|Molecular Vision 2005;
Received 18 January 2005 | Accepted 10 May 2005 | Published 12 May 2005
Lipid differences in rod outer segment membranes of rats with P23H and S334ter opsin mutations
Rex E. Martin,1,2,3
Steven J. Fliesler,4,5 Richard S.
Brush,2,3 Michael J. Richards,4 Stephen A. Hopkins,2,3
Robert E. Anderson1,2,3
Departments of 1Cell Biology and 2Ophthalmology, and the 3Dean A. McGee Eye Institute, University of Oklahoma Health Sciences Center, Oklahoma City, OK; Departments of 4Ophthalmology and 5Pharmacological and Physiological Science, Saint Louis University School of Medicine, St. Louis, MO
Correspondence to: Robert E. Anderson, MD, PhD, Department of Cell Biology, Dean A. McGee Eye Institute, University of Oklahoma Health Sciences Center, 608 Stanton L. Young Boulevard, Oklahoma City, OK, 73104; Phone: (405) 271-8244; FAX: (405) 271-8128; email: firstname.lastname@example.org
Purpose: Retinal degenerations and diets low in n-3 fatty acids are associated with decreased docosahexaenoic acid (22:6n-3) in retina and plasma and with sterol abnormalities in retina and sperm. Using wild type (WT) and transgenic rats with P23H and S334ter opsin mutations, we evaluated retinal cholesterol levels, cholesterol synthesis, and fatty acid compositions of phospholipid classes in animals fed diets enriched in n-3 or n-6 polyunsaturated fatty acids.
Methods: Pregnant WT and heterozygous P23H and S334ter transgenic (TG) rats were fed safflower (safflower oil [SO], high n-6, trace n-3 fatty acids) or flaxseed oil (flaxseed oil [FO], high n-3, moderate n-6 fatty acids) diets beginning at E15, and pups were continued on the diets after weaning. Rod outer segment (ROS) membranes were prepared from 55-day-old rats, and the ratios of total fatty acid to cholesterol and the fatty acid compositions of phosphatidylcholine (PC), phosphatidylethanolamine (PE), and phosphatidylserine (PS) in ROS were determined. Intravitreal injections of [3H]acetate were given to 35-day-old WT and TG rats fed standard chow-diets. Endogenous cholesterol mass and de novo [3H]cholesterol synthesis were measured and normalized to total ROS fatty acid content. Multivariate analysis of variance (ANOVA) with post hoc Newman-Keuls tests were used to determine statistical differences.
Results: The relative levels of PC, PE, and PS were similar in all three rat strains independent of diet. Total lipids, PC, PS, and PE of ROS FO fed rats had higher levels of 22:6n-3 and lower levels of 22:5n-6 than those fed SO. Rats fed SO had higher levels of 22:5n-6 than those fed FO. Significant increases in 18:1n-9 were seen in PC and PS of P23H and S334ter rats; arachidonate (20:4n-6) increased only in PE. These changes were independent of diet. ROS membranes of transgenic rats were cholesterol enriched, relative to WT ROS, yet retinal cholesterol synthesis was not altered. Plasma cholesterol levels of transgenic rats were not different from those of WT rats.
Conclusions: Endogenous levels of cholesterol, 18:1n-9, 20:4n-6, 22:5n-6, and 22:6n-3 were altered in ROS membranes of P23H and S334ter compared to WT rats. There appear to be two pools of 22:6n-3 in rat ROS, one that is sensitive to retinal degenerations and one that is not. The stress induced reduction in 22:6n-3 was not specific to any phospholipid class and was not caused by alteration of relative amounts of PC, PS, or PE in the membrane. Elevated retinal cholesterol may be a result of either an increased half life or an increased uptake of cholesterol from the blood.
While much is known regarding the mechanism of phototransduction, many questions remain regarding the roles of docosahexaenoic acid (DHA; 22:6n-3) and cholesterol in the retina in general, and in photoreceptor outer segments in particular. For example, why is there such an abundance of 22:6n-3 in photoreceptors ? Why is there more cholesterol in the basal disks and plasma membrane of the rod outer segment (ROS) than in the more distal disks [2,3]? What role does cholesterol play in the rod outer segment (ROS) ? Why is there a preferential loss (relative to other fatty acids) of 22:6n-3 under conditions of photoreceptor stress [5-12]? The present study focuses on how 22:6n-3, other fatty acids, and cholesterol are affected in the course of photoreceptor degeneration.
When 22:6n-3 and its n-3 fatty acid precursors are withheld from the diet, there are fewer of these fatty acids in the plasma, brain, and retina; the development of the brain and retina are adversely affected; and the 22:6n-3 is partially replaced by docosapentaenoic acid (22:5n-6) [11-20]. In studies of rat brain, it was shown that n-3 fatty acid deprivation led to 22:6n-3 loss from phosphatidylcholine (PC), phosphatidylethanolamine (PE), and phosphatidylserine (PS). Significantly, however, there was a relative loss of PS and an increase in PC, which contains relatively low levels of 22:6n-3 compared to PE and PS [14,21]. This diet associated reorganization of phospholipid abundance in brain lead us to hypothesize that, during dietary n-3 fatty acid deprivation or photoreceptor stress, a similar loss of PS and gain in PC would be seen in retina.
