Fish Physiology and Biochemistry vol. 14 no. 2 pp 139-151 (1995) Kugler Publications, Amsterdam/New York

Effects of different dietary arachidonic acid : docosahexaenoic acid ratios on phospholipid fatty acid compositions and prostaglandin production in juvenile turbot (Scophthalmus maximus) J. Gordon Bell, John D. Castell*, Douglas R. Tocher, Fiona M. MacDonald and John R. Sargent N.E.R.C. Unit of Aquatic Biochemistry, School of Natural Sciences, University of Stirling, Stirling FK9 4LA, Scotland, U.K.; *Department of Fisheriesand Oceans, P.O. Box 550, 1707 Lower Water Street, Halifax, Nova Scotia, CanadaB3J 2S7 Accepted: November 24, 1994 Keywords: turbot, arachidonic acid, docosahexaenoic acid, phospholipid, prostaglandin Abstract Five purified diets containing AA (20:4n-6) at 0.02-0.78% dry weight and DHA (22:6n-3) at 0.93-0.17% dry weight were fed to duplicate groups of juvenile turbot (Scophthalmus maximus) of initial weight 0.87 g for a period of 11 weeks. The dietary DHA:AA ratio ranged from 62 to 0.2. Incorporation of AA into liver phospholipids increased with increasing dietary AA input. Phospholipids from fish fed diets containing 0.02, 0.06 and 0.11°7 of dry weight as AA generally contained less AA compared to fish fed fish oil while those fed diets containing 0.35 and 0.78% of dry weight as AA had higher AA levels in their phospholipids. The highest levels of AA were found in PI but the greatest percentage increase in AA incorporation was in PE and PC. Brain phospholipid fatty acid compositions were less altered by dietary treatment than those of liver but DHA content of PC and PE in brain was substantially lower in fish fed 0.93% pure DHA compared to those fed fish oil. This suggests that dietary DHA must exceed 1% of dry weight to satisfy the requirements of the developing neural system in juvenile turbot. In both tissues, (20:5n-3) concentration was inversely related to both dietary and tissue PI AA concentration. Similar dietary induced changes in AA, EPA and DHA concentrations occurred in the phospholipids of heart, gill and kidney. PGE 2 and 6-ketoPGF,, were measured in homogenates of heart, brain, gill and kidney. In general, fish fed the lowest dietary AA levels had reduced levels of prostaglandins in their tissue homogenates while those fed the highest level of AA had increased prostaglandin levels, compared to fish fed fish oil. In brains, the PGE 2 concentration was only significantly increased in fish fed the highest dietary AA.

Introduction The EFA requirements of most freshwater fish can be met by supplying 18:2(n-6) and 18:3(n-3) in their diet. The Cl8 precursors can then be metabo-

lized, by a series of desaturation and elongation reactions, to their long-chain PUFA derivatives which are required to maintain functional cell membrane physiology (Henderson and Tocher 1987; Sargent et al. 1989). In marine fish such as

Abbreviations: AA, arachidonic acid; DHA, docosahexaenoic acid; EFA, essential fatty acid; EPA, eicosapentaenoic acid; HPTLC, high performance thin-layer chromatography; HUFA, highly unsaturated fatty acid; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PGE, prostaglandin E; PGF, prostaglandin F; PI, phosphatidylinositol; PS, phosphatidylserine; PUFA, polyunsaturated fatty acid; TLC, thin-layer chromatography. Correspondence to: J.G. Bell, Tel. 0786 467821, Fax. 0786 464994.

