| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Department of Physiology, Keio University School of Medicine, Tokyo; 2Department of Comparative Pathophysiology, Graduate School of Agriculture and Life Sciences, University of Tokyo, Tokyo; 3Department of Biological Sciences, Faculty of Medicine, Kyoto University, Kyoto; and 4Medical Research Institute, Tokyo Women's Medical University, Tokyo, Japan
Submitted 25 September 2006; accepted in final form 2 April 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
-conotoxin GVIA-sensitive component (N-type) and a -Aga IVA-sensitive component (P/Q-type). Tetrodotoxin-sensitive Na+ currents and dihydropyridine-sensitive Ca2+ currents (L-type) were not observed. Immunoreactivity for the Na channel subunit (Pan Nav), the K channel subunits (the A-current subunits [Kv. 3.3 and Kv 3.4]) and the Ca channel subunits (
1A [P/Q-type], 1B [N-type] and 1C [L-type]) was detected in the membrane fraction of the mouse retina by Western blot analysis. Immunoreactivity for the Kv. 3.3, Kv 3.4, 1A [P/Q-type], and 1B [N-type] was colocalized with the GFP signals. Immunoreactivity for 1C [L-type] was not colocalized with the GFP signals. Immunoreactivity for Pan Nav did not exist on the membrane surface of the GFP-positive cells. Our findings indicate that signal propagation in cholinergic amacrine cells is mediated by a combination of two types of voltage-gated K+ currents (the A current and the delayed rectifier K+ current) and two types of voltage-gated Ca2+ currents (the P/Q-type and the N-type) in the mouse retina. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
). Direction selectivity is thought to be a product of the dendritic computation at the direction-selective ganglion cells (Sterling 2002; Wässle 2004). The dendritic computation for direction selectivity occurs presynaptically (Borg-Graham 2001; Fried et al. 2002, 2005; Taylor and Vaney 2002) and postsynaptically (Taylor and Vaney 2002). As a presynaptic mechanism, Fried et al. (2002, 2005) reported that synaptic inputs onto direction selective ganglion cells are different between the preferred direction and the null direction.
Recent studies demonstrated that direction selectivity is lost when cholinergic amacrine cells are ablated from the retinal circuit (Amthor et al. 2002; Yoshida et al. 2001). Cholinergic amacrine cells make a synaptic contact with direction-selective ganglion cells (Famiglietti 1991) and release GABA and ACh in a Ca2+-dependent manner onto the dendrites of direction selective ganglion cells (Fried et al. 2005; Zheng et al. 2004). Because cholinergic amacrine cells are a presynaptic component, there is a possibility that the release of GABA and ACh from the cholinergic amacrine cells onto the directionally selective ganglion cells itself is direction selective. In fact, Euler et al. (2002) demonstrated that intradendritic Ca2+ increase by the light stimuli in the cholinergic amacrine cells has direction selectivity. Because the intradendritic Ca2+ increase of cholinergic amacrine cells was not affected by GABAA antagonist (Euler et al. 2002), it is likely that the direction-selective intradendritic Ca2+ increase is generated by the intrinsic membrane properties of cholinergic amacrine cells. Thus the systematic analysis of membrane properties of cholinergic amacrine cells is necessary to understand the presynaptic mechanism of direction selectivity.
Because amacrine cells consist of a group of axonless neurons with different cell morphologies (Kolb and Nelson 1981, 1984; Kolb et al. 1981; MacNeil and Masland 1998) and responses to light (Dacheux and Raviola 1986; Djamgoz et al. 1990; Kolb and Nelson 1984), the characterization of the membrane properties of individual amacrine cell subtypes have an inherent difficulty. Despite the inherent difficulty in identifying individual subtypes, the membrane properties of the cholinergic amacrine cells have been reported in rabbits (Cohen 2001; Taylor and Wässle 1995; Zhou and Fain 1996), mice (Ozaita et al. 2004), and ferrets (Aboelela and Robinson 2004). For the voltage-gated Na channels, Cohen (2001) reported the presence of a tetrodotoxin-sensitive current in rabbits, whereas three previous papers (Ozaita et al. 2004; Taylor and Wässle 1995; Zhou and Fain 1996) reported the absence of a tetrodotoxin-sensitive current in both rabbits and mice. For the voltage-gated Ca channels, Cohen (2001) reported that P/Q type and N-type are the major components in rabbits. For voltage-gated K channels, Ozaita et al. (2004) reported the unique distribution of Kv3.1 and Kv3.2 in the cholinergic amacrine cells of mice. Ozaita et al. (2004) further reported that delayed rectifier K channels have a density gradient along the dendrites. These reports provided much progress toward the understanding of the signal processing in cholinergic amacrine cells. In particular, the finding of the density gradient of delayed rectifier K channels (Ozaita et al. 2004) is favorable to explain the direction selectivity of the intradendritic Ca2+ increase (Euler et al. 2002). However, the types of voltage-gated Ca channels in the cholinergic amacrine cells have not been determined in mouse retina. In addition, because of the difficulties in accessing OFF-type cholinergic amacrine cells, previous electrophysiological studies have focused on the membrane properties of ON-type cholinergic amacrine cells (Aboelela and Robinson 2004; Cohen 2001; Ozaita et al. 2004; Taylor and Wässle 1995; Zhou and Fain 1996).
