Protective Effect of Docosahexaenoic Acid on Oxidative Stress-Induced Apoptosis of Retina Photoreceptors

Abstract

purpose. In a recent study, it was demonstrated that docosahexaenoic acid (DHA) promotes the survival of retinal photoreceptors in vitro, delaying apoptosis. However, lipid enrichment in DHA is known to contribute to retina vulnerability to oxidative stress. In this study, the effect of oxidative damage on rat retina neurons in vitro and whether DHA enhances or diminishes this damage were investigated.

methods. Rat retina neurons in 3-day cultures, with or without DHA, were treated with the oxidant paraquat. After 24 hours, apoptosis, mitochondrial membrane integrity, and Bcl-2 and Bax expression were immunocytochemically determined.

results. Paraquat induced apoptosis in amacrine and photoreceptor neurons, major neuronal types in the culture. Neuronal apoptosis was accompanied by mitochondrial membrane depolarization, an increase in the amount of photoreceptors expressing Bax, and a decrease in those expressing Bcl-2. Addition of DHA reduced photoreceptor apoptosis by almost half, simultaneously preserving their mitochondrial membrane integrity. DHA blocked the paraquat-induced increase in Bax expression and remarkably upregulated Bcl-2 expression. Glia-derived neurotrophic factor, a photoreceptor trophic factor, only slightly increased Bcl-2 expression and did not protect photoreceptors from oxidative damage. Similarly, other fatty acids tested did not prevent photoreceptor apoptosis.

conclusions. These results show that oxidative damage induces apoptosis in retinal neurons during their early development in culture and suggest that the loss of mitochondrial membrane integrity is crucial in the apoptotic death of these cells. DHA activates intracellular mechanisms that prevent this loss and by modulating the levels of pro- and antiapoptotic proteins of the Bcl-2 family selectively protect photoreceptors from oxidative stress.

Oxidative stress is thought to be involved in the pathogenesis of many ocular diseases. The retina is extremely prone to oxidative damage due to its relatively high oxygen consumption and its constant exposure to light. Photooxidative mechanisms play a significant role in photoreceptor death due to excessive exposure to light, ¹ which results in apoptosis of these cells in rat retina. ^{2 3 4} Protection from photo-oxidative damage has been extensively used to identify several photoreceptor trophic factors. ⁵ Further evidence points to the involvement of oxidative stress in photoreceptor apoptosis: antioxidants ameliorate light-induced retinal degeneration, and an early and sustained increase in intracellular reactive oxygen species accompanies photoreceptor apoptosis in vitro, ⁶ whereas prolonged oxygen exposure leads to an altered rod photoreceptor layer in infant rats. ⁷  

Retina photoreceptors are highly enriched in docosahexaenoic acid (DHA), ⁸ which amounts to approximately 50% of the fatty acids in photoreceptor rod outer segments. Vital retinal functions depend on the existence of an acceptable proportion of DHA in retinal lipids. ⁹ This fatty acid is required for satisfactory development of vision ¹⁰ and its deficiency causes loss of visual acuity. ¹¹ Adequate levels are essential for proper rhodopsin functionality ¹² and visual transduction. ¹³ In addition, we have recently shown that DHA prevents the apoptosis of photoreceptor cells that otherwise inevitably occurs during their early development in vitro. ^{14 15 16} DHA also enhances photoreceptor differentiation, promoting opsin expression and the formation of apical processes, and simultaneously leads to accumulation of this acid in neuronal lipids. ¹⁶  

At the same time, the multiplicity of double bonds in DHA renders it extremely sensitive to free radical damage during oxidative stress. ¹⁷ Its abundance in retina photoreceptors is believed to contribute to making these cells main targets for oxidative damage. However, the relationship between DHA levels and oxidative stress remains controversial. ¹⁸ A DHA-enriched diet increases peroxidation products in plasma and several tissues ¹⁹ and augments oligodendroglial cell vulnerability to oxidative stress, ²⁰ whereas rats fed a diet deficient in DHA precursors show less retinal damage after exposure to light. ²¹ In contrast, DHA-supplemented human lymphocytes are less vulnerable to hydrogen-peroxide–induced oxidative damage. ²²  

