Low docosahexaenoic acid levels in rod outer segment membranes of mice with rds/peripherin and P216L peripherin mutations.

PURPOSE: Humans with retinitis pigmentosa and dogs with progressive rod-cone degeneration (prcd) have lower than normal blood levels of long-chain polyunsaturated fatty acids, including docosahexaenoic acid (DHA), the major fatty acid found in retinal rod outer segments (ROS). In addition, prcd-affected dogs have lower levels of DHA in their ROS than control animals. The present study was designed to determine whether mice that are heterozygous for the rds mutation and transgenic mice heterozygous for a specific rds/peripherin mutation (P216L) have lower DHA levels in their ROS and other tissues than do control mice. METHODS: Wild-type (rds(+/+)) mice, mice with the rds(-/-) (null) and rds(+/-) mutations, and mice with the P216L rds/peripherin mutation on the rds(+/-) background were maintained in the vivarium under identical husbandry conditions, and tissues were removed from each group for analysis at approximately 2 months of age. Fatty acid compositions of total lipids from plasma, red blood cells, liver, and ROS were determined by gas-liquid chromatography. ROS purity from each group was determined by SDS-PAGE with silver staining. The morphologic status of retinas representing each genotype was analyzed by light and electron microscopy. RESULTS: There was no difference between rds(+/-), P216L on rds(+/-), and rds(+/+) (control) animals in the fatty acid composition of plasma, expressed as relative mole percent or as nanomoles fatty acid per milliliter of plasma. Small but statistically significant differences were found in 18:0 and C-22 polyunsaturated fatty acids of red blood cells. In the liver, the control animals had higher levels of 20:4n-6. In contrast, the ROS of control animals had levels of DHA that were 1.4 times that of ROS from either rds(+/-) or P216L on rds(+/-) mice of the same age. The reduction in DHA was not accompanied by an increase in 22:5n-6, which always occurs in neural tissues of animals deprived of n-3 fatty acids. SDS-PAGE of the three ROS membrane preparations showed that they were of identical purity. CONCLUSIONS: Mice heterozygous for the spontaneous rds/peripherin mutation or mice carrying the P216L mutation on this heterozygous background have a statistically significant reduction of DHA in their ROS membranes. The authors propose that reduction in DHA is an adaptive response to metabolic stress caused by the mutation.

A hypothesis to explain the reduced blood levels of docosahexaenoic acid in inherited retinal degenerations caused by mutations in genes encoding retina-specific proteins.

Some humans and animals with inherited retinal degenerations (RD) have lower blood levels of docosahexaenoic acid (22:6n-3) than controls. As a result of recent studies, clearly the low blood 22:6n-3 phenotype is found in multiple RD phenotypes and no mutation thus far identified in humans or animals is involved in lipid metabolism. Therefore, it seems reasonable to suggest that the primary defect is not in 22:6n-3 metabolism, but rather in some common convergent pathway that ultimately leads to the reduction of blood and tissue 22:6n-3 levels. One possibility is that the different mutations produce a metabolic stress that provokes structural and biochemical adaptive changes in photoreceptor cells and their rod outer segments. If the stress is oxidant, the retina could downregulate 22:6n-3 and upregulate antioxidant defenses. How such a stress could lead to changes in blood levels of 22:6n-3 is not obvious. However, the consistent finding of the 22:6n-3 phenotype in many different retinal degeneration genotypes suggests that some form of communication exists between the retina and other tissues that serves to reduce blood levels of 22:6n-3.

Polyunsaturated fatty acids are lower in blood lipids of Usher's type I but not Usher's type II.