Recent progress in gene manipulation techniques has enabled us to identify cholinergic amacrine cells by tagging them with green fluorescent protein (GFP) (Kaneda et al. 2004; Yoshida et al. 2001). In the present study, we systematically characterized the voltage-gated ionic channels of both the ON- and the OFF-type cholinergic amacrine cells in transgenic mouse retina using both electrophysiological and immunohistochemical techniques. We found that signal processing in cholinergic amacrine cells is mediated by a combination of K and Ca channels. Our data also suggest that P/Q- and N-type Ca channels act as a pathway for Ca2+ influx in cholinergic amacrine cells of the mouse retina. Our findings support the hypothesis that the voltage-gated Ca channels, especially P/Q-type, of cholinergic amacrine cells play an important role in the formation of the direction-selective intradendritic Ca2+ increase in cholinergic amacrine cells of the mouse retina.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Animals
This study used wild-type mice (C57BL/6N) for Western blot analysis and the IG-8 line of heterozygous transgenic mice (C57BL/6N) for the patch-clamp study and the immunohistochemical staining. The specific localization of GFP signals in the cholinergic amacrine cells of the retina was previously reported for this transgenic line (Kaneda et al. 2004; Yoshida et al. 2001).
Preparations
Details of the slice preparation (Satoh et al. 1998) and dissociated cell preparation (Kaneko et al. 1989) methods were described in previous papers. In brief, the animals were killed by neck dislocation, both eyes were enucleated and hemisected, and the retina was isolated from the sclera. For the slice preparation, the detached retina was placed on a membrane filter (pore size, 0.45 µm; Advantec Toyo, Tokyo, Japan) with the photoreceptor side up and sliced at a thickness of 150–200 µm in Ringer solution (which contained, in mM, 115 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, 1.1 NaH2PO4, 26 NaHCO3, and 10 glucose; pH 7.4 when bubbled with 95% O2-5% CO2). The slices were kept in Ringer solution until use. For the dissociated cell preparation, the detached retina was incubated for 15–20 min in an external solution (which contained, in mM, 135 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, 10 glucose, and 5 HEPES; pH adjusted to 7.4) containing 2.5 U/ml papain (Worthington Biochemical, Freehold, NJ) and its activator, L-cysteine (0.1 mg/ml) bubbled with 100% O2 at 37°C. Enzyme treatment was stopped by washing the retina with the external solution containing 0.1 mg/ml of bovine serum albumin. The retina was then triturated with a Pasteur pipette, and the cell suspension was dispensed on a concanavalin-A-coated cover glass in a plastic petri dish. After the cells had attached to the cover glass (
30 min), the dish was filled with the external solution and kept at room temperature until use. Cholinergic amacrine cells were useful for whole cell recordings 
3 h after dissociation. Recordings were carried out in Ringer solution for the slice preparation and the external solution for the dissociated cell preparation. All experiments were carried out at room temperature.
Current recordings
Whole cell patch-clamp recordings were performed using cholinergic amacrine cells identified by a GFP fluorescent signal when viewed under a fluorescent microscope (BX51WI; Olympus, Tokyo, Japan). In slice preparation, cholinergic amacrine cells died within 1 h after slicing, and it was difficult to hold the cells for long periods of time under whole cell recording. As there were no difficulties for current recordings from other types of amacrine cells in the same preparation, the fragility of cholinergic amacrine cells was thought to be due to severe damage to dendrites during slicing procedures or damage produced by the emission of GFP. Therefore to maximize the chance for stable whole cell recordings from cholinergic amacrine cells, we used the patch pipettes with relatively high resistance. Patch pipettes were pulled from borosilicate glass (Hilgenberg GmbH, Marsfeld, Germany) using a two-stage electrode puller (PP-83; Narishige, Tokyo, Japan) and they exhibited a resistance of between 15 and 18 M
when filled with a K intra-pipette solution (which contained, in mM, 120 KCl, 0.5 CaCl2, 5 HEPES, 5 EGTA, 5 ATP-2Na, and 1 GTP-3Na; pH adjusted to 7.3 with KOH). Voltage-gated Ca2+ current recordings were performed using a K+ intra-pipette solution or a Cs+ intra-pipette solution (which contained, in mM, 120 CsCl, 0.5 CaCl2, 5 HEPES, 5 EGTA, 5 ATP-2Na, and 1 GTP-3Na; pH adjusted to 7.3 with CsOH). An Ag-AgCl pellet submerged in an NaCl well and connected to a recording chamber via a 150 mM NaCl agar-bridge was used as a reference electrode. When tetraethylammonium-Cl was added to the perfusate, NaCl was replaced with equimolar tetraethylammonium-Cl. For P/Q- and N-type Ca channel blockers, -conotoxin AgaIVA and -conotoxin GVIA were dissolved with Ringer solution containing 0.1 mg/ml cytochrome c. Simultaneous application of these toxins was carried out only in the dissociated cell preparation. Current and voltage signals were recorded using a patch-clamp amplifier (Axopatch-200B; Axon Instruments, Foster City, CA), sampled at 10 kHz through a DigiData 1322A Interface using pCLAMP software (version 8.0, Axon Instruments), and were stored in a personal computer after passing the data through a low-pass Bessel filter (<5 kHz). The cells were continuously superfused with the external solution. The Y-tube system was used to apply drugs. Holding potentials were corrected for the liquid junction potentials. Current-voltage relationships were plotted against the peak current amplitude after the subtraction of leak current. Leak current produced by the ohmic conductance was estimated from the holding currents at a holding potential of –72 and –62 mV. It should be noted that electrodes with relatively high resistance can be a possible cause of potential drop when large amplitude of current was recorded.
