Oxidative Stress, Hypoxia, and Ischemia-Like Conditions Increase the Release of Endogenous Amino Acids by Distinct Mechanisms in Cultured Retinal Cells

Abstract: The aim of this study was to elucidate the mechanisms by which retinal cells release endogenous amino acids in response to ascorbate/Fe²⁺-induced oxidative stress, as compared with chemical hypoxia or ischemia. In the absence of stimulation, oxidative stress increased the release of aspartate, glutamate, taurine, and GABA only when Ca²⁺ was present. Under hypoxia or ischemia, the release of aspartate, glutamate, glycine, alanine, taurine, and GABA increased mainly by a Ca²⁺-independent mechanism. The increased release observed in N -methyl- d -glucamine⁺ medium suggested the reversal of the Na⁺-dependent amino acid transporters. Upon oxidative stress, the release of aspartate, glutamate, and GABA, occurring through the reversal of the Na⁺-dependent transporters, was reduced by about 30%, although the release of taurine was enhanced. An increased release of [³H]arachidonic acid and free radicals seems to affect the Na⁺-dependent transporters for glutamate and GABA in oxidized cells. All cell treatments increased [Ca²⁺]_i (1.5 to twofold), although no differences were observed in membrane depolarization. The energy charge of cells submitted to hypoxia or oxidative stress was not changed. However, ischemia highly potentiated the reduction of the energy charge, as compared with hypoglycemia or hypoxia alone. The present work is important for understanding the mechanisms of amino acid release that occur in vivo upon oxidative stress, hypoxia, or ischemia, frequently associated with the impairment of energy metabolism.     

In Vitro Release of Endogenous Excitatory Sulfur-Containing Amino Acids from Various Rat Brain Regions

Efflux of various amino acids from rat brain slices was determined under resting or depolarizing conditions. Slices of neocortex, hippocampus, striatum, cerebellum, mesodiencephalon, pons-medulla, and spinal cord were depolarized by K+ (50 mM) or veratrine (33 micrograms/ml). The 4-N,N-dimethylamino-azobenzene-4'-isothiocyanate (DABITC) derivatization method of Chang [Biochem. J. 199, 537-545 (1981)] for HPLC was adapted for analysis of amino acids and peptides in superfusion solutions. It allowed the separation and simultaneous detection of the sulfur-containing amino acids cysteine sulfinic acid (CSA), cysteic acid (CA), homocysteine sulfinic acid (HCSA), and homocysteic acid (HCA) at the picomole level. All four were shown to be released on depolarization in a Ca2+-dependent manner from brain slices. CSA and HCSA were released from cortex, hippocampus, mesodiencephalon, and, for HCSA only, striatum. HCA release, observed in all regions, was most prominent in cortex and hippocampus. CA was slightly increased by depolarization in hippocampus and mesodiencephalon. These sulfur-containing amino acids have been shown to exert an excitatory action on CNS neurons. The fact that these sulfur-containing amino acids are released as endogenous substances from nervous tissue supports the hypothesis that they play a role in CNS neurotransmission.     

A Retinal Ganglion Cell Neurotrophic Factor Purified from the Superior Colliculus

Dissociated neonatal rat retinal ganglion cells can be maintained by the addition of an extract from the neonatal superior colliculus. This extract can support 95% of ganglion cells over 24 h in culture; in addition it promotes the expression of neurites from these cells. This report describes the purification of a neurotrophic factor from the superior colliculus which supports the survival of 80% of retinal ganglion cells over 24 h in vitro. The purification procedure involves a combination of dye-ligand, anion-exchange, and molecular sieve chromatography. The purified neurotrophic factor has a Stokes radius of approximately 200 A using molecular sieve chromatography in the presence of a chaotropic agent. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the purified factor indicates that it is a glycoprotein that migrates with a molecular mass greater than 400 kDa. Further characterization of this high-molecular-mass glycoprotein by enzymatic digestion demonstrated that it is a chondroitin sulfate proteoglycan. This factor is clearly distinguishable from other neurotrophic factors that have an effect on retinal ganglion cells such as brain-derived neurotrophic factor and fibroblast growth factor. The chondroitin sulfate proteoglycan from the neonatal superior colliculus is the first proteoglycan to be identified as a neurotrophic factor.     

