Zinc modulation of AMPA receptors may be relevant to splice variants in carp retina

Neuroscience Letters 259 (1999) 177–180

Zinc modulation of AMPA receptors may be relevant to splice variants in carp retina Ying Shen1, Xiong-Li Yang* Shanghai Institute of Physiology and Key Laboratory of Neurobiology, Chinese Academy of Sciences, 320 Yue-Yang Road, Shanghai 200031, People’s Republic of China Received 28 October 1998; received in revised form 17 November 1998; accepted 18 November 1998

Abstract With the use of the whole-cell patch clamp technique, we examined effects of zinc on AMPA receptors of isolated carp retinal horizontal cells, predominantly consisting of flop splice variants. We found that zinc ranging from 30 mM to 1 mM failed to modulate glutamate-induced currents of these cells, which is clearly distinct from the results previously obtained in superior colliculus neurons and Xenopus ooctyes. Furthermore, glutamate responses remained unchanged when zinc was co-applied with PEPA, a flop variant-preferential AMPA receptor potentiator. With the co-application of cyclothiazide, a flip variant-preferential AMPA receptor potentiator, however, a dual effect could be observed: zinc potentiated glutamate responses at low concentrations, but inhibited them at higher concentrations. These results suggest that the action of zinc on AMPA receptors may be splice variant-relevant.  1999 Elsevier Science Ireland Ltd. All rights reserved

Keywords: Zinc; AMPA receptor; Cyclothiazide; PEPA; Glutamate; Horizontal cell; Retina

Recent evidence suggests that zinc presents in synaptic vesicles of a subset of glutamatergic boutons in certain brain regions and is believed to function as an endogenous modulator of ligand- and voltage-gated ion channels when coreleased with glutamate into the synapse [14]. It was shown that zinc had different effects on NMDA and non-NMDA receptors expressed in Xenopus oocytes [11]. Zinc generally inhibits NMDA receptor-mediated currents. In contrast, zinc has a dual effect on non-NMDA receptor-mediated currents: it potentiates both AMPA and kainate responses at low concentrations, but inhibits them at higher concentrations [1,11]. It was further shown that the action of zinc on NMDA receptors may correlate with receptor structures or splice variants [6]. With the use of the Xenopus oocyte expression system, it was previously reported that the effects of zinc on two similar homomeric glutamate receptors, GluR1 and GluR3, differed in that GluR3 but not * Corresponding author. Tel.: +86 21 64335120; fax: +86 21 64332445; e-mail: [email protected] 1 Graduate student from the Department of Biological Science and Technology, Zhejiang University, Hangzhou 310027, China.

0304-3940/99/$ - see front matter PII S0304- 3940(98) 00938- 0

GluR1 could be potentiated by low concentration of zinc [2,3]. In the retina zinc has been shown to abundantly localized near the terminals of photoreceptors [15], providing a possibility that zinc may be released from the terminals and thus modulate the activities of retinal second-order neurons: horizontal and bipolar cells. We previously showed that glutamate receptors on carp horizontal cells may be AMPA-preferring ones [10], which predominantly consist of flop splice variants [13]. In the present work, we further examined the action of zinc on AMPA receptor-mediated responses of these cells. Our findings suggest that the zinc effect may be relevant to splice variants composing the AMPA receptors. Horizontal cells were acutely dissociated from adult crucian carp (Carassius auratus) retinas by enzymatic and mechanical treatments as described previously [10,13]. Briefly, carp (14–20 cm in length) were dark-adapted for a minimum of 20 min. After fish had been anaesthetized, eyes were enucleated and then rinsed with 70% ethanol and sterile physiological saline at 4°C. Chopped retina pieces were incubated for 30 min at room temperature in 4 ml

