Immunocytochemical and electrophysiological characterization of GABA receptors in the frog and turtle retina

Vision Research 41 (2001) 691– 704 www.elsevier.com/locate/visres

Immunocytochemical and electrophysiological characterization of GABA receptors in the frog and turtle retina L. Vitanova a, P. Kupenova a, S. Haverkamp b, E. Popova a, L. Mitova a, H. Wa¨ssle b,* b

a Department of Physiology, Medical Uni6ersity, Sofia, Bulgaria Max-Planck-Institute for Brain Research, Frankfurt am Main, Frankfurt, Germany

Received 26 September 2000; received in revised form 15 November 2000

Abstract The expression of GABA receptors (GABARs) was studied in frog and turtle retinae. Using immunocytochemical methods, GABAARs and GABACRs were preferentially localized to the inner plexiform layer (IPL). Label in the IPL was punctate indicating a synaptic clustering of GABARs. Distinct, but weaker label was also present in the outer plexiform layer. GABAAR and GABACR mediated effects were studied by recording electroretinograms (ERGs) and by the application of specific antagonists. Bicuculline, the GABAAR antagonist, produced a significant increase of the ERG. Picrotoxin, when co-applied with saturating doses of bicuculline, caused a further increase of the ERG due to blocking of GABACRs. The putative GABACR antagonist Imidazole-4-acidic acid (I4AA) failed to antagonize GABACR mediated inhibition and, in contrast, appeared rather as an agonist of GABARs. © 2001 Elsevier Science Ltd. All rights reserved. Keywords: GABAC receptors; Electroretinogram; Bicuculline; Picrotoxin; Imidazole-4-acidic acid (I4AA)

1. Introduction GABA is the main inhibitory transmitter in the retina, as well as in other parts of the central nervous system. It exerts its effects through three distinct classes of membrane receptors: the ionotropic GABAA and GABAC, and the metabotropic GABAB receptors. The recently described GABAC receptors are chloride channels, composed of several rho (r) subunits. GABAC receptors differ physiologically and pharmacologically from GABAA receptors, which are also chloride channels. GABAC receptors are more sensitive to GABA than GABAA receptors. Unlike GABAA receptors, they mediate small, but sustained chloride currents, which do not desensitize during maintained application of GABA. GABAC currents are not blocked by bicuculline, the selective antagonist of GABAA receptors,

* Corresponding author. Present address: Max-Planck-Institut fu¨r Hirnforschung, Neuroanatomische Abteilung, Deutschordenstrasse 46, D-60528 Frankfurt, Germany. Tel.: + 49-69-96769211; fax: +4969-96769206. E-mail address: [email protected] (H. Wa¨ssle).

and are not modulated by benzodiazepines, barbiturates or neurosteroids (Feigenspan, Wa¨ssle, & Bormann, 1993; Feigenspan & Bormann, 1994; Johnston, 1996; Enz & Cutting, 1998; Feigenspan & Bormann, 1998; Chebib & Johnston, 1999; Bormann, 2000). The retinal GABAC receptors were intensively studied and seem to be widely distributed among different vertebrate species. GABAC receptors have been described in the retina of Pisces: white perch, goldfish, carp, catfish (Qian & Dowling, 1993; Matthews, Ayoub, & Heidelberger, 1994; Qian & Dowling, 1994; Dong & Werblin, 1995; Qian & Dowling, 1995; Zhang & Yang, 1998); Amphibia: frog (Du & Yang, 2000), salamander (Lukasiewicz, Maple, & Werblin, 1994; Lukasiewicz & Werblin, 1994; Gao et al., 2000); Reptiles: turtle (Liu & Lasater, 1994) and Mammalia: rat, ferret, mouse, rabbit, pig, ox and man (Cutting et al., 1991; Feigenspan et al., 1993; Lukasiewicz & Wong, 1997; Massey, Linn, Kittila, & Mirza, 1997; Fletcher, Koulen, & Wa¨ssle, 1998; Greka, Koolen, Lipton, & Zhang, 1998). Several subunits of retinal GABAC receptors have been cloned, including the GABAC receptor r1, r2 and r3 subunits of the human retina,

0042-6989/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 2 - 6 9 8 9 ( 0 0 ) 0 0 2 9 4 - 7

692

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

suggesting the existence of different isoforms of GABAC receptors (Polenzani, Woodward, & Miledi, 1991; Wang, Guggino, & Cutting, 1994; Ogurusu, Taira, & Shingai, 1995; Wang, Hackam, Guggino, & Cutting, 1995; Ogurusu & Shingai, 1996). The GABAC receptors of the retina have mainly been localized to bipolar cell terminals (Lukasiewicz et al., 1994; Lukasiewicz, 1996; Lukasiewicz & Shields, 1998a) however, they have also been found on bipolar cell dendrites (Wa¨ssle, Koulen, Brandsta¨tter, Fletcher, & Becker, 1998; Du & Yang, 2000). GABAC receptors have also been described on horizontal cells of some species (Qian & Dowling, 1993; Dong, Picaud, & Werblin, 1994; Dong & Werblin, 1996; Jung, Lee, Paik, & Bai, 1999), on ganglion cells (Zhang & Yang, 1998) and on photoreceptors (Picaud et al., 1998). In a series of previous experiments (Popova & Penchev, 1990; Kupenova, Vitanova, Mitova, & Belcheva, 1991; Popova, Kupenova, Vitanova, & Mitova, 1995; Kupenova, Vitanova, Popova, & Mitova, 1997; Vitanova, Kupenova, Popova, & Mitova, 1997) it has been shown by electrophysiological methods that GABA is involved in light adaptation, contrast coding and spectral sensitivity of both frog and turtle retina. Now the question arises, which of these functions of GABA are mediated by GABAC receptors. We first studied the retinal distribution of GABAA and GABAC receptors using immunocytochemical methods. From double labeled sections we could show that GABAA and GABAC receptors are differentially distributed. Using bipolar cell markers together with GABAC receptor antibodies we were able to show that bipolar cell axon terminals express GABAC receptors. The functional consequences of the expression of GABAA and GABAC receptors were studied by ERG recordings and by the application of the known GABA antagonists bicuculline, I4AA and picrotoxin. The experiments demonstrate that both GABAA and GABAC receptors are involved with inhibitory circuits in frog and turtle retinae.

2. Material and methods The electrophysiological experiments were carried out on frog (Rana ridibuna) and freshwater turtle (Emys orbicularis) retinae. After deep anaesthesia in cold water, containing 1 g l − 1 tricain (MS-222, Sandoz, Basel, Switzerland), the frog and turtle eyecups were excised and further processed as described below. The animals were sacrified immediately afterwards. The immunocytochemical experiments were performed on frog as well as on another species of turtle (Pseudemys scripta elegans). The tissue was a kind gift of Dr Josef Ammermu¨ller, Universita¨t Oldenburg.

