Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Aug 1.
Published in final edited form as: J Comp Neurol. 2021 Apr 29;529(11):3098–3111. doi: 10.1002/cne.25156

Synaptic inputs to broad thorny ganglion cells in macaque retina

Andrea S Bordt 1,2, Sara S Patterson 3, Rebecca J Girresch 4, Diego Perez 1, Luke Tseng 1, James R Anderson 5, Marcus A Mazzaferri 2, James A Kuchenbecker 2, Rodrigo Gonzales-Rojas 6, Ashley Roland 6, Charis Tang 6, Christian Puller 2,7, Alice Z Chuang 8, Judith Mosinger Ogilvie 4, Jay Neitz 2, David W Marshak 1
PMCID: PMC8193796  NIHMSID: NIHMS1692873  PMID: 33843050

Abstract

In primates, broad thorny retinal ganglion cells are highly sensitive to small, moving stimuli. They have tortuous, fine dendrites with many short, spine-like branches that occupy three contiguous strata in the middle of the inner plexiform layer. The neural circuits that generate their responses to moving stimuli are not well-understood, and that was the goal of this study. A connectome from central macaque retina was generated by serial block-face scanning electron microscopy, a broad thorny cell was reconstructed, and its synaptic inputs were analyzed. It received fewer than 2% of its inputs from both ON and OFF types of bipolar cells; the vast majority of its inputs were from amacrine cells. The presynaptic amacrine cells were reconstructed, and seven types were identified based on their characteristic morphology. Two types of narrow-field cells, knotty bistratified Type 1 and wavy multistratified Type 2, were identified. Two types of medium-field amacrine cells, ON starburst and spiny, were also presynaptic to the broad thorny cell. Three types of wide-field amacrine cells, wiry Type 2, stellate wavy, and semilunar Type 2, also made synapses onto the broad thorny cell. Physiological experiments using a macaque retinal preparation in vitro confirmed that broad thorny cells received robust excitatory input from both the ON and the OFF pathways. Given the paucity of bipolar cell inputs, it is likely that amacrine cells provided much of the excitatory input, in addition to inhibitory input.

Keywords: connectomics, electron microscopy, interneuron, motion sensitivity, primate, RRID:SCR_001622, RRID:SCR_003584, RRID:SCR_005986, RRID: SCR_017350, vision

1 |. INTRODUCTION

Movements of the body, head, and eyes generate moving stimuli on the retina, as do moving objects. Encoding this signal in the retina is the first step in visually guided reflexes, perception, and behavior (Wei, 2018). The neural circuits that generate motion sensitivity in ganglion cells, the projection neurons of the retina, have been studied extensively in mice and rabbits, and it is clear that input from amacrine cells, local circuit neurons of the inner retina that are generally inhibitory, plays a major role (Chen & Wei, 2018). The contributions of amacrine cells to neural circuits that generate motion sensitivity in primate retinal ganglion cells are not as well-characterized, however. The focus of this work was on one type of motion-sensitive ganglion cell in macaque retina, the broad thorny cell. Broad thorny cells project to koniocellular Layer 3, located between the magnocellular and parvocellular layers, of the lateral geniculate nucleus (LGN) in marmosets (Percival et al., 2013) and to the LGN in macaques (Dacey et al., 2003). They also project to the superior colliculus in macaques and marmosets (Kwan et al., 2019; Peterson & Dacey, 2000; Rodieck & Watanabe, 1993). Broad thorny cells respond to small, moving stimuli, firing vigorously when a small spot enters or leaves the receptive field center, but they are insensitive to movements in the background. These characteristics make them particularly well-suited to guide “catch-up” saccades during smooth pursuit eye movements (Puller et al., 2015).

There have been two previous studies of the synaptic inputs to broad thorny ganglion cells from bipolar cells, excitatory local circuit neurons that convey signals from the photoreceptors to the inner retina. Serial ultrathin sections of central macaque retina were analyzed by transmission electron microscopy (TEM), and neurons that may be broad thorny cells and their presynaptic bipolar cells were reconstructed from serial, vertical sections. The bipolar cells were diffuse, receiving input from several cones, and they were identified as types DB2 and DB3 (Calkins & Sterling, 2007). These both have OFF responses to light, depolarizing to decrements in light intensity (Puthussery et al., 2013, 2014), and together, they provided 30% of the synaptic input to the putative broad thorny cells. Retrogradely labeled broad thorny ganglion cells in marmoset retinas were studied using light microscopic immunolabeling with markers of synaptic transmission. Using this technique, broad thorny cells had essentially the same spatial density of synaptic inputs from bipolar cells as two other, more common retinal ganglion cell types, midget cells and parasol cells (Percival et al., 2011). One of the presynaptic cell types was identified as DB3a (Masri et al., 2016).

The amacrine cells presynaptic to broad thorny cells have not been identified, but this is essential because amacrine cells contribute to motion sensitivity, surround inhibition, and adaptation to contrast (Appleby & Manookin, 2019; Puller et al., 2015). Based on light microscopic immunolabeling in marmoset retina, two amacrine cells with medium-sized dendritic fields, spiny cells, and starburst cells have been proposed as candidates (Weltzien et al., 2014; Masri et al., 2019). Here, two presynaptic narrow-field amacrine cells, knotty bistratified Type 1 and wavy multistratified Type 2, were identified. In addition, inputs from wiry Type 2, semilunar Type 2, stellate wavy, and other wide-field amacrine cells were identified. However, the broad thorny cell received a much smaller proportion of input from bipolar cells than expected. Less than 2% of its input was from bipolar cells, mainly the ON type that respond to increments in light intensity.

Broad thorny ganglion cells receive robust excitatory and inhibitory synaptic inputs both at light increments and decrements (Puller et al., 2015), but the neural circuits that generate these responses are not well-understood. Electrophysiological experiments were designed to determine how the ON and OFF pathways contribute to the responses. When the ON pathway was blocked with pharmacological agents, both inhibitory and excitatory inputs at light onset were eliminated while robust responses were still elicited at light offset. The electrophysiological experiments rule out synaptic effects like disinhibition as the sole source of the responses to decrements in light intensity.

2 |. METHODS

2.1 |. Electron microscopy

Retinal tissue was obtained from a terminally anesthetized, adult male macaque (Macaca nemestrina) through the Tissue Distribution Program at the Washington National Primate Center. All procedures were approved by the Institutional Animal Care and Use Committee at the University of Washington. Central retinal tissue was processed for serial block-face scanning electron microscopy (SBFSEM) as previously described (Patterson et al., 2019). Briefly, a 1 × 1 mm square was fixed in glutaraldehyde, stained en bloc with osmium ferrocyanide, uranyl acetate and lead aspartate, and then embedded in epoxy resin. The selected area, ~2 mm temporal to the center of the fovea, was particularly well-suited for connectomic analysis because the neurons were small, but even relatively rare cell types were represented. The images were acquired using a Zeiss Sigma VP field emission scanning electron microscope equipped with a 3View system (Gatan, Inc.).

