João Paulo Coimbra

Scene from above: Retinal ganglion cell topography and spatial resolving power in the giraffe (Giraffa camelopardalis)

The giraffe (Giraffa camelopardalis) is a browser that uses its extensible tongue to selectively collect leaves during foraging. As the tallest extant terrestrial mammal, its elevated head height provides panoramic surveillance of the environment. These aspects of the giraffes ecology and phenotype suggest that vision is of prime importance. Using Nissl‐stained retinal wholemounts and stereological methods, we quantitatively assessed the retinal specializations in the ganglion cell layer of the giraffe. The mean total number of retinal ganglion cells was 1,393,779 and their topographic distribution revealed the presence of a horizontal visual streak and a temporal area. With a mean peak of 14,271 cells/mm2, upper limits of spatial resolving power in the temporal area ranged from 25 to 27 cycles/degree. We also observed a dorsotemporal extension (anakatabatic area) that tapers toward the nasal retina giving rise to a complete dorsal arch. Using neurofilament‐200 immunohistochemistry, we also detected a dorsal arch formed by alpha ganglion cells with density peaks in the temporal (14–15 cells/mm2) and dorsonasal (10 cells/mm2) regions. As with other artiodactyls, the giraffe shares the presence of a horizontal streak and a temporal area which, respectively, improve resolution along the horizon and in the frontal visual field. The dorsal arch is related to the giraffes head height and affords enhanced resolution in the inferior visual field. The alpha ganglion cell distribution pattern is unique to the giraffe and enhances acquisition of motion information for the control of tongue movement during foraging and the detection of predators. J. Comp. Neurol. 521:2042–2057, 2013.

Photoreceptor types, visual pigments, and topographic specializations in the retinas of hydrophiid sea snakes

Sea snakes have evolved numerous anatomical, physiological, and behavioral adaptations to suit their wholly aquatic lifestyle. However, although sea snakes use vision for foraging and mate selection, little is known about their visual abilities. We used microspectrophotometry, light microscopy, and scanning electron microscopy to characterize the retinal photoreceptors of spine‐bellied (Lapemis curtus) and horned (Acalyptophis peronii) sea snakes. Both species have three types of visual pigment sensitive to short (SWS; wavelength of maximum absorbance, λmax 428–430 nm), medium (MWS; λmax 496 nm), and long wavelengths of light (LWS; λmax 555–559 nm) in each of three different subtypes of cone‐like single photoreceptor. They also possess a cone‐like double photoreceptor subtype, both the principal and accessory member of which contain the LWS visual pigment. Conventional rods were not observed, although the MWS photoreceptor may be a “transmuted” rod. We also used stereology to measure the total number and topographic distribution of neurons in the ganglion cell layer of L. curtus, the olive sea snake (Aipysurus laevis), and the olive‐headed sea snake (Disteira major). All species have a horizontal visual streak with specialized areas in the nasal and temporal retina. Both L. curtus and D. major also have a specialized area in the ventral retina, which may reflect differences in habitat usage and/or foraging behavior compared to A. laevis. Maximal spatial resolution was estimated at 1.1, 1.6, and 2.3 cycles deg−1 in D. major, L. curtus, and A. laevis, respectively; the superior value for A. laevis may reflect its specialized crevice‐foraging hunting technique. J. Comp. Neurol. 520:1246–1261, 2012.

