Fish Physiology and Biochemistry vol, 1 no. l pp 27-35 (1986) Kugler Publications, AmsterdamtBerkeley
The effect of dietary iipids on the trout erythrocyte membrane C. Leray, G. Nonnotte, L. Nonnotte C N R S , B. P. 20 CR, 67037 Strasbourg-C~dex, France
Keywords: rainbow trout, erythrocyte, osmotic fragility, phospholipid, fatty acid, diet, morphology
Abstract
Rainbow trout were fed either a commercial diet or semi-purified diets containing 8% by weight of either cod liver oil, grape seed oil or hydrogenated coconut oil. Important changes in the fatty acid composition of erythrocyte membrane phospholipids were induced by these dietary fats. No changes were seen in the cholesterol/phospholipid ratio and in the cellular ATP level. Shape changes (crenation of cell margin and shrinkage) were only seen in trout fed hydrogenated coconut oil in connection with an accumulation of high amounts of (n-9) fatty acids including 20:3 (n-9) phospholipids. The compositional changes affect the stability of the erythrocyte membrane. An increased rate of osmotic hemolysis appeared to be associated with an increased unsaturated fatty acid content of the membranes.
Introduction
Extensive dietary studies have shown the essentiality and the requirement of (n-3) fatty acids by rainbow trout, and physiological changes induced by raising trout on (n-3) fatty acid deficient diet were described (Castell et al. 1972a, b). Numerous nutritional studies have revealed that these compounds have important functions in maintaining proper growth and healthy conditions (Yu and Sinnhuber t976; Watanabe 1982) and reproduction processes (Leray et al. 1985) but specific effects on membrane structure and function are poorly understood. Trout raised on (n-3) fatty acid deficient diet have been reported to develop mitochondrial swelling (Sinnhuber et al. 1972) and a decrease in ion permeability of intestinal brush border membrane (Di Costanzo et al. 1983). Recent reports suggest that the osmotic sensitivity of rat erythrocytes is related to their essential fat-
ty acid (Ehrstr6m et al. 1981) or cholesterol content (Bruckdorfer et al. 1969), and that shape changes in human erythrocytes are induced by exogenous lipids (Raz and Livne 1973; Rao et al. 1979; Lange et at. I980; Fujii et al. 1984) or alteration of phospholipid metabolism (Ferrell and Huestis 1984). Thus, it appears that erythrocytes represent a potentially useful and efficient model for the study of the correlations between defined changes in membrane composition, electron microscopic morphology and mechanical properties (Farnsworth et al. 1965; Araki and Rifkind 1981). Since previous studies in trout have shown that dietary manipulations modify the lipid composition of cellular membranes (Leray and Florentz 1983), we investigated the relationships between membrane lipids, osmotic fragility, rate of hemolysis and shape of erythrocytes from trout raised on diets supplemented with lipids of various compositions.
28 Material and methods
Animals Rainbow trout (Salmo gairdneri) with initial weight between 300 and 800 g were maintained in outdoor running water at 13-14°C. They were fed once a day (1070 live weight) for 6 months either a commercial diet (Kraft universal, diet K) or 3 different experimental diets (A, C and N) prepared according to CasteI1 et aL (1972b) and having different lipid moieties as fatty acid sources (Table 1). Blood was sampled by caudal puncture and the erythrocytes were washed two times by physiological saline. White blood cells were carefully removed after the last centrifugation.
Nucleotides analysis Perchloric acid extracts from erythrocyte suspensions were processed and analyzed by H P L C as previously described (Leray 1979).
Lipid analysis Erythrocytes were lysed by brief sonication (2 x I0 s.) in physiological saline. The membranes were separated by centrifugation and washed two times by additional NaC1 solution. Membrane lipids were extracted with 20 volumes of chloroform/methanol (2:1, v/v) and washed with 0.9% KCI according to Folch et al. (1957). Phospholipids were purified by TLC with the following solvent system, ether/methanol/acetic acid (90:2:1, v/v) and their fatty acid composition was analyzed by gas-liquid chromatography (glass wall-coated open-tubular column, 0.35 m m i d x 50 m, Carbowax 20M - Perkin Elmer Sigma 1 gas chromatograph with an in-line Sigma 10 chart integrator). The fatty acid composiTable 1. Fatty acid composition of the dietary oils Fatty acid
Sat urated n-9 n-7 n-6 n-3
tion is reported as molar percentage o f butyl esters. The standard nomenclature of fatty acids was used i.e. 22:6 ( n - 3 ) means 22 carbons and 6 double bonds which are specified from the methyl and, the first one being after the 3rd carbon ( n - 3 ) . Due to an unknown interference, cholesterol was determined after purification from neutral lipids by T L C with the solvent system hexane/ether/acetic acid (90:30:1, v/v) (Kates 1972). Total phospholipid fraction was quantified by phosphorus determination according to Bartlett (1959) and protein amount was estimated according to Lowry et al. (1951).
Diet K
A
C
N
34.0 21.1 5.6 23.6 12
4.6 52.5 0 0.6 29.2
12,4 17.8 0,6 67.2 0
100 0 0 0 0
K, commercial diet; A, cod liver oil; G, grapeseed oil; N, hydrogenated coconut oil. Data are expressed in mole °7o. 1107o fat are provided in diet K and 8% for diets A, C and N.
Osmotic sensitivity tests The osmotic fragility was determined as described by Ehrstr6m et al. (1981) from the degree of hemolysis measured 30 minutes after mixing blood samples to a dilution series of buffered saline. The time course of hemolysis was measured in hypotonic solution (50 mM NaC1, 10 mM Na2HPO4, pH 7.4) at 20°C. A Hamilton syringe with a 30 g brass cylinder added on the piston end was used to inject vertically 50 #1 of an isotonic erythrocyte suspension (2°70 hematocrit) in a well stirred tbermostated cuvette containing 2 ml of hypotonic solution. The change in turbidity at 690 nm was recorded on a Kontron Uvikon 820 spectrophotometer. The linear slope observed between 2 and 5 sec. after the beginning of the mixing (about 20% of the decrease in absorbance occurs during that time) was used to calculate the rate of hemolysis. This rate was expressed in percent of initial absorbance decrease per second.
Scanning electron microscopy An erythrocyte suspension (0.5% hematocrit) was made in a cold buffered saline (10raM Na2HPO4, pH 7,4) containing 1% glutaraldehyde. After l h at
29 erythrocytes were washed three times with phosphate buffer and finally fixed with 1% osmic acid in the same buffer. Fixed cells were dehydrated sequentially with ethanol and propylene oxide, spread over a cover-glass and air dried. After coating with gold, cells were examined in a Cambridge stereoscan 100 scanning electron microscope.