In this work, we investigated the effect of DHA supplementation in retina neurons subjected to oxidative stress. Our purpose was to determine whether enrichment of neuronal lipids in this polyunsaturated fatty acid increased neuronal susceptibility to oxidative damage or whether DHA’s biological effect as a trophic factor prevailed, hence protecting photoreceptors from oxidative stress. To study the effect of oxidative stress on retinal neurons, we used an in vitro model, adding the oxidant paraquat to pure rat retina neurons in culture. Paraquat, an anion superoxide generator, induces lipid peroxidation ²³ and causes neurodegenerative damage of peripheral or central nervous system. It also produces apoptosis in PC12 cells ²⁴ and is cytotoxic to rat cortical neurons in culture, impairing cellular energy production. ²⁵ In this work, treatment with paraquat triggered the apoptotic death of amacrine and photoreceptor neurons, simultaneously leading to mitochondrial membrane depolarization. Addition of DHA selectively reduced photoreceptor apoptosis after oxidative damage, preserving mitochondrial membrane integrity in these cells by a mechanism in which upregulation of Bcl-2 expression seemed to be involved. 

Materials and Methods

Albino Wistar rats bred in our own colony were used in all the experiments. All proceedings concerning animal use were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Plastic culture 35-mm diameter dishes (Nunc) were purchased from InterMed (Naperville, IL). Dulbecco’s modified Eagle’s medium (DMEM) was from Gibco-Life Technologies. Bovine serum albumin (Fraction V; fatty acid-free; low endotoxin, tissue culture tested); paraquat dichloride (methyl viologen, 1,1′-dimethyl-4,4′-bipyridinium dichloride); poly-dl-ornithine; trypsin; trypsin inhibitor; transferrin; hydrocortisone; putrescine; insulin; selenium; gentamicin; 4,6-diamidino-2-phenylindole (DAPI); monoclonal anti-syntaxin clone HPC-1; fluorescein-conjugated secondary antibodies; palmitic, oleic, arachidonic, and docosahexaenoic acids; and paraformaldehyde were from Sigma Chemical Co. (St. Louis, MO). A tyramide signal amplification kit was purchased from NEN (Boston, MA); glia-derived neurotrophic factor (GDNF) from Peprotech, Inc. (Rocky Hill, NJ); secondary antibodies, green fluorescence (Alexa 488)–conjugated-goat anti-mouse and a fluorescent probe (MitoTracker Red CMXRos) from Molecular Probes, Inc.; and monoclonal antibodies for Bax (sc-7480) and Bcl-2 (sc-7382) from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The monoclonal antibody against rhodopsin, Rho4D2, was generously supplied by Robert Molday (University of British Columbia, Vancouver, British Columbia, Canada). Solvents were HPLC grade, and all other reagents were analytical grade. 

Neuronal Cultures

Purified cultures of rat retinal neurons were prepared as previously described. ^{14 15 26} Approximately 0.5 × 10⁵ cells/cm² were seeded on 35-mm diameter dishes, which had previously been sequentially treated with polyornithine and schwannoma-conditioned medium, to allow cell attachment to substrata and to promote neurite outgrowth, respectively. ²⁷  

Addition of Fatty Acids and GDNF

Docosahexaenoic, palmitic (16:0), oleic (18:1), and arachidonic (20:4 n-6) acids, at 6.7 μM final concentration, were added at day 1 in culture in a complex with bovine serum albumin (BSA) in a 2:1 (fatty acid: BSA) molar ratio. ¹⁴ The same volume and concentration of BSA solution was simultaneously added to control cultures. GDNF was added in DMEM at a final concentration of 4 ng/mL immediately after cells were seeded. 

Addition of Paraquat

Paraquat (2.4–240 μM, in a calcium-magnesium–free solution) was added to 3-day cultures, which were then incubated for 24 hours before fixation. 