Retinal Ganglion Cell Topography and Spatial Resolving Power in Penguins

Penguins are a group of flightless seabirds that exhibit numerous morphological, behavioral and ecological adaptations to their amphibious lifestyle, but little is known about the topographic organization of neurons in their retinas. In this study, we used retinal wholemounts and stereological methods to estimate the total number and topographic distribution of retinal ganglion cells in addition to an anatomical estimate of spatial resolving power in two species of penguins: the little penguin, Eudyptula minor, and the king penguin, Aptenodytes patagonicus. The total number of ganglion cells per retina was approximately 1,200,000 in the little penguin and 1,110,000 in the king penguin. The topographic distribution of retinal ganglion cells in both species revealed the presence of a prominent horizontal visual streak with steeper gradients in the little penguin. The little penguin retinas showed ganglion cell density peaks of 21,867 cells/mm2, affording spatial resolution in water of 17.07–17.46 cycles/degree (12.81–13.09 cycles/degree in air). In contrast, the king penguin showed a relatively lower peak density of ganglion cells of 14,222 cells/mm2, but – due to its larger eye – slightly higher spatial resolution in water of 20.40 cycles/degree (15.30 cycles/degree in air). In addition, we mapped the distribution of giant ganglion cells in both penguin species using Nissl-stained wholemounts. In both species, topographic mapping of this cell type revealed the presence of an area gigantocellularis with a concentric organization of isodensity contours showing a peak in the far temporal retina of approximately 70 cells/mm2 in the little penguin and 39 cells/mm2 in the king penguin. Giant ganglion cell densities gradually fall towards the outermost isodensity contours revealing the presence of a vertically organized streak. In the little penguin, we confirmed our cytological characterization of giant ganglion cells using immunohistochemistry for microtubule-associated protein 2. This suite of retinal specializations, which is also observed in the closely related procellariiform seabirds, affords the eyes of the little and king penguins panoramic surveillance of the horizon and motion detection in the frontal visual field.

Topographic specializations in the retinal ganglion cell layer correlate with lateralized visual behavior, ecology, and evolution in cockatoos

Cockatoos are a unique avian group inhabiting a diversity of arboreal and terrestrial microhabitats. Most species display strong lateralized visual behaviors using their left eye/foot to assist with food manipulation during foraging. In this study, we used retinal wholemounts and stereological methods to investigate whether the topographic distribution of retinal ganglion cells in cockatoos reflects their lateralized behaviors and microhabitat diversity. We found that all species studied possess a horizontal visual streak and a shallow central fovea that afford increased spatial resolution in the lateral visual field. Arboreal cockatoos have a well‐defined dorsotemporal area, in contrast to terrestrial cockatoos, in which this specialization is inconspicuous or absent. Terrestrial cockatoos also have a triangular extension of increased ganglion cell density directed toward the dorsotemporal retinal periphery. Both the dorsotemporal area and the triangular extension enhance spatial resolution in the frontal and inferior visual fields, which potentially assists with binocular coordination during foraging. We found significantly higher ganglion cell densities in the left (52,000–72,000 cells/mm2) compared with the right (42,500–50,000 cells/mm2) perifoveal region of species that have strong left eye–left foot lateralized behaviors. In contrast, cockatoo species that show no lateralized behaviors have equivalent retinal ganglion cell densities in both left and right perifoveal regions (42,500–52,500 cells/mm2). Retinal ganglion cell peak densities in the dorsotemporal area showed no significant difference between left and right eyes for any species, suggesting that cockatoos use both eyes to extract information in the binocular visual field, independent of the degree of lateralization. J. Comp. Neurol. 522:3363–3385, 2014.

Variations in retinal photoreceptor topography and the organization of the rod-free zone reflect behavioral diversity in Australian passerines

The avian retina possesses one of the most diverse complements of photoreceptor types among vertebrates but little is known about their spatial distribution. Here we used retinal wholemounts and stereological methods to present the first complete maps of the topographic distribution of rods and cones in four species of Australian passerines with diverse trophic specializations. All species studied have one central and one temporal rod‐free zone. In the insectivorous yellow‐rumped thornbill, the central rod‐free zone is unusually large, occupying ∼17% (56°) of the retinal area (angular subtense), whereas in nectarivorous and frugivorous species it represents only ∼0.1% (5–7°) to 0.3% (10°) of the retinal area (angular subtense). In contrast, the temporal rod‐free zone varies little between species (∼0.02–0.4%; 2–10°). In all species, rods follow a pronounced dorsoventral gradient with highest densities in the ventral retina. The topographic distribution of cones is concentric and reveals a central fovea and a temporal area. In the yellow‐rumped thornbill, cone densities form an extended plateau surrounding the fovea, beyond which densities fall rapidly towards the retinal periphery. For the other species, cone densities decline gradually along a foveal to peripheral gradient. Estimates of spatial resolving power calculated using cone peak densities are higher in the central fovea (19–41 cycles/degree) than in the temporal area (9–15 cycles/degree). In conclusion, we suggest that the unusual organization of the rod‐free zone and the distinct topographic distribution of rods and cones correlate with specific ecological needs for enhanced visual sensitivity and spatial resolution in these birds. J. Comp. Neurol. 523:1073–1094, 2015.

