Regional variation in the organization and connectivity of the first synapse in the primate night vision pathway

Abstract

Sensitivity of primate daylight vision varies across the visual field. This is attributed to regional variations in cone photoreceptor density and synaptic connectivity of the underlying circuitry. In contrast, we have limited understanding of how synapse organization of the primate night vision pathway changes across space. Using serial electron microscopy, we reconstructed the first synapse of the night vision pathway between rod photoreceptors and second-order neurons, at multiple locations from the central part of the primate retina, fovea, to the periphery. We find that most facets of the rod synapse connectivity vary across retinal regions. However, rod synaptic divergence and convergence patterns do not change in the same manner across locations. Moreover, patterns of rod synapse organization are tightly correlated with photoreceptor density. Such regional heterogeneities revise the connectivity diagram of the primate rod synapse which will shape synapse function and sensitivity of the night vision pathway across visual space.

Keywords: cellular neuroscience; sensory neuroscience.

Graphical abstract

Figure 1

Rod synapse organization across locations in primate macaque retina (A) Schematic representation of the rod to rod bipolar cell (RBC) ribbon synapse in the mammalian retina. Each RBC pools inputs from multiple rod photoreceptors. The signal from the rod photoreceptors is modified by input from horizontal cell (HC) processes. Inset: Schematic of a rod terminal (rod spherule or RS) demonstrating a classical ribbon synapse flanked by two central RBC processes and two lateral HC processes. (A′) Schematic representation of the different regions of the macaque retina. (B) 3D rendering of a macaque rod spherule (gray) with one ribbon (red) (top-left); 2 ribbons (middle-left), and one ribbon forming two synaptic units by contacting two distinct sets of postsynaptic partners (bottom-left). Corresponding single-plane annotated EM images displaying the spherules and their ribbon synapses as seen during reconstructions. Four consecutive images depicted for each scenario (right panels). (C) Number of ribbons per individual rod spherule for all four regions of interest (parafovea, central, mid-periphery, and periphery) in the macaque retina. Parafovea: 1.63 ± 0.03 (n = 88), central: 1.05 ± 0.02 (n = 93), mid-periphery: 1.05 ± 0.02 (n = 117), and periphery: 1.27 ± 0.05 (n = 92). n = number of rod spherules reconstructed. The size of the circle represents the relative frequency of occurrence of rod spherules containing 1, 2, or 3 ribbons with respect to the total number of rod spherules in a specific region (see Table S1 and STAR methods). Error bars represents mean ± SEM. Statistics: Tukey’s multiple comparison test was performed. (D) Number of synaptic units per rod spherule across the four regions of interest. Parafovea: 1.65 ± 0.05 (n = 88), central: 1.16 ± 0.04 (n = 93), mid-periphery: 1.21 ± 0.04 (n = 117), and periphery: 1.46 ± 0.05 (n = 92). n = number of rod spherules reconstructed. The size of the circle represents the relative frequency of occurrence of rod spherules containing 1, 2, or 3 synaptic units with respect to the total number of rod spherules in a specific region (see Table S1 and STAR methods). Error bars represents mean ± SEM. Statistics: Tukey’s multiple comparison test was performed. (E) Number of synaptic units per ribbon in each rod spherule across the four regions of interest. Parafovea: 1.01 ± 0.01 (n = 143), central: 1.1 ± 0.03 (n = 98), mid-periphery: 1.17 ± 0.04 (n = 121), and periphery: 1.16 ± 0.03 (n = 116). n represents the number of ribbon synapses documented at each region. The size of the circle represents the relative frequency of occurrence of ribbon synapses containing 1 or 2 synaptic units with respect to the total number of ribbon synapses in a specific region (see Table S1 and STAR methods). Error bars represent mean ± SEM. Statistics: Tukey’s multiple comparison test was performed.

