Changes in the fatty acid composition of phospholipids from turbot (Scophthalmus maximus) in relation to dietary polyunsaturated fatty acid deficiencies

Comp. Biochem. Physiol. Vol. 81B, No. 1, pp. 193-198, 1985 Printed in Great Britain

0305-0491/85 $3.00+0.00 © 1985 Pergamon Press Ltd

C H A N G E S IN THE F A T T Y A C I D C O M P O S I T I O N OF P H O S P H O L I P I D S F R O M T U R B O T (SCOPHTHALMUS MAXIMUS) IN R E L A T I O N TO D I E T A R Y POLYUNSATURATED FATTY ACID DEFICIENCIES M. V. BELL,R. J. HENDERSON and J. R. SARGENT N.E.R.C. Institute of Marine Biochemistry, St. Fittick's Road, Aberdeen AB1 3RA, UK (Tel: 0224 875695) (Received 28 August 1984) Young turbot (1 20 g) were maintained for not less than 14 weeks on three diets: (1) a control diet containing normal amounts of polyunsaturated fatty acids (PUFA); (2) a diet totally deficient in PUFA: (3) a diet deficient in the (n-6) series of PUFA but containing (n-3) PUFA. 2. At 14 weeks the fatty acid compositions of the phospholipids from liver, gut, gills and muscle were analysed. Large changes in the amounts of PUFA in the phospholipids were found. 3. Fish maintained on the totally PUFA deficient diet 2 had retained arachidonic acid, 20:4(n-6), and docosahexaenoic acid, 22:6(n-3), at the expense of eicosapentaenoic acid, 20: 5(n-3). 4. Fish maintained on the (n-6) PUFA-deficient diet (3) contained decreased amounts of 20:4(n-6) and 22:6(n-3) while retaining 20: 5(n-3). 5. In all cases phosphatidylinositol had the lowest n-3/n-6 ratios. These results are discussed in terms of PUFA function.

Abstract--l.

(Bell et al., in press). In this study we describe changes in the fatty acid composition of PtdCho, PtdEtn, PtdSer, Ptdlns and SM brought about by dietary P U F A deficiencies in the turbot.

INTRODUCTION Dietary fat in animals influences the fatty acid composition of depot neutral lipid, the fatty acid composition of membrane phospholipids, the head group composition of membrane phospholipids and the ratio of membrane phospholipid to membrane protein (Stubbs and Smith, 1984; Clandinin et al., 1983). For example, in rodents both the fatty acid composition and the polar head group composition of mitochondrial phospholipids respond rapidly to changes in dietary fatty acid composition (Innis and Clandinin, 1981a, b). Changes in the fatty acid composition of dietary fat altered the phospholipid class composition of brain synaptosomal and liver plasma membranes (Clandinin et al., 1983). Changes such as these alter the physicochemical properties and the biochemical functions of the biomcmbrane in question (Lee, 1977; Innis and Clandinin, 198 lc; Di Costanzo et al., 1983; Stubbs and Smith, 1984). We have found that, in a variety of marine fish species where the polyunsaturated fatty acids of the major phospholipids (PtdCho and PtdEtn) are predominantly of the (n-3) series, phosphatidylinositol (Ptdlns) contains large amounts of arachidonic acid, 20:4(n-6). In these species the ratio of (n-3)/(n-6) polyunsaturates in PtdIns is about one order of magnitude lower than the bulk phospholipids (Bell et al., 1983; Tocher and Sargent, 1984). In a previous study we described the effects of dietary P U F A * deficiencies on the growth and survival of juvenile turbot together with the development of certain nutritional pathologies, particularly in gills

MATERIALS AND METHODS

Chemicals All solvents were of HPLC grade and were obtained from Rathburn Chemicals Ltd., Walkerburn, Peeblesshire, UK. Butylated hydroxytoluene (BHT) was obtained from Sigma Chemical Co., Poole, Dorset, UK and 2,7-dichlorofluorescein from BDH Chemicals Ltd, Poole, Dorset, UK. Thinlayer silica gel G plates (20 x 20 cm, 0.25 mm thick) were obtained from E. Merck, Darmstadt, FRG. Fish Young turbot (Scophthalmus maximus) were netted close inshore in Cardigan Bay, UK in October 1982 and transported immediately to the aquarium in Aberdeen. The fish were distributed uniformly into three fibre-glass tanks, 102 fish of 1-20 g weight per tank, each containing 120 1 of circulating seawater at 10-13°C. The three groups offish were maintained separately on the diets described below, all diets being offered to the fish at a daily rate of 2 g/100 g fish. Fish were used after at least 10 weeks on the experimental diets. For diet 3 fish, the results were obtained from a pool of fish which had been maintained for 10 weeks on diet 3A followed by an average of 4 weeks on diet 3B. Diet 3 was changed because of the high mortality of the fish on diet 3A (Bell et al., in press). Diets The composition of the diets, described previously by Bell et al. (1984), is reproduced here for convenience (Table 1). The fish oils and (n-3) PUFA concentrates used in the diets were also described by Bell et al. (1984) (Table 2).

