Retinal S antigen identified as the 48K protein regulating light-dependent phosphodiesterase in rods

Amplification of phosphodiesterase activation is greatly reduced by rhodopsin phosphorylation

In the vertebrate rod outer segment (ROS), the light-dependent activation of a GTP-binding protein (G-protein) and phosphodiesterase (PDE) is quenched by a process that requires ATP [Liebman, P.A., & Pugh, E.N. (1979) Vision Res. 19, 375-380]. The ATP-dependent quenching mechanism apparently requires the phosphorylation of photoactivated rhodopsin (Rho*); however, a 48-kilodalton protein (48K protein) has also been proposed to participate in the inactivation process. Purified species of phosphorylated rhodopsin containing 0, 2, or greater than or equal to 4 (high) phosphates per rhodopsin (PO4/Rho) were reconstituted into phosphatidylcholine (PC) vesicles and reassociated with a hypotonic extract from isotonically washed disk membranes that were depleted of 48K protein; PDE activation, in response to bleaching from 0.01% to 15% of the rhodopsin present, was measured. PDE activity was reduced by at least 30% at high fractional rhodopsin bleaches and by greater than 80% at low fractional rhodopsin bleaches in high PO4/Rho samples when compared to the activity measured in O PO4/Rho controls. A phosphorylation level of 2 PO4/Rho produced PDE activities that were intermediate between O PO4/Rho and high PO4/Rho samples at low bleaches, but were identical with the O PO4/Rho samples at high rhodopsin bleaches. Rhodopsin phosphorylation is thus capable of producing a graded inhibition of light-stimulated PDE activation over a limited range of (near physiological) bleach levels. This effect become less pronounced as the bleach levels approach those that saturate PDE activation. These results are consistent with increasing levels of phosphorylation, producing a reduction of the binding affinity of G-protein for Rho*.

Retinoblastoma. A clinical, immunohistochemical, and electron microscopic case report

A 4-year, 9-month-old boy had a history of leukocoria of the right eye for approximately six months prior to admission. The other eye was normal. There was no family history of retinoblastoma. Funduscopy disclosed a large white mass extending from the nasal pars plana to the mid-pupillary zone and the posterior pole with a near total retinal detachment in the superior temporal quadrant. A B-scan ultrasound showed an echo dense area of the anterior portion of the mass. A CT scan showed intraocular tissue densities with no evidence of optic nerve involvement or extraocular extension. Immunohistochemistry of a fresh frozen portion of tumor revealed reactivity with antibodies directed against interphotoreceptor retinoid-binding-protein (IRBP), neuron-specific enolase, glial fibrillary acidic protein, S-antigen, focal reactivity with opsin, and scattered cytoplasmic staining for cyclic GMP and cyclic GMP phosphodiesterase. Biochemical analysis of fresh frozen tissue samples confirmed the presence of IRBP. Transmission electron microscopy disclosed occasional Flexner-Wintersteiner rosettes connected by zonula adherens-like junctions. These showed inner segment-like structures containing prominent mitochondria, portions of cilia and fragments of outer segment material. These data, along with the immunocytochemistry indicates a predominant neuronal nature of the tumor cells with significant photoreceptor-like differentiation.

Rabbit ocular and pineal autoimmune response to retina antigens

Different forms of experimental autoimmune uveitis (EAU) can be produced by varying protocols to present different autoantigens to several species of experimental animal. We have studied the clinical, histological and serological responses of rabbits to footpad injection of various fractions of retina extract. Rabbits injected with retina extract or S antigen developed posterior uveitis. However, rabbits injected with retina extract, also developed an anterior uveitis and pinealitis not seen in rabbits receiving S antigen. The Serological response of rabbits to retina extract was different than that to purified S antigen. Antisera of rabbits receiving retina extract reacted with rabbit retina and pineal gland as well as with guinea pig retina but not with guinea pig pineal gland. In contrast anti-S antigen sera reacted with rabbit retina and guinea pig retina and pineal gland but not with rabbit pineal gland. Gel filtration chromatography of the ammonium sulfate supernate of retina extract was used to differentiate the antigens with which these two sera reacted. An analysis of these experiments gives preliminary evidence of an autoantigen(s) of rabbit retina and pineal gland that is not S antigen. The existence of multiple autoantigens common to retina and pineal gland in various species is significant in that it further underscores the relationship of these tissues. Furthermore, it is not unrealistic to expect more than one autoantigen of retina or uvea to be involved in autoimmune uveitis.

Melatonin: parallels in pineal gland and retina

It is apparent that several relationships exist between the pineal gland and retina. The similarities in development and morphology have been obvious for many years. A recent resurgence of interest in this field has led to a further understanding of many functional similarities between these two organs. A notable feature of the pineal gland and retina is their common ability to synthesize the indolamine hormone, melatonin. Many investigators suspect that the cyclic rhythm of retinal melatonin synthesis may be related to other cyclic events which normally occur in the retina.

Codistribution of S-antigen and opsin in light-adapted retinal rods

The localization of S-antigen in ultra-cryosections of rat retina has been studied and compared with that of opsin by immunoelectron microscopy. Monospecific antisera to both proteins, and Protein A-coated gold particles of two sizes were used for (double) labeling. In the light-adapted retina, S-antigen showed codistribution with opsin in the rod outer segments. S-antigen labeling was diminished in dark-adapted retina. In retina pigment epithelium, multi-layered bodies were found containing S-antigen and rhodopsin representing spent rod outer segment fragments. In addition to theoretical considerations, there is indication that S-antigen is closely associated with the cytoplasmic surface of rod outer segment disc membranes. It seems moreover likely that S-antigen and opsin are closely associated in the illuminated retina. An important reason for this assumption was the observation that double labeling experiments showed interference of the antibodies to both antigens probably due to steric hinderance. The present results are complementary to, and in agreement with biochemical data from literature and point to an important function of S-antigen in the process of phototransduction.