2.2 |. Connectomic analysis

A macaque retinal volume, sectioned in the horizontal plane and acquired at a resolution of 7.5 nm/pixel, was studied. It contained 937 70 nm sections, spanning from the ganglion cell layer (GCL) to the inner nuclear layer (INL). This connectome was also used in a recent study of ON parasol ganglion cells (Patterson, Bordt, et al., 2020). Image registration was performed using Nornir (http://nornir.github.io RRID:SCR_003584), and the image tiles were reassembled into cohesive digital volumes and hosted on a 24-core server at the University of Washington.

The serial EM volume was annotated using the web-based, multiuser Viking software described previously (Anderson et al., 2011; http://connectomes.utah.edu RRID:SCR_005986.) Briefly, profiles of processes were annotated by placing circular discs with the same diameter at their centers of mass and linking them to annotations on adjacent sections. Synaptic densities were annotated with lines and linked to the neurons in which they were located. Neurons and other structures, such as synaptic densities, were numbered consecutively. The boundary between the INL and the inner plexiform layer (IPL) was designated as 0% and the IPL-GCL boundary as 100% depth.

The major cell types were identified using ultrastructural criteria (Dowling & Boycott, 1966; Tsukamoto & Omi, 2015, 2016). Axon terminals of bipolar cells contained numerous synaptic vesicles and synaptic ribbons. Because the contrast of the synaptic ribbons in this connectome is not as high as in images from TEM, we confirmed the identity of the presynaptic bipolar cells by reconstructing the axon terminal and, wherever possible, the soma and primary dendrite. Axons and dendrites of amacrine cells contained fewer synaptic vesicles, and they were typically clustered at synapses. One broad thorny ganglion cell and the local circuit neurons providing its input were annotated. Whenever possible, the presynaptic neurons were identified by their morphology.

2.3 |. Anatomical data analysis

Data analysis and 3D rendering were performed using an open-source Matlab (Mathworks, RRID: SCR_001622) program https://github.com/neitzlab/sbfsem-tools RRID: SCR_017350. The image rendering was performed using the RenderApp function (Bordt et al., 2019). Using the synapseSphere function, synapses were rendered as unit spheres centered at each synapse annotation’s X, Y, and Z coordinates then scaled to optimize visibility. Processes of amacrine cells, axons of broad thorny cells, and axon terminals of bipolar cells were analyzed using the DendriteDiameter and IPLDepth functions (Patterson, Bordt, et al., 2020). The diameters of individual amacrine cell dendrites were computed with the single Dendrite Diameter function in SBFSEM-tools, described here for the first time. The annotations associated with a neuron and the links between the annotations were represented as the nodes and edges of a graph. The annotations of an individual branch were obtained by calculating the shortest path along the graph between two user-specified annotations at either end of the branch. Statistics were then calculated from the diameters of all annotations associated with the branch. Four representative distal dendrites from identified wide-field amacrine cells were analyzed.

2.4 |. Statistical analysis

The processes of a subset of wide field amacrine cells presynaptic the broad thorny cell were assigned to groups using the clustering algorithm k-means (Maechler et al., 2013). The cytoplasmic electron density, light or intermediate, was converted to a numeric value, 1 or 2, respectively. Four features, IPL depth, process diameter (mean and standard deviation), and cytoplasmic electron density, were standardized for comparability. First, the number of clusters was determined by a plot of the within groups sum of squares by number of clusters extracted, and the optimal number of clusters was identified by a plateau in the plot. An analysis of variance (ANOVA) based on a one-way ANOVA with Dunnett multiple comparison or a two-sample t-test was used to compare each feature among groups. A p-value <.05 was considered statistically significant. Statistical calculations were performed using the Cluster package in R version 3.6.0 (R Project for Statistical Computing, RRID:SCR_001905).

2.5 |. Electrophysiology

Eyes were dissected from terminally anesthetized adult female macaques (M. nemestrina), obtained through the Tissue Distribution Program of the National Primate Research Center at the University of Washington. All procedures were approved by the University of - Washington Institutional Animal Care and Use Committee. Small pieces (2–4 mm on a side) of peripheral retina (>5 mm away from the fovea) were dissected away from the supporting layers under infrared light in warm (~32°C) carboxygenated Ames’ medium (Sigma-Aldrich). Retinal pieces were mounted photoreceptor-side down onto a polylysine-coated glass slip and transferred to the recording chamber, where the tissue was continuously superfused with warm Ames medium.

Broad thorny ganglion cells were identified based on their characteristic soma shape and ON-OFF light response pattern (Puller et al., 2015). Light-evoked spikes were recorded in cell-attached mode with glass pipettes filled with Ames medium. Light-evoked excitatory or inhibitory ganglion cell synaptic input was characterized at holding potentials of −60 or 10–30 mV, respectively, using whole-cell voltage-clamp recordings. The pipettes were filled with a solution containing (in mM) 105 CsCH3SO3, 10 TEA-Cl, 20 HEPES, 10 EGTA, 2 QX-314, 5 Mg-ATP, and .5 Tris-GTP (pH ~ 7.3 with CsOH, ~280 mOsm). Series resistance for voltage-clamp recordings was compensated by 50% at the time of recording and the remaining 50% offline. The membrane potential was corrected offline for the ~10 mV liquid junction potential between the recording solution and the extracellular medium. The pipette solution also contained .1% Lucifer yellow CH (Sigma-Aldrich) to reveal cellular morphology, confirming the cell type.

Light responses were recorded in control conditions and while signal transmission from cones to ON bipolar cells was blocked via superfusion of the tissue with a mixture of the mGluR antagonist LY341495 (7.5 μM) and agonist L-AP4 (5 μM) in Ames (Ala-Laurila et al., 2011).

Stimuli from an oLED monitor (eMagin Corporation) were focused on the photoreceptors through a condenser lens. The monitor illuminated a circular area (~1 mm in diameter) centered on the recorded cell’s soma (vertical refresh, 60 Hz). All recordings were performed at a background in the photopic regime (quantal catch in R*/cone/sec: L/M-cone, ~13 × 103). Signals were sampled at 10 kHz with an ITC-18 analog-digital board (HEKA Instruments), amplified with a Multi-clamp 700B amplifier (Molecular Devices), and Bessel filtered at 3 kHz. Data analyses were performed in MATLAB (MathWorks, RRID: SCR_001622).

The stimulus was either a black spot centered on the soma (diameter, 288 μm) and presented for 1 s, or a 2 Hz square-wave modulated spot (diameter, 315 μm, 100% Michelson contrast), all presented at a photopic background (see Puller et al., 2015 for details). Measurements of the OFF-response kinetics before and after drug application were taken by analyzing synaptic currents averaged across five cycles of the square-wave stimulus.