2.1. Immunocytochemistry After opening along the ora serrata, the eyecups were fixed for 15–30 min in 4% paraformaldehyde in phosphate buffer (PB; pH 7, 4). The isolated retinae were washed in PB, dissected and cryoprotected in increasing sucrose concentrations of 10, 20 and 30%. Afterwards, either the isolated retinae or the whole eyecups were sectioned vertically on a freezing microtome at 14 mm thickness and collected on gelatine coated slides. Immunocytochemistry was carried out by the indirect fluorescence method (for details see Haverkamp & Wa¨ssle, 2000). The sections were incubaded in primary antibodies overnight. The following primary antibodies were used: 1. A polyclonal antiserum against the GABAC r1 subunit (Enz, Brandsta¨tter, Wa¨ssle, & Bormann, 1996), recognizing r1, r2 and r3 subunits of the receptor (1:100, made in rabbit). 2. A polyclonal antiserum against carnosine, serving to label the bipolar cells in frog retina (1:500, made in rabbit). It was a kind gift of Dr Frank Margolis (NIH, Bethesda). 3. A monoclonal antibody against rabbit olfactory bulb (ROB) (or Mab5A10, see Onoda & Fujita, 1987; Onoda, 1988), used to label a population of bipolar cells in frog retina (1:400). It was kindly provided from Dr S. Fujita (Mitsubishi Kasei Inst. Life Sci., Machida, Japan). 4. A monoclonal antibody against the a-isoform of protein kinase C (PKC), serving to label a population of bipolar cells in turtle retina (from mouse; 1:100) (Amersham, Arlington Heights, IL, USA). 5. A polyclonal antibody against PKCa (1:10.000, made in rabbit) (Sigma-Aldrich, Taufkirchen, Germany) 6. A monoclonal antibody against the b2/b3 subunit of the GABAA receptor (bd 17, 1:500 (Roche Diagnostics, Mannheim, Germany). The binding of primary antibodies was revealed by secondary antibodies, in which the sections were incubated for 1 h. The following secondary antibodies were used: goat anti-rabbit (1:500) and goat anti-mouse (1:500), coupled to either FITC (green fluorescence; Dianova, Hamburg, Germany), Alexa TM488 (green fluorescence; Molecular Probes, Eugene, Oregon, USA), Alexa TM594 (red fluorescence, Molecular Probes, Eugene, Oregon, USA) or CY3 (red fluorescence; Dianova, Hamburg, Germany). In double labeling experiments the sections were incubated in a mixture of primary antibodies, followed by incubation in a mixture of secondary antibodies. In the control experiments one of the primary antibodies was omitted and in this case only the immunoreactivity due to the remaining antibody was detected.

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

The sections were examined and photographed using a Zeiss photomicroscope (Axiophot, Zeiss, Oberkochen, Germany). Some of the double labeled sections were viewed by means of Leica TCS SP confocal microscope. All the images were processed using the program, Photoshop 4.0.1 (Adobe).

2.2. Electrophysiology The electrophysiological experiments were carried out on excised open eyecup preparations of 32 frogs (Rana ridibunda) and 24 freshwater turtles (Emys orbicularis). The preparations were continuously perfused with Ringer solution and supplied with moistened O2.

2.2.1. Recording The electroretinogram (ERG) was recorded by means of non-polarized Ag/AgCl electrodes at a bandpass of 0.1 – 1000 Hz (Tektronix 5113 Inc.). The ERG b- and d-waves which reflect the activity of ON- and OFF bipolar cells (Dick, Miller, & Dacheux, 1979), were chosen as appropriate electrophysiological signals. 2.2.2. Stimulation Diffuse white light stimuli, with on and off periods of 5 and 25 s, respectively, were presented intermittently. The background intensity was 1.2× 104 quanta× s − 1 ×mm − 2 in the frog experiments and 1.2×104 quanta× s − 1 ×mm − 2 or 1.2×105 quanta ×s − 1 × mm − 2 in the turtle experiments. The test stimuli intensities varied in the range of 7.25×103 – 2.3 ×105 quanta× s − 1 ×mm − 2 in the frog experiments and 2× 104 – 6.3×106 quanta ×s − 1 ×mm − 2 in the turtle experiments. 2.2.3. Antagonists of GABA receptors In order to block GABAA receptor mediated influences, bicuculline (Fluka) was applied in concentrations of 10, 20, 40, 100, 200, 300 and 400 mM. Several substances have been described to antagonize GABAC receptor actions: I4AA (imidazole-4-acetic acid), 3APMPA (3-amino-propyl[methyl]phosphonic acid) and picrotoxin (Johnston, 1996). However, none appears to be selective for GABAC receptors only (Ragozzino et al., 1996). We applied I4AA at concentrations of 50, 100 and 200 mM (Sigma-Aldrich, Taufkirchen, Germany), Picrotoxin at concentrations of 10, 20, 50 and 70 mM (Sigma-Aldrich, Taufkirchen, Germany) and g-amino-n-butyric acid (GABA) at concentrations of 0.5, 1, 2.5, 3.5, 4 and 5 mM (Sigma-Aldrich, Taufkirchen, Germany). The substances were dissolved in Ringer solution and applied alone or in combinations through perfusion (see below). 2.2.4. Data acquisition The amplitudes of the ERG b- (on-response) and

693

d-wave (off-response) were measured from peak to peak. The values obtained during Ringer solution perfusion were considered as control values. To evaluate the influence of a certain substance, the amplitudes of the waves, when the effect of this substance was fully developed, were expressed as a percent of the control values. When a combination of two substances was used, the effect of the combined action was expressed as a percent to the control values or was normalized to the effect of the substance first applied. In cases when oscillations of the amplitudes occured, the amplitudes were averaged for a certain period of time. The Student’s t-test was used for statistical evaluation of the data.

3. Results

3.1. Immunocytochemistry 3.1.1. Distribution of GABAA and GABAC receptors in frog and turtle retinae A section through a frog retina that was double labeled for GABAA receptors (b2/3 subunits) and GABAC receptors is shown in Fig. 1A and B. Both the GABAA (Fig. 1A) and the GABAC (Fig. 1B) receptor labeling is most prominent in the inner plexiform layer (IPL). The label has a distinct punctate appearance and in accordance with results from other retinae (Fletcher et al., 1998; Koulen, Brandsta¨tter, Enz, Bormann, & Wa¨ssle, 1998) we interprete the puncta as an aggregation of the receptors in postsynaptic densities of GABA-ergic synapses. The density of puncta is not uniform across the IPL and bands of higher and lower densities can be discerned. Comparison of GABAA and GABAC immunoreactive bands shows that they occur at different heights within the IPL and do not coincide. In the inner part of the IPL staining for the GABAA receptor appears to be absent. We also compared, using higher magnification, the position of individual GABAA and GABAC immunoreactive puncta and found that they did not coincide either (not shown). Thus, comparable to other retinae (Fletcher et al.; Koulen et al.), GABAA and GABAC receptors of the frog retina appear to occur at different synapses. There was also specific, but weaker, labeling for both GABAA (Fig. 1A) and GABAC receptors (Fig. 1B) in the outer plexiform layer (OPL). In a recent analysis of GABAA and GABAC receptors of the primate retina a comparable labeling pattern was found and it has been shown that the expression is on bipolar cell dendrites close to the photoreceptor terminals (Haverkamp, Gru¨nert, & Wa¨ssle, 2000). Whether the label of the photoreceptors, particularly in Fig. 1B, is specific, can-

