependent phosphorylation of the NR1 subunit in mice. Br J Pharmacol 154:1125–1134. Klette KL, Lin Y, Clapp LE, DeCoster MA, Moreton JE, Tortella FC (1997) Neuroprotective sigma ligands attenuate NMDA and transACPD-induced calcium signaling in rat primary neurons. Brain Res 756:231–240. Kume T, Nishikawa H, Taguchi R, Hashino A, Katsuki H, Kaneko S, Minami M, Satoh M, Akaike A (2002) Antagonism of NMDA receptors by sigma receptor ligands attenuates chemical ischemia-induced neuronal death in vitro. Eur J Pharmacol 455:91–100. Liu LL, Wang L, Zhong YM, Yang XL (2010) Expression of sigma receptor 1 mRNA and protein in rat retina. Neuroscience 167: 1151–1159. Lupardus PJ, Wilke RA, Aydar E, Palmer CP, Chen Y, Ruoho AE, Jackson MB (2000) Membrane-delimited coupling between sigma receptors and K⫹ channels in rat neurohypophysial terminals requires neither G-protein nor ATP. J Physiol 526(Pt 3):527–539. Margolis DJ, Detwiler PB (2007) Different mechanisms generate maintained activity in ON and OFF retinal ganglion cells. J Neurosci 27:5994 – 6005. Martin PM, Ola MS, Agarwal N, Ganapathy V, Smith SB (2004) The sigma receptor ligand (⫹)-pentazocine prevents apoptotic retinal ganglion cell death induced in vitro by homocysteine and glutamate. Brain Res Mol Brain Res 123:66 –75. Martin WR, Eades CG, Thompson JA, Huppler RE, Gilbert PE (1976) The effects of morphine- and nalorphine-like drugs in the nondependent and morphine-dependent chronic spinal dog. J Pharmacol Exp Ther 197:517–532. Martina M, Turcotte ME, Halman S, Bergeron R (2007) The sigma-1 receptor modulates NMDA receptor synaptic transmission and plasticity via SK channels in rat hippocampus. J Physiol 578: 143–157. Masland RH (2001) The fundamental plan of the retina. Nat Neurosci 4:877– 886. Maurice T, Gregoire C, Espallergues J (2006) Neuro(active)steroids actions at the neuromodulatory sigma1 (sigma1) receptor: biochemical and physiological evidences, consequences in neuroprotection. Pharmacol Biochem Behav 84:581–597. Maurice T, Su TP (2009) The pharmacology of sigma-1 receptors. Pharmacol Ther 124:195–206. Mayer ML, Westbrook GL, Guthrie PB (1984) Voltage-dependent block by Mg2⫹ of NMDA responses in spinal cord neurones. Nature 309:261–263. Meyer DA, Carta M, Partridge LD, Covey DF, Valenzuela CF (2002) Neurosteroids enhance spontaneous glutamate release in hippocampal neurons. Possible role of metabotropic sigma1-like receptors. J Biol Chem 277:28725–28732. Monnet FP (2005) Sigma-1 receptor as regulator of neuronal intracellular Ca2⫹: clinical and therapeutic relevance. Biol Cell 97: 873– 883.