* Abbreviations used: PUFA, polyunsaturated fatty acid; PtdCho, phosphatidycholine; PtdEtn, phosphatidylethanolamine; PtdSer, phosphatidylserine; Ptdlns, phosphatidylinositol; SM, sphingomyelin. CBP(B>81 :~-M

Extraction and purification of phospholipids Healthy fish from the control diet (diet 1) were killed by transecting the spinal column, then destroying the brain. The 193

194

M. V. BELL. R. J. HENDERSON and J. R. SARGENT Table I. The percentage compositions of the diets

Control Diet (I) (n-3)÷(n-6) PUPA

Lxperimental Diet (2) NO PUEA

Experimental Diet (3) (n-3) PUFA

FiSh Protein

60

60

60

Dextrin

21.4

21.4

21.4

Carboxyme thy icellulose

2.5

2.5

2.5

Vit a~itn mix

2.8

2.8

2.8

Mineral Mix

3.0

3.0

3.0

8.7

Fish oil

10.0

Pahnitic acid

-

10.0

(n-3) PUFA concentrate

-

-

1.3

Santoquin

0.05

0.05

0.05

Inosine

0.27

0.27

0.27

total (n-3) PUFA %

2.5

0

1.3

total (n-6) PUFA %

0,23

0

<0.003

(n-3) / (n-6) 20:5/22:6

11,1

0

1.8

0

430:1 13.8 (At, 2.2 (B)

Table 2. Polyunsaturated fatty acid composition of the fish oil and concentrates used during the preparation of diet 1, the control diet, and diets 3A and 3B. the (m6) PUFA-deficient diets (diet 3) % composition

Contzol (I)

3A

19,8

± 3.1

3B

16:4(n-3)

1.8

18:2(n-6)

1.4

0

18:4 (n-3)

2.1

0

0

20:4(n-6)

0.9

0,26 -+ 0.30

0.05 f 0.05

20:5(n-3)

13.0

62.1

± 7.0

22:5(n-3)

1.5

3.2

+ N.7

22:6(n-3)

7.2

4.5

± 2.7

20:5/22:6

1.9

13.8

29.1

+ 6.8 0

43.4

f 6,5

0.45 _+ 0.45 19.3

± 2.1

2.2

Means + S.D. for 3 batches of 3A and mean + range of 2 batches of 3B.

livers, guts, gills and two fillets of muscle from each fish were removed and placed on ice. It was necessary to use 15-20 fish from each diet to obtain sufficient material. The intestine from the stomach to the anus was taken, opened and washed free of contents with 0.9~ (w/v) NaC1, then blotted dry. Gill filaments were removed from the arches and the muscle was freed of skin and subcutaneous fat. The tissues were finely chopped with scissors, resuspended in 2 ml of 0.9~ (w/v) KCI and homogenised in 20 ml methanol (6 strokes, Teflon in glass homogeniser, 0.2 m m clearance, TriR model K43 drive, speed 6500-11,000 rev/min range). All solvents, except those used for chromatography, contained 0.013/,, (w/v) BHT. Forty millilitres of CHC13 was then added to the homogenate and the suspension shaken vigorously then filtered. Thirteen millilitres of water were added to the filtrate, the suspension shaken and then centrifuged (700 g, 5 min) to give two phases. The lower CHC13 layer was filtered through phase-separating paper (Whatman, England), concentrated under vacuum at 30°C and finally dried under a stream of nitrogen at room temperature before desiccating in v a c u o overnight. The turbot on both of the PUFA-deficient diets showed high mortality and the fish which died were frozen in liquid

