Purpose

To evaluate retinal sheet transplants in S334ter-line-3 retinal degenerate rats by comparing visual responses recorded electrophysiologically with morphology based on light and electron microscopy.

Methods

S334ter-line-3 retinal degenerate rats (n = 7) received retinal sheet transplants between postnatal days 28 and 31. The donor tissue was derived from transgenic embryonic day 19 (E19) rat retinae expressing human placental alkaline phosphatase (hPAP). Fresh retinal sheets were gently transplanted into the subretinal space of the left eye with the help of a custom-made implantation tool. Selected rats (n = 5) were subjected to electrophysiologic evaluation of visual responses from the superior colliculus about 84-121 days after surgery. Transplanted eyes were processed for light microscopy (LM) and electron microscopy (EM) evaluations.

Results

All the transplanted rats that were evaluated for visual responses in the brain showed responses to very low light stimulation (-3.42 to -2.8 log cd/m(2)) of the eye in a small area of the superior colliculus corresponding with the placement of the transplant in the host retina. Histologic evaluation showed that most of the transplants contained well-laminated areas with correct polarity in the subretinal space. Inside the transplant areas, rosettes of photoreceptors with inner and outer segments were found. In the laminated areas, the outer segments of photoreceptors were facing the host retinal pigment epithelium (RPE). Immunohistochemical evaluation of hPAP donor cells revealed areas with specific staining of the transplants in the subretinal space. Electron microscopic evaluation showed a glial demarcation membrane between the host and the transplant, however, processes originating from the transplant were observed inside the host retina.

Conclusions

Sheets of E19 rat retina transplanted into the subretinal space of S334ter-line-3 rats survived without immune rejection and continued to show visual function when tested after 3 months. Well-developed photoreceptors and many synapse types were seen within the transplants. hPAP staining showed a certain degree of integration between the host retina and the transplant suggesting that transplanted photoreceptors contributed to the restored light sensitivity.

This study aimed to test the hypothesis that visual responses in the superior colliculus (SC) originate from synaptic connections between fetal retinal transplants and degenerating host retinas. Sheets of embryonic day 19 rat retina expressing human placental alkaline phosphatase were transplanted to the subretinal space of 3- to 4-week-old S334ter-line-3 rats with fast retinal degeneration. Several months later, visual responses were recorded from the SC. Attenuated pseudorabies virus that is specifically transferred between neurons at synapses (strains PRV-152, expressing green fluorescent protein (GFP) or BaBlu, expressing Escherichia colibeta-galactosidase) was injected into the visually responsive site of the SC. After survival times of 1-2 days, the virus was detected in the retina by immunohistochemistry in combination with different retinal cell markers, such as protein kinase C, recoverin, calcium-calmodulin-dependent protein kinase II and glutamine synthetase. Transplanted rats had a mean response threshold of -3.1 log cd/m(2) in a small area of the SC corresponding to the location of the graft in the retina. By 30 h after injection into this SC area, the virus traced back to host ganglion cells overlying the transplant and in close proximity to the transplant. By 2 days after injection, extensive virus label was found in the host retina and many cells in the transplant were also labeled. Virus-labeled cells in the transplant were double labeled for neuronal and glial cell markers. This study provides anatomical evidence that synaptic connections between fetal retinal transplants and host retinas contribute to the visual responses in the SC.

Optical coherence tomography (OCT), a non-invasive method, was used for qualitative assessment of fetal retinal sheet transplants by non-invasive imaging. Rhodopsin-mutant S334ter-line-3 rats with fast retinal degeneration (28-37-day old) were transplanted with fetal retinal sheets from embryonic day (E) 18-19 pigmented normal rats. Retinal thickness measurements from transplanted (n = 51), no surgery control (n = 8), and normal pigmented rat eyes (n = 6) were obtained using a Zeiss stratus OCT-3 scanning instrument. Frozen retinal sections were stained with hematoxylin/eosin. S334ter-line-3 rats showed significant reduction in OCT retinal thickness (p<0.001) compared to normal pigmented rats at the age of 21 days. In 62% of the transplanted rats, OCT scanning revealed the presence of a subretinal graft, which was confirmed by subsequent histology. Retinal thickness in the transplant area was significantly increased compared to the area outside the transplant and to non-transplanted eyes (p<0.001). While most of the transplants with single-band OCT images (87%) had rosetted transplants, a considerable proportion of transplants having a multi-band OCT image were found to have well-laminated areas in the graft after histological evaluation. Following retinal transplantation in rodents, OCT imaging data correlated mostly with transplant morphology. OCT is a useful technique for in vivo screening and evaluation of retinal transplants. This technique determines surgical outcomes at a much earlier stage.