Immunocytochemical Methods

Cultures were fixed for at least 1 hour with 2% paraformaldehyde in phosphate-buffered saline (PBS; 0.9% NaCl in 0.01 M NaH₂PO₄ [pH 7.4]), followed by permeation with Triton X-100 (0.1%) for 15 minutes. Neuronal cell types were identified by immunocytochemistry with the monoclonal anti Rho4D2, for photoreceptor cells and anti-syntaxin clone HPC-1, for amacrine cells, ^{28 29} as previously described. ¹⁴ Green fluorescence (Alexa 488; Molecular Probes)–conjugated goat anti-mouse was used as the secondary antibody. Tyramide signal amplification was occasionally used to improve visualization, according to the procedure described by the manufacturer (NEN). Controls for immunocytochemistry were performed by omitting either the primary or the secondary antibody. Apoptosis was determined by evaluating nuclei integrity after incubating the cells with DAPI for 20 minutes. The amount of photoreceptors or amacrine cells showing apoptotic nuclei was determined in cultures double-labeled with DAPI and with Rho4D2 or HPC-1, to unambiguously identify cells as photoreceptors or amacrine neurons, respectively. The total number of apoptotic photoreceptors was then calculated, taking into account the percentage of Rho4D2-labeled cells at that time in culture. To assess mitochondrial membrane integrity, cultures were incubated for 30 minutes before fixation with the fluorescent probe (0.1 μg/mL; MitoTracker; Molecular Probes), which is concentrated in active mitochondria, labeling them with a bright fluorescence. The amount of photoreceptors displaying fluorescent mitochondria was determined in relation to the total number of photoreceptors. The number of photoreceptors expressing either Bax or Bcl-2 in relation to the total number of photoreceptors was quantitated by immunocytochemistry, using specific monoclonal antibodies. 

Statistical Analysis

For cytochemical studies, 10 fields per sample, randomly chosen, were analyzed in each case. Data represent the average of two or three experiments, with three to four dishes for each condition ± SD. Statistical significance was determined by Student’s two-tailed t-test. 

Results

Effect of Paraquat on Neuronal Cell Apoptosis

Two major cell types were found in retina neuronal cultures after 4 days in vitro: photoreceptors and amacrine neurons (Figs. 1 2 ^{14 26} ). Photoreceptors have a small, round cell body (3–5 μm) and show a single neurite at one end, usually ending in a conspicuous synaptic “spherule,” and sometimes a connecting cilium at the opposite end. Opsin is diffusely distributed over the cell body (Figs. 1A 1C) , which is usually darker than that of amacrine cells. Amacrine neurons are larger than photoreceptors (7–20 μm) and have two or more neurites (Figs. 2A 2C 2E) . 

A 24-hour treatment with 24 μM paraquat prompted dramatic changes in neuronal morphology. Many cells tended to lose their characteristic shape, acquiring a round appearance and showing few or no neurites (Figs. 2E 2F) . The characteristic nuclear fragmentation of apoptotic cells was clearly observed in photoreceptors (Figs. 1G 1H) and amacrine cells (Fig. 2D) . In addition, condensed nuclei, nuclear fragments, and, in some cases, enucleated cells were also found (Figs. 1H 1K 2D) . Oxidative damage markedly affected both neuronal types. Apoptotic photoreceptors increased from almost 44,000 in the control to more than 122,000 cells per dish after addition of paraquat, whereas apoptotic amacrine neurons increased from approximately 7,200 in control conditions to 61,000 cells per dish in paraquat-treated cultures (Fig. 3) . 

When paraquat concentrations ranging from 2.4 to 240 μM were added to the cultures, the smallest concentration tested was sufficient to induce apoptosis in a large number of both amacrine and photoreceptor cells (Fig. 4) . Higher paraquat concentrations increased the amount of apoptotic neurons; however, a 100-fold higher oxidant concentration did not double this amount. At this high (240 μM) concentration of paraquat, necrotic neurons were also observed (not shown). 

DHA Protection of Photoreceptors from Oxidative Stress-Induced Apoptosis

Confirming previous findings in our laboratory, ¹⁵ in 4-day cultures without photoreceptor trophic factors, approximately 20% of photoreceptors became apoptotic (Fig. 5A) , whereas DHA supplementation reduced this percentage by half. We investigated whether addition of this polyunsaturated fatty acid enhances cell vulnerability or protects photoreceptors from oxidative damage and found that DHA supplementation prevented oxidative stress-induced apoptosis in photoreceptors. Whereas more than 65% of photoreceptors were apoptotic after treatment with 24 μM paraquat in cultures without DHA, this percentage decreased to less than 40% in DHA-supplemented cultures. DHA still prevented photoreceptor apoptosis in cultures treated with 240 μM paraquat (Fig. 5A) . This protective effect was highly selective for photoreceptors, because paraquat-induced apoptosis of amacrine cells was not diminished by addition of DHA (Fig. 5B) . 