We previously showed that: First, the rate of photoreceptor degeneration could neither be accelerated nor slowed by adding (or taking away) 22:6n-3 from the ROS . Second, there is a direct relationship between the rate of degeneration and 22:6-n-3 level in ROS . Third, there appear to be two pools of DHA in ROS. One pool that is lost during degeneration and another that is retained .
To determine if 22:6n-3 loss is due to loss from all membrane lipids or to changes in the relative amounts of PC, PE, and PS, we took advantage of an increasingly well characterized model of photoreceptor damage and 22:6n-3 loss, P23H and S334ter transgenic rats. The rats were fed diets containing either n-3 fatty acid enriched flaxseed oil (FO) or n-6 fatty acid enriched safflower oil (SO). Rod outer segments (ROS) were collected from retinas of these transgenic rats and their wild type (WT) counterparts at 55 days of age, and the fatty acid composition of PC, PE, and PS in the ROS was then determined.
Retinal degenerations are associated with lipid abnormalities other than those related to fatty acid metabolism. Perturbations in cholesterol metabolism have been reported to be associated with retinal degeneration [22,23]. Using Royal College of Surgeons (RCS) rats, it was shown that the ratios of cholesterol to phospholipid decreased in ROS and increased in plasma membrane of rat photoreceptors . If, as hypothesized above, there is a loss of PS from the disk membranes (rod outer segments, ROS) of P23H or S334ter transgenic rats, then that could explain the increase in cholesterol content (relative to phospholipid) that was seen in the RCS rats. To test this hypothesis in "normal" rat chow-fed P23H and S334ter rats, two determinations were made: First, the mole ratios of cholesterol to fatty acid in the ROS, and second, the rates of de novo cholesterol synthesis in the retina.
Breeding pairs of homozygous transgenic rats with P23H (line 3) and S334ter (line 4) rhodopsin mutations were the generous gift from Dr. Matthew LaVail (University of California, San Francisco, CA). These rats were used to establish breeding colonies of homozygous transgenic rats at the Dean A. McGee Eye Institute. Homozygous P23H rats carry two copies of a mouse opsin gene that encodes a proline to histidine substitution at residue 23. S334ter rats express a mouse opsin gene that bears a termination codon at residue 334, which causes a C-terminal truncation of opsin . Homozygous mutant males were bred to WT female Sprague-Dawley (SDCD) rats so that they carried a single copy of the transgene and two copies of the wild type rhodopsin gene. Pregnant rats carrying heterozygous litters were maintained in dim-cyclic light (5 lux, 12 h on/off) and, from embryonic day 15 through weaning, given one of two isocaloric, semisynthetic diets (Dyets Inc., Bethlehem, PA). Their pups remained on that diet for the duration of the experiment (55 days). At postnatal day 5 (P5), litters were culled or fostered (as appropriate) to 10 pups. Rats were fed chows containing 10% (by weight) dietary fat exclusively derived from flaxseed (FO; n-3 enriched) or safflower oil (SO, devoid of n-3 fats). The fatty acid content of the diets was determined by gas-liquid chromatographic (GLC) analysis . All procedures adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and were in accordance with the University of Oklahoma Institutional Animal Care and Use Committee (IACUC) approved protocol.
Rats were bred and raised as described above, except both the dams and the pups were fed a standard Purina rat chow. At 35 days of age (P35), 4 rats from each litter were selected for experimentation. Intravitreal injections were performed essentially as described previously . In brief, each rat was sedated by intramuscular injection of a mixture of a Ketamine (120 mg/kg) and xylazine (6 mg/kg). Both pupils were dilated with 2.5% phenylepherine HCl (Bausch & Lomb, Tampa, FL) and topical drops of anesthetic (0.5% proparacaine HCl, AK-Taine; Bausch & Lomb) were applied to the cornea. The left eye of each rat was injected intravitreally with 0.5 mCi [3H]acetate (Na salt; American Radiolabeled Chemicals, St. Louis, MO) in 5 μl of PBS [50 mM phosphate and 150 mM NaCl (pH 7.4); Ca and Mg free]. The temporal conjunctiva was partially reflected and an entry port was made with a 27 gauge surgical needle at the temporal equator just posterior to the limbus. The injection was made with a 10 μl Hamilton syringe fitted with a blunt tipped, 30 gauge needle. Air (1 μl) followed the injection of the 5 μl of PBS/[3H]acetate. Visualization of the intraocular air bubble ensured complete delivery of [3H]acetate and aided closure of the injection site. Rats were kept in an exhaust hood suitable for use with radioactive materials according to their accustomed lighting regimen, and were euthanized with CO2 24 h later. Retinas and livers were dissected and frozen in liquid N2. Whole blood (about 2 ml) was collected by cardiac puncture, from which plasma was obtained by centrifugation (see "Plasma and ROS preparation" below), and all tissues were stored at -80 °C until further processing.
Endogenous and newly synthesized cholesterol determination
The determination of cholesterol mass and specific activity of radiolabeled cholesterol was performed as described in detail elsewhere . In brief, tissues were saponified in methanolic KOH, and the nonsaponifiable lipids were extracted with petroleum ether and analyzed by reverse phase radio-HPLC. Cholesterol mass was quantified by UV peak area integration (detection at 205 nm), in comparison with the empirically determined response factor for a cholesterol standard (Sigma, St. Louis, MO), within the linear range of the assay (reproducibility, ±3%). Radiolabeled cholesterol was quantified by post-UV detection using a Model C505 Flow Scintillation Analyzer (Packard Instrument Co., Meriden, CT).