Western blot analysis
Samples for Western blot analysis were homogenized in HEPES buffer (5 mM HEPES and 0.32 M sucrose, pH 7.4) with a 1X protease inhibitor cocktail (Roche, Molecular Biochemicals), centrifuged at 800 g for 10 min at 4°C to remove the nucleus. The supernatants were centrifuged at 10,000 g for 30 min at 4°C. The pellet was collected and solved in PBS (which contained, in mM, 137 NaCl, 2.7 KCl, 2.5 CaCl2,1 MgCl2, and 10 phosphate buffer, pH 7.4) containing 0.1% SDS, 0.5% sodium deoxycholate, 1% NP-40, and a 1X protease inhibitor cocktail (Roche, Molecular Biochemicals), centrifuged at 10,000 g for 30 min at 4°C. The supernatants were collected and run on 7% polyacrylamide gels (20 µg protein per lane) and then transferred onto a PVDF membrane (Amersham, Buckinghamshire, UK). After blocking, membranes were processed through sequential incubations with rabbit anti-Pan Nav, rabbit anti-Kv.3.3, rabbit anti-Kv.3.4, or rabbit anti-calcium channel subunits [1A (P/Q-type), 1B (N-type), or 1C (L-type)] overnight, and then with horseradish peroxidase-conjugated anti-rabbit Ig G (Vector Laboratories, Burlingame, CA) for 1 h. The rabbit polyclonal antibodies used were raised against residues 865–881 of 1A subunit (VGCC, CNA1; Accession No. P54282), residues 851–867 of 1B subunit (VGCC, CNB1; Accession No. Q02294), residues 848–865 of 1C subunit (VGCC, CNC1; Accession No. P22002) of rat brain voltage-gated calcium channel of rat protein (with additional N-terminal lysine and tyrosine), residues 1500–1518 of rat Nav1.1 (Accession No. P04774), residues 701–718 of rat Kv3.3 (Accession No. Q01956), and residues 177–195 of rat Kv3.4 (Accession No. Q63734). Immunoreactive proteins on the membranes were visualized using an ECL plus kit (Amersham). The specificity for immunoreactive bands for Pan-Nav and calcium channel 1C subunit was confirmed by the preadosorption test. Rabbit anti-Pan Nav, rabbit anti-Kv.3.3, and rabbit anti-Kv.3.4 were purchased from Alomone (Jerusalem, Israel). Rabbit anti-calcium channel subunits (1A (P/Q-type), 1B (N-type), and 1C (L-type)) were purchased from Sigma (Saint Louis, MO).
Immunohistochemical staining
The enucleated eyes were opened at the equator, and the retinae were removed and fixed with 4% paraformaldehyde (wt/vol) in 0.1 M phosphate buffer for 2 h at room temperature. After this fixation, the retinae were immersed in 30% sucrose at 4°C overnight for cryoprotection. Cryoprotected retinae were then sectioned into 10-µm–thick sections using a cryostat (MICROM HM560; Carl Zeiss, Jena, Germany). Isolated cell preparations were fixed with 4% paraformaldehyde (wt/vol) in 0.1 M phosphate buffer for 10–15 min at room temperature. The immunostaining procedures are described in detail in our previous papers (Ishii et al. 2003; Kaneda et al. 2004). Briefly, the slice preparations were allowed to react with the primary polyclonal antibodies (working dilution, 1:200) in 0.1 M phosphate buffer containing 0.3% TritonX-100 for 10–12 h at room temperature. Isolated cell preparations were reacted with a cocktail of one type of primary antibody and chicken anti-GFP (working dilution, 1:5,000; Chemicon, Norcross, GA). The sections were then allowed to react with Alexa546-conjugated goat anti-rabbit IgG (1:500; Molecular Probes, Eugene, OR) in 0.1 M phosphate buffer for 2 h at room temperature. Isolated cell preparations were reacted with a cocktail of Alexa488-conjugated goat anti-chicken IgG (1:500; Molecular Probes) and Alexa546-conjugated goat anti-rabbit IgG (1:500) in 0.1 M phosphate buffer for 2 h at room temperature.
Light microscopy
Fluorescent images were observed under a confocal microscope (LSM-510; Carl Zeiss, Jena, Germany) at an excitation wavelength of 488 nm for GFP and Alexa 488 (band-pass: 505–530 nm) or 543 nm for Alexa546 (band-pass: 560–615 nm). The fluorescent images were scanned at a slice width of 0.5–0.6 µm and digitized using a computer. Image sampling and processing were performed using the LSM5 image browser's accompanying software. In the present study, we used "multitrack mode" which protects the possible contamination of signals of Alexa 488 or GFP with signals of Alexa546 when samples were excited at a wavelength of 543 nm. Recording levels of each signal were adjusted under the "palette mode" according to the recommended procedure by Carl Zeiss. The sampled images were further processed in an off-line computer using Photoshop 6.0 (Adobe, San Jose, CA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
In the present study, we observed two types of responses to voltage steps (Fig. 1A). When voltage steps only activated a large sustained outward current, the response was classified as type 1. When a large sustained outward current accompanied a transient outward current, the response was classified as type 2. Voltage steps were applied to 94 cells, and the pooled data were classified according to the above-mentioned criteria (Fig. 1B). In slice preparations, 67% of the examined cells were classified as type 1. In contrast, 85% of the dissociated cell preparations were classified as type 2. We measured the amplitude of the sustained outward current at 40 ms from the onset of the step pulses (step to 10 mV) and averaged the amplitude (Fig. 1C). The current amplitude of the slice preparations (type 1) was larger than that of the dissociated cell preparations. We also examined the threshold to activate the sustained outward current and the transient outward current (Fig. 1D). In the present study, the threshold of the transient outward currents was defined as the potential where the transient current was clearly detected. In both the slice preparations and the dissociated cell preparations, the threshold potential of the type 1 responses was more negative than that of the type 2 responses. In addition, the threshold of the transient current (type 2 peak) was more positive than the threshold of the sustained current in the type 2 responses.