A Bioluminescence Method for the Measurement of l-Glutamate: Applications to the Study of Changes in the Release of l-Glutamate from Lateral Geniculate Nucleus and Superior Colliculus After Visual Cortex Ablation in Rats

We have developed a rapid, simple, specific, and very sensitive bioluminescence method for the measurement of L-glutamate (L-Glu). Oxidation of L-Glu by glutamate dehydrogenase has been coupled with bacterial FMN reductase and luciferase. Light production (i.e., peak height or integral) was linear from less than 0.5 to 500 pmol of L-Glu. Potential interfering substances that may be encountered in brain tissue have been identified. The most potent inhibitors were ascorbate and the biogenic amines. Procedures that conferred long-term stability of the reagent mixture (greater than 8 h) were established. Bioluminescence analysis of L-Glu content in brain tissue extracts, fractions from release experiments, and human CSF corroborated respective results obtained by HPLC analysis. In this study, we have applied the method to monitor changes in the KCl-evoked release of endogenous L-Glu from milligram amounts of brain tissue, i.e., from lateral geniculate nucleus and superior colliculus after visual cortex ablation.     

Release of Endogenous Amino Acids from the Hippocampus and Brain Stem from Developing and Adult Mice in Ischemia

The release of neurotransmitters and modulators has been studied mostly using labeled preloaded compounds. For several reasons, however, the estimated release may not reliably reflect the release of endogenous compounds. The basal and K⁺-evoked release of the neuroactive endogenous amino acids GABA, glycine, taurine, l-glutamate and l-aspartate was now studied in slices from the hippocampus and brain stem from 7-day-old and 3-month-old mice under control and ischemic conditions. The release of synaptically not active l-glutamine, l-alanine, l-threonine and l-serine was assessed for comparison. The estimates for the hippocampus and brainstem were markedly different and also different in developing and adult mice. GABA release was much greater in 3-month-old than in 7-day-old mice, whereas with taurine the situation was the opposite, in the hippocampus in particular. K⁺ stimulation enhanced glycine release more in the mature than immature brain stem while in the hippocampus the converse was observed. Ischemia enhanced the release of all neuroactive amino acids in both brain regions, the effects being relatively most pronounced in the case of GABA, aspartate and glutamate in the hippocampus in 3-month-old mice, and taurine in 7-day-old and glycine in 3-month-old mice in the brain stem. These results are qualitatively similar to those obtained on earlier experiments with labeled preloaded amino acids. However, the magnitudes of the release cannot be quite correctly estimated using radioactive labels. In developing mice only taurine release may counteract the harmful effects of excitatory amino acids in ischemia in both hippocampus and brain stem.     

Release of Endogenous and Accumulated Exogenous Amino Acids from Slices of Normal and Climbing Fibre-Deprived Rat Cerebellar Slices

Efflux of various amino acids from slices of rat cerebellar hemispheres was determined under resting or depolarizing conditions. It was increased under high K+(50 mM) as compared to low K+ (5 mM) conditions by 1258 pmol/mg protein for aspartate, 478 for gamma-aminobutyric acid (GABA), 44,693 for glutamate, and 615 for glycine. These were significantly higher than the corresponding values obtained under low-Ca2+ (0.1 mM), high-Mg2+ (12 mM) conditions, whereas for 11 other amino acids the K+-induced efflux was similar under normal and low-Ca2+ concentrations. The K+-induced efflux of exogenously accumulated L-[3H]aspartate, D-[3H]aspartate, and L-[3H]glutamate was higher by factors of 2, 5.8, and 6.3, respectively, under normal Ca2+ conditions, as compared with low-Ca2+, high-Mg2+ conditions. After climbing fibre degeneration induced by destruction of the inferior olive with 3-acetylpyridine, release of endogenous aspartate and exogenous L-[3H]glutamate and D-[3H]aspartate was significantly reduced, by 26%, 38%, and 27%, respectively. These results support the hypothesis that climbing fibres may use aspartate or a related compound as a neurotransmitter. In rat cerebellar tissue, L-[3H]glutamate and L-[3H]aspartate differ in several aspects: (1) L-[3H]glutamate uptake was 4 times higher than that of L-[3H]aspartate; (2) fractional rate constant of K+- evoked release of L-[3H]aspartate was 7% X 2.5 min-1, and of L-[3H]glutamate 36% X 2.5 min-1; and (3) specific activity of L-[3H]glutamate in the eluate collected during K+ stimulation was 3.5 times the value in the tissue, whereas for L-[3H]aspartate, specific activities in the eluate and tissue were similar.     