PURPOSE: Previous studies have shown that persons with retinitis pigmentosa and Usher's syndrome have lower blood levels of long-chain polyunsaturated fatty acids (PUFAs). In this study, the fatty acid composition of phospholipids from plasma and red blood cells (RBCs) was compared in persons with Usher's syndrome type I; Usher's syndrome type II; or no retinal disease (control subjects). METHODS: Blood was drawn from fasting volunteers and separated into plasma and RBCs by centrifugation. Lipids were extracted and phospholipids were obtained by thin-layer chromatography. Fatty acid methyl esters were prepared and analyzed by gas-liquid chromatography. RESULTS: There were no differences in plasma or RBC phospholipid fatty acid composition between control subjects (n = 54) and persons with Usher's syndrome type II (n = 20). However, all 20- and 22-carbon PUFA levels from RBCs of persons with Usher's syndrome type I were lower than those from control subjects and persons with Usher's Syndrome type II. Likewise, plasma levels of 20:3n-6, 20: 5n-3, and 22:6n-3 were lower in Usher's syndrome type I compared with the control group. In contrast, plasma levels of 18:1n-9 and RBC levels of 16:0 and 18:1n-9 were higher in the group with Usher's syndrome type I. CONCLUSIONS: Plasma and RBCs from Usher's syndrome type I, but not type II, have lower levels of long-chain PUFAs than plasma and RBCs from control subjects.

Effect of diet on the fatty acid and molecular species composition of dog retina phospholipids.

Dogs were born to mothers fed commercial diets low or enriched in n-3 fatty acids and raised on those diets until they were about 50 d old. Retinas were removed, lipids were extracted, and total phospholipids were analyzed for fatty acid and molecular species composition. Animals from the low n-3 group had significantly lower retinal levels of 22:6n-3 and higher levels of n-6 fatty acids, especially 20:4n-6 and 22:5n-6. There was no difference in the retinal levels of 18:2n-6, and only small differences were found in saturated and monounsaturated fatty acids. The most dramatic differences in molecular species occurred in 22:6n-3-22:6n-3 (4.7 vs. 0.8%) and 18:0-22:6n-3 (27.6 vs. 14.4%); total molecular species containing 22:6n-3 were significantly lower in the low n-3 group (45.5 vs. 24.0%). Molecular species containing 20:4n-6 and 22:5n-6 were greater in the low n-3 animals (13.0 vs. 25.7%), as were molecular species containing only saturated and monounsaturated fatty acids (40.8 vs. 35.4%). These results show that modest differences in the amount of n-3 fatty acids in the diets of dogs can have profound effects on the fatty acid and molecular species composition of their retinas.

Diets enriched in docosahexaenoic acid fail to correct progressive rod-cone degeneration (prcd) phenotype.

PURPOSE: Results of a previous study show abnormal plasma lipids in progressive rod-cone degeneration (prcd)-affected dogs, with lower docosahexaenoic acid (DHA; 22:6n-3) and cholesterol levels but no differences in other plasma fatty acids, lipids, triglycerides, and fat-soluble vitamins. There is also an increase of the DHA precursor 22:5n-3, so that the ratio of 22:5n-3 to 22:6n-3 is higher in affected than in normal dogs. Because DHA is the predominant esterified fatty acid in rod outer segment (ROS) phospholipids, these findings suggest a possible causal association between abnormal plasma lipid levels and retinal degeneration. In the current study, dietary supplements rich in 22:6n-3 were used to determine whether plasma, liver, and rod outer segment phospholipid composition can be altered to modify the prcd disease phenotype. METHODS: prcd-affected and normal control dogs were given DHA-enriched supplements for short (7- and 25-day) and long (21-week) periods, and the fatty acid composition of plasma, liver, and rod outer segment phospholipids were examined. In the long-term study, electroretinography and morphology were used to assess modification of the retinal degeneration phenotype. RESULTS: Administration of DHA-enriched supplements resulted in increases in plasma DHA and n-3 polyunsaturated fatty acids and in decreases in some n-6 fatty acids in normal and prcd-affected dogs. Similar increases in DHA and n-3 fatty acids were observed in the liver, but affected dogs had significantly higher levels at all supplementation time points examined. In contrast, the ROS of affected dogs had statistically lower (approximately 20%) DHA levels, and these levels could not be increased with dietary supplementation. The disease phenotype could not be modified by DHA-enriched supplements. CONCLUSIONS: Regardless of the sustained three- to fourfold elevation in plasma and liver DHA that occurs as the result of supplementation, the ROS DHA levels remain unchanged, and the prcd disease phenotype is not modified by the dietary manipulation. These findings could be the result of a reduction in the synthesis of DHA-containing phospholipids in the retinas of affected dogs; or, alternatively, there could be a reduction in DHA uptake, transport, or storage within the retinal pigment epithelium-photoreceptor complex.