|
A previous study reported that cholinergic amacrine cells of the mouse retina exhibit delayed rectifier K+ currents but do not exhibit A currents (Ozaita et al. 2004). In the present study, however, we observed a transient component to the outward current in both the slice preparation and the dissociated cell preparation. Therefore we examined whether 4-aminopyridine, an A current blocker, inhibited the transient outward current in dissociated cell preparations. When 3 mM 4-aminopyridine was added to the perfusate, the transient component of the outward currents was inhibited (Fig. 2A). The A current obtained by computer subtraction had both the transient and the sustained components. The current-voltage relationship of the A current had a threshold at –42 mV and increased with increasing step pulses (Fig. 2C). The further addition of 30 mM tetraethylammonium, a delayed rectifier K+ current blocker, inhibited the 4-aminopyridine-insensitive sustained outward current (Fig. 2B). The current-voltage relationship of the delayed rectifier K+ current had a threshold at –32 mV and increased with increasing step pulses (Fig. 2D).
|
Voltage-gated Na+ currents were not detected in cholinergic amacrine cells of the mouse retina in previous research (Ozaita et al. 2004). In the present study, we examined whether active voltage-gated Na currents existed in our preparations. However, no tetrodotoxin-sensitive components were detected when 1 µM tetrodotoxin was applied to the cells (Fig. 3A).
|
Voltage-gated Ca2+ currents were isolated by perfusing the cells with 4-aminopyridine and tetraethylammonium (Fig. 2B, middle). The voltage-gated Ca2+ currents consisted of sustained inward currents and were reversibly inhibited by 100 µM Cd (Fig. 3B). The current-voltage relationship of the voltage-gated Ca2+ currents had a threshold at –52 mV and peaked at –12 mV (Fig. 3C). To identify the subtypes of the voltage-gated Ca2+ currents, we examined three pharmacological agents in the dissociated cell preparation. The voltage-gated Ca2+ currents were inhibited by the application of 200 nM -AgaIVA, a P/Q-type blocker (Fig. 4A). A current obtained by computer subtraction was observed as a sustained current. When 3 µM -conotoxin GVIA, an N-type blocker, was applied in the presence of 200 nM -AgaIVA, no detectable inward current was recorded (Fig. 4B). A current obtained by computer subtraction had both the transient and the sustained components. In contrast, 10 µM nimodipine, an L-type blocker, did not exert any inhibitory actions on the voltage-gated Ca2+ currents (Fig. 4C). These pharmacological actions of Ca channel blockers were consistent in the slice preparations. These observations indicate that the voltage-gated Ca2+ currents are mediated by the P/Q- and N-type. We measured the current amplitude at peak in the presence and absence of toxins and assessed the ratio of P/Q- and N-type to the total Ca2+ currents. As actions of -AgaIVA or -conotoxin GVIA were examined in different slice preparations, the total percentage of P/Q- and N-type did not become 100%. In the present study, the voltage-gated Ca2+ currents were inhibited by 0.52 ± 0.22% with -AgaIVA (2 dissociated cells and 2 slices) and 0.28 ± 0.10% with -conotoxin GVIA (2 dissociated cells and 3 slices).
|
A single band was detected for Kv.3.3, Kv.3.4, calcium channel subunits 1A (P/Q-type), and calcium channel subunits 1B (N-type) in the membrane fraction (Fig. 5). Molecular weights of these bands approximately corresponded to the predicted value. Therefore the immunoreactivity detected by antibodies for Kv3.3, Kv3.4, calcium channel subunits 1A (P/Q-type), and calcium channel subunits 1B (N-type) reflect the distribution of the channel proteins in the mouse retina. There were two bands for Pan Nav and three bands for calcium channel subunits 1C (L-type). The band with high molecular weight for Pan Nav or for calcium channel subunits 1C corresponded to the predicted molecular weights. This result reflects the fact that these channel proteins also exist in the mouse retina. One band with low molecular weight (approximately 45 kDa) for Pan Nav and two bands with low molecular weights (approximately 70 and 75 kDa) for calcium channel subunits 1C were also detected in the cytosolic fraction. Presumably, these bands with low molecular weights are contamination of cytosolic protein. When membrane fractions containing nuclear components were run on the gel, additional bands (30 and 90 kDa) were detected for Pan Nav.
|
To confirm the specificity of immunoreactivity for individual antibodies, three control experiments were carried out on the slice preparations. Sections incubated with goat serum did not show any notable fluorescent image. Sections reacted with the Alexa546-conjugated goat anti-rabbit IgG without the primary antibody showed no anatomically specific fluorescent image. For all secondary antibodies used in the present study, we confirmed the absence of any notable fluorescent image. When the primary antibody was preincubated with the antigen peptide to occlude the epitope before immunohistochemical staining, we could not observe any specific immunoreactivity. In the dissociated cell preparation, we stained cells with chicken anti-GFP only and incubated the cells with Alexa488-conjugated goat anti-chicken IgG. When Alexa488-labeled cells were examined under the confocal microscope using "multitrack mode, " we could not detect any significant fluorescent signals when Alexa488 was excited by a laser with a wavelength of 543 nm (band-pass: 560–615 nm).
