Dietary Deficiency of N-3 Fatty Acids Alters ... - Semantic Scholar
Dietary Deficiency of N-3 Fatty Acids Alters Rhodopsin Content and Function in the Rat Retina Ronald A. Bush,* Armand Malnoe\\ Charlotte E. Reme,* and Theodore P. WilliamsX Purpose. To investigate the possibility that previously demonstrated reductions in photoreceptor sensitivity to light in n-3 fatty-acid-deficient rats can be explained by alterations in rhodopsin content, and/or function. Methods. Sprague-Dawlcy rats were reared throughout gestation, lactation, and up to 24 weeks of age on a diet containing sarTlower oil (n-3 fatty-acid-deficient) or soybean oil as the sole source of lipids. Dark-adapted content and in vivo regeneration of rhodopsin after bleaching were measured by detergent extraction. The regeneration rate constants and number of photons absorbed by rhodopsin under steady-state bleach conditions were calculated from these values. The rate of metarhodopsin II (Mil) formation in vitro was determined by flash bleaching extracted pigment and native rod outer segment membranes. Rod outer segment length and photoreceptor cell density were determined in histologic sections through the inferior central retina. Results. Dark-adapted rhodopsin content of retinas from rats reared on sarTlower oil was 12% to 15% higher than that of rats raised on soybean oil at every age measured. The rate of rhodopsin regeneration was significantly slower in rats reared on sarTlower oil while the level of steady-state bleach was the same. This meant that the rats reared on salllowcr oil absorbed about one half as many photons during light exposure. The rate of metarhodopsin II formation in vitro was unaffected by n-3 fatly acid deficiency. No difference in either rod outer segment length or cell number was detected. Conclusion. A reduced capacity for photon absorption by rhodopsin could play a role in lowering retinal sensitivity to light in n-3 fatty-acid-deficient rats. Invest Ophihalmol Vis Sci. 1994; 35:91-100
JLt has been known for some time that phospholipids play an important role in rhodopsin function.1"3 In vitro investigations have revealed some of the specific characteristics of lipids, such as acyl chain length and degree of desaturation,4 thai influence rhodopsin function. The importance of a specific fatty acid, docosahexaenoic acid (DHA), has also been recently demonstrated.5 From the *Deparlment of Ophthalmology, University of Ziirirh. Zurich, Switzerland, the fNestle Research Centra, Nestec. Ltd., Ijiusanne, Switzerland, and the ^Department of Biological Science. Florida State University, Tallahassee, Florida. Dr. Hush is currently affiliated with the University of Michigan, Kellogg F.ye Center, Ann Arbor, Michigan. This research was supported by a grant from the Swiss National Science Foundation, Nr. 31-9156.87. Presented m part at the annual meeting of the Association for Research in Vision and Ophthalmology, Samsola. Florida, April 28, 1991. Submitted for publication: March 17, 1993; revised June 28, 1993; accepted June 30, 1993. Proprietary interest category: N. Reprint requests: Ronald A. Hush, University of Michigan, Kellogg Fye Center, 11)00 Wall Street. Ann Arbor. Ml -18105.
Investigative Ophthalmology & Visual Science, January 1994. Vol. 3">, No. 1 Copyright © Association for Research in Vision and Ophthalmology
DHA (22:6n-3) is the major polyunsat united fatty acid in the phospholipids of the rod outer segment (ROS) membranes. Because DHA itself or its precursor, cv-linolenic acid (18:3n-3), must be furnished in the diet, dietary deprivation of these fatty acids can reduce membrane DHA content, especially if the deprivation is continued for two generations.°~° DHA is closely associated with rhodopsin in ROS membranes 10 " and a deficiency in n-3 fatty acids alters visual function in both animals7i812~M and humans.15 In rats, a deficiency is also associated with reduced susceptibility to light-induced photoreceptor cell death16 and ROS disk disruptions.9 In addition, lightstimulated disk shedding was shown to be reduced in rats made deficient in n-3 fatty acids.0 The absorption of light by the visual pigment is the first step in vision and in certain nonvisual processes in the retina. In rats, the wavelength dependence of the scotopic ERG response,17 light damage with low to moderate levels of illumination (see reference 18 for
91
92
Investigative Ophthalmology 8c Visual Science, January 1994, Vol. 35, No. 1
review), and light-elicited disk shedding 19 is similar to the action spectrum of rhodopsin. If n-3 fatty acid deficiency alters the function of the visual pigment this could explain its effects on these diverse light-stimulated retinal processes. In this study we present evidence that reduction of retinal DHA content by dietary means alters rhodopsin content and function in vivo.
METHODS Experimental Animals and Diets All experiments were conducted in accordance with the ARVO Resolution on the Use of Animals in Research. Sprague-Dawley rats (Ifl'a Credo, France) were individually housed under dim (10 lux) cyclic illumination and fed semipurified diets containing soybean oil (SO) or safflower oil (SFO, n-3 deficient) as the sole source of lipids as described previously. 920 The fatty acid composition of the diets is presented in Table 1. The dietary regimens were extended over two generations of animals and the experiments were performed using 4, 8, 12, and 24-week-old second-generation male offspring.