Results
The hemolytic properties of animals fed control diet (A), commercial diet (K), (n-3) deficient diet (C) and unsaturated fatty acid free diet (N) are presented in Table 2. Osmotic fragility test was processed only for K and C groups and osmotic hemolysis rate was determined for the four groups. Erythrocytes from animals fed balanced diets (A, K) showed a significant increase of resistance to osmotic lysis (fragility test or hemolysis rate) when compared to the two experimental diets (C, N). The cholesterol/phospholipid molar ratios in the erythrocytes of the three experimental groups (Table 3) were not significantly different from that measured in erythrocytes of the control group (Diet
Table 2. The effect of dietary fats on osmotic fragility and hemolysis rate of trout erythrocytes. Diet
Osmotic fragility
Osmotic hemolysis rate
K A C N
22.04 _+0.38 -26.31 ± 1.64" --
5.58 _+0.96 6.99 _+0.55 9.86_+ 1.08" 11.33-+0.21"*
Osmotic fragility expressed as °70 saline producing 50°70 hemolysis, osmotic hemolysis rate expressed as 070 decrease of initial absorbance per sec. Mean value _+ S.D. of four samples. K, commercial diet; A, cod liver oil; C, grapeseed oil; N, hydrogenated coconut oil, * P < 0.05, ** P < 0.01 as compared to group K (control group). - : no observations.
Table 3. The effect of dietary fats on the cholesterol/phospholipid ratio of trout erythrocytes Diet
CHOL/PL
K A C N
0.60 _+0.08 0.67 -4-0.07 0.42 _+0.05 0.41 _+0.06
Data are expressed on a molar basis as the mean ± S.E. for four samples. K, commercial diet; A, cod liver oil; C, grapeseed oil; N, hydrogenated coconut oil; CHOL, cholesterol; PL, phospholipid.
Table 4. The effect of dietary fats on fatty acid composition of trout erythrocytes Fatty acid
14:0 16:0 18:0 18:1 20:1 20:3 22:1 18:2 20:4 22:5 18:3 20:5 22:6
(n-9) (n-9) (n-9) (n-9) (n-6) (n-6) (n-6) (n-3) (n-3) (n-3)
Diet K
A
C
N
1.90 __.0.33 39.29 _+3.7 10.35 + 3.62 9.97 _+0.73 1.56_+0.35 -5.64 + 0.86 3.57 _+0.32 6.64± 1.01 1.03 -+0.50 -1.72±0.50 10.72 -+ 1.71
2.98 + 0.17* 34.45 ± 2.74 10.64 +__2.94 3.03 +_ 1.44'* 1.68_+0.12 -0.62 _+0.35** -4.40_+0.33 -2.03 _+0.34 5.31 -+0.90* 22.43 _+0.67***
1.40 +_0,07 30.33 _+2.50 8.21 +__0.09 7.73 _+0.54 ---9.35 _+0.86"** 15.55 _+ 1.18"* 8.41 _+0.36*** --8.18 _+0.92
3.15 ± 0.30* 33.74+ 1.77 6.03 _+0.27 16.93 ± 0.88"** 1.10_+0.10 4.61 ±0.60 0.76 _+0.16"* 1.30 -+ 0.07*** 6.55 _+0.35 1.08 +0.10 -2.14_+0.17 20.52 +_ 1.33**
The data are expressed as the mean mola~ percent of total _ S.E. for four samples. K, commercial diet; A, cod liver oil; C, grapeseed oil; N, hydrogenated coconut oil. * p < 0.05, ** p < 0.01, *** p < 0.00t as compared to group K (control group). - : value lower than 0.3%.
30 Table 5. The effect of dietary fats on the parameters of the phospholipid acyl chains of trout erythrocytes
Fatty acid
Diet K
A
C
N
sat. (n-9) (n-7) (n-6) (n-3)
53.62 -+3.05 17.69 -+1.72 2.04+_0.28 12.55 _-+4. t4 11.86_+0.69
52.02 _+5.50 5.33 +__1.40** 10.00_+ 1.72"* 5.38 -+0.49 29.77 _+1.42"**
39.94 _+2.50* 11.82 _+1.00" 1.31 _+0.t3 38.27 _+1.90"* 8.66_+ 1.07"
42.91 -+0.92* 25.00 _+0.50** -9.29 _+0.63 23.00+_ 1.28"**
unsat./sat.
0.88 +_O.12
0,97 -+O.19
1.53 -+O.14*
1.33 _+0.05*
DB index
143 + 12
201 + I0"*
201 +_10"*
205 + 6**
The fatty acid data are expressed as the mean molar percent of total _+ S.E. for four samples. K, commercial diet; A, cod liver oil; C, grapeseed oil; N, hydrogenated coconut oil. DB index: number of double bonds per 100 moles fatty acid. * p < 0.05, ** p < 0.0I, • ** p < 0.001 as compared to group K (control group).
K). The fatty acid c o m p o s i t i o n of erythrocyte membranes o f the four groups are presented in Table 4 and the various indices are given in T a b l e 5. The erythrocytes of A-fed trout had the highest ( n - 3 ) fatty acid level ( m a i n l y as 22:6) a n d the lowest ( n - 6 ) fatty acid level. The reverse c o m p o s i t i o n was observed in C-fed trout, the level of the ( n - 6 ) fatty acids being the highest and c o m p o s e d of 18:2, 20:4 and 22:5. The erythrocytes of N-fed trout were characterized by a high level of ( n - 9 ) fatty acids (mainly as 18:1) and by the presence of the u n u s u a l 20:3 ( n - 9 ) . The erythrocytes of K-fed trout had a high ( n - 9 ) fatty acid level a n d their ( n - 6 ) a n d ( n - 3 ) fatty acid c o n t e n t s had i n t e r m e d i a t e values between those o f A- a n d C-fed trout. The u n s a t u r a ted/s~iturated ratio was higher in the erythrocytes o f C- a n d N-fed t r o u t t h a n in A- a n d K-fed trout while the d o u b l e b o n d index was the same in erythrocytes o f A-, C- a n d N-fed trout a n d lower in those o f K-fed trout.
Table 6. The effect of dietary fats on nucleotide contents of trout erythrocytes
Diet
ATP
ATP/ADP
K A C N
2039+ 168 2017 _+66 2233 _+94 1941 _+122
4.96+_0.30 2.77 _+0.15*** 3.64 _+0.08*** 3.73_+0.43
ATP contents are expressed in nmole per ml packed cells. Mean value _+ S.D. of four samples. K, commercial diet; A, cod liver oil; C, grapeseed oil; N, hydrogenated coconut oil. ** p < 0.01, *** p < 0.001 as compared to group K (control group).
throcytes from N-fed trout had altered shapes (Fig. 3 - 5 ) , 4 2 % having crenated m a r g i n s (discoechinocytes) a n d 30O/o being characterized by a s h r u n k e n appearance. The A T P content and A T P / A D P
ratio o f ery-
throcytes from A-, K-, C- a n d N-fed t r o u t are given
S c a n n i n g electron micrographs o f fresh erythrocytes from K- a n d N-fed t r o u t are presented in Fig.
in T a b l e 6. No difference in the A T P c o n t e n t s could be observed between the four g r o u p s but differ-
1 - 5 . Erythrocytes from A- or K-fed trout had the same m o r p h o l o g y a n d only o n e representative sample is presented (Fig. 1 - 2 , K-fed trout). They can be considered the typical teleost discocytes. Erythrocytes from C-fed trout (not shown
ences were observed at the level o f A T P / A D P ratios. Erythrocytes from A-fed trout had the
had little more a p p a r e n t nuclei a n d slightly swollen margins when c o m p a r e d to control discocytes. Ery-
lowest A T P / A D P ratio, erythrocytes from K-fed trout having the highest value a n d those from Ca n d N-fed trout having i n t e r m e d i a t e values.