Occasionally, there was autofluorescence in the outer segment of photoreceptors in the slice preparation when the wavelength of 488 nm was used for excitation. The autofluorescence became weak when the wavelength of 546 nm was used for excitation, but we still observed a faint autofluorescence. We do not think this was due to aging of the retina during preservation because we observed autofluorescence in only some of the serial sections. However, the presence of autofluorescence made it difficult to assess precisely any immunoreaction at the outer segment of the photoreceptors. In the present study, positive immunoreaction in this location was judged as autofluorescence unless it was consistently strong throughout the series of sections. The control experiments indicate that the immunoreaction for the individual voltage-gated ionic channel is specific to each voltage-gated ionic channel subunit antibody.
Based on the Western blot analysis of membrane fraction, the immunoreactivity detected by antibodies for Kv3.3, Kv3.4, calcium channel subunits 1A (P/Q-type), and calcium channel subunits 1B (N-type) was thought to reflect the distribution of these channel subunits because the high specificity of antibodies was approved. For Pan Nav and calcium channel subunits 1C (L-type), it should be noted that the distribution of immunoreactivity does not completely reflect the distribution of channel proteins as these antibodies can detect additional cytosolic proteins. Immunoreactivity detected in the nuclear region was not considered as the distribution of channel protein.
Distribution of immunoreactivity for voltage-gated Na channels
In the dissociated cell preparations, immunoreactivity for Pan Nav seemed to accumulate in the nuclear regions of the GFP-positive cells (Fig. 6, A–C). It is likely that this immunoreactivity corresponds to the different proteins which were detected as 30 and 90 kDa in the nuclear fraction by Western blot analysis. When images of the GFP signal and the Pan Nav immunoreactivity were merged, the two signals were shown to have different subcellular localizations (Fig. 6, A–C, and Table 1). On the other hand, Pan-Nav immunoreactivity outlined the cell surfaces of some GFP-negative cells in the dissociated preparations. In the slice preparations, Pan Nav immunoreactivity was observed in the inner nuclear layer and the ganglion cell layer (Figs. 7A and 8, A and B). When images of the GFP signal and Pan Nav immunoreactivity were merged, Pan Nav immunoreactivity was observed in the nuclear region of GFP-positive cells in both the inner nuclear layer and the ganglion cell layer. No significant colocalization of GFP signals and Pan Nav immunoreactivity was observed in the inner plexiform layer.
|
|
|
|
Delayed rectifier K channels and A-type K channels are reportedly formed by different K channel subunits (Kv) (Rudy et al. 1999). Kv3.1 and Kv3.2 form delayed rectifier K channels, whereas Kv3.3 and Kv3.4 form A-type K channels. As the immunoreactivity for Kv3.1 and Kv3.2 in the cholinergic amacrine cells of the mouse retina was precisely described in the previous paper (Ozaita et al. 2004), we herein focused on the immunohistochemical distribution of Kv3.3 and Kv3.4. The distribution of immunoreactivity for Kv3.1 and Kv3.2 in the transgenic mouse retina was similar to the distribution of immunoreactivity for Kv3.1 and Kv3.2 in wild-type mouse retina (Ozaita et al. 2004).
In the dissociated cell preparations, immunoreactivity for Kv3.3 appeared as small puncta and was distributed throughout the somas (Fig. 6, D–F). Significant immunoreactivity for Kv3.3 was colocalized with the majority of GFP-positive cells (Fig. 6, D–F, and Table 1) although the immunoreactivity for Kv3.3 was very weak. In the slice preparations, immunoreactivity for Kv3.3 existed in the outer plexiform layer, the inner nuclear layer, the inner plexiform layer, and the ganglion cell layer (Figs. 7B and 8, C and D). In the inner nuclear and the ganglion cell layers, immunoreactivity for Kv3.3 outlined the somas. Immunoreactivity for Kv3.3 seemed to be colocalized with the GFP signals. In the inner plexiform layer, immunoreactivity for Kv3.3 seemed to exist in the dendritic region of the cholinergic amacrine cells.
In dissociated cell preparations, immunoreactivity for Kv3.4 was distributed throughout the somas (Fig. 6, G–I). Colocalization of Kv3.4 immunoreactivity and the GFP signal was detected in 26.7% of the cells examined by visual inspection because the Kv3.4 immunoreactivity was very weak (Table 1). In the slice preparations, immunoreactivity for Kv3.4 was observed in both the outer plexiform layer and the inner plexiform layer (Figs. 7C and 8, E and F). In the inner plexiform layer, immunoreactivity for Kv3.4 was distributed throughout the layer and overlapped with the GFP signal. Colocalization of Kv3.4 immunoreactivity and the GFP signal was detected in some of the cells in the inner nuclear layer and in the ganglion cell layer.
Distribution of immunoreactivity for voltage-gated Ca channels
In the dissociated cell preparations, immunoreactivity for the 1C subunit (L-type) was found in bipolar cells but not in GFP-positive cells (Fig. 6, J–L, and Table 1). In the slice preparations, immunoreactivity for the 1C subunit was exclusively found in bipolar cells (Fig. 7D). In the inner plexiform layer, axon terminals with immunoreactivity for the 1C subunit existed in sublamina b but not in sublamina a, suggesting that L-type Ca channels exist in ON-type bipolar cells. No remarkable overlap of the GFP signal and 1C subunit immunoreactivity was observed. The selective localization of the 1C subunit in bipolar cells of the rat retina had been previously reported (Xu et al. 2002). However, the distribution of L-type Ca channels on bipolar cells might be restricted to the axon terminal as the L-type Ca2+ current has been recorded in the axon terminal in the mouse retina (de la Villa et al. 1998; Satoh et al. 1998).
