Exp Neurobiol 2017; 26(6): 329-338
Published online December 31, 2017
© The Korean Society for Brain and Neural Sciences
Jong Woo Lee1, Min Young Lim1,2, Yong Soo Park1,3, Su Jin Park1,3 and In-Beom Kim1,3,4*
1Department of Anatomy, College of Medicine, The Catholic University of Korea, Seoul 06591, 2Gimpo Hangil Eye Center, Gimpo 10110, 3Catholic Neuroscience Institute, College of Medicine, The Catholic University of Korea, Seoul 06591, 4Catholic Institute for Applied Anatomy, College of Medicine, The Catholic University of Korea, Seoul 06591, Korea
Correspondence to: *To whom correspondence should be addressed.
TEL: 82-2-2258-7263, FAX: 82-2-536-3110
Dopaminergic amacrine cells (DACs) are among the most well-characterized neurons in the mammalian retina, and their connections to AII amacrine cells have been described in detail. However, the stratification of DAC dendrites differs based on their location in the inner plexiform layer (IPL), raising the question of whether all AII lobules are modulated by dopamine release from DACs. The present study aimed to clarify the relationship between DACs and AII amacrine cells, and to further elucidate the role of dopamine at synapses with AII amacrine cell. In the rabbit retina, DAC dendrites were observed in strata 1, 3, and 5 of the IPL. In stratum 1, most DAC dendritic varicosities—the presumed sites of neurotransmitter release—made contact with the somata and lobular appendages of AII amacrine cells. However, most lobular appendages of AII amacrine cells localized within stratum 2 of the IPL exhibited little contact with DAC varicosities. In addition, double- or triple-labeling experiments revealed that DACs did not express the GABAergic neuronal markers anti-GABA, vesicular GABA transporter, or glutamic acid decarboxylase. These findings suggest that the lobular appendages of AII amacrine cells are involved in at least two different circuits. We speculate that the circuit associated with stratum 1 of the IPL is modulated by DACs, while that associated with stratum 2 is modulated by unknown amacrine cells expressing a different neuroactive substance. Our findings further indicate that DACs in the rabbit retina do not use GABA as a neurotransmitter, in contrast to those in other mammals.
Dopamine is the predominant catecholamine transmitter in the mammalian retina . At present, the most effective marker of dopaminergic neurons in the retina is an antibody against tyrosine hydroxylase (TH), the rate-limiting enzyme for catecholamine synthesis, because norepinephrine and epinephrine typically occur only in trace amounts in the retina . Dopaminergic neurons of the retina form a dense dendritic plexus at stratum 1 of the inner plexiform layer (IPL). In rabbit and some other species, dopaminergic neurons are typical amacrine cells whose dendrites do not extend beyond the IPL . In other species, they are interplexiform cells, which send additional processes to the outer plexiform layer (OPL) . Dopaminergic amacrine cells (DACs) are a particularly well-described subtype of retinal neurons, and are among the wide-field amacrine cells that receive input from the cone bipolar cells [5,6,7]. DACs in the retina establish synapses on AII amacrine cells [6,7,8,9,10], which are involved in the rod pathway that carries dim-light signals to ganglion cells. The balance between rod and cone inputs to the ganglion cells is under the control of AII amacrine cells, which are inhibited by DACs [11,12]. Light increases dopamine release, and this light-induced release of dopamine is believed to play a role in the inhibitory mechanisms underlying light adaptation [7,12,13]. Thus, dopamine acts as a neurotransmitter and modulator in the retina.
AII amacrine cells are crucial interneurons of the scotopic pathway, as these cells transfer rod signals to the photopic pathway and represent the main synaptic target of DACs. AII amacrine cells have distinct bistratified dendrites: Lobular appendages arising from the main dendrite are distributed over sublamina
Colocalization of dopamine with other neurotransmitters appears to be common in the brain. For example, γ-aminobutyric acid (GABA) is localized in some periglomerular cells of the olfactory bulb  and in a subpopulation of dopamine-containing neurons of the substantia nigra , whereas glutamate is localized in the remaining nigral neurons  and ventral tegmental neurons . In the retina of various mammalian species, DACs contain GABA [20,24,25]. A previous study in the rodent retina has demonstrated that synapses between DAC and AII amacrine cells are GABAergic . Nevertheless, additional evidence indicates that only very few of DACs contain very low levels of GABA , suggesting that DAC synapses are not GABAergic in all mammalian retinas. Moreover, recent reports have demonstrated that DACs also express glutamate [28,29].