694

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

not be answered at present, because photoreceptors have been found in the past to be sticky for many antibodies, including those against GABAC (Enz et al., 1996). However, it has to be emphasized that GABAC receptors have been recently reported in photoreceptors of the mammalian retina (Picaud et al., 1998; Pattnaik, Jellali, Dreyfus, Sahel, & Picaud, 2000). We also observed GABAC labeling in a sparse population of amacrine cells in the inner nuclear layer (Fig. 1B, oblique arrow). A section through a turtle retina that was double

labeled for GABAA and GABAC receptors is shown in Fig. 2A and B, respectively. The pattern of immunostaining is similar to that of other vertebrate retinae including the frog (Fig. 1). Both receptors are preferentially expressed in synaptic hot spots in the IPL. However, prominent staining for GABAC receptors is also present in the OPL (horizontal arrow in Fig. 2B). Comparable to other retinae the stratification pattern in the IPL is different for GABAA and GABAC receptors. This holds true also for individual puncta, suggesting

Fig. 1. Micrographs of a vertical section through the frog retina that was double labeled for the b2/3-subunits of the GABAA receptor (A) and for the r-subunits of the GABAC receptor (B). The Nomarski micrographs were taken from different parts of the section and show the retinal layers (IS: photoreceptor inner segments; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer; scale bar: 50 mm). (A) GABAA receptor immunoreactivity is prominent in the IPL and has a punctate appearance. Some amacrine cell bodies are weakly labeled (small vertical arrows). Weak label can also be detected in the OPL (horizontal arrow). (B) GABAC receptors are concentrated in synaptic hot spots in the IPL. The density of hot spots across the IPL is not uniform and several horizontal bands (strata) of higher and lower density can be detected. An amacrine cell body (oblique arrow) is strongly labeled. Diffuse and weaker label is also present in the OPL (horizontal arrow) and at the photoreceptor inner segments.

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

695

Fig. 2. Micrographs of a vertical section through the turtle retina that was double labeled for the b2/3-subunits of the GABAA receptor (A) and for the r-subunits of the GABAC receptor (B). The Nomarski micrograph shows the retinal layers (conventions as in Fig. 1). (A) GABAA receptor immunoreactivity is restricted to the IPL and has a punctate appearance. (B) GABAC receptor immunoreactivity in the IPL is punctate with a broad band of higher density of puncta in the center of the IPL. In the OPL (horizontal arrows), GABAC receptor immunofluorescence is patchy. The patches represent an aggregation of GABAC receptors underneath the cone pedicles (scale bar: 50 mm).

the two GABA receptor types are involved with different synapses.

3.1.2. Localization of GABAA and GABAC receptors in the IPL Next we wanted to determine whether the GABAC receptor hot spots of the frog IPL coincide with bipolar cell axon terminals. Two markers were applied that label specific populations of bipolar cells in the frog retina: antibodies against carnosine and antibodies against rabbit olfactory bulb (ROB) (Onoda & Fujita, 1987; Panzanelli, Cantino, & Sassoe`-Pognetto, 1997). Carnosine immunoreactivity was prominent in photore-

ceptors and bipolar cell bodies, and their axons descending into the IPL were also strongly labeled. However, it was difficult to discern individual axon terminals (not shown). Labeling with the ROB antibody was more selective and comparable to other vertebrate retinae appears to be restricted to ON-bipolar cells (Haverkamp & Wa¨ssle, 2000). A section that was double labeled for ROB and GABAC receptors is shown in Fig. 3. ROB immunofluorescence is restricted to bipolar cells (Fig. 3A): their dendrites in the OPL, their cell bodies in the inner nuclear layer (INL) and their axon terminals in the inner half of the IPL are distinctly labeled. Punctate GABAC immunoreactivity

696

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

is found throughout the IPL and comparison of Fig. 3A and B shows that the density of puncta is higher at the bipolar cell axon terminals (arrows). In the OPL there is also a correspondence between the labelled bipolar cell dendritic plexus and the weak GABAC immunofluorescence (vertical arrows in Fig. 3A and B). In order to show in more detail the coincidence of GABAC immunoreactive puncta with ROB labeled bipolar cell axon terminals, high power micrographs of the IPL of double labeled sections are shown in Fig. 4. Comparison of Fig. 4A and B shows that the ‘ringlike’ axon terminal (arrow in Fig. 4A) is decorated by GABAC puncta (arrow in Fig. 4B). Similarly, the axon terminal indicated by the arrow in Fig. 4C is covered by GABAC immunoreactive puncta in Fig. 4D. A more detailed comparison of these micrographs shows that most axonal varicosities are covered by GABAC immunoreactive puncta. It is known that a population of bipolar cells in the turtle retina can be immunostained by antibodies against PKCa (Zhang, Dekorver, & Kolb, 1992; Ammermu¨ller & Kolb, 1996). The axon terminals of these bipolar cells stratify mainly in strata 3 and 4 of the IPL. A section that was double labeled for GABAC and PKCa receptors is shown in Fig. 5A and B, respectively. The bipolar cell axonal branches in stratum 3 appear to coincide with a band of high density of GABAC receptors, while those in stratum 4 coincide with a band of low GABAC expression. Observation of the section at high magnification showed that the PKCa labeled axonal branches of stratum 3 actually coincide with GABAC immunoreactive puncta (not shown). We also double labeled a section through the

turtle retina for PKCa and GABAA receptors (Fig. 6). Comparison of Fig. 6A and B shows that the bipolar cell axon terminals (Fig. 6B) occupy a stratum within the IPL that is also covered by GABAA receptors (Fig. 6A). Observation at high magnification confirmed that GABAA immunoreactive puncta are in register with PKCa immunoreactive bipolar cell axon terminals (not shown).

3.1.3. Conclusion The immunocytochemical experiments demonstrate the presence of GABAA and GABAC receptors in both the frog and the turtle retina. Immunoreactivity in the IPL is punctate, suggesting an aggregation of the receptors at synapses. GABAC receptors are preferentially found on bipolar cell axon terminals. GABAC receptors are also present in the OPL, most likely on bipolar cell dendrites. Comparison with results from mammalian retinae shows that the distribution of GABAA and GABAC receptors across vertebrate retinae appears to be similar (Enz et al., 1996; Fletcher et al., 1998; Koulen et al., 1998; Haverkamp et al., 2000). 3.2. Electrophysiology In order to study the more global effects of GABAA and GABAC receptors in the frog and turtle retina ERG recordings were performed and a pharmacological separation of the actions of these receptors by means of their antagonists, bicuculline, I4AA and picrotoxin, was undertaken. Two main sets of experiments were performed: (1) application of I4AA; and (2) application of picrotoxin. They will be described consecutively.

Fig. 3. Low power micrographs of a vertical section through a frog retina that was double labeled for ROB (A) and for the r-subunits of the GABAC receptor (B). The Nomarski micrograph (C) shows the retinal layers (conventions as in Fig. 1). (A) ROB immunoreactivity is found at bipolar cell bodies in the INL (horizontal arow), at their axon terminals in the inner IPL and at their dendrites in the OPL. Comparable to other vertebrate retinae ON-bipolar cells appear to be labeled. (B) GABAC immunoreactivity in the IPL is punctate. Groups of puncta (two arrows) appear to be in register with axon terminals of bipolar cells in A (two arrows). In the OPL diffuse GABAC label (vertical arrow in B) appears to be in register with ON bipolar cell dendrites (vertical arrow in A) (scale bar: 50 mm).