In the dissociated cell preparations, immunoreactivity for the 1B subunit (N-type) existed in GFP-positive cells (Fig. 6, M–O, and Table 1). Immunoreactivity for the 1B subunit was relatively strong in the nuclear region and weak on the cell surface. In the slice preparations, immunoreactivity for the 1B subunit was observed in the outer plexiform layer, the inner nuclear layer, the inner plexiform layer and in the ganglion cell layer (Figs. 7E and 8, G and H). In the outer plexiform layer, immunoreactivity for the 1B subunit was observed as puncta. In the inner nuclear layer, strong immunoreactivity for the 1B subunit was found in cells located on the vitreous side, presumably amacrine cells. Immunoreactivity for the 1B subunit was distributed throughout the inner plexiform layer. Immunoreactivity for the 1B subunit was distributed diffusely on the scleral side, whereas immunoreactivity for the 1B subunit was punctuate on the vitreous side. Immunoreactivity for the 1B subunit was observed in the nuclear region of GFP-positive cells in both the inner nuclear layer and the ganglion cell layer. In the inner plexiform layer, the GFP signal overlapped with immunoreactivity for the 1B subunit, which was diffusely distributed. Immunoreactivity for the 1B subunit appearing as puncta was located closer to the vitreous side than the GFP signal. Further characterization of the immunoreactivity for the 1B subunit was not carried out.
In the dissociated cell preparations, immunoreactivity for the 1A subunit (P/Q-type) was dotted and colocalized with the GFP signal (Fig. 6, P–R, and Table 1). In the slice preparations, immunoreactivity for the 1A subunit was found in the outer plexiform layer, the inner nuclear layer, the inner plexiform layer and in the ganglion cell layer (Figs. 7F and 8, I and J). In the inner nuclear layer, immunoreactivity for the 1A subunit outlined the somas and was strong on the vitreous side. In the ganglion cell layer, similarly outlined somas were also observed. Immunoreactivity for the 1A subunit overlapped with the GFP-signal in the inner nuclear layer and the ganglion cell layer. In the inner plexiform layer, immunoreactivity for the 1A subunit was diffusely distributed throughout the layer and overlapped with the GFP signal. The presence of the 1A and 1B subunits in the inner plexiform layer of the rat retina has also been reported (Xu et al. 2002).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Yoshida et al. 2001). Our data show that the membrane potentials of the cholinergic amacrine cells in the mouse retina are controlled by a combination of voltage-gated K+ currents (the A current and the delayed rectifier K+ current) and voltage-gated Ca2+ currents (the N- and P/Q types). In particular, our data support the notion that delayed rectifier K channels and the P/Q-type Ca channels are important for the formation of unique membrane properties of cholinergic amacrine cells. Types of voltage-gated K channels
In the present study, we identified the A current and the delayed rectifier K+ current electrophysiologically. Because transient currents in type 2 responses had a higher threshold than sustained currents in the same response, in addition to having different kinetics, it is likely that transient currents reflect the properties of the A current and sustained currents reflect that of the delayed rectifier K+ currents. We found type 2 responses frequently in the dissociated cell preparations. Because the dendrites of the cholinergic amacrine cells are lost during the dissociation procedure, it is likely that the frequent observation of type 2 responses in the dissociated cell preparations reflect the fact that both the A current and the delayed rectifier K+ current exist in the soma. However, in the majority of cells in the slice preparations, the responses to voltage steps consisted of only a large outward K+ current with a relatively low threshold ("type 1 responses"). Ozaita et al. (2004) reported that delayed rectifier K+ channels are abundant in the proximal dendrites of cholinergic amacrine cells in the mouse retina. In fact, we found that the amplitude of type 1 responses in the slice preparations was very large, presumably reflecting the massive activation of the delayed rectifier K+ currents in the proximal dendrites. Probably the A currents in the soma become detectable when the massive activation of the delayed rectifier K+ currents in the dendrites does not mask the A current. Such a masking effect on the A current by massive activation of the delayed rectifier K+ current in the proximal dendrites may support the notion that the signals of each dendrite are processed locally in the cholinergic amacrine cells (Euler et al. 2002).
Types of voltage-gated Ca channels
The voltage-gated Ca channels in the cholinergic amacrine cells were inhibited by -AgaIVA (P/Q-type Ca channel blocker) and -conotoxin GVIA (N-type Ca channel blocker), but not nimodipine (L-type Ca channel blocker). In addition, simultaneous application of -AgaIVA and -conotoxin GVIA completely inhibited the voltage-gated Ca2+ currents. We calculated the contribution of P/Q-type and N-type to the total Ca2+ currents and found that contribution of P/Q-type was larger than the contribution of N-type. In addition, GFP signals in the inner plexiform layer overlapped heavily with P/Q-type immunoreactivity but weakly with N-type immunoreactivity. Our findings support the hypothesis that P/Q-type and N-type Ca channels, especially P/Q-type, work as a pathway of Ca2+ influx into cholinergic amacrine cells of the mouse retina. Euler et al. (2002) reported that the increase in amplitude of intradendritic Ca2+ in cholinergic amacrine cells differs between preferred direction and null direction. Whether the intradendritic Ca2+ increase in cholinergic amacrine cells is inhibited by P/Q-type or N-type Ca channel blockers would be an interesting study topic.
Jensen (1995) reported that direction selectivity in ganglion cells is strongly affected by -AgaIVA, slightly affected by -conotoxin GVIA, and not affected by nicardipine (L-type Ca channel blocker) in rabbit retina. This report raised a possibility that P/Q-type and N-type, especially P/Q-type Ca channels, play a crucial role in the formation of direction selectivity at the ganglion cell level. Because voltage-gated Ca channels in ganglion cells have been reported to be T- and L-type (Karschin and Lipton 1989) or L-type (Kaneda and Kaneko 1991; Lasater and Withokovsky 1990; Lukasiewicz and Werblin 1988), it is likely that P/Q-type or N-type Ca channels are located in the presynaptic terminals with reference to direction-selective ganglion cells. Because cholinergic amacrine cells, a presynaptic neuron in direction-selective ganglion cells (Famiglietti 1991), are important for the formation of direction selectivity (Amthor et al. 2002; Yoshida et al. 2001), it is interesting to see if the voltage-gated Ca channels in cholinergic amacrine cells contribute as a presynaptic mechanism for direction selectivity.
