Morphological properties of wiry amacrine cells. A–C: z-projections of confocal image stacks of 2 ON (A and B)- and 1 OFF (C)-type wiry amacrine cell in whole-mounted retinas with overlapping parasol cells (green circles in B and C). Arrowhead in B points to the axon of the parasol cell. D: branching pattern of primary dendrites. E: z-projection of confocal image stack of a single dendrite (green) from an OFF wiry amacrine cell. The dendrite was very thin (diameter ∼0.35 μm) and exhibited putative synaptic varicosities (diameter ∼0.6–1.0 μm) along its length. F–I: xy-projections of single dendrites (green) from whole-mounted retinas, double-labeled with DAPI (red) to identify inner and outer borders of the inner plexiform layer (IPL). Wiry cells stratified at 3 distinct levels within the IPL. I: xy-projection of a single dendrite (green) from a center-stratifying ON wiry cell in whole-mounted retina, double-labeled with ChAT (red). Dendrite is located in close proximity above ON starburst amacrine cells. Scale bars: 100 μm (A–C), 10 μm (D–I).

Basic light responses of wiry amacrine cells. A: Voltage responses of an ON-type and an OFF-type wiry amacrine cell to a 0.5-s, +100% contrast spot centered on the soma. Inset, 60-ms region of peak OFF cell response exhibiting high-frequency spikelets riding on top of a sustained depolarization. B: responses of cells in A to a −100% contrast spot. Inset, 60-ms region of peak ON cell response exhibiting spikelets. C: voltage response in an ON-type wiry cell to a 0.5-s family of 11 injected currents (50-pA increments from −200 pA to +300 pA). D: linear filters of ON and OFF wiry cells calculated with full-field Gaussian noise. E: nonlinear input-output functions for cells in D. F: kinetic properties derived from temporal filters of 7 ON wiry cells, 4 OFF wiry cells, 6 ON parasol cells, and 3 OFF parasol cells. Error bars denote mean and SD. Left: time to filter peak for the temporal filters in milliseconds. A smaller value indicates more brisk kinetics. Center: time to zero crossing of the temporal filter in milliseconds. A smaller value indicates more transient kinetics. Right: biphasic index of temporal filters. This value is the trough divided by the peak of the temporal filter.

Wiry cells integrate over an extensive area of space. A: spatial receptive field of a wiry amacrine cell determined with spatio-temporal white noise. B: maximum-intensity projection of confocal images from cell in A injected with Lucifer yellow. C: merge of spatial receptive field in A and dendritic morphology in B. D: normalized cross-correlation of spatial receptive field and dendritic morphology as a function of rotation angle for 5 wiry cells. Plot shows means ± SE.

Spatio-temporal dependence of wiry cell receptive fields. A–E: slices through the spatio-temporal receptive field at different points in time. F: example temporal filters taken from points in the receptive field in E as indicated by arrows. G: receptive field map of cell in A–E in which the peak of each temporal filter is color-coded. H: timing of temporal filter peak in dendrites as a function of distance from center. Generally, distal dendrites peaked earlier than proximal dendrites (n = 5 cells; mean ± SE). Cyan and red lines show prediction based on passive membrane properties with fast or slow membrane time constants, respectively. I: gain of temporal filter in dendrites as function of distance from center for cells in H. Cyan and red lines indicate the prediction of the passive models corresponding to H.

Propagation of signals from distal dendritic processes. A: voltage responses of an OFF wiry amacrine cell to a black bar (contrast −100%, width 90 μm, height 270 μm, duration 0.5 s) presented either proximally or distally (0.56 mm) to the soma. B: voltage recording from an A1 amacrine cell stimulated with a contrast increment (contrast 100%, duration 0.5 s). C: example spikes from an A1 amacrine cell (AC) and a wiry amacrine cell. D: responses of an ON wiry cell to a white spot centered on the soma (contrast −100%, diameter 0.52 mm, duration 0.5 s) under control conditions, with voltage-gated Na⁺ (Na_V) channels blocked (TTX 0.5 μM), and with NMDA spikes blocked (R-CPP 15 μM, MK-801 10 μM). E: responses of cell in B to a white ring centered on the soma (inner diameter 0.78 mm, outer diameter 0.91 mm). F: peak responses for conditions in E. Spikelets seen under control condition were not suppressed by TTX but were abolished by NMDA receptor antagonists. G: response amplitude to the distal ring relative to the proximal spot for the 3 conditions in D and E averaged across 4 wiry cells and recordings with intracellular MK-801 (1 mM; n = 3 cells) (mean ± SE). H: example spikelets in the same cell under control conditions and with Na_V blockade. I: average spikelet across 4 cells for conditions in H. Dashed line indicates the fitted slope during the spikelet's initial rising phase. J and K: spikelet amplitude (J) and initial slope of spikelets (K) across 4 cells for conditions in H. Circles indicate individual cells; error bars indicate SE.

Wiry cells are ON-OFF to moving objects. A: spatial receptive field of an ON-type wiry amacrine cell. B: schematic showing location of a moving square relative to the stationary receptive field. C: responses of cell in A to outwardly moving black and white squares. Solid and dashed arrows indicate onset and offset of motion, respectively. D: responses of the same cell to inwardly moving black and white squares. E: wiry cell responses to outward squares of varying velocity (n = 3 cells). F: responses to expanding ring at multiple velocities (n = 5 cells). G: responses of an OFF-type wiry amacrine cell to a rectangle (90 μm × 270 μm; contrast ±100%; speed 1.084 mm/s, 5.42°/s) moving perpendicular to long axis and inwardly along wiry cell's dendrite. H: responses of cell in G to motion perpendicular to the same dendrite.