In the present study, we aimed to further elucidate the relationship between DACs and AII amacrine cell, as well as the role of dopamine at synapses between the two cell types. In addition, we aimed to determine whether DACs in the rabbit retina use GABA as a neurotransmitter, similar to those in other mammals. Our findings demonstrated that DACs exert a substantial influence on AII amacrine cells, although DACs may not form synapses at all AII lobules. We further observed that DACs in the rabbit retina do not contain any reliable GABAergic markers, such as GABA, vesicular GABA transporter (VGAT), or the catalytic enzyme glutamic acid decarboxylase (GAD). Thus, our findings indicate that DACs in the rabbit retina are not GABAergic, in contrast to those of other mammals.
Eight adult albino rabbits of both sexes (weight: 3~5 kg) were used for the experiments of the present study. Retinas were prepared in accordance with previously described methods . Briefly, animals were deeply anesthetized via an intraperitoneal (i.p.) injection of urethane (loading dose: 1.5 g/kg). Following enucleation, animals were sacrificed
The retina was removed from the sclera, placed onto nitrocellulose filter paper, and fixed in 4% paraformaldehyde for 2 h. After fixation, the retina was washed with phosphate-buffered saline (PBS) containing 0.5% Triton and 0.1% sodium azide.
Immunohistochemistry was conducted in accordance with previously described methods . Briefly, whole-mount retina or free-floating vibratome vertical preparations were blocked with 3% donkey serum in PBS containing 0.5%Triton X-100/0.1% sodium azide for 2 h to reduce nonspecific labeling. The tissues were then incubated in mixture of primary antibodies—usually mouse monoclonal antibodies to TH (1:3,000; provided by Dr. John Haycock, Louisiana State University, Baton Rouge, LA) and goat polyclonal antibodies to calretinin (CR) (1:4,000; Chemicon, El Segunda, CA)—in the presence of 1% donkey serum/PBS/with 0.5%Triton X-100/0.1% sodium azide for 5~10 days. Samples that lacked primary antibodies were used as blank controls. Following extensive washing with PBS containing 0.5%Triton X-100/0.1% sodium azide, the tissue was incubated overnight with a 1:200 dilution of secondary antibodies conjugated to the appropriate fluorochromes, which were Cy-3 (Jackson Labs, West Grove, PA), and Alexa 488 (Molecular Probes, Eugene, OR). After several washes in PBS, the fluorescent specimens were mounted with Vectashield mounting medium (Vector, Burlingame, CA). Digital images (1,024×1,024 pixels) were acquired using a Zeiss LSM 510 confocal microscope and processed in Photoshop (Adobe Systems, San Jose, CA).
For triple-labeling experiments, the retina was incubated in antibodies against TH (1: 3,000), CR (1:4,000), and one of the following, as appropriate: vesicular glutamate transporter 1 (VGLUT1) (guinea pig polyclonal, 1:1,000; Chemicon), VGAT (guinea pig polyclonal, 1:1,000, Chemicon), GABA (rabbit polyclonal, 1:1,000; provided by Dr. David Pow, University of Queensland, Brisbane, Austrailia), or GAD65/67 (rabbit polyclonal, 1:1,000; Chemicon). Appropriate secondary antibodies conjugated with Cy-3, Cy-5, or Alexa 488, were used.
Image-Pro Plus and custom software were used to analyze the association between DAC varicosities and AII lobules . Boxes (50×50 pixels) were clipped from the image by centering a sampling box around the CR-immunoreactive AII lobules in stratum 1 of the IPL. Alignment and averaging of these boxes produced a plot of color intensity against the location of pixels in the
An antibody against TH clearly labeled the somata and dendrites of DACs, as previously reported [2,32]. In the rabbit retina, DACs exhibited large somata and dendrites, which were located adjacent to the inner nuclear layer (INL) in stratum 1 of the IPL (Fig. 1A). In addition, a small number of TH-labeled dendrites were observed in strata 3 and 5 of the IPL. While these rather sparse processes may have represented dendrites arising from DACs, they may also have represented those of type II catecholaminergic amacrine cells . In whole-mount preparations, each DAC soma gave rise to 3 to 5 primary dendrites, as well as fine dendrites that formed ring-like structures around AII amacrine somata labeled with an antibody against CR (Fig. 1B). These findings are in accordance with the previously reported properties of typical DACs [2,8,9].