Release of Excitatory Amino Acids from Cultured Hippocampal Astrocytes Induced by a Hypoxic-Hypoglycemic Stimulation

An excess release of excitatory amino acids (EAA) is an important factor for postischemic brain damage. In the present communication, we demonstrate that cultured hippocampal cells release EAA after hypoxic-hypoglycemic treatment. The amounts of EAA released from astrocytes were appreciably above those released from neurons. Furthermore, the amount of aspartate released from astrocytes was comparable to that of glutamate, although the endogenous content of aspartate was one-fifth that of glutamate. The endogenous content of aspartate in astrocytes increased even after hypoxic-hypoglycemic treatment. These results suggests that ischemic neuronal death is due, at least in part, to the excitotoxicity of aspartate and glutamate derived from surrounding astrocytes.     

Content and In Vitro Release of Endogenous Amino Acids in the Area of the Nucleus of the Solitary Tract of the Rat

We sought to identify amino acid neurotransmitter candidates within the nucleus of the solitary tract in rats. Twenty endogenous amino acids were quantified by reverse-phase HPLC with fluorescence detection (30-fmol limit). Micropunches (1 mm) of the intermediate area of the solitary nucleus were prepared, and the amino acid content determined. Of all the components measured, the putative transmitters Glu, Gly, gamma-aminobutyric acid, taurine, Asp, and Ala appeared in greatest concentrations. Bilateral micropunches superfused in vitro with buffered medium containing 56 mM potassium released Glu, gamma-aminobutyric acid, and Gly in a significant manner (p less than 0.05) compared with basal levels. With Glu, 78% was calcium-dependent and, therefore, presumably from nerve endings; 99% of gamma-aminobutyric acid and 42% of Gly were dependent on calcium. After removal of the nodose ganglion, a bilateral decrease in the calcium-dependent release of Glu and gamma-aminobutyric acid, but not Gly, was observed; decreases were significant ipsilateral to the site of ablation. We conclude that (a) Glu is a transmitter of primary afferents in the nucleus of the solitary tract; (b) glutamatergic afferents may interact with gamma-aminobutyric acid system(s) in this region; (c) Gly also may participate in the mediation and/or modulation of cardiovascular or other visceral reflexes; and (d) amino acid neurotransmission may play an integral role in the neurogenic control of arterial pressure.     

Endogenous Amino Acid Release from Cultured Cerebellar Neuronal Cells: Effect of Tetanus Toxin on Glutamate Release

Endogenous amino acid release was measured in developing cerebellar neuronal cells in primary culture. In the presence of 25 mM K+ added to the culture medium, cerebellar cells survived more than 3 weeks and showed a high level of differentiation. These cultures are highly enriched in neurons, and electron-microscopic observation of these cells after 12 days in vitro (DIV) confirmed the presence of a very large proportion of cells with the morphological characteristics of granule cells, making synapses containing many synaptic vesicles. Synaptogenesis was also confirmed by immunostaining the cells with antisera against synapsin I and synaptophysin, two proteins associated with synaptic vesicles. From these cultures, endogenous glutamate release stimulated by 56 mM K+ was already detected after only a few days in culture, the maximal release value (1,579% increase over basal release) being reached after 10 DIV. In addition to that of glutamate, the release of aspartate, asparagine, alanine, and, particularly, gamma-aminobutyric acid (GABA) was stimulated by 56 mM K+ after 14 DIV, but to a lesser extent. No increase in serine, glutamine, taurine, or tyrosine release was observed during K+ depolarization. The effect of K+ on amino acid release was strictly Ca2+-dependent. Stimulation of the cells with veratridine resulted in a qualitatively similar effect on endogenous amino acid release. In the absence of Ca2+, 30% of the veratridine effect persisted. The Ca2+-dependent release was quantitatively similar after stimulation by veratridine and K+. Treatment of cerebellar cells with tetanus toxin (5 micrograms/ml) for 24 h resulted in a total inhibition of the Ca2+-dependent component of the glutamate release evoked by K+ or veratridine. It is concluded that glutamate is the main amino acid neurotransmitter of cerebellar cells developed in primary culture under the present conditions and that glutamate is probably mainly released through the exocytosis of synaptic vesicles.     