140 turbot (Scophthalmus maximus) the A5-desaturation reaction, which converts 20:3(n-6) to 20:4(n6) (AA), and also 20:4(n-3) to 20:5(n-3) (EPA), appears to be absent or at least extremely low. Thus, these fish require the HUFAs 20:5(n-3) and 22:6(n-3) (DHA) to be supplied in the diet (Owen et al. 1975). Very low A5-desaturase activity in turbot, has been confirmed in studies with turbot cells in primary and established cultures which have also shown that the cells have significant elongation activity (Tocher 1993; Tocher and MacKinlay 1990). While the EFA requirement for turbot has been described as 0.8% of the diet as (n-3) HUFA (Kanazawa 1985), studies by Bell et al. (1985a) showed that extensive mortalities occurred in turbot fed diets with a low DHA/ EPA ratio (0.07) but not in fish fed a high DHA/ EPA ratio (0.45), suggesting metabolic conversion of EPA to DHA is too low to supply the DHA levels required for normal growth and development. Although desaturation and elongation of EPA to DHA occurs in primary cultures of turbot cells, the conversion is slow (Tocher 1993) and recent studies in turbot juveniles have suggested that dietary DHA is required for proper development of neural tissues (Mourente and Tocher 1992). To date no (n-6) PUFA requirement has been established for turbot, or any other marine fish species although juvenile turbot fed diets deficient in (n-6) PUFA had increased mortality and developed an extensive pathology in gill epithelium, compared to those given fish oil (Bell et al. 1985a). The apparent deficiency in A5-desaturase dictates that turbot would require AA to be supplied in the diet. AA is the precursor of a wide variety of biologically active compounds, known collectively as eicosanoids, and including prostaglandins, thromboxanes and leukotrienes (Johnston et al. 1983). AA-derived eicosanoids have been identified in many fish species and have a wide range of physiological functions including control of fluid and electrolyte fluxes, the cardiovascular system, reproductive function and control of the neural system (Mustafa and Srivastava 1989). EPA is also a precursor for eicosanoid formation (Terano et al. 1986) and is usually much more abundant in fish tissues than AA. Despite this, plaice (Pleuronectes

platessa) neutrophils produced 3-4 times as much AA-derived eicosanoids than EPA-derived species, suggesting that AA is the prefered substrate for eicosanoid synthesis, even in an organism with considerably more EPA in cell phospholipids than AA (Tocher and Sargent 1987). Given the importance of AA in eicosanoid formation and the vital range of physiological roles attributed to eicosanoids we investigated the effects of a diet depleted in AA on juvenile turbot. The aim was to feed five diets containing AA and DHA as the only long-chain HUFA with AA ranging from 0.02 to 0.78% of the diet and to determine the effects on tissue phospholipid fatty acid compositions and the production of two ubiquitous prostaglandins, PGE 2 and 6-ketoPGFia.

Materials and methods Materials TLC plates (20 x 20 cm x 0.25 mm) and HPTLC plates (10 x 10 x 0.15 mm), precoated with silica gel 60 were obtained from Merck (Darmstadt, Germany). All solvents were of HPLC grade except those used in extraction of dietary components, which were reagent grade, and were obtained from Rathburn Chemicals Ltd. (Walkerburn, U.K.). Hydrogenated coconut oil was obtained from ICN Biomedicals Ltd. (High Wycombe, U.K.). Oleic acid (SLR grade) was obtained from Fisons Ltd. (Loughborough, U.K.). Pure DHA (99% free acid, n-6 < 0.1%) was obtained from Cascade Biochemicals Ltd. (Reading, U.K.). Pure AA-methyl ester (99%) was obtained from Nu Chek Prep, Inc. (Elysian, U.S.A.). Octadecyl silyl 'Sep-Pak' minicolumns were obtained from Millipore (U.K.) Ltd., Watford.