Lipid Analysis The fatty acid composition of the retinal phospholipids was measured in 8-week old SO and SFO animals. The method has been described previously by Malnoe et al.20
Dark-Adapted Rhodopsin Dark-adapted rhodopsin levels were determined for rats from both dietary groups at 4, 8, 12, and 24 weeks TABLE l. Fatty Acid Composition of Diets Diet Soybean Oil (SO) Fatty Acid 14:0
16:0 16:ln-7 18:0 18:ln-9 18:2n-6 18:3n-6 18:3n-3 20:0 20:ln-9 22:0 22:ln-9 PUFA, sum n-3/11-6
Safflower Oil (SFO)
Weight % of Total Fatty Acids
0.25 10.16 0.33 3.67 36.90 41.50 0.76 4.19 0.44 0.52 0.89 0.33 46.45 0.099
0.33 7.84 0.27 2.59 13.60 73.82 1.12 0.22 0.21 ND ND ND
75.16 0.003
ND, none detected; PUFA, polyunsaturated fatty acids (n—3 + 71-6).
of age alter being dark-adapted overnight (12 to 15 hours). All procedures were carried out in dim red light. After killing the animals by decapitation their retinas were removed by extrusion through a slit in the cornea and immediately frozen in liquid nitrogen. Rhodopsin was extracted from the retinas using 15% Triton X-100 according to the method of Fulton, Manning, Baker, Schukar, and Bailey.21 The prebleach and postbleach absorbance of the extract was measured from 360 to 700 nm using a Beckman (Fullerton, CA) DU-70 spectrophoiometer and the difference in absorbance at 500 nm was used to calculate the number of nanomoles of rhodopsin per eye. Bleaching of the samples was done in the presence of 1 % hydroxylamine by a while fluorescent light source (900 /xw/ cm2) at a distance of 30 cm for 5 minutes.
Regeneration Rates Eight-week-old rats from both dietary groups were dark-adapted overnight. They were then put in reflective cages in groups of 3 or 4 and exposed to 80 lux cool white fluorescent light for 1 hour. The light intensity did not vary more than 20% in any direction. After 1 hour they were assumed to have reached steady state 22 and the lights were tinned off. The retinas were immediately removed from the first group of rats for measurement of steady-state rhodopsin levels. Preliminary experiments using rats from both groups showed that 80 lux produced a steady-state bleach level that could be measured easily and compared between groups. After light exposure additional rats from both dietary groups were allowed to remain in the dark for 1/2 and 1 hour before being killed. Their retinas were immediately frozen in liquid nitrogen for later measurement of rhodopsin regeneration that had taken place in vivo. Because the regeneration of rat visual pigment in the dark is a first-order process 25 " 2:> a plot of the natural log of the fraction of the total pigment in the bleached state with time yields a straight line. The slope of this line is the regeneration rate constant, kr. See Penn and Williams25 for a discussion of the validity of assuming first-order kinetics. Another reason for choosing a 1-hour exposure time was our wish to study photochemical processes occurring during the time in which acute structural alterations in ROS membranes are initiated.9'20 It was necessary, however, to reduce the intensity from our previous study to avoid photoreceptor damage, which may have made interpretation of our results difficult.
Photon Counting The number of photons absorbed per eye per second in vivo at steady state in 80 lux was determined as described by Penn and Williains2:) for 8-week-old rats. Briefly, the reaction between light and rhodopsin can be described by the chemical equation
Rhodopsin in N-3 Fatty Acid Deficiency I-kr
where R = unbleached rhodopsin, I = light intensity, k,= the bleaching rate constant, B = bleached rhodopsin, and kr = the regeneration rate constant. The light intensity at the retina and the bleaching rate are unknown. However, the rhodopsin content of the retina and the regeneration rate in the dark can be measured as described earlier. The amount of bleached rhodopsin can be calculated by subtracting the measured value at any time from the dark-adapted value. At steady state the number of molecules being bleached per second is equal to the number being regenerated per second or, mathematically, R ss -(J'k,) = BSS'K,., where Rss = unbleached rhodopsin at steady state and Bss = rhodopsin bleached at steady .state. Perm and Williams2'' discussed the justification for using this simplified expression as an alternative to more complex ones presented elsewhere. 2 ' 28 The term on the right can be easily calculated from measured values and represents the number of photons caught per unit time, which result in bleaching of the visual pigment. Values will be reported as "number per 12 hours of light exposure" to facilitate comparison with those reported by Penn and Williams.2'
Metarhodopsin II Rate The rate of metarhodopsin II (Mil) formation in vitro was determined using rhodopsin from 12- and 24week-old SFO- and SO-reared rats. No difference between these two age groups was detected so their results were combined, (see Table 5) The change in absorbance at 400 inn in retinal detergent extracts and ROS membrane preparations with time after a brief light flash was measured. The detergent used was 2% oct:yl-/3-D-giucopyranoside in potassium phosphate buffer, pH 7.0. This detergent was used rather than Triton X-100 to avoid, as far as possible, disturbing the lipid environment of rhodopsin. Extraction with Triton can result in Mil rates that are 16-fold higher than native membranes, whereas some milder detergents have only a small effect on Mil rates. 21 ROS membranes were prepared from 24-week-old animals according to Penn and Anderson, 50 suspended in 67 niM potassium phosphate buffer, pH 7.0, and briefly sonicated before measurement. Measurements were made at 21°C. Samples were flash-bleached using a Vivitar (Santa Monica, CA) 2500 photoflash with a 500 //S duration at half height through a broadband yellow filter. The change in absorbance of the sample at 400 mn was monitored by placing a 400 nm interference filter in front of the detector window of the Beckman DU-70 spectrophotometer and recording the output
93 of the photornultiplier tube with time after the flash using a Tektronix 2230 digital storage oscilloscope, Tekt ionic, Beaverton, OR. The signal was transferred to a computer and stored for later plotting and analysis. In most experiments two retinas from a single animal were used per sample and the flash produced a 20.6 ± 2.6% bleach. In experiments using single retinas the flash intensity was increased to give a larger signal and resulted in a 37 ± 3.3% bleach. There were no differences in results using the two flash intensities.