nitrogen as soon as possible after death and stored at 70 ~C. These fish were used as the source of material for turbot on the PUFA-deficient diets. When required the fish were thawed and the liver, gut, gills and muscle removed as for control fish. In the case of liver from diet 1 the large a m o u n t of neutral lipid present was removed by t.l.c, using hexane/diethyl ether/glacial acetic acid 80:20:2 (v/v) as developing solvent. Polar lipids were eluted from the origin using chloroform/ methanol/water 5 : 5 : 1 (v/v) and concentrated under vacuum before final removal of solvent under nitrogen. Phospholipids were purified by t.l.c, in two dimensions using chloroform/methanol/water/0.88 a m m o n i a 130:70: 8:0.5 (v/v) in the first dimension and chloroform/acetone/ methanol/glacial acetic acid/water 10:4:2:2:1 (v/v) in the second dimension (Parsons and Patten, 1967). After development the plates were sprayed lightly with O.l')i, (w/v) 2,7-dichlorofluorescein in 951~/oMeOH containing 0.1"~, (w/v) BHT, and visualised under u.v. light. Individual zones were scraped from the plates and eluted with 10 ml of chloroform/ methanol/water 5: 5 : 1 (v/v). The solvent was removed under vacuum and the phospholipids desiccated in vacuo for 3 hr.

195

Dietary PUFA deficiencyin turbot Preparation and analysis of fatty acid methyl esters The fatty acyl chains of the phospholipids were transmethylated and the methyl esters purified as described previously (Bell et al., 1983). Analysis of fatty acid methyl esters was performed on a Packard 429 gas chromatograph (Packard Instruments Ltd., Caversham, UK) equipped with an open fused silicacapillary, 50 m in length, 0.32 mm i.d., and coated with liquid-phase CP Wax 51 (Chrompack, Middelburg, The Netherlands). The chromatogram program has been described (Bell et al., 1983).

RESULTS AND DISCUSSION

The liver, gut, gills and muscle were chosen for analysis for the following reasons. Liver contains large amounts of intracellular biomembranes. Intestinal mucosa has a rapid cellular turnover, so that changes in dietary fatty acid intake might be expected to manifest themselves quickly in this tissue. The gills are the osmoregulatory tissue in teleost fish and PtdIns turnover is important in control of salt secretion (Simpson and Sargent, manuscript in preparation). Muscle makes up the bulk of the body tissue in a healthy fish and therefore is more representative of the total body situation. The percentages of the main polyunsaturated fatty acids, arachidonate 20:4(n-6), eicosapentaenoate 20: 5(n-3) and docosahexaenoate 22:6(n-3), found in PtdCho, PtdEtn, PtdSer, PtdIns and SM in the liver, gut, gills and muscle from fish fed on diets 1, 2 and 3, are shown in Tables 3-5 respectively. The total saturates, total monounsaturates, saturate to monounsaturate ratio, total (n-3), total (n-6) and n-3 to n-6 ratio are also shown in Tables 3 5. The saturates 14:0, 16:0 and 18:0 and the monounsaturates 16: l(n-7) and 18: l(n-9) were the other major fatty acids found. Other fatty acids which were present at the levels of several percent in most of the phospholipids were 16: l(n-9), 18: l(n-7), 18:2(n-6), 22:4(n-6)

and 22:5(n-3), while 15:0, 16:4(n-3), 20:0, 20:1 isomers, 24: 0 and 24:1 isomers were present in certain phospholipids. Small amounts of less than 1~ of iso-16:0, 17:0, 18:3(n-6),- 18:3(n-3), 18:4(n-3), 20:2(n-6), 20:3(n-6), 20:3(n-3), 20:4(n-3) and 22:4(n-3) were found in some lipids. All these minor fatty acids are included in the totals for saturates, monounsaturates, n-3 and n-6 PUFAs. Certain phospholipids showed particular characteristics in all the analyses, despite some variations between tissues and diets. Thus PtdCho was rich in 16: 0 and 18: l(n-9), with either 20: 5(n-3) or 22: 6(n-3) the major PUFA (e.g. Table 3). PtdEtn was always rich in PUFA, especially 22: 6(n-3) (Tables 3 and 4). Some dimethyl acetals produced from plasmalogens were found in PtdEtn. Amounts were variable but the gills were the tissue richest in plasmalogens, with up to 7~ in PtdEtn from gills of fish from diet 1. The dimethyl acetals of 16:0, 18:0, 18:1(n-9) and 18: l(n-7) acids were found. PtdSer was rich in saturates, particularly 18:0, with 22: 6 (n-3) the predominant PUFA (e.g. Tables 3 and 5). Ptdlns was also rich in 18:0 but arachidonic acid 20:4(n-6) was usually the major PUFA, with the exception of the tissues of fish on diet 3 (Tables 3-5). Consequently the (n-3)/(n-6) ratio of Ptdlns was much lower than that of the other phospholipids, confirming the findings of Bell et al. (1983) and Tocher and Sargent (1984). Ptdlns from muscle from all diets had high saturate/monosaturate ratios (4.5-4.8) compared to the other phospholipids, which showed ratios between 1 and 2.5 (Tables 3-5). SM showed a more variable composition than the other phospholipids, with some unique features. 14:0 was often very important, with up to 30~ in gut from fish on diet 1, and there were significant amounts of 15:0 (up to 7.3~, but low in diet 2 fish) and 24:1 isomers (up to 30.7~ in gut from diet 2 fish). SM from fish on diet 1 contained a high proportion of saturates

