Survival and Synapse Formation of Transplanted Rat Rods

Isolated rods enzymatically removed from normal adult rat retina have been transplanted to the subretinal space of adult rats with a retinal dystrophy winich has destroyed almost all the photoreceptors. These transplanted rods survive for months after transplantation during which time they form synapses with other retinal cells. Rod spherules with large amounts of synaptic vesicles and synaptic ribbons are found forming discreet contacts with pre- and postsynaptic densities in arrangements closely resembling those seen in the normal retina.


INTRODUCTION
Solitary photoreceptors /11,14,18,19/ can be enzymatically isolated from the adult retina and maintained in vitro where they remain responsive to light /1/ and in the ease of salamander rods /16/ form synapses with second-order retinal neurons. We have used this method to isolate solitary photoreceptors, predominantly rods, from normal pigmented adult rats and transplanted them to the subretinal space of 4to 6-month-old albinotic rats with an inherited retinal dystrophy, which causes destruction of virtually all the photoreceptors at about 3 to 4 months of age/2,5/.
In order to easily identify the transplant site we also dissociated pigmented retinal epithelial cells from normal rats and added them to the solution of photoreceptor cells just prior to surgery. These pigmented cells provide a marker for the transplant site because the dystrophic rat retina is albinotic and therefore the pigmented cells stand out within the small retinal area, about a millimeter in diameter, receiving the transplanted cells. The retinal dystrophy in this strain of rats is due to a genetic defect that prevents the retinal epithelium from phagocytozing the growing tips of the rods and this subsequently leads to VOL. 2, NO. 2, 1991 91 their destruction /6,10/. Therefore introducing normal retinal epithelial cells not only serves as a marker but may also facilitate the survival of the transplanted photoreceptors/9,13/. MATERIALS  The globes were hemisected at the ora serrata. The lens and vitreous were removed. The neural retina was then peeled away from retinal pigmented epithelium and placed in Hank's solution without Mg +/ and containing papain (5 units/ml, Worthington Biochemical, Freehold, NJ), cysteine (2.7 mM) and 40 mM glucose. The tissue was intermittently agitated for 10 minutes in a 95% O:/5% CO: environment and then gently triturated with a wide base glass pipette releasing dissociated photoreceptors into the solution. The cell suspensions obtained were examined periodically by phase contrast microscopy to determine optimum dissociation and purity. In order to establish the purity of our solutions of transplanted photoreceptors we centrifuged and pelleted the dissociated material and examined it by electronmicroscopy. This revealed only rod photoreceptor organelles, including outer and inner segments, nuclei, all typical of rods together with rod axons and terminals. We found no evidence of other cells and no synapses contaminating these samples.

Isolation of Pigmented Retinal Epithelial Cells
The methods to dissociate retinal epithelial cells are similar to those we have used previously/8,9/. The pigmented retinal epithelial cells were obtained from the bisected normal rat eye by removing the neural retina and submerging the eye cup in Hank's solution, Ca / / and Mg + + free, and treated with 0.25% trypsin (GIBCO, Grand Island, NY) for 1 hr at 37C. The trypsin solution is removed and replaced with minimum essential medium (MEM) solution. The retinal epithelial cells were pipetted off Bruch's membrane. An aliquot of the cell solution is used to determine the concentration and purity of the dissociated epithelial cells using a hemocytometer. The cell solution is centrifuged at 1000 RPM for 5 minutes and the supernatant removed in order to concentrate the cells.

Transplantation of Dissociated Cells
The cells are transplanted with a glass micropipette having a tip diameter of 120-130 microns as described previously /7,8,9,15/. This pipette is filled with a balanced salt solution and then a small air bubble to keep the transplant solution from being diluted into the larger volume of saline in the pipette. The pipette is then used to aspirate a small amount, about 5 microliters, of a solution of dissociated retinal pigmented epithelial cells, prepared earlier and subsequently a similar volume of solution containing the dissociated photoreceptors. The micropipette is then introduced into the subretinal space transchoroidally through a small incision in the sclera at the equator of the eye of an anesthetized fourto five-month-old dystrophic rat and the cell solution up to the air bubble in the micropipette injected into the subretinal space. One eye receives the transplanted cells; the other eye serves as a control. Because of its pigmentation, the transplant site can be examined immediately or at a later date by ophthalmoscopy through the dilated pupil.