Histology F.yes removed from 8-week-old animals were prepared for histology as described previously.9 Outer nuclear layer thickness and ROS lengths were measured in seclions through the inferior central retina. Ten equally spaced measurements were made from the optic nerve to the periphery on one section from each animal (n = 6 per dietary group).
RESULTS Phospholipid Analysis The fatty acid content of the retinal phospholipids in 8-week-old animals is shown in Table 2 expressed as percentage distribution (w/w). There are only minor differences between these values and those reported in our previous studies.9-'20 The most striking effect of n-3 fatty acid deprivation was the significant reduction in the n-3/n-6 ratio in all phospholipid fractions (except phosphatidylserine) without significant change in the total percentage of polyunsaturated fatty acids (n3+n-6). Most of the reduction in the n-3/n-6 ratio in SFO rats was attributable to a decrease in 22:6n-3 to 19-25% of the value in SO rats, which was compensated for by an increase in 22:5n-6. Although retinal fatty acids were only measured at 8 weeks this pattern persists even after 6 months on these diets. 20
Rhodopsin Content Whole retina rhodopsin increased rapidly from 4 to 12 weeks of age in both groups but increased only slightly between 12 and 24 weeks (Fig. .1). The levels in the retinas of rats reared on the SFO diet were 12% to 1 5% higher {P < 0.0 1) at each age than in the retinas of rats raised on SO.
Histology To determine whether differences in rhodopsin content could be explained by either photoreceptor cell number or ROS length these parameters were compared in 8-week-old SO- and SFO-reared rats. As indicated in Table 3, no statistically significant difference existed in ROS length or outer nuclear layer thickness between the two groups in the region of the inferior central retina.
ect of N-3 Fatty Acid Deprivation on Fatty Acid Content of Retinal Phospholipids Phosphatidylethanolamine
Phosphatidylcholine SO
0.44 ± 0.04 0.23 ± 0.02 33.16 ± 0.86 1.37 ±0.05 0.28 ±0.01 19.21 ± 0.34 14.97 ±0.17 0.61 ±0.03 0.20 ± 0.01 0.10 ± 0.01 5.28 ±0.12 0.09 ±0.01 0.64 ±0.01 0.38 ± 0.04 0.25 ± 0.01 22.81 ± 1.40 30.36 ± 1.29 3.23 ± 0.25
SFO
0.40 0.23 30.79 1.32 0.25 19.20 13.97 0.62 0.21 0.08 6.20
± 0.04 ±0.00 ± 0.55 ± 0.09 ±0.01 ± 0.20 ± 0.13* ± 0.04 ±0.01 ± 0.00 ±0.14*
ND
1.34 ± 0.05 19.57 ± 0.48* ND
5.81 ± 0.32* 33.83 ± 0.70 0.21 ± 0.01*
Phosphatidylserine
Phosphatidylinos
SO
SFO
SO
SFO
SO
0.14 ± 0.02
0.13 ± 0.02
0.21 ±0.04
0.13 ± 0.02
ND
ND
5.61 ±0.16 0.60 ± 0.08 0.23 ± 0.02 24.02 ± 0.34 4.98 ± 0.46 0.51 ±0.14 0.30 ±0.01 0.16 ± 0.02 9.67 ±0.18 0.09 ± 0.00 2.45 ± 0.05 0.83 ± 0.08 0.61 ±0.02 49.81 ± 0.74 64.43 ± 0.50 3.64 ±0.14
ND
6.79 0.51 0.22 23.62 4.47 0.48 0.31 0.15 10.90
± ± ± ± ± ± ± ± ±
ND
0.16* 0.08 0.00 0.22 0.43 0.04 0.02 0.00 0.23*
ND
3.82 ±0.10* 38.26 ± 0.51* ND
10.33 ± 0.32* 64.26 ± 0.65 0.19 ± 0.01*
ND
3.57 ± 0.33 0.24 ± 0.04 ND
37.92 3.52 0.50 0.29
± ± ± ±
ND
1.02 0.20 0.04 0.03
ND
3.12 ± 0.07 ND
3.69 1.13 1.07 44.74 54.54 5.26
2.48 ±0.19* 0.30 ± 0.03
±0.10 ± 0.09 ± 0.06 ± 0.95 ± 0.88 ±0.21
31.93 3.02 0.34 0.22
±0.87* ± 0.50 ±0.01* ± 0.02
ND
1 1.98 ± 0.45 ND ND
36.60 7.75 0.89 0.90
± 1.23 ±0.19 ± 0.02 ± 0.08
ND
ND
3.06 ± 0.08
31.38 ±0.85
ND