Plasma and ROS preparation
Blood samples were centrifuged (10 min at 2,000x g) in EGTA containing tubes to obtain at least 100 μl of plasma for lipid extraction. Retinas were harvested from eyes as described , and 4 retinas were pooled for ROS membrane preparation essentially as previously described [28,29]. In brief, retinas were homogenized in 10 mM Tris-acetate (pH 7.4) containing 70 mM NaCl, 2 mM MgCl2, and 0.1 mM EGTA, and layered onto discontinuous sucrose gradients, and centrifuged at 82,000x g for 70 min. The ROS enriched bands were collected from discontinuous sucrose gradients at the 1.11-1.13 g/ml (28%-34%) interface and diluted with 10 volumes of 10 mM Tris-acetate (pH 7.4) containing 70 mM NaCl, 5 mM MgCl2 and 0.1 mM EGTA, and centrifuged at 27,000x g for 30 min. The ROS membrane pellet was resuspended in H2O for lipid extraction and fatty acid analysis (see the "Lipid extraction" section below). The remaining material not in the ROS enriched band ("rest of retina") was collected similarly by centrifugation and saved for comparative lipid analysis (see the "Lipid extraction" section). Individual ROS preparations were of similar volume, protein content, and purity [28,29]. All procedures were performed at 0-4 °C.
Lipids from plasma and retinal membrane specimens were extracted in chloroform:methanol (C:M; 1:1) following the method of Bligh and Dyer . Crude lipid extracts were obtained from 200 μl plasma or 250 μl aqueous suspensions of ROS membranes. Proteins in the crude lipid extracts were collected by centrifugation (1,000x g for 5 min), washed with 2 volumes of C:M, and the wash was combined with the crude lipid extract. Water containing 0.33 mM diethylenetriaminepentaacetic acid (DTPA) was added, and the organic layer was removed. The aqueous layer was then re-extracted with chloroform. This two phase system was capped under N2, mixed, and centrifuged for 10 min at 1000 g. The upper (aqueous) phase was discarded, leaving the purified lipid extract, which was stored under N2 in a known volume of C:M (2:1, by volume).
Thin layer chromatography
Lipid classes were separated using high performance thin layer chromatography (TLC) plates (Analtech, Newark, DE) and the two dimensional, three solvent system method described previously [31,32]. Prior to use, the plates were saturated with 3% magnesium acetate and activated for 2 h at 110 °C. Lipid spots were localized by spraying the plates with a methanolic solution of dichlorofluorescein.
Fatty acid derivatization and gas-liquid chromatography
Dichlorofluorescein stained plates were scraped, and fatty acids from individual lipid spots were converted to their corresponding methyl esters (FAMEs) prior to gas-liquid chromatography (GLC) analysis. Regions of silica scraped from TLC plates (corresponding to the resolved lipid species) or an aliquot of the lipid extract was added to a screw top test tube and known amounts of pentadecanoic acid (15:0), heptadecanoic acid (17:0), and heneicosanoic acid (21:0) were added as internal standards. Toluene (200 μl) and 2% methanolic H2SO4 (1 ml) were added, and the mixture was sealed under N2 with Teflon lined caps, sonicated, and heated at 100 °C for 45 min. Tubes were cooled on ice, 1.2 ml of H2O was added, and the FAMEs were extracted with three 2.4-ml portions of hexane. The fatty acid compositions of retina and plasma were determined by GLC with a DB-225 capillary column (30 m with 0.25 mm internal diameter; J&W Scientific, Folsom, CA) and an Agilent 6890N gas-liquid chromatograph that was equipped with a model 7683 autosampler (Wilmington, DE). The samples were dissolved in 20 μl nonane and 3 μl of each was injected at 250 °C with the split ratio set to 20:1. The column temperature was programmed to hold at 160 °C for 1 min, then raised to 220 °C at 1 °C/min, and held at 220 °C for 10 min. Helium carrier gas flowed linearly at 42.4 cm/sec (1 ml/min). The hydrogen flame ionization detector temperature was maintained at 270 °C. The chromatographic peaks were integrated and processed with ChemStation® software (Agilent Technologies). FAMEs were identified by comparison of their relative retention times with authentic standards, and the relative mole percentages were calculated. The mass of each fatty acid was determined relative to the internal standards, and results are reported in nanomoles or as relative mole percent of total fatty acids.
All values are expressed as mean±standard deviation. There were 6 independent ROS fatty acid preparations (determinations) per condition, and there were 4 retinas taken from 2 rats in a single litter per preparation. Experiments on retinal [3H]cholesterol synthesis used 6 independent preparations per condition, and there were 11-18 preparations per condition in the analyses of endogenous plasma cholesterol and fatty acids. Multivariate analysis of variance (ANOVA) with post hoc Newman-Keuls tests confirmed statistical differences in fatty acid content between animal strains and diets (p<0.05). Student's t-test confirmed differences in endogenous cholesterol mass and [3H]cholesterol specific activities (dpm/nmol) in tissue specimens(p<0.05).