The Retina of Ansorge's Cusimanse (Crossarchus ansorgei): Number, Topography and Convergence of Photoreceptors and Ganglion Cells in Relation to Ecology and Behavior

The family Herpestidae (cusimanses and mongooses) is a monophyletic radiation of carnivores with remarkable variation in microhabitat occupation and diel activity, but virtually nothing is known about how they use vision in the context of their behavioral ecology. In this paper, we measured the number and topographic distribution of neurons (rods, cones and retinal ganglion cells) and estimated the spatial resolving power of the eye of the diurnal, forest-dwelling Ansorges cusimanse (Crossarchus ansorgei). Using retinal wholemounts and stereology, we found that rods are more numerous (42,500,000; 92%) than cones (3,900,000; 8%). Rod densities form a concentric and dorsotemporally asymmetric plateau that matches the location and shape of a bright yellow tapetum lucidum located within the dorsal aspect of the eye. Maximum rod density (340,300 cells/mm2) occurs within an elongated plateau below the optic disc that corresponds to a transitional region between the tapetum lucidum and the pigmented choroid. Cone densities form a temporal area with a peak density of 44,500 cells/mm2 embedded in a weak horizontal streak that matches the topographic distribution of retinal ganglion cells. Convergence ratios of cones to retinal ganglion cells vary from 50:1 in the far periphery to 3:1 in the temporal area. With a ganglion cell peak density of 13,400 cells/mm2 and an eye size of 11 mm in axial length, we estimated upper limits of spatial resolution of 7.5-8 cycles/degree, which is comparable to other carnivores such as hyenas. In conclusion, we suggest that the topographic retinal traits described for Ansorges cusimanse conform to a presumed carnivore retinal blueprint but also show variations that reflect its specific ecological needs.

Topographic specializations in the retinal ganglion cell layer of Australian passerines

Thornbills, honeyeaters, and silvereyes represent an abundant group of Australian passerines, whose diversity in niche differentiation suggests a pivotal role for vision. Using stereological methods and retinal wholemounts, we studied the topographic distribution of neurons in the ganglion cell layer of insectivorous, nectarivorous, and frugivorous species occupying terrestrial and arboreal microhabitats. All species studied have a central convexiclivate fovea (peak densities from 130,000 to 160,000 cells/mm2), which is shallow in the terrestrial/insectivorous yellow‐rumped thornbill and deep in the arboreal/nectarivorous honeyeaters and frugivorous silvereye. Surrounding the fovea, neuronal densities in the ganglion cell layer form a broadly ovoid and asymmetric plateau in the yellow‐rumped thornbill and a more restricted, circular and symmetric plateau in the other species. These differences in the plateau organization may reflect specific needs to locate food on the ground or among dense vegetation. We also found a temporal area (peak densities from 43,000 to 54,000 cells/mm2) across species, which increases spatial resolution in the frontal visual field and assists with foraging. Using microtubule‐associated protein 2 (MAP2) immunohistochemistry, we detected a higher concentration of giant ganglion cells forming an area gigantocellularis in the temporal retina of all species. Giant ganglion cell densities also form a horizontal streak in all species, except in the yellow‐rumped thornbill, which has an unusual increase toward the retinal periphery. In the yellow‐rumped thornbill and silvereye, giant ganglion cells also peak in the nasal retina. We suggest that these topographic variations afford differential sampling of motion signals for the detection of predators. J. Comp. Neurol. 522:3609–3628, 2014.

Classification of retinal ganglion cells in the southern hemisphere lamprey Geotria australis (Cyclostomata)