6.72 ± 0.24* 41.59 ± 1.26*
12.0
33.4 8.1 0.9 0.9
35.8
ND
1.90 ± 0.07 ND
ND
ND
10.21 ± 0.42* 62.14 ± 1.53* 0.20 ± 0.01*
8.59 ± 1.97 43.66 ±1.19 0.25 ± 0.06
1.9 5.0
1.6 46.3 0.0
for two generations on diets containing soybean oil (SO) or safllower oil (SFO) as the sole source of lipids. The measurements were made on the retinas of second gene eks of age. Values are means ± SKM (w = 4) expressed as % (w/w) of identified fatty acid:>. Significant differences between diets for each phospholipid class are indica none detected.
Rhodopsin in N-3 Fatty Acid Deficiency
95
2.50
o
2.00
a. o •o o JI on
• — •
1.50
Soybeon oil
o — o Sofflower oil
1.00
12
16
20
o
24
Weeks of age FIGURE 1. The effect of dietary n-3 fatty acid deprivation on whole retina rhodopsin content. Values are means ± SD (n = 6 except at 24 weeks n - 3). *Significantly different (P < 0.01) from rats fed soybean oil. For each dietary treatment up to 1 2 weeks of age there is also a statistically significant age-related increase in rhodopsin content (analysis of variance, P < 0.05).
o
•
Soybean oil - 1 . 2 8 ± 0.17
-3 O Safflower oil -0.71 ± 0.18
c -4
0.00
0.25
0.50
0.75
1.00
Time (hrs.) Rhodopsin Regeneration and Photon Catch The rate of regeneration of rhodopsin in vivo from a steady-state bleach in 80 lux was significantly slower in SFO-reared rats (P < 0.03). Figure 2 plots the regeneration of visual pigment for each group as described in Methods. The regeneration rate constants, represented by the slopes of the least squares fit to the data points, are given in the figure. Based on these values and the level of steady-state bleach the photon catching ability for the rats in the two groups can be calculated as described in Methods. Table 4 shows that the retinas of SFO-reared animals caught nearly 50% fewer photons than those of the SO-reared group. This was mainly due to a reduced pigment regeneration rate because the amount of pigment bleached at steady state was essentially the same in both groups. The fraction of total dark-adapted pigment bleached at steady state was 59 ± 11% and 47 ± 9% for SO and SFO rats, respectively. However, this difference was not statistically significant (P > 0.10).
3. Effect of Dietary N-3 Fatty Acid Deprivation on Retinal Histology in the Rat
TABLE
Diet
ONL Thickness (\im)
ROS Length (\im)
Soybean oil Safflower oil
36.4 ± 2.2 37.0 ± 2.6
31.2 ± 4.0 32.3 ± 2.3
Outer nuclear layer (ONL) thickness and rod outer segment (ROS) length were measured in 1.5-/im thick sections through the inferior central retinas of 8-week old rats reared on soybean oil and safflower oil. Values are the mean ± SD (n = 6). There is no statistically significant, difference between dietary groups.
FIGURE 2. The effect of dietary n-3 fatty acid deprivation on rhodopsin regeneration rates. First-order rate plots of the regeneration of visual pigment in the dark in vivo show a statistically significant slower rate for SFO-reared rats (analysis of variance, P < 0.03). Regeneration was measured in 8-week-old rats after exposure to 80 lux for J hour. The y-axis is the natural log of the fraction bleached. The slopes of the lines are the regeneration rate constants and are given at the bottom of the graph (± SEM). Rho x = dark-adapted rhodopsin; rho, = rhodopsin at time l.
