Interphotoreceptor retinoid-binding protein (IRBP) in progressive rod-cone degeneration (prcd)—Biochemical, immunocytochemical and immunologic studies

Exp. Ege Res. (1991) 53. 389-398

Interphotoreceptor Retinoid-binding Protein (IRBP) Progressive Rod-cone Degeneration (prcd)-Biochemical, lmmunocytochemical and Immunologic Studies BARBARA WIGGERT”, GEETHA KUTTY”, IGAL GERY”, GERALD J. CHADER”

in

KENNETH 0. LONGb*, LILA INOUYE”, and GUSTAV0 D.AGUIRREb,ct

a Laboratory of Retinal Cell and Molecular Biology, National Eye Institute, NIH, Bethesda, MD 20892 and b Section of Medical Genetics, School of Veterinary Medicine and c Scheie Eye Institute, Department of Ophthalmology, School of Medicine, University of Pennsylvania, Philadelphia, PA 79704, U.S.A. (Received

3 October

1990 and accepted

in revised

form

7 December

1990)

Interphotoreceptor retinoid-binding protein (IRBP) is synthesized and secreted by photoreceptor cells and is thought to facilitate the transport of retinoids during the visual cycle as well as fatty acids essential to the maintenance of normal outer segment membranes. Proteins such as IRBP, which are unique to the photoreceptor cells in the retina, are prime candidates in the consideration of biochemical defects which could contribute to photoreceptor cell degeneration in man and animals. In this study, the association between IRBP and retinal degeneration was examined using the progressive rod-cone degeneration @cd) mutant retina in dogs as an animal model. This study shows that loss of IRBP is not an early occurrence in prcd. IRBP is present in relatively normal amounts and distribution even at 1.7 years of age, a time when there is extensive visual cell disease and degeneration. By 2.7-3.0 years of age, IRBP loss correlates with the severity of the disease and concomitant loss of photoreceptor cells. IRBP immunoreactivity was present in the interphotoreceptor matrix (IPM) as long as inner segments were present to a significant degree. The late loss of IRBP immunoreactivity seems to be, therefore, the result of advanced degeneration and end-stage atrophy of the retina. In addition, immunological studies were carried out in order to examine the possible role of an autoimmune response against IRBP in the disease cascade. Normal, heterozygote and prcd-affecteddogs had measurable antibody titers to IRBP. but there was no correlation between disease state and antibody levels. Kry words : dog ; immunocytochemistry : IRBP : progressive rod-one degeneration : retina ; retinal degeneration ; retinitis pigmentosa

1, Introduction Interphotoreceptor retinoid-binding protein (IRBP) is a major component of the retinal extracellular matrix (Adler and Martin, 1982; Bunt-Milam and Saari, 1983; Pfeffer et al., 1983; Fong et al., 1984). It is the only retinoid-binding species in the interphotoreceptor space (Pfeffer et al., 1983). and many of its characteristics indicate that it may play a role in intercellular retinoid transport between the neural retina and the pigment epithelium (Chader et al., 198 3). It may also function as an intracellular fatty acid carrier since endogenous fatty acids are covalently and non-covalently bound to the isolated protein (Bazan et al.. 1985). As is the case with opsin, IRBP appears to be synthesized primarily by rod photoreceptor cells (Hollyfield et al., 19 85 ; van Veen et al., 19 86), but, in contrast to opsin which is a membrane * Current address: Department of Biological Science, California Lutheran University. 60 West Olsen Road. Thousand Oaks, CA

91 360-2787. 1J.S.A. i For correspondence and reprint requests at: School of Veterinary Medicine, Iiniversity of Pennsylvania, 3850 Spruce St.. Philadelphia. PA 19 104. U.S.A.

0014-48

~5/91/090389+10

$03.00/O

protein. IRBP is a soluble protein which is secreted into the interphotoreceptor matrix (IPM). Because of its importance in normal photoreceptor physiology, abnormalities in IRBP function, resulting from changes in protein concentration, distribution or altered affinity for the ligand, could be important, either directly or indirectly, in the pathogenesis of visual cell degenerative diseases. Studies of human donor eye tissues with retinitis pigmentosa (RP) and choroideremia have suggested an association between IRBP and disease since IRBP levels are absent or much lower than could be accounted for by the ongoing photoreceptor degeneration (Rodrigues et al., 1984 ; Bridges et al., 1985). This association is supported by experimental studies with mice that are doubly homozygous at the rd and rds loci and in Abyssinian cats with progressive retinal atrophy; in both cases, IRBP loss significantly precedes visual cell loss (van Veen et al., 1988; Narfstrom et al., 1989). In contrast, other mutant strains of animals which serve as models for RP, e.g. rd and rds mice and the RCS rat, fail to show a causal association between IRBP and disease; in these mutants, changes in IRBP follow the extensive degeneration and loss of visual cells (Eisenfeld, Bunt0 199 1 Academic Press Limited