Lipid class and fatty acid determinations
We showed previously that ROS from P23H and S334ter transgenic rats had lower levels of 22:6n-3 than ROS from WT rats [11,12]. Moreover, the degree of DHA loss has been directly correlated with the severity of the retinal degeneration . Here we sought to determine if the reduction in 22:6n-3 was due to an increase in PC, which has the lowest levels of 22:6n-3, or to a loss of 22:6n-3 in the phospholipid classes. Figure 1 compares the fatty acid content of PC, PE, and PS from ROS of WT, P23H, and S334ter rats that were fed either FO (n-3 enriched) or SO (n-3 deficient). Indicating that neither diet nor disease could alter the relative abundance of major phospholipids in ROS, approximately 15%, 45%, and 40% of the combined total fatty acids were, respectively, esterified in PS, PC, and PE (Figure 1A,B). However, there was, in both mutant strains, a significant reduction in 22:6n-3 (compared to WT) in individual phospholipids of P23H and S334ter. Thus, the reduction of 22:6n-3 in the ROS of mutant animals was specific for the fatty acid itself, and was not attributable to the loss of a particular phospholipid.
There were changes in other fatty acids within each phospholipid class, some of which were related to degeneration, while others were related to diet. Degeneration specific increases were seen in the levels of arachidonate (20:4n-6) and oleate (18:1n-9). Specifically, 20:4n-6 increased significantly in PE, but not PS or PC, in transgenic rats fed either FO or SO. The 20:4n-6 content of PE was higher only in the S334ter TG animals compared to WT animals (respectively, 28% and 56% in FO fed animals and 17% and 30% in SO fed animals). This increase was independent of diet. Similarly, 18:1n-9 concentrations were insensitive to diet, but were sensitive to retinal degeneration (Figure 1). The 18:1n-9 content of PC was higher in the P23H and S334ter TG animals (22% and 43% in FO fed rats and 36% and 59% in SO fed rats, respectively), but significant increases in PS-18:1n-9 were seen only in the S334ter rats (97% and 104% in SO fed and FO fed animals, respectively).
When mutant animals with degenerating photoreceptors are fed laboratory chow, the relative levels of 22:5n-6 and 22:6n-3 in ROS are lower than those of the WT animal [11,12]. In the current study however, 22:6n-3 was diminished only when the rats were fed FO, and 22:5n-6 was diminished only when SO was fed. There was a tendency for the change to be more substantial in the S334ter rats than in P23H rats. ROS phospholipids from FO fed P23H rats had less 22:6n-3 than ROS from WT animals (25%, 30%, and 14% in PS, PC, and PE, respectively). In the S334ter rats, the corresponding values were 37%, 50%, and 54%, respectively. ROS from P23H, SO fed rats had 19%, 28%, and 21% less 22:5n-6 than their WT counterparts in PS, PC, and PE, respectively. The corresponding values for loss of 22:5n-6 in S334ter rats were 68%, 78%, and 71%, respectively. Collectively, these results show that the opsin mutations caused 20:4n-6 content to increase in PE, while 22:6n-3 and 22:5n-6 were lower in PC, PE, and PS.
To standardize values between samples, the ratios of total cholesterol (cholesterol esters plus free cholesterol) to total fatty acid were determined (cholesterol/fatty acid mole ratios). Figure 2 shows that the diets had no effect on these ratios. Also, the data clearly show that there was an increase in the relative abundance of cholesterol in the retinas of the TG rats. The cholesterol/fatty acid mole ratios of ROS membranes from P23H rats were 55% greater than in WT rats; values in S334ter rats were 73% greater than in WT. The elevated cholesterol/fatty acid mole ratios that were seen in ROS membranes of the TG rats led to the hypothesis that it could be attributed to an increase in either retinal cholesterol synthesis or in uptake of plasma cholesterol. To test this hypothesis, we used standard chow-fed, age matched, WT and transgenic rats to determine levels of plasma cholesterol and the rates at which the retinas synthesize cholesterol. HPLC analysis of plasma showed that the transgenes had no effect on plasma cholesterol levels (Figure 3). To examine retinal cholesterol synthesis, [3H]acetate was injected intraocularly. The content of [3H]cholesterol in blood and retinas 24 h later was determined. Figure 4 illustrates that intraocularly injected [3H]acetate was converted to [3H]cholesterol in the retina, with comparable specific activity being achieved in ROS membranes and residual retina fractions. Genotype made no difference in the amount of [3H]cholesterol recovered from plasma, which represented on average, <0.5% of the cholesterol specific activity in retinas. However, there was a trend (although statistically insignificant) toward lower retinal cholesterol synthesis in P23H and S334ter TG rats, compared to WT rats, both in the ROS and the "rest of retina" fractions.
Lipid and fatty acid determinations
Numerous studies show that, whether due to strong light or genetic mutation, photoreceptor stress is accompanied by a relative loss of 22:6n-3 from the ROS membrane [6,7,9-12,19,33-37]. This reduction is thought to be a cellular response that diminishes the potential for lipid peroxidation and free radical damage  and may represent a protective mechanism to reduce the likelihood of photoreceptor damage or stress. Bush et al.  showed that, in times of ROS 22:6n-3 depletion, photon absorption by rhodopsin is reduced and thus the retina's sensitivity to light is also reduced. Recent studies in n-3 deficient ROS by Niu et al.  showed a reduced rhodopsin activation, reduced rhodopsin-transducin (Gt) coupling, reduced cGMP phosphodiesterase activity, and slower formation of metarhodopsin II-Gt complex, relative to the animals fed the n-3 adequate diet. Thus, we would expect reduced efficiency in visual transduction in the P23H and S334ter animals. Anderson and Penn  recently hypothesized that the reduction in 22:6n-3 in environmental and genetic stressed animals is a neuroprotective response to control the number of photons captured by rhodopsin and the efficiency of visual transduction, under conditions where excessive activation of the transduction cascade could lead to cell death.