2.6 |. Figures

Figures were prepared using Adobe Photoshop CS6 and SBFSEM-tools. Whenever possible, the color palette was selected so that the figures could be interpreted by individuals with all of the common forms of color blindness http://mkweb.bcgsc.ca/colorblind/. The code and data used to generate the figures in this study will be made available upon request.

3 |. RESULTS

The morphology and stratification pattern indicated that cell 103 in the temporal connectome was the broad thorny type (Dacey et al., 2003; Puller et al., 2015; Yamada et al., 2005). The dendritic arbor was asymmetric, and the dendrites were tortuous, branched frequently, and had numerous spine-like terminals. The dendrites terminated in a broad band extending from 30% to 80% of the IPL depth, with a peak at 60% (Figure 1). The axon was ~.2 μm in diameter (median = .209 μm) with varicosities .5 to .6 μm in diameter.

FIGURE 1.

FIGURE 1

Open in a new tab

Broad thorny ganglion cell 103 and the approximate boundaries of the connectome (white) are shown as they would appear in a flat mount preparation viewed from the scleral side of the retina. The cell is colored according to depth in the inner plexiform layer (IPL), with red as the inner nuclear layer and violet as the ganglion cell layer; the dendrites stratify broadly in the three middle strata. Some dendrites on the left were truncated when they reached the edge of the connectome. Note the varicose axon at the lower left (arrowhead). Scale bar = 100 μm [Color figure can be viewed at wileyonlinelibrary.com]

Broad thorny ganglion cell 103 received 1094 synapses, and bipolar cells provided only 1.4% (15/1094) of the inputs (Figure 2). The remaining 98.6% of the synapses (1079/1094) were from amacrine cells (Tables S1 and S2). Most of these synapses, 82% (887/1079), came from narrow-field or medium-field amacrine cells. A motif seen frequently was two or three of these amacrine cell processes converging onto a relatively thin ganglion cell dendrite (Figure 3). Axons and dendrites of wide-field amacrine cells, whose processes were unbranched, straight and 100 μm or longer, also provided a considerable amount of input to the broad thorny ganglion cell. They provided 18% (192/1079) of its amacrine cell input, and some of them were also presynaptic to the ON parasol cells studied previously (Patterson, Bordt, et al., 2020). The distribution of synapses onto broad thorny cell 103 is illustrated in Figure 4.

FIGURE 2.

FIGURE 2

Open in a new tab

Two bipolar cell axon terminals (green) make ribbon synapses (arrowheads) onto dendrites of broad thorny ganglion cell 103 (yellow). The synapses are shown at low (a) and somewhat higher (b) magnification. In this and in all subsequent figures, the markers are located in the presynaptic cells. Synapses like these provided less than 2% of the input to this ganglion cell. Scale bars = 500 nm [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 3.

FIGURE 3

Open in a new tab

It was common to observe two (a) or three (b) amacrine cell dendrites (orange) converging onto spines of ganglion cell 103 (yellow) and making synapses (arrowheads). Scale bar = 500 nm [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 4.

FIGURE 4

Open in a new tab

(a) Distribution of synapses from bipolar cell axons (green) onto broad thorny ganglion cell 103. Inputs from ON bipolar cells (light green) are more numerous than those from OFF bipolar cells (dark green). Note that only 15 of more than 1000 synapses originated from bipolar cells and that the distribution appeared to be random. (b) Inputs from amacrine cell axons and dendrites (red) were far more common. The box indicates the area shown at higher magnification in panel (c) Note that there are four clusters of synapses (arrowheads) onto cell 103. Inputs from wide-field amacrine cells are shown in panel (d). These also appear to be randomly distributed. Panels a, b, and d scale bars = 50 μm; panel c scale bar = 5 μm [Color figure can be viewed at wileyonlinelibrary.com]

3.1 |. Presynaptic bipolar cells

Fourteen bipolar cells made one synapse each, and one bipolar cell made two synapses onto broad thorny ganglion cell 103 (Figure 4). Based on the depth of their axon terminals in the IPL, 12 of these were expected to have ON responses to light and three were expected to have OFF responses (Figures 5 and 6). The ON cells terminated at 60% (30586, 39248, 41007, 41946, 41953, 43602, 43616), 70% (40761), 80% (42569), or 90% (40029, 42040) of the depth in the IPL. The presynaptic OFF bipolar cells terminated at IPL depths of 40% (41145, 42721), 30% (43119), and 20% (43128).

FIGURE 5.

FIGURE 5

Open in a new tab

Bipolar cell inputs to broad thorny ganglion cell 103 as they would appear in a flat mount preparation. Thirteen bipolar cells made one synapse each, and one, a giant bipolar cell, made two synapses. The depth code in this and subsequent figures is the same as in Figure 1. Scale bar = 100 μm [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 6.

FIGURE 6

Open in a new tab

Representative presynaptic bipolar cells as they would appear in vertical sections. Synapses onto cell 103 are indicated with white circles. (a) Diffuse bipolar cell 40761 ramifying at 60–70% depth. (b) Diffuse bipolar cell 41145 ramifying at 35–45%. (c) Diffuse bipolar cell 42040 ramifying at 90–100% depth, (d). Diffuse bipolar cell 42569 ramifying at 70–80%, (e) giant bipolar cell 30586 axon terminal ramifying at 50–60% depth. Scale bar = 10 μm [Color figure can be viewed at wileyonlinelibrary.com]

3.2 |. Presynaptic amacrine cells

Two types of narrow-field amacrine cells presynaptic to broad thorny cell 103 were identified morphologically. Cells 43016, 38098, 42714 and 43233 had somas in the INL and small dendritic fields, ~35 μm in diameter (Figure 7). They had varicose dendrites ramifying broadly in the center of the IPL, with peaks in density at 20–30% and 40–60% of the IPL depth. These resembled the knotty bistratified Type 1 cells of macaque retina (Mariani, 1990) and the A4 cells of human retina (Kolb et al., 1992). They also resembled the cells that contain immunoreactive vesicular glutamate transporter 3 (vGluT3) in baboon retina (Marshak et al., 2015). Each of these three knotty bistratified Type 1 cells made one synapse onto broad thorny cell 103.

FIGURE 7.