Table 3. Majorfatty acidsin phospholipidsfrom tissuesof turbot maintainedon diet 1

Lipid

Tissue

20:4

20:5

22:6

total sat

total monounsat

sat monounsat

total (n-3)

total (n-6)

PtdCho

Liver Gut Gills Muscle

1.5 1.9 2.4 1.9

12.6 12.5 10.7 15.0

15.9 17.5 8.8 7.3

39.8 35.8 35.2 42.1

19.9 20.6 32.1 25.2

2.0 1.7 1.1 1.7

32.5 35.6 23.4 25.7

2.9 3.1 4.7 3.3

11.2 11.6 5.0 7.8

PtdEtn

Liver Gut Gills Muscle

1.5 2.3 4.7 1.7

10.9 9.1 11.4 10.0

23.0 27.2 20.6 21.2

28.5 25.9 25.0 29.2

23.4 23.5 23.6 26.7

1.2 1.1 I .I 1.1

38.9 41.3 37.4 37.1

3.8 4.6 6.9 4.0

10.1 9.0 5.4 9.2

PtdSer

Liver Gut Gills Muscle

2.5 2.4 3.4 2.5

7.1 5.5 6.5 4.5

16.4 22.8 20.7 20.7

81.5 38.2 40.6 44.0

15.7 18.9 14.9 14.2

3.3 2.0 2.7 3.1

26.4 34.4 34.5 32.8

3.1 4.4 5.5 3.6

8.5 7.9 6.2 9.3

Ftdlns

Liver Gut Gills Muscle

28.5 21.7 22.1 12.3

11.2 8.3 9.7 9.2

4.2 13.4 7.4 16.3

37.9 38.9 42.5 45.6

15.1 12.3 11.1 9.5

2.5 3.2 3.8 4.8

17.2 25.3 20.5 30.2

29.3 22.4 23.3 12.3

0.6 I .I 0.9 2.5

SM

Liver Gut Gills Muscle

10.7 7.4 4.4 2.5

6.5 7.0 6.0 4.9

6.9 9.5 7.5 7.6

50.2 53.6 51.7 50.0

17.0 12.3 17.6 20.6

3.0 4.4 2.9 2.4

15.0 18.8 15.4 14.7

11.4 8.8 5.1 3.5

1.3 2.1 3.0 4.3

Data are as wt ~.

n-3 n----~

M. V. BELL,R.J. HENDERSONand J. R. SARGENT

196

Table 4. Major fatty acids in phospholipids from tissues of turbot maintained on diet 2

Lipid

Tissue,

20:4

20:5

22:6

total sat

total monounsat

sat monounsat

total (n-3)

total (n-6)

