PMC

Different ganglion cell types have different sodium- channel bands. A: each point represents a single sodium-channel band from either a DS (filled) or BT (open) cell and plots the length vs. distance from the soma for the band. B: examples of sodium-channel bands from 2 unidentified ganglion cell types. Note the variability in length and location of the bands. C: sodium-channel bands from additional, unidentified ganglion cell types are shown alongside the bands from DS and BT cells. Scale bar in B: 25 μm.

High-density sodium-channel and low-threshold regions are spatially coextensive. Each thin dashed line represents a measured sodium-channel band. The length of the line corresponds to the length of the band and the distance between the line and the soma corresponds to the distance between the band and the soma. The thick dashed line represents the arithmetic mean of all high-density regions. The solid lines are identical to those in : each thin line corresponds to the length and position of each region of low threshold. The thick line represents the mean size and location of all low-threshold regions. The axon bend region is identical to that of and represents a distance of 6 μm from the soma. The top and bottom arrows indicate the low-threshold region and the sodium-channel band, respectively, from the same cell (see ).

The sodium-channel band is coextensive with the region of low threshold. A: threshold map of a DS ganglion cell. The map includes threshold measurements over the soma/proximal axon region (right) as well as along more distal sections of the axon (columns in the middle and the left). The threshold values given by each color are identical to those in . The circle indicates the position of the soma. B: overlay of the threshold map with the dye-filled ganglion cell. “X”'s indicate the position of lowest threshold in each column and were aligned to the corresponding portion of the distal axon in the dye-filled cell. C: a higher-magnification view of the soma/proximal axon region from B (indicated by the rectangular box in B). The position of the sodium-channel band, indicated by the arrows, is in the approximate center of the low-threshold region. Scale bar in A, B, and C: 50 μm.

Determining threshold at a single location. A: ganglion cell responses to cathodic stimulus pulses measured with cell-attached patch clamp (voltage clamp). A₁: responses to a 16-μA pulse (3 repetitions): arrows indicate the onset and offset of the stimulus artifact. A₂: responses to three 18-μA pulses: an action potential (arrow) is elicited by one of the 3 pulses. A₃ and A₄: responses to 20- and 22-μA pulses: action potentials (arrows) are elicited by 2 and 3 of the pulses, respectively. B: number of elicited spikes is plotted as a function of stimulus pulse amplitude (“×”). The amplitude level at which 2 of 3 pulses elicits spiking is ambiguous and so the sigmoid function was fit to the data points and used to assign a value of threshold; solid lines intersect the sigmoid curve to determine the threshold level. Scale bar in A₁: 200-pA vertical and 1-ms horizontal applies to A₂–A₄.

Threshold maps reveal regions of high and low sensitivity. A: threshold map for a directionally selective (DS) ganglion cell. Each colored square represents a threshold measurement for that location of the stimulating electrode (scale at bottom). The circle indicates the approximate location of the soma. B: threshold map for a local edge detector (LED). Note the difference in size and location of the low-threshold region. C: population results from threshold measurements in brisk transient/alpha (BT, n = 7), DS (n = 10), and LED (n = 5) ganglion cells. The red horizontal lines indicate the median threshold for each type; the top and bottom horizontal edges of the box represent the upper and lower quartiles. The whiskers indicate the lower and upper extents of the data for each type. Thresholds were statistically different for all 3 types (see text). Scale bar: 100 μm applies to A and B.

Dense sodium-channel staining is limited to a small region of the axon. A: confocal image stack of a green fluorescent protein (GFP)–filled DS ganglion cell. The characteristic morphological features include a relatively large soma as well as bistratified and recursive dendritic processes. B: higher-magnification view of the soma and initial portion of the axon. C: immunochemical staining for Ankyrin G (red) reveals several long, thin segments (arrows). The small, punctuate staining is an artifact associated with nonspecific binding of one of the antibodies. D: overlay of B and C reveals that one of the long Ankyrin G segments is coextensive with a portion of the axon. Vertical lines from C help clarify the edges of the Ankyrin G staining. E: the dense sodium-channel region is colocalized to the region where the axon diameter decreases. F: close-up view of the soma and axon from a DS ganglion cell (green) with pan sodium antibody (PAN) immunostaining (red). The solid vertical lines indicate the approximate extent of the dense sodium-channel region and reveal that the dense sodium-channel region fills the tapered portion of the axon (compare the axon diameter at the right and left solid lines in E). The dashed vertical line indicates the edge of the soma. The distance between the 2 solid lines was used to determine the length of the band; the distance between the dashed line and the closest solid line was used to determine the distance between the soma and the band. Scale bar in A: 25 μm. Scale bar in B: 25 μm also applies to C and D. Scale bar in E: 20 μm, also applies to F.

Low-threshold regions are spatially distinct from the soma and axon bend. A, top: the trajectory of the proximal axon was estimated by drawing a line from the soma to the start of the distal axon (single column of threshold measurements at left; see also and ). The scale correlating color to amplitude is identical to that of . Bottom: measured thresholds along this line (“×”) were fit with a second-order curve (solid line); the minimum point along the fit curve was set as the center of the low-threshold region. The length of the axon for which thresholds were within 2 μA of minimum determined the length of the low-threshold region. B: each thin line corresponds to the low-threshold region from a threshold map for a different DS cell. The length of each line is proportional to the length of the low-threshold region and the distance between the closest point of the region and the closest edge of the soma is proportional to the distance between the region and the soma. The thick horizontal line represents the mean (size and location) of all low-threshold regions. The axon bend region, to the left of the soma, represents a 6-μm-wide region adjacent to the soma (see text). C: 2 adjacent ganglion cells (green) extend dendritic processes down into the inner plexiform layer and axons up to the nerve fiber layer (NFL, arrowhead). For the cell on the right (DS), the axon emerges from the vitreal end of the soma and ascends directly to the NFL; the bend (arrow) occurs within the lateral boundaries of the soma. For the cell on the left (non-DS), the axon emerges from the approximate midline of the soma and ascends slowly to the NFL; the bend is about 45 μm from the soma (arrow). Scale bars in A and C are both 50 μm.