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

697

Fig. 4. High power micrographs of two vertical sections through the IPL of a frog retina that were double labeled for ROB and the r-subunits of the GABAC receptor. (A) ROB labeled axon terminals form varicosities in the inner IPL (arrow). (B) same section as in A, immunolabeled for GABAC receptors. The puncta, indicated by the arrow, coincide with the ‘ringlike’ axon terminal in A and thus most likely represent synapses onto this terminal expressing GABAC receptors. (C) ROB labeled axon terminals form distinct varicosities (arrow). (D) same section as in C, immunolabeled for GABAC receptors. The arrow points to a group of puncta that are colocalized with the axonal varicosities in C (scale bar: 25 mm).

3.2.1. I4AA experiments The effects of I4AA on the ERG were studied using three different concentrations — 50, 100 and 200 mM. Since it has been shown that I4AA antagonizes specifically GABAC receptors (Qian & Dowling, 1994; Picaud et al., 1998), we expected to see a potentiation of the amplitudes of the ERG waves. However, at all concentrations tested, I4AA caused a suppression of the amplitudes of the ERG waves both in frog and turtle retinae (Fig. 7). There was a tendency for saturation of this suppressive effect at a concentration of 200 mM I4AA. It might be supposed that the effects of I4AA, applied alone, depend considerably on the endogeneous content of GABA, which could vary at different functional states of the retina. Therefore, we measured the effects of exogeneously applied GABA and compared it with the combined action of I4AA and GABA, applied simultaneously. Such an approach should reveal the antagonizing action of I4AA on exogeneously applied GABA. In a first step, we applied GABA alone and

measured dose-response curves (Fig. 8A and B, left curves). GABA application caused a reduction of both the b- and d-wave of the ERG. Next we applied I4AA alone, and in combination with GABA (Fig. 8A and B, bar diagrams). We did not observe any signs that I4AA, when applied together with GABA, was able to antagonize the suppressive effect of GABA application. In contrast, in the frog retina (Fig. 8A) application of I4AA caused a further decrease of the ERG amplitudes. The data obtained so far can be explained by two possible alternatives: (1) no GABAC receptors exist in frog and turtle retinae; (2) I4AA does not antagonize the GABAC receptors in frog and turtle retinae, but rather acts as an agonist. The immunocytochemical data presented above as well as the experiments applying picrotoxin reported below demonstrate the presence of GABAC receptors in these retinae, and therefore we have to conclude that at least in frog and turtle retinae I4AA is not a specific antagonist for the effects mediated by GABAC receptors.

698

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

3.2.2. Picrotoxin experiments Picrotoxin (PT) is an open channel blocker of ligand gated Cl− channels. PT is well known as an antagonist of both GABAA and GABAC receptors (for review see Feigenspan & Bormann, 1994, 1998; Bormann, 2000). In the range of concentrations used here it is a selective GABA receptor antagonist (Miller, Frumkes, Slaughter, & Dacheux, 1981; Handford et al., 1996; Schofield et al., 1996). In order to separate its effects on GABAA and GABAC receptors we first applied bicuculline (Bic) alone and afterwards we applied Bic together with PT. Original ERG traces from such an experiment are shown in Fig. 9A and B. Both the b-waves (Fig. 9A and B, left) and the d-waves (Fig. 9A and B, right) are increased during the application of Bic. The additional application of PT causes a further increase of the ERG amplitudes. In order to quantify the actions of Bic and PT, we first measured dose response curves with both drugs applied separately. The dose response curve for Bic application should define the concentration of Bic that effectively blocked all GABAA receptor mediated effects. The results for the turtle retina are shown in Fig. 11A. There is a continuous increase of the ERG amplitude which saturates at a Bic concentration of 300 –400 mM. In the frog retina saturating effect was already present for a Bic concentration of 50 mM. Higher concentrations of Bic caused a decline of the dose response curve, most likely because of a ‘desensitization’ effect (not shown). The dose response curve for PT, measured in the frog retina, is shown in Fig. 10A. It saturated between 50 and 70mM. A comparable result has been measured in the turtle retina (not shown).

The quantitative results of the co-application of Bic and PT are shown in the bar diagrams (Fig. 10B and C, Fig. 11B and C). Both the b- and d-waves in frog and turtle retinae were significantly increased by saturating concentrations of Bic. When Bic and PT were applied together (Bic+ PT in Figs. 10 and 11) a further increase of the ERG amplitudes was observed. The average additional increase caused by PT application was to 141.69 2.4% and to 1759 4.5% for the b- and d-waves, respectively in the frog retina (100% representing the ERG amplitude during application of Bic alone). In the case of the turtle retina the respective increase was to 139.79 2.6% (b-wave) and to 1579 6.1% (d-wave). In addition to its potency to increase the amplitudes of the ERG waves, picrotoxin also changed the temporal characteristics of the response. The time course of both b- and d-waves in frog and turtle retinae became slower and the duration of the ERG waves increased (Fig. 9). Under the influence of PT the peak latencies of b- and d-waves in the frog retina increased by 120 –150 ms, as compared to the peak latencies of the corresponding waves before drug application and the responses became more tonic. These changes possibly reflect the specific temporal characteristics of GABAC receptors, antagonized by PT.

4. Discussion

4.1. Localization of GABAC receptors In close agreement with the results from other vertebrate retinae (for review see Wa¨ssle et al., 1998) we

Fig. 5. Micrograph of a vertical section through a turtle retina that was double labeled for the r-subunits of the GABAC receptor (A) and for PKCa (B). The Nomarski micrograph (C) shows the retinal layers (conventions as in Fig. 1). (A) GABAC immunoreactive puncta are concentrated in a broad band in the center of the IPL. (B) the cell bodies, descending axons and axon terminals of the PKCa-immunoreactive bipolar cells are strongly labeled. The axonal varicosities occupy the center of the IPL. In the OPL the dendrites of the bipolar cells and Landold clubs extending into the ONL are labeled. The small arrows indicate dendrites that are in register with GABAC label in A (scale bar: 50 mm).