Absence of voltage-gated Na channels
In the present study, we could not detect any tetrodotoxin-sensitive currents using the patch-clamp technique as reported in a previous study on the mouse retina (Ozaita et al. 2004) and rabbit retina (Taylor and Wässle 1995). Pan Nav immunoreactivity was detected by Western blot analysis in the membrane fraction but not detected on the membrane surface of cholinergic amacrine cells of the mouse retina. In the rabbit retina, as the voltage-gated Na+ current of the cholinergic amacrine cells disappears after the eye opening in the rabbit retina (Zhou and Fain 1996), it is likely that such a developmental change in the voltage-gated Na channels can occur in the cholinergic amacrine cells of the mouse retina. However, the possible expression of voltage-gated Na+ channels in distal dendrites of the mouse retina has not been completely excluded because Cohen (2001) reported the presence of tetrodotoxin-sensitive action potentials in the cholinergic amacrine cells of the rabbit retina.
Comparison of voltage-gated ionic channels between ON- and OFF-type cholinergic amacrine cells
In two interesting reports on cholinergic amacrine cells, Euler et al. (2002) demonstrated the direction selectivity of intradendritic Ca2+ increases, and Ozaita et al. (2004) reported a density gradient of delayed rectifier K channels along the dendrites. However, because of difficulties in accessing OFF-type cholinergic amacrine cells, previous electrophysiological studies have focused on the membrane properties of ON-type cholinergic amacrine cells (Aboelela and Robinson 2004; Cohen 2001; Ozaita et al. 2004; Taylor and Wässle 1995; Zhou and Fain 1996). In the present study, we examined the voltage-gated ionic channels of ON- and OFF-type cholinergic amacrine cells both electrophysiologically and immunohistochemically. We could not detect any remarkable differences in the voltage-gated ionic currents between ON- and OFF-type cholinergic amacrine cells. Furthermore, no remarkable differences in the voltage-gated ionic channels of ON- and OFF-type cholinergic amacrine cells have been reported in immunohistochemical studies of rat retina (Tian et al. 2003; Xu et al. 2003). The unique signal processing that has been reported for ON-type cholinergic amacrine cells is in all probability basically preserved in OFF-type cholinergic amacrine cells. However, the similarity of the voltage-gated ionic currents does not indicate that ON- and OFF-type cholinergic amacrine cells use the same signal processing system because there is a difference in the distribution of P2X2 purinergic receptors between ON- and OFF-type cholinergic amacrine cells in the mouse retina (Kaneda et al. 2004). Details of the signal processing in OFF-type cholinergic amacrine cells should be studied in future experiments.
Synaptic inputs onto direction selective ganglion cells themselves are direction selective (Fried et al. 2002, 2005). In the present experiments, our data raised a possibility that the voltage-gated Ca channels, especially P/Q-type, can work as a pathway of direction-selective intradendritic Ca2+ increase (Euler et al. 2002). Because release of GABA and ACh from cholinergic amacrine cells is a Ca2+-dependent process (Fried et al. 2005; Zheng et al. 2004), it is interesting to see if the unique membrane properties of cholinergic amacrine cells are essential to form the presynaptic mechanisms of direction selectivity.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: M. Kaneda, Department of Physiology, Keio University School of Medicine, Tokyo 160-8582, Japan (E-mail: mkaneda{at}sc.itc.keio.ac.jp)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Amthor FR, Keyser KT, Dmitrieva NA. Effects of destruction of starburst-cholinergic amacrine cells by the toxin AF64A on rabbit retinal directional selectivity. Vis Neurosci 19: 495–509, 2002.[CrossRef][ISI][Medline]
Ariel M, Daw N. Pharmacological analysis of directionally sensitive rabbit retinal ganglion cells. J Physiol 324: 161–185, 1982.
Barlow HB, Hill RM. Selective sensitivity to direction of movement in ganglion cells of rabbit retina. Science 139: 412–414, 1963.
Borg-Graham LJ. The computation of directional selectivity in the retina occurs presynaptic to the ganglion cell. Nat Neurosci 4: 176–183, 2001.[CrossRef][ISI][Medline]
Cohen ED. Voltage-gated calcium and sodium currents of starburst amacrine cells in the rabbit retina. Vis Neurosci 18: 799–809, 2001.[CrossRef][ISI][Medline]
Dacheux RF, Raviola E. The rod pathway in the rabbit retina: a depolarizing bipolar and amacrine cell. J Neurosci 6: 331–345, 1986.[Abstract]
de la Villa P, Vaquero CF, Kaneko A. Two types of calcium currents of the mouse bipolar cells recorded in the retinal slice preparation. Eur J Neurosci 10: 317–323, 1998.[CrossRef][ISI][Medline]
Djamgoz MBA, Spadavecchia L, Usai C, Vallerga S. Variability of light-evoked response pattern and morphological characterization of amacrine cells in goldfish retina. J Comp Neurol 301: 171–190, 1990.[CrossRef][ISI][Medline]
Euler T, Detweiler PB, Denk W. Directionally selective calcium signals in dendrites of starburst amacrine cells. Nature 418: 845–852, 2002.[CrossRef][Medline]
Famiglietti EV. Synaptic organization of starburst amacrine cells in rabbit retina: analysis of serial thin sections by electron microscopy and graphic reconstruction. J Comp Neurol 309: 40–70, 1991.[CrossRef][ISI][Medline]
Fried SI, Münch TA, Werblin FS. Directional selectivity is formed at multiple levels by laterally offset inhibition in the rabbit retina. Neuron 46: 117–127, 2005.[CrossRef][ISI][Medline]
Fried SI, Münch TA, Werblin FS. Mechanisms and circuitry underlying directional selectivity in the retina. Nature 420: 411–414, 2002.[CrossRef][Medline]
Ishii K, Kaneda M, Li H, Rockland KL, Hashikawa T. Neuron specific distribution of P2X7 purinergic receptors in the monkey retina. J Comp Neurol 459: 267–277, 2003.[CrossRef][ISI][Medline]
Jensen PJ. Effects of Ca2+ channel blockers on directional selectivity of rabbit retinal ganglion cells. J Neurophysiol 74: 12–23, 1995.