Lipids of plasma, retina, and retinal pigment epithelium in Swedish briard dogs with a slowly progressive retinal dystrophy.

Reduced blood levels of long chain polyunsaturated fatty acids have been reported in humans and animals with inherited retinal degenerations. The lipid and fatty acid compositions of plasma, retina, and retinal pigment epithelium of the Swedish Briard dog, which has a very slowly progressive retinal dystrophy that is inherited in an autosomal recessive manner were analysed. The lipid class composition of the pigment epithelium was not different between affected and normal dogs; however, significant differences were found between the retinas of the two groups. Affected dogs had relatively more phosphatidylethanolamine and phosphatidylinositol and less phosphatidylcholine than normal dogs. There was no difference in the fatty acid compositions of plasma and retinal pigment epithelium between affected and normal dogs. However, the retinas of affected dogs had significantly lower levels of 22:5n-3 and 22:6n-3 and higher levels of 18:2n-6, 20:4n-6, and 22:5n-6. The total n-3 fatty acid content was significantly lower in affected retinas (P < 0.001), whereas the content of n-6 fatty acids was significantly higher in affected retinas (P < 0.001). These studies provide evidence for yet another animal model of inherited retinal degeneration with a defect in retinal polyunsaturated fatty acid metabolism. The fatty acid pattern in affected dogs resembles that seen in the retina in n-3 fatty acid deficiency.

Plasma lipid changes in PRCD-affected and normal miniature poodles given oral supplements of linseed oil. Indications for the involvement of n-3 fatty acids in inherited retinal degenerations.

It has previously been shown that miniature poodles with progressive rod-cone degeneration (PRCD) have lower plasma levels of docosahexaenoic acid (22:6n-3) than normal poodles and it has been suggested that affected animals have a defect in the metabolism of 22:6n-3. To test this hypothesis in vivo, PRCD-affected and normal miniature poodles were given daily oral supplements of linseed oil (enriched in 18:3n-3). Blood was drawn from food-deprived animals at predetermined times before, during and after supplementation, and plasma lipid fatty acids were analysed. There were no differences in the levels of 18:3n-3, 20:5n-3, and 22:5n-3 between affected and normal dogs. Therefore, there appears to be no abnormality in the elongation and desaturation system that takes 18:3n-3 to 22:5n-3. Surprisingly, the plasma level of 22:6n-3 was reduced in both groups following supplementation, but to a significantly greater extent in affected dogs. This resulted in a significantly higher 22:5n-3/22:6n-3 ratio in affected animals. These results support the earlier suggestion of an abnormality in 22:6n-3 metabolism in PRCD-affected miniature poodles. To determine the effect of n-3 supplementation on polyunsaturated fatty acid metabolism in dogs (not as a function of disease), results from both groups of dogs were pooled and compared at times before and near the end of supplementation. Dietary 18:3n-3 led to predictable increases in 18:3n-3, 20:5n-3, and 22:5n-3, but to a decrease in 22:6n-3.(ABSTRACT TRUNCATED AT 250 WORDS)

Lipids of human retina, retinal pigment epithelium, and Bruch's membrane/choroid: comparison of macular and peripheral regions.