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

699

Fig. 6. Micrographs of a vertical section through a turtle retina that was doubled labeled for the b2/3 subunits of the GABAA receptor (A) and for PKCa (B). The Nomarski micrograph (C) shows the retinal layers (conventions as in Fig. 1). (A) GABAA receptors are found throughout the IPL. (B) The PKCa labeled bipolar cell axons terminate in a broad band in the center of the IPL. (Scale bar: 50 mm).

observed GABAC receptors in the frog and the turtle retina in both plexiform layers. Labeling in the IPL was punctate which we interpret, based on our results in other species, as an aggregation of GABAC receptors at GABAergic synapses (Fletcher et al., 1998; Koulen et al., 1998). We could also show by double labeling experiments that many puncta coincide with the axon terminals of bipolar cells (Fig. 4) and thus represent synapses from GABA-ergic amacrine cells onto bipolar cell axon terminals. We also observed in the frog retina (Fig. 1B) that a sparse population of amacrine cell bodies expressed GABAC-immunoreactivity. It is therefore possible that in a few instances GABAC receptors are also expressed at amacrine to amacrine synapses. GABAC receptor immunoreactivity across the IPL was not uniform but several layers of higher and lower density of puncta could be discerned. This likely represents differences among the axon terminals of different bipolar cell types which terminate in different IPL strata where they receive different amounts of GABAC receptor mediated input (Sterling, 1998; Boycott & Wa¨ssle, 1999; Du & Yang, 2000; Wu et al., 2000). Rod bipolar cells and their axon terminals close to the ganglion cell layer receive the most prominent input through GABAC receptors in the mammalian retina. However, both in the frog and in the turtle retina (Fig. 1B and Fig. 2B) a band in the center of the IPL showed the strongest expression of GABAC receptors. A comparable concentration of GABAC receptors in the center of the IPL has also been observed in the chicken retina (Koulen et al., 1997). Both in the turtle and in the chicken retina this band of high GABAC receptor expression shows partial overlap with the axon terminals of PKCa-immunolabeled bipolar cells (Fig. 5; Koulen et al., 1997). The presence of GABAC im-

munoreactive puncta both in the outer and inner part of the IPL suggests that both OFF- and ON-bipolar cell axon terminals express GABAC receptors (Du & Yang, 2000; Wu et al., 2000). We also observed specific expression of GABAC receptors in the OPL. Close inspection of micrographs that were double labeled for bipolar cell markers and GABAC receptors suggests the GABAC receptors are localized on bipolar cell dendrites. There they would receive GABA-ergic input from horizontal cells. We have recently shown in the monkey retina that GABA receptors and the horizontal cell processes that release GABA are in close apposition underneath the cone pedicles (Haverkamp et al., 2000). Since horizontal cells exhibit a non vesicular, Ca2 + independent release of GABA, the GABA receptors in the OPL are not clustered at synapses, but show an aggregation along the dendritic terminals of bipolar cells. This might explain, why in Fig. 1B and Fig. 2B GABAC receptors in the OPL do not have a crisp, punctate appearance comparable to the IPL, but instead are more diffusely distributed. Our GABAC-labeling of bipolar cell dendrites in the OPL is in agreement with recent electrophysiological results that have described GABAC receptor mediated Cl−-currents upon application of GABA to bipolar cell dendrites in the bullfrog retina (Du & Yang, 2000). The GABAC receptor immunofluorescence observed in the cone inner segments of Fig. 1B may indicate that cone photoreceptors of the frog retina also express GABAC receptors. GABAC receptors of cones have been described in some mammalian retinae (Picaud et al., 1998; Pattnaik et al., 2000). However, in our comparative analysis of several mammalian (cat, mouse, rat, rabbit, monkey) and non-mammalian retinae

700

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

(goldfish, chicken, frog and turtle) we found staining of photoreceptors extremely variable across species (Enz et al., 1996; Koulen et al., 1998; Wa¨ssle et al., 1998; Haverkamp et al., 2000), while bipolar cell staining was remarkably similar. One has to postulate, therefore, alternative techniques and different approaches before accepting GABAC receptor expression in frog photoreceptors.

4.2. Localization of GABAA receptor Molecular cloning studies have demonstrated that GABAA receptors are composed of five structurally related subunits that form the Cl−-conducting pore (for review see Macdonald & Olsen, 1994). To date six a subunits, four b subunits, three g subunits, and one d

Fig. 8. The influence of GABA, of I4AA and of the co-application of GABA+I4AA on the frog (A) and turtle (B) ERGs. Left diagrams show the reduction of the b- and d-waves under the influence of GABA applied at different concentrations. The bar diagrams show the reduction of the ERG waves during the application of I4AA alone, of GABA alone and of I4AA and GABA applied together. The following concentrations were used: GABA, 5 mM in A and 3.5 mM in B; I4AA, 200 mM in A and 100 mM in B.

Fig. 7. The influence of I4AA on the frog (A) and turtle (B) electroretinograms (ERGs). The 100% amplitudes represent ERG recordings before the application of I4AA. Three different concentrations of I4AA were applied (50, 100 and 200 mM). Application of I4AA reduced both the b- and the d-wave of the ERG. The mean 9 SEM are represented.

subunit have been described, and their expression in the mammalian retina has been studied with subunit specific antibodies (Greferath et al., 1995). The GABAA receptor subunits were preferentially found in the IPL and multiple isofoms were selectively clustered at specific synapses (Koulen, Sassoe`-Pognetto, Gru¨nert, & Wa¨ssle, 1996). As a rule, a1, a2 and a3 subunits were never found together in the same postsynaptic aggregates. The b2/3 subunits were found in different combinations with a and g subunits, however, not all GABAA receptors contain the b2/3 subunits (Sieghart, 1995). This shows that in the present study we have labeled

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

701

al., 1998). The same wide spread cellular distribution of the b2/3 subunit also has to be postulated for the frog and turtle retinae. The observation of sections double labeled for ROB and PKCa, respectively, and for the b2/3 subunits indicated the coincidence of GABAA receptor immunoreactive puncta and bipolar cell axon terminals. This suggests that bipolar cell axon terminals of the frog and of the turtle retina express both GABAA — possibly with different subunit combinations — and GABAC receptors.

4.3. Electrophysiological recordings

Fig. 9. ERG records from frog (A) and turtle (B) retinae. The b-waves, recorded at light ON are shown to the left, the d-waves, recorded at light OFF are shown to the right. The traces marked with an R represent control records with normal Ringer solution. The traces marked with Bic were recorded during the application of saturating concentrations of bicuculline (50 mM in A and 300 mM in B). The traces marked with Bic +PT represent recordings at which Bic and picrotoxin were applied together (Bic, 50 mM in A, 300 mM in B; PT, 50 mM in A and B). Calibrations A: 0.2 s and 100 mV for trace R and 200 mV for traces Bic and Bic + PT; B: 0.2 s and 50 mV for trace R and 100 mV for traces Bic and Bic +PT.

only a subset of the GABAA receptors, most likely those with subunit composition of a1, b2/3, g2 (Greferath, Gru¨nert, Mu¨ller, & Wa¨ssle, 1994; Greferath et al., 1995). This subunit composition is also common in other parts of the brain (Sieghart, 1995). In the mammalian retina a1, b2/3 and g2 subunits have been localized to bipolar cells, to amacrine cells and to ganglion cells (Greferath et al., 1994; 1995; Fletcher et

Kusama et al. (1993) have reported that I4AA selectively antagonizes the action of GABA on GABAC receptor r1 subunits expressed in Xenopus oocytes. However, it has also been shown that I4AA is a GABAA receptor agonist (Nistri & Constantia, 1979; Kemp, Marshall, & Woodruff, 1986). Qian and Dowling (1994) found that GABAC responses recorded from the horizontal cells of the white perch retina were completely blocked by the application of I4AA (Qian & Dowling, 1994). In bipolar cells of the white perch retina both GABAA and GABAC receptor mediated responses were observed which were completely suppressed by the co-application of Bic+ I4AA. GABAC receptors in cones of the pig retina were also blocked by the application of I4AA (Picaud et al., 1998). More recently it was shown for the GABAC r-subunits of the white perch retina that I4AA behaved as an antagonist of A-type r receptors and as a partial agonist of B-type r receptors (Qian, Dowling, & Ripps, 1998). Using site directed mutagenesis it could be shown that residues at

Fig. 10. (A) the influence of varying concentrations of picrotoxin on the ERG b- and d-wave amplitudes of frog retinae. (B) the increase of the frog ERG b-wave as a result of the application of Bic, and of Bic together with picrotoxin. (C) the increase of the frog ERG d-wave as a result of the application of Bic, and of Bic together with picrotoxin (Bic: 50 mM; PT: 50 mm). R represent control records with normal Ringer solution. The records in B and C were from the same retina, the data in A were from a different retina.