The effect of other fatty acids on photoreceptor apoptosis triggered by paraquat was then determined. In cultures that had been supplemented at day 1 with either palmitic, oleic, or arachidonic acid, photoreceptor apoptosis after treatment with paraquat was the same as in fatty-acid–free cultures (Table 1) , suggesting that only DHA could effectively prevent oxidative stress-induced photoreceptor apoptosis. 

As does DHA, GDNF promotes survival of photoreceptors during their early development in vitro and delays apoptosis both in vitro and in the rd mouse retina. ^{30 31 32} Hence, we evaluated whether GDNF was equally effective in preventing oxidative stress-induced apoptosis. After treatment with paraquat, the percentage of apoptotic photoreceptors was the same in GDNF-supplemented and unsupplemented cultures (Table 1) , suggesting that at least at this time of development, this molecule granted no protection from oxidative damage. 

Protective Effect of DHA on Mitochondrial Membrane Depolarization Induced by Paraquat in Photoreceptors

Given the key role played by mitochondria in the onset of apoptosis, ³³ we investigated their involvement in paraquat-induced apoptotic death. In control, 4-day cultures (Figs. 6A 6B) , most cells showed brightly fluorescent mitochondria. In contrast, cells preserving labeled mitochondria were noticeably diminished in paraquat-treated cultures (Figs. 6C 6D) . In control cultures, approximately 70% and 90% of photoreceptors and amacrine cells, respectively, maintained mitochondrial membrane integrity (Fig. 7) . In contrast, only 34% of photoreceptors and 45% of amacrine cells retained mitochondrial membrane integrity after addition of paraquat (Fig. 7) , suggesting a close correlation between this loss and development of apoptosis. 

Addition of DHA prevented mitochondrial membrane depolarization in photoreceptors (Figs. 6E 6F) . When no paraquat was added, the percentage of photoreceptors maintaining mitochondrial membrane integrity was similar in cultures with or without DHA (Fig. 7A) . After treatment with paraquat, the percentage of photoreceptors retaining mitochondrial membrane integrity in DHA-supplemented cultures was nearly double that found in cultures without DHA (Fig. 7A) , amounting to 57% and 34% of total photoreceptors, respectively. In contrast, the paraquat-induced decrease in the amount of amacrine cells maintaining mitochondrial membrane integrity was similar in DHA-supplemented and unsupplemented cultures (Fig. 7B) , suggesting that DHA protection of mitochondrial membrane integrity is selective for photoreceptors. 

Effect of Paraquat and DHA on the Expression of Bax and Bcl-2

The relative levels of Bax and Bcl-2, a pro- and an antiapoptotic protein of the Bcl-2 family, respectively, are crucial in regulating mitochondrial membrane permeability. ^{33 34} To investigate whether these proteins had a role in paraquat-induced apoptosis and in DHA’s protective effect, their expression was determined in the different culture conditions. Only 2% of total photoreceptors expressed Bax in control and DHA-supplemented cultures (Fig. 8A) . Treatment with paraquat significantly increased this expression in cultures without DHA, where approximately 12% of photoreceptors expressed Bax. DHA supplementation completely inhibited this increase, with the levels of Bax expressing-photoreceptors remaining the same as in control conditions. 

DHA strikingly increased the amount of photoreceptors expressing Bcl-2 (Figs. 9A 9B) , from approximately 35% of photoreceptors in control cultures to more than 60% in DHA-supplemented cultures (Fig. 8B) . In cultures without DHA, paraquat induced a slight decrease in expression of Bcl-2. On the contrary, the DHA-induced increase in Bcl-2 expression was not significantly diminished by paraquat. Approximately 55% of photoreceptors still expressed Bcl-2 in DHA-supplemented, paraquat-treated cultures (Fig. 8B) . In every culture condition, only a few (approximately 5%) Bcl-2 expressing photoreceptors nevertheless became apoptotic (Fig. 9) . As a whole, the described results and the corresponding Bcl-2/Bax ratio (Fig. 8C) closely correlated with the extent of photoreceptor apoptosis. When compared with the control, this ratio dramatically decreased after treatment with paraquat. In contrast, the Bcl-2/Bax ratio doubled in DHA-supplemented cultures and was not significantly decreased after treatment with paraquat of these cultures. 