PURPOSE: To understand the difference between macular and peripheral regions, tissue samples of human retina, retinal pigment epithelium, and Bruch's membrane/choroid were dissected and analyzed for lipid composition. METHOD: To facilitate dissections and enhance the recovery of tissues, eyecups were prefixed for 1 hour in 10% formalin (pH 7). Lipids were extracted from 4-mm trephined punches of tissues and analyzed by high-performance liquid chromatography. After separation of neutral lipids and phospholipids, total fatty acids in both lipid classes were quantitated. RESULTS: The major phospholipid classes in retina, retinal pigment epithelium, and Bruch's membrane/choroid were phosphatidyl choline, phosphatidyl ethanolamine, phosphatidyl inositol, and phosphatidyl serine; the major fatty acids were palmitic (16:0), stearic (18:0), and oleic (18:1). Although the three tissues had similar total fatty acid and phospholipid components, their relative compositions were different. Neutral lipid/phospholipid ratios in retinal pigment epithelium and Bruch's membrane/choroid were almost three times higher than in the retina. CONCLUSIONS: This study provides information about the lipid composition of macular and peripheral regions of the human retina, retinal pigment epithelium, and Bruch's membrane/choroid. The methodology employed enabled study of lipids in small amounts of tissue, which will be of value in investigating the biochemical aspects of age-related macular degeneration.

Plasma lipid abnormalities in the miniature poodle with progressive rod-cone degeneration.

The miniature poodle with progressive rod-cone degeneration (prcd) is a model for human retinitis pigmentosa (RP). Since previous studies from several laboratories have shown abnormalities in plasma lipids in human RP, we examined the plasma lipids of prcd-affected animals. Fasting blood was drawn on three separate occasions from affected and control miniature poodles and on one occasion from normal Irish setters and those affected with a different inherited retinal degeneration (rod-cone dysplasia). Plasma phospholipids from prcd-affected animals had significantly lower levels of docosahexaenoic acid (22:6 omega 3) and cholesterol, compared to control miniature poodles. No differences were observed in plasma levels of phospholipids, vitamin E, or vitamin A, and no lipid differences were found between control and affected Irish setters. The ratios of 22:5 omega 3 to 22:6 omega 3 and of 22:4 omega 6 to 22:5 omega 6 were significantly elevated in prcd-affected poodles compared to controls. Since the conversion of 22:5 omega 3 to 22:6 omega 3 and of 22:4 omega 6 to 22:5 omega 6 is catalysed by a delta 4-desaturase, these results are consistent with a defect in desaturase activity in the prcd-affected poodle.

Fatty acid metabolism in normal miniature poodles and those affected with progressive rod-cone degeneration (prcd).

It is possible that a genetic defect observed in the prcd poodle involves the abnormality of an enzyme which functions in phospholipid or lipoprotein metabolism. In our studies thus far, we have been unable to detect any defect in retinal phospholipid biosynthesis, but we have noted a decrease in plasma levels of 22:6w3 which may be a result of an enzyme defect in liver biosynthesis of 22:6w3 from its dietary precursor, linolenic acid, or some defect in the blood lipoprotein transport of this essential fatty acid. If 22:6w3 is essential to the normal elaboration and functioning of the photoreceptor outer segment, it is possible that decreased access to this fatty acid due to lower blood levels of 22:6w3 could cause photoreceptor abnormalities. Further studies are needed to confirm the possible defect in delta-4 desaturase activity and possible dietary modification of the course of this prcd retinal degeneration.

Oil droplets of the retinal epithelium of the rat.

Oil droplets are observed in the retinal pigment epithelium (RPE) of both light- and dark-adapted rats. The accumulation of these droplets is strongly correlated with the clearance of large phagosomes that are found on a circadian basis in the RPE. Although strongly correlated with phagosomes in this way, the composition of the oil droplets does not resemble the lipid composition of rod outer segments (ROS). The important differences are that most of the 18:0 fatty acid (stearic) and virtually all of the polyunsaturated ones (principally docosahexaenoic, 22:6) are absent from the droplets. The major (greater than 95%) constituents of the droplets are triglycerides admixed with small amounts of vitamin A esters (less than 5%). This result suggests that certain of the ROS lipids, especially the polyunsaturated fatty acids, are cleared quickly from the RPE and are, perhaps, recycled to the neural retina. Such a process would account for the known ability of the retina to conserve the polyunsaturates despite dietary deficiencies.