Effect of Amygdaloid Kindling on the Content and Release of Amino Acids from the Amygdaloid Complex: In Vivo and In Vitro Studies

Abstract: The tissue content and the interstitial fluid levels of glutamate, aspartate, GABA, glutamine, glycine, and serine were studied in amygdaloid-kindled rat brain. Interstitial levels were studied in vivo before and during stage 5 full limbic seizures using microdialysis. Slices of amygdala from kindled and sham-operated animals were used to study baseline and KCl-evoked release in vitro. The contents of these amino acids were measured in slices of amygdala, hippocampus, and cerebral cortex from kindled and sham-operated animals. Kindled brains showed two- to threefold higher levels of glutamate, aspartate, and GABA and 12-fold higher levels of glutamine than sham-operated controls. Correlating with this, interstitial fluid levels of glutamate were two- to threefold higher from kindled amygdala than from control both in vivo (microdialysis) and in vitro (superfusion). GABA levels in interstitial fluid from kindled amygdala were reduced by 67% compared with control amygdala.     

Desmethylimipramine Enhances the Release of Endogenous GABA and Other Neuro trans mitter Amino Acids from the Rat Thalamus

The influence of desmethylimipramine (DMI) on the release of endogenous gamma-aminobutyric acid (GABA) and some other amino acids from the rat thalamus was studied with a push-pull perfusion technique. Following HPLC the amino acids were fluorimetrically estimated. Added to the perfusion medium at a concentration of 10 mumol L-1, DMI caused a 5- to 10-fold increase in the release of GABA. Similar effects were found with imipramine, trimeprimine, haloperidol, and propranolol. The elevation of GABA release induced by DMI was Ca dependent. The release of aspartate and glutamate was also increased by DMI, but in contrast to K ions, DMI did not reduce the thalamic output of glutamine.     

Comparison of the Effects of a Convulsant Barbiturate on the Release of Endogenous and Radiolabeled Amino Acids from Slices of Mouse Hippocampus

The convulsant barbiturate 5-(2-cyclohexylidene-ethyl)-5-ethyl barbituric acid (CHEB) stimulates the spontaneous release of endogenous and radiolabeled acetylcholine (ACh) from mouse hippocampal slices in vitro. In order to determine if the ability of CHEB to release ACh was unique to this neurotransmitter, we have studied the action of this drug in vitro on the release of both radiolabeled and endogenous putative neurotransmitter and non-transmitter amino acids in the hippocampus. Although CHEB stimulated the spontaneous release of both [3H]gamma-n-aminobutyric acid (GABA) and endogenous GABA, CHEB had different effects on the spontaneous release of radiolabeled and endogenous L-glutamate and L-aspartate: L-[3H]glutamate release was inhibited by CHEB, but endogenous L-glutamate release was unaffected by CHEB, but endogenous L-aspartate release was stimulated. The spontaneous release of the amino acids L-alanine and glycine (not thought to be neurotransmitters in the hippocampus) was not affected by CHEB. The results of this study indicate that CHEB does not always stimulate the release of all putative neurotransmitters. The ability of this drug to release ACh, GABA, and L-aspartate may be the result of some specific interaction of CHEB with nerves using these neurotransmitters in the hippocampus. In addition, the results suggest some problems that may be encountered when radiolabeled substances are used to study neurotransmitter release.     

Effect of β-Bungarotoxin on the Release of Endogenous Amino Acids from the Sensorimotor Cortex

Abstract: β-Bungarotoxin, a snake neurotoxin purified from the venom of Bungarus multicinctus , caused a significant increase in the in vivo release of glutamate from the superfused sensorimotor cortex of awake animals. A smaller effect on GABA release was observed, but no change was detected in the release of six other amino acids measured. The effects on glutamate and GABA release were entirely blocked by tetrodotoxin (1 μM) and were reversible when the cortical tissue was washed with saline.     

