Cone Properties of Retinal Margin Cells in the Monkey (Macaca mulatta)

Abstract

purpose. To characterize a cell population in the monkey retinal margin that was labeled with a cone-specific antibody and to determine the presence of additional markers.

methods. Retinal whole-mount preparations from infant and adult rhesus monkeys (Macaca mulatta) were immunolabeled by incubation overnight with the primary antibodies 7G6, a cone-specific antibody; SV2, a synaptic-vesicle antibody; and opsin antibodies that recognize either the short or long/middle wavelength–sensitive opsins.

results. The retinal margin cells labeled by 7G6 lay within 1 mm of the ora serrata and differed from 7G6-labeled cones in the central retina. The margin cells possessed a soma, a fiber process, and a terminal enlargement that lay in the plane of the retina; no outer segment was discernible. A total of 5400 and 7252 margin cones cells were found in each of two monkeys. The terminal enlargement and soma of the labeled margin cells also showed SV2 immunoreactivity. Surprisingly, opsin immunoreactivity extended throughout the margin cell, which is consistent with the absence of a discernible outer segment.

conclusions. Cells with immunoreactive cone properties were found in the margin of the monkey retina. The absence of an outer segment and the presence of somatic opsin and SV2 are reminiscent of features observed in the central cones of fetal monkey retinas. These results suggest that a subpopulation of cones in the retinal margin might fail to mature completely and thus retain juvenile characteristics into adulthood.

The margin of the retina in vertebrates is the last retinal area to differentiate during development, and in a number of cold-blooded vertebrates, the cells in this area retain embryonic features such as the ability to proliferate. ^{1 2} In fish, where eye growth continues throughout adulthood, new cells are added at the retinal margin. In adult newts, which are capable of retinal regeneration, a major site of proliferation is the retinal margin. ³ Thus, in cold-blooded vertebrates the margin retains progenitor cells that can be stimulated to divide in the adult. The developmental potential of retinal margin cells in primates has received less attention. In primates, as in fish and amphibians, the margin of the retina becomes increasingly thin, and the characteristic laminar organization of the retina is progressively degraded. In addition, the cellular composition and interconnections of the margin remain poorly defined. In monkey retinal whole mounts treated with a cone-selective antibody, we detected a population of cells that possessed immunoreactive properties of central cones but was oriented differently in that the cells lay in the plane of the retina, had their long axis parallel to the retinal surface, and lacked a distinct outer segment. ⁴ The present article reports the distribution of these conelike cells that we will refer to as margin cones and the presence of other immunologic markers found in central retinal cones. 

Materials and Methods

Tissue Preparation

Eyes were obtained from 18 rhesus macaque monkeys (Macaca mulatta) in adherence with the ARVO Statement on the Use of Animals in Ophthalmic and Vision Research. The monkeys ranged in age from 1 month to 19 years and were euthanatized after terminal experiments at the Yerkes Regional Primate Center of Emory University. No monkey was killed solely for the purposes of the experiments reported here. The globes were removed within 10 to 15 minutes of death and placed in Hanks’ balanced salt solution at 4°C. Retinas were removed within 1 hour of enucleation, separated from the retinal pigment epithelium, and fixed in 4% paraformaldehyde for at least 1 hour. Complete retinal whole mounts or retinal pieces were rinsed with phosphate-buffered saline (PBS) and then labeled with the antibody 7G6 in the presence of 0.1% Triton X-100 at 4°C overnight. Antibody 7G6 is a mouse monoclonal antibody known to label cones in the central retina. ⁵ The tissue was then incubated with secondary antibody from a Vectastain Elite ABC kit (Vector, Burlingame, CA) and treated according to the manufacturer’s instructions to visualize labeled cells. Other whole-mount preparations were double-labeled, either with 7G6 plus SV2, a mouse monoclonal antibody to synaptic vesicles ⁶ or with 7G6 plus rabbit antibodies that recognize the short wavelenth–sensitive (S) opsins or the long/middle wavelength–sensitive (L/M) opsins. ⁷ For double labeling with the two mouse primary antibodies, 7G6 and SV2, the 7G6 antibody was directly conjugated with Cy2, a green fluorescent probe (Amersham Pharmacia Biotech, Piscataway, NJ) following the manufacturer’s instructions. The retinal whole mounts were first incubated with SV2 in the presence of 0.1% Triton X-100 at 4°C overnight and then with rabbit anti-mouse IgG conjugated to Texas red for 2 hours. Next, the tissue was incubated with 7G6 directly conjugated to Cy2 in the presence of 0.1% Triton X-100 at 4°C overnight. For double labeling with 7G6 and the opsin antibodies, conventional indirect methods were used because the primary antibodies were from different species. The double-labeled retinal tissue was mounted on slides with the photoreceptors side up and viewed with a conventional microscope or with a confocal laser scanning microscope (Multiprobe 2001; Molecular Dynamics, Sunnyvale, CA) in the presence of antifade reagent (Bio-Rad, Hercules, CA). 