n-3 n-6

PtdCho

Liver Gut Gills Muscle

1.5 3.7 3.7 4.3

4.0 3.8 4.7 12.9

18.4 12.2 9.4 12.0

35.5 40.4 32.8 23.2

30.2 31.3 39.8 23.2

1.2 1.3 0.8 1.6

26.0 18.5 16.7 28.5

4.0 6.7 5.5 6.1

6.4 2.7 3.0 4.7

PtdEtn

Liver Gut Gills Muscle

3.2 7.1 8.2 6.0

6.4 5.6 8.7 13.9

27.6 25.4 23.1 23.1

25.7 25.9 23.9 26.5

25.5 21.8 23.6 19.8

1.0 1.2 3.0 1.3

37.7 34.5 36.3 41.3

5.2 10.1 10.6 8.4

7.3 3.4 3.4 4.9

ptdSer

Liver Gut Gills Muscle

2,8 3.8 2.8 3.9

2.4 2.1 2.3 3.9

13.3 20.8 16.9 15.8

50.9 38.3 47.8 38.0

21 .I 19.7 14.5 17.3

2.4 1.9 3.3 2.2

20.3 27.0 24.2 25.3

4.2 6.2 5.0 6.0

4.9 4.3 4.8 4.3

Ptdlns

Liver Gut Gills Muscle

21.7 20.0 11.3 14.3

11.5 5.6 4.4 6.4

2.8 12.1 10.4 13.1

38.2 40.3 39.7 48.9

21.2 15.3 21.3 10.7

1.8 2.6 1.9 4.6

14.9 19.2 16.8 21.7

22.7 21.7 12.2 14.3

0.7 0.9 1.4 1.5

SM

Liver Gut Gills Muscle

3.0 2.5 7.0 4.2

3.9 1.1 4.5 7.2

14.8 5.0 5.3 42.6

34.9 31.3 40.9 36.5

29.2 41 .I 19.5 22.9

1.2 0.8 2.1 1.6

20.8 7.1 11.7 22.0

4.6 5.1 8.3 5.9

4.6 1.4 1.4 3.8

Data are as wt %;.

Table 5. Major fatty acids in phospholipids from tissues of turbot maintained on diet 3

Lipid

Tissue

20:4

20:5

22:6

total sat

total monounsat

sat monounsat

total (n-3)

total (n-6)

n-3 n-6

PtdCho

Liver Gut Gills Muscle

0.4 I .I 1.8 1.4

11.9 9.8 7.5 13.0

8.0 4.9 4.7 5.6

45.0 46.0 41.6 40.7

26.6 32.6 37.8 32.8

1.7 ] .4 I .I 1.2

23.5 16.5 14.6 21.0

2.4 2.5 3.2 2.8

10.0 6.6 4.6 7.5

PtdEtn

Liver Gut Gills Muscle

0.7 2.9 5.5 2.3

16.3 16.6 13.4 15.0

18.6 21.6 15.4 20.6

31.4 30.3 30.1 29.4

24.9 17.7 25.0 22.1

1.3 1.7 1.2 1.3

40.1 44.2 33.3 40.7

2.3 4.8 7.9 4.2

17.3 9.3 4.2 9.7

PtdSer

Liver Gut Gills Muscle

0.4 1.0 1.8 1.4

5.7 6.2 4.3 5.9

17.7 15.1 15.1 13.3

52.1 49.6 49.5 47.0

15.1 18.0 15.0 19.7

3.5 2.8 3.3 2.4

29.9 26.3 26.8 25,6

1.4 2.4 2.7 2.8

21.8 11.2 7.3 9.2

PtdIns

Liver Gut Gills Muscle

4.6 5.5 9.9 8.4

24.6 11.4 8.7 11.6

2.2 3.3 4.9 8.8

46.2 46.0 55.1 54.0

19.8 30.7 15.7 12.1

2.3 1.5 3.5 4,5

28,1 16,0 15.8 22,8

4.9 5.0 10.2 8.4

5.7 2.7 1.5 2.7

SM

Liver Gut Gills Muscle

0.4 0.9 0.9 2.9

5.2 6.9 1.6 5.8

3.6 7.0 1.8 6.0

33.6 39.1 52.8 40.0

47.2 31.9 23.9 24.0

0.7 3.2 2.2 1.7

10,5 16,4 4,6 13,5

1.3 1.5 1.3 3.4

7.9 10.9 3.6 4.0

Data are as wt o%.

(Table 3) but fish tissues from the other two diets had a much decreased proportion of saturates, particularly 14:0, and conversely increased monounsaturates in SM (Tables 4 and 5). The P U F A content of SM was also very variable, being as low as 5.9~o in the gills from diet 3 fish (Table 5), and in some cases the (n-3)/(n-6) ratios were almost as low as those for PtdIns. This was largely due to a low content of (n-3) P U F A rather than an especially high (n-6) P U F A content.