Animals and diets Three hundred juvenile turbot (initial mean weight 0.87 g), which had been weaned onto a dry pelleted diet for approximately two weeks, were distributed randomly into twelve tanks containing 25 1 of sea

141 Table 1. Fatty acid compositions of diets with varying DHA/AA ratio. Diet Fatty acid

l(control)

14:0 16:0 18:0

6.6 15.2 3.8

6.1 6.1 4.3

5.8 6.1 4.5

6.7 5.7 4.2

7.2 6.2 4.7

8.9 7.6 7.1

Total saturates

27.5

22.0

23.2

28.2

29.5

28.9

16:1n-7 18:In-9 18:ln-7 20:ln-9 22:1n-1l 24:1

5.2 13.9 2.7 8.2 11.3 1.1

3.4 49.3 6.1 0.2

3.1 47.7 5.7 0.2

3.0 42.7 5.2 0.2

2.9 44.0 5.4 0.2

2.5 43.0 5.3 0.1

-

-

0.2

0.3

0.2

0.3

0.4

44.8

60.6

58.8

52.0

53.5

52.3

18:2n-6 20:2n-6 20:4n-6

1.4 0.2 0.9

5.4 0.2 0.1

5.2 0.1 0.4

5.1 0.1 0.7

5.1 0.2 2.3

4.9 0.2 5.2

Total n-6 3

2.9

5.6

5.7

5.8

7.5

11.0

18:3n-3 18:4n-3 20:4n-3 20:5n-3 22:5n-3 22:6n-3

1.0 2.5 0.8 5.7 1.3 8.4

0.7

0.6

0.8

0.7

0.6

-

-

-

-

-

-

6.2

5.7

5.5

4.3

1.1

19.7 22.6 6.8 9.3

6.7 12.1 1.2 62.0

6.4 12.1 1.1 14.3

6.3 12.2 1.1 7.9

5.0 12.6 0.7 1.9

1.7 12.8 0.2 0.2

Total monoenes

Total n-3 Total PUFA n-3/n-6 DHA/AA

2

2

3

4

5

6

-

-

Results are expressed as % by weight and represent means of 3 determinations; includes 12:0, 15:0, 17:0, 20:0 and 22:0; 2 includes 16:1n-9, 20:ln-11; 20:ln-7 and 22:1n-9; 3 includes 18:3n-6, 22:4n-6 and 22:5n-6; - = not determined.

water. The sea water was constantly recirculated and filtered and ammonia and nitrite concentrations were monitored daily. The tanks were subjected to a constant photoperiod of 12 h light, 12 h dark and the water temperature varied from 17-20°C. The fish were fed by hand, to satiation, 3-4 times each day, seven days per week. Fish were starved for 24h prior to weighing at intervals of two weeks. The purified diet formulation utilized casein as protein source, which along with the starch binder and a-cellulose bulking agent were extracted twice with 2 volumes of chloroform/methanol (2:1, v/v) followed by one extraction with 2 volumes of ethanol to remove lipids. The diets were formulated to satisfy the nutritional requirements of turbot and

contained 55% protein and 15% lipid. Diet I was a control diet in which the lipid component was supplied by a North Sea fish oil (Fosol, Seven Seas Ltd., Hull, U.K.). The basic lipid component in diets 2-6 comprised 7 g/100 g of hydrogenated coconut oil and 7 g/100 g of oleic acid. The oleic acid also contained 18:1(n-7) (8%), 16:1(n-7) (5%), 18:2(n-6) (7%) and 18:3(n-3) (1%). The remaining 1% of the lipid component of the diet comprised varying amounts of pure AA and DHA. The fatty acid compositions of the experimental diets are shown in Table 1. A more detailed description of dietary formulation and preparation and fish husbandry has been published elsewhere (Castell et al. 1994).

142 Sampling procedure Fish were killed by severing the spinal column posterior to the brain and samples of liver, gill, heart, brain, kidney and eyes for phospholipid fatty acid analysis were dissected from 18 fish per dietary treatment (9 fish per tank) and frozen immediately in liquid nitrogen. Triplicate samples (4 fish per sample; 6 fish per tank) of brain, heart, gill and kidney were dissected for eicosanoid analysis and homogenized immediately in 3 ml of modified Hanks medium (Moon et al. 1985) containing 15% ethanol (v/v) and 0.15 ml of 2M formic acid. Homogenates were frozen in liquid nitrogen and stored at -20°C until the eicosanoids were extracted.

