Irregular S-cone mosaics in felid retinas Spatial interaction with axonless horizontal cells, revealed by cross correlation

Abstract

In most mammals short-wavelength-sensitive (S) cones are arranged in irregular patterns with widely variable intercell distances. Consequently, mosaics of connected interneurons either may show some type of correlation to photoreceptor placement or may establish an independent lattice with compensatory dendritic organization. Since axonless horizontal cells (A-HC's) are supposed to direct all dendrites to overlying cones, we studied their spatial interaction with chromatic cone subclasses. In the cheetah, the bobcat, and the leopard, anti-S-opsin antibodies have consistently colabeled the A-HC's in addition to the S cones. We investigated the interaction between the two cell mosaics, using autocorrelation and cross-correlation procedures, including a Voronoi-based density probe. Comparisons with simulations of random mosaics show significantly lower densities of S cones above the cell bodies and primary dendrites of A-HC's. The pattern results in different longwavelength-sensitive-L-and S-cone ratios in the central versus the peripheral zones of A-HC dendritic fields. The existence of a related pattern at the synaptic level and its potential significance for color processing may be investigated in further studies.

Statistical mechanics characterization of neuronal mosaics

“…Such results provide circumstantial evidence in favor of the fact that irregular rather than regular patterns appear to be the rule among mammals with S cone mosaics [11]. Random distribution in the S cone mosaics have been identified in species like marsupials, rabbits, cats, horses, rats, cheetahs and guinea pigs [10,12,13,14,15] 4 which, like the agouti, have rod-dominant retina. On the other hand, regular S mosaics are found only in animals with cone-dominant retina [16,17,18], possibly accounting for enhanced visual sampling properties compatible with diurnal arboreal habitats.…”

V-Proportion: A method based on the Voronoi diagram to study spatial relations in neuronal mosaics of the retina

“…In contrast to this technique, the V-Proportion measurement is based on polygonal domains around the cells, which generally provide more adequate spatial approximations of dendritic fields than circles [14]. However, since our Voronoi procedure is based solely on the locations of cell bodies and does not produce concave corners, it may also fail for cells with extreme dendritic domains.…”

“…Finally, Ahnelt et al [14] apply a first version of the V-Proportion to analyze interactions between irregular S-cone mosaics and axonless horizontal cells. In comparison to [14], this paper presents several improvements.…”

“…In comparison to [14], this paper presents several improvements. First, we solve the edge problem by ignoring points lying on open Voronoi polygons.…”

“…A positive spatial dependency between two different populations of neurons may be an indicator of some connection patterns between them [11,12,13,14]. The spatial information is moreover helpful to study dependencies during development [15].…”