31
Figs. 1-5. The effect of dietary fats on the morphology of trout erythrocytes. Fig. 1. Normal discocytes, K diet ( x 2600). Fig. 2. Normal discocytes, K diet (x 5000). Fig. 3. Altered shapes with N-diet (x 2600).
Fig. 4. Shrunken erythrocyte, N-diet (× 6000). Fig. 5. Discoechinocyte, N-diet (x 6000).
32 Discussion
The results of the present study show that a six month feeding of various oils is able to profundly alter the lipid composition of trout erythrocyte membrane. This is in contrast with results from mammalian erythrocytes where only minor changes were detected in the membrane fatty acid pattern (Rao et al. 1979; Vajreswari et al. 1983; Benga et aL 1984) even after feeding animals for several generations (Ehrstr6m et al. 1981). Concurrently, the basic features of the fatty acid composition are largely modified by the diet as regards either the molar percentage of saturated fatty acids or the average unsaturation index. It is noteworthy that the lipid compositional changes induced by essential fatty acid deficient diet at the level of trout erythrocytes are similar to those we have reported for trout intestinal mucosa (Di Costanzo et al. 1983; Leray and Florentz 1983). The analysis of the influence of fully saturated dietary fats in the present work suggests that in addition to the appearance of eicosatrienoic acid (20:3, n - 9 ) which is characteristic of this deficiency state (Farnsworth et al. 1965; Walker et al. 1967) there is a marked incorporation of essential ( n - 3 ) fatty acids in membrane phospholipids. This specific molecular regulation which must involve transport from other tissues deserves further investigations. As it was reported in rat after dietary manipulations (Ehrstr6m et al. 1981) the cholesterol/phospholipid ratio of trout erythrocytes is not significantly affected by the quality of ingested fats. lri a number of investigations there have been attempts to correlate the osmotic fragility of erythrocytes to their lipid and fatty acid composition (Bruckdorfer et aL 1969; Raz and Livne 1973; Lange et al. 1980; Benga et al. 1984; Kuypers et al. 1984). As a result, the stabilization of erythrocyte membrane has been generally attributed to increased cholesterol/phospholipid ratio and increased unsaturated fatty acids. In the present investigations, however, the effect of cholesterol levels on the differences observed in the rate of osmotic hemolysis may be discounted since the cholesterol/ phospholipid ratio was practically constant for all four groups of animals. An important factor in
determining the stability of the erythrocytes membrane is that of its fatty acid composition, in particular the level of unsaturated fatty acids. The abundance of unsaturated fatty acids (Raz and Livne 1973; Shand and Noble 1981; Vajreswari et al. 1983) was shown to be related to a greater extent of membrane expansion when cells are exposed to hypotonic solutions. This increase in membrane surface allows an increase in cellular critical volume which is positively correlated with the osmotic resistance (Anderson and Lovrien 1977) when the resulting hemolysis is studied as a function of salt concentration. On the other hand, the rate of osmotic hemolysis measures the cell rupture process which follows the swelling process and thus may be independent of the osmotic fragility (Araki and Rifkind 1981). This last biophysical study has indicated that the rupturing rate of the erythrocyte membrane is closely associated with the fluidity of the lipid bilayer and thus affords estimation of intermolecular interactions within the bilayer. While these relationships were demonstrated in altering membrane fluidity by treatment with lipidsoluble compounds (Araki and Rifkind 1981), similar results can be expected when the flexibility of the hydrocarbon chains forming phospholipid molecules are experimentally or naturally modified. From a large number of investigations using several technical approaches it has been stated that both the length of the chains and the number of double bonds per chain are indicators of membrane fluidity (Deuticke 1977; Brenner 1984). In contrast to the above-mentioned studies, the comparisons of phospholipid fatty acid compositions of the erythrocytes from the four groups of trout in the present investigations demonstrate that the level of unsaturated fatty acids and the degree of unsaturation are positively correlated with the rates of osmotic hemolysis (rate of osmotic hemolysis = 3.81 + 3.92 x (unsat./sat.), r = 0.523, n = 16, P = 0.036). The analysis of results on osmotic fragility determined in two groups (diets K and C) allowed similar conclusions. Furthermore, changes in the fatty acid pattern of trout erythrocyte lipids do not suggest relationships between any specific constituents and osmotic fragility. Thus, no direct in-
33 fluence on the cell rupture process can be inferred from the relative amount of essential (n-3), nonessentiel (n-6) or non-essentiel (n-9) fatty acids as it was suggested by Ehrstr6m et al. (1981) and Benga et al. (1984). Our experimental results in trout erythrocytes emphasize the importance of the specific effects on membrane structure and function caused by changes in fatty acid composition. These changes resulting from differences in dietary composition are of particular interest when compared with the rather limited alterations described in mammalian erythrocytes (Ehrstr6m et al. 198 I). The hypothesis of a close association between the lipid unsaturation and the erythrocyte rupture process seems to be supported by recent results on the importance of the degree of unsaturation of phosphatidylcholine in human erythrocytes (Kuypers et al. 1984). Thus, the replacement of native phosphatidylcholine by more unsaturated species has been found to modify considerably the membrane properties resulting in a progressive increase in osmotic fragility and K + permeability. It is noticeable that in this last study the osmotic fragility was measured as the degree of hemolysis of erythrocytes in a continuously decreasing salt concentration, approach which is not very different from that used in our study and emphasizes the importance of dynamic tests in the appreciation of membrane perturbations. It has been shown that erythrocyte shape is controlled by several and frequently related biochemical mechanisms. Since the early observation of echinocytosis (called also crenation) after ATP depletion (Nakao et al. 1960), several workers have proposed that this metabolic control is exerted either by the phosphorylation of spectrin (Birchmeier and Singer 1977) or by lipid bilayer imbalance secondary to phosphoinositide metabolism (Ferrell and Huestis 1984). Various amphipathic molecules are able to induce a spiny shape in red cells similar to that produced by ATP depletion (Deuticke 1968; Fujii et al. 1984). All these morphological changes have been convincingly explained by differential variations in the relative areas of the plasma membrane monolayers (Sheetz and Singer 1974; Lange and Slayton 1982). Our observation of a discoechinocytosis and
shrinkage in erythrocytes from trout raised on saturated lipid rich-diet without changes in ATP level cannot be presently interpreted in term of membrane structural organization until further investigations have been carried out. No explanation can be given for the observed variations of the ATP/ADP ratios in the erythrocytes of the A- and C-fed trout when compared to the control group (K-fed trout) and no correlation with shape changes can be established. In spite of the complexity of these phenomena, the association of structural changes with the presence of large amounts of (n-9) fatty acids (mainly 18:l and 20:3) suggests the hypothesis of specific accumulation of these compounds in the outer membrane leaflet, expanding its area by local increase of unsaturation and thereby producing the cr.enated margin and the skrunken appearance. The study of the distribution of phospholipids in the two halves of the membrane bilayer would help to explain the shape changes induced in the circulating blood red cells of N-fed trout. In conclusion, the results of the present study show the clear influence of the dietary lipids on the biochemical composition of trout erythrocyte membrane. The analysis of the correlations between the different fatty acid patterns and the osmotic sensitivities suggests that lipid unsaturation is the main modifier of the cell membrane properties. While the mechanisms involved in these interactions are presently unknown, our results do demonstrate the importance of proper fatty acid feeding in maintaining the normal physiological functions and discoid shape of trout red cells. Thus, it appears that the system here reported is potentially useful since characteristic chemical changes can be produced with associated modifications in the morphology and physical state of the erythrocyte membrane.