 1999 Elsevier Science Ireland Ltd. All rights reserved

178

Y. Shen, X.-L. Yang / Neuroscience Letters 259 (1999) 177–180

Hank’s solution with 1 mg/ml papain (15 units/mg; Worthington Biochemical, Freehold, NJ) and 4 mg Lcystein. Hank’s solution contained (in mM): 137 NaCl, 3 KCl, 2 CaCl2, 1 MgSO4, 1 NaPyruvate, 1 NaH2PO4, 0.5 NaHCO3, 20 HEPES, 16 Glucose, pH 7.4 adjusted with NaOH. After rinsed with clean Hank’s solution, the papain-treated retina pieces were kept in normal Hank’s solution for up to 8 hours at 4°C. Cells were freshly dissociated from the stored retina pieces by gentle trituration with fire polished Pasteur pipettes, and the cell suspension was placed onto a plastic dish mounted on the stage of a phase contrast microscope (CK-2; Olympus Optical, Tokyo, Japan). Horizontal cells were easily identified by their typical morphology and by the characteristic inward rectifying potassium current [10]. In this work, we focused on H1 cells, which are cone-dominated horizontal cells and can be easily identified by their small flat stellate cell body and short thick dendrites [10]. Whole-cell membrane currents were achieved with the use of the conventional patch clamp technique [4] and recorded with pipettes of 5–7 MQ resistance when filled with a solution containing (in mM): 60 KCl, 60 KF, 1 CaCl2, 1 MgCl2, 10 K3-EGTA, 10 HEPES, pH 7.3 adjusted with HCl. Cells were voltage-clamped at −60 mV and whole-cell recordings were made with an Axopatch 200A amplifier (Axon Instruments, Foster City, CA), interfacing to an IBM 486 compatible personal computer through an A/D converter (TL-1; Axon Instrument). The analog signals of cell response were low-pass filtered at 1.0 kHz, digitized using the pClamp 5.7.2 (Axon Instrument), and stored in PC for further off-line analysis. A rapid solution changer (RSC-100, Biologica Science Instruments, Claix, France) was used to apply drug solutions to the cells. All the drug solutions were made with Ringer’s solution containing (in mM): 160 NaCl, 5 KCl, 3 CaCl2, 1 MgCl2, 10 HEPES, 16 glucose, pH 7.4 adjusted with NaOH. Exchange time between two adjacent injectors was set at 10 ms. The injectors (i.d. 300 mM) were positioned ~200 mm from the recorded cells. Each injector delivered drug solution at a flow rate of ~0.25 ml/min. The procedure of drug application was controlled by the analog output from the pClamp program. Glutamate and cyclothiazide were purchased from Research Biochemicals (Natick, MA). Aniracetam was from Tocris Cookson (Bristol, UK). 4-[2(Phenylsulfonylamino)ethylthio]-2,6-difuloro-phenoxyacetamide (PEPA) was a gift from Dr. M. Sekiguchi (Tokyo, Japan) and all other chemicals were from Sigma Chemical Company (St. Louis, MO). Glutamate-evoked currents from the horizontal cells desensitized rapidly [10]. In the present work the amplitude of the equilibrium responses was the mean value of those measured for the last 100 sampling points prior to the termination of agonist application. The time constant tdesen of glutamate receptor-mediated current decay was fitted with a single exponential function: A × exp[−(t − k)/t] + C, where A is the amplitude, k is the start fitting time, and C is an offset. Data fitting and statistics, including paired student’s

t-test, were performed with Clampfit 6.02 and Sigmaplot 1.02, respectively. Potentiation or suppression of the responses by modulators was represented as folds of control values obtained in the absence of the modulators. Unless stated otherwise, data were presented as the mean ± SD in the text and mean ± SE in illustrations. In this work the responses of H1 cells to 3 mM glutamate, little higher than the EC50 (1.08 mM) of these responses [10], were chosen for testing the action of zinc. External application of 3 mM glutamate to H1 cells induced an inward current (Fig. 1A), which rose rapidly to a peak (421.10 ± 112.22 pA; 10~90% rise time, 3.69 ± 0.54 ms; n = 63) and then decayed to a much lower steady level (55.65 ± 16.72 pA) with tdesen of 6.11 ± 1.05 ms. Zinc was then introduced into the perfusate 5 s before and maintained during glutamate application. It was unexpectedly found that the glutamate response remained intact with the application of zinc of either low (30 mM) (Fig. 1B) or high (3 mM) concentration (Fig. 1C). Paired student’s tests for all four parameters of the glutamate responses (10~90% rise time, tdesen, peak and equilibrium current) did not show any statistically significant difference after zinc application (P . 0.05, n = 63). These findings were clearly distinct from the results reported previously in AMPA receptors of superior colliculus neurons and expressed in Xenopus oocytes [1,11]. Although native receptors generally have heteromeric composition, glutamate responses of carp H1 cells consist of a major component mediated by the AMPA receptor flop splice variant, with a minor contribution from the flip splice variant [13]. This subunit composition raised a possibility that zinc modulation might be relevant to splice variants. To explore this possibility, PEPA, a flop variant-preferential AMPA receptor potentiator [12], and cyclothiazide, a flip variant-preferential AMPA receptor potentiator [8], were used to ‘amplify’ the corresponding components of glutamate responses mediated by the flop and flip variants respectively, and then the effects of zinc on these ‘amplified’ components were examined. In these experiments, either PEPA or cyclothiazide was introduced into the perfusate 5 s before and maintained during glutamate application. Fig. 1D shows the response of another H1 cell to 3 mM glutamate, which was greatly potentiated following 10 mM PEPA application and the desensitization was completely blocked (Fig. 1E). The mean equlibrium current obtained after PEPA application was 546.19 ± 183.53 pA (n = 30). When 30 mM zinc was added with 10 mM PEPA (Fig. 1F), the glutamate response was almost not changed, as compared to control (Fig. 1E) (P . 0.05, n = 30). When zinc concentration was increased to 300 mM, similar results were obtained (data not shown). This result strengthened the possibility that zinc failed to modulate the flop variantmediated responses. The effects of zinc on the cyclothiazide-‘amplified’ component, as shown in Fig. 2, were clearly different. Like reported previously [13], both peak and equilibrium currents of the cells to 3 mM glutamate