Histology
The rats were euthanized with an overdose of nembutal at different times after transplantation surgery. The eyes were removed and incised at the limbus, placed in 3% glutaraldehyde buffered Earle's solution and kept at 40C for 24-48 hours. Then they were washed in buffer and dissected with the aid of a surgical microscope in order to identify the transplant site by finding the intraretinal pigmentation produced by the co-transplanted retinal epithelium. This area was cut out, dehydrated, embedded in epoxy resin and sectioned for examination by light and electronmicroscopy. A similar section from the control eye was processed in parallel. We have examined 41 eyes which received transplanted rods: 33 within 1-2 weeks and 8 within 1-2 months after transplantation surgery. In earlier experiments both eyes were used for transplantation; in later ones the contralateral eye served as a control. In the former, our control was an adjacent area of retina which had not received any transplanted cells. Each block was sectioned serially until the pigmented epithelial cells were found by light microscopy, usually together with a group of rod cell nuclei (see Fig. 2). At this point ultra thin sections were taken and examined by electronmicroscopy. Electronmicrographic photomontages were made of the area and examined for rod organelles.' outer and inner segments, nuclei and synaptic terminals and vesicles, synaptic ribbons and contacts. These were compared with similar montages from the control retinas.
The rats used in these experiments are from a congenic Royal College of Surgeons strain of rats, tanhooded pink eyed, homozygous (p/p) for the pink eye dilution gene as recipients and pigmented (p/+) rats as donors, obtained through the courtesy of M. LaVail. Figure 1 illustrates by light microscopy normal (A) compared to four-month-old dystrophic (B) retina. In the dystrophic retina the entire photoreceptor layer has disappeared leaving the neural retina at about half the thickness of the normal retina. This is most obvious from the absence of the darkly staining photoreceptor nuclei, about 7 to l0 layers thick, as well as the loss of inner and outer segment organelles and the external plexiform layer where the photoreceptors synapse with other retinal neurons. In the dystrophic retina all of these structures are lost. Figure 2 shows transplant sites at one (A) and two (B) weeks and at one (C) and two (D) months after transplantation surgery. In all cases the rod cell nuclei found in these areas marked by the transplanted pigmented epithelial cells are much more numerous than anything we can find elsewhere in this retina or in the contralateral control eye (Fig. 1B). In some experiments ( Fig. 2A)there are as many as three rows of rod cell nuclei in the same place where the pigmented epithelial cells are located. Elsewhere in the retina, these nuclei and pigmented cells were absent. The relatively thin neural retina invariably reattaches to the underlying epithelial layer. Sometimes some of the host retinal epithelial cells are dislodged by the stream of the transplantation solution and pigmented donor epithelial cells insert themselves in this layer ( Fig. 2C and D). There is little indication of outer segment material at the light microscopic level within the retina. There is also no evidence of inflammation or rejection of the transplant material.

RESULTS
Electronmicroscopy of the transplant site shows typical rod cell nuclei with some outer segment material (Fig. 3). The cytoplasm of these cells appears healthy, in Figures 3 and 4 at one month after transplantation. Rod cell nuclei, if any, cannot be found in these numbers in the control retinas. The melanin granules within the retinal pigmented epithelial cell, the cytoplasm of which is visible in the upper left corner of Figure 4 (arrow), are unequivocal evidence that this is the transplant site.
Outer segments are relatively rare in the transplant sites especially at one to two months after transplantation surgery. The best example we have found is shown in Figure 5. This shows a relatively short outer segment connected to its ciliary stalk and rootlet in a seven-month-old dystrophic rat. Interestingly, a melanin granule identifies the cytoplasm of a transplanted retinal pigment epithelial which abuts the outer segment. This close association of a retinal epithelial cell to the transplanted outer segment could be responsible for its longer survival than those of other transplanted rods. Outer segment material has never before been seen by us or reported by others in dystrophic rats of this age/12/.
The most striking characteristic of the transplant site, especially at one or two months after surgery, is the abundance of synaptic structures, i.e., vesicles, synaptic ribbons and pre-and postsynaptic densities present. Figure 6 illustrates examples of such synapses cut in both radial and tangential directions through rod spherules in a six-and-a-half month-old dystrophic rat retina, two months after photoreceptor transplantation. The postsynaptic structures do not have the cytoplasmic characteristics of either photoreceptor or pigmented retinal epithelial cells, having a relatively clear cytoplasm with either a few or no vesicles.
Measurements of the number of ribbon synapses that could be identified in electronmicrographic photomontages at different times after transplantation show an increase from 3/1000 pm 2 at 1-2 weeks to 19/1000/.tm 2 at 1-2 months after transplantation. We could find no synapses within the outer plexiform layer of the control retinas. On the other hand, the number of outer segments we observed decreased progressively with time after transplantation.

DISCUSSION
The results indicate that dissociated rods .from adult rats can survive in the subretinal space of the dystrophic rat retina for at least two months. The evidence that they are transplanted cells is the These results are an evolution of earlier research demonstrating that photoreceptors, isolated from the retina, remain functional/11,18,19/ and even form synapses/16/in vitro. The formation of synapses on host neurons is crucial to advancing the use of this method to examine retinal cell biology as well as possibly correcting those retinal degenerations which uniquely destroy photoreceptors but leave the rest of the retinal machinery for vision relatively intact.
Another crucial factor for making this technique potentially therapeutic is the maintenance of normal outer segments. We found that outer segments decreased in number with time after transplantation. This may be due in part to the poor alignment of these organelles with the retinal epithelium which is a drawback of using dissociated cells. The retinal epithelium in the dystrophic retina is itself defective and this may add to the problem in the particular animal model we have used. Our choice of this model was because it contains a retina in which the photoreceptor cells have disappeared. Therefore, identifying the transplanted rod cells was enormously facilitated.
The use of solutions of dissociated retinal epithelial /7/or neuroretinal/20/cells has advantages in transplantation. Rather than using segments or slices of organized tissue, solutions of cells can be used and injectedwith micropipettes or cannulas. This minifies the surgical instrumentation and reduces the trauma required to reach a target in the central nervous system. It also facilitates microscopic examination of an aliquot from the cell suspension prior to trans-plantation, which can provide insights into the viability and purity of the inoculum.
The use of single cell types, as done in this study, seems particularly attractive to neuro-transplantation because in the nervous system it is the number and organization of specific cell types that determine function. The photoreceptor is a particularly specialized neuron which is frequently selectively damaged by genetic defects and/or toxic factors. There are a number of retinal degenerations where destruction of this cell system alone prevents the remainder of the visual system from functioning. Under these circumgtances photoreceptor transplantation could become a feasible treatment.