Metarhodopsin II Formation The rate of formation of Mil in vitro was not significantly affected by diet. Figure 3 shows the averaged and normalized output of the photomultiplier tube indicating change in absorbance at 400 nm with time after a flash bleach for each type of preparation (pigment extract or native ROS membranes). The curves from SO- and SFO-reared rats have identical time courses indicating that there was no difference in the rate of formation of Mil. Table 5 gives the average rates for the pigment extracts and native membranes from the two dietary groups derived from these plots. The value of l/ti / 2 has been used as a measure of rate and is sufficient for comparing the SFO and SO groups. The advantages and disadvantages of using 1/ t1/2 are discussed in Baker et al.29 There was no difference between the values for pigment extracts from fresh and frozen retinas so these have been combined. Table 5 and Figure 3 indicate that the SFO diet did not affect the rate of Mil formation in vitro. There was also no difference between SO and SFO rats in total amount of Mil formed (change in absorbance per nmol rhodopsin bleached, data not shown).
96
Investigative Ophthalmology & Visual Science, January 1994, Vol. 35, No. 1
TABLE 4.
Effect of Dietary Deprivation of N-3 Fatty Acids on the Photon-Catching Ability of the Rat Retina 5
Bt
K
molecules eye-'xio-'
Diet Soybean oil SafTlower oil
1.19 ± . 0 7 1.37 ± .09
0.49 ± .13 0.72 ± .13
sec ' x/0-
JLt has been known for some time that phospholipids play an important role in rhodopsin function.1"3 In vitro investigations have revealed some of the specific characteristics of lipids, such as acyl chain length and degree of desaturation,4 thai influence rhodopsin function. The importance of a specific fatty acid, docosahexaenoic acid (DHA), has also been recently demonstrated.5 From the *Deparlment of Ophthalmology, University of Ziirirh. Zurich, Switzerland, the fNestle Research Centra, Nestec. Ltd., Ijiusanne, Switzerland, and the ^Department of Biological Science. Florida State University, Tallahassee, Florida. Dr. Hush is currently affiliated with the University of Michigan, Kellogg F.ye Center, Ann Arbor, Michigan. This research was supported by a grant from the Swiss National Science Foundation, Nr. 31-9156.87. Presented m part at the annual meeting of the Association for Research in Vision and Ophthalmology, Samsola. Florida, April 28, 1991. Submitted for publication: March 17, 1993; revised June 28, 1993; accepted June 30, 1993. Proprietary interest category: N. Reprint requests: Ronald A. Hush, University of Michigan, Kellogg Fye Center, 11)00 Wall Street. Ann Arbor. Ml -18105.
Investigative Ophthalmology & Visual Science, January 1994. Vol. 3">, No. 1 Copyright © Association for Research in Vision and Ophthalmology
DHA (22:6n-3) is the major polyunsat united fatty acid in the phospholipids of the rod outer segment (ROS) membranes. Because DHA itself or its precursor, cv-linolenic acid (18:3n-3), must be furnished in the diet, dietary deprivation of these fatty acids can reduce membrane DHA content, especially if the deprivation is continued for two generations.°~° DHA is closely associated with rhodopsin in ROS membranes 10 " and a deficiency in n-3 fatty acids alters visual function in both animals7i812~M and humans.15 In rats, a deficiency is also associated with reduced susceptibility to light-induced photoreceptor cell death16 and ROS disk disruptions.9 In addition, lightstimulated disk shedding was shown to be reduced in rats made deficient in n-3 fatty acids.0 The absorption of light by the visual pigment is the first step in vision and in certain nonvisual processes in the retina. In rats, the wavelength dependence of the scotopic ERG response,17 light damage with low to moderate levels of illumination (see reference 18 for
91
92
Investigative Ophthalmology 8c Visual Science, January 1994, Vol. 35, No. 1
review), and light-elicited disk shedding 19 is similar to the action spectrum of rhodopsin. If n-3 fatty acid deficiency alters the function of the visual pigment this could explain its effects on these diverse light-stimulated retinal processes. In this study we present evidence that reduction of retinal DHA content by dietary means alters rhodopsin content and function in vivo.
METHODS Experimental Animals and Diets All experiments were conducted in accordance with the ARVO Resolution on the Use of Animals in Research. Sprague-Dawley rats (Ifl'a Credo, France) were individually housed under dim (10 lux) cyclic illumination and fed semipurified diets containing soybean oil (SO) or safflower oil (SFO, n-3 deficient) as the sole source of lipids as described previously. 920 The fatty acid composition of the diets is presented in Table 1. The dietary regimens were extended over two generations of animals and the experiments were performed using 4, 8, 12, and 24-week-old second-generation male offspring.
Lipid Analysis The fatty acid composition of the retinal phospholipids was measured in 8-week old SO and SFO animals. The method has been described previously by Malnoe et al.20
Dark-Adapted Rhodopsin Dark-adapted rhodopsin levels were determined for rats from both dietary groups at 4, 8, 12, and 24 weeks TABLE l. Fatty Acid Composition of Diets Diet Soybean Oil (SO) Fatty Acid 14:0
16:0 16:ln-7 18:0 18:ln-9 18:2n-6 18:3n-6 18:3n-3 20:0 20:ln-9 22:0 22:ln-9 PUFA, sum n-3/11-6
Safflower Oil (SFO)
Weight % of Total Fatty Acids
0.25 10.16 0.33 3.67 36.90 41.50 0.76 4.19 0.44 0.52 0.89 0.33 46.45 0.099
0.33 7.84 0.27 2.59 13.60 73.82 1.12 0.22 0.21 ND ND ND
75.16 0.003
ND, none detected; PUFA, polyunsaturated fatty acids (n—3 + 71-6).
