Exp. Eye Res. (1992) 54. 933-939
Uptake
of 22-Carbon NAN
a Department
WANG”*,
Acids
into
REX D.WIEGANDbAND
of Biochemistry,
(Received
Fatty
Rockviiie
ROBERT
and b Cullen Eye Institute, TX 77030, U.S.A. 22 June
Rat Retina
1991 and accepted
Baylor
and Brain
E.ANDERSON”.b College
in revised
of Medicine,
form 31 July
Houston,
1991)
Rat retina accumulates high levels of 22-carbon (Cp?) polyunsaturated fatty acids (PUFA). especially docosahexaenoic acid (DHA, 22 : 6~3), in rod outer segment (ROS) phospholipids (PL). However, plasma. the source of retina lipids, is enriched in 20-carbon (C,,) fatty acids instead of C,, PUFA. This suggests that the retina has a mechanism(s) for selective uptake of C,, PUFA from the blood. It is not known if the selective uptake is specific for the carbon number alone, or if the number of double bonds is also important. To address this question. the following study was carried out using erucic acid (22 : 109) as a metabolic marker molecule. Albino rats were raised from birth on a diet containing loo/, (by weight) of either rapeseed oil (43 y0 22 : 1~9) or blended canola oil (0.40/, 22 : 1~9 J. At 4 months of age. plasma, liver, adrenal gland, brain and retina were collected, lipids were extracted, and fatty acids were determined. In those rats fed rapeseed oil, 22 : 109 was incorporated into the lipids of plasma (2.3 %). liver (06%,),). and adrenal gland (17.6%). indicating that this fatty acid was absorbed, transported. and metabolized by the rats. However. 22 : 109 was not incorporated into the lipids of retinal ROS or brain. Our results suggest that both the carbon number and degree of unsaturation are important determinants in the selective uptake of C,, fatty acids from plasma into both the brain and the retina. Key words : brain : retina ; rod outer segments : membrane lipids : phospholipids : fatty acids : rapeseed oil : erucic acid : canola oil : plasma.
1. Introduction The rod outer segments (ROS) are made up of nearly equal amounts of proteins and lipids. Over 90% of the
integral proteins is the visual pigment rhodopsin (Hargrave, 1982). The lipid bilayer, which provides the critical environment for rhodopsin to function. has a rather simple lipid composition (Fliesler and Anderson, 1983). About 90-95 % of ROS membrane lipids are phospholipids (PL) : cholesterol. diacylglycerols, and free fatty acids account for the remaining lipids. The major phospholipid classes are phosphatidylcholine (PC ; 4045 %). phosphatidylethanolamine (PE ; 30-40 %), and phosphatidylserine (PS ; 10-l 5 %). Phosphatidylinositol, sphingomyelin (SPH) and lysophospholipids make up about 5% of the total phospholipids. The ROS membranes are enriched in 22-carbon (C,,) polyunsaturated fatty acids (PUFA). The major PIJFA is docosahexaenoic acid (DHA, 22 : 6~3). which can account for as much as 50 mol% of the total fatty acids in PE and PS (Fliesler and Anderson, 1983 ). In vertebrates, C,, PUFA (both w3 series and 06 series) are essential fatty acids (EFA) and can only be synthesized by the animal from short-chain precursors (18 : 3w3 or 18 : 206), which must be obtained from the diet. A deficiency of w3 fatty acids results in functional changes in the retina (Wheeler, Benolken and Anderson. 1975 : Lamptey and Walker, 1976 ;
Neuringer et al., 1984, 1986). The importance of C,, PUFA in the retina is also reflected in the strong ability of this tissue to conserve them (especially DHA) during dietary deficiency of w3 fatty acids (Futterman. Downer and Hendrickson, 1971 ; Anderson and Maude, 1972 : Tinoco, 1982 : Stinson. 1990: Wiegand et al., 1991). The mechanism
for enrichment
of 22: 6~3 in the
retina is not clear. Dietary precursors (e.g. 18: 3~3) are converted to their corresponding long-chain products
(e.g. 20: 503,
22: 5~3. or 22:6w3)
in the
liver, incorporated into lipoproteins, and transported in the blood to other tissues. Present evidence indicates that the retina incorporates 22: 6~3 from the blood rather than from elongation of appropriate precursors (Scott and Bazan, 1989 : Alvarez and Anderson. 1990: Strand and Bazan. 1990; Wetzel et al., 1990; Li, Wetzel and O’Brien, 1991). Since the level of C,, PUFA in the plasma is low compared to C,,, PUFA. the retina must have a mechanism(s) for selective uptake of C,, PUFA. However, it is not known if this uptake of C,, PUFA is specific for carbon number alone. or if the number of double bonds is also important. To address this question, we raised rats on diets containing erucic acid (22: lw9). which has 22 carbons but only one double bond, and followed its incorporation into tissue lipids. Since erucic acid is not usually found in mammalian retina and cannot be converted to the C,, PUFA (w3 and w6 series), it acted in this experiment as a metabolic marker molecule.