Role of Dendritic Dopamine of the Substantia Nigra in the Modulation of Nigrocollicular γ-Aminobutyric Acid Release: In Vivo Studies in the Rat

The release of gamma-[3H]aminobutyric acid ([3H]GABA) newly synthesized from [3H]glutamine was estimated in the superior colliculus of ketamine-anesthetized rats superfused via a push-pull cannula. A significant amount of [3H]GABA was spontaneously released in the superior colliculus (582 +/- 49 pCi/10 min). A major part of the large K(+)-evoked increase of the [3H]GABA release was Ca2+ dependent. When neuronal activity of the substantia nigra was enhanced by nigral application of K+ (30 mM) or bicuculline (10(-4) M), a persistent increase of the collicular [3H]GABA release was observed (60 and 80%, respectively). Conversely, when nigral activity was reduced by nigral application of GABA (10(-4) M) or superfusion with a Ca(2+)-free medium, a sustained decrease of the collicular [3H]GABA release was observed (-30 and -40%, respectively). Following the nigral application of a selective D2-receptor agonist. RU 24926 (10(-6) M), for 30 min in 6-hydroxydopamine-lesioned rats, a phasic increase (60%) of the collicular [3H]GABA release was detected. This effect could result from an activation of nigrocollicular GABAergic neurons by D2-receptor stimulation, because nigral activity and collicular release of [3H]GABA changed in a parallel direction.     

Sulphur-Containing Amino Acids Modulate Noradrenaline Release from Hippocampal Slices

Abstract: The l - and d -enantiomers of the sulphur-containing amino acids (SAAs)—homocysteate, homocysteine sulphinate, cysteate, cysteine sulphinate, and S-sulphocysteine—stimulated [³H]noradrenaline release from rat hippocampal slices in a concentration-dependent manner. The relative potencies of the l -isomers (EC₅₀ values of 1.05–1.96 mM) were of similar order to that of glutamate (1.56 mM), which was 10-fold lower than that of NMDA (0.15 mM), whereas the d -isomers exhibited a wider range of potencies (0.75 to >5 mM). All stimulatory effects of the SAAs were significantly inhibited by the voltage-sensitive Na⁺ channel blocker tetrodotoxin (55–71%) and completely blocked by addition of Mg²⁺ or Co²⁺ to the incubation medium. All SAA-evoked responses were concentration-dependently antagonized by the selective NMDA receptor antagonist d -(?)-2-amino-5-phosphonopentanoic acid (IC₅₀ values of 3.2–49.5 µM). 6-Cyano-7-nitroquinoxaline-2,3-dione (CNQX), a non-NMDA receptor antagonist, at 100 µM inhibited the [³H]noradrenaline release induced by glutamate and NMDA (65 and 76%, respectively) and by all SAAs studied (65–85%), whereas 10 µM CNQX only inhibited the effects of S-sulpho-l -cysteine and l - and d -homocysteate (33, 32, and 44%, respectively). However, the more selective AMPA/kainic acid receptor antagonist 6-nitro-7-sulphamoylbenzo(f)quinoxaline-2,3-dione (100 µM), which did not antagonize the [³H]noradrenaline release induced by glutamate and NMDA, reduced only the S-sulpho-l -cysteine-evoked response (25%). Thus, the stimulation of Ca²⁺-dependent[³H]noradrenaline release from hippocampal slices elicited by the majority of the SAAs appears to be mediated by the NMDA receptor.     

In Vitro Release of Endogenous Dopamine from the Striatum of the Weaver Mutant Mouse

The weaver mutant mouse has a genetically determined defect in the nigrostriatal dopaminergic system. The present study was undertaken to test the hypothesis that in the weaver mutant mouse, striatal nerve terminals undergo compensatory changes in response to this deficiency. To test this hypothesis, we studied the basal and stimulated release of dopamine from striatal slices of weaver mutant mice and matched controls. By using a superfusion system and concentrating the superfusate by passage over alumina, resting dopamine release could be determined in the weaver mutant despite the fact that striatal tissue content of dopamine in these mice is reduced by greater than 75% compared with control mice. Fractional resting release of dopamine in weaver striatal slices was significantly elevated compared with that in controls, suggesting that the release mechanisms in the weaver may be adapting to overcome the dopamine deficit. Potassium-evoked release (24 and 48 mM potassium) was not significantly different between the two genotypes. In contrast, amphetamine-evoked release (1 microM) was significantly greater in the weaver mice than in controls. In both genotypes, release evoked by amphetamine was completely inhibited by cocaine, implicating the dopamine uptake carrier in this release process. These findings suggest that fundamental differences in dopamine release mechanisms exist between weaver and control mice and support the hypothesis that compensatory mechanisms may develop in neurons in response to dopamine deficits.     