B. WIGGERT

390

Milam and Saari, 1985 ; Carter-Dawson et al., 1986; van Veen et al., 1988). We have examined the association between IRBP and hereditary retinal degeneration in another animal model whose disease shares many similarities, both clinically and pathologically, with the RP group of diseases. Progressive rod-cone degeneration (prcd) in the miniature poodle is a recessively inherited retinal disease in which the visual cells differentiate normally, but then slowly degenerate, with the rate of degeneration being more extensive and rapid for rods than for cones. The disease had a predictable spatial distribution, and characteristic functional, structural and renewal abnormalities have been found (Aguirre et al., 1982; Aguirre and O’Brien, 1986: Aguirre and Acland, 1988). We reasoned that this mutant, because of the slower course of degeneration and the selective preservation of photoreceptors that results from retardation of disease in specific regions of the fundus, provides an experimental condition that more closely simulates RP in man. As a corollary to our immunochemical and immunocytochemical studies, we also have examined the role of IRBP as a potential autoantigen in prcd. IRBP is a highly uveitogenic protein that induces autoimmune ocular disease when injected intradermaIly (Broekhuyse. Winkens and Kuhlmann, 1986; Hirose et al., 1986). Because it is a soluble protein. we speculated that it could be released from the interphotoreceptor matrix during the course of the retinal degeneration and sensitize the host: this, in turn, could potentially accentuate the rate of visual cell degeneration.

ET AL.

12-16 weeks (4). h and 2.1 yr; (b) prcd-affected, months, 9 months, 1.7 yr and 2 yr (2). 2.7 yr, 3 yr. 5-4 yr and 6.2 yr (2). For these studies the eyecup was divided into eight segments consisting of central and peripheral regions of the nasal and temporal quadrants from the superior and inferior hemispheres (Fig. 1 I. The retina-RPE-choroid-sclera unit was isolated by compression cutting with a razor blade on a waxed dish placed under a dissecting microscope. This method was used to minimize loss of soluble interphotoreceptor matrix constituents by sealing the cut tissue edges by crushing. The tissue samples were placed in screw-capped vials, plunged into liquid nitrogen, and stored at - 70°C until use. For analysis, the retina-pigment epithelium complex was removed from the eyecup samples and homogenized

in 0.5 ml of Tris buffer

(10 mM Tris. 2 mM

EDTA, 0.5 M NaCl, pH 7.5) then centrifuged at 110000 g for 1 hr to obtain the soluble supernatant. SDS-polyacrylamide gel electrophoresis followed by Western blotting was performed as previously described (Wiggert et al., 1986). The primary antibody was a 1:200 dilution of goat anti-bovine IRBP. Horseradish peroxidase conjugated to rabbit anti-goat F(ab’)Z fragment (Cappel Scientific, West Chester. PA) was the second antibody used at a 7 : 1500 dilution. Protein concentrations were determined by the Bradford dye-binding method (Bradford, 1976) using a micro Bio-Rad (Richmond, CA) protocol. Left eye Dorsal Nasal

Temooral

2. Materials and Methods Experimental Dogs Non-affected (homozygous or heterozygous) and yrcd-affected miniature poodles and m. poodle crosses were produced by one of the investigators (G. D. A. ) or obtained from the NEI/NIH supported retinal degeneration research colony (EY-068 5 5). Both groups of animals were housed in the same animal facility in indoor/outdoor runs (Federated Medical Resources, Honeybrook, PA). Additional normal dogs, primarily beagles. were used for selected studies and are identified in each specific methods section. Enucleations were performed from the light-adapted dogs under pentobarbital anesthesia, after which the dogs were immediately killed with a barbiturate overdose. All dogs were maintained and treated in accordance with the ARVO Resolution on the Use of Animals in Research. Immunochemistry For the immunochemical portion of the experiment, one or both eyes from 14 prcd-affected and nine phenotypically normal dogs were used at the following ages: (a) normal, 12-16 weeks (S), 9 months (2), 1 yr.