* For correspondence at: Department of Biochemistry, Baylor College of Medicine. One Baylor Plaza, Houston, TX 77030. IJ.S.A. 0014-4835/92/060933+0/
$03.00/0
0 1992 Academic Press Limited
N. WANG
934
ET AL
2. Materials and Methods Diefs Two diets were formulated and pelleted using AIN7hA basal diet (containing vitamin-free casein) supplemented with 10% (by weight) of either rapeseed oil or blended canola oil (Dyets Inc., Bethlehem, PA). High erucic acid rapeseed oil was the oil used in the rapeseed oil (KG) diet. Canola oil was blended with linseed oil ( 1.910 kg canola oil plus 91.0 g linseed oil) to make the level of 18: 30~3 the same as that in rapeseed oil. The blended canola oil was the oil used in the canola oil (CO) diet. The antioxidant tertiary butylhydroquinone (TBHQ) was added at 0.02 % final concentration in both diets. The pelleted diets were stored at - 20°C until used. Several pellets from each diet group were extracted for total lipid and fatty acid analyses. Food pellets were completely replaced every 4 days, because we have found that there is no loss of 18 : 30)3 from similar diets containing linseed oil ( 10 %, by weight) during this time period (Wiegand et al., 1991). Animals All procedures were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals and the ARVO Resolution on the Use of Animals in Research. Four late-term, pregnant Albino rats (1 S-2 1 days gestation : Harland Sprague Dawley Inc., Indianapolis, IN) were housed at 20 + 1°C under cyclic lighting conditions (12 hr light/dark), with average illumination of 1 lx. Two of the dams were fed the RO diet and the other two the CO diet. Pups were born within 4 days of arrival. The nursing mothers were kept for 3 weeks with their pups and then killed. The young rats were separated according to their sex and diet group. Diets and tap water were given ad libitum. Tissue Collection At 16 weeks of age, the rats were anesthetized deeply by intraperitoneal injection of Nembutal (50 mg kg-’ body weight). Retinas were extruded through a slit made across the entire cornea, quickfrozen in liquid nitrogen, and stored in liquid nitrogen until analysed. Blood was removed via cardiac puncture, collected in heparinized tubes. and kept on ice. The rats were then killed by an overdose injection of Nembutal. Two adrenal glands, a section of liver, and the left cerebrum were removed from each rat and stored at - 20°C until analysed. After sitting on ice for 30 min. blood elements were removed by centrifugation and plasma was collected and stored at - 20°C until analysed. Lipid analyses were performed on tissues obtained from the same three animals of each diet group (II = 3).
KOS were prepared from the pooled retinas from each rat using a modified method (Stinson, Wiegand and Anderson, 1991) of the discontinuous sucrose gradient centrifugation procedure originally described by Papermaster and Dreyer (1974). Previous studies from our laboratory have shown that, in the polyacrylamide gel electrophoresis of ROS prepared by this method, X 5-90% of the total proteins is opsin (Stinson et al.. 1991). Lipid analyses were performed on KOS from three animals of each diet group (rt = 3). Total Lipid Extraction and Lipid Class Separation Total lipids (TL) were extracted from various tissues by the method of Bligh and Dyer ( 19 59) and separated into total phospholipids (PL), cholesterol, free fatty acids (FFA). triglycerides (TG), and cholesterol esters (CE) by one-dimensional, thin-layer chromatography (1-D TLC) on silica gel 60 (EM Science, Curtin Matheson Scientific Inc., Houston, TX) using a solvent system of hexane/diethyl ether/glacial acetic acid (75 : 2 5 : 1. by volume). The phospholipid classes PC, PE. PS and SPH were resolved by 2-D TLC on silica gel HR (Anderson, Maude and Feldman. 1969 ). Lipid classes were localized on the chromatoplate by spraying with 0.05 % 2 ‘. 7’-dichlorofluorescein in 7 5 ‘1, aqueous methanol and viewed under ultraviolet light. Tranmethylation and Fatty Acid Methyl Esters (FAME) Analysis Transmethylation was accomplished by heating the lipids at 95-100°C for 2 hr in 1 ml anhydrous methanol containing 2.5 % concentrated sulfuric acid in the presence of known amounts of internal standards (15:0, 17:O. and 21 :O). FAME were extracted with three volumes of hexane and the combined hexane extract was washed once with water. After methylation of total lipids or cholesterol esters, the free cholesterol was separated from FAME by I-D TLC on silica gel 60 (EM Science, Curtin Matheson Scientific Inc.) using a solvent system of hexane/diethyl ether (80:20. by volume). FAME were analysed and quantitated on a Varian 3 700 gas chromatograph (GC) equipped with a DB22 5 (30 m x 0.25 mm internal diameter: J & W Scientific. Folsom, CA) fused silica capillary column operating in a split injection mode of 1 : 30. The procedure is previously described by Stinson et al. (1991). Statistical Analysis Comparisons of fatty acid composition (mols,), food intake, and body weight were made using Student’s ttest. Since the fatty acid composition (expressed as mol%) is not distributed normally, statistical analyses