Polyunsaturated fatty acids and vitamin E in rat rod outer segments during light damage.

Previous evidence suggests that lipid peroxidation may initiate photoreceptor damage induced by constant light exposure. In order to investigate the role of the antioxidant vitamin E in light damage, Long-Evans (pigmented) rats were atropinized and exposed to constant fluorescent light (Vita-Lite) of 10-20 foot candles for intervals up to 5 days. Following light exposure, retinal rod outer segments (ROS) were prepared and their lipids extracted. Retinas processed in parallel for morphological examination showed progressive ROS deterioration and selective loss of photoreceptor cells at 3 and 5 days of constant light. Similar to previous observations in undilated albino rats, constant illumination resulted in the specific loss of docosahexaenoic acid (22:6 omega 3) in the ROS. A novel finding in this study was an increase in the content of vitamin E relative to lipid phosphorus, stearic acid, and docosahexaenoic acid in the ROS of constant light-exposed animals.

Rabbit photoreceptor outer segments contain high levels of docosapentaenoic acid.

Docosapentaenoic acid (22:5 omega 6), a minor constituent (less than 4%) of photoreceptor outer segment membranes in all vertebrate species examined to date, comprises 23% of the fatty acids in total lipids from rabbit outer segment membranes. This fatty acid is a significant constituent of each of the three major phospholipid classes in these membranes. The levels of docosahexaenoic acid (22:6 omega 3), the major polyunsaturated fatty acid of most other vertebrate outer segments, was 20%. The sum of 22:5 omega 6 and 22:6 omega 3 in rabbit outer segment membrane lipids is similar to the amount of 22:6 omega 3 usually found in membranes from other vertebrate species.

Effect of light on the metabolism of lipids in the rat retina.

The effect of light on the in vitro incorporation of a variety of radioactive precursors into glycerolipids was tested in isolated retinas of albino rats. There was an increase in the incorporation of [2-3H]myo-inositol, 32Pi, [2-3H]glycerol, and [methyl-3H]choline into retinal phospholipids in light compared to that in darkness. [2-3H]myo-Inositol was incorporated primarily into phosphatidylinositol. 32Pi was incorporated primarily into the phosphoinositides, although there were significant increases in the specific activities of all retinal phospholipids in light compared to those in darkness. Likewise, [2-3H]glycerol incorporation into all retinal phospholipids and diglycerides was greater in light than in the dark. There was no effect of light on the incorporation of [2-3H]ethanolamine into phosphatidylethanolamine or of [3-3H]serine into phosphatidylserine, although these phospholipids were labeled to a greater extent in light with [2-3H]glycerol. There was no effect of light on the incorporation of [3H]palmitic acid into diglycerides and phospholipids, with the exception of phosphatidylinositol. Light also had no effect on the uptake of [2-3H]glycerol, [2-3H]inositol, or [methyl-3H]choline into the retina. We conclude from these studies that light stimulates the phosphoinositide effect in the rat retina. Although some of the results are consistent with a stimulation of de novo synthesis of all lipid classes, our studies with [3H]palmitate, [2-3H]ethanolamine, and [3-3H]serine do not support this conclusion.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of lipid peroxidation on rhodopsin regeneration.