The diet totally deficient in P U F A (diet 2) had some unexpected effects on the phospholipid composition (Table 4). The levels of arachidonic acid were increased in PtdCho from gut, gill and muscle, and in PtdEtn. In contrast, the levels of 20:4(n-6) were depressed in PtdIns from liver, gut and especially gills (Table 4). The elevated levels of 20 : 4(n-6) in PtdCho and PtdEtn mean that Ptdlns no longer contains the largest proportion of arachidonic acid. Levels of 20:5(n-3) were much decreased in the phospholipids

Dietary PUFA deficiency in turbot from liver, gut and gills, where they were sometimes more than halved, but were largely unchanged in muscle, where small decreases in PtdCho, PtdSer and PtdIns were balanced by increases in PtdEtn and SM (Table 4). There was no pattern to the changes that occurred in the levels of 22: 6(n-3), with slight increases in some phospholipids balanced by decreases in others. As a result of the general increase in arachidonic acid, decrease in eicosapentaenoic acid and no change docosahexaenoic acid, the n-3/n-6 ratios were decreased in all the phospholipids except PtdIns, which only showed very slight changes (Tables 3 and 4). The saturate/monosaturate ratio was generally decreased in the phospholipids of the fish fed on diet 2, but all the changes were small (Table 4). The fish on both the P U F A deficient diets were very thin and this was particularly the case in diet 2, the totally PUFAdeficient fish. Thus, during the weight loss that occurred there was selective retention of 20: 4(n-6) and 22: 6(n-3) at the expense of 20: 5(n-3). The results obtained from the fish on the (n-6) PUFA-deficient diet (diet 3) at first sight appear to contradict the above result as the levels of 20:4(n-6) and 22:6(n-3) were decreased in fish fed on diet 3 (Table 5). It is, however, necessary to consider the dietary history of these fish. For the first 10 weeks the (n-6) PUFA-deficient diet (3A) was probably also deficient in 22: 6(n-3) as the 20: 5(n-3)/22: 6(n-3) ratio was 13.8. Thereafter the 20: 5/22:6 ratio was 2.2 (diet 3B), which is very close to the value for the natural fish oil used in diet 1 (Tables 1 and 2). The diet 3 fish which were analyzed died after the change to diet 3B at an average of 14 weeks, that is, after 4 weeks on diet 3B, following exposure to low dietary 22:6(n-3) for 10 weeks. Diet 3 was changed because of the very high mortality resulting from diet 3A. There were differences between the four tissues of the fish fed on diet 3, with muscle only losing a small amount of arachidonic acid and progressively larger decreases in gills, gut and liver (Table 5). Liver levels were decreased by 85~ in PtdSer and PtdIns and by 96~ in SM. The amount of20:4(n-6) was decreased in most phospholipids, exceptions being the PtdEtn of gut, gills and muscle, which showed slight increases (Table 5), the largest increase being 25~ in gut. The decrease was most marked for PtdIns from all tissues: down 85~ in liver, 75~ in gut, 55~o in gills and 31~o in muscle, and for SM from all but muscle. These two phospholipids had the highest contents of arachidonic acid in fish from the control diet (Table 3) and therefore had the most to lose. In the PtdCho and PtdEtn of several tissues 18 : 2(n-6) was the major (n-6) PUFA, thus emphasising the enrichment of PtdIns with 20 : 4(n-6). There were also changes in the (n-3) P U F A content of the phospholipids from the fish fed on the (n-6) PUFA-deficient diet. The levels of 20:5(n-3) were decreased in PtdCho (by 30~o in gills) but elevated in PtdEtn in all tissues (up 92~ in gut) (Table 5). 20:5(n-3) was also elevated in PtdIns from liver, gut and muscle, presumably to compensate for the decrease in arachidonate. However, 22: 6(n-3) levels were decreased in all phospholipids except PtdSer from liver, often by more than half (Table 5). There was some 16: 4(n-3) in the (n-3) P U F A concentrate used in diet 3 and small amounts (between 0.3 and 0.7~) of