702

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

Fig. 11. (A) the influence of varying concentrations of bicuculline on the ERG b- and d-wave amplitudes of the turtle retina. (B) the increase of the turtle ERG b-wave as a result of the application of Bic, and Bic together with picrotoxin. (C) the increase of the turtle ERG d-wave as a result of the application of Bic, and Bic together with picrotoxin (Bic: 300 mM; PT: 50 mM). R represents control records with normal Ringer solution. All recoreds were from the same retina.

amino acid position 320 in the second transmembrane domain were responsible for the different actions of I4AA (Qian, Dowling, & Ripps, 1999). When the site contains a proline residue, as in r1 subunits, I4AA acts as a partial GABA agonist. If a serine residue is present at this site (as in r2 subunits) I4AA acts as an antagonist of GABA mediated responses. When we applied I4AA to frog or turtle retinae, we observed a reduction of the ERG amplitudes, similar to the reduction recorded during the application of GABA (Figs. 7 and 8). This would suggest I4AA acts as an agonist and this response is mediated by r1 subunits. KapoustaBruneau (2000) has recorded the b-wave of the ERG of the isolated rat retina. She applied another GABAC receptor antagonist, 3-aminorpopyl phosphonic acid (3-APA), and observed, comparable to our results, a reduction of the b-wave. This result suggests that both I4AA and 3-APA act as partial agonists at certain isoforms of GABAC receptors. Du and Yang (2000) have recently measured GABAevoked currents from single bipolar cells of a bullfrog retinal slice preparation. GABAC responses evoked from dendrites could be blocked by the application of I4AA, while GABAC responses evoked at axons were resistant to I4AA. This would suggest that bipolar cell dendrites express the r2 subunit, while r1 subunits would predominate at bipolar cell axon terminals. Fig. 1B and Fig. 2B show strong immunostaining over the axon terminals in the IPL, and only weak label in the OPL. Taken together, all these results suggest that the r1 subunit of the GABAC receptor appears to play the dominant role in frog and turtle retinae. Two different subtypes of GABAC receptors have also recently been reported for bipolar cells of the tiger salamander retina

(Gao et al., 2000): one exhibits stronger I4AA sensitivity than the other, but both are antagonized by picrotoxin. In agreement with results from other vertebrate retinae (tiger salamander: Lukasiewicz & Werblin, 1994; catfish: Dong & Werblin, 1996; ferret: Lukasiewicz & Wong, 1997; carp: Zhang & Yang, 1998; frog: Du & Yang, 2000) we found that picrotoxin (PT) is an efficient blocker of all GABAC receptor mediated responses. It is interesting in this context that for the rat retina, picrotoxin was not a potent antagonist of GABAC receptors (Feigenspan et al., 1993). Once again, site directed mutagenesis demonstrated that in the second transmembrane domain, at amino acid position 314 of the r2 subunit of the rat, a methionine in place of a threonine is the determinant of the PT resistance of the rat GABAC receptor (Zhang, Pan, Zhang, Brideau, & Lipton, 1995). The application of PT not only enhanced the amplitudes of the ERG b- and d-waves, but increases dramatically their duration (Fig. 8). These results confirm our early observations with respect to the sustained type of inhibition exerted by GABA in frog and turtle retinae (Belcheva & Vitanova, 1974, 1978; Vitanova, Penchev, Kupenova, & Belcheva, 1987). The present results suggest that these inhibitory influences have been mediated by GABAC receptors. This is in agreement with the results of Dong and Werblin (1996), who found that the blockade of GABAC receptors in tiger salamander retina caused a transformation of the phasic discharges of ON-OFF amacrine and ganglion cells into sustained responses. They also support the hypothesis that sustained GABA inhibition is mediated by GABAC receptors (Feigenspan & Bormann, 1994; Lukasiewicz & Shields, 1998b).

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

Acknowledgements We are grateful to F. Boij and D. Benzaid for excellent technical assistance; I. Odenthal for typing the manuscript; and Dr B. O’Brien for critically reading and improving the English text. This work was supported by the Deutsche Forschungsgemeinschaft (SFB 269/B4) and by the Council for Medical Science, Medical University Sofia, Bulgaria. References Ammermu¨ller, J., & Kolb, H. (1996). Functional architecture of the turtle retina. Progress in Retinal and Eye Research, 15, 393– 433. Belcheva, S., & Vitanova, L. (1974). Effects of some antagonists of the inhibitory transmitters on the bioelectric activity of the retinal cells. Agressology, 15, 461–469. Belcheva, S., & Vitanova, L. (1978). Effect of picrotoxin and strychnine on the bioelectric activity of the turtle’s retinal neurons. Agressology, 19, 309–320. Bormann, J. (2000). The ‘ABC’ of GABA receptors. Trends in Pharmacological Sciences, 21, 16–19. Boycott, B. B., & Wa¨ssle, H. (1999). Parallel processing in the mammalian retina — the proctor lecture. In6estigati6e Ophthalmology and Visual Science, 40, 1313–1327. Chebib, M., & Johnston, G. A. R. (1999). The ‘ABC’ of GABA receptors: a brief review. Clinical and Experimental Pharmacology and Physiology, 26, 937–940. Cutting, G. R., Lu, L., O’Hara, B. F., Kasch, L. M., MontroseRafizadeh, C., Donovaan, D. M., Shimada, S., Antonarakis, S. E., Guggino, W. B., Uhl, G. R., & Kazazian, H. H. (1991). Cloning of the g-aminobutyric acid (GABA) r1 cDNA: A GABA receptor subunit highly expressed in the retina. Proceedings of the National Academy of Sciences of the USA, 88, 2673–2677. Dick, E., Miller, R., & Dacheux, R. (1979). Neuronal origin of b- and d-wave in the I-type of ERG. In6estigati6e Ophthalmology and Visual Science, 34 (Supplementary), N12 Abstract. Dong, C. J., & Werblin, F. S. (1995). Zinc downmodulates the GABAC receptor current in cone horizontal cells acutely isolted from the catfish retina. Journal of Neurophysiology, 73, 916– 919. Dong, C. J., & Werblin, F. S. (1996). Use-dependent and use-independent blocking actions of picrotoxin and zink at the GABAC receptor in the retinal horizontal cells. Vision Research, 36, 3997– 4005. Dong, C. J., Picaud, S. A., & Werblin, F. S. (1994). GABA transporters and GABAC-like receptors on catfish cone- but not roddriven horizontal cells. Journal of Neuroscience, 14, 2648– 2658. Du, J.-L., & Yang, X.-L. (2000). Subcelular localization and complements of GABAA and GABAC receptors on bullfrog retinal bipolar cells. Journal of Neurophysiology, 84, 666–676. Enz, R., & Cutting, G. R. (1998). Molecular composition of GABAC receptors. Vision Research, 38, 1431–1441. Enz, R., Brandsta¨tter, J. H., Wa¨ssle, H., & Bormann, J. (1996). Immunocytochemical localization of the GABAC receptor r subunits in the mammalian retina. Journal of Neuroscience, 16, 4479– 4490. Feigenspan, A., & Bormann, J. (1994). Differential pharmacology of GABAA and GABAC receptors on rat retinal bipolar cells. European Journal of Pharmacology, 288, 97–104. Feigenspan, A., & Bormann, J. (1998). GABA-gated Cl− channels in the rat retina. Progress in Retinal and Eye Research, 17, 99 – 126. Feigenspan, A., Wa¨ssle, H., & Bormann, J. (1993). Pharmacology of GABA receptor Cl− channels in rat bipolar cells. Nature, 361, 159– 162.

