Age-related macular degeneration is one of the most serious diseases in elderly people because of its disasterous visual outcome and its prevalence. Even if the submacular and choroidal neovascular membranes could be surgically excised, severe damage or evacuation of retinal pigment epithelium is inevitable in the operated area. Pigmentary dystrophy is also a devastating hereditary eye disease with severe visual disturbance. Up to now, there have been no effective treatments for either of them. We conducted basic experiments on retinal pigment epithelium (RPE) culture, transplantation of the cells to the subretinal space of animals, especially, the Royal College of Surgeon's (RCS) rat, a model of hereditary retinal degeneration, and observed their effects in preventing photoreceptor cell death. 1) We reviewed recent reports of RPE function in relation to cytokine production and autocrine/paracrine function of these ligands. Some cytokines with strong mitogenic effects as nerve trophic/growth factors were able to rescue photoreceptor cell death in dystrophic, ischemic, and light-damaged retinas in the rats. We transplanted allograft pigmented RPE from Long Evans rats or xenograft, human and bovine RPE into the subretinal space of RCS rats, and could observe the retardation of the photoreceptor cell death. 2) As a source of human transplantable RPE in clinical practice, we could use patients' own RPE cells as autografts or those from aborted human fetus eyes as allografts. At present, we cannot use RPE cells from different species as xenografts. We tried to obtain enough RPE cells for culture in vitro from patients with large or giant retinal tears, but were unsuccessful. Cells were easily obtained from fetus eyes, and could be cultured and transplanted as fresh, primary, or multiple passage cells. We also tried cryopreservation of these cells for up to 3 months. Enzymatic expression of tyrosinase, tyrosinase related protein I and II and some other enzymes was examined by proliferating chain reaction to detect possible transformation during the procedure. The cell characteristics were well preserved. In the future, if these RPE cells could be safely kept and available in deep-frozen condition, we could use them clinically at the appropriate time and in appropriate numbers for patients as an "RPE bank" just like an "eye bank" for corneal transplantation. 3) Immunological reaction is very important if we consider this technique for clinical application. Up to now, in experimental animals, no immunological reaction has been reported even for xenograft human RPE in rats, in funduscope and histological examination, because the intraocular space is an immunologically privileged site. But transplantation of human RPE cells with a collagen sheet into the anterior chamber in rabbits caused a definite reaction detected by suppression of the electroretinogram and macrophage infiltration into the subretinal space, not only in the operated eye but also in the contralateral non-operated eye. These results suggest that we must be cautious in clinical use of heterogeneous RPE transplantation. The expression of MHC class II cells was observed in the course of photoreceptor cell degeneration in the RCS rats but it was suppressed if they were rescued by the transplantation of human cultured RPE in these animals. 4) For clinical application of this technique, autografts are naturally much better than the xeno grafts or allografts. We tried to use iris pigment epithelium (IPE) for transplantation because it consists of pigmented cells of neural origin and enough could be obtained with ease by peripheral iridectomy. We also tried transfection of a vector (pCNX2) or vector-inserted cDNA of rat bFGF into the rat IPE and transplanted into the subretinal space of RCS rats. These transfected cells expressed strong mRNA of bFGF. The photoreceptors were well preserved and immunological reaction could not be detected by funduscopical or histological examinat