Y. Shen, X.-L. Yang / Neuroscience Letters 259 (1999) 177–180

179

was quite similar to what reported in AMPA receptors of superior colliculus neurons and expressed in Xenopus oocytes [1,11]. In contrast, the equilibrium current and tdesen of these responses were hardly changed by zinc (P . 0.05, n = 38). In hippocampus recent evidence indicates that the CA3 area, receiving synaptic inputs from mossy fibers, synthesizes only the flip version of AMPA receptors [9]. For comparison, we examined the effects of zinc on isolated hippocampal neurons which were pharmacologically demonstrated to be flip variant-dominated. In these neurons, glutamate induced a desensitizing inward current (Fig. 3A). The current was only slightly increased by 10 mM PEPA (Fig. 3B), but substantially potentiated by 10 mM cyclothiazide (Fig. 3C), suggesting that the receptors were flip variant-dominant. For these cells 30 mM zinc did cause a big potentiation of the glutamate response (Fig. 3D), which was consistent with the observation that the excitatory transmission at the CA3 neuron synapses could be modulated by zinc [5]. This result was also similar to those obtained in superior colliculus neurons [1]. Taken together, all these suggest that zinc modulation of the AMPA receptor may be relevant to splice variants and zinc may exert differential effects on glutamatergic transmission mediated by the receptors consisting of different splice variants in distinct regions of the brain.

Fig. 1. Zinc failed to modulate PEPA-potentiated glutamate responses of H1 cells. (A) Response of an isolated H1 cell to 3 mM glutamate (Glu). (B,C) Glutamate responses of the cell were not changed by zinc (Zn) of 30 mM and 1 mM. (D) Response of another H1 cell to 3 mM glutamate. (E) 10 mM PEPA greatly potentiated the response. (F) In the presence of 10 mM PEPA, the addition of 30 mM zinc hardly changed the response. Note different scales for A, B, C and D, E, F.

were potentiated by cyclothiazide to different extents (peak current 911.65 ± 23.64 pA, equilibrium current 81.79 ± 22.07 pA, n = 38). In the presence of 10 mM cyclothiazide, the addition of 30 mM zinc further potentiated the peak current (n = 15) (Fig. 2A). When zinc concentration was increased to 100 mM, the response was still potentiated, though with less extent (n = 10) (Fig. 2B). It was noteworthy that the response was suppressed with the addition of 300 mM zinc (n = 13) (Fig. 2C). Peak currents recorded after zinc application of the three concentrations were 1.35 ± 0.17, 1.11 ± 0.11 and 0.68 ± 0.17 folds of control, respectively (Fig. 2D) and the differences were all statistically significant (P , 0.001, P , 0.05, P , 0.001, respectively). In other words, zinc had a dual effect on the cyclothiazide-potentiated glutamate responses: potentiating them at the low concentration (30 mM), but suppressing them at the higher concentration (300 mM). This dual effect

Fig. 2. Dual effect of zinc on cyclothiazide-potentiated glutamate responses of H1 cells. (A–C) Responses of three isolated H1 cells to 3 mM glutamate (Glu) recorded in normal Ringer’s (control), in the presence of 10 mM cyclothiazide (CTZ), and under co-application of cyclothiazide and zinc (CTZ + Zn). Zinc concentration was 30 mM (A), 100 mM (B) and 300 mM (C), respectively. Control response peaks are marked by arrows. (D) Modulatory effects of zinc on cyclothiazide-potentiated glutamate responses, represented as folds of control values obtained in the absence of the modulators, are plotted against zinc concentrations.