Ventral

FIG. 1. Schematic illustration of the dissection technique used to separate the eye into ‘quadrants ’ for Western analysis. The globe was cut with a razor blade anterior to the equator and the anterior segment and vitreous removed. The eyecup was placed on a waxed dish, and the segments illustrated in the tigure were isolated by cutting through the retina-RPE-choroid-sclera as a unit. Note that segments do not represent equal retinal areas: this results from the asymmetric location of the optic disc in the eyecup. C. central: P, peripheral: I, inferior ; S, superior; N, nasal : T. temporal.

IRBP

AND

391

prcd

Northern Analysis RNA was prepared from normal (1.1 yr) and prcdaffected retina (1.1 yr) using the acid-guanidinium thiocyanate method of Chomczynski and Sacchi (1987). The methods have been previously described and are summarized in this section (Inouye et al., 1989). Twenty micrograms total RNA was loaded per lane and size-fractionated in a 0.8% agarose-2.2 M formaldehyde denaturing gel. After electrophoresis, the gel was capillary transferred to nitrocellulose and the filter was baked in a vacuum oven for 1 hr. The bovine IRBP cDNA probe PTMRZ was labeled with 32P by random priming. The filter was prehybridized for 2 hr at 42°C in 40% formamide, 5 x SSPE,5 x Denhardt’s solution, 1% SDS and 100 ,ug ml-’ salmon sperm DNA. A fresh aliquot of the same mix was used for hybridization ; labeled probe was added to the hybridization mix at 1 x lo6 cpm ml-’ and hybridization was allowed to proceedovernight at 42’C. Posthybridization washes were done for 15 min ( x 2) at room temperature in 2 x SSC,0.1% SDS,followed by two washes in 0.1 x SSC,0.1% SDSat 42°C and two washes in 0.1 x SSC,0.1% SDSat 52°C. Immunocytochemistry Immunocytochemical examination of the retina was performed in the eyes of three normal and 15 prcd-affecteddogs at the following ages: (a) normal, 2 months, 5 months and 9 yr: (b) prcd-affected, 8-14 weeks (3), 5 months, 10 months, 1.3 yr, 2-2.7 yr (4), 3.1 yr, 4.3 yr (2) and 64 yr (2). Eyecups or selected regions of eyecupswere fixed for approximately 3 hr in 0.1 M cacodylate-buffered 4% formaldehyde with 3% sucrose, pH 7.2-7.4, and were then embedded in acrylamide and frozen (Johnson and Blanks, 1984). Ten-micron thick cryosections were collected on gelatin-coated cover slips and air dried. Sections were

blocked in PBS containing 0.3% TritonX-100, BSA (1 mg ml-l) or goat serum (for rabbit antibodies) for 15-30 min. Antibodies used were goat anti-bovine

IRBP and rabbit anti-monkey IRBP ; for some sections the antibodies had been affinity purified (Kapoor and FIG. 2. Nomarski (A) and immunofluorescence (B.C) photomicrographs of normal dog retina. IRBP immunoreactivity (B) is retricted to the interphotoreceptor space, between the apex of the retinal pigment epithelium (RPE) and the outer limiting membrane (arrowhead). The reaction appears more intense along the RPE apical surface, and is lessintense in the proximal region of the inner segment layer (B). The fluorescent profiles in the ganglion cell layer (B) are vascular elements which also label in the control (C). Lipofuscin fluorescence is present in the RPE (C, horizontal arrow). *, choroidal blood vessel: PR, photoreceptor layer ; ONL, outer nuclear layer, INL, inner nuclear layer: oblique arrow (A), internal limiting membrane. B-primary = nonaffinity purified goat anti-bovine IRBP antisera ; secondary = rabbit anti-goat RITC: C-primary = goat control serum: secondary = rabbit anti-goat RITC. x 500.

B. WIGGERT

392

I

/ 1 j !

ET AL.

Cho-Chung, 1983). Sectionswere incubated overnight at 4°C with the primary antisera (1: 1000, 1: 2000) or with affinity purified antisera (1: 500 to 1: 1000) in PBSwith 0.3 % Triton X-100. Antisera were visualized using either a Vectastain ABC peroxidase kit (Vector Laboratories, Burlingame, CA) or the corresponding FITC- or RITC-labeled F(ab’)Z secondaries (Cappel, West Chester, PA). Where it was necessary to correlate the immunocytochemical results with morphology, e.g. in the prcd retina where staging of disease was critical for interpretation, the peroxidase reaction was used. For controls, pre-immune serum was used at the above concentrations. Background labeling was highest in the control rabbit serum, but control goat serum yielded little background label. Affinity purified serum was quite clean; controls for secondary antisera were blank. Immunology