We have developed an in vitro system in which lipid peroxidation can be produced in a predictable fashion and have studied the effects of peroxidation on regenerability of rhodopsin in rod outer segments (ROS). ROS isolated by sucrose flotation from dark-adapted retinas of Rana pipiens were suspended in various concentrations of FeSO4 and 1 or 2 mM ascorbic acid for 10 minutes. An increase in lipid hydroperoxides (measured as conjugated dienes) and a decrease in 22:6 omega 3 were determined for each Fe+2 concentration and used as an index of peroxidation. Following incubation with and without FeSO4, ROS were pelleted and rhodopsin was regenerated with 15 micron 11-cis-retinal. The regenerability of rhodopsin in ROS in which significant lipid peroxidation had taken place was reduced by 40-50% compared to controls. DTPA (diethylenetriamine pentaacetic acid), EDTA, and EGTA protected against lipid peroxidation by Fe+2 and allowed almost complete regeneration of rhodopsin. Incubation with DTPA also prevented the destruction of vitamin E in ROS. Other compounds containing primary amine groups did not protect against peroxidation or loss of regenerability. These experiments indicate that lipid peroxidation is associated with a loss of regenerability of the visual pigment rhodopsin.

Phosphoinositide metabolism in the retina: localization to horizontal cells and regulation by light and divalent cations.

Isolated retinas from Xenopus laevis incorporated greater amounts of [3H]inositol and 32Pi into phosphoinositides when incubated in light than did control retinas incubated in the dark. Inositol was primarily incorporated into phosphatidylinositol (83-86%), while phosphate labeled the polyphosphoinositides (72-79%). The incorporation of radioactive glycerol, serine, choline, or ethanolamine into retinal lipids was unaffected by light. Following incubation with [3H]inositol, the cell type involved in the light response was identified by light and electron microscope autoradiography to be the horizontal cell. These results are consistent with a classic phosphatidylinositol effect in the retina. An interesting feature of this response is that the stimulus (light) is received in the photoreceptor cell and the effect is manifest in the horizontal cell.

Membrane addition to rod photoreceptor outer segments: light stimulates membrane assembly in the absence of increased membrane biosynthesis.

The effect of light on the biosynthesis and the assembly of rod photoreceptor outer segment membranes was analyzed in vitro using retinas from Xenopus laevis. The number of open discs at the base of the outer segment was used as a morphologic index to evaluate relative differences in rates of membrane assembly. Assembly was stimulated in vitro by light, as evidenced by a greater number of open discs that accumulated in light-exposed retinas than that in retinas maintained under the same conditions but in darkness. Quantitative autoradiography indicated, however, that unlike membrane assembly, the incorporation of 3H-leucine or 3H-mannose into proteins by rods was nearly identical in light-stimulated and dark-maintained retinas. Likewise, the synthesis of opsin, a protein destined for the rod outer segment, was not affected by light. We conclude from these studies that the enhanced rate of outer segment membrane assembly that occurs during the early light phase of the diurnal cycle takes place in the absence of an increase in the rate of biosynthesis of opsin, the major outer segment membrane protein.

Metabolism of phosphatidylethanolamine in the frog retina.

The synthesis and the turnover of phosphatidylethanolamine in frog retinal rod outer segments and microsomes were studied by monitoring the incorporation of five radioactive precursors: 32PO4, 33PO4 [3H]glycerol, [3H]serine, and [3H]ethanolamine. 1. Labeled serine was actively incorporated into phosphatidylethanolamine. The kinetics of the labeling patterns in both microsomes and rod outer segments was consistent with formation via decarboxylation of phosphatidylserine. 2. Ethanolamine was found to be an ineffective precursor of phosphatidylethanolamine, suggesting that the major pathway for phosphatidylethanolamine synthesis in the retina is via the decarboxylation reaction. 3. An active methylation of phosphatidylethanolamine to phosphatidylcholine was observed in both retinal microsomes and rod outer segments. 4. The kinetics of labeling of phosphatidylethanolamine in the rod outer segments was different for the various isotopic precursors, and was found to depend on the relative turnover times of the precursor pools. Glycerol was the only precursor that gave a true pulse of radioactivity. 5. The specific activity of phosphatidylethanolamine derived from labeled glycerol declined exponentially, demonstrating that the labeled lipid was diffusely distributed throughout the rod outer segments. The half-life of phosphatidylethanolamine in the rod outer segments was determined to be 18 days. Comparison of this value to the turnover time of rod outer segment integral proteins revealed that rod outer segment lipid is renewed at a faster rate than protein.