of age alter being dark-adapted overnight (12 to 15 hours). All procedures were carried out in dim red light. After killing the animals by decapitation their retinas were removed by extrusion through a slit in the cornea and immediately frozen in liquid nitrogen. Rhodopsin was extracted from the retinas using 15% Triton X-100 according to the method of Fulton, Manning, Baker, Schukar, and Bailey.21 The prebleach and postbleach absorbance of the extract was measured from 360 to 700 nm using a Beckman (Fullerton, CA) DU-70 spectrophoiometer and the difference in absorbance at 500 nm was used to calculate the number of nanomoles of rhodopsin per eye. Bleaching of the samples was done in the presence of 1 % hydroxylamine by a while fluorescent light source (900 /xw/ cm2) at a distance of 30 cm for 5 minutes.
Regeneration Rates Eight-week-old rats from both dietary groups were dark-adapted overnight. They were then put in reflective cages in groups of 3 or 4 and exposed to 80 lux cool white fluorescent light for 1 hour. The light intensity did not vary more than 20% in any direction. After 1 hour they were assumed to have reached steady state 22 and the lights were tinned off. The retinas were immediately removed from the first group of rats for measurement of steady-state rhodopsin levels. Preliminary experiments using rats from both groups showed that 80 lux produced a steady-state bleach level that could be measured easily and compared between groups. After light exposure additional rats from both dietary groups were allowed to remain in the dark for 1/2 and 1 hour before being killed. Their retinas were immediately frozen in liquid nitrogen for later measurement of rhodopsin regeneration that had taken place in vivo. Because the regeneration of rat visual pigment in the dark is a first-order process 25 " 2:> a plot of the natural log of the fraction of the total pigment in the bleached state with time yields a straight line. The slope of this line is the regeneration rate constant, kr. See Penn and Williams25 for a discussion of the validity of assuming first-order kinetics. Another reason for choosing a 1-hour exposure time was our wish to study photochemical processes occurring during the time in which acute structural alterations in ROS membranes are initiated.9'20 It was necessary, however, to reduce the intensity from our previous study to avoid photoreceptor damage, which may have made interpretation of our results difficult.
Photon Counting The number of photons absorbed per eye per second in vivo at steady state in 80 lux was determined as described by Penn and Williains2:) for 8-week-old rats. Briefly, the reaction between light and rhodopsin can be described by the chemical equation
Rhodopsin in N-3 Fatty Acid Deficiency I-kr
where R = unbleached rhodopsin, I = light intensity, k,= the bleaching rate constant, B = bleached rhodopsin, and kr = the regeneration rate constant. The light intensity at the retina and the bleaching rate are unknown. However, the rhodopsin content of the retina and the regeneration rate in the dark can be measured as described earlier. The amount of bleached rhodopsin can be calculated by subtracting the measured value at any time from the dark-adapted value. At steady state the number of molecules being bleached per second is equal to the number being regenerated per second or, mathematically, R ss -(J'k,) = BSS'K,., where Rss = unbleached rhodopsin at steady state and Bss = rhodopsin bleached at steady .state. Perm and Williams2'' discussed the justification for using this simplified expression as an alternative to more complex ones presented elsewhere. 2 ' 28 The term on the right can be easily calculated from measured values and represents the number of photons caught per unit time, which result in bleaching of the visual pigment. Values will be reported as "number per 12 hours of light exposure" to facilitate comparison with those reported by Penn and Williams.2'
Metarhodopsin II Rate The rate of metarhodopsin II (Mil) formation in vitro was determined using rhodopsin from 12- and 24week-old SFO- and SO-reared rats. No difference between these two age groups was detected so their results were combined, (see Table 5) The change in absorbance at 400 inn in retinal detergent extracts and ROS membrane preparations with time after a brief light flash was measured. The detergent used was 2% oct:yl-/3-D-giucopyranoside in potassium phosphate buffer, pH 7.0. This detergent was used rather than Triton X-100 to avoid, as far as possible, disturbing the lipid environment of rhodopsin. Extraction with Triton can result in Mil rates that are 16-fold higher than native membranes, whereas some milder detergents have only a small effect on Mil rates. 21 ROS membranes were prepared from 24-week-old animals according to Penn and Anderson, 50 suspended in 67 niM potassium phosphate buffer, pH 7.0, and briefly sonicated before measurement. Measurements were made at 21°C. Samples were flash-bleached using a Vivitar (Santa Monica, CA) 2500 photoflash with a 500 //S duration at half height through a broadband yellow filter. The change in absorbance of the sample at 400 mn was monitored by placing a 400 nm interference filter in front of the detector window of the Beckman DU-70 spectrophotometer and recording the output
93 of the photornultiplier tube with time after the flash using a Tektronix 2230 digital storage oscilloscope, Tekt ionic, Beaverton, OR. The signal was transferred to a computer and stored for later plotting and analysis. In most experiments two retinas from a single animal were used per sample and the flash produced a 20.6 ± 2.6% bleach. In experiments using single retinas the flash intensity was increased to give a larger signal and resulted in a 37 ± 3.3% bleach. There were no differences in results using the two flash intensities.