Acknowledgements We thank Mrs G. Gutbier and Mr. J.C. Barthe for their technical assistance.
34
References cited Anderson, P.C. and Lovrien, R.E. 1977. Human red cell hemolysis rates in the subsecond to seconds range. An analysis, Biophys. J. 20: 181-191. Araki, K. and Rifkind, J.M. 1981. The rate of osmotic hemolysis. A relationship with membrane bilayer fluidity. Biochim. Biophys. Acta 645: 81-90. Bartlett, G.R. 1959. Phosphorus assay in column chromatography. J. Biol. Chem. 234: 466-468. Benga, G., Travis, B.D., Pop, V.I., Popescu, O., Toader, S. and Holmes, R.P. 1984. The effect of the saturation and isomerization of dietary fatty acids on the osmotic fragility and water diffusional permeability of rat erythrocytes. Biochim. Biophys. Acta 775: 255-259. Birchmeier, W. and Singer, S.J. 1977. On the mechanism of ATP-induced shape changes in human erythrocyte membranes. I1. The role of ATP. J. Cell. Biol. 73: 647-659. Brenner, R.R. 1984. Effect of unsaturated acids on membrane structure and enzyme kinetics. Prog. Lipid Res. 23: 69-96. Bruckdorfer, K.R., Demel, R.A., De Gier, J. and Van Deenen, L.L.M. 1969. The effect of partial replacements of membrane cholesterol by other steroids on the osmotic fragility and glycerol permeability of erythrocytes. Biochim. Biophys. Acta 183: 334-345. Castell, J.R., Sinnhuber, R.O., Lee, D.J. and Wales, J.H. 1972a. Essential fatty acids (EFA) in the diet of rainbow trout: physiological symptoms of EFA deficiency. J. Nutr. 102: 87-92. Castell, J.R., Sinnhuber, R.O., Wales, J.H. and Lee, D.J. 1972b. Essential fatty acids in the diet of rainbow trout (Salmo gairdneri): growth, feed conversion and some gross deficiency symptoms. J. Nutr. 102: 77-85. Deuticke, B. 1968. Transformation and restoration of biconcave shape of human erythrocytes induced by amphiphilic agents and changes of ionic environment. Biochim. Biophys. Acta 163: 494-500. Deuticke, B. t977. Properties and structural basis of simple diffusion pathways in the erythrocytes membrane. Rev. Physiol. Biochem. Pharmacol, 78: 1-97. Di Costanzo, G., Duportail, G., Florentz, A. and Leray, C. 1983. The brush border membrane of trout intestine: influence of its lipid composition on ion permeability, enzyme activity and membrane fluidity. Mol. Physiol. 4: 279-290. Ehrstr6m, M., Harms-Ringdahl, M. and Ailing, C. 1981. Osmotic fragility and fluidity of erythrocyte membranes from rats raised on an essential fatty acid deficient diet. Biochim. Biophys Acta 644: 175-182. Farnsworth, P., Danon, D. and Gellhorn, A. 1965. The effect of fat-free diet on the rat erythrocyte membrane. Brit. J. Haemat. 11 : 200-205. Ferrell, J.E. and Huestis, W.H. 1984. Phosphoinositide metabolism and the morphology of human erythrocytes. J. Cell Biol. 98: 1992-1998. Folch, J., Lees, M. and Sloane-Stanley, G.N. 1957. A simple method for the isolation and purification of total lipids from
animal tissues. J. Biol. Chem. 226: 497-509. Fijii, T., Tamura, A., Tanaka, T., Nagasawa-Fujimori, H. and Hiromi, K. 1984. Light scattering stopped-flow studies on the time course of shape changes induced in human erythrocytes by several amphiphilic lipids. Biochem. lnternat. 8: 83-88. Kates, M. 1972. Techniques of Lipidology. North-Holland Publishing Company, Amsterdam. Kuypers, F.A., Roelofsen, B., Op den Kamp, J.A.F. and Van Deenen, L.L.M. 1984. The membrane of intact human erythrocytes tolerates only limited changes in the fatty acid composition of its phosphatidylcholine. Biochim. Biophys. Acta 769: 337-347. Lange, Y. and Slayton, J. 1982. Interaction of cholesterol and lysophosphatidylcholine in determining red cell shape. J. Lipid Res. 23: 1121-1127. Langue, L.G., Van Meer, G., Op den Kamp, J.A.P. and Van Deenen, L.L.M. 1980. Hemolysis of rat erythrocytes by replacement of the natural phosphatidylcholine by various phosphatidylcholines. Eur. J. Biochem. 110:115-121. Leray, C. 1979. Patterns of purine nucleotides in fish erythrocytes. Comp. Biochem. Physiol. 64B: 77-82. Leray, C. and Florentz, A. 1983. Biochemical adaptation of trout intestine related to its ion transport properties. Influences of dietary salt and fatty acids, and environmental salinity. In Intestinal Transport, pp. 354-368. Edited by M. Gilles-Baillien and G. Gilles. Springer-Verlag, Berlin. Leray, C. Nonnotte, G., Roubaud, P. and Leger, C. 1985. Incidence of (n-3) essential fatty acid deficiency on trout reproduction processes. Reprod. Nutr. Dev. 25: 561-581. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randal, R.J. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265-275. Nakao, M., Nakao, T. and Yamazoe, S. 1960. Adenosine triphosphate and maintenance of shape of human red cells. Nature, Lond. 187: 945-946. Rao, G.A., Siler, K. and Larkin, E.C. 1979. Diet-induced alterations in the discoid shape and phospholipid fatty acid compositions of rat erythrocytes. Lipids 14: 30-38. Raz, A. and Livne, A. 1973. Differential effects of lipids on the osmotic fragility of erythrocytes. Biochim. Biophys. Acta 311: 222-229. Shand, J.H. and Noble, R.C. 1981. The maternal diet and its effect on the lipid composition and osmotic fragility of neonatal ovine erythrocytes. Biol. Neonate 40: 150-159. Sheetz, M.P. and Singer, S.J, 1974. Biological membranes as bilayer couples. A mechanism of drug-erythrocyte interactions. Proc. Natl. Acad. Sci. 71: 4457-4461. Sinnhuber, R.O., Castelt, J.D. and Lee, D.J. 1972. Essential fatty acid requirement of the rainbow trout, Satrno gairdneri. Fed. Proc. 31: 1436-t441. Vajreswari, A., Rao, P.S., Kaplay, S.S. and Tulpule, P.G. 1983. Erythrocyte membrane in rats fed high erucid acidcontaining mustard oil: osmotic fragility, lipid composition, and (Na +, K + )- and (Ca 2+ , Mg2÷ )-ATPases. Biochem. Med. 29: 74-84. Walker, B.L. and Yurkowski, M, 1967. Effect of cell age on
35 erythrocyte fatty acid composition in rats on different dietary regimes. Biochem. J. 103: 218-224. Watanabe, T. 1982. Lipid nutrition in fish. Comp. Biochem. Physiol. 73B: 3-15.
Yu, T.C. and Sinnhuber, R.O. 1976. Growth response of rainbow trout (Salmo gairdneri) to dietary w3 and ~6 fatty acids. Aquaculture 8: 309-317.