We previously showed that the ROS levels of 22:6n-3 are lower in P23H and S334ter rats compared to WT, and that the loss of 22:6n-3 from ROS membranes is directly related to the severity of the photoreceptor degeneration . Subsequently, we showed that supplementation with FO does not prevent or slow the retinal degeneration, although there is a higher level of 22:6n-3 in ROS total lipids of FO compared to SO supplemented animals. In the current study, we determined that the reduction of 22:6n-3 in ROS from the mutant rats is due to loss from the three major phospholipid classes rather than to a relative increase in PC, which contains lower levels of 22:6n-3 than PE and PS, Figure 1. This suggests that relatively less 22:6n-3 is incorporated into newly synthesized glycerolipids in the mutant retinas compared to WT. Whether this is due to lower levels of 22:6n-3 in the fatty acid CoA pools is not known, but seems unlikely since the FO supplemented rats should have a large 22:6n-3 pool. The alternative is that the stress of the mutation alters the enzyme activities of the acyl transferases that catalyze the esterification of acyl CoAs into glycerolipids.
What is the fate of the 22:6n-3 that is "lost" when the retina is injured or stressed? Bazan  showed that the first biochemical event to follow neuronal damage is the release of free fatty acids from membrane lipids. These fatty acids are either rapidly re-esterified into phospholipids or are metabolized to biologically active molecules that can be either beneficial or detrimental. Classes of molecules termed "neuroprostanes" are produced via free radical induced oxidation of 22:6n-3; these reactive molecules adduct covalently to proteins, thus modifying their function . Gu et al.  found carboxyethyl pyrrole (CEP) protein adducts derived from 22:6n-3 that are thought to be markers of immune mediated pathology in macular degeneration. There is also a lipoxygenase derived metabolite of 22:6n-3 (neuroprotectin-1) that is protective in transient focal ischemia . In oxidatively stressed cultured retinal pigment epithelial cells, neuroprotectin-1 is actually more abundant than free 22:6n-3, and it inhibits apoptotic signaling . Further experiments will be required to assess the ramifications of 22:6n-3 reduction during photoreceptor stress.
Like the fatty acids mentioned above, the levels of cholesterol changed in membranes of degenerating photoreceptor cells. In Royal College of Surgeons (RCS) rats, Boesze-Battaglia et al.  examined the cholesterol/phospholipid ratios in ROS plasma membrane and disk membrane fractions. They showed that the ROS plasma membrane has about half the cholesterol content (relative to phospholipid) as in the normal rat (0.40): Ratios in ROS disk membranes are little different. However, our findings are quite different: We observe 55% and 73% more cholesterol (relative to fatty acid) in ROS membranes of P23H and S334ter TG rats, respectively, compared with WT rat ROS. Although we analyzed whole (not fractionated) ROS membranes, rather than plasma membranes and disk membrane fractions, and rats from a different strain, the apparent discrepancy between the findings in these two studies is difficult to explain. One might speculate that the difference may be attributable to the fact that, unlike the P23H and S334ter opsin mutations (which directly affect the rod photoreceptor cell), the defective protein in RCS rats (Mertk-/-) is expressed in the retinal pigmented epithelium (RPE), and it is the pathological accumulation of debris between the photoreceptors and the pigmented epithelium that is thought to compromise visual physiology .
Taken together, Figure 1A,B and Figure 2A,B suggest that ROS cholesterol levels change independent of phospholipid (or fatty acid) abundance. Because cholesterol is more concentrated at the base of the ROS than in its more distal regions (proximal to the RPE) [3,48,49], the altered cholesterol content could be attributed to a preferential loss of the relatively cholesterol deficient end of the ROS, leaving a shorter, more cholesterol enriched outer segment in the mutant rats. Additional experiments will be required to confirm the validity of this speculation. In considering what could cause the cholesterol levels in photoreceptors of a degenerating retina to change, we evaluated the possibility of enhanced de novo synthesis in the retina. We measured endogenous cholesterol mass in plasma and neural retina as well as the formation and specific activity of radiolabeled cholesterol in plasma and retina following intravitreal injection of [3H]acetate. Taken together, Figure 3 and Figure 4 show that these retinal degenerations did not alter endogenous plasma cholesterol and, if anything, retinal cholesterol synthesis was slightly reduced in the TG mutants, relative to WT rats.