180

Y. Shen, X.-L. Yang / Neuroscience Letters 259 (1999) 177–180

[2]

[3]

[4]

[5] [6]

Fig. 3. Zinc modulation at a flip variant-dominant hippocampal neuron. (A) Response of an isolated hippocampal cell to 1 mM glutamate (Glu). (B) Glutamate response of the cell was slightly potentiated by 10 mM PEPA. (C) Glutamate response was greatly potentiated by 10 mM cyclothiazide (CTZ). (D) 30 mM zinc (Zn) substantially potentiated the glutamate response.

Compelling evidence suggests that the flip/flop region of alternative splicing of glutamate receptors located at the end of L3 loop is extracellular [7]. In NMDA receptors, zinc was found to only potentiate the splice variants lacking a 63-bp insertion near 5’ end, indicating that the binding of zinc on these subunits is relevant to extracellular structure [6]. As the AMPA receptor shares the same topology with the NMDA receptor, it is reasonable to speculate that this splice variant-relevant modulation of zinc may be due to the difference of extracellular spatial conformation between the flip and flop variants. This work was supported by grants from the State Commission of Science and Technology of China (Climbing Project), the National Foundation of Natural Science of China (Nos. 39670253 and 39770256) and the National Research Center of Life Science of China (Shanghai). We are grateful to Dr. M. Sekiguchi for kindly giving us PEPA as a gift.

[7] [8]

[9]

[10]

[11]

[12]

[13]

[14]

[15] [1] Bresink, I., Ebert, B., Parsons, C.G. and Mutschler, E., Zinc

changes AMPA receptor properties: Results of binding studies and patch clamp recordings, Neuropharmacology, 35 (1996) 503–509. Dreixler, J.C. and Leonard, J.P., Subunit-specific enhancement of glutamate receptor responses by zinc, Mol. Brain Res., 22 (1994) 144–150. Dreixler, J.C. and Leonard, J.P., Effects of external calcium on zinc modulation of AMPA receptors, Brain Res., 752 (1997) 170–174. Hamill, O.P., Marty, A., Neher, E., Sakmann, B. and Sigworth, F.J., Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches, Pflu¨gers Arch., 391 (1981) 85–100. Hesse, G.W., Chronic zinc deficiency alters neuronal function of hippocampal mossy fibers, Science, 205 (1979) 1005–1007. Hollmann, M., Boulter, J., Maron, C., Beasley, L., Sullivan, J., Pecht, G. and Heinemann, S., Zinc potentiates agonist-induced currents at certain splice variants of the NMDA receptor, Neuron, 10 (1993) 943–954. Hollmann, M. and Heinemann, S., Cloned glutamate receptors, Annu, Rev. Neurosci., 17 (1994) 31–108. Johansen, T.H., Chaudhary, A. and Verdoorn, T.A., Interactions among GYKI-52446, cyclothiazide, and aniracetam at recombinant AMPA and kainate receptors, Mol. Pharmacol., 48 (1995) 946–955. Lomeli, H., Mosbacher, J., Melcher, T., Ho¨ger, T., Geiger, J.R.P., Kuner, T., Monyer, H., Higuchi, M., Bach, A. and Seeburg, P.H., Control of kinetics properties of AMPA receptor channels by nuclear RNA editing, Science, 266 (1994) 1709– 1713. Lu, T., Shen, Y. and Yang, X.L., Desensitization of AMPA receptors on horizontal cells isolated from crucian carp retina, Neurosci. Res., 31 (1998) 123–135. Rassendren, F.-A., Lory, P., Pin, J.-P. and Nargeot, J., Zinc has opposite effects on NMDA and non-NMDA receptors expressed in Xenopus oocytes, Neuron, 4 (1990) 733–740. Sekiguchi, M., Fleck, M.W., Mayer, M.L., Takeo, J., Chiba, Y., Yamashita, S. and Wada, K., A novel allosteric potentiator of AMPA receptors: 4-[2-(phenylsulfonylamino)ethylthio]-2,6difluoro-phenoxyacetamide, J. Neurosci., 17 (1997) 5760– 5771. Shen, Y., Lu, T. and Yang, X.L., Modulation of desensitization at glutamate receptors in isolated crucian carp horizontal cells by concanavalin A, cyclothiazide, aniracetam and PEPA, Neuroscience, (1999) in press. Smart, T.G., Xie, X. and Krishek, B.J., Modulation of inhibitory and excitatory amino acid receptor ion channels by zinc, Prog. Neurobiol., 42 (1994) 393–441. Wu, S.M., Qiao, X.X., Noebeis, J.L. and Yang, X.L., Localization and modulatory actions of zinc in vertebrate retina, Vision Res., 33 (1993) 2611–2616.