The effect of dietary iipids on the trout erythrocyte membrane C. Leray, G. Nonnotte, L. Nonnotte C N R S , B. P. 20 CR, 67037 Strasbourg-C~dex, France
Keywords: rainbow trout, erythrocyte, osmotic fragility, phospholipid, fatty acid, diet, morphology
Abstract
Rainbow trout were fed either a commercial diet or semi-purified diets containing 8% by weight of either cod liver oil, grape seed oil or hydrogenated coconut oil. Important changes in the fatty acid composition of erythrocyte membrane phospholipids were induced by these dietary fats. No changes were seen in the cholesterol/phospholipid ratio and in the cellular ATP level. Shape changes (crenation of cell margin and shrinkage) were only seen in trout fed hydrogenated coconut oil in connection with an accumulation of high amounts of (n-9) fatty acids including 20:3 (n-9) phospholipids. The compositional changes affect the stability of the erythrocyte membrane. An increased rate of osmotic hemolysis appeared to be associated with an increased unsaturated fatty acid content of the membranes.
Introduction
Extensive dietary studies have shown the essentiality and the requirement of (n-3) fatty acids by rainbow trout, and physiological changes induced by raising trout on (n-3) fatty acid deficient diet were described (Castell et al. 1972a, b). Numerous nutritional studies have revealed that these compounds have important functions in maintaining proper growth and healthy conditions (Yu and Sinnhuber t976; Watanabe 1982) and reproduction processes (Leray et al. 1985) but specific effects on membrane structure and function are poorly understood. Trout raised on (n-3) fatty acid deficient diet have been reported to develop mitochondrial swelling (Sinnhuber et al. 1972) and a decrease in ion permeability of intestinal brush border membrane (Di Costanzo et al. 1983). Recent reports suggest that the osmotic sensitivity of rat erythrocytes is related to their essential fat-
ty acid (Ehrstr6m et al. 1981) or cholesterol content (Bruckdorfer et al. 1969), and that shape changes in human erythrocytes are induced by exogenous lipids (Raz and Livne 1973; Rao et al. 1979; Lange et at. I980; Fujii et al. 1984) or alteration of phospholipid metabolism (Ferrell and Huestis 1984). Thus, it appears that erythrocytes represent a potentially useful and efficient model for the study of the correlations between defined changes in membrane composition, electron microscopic morphology and mechanical properties (Farnsworth et al. 1965; Araki and Rifkind 1981). Since previous studies in trout have shown that dietary manipulations modify the lipid composition of cellular membranes (Leray and Florentz 1983), we investigated the relationships between membrane lipids, osmotic fragility, rate of hemolysis and shape of erythrocytes from trout raised on diets supplemented with lipids of various compositions.
28 Material and methods
Animals Rainbow trout (Salmo gairdneri) with initial weight between 300 and 800 g were maintained in outdoor running water at 13-14°C. They were fed once a day (1070 live weight) for 6 months either a commercial diet (Kraft universal, diet K) or 3 different experimental diets (A, C and N) prepared according to CasteI1 et aL (1972b) and having different lipid moieties as fatty acid sources (Table 1). Blood was sampled by caudal puncture and the erythrocytes were washed two times by physiological saline. White blood cells were carefully removed after the last centrifugation.
Nucleotides analysis Perchloric acid extracts from erythrocyte suspensions were processed and analyzed by H P L C as previously described (Leray 1979).
Lipid analysis Erythrocytes were lysed by brief sonication (2 x I0 s.) in physiological saline. The membranes were separated by centrifugation and washed two times by additional NaC1 solution. Membrane lipids were extracted with 20 volumes of chloroform/methanol (2:1, v/v) and washed with 0.9% KCI according to Folch et al. (1957). Phospholipids were purified by TLC with the following solvent system, ether/methanol/acetic acid (90:2:1, v/v) and their fatty acid composition was analyzed by gas-liquid chromatography (glass wall-coated open-tubular column, 0.35 m m i d x 50 m, Carbowax 20M - Perkin Elmer Sigma 1 gas chromatograph with an in-line Sigma 10 chart integrator). The fatty acid composiTable 1. Fatty acid composition of the dietary oils Fatty acid
Sat urated n-9 n-7 n-6 n-3
tion is reported as molar percentage o f butyl esters. The standard nomenclature of fatty acids was used i.e. 22:6 ( n - 3 ) means 22 carbons and 6 double bonds which are specified from the methyl and, the first one being after the 3rd carbon ( n - 3 ) . Due to an unknown interference, cholesterol was determined after purification from neutral lipids by T L C with the solvent system hexane/ether/acetic acid (90:30:1, v/v) (Kates 1972). Total phospholipid fraction was quantified by phosphorus determination according to Bartlett (1959) and protein amount was estimated according to Lowry et al. (1951).
Diet K
A
C
N
34.0 21.1 5.6 23.6 12
4.6 52.5 0 0.6 29.2
12,4 17.8 0,6 67.2 0
100 0 0 0 0
K, commercial diet; A, cod liver oil; G, grapeseed oil; N, hydrogenated coconut oil. Data are expressed in mole °7o. 1107o fat are provided in diet K and 8% for diets A, C and N.
Osmotic sensitivity tests The osmotic fragility was determined as described by Ehrstr6m et al. (1981) from the degree of hemolysis measured 30 minutes after mixing blood samples to a dilution series of buffered saline. The time course of hemolysis was measured in hypotonic solution (50 mM NaC1, 10 mM Na2HPO4, pH 7.4) at 20°C. A Hamilton syringe with a 30 g brass cylinder added on the piston end was used to inject vertically 50 #1 of an isotonic erythrocyte suspension (2°70 hematocrit) in a well stirred tbermostated cuvette containing 2 ml of hypotonic solution. The change in turbidity at 690 nm was recorded on a Kontron Uvikon 820 spectrophotometer. The linear slope observed between 2 and 5 sec. after the beginning of the mixing (about 20% of the decrease in absorbance occurs during that time) was used to calculate the rate of hemolysis. This rate was expressed in percent of initial absorbance decrease per second.
Scanning electron microscopy An erythrocyte suspension (0.5% hematocrit) was made in a cold buffered saline (10raM Na2HPO4, pH 7,4) containing 1% glutaraldehyde. After l h at
29 erythrocytes were washed three times with phosphate buffer and finally fixed with 1% osmic acid in the same buffer. Fixed cells were dehydrated sequentially with ethanol and propylene oxide, spread over a cover-glass and air dried. After coating with gold, cells were examined in a Cambridge stereoscan 100 scanning electron microscope.
Results
The hemolytic properties of animals fed control diet (A), commercial diet (K), (n-3) deficient diet (C) and unsaturated fatty acid free diet (N) are presented in Table 2. Osmotic fragility test was processed only for K and C groups and osmotic hemolysis rate was determined for the four groups. Erythrocytes from animals fed balanced diets (A, K) showed a significant increase of resistance to osmotic lysis (fragility test or hemolysis rate) when compared to the two experimental diets (C, N). The cholesterol/phospholipid molar ratios in the erythrocytes of the three experimental groups (Table 3) were not significantly different from that measured in erythrocytes of the control group (Diet
Table 2. The effect of dietary fats on osmotic fragility and hemolysis rate of trout erythrocytes. Diet
Osmotic fragility
Osmotic hemolysis rate
K A C N
22.04 _+0.38 -26.31 ± 1.64" --
5.58 _+0.96 6.99 _+0.55 9.86_+ 1.08" 11.33-+0.21"*
Osmotic fragility expressed as °70 saline producing 50°70 hemolysis, osmotic hemolysis rate expressed as 070 decrease of initial absorbance per sec. Mean value _+ S.D. of four samples. K, commercial diet; A, cod liver oil; C, grapeseed oil; N, hydrogenated coconut oil, * P < 0.05, ** P < 0.01 as compared to group K (control group). - : no observations.