Conceivably, the observed increase in ROS cholesterol content could be due to either an increase in the half life of cholesterol in the ROS or to increased uptake of cholesterol by the retina (i.e., from the blood). However, data from the present study can neither confirm nor exclude either of these possibilities. There is precedent for cholesterol level alterations in association with retinal degeneration. For example, in male patients affected either with Usher's Syndrome Type II or with non-Usher's retinitis pigmentosa (RP), sperm contain significantly elevated levels of cholesterol, but not of desmosterol, an immediate precursor to cholesterol and normally the major sterol of sperm [50,51]. RP is characterized by degeneration of rod photoreceptors, while Usher's Syndrome involves a primary defect in the neurosensory cells of the inner ear with a secondary, RP-like defect occurring in a subset of affected individuals. Perhaps not coincidentally, retinal photoreceptor cells, the neurosensory cells of the inner ear, and sperm are all ciliated cells. While the relationship between these facts is not understood at present, it serves to illustrate the fact that modulation of the steady state levels of cholesterol in retinal photoreceptor cells does occur and is not restricted to transgenic mice. It should be noted that elevated cholesterol content can compromise the visual transduction cascade. For example, Boesze-Battaglia and Albert  showed that photic stimulation of rod cGMP dependent phosphodiesterase (cGMP-PDE) is hindered by a high cholesterol environment: The ROS plasma membrane (which contains >3 fold more cholesterol than disk membranes) exhibits little or no ability to activate cGMP-PDE. Yet when the majority of its cholesterol is oxidized to cholestenone, the cGMP-PDE activity becomes comparable to that in disk membranes. In addition, it has been demonstrated recently  that the metarhodopsin I intermediate of the visual pigment bleaching cascade is stabilized when cholesterol is present during reconstitution of ROS derived membranes. These findings predict that localization of rhodopsin and other phototransduction cascade components in a high cholesterol environment would hinder the kinetics of phototransduction. Indeed, such components (e.g., transducin, cGMP-PDE) and structural proteins (e.g., ROM-1, caveolin-1) are known to partition into cholesterol enriched "raft" domains in ROS membranes [54-57]. To the extent that changes in the steady state levels of cholesterol can alter the relative percentage of total ROS membrane area occupied by rafts, such changes conceivably could alter the relative partitioning of these ROS resident proteins into rafts or non-raft membrane domains and, concomitantly, modulate the phototransduction cascade. In addition, we cannot exclude the possibility that alterations in fatty acid content directly affect lipid raft composition and formation, which in turn may impact the incorporation and intramembrane distribution (raft and non-raft domains) of ROS destined proteins. Hence, in addition to the shortening of ROS observed in the P23H and S334ter retinal degeneration models, the electrophysiological function of the remnant photoreceptors may be further impacted by alterations in the lipid composition of their ROS membranes.
This research was supported by NIH grants EY04149 (REA), EY00871 (REA), EY12190 (REA), RR017703 (REA), and EY07361 (SJF); by funding from the Presbyterian Health Foundation (REA); by unrestricted departmental grants from Research to Prevent Blindness (REA and SJF); and by a grant from the Foundation Fighting Blindness (REA).
1. Fliesler SJ, Anderson RE. Chemistry and metabolism of lipids in the vertebrate retina. Prog Lipid Res 1983; 22:79-131.
2. Boesze-Battaglia K, Hennessey T, Albert AD. Cholesterol heterogeneity in bovine rod outer segment disk membranes. J Biol Chem 1989; 264:8151-5.
3. Boesze-Battaglia K, Fliesler SJ, Albert AD. Relationship of cholesterol content to spatial distribution and age of disc membranes in retinal rod outer segments. J Biol Chem 1990; 265:18867-70.
4. Fliesler SJ, Keller RK. Isoprenoid metabolism in the vertebrate retina. Int J Biochem Cell Biol 1997; 29:877-94.
5. Anderson RE, Maude MB, Nilsson SE, Narfstrom K. Plasma lipid abnormalities in the abyssinian cat with a hereditary rod-cone degeneration. Exp Eye Res 1991; 53:415-7.
6. Bicknell IR, Darrow R, Barsalou L, Fliesler SJ, Organisciak DT. Alterations in retinal rod outer segment fatty acids and light-damage susceptibility in P23H rats. Mol Vis 2002; 8:333-40 <http://www.molvis.org/molvis/v8/a40/>.
7. Aguirre GD, Acland GM, Maude MB, Anderson RE. Diets enriched in docosahexaenoic acid fail to correct progressive rod-cone degeneration (prcd) phenotype. Invest Ophthalmol Vis Sci 1997; 38:2387-407.
8. Anderson RE, Maude MB, Narfstrom K, Nilsson SE. Lipids of plasma, retina, and retinal pigment epithelium in Swedish briard dogs with a slowly progressive retinal dystrophy. Exp Eye Res 1997; 64:181-7.
9. Anderson RE, Maude MB, Bok D. Low docosahexaenoic acid levels in rod outer segment membranes of mice with rds/peripherin and P216L peripherin mutations. Invest Ophthalmol Vis Sci 2001; 42:1715-20.
10. Anderson RE, Sieving PA, Maude MB, Naash MI. Mice with G90D rhodopsin mutations have lower DHA in rod outer segment membranes than control mice. In: Anderson RE, Hollyfield JG, LaVail MM, editors. New insights into retinal degenerative diseases. New York: Plenum; 2001. p. 235-45.
11. Anderson RE, Maude MB, McClellan M, Matthes MT, Yasumura D, LaVail MM. Low docosahexaenoic acid levels in rod outer segments of rats with P23H and S334ter rhodopsin mutations. Mol Vis 2002; 8:351-8 <http://www.molvis.org/molvis/v8/a42/>.
12. Martin RE, Ranchon-Cole I, Brush RS, Williamson CR, Hopkins SA, Li F, Anderson RE. P23H and S334ter opsin mutations: Increasing photoreceptor outer segment n-3 fatty acid content does not affect the course of retinal degeneration. Mol Vis 2004; 10:199-207 <http://www.molvis.org/molvis/v10/a25/>.