When the effect on Bcl-2 expression of adding GDNF was determined at day 4, the percentage of photoreceptors expressing this protein was 37.3% ± 2.1% and 43.9% ± 2.3% (n = 3, P < 0.01), in control and GDNF-supplemented cultures, respectively. Hence, GDNF promoted only a very slight increase in the amount of photoreceptors expressing Bcl-2, distinctly contrasting with the remarkable upregulation of Bcl-2 expression triggered by DHA in these cells. 

Discussion

The present results show that the experimental model used is an efficient tool for investigating neuronal death triggered by oxidative damage and the protective effects of trophic factors. Paraquat induced the apoptotic death of amacrine and photoreceptor neurons, and this increase in apoptosis closely correlated with the loss of mitochondrial membrane permeability in these neurons. A notable finding was that supplementation of the cultures with DHA effectively prevented photoreceptor apoptosis caused by oxidative damage. 

Although both neuronal types were highly vulnerable to oxidative stress, photoreceptors seemed to be the more sensitive, because the smallest paraquat concentration used was enough to induce apoptosis in approximately half the photoreceptors present in the culture. The small differences in the extent of apoptosis on large increases in the concentration of paraquat used could be attributable to the cyclic chain reactions characteristic of lipid peroxidation processes. ³⁵ However, a neuronal subpopulation resistant to oxidative-stress–induced apoptosis may also exist in these cultures, because even at a paraquat concentration high enough to induce a certain amount of necrosis, a small group of photoreceptors and a large subpopulation of amacrine neurons survived. 

We have previously shown that during early development in vitro, addition of DHA postpones the apoptosis of photoreceptor cells, improving survival and enhancing differentiation, ^{14 15 16} with a simultaneous enrichment of this acid in neuronal lipids. Similarly, DHA’s antiapoptotic effect on mouse neuroblastoma cells apparently requires that it accumulates in lipids. ³⁶ The effect of such an accumulation on cell vulnerability to oxidative stress remains unclear. ^{19 37 38} Addition of DHA and its increase in neuronal lipids augment lipid radical formation and often enhance susceptibility to oxidative stress. ^{39 40} However, our present results show that for photoreceptors, the protective effect of DHA prevailed, with its addition preventing oxidative-stress–induced apoptosis. Other fatty acids granted no protection to photoreceptors after oxidative damage, which was consistent with the previously reported inability of these acids to prevent photoreceptor apoptosis during early development in vitro. ¹⁵ In contrast, GDNF, which is an efficient survival factor for photoreceptors during development, ^{30 31} was unable to prevent oxidative stress-induced apoptosis in these cells. Apoptosis of amacrine neurons, which in the insulin-containing culture medium used is negligible at this time in culture, ⁴¹ was considerably increased by addition of paraquat. Remarkably, the increase in apoptosis in amacrine cells was not precluded by DHA, suggesting that the operative mechanisms leading to DHA’s activation of survival pathways are only found in photoreceptors. 

Our results suggest that paraquat induces apoptosis in both amacrine and photoreceptor neurons by a similar mechanism, in which mitochondrial membrane depolarization appears to be involved. Most researchers agree that mitochondria have a central role in the triggering of programmed cell death. ^{33 34} Membrane lipid peroxidation can lead to this organelle dysfunction ⁴² and paraquat has been shown to affect mitochondrial membrane potential on primary cortical neurons. ²⁵ In the retina, alteration of mitochondria permeability participates in rod apoptosis induced by lead and calcium ⁴³ and in photoreceptor apoptosis due to the lack of trophic factors during development in vitro. ^{30 31} Therefore, the loss of mitochondrial membrane integrity seems to be linked to photoreceptor apoptosis, both during development and after toxic insults. Conversely, postponement of photoreceptor apoptosis by DHA and GDNF during early development in vitro is associated with the maintenance of mitochondrial membrane integrity. ^{30 31} Similarly, our present results suggest that DHA’s ability to prevent mitochondrial membrane depolarization is crucial for photoreceptor protection after oxidative damage. In contrast, and consistent with its lack of protective effect on amacrine neurons, DHA did not prevent mitochondrial membrane depolarization in these neurons after treatment with paraquat. 

