Thorny ganglion cells in marmoset retina: Morphological and neurochemical characterization with antibodies against calretinin
In primates, including humans, Old World and New World monkeys, retinal ganglion cell types have been distinguished by morphological criteria such as dendritic field size, branch density, and stratification in the inner plexiform layer, using Golgi impregnated tissue (Boycott & Dowling, 1969; Kolb, Linberg, & Fisher, 1992; Polyak, 1941) or intracellular injection (Ghosh, Goodchild, Sefton, & Martin, 1996; Peterson & Dacey, 1999; Watanabe & Rodieck, 1989; Yamada, Bordt, & Marshak, 2005). More recently, retrograde labeling in combination with photofilling (Dacey, Peterson, Robinson, & Gamlin, 2003; Szmajda, Grünert, & Martin, 2008) and particle mediated gene transfer (Masri, Percival, Koizumi, Martin, & Grünert, in press; Moritoh, Komatsu, Yamamori, & Koizumi, 2013) have been used and at least 17 types of ganglion cell types have now been identified in primate retina.
In addition, retinal ganglion cell types have been distinguished based on their projection to specific brain areas (Dacey et al., 2003; Leventhal, Rodieck, & Dreher, 1981; Perry, Oehler, & Cowey, 1984; Perry & Cowey, 1984; Rodieck & Watanabe, 1993; Szmajda et al., 2008). These studies demonstrated that the large majority of primate ganglion cells are either midget cells which project to the parvocellular layers of the lateral geniculate nucleus (LGN), or parasol cells which project to the magnocellular layers of the LGN. The third most frequent ganglion cell type, the small bistratified ganglion cell (Dacey, 1993a; Dacey & Lee, 1994) projects to the koniocellular layers of the LGN (Martin, White, Goodchild, Wilder, & Sefton, 1997; Roy et al., 2009; Szmajda et al., 2008).
Despite this knowledge about the morphological diversity and the central projections of retinal ganglion cells, only limited information is available on the density and spatial distribution of ganglion cell types within the retina, whereas cell specific markers have long been used to study spatial density and distribution of retinal interneurons, that is, horizontal, bipolar and amacrine cell types (Grünert, Martin, & Wässle, 1994; Vaney, 1990; Wässle, Boycott, & Röhrenbeck, 1989). The problem is that there are only very few known markers for any ganglion cell populations in primate retina, and most of these markers do not reveal the dendritic tree of the cell. Thus, the spatial density and distribution for most ganglion cell types so far is based on estimates (Dacey, 2004; Masri et al., in press). An exception are the melanopsin‐expressing ganglion cells, whose spatial distribution has been well studied with antibodies against melanopsin (Dacey et al., 2005; Jusuf, Lee, Hannibal, & Grünert, 2007; Liao et al., 2016; Nasir‐Ahmad, Lee, Martin, & Grünert, in press).
Antibodies against calcium binding proteins such as parvalbumin and calretinin are known to label cell populations in the retina of a variety of mammals, and have been used to classify ganglion cells in mouse and rabbit retinas (Kim & Jeon, 2006; Kwon, Lee, & Jeon, 2014; Lee, Lee, & Jeon, 2010; Yi, Yu, Lee, Lee, & Jeon, 2012). In primate retina, the expression of calcium binding proteins shows common pattern as well as species specific differences (Chiquet, Dkhissi‐Benyahya, & Cooper, 2005; Pasteels, Rogers, Blachier, & Pochet, 1990).