13. Galli C, Trzeciak HI, Paoletti R. Effects of essential fatty acid deficiency on myelin and various subcellular structures in rat brain. J Neurochem 1972; 19:1863-7.
14. Murthy M, Hamilton J, Greiner RS, Moriguchi T, Salem N Jr, Kim HY. Differential effects of n-3 fatty acid deficiency on phospholipid molecular species composition in the rat hippocampus. J Lipid Res 2002; 43:611-7.
15. Anderson RE, Maude MB. Lipids of ocular tissues. 8. The effects of essential fatty acid deficiency on the phospholipids of the photoreceptor membranes of rat retina. Arch Biochem Biophys 1972; 151:270-6.
16. Anderson GJ, Connor WE, Corliss JD, Lin DS. Rapid modulation of the n-3 docosahexaenoic acid levels in the brain and retina of the newly hatched chick. J Lipid Res 1989; 30:433-41.
17. Anderson GJ, Connor WE, Corliss JD. Docosahexaenoic acid is the preferred dietary n-3 fatty acid for the development of the brain and retina. Pediatr Res 1990; 27:89-97.
18. Wiegand RD, Koutz CA, Stinson AM, Anderson RE. Conservation of docosahexaenoic acid in rod outer segments of rat retina during n-3 and n-6 fatty acid deficiency. J Neurochem 1991; 57:1690-9.
19. Wiegand RD, Koutz CA, Chen H, Anderson RE. Effect of dietary fat and environmental lighting on the phospholipid molecular species of rat photoreceptor membranes. Exp Eye Res 1995; 60:291-306.
20. Ahmad A, Murthy M, Greiner RS, Moriguchi T, Salem N Jr. A decrease in cell size accompanies a loss of docosahexaenoate in the rat hippocampus. Nutr Neurosci 2002; 5:103-13.
21. Hamilton L, Greiner R, Salem N Jr, Kim HY. n-3 fatty acid deficiency decreases phosphatidylserine accumulation selectively in neuronal tissues. Lipids 2000; 35:863-9.
22. Boesze-Battaglia K, Organisciak DT, Albert AD. RCS rat retinal rod outer segment membranes exhibit different cholesterol distributions than those of normal rats. Exp Eye Res 1994; 58:293-300.
23. Connor WE, Weleber RG, DeFrancesco C, Lin DS, Wolf DP. Sperm abnormalities in retinitis pigmentosa. Invest Ophthalmol Vis Sci 1997; 38:2619-28.
24. Steinberg RH, Flannery JG, Naash M, Oh P, Matthes MT, Yasumura D, Lau-Villacorta C, Chen J, LaVail MM. Transgenic rat models of inherited retinal degeneration caused by mutant opsin genes. Invest Ophthalmol Vis Sci 1996; 37:S698.
25. Fliesler SJ, Keller RK. Metabolism of [3H]farnesol to cholesterol and cholesterogenic intermediates in the living rat eye. Biochem Biophys Res Commun 1995; 210:695-702.
26. Fliesler SJ, Richards MJ, Miller C, Peachey NS. Marked alteration of sterol metabolism and composition without compromising retinal development or function. Invest Ophthalmol Vis Sci 1999; 40:1792-801.
27. Winkler BS. The electroretinogram of the isolated rat retina. Vision Res 1972; 12:1183-98.
28. Papermaster DS, Dreyer WJ. Rhodopsin content in the outer segment membranes of bovine and frog retinal rods. Biochemistry 1974; 13:2438-44.
29. Wiegand RD, Giusto NM, Rapp LM, Anderson RE. Evidence for rod outer segment lipid peroxidation following constant illumination of the rat retina. Invest Ophthalmol Vis Sci 1983; 24:1433-5.
30. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959; 37:911-7.
31. Martin RE. Docosahexaenoic acid decreases phospholipase A2 activity in the neurites/nerve growth cones of PC12 cells. J Neurosci Res 1998; 54:805-13.
32. Martin RE, Wickham JQ, Om AS, Sanders J, Ceballos N. Uptake and incorporation of docosahexaenoic acid (DHA) into neuronal cell body and neurite/nerve growth cone lipids: evidence of compartmental DHA metabolism in nerve growth factor-differentiated PC12 cells. Neurochem Res 2000; 25:715-23.
33. Wiegand RD, Joel CD, Rapp LM, Nielsen JC, Maude MB, Anderson RE. Polyunsaturated fatty acids and vitamin E in rat rod outer segments during light damage. Invest Ophthalmol Vis Sci 1986; 27:727-33.
34. Penn JS, Anderson RE. Effect of light history on rod outer-segment membrane composition in the rat. Exp Eye Res 1987; 44:767-78.
35. Penn JS, Anderson RE. Effects of light history on the rat retina. In: Osborne NN, Chader G, editors. Progress in retinal research. Vol 11. New York: Pergamon Press; 1992. p. 75-98.
36. Li F, Cao W, Anderson RE. Protection of photoreceptor cells in adult rats from light-induced degeneration by adaptation to bright cyclic light. Exp Eye Res 2001; 73:569-77.
37. Li F, Cao W, Anderson RE. Alleviation of constant-light-induced photoreceptor degeneration by adaptation of adult albino rat to bright cyclic light. Invest Ophthalmol Vis Sci 2003; 44:4968-75.