Table 3. The effect of dietary fats on the cholesterol/phospholipid ratio of trout erythrocytes Diet
CHOL/PL
K A C N
0.60 _+0.08 0.67 -4-0.07 0.42 _+0.05 0.41 _+0.06
Data are expressed on a molar basis as the mean ± S.E. for four samples. K, commercial diet; A, cod liver oil; C, grapeseed oil; N, hydrogenated coconut oil; CHOL, cholesterol; PL, phospholipid.
Table 4. The effect of dietary fats on fatty acid composition of trout erythrocytes Fatty acid
14:0 16:0 18:0 18:1 20:1 20:3 22:1 18:2 20:4 22:5 18:3 20:5 22:6
(n-9) (n-9) (n-9) (n-9) (n-6) (n-6) (n-6) (n-3) (n-3) (n-3)
Diet K
A
C
N
1.90 __.0.33 39.29 _+3.7 10.35 + 3.62 9.97 _+0.73 1.56_+0.35 -5.64 + 0.86 3.57 _+0.32 6.64± 1.01 1.03 -+0.50 -1.72±0.50 10.72 -+ 1.71
2.98 + 0.17* 34.45 ± 2.74 10.64 +__2.94 3.03 +_ 1.44'* 1.68_+0.12 -0.62 _+0.35** -4.40_+0.33 -2.03 _+0.34 5.31 -+0.90* 22.43 _+0.67***
1.40 +_0,07 30.33 _+2.50 8.21 +__0.09 7.73 _+0.54 ---9.35 _+0.86"** 15.55 _+ 1.18"* 8.41 _+0.36*** --8.18 _+0.92
3.15 ± 0.30* 33.74+ 1.77 6.03 _+0.27 16.93 ± 0.88"** 1.10_+0.10 4.61 ±0.60 0.76 _+0.16"* 1.30 -+ 0.07*** 6.55 _+0.35 1.08 +0.10 -2.14_+0.17 20.52 +_ 1.33**
The data are expressed as the mean mola~ percent of total _ S.E. for four samples. K, commercial diet; A, cod liver oil; C, grapeseed oil; N, hydrogenated coconut oil. * p < 0.05, ** p < 0.01, *** p < 0.00t as compared to group K (control group). - : value lower than 0.3%.
30 Table 5. The effect of dietary fats on the parameters of the phospholipid acyl chains of trout erythrocytes
Fatty acid
Diet K
A
C
N
sat. (n-9) (n-7) (n-6) (n-3)
53.62 -+3.05 17.69 -+1.72 2.04+_0.28 12.55 _-+4. t4 11.86_+0.69
52.02 _+5.50 5.33 +__1.40** 10.00_+ 1.72"* 5.38 -+0.49 29.77 _+1.42"**
39.94 _+2.50* 11.82 _+1.00" 1.31 _+0.t3 38.27 _+1.90"* 8.66_+ 1.07"
42.91 -+0.92* 25.00 _+0.50** -9.29 _+0.63 23.00+_ 1.28"**
unsat./sat.
0.88 +_O.12
0,97 -+O.19
1.53 -+O.14*
1.33 _+0.05*
DB index
143 + 12
201 + I0"*
201 +_10"*
205 + 6**
The fatty acid data are expressed as the mean molar percent of total _+ S.E. for four samples. K, commercial diet; A, cod liver oil; C, grapeseed oil; N, hydrogenated coconut oil. DB index: number of double bonds per 100 moles fatty acid. * p < 0.05, ** p < 0.0I, • ** p < 0.001 as compared to group K (control group).
K). The fatty acid c o m p o s i t i o n of erythrocyte membranes o f the four groups are presented in Table 4 and the various indices are given in T a b l e 5. The erythrocytes of A-fed trout had the highest ( n - 3 ) fatty acid level ( m a i n l y as 22:6) a n d the lowest ( n - 6 ) fatty acid level. The reverse c o m p o s i t i o n was observed in C-fed trout, the level of the ( n - 6 ) fatty acids being the highest and c o m p o s e d of 18:2, 20:4 and 22:5. The erythrocytes of N-fed trout were characterized by a high level of ( n - 9 ) fatty acids (mainly as 18:1) and by the presence of the u n u s u a l 20:3 ( n - 9 ) . The erythrocytes of K-fed trout had a high ( n - 9 ) fatty acid level a n d their ( n - 6 ) a n d ( n - 3 ) fatty acid c o n t e n t s had i n t e r m e d i a t e values between those o f A- a n d C-fed trout. The u n s a t u r a ted/s~iturated ratio was higher in the erythrocytes o f C- a n d N-fed t r o u t t h a n in A- a n d K-fed trout while the d o u b l e b o n d index was the same in erythrocytes o f A-, C- a n d N-fed trout a n d lower in those o f K-fed trout.
Table 6. The effect of dietary fats on nucleotide contents of trout erythrocytes
Diet
ATP
ATP/ADP
K A C N
2039+ 168 2017 _+66 2233 _+94 1941 _+122
4.96+_0.30 2.77 _+0.15*** 3.64 _+0.08*** 3.73_+0.43
ATP contents are expressed in nmole per ml packed cells. Mean value _+ S.D. of four samples. K, commercial diet; A, cod liver oil; C, grapeseed oil; N, hydrogenated coconut oil. ** p < 0.01, *** p < 0.001 as compared to group K (control group).
throcytes from N-fed trout had altered shapes (Fig. 3 - 5 ) , 4 2 % having crenated m a r g i n s (discoechinocytes) a n d 30O/o being characterized by a s h r u n k e n appearance. The A T P content and A T P / A D P
ratio o f ery-
throcytes from A-, K-, C- a n d N-fed t r o u t are given
S c a n n i n g electron micrographs o f fresh erythrocytes from K- a n d N-fed t r o u t are presented in Fig.
in T a b l e 6. No difference in the A T P c o n t e n t s could be observed between the four g r o u p s but differ-
1 - 5 . Erythrocytes from A- or K-fed trout had the same m o r p h o l o g y a n d only o n e representative sample is presented (Fig. 1 - 2 , K-fed trout). They can be considered the typical teleost discocytes. Erythrocytes from C-fed trout (not shown
ences were observed at the level o f A T P / A D P ratios. Erythrocytes from A-fed trout had the
had little more a p p a r e n t nuclei a n d slightly swollen margins when c o m p a r e d to control discocytes. Ery-
lowest A T P / A D P ratio, erythrocytes from K-fed trout having the highest value a n d those from Ca n d N-fed trout having i n t e r m e d i a t e values.
31
Figs. 1-5. The effect of dietary fats on the morphology of trout erythrocytes. Fig. 1. Normal discocytes, K diet ( x 2600). Fig. 2. Normal discocytes, K diet (x 5000). Fig. 3. Altered shapes with N-diet (x 2600).