Results

7G6-Labeled Cells at the Retinal Margin

The antibody 7G6 labeled a population of cells with unique morphologic characteristics in the very far periphery or margin of the retina, that is, within 1 mm of the ora serrata (Fig. 1) . This distinct population of cells lay in the plane of the retina with their long axis parallel to the retinal surface, and frequently an entire cell was in the plane of focus within the whole-mount. This contrasts with cones in posterior retina, which lie more or less perpendicular to the retinal surface. Most of the labeled cells in the margin possessed a soma with no obvious outer segment, an elongated process, and a terminal enlargement. Other labeled cells seemed to consist only of a soma. These cells were as heavily labeled as cones in central retina. The intensity of the labeling of the margin cells suggests that they were related to central cones. Therefore, we will refer to these cells as margin cones. 

Numbers and Distribution of Margin Cones

Margin cones were observed in all 18 retinas examined. These retinas ranged in age from 1 month to 19 years. Typically, labeled cell bodies were not contiguous to one another, and the process and terminal enlargement lay in a position central to the cell body. Margin cones were more prominent in some retinas than in others, but their prominence did not seem to depend on age. We counted the 7G6-labeled cells in the retinal margin from two animals. The retinas of one animal (age 9 years) contained 2472 and 2928 margin cones, and of the other (age 4.5 years) 3457 and 3795 margin cones (Fig. 2) . The distribution of the margin cones within a single retina was not uniform, as shown by the numbers of cells in each quadrant of the whole mount (Fig. 2) , but no obvious pattern was discerned. 

Double-Label Studies with 7G6 and SV2 Antibodies

The presence of other markers associated with cones in the central retina was investigated with double-labeling experiments. The double-labeling procedure with the two mouse antibodies, 7G6 and SV2, gave excellent separation of labeling, as judged by the specificity of labeling of central retinal sections (not shown). Confocal microscopic images of double-labeled whole mounts of peripheral retina show the pattern of 7G6 labeling (Fig. 3A ) and of SV2 labeling of the same field (Fig. 3B) . The labeling with SV2 antibody was present not only in the terminal region but extended along the process and was even observed in the soma. When the labeled images are superimposed (Fig. 3C) and enlarged (Fig. 3D) , both labels overlapped most extensively in the terminal region. Therefore, the margin cones, identified by 7G6 labeling, contain synaptic vesicle protein detected by SV2. 

Antibodies to Cone Opsins

The presence of cone opsins was investigated within the population of margin cones by double labeling with 7G6 and cone opsin antibodies. The vast majority of these cones expressed the L/M-opsin (Figs. 4A 4B) . In a survey of 10 fields that contained L/M-labeled cells, 92% (490/530) of the 7G6-labeled cells showed L/M antiopsin reactivity; no cells labeled with L/M-opsin antibody were negative for 7G6. The L/M-opsin antibody was clearly distributed throughout the soma, process, and terminal enlargement. As reported earlier, the absence of an outer segment from these margin cones was a consistent feature. In central cones, the labeling with opsin antibody was highly localized in the outer segment (not shown). In contrast, double labeling with 7G6 and S-opsin antibody indicated a small proportion of the margin cones expressed S-opsin (Figs. 5A 5B) . In a survey of 10 fields, 2.7% (11/401) of the 7G6-labeled margin cells were also labeled with the S-opsin antibody. No cells positive for the S-opsin antibody were negative for 7G6. Like the labeling with the L/M-opsin antibody, the labeling with S-opsin antibody was found on the soma, process, and terminal enlargement. 

Discussion

The cone-selective antibody, 7G6, identified a population of cells in the margin of the monkey retina that was specifically and intensely labeled, as were cones in the central retina. Given the high specificity of 7G6 for central cones, we conclude that these margin cells are related to central cones. The long axis of the margin cone lay in the plane of the retina and was orthogonal to that of central cones. The presence of these margin cones was detected in retinas that ranged in age from 1 month to 19 years. On the basis of this observation, we conclude that these cells are present throughout the life of the retina and do not represent an aging process, ⁸ given their presence in young retinas. The high specificity of 7G6 for cones facilitated the identification of the margin cones. In the flat-mount preparations, these cells have a striking appearance and seem healthy at the light microscopic level. The antigen is one of the earliest found in cones in central retina, ⁵ and efforts are underway to identify it. 

In addition to 7G6 labeling, margin cones shared other properties with cones in central retina. Both cell populations are labeled with SV2, but the labeling was not as highly localized in the margin cones. In the central retina, SV2 labeling is confined to the terminal region. In developing cones, ^{9 10} in addition to labeling of the terminal, diffuse labeling was observed in the soma. The finding of SV2 labeling in peripheral cones within the age group in this study, 1 month to 19 years, prompted the notion that the margin cones might be halted at a stage of development. 