FIGURE 7

Open in a new tab

Knotty bistratified Type 1 amacrine cell 43016 was one type of narrow-field amacrine cell presynaptic to broad thorny ganglion cell 103 (white circle). It is shown as it would appear in flat mount and in a vertical section. Note that the soma is in the inner nuclear layer and that the dendrites ramify in two distinct strata near the center of the inner plexiform layer (IPL). The inset color scheme in this and subsequent figures is the same as in Figure 3. Arrowheads in the electron micrographs represent synaptic specializations in this and subsequent figures. Scale bars = 10 μm; inset scale bar = 500 nm [Color figure can be viewed at wileyonlinelibrary.com]

Three presynaptic amacrine cells were identified morphologically as wavy multistratified Type 2 (Mariani, 1990). Amacrine cell 42338 made three synapses onto broad thorny cell 103, including one onto a spine where two other amacrine cells also made synapses. Its soma was located in the INL, and its dendritic arbor was larger than those of the knotty bistratified cells. The exact dimensions were uncertain because the soma was located near the edge of the connectome, and several of its dendrites were truncated. Its dendrites were relatively tortuous and ramified broadly around the center of the IPL, with a peak at 50% depth. The morphology of cell 43446 was very similar, and its dendritic arbor was also truncated; it made three synapses onto broad thorny cell 103. Cell 40879 made five synapses onto broad thorny ganglion cell 103, including two onto spines alongside -other synapses. It was morphologically very similar to the other two wavy multistratified cells and had a more completely annotated dendritic arbor, whose diameter was ~60 μm (Figure 8).

FIGURE 8.

FIGURE 8

Open in a new tab

Wavy multistratified Type 2 amacrine cell 40879 was another type of narrow-field amacrine cell presynaptic to broad thorny ganglion cell 103. It is shown as it would appear in flat mount and in a vertical section. It made five synapses onto cell 103. The soma was in the inner nuclear layer, and the dendritic arbor was larger in diameter than that of the knotty bistratified cell illustrated in Figure 7. Scale bars = 10 μm; inset scale bar = 500 nm [Color figure can be viewed at wileyonlinelibrary.com]

Two types of medium-field amacrine cells presynaptic to the broad thorny ganglion cells were identified (Grünert & Martin, 2020). Cell 40654 was identified as the spiny amacrine cell type (Kolb et al., 1992; Mariani, 1990). It had a dendritic field ~150 μm in diameter, and its soma was located in the GCL. Its dendrites were tortuous, branched frequently, and had small spines (Figure 9). The dendrites ramified broadly in the center of the IPL, with a peak density at 50% of the IPL depth. Cell 40654 made five synapses onto broad thorny cell 103.

FIGURE 9.

FIGURE 9

Open in a new tab

Spiny amacrine cell 40654 was a medium-field amacrine cell that made five synapses (white circles) onto broad thorny ganglion cell 103. Note that the soma is in the inner nuclear layer and the dendrites have small spines and mainly ramify in the center of the inner plexiform layer (IPL). Scale bar = 10 μm; inset scale bar = 500 nm [Color figure can be viewed at wileyonlinelibrary.com]

Amacrine cell 3111 was classified as a medium-field amacrine cell based on previous descriptions of starburst cells in primate retinas (Kolb et al., 1992; Mariani, 1990; Rodieck, 1989). It had a small soma in the GCL, and its proximal dendrites were very thin, tortuous and relatively electron dense. The distal dendrites were enlarged, and the majority were found at a depth of 70% in the IPL. It made one synapse onto cell 103 (Figure 10).

FIGURE 10.

FIGURE 10

Open in a new tab

ON starburst amacrine cell 3111 was another medium-field amacrine cell presynaptic to ganglion cell 103. Although it was not completely reconstructed, it was classified based on its thin, tortuous dendrites ramifying at 65–70% depth in the inner plexiform layer (IPL) that were enlarged at their distal ends. Note that the soma is located in the ganglion cell layer. Scale bar = 10 μm; inset scale bar = 500 nm [Color figure can be viewed at wileyonlinelibrary.com]

Three types of wide-field amacrine cells presynaptic to broad thorny cell 103 were identified. The first type identified were wiry Type 2, a common type of wide-field amacrine cell in primate retina (Kolb et al., 1992; Mariani, 1990). Two of them (4651 and 182) also made synapses onto ON parasol cells (Patterson, Bordt, et al., 2020), and one (41444) is described here for the first time. Wiry Type 2 amacrine cells had ovoid somas that were located in the GCL. They had one or two short primary dendrites, relatively short secondary dendrites and tertiary dendrites each that bifurcated, giving rise to long, relatively straight distal dendrites that did not branch again. Their distal dendrites had a small diameter (mean = .352 μm, SD = .11 μm) and ramified in a narrow stratum of the IPL at 60% depth. Together, the three identified wiry amacrine Type 2 cells made 11 synapses onto broad thorny cell 103 (Figure 11).

FIGURE 11.

FIGURE 11

Open in a new tab

Wiry Type 2 amacrine cell 182 was a wide-field amacrine cell that made four synapses (white circles) onto broad thorny ganglion cell 103. Note that the soma was located in the ganglion cell layer and that the long, unbranched distal dendrites ramified at 60% depth in the inner plexiform layer (IPL). Scale bar = 10 μm; inset scale bar = 500 nm [Color figure can be viewed at wileyonlinelibrary.com]

Two other types of wide-field amacrine cells presynaptic to broad thorny cell 103 were identified. Cell 4781, which made two synapses onto cell 103, was a semilunar Type 2 cell that also made synapses onto ON parasol cells (Patterson, Bordt, et al., 2020). It had a relatively large soma and an extensive dendritic arbor centered at 59% IPL depth. Its dendrites had a larger diameter (mean = .66 μm, SD = .19 μm) and branched at larger angles than those of wiry cells. Thinner axons originated from some of the dendrites and the soma (Figure 12).

FIGURE 12.

FIGURE 12

Open in a new tab

Semilunar Type 2 amacrine cell 4781 was another wide-field amacrine cell presynaptic to broad thorny ganglion cell 103. The soma was located in the inner plexiform layer (IPL) and the dendrites were larger in diameter than those of the wiry Type 2 cells illustrated in Figure 11. Thinner axons (arrowheads) arose from both the soma and the distal tips of the dendrites. Scale bar = 10 μm; inset scale bar = 500 nm [Color figure can be viewed at wileyonlinelibrary.com]

The third type of identified presynaptic wide-field amacrine cell, 4308, was similar in some respects to semilunar Type 2 cell 4781, but the dendrites had a smaller diameter (mean = .61 μm, SD = .12 μm), and the cytoplasm was more electron lucent. One slender, varicose process originating from a secondary dendrite appeared to be an axon, and similar processes emerged from the tips of two distal dendrites. The sparse dendritic arbor was centered at 26% depth in the IPL. Cell 4308 made three synapses onto broad thorny cell 103 (Figure 13). Of the morphological types described previously in macaque and human retina, this amacrine cell most closely resembled the stellate wavy cell (Kolb et al., 1992; Mariani, 1990).

FIGURE 13.