UPTAKE
OF 22-CARBON
FATTY
ACIDS
IN RAT TISSUES
were performed using the arc-sine transformation described by Zar ( 1984). 3. Results Fatty Acid Composition of Diets As Table I indicates, both diet groups had similar fatty acid compositions, with two exceptions. The CO diet was enriched in 18 : 1~~19at 5 5.2 %, while the RO diet contained only 14.2x, and the RO diet was enriched 22 : 1t1j9 at 43.4 ‘% compared to the CO diet which contained only 0.4 “/“. Although 18 : 206 was slightly higher in the CO diet than in the RO diet (21.6% vs. lS.iY,), the levels of 18: 3~3 (8.2% in CO. 7.8% in RO) were similar in each diet group. Food Consllmption and Weight Gain There was no significant difference in the food consumption and weight gain between the two diet groups, although differences did exist between male and female rats in each group. Data presented in Table I1 were obtained from 9-week-old rats. Fatty Arid Composition of Tissue Lipids The fatty acid composition of rat plasma lipids after TABLE 1
Fatty acid composition (mol%) of dietary total lipids Fatty acid 16:O 18:O 1 8 : lr09 18 : 1o 7
18 : 2hJh 1 8 : 3~ 3 20: 109 22: lhJ9 24: lb19
co*
Rot
5.7 1.8 55.2 3.0 21.6 8.2 I.2 0.4 0.1
4.4 I.1 14.2 Ia 15.7 7.8 6.4 43.4 0.8
Others
2.8
5.2
* Canola oil. t Rapeseed oil.
TABLE
935
RO or CO supplementation is shown in Table III. In ROsupplemented animals, 22 : 1~9 made up 2.3 % of total fatty acids in plasma. The highest level of 22 : l(d9 was present in the TG fraction (7.5 o/o).However, in terms of total mmol ml-’ of plasma, PL and TG carried the majority of 22: 1~9 in plasma (51.9 and 48.1 nmol ml-’ plasma, respectively). The plasma levels of 22 : 603, the predominant C,, PUFA in ROS, were 3.0 and 3.7% of the total lipids in the CO group and KO group, respectively. In PL, where the majority of 22 :6w3 was found, 22:6w3 reached 5.1% in the RO group, about 2.7 times the level of 22 : 10~9(1.9%). All other C,, fatty acids were found at less than 1.0%. In contrast, C,,, fatty acids were found at high concentrations in the plasma, making up more than 20% of the total fatty acids in plasma in both diet groups, and up to 53 % of the fatty acids in CE. The ratio of 20-carbon fatty acids over 22-carbon fatty acids (CC,,,/CC,,) is between 4 and 6 for TL and approximately 3 for PL. CC&C,, ratio is low for TG (below 2.0) : however, in CE the ratio is more than 2 3. Though 18 : 109 levels were about four-fold higher in the CO diet than in the RO diet (Table 1). the levels of 18: 1w9 in plasma lipids were only slightly higher in the CO group (Table III). Erucic acid was not detected in the three major PL classes (PC, PE and PS) of ROS membranes (Table IV). In addition, diet produced no significant difference in the overall fatty acid composition of the PL classes. 22 : hcti3 was the major PUFA in ROS. making up more than half of the total fatty acids in PS and PE, and about 30% of the fatty acids in PC. The next major components were the saturated fatty acids ( 18 : 0 and 16: 0). with levels up to 50% in PC and 30(x, in PE and PS. Although 20:4(,/6 was found in ROS IX. its level is much lower in ROS than in plasma PI, ( < 6% in ROS, Table IV: > 15 “/o in plasma PL. Table III). The CC&C,, ratio in ROS PL is only 0.1. about onethirtieth of that found in plasma PI,. These observations clearly demonstrated the accumulation of C,, PIJFA over C,,, PUFA in ROS membrane lipid. Fatty acid composition of brain lipids is shown in Table V. In the four phospholipid classes, PC, PE, PS and SPH. very low levels of 22 : 109 (0 3 and 04 y0 in II
Food consumption and body weight in V-week-old r&s Gender Food consumption (g day-’ per rat)
Male Female P-value
Body weight (g)
Male Female P-value
* Canola oil. Values are means k S.E.M. t Number of animals examined. f Rapeseed oil. Values are means F S.E.M.