It is still not established whether permeabilization of the outer mitochondrial membrane is due to nonspecific rupture or to the opening or formation of large channels. ⁴⁴ As a lipid molecule, DHA could have a direct, structural effect, modulating membrane properties and hence permeability or channel opening. As a trophic factor, it could act in a more indirect manner—for instance, regulating the expression of anti- and proapoptotic proteins from the Bcl-2 family, which are strongly involved in the control of mitochondrial functionality. ^{33 44 45} Ectopic expression of Bax in photoreceptors results in extensive cell death. ⁴⁶ Bax levels are directly related to the permeabilization of the mitochondrial membrane and the ratio of Bcl-2 to Bax is apparently crucial for preventing it. ⁴⁷ Acute oxidative stress reduces expression of Bcl-2 in retinal frog photoreceptors, ⁴⁸ whereas overexpression of Bcl-2 and a higher Bcl-2-to-Bax ratio protect a photoreceptor-like cell line from photooxidative damage. ⁴⁹ In the current study, oxidative stress altered the expression of Bax and Bcl-2 in photoreceptors, leading to a substantial reduction in the Bcl-2-to-Bax ratio. Such changes may contribute to destabilization of mitochondria in these cells. The present results suggest that the upregulation of Bcl-2 in these cells and the simultaneous inhibition of the oxidative-stress–induced increase in Bax levels, thus preserving mitochondrial membrane integrity, could be at least one of the mechanisms activated by DHA to prevent apoptosis in photoreceptors. Consistent with this hypothesis, GDNF, which was unable to induce an increase in Bcl-2 levels similar to that of DHA, was therefore ineffective in preventing oxidative-stress–induced apoptosis in photoreceptors. 

These results suggest that in these cells, the pathways leading to oxidative-stress–induced apoptosis and to apoptosis resulting from the absence of trophic molecules during development may either differ from each other upstream of mitochondrial membrane permeabilization, or they may be the same, with only DHA being able to block them effectively, at least at this time in development. Several mechanisms have been proposed to explain DHA’s action as a cellular antioxidant in certain experimental conditions: regulation of the levels of reactive oxygen species due to its chemical structure, modulation of membrane properties, or activation of cellular antioxidant enzymes. ^{18 22} Similar chemical or membrane-derived protective mechanisms may be operative in photoreceptors. However, our results suggest that DHA also functions as an effective survival factor for retina photoreceptors, acting at an early stage to slow down the death cascade leading to apoptosis, both during development and when induced by an oxidative insult. 

 Supported by the National Research Council (CONICET), the Fund for Research in Science and Technology (FONCYT), and the Universidad Nacional del Sur, Argentina. NPR and LEP are CONICET research career members, OLG has a CONICET doctoral fellowship, and RG has a fellowship from the Universidad Nacional del Sur.

 Submitted for publication September 4, 2002; revised November 22, 2002; accepted December 11, 2002.

 Disclosure: N.P. Rotstein, None; L.E. Politi, None; O.L. German, None; R. Girotti, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Paraquat-induced apoptosis of photoreceptors in culture. Photomicrographs of rat retinal neurons cultured for 4 days in a chemically defined medium (control: A, B, E, F, I, J) or treated at day 3 with 24 μM paraquat for 24 hours. (C, D, G, H, K, L). Fluorescence photomicrographs of photoreceptors, identified with the monoclonal antibody Rho4D2 (A–D, open arrows) and their nuclei, labeled with the DNA marker DAPI (E–H). Phase micrographs of these cultures are also shown (I, J, K, L). In paraquat-treated cultures, the presence of fragmented and condensed nuclei is clearly observed (G, H, white solid arrows; K, L, black arrows). Note the presence of an enucleated cell (G, K, arrowhead) and of an isolated nuclear fragment (G, K). Scale bars: (B, D, F, H) 15 μm; (A, C, E, G, I–K) 20 μm.

Figure 2.

 

Paraquat-induced apoptosis in amacrine neurons in culture. Fluorescence (A–D) and phase (E, F) photomicrographs of rat retinal neurons in control (A, C, E) or paraquat-treated (B, D, F) cultures. Amacrine neurons (open arrows) were identified with the monoclonal antibody HPC-1 (A, B) and their nuclei integrity observed with DAPI (C, D). Note the clear nuclei fragmentation (D, white solid and white open arrows) the presence of nuclear fragments (D, solid arrow; F, black solid arrow) and the loss of neurites (compare E and F) after treatment with paraquat. Scale bar, 20 μm.