FIGURE 13

Open in a new tab

Stellate wavy amacrine cell 4308 was a third type of wide-field amacrine cell presynaptic to ganglion cell 103. It made three synapses (white circles). The soma was located in the ganglion cell layer, and the dendrites were intermediate in diameter between those of wiry and semilunar cells. Thinner axons (arrowheads) arose from secondary and distal dendrites, and both types of processes ramified between 20% and 30% depth in the inner plexiform layer (IPL), on average. Scale bar = 10 μm; inset scale bar = 500 nm [Color figure can be viewed at wileyonlinelibrary.com]

Most types of wide-field cells presynaptic to broad thorny cell 103 could not be followed to their somas, however. The processes of a subset of wide-field amacrine cells without somas in the volume (n = 98) were analyzed using the IPLDepth and DendriteDiameter functions in SBFSEM-tools. They were also classified as light or intermediate in cytoplasmic electron density. Based on these parameters, dendrites were grouped using a cluster analysis. A plot of the within groups sum of squares by number of groups was used to determine the appropriate number of clusters (not illustrated), and six groups were identified.

Three groups of processes with light cytoplasmic electron density were identified. The first light group (1) had a mean diameter of .52 μm (n = 11, SD = .087 μm) and ramified at 50% depth in the IPL, on average. The second light group (2) had a smaller diameter (n = 15, mean = .425 μm, SD = .028 μm), and they ramified at 60.5% depth in the IPL. The differences between the two groups were significant at the .05% level for both stratification depth and diameter. The dendrites in the third light group (6) were the most common and had a slightly larger diameter than those in the second (n = 28, mean = .48 μm, SD = .068 μm) and ramified at 29% of the IPL depth.

There were also three groups of processes with darker cytoplasmic electron density. The first group (3) had a mean diameter of .46 μm (n = 14, SD = .061) and ramified at 28% IPL depth. Another group of processes with intermediate cytoplasmic electron density (4) was clearly different, having a larger average diameter (n = 8, mean = .56 μm, SD = .086 μm) and ramifying at 44% depth in the IPL, on average. The third group (5) was the most common and had the smallest diameter (n = 22, mean = .45 μm, SD = .073 μm), and they ramified at a depth of 62%, on average. The wide-field amacrine cell processes in Group 5 were morphologically similar to the dendrites of the identified wiry Type 2 cells. Because wiry Type 2 somas have a high spatial density and their long dendrites form a very dense plexus (Majumdar et al., 2008), they were likely to be the source.

Although the synaptic inputs and outputs of the amacrine cells presynaptic to broad thorny cell 103 were not analyzed completely, it was apparent that they also made synaptic contacts with one another and with parasol cells identified previously (Patterson, Bordt, et al., 2020). Wiry Type 2 cell 182 made nine synapses onto four ON parasol cells and one onto another wiry amacrine cell, 4315. Wiry Type 2 amacrine cell made four synapses onto two ON parasol cells. Wiry Type 2 amacrine cell 41444 made a synapse onto wiry Type 2 amacrine cell 182. If the processes of the wiry Type 2 amacrine cells that were tentatively identified were included in this analysis, there would be many more examples. Semilunar Type 2 cell 4781 made 13 synapses onto three ON parasol cells.

The same pattern was observed with smaller amacrine cells. One of the knotty bistratified Type 1 cells, 43016, made two synapses onto an ON parasol cell. A second knotty bistratified Type 1 cell, 38098, made five synapses onto the same ON parasol cell and also one onto wiry Type 2 cell 4651. A third knotty bistratified Type 1 cell, 42714, made a synapse onto OFF parasol cell 61. Taken together, these findings suggest that the same interconnected amacrine cells contribute to the neural circuits providing input to at least two types of motion-sensitive retinal ganglion cells.

3.3 |. Electrophysiology

Excitatory and inhibitory synaptic currents were recorded from three broad thorny ganglion cells under control conditions and during blockade of the ON pathway to gain insight into possible effects of ON pathway signals on the OFF responses of the cells (Figure 14). Synaptic currents in control conditions were consistent with a previous report (Puller et al., 2015). During drug application, excitatory and inhibitory inputs persisted at light decrements but were abolished at light increments. We analyzed the OFF responses to a square-wave modulated spot to obtain further insight into the effects of ON pathway blockade on the response kinetics of the cells (n = 3, all measurements provided as mean ± SEM). The peak times relative to the stimulus transition time were only slightly shifted. Control excitation peaked at 117 ± 11 ms and inhibition at 55 ± 4 ms. When the drugs were applied, excitation peaked at 106 ± 7 ms, and inhibition at 59 ± 6 ms. The effect of the drugs on the decay time constant τ (ms) appeared smaller for the excitation (control 187 ± 28 ms, drug 192 ± 19 ms) than for the inhibition (control 388 ± 109 ms, drug 198 ± 46 ms).

FIGURE 14.

FIGURE 14

Open in a new tab

OFF responses of broad thorny ganglion cells do not depend on the ON pathway. (a) Whole-cell voltage-clamp responses of a broad thorny cell in control conditions at two different holding potentials to reveal excitatory and inhibitory synaptic currents. The stimulus is indicated above the recording traces. (b) Recording of the cell as above but during application of L-AP4 (5 μM) and LY341495 (7.5 μM) to block the ON pathway [Color figure can be viewed at wileyonlinelibrary.com]

4 |. DISCUSSION

The major finding from this study was that broad thorny ganglion cells from central macaque retina received a very small percentage of their inputs from bipolar cells, smaller than any retinal ganglion cell described previously. This result is not attributable to under-sampling of bipolar cells. ON parasol cells in the same volume received a much larger proportion of their input from bipolar cells (Patterson, Bordt, et al., 2020). Typically, light responses of retinal ganglion cells are driven by bipolar cell input, and amacrine cells are usually thought to have more of a modulatory role. However, this is clearly not the case for broad thorny cells. They have highly specific responses compared with parasol ganglion cells, which respond to any moving textures (Puller et al., 2015). It may be that a preponderance of amacrine cell input is required to produce more nuanced response properties.

Broad thorny cells of macaque retina were different in this respect from broad thorny ganglion cells of marmoset retina, which received approximately the same proportion of inputs from bipolar cells as other, more common, types of retinal ganglion cells in a light microscopic study (Percival et al., 2011). It is uncertain whether this is attributable to methodological differences between their study and ours or to a difference between the two species. Old World and New World monkeys may be different in this respect, having diverged 40 million years ago. An alternative explanation is that the resolution of conventional light microscopy may be insufficient to distinguish synapses directly onto a labeled ganglion cell dendrite from those onto other processes nearby.

The percentage of bipolar cell input to the broad thorny cell was also different from the value reported for garland cells from macaque parafovea, which, in retrospect, were likely broad thorny cells. These received 30% of their input from bipolar cells, a value much lower than the percentages reported for midget cells or parasol cells at this eccentricity (Calkins & Sterling, 1996; Calkins & Sterling, 2007). Broad thorny cell 103 received less than 2% of its input from bipolar cells. Taken together, the two studies of broad thorny cells in macaques suggest that the proportion of bipolar cell input increases with proximity to the center of the fovea, but it remains lower than that of any other known type of primate retinal ganglion cell.