For determination of humoral immunity to IRBP, fresh blood was collected from homozygous normal (n = 8; l-10 yr of age), heterozygous (n = 3; l-9 yr of age) and homozygous affected (n = 5: 0.8-6 yr of age) miniature poodle and beagle-m. poodle crossbred dogs.The blood was allowed to clot and the serum was separated and stored at - 70°C until assayed. AntiIRBP titers were quantitated by the enzyme-linked immunosorbent assay (ELISA) as previously described (Mochizuki et al., 1985) except that labeled goat antidog immunoglobulin was used as the secondary antibody. Sera were tested at dilutions of 1: 10, 1: 40 and 1: 160; absorbance was measured at 410 run. Assays were run using dog serum albumin as the blocking protein. 3. Results lmmunocytochemistry

The immunofluorescent localization of IRBP in normal adult dog retina is shown in Fig. 2. As in the case of other vertebrates, IRBP immunoreactivity is restricted to the interphotoreceptor space,between the apex of the retinal pigment epithelium (RPE) and the outer limiting membrane (OLM). The diseasein p&-affected animals is not uniformly distributed across the retina, but has a very specific spatial distribution that is temporally modified. In affected retinas, rod disease initially occurs in the posterior pole and equatorial regions, with preservation of normal retinal structure peripherally. Subsequently, photoreceptor degeneration occurs in a sequence that is quadrant specific: that is, degenerFIG. 3. IRBP immunoreactivity in the early prcd disease stages (A). At S-O (2 months) the IPM shows normal distribution of IRBP immunoreactivity, but the intensity of the reaction is low: this is a normal variation in the normal and diseased retina. B, At S-l (9.5 months) the outer

segments are vesiculated, but IRBP is localized normally. C, Higher magnification of (B) to illustrate the disruption present in the OS layer. Arrowheads mark the OLM. (A.B)

x 620. (C) x 1550.

IRBP

AND

prcd

FIG. 4. IRBP immunoreactivity in the degeneration phase of p-cd. A, Rod inner and outer segment material is reduced at S2 (posterior pole, 2.5 yr) ; IRBP is localized from the OLM to the RPE. The increase in label at the level of the OLM (arrowhead) in comparison to normal or early disease stages may result from the increased space available due to cell loss in the photoreceptor layer. A macrophage showing IRBP immunoreactivity (open arrow) is present in the IPM. B, At S-4 (posterior pole, 3.5 yr), IRBP is localized in the narrowed sub-retinal space, and, to a lesser extent, to the cone inner segments. The outer

nuclear layer is two nuclei wide, and distinct cone inner segments (fine arrows) predominate. x 15 50. ation occurs earlier and progresses more rapidly in the inferior quadrant than in the superior or temporal quadrants (inferior > superior > temporal) (for review see Aguirre and Acland, 1988). Details and staging of the disease have been described previously (Aguirre and O’Brien, 1986; Aguirre and Acland, 1988). IRBP localization in retinas of prcd-affected poodles at various ages and disease stages is shown in Figs 3-5. IRBP localization is normal when the photoreceptors are structurally intact [Stage (S)-0] or when rod disease is limited to disorientation and vesiculation of the outer segments (S-l) (Fig. 3). Rod disease is followed by degeneration, and the retina shows a reduction in cell number and in the amount of ROS material (S-2). At this stage, macrophages invade the sub-retinal space ; they are IRBP immunoreactive, presumably from the phagocytosis of IPM constituents along with the rod debris. IRBP is localized to the interphot.orecept.or space at this stage as well as in S4, when the subretinal space is substantially reduced in size, the outer nuclear layer is narrowed and cone inner segments are the predominant distal photoreceptor structure that remains (Fig. 4). The retina at this stage shows IRBP immunoreactivity in the IPM and also within the inner segments of the remaining photoreceptor cells [Fig. 4(B)]. Degeneration is followed by atrophy (S-5 through S-8) where the photoreceptor and outer nuclear layers have disappeared and the inner retinal layers progressively