Histology F.yes removed from 8-week-old animals were prepared for histology as described previously.9 Outer nuclear layer thickness and ROS lengths were measured in seclions through the inferior central retina. Ten equally spaced measurements were made from the optic nerve to the periphery on one section from each animal (n = 6 per dietary group).
RESULTS Phospholipid Analysis The fatty acid content of the retinal phospholipids in 8-week-old animals is shown in Table 2 expressed as percentage distribution (w/w). There are only minor differences between these values and those reported in our previous studies.9-'20 The most striking effect of n-3 fatty acid deprivation was the significant reduction in the n-3/n-6 ratio in all phospholipid fractions (except phosphatidylserine) without significant change in the total percentage of polyunsaturated fatty acids (n3+n-6). Most of the reduction in the n-3/n-6 ratio in SFO rats was attributable to a decrease in 22:6n-3 to 19-25% of the value in SO rats, which was compensated for by an increase in 22:5n-6. Although retinal fatty acids were only measured at 8 weeks this pattern persists even after 6 months on these diets. 20
Rhodopsin Content Whole retina rhodopsin increased rapidly from 4 to 12 weeks of age in both groups but increased only slightly between 12 and 24 weeks (Fig. .1). The levels in the retinas of rats reared on the SFO diet were 12% to 1 5% higher {P < 0.0 1) at each age than in the retinas of rats raised on SO.
Histology To determine whether differences in rhodopsin content could be explained by either photoreceptor cell number or ROS length these parameters were compared in 8-week-old SO- and SFO-reared rats. As indicated in Table 3, no statistically significant difference existed in ROS length or outer nuclear layer thickness between the two groups in the region of the inferior central retina.
ect of N-3 Fatty Acid Deprivation on Fatty Acid Content of Retinal Phospholipids Phosphatidylethanolamine
Phosphatidylcholine SO
0.44 ± 0.04 0.23 ± 0.02 33.16 ± 0.86 1.37 ±0.05 0.28 ±0.01 19.21 ± 0.34 14.97 ±0.17 0.61 ±0.03 0.20 ± 0.01 0.10 ± 0.01 5.28 ±0.12 0.09 ±0.01 0.64 ±0.01 0.38 ± 0.04 0.25 ± 0.01 22.81 ± 1.40 30.36 ± 1.29 3.23 ± 0.25
SFO
0.40 0.23 30.79 1.32 0.25 19.20 13.97 0.62 0.21 0.08 6.20
± 0.04 ±0.00 ± 0.55 ± 0.09 ±0.01 ± 0.20 ± 0.13* ± 0.04 ±0.01 ± 0.00 ±0.14*
ND
1.34 ± 0.05 19.57 ± 0.48* ND
5.81 ± 0.32* 33.83 ± 0.70 0.21 ± 0.01*
Phosphatidylserine
Phosphatidylinos
SO
SFO
SO
SFO
SO
0.14 ± 0.02
0.13 ± 0.02
0.21 ±0.04
0.13 ± 0.02
ND
ND
5.61 ±0.16 0.60 ± 0.08 0.23 ± 0.02 24.02 ± 0.34 4.98 ± 0.46 0.51 ±0.14 0.30 ±0.01 0.16 ± 0.02 9.67 ±0.18 0.09 ± 0.00 2.45 ± 0.05 0.83 ± 0.08 0.61 ±0.02 49.81 ± 0.74 64.43 ± 0.50 3.64 ±0.14
ND
6.79 0.51 0.22 23.62 4.47 0.48 0.31 0.15 10.90
± ± ± ± ± ± ± ± ±
ND
0.16* 0.08 0.00 0.22 0.43 0.04 0.02 0.00 0.23*
ND
3.82 ±0.10* 38.26 ± 0.51* ND
10.33 ± 0.32* 64.26 ± 0.65 0.19 ± 0.01*
ND
3.57 ± 0.33 0.24 ± 0.04 ND
37.92 3.52 0.50 0.29
± ± ± ±
ND
1.02 0.20 0.04 0.03
ND
3.12 ± 0.07 ND
3.69 1.13 1.07 44.74 54.54 5.26
2.48 ±0.19* 0.30 ± 0.03
±0.10 ± 0.09 ± 0.06 ± 0.95 ± 0.88 ±0.21
31.93 3.02 0.34 0.22
±0.87* ± 0.50 ±0.01* ± 0.02
ND
1 1.98 ± 0.45 ND ND
36.60 7.75 0.89 0.90
± 1.23 ±0.19 ± 0.02 ± 0.08
ND
ND
3.06 ± 0.08
31.38 ±0.85
ND
6.72 ± 0.24* 41.59 ± 1.26*
12.0
33.4 8.1 0.9 0.9
35.8
ND
1.90 ± 0.07 ND
ND
ND
10.21 ± 0.42* 62.14 ± 1.53* 0.20 ± 0.01*
8.59 ± 1.97 43.66 ±1.19 0.25 ± 0.06
1.9 5.0
1.6 46.3 0.0
for two generations on diets containing soybean oil (SO) or safllower oil (SFO) as the sole source of lipids. The measurements were made on the retinas of second gene eks of age. Values are means ± SKM (w = 4) expressed as % (w/w) of identified fatty acid:>. Significant differences between diets for each phospholipid class are indica none detected.