197

this fatty acid appeared in the phospholipids from liver and gut. Gills showed the lowest n-3/n-6 ratios for all the phospholipids from tissues of fish fed on diet 3 (Table 5). The saturate/monounsaturate ratio remained generally constant in fish fed on diet 3, with balanced increases in both saturates and monounsaturates to offset the decrease in PUFAs (Table 5). The exception to this was SM, which had greatly increased amounts of long-chain monounsaturates, resulting in large decreases in the saturate/monounsaturate ratio. To summarize, therefore, when turbot are fed a diet totally deficient in PUFA, both 20:4(n-6) and 22:6(n-3) are selectively retained at the expense of 20:5(n-3). This suggests that 20:4(n-6) and 22:6(n-3) have more important biochemical functions in the phospholipids than 20: 5(n-3). The loss of weight and tissue degradation which occurred in the experimental fish would allow for selective catabolism of 20: 5(n-3) and retention of 20:4(n-6) and 22: 6(n-3), but the data are also consistent with some limited elongation of 20: 5(n-3) to 22: 6(n-3). However, in the fish fed a diet adequate in 20: 5(n-3)but deficient in 20:4(n-6) and probably deficient in 22: 6(n-3), loss of both 20 : 4(n-6) and 22:6(n-3) occurred, supporting the finding of Cowey et al. (1976) that turbot cannot elongate 20:5 to 22:6. The avidity with which P U F A are retained in the phospholipids is illustrated by the relatively constant P U F A levels in the phospholipids from the fish maintained on the different dietary regimes. This was particularly noticeable for PtdEtn, which contained 40-45~o P U F A in all tissues from all diets. However, there were marked differences in the composition of this PUFA, showing that turbot are unable to maintain a constant chain length and desaturation of the P U F A in the face of dietary changes, unlike mammals, which can further elongate and desaturate dietary fatty acids to maintain phospholipid composition. Turbot therefore require a correct balance of the three main PUFAs 20: 4(n-6), 20: 5(n-3) and 22: 6(n-3) in the diet. Changes in fatty acid composition of membrane phospholipids can exert many biochemical effects. Essential fatty acid deficiency altered the phospholipid to protein ratio, and the cholesterol and PtdCho contents of trout intestine (Di Costanzo et al., 1983). The substrate affinity of brush border alkaline phosphatase was found to be inversely correlated with essential fatty acid content. However, in spite of important membrane modification at the level of lipid composition and ion permeability, no changes in membrane fluidity were found (Di Costanzo et al., 1983). Precisely what changes occur at the biochemical level in these fish, and how they generate the resulting pathologies, remains to be defined.

REFERENCES

Bell M. V., Simpson C. M. F. and Sargent 3. R. (1983) (n-3) and (n-6) polyunsaturated fatty acids in the phosphoglycerides of salt-secreting epithelia from two marine fish species. Lipids 18, 720-726. Bell M. V., Henderson R. J., Pirie B. J. S. and Sargent J. R. (1984) Effects of dietary polyunsaturated fatty acid deficiencies on mortality, growth and gill structure in the

198

M. V. BELL, R. J. HENDERSONand J. R. SARGENT

turbot (Scophthalmus maximus, Linnaeus). J. Fish Biol. (in press). Cowey C. B., Adron J. W,, Owen J. M. and Roberts R. J. (1976) The effects of different dietary oils in tissue fatty acids and tissue pathology in turbot, Scophthalmus maximus. Comp. Biochern. Physiol. 53B, 399403. Clandinin M. T., Foot M. and Robson L. (1983) Plasma membrane: can its structure and function be modulated by dietary fat? Comp. Biochem. Physiol. 76B, 335-339. 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. Molec. Physiol. 4, 279-290. Innis S. M. and Clandinin M. T. (1981a) Dynamic modulation of mitochondrial inner-membrane lipids in rat heart by dietary fat. Biochem. J. 193, 155-167. Innis S. M. and Clandinin M. T. (1981b) Mitochondrial-

membrane polar-head group composition is influenced by diet fat. Biochem. J. 198, 231-234. Innis S. M. and Clandinin M. T. (1981a) Dynamic modulation of mitochondrial membrane physical properties and ATPase activity by diet lipid. Biochem. J. 198, 167-175, Lee A. G. (1977) Lipid phase transitions and phase diagrams II. Mixtures involving lipids. Biochim. biophys. Acta 472, 285-344. Parsons J. G. and Patton S. (1967) Two-dimensional thinlayer chromatography of polar lipids from milk and mammary tissue. J. Lipid Res. 8, 69~%698. Stubbs C. D. and Smith A. D. (1984) The modification of mammalian membrane polyunsaturated fatty acid composition in relation to membrane fluidity and function. Biochim. biophys. Acta 779, 89-137. Tocher D. R. and Sargent J. R. (1984) Analyses oflipids and fatty acids in ripe roes of some N.W. European marine fish. Lipids 19, 492~1-99.