Robust excitatory inputs have been recorded from peripheral broad thorny ganglion cells using voltage clamp (Puller et al., 2015). The experiments reported here indicate that both the excitation and the inhibition recorded at the onset of light stimuli are conveyed via the ON pathway. Likewise, ON-OFF ganglion cells in marmoset retina receive excitatory input from both ON and OFF pathways (Protti et al., 2014). The changes in the OFF response kinetics after blockade of the ON pathway suggest the possibility that a cessation of crossover inhibition from the ON pathway contributes to the OFF excitatory responses in broad thorny cells, but the ON pathway is clearly not the sole source for generating responses at light decrements. Further experiments, including a greater number of cells and additional pharmacological agents, would be required to gain additional insight into the interactions of ON and OFF pathways in the neural circuit providing input to broad thorny ganglion cells.

Nevertheless, the results indicate that there are direct, light-stimulated excitatory synaptic inputs via both the ON and OFF pathways. Because there are so few inputs from bipolar cells to the thorny amacrine cell, amacrine cells are likely to provide most of the excitatory input. An additional synapse in the input pathway would account for the finding that light responses of broad thorny cells have longer latencies than other types of third-order neurons ramifying in the center of the IPL (Puller et al., 2015). This excitatory amacrine cell input is expected to be particularly important in the OFF pathway because the majority of the bipolar cell inputs to broad thorny ganglion cells came from ON bipolar cells.

The amacrine cells presynaptic to broad thorny cells described in this study are likely sources of the excitatory input. Acetylcholine from the ON starburst cells acting at nicotinic receptors and glutamate from the knotty bistratified Type 1 cells acting at ionotropic receptors are both excitatory (Marshak, 2016). Glycine acting at certain subtypes of NMDA receptors might also be excitatory (Otsu et al., 2019). If macaque retinas have the amacrine cells described recently in human retina that contain neither glycine nor GABA, these would be another possible source of excitatory input (Yan et al., 2020).

One type of narrow-field amacrine cell presynaptic to broad thorny ganglion cell 103 was identified morphologically as knotty bistratified Type 1 (Mariani, 1990). It also closely resembled the A4 type from human retina (Kolb et al., 1992). In baboon retina, these cells contain immunoreactive vGluT3, and they frequently make synapses onto retinal ganglion cell dendrites (Marshak et al., 2015). In mice, they make both glycinergic inhibitory synapses and glutamatergic excitatory synapses (Marshak, 2016). A second type of narrow-field amacrine cell presynaptic to broad thorny ganglion cell 103 was identified morphologically as the wavy multistratified Type 2 (Mariani, 1990). Two of these cells contributed to the pairs of synapses found on the spines of broad thorny cell 103. These synapses from wavy multistratified cells might provide crossover inhibition to broad thorny cells (Werblin, 2010).

Broad thorny cell 103 also received a single synapse from an ON starburst amacrine cell, which was identified by its dendritic morphology and stratification depth in the IPL (Kolb et al., 1992; Mariani, 1990; Rodieck, 1989; Rodieck & Marshak, 1992). It is possible that it also receives cholinergic input at unspecialized sites, as reported recently in the mouse retina (Sethuramanujam et al., 2021).

Another medium-field amacrine cell presynaptic to broad thorny cell 103 resembled the stratified amacrine cells first described using the Golgi method in macaque central retina (Boycott & Dowling, 1969). These were named spiny in a later study of macaque retina using the same technique, and roughly equal numbers of these cells had somas in the GCL, like the spiny amacrine cell presynaptic to broad thorny cell 103 (Mariani, 1990). The first descriptions of the amacrine cells containing immunoreactive vasoactive intestinal peptide (VIP) identified those cells as spiny (Lammerding-Kӧppel et al., 1991; Marshak, 1989). However, more recently, the VIP-positive cells have been classified as the stellate varicose type, instead (Grünert & Martin, 2020). The spiny amacrine cells have a very high spatial density in marmoset retina, where they can be labeled using an antibody to secretagogin, and they constitute a single population, regardless of the positions of their somas (Weltzien et al., 2014).

Wide-field amacrine cells provided a major input to the broad thorny cell 103. These were identified as wiry Type 2 cells, semilunar Type 2 cells, and stellate wavy cells. Their inputs may account for two findings reported in an earlier electrophysiological study of broad thorny cells in macaque retina. Postsynaptic inhibition of broad thorny cells increases with the stimulus size, reaching a maximum at sizes larger than the dendritic field diameter, and broad thorny cell light responses are suppressed by global stimulus motion (Puller et al., 2015).

Semilunar Type 2 cells are coupled to ON parasol cells by gap junctions (Greschner et al., 2016; Jacoby et al., 1996), and they were also presynaptic to broad thorny cell 103. This pathway might enable ON parasol cells to inhibit broad thorny cells. Another prediction is that the axons of broad thorny ganglion cells should have relatively slow conduction velocities because of their narrow diameter. In this respect, they would resemble the “rarely encountered” ganglion cells of macaque retina (Schiller & Malpeli, 1977).

The amacrine cells presynaptic to broad thorny ganglion cell 103 were not specialized to provide inputs to this type of ganglion cell; many also provided input to parasol cells and to each other. In this respect, they were different from the highly specialized A12 amacrine cells that convey input from short wavelength-sensitive cone bipolar cells to intrinsically-photosensitive ganglion cells in macaque retina (Patterson, Kuchenbecker, et al., 2020) and other types of well-characterized primate amacrine cells that direct most of their output to a single type of postsynaptic cell (Grünert & Martin, 2020). The dendrites of broad thorny cell 103 extended to the boundaries of the connectome, stratified broadly and received over 1000 synapses from amacrine cells. But the identified presynaptic cells provided, at most, six synapses to cell 103. The same trend was seen with the unidentified amacrine cells. Taken together, these findings suggest that this ensemble of neurons contributes to the responses of many motion-sensitive ganglion cells.

Supplementary Material

tS2
tS1

ACKNOWLEDGMENTS

This research was supported by NIH grants EY027859, EY028927, EY002576, NS099578, EY007031, EY007125, P51-OD010425/ORID, P30-EY014800, and P30-EY001730, the German Research Foundation (DFG grant PU 469/2-1) and by Research to Prevent Blindness.

Funding information

Deutsche Forschungsgemeinschaft, Grant/Award Number: PU 469/2-1; National Institutes of Health, Grant/Award Numbers: EY002576, EY007031, EY007125, EY027859, EY028927, NS099578, P30-EY001730, P30-EY014800, P51-OD010425/ORID; Research to Prevent Blindness, Grant/Award Number: N/A

Footnotes

PEER REVIEW

The peer review history for this article is available at https://publons.com/publon/10.1002/cne.25156.

DATA AVAILABILITY STATEMENT

The serial EM volume was annotated using the web-based, multiuser Viking software described previously (Anderson et al., 2011; http://connectomes.utah.edu RRID:SCR_005986). The annotated volume is available in a read-only format. Data analysis and 3D rendering were performed using an open-source Matlab (Mathworks, RRID: SCR_001622) program https://github.com/neitzlab/sbfsem-tools RRID: SCR_017350. The code and data used to generate the figures in this study will be made available upon request.