Lipid extraction and analysis Extraction of total lipid from samples of tissues and diets was performed by the method of Folch et al. (1957) as described in detail previously (Bell et al. 1991). Separation of total lipid into PC, PE, PS and PI fractions was performed by TLC. Samples of total lipid were applied to TLC or HPTLC plates at a concentration of 1 mg cm - l. The plates were developed in methyl acetate/isopropanol/chloroform/methanol/0.25 % aqueous KCI (25:25:25:10: 9 by volume) as described by Vitiello and Zanetta (1978). The plates were sprayed with 0.1% 2',7'dichlorofluorescein in 97% methanol containing 0.05% butylated hydroxytoluene and the lipid classes visualized under UV light. Lipid classes were scraped from the plates and acid-catalyzed transmethylation was performed overnight at 50 0 C as described by Christie (1982). The fatty acid methyl esters were separated and quantified by gas-liquid chromatography as described previously (Bell and Sargent 1992). Individual methyl esters were identified by comparison with commercial standards and by reference to published data (Ackman 1980; Bell et al. 1983). In all analyses of tissue phospholipid fatty acid compositions, values are derived from pooled tissue samples. The small size of the fish, and consequently of the organs obtained from them, made pooling

necessary in order to obtain sufficient lipid to carry out an analysis of phospholipid fatty acid compositions. The need to use pooled tissues did not allow for more than one sample per dietary treatment and thus no statistical analysis of phospholipid fatty acid data was performed.

Extraction and measurement of prostaglandins The frozen tissue homogenates for eicosanoid analysis were thawed then centrifuged to remove precipitated protein. The supernatants were isolated using octadecyl silyl 'Sep-Pak' mini-columns largely as described by Powell (1982). The supernatant was applied to the column which had been prewashed with 5 ml of methanol and 10 ml of distilled water. The column was then washed successively with 10 ml of distilled water, 5 ml of 15% (v/v) ethanol and 5 ml of hexane/chloroform (65:35, v/v) before elution of prostanoids with 10 ml ethyl acetate. The final extract was dried under nitrogen and redissolved in assay buffer. PGE 2 and 6-ketoPGFIa, the stable metabolite of prostacyclin were measured by enzyme immunoassay using kits supplied by Cascade Biochemicals Ltd. (Reading, U.K.).

Statistical analysis Significance of differences (p < 0.05) in prostaglandin production between dietary treatments were determined by ANOVA. Analyses were performed using a Statgraphics (system 3.0) computer package. Differences between means were determined by Tukeys' test.

Results Growth data from this experiment was obtained after 10 weeks on the diets at which point the final weights ranged from 1.81 g (Diet 2) to 2.81 g (Diet 1). The only significant differences in final weights were that fish fed diet 2 weighed significantly less than those fed diets 1 and 6. A more complete

143 Table 2. Fatty acid compositions of phosphatidylcholine from livers of turbot fed diets with varying DHA/AA ratio. Diet Fatty acid

I (control)