co*
nt
17.3kO.6
9
14.1 f0.7 < 0.001
4
294* 5 217+6 < 0.001
Y 4
nt
P-value
17.9 & 0.7 13~0$03 < 0001
9 4
> 0.1 > 0.1
284-t 5 Xi& < 0~001
Y 3
> 0.2 > O-5
ROS
N.WANG
CE
TG
PL
TL
ET AL
co
RO
CO
KO
co
RO
CO
RO
(ml%) 16:O 18:O 18 : lW9 18:2w6 18:3/03 20: lw9 20:4w6 20:5w3 22:1w9 22~4~16 22:hw3 24: lW9 Others
17.9 f2.0 12.ort1.9 18.4* 1.6 16.7$1.1 1.3 f-o.2 0.3rto.o 18.9i2.7 2.1kO.3 0~0 * 0.0 0.0 * 0.0 3-O&0.8 0.6 * 0.1 8.8
15.3 k2.2 15.9f23 12.9 * 1.9 12.4f 1.5 0.6 k 0.2 1~0*01 23.Ok2.5 3.1 kO.5 2.3kO.4 0.0 * 0.0 3.7*0-5 1.9kO.3 7.7
21.3 + 3.3 22.1 k 3.6 8.3k1.2 16,8* 1.6 0.2-co.o 0.3 li: 0.0 15.4* 1.9 1.3 kO.2 0.1 + 0.0 0.1 * 0.0 4.2fl.0 1.3 kO.1 8.6
17.7f 3.1 26.4k4.1 7.3 * 0.8 11.7f1.6 0.2 * 0.0 1.2fO.l 15,9*1,3 1.8 +0.2 1.9 *0.1 0.1 kO.0 5.1 kO.3 3.9kO.2 6.9
18.3kO.3 1.8kO.2 45.2+0.5 15.0*0.1 3.3kO.2 0.7+0.1 1.1 +0.2 1.1 iO.1 0.1 f 0.0 0.1 * 04 I.1 +0.1 0.0 + 0.0 12.2
19.4* I.2 2.4*0.2 40.8+ 1.4 9.1 * 1.0 1.8 k 0.5 2.6 f 0.2 1.2 fO.3 0.7*0.1 7.5k1.5 0.0 & 0.0 0~7$0~1 0.2 * 0~0 13.5
8.7-t I.1 0.4 * 04 10.4+ 1.3 19.2* 1.9 0.5 I: 0.1 o-o & 0.0 47.7k4.4 5.1 f 0.5 04 f 04 0~0 & 0~0 1.7i-o-4 0.0 jy 0~0 A.3
7.92 1.2 O.‘it-0.1 9.9 * 0.7 lh.O& 1.9 0.6 * 0.0 0.2 * 0.1 48.1 + 3.1 7.5 i: 0.8 0.2 * oa 04 + 04 2.2 * 0.1 0.0 * 0.0 A.7
(nmol ml-‘) 22 : lW9 22 : 6w3
1.8kO.l 132f38
51,9+13.7 132k23
1.6 k 0.3 16.9& 1.6
48.1k21.3 4.Ok1.5
0.0 Ifr 0.0
211k67
23.3f
5.3
2.7kO.8 30.3& 7.4
3.4kO.3
2,6+0.1
1.9kO.l
0.7*0.1
32.9*
3.7
23.6kO.8
Fatty
acid
0.0 * 0.0 6.1 kO.4
cc&c,,*
133*47 189k42 4.3kO.4
Values are means of analyses from three animals f S.E.M. Abbreviations: TL. total lipids: PL. phospholipids, TC. triglycerides: CE, cholesterol esters; CO, canola oil; RO. rapeseed oil. * Sum ofmolO/, of 20 carbon fatty acids f2O:O. 20: lw9, 20: 1~7. 20:2, 20: 3~9. 20: 3~6, 20:4~6. 70:403, and 20: 5~31 divided by sum ofmol%, of 22 carbon fatty acids (22:0, 22:l~ll. 22: 109, 22:l~i. 22:4oh. 22:5ruh. 22:5~3. and 22:6~3).
