Neurofilaments in Retinas of Normal Mice and of Mice with Hereditary Photoreceptor Loss. In: Sheffield JB, Hilfer SR, editors. Heredity and Visual Development. New York: Springer; 1985. pp. 43-62. [DOI: 10.1007/978-1-4612-5134-7_3]

Development of A-type (axonless) horizontal cells in the rabbit retina

The development of A-type horizontal cells (HC) was studied in the rabbit retina between embryonic day (E)24 and adulthood [the day of birth was called postnatal day (P)1 and corresponds to E31-32]. The cells were visualized by several methods 1) by immunolabeling with antibodies to neurofilament 70,000 (NF-70kD), 2) by immunolabeling with antibodies to a calcium binding protein (CaBP-28kD), 3) by two different methods of silver impregnation, and 4) by histochemical demonstration of NADH-diaphorase activity. Most methods labeled A-type HC only in the dorsal retina; thus, our study is restricted to HC of this region. HC densities were determined at each developmental stage. The cells were drawn at scale, and size, quotient of symmetry, and topographical orientation of dendritic trees were studied by image analysis. The growth of HC dendritic fields was correlated with data on the postnatal local retinal expansion, which is known to be driven by the intraocular pressure (after cessation of retinal cell proliferation at P9). This expansion was evaluated in an earlier paper (Reichenbach et al. [1993] Vis. Neurosci. 10:479-498) by using local subpopulations of Müller cells as "markers" of distinct topographic regions of the retinae. After E24, when the final number of HC is established, we can discriminate three distinct developmental stages of A-type HC. During the first stage, between E24 and E27, the young cells are often vertically oriented and may extend their first short dendrites within (the primordia of) both plexiform layers. The irregular HC mosaic at E24 shows a significant difference to all other stages. The second stage begins after birth when the dendritic trees of the cells are already restricted to the outer plexiform layer. Between P3 and P9, their dendritic trees enlarge more than the surrounding retinal tissue expands, and the coverage factor almost doubles from 2.5 to 4.4. The third stage occurs after P9 when the growth rate of dendritic tree areas corresponds to that of the local retinal tissue expansion caused by "passive stretching" of the postmitotic tissue, and the coverage factor remains constant. This is compatible with the view that mature synaptic connections of A-type HC are mostly established after the first week of life and are then maintained.

Growth and development of the mouse retinal pigment epithelium. I. Cell and tissue morphometrics and topography of mitotic activity

A computer-assisted morphometric and kinetic analysis of retinal pigment epithelium (PE) development was carried out in C57BL/6J and hybrid mice from Embryonic Day 13 (E13) to Postnatal Day 250 (P250). Total cell number rose from 14,000 at E15 at the rate of about 4000 cells/day to P1 and then at about 1500 cells/day to reach a stable level of 54,000 cells at P15. Compared to the 4-fold rise in cell number, PE area increased about 10-fold, in part through cell hypertrophy which continued beyond P15. Cell concentration increased with distance from the optic nerve head during development, but the gradient disappeared by P20 except for a consistent population of small cells around the optic nerve head and a late-appearing population of very large cells at the ora serrata. Binucleate cells constitute 2.1% of the PE cell population at P1 and 26% at P30, almost all of them located in the posterior 75% of the PE where they comprise 70% of the cell population at some radial positions. Mitotic cells, detected by fluorescent monoclonal antibody R3, are distributed across the entire PE at E13. As the eye grows the mitotic zone occupies a progressively smaller and more distal proportion of the increasing radius; by P5 only the region near the ora serrata is highly active, with some additional mitotic cells trailing into a broad central zone. From P7 to P15 nuclear divisions persist only centrally to generate the youngest uninucleate and binucleate cells. The mouse PE thus shows a pattern of edge-biased interstitial growth (in contrast to amphibians with strict edge growth).

Light-dependent antibody labelling of photoreceptors

Monoclonal antibodies are tools widely used to analyse the structure of the nervous system. Whereas some labelling patterns are highly reproducible, others appear to vary from one preparation to the next, as we noticed in particular for some antibodies with respect to photoreceptor labelling. To establish whether some of this variability can be linked to functional criteria, we tested for light-dependence. We found two antibodies that label photoreceptor outer segments, only when the retina has been illuminated, and a third antibody that has a selective affinity for dark-adapted outer segments. The two antibodies against light-activated sites are primarily directed against the highly phosphorylated neurofilament subunit at relative molecular mass 200,000 (200K). One of them, RT97, recognizes on immunoblots, in addition to neurofilaments, a light-activated epitope on a protein that resembles the photopigment rhodopsin, presumably a phosphorylation-dependent site. Antibodies like those described here may allow the study of physiological processes such as light and dark adaptation using morphological techniques.