2

3

4

5

6

Total saturates Total monoenes

29.9 17.6

26.2 30.4

28.1 33.7

24.1 37.1

29.0 32.3

28.3 30.7

1.8 0.3 0.2 1.7 0.4 0.5 5.0

8.9 0.8 0.1 0.3 t 0.4 10.6

9.3 0.7 t 0.4 t 0.2 10.7

8.8 0.9 0.2 0.8 t t 10.8

9.5 1.0 0.2 3.4 0.3 0.2 14.8

8.8 0.8 0.2 13.6 0.8 t 24.3

18:3n-3 18:4n-3 20:4n-3 20:5n-3 22:5n-3 22:6n-3 Total n-3 Total PUFA

0.7 0.7 0.8 11.8 2.4 26.7 43.1 48.1

0.7 0.3 0.3 1.9 0.8 25.6 29.6 40.2

0.7 0.3 0.1 0.9 0.4 21.6 24.0 34.7

0.7 0.5 0.2 1.0 0.3 19.6 22.3 33.1

0.6 0.4 0.2 1.2 0.6 16.5 19.5 34.3

0.5 0.3 0.2 1.4 0.5 8.7 11.6 35.9

n-3/n-6 20:4/20:5 DHA/AA

8.6 0.1 15.7

2.8 0.2 85.3

2.2 0.4 54.0

2.1 0.8 24.5

1.3 2.8 4.9

0.5 9.7 0.6

18:2n-6 20:2n-6 20:3n-6 20:4n-6 22:4n-6 22:5n-6 Total n-6

Results are expressed as % by weight; values are obtained from livers pooled from 18 fish per dietary treatment; t = value < 0.05%.

description of the growth data, mortalities and carcass total lipid fatty acid compositions is published elsewhere (Castell et al. 1994). Although levels of saturated fatty acids in liver PC were similar between treatments the fish fed diets 2-6 contained considerably more monoenoic fatty acids, largely due to increased 18:l(n-9) which was the major dietary fatty acid in these treatments (Table 2). Fish fed diets 2-6 also had over 4 times as much 18:2(n-6) in liver PC compared to diet 1. 18:2(n-6) was a contaminating fatty acid in the oleic acid used in preparing diets 2-6. Fish fed diets 2-6 also contained more 20:2(n-6) compared to fish fed diet 1. AA increased in liver PC from diet 2 to diet 6 with diets 2-4 having levels less than diet 1 (5- to 2-fold) and diets 5 and 6 having greater levels (2- and 8-fold respectively). The percentage of 22:4(n-6) was also increased in fish fed diet 6. DHA levels in liver PC in diet 1 and diet 2 fish were similar but values declined in diets 3-6 with the level of DHA in diet 6 fish around one

third that of diet fish. EPA and 22:5(n-3) were also reduced in fish on diets 2-6 compared to diet 1. The (n-3)/(n-6) PUFA ratio decreased whereas the 20:4/20:5 ratio increased with increasing dietary AA (Table 2). The percentage of AA in liver PE was reduced in fish fed diets 2-4 (3- to 1.5-fold) and increased in diets 5 and 6 (3- and 6-fold respectively), compared to diet (Table 3). The percentages of 22:4(n-6) were also increased in fish fed diets 5 and 6 compared to diet 1 but 22:5(n-6) was lower than diet 1 in all experimental diets. The level of DHA was slightly increased in fish fed diets 2 and 3 and decreased in diets 4-6, compared to diet 1, although the reduction in DHA in diet 6 fish was not as great as in PC. The (n-3)/(n-6) PUFA ratio decreased progressively from diets 2 to 6 while the ratio of 20:4/20:5 showed the opposite trend, being 60 times greater in diet 6 fish compared to the ratio in fish fed diets I or 2. In general, the dietary induced changes in PS were

144 Table 3. Fatty acid compositions of phosphatidylethanolamine from livers of turbot fed diets with varying DHA/AA ratio. Diet Fatty acid

I (control)

2

3

4

5

6

Total saturates Total monoenes

22.8 22.2

17.5 26.0

16.9 28.4

16.8 31.5

17.7 29.4

17.8 27.4

1.1 0.4 2.0 0.2 1.1 4.8

5.5 0.5 0.6 t 0.4 7.2

5.4 0.4 0.7 t 0.2 7.0

5.4 0.5 1.3 t t 7.4

6.2 0.6 5.7 0.3 t 12.8

4.5 0.6 19.2 1.1 0.2 25.8

18:3n-3 20:4n-3 20:5n-3 22:5n-3 22:6n-3 Total n-3 Total PUFA

0.2 0.7 5.3 2.1 36.8 45.3 50.1

0.3 t 1.5 0.8 41.5 44.3 51.5

0.5 t 0.9 0.4 38.2 40.3 47.3

0.5 t 1.0 0.4 33.9 36.4 43.8

0.6 t 0.8 0.5 32.1 34.3 47.1

0.7 0.1 0.8 0.8 20.5 22.9 48.7

n-3/n-6 20:4/20:5 DHA/AA

9.4 0.4 18.4

6.2 0.4 69.2

5.8 0.8 54.6

4.9 1.3 26.1

2.7 7.1 5.6

0.9 24.0 1.1

18:2n-6 20:2n-6 20:4n-6 22:4n-6 22:5n-6 Total n-6

Footnotes as described in Table 2.