A visual discrimination apparatus was developed to evaluate the visual sensitivity of normal pigmented rats (n=13) and S334ter-line-3 retinal degenerate (RD) rats (n=15). The apparatus is a modified Y maze consisting of two chambers leading to the rats' home cage. Rats were trained to find a one-way exit door leading into their home cage, based on distinguishing between two different visual alternatives (either a dark background or black and white stripes at varying luminance levels) which were randomly displayed on the back of each chamber. Within 2 weeks of training, all rats were able to distinguish between these two visual patterns. The discrimination threshold of normal pigmented rats was a luminance level of -5.37+/-0.05 log cd/m(2); whereas the threshold level of 100-day-old RD rats was -1.14+/-0.09 log cd/m(2) with considerable variability in performance. When tested at a later age (about 150 days), the threshold level of RD rats was significantly increased (-0.82+/-0.09 log cd/m(2), p<0.03, paired t-test). This apparatus could be useful to train rats at a very early age to distinguish between two different visual stimuli and may be effective for visual functional evaluations following therapeutic interventions.

The aim of this study was to evaluate the functional efficacy of retinal progenitor cell (RPC) containing sheets with BDNF microspheres following subretinal transplantation in a rat model of retinal degeneration. Sheets of E19 RPCs derived from human placental alkaline phosphatase (hPAP) expressing transgenic rats were coated with poly-lactide-co-glycolide (PLGA) microspheres containing brain-derived neurotrophic factor (BDNF) and transplanted into the subretinal space of S334ter line 3 rhodopsin retinal degenerate rats. Controls received transplants without BDNF or BDNF microspheres alone. Visual function was monitored using optokinetic head-tracking behavior. Visually evoked responses to varying light intensities were recorded from the superior colliculus (SC) by electrophysiology at 60days after surgery. Frozen sections were studied by immunohistochemistry for photoreceptor and synaptic markers. Visual head tracking was significantly improved in rats that received BDNF-coated RPC sheets. Relatively more BDNF-treated transplanted rats (80%) compared to non-BDNF transplants (57%) responded to a "low light" intensity of 1cd/m2 in a confined SC area. With bright light, the onset latency of SC responses was restored to a nearly normal level in BDNF-treated transplants. No significant improvement was observed in the BDNF-only and no surgery transgenic control rats. The bipolar synaptic markers mGluR6 and PSD-95 showed normal distribution in transplants and abnormal distribution of the host retina, both with or without BDNF treatment. Red-green cones were significantly reduced in the host retina overlying the transplant in the BDNF-treated group. In summary, BDNF coating improved the functional efficacy of RPC grafts. The mechanism of the BDNF effects--either promoting functional integration between the transplant and the host retina and/or synergistic action with other putative humoral factors released by the RPCs--still needs to be elucidated.

Retinal degenerative conditions increase susceptibility to light damage, but rapid retinal degeneration (RD) models show less susceptibility to cyclic dim light. We investigated whether constant blue light (BL) exposure can eliminate the residual visual responses in a comparatively rapid RD rat model. Pigmented rhodopsin mutant S334ter line-3 rat pups (21 days old) were exposed for 5-6 consecutive days to constant BL. Visual behavior was evaluated with an optokinetic head tracking apparatus. Electrophysiological recordings were made from the superior colliculus (SC). S-antigen, red-green opsin and rhodopsin immunoreactive residual photoreceptors were counted. Following BL exposure, head tracking was significantly reduced at 0.25 cycles degree(-1) in 38-day-old line 3 rats. With a 0.125 cycles degree(-1) stimulus, the head tracking performance of 80-day-old BL rats were similar to that of 220-day-old no-BL-treated line-3 rats. SC recordings also revealed a significant decrease in the residual photoreceptor activity. Histological evaluation showed reduction of the rod population in the central area of the light-damaged retina. Exposure to constant BL considerably reduces the residual visual responses in a rapid degenerating RD rat model.