TABLE IV
Fatty acid composition of ROS phospholipids PE
PC Fatty acid
PS
co
RO
co
RO
CO
RO
29.6 & 1.0 0.1 + 0.0
29.4+0.3 0.1 f0.0 21.3kO.5 9.2kO.4 0.4+0.0 0.7-tO.l 3.1 * 0.1 0.0 * 0.0 0.0 * 0.0 0.2 f 0.0 0.2 * 04 30.9 * 0.3 0.0 * 0.0 4.6
5,0+0,2 4.1 kO.3 26.2kO.3 2.2 kO.2 0.3kO.l 0.6kO.2 5.4kO.4 0.0 * 0.0 0.0 * 0.0 1.2kO.l 0.3+0-o 50.7+1-S 0.0 + 0.0 4.0
5.0&0~1 4.5kO.4 26.8kO.6 2.0fO.l 0.2 * 0.0 0.6kO.l 5.orfio.3 0.0 * 0.0 0.0 * 0.0 1.2fO.l 0.2 kO.0 50.6f0.7 0.0 * 04 3.8
I.1 f0.4 0.0 * 0.0 30.7-t I.0 0.8kO.2 0.0 i: 0.0 1.1kO.5 2.3kO.4 0.0 + 04 0.0 * 04 2-6kO.l 1.1 f0.0 53.7k2.4 0.0 f 0.0 6.6
0.8kO.l 0.0 * 0.0 30.4kO.9 0.7*0.1 0.0 f 0.0 l.lkO.3 1.7kO.l 0.0 * 0.0 0.0 * 0.0 L-3+0.1 1.2&0~0 56.7kO.8 0.0 * 04 6.0
0.1 kO.0
0.1 +o.o
0.06 rfI 0.02
0.0 5 &- 0.00
(mol%) 16:0 18:DMAs 18:0 18:1w9 l8:2w6 2O:lw9 20:4w6 20:5w3 22:lw9 22 :4w6 22: h3 22:6w3 34:1w9 Others
20.9f0.5 9.4fO.3 0.5 kO.1 0.8kO.l 3.3 kO.2 0.0 f 0.0 o.o+o.o 0.2kO.O 0.2 * 04 30,3*1,5 0.0 f 0.0 4.7
CC,,,pC,,*
0.1 +o.o
0.1 * 0.0
Values are means of analyses from three animals * S.E.M. Abbreviations : PC, phosphatidylcholine. PE. phosphatidylethanolamine ; PS. phosphatidylserine : CO, canola oil: RC, rapeseed oil. * Sum of mol% of 20 carbon fatty acids (2O:O. 20: 109, 20: 1~7. 20:2. 20:3w9, 20:306. 20:4w6. 20:403. and 20:5w3) divided by sum , 22:1(,~9. 22x1~7. 22:40~6. 22:5m6, 22x5~3. and 2216~31. of mol% of 22 carbon fatty acids (22:O. 22:loll
CO and RO groups, respectively) were detected only in PS from both animal groups, and there is no significant difference
between
them
(P > 0.2).
The
major
fatty
acids in the brain were saturates (18 :0 and 16:O)
followed by monoenes (18 : 109). 22 : 6~3 and 20: 4~6 were the major PUFA in the brain. The levels of 22 : 6~3 were higher in the brain than in the plasma, while the reverse was true for 20 : 4~6 (Tables III and
UPTAKE
OF 22-CARBON
FATTY
ACIDS
IN RAT TISSUES
937
TABLE
acid composition 01 brain lipids
Fatty PC Fatty
V
PE
PS
SPH
co
RO
co
RO
co
RO
co
49.2 +0.7 0.2 f 0.0 12.0 f 0.2 21.1 +0.4 0.7 * 0.0 0.7*0.0 52 +@3 0.0 rt_0.0 0.5 * 0.0 2.9TO.l 04 & 04 7.4
49.6 k 0.8 0.2 + 0.0 12.0*0.2 21.0*0.4 0.5 +o.o 0.8 + 0.0 5.3 f0.1 o~o+o.o 0.5 f 0.0 2.9kO.l 0.0 * 0.0 7.3
5.6 + 0.2 14.5&0.3 20.9 +0.2 11.2+0.7 0.2 f 0.0 1.3 kO.2 12~5~08 0.0 IfI 0.0 4.6 k 0.2 18.9 kO.2 0.4 + 0.2 97
56 * 0.2 14.4f0.3 20.7f0.3 11.4kO.8 0.2 * 0.0 1.3 f0.2 12.3 +0.4 0.0 + 0.0 4.3 kO.1 18.611.1 0.7rto.3 10.4
1,9*0,1 0.4 + 0.2 45.7kO.3 13.7kl.l 0.1 + 0.0 1.6kO.2 2.S kO.1 0.3 * 0.1 3.0 * 0.2 26.9* 1.2 0.2 + 0.0 3.7
1.8 * 0.0 0.3 * 0.0 45.9-to.5 14.2 + 1.2 0.1 f0.1 1.4f0.2 2.5 kO.2 0.4 * 0.1 2.8 + 0.0 26.8kl.l 0.2 * 00 3.7
4.3 f 0.2 0.0 * 0.0 78.7k2.2 oa + 0.0 0~0 * 0.0 1.8kO.l 0.0 + 0.0 0.0 2 0.0 0.0 * 0.0 04 + 0.0 6.5+- 1.4 8.7
0.6 i 0.0
0.6 k 0.0
0.2 * 0.0
0.2 * 04
3.0 * 04
acid
RO
(mol%) 16:0 18:DMAs 18:O 1 8 : 1 w9 18:2w6 20: 1~9 20 : 4~6 22 : lw9 22 :4w6 22 : 6~3 24: lf09 Others X2,X,,*
1.9 f 0.1
2.OkO.l
4.0 * 0.0 ‘r 78.6 k 0.0 f 0.0 * 2.3 If 04 & 04 * 0.0 + 0.0 * 7.0 & 8.2
0.1 0.0 2.9 0.0 Owl1 0.8 04 1 04 04 0.0 1.6
3.4 2 0.6