Fig. 4. Shrunken erythrocyte, N-diet (× 6000). Fig. 5. Discoechinocyte, N-diet (x 6000).
32 Discussion
The results of the present study show that a six month feeding of various oils is able to profundly alter the lipid composition of trout erythrocyte membrane. This is in contrast with results from mammalian erythrocytes where only minor changes were detected in the membrane fatty acid pattern (Rao et al. 1979; Vajreswari et al. 1983; Benga et aL 1984) even after feeding animals for several generations (Ehrstr6m et al. 1981). Concurrently, the basic features of the fatty acid composition are largely modified by the diet as regards either the molar percentage of saturated fatty acids or the average unsaturation index. It is noteworthy that the lipid compositional changes induced by essential fatty acid deficient diet at the level of trout erythrocytes are similar to those we have reported for trout intestinal mucosa (Di Costanzo et al. 1983; Leray and Florentz 1983). The analysis of the influence of fully saturated dietary fats in the present work suggests that in addition to the appearance of eicosatrienoic acid (20:3, n - 9 ) which is characteristic of this deficiency state (Farnsworth et al. 1965; Walker et al. 1967) there is a marked incorporation of essential ( n - 3 ) fatty acids in membrane phospholipids. This specific molecular regulation which must involve transport from other tissues deserves further investigations. As it was reported in rat after dietary manipulations (Ehrstr6m et al. 1981) the cholesterol/phospholipid ratio of trout erythrocytes is not significantly affected by the quality of ingested fats. lri a number of investigations there have been attempts to correlate the osmotic fragility of erythrocytes to their lipid and fatty acid composition (Bruckdorfer et aL 1969; Raz and Livne 1973; Lange et al. 1980; Benga et al. 1984; Kuypers et al. 1984). As a result, the stabilization of erythrocyte membrane has been generally attributed to increased cholesterol/phospholipid ratio and increased unsaturated fatty acids. In the present investigations, however, the effect of cholesterol levels on the differences observed in the rate of osmotic hemolysis may be discounted since the cholesterol/ phospholipid ratio was practically constant for all four groups of animals. An important factor in
determining the stability of the erythrocytes membrane is that of its fatty acid composition, in particular the level of unsaturated fatty acids. The abundance of unsaturated fatty acids (Raz and Livne 1973; Shand and Noble 1981; Vajreswari et al. 1983) was shown to be related to a greater extent of membrane expansion when cells are exposed to hypotonic solutions. This increase in membrane surface allows an increase in cellular critical volume which is positively correlated with the osmotic resistance (Anderson and Lovrien 1977) when the resulting hemolysis is studied as a function of salt concentration. On the other hand, the rate of osmotic hemolysis measures the cell rupture process which follows the swelling process and thus may be independent of the osmotic fragility (Araki and Rifkind 1981). This last biophysical study has indicated that the rupturing rate of the erythrocyte membrane is closely associated with the fluidity of the lipid bilayer and thus affords estimation of intermolecular interactions within the bilayer. While these relationships were demonstrated in altering membrane fluidity by treatment with lipidsoluble compounds (Araki and Rifkind 1981), similar results can be expected when the flexibility of the hydrocarbon chains forming phospholipid molecules are experimentally or naturally modified. From a large number of investigations using several technical approaches it has been stated that both the length of the chains and the number of double bonds per chain are indicators of membrane fluidity (Deuticke 1977; Brenner 1984). In contrast to the above-mentioned studies, the comparisons of phospholipid fatty acid compositions of the erythrocytes from the four groups of trout in the present investigations demonstrate that the level of unsaturated fatty acids and the degree of unsaturation are positively correlated with the rates of osmotic hemolysis (rate of osmotic hemolysis = 3.81 + 3.92 x (unsat./sat.), r = 0.523, n = 16, P = 0.036). The analysis of results on osmotic fragility determined in two groups (diets K and C) allowed similar conclusions. Furthermore, changes in the fatty acid pattern of trout erythrocyte lipids do not suggest relationships between any specific constituents and osmotic fragility. Thus, no direct in-
33 fluence on the cell rupture process can be inferred from the relative amount of essential (n-3), nonessentiel (n-6) or non-essentiel (n-9) fatty acids as it was suggested by Ehrstr6m et al. (1981) and Benga et al. (1984). Our experimental results in trout erythrocytes emphasize the importance of the specific effects on membrane structure and function caused by changes in fatty acid composition. These changes resulting from differences in dietary composition are of particular interest when compared with the rather limited alterations described in mammalian erythrocytes (Ehrstr6m et al. 198 I). The hypothesis of a close association between the lipid unsaturation and the erythrocyte rupture process seems to be supported by recent results on the importance of the degree of unsaturation of phosphatidylcholine in human erythrocytes (Kuypers et al. 1984). Thus, the replacement of native phosphatidylcholine by more unsaturated species has been found to modify considerably the membrane properties resulting in a progressive increase in osmotic fragility and K + permeability. It is noticeable that in this last study the osmotic fragility was measured as the degree of hemolysis of erythrocytes in a continuously decreasing salt concentration, approach which is not very different from that used in our study and emphasizes the importance of dynamic tests in the appreciation of membrane perturbations. It has been shown that erythrocyte shape is controlled by several and frequently related biochemical mechanisms. Since the early observation of echinocytosis (called also crenation) after ATP depletion (Nakao et al. 1960), several workers have proposed that this metabolic control is exerted either by the phosphorylation of spectrin (Birchmeier and Singer 1977) or by lipid bilayer imbalance secondary to phosphoinositide metabolism (Ferrell and Huestis 1984). Various amphipathic molecules are able to induce a spiny shape in red cells similar to that produced by ATP depletion (Deuticke 1968; Fujii et al. 1984). All these morphological changes have been convincingly explained by differential variations in the relative areas of the plasma membrane monolayers (Sheetz and Singer 1974; Lange and Slayton 1982). Our observation of a discoechinocytosis and
shrinkage in erythrocytes from trout raised on saturated lipid rich-diet without changes in ATP level cannot be presently interpreted in term of membrane structural organization until further investigations have been carried out. No explanation can be given for the observed variations of the ATP/ADP ratios in the erythrocytes of the A- and C-fed trout when compared to the control group (K-fed trout) and no correlation with shape changes can be established. In spite of the complexity of these phenomena, the association of structural changes with the presence of large amounts of (n-9) fatty acids (mainly 18:l and 20:3) suggests the hypothesis of specific accumulation of these compounds in the outer membrane leaflet, expanding its area by local increase of unsaturation and thereby producing the cr.enated margin and the skrunken appearance. The study of the distribution of phospholipids in the two halves of the membrane bilayer would help to explain the shape changes induced in the circulating blood red cells of N-fed trout. In conclusion, the results of the present study show the clear influence of the dietary lipids on the biochemical composition of trout erythrocyte membrane. The analysis of the correlations between the different fatty acid patterns and the osmotic sensitivities suggests that lipid unsaturation is the main modifier of the cell membrane properties. While the mechanisms involved in these interactions are presently unknown, our results do demonstrate the importance of proper fatty acid feeding in maintaining the normal physiological functions and discoid shape of trout red cells. Thus, it appears that the system here reported is potentially useful since characteristic chemical changes can be produced with associated modifications in the morphology and physical state of the erythrocyte membrane.