Release by Electrical Stimulation of Endogenous Glutamate, γ-Aminobutyric Acid, and Other Amino Acids from Slices of the Rat Medulla Oblongata

Evidence was obtained for the release of amino acids by electrical stimulation of slices of regions of the rat medulla oblongata: rostral ventrolateral, caudal ventrolateral and caudal dorsomedial. There was a Ca2+-dependent, tetrodotoxin-sensitive increase in the efflux of aspartate, glutamate, gamma-aminobutyric acid (GABA), glycine, and beta-alanine in all regions examined. There were distinct regional differences in the relative amounts of amino acids released. These results provide evidence for the possible neurotransmitter role of aspartate, glutamate, GABA, glycine, and beta-alanine in these regions of the rat medulla oblongata.     

Effects of Ethanol In Vitro and In Vivo on the Release of Endogenous Catecholamines from Specific Regions of Rat Brain

Blocks of tissue from the hypothalamus, olfactory bulb, or striatum of rats were incubated in vitro to study the basal and potassium-stimulated release of endogenous catecholamines. When ethanol (100-250 mM) was added to these preparations in vitro no changes in release were observed. When ethanol (3.0 g X kg-1) was injected intraperitoneally in vivo, however, and 3,4-dihydroxyphenylethylamine (DA, dopamine) release was measured in vitro at various times after drug administration, significant increases in the basal release and decreases in the potassium-stimulated release were observed in striatum and olfactory bulb. In striatum, these changes showed a more rapid onset and a longer duration than in olfactory bulb. In both brain regions, DA release did not differ from controls at 4-6 h after the ethanol injection, although blood ethanol concentrations remained elevated. This may imply the tissue's acquisition of acute functional tolerance to the drug. Similar increases and decreases in the basal and the potassium-induced release of DA from striatal tissues were also found at 1 h after injection of a lower dose of ethanol (1.0 g X kg-1). In terms of behavior, this lower dose of ethanol produced only mild intoxication and ataxia, in contrast to the loss of righting reflex following the higher dose.     

Release of Endogenous Amino Acids from Striatal Neurons in Primary Culture

Following partial purification, the characteristics of a cytosol protein kinase were investigated. The protein kinase was purified by ammonium sulfate precipitation and diethylaminoethyl-cellulose, ATP-agarose, and hydroxyapatite chromatography. Analysis of the purified protein kinase preparation by polyacrylamide gel electrophoresis revealed three major protein bands. The cytosol protein kinase was purified approximately 442-fold, as calculated from the cyclic nucleotide independent protein kinase activity in the 40,000 g supernatant. The activity of the kinase was found to be independent of either cyclic AMP or cyclic GMP. Moreover, the kinase activity was unaffected by the addition of the endogenous protein kinase inhibitor, or the regulatory subunit from the type II cyclic AMP-dependent protein kinase from bovine heart. The molecular weight of the enzyme was determined to be 95,000 by Sephadex G-200 gel filtration. The activity of the kinase was increased approximately twofold in the presence of 10 microM Ca+2 and calmodulin. This increase was reversed by the addition of EGTA. The subcellular distribution of the protein kinase was also examined. The soluble fraction from nerve terminal was found to have the highest concentration of the kinase activity.     

Apoptosis of Retinal Photoreceptors During Development In Vitro: Protective Effect of Docosahexaenoic Acid

Abstract: When rat retinal cells are cultured in a serum-free medium, the photoreceptor cells start dying after 7 days. The addition of docosahexaenoic acid (DHA) to the cultures prevents the selective death of photoreceptors. Here it is shown that, unlike other retinal neurons, photoreceptors die through an apoptotic pathway. Hallmarks of apoptosis, such as nuclear fragmentation and condensation and DNA cleavage forming a ladder pattern on an agarose gel, were observed. The timing and high selectivity of the triggering of photoreceptor cell apoptosis suggest the existence of a programmed cell death. Compared with other fatty acids, DHA not only was the most effective in promoting photoreceptor survival, but also the only one to decrease the number of apoptotic nuclei. The results suggest that DHA is important among the factors preventing apoptosis of photoreceptors in the developing retina. A limitation in the availability of this fatty acid might trigger apoptosis as a result of the failure to develop functional photoreceptor outer segments.