become gliotic and disorganized. IRBP immunoreactivity is absent from the IPM in the atrophic stages of the disease (Fig. 5). Western Analysis Western blotting of soluble proteins in retina-RPE samples from a normal 2.1 year-old miniature poodle demonstrated a band of IRBP immunoreactivity at approximately the same molecular weight (140 kD) as purified bovine IRBP (Fig. 6). Retina-RPE samples from one 2.0-year-old affected poodle were similar to the normal samples, but in the other 2Gyear-old there was a loss of IRBP immunoreactivity from one of the inferior retinal quadrants (Fig. 7). In the case of a 2.7-year-old (Fig. 8) and a 3.0-year-old (Fig. 9) affected poodle, the 140~kD band of IRBP immunoreactivity was detectable in reduced amounts in the superior quadrants of the retinas, but barely or not at all in the inferior quadrants. There was no detectable IRBP immunoreactivity in the retina of a 5.4-year-old affected poodle (data not shown). Lower molecular weight bands such as those present in the 2.7 and 3.0year-old affected poodles (Figs 8 and 9) were also present in a few normal samples (data not shown). Such bands were usually observed only when blots were developed for a longer period of time in order to maximize the chances of detecting the 140-kD band as was the case in the 2.7 and 3.0-year-old affected dogs.

B. WIGGERT

ET AL.

was 1.1 yr of age and primarily affected with photoreceptor disease rather than more extensive degeneration. The results show that IRBP mRNA is the same in normal and affected animals at 1.1 yr. lmmunology The normal, heterozygote and prcd-affected dogs used in this study had measurable antibody titers to IRBP, but no correlation was found between disease status and antibody levels in the small sample examined. Positive ELISA responses (absorbance 0.2 at 410 nm) were found in most of the normal dogs when tested at 1: 10 dilution and in only one out of eight samples at 1: 160. Similar serum antibody levels to IRBP were present in the prcd-affected and heterozygous animals (Table I). More ELISA-positive antibody levels were found in all three groups (data not shown) when tested against bovine serum albumin (BSA). thus re-emphasizing the need to use canine serum in the assay (Long et al., 1988). 4. Discussion

FIG. 5. IRBP immunoreactivity in the atrophic phase of

A, The equatorial retina from the tapetal zone of a 3,1year-old dog shows S-5 disease. The RPE abutts the one nuclei wide outer nuclear layer (ONL) with little or no inner segment/outer segment material remaining. No IRBP immunoreactivity is present. The peripheral area of the same retina is at S-6 and shows the onset of gliosis and disorganization of the retina. B, The far periphery at 4.3 yr shows S-7 disease: the inner nuclear layer abutts the RPE, and no labeled cells are present. Fine arrow (B) marks the internal limiting membrane: Ch, choroid. Note that RPE pigmentation is normally variable and determined by spatial position. x 15 50. prrd.

Northern Analysis The bovine IRBP cDNA probe hybridized with a Northern blot of normal canine total retinal RNA detects a message at 4.6 kb (Fig. 10). The message is readily detectable in 20 yg total RNA, indicating that IRBP message is abundant in canine retina. This result agrees with previous findings of conservation of retinal IRBP message among several other mammalian species (Inouye et al., 1989). The p-cd retina analysed

Proteins unique to photoreceptor cells in the retina are prime candidates for defects associated with inherited retinal degenerations in man and animals. IRBP, which is synthesized and secreted primarily by rod cells, is such a protein. Since IRBP is thought to function in aiding the transport of retinoids during the visual cycle as well as fatty acids critical to the maintenance of normal outer segment membranes, loss of IRBP could have devastating consequences to photoreceptor function and viability: e.g. the accumulation of free retinoids which are known to be membranolytic or a deficit of essential fatty acids needed for ROS disc assembly. The present study indicates that loss of IRBP is not an early occurrence in prcd. Up to approximately 2.0 years of age, Western blotting of soluble proteins from affected retinas showed IRBP to be relatively similar in amount and distribution to a normal 2.1year-old control retina. Similarly, Northern analysis indicated that IRBP message is the same in normai and affected animals. By 2.7-3.0 years of age, however, IRBP is detectable primarily in the superior quadrants of affected retina and no immunoreactive protein is present in the blots from the inferior quadrants. This correlates with the more extensive disease and degeneration present in the inferior quadrants at this age (Aguirre and Acland. 1988 1. Loss of IRBP immunoreactivity from all quadrants of the eye is associated with the severe degeneration and atrophy present in the retina with advanced disease. The Western blotting results of the older prodaffected retina also showed that the distinct 140-kD IRBP immunoreactive band was frequently replaced by lower molecular weight bands. The presence of a similar finding in a few of the normal samples

IRBP

AND

prcd

395

ITP

INC

ITC

SNP

STP

SNC

STC

INP

Bovine IRBP

kD

116-2

66.2

FIG. 6. Western blot of soluble proteins from quadrants of a 2.1-year-old normal miniature poodle retina-RPE. S, Superior: I, inferior: T. temporal : N, nasal; C, central: P, peripheral. IRBP immunoreactivity is present in all quadrants as a protein band of approximately the same molecular weight (140 kD) as purified bovine IRBP.