Acknowledgements We thank Mrs G. Gutbier and Mr. J.C. Barthe for their technical assistance.
34
References cited Anderson, P.C. and Lovrien, R.E. 1977. Human red cell hemolysis rates in the subsecond to seconds range. An analysis, Biophys. J. 20: 181-191. Araki, K. and Rifkind, J.M. 1981. The rate of osmotic hemolysis. A relationship with membrane bilayer fluidity. Biochim. Biophys. Acta 645: 81-90. Bartlett, G.R. 1959. Phosphorus assay in column chromatography. J. Biol. Chem. 234: 466-468. Benga, G., Travis, B.D., Pop, V.I., Popescu, O., Toader, S. and Holmes, R.P. 1984. The effect of the saturation and isomerization of dietary fatty acids on the osmotic fragility and water diffusional permeability of rat erythrocytes. Biochim. Biophys. Acta 775: 255-259. Birchmeier, W. and Singer, S.J. 1977. On the mechanism of ATP-induced shape changes in human erythrocyte membranes. I1. The role of ATP. J. Cell. Biol. 73: 647-659. Brenner, R.R. 1984. Effect of unsaturated acids on membrane structure and enzyme kinetics. Prog. Lipid Res. 23: 69-96. Bruckdorfer, K.R., Demel, R.A., De Gier, J. and Van Deenen, L.L.M. 1969. The effect of partial replacements of membrane cholesterol by other steroids on the osmotic fragility and glycerol permeability of erythrocytes. Biochim. Biophys. Acta 183: 334-345. Castell, J.R., Sinnhuber, R.O., Lee, D.J. and Wales, J.H. 1972a. Essential fatty acids (EFA) in the diet of rainbow trout: physiological symptoms of EFA deficiency. J. Nutr. 102: 87-92. Castell, J.R., Sinnhuber, R.O., Wales, J.H. and Lee, D.J. 1972b. Essential fatty acids in the diet of rainbow trout (Salmo gairdneri): growth, feed conversion and some gross deficiency symptoms. J. Nutr. 102: 77-85. Deuticke, B. 1968. Transformation and restoration of biconcave shape of human erythrocytes induced by amphiphilic agents and changes of ionic environment. Biochim. Biophys. Acta 163: 494-500. Deuticke, B. t977. Properties and structural basis of simple diffusion pathways in the erythrocytes membrane. Rev. Physiol. Biochem. Pharmacol, 78: 1-97. Di Costanzo, G., Duportail, G., Florentz, A. and Leray, C. 1983. The brush border membrane of trout intestine: influence of its lipid composition on ion permeability, enzyme activity and membrane fluidity. Mol. Physiol. 4: 279-290. Ehrstr6m, M., Harms-Ringdahl, M. and Ailing, C. 1981. Osmotic fragility and fluidity of erythrocyte membranes from rats raised on an essential fatty acid deficient diet. Biochim. Biophys Acta 644: 175-182. Farnsworth, P., Danon, D. and Gellhorn, A. 1965. The effect of fat-free diet on the rat erythrocyte membrane. Brit. J. Haemat. 11 : 200-205. Ferrell, J.E. and Huestis, W.H. 1984. Phosphoinositide metabolism and the morphology of human erythrocytes. J. Cell Biol. 98: 1992-1998. Folch, J., Lees, M. and Sloane-Stanley, G.N. 1957. A simple method for the isolation and purification of total lipids from
animal tissues. J. Biol. Chem. 226: 497-509. Fijii, T., Tamura, A., Tanaka, T., Nagasawa-Fujimori, H. and Hiromi, K. 1984. Light scattering stopped-flow studies on the time course of shape changes induced in human erythrocytes by several amphiphilic lipids. Biochem. lnternat. 8: 83-88. Kates, M. 1972. Techniques of Lipidology. North-Holland Publishing Company, Amsterdam. Kuypers, F.A., Roelofsen, B., Op den Kamp, J.A.F. and Van Deenen, L.L.M. 1984. The membrane of intact human erythrocytes tolerates only limited changes in the fatty acid composition of its phosphatidylcholine. Biochim. Biophys. Acta 769: 337-347. Lange, Y. and Slayton, J. 1982. Interaction of cholesterol and lysophosphatidylcholine in determining red cell shape. J. Lipid Res. 23: 1121-1127. Langue, L.G., Van Meer, G., Op den Kamp, J.A.P. and Van Deenen, L.L.M. 1980. Hemolysis of rat erythrocytes by replacement of the natural phosphatidylcholine by various phosphatidylcholines. Eur. J. Biochem. 110:115-121. Leray, C. 1979. Patterns of purine nucleotides in fish erythrocytes. Comp. Biochem. Physiol. 64B: 77-82. Leray, C. and Florentz, A. 1983. Biochemical adaptation of trout intestine related to its ion transport properties. Influences of dietary salt and fatty acids, and environmental salinity. In Intestinal Transport, pp. 354-368. Edited by M. Gilles-Baillien and G. Gilles. Springer-Verlag, Berlin. Leray, C. Nonnotte, G., Roubaud, P. and Leger, C. 1985. Incidence of (n-3) essential fatty acid deficiency on trout reproduction processes. Reprod. Nutr. Dev. 25: 561-581. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randal, R.J. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265-275. Nakao, M., Nakao, T. and Yamazoe, S. 1960. Adenosine triphosphate and maintenance of shape of human red cells. Nature, Lond. 187: 945-946. Rao, G.A., Siler, K. and Larkin, E.C. 1979. Diet-induced alterations in the discoid shape and phospholipid fatty acid compositions of rat erythrocytes. Lipids 14: 30-38. Raz, A. and Livne, A. 1973. Differential effects of lipids on the osmotic fragility of erythrocytes. Biochim. Biophys. Acta 311: 222-229. Shand, J.H. and Noble, R.C. 1981. The maternal diet and its effect on the lipid composition and osmotic fragility of neonatal ovine erythrocytes. Biol. Neonate 40: 150-159. Sheetz, M.P. and Singer, S.J, 1974. Biological membranes as bilayer couples. A mechanism of drug-erythrocyte interactions. Proc. Natl. Acad. Sci. 71: 4457-4461. Sinnhuber, R.O., Castelt, J.D. and Lee, D.J. 1972. Essential fatty acid requirement of the rainbow trout, Satrno gairdneri. Fed. Proc. 31: 1436-t441. Vajreswari, A., Rao, P.S., Kaplay, S.S. and Tulpule, P.G. 1983. Erythrocyte membrane in rats fed high erucid acidcontaining mustard oil: osmotic fragility, lipid composition, and (Na +, K + )- and (Ca 2+ , Mg2÷ )-ATPases. Biochem. Med. 29: 74-84. Walker, B.L. and Yurkowski, M, 1967. Effect of cell age on
35 erythrocyte fatty acid composition in rats on different dietary regimes. Biochem. J. 103: 218-224. Watanabe, T. 1982. Lipid nutrition in fish. Comp. Biochem. Physiol. 73B: 3-15.
Yu, T.C. and Sinnhuber, R.O. 1976. Growth response of rainbow trout (Salmo gairdneri) to dietary w3 and ~6 fatty acids. Aquaculture 8: 309-317.