703

Fletcher, E., Koulen, P., & Wa¨ssle, H. (1998). GABAA and GABAC receptors on mammalian rod bipolar cells. Journal of Comparati6e Neurology, 396, 351– 365. Gao, F., Maple, B. R., & Wu, S. M. (2000). I4AA-sensitive chloride current contributes to the center light responses of bipolar cells in the riger salamander retina. Journal of Neurophysiology, 83, 3473– 3482. Greferath, U., Gru¨nert, U., Mu¨ller, F., & Wa¨ssle, H. (1994). Localization of GABAA receptors in the rabbit retina. Cell and Tissue Research, 276, 295– 307. Greferath, U., Gru¨nert, U., Fritschy, J. M., Stephenson, A., Mo¨hler, H., & Wa¨ssle, H. (1995). GABAA receptor subunits have differential distributions in the rat retina: in situ hybridization and immunohistochemistry. Journal of Comparati6e Neurology, 353, 553– 571. Greka, A., Koolen, J. A., Lipton, S. A., & Zhang, D. X. (1998). Cloning an characterization of mouse GABAC receptor subunits. Neuroreport, 9, 229– 232. Handford, C. A., Lynch, J. W., Baker, E., Webb, G. C., Ford, J. C., Sutherland, G. R., & Schofield, P. R. (1996). The human glycine receptor b subunit: primary structure, functional characterization and chromosomal localization of the human and murine genes. Molecular Brain Research, 35, 211– 219. Haverkamp, S., & Wa¨ssle, H. (2000). Immunocytochemical analysis of the mouse retina. Journal of Comparati6e Neurology, 424, 1 – 23. Haverkamp, S., Gru¨nert, U., & Wa¨ssle, H. (2000). The cone pedicle, a complex synapse in the retina. Neuron, 27, 85 – 95. Johnston, G. A. R. (1996). GABAC receptors: relatively simple transmitter-gated channels? Trends in Pharmacological Sciences, 17, 319– 323. Jung, C.-S., Lee, S.-L., Paik, S.-S., & Bai, S.-H. (1999). The GABAC receptor is present in cone-horizontal cell axon terminals isolated from catfish retina. Neuroscience Letters, 260, 185– 188. Kapousta-Bruneau, N. V. (2000). Opposite effects of GABAA and GABAC receptor antagonists on the b-wave of ERG recorded from the isolated rat retina. Vision Research, 40, 1653– 1665. Kemp, J. A., Marshall, G. R., & Woodruff, G. N. (1986). Quantitative evaluation of the potencies of GABA-receptor agonists and antagonists using the rat hippocampal slice preparation. British Journal of Pharmacology, 87, 677– 684. Koulen, P., Sassoe`-Pognetto, M., Gru¨nert, U., & Wa¨ssle, H. (1996). Selective clustering of GABAA and glycine receptors in the mammalian retina. Journal of Neuroscience, 16, 2127– 2140. Koulen, P., Brandsta¨tter, J. H., Kro¨ger, S., Enz, R., Bormann, J., & Wa¨ssle, H. (1997). Immunocytochemical localization of the GABAC receptor r subunits in the cat, goldfish, and chicken retina. Journal of Comparati6e Neurology, 380, 520– 532. Koulen, P., Brandsta¨tter, J. H., Enz, R., Bormann, J., & Wa¨ssle, H. (1998). Synaptic clustering of GABAC receptor r-subunits in the rat retina. European Journal of Neuroscience, 10, 115– 127. Kupenova, P., Vitanova, L., Mitova, L., & Belcheva, S. (1991). Participation of the GABAergic system of the turtle retina in the light adaptation process. Acta Physiologica Scandina6ica, 141, 203– 210. Kupenova, P., Vitanova, L., Popova, E., & Mitova, L. (1997). Influence of picrotoxin and strychnine on the spectral sensitivity of the turtle ERG b- and d-wave: I. Dark adaptation. Acta Physiologica Scandina6ica, 159, 217– 225. Kusama, T., Spivak, C. E., Whiting, P., Dawson, V. L., Schaeffer, J. C., & Uhl, G. R. (1993). Pharmacology of GABA r1 and GABA a/b receptors expressed in Xenopus oocytes and COS cells. British Journal of Pharmacology, 109, 200– 206. Liu, Y., & Lasater, E. M. (1994). Multiple GABA receptors of ganglion cells in the turtle retina. In6estigati6e Ophthalmology and Visual Science, 35 (Supplementary), 2154 Abstract.