In the present study, we aimed to demonstrate the relationship between DAC varicosities and both types of AII dendrites—lobules and arboreal dendrites—which are associated with chemical and electrical synapses, respectively. Fig. 2A shows the relationship between DAC varicosities and AII lobules in stratum 1 of the IPL. In this stratum, DAC dendrites (green) with varicosities form a dense plexus, and the base of AII somata and lobules (red) can be observed. The merged image indicates that almost all AII lobules and AII somata appear to be localized close to DAC varicosities. Therefore, these findings demonstrate that AII lobules form synapses with DAC varicosities in stratum 1 of the IPL.
The relationships between DAC varicosities and AII arboreal dendrites in strata 3 and 5 of the IPL are shown in Figs. 2B and 2C, respectively. Although a few DAC dendrites with varicosities were observed in stratum 3, several AII primary and secondary arboreal dendrites were identified in this region. DAC varicosities with a diameter >0.5 µm appeared to associate with AII dendrites in stratum 3. Among DAC varicosities in stratum 3, 66% (76/116) made contact with AIIs. In stratum 5, several DAC dendrites were visible, along with a dense arboreal meshwork of AII cells, 67% (104/156) made contact with AII dendrites.
We further quantitatively assessed the relationship between DAC varicosities and AII lobules in stratum 1 of the IPL, using custommade software . A cursor was centered on each AII lobule, and a 50-×-50-pixel box was clipped from the image. During this manual selection process, the red and green channels were turned off to reduce operator bias. These image sections were then aligned and averaged. In effect, this method enables one to analyze the average distribution of labeling in the other two channels around a repeated neuronal structure. In the present study, we evaluated the average distribution of DAC varicosities around AII lobules (Fig. 3). Fig. 3A shows the central peak of AII lobules with very low noise, while Fig. 3B shows the average distribution of DAC varicosities around the AII lobules. In Fig. 3A, a distinct volcanic caldera can be observed around a central cavity coincident with the average AII lobular peak. In this case, the height of the caldera immediately adjacent to the AII lobules exceeds the local noise at a distance from the central cavity. This result demonstrates that, on average, there is a high probability of encountering DAC varicosities immediately adjacent to an AII lobule. Controls for this analysis were developed by rotating the DAC 180o from the AII image. Following rotation, we observed no peaks in association with the AII lobular signal (Fig. 3C), indicating that AII lobules form synapses with DAC varicosities in this stratum.
AII amacrine cells are crucial interneuron of the scotopic pathway, and their morphology has been well-characterized in previous studies [14,19,33]. Fig. 4A shows an AII amacrine cell with distinct bistratified dendrites in a vertical section of the rabbit retina: Lobular appendages arising from the main dendrite are distributed through strata 1 and 2 of the IPL, while the finer arboreal dendrites penetrate strata 1 and 2, and stratified throughout strata 3 to 5, in accordance with the findings of previous studies [14,33]. Previous studies have suggested that the synapses between AII lobules and the varicosities of DACs in stratum 1 are dopaminergic [6,7,8,9,10]. However, DAC dendrites very narrowly stratify in the most distal portion of stratum 1 in the IPL, while AII lobules are localized throughout strata 1 and 2. Therefore, we hypothesized that AII lobules exhibit two types of synaptic circuits in sublamina
Fig. 4B and 4C include stack images (thickness: 1 µm) obtained at different focal planes in the same field of a whole-mount preparation, showing the relationship among AII lobules (red), DAC varicosities (green) and the terminals of OFF-cone bipolar cells (blue) according to the level of sublamina
DACs are known to contain GABA and are thus regarded as a GABAergic subpopulation in the mammalian retina [20,24,25]. Recently, Contini and Raviola demonstrated that synapses formed between DACs and AII amacrine cells in the rodent retina are GABAergic . However, some evidence suggests that that only very few of DACs contain very low level of GABA . Moreover, recent reports have demonstrated that DACs express glutamate rather than GABA [28,29]. Thus, we aimed to confirm whether DAC synapses in the rabbit retina are GABAergic.