Kaneda M, Ishii K, Morishima Y, Akagi T, Yamazaki Y, Nakanishi S, Hashikawa T. OFF-cholinergic-pathway-selective localization of P2X2-purinoceptors in the mouse retina. J Comp Neurol 476: 103–111, 2004.[CrossRef][ISI][Medline]
Kaneda M, Kaneko A. Voltage-gated Calcium currents in isolated retinal ganglion cells of the cat. Jpn J Physiol 41: 35–48, 1991.[CrossRef][ISI][Medline]
Kaneko A, Pinto LH, Tachibana M. Transient calcium current of retinal bipolar cells of the mouse. J Physiol 410: 613–629, 1989.
Karschin A, Lipton SA. Calcium channels in solitary retinal ganglion cells from post-natal rat. J Physiol 418: 379–396, 1989.
Kolb H, Nelson R. Amacrine cells of the cat retina. Vision Res 21: 1625–1633, 1981.[CrossRef][ISI][Medline]
Kolb H, Nelson R. Neural architecture of the cat retina. Prog Retinal Res 3: 21–60, 1984.[CrossRef]
Kolb H, Nelson R, Mariani A. Amacrine cells, bipolar cells and ganglion cells of the cat retina: a Golgi study. Vision Res 21: 1081–1114, 1981.[CrossRef][ISI][Medline]
Lasater EM, Withokovsky P. Membrane currents of spiking cells isolated from turtle retina. J Comp Physiol [A] 167: 11–21, 1990.[Medline]
Lee S, Zhou JZ. The synaptic mechanism of direction selectivity in distal processes of starburst amacrine cells. Neuron 51: 787–799, 2006.[CrossRef][ISI][Medline]
Lukasiewicz P, Werblin F. A slowly inactivating potassium current truncates spike activity in ganglion cells of the tiger salamander retina. J Neurosci 8: 4470–4481, 1988.[Abstract]
MacNeil MA, Masland RH. Extreme diversity among amacrine cells: implication for function. Neuron 20: 971–982, 1998.[CrossRef][ISI][Medline]
Ozaita A, Petit-Jacques J, Völgyi B, Ho CS, Joho RH, Bloomfield SA, Rudy B. A unique role for Kv3 voltage- gated potassium channels in starburst amacrine cell signaling in mouse retina. J Neurosci 24: 7335–7343, 2004.
Rudy B, Chow A, Lau D, Amarilo Y, Ozaita A, Saganich M, Moreno H, Nadal MS, Hernandez-Pineda R, Hernandez-Cruz A, Erisir A, Leonard C, Vega-Saenz de Miera E. Contribution of Kv3 channels to neuronal excitability. Ann NY Acad Sci 868: 304–343, 1999.
Satoh H, Aoki K, Watanabe S-I, and Kaneko A. L-type calcium channels in the axon terminal of mouse bipolar cells. Neuroreport 13: 2161–2165, 1998.[CrossRef]
Satoh H, Kaneda M, Kaneko A. Intracellular chloride concentration is higher in rod bipolar cells than in cone bipolar cells of the mouse retina. Neurosci Lett 310: 161–164, 2001.[CrossRef][ISI][Medline]
Sterling P. How neurons compute direction. Nature 420: 375–376, 2002.[CrossRef][Medline]
Taylor WR, He S, Levick WR, Vaney DI. Dendritic computation of direction selectivity by retinal ganglion cells. Science 289: 2347–2350, 2000.
Taylor WR, Vaney DI. Diverse synaptic mechanisms generate direction selectivity in the rabbit retina. J Neurosci 22: 7712–7720, 2002.
Taylor WR, Wässle H. Receptive field properties of starburst cholinergic amacrine cells in the rabbit retina. Eur J Neurosci 7: 2308–2321, 1995.[CrossRef][ISI][Medline]
Tauchi M, Masland RH. The shape and arrangement of the cholinergic neurons in the rabbit retina. Proc R Soc Lond B Biol Sci 223: 101–119, 1984.[Medline]
Tian M, Zhao J-W, Yang X-L, Xie J-X. Voltage-gated K+ channel subunits on cholinergic and dopaminergic amacrine cells. Neuroreport 14: 1763–1766, 2003.[CrossRef][ISI][Medline]
Wässle H. Parallel processing in the mammalian retina. Nature Rev 5: 1–11, 2004.[CrossRef]
Xu H-P, Zhao J-W, Yang X-L. Expression of voltage-dependent calcium channel subunits in the rat retina. Neurosci Lett 329: 297–300, 2002.[CrossRef][ISI]