Values are means of analyses from three animals k S.E.M. Abbreviations : PC, phosphatidylcholine, PE. phosphatidylethanolamine ; PS. phosphatidylserine: SPH. sphingolipids: CO. canola oil: RO. rapeseed oil. * Sum of molo/,of 20 carbon fatty acids (2O:O. 20: 1~9, 2O:lw7. 20:-.7 20: 309. 20: 3w,h. 20:4& 20:4w3, and 20: 5~31 divided by sum of mol% of 22 carbon fatty acids (22:O. 22:lwll. 22: 1~9. 22: 1~7. 77:4ok1. 22: jr&. 22:503. and 2L:hc31. TABLE
VI
4. Discussion
Fatty acid composition (moZ%) of tissue total lipids Adrenal Fatty
acid
16:0 18:O 18 : 1OJ9 18:2w6 18:303 20: 1~9 20 : 4~6 20: 5w3 22 :lw9 22 : 4w6 2 2 : 61,)3 24: 1~9 Others
gland
Liver
co*
RO;
co
RO
16.4kO.2 7.0 * 0.7 40.0f I.6 10.6 + 0.8 1.9 * 0.2 1.1 t-O.0 7.0-t I.0 0.3 +01 0.3 + 0.0 2.3 $0.4 0.9 -fi 0.2 0.2 + 0.0 11.8
13.7kO.3 6.5 kO.6 21.7kl.l 5.0 * 0.2 1.1 io.1 5.1* 0.3 8.7-t 1.3 0.5 f0.0 17.6 k 1.4 3.610.2 1.6$0.0 2.5kO.2 12.5
21.1 k 1.9 15.2+2.0 19.7f0.4 11.3kO.6 1.0+04 0.4kO.l 13.9-&o.]. 1.7kO.O O.O+O.O 0.0fO.O 5.7kO.8 0.3 rfr 0.0 9.7
2O.OIf: 1.7 17.9k2.1 21.1 kO.4 7.3 & 1.1 0.4+0.1 0.9 + 0.0 13.4kO.3 2.1 kO.1 0.6kO.O o.o*o.o 6.7kO.6 0.8 f 0.0 8.8
Values are means of analyses from three animals + S.E.M. * Canola oil. ; Rapeseed oil.
V). The CC&C,, ratio is also lower in brain than in plasma PL, except for the SPH fraction. SPH, a minor component compared to PC or PE, has a rather different composition and is enriched in 24: lo9 (6.5 and 7.0% in CO and RO groups, respectively). No difference in the fatty acid composition of the brain phospholipids was observed between the two diet groups. Fatty acid composition of liver and adrenal gland is given in Table VI. Though liver has only 06% of 22 : 109. the adrenal gland accumulates a high level of 22 : 1~9 ( 17.6 %) in RO-supplemented animals.
Previous studies (Wiegand et al., 199 I ) in our laboratory have shown that rats fed diets supplemented with linseed oil (53% 18:3w3, 20% 18:2w6), safflower oil ( < 0.1 y0 18: 3~3, 83% 18:2~,)6), or hydrogenated coconut oil (no 18 : 303 or 18 : 2~6) had elevated plasma levels of 20 : 5~3.20: 40)6, or 20: 3cti9, respectively. However, neither 20: 5~3 nor 20: 3(ti9 was incorporated into ROS phospholipids, and the level of 20:4w6 was not elevated in the ROS of the safflower oil group. The level of 22: 6(,)3 was high in the KOS of both the linseed oil (3 6 % in PC, 54 “/oin PE. 54 % in PS) and hydrogenated coconut oil (34 “/oin PC, 49 %,in PE, 5 1% in PS) groups, confirming numerous reports that the retina conserves this fatty acid during w3 fatty acid deficiency (Futterman et al., 19 71 ; Anderson and Maude, 1972 : Tinoco, 1982 ; Stinson. 1990; Wiegand et al.. 1991). In the safflower oil group, the level of 22 : 60 3 was decreased (2 5 % in PC, 34 % in PE. 3 3 % in PS), but the loss was compensated for by an increase in 22: 50)6 (Wiegand et al.. 1991). This indicates that the retina can selectively take up C,, PUFA from the plasma. In order to determine if this selective uptake is based on the number of carbons in the fatty acid alone, or if the number of double bonds is also an important determinant, we designed an experiment using 22: lw9 as a metabolic marker molecule. In the present study, we found that 22: l(f)9 was incorporated into the lipids of plasma, liver and adrenal gland in rats fed RO diet. The finding of lower levels of 22 :lo9 in liver than those in plasma is in accordance with a previous report (Walker. 19 72).