Phosphodiesterase activation by photoexcited rhodopsin is quenched when rhodopsin is phosphorylated and binds the intrinsic 48-kDa protein of rod outer segments

Each photoexcited rhodopsin (R*) molecule catalyzes binding of GTP to many copies of the guanine nucleotide-binding protein transducin, which, in its GTP-binding form, then activates cGMP phosphodiesterase (PDEase). Subsequent deactivation of this light-activated enzyme cascade involves hydrolysis of the GTP bound to transducin, as well as decay of the activating capacity of R*. We report here that deactivation of PDEase in rod outer segment suspensions is highly enhanced by addition of ATP and purified 48-kDa protein, which is an intrinsic rod outer segment protein that is soluble in the dark but binds to photolyzed rhodopsin that has been phosphorylated by rhodopsin kinase and ATP [Kühn, H., Hall, S.W. & Wilden, U. (1984) FEBS Lett. 176, 473-478]. To analyze the mechanism by which ATP and 48-kDa protein deactivate PDEase, we used an ATP-free system consisting of thoroughly washed disk membranes, whose rhodopsin had been previously phosphorylated and chromophore-regenerated, and to which purified PDEase and transducin were reassociated. Such phosphorylated membranes exhibited a significantly lower (by a factor less than or equal to 5) light-induced PDEase-activating capacity than unphosphorylated controls. Addition of purified 48-kDa protein to phosphorylated membranes further suppressed their PDEase-activating capacity; suppression could be as high as 98% (as compared to unphosphorylated membranes), depending on the amount of 48-kDa protein and the flash intensity. Addition of ATP had little further effect. In contrast, PDEase activation or deactivation with unphosphorylated control membranes was not influenced by 48-kDa protein, even in the presence of ATP, provided rhodopsin kinase was absent. Our data suggest that 48-kDa protein binds to phosphorylated R* and thereby quenches its capacity to activate transducin and PDEase.

Induction of experimental autoimmune uveoretinitis and pinealitis by IRBP. Comparison to uveoretinitis induced by S-antigen and opsin

Microgram quantities bovine IRBP (interphotoreceptor retinoid binding protein) injected in Freund's complete adjuvant induced severe autoimmune uveoretinitis and pinealitis in Lewis rats. At low doses the onset was accelerated and intensified by co-injection of Hemophilus pertussis bacteria. Wistar, BN and PVG rats were less susceptible, while the eyes of athymic, nude rats did not respond. The disease developed similar to but faster than S-antigen-induced uveoretinitis, while its onset was one day earlier and the reactions were slightly more severe. As distinct from these two types of uveoretinitis, opsin (in much higher doses) caused milder reactions in the anterior segment, while retinitis dominated. In each type of inflammation the photoreceptor cell layer was totally destroyed. All three ocular diseases were inhibited by cyclosporine treatment, which indicates that T cell-dependent mechanisms are essential for the development.

Pinealocyte projections into the mammalian brain revealed with S-antigen antiserum

Neural processes from mammalian pinealocytes have been discovered in several brain areas. These processes were visualized immunocytochemically in the Djungarian hamster, Phodopus sungorus, with an antiserum against bovine retinal S-antigen and traced as far as the region of the posterior commissure and habenular nuclei. This result indicates that pineal-to-brain connections exist in the mammal, and that the mammalian pineal gland, currently thought of only as a neuroendocrine organ, may communicate directly with select brain regions by way of these projections. The existence of mammalian pinealocyte projections is consistent with the view that these cells are not of glial origin but are derivatives of photoreceptor cells of the pineal complex of lower vertebrates that transmit signals to the brain by neural projections.

Sequence analysis of bovine retinal S-antigen. Relationships with alpha-transducin and G-proteins

S-Antigen is a major soluble protein of the retina and pineal. It is capable of inducing experimental autoimmune uveitis (EAU) in laboratory animals and also seems to play an important role in the visual cycle. The results of partial cDNA sequence analysis reveal interesting homologies with alpha-transducin, a GTP-binding protein of retina and other purine nucleotide-binding proteins. In particular S-antigen shows over 50% identity to the proposed pertussis toxin ADP-ribosylation site of alpha-transducin. It also contains the Gly-X-X-X-X-Gly-Lys pattern common to phosphoryl binding sites. A possible relationship between S-antigen and purine nucleotide-binding proteins is discussed. There is also evidence for a repetitious beta-structure in the C-terminal half of S-antigen, with a monoclonal antibody epitope in a helical region at the C-terminus.

S-antigen is coded by [poly A+] mRNA from bovine retina

Total RNA, [poly (A)-] mRNA and [poly (A)+] mRNA purified from bovine retina were translated in vitro in a rabbit reticulocyte lysate system. Immunoprecipitation of translation products with antibodies to the retinal S-antigen (a photoreceptor specific protein involved in autoimmune retinal disease) revealed this protein as a 50,000 daltons band comigrating with purified S-antigen. This indicates that the S-antigen is synthesized in the retina and is not a maturation or degradation product of a larger protein. Its messenger RNA is the polyadenylated RNA, as for some other proteins expressed in nervous tissue.