Figure 2

Synaptic divergence at the rod synapse across retinal eccentricity (A) 3D rendering of a parafoveal rod spherule (RS) with 2 ribbons contacting 3 RBC (RBC1-3) and 3 HC (HC1-3) partners. The top panel shows the 3D reconstructed image whereas the bottom panel shows annotated single-plane images of invaginating contact with the different postsynaptic partners. The image on the extreme left is an example raw image of the terminal. Typically, parafoveal rods contact 6 postsynaptic partners and contain two ribbons. Scale bar: 1 μm for single-plane images and 2 μm for 3D renderings. (B) 3D rendering of a central rod spherule with 1 ribbon contacting 2 RBC (RBC1-2) and 2 HC (HC1-2) partners. The top panel shows the 3D reconstructed image whereas the bottom panel shows annotated single-plane images of invaginating contact with the different postsynaptic partners. The image on the extreme left is an example raw image of the terminal. Typically, central rods utilize a single ribbon to connect with 4 postsynaptic partners. (C) 3D rendering of a mid-peripheral rod spherule with a single ribbon contacting 2 RBC (RBC1-2) and 2 HC (HC1-2) partners. The top panel shows the 3D reconstructed image whereas the bottom panel shows annotated single-plane images of invaginating contact with the different postsynaptic partners. The image on the extreme left is an example raw image of the terminal. Typically, mid-peripheral rods utilize a single ribbon to connect with 4 postsynaptic partners. (D) 3D rendering of a peripheral rod spherule containing 1 ribbon apposed to 3 RBC (RBC1-3) and 2 HC (HC1-2) partners. The top panel shows the 3D reconstructed image whereas the bottom panel shows annotated single-plane images of invaginating contact with the different postsynaptic partners. The image on the extreme left is an example raw image of the terminal. Typically, peripheral rod spherules contact 5 postsynaptic partners at one ribbon. (E) Total number of postsynaptic partners (RBC+HC) contacting a rod spherule across eccentricities. The bar represents the average with the different sized circles representing the number of instances of each occurrence. Parafovea: 5.81 ± 0.11 (n = 88), central: 4.03 ± 0.07 (n = 104), mid-periphery: 4.13 ± 0.05 (n = 117), and periphery: 5.03 ± 0.09 (n = 92). In this plot and in plots shown in panel F and G, the size of the circle represents the relative frequency of occurrence of rod spherules containing the different numbers of postsynaptic partners with respect to the total number of rod spherules (see Table S2 and STAR methods). n represents number of rod spherules. Error bars represents mean ± SEM. Statistics: Tukey’s multiple comparison test was performed. (F and G) Number of RBCs (F) and HCs (G) contacting a rod spherule across different regions of the macaque retina. Number of RBCs in (F) - Parafovea: 3.3 ± 0.07 (n = 88), central: 2 ± 0.04 (n = 104), mid-periphery: 2.11 ± 0.04 (n = 117), and periphery: 2.84 ± 0.07 (n = 92). Number of HCs in (G) - Parafovea: 2.51 ± 0.07 (n = 88), central: 2.06 ± 0.04 (N = 104), mid-periphery: 2.02 ± 0.02 (n = 117), and periphery: 2.2 ± 0.04 (n = 92). n represents number of rod spherules. Error bars represents mean ± SEM. Statistics: Tukey’s multiple comparison test was performed.