INP

ITP

INC

ITC

SNP

STP

SNC

STC

Borne IRBP

kD 200-

/ 16.292.566.2-

45.0-

FIG. 7. Western blot of soluble proteins from quadrants of a 2.0-year-old affected miniature poodle retina-RPE. S, Superior: I, inferior: T, temporal: N, nasal: C, central; P. peripheral. Except for the loss of an IRBP immunoreactive band from the ITP quadrant, the immunoblots are similar to the normal samples seen in Fig. 6. suggested the lower molecular weight bands may be the result of breakdown of the protein or of nonspecific interaction of the primary or secondary antibodies. Finding a prominent 4*6-kb message on Northern analysis of normal and prcd-affected retinas argues in favor of limited proteolysis or non-specific antibody interaction as the cause for the lower molecular bands present in the immunoblots. Our immunocytochemical results demonstrated that IRBP localization is stage-dependent rather than simply age-dependent. This finding correlates with the immunochemical results, since prcd is characterized by marked spatial differences in disease progression.

Regardless of the age of the affected animal, IRBP could be localized in the IPM as long as inner segments were present to a significant degree. Once inner segments were lost, IRBP immunoreactivity disappeared from the sub-retinal space. This is somewhat similar to the situation in the rds mouse mutant in which IRBP is synthesized and secreted into the IPM in spite of the almost total lack of outer segments (van Veen et al., 1988). In contrast to the rds mouse. the prcd-affected retina did not demonstrate significant intracellular localization of IRBP at late stages of the retinal degenerative disease. Immunochemical studies of RP-affected eyes also have showed the presence of

B. WIGGERT

396

ET AL

Bovcne

INP

ITP

STP

ITC

ST0

SNP

SNC

IRBP

kD

Il6.292*5-

6&Z-

45-

FIG. 8. Western blot of soluble proteins from quadrants of a 2.7-year-old affected miniature poodle retina-KPE. S, Superior ; I, inferior: T. temporal: N. nasal: C, central : P, peripheral. lRBP immunoreactivity is reduced in the superior quadrants and barely detectable in the inferior quadrants. Note that lower molecular weight (90 kD) immunoreactive bands are visible in retinal segments isolated from most quadrants when the blots are developed for a longer time period.

ITP

INC

Bovine

ITC

SNP

STP

SNC

STC

INP

IRBP

kD 200-

Il6.292,5-

66.2-

45.0-

FIG. 9. Western blot of soluble proteins from quadrants of a 3.0-year-old affected miniature poodle retina-KPE. S. Superior; I, inferior: T. temporal : N, nasal ; C, central : P, peripheral. The pattern of IRBP immunoreactivity is similar to that seen in the 2.7-year-old affected samples. Lower molecular weight (90 kD) immunoreactive bands are visible in retinal segments isolated from most quadrants when the blots are developed for a longer time period.

IRBP

in retinal

homogenates,

but

not

IPM

prep-

arations. These eyes, although having advanced degeneration, still showed the retention of some photoreceptor cells (Schmidt et al., 1988). Since immunocytochemical results were not presented in that study, it is not clear whether the IRBP was localized intracellularly, in the IPM or in both sites, and the correlation between IRBP levels and disease severity

could not be made.

The detection of antibodies to IRBP in the sera of dogs tested in this study is similar to our previous

finding of antibodies to S-antigen in both normal and diseased (rod-cone dysplasia 1) dogs (Long et al.. 1988). It is further proposed that these antibodies are analogous to antibodies against S-antigen (Doekes et al.. 1987). or against a variety of other tissue-derived antigens (Guilbert, Guillaume and Avrameas. 1982 ). The mechanism of production of these ‘natural antibodies ’ and their function remains obscure. In summary. our studies demonstrate that loss o! IRBP is not causally related to the early visual cell disease and degeneration present in prcd-affected

IRBP

AND

prcd Lane

397

miniature poodle retina. Additionally, the lack of significant antibody titers to this potent uveitogenic protein would argue against a humoral immunologic basis to the disease process. Our findings do not rule out, however, the possibility that the gradual loss of IRBP may contribute to, or modulate. the visual cell degenerative process which occurs later in the disease.