Table 4. Fatty acid compositions of phosphatidylserine from livers of turbot fed diets with varying DHA/AA ratio. Diet Fatty acid

1 (control)

2

3

4

5

6

Total saturates Total monoenes

34.5 16.7

31.5 17.8

30.0 18.3

28.7 18.8

26.7 19.6

29.2 17.1

0.6 0.4 1.9 0.5 1.2 4.9

2.2 0.4 0.6 t 0.6 3.8

2.1 0.4 0.5 t t 3.0

2.3 0.5 0.8 t t 3.6

2.2 0.4 5.7 0.7 0.3 9.9

2.1 0.5 7.3 3.5 0.6 14.0

18:3n-3 20:5n-3 22:5n-3 22:6n-3 Total n-3 Total PUFA

0.4 1.5 3.9 34.3 40.5 45.4

0.3 0.7 1.8 42.7 45.5 49.3

0.3 0.4 0.9 45.1 46.7 49.7

t 0.5 0.9 44.3 45.7 49.3

0.2 0.4 0.9 37.3 39.2 49.4

0.2 0.4 2.0 35.9 38.5 52.5

n-3/n-6 20:4/20:5 DHA/AA

8.3 1.3 18.1

12.0 0.9 71.2

15.6 1.3 90.2

12.7 1.6 55.4

4.0 14.3 6.5

2.8 18.3 4.9

18:2n-6 20:2n-6 20:4n-6 22:4n-6 22:5n-6 Total n-6

Footnotes as described for Table 2.

145 Table 5. Fatty acid compositions of phosphatidylinositol from livers of turbot fed diets with varying DHA/AA ratio. Diet Fatty acid

l(control)

2

3

4

5

6

Total saturates Total monoenes

32.8 16.3

21.0 34.8

19.7 32.6

21.9 31.0

22.0 27.2

21.0 28.1

18:2n-6 20:2n-6 20:3n-6 20:4n-6 Total n-6

0.6 t t 36.8 37.4

8.2 1.0 0.7 8.5 18.4

9.2 1.2 0.7 11.3 22.4

4.8 1.1 1.2 20.8 27.9

4.3 0.7 0.2 33.6 38.8

5.9 0.6 t 35.3 42.1

18:3n-3 20:4n-3 20:5n-3 22:5n-3 22:6n-3 Total n-3 Total PUFA

t 0.3 5.0 1.1 5.5 11.9 49.3

0.6 t 5.7 0.9 16.1 24.0 42.4

0.5 0.3 3.1 0.4 16.3 20.8 43.2

0.2 t 2.3 0.3 9.6 12.4 40.3

0.3 t 0.7 0.3 9.6 10.9 49.7

0.4 t 0.5 0.4 4.9 6.6 48.7

0.3 7.4 0.2

1.3 1.5 1.9

0.9 3.7 1.4

0.4 9.0 0.5

0.3 48.0 0.3

0.2 70.6 0.1

n-3/n-6 20:4/20:5 DHA/AA

Footnotes as described in Table 2.