N WANG ET AL
Since dietary lipids are processed in the liver before being transported to other body organs, the liver may metabolize part of the dietary 22 : 1WY to prevent its accumulation. Chain shortening of 22 : 1019 has been demonstrated in rat liver (Ong, Bezard and Lecerf, 1Y 7i ; Christiansen. Christiansen and Bremer. 19 79 : Thomassen and Christiansen. 19 84 ; Holmer and Ronneberg. 19 86 : Ronneberg. Hcllmer and Lambertsen, 19 8 7), small intestine (Thomassen, Helgerud and Norum, 1985). heart (Pinson and Padieu. 1974). and cultured fibroblasts (Christensen, Hagve and Christophersen, 19 88 : Christensen et al., 19 89). with the predominant product being 18 : 1(!)9. Our finding that 18 : 1~~19levels were similar in plasma and liver between the two diet groups, although there was almost a four-fold difference in dietary levels ( 5 5.2 y0 in CO diet, 14.2 “/oin RO diet, Table I), suggests that some of the 18 : 109 in RO-fed rats may have been derived from 22 : 109. Chain elongation of 22 : 1~9 to 24: 109 has been shown in rat brain (Fulco and Mead, 1961: Kishimoto and Radin, 1963 ; Bourre. Daudu and Baumann, 19 76) and rat lung microsomes (Lecerf and Bodin, 1984). The higher level of 24: 109 in brain SPH. plasma and liver in RO group may also be derived from 22 : 109. although contribution from the diet content (0.8 % in RO group, 01 “/o in CO group) cannot be ruled out. Despite these possible retroconversions, there is still an ample supply of 22: 1~9 as compared to the level of 22 : 6w 3 in the circulating plasma. The plasma level of 22 : 6~3 in both dietary groups was comparable to the level of 22: 1~9 in the RO group. The accumulation of 22: 1~9 in the adrenal gland indicates that 22 : 109 in the plasma was taken up into tissues and incorporated into lipids. However, retinal ROS and brain, which have a high content of C,, PUFA, did not incorporate 22 : 109 into PL. These results suggest that, although retina and brain have mechanisms for the selective uptake of C,, PUFA. chain length alone is not sufficient for the selectivity shown by these two tissues. Degree of unsaturation and acyl chain length are both important determinants in the selective uptake mechanism. The retinal pigment epithelium (RPE), a layer of cuboidal epithelial cells, lies between the ROS and the choriocapillaris. the blood supply for the photoreceptor cells. The RPE cells form tight junctional complexes between each other and therefore block the direct access of blood components to photoreceptor cells. Thus, fatty acids as well as other components must traverse the RPE before they can reach the ROS. We do not know if the selective uptake of C,, PUFA into the retina is at the RPE-choriocapillaris boundary or the RPE-ROS boundary. Studies are currently underway in our laboratory to differentiate between these two possibilities. Acknowledgements This research was supported by grants from NIH/NEI
lEYO414Y. EYOOXTI, and l3’025.20). the Kl’ I:oundation Fighting Blindness. Kesearch to Prevent Blindness, Inc., and the Retina Kesearch Foundation. Kobert 1’. i\nderson is the recipient of a Senior Scientific Investigator Award from Kesearch to Prevent Blindness, Inc. We would like to thank Janice bt’ilson for preparing the manuscript.
References Alvarez, R. A. and Anderson. K. E. ( IYYO). Kat retina can elongate and desaturate 20 : 5~ 3 to 22 :ho~3-but why? Invest. Ophthalmal. Vis. Sci. 31, 579. Anderson, R. E. and Maude. M. B. (1972). Lipids of ocular tissues. VIII. The effects of essential fatty acid deficiency on the phospholipids of the photoreceptor membranes of rat retina. Arch. Biochem. Biophys. 151, 2 70-76. Anderson, R. E.. Maude. M. B. and Feldman, G. L. ( 1Y6Y i. Lipids of ocular tissues. 1. The phospholipids of mature rabbit and bovine lens. Biochin~. Bioplgs. Ada 187. 345-53.