I

Acknowledgements The authors are indebted to Susan Nitroy for expert technical assistance. This work was supported in part by NEI/NM grants EY-01244,06855, PHS Training Grant EY07035, the Retinitis Pigmentosa Foundation Fighting Blindness, the Frances V. R. Seebe Trust and the CERF-PRA Research Fund.

References Adler, A. and Martin, K. (1982). Retinoid-binding proteins in bovine interphotoreceptor matrix. Biochem. Biophys. Res. Commun. 108, 1601-8. Aguirre, G. and Acland, G. (1988). Variation in retina1 degeneration phenotype inherited at the prcd locus. Exp. Eye Res. 46, 663-87. Aguirre, G., Alligood, H., O’Brien, P. and Buyukmichi, N. (1982). Pathogenesis of progressive rod-cone degeneration in miniature poodles. Invest. Ophthalmol. Vis. Sci. 23, 610-30.

Aguirre, G. and O’Brien, P. (1986). Morphological and biochemical studies of canine progressive rod-cone degeneration. 3H Fucose autoradiography. Invest. Ophthalmol. Vis. Sci. 27, 635-55. Bazan, N., Reddy, T.-S., Redmond, T., Wiggert, B. and Chader. G. (1985). Endogenous fatty acids are covalently and noncovalently bound to interphotoreceptor retinoid-binding protein in the monkey retina. J. Biol. Chem. 260, 13677-80. Bradford, M. ( 19 76). A rapid and sensitive method for the quantitation of microgram quantities of protein using the principle of protein dye binding. Anal. Biochem. 72, FIG. 10. Detection of IRBP transcripts in canine retinal RNA. Arrow indicates message at 4-6 kb. Lane 1 is normal retina and lane 2 is prcd retina. Twenty micrograms of total retinal RNA was loaded. Bars denote RNA size standards at 9.5. 7.5, 4.4 and 2.4 kb.

TABLE

I

Determination of humoral immunity to IRBP in prcdaffected, normal and heterozygous dogs Serum dilution Group Normal @OS.*) Heterozygote

1: 10

1:40

1:160

0.33 (7/8)

0.16 (l/8)

0.09 (l/8)

0.23 (2/3)

0.12 (O/3)

0.08 (O/3)

0.34 (3/4)

0.16 (2/4)

0.14 (l/4)

(pas.*)

Affected (Pas.*)

Values are expressed as mean absorbance values at each serum dilution. * (Positive/Total). Positive ELISA responses are those with absorbance greater than 0.2 at 410 nm. 2;

248-54.

Bridges. C. D., O’Gorman. S.. Fong. S.-L., Alvarez, R. A. and Berson. E. (1985). Vitamin A and interstitial retinolbinding protein in an eye with recessive retinitis pigmentosa. Invest. Ophthalmol. Iris. Sci. 26. 684-91. Broekhuyse, R., Winkens. H. and Kuhlmann, E. (1986). Induction of experimental autoimmune uveoretinitis and pinealitis by IRBP. Comparison to uveoretinitis induced by S-antigen and opsin. Curr. Eye Res. 5, 231-40. Bunt-Milam, A. and Saari, J. (1983). Immunocytochemical localization of two retinoid-binding proteins in the vertebrate retina. J. Ceil Biol. 97. 703-12. Carter-Dawson, L.. Alvarez, R.. Fong. S-L.. Lieu, G.. Sperling, H. and Bridges, C. D. (1986). Rhodopsin, 1 1-cis vitamin A, and interstitial retinol-binding protein (IRBP) during retinal development in normal and rd mutant mice. Dev. Biol. 116, 431-8. Chader, G., Wiggert, B., Lai, Y.-L., Lee, I,. and Fletcher, R. (198 3 ). Interphotoreceptor retinoid-binding protein : a possible role in retinoid transport in the retina. In Progress in Retinal Research. (Eds Osborne, N. and Chader. G.) Pp. 162-89. Pergamon Press. Oxford. Chomczynski, P. and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanatephenol-chloroform extraction. Anal. Biochem. 132, 6-l 3. EER 53