38. Anderson RE, Maude MB, Alvarez RA, Acland G, Aguirre GD. A hypothesis to explain the reduced blood levels of docosahexaenoic acid in inherited retinal degenerations caused by mutations in genes encoding retina-specific proteins. Lipids 1999; 34:S235-7.
39. Bush RA, Malnoe A, Reme CE, Williams TP. Dietary deficiency of N-3 fatty acids alters rhodopsin content and function in the rat retina. Invest Ophthalmol Vis Sci 1994; 35:91-100.
40. Niu SL, Mitchell DC, Lim SY, Wen ZM, Kim HY, Salem N Jr, Litman BJ. Reduced G protein-coupled signaling efficiency in retinal rod outer segments in response to n-3 fatty acid deficiency. J Biol Chem 2004; 279:31098-104.
41. Anderson RE, Penn JS. Environmental light and heredity provoke adaptive changes in retinal DHA levels that affect retinal function. Lipids. In press 2005.
42. Bazan NG Jr. Effects of ischemia and electroconvulsive shock on free fatty acid pool in the brain. Biochim Biophys Acta 1970; 218:1-10.
43. Fam SS, Murphey LJ, Terry ES, Zackert WE, Chen Y, Gao L, Pandalai S, Milne GL, Roberts LJ, Porter NA, Montine TJ, Morrow JD. Formation of highly reactive A-ring and J-ring isoprostane-like compounds (A4/J4-neuroprostanes) in vivo from docosahexaenoic acid. J Biol Chem 2002; 277:36076-84.
44. Gu X, Meer SG, Miyagi M, Rayborn ME, Hollyfield JG, Crabb JW, Salomon RG. Carboxyethylpyrrole protein adducts and autoantibodies, biomarkers for age-related macular degeneration. J Biol Chem 2003; 278:42027-35.
45. Marcheselli VL, Hong S, Lukiw WJ, Tian XH, Gronert K, Musto A, Hardy M, Gimenez JM, Chiang N, Serhan CN, Bazan NG. Novel docosanoids inhibit brain ischemia-reperfusion-mediated leukocyte infiltration and pro-inflammatory gene expression. J Biol Chem 2003; 278:43807-17. Erratum in: J Biol Chem 2003; 278:51974.
46. Mukherjee PK, Marcheselli VL, Serhan CN, Bazan NG. Neuroprotectin D1: a docosahexaenoic acid-derived docosatriene protects human retinal pigment epithelial cells from oxidative stress. Proc Natl Acad Sci U S A 2004; 101:8491-6.
47. Vollrath D, Feng W, Duncan JL, Yasumura D, D'Cruz PM, Chappelow A, Matthes MT, Kay MA, LaVail MM. Correction of the retinal dystrophy phenotype of the RCS rat by viral gene transfer of Mertk. Proc Natl Acad Sci U S A 2001; 98:12584-9.
48. Andrews LD, Cohen AI. Freeze-fracture evidence for the presence of cholesterol in particle-free patches of basal disks and the plasma membrane of retinal rod outer segments of mice and frogs. J Cell Biol 1979; 81:215-28.
49. Andrews LD, Cohen AI. Freeze-fracture studies of photoreceptor membranes: new observations bearing upon the distribution of cholesterol. J Cell Biol 1983; 97:749-55.
50. Lin DS, Connor WE, Wolf DP, Neuringer M, Hachey DL. Unique lipids of primate spermatozoa: desmosterol and docosahexaenoic acid. J Lipid Res 1993; 34:491-9.
51. Connor WE, Lin DS, Wolf DP, Alexander M. Uneven distribution of desmosterol and docosahexaenoic acid in the heads and tails of monkey sperm. J Lipid Res 1998; 39:1404-11.
52. Boesze-Battaglia K, Albert AD. Cholesterol modulation of photoreceptor function in bovine retinal rod outer segments. J Biol Chem 1990; 265:20727-30.
53. Ruprecht JJ, Mielke T, Vogel R, Villa C, Schertler GF. Electron crystallography reveals the structure of metarhodopsin I. EMBO J 2004; 23:3609-3620.
54. Seno K, Kishimoto M, Abe M, Higuchi Y, Mieda M, Owada Y, Yoshiyama W, Liu H, Hayashi F. Light- and guanosine 5'-3-O-(thio)triphosphate-sensitive localization of a G protein and its effector on detergent-resistant membrane rafts in rod photoreceptor outer segments. J Biol Chem 2001; 276:20813-6.
55. Nair KS, Balasubramanian N, Slepak VZ. Signal-dependent translocation of transducin, RGS9-1-Gbeta5L complex, and arrestin to detergent-resistant membrane rafts in photoreceptors. Curr Biol 2002; 12:421-5.
56. Boesze-Battaglia K, Dispoto J, Kahoe MA. Association of a photoreceptor-specific tetraspanin protein, ROM-1, with triton X-100-resistant membrane rafts from rod outer segment disk membranes. J Biol Chem 2002; 277:41843-9.
57. Elliott MH, Fliesler SJ, Ghalayini AJ. Cholesterol-dependent association of caveolin-1 with the transducin alpha subunit in bovine photoreceptor rod outer segments: disruption by cyclodextrin and guanosine 5'-O-(3-thiotriphosphate). Biochemistry 2003; 42:7892-903.