21

Monnet FP, Debonnel G, Junien JL, De Montigny C (1990) N-methylD-aspartate-induced neuronal activation is selectively modulated by sigma receptors. Eur J Pharmacol 179:441– 445. Morin-Surun MP, Collin T, Denavit-Saubie M, Baulieu EE, Monnet FP (1999) Intracellular sigma1 receptor modulates phospholipase C and protein kinase C activities in the brainstem. Proc Natl Acad Sci U S A 96:8196 – 8199. Mtchedlishvili Z, Kapur J (2003) A presynaptic action of the neurosteroid pregnenolone sulfate on GABAergic synaptic transmission. Mol Pharmacol 64:857– 864. Myers AM, Charifson PS, Owens CE, Kula NS, McPhail AT, Baldessarini RJ, Booth RG, Wyrick SD (1994) Conformational analysis, pharmacophore identification, and comparative molecular field analysis of ligands for the neuromodulatory sigma 3 receptor. J Med Chem 37:4109 – 4117. Ola MS, Moore P, El-Sherbeny A, Roon P, Agarwal N, Sarthy VP, Casellas P, Ganapathy V, Smith SB (2001) Expression pattern of sigma receptor 1 mRNA and protein in mammalian retina. Brain Res Mol Brain Res 95:86 –95. Osborne NN, Ugarte M, Chao M, Chidlow G, Bae JH, Wood JP, Nash MS (1999) Neuroprotection in relation to retinal ischemia and relevance to glaucoma. Surv Ophthalmol 43(Suppl 1):S102–S128. Pang JJ, Gao F, Wu SM (2003) Light-evoked excitatory and inhibitory synaptic inputs to ON and OFF alpha ganglion cells in the mouse retina. J Neurosci 23:6063– 6073. Quirion R, Bowen WD, Itzhak Y, Junien JL, Musacchio JM, Rothman RB, Su TP, Tam SW, Taylor DP (1992) A proposal for the classification of sigma binding sites. Trends Pharmacol Sci 13:85– 86. Sagdullaev BT, McCall MA, Lukasiewicz PD (2006) Presynaptic inhibition modulates spillover, creating distinct dynamic response ranges of sensory output. Neuron 50:923–935. Senda T, Mita S, Kaneda K, Kikuchi M, Akaike A (1998) Effect of SA4503, a novel sigma1 receptor agonist, against glutamate neurotoxicity in cultured rat retinal neurons. Eur J Pharmacol 342:105–111. Shen Y, Liu XL, Yang XL (2006) N-methyl-D-aspartate receptors in the retina. Mol Neurobiol 34:163–179. Shimazu S, Katsuki H, Takenaka C, Tomita M, Kume T, Kaneko S, Akaike A (2000) sigma receptor ligands attenuate N-methyl-Daspartate cytotoxicity in dopaminergic neurons of mesencephalic slice cultures. Eur J Pharmacol 388:139 –146. Shpakov AO, Pertseva MN (2006) Molecular mechanisms for the effect of mastoparan on G proteins in tissues of vertebrates and invertebrates. Bull Exp Biol Med 141:302–306. Skeberdis VA, Lan J, Opitz T, Zheng X, Bennett MV, Zukin RS (2001) mGluR1-mediated potentiation of NMDA receptors involves a rise in intracellular calcium and activation of protein kinase C. Neuropharmacology 40:856 – 865. Smith SB (2002) Diabetic retinopathy and the NMDA receptor. Drug News Perspect 15:226 –232. Smith SB, Duplantier J, Dun Y, Mysona B, Roon P, Martin PM, Ganapathy V (2008) In vivo protection against retinal neurodegeneration by sigma receptor 1 ligand (⫹)-pentazocine. Invest Ophthalmol Vis Sci 49:4154 – 4161. Soriani O, Foll FL, Roman F, Monnet FP, Vaudry H, Cazin L (1999) A-Current down-modulated by sigma receptor in frog pituitary melanotrope cells through a G protein-dependent pathway. J Pharmacol Exp Ther 289:321–328. Soriani O, Vaudry H, Mei YA, Roman F, Cazin L (1998) Sigma ligands stimulate the electrical activity of frog pituitary melanotrope cells through a G-protein-dependent inhibition of potassium conductances. J Pharmacol Exp Ther 286:163–171. Sucher NJ, Lipton SA, Dreyer EB (1997) Molecular basis of glutamate toxicity in retinal ganglion cells. Vis Res 37:3483–3493. Tchedre KT, Huang RQ, Dibas A, Krishnamoorthy RR, Dillon GH, Yorio T (2008) Sigma-1 receptor regulation of voltage-gated calcium channels involves a direct interaction. Invest Ophthalmol Vis Sci 49:4993–5002.

22

X.-J. Zhang et al. / Neuroscience 177 (2011) 12–22

Tchedre KT, Yorio T (2008) Sigma-1 receptors protect RGC-5 cells from apoptosis by regulating intracellular calcium, Bax levels, and caspase-3 activation. Invest Ophthalmol Vis Sci 49:2577–2588. Ueda H, Yoshida A, Tokuyama S, Mizuno K, Maruo J, Matsuno K, Mita S (2001) Neurosteroids stimulate G protein-coupled sigma receptors in mouse brain synaptic membrane. Neurosci Res 41:33– 40. Yamamoto H, Yamamoto T, Sagi N, Klenerova V, Goji K, Kawai N, Baba A, Takamori E, Moroji T (1995) Sigma ligands indirectly modulate the NMDA receptor-ion channel complex on intact neuronal cells via sigma 1 site. J Neurosci 15:731–736.

Yang XL (2004) Characterization of receptors for glutamate and GABA in retinal neurons. Prog Neurobiol 73:127–150. Zawalich WS, Zawalich KC (1996) Regulation of insulin secretion by phospholipase C. Am J Physiol 271:E409 –E416. Zhang J, Diamond JS (2006) Distinct perisynaptic and synaptic localization of NMDA and AMPA receptors on ganglion cells in rat retina. J Comp Neurol 498:810 – 820. Zhang J, Diamond JS (2009) Subunit- and pathway-specific localization of NMDA receptors and scaffolding proteins at ganglion cell synapses in rat retina. J Neurosci 29:4274 – 4286.

(Accepted 29 December 2010) (Available online 4 January 2011)