SUPPORTING INFORMATION

Additional supporting information may be found online in the Supporting Information section at the end of this article.

REFERENCES

  1. Abbott CJ, Percival KA, Martin PR, & Grünert U (2012). Amacrine and bipolar inputs to midget and parasol ganglion cells in marmoset retina. Visual Neuroscience, 29(3), 157–168. 10.1017/S095252381200017X [DOI] [PubMed] [Google Scholar]
  2. Ala-Laurila P, Greschner M, Chichilnisky EJ, & Rieke F (2011). Cone photoreceptor contributions to noise and correlations in the retinal output. Nature Neuroscience, 14(10), 1309–1316. 10.1038/nn.2927 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Anderson JR, Mohammed S, Grimm B, Jones BW, Koshevoy P, Tasdizen T, Whitaker R, & Marc RE (2011). The Viking viewer for connectomics: Scalable multi-user annotation and summarization of large volume data sets. Journal of Microscopy, 241(1), 13–28. 10.1111/j.1365-2818.2010.03402.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Appleby TR, & Manookin MB (2019). Neural sensitization improves encoding fidelity in the primate retina. Nature Communications, 10(1), 4017. 10.1038/s41467-019-11734-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bordt AS, Perez D, Tseng L, Liu WS, Neitz J, Patterson SS, Famiglietti EV, & Marshak DW (2019). Synaptic inputs from identified bipolar and amacrine cells to a sparsely branched ganglion cell in rabbit retina. Visual Neuroscience, 36, E004. 10.1017/S0952523819000014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Boycott BB, & Dowling JE (1969). Organization of the primate retina: Light microscopy. Philosophical Transactions of the Royal Society of London, Series B Biological Sciences, 225, 109–184. [DOI] [PubMed] [Google Scholar]
  7. Calkins DJ, & Sterling P (1996). Absence of spectrally specific lateral inputs to midget ganglion cells in primate retina. Nature, 381(6583), 613–615. 10.1038/381613a0s [DOI] [PubMed] [Google Scholar]
  8. Calkins DJ, & Sterling P (2007). Microcircuitry for two types of achromatic ganglion cell in primate fovea. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 27(10), 2646–2653. 10.1523/JNEUROSCI.4739-06.2007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chen Q, & Wei W (2018). Stimulus-dependent engagement of neural mechanisms for reliable motion detection in the mouse retina. Journal of Neurophysiology, 120(3), 1153–1161. 10.1152/jn.00716.2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dacey DM, Peterson BB, Robinson FR, & Gamlin PD (2003). Fireworks in the primate retina: In vitro photodynamics reveals diverse LGN-projecting ganglion cell types. Neuron, 37(1), 15–27. 10.1016/s0896-6273(02)01143-1 [DOI] [PubMed] [Google Scholar]
  11. Dowling JE, & Boycott BB (1966). Organization of the primate retina: Electron microscopy. Proceedings of the Royal Society B, 166(1002), 80–111. 10.1098/rspb.1966.0086 [DOI] [PubMed] [Google Scholar]
  12. Greschner M, Heitman AK, Field GD, Li PH, Ahn D, Sher A, Litke AM, & Chichilnisky EJ (2016). Identification of a retinal circuit for recurrent suppression using indirect electrical imaging. Current Biology, 26(15), 1935–1942. 10.1016/j.cub.2016.05.051 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Grünert U, & Martin PR (2020). Cell types and cell circuits in human and non-human primate retina. Progress in Retinal and Eye Research, 100844. 10.1016/j.preteyeres.2020.100844 [DOI] [PubMed] [Google Scholar]
  14. Jacoby R, Stafford D, Kouyama N, & Marshak D (1996). Synaptic inputs to ON parasol ganglion cells in the primate retina. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 16 (24), 8041–8056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kolb H, Linberg KA, & Fisher SK (1992). Neurons of the human retina: A Golgi study. The Journal of Comparative Neurology, 318(2), 147–187. 10.1002/cne.903180204 [DOI] [PubMed] [Google Scholar]
  16. Kwan WC, Mundinano IC, de Souza MJ, Lee SC, Martin PR, Grünert U, & Bourne JA (2019). Unravelling the subcortical and retinal circuitry of the primate inferior pulvinar. The Journal of Comparative Neurology, 527(3), 558–576. [DOI] [PubMed] [Google Scholar]
  17. Lammerding-Kӧppel M, Thier P, & Koehler W (1991). Morphology and mosaics of VIP-like immunoreactive neurons in the retina of the rhesus monkey. Journal of Comparative Neurology, 312(2), 251–263. 10.1002/cne.903120208 [DOI] [PubMed] [Google Scholar]
  18. Maechler M, Rousseeuw P, Struyf A, Hubert M, & Hornik K (2013). Cluster: Cluster analysis basics and extensions. R package (Version 1.14.4). [Google Scholar]
  19. Majumdar S, Wässle H, Jusuf PR, & Haverkamp S (2008). Mirror-symmetrical populations of wide-field amacrine cells of the macaque monkey retina. The Journal of Comparative Neurology, 508(1), 13–27. 10.1002/cne.21666 [DOI] [PubMed] [Google Scholar]
  20. Mariani AP (1990). Amacrine cells of the rhesus monkey retina. The Journal of Comparative Neurology, 301(3), 382–400. 10.1002/cne.903010305 [DOI] [PubMed] [Google Scholar]
  21. Marshak DW (1989). Peptidergic neurons of the macaque monkey retina. Neuroscience Research Supplements, 10, S117–S130. 10.1016/0921-8696(89)90014-5s [DOI] [PubMed] [Google Scholar]
  22. Marshak DW (2016). A tale of two neurotransmitters. Visual Neuroscience, 33, E017. 10.1017/S0952523816000146 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Marshak DW, Chuang AZ, Dolino DM, Jacoby RA, Liu WS, Long YE, Sherman MB, Suh JM, Vila A, & Mills SL (2015). Synaptic connections of amacrine cells containing vesicular glutamate transporter 3 in baboon retinas. Visual Neuroscience, 32, E006. 10.1017/S0952523815000036 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Masri RA, Percival KA, Koizumi A, Martin PR, & Grünert U (2016). Connectivity between the OFF bipolar type DB3a and six types of ganglion cell in the marmoset retina. The Journal of Comparative Neurology, 524(9), 1839–1858. 10.1002/cne.23925 [DOI] [PubMed] [Google Scholar]
  25. Masri RA, Percival KA, Koizumi A, Martin PR, & Grünert U (2019). Survey of retinal ganglion cell morphology in marmoset. Journal of Comparative Neurology, 527(1), 236–258. 10.1002/cne.24157 [DOI] [PubMed] [Google Scholar]