Rhodopsin in N-3 Fatty Acid Deficiency
95
2.50
o
2.00
a. o •o o JI on
• — •
1.50
Soybeon oil
o — o Sofflower oil
1.00
12
16
20
o
24
Weeks of age FIGURE 1. The effect of dietary n-3 fatty acid deprivation on whole retina rhodopsin content. Values are means ± SD (n = 6 except at 24 weeks n - 3). *Significantly different (P < 0.01) from rats fed soybean oil. For each dietary treatment up to 1 2 weeks of age there is also a statistically significant age-related increase in rhodopsin content (analysis of variance, P < 0.05).
o
•
Soybean oil - 1 . 2 8 ± 0.17
-3 O Safflower oil -0.71 ± 0.18
c -4
0.00
0.25
0.50
0.75
1.00
Time (hrs.) Rhodopsin Regeneration and Photon Catch The rate of regeneration of rhodopsin in vivo from a steady-state bleach in 80 lux was significantly slower in SFO-reared rats (P < 0.03). Figure 2 plots the regeneration of visual pigment for each group as described in Methods. The regeneration rate constants, represented by the slopes of the least squares fit to the data points, are given in the figure. Based on these values and the level of steady-state bleach the photon catching ability for the rats in the two groups can be calculated as described in Methods. Table 4 shows that the retinas of SFO-reared animals caught nearly 50% fewer photons than those of the SO-reared group. This was mainly due to a reduced pigment regeneration rate because the amount of pigment bleached at steady state was essentially the same in both groups. The fraction of total dark-adapted pigment bleached at steady state was 59 ± 11% and 47 ± 9% for SO and SFO rats, respectively. However, this difference was not statistically significant (P > 0.10).
3. Effect of Dietary N-3 Fatty Acid Deprivation on Retinal Histology in the Rat
TABLE
Diet
ONL Thickness (\im)
ROS Length (\im)
Soybean oil Safflower oil
36.4 ± 2.2 37.0 ± 2.6
31.2 ± 4.0 32.3 ± 2.3
Outer nuclear layer (ONL) thickness and rod outer segment (ROS) length were measured in 1.5-/im thick sections through the inferior central retinas of 8-week old rats reared on soybean oil and safflower oil. Values are the mean ± SD (n = 6). There is no statistically significant, difference between dietary groups.
FIGURE 2. The effect of dietary n-3 fatty acid deprivation on rhodopsin regeneration rates. First-order rate plots of the regeneration of visual pigment in the dark in vivo show a statistically significant slower rate for SFO-reared rats (analysis of variance, P < 0.03). Regeneration was measured in 8-week-old rats after exposure to 80 lux for J hour. The y-axis is the natural log of the fraction bleached. The slopes of the lines are the regeneration rate constants and are given at the bottom of the graph (± SEM). Rho x = dark-adapted rhodopsin; rho, = rhodopsin at time l.
Metarhodopsin II Formation The rate of formation of Mil in vitro was not significantly affected by diet. Figure 3 shows the averaged and normalized output of the photomultiplier tube indicating change in absorbance at 400 nm with time after a flash bleach for each type of preparation (pigment extract or native ROS membranes). The curves from SO- and SFO-reared rats have identical time courses indicating that there was no difference in the rate of formation of Mil. Table 5 gives the average rates for the pigment extracts and native membranes from the two dietary groups derived from these plots. The value of l/ti / 2 has been used as a measure of rate and is sufficient for comparing the SFO and SO groups. The advantages and disadvantages of using 1/ t1/2 are discussed in Baker et al.29 There was no difference between the values for pigment extracts from fresh and frozen retinas so these have been combined. Table 5 and Figure 3 indicate that the SFO diet did not affect the rate of Mil formation in vitro. There was also no difference between SO and SFO rats in total amount of Mil formed (change in absorbance per nmol rhodopsin bleached, data not shown).
96
Investigative Ophthalmology & Visual Science, January 1994, Vol. 35, No. 1
TABLE 4.
Effect of Dietary Deprivation of N-3 Fatty Acids on the Photon-Catching Ability of the Rat Retina 5
Bt
K
molecules eye-'xio-'
Diet Soybean oil SafTlower oil
1.19 ± . 0 7 1.37 ± .09
0.49 ± .13 0.72 ± .13
sec ' x/0-