Bligh, E. G. and Dyer. W. J. ( 1959). A rapid method of total lipid extraction and purification. Carl. 1. Hiochem. Physiol. 37. 91 1-l 7. Bourre. J.. Daudu. 0. and Baumann. N. (1976). Nervonic acid biosynthesis by erucyl-CoA elongation in normal and quaking mouse brain microsomes. Elongation of other unsaturated fatty acyl-CoAs (mono and polyunsaturated). Biochiv~. Biophys. Acts 424, l-i. Christensen. E.. Grann. M.. Hagve. T. A., Kase, B. F. and Christophersen. B. 0. ( 1989). Adrenoleukodystrophy. The chain shortening of erucic acid [22: l(n-Y)] and adrenic acid [22:4(n-6)] is deficient in neonatal adrenoleukodystrophy and normal in X-linked adrenoleukodystrophy skin fibroblast. Riochin~. Biop,phys.Acta 1002. 79-83. Christensen, E., Hagve. T. and Christophersen. B. 0. ( 1988). The Zellweger syndrome : deficient chain-shortening of erucic acid [22 : l(n-Y)] and adrenic acid [22 : 4(n-6)] in cultured skin fibroblast. Riochim. Biophys. Acta 959, 95-9. Christiansen, R. Z.. Christiansen. E. N. and Bremer. J. ( 1979). The stimulation of erucate metabolism in isolated rat hepatocytes by rapeseed oil and hydrogenated marine oil-containing diets. Biorhim. Biophys. Acta 573, 417-29. Fliesler. S. J. and Anderson, R. E. (198 3). Chemistry and metabolism of lipids in the vertebrate retina. In Progress in Lipid Research. Vol. 22. (Ed. Holman. R. T.). Pp. 79-l 31. Pergamon Press: London. Fulco. A. J. and Mead, J. F. (1961). The biosynthesis of lignoceric, cerebronic. and nervonic acids. I. Bid. Chem. 236, 231 h-20. Futterman. S.. Downer. J. 1~.and Hendrickson, A. (1971). Effect of essential fatty acid deficiency on the fatty acid composition, morphology, and electroretinographic response of the retina. Inwst. Ophthnhol. 10. 1 51-6. Hargrave. P. A. (1982). Rhodopsin chemistry. structure. and topography. In Progress in Retinal Research. Vol. 1. (Eds Osborne, N. and Chader. G.). Pp. l-52. Pergamon Press : Oxford. Helmer. G. and Rsnneberg. R. (1986). Influence of dietary fat on metabolism of (14-“Cl erucic acid in the perfused rat liver. Distribution of metabolites in lipid classes. Lipids 21. 39 j-400. Kishimoto. Y. and Radin. N. S. ( 1963). Biosynthesis of nervonic acid and its homologues from carboxyl-labeled oleic acid. 1. Lipid Res. 4. 444-i. Lamptey, M. S. and Walker, B. L. (1 Y 76). A possible essential role for dietary linolenic acid in the development of the young rat. 1. N&r. 106. 86-Y 3.
UPTAKE
OF 22-CARBON
FATTY
ACIDS
IN RAT TISSUES
Lecerf. J. and Bodin. J. ( 1984). Effects of high erucic acid diet on sphingomyelin biosynthesis in rat lung microsomes. 1. Physiol. 79, 345-51. Li. J.. Wetzel. M. G. and O’Brien, P. J. ( 199 1). Transport of o3 fatty acids from the intestine to the retina. Invest. Ophtl~ulrnol. I/is. Sci. 32, 701. Neuringer. M.. Connor, W. E.. Lin. D. S.. Barstad. L. and Luck. S. (1986). Biochemical and functional effects of prenatal and postnatal omega-3 fatty acid deficiency on retina and brain in rhesus monkey. Proc. ,N&l. .4c(rd.Sci. U.S.A. 83, 302 1-5. Neuringer. M.. Connor, W. E., Van Petten. C. and Barstad, L. ( 19841. Dietary omega-3 fatty acid deficiency and visual loss in infant rhesus monkey. I. Clirt. rtwst. 73, 272-6. Nilsson. A., Thomassen. M. S. and Christiansen, E. (1984). Long-chain acyl-CoA levels in liver from rats fed highfat diets: is it of significance for an increased peroxisomal /I-oxidation ? Lipids 19. 18 7-94. Ong. N.. Bezard. J. and Lecerf, J. (1977). Incorporation and metabolic conversion of erucic acid in various tissues of the rat in short-term experiments. Lipids 12. 563-9. Papermaster. D. S. and Dreyer, W. J. (1974). Rhodopsin content in the outer segment membranes of bovine and frog retinal rods. Riochetnistrg 13, 2438-44. Pinson. A. and Padieu, P. (1974). Erucic acid oxidation by beating heart cells in culture. FEBS Letts. 39. 88-W. Ronneberg. R., Holmer. G. and Lambertsen. G. (1987). Comparative metabolism of erucic and oleic acid in hepatocytes from rats fed partially hydrogenated marine oil or palm oil. Am. Nutr. Metah. 31, 160-Y. Scott, B. L. and Bazan. N. G. ( 1989). Membrane docosahexaenoate is supplied to the developing brain and retina by the liver. Proc. ,Natl. Arad. Sci. U.S.A. 86. 290 3--T.
939
Stinson. A. M. ( 1990). Mechanisms of Corrservatiorr of‘Docosahrxaenoic Acid in Rut Retinas during Essential Fatty Acid Deprivation. Ph.D. Thesis, Baylor College of Medicine, Houston. Stinson, A. M.. Wiegand, R. D. and Anderson, R. E. ( 199 I ). Fatty acid and molecular species compositions of phospholipids and diacylglycerols from rat retinal membranes. Erp. Eye Res. 52. 213-l 8. Strand, V. C. and Bazan, N. G. (1990). The developing mouse retina has the capacity for desaturation (A-4 desaturase) or retroconversion of eicosapentaenoic acid. Invrst. Ophthtrh?m/. Vis. Sci. 31, 579. Thomassen. M. S.. Helgerud. P. and Norum. I