  26. Otsu Y, Darcq E, Pietrajtis K, Mátyás F, Schwartz E, Bessaih T, Gerges A, Rousseau CV, Grand T, Dieudonné S, Paoletti P, Acsády L, Agulhon C, Kieffer BL, & Diana MA (2019). Control of aversion by glycine-gated GluN1/GluN3A NMDA receptors in the adult medial habenula. Science (New York, N.Y.), 366(6462), 250–254. 10.1126/science.aax1522 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Patterson SS, Bordt AS, Girresch RJ, Linehan CM, Bauss J, Yeo E, Perez D, Tseng L, Navuluri S, Harris NB, Matthews C, Anderson JR, Kuchenbecker JA, Manookin MB, Ogilvie JM, Neitz J, & Marshak DW (2020). Wide-field amacrine cell inputs to ON parasol ganglion cells in macaque retina. The Journal of Comparative Neurology, 528(9), 1588–1598. 10.1002/cne.24840 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Patterson SS, Kuchenbecker JA, Anderson JR, Bordt AS, Marshak DW, Neitz M, & Neitz J (2019). An S-cone circuit for edge detection in the primate retina. Scientific Reports, 9(1), 11913. 10.1038/s41598-019-48042-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Patterson SS, Kuchenbecker JA, Anderson JR, Neitz M, & Neitz J (2020). A color vision circuit for non-image-forming vision in the primate retina. Current Biology, 30(7), 1269–1274.e2. 10.1016/j.cub.2020.01.040 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Percival KA, Martin PR, & Grünert U (2011). Synaptic inputs to two types of koniocellular pathway ganglion cells in marmoset retina. The Journal of Comparative Neurology, 519(11), 2135–2153. 10.1002/cne.22586 [DOI] [PubMed] [Google Scholar]
  31. Percival KA, Martin PR, & Grünert U (2013). Organisation of koniocellular-projecting ganglion cells and diffuse bipolar cells in the primate fovea. The European Journal of Neuroscience, 37(7), 1072–1089. 10.1111/ejn.12117 [DOI] [PubMed] [Google Scholar]
  32. Peterson BB, & Dacey DM (2000). Morphology of wide-field bistratified and diffuse human retinal ganglion cells. Visual Neuroscience, 17(4), 567–578. 10.1017/s0952523800174073 [DOI] [PubMed] [Google Scholar]
  33. Protti DA, Di Marco S, Huang JY, Vonhoff CR, Nguyen V, & Solomon SG (2014). Inner retinal inhibition shapes the receptive field of retinal ganglion cells in primate. The Journal of Physiology, 592 (1), 49–65. 10.1113/jphysiol.2013.257352 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Puller C, Manookin MB, Neitz J, Rieke F, & Neitz M (2015). Broad thorny ganglion cells: A candidate for visual pursuit error signaling in the primate retina. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 35(13), 5397–5408. 10.1523/JNEUROSCI.4369-14.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Puthussery T, Percival KA, Venkataramani S, Gayet-Primo J, Grünert U, & Taylor WR (2014). Kainate receptors mediate synaptic input to transient and sustained OFF visual pathways in primate retina. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 34(22), 7611–7621. 10.1523/JNEUROSCI.4855-13.2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Puthussery T, Venkataramani S, Gayet-Primo J, Smith RG, & Taylor WR (2013). NaV1.1 channels in axon initial segments of bipolar cells augment input to magnocellular visual pathways in the primate retina. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 33(41), 16045–16059. 10.1523/JNEUROSCI.1249-13.2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Rodieck RW (1989). Starburst amacrine cells of the primate retina. Journal of Comparative Neurology, 285(1), 18–37. 10.1002/cne.902850104s [DOI] [PubMed] [Google Scholar]
  38. Rodieck RW, & Marshak DW (1992). Spatial density and distribution of choline acetyltransferase immunoreactive cells in human, macaque, and baboon retinas. The Journal of Comparative Neurology, 321(1), 46–64. 10.1002/cne.903210106 [DOI] [PubMed] [Google Scholar]
  39. Rodieck RW, & Watanabe M (1993). Survey of the morphology of macaque retinal ganglion cells that project to the pretectum, superior colliculus, and parvicellular laminae of the lateral geniculate nucleus. The Journal of Comparative Neurology, 338(2), 289–303. 10.1002/cne.903380211 [DOI] [PubMed] [Google Scholar]
  40. Schiller PH, & Malpeli JG (1977). Properties and tectal projections of monkey retinal ganglion cells. Journal of Neurophysiology, 40(2), 428–445. 10.1152/jn.1977.40.2.428 [DOI] [PubMed] [Google Scholar]
  41. Sethuramanujam S, Matsumoto A, de Rosenroll G, Murphy-Baum B, McIntosh JM, Jing M, Li Y, Berson D, Yonehara K, & Awatramani GB (2021). Rapid multidirected cholinergic transmission in the central nervous system. Nature Communications, 12, 1374. 10.1038/s41467-021-21680-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Tsukamoto Y, & Omi N (2015). OFF bipolar cells in macaque retina: Type-specific connectivity in the outer and inner synaptic layers. Frontiers in Neuroanatomy, 9, 122. 10.3389/fnana.2015.00122 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Tsukamoto Y, & Omi N (2016). ON bipolar cells in macaque retina: Type-specific synaptic connectivity with special reference to OFF counterparts. Frontiers in Neuroanatomy, 10, 104. 10.3389/fnana.2016.00104 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wei W (2018). Neural mechanisms of motion processing in the mammalian retina. Annual Review of Vision Science, 4, 165–192. 10.1146/annurev-vision-091517-034048 [DOI] [PubMed] [Google Scholar]
  45. Weltzien F, Dimarco S, Protti DA, Daraio T, Martin PR, & Grünert U (2014). Characterization of secretagogin-immunoreactive amacrine cells in marmoset retina. The Journal of Comparative Neurology, 522(2), 435–455. 10.1002/cne.23420 [DOI] [PubMed] [Google Scholar]
  46. Werblin FS (2010). Six different roles for crossover inhibition in the retina: Correcting the nonlinearities of synaptic transmission. Visual Neuroscience, 27(1–2), 1–8. 10.1017/S0952523810000076 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Yamada ES, Bordt AS, & Marshak DW (2005). Wide-field ganglion cells in macaque retinas. Visual Neuroscience, 22(4), 383–393. 10.1017/S095252380522401X [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Yan W, Peng Y-R, van Zyl T, Regev A, Shekhar K, Juric D, & Sanes JR (2020). Cell atlas of the human fovea and peripheral retina. Scientific Reports, 10(1), 9802. 10.1038/s41598-020-66092-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

tS2
NIHMS1692873-supplement-tS2.docx (12.3KB, docx)
tS1
NIHMS1692873-supplement-tS1.docx (37.2KB, docx)

RESOURCES