Lampreys are one of two extant representatives of the earliest group of vertebrates, the agnathans or jawless fishes. The single species of the southern hemisphere lamprey family Geotriidae, Geotria australis, possesses the potential for pentachromatic color discrimination opposed to the mono‐ or dichromacy found in other lampreys. However, little is known of the retinal ganglion cell types that contribute to visual processing in G. australis. A quantitative morphological approach was used to distinguish and describe retinal ganglion cell types in G. australis. The morphology of retinal ganglion cells was revealed by retrograde biocytin labeling from the optic disc. Cells were digitally reconstructed, and somatic area and position and dendritic field size, density, tortuosity, and stratification were subjected to quantitative morphometric analyses. Cluster analysis, in conjunction with similarity profile analysis (SIMPROF), statistically identified five discrete monostratified retinal ganglion cell types, one of which may comprise two subtypes. Two bistratified types were identified separately, including a biplexiform and a bistratified subtype. The use of cluster analysis with SIMPROF provided a robust statistical technique for objectively identifying cell types whose characteristics were similar and significantly different from those of other types and thus provides an objective resolution of the problems posed by “lumpers vs. splitters” when designating cell types. The diversity of retinal ganglion cells suggests that visual information in the lamprey G. australis is processed in parallel streams, as in gnathostomes. These findings, together with the results of previous studies, indicate that the visual system of the lamprey G. australis represents the upper limit of visual complexity in extant agnathans. J. Comp. Neurol. 522:750–771, 2014.

The Topographic Organization of Retinal Ganglion Cell Density and Spatial Resolving Power in an Unusual Arboreal and Slow-Moving Strepsirhine Primate, the Potto (Perodicticus potto).

The potto (Perodicticus potto) is an arboreal strepsirhine found in the rainforests of central Africa. In contrast to most primates, the potto shows slow-moving locomotion over the upper surface of branches, where it forages for exudates and crawling invertebrates with its head held very close to the substrate. Here, we asked whether the retina of the potto displays topographic specializations in neuronal density that correlate with its unusual lifestyle. Using stereology and retinal wholemounts, we measured the total number and topographic distribution of retinal ganglion cells (total and presumed parasol), as well as estimating the upper limits of the spatial resolution of the potto eye. We estimated ∼210,000 retinal ganglion cells, of which ∼7% (∼14,000) comprise presumed parasol ganglion cells. The topographic distribution of both total and parasol ganglion cells reveals a concentric centroperipheral organization with a nasoventral asymmetry. Combined with the upwardly shifted orbits of the potto, this nasoventral increase in parasol ganglion cell density enhances contrast sensitivity and motion detection skywards, which potentially assists with the detection of predators in the high canopy. The central area of the potto occurs ∼2.5 mm temporal to the optic disc and contains a maximum ganglion cell density of ∼4,300 cells/mm2. We found no anatomical evidence of a fovea within this region. Using maximum ganglion cell density and eye size (∼14 mm), we estimated upper limits of spatial resolving power between 4.1 and 4.4 cycles/degree. Despite their reported reliance on olfaction to detect exudates, this level of spatial resolution potentially assists pottos with foraging for small invertebrates and in the detection of predators.

Retinal ganglion cell topography and spatial resolving power in African megachiropterans: Influence of roosting microhabitat and foraging

Megachiropteran bats (megabats) show remarkable diversity in microhabitat occupation and trophic specializations, but information on how vision relates to their behavioral ecology is scarce. Using stereology and retinal wholemounts, we measured the topographic distribution of retinal ganglion cells and determined the spatial resolution of eight African megachiropterans with distinct roosting and feeding ecologies. We found that species roosting in open microhabitats have a pronounced streak of high retinal ganglion cell density, whereas those favoring more enclosed microhabitats have a less pronounced streak (or its absence in Hypsignathus monstrosus). An exception is the cave‐dwelling Rousettus aegyptiacus, which has a pronounced horizontal streak that potentially correlates with its occurrence in more open environments during foraging. In all species, we found a temporal area with maximum retinal ganglion cell density (∼5,000–7,000 cells/mm2) that affords enhanced resolution in the frontal visual field. Our estimates of spatial resolution based on peak retinal ganglion cell density and eye size (∼6–12 mm in axial length) range between ∼2 and 4 cycles/degree. Species that occur in more enclosed microhabitats and feed on plant material have lower spatial resolution (∼2 cycles/degree) compared with those that roost in open and semiopen areas (∼3–3.8 cycles/degree). We suggest that the larger eye and concomitant higher spatial resolution (∼4 cycles/degree) in H. monstrosus may have facilitated the carnivorous aspect of its diet. In conclusion, variations in the topographic organization and magnitude of retinal ganglion density reflect the specific ecological needs to detect food/predators and the structural complexity of the environments. J. Comp. Neurol. 525:186–203, 2017.