The labeling of margin cells with the antibodies to cone opsins also revealed similarities and differences with central cones. The main difference was the diffuse pattern of labeling in the margin cones compared with that in mature central cones, where labeling is almost exclusively in the outer segment. The diffuse labeling by opsin antibodies was, however, similar to that seen during retinal development ^{10 11 12} and reminiscent of SV2 labeling described above. Diffuse visual pigment labeling has also been observed in association with disease states. Milam et al. ¹³ reported widespread delocalization of rhodopsin in rods obtained from patients with retinitis pigmentosa. It seems that mutations in rhodopsin and/or the absence of a mature or intact outer segment can cause massive rerouting of visual pigment to compartments other than the outer segment. Another similarity lay in the proportion of peripheral cones expressing L/M- or S-opsin. The percentage of 7G6-labeled cells that showed L/M-opsin antibody labeling was similar to that reported for the central retinal cones, roughly 90%. The percentage of 7G6-labeled margin cones that showed S-opsin antibody labeling was just under 3%, again roughly mirroring the proportion of S-opsin central cones. Labeling with L/M- or S-opsin antibodies, however, accounted for only approximately 95% of the 7G6-labeled cones. A proportion of the remaining 5% might have arisen from failure to count cells that were lightly labeled with the opsin antibodies or from the presence of cones that were truly negative for opsins. If a population of L/M- and S-opsin–negative cones did exist, the question that arises is whether or not this population remained permanently negative or represented progenitor cells that eventually expressed one of the opsins. 

Williams ¹⁴ reported the presence of a band of cones in the margin of the human retina in a similar position as the margin cones that we identified in the monkey retina. We wondered whether the human cells are counterparts of the monkey cells. Williams identified the human margin cones on the basis of cell size and labeling with opsin antibodies, whereas we relied on 7G6 labeling. Two major differences in the two cell types are the presence of the outer segment and the lack of opsin labeling in the soma of the human margin cones. With our current information, it is difficult to reconcile these differences. It might be necessary to immunolabel the human retina with 7G6 to provide further insight into the relationship between the band of cones in humans and the margin cones in monkeys. 

We have shown that the margin cones possess essential parts of the transduction machinery, the opsins and, possibly, the molecular apparatus for neurotransmitter storage and release as detected by SV-2 labeling. Whether the margin cones are capable of transducing light and transmitting light information to other retinal cells warrants further investigation. 

 Supported by grants RO1 NS-35510 (PRM), U54 NS-34194 (PRM), and RCMI grant 5G12RR03034.

 Submitted for publication March 2, 1999; revised November 2, 1999 and January 11, 2000; accepted February 2, 2000.

 Commercial relationships policy: N.

 

Figure 1.

 

Whole-mount labeling of retinal margin cells with 7G6, a cone-specific antibody. A population of heavily labeled cells can be seen at the retinal margin. The cells lie in the plane of the retina with their long axis parallel to the retinal surface. Most of the labeled cells possess a soma without an obvious outer segment, a fiber process, and a terminal enlargement. Some cells show only the soma. Calibration bar, 20 μm.

Figure 2.

 

Distribution of 7G6-labeled margin cones. Whole retinas were labeled with 7G6, cut as shown, and mounted on slides. The total number of labeled cells is shown for each quadrant. Margin cones were found within 1 mm of the periphery. Local variations in the density of cell labeling was observed. F, fovea; O, optic nerve head.

Figure 3.

 

Confocal microscopic images of the same field of a retinal whole-mount preparation showing margin cones double-labeled with 7G6 and SV2, an antibody against synaptic vesicle protein. (A) Cells labeled with 7G6 directly conjugated to Cy2 (green). (B) Cells labeled with SV2 followed by rabbit anti-mouse secondary antibody conjugated to Texas red. Heavy labeling is seen in the terminal region and in the distal part of the fiber process. When present, the labeling in the perinuclear region was moderate. (C) The superimposed images from (A) and (B). Areas that are double-labeled with 7G6 and SV2 appear yellow. Calibration bar, 20 μm. (D) Enlargement of the double-labeled cell shown inside the box in (C). Calibration bar, 5 μm.

Figure 4.

 

Confocal microscopic images of a whole-mount preparation of retinal margin tissue double-labeled with 7G6 and L/M-opsin antibody. Those two antibodies were from different species and were applied at the same time; they were followed by anti-mouse IgG conjugated to Texas red and anti-rabbit IgG conjugated to FITC. L/M-opsin antibody labeling (B) was seen in most of the 7G6-labeled cells (A). Note that the entire cell is labeled with the L/M-opsin antibody. The asterisk shows cells labeled with 7G6 but not with L/M-opsin antibody. Calibration bar, 20 μm.

Figure 5.

 

Confocal microscopic images of a whole-mount preparation of retinal margin tissue double-labeled with 7G6 and S-opsin antibody. The double-labeling procedure was same as that described for Figure 4 . Labeling with S-opsin antibody (B) was seen in a small subpopulation of 7G6-labeled cells (A). The asterisk shows cells that were labeled with 7G6 but not with S-opsin antibody. Calibration bar, 20 μm.

We thank the Yerkes Regional Primate Research Center of Emory University for providing monkey eyes. We also thank Kathy Buckley, PhD, for the kind gift of SV2 antibody and Jeremy Nathans, MD, PhD, for the generous gift of the opsin antibodies.