We first investigated whether GABA is localized in DAC varicosities using an antibody against VGAT, a transporter that delivers GABA to synaptic vesicles at presynaptic terminals . Fig. 5 depicts images obtained from a whole-mount preparation triplelabeled with antibodies against TH (green), CR (red), and VGAT (blue). VGAT immunoreactivity was not observed in DAC varicosities, soma, or dendrites; however, VGAT-labeled cells were observed in all lobules (but not somata) in stratum 1 of the IPL (Fig. 5A, arrowheads). In stratum 5, VGAT did not colocalize with TH and CR, indicating that DAC processes and AII arboreal dendrites in this stratum do not contain GABA (Fig. 5B).
We then aimed to confirm whether DACs of the rabbit retina are GABAergic
Recent studies have revealed that DACs innervate AII amacrine cells in a compartment-specific manner and make contact with various types of amacrine cells, including AII cells [35,36]. In the present study, we investigated the relationship between DACs and AII amacrine cells in the rabbit retina, as well as the neurochemical property of DACs, using immunohistochemistry and confocal microscopy. In accordance with the findings of previous studies, our results indicate that DACs exert a substantial influence on AII somata and lobules. Interestingly, however, we also identified AII lobules that were not contacted by DAC varicosities in stratum 2 of sublamina
In the present study, the stratification and morphological characteristics of DAC somata and dendrites were consistent with those reported in previous reports [2,32]. The characteristic rings surrounding AII somata at the border between the INL and IPL were clearly observed. Although the synaptic circuits associated with these characteristic rings have been well-described, the precise circuitry associated with synapses between DACs and AII cells remains to be elucidated.
The lobular appendages are the major chemical synaptic output sites of AII amacrine cells [15,17]. These lobules form reciprocal inhibitory synapses with the terminals of OFF-cone bipolar cells in strata 1 and 2 [17,18,37,38]. These synapses account for 90% of the output from AII amacrine cells, which rarely synapse directly onto OFF-ganglion cells in the rabbit retina . Both AII somata and dendrites are postsynaptic to other amacrine cells in stratum 1. Double labeling studies have revealed that the dendrites of DACs contact lobules and encircle AII somata, forming characteristic rings [6,8,10]. These DAC inputs are believed to mediate changes in AII/AII coupling that occur during light adaptation [11, 12].
In the present study, we demonstrated that AII lobules in stratum 1 of sublamina
GABA is colocalized with dopamine in DACs of the retina in various mammalian species [20,24,25]. A clever experiment by Contini and Raviola demonstrated that DAC varicosities express both GABA and VGAT, and that postsynaptic GABAA receptors are localized between the dendrites of DACs and AII somata in the rodent retina, indicating that DAC synapses onto AII amacrine cells are GABAergic . While postsynaptic dopamine receptors are not localized on the surface of AII somata, vesicular monoamine transporter-2 colocalizes with VGAT in the same DAC varicosities, suggesting that dopamine diffuses far from the synapse and exerts paracrine effects at the extrasynaptic sites of metabotropic receptors [13,25,42,43].
However, in the present study, DACs in the rabbit retina did not express GABAergic markers such as GABA or GAD (an enzyme involved in GABA synthesis). VGAT, a transporter that delivers GABA to synaptic vesicles, was not localized in the soma, dendrites, or varicosities of DACs. However, previous studies have indicated that glycinergic AII lobules contain VGAT, indicating that VGAT also delivers glycine to synaptic vesicles . In accordance with the findings of previous studies [27,28,29], our results support the notion that DACs in the rabbit retina contain dopamine only, and suggest that synapses between DAC varicosities and AII somata/lobules in stratum 1 of the IPL are dopaminergic rather than GABAergic. Taken together, these findings indicate that synaptic input from DACs in the rabbit retina may be mediated only by dopamine, which is known to modulate the permeability of tracers for AII-AII gap junctions [44,45]. Further study should be needed to provide the physiologic implication of the absence of GABAergic neurotransmission from DACs in the rabbit retina in contrast to those in other mammals.
Lastly, we want to comment on distinct characteristics of DACs in the rabbit retina. As mentioned above, previous studies have revealed that DACs are present in various species and share common morphological and synaptic features [5,6,7,8,9,10]. However, rabbit DACs are typical amacrine cells whose dendrites do not extend beyond the IPL , while those in other species are interplexiform cells that send additional processes to the OPL . In addition, rabbit DACs do not use GABA as a co-neurotransmitter. These specific features suggest that synaptic wiring and the mechanisms underlying visual processing in the retina differ between rabbits and other mammals. This hypothesis is in accordance with the findings of a recent report by Ding et al., who demonstrated that direction-selectivity—a common mechanism in the mammalian retina—occurs