Figure 3

Synapse organization and divergence at rod photoreceptors in mouse and marmoset retina (A and B) 3D rendering of typical organization of the rod terminal ribbon synapse and apposed postsynaptic partners in mouse (A, top) and marmoset (B, top) rod axon terminals. Mouse rods typically have 4 postsynaptic partners (2 RBCs and 2 HCs) whereas marmoset rods have significantly higher degree of synaptic divergence with the most common configuration of ∼8 postsynaptic partners (typically 5 RBCs and 3 HCs). Bottom panel (A and B) depicts corresponding single-plane annotated images depicting invaginating contact of each postsynaptic partner at the rod spherule. (C–H) Quantification of the number of ribbons per rod spherule Mouse: 1.08 ± 0.04 (n = 50), Marmoset: 1.82 ± 0.125 (n = 22) (C), number of synaptic units per rod spherule Mouse: 1.14 ± 0.05 (n = 50), Marmoset: 2.32 ± 0.12 (n = 22) (D), number of synaptic units per rod ribbon Mouse: 1.06 ± 0.03 (n = 54), Marmoset: 1.28 ± 0.08 (n = 40) (E), number of total postsynaptic partners contacting a rod spherule Mouse: 4.02 ± 0.05 (n = 50), Marmoset: 7.86 ± 0.3 (n = 22) (F), number of RBC partners contacting a rod spherule Mouse: 2.02 ± 0.04 (n = 50), Marmoset: 5.32 ± 0.13 (n = 22) (G) and number of HC partners contacting a rod spherule (H) for mouse vs. marmoset rod spherules Mouse: 2 ± 0.04 (n = 50), Marmoset: 2.55 ± 0.13 (n = 22). The size of the circle represents the relative frequency of occurrence as plotted in Figures 1 and 2 (see Table S3, STAR methods). Scale bar: 1 μm for single-plane images and 2 μm for 3D renderings. n = number of rod spherules reconstructed. Error bars represents mean ± SEM. Statistics: An unpaired, two-tailed student's t test was performed.

Figure 4

Rod convergence to rod bipolar cells across retinal eccentricities (A) 3D reconstructions of rod bipolar cells (RBCs) that contact rod spherules at different locations of the macaque retina—parafovea (magenta), central (brown), mid-peripheral (green), and peripheral (blue). Top panel: side view of RBC soma and dendritic arbor; bottom panel: top-down view of RBC arbor. (B) 3D rendering of all rod spherules contacting an individual RBC at different retinal locations—parafovea (magenta, rods in shades of pink), central (brown, rods in shades of orange), mid-peripheral (green, rods in lighter shades of green), and peripheral (blue, rods in lighter shades of blue) retina. Top panel: side view; bottom panel: top-down view. (Scale bar: 8 μm). (C) Quantification of RBC dendritic arbor area across retinal regions. Parafovea: 894.36 ± 32.13 μm² (n = 3), central: 290.07 ± 14.85 μm² (n = 3), mid-periphery: 345.53 ± 55.32 μm² (n = 3), and periphery: 1714.17 ± 343.10 μm² (n = 3). The circles represent individual data points. n = number of RBCs reconstructed. Error bars represents mean ± SEM. Statistics: Tukey’s multiple comparison test was performed. (D) Comparison of number of rods converging onto a single RBC across different locations of macaque retina. Parafovea: 34.67 ± 2.40 (n = 3), central: 36.50 ± 2.06 (n = 4), mid-periphery: 47.67 ± 0.33 (n = 3 RBCs), and periphery: 52 ± 0.58 (n = 3). The circles represent individual data points. n = number of RBCs reconstructed. Error bars represents mean ± SEM. Statistics: Tukey’s multiple comparison test was performed.

Figure 5

Overlap in rod connectivity and dendritic area between neighboring rod bipolar cells across retinal locations (A) Dendritic arbor of two neighboring RBCs that share rod input across different regions of the primate retina. Side (top panels) and top-down (bottom panels) view of the 3D reconstructed profile of neighboring RBCs with overlapping dendritic arbors and sharing rod inputs in the parafovea (magenta), central (brown), mid-peripheral (green), and peripheral (blue) retina. (B) Dendritic arbor overlaps between neighboring RBCs across different regions of the macaque retina. Each oval depicts the dendritic arbor of an individual RBC. The dendritic overlap has been estimated from the overlap between two ellipses. (C) Side (top panels) and top-down (bottom panels) view of rods (in gray) shared between neighboring RBCs at the parafovea, central, mid-peripheral, and peripheral region of the macaque retina. (D) Comparison of the fraction of rods shared per unit shared dendritic area between two neighboring RBCs at the different regions of interest in the macaque retina. Scale bar: 8 μm.