704

L. Vitano6a et al. / Vision Research 41 (2001) 691–704

Lukasiewicz, P. (1996). GABAC receptors in the vertebrate retina. Molecular Neurobiology, 12, 181–194. Lukasiewicz, P. D., & Werblin, F. S. (1994). A novel GABA receptor modulates synaptic transmission from bipolar to ganglion and amacrine cells in the tiger salamander retina. Journal of Neuroscience, 14, 1213– 1223. Lukasiewicz, P. D, & Wong, R. O. L. (1997). GABAC receptors on ferret retinal bipolar cells: a diversity of subtypes in mammals? Visual Neuroscience, 14, 989–994. Lukasiewicz, P. D., & Shields, C. R. (1998a). A diversity of GABA receptors in the retina. Seminars in Cell and De6elopmental Biology, 9, 293– 299. Lukasiewicz, P. D., & Shields, C. R. (1998b). Different combination of GABAA and GABAC receptors confer distinct temporal properties to retinalsynaptic responses. Journal of Neurophysiology, 79, 3157– 3167. Lukasiewicz, P., Maple, B. R., & Werblin, F. S. (1994). A novel GABA receptor on bipolar cell terminals in the tiger salamander retina. Journal of Neuroscience, 14, 1201–1212. Macdonald, R. L., & Olsen, R. W. (1994). GABAA-receptor channels. Annual Re6iews of Neuroscience, 17, 569–602. Massey, S., Linn, D., Kittila, C., & Mirza, W. (1997). Contribution of GABAA receptors and GABAC receptors to acetylcholine release and directional selectivity in the rabbit retina. Visual Neuroscience, 14, 939–948. Matthews, G., Ayoub, G. S., & Heidelberger, R. (1994). Presynaptic inhibition by GABA is mediated via two distinct GABA receptors with novel pharmacology. Journal of Neuroscience, 14, 1079– 1090. Miller, R. F., Frumkes, T. E., Slaughter, M., & Dacheux, R. F. (1981). Physiological and pharmacological basis of GABA and glycine action on neurons of mudpuppy retina. I. Receptors, horizontal cells, bipolars and G-cells. Journal of Neurophysiology, 45, 743– 763. Nistri, A., & Constantia, A. (1979). Pharmacological characterization of different types of GABA and glutamate receptors in vertebrates and invertebrates. Progress in Neurobiology, 13, 117–235. Ogurusu, T., & Shingai, R. (1996). Cloning of a putative g-aminobutyric acid (GABA) receptor subunit r3 cDNA. Biochimica et Biophysica Acta, 1305, 15–18. Ogurusu, T., Taira, H., & Shingai, R. (1995). Identification of GABAA receptor subunit in the rat retina: cloning of the rat GABAA receptor r2 subunit cDNA. Journal of Neurochemistry, 65, 964– 968. Onoda, N. (1988). A monoclonal antibody specific for a subpopulation of retinal bipolar cells in vertebrates. Neuroscience Research, 8 (Supplementary), S113–S125. Onoda, N., & Fujita, S. C. (1987). A monoclonal antibody specific for a subpopulation of retinal bipolar cells in frog and other vertebrates. Brain Rsearch, 416, 359–363. Panzanelli, P., Cantino, D., & Sassoe`-Pognetto, M. (1997). Co-localization of carnosine and glutamate in photoreceptors and bipolar cells of the frog retina. Brain Research, 758, 143–152. Pattnaik, B., Jellali, A., Dreyfus, H., Sahel, J., & Picaud, S. (2000). GABAC receptors are localized with microtubule-associated protein 1B in mammalian cone photoreceptors. Journal of Neuroscience, 20, 6781– 6788. Picaud, S., Pattnaik, B., Hicks, D., Forster, V., Fontaine, V., Sahel, J., & Dreyfus, H. (1998). GABAA and GABAC receptors in adult porcine cones: evidence from a photoreceptor — glia co-culture model. Journal of Physiology, 513, 33–42. Polenzani, L., Woodward, R. M., & Miledi, R. (1991). Expression of mammalion g-aminobutyric acid receptors with distinct pharmacology in Xenopus oocytes. Proceedings of National Academy of Science USA, 88, 4318–4322. Popova, E., & Penchev, A. (1990). GABAergic control on the intensity-response function of ERG waves under different background illumination. Biomedica Biochimica Acta, 49, 1005–1013.

Popova, E., Kupenova, P., Vitanova, L., & Mitova, L. (1995). Effect of 4-amino-phosphono-butyric acid (APB) on ERG off-response after flycinergic and GABAergic blockade. Vision Research, 35, 1945– 1949. Qian, H., & Dowling, J. E. (1993). Novel GABA responses from rod-driven retinal horizontal cells. Nature, 361, 162– 164. Qian, H., & Dowling, J. E. (1994). Pharmacology of novel GABA receptors found on rod horizontal cells of the white perch retina. Journal of Neuroscience, 14, 4299– 4307. Qian, H., & Dowling, J. E. (1995). GABAA and GABAC receptors on hybrid bass retinal bipolar cells. Journal of Neurophysiology, 75, 1920– 1928. Qian, H., Dowling, J. E., & Ripps, H. (1998). Molecular and pharmacological properties of GABA-r subunits from white perch retina. Journal of Neurobiology, 37, 305– 320. Qian, H., Dowling, J. E., & Ripps, H. (1999). A single amino acid in the second transmembrane domain of GABA r subunits is a determinant of the response kinetics of GABAC receptors. Journal of Neurobiology, 40, 67 – 76. Ragozzino, D., Woodward, R. M., Murata, Y., Eusebi, F., Overman, L. E., & Miledi, R. (1996). Disign and in vitro pharmacology of a selective g-aminobutyric acid C receptor antagonist. Molecular Pharmacology, 50, 1024– 1030. Schofield, P. R., Lynch, J. W., Rahendra, S., Pierce, K. D., Handford, C. A., & Barry, P. H. (1996). Molecular and genetic insights into ligand binding and signal transduction at the inhibitory glycine receptor. In Cold spring harbor symposia on quantitati6e biology, vol. 61 (pp. 333– 342). Cold Spring Harbor Laboratory Press. Sieghart, W. (1995). Structure and pharmacology of g-aminobutyric acidA receptor subtypes. Pharmacological Re6iews, 47, 181–234. Sterling, P. (1998). Retina. In G. M. Shepherd, The synaptic organization of the brain (pp. 205– 253). New York: Oxford University Press. Vitanova, L., Kupenova, P., Popova, E., & Mitova, L. (1997). Influence of picrotoxin and strychnine on the spectral sensitivity of the turtle ERG b- and d-wave: II. Light adaptation. Acta Physiologica Scandina6ica, 159, 227– 235. Vitanova, L., Penchev, A., Kupenova, P., & Belcheva, S. (1987). Influence of picrotoxin on the retinal sensitivity at different level of background illumination. Physiologica Bohemoslo6aca, 36, 463– 471. Wa¨ssle, H., Koulen, P., Brandsta¨tter, J. H., Fletcher, E. L., & Becker, C.-M. (1998). Glycine and GABA receptors in the mammalian retina. Vision Research, 38, 1411– 1431. Wang, T. L., Guggino, W. B., & Cutting, G. R. (1994). A novel g-aminobutyric acid receptor subunit (r2) cloned from human retina forms bicuculline-insensitive homo-oligomeric receptors in Xenopus oocytes. Journal of Neuroscience, 14, 6524– 6531. Wang, T. L., Hackam, A. S., Guggino, W. B., & Cutting, G. R. (1995). A single amino acid in g-aminobutyric acid r1 receptors affects competitive and noncompetitive components of picrotoxin inhibition. Proceedings of the National Academy of Science USA, 92, 11 751– 11 755. Wu, S. M., Gao, F., & Maple, B. R. (2000). Functional architecture of synapses in the inner retina: segregation of visual signals by stratification of bipolar cell axon terminals. Journal of Neuroscience, 20, 4462– 4470. Zhang, D. Q., & Yang, X. L. (1998). GABA modulates color-opponent bipolar cells in carp retina. Brain Research, 792, 319–323. Zhang, L. I., Dekorver, L., & Kolb, H. (1992). Immunocytochemical staining with antibodies against protein kinase C and its isozymes in the turtle retina. Journal of Neurocytology, 21, 833– 847. Zhang, D., Pan, Z.-H., Zhang, X., Brideau, A. D., & Lipton, S. A. (1995). Cloning of a g-aminobutyric acid type C receptor subunit in rat retina with a methionine residue critical for picrotoxinin channel block. Proceedings of the National Academy of Science USA, 92, 11 756– 11 760.