Figure 3.

 

Effect of paraquat on apoptosis in retinal neurons. Three-day retinal cultures were treated with (PQ) or without (control) 24 μM paraquat, as described in Figure 1 . Data represent the number of apoptotic photoreceptor and amacrine neurons, quantitated by determining nuclei integrity with DAPI in cells identified with the monoclonal antibodies Rho4D2 and HPC-1, respectively. *P < 0.01 versus the control group.

Figure 4.

 

Effect of paraquat concentration on the apoptosis of retinal neurons. Cultures were treated with paraquat (PQ) concentrations ranging from 2.4 to 240 μM for 24 hours, and the number of apoptotic photoreceptors (○) and amacrine neurons (▪) was then determined, as described in Figure 3 .

Figure 5.

 

Protective effect of DHA on oxidative-stress–induced apoptosis in photoreceptors. One-day cultures were supplemented or were without 6.7 μM DHA, and then treated at day 3 without (−) or with (+) either 24 or 240 μM paraquat (PQ). Data are the percentage of apoptotic (A) photoreceptors and (B) amacrine neurons. *P < 0.01 versus the control group.

Table 1.

 

Effect of Fatty Acids and GDNF on the Oxidative Stress-Induced Apoptosis of Photoreceptors

	Apoptotic Photoreceptors (%)
1) Effect of fatty acid supplementation on paraquat-induced apoptosis
−	71.5 ± 3.8
Palmitic acid	75.7 ± 1.9
Oleic acid	72.2 ± 5.5
Arachidonic acid	71.3 ± 5.7
DHA	39.1 ± 4.8*
2) Lack of protection by GDNF against paraquat-induced apoptosis
Control (−paraquat), †	26.4 ± 2.9
+GDNF (−paraquat), †	24.5 ± 3.6
+Paraquat	70.7 ± 7.2*
+GDNF+paraquat	67.1 ± 7.0*

Figure 6.

 

Fluorescent labeling of neuronal mitochondria shows the effect of treatment with paraquat and DHA supplementation. Phase-contrast (left) and fluorescence (right) photomicrographs of retinal cultures labeled with a fluorescent probe (B, D, F, solid, thin arrows) showing amacrine neurons (open arrows) and photoreceptors (arrowheads) in control (A, B), paraquat-treated (C, D), and DHA-supplemented, paraquat-treated cultures (E, F) Note the bright fluorescent mitochondria in control (B) and DHA-supplemented, paraquat-treated cultures (F) and the almost complete disappearance of fluorescence-labeled mitochondria in paraquat-treated cultures (D). Scale bars: (A, C, E) 20 μm; (B, D, F) 15 μm.

Figure 7.

 

Effect of oxidative stress and DHA on mitochondrial membrane integrity of photoreceptor and amacrine neurons. Data are the percentage of (A) photoreceptors and (B) amacrine neurons retaining mitochondrial membrane integrity, determined by labeling with a fluorescent probe, in cultures without (−PQ) or with (+PQ) paraquat and with or without DHA. *P < 0.01 versus the control group.

Figure 8.

 

Effect of oxidative stress and DHA on Bax and Bcl-2 expression in photoreceptors. Data are the percentage of photoreceptors expressing (A) Bax and (B) Bcl-2, quantitated by using specific monoclonal antibodies, and (C) the Bcl2-to-Bax ratio, in cultures without (−PQ) or with (+PQ) paraquat and with or without DHA. *P < 0.01 versus the control group.

Figure 9.

 

DHA-induced increase in photoreceptors expressing Bcl-2. Fluorescence (A–D) and phase-contrast (E, F) photomicrographs of retinal neurons in control (A, C, E) or DHA-supplemented (B, D, F) cultures. (A, B) Bcl-2 expression in photoreceptors (open arrows) determined with a specific monoclonal antibody; (C, D) nuclei labeled with DAPI. The increase in the amount of photoreceptors expressing Bcl-2 was clearly visible in DHA-supplemented cultures. Scale bars: (A–D) 15 μm; (E, F) 20 μm.

The authors thanks Beatriz de los Santos for excellent technical assistance and Nestor Carri for his generous supply of GDNF.