398

Doekes, G., van der Gaag, R.. Rothova, A., van Kooyk, Y., Boersma. L., Zaal, M. J. M.. Dijkman, G., Fortuin. M. E., Baarsma, G. S. and Kijlstra, A. (1987). Humoral and cellular immune responsiveness to human S-antigen in uveitis. Curr. Eye Res. 6, 909-19. Eisenfeld. A., Bunt-Milam, A. and Saari. J. (1985). Immunocytochemical localization of interphotoreceptor retinoid-binding protein in developing normal and RCS rat retinas. Invest. Ophthalmol. Vis. Sci. 26, 775-8. Fong, S.-L., Lieu, G., Landers. R., Alvarez, R. and Bridges, C. (I 984). Purification and characterization of a retinolbinding glycoprotein synthesized and secreted by bovine neural retina. J. Biol. Chem. 259, 653442. Guilbert, B., Guillaume, D. and Avrameas. S. (1982). Naturally occurring antibodies against nine common antigens in human sera. 1. Immunol. 128. 2779-87. Hirose. S., Kuwabara, T.. Nussenblatt. R., Wiggert, B.. Redmond, M. and Gery. I. (1986). Uveitis induced in primates by interphotoreceptor retinoid-binding protein. Arch. Ophthalmol. 104, 1698-702. Hollyfield, J., Fliesler, S., Rayborn. M., Fang, S.-L., Landers, R. and Bridges, C. D. (1985). Synthesis and secretion of interstitial retinol-binding protein by the human retina. Invest. Ophthalmol. Vis. Sci. 26, 58-67. Inouye. L., Albini, A., Chader, G., Redmond. M. and Nickerson, J. (1989). mRNA for interphotoreceptor retinoid-binding protein (IRBP) : distribution and size diversity in vertebrate species. Exp. Eye Res. 49. 171-80. Johnson, L. V. and Blanks. J. C. (1984). Application of acrylamide as an embedding medium in studies of lectin and antibody binding in the vertebrate retina. Curr. Eye Res. 3. 969-73. Kapoor, C. and Cho-Chung. U. (1983). Affinity purification of regulatory subunits of CAMP-dependent protein kinase using cross-linked immunoabsorbent. 1. lmmunol. Methods 57, 215-20. Long, K.. Philp, N., Gery. I. and Aguirre, G. (1988). SAntigen in a hereditary visual cell disease: Immuno-

6. WIGGERT

ET AL.

cytochemical and immunological studies. Invest. Opkthalmol. Vis. Sci. 29, 1594-606. Mochizuki, M.. Kuwabara, T., McAllister, C., Nussenblatt. R. and Gery, I. (1985). Adoptive transfer of experimental autoimmune uveoretinitis in rats. Invrst. Ophthalmol. Vis. Sci. 26, 1-9. Narfstrom, K.. Nilsson, S.-E., Wiggert, B.. Lee. I,.. Chader, G. and van Veen, T. (1989). Reduced level of interphotoreceptor retinoid-binding protein (IRBP). a possible cause for retinal degeneration in the Abyssinian cat. Cell Tissue Res. 27, 631-9. Pfeffer, B.. Wiggert, B., Lee, L.. Zonnenberg. B.. Newsome. I). and Chader, G. (1983). The presence of a soluble interphotoreceptor retinoid-binding protein in the retinal interphotoreceptor space. 1. Cell. Physiol. 117, 3 3 3-41. Rodrigues. M.. Ballintine, E., Wiggert, B., Lee. L.. Fletcher, R. T. and Chader, G. (1984). Choroideremia: A clinical, electron microscopic and biochemical report. Ophthalmology 91, 873-83. Schmidt, S., Heth. C.. Edwards, R.. Brandt. J.. Adler, A., Spiegel, A.. Shichi, T. and Berson. E. (1988). Identification of proteins in retinas and IPM from eyes with retinitis pigmentosa. Invest. Ophfhalmol. Vis. Sci. 29: 1585-93. van Veen, T.. Ekstrom, P., Wiggert, B.. Lee, L.. Hirose. Y.. Sanyal. S. and Chader, G. (1988). A developmental study of interphotoreceptor retinoid-binding protein (IRBP) in single and double homozygous rd and rds mutant mouse retinae. Exp. Eye RPS.47. 291-305. van Veen. T., Katial, A., Shinohara, T.. Barrett, D.. Wiggert. B., Chader. G. and Nickerson, J. (1986). Retinal photoreceptor neurons and pinealocytes accumulate mRNA for interphotoreceptor retinoid-binding protein (IRBP). FEBS Lett. 208, 133-7. Wiggert, B., Lee, L.. Rodrigues, M.. Hess, H., Redmond, T. and Chader, G. (1986). Immunochemical distribution of interphotoreceptor retinoid-binding protein in selected species. Invest. Ophthalmol. Vis. Sci. 27, 1041-9.