less than those occurring in PC or PE (Table 4). AA was reduced in liver PS in fish fed diets 2-4 and increased in diets 5 and 6 (3- and 3.8-fold respectively) in comparison with diet I (Table 4). The percentage of 22:4(n-6) was increased 7-fold in diet 6 fish compared to those fed diet 1. DHA was increased in liver PS in fish fed diets 2-6 compared to those fed diet 1. The (n-3)/(n-6) PUFA ratio was increased in fish fed diets 2-4 and decreased in diets 5 and 6 compared to fish fed diet 1 while the 20:4/20:5 ratio was increased in diets 5 and 6 compared to diets 1-4. The percentages of AA in liver PI were lower in fish fed diets 2-6 compared to those fed diet 1 (Table 5). The reduction ranged from 4.3-fold in diet 2 to only 1.04-fold in diet 6. The DHA level was increased in fish fed diets 2-5 and slightly reduced in diet 6 compared to fish fed diet 1. The (n-3)/ (n-6) PUFA ratio of liver PI was relatively unaltered although diet 2 and 3 fish had a higher ratio than the other dietary treatments. The percentage of EPA decreased considerably from diet 2 to diet 6 with increasing incorporation of AA. The effect of this inverse relationship caused a large increase

in 20:4/20:5 ratio from diet 2 to diet 6 in liver PI (Table 5). The percentages of AA, EPA and DHA in brain phospholipids are shown in Fig. 1. In brain PC, the percentage of AA was lower in fish fed diets 2-4 but up to 2-fold higher in diets 5 and 6 compared to fish fed diet 1. The DHA levels of brain PC from fish fed diets 2-6 were all similar and lower than that in fish fed diet 1. As with PC the percentage of AA in brain PE was slightly reduced in fish fed diets 2-4 and increased in diets 5 and 6 (2- and 4-fold respectively), compared to fish fed diet 1. The percentage of DHA in brain PE decreased from diet 2 to diet 6, reflecting dietary input but even in fish fed diet 2 the level of DHA was only 84% of that in fish fed fish oil (diet 1). Brain PS composition showed the least changes of any phospholipid class with only fish fed diets 5 and 6 showing increased percentages of AA (5- and 8-fold respectively) and 22:4(n-6) (2- and 4.5-fold respectively) compared to fish fed diet 1. In brain PI, fish fed diets 2-4 had slightly reduced percentages of AA and diets 5 and 6 had increased percentages of AA compared to fish fed

146

i

'e

'

0 o i

e.

e

§

PC

PE

PS

PI

Phospholipid class

Phospholipid class

Fig. 1. Effects of diets with varying DHA/AA ratio on the relative contents of 20:4(n-6), 20:5(n-3) and 22:6(n-3) in the major phospholipids in turbot brain. Values are expressed as percentage of total weight and were obtained from the analyses of 18 brains per dietary treatment.

Fig. 2. Effects of diets with varying DHA/AA ratio on the relative contents of 20:4(n-6), 20:5(n-3) and 22:6(n-3) in the major phospholipids in turbot gills. Values are expressed as percentage of total weight and were obtained from the analyses of gills from 18 fish per dietary treatment.

*, Diet 1;

· , Diet 1;

, diet 2;

, diet 3; , diet 4;

, diet 5; 0, diet 6.

diet 1. As with liver PI, the percentage of EPA in brain PI decreased with increasing dietary AA. The percentages of AA, EPA and DHA in phospholipids from gill, kidney and heart are shown in Figs. 2-4. The dietary induced alterations in these tissues were largely similar to those described for liver and brain phospholipids. Heart homogenates from fish fed diet 2 con-

, diet 2;

, diet 3;

, diet 4; [D, diet 5;

, diet 6.

tained significantly lower PGE2, compared to fish fed diet 5 and 6 while fish fed diets 1-4 had significantly lower PGE 2 compared to those fed diet 6 (Table 6). In brain homogenates, fish fed diet 6 had significantly greater levels of PGE 2, compared to all other diets. In kidney, fish fed diets 2 and 3 had significantly lower levels of PGE2, compared to those fed diets 1 and 6. The concentrations of

147 .A

0

-

o

g e.

C

e

u

PC

PE

PS

PI

PC

Phospholipid class

PE

PS

PI

Phospholipid class

Fig. 3. Effects of diets with varying DHA/AA ratio on the relative contents of 20:4(n-6), 20:5(n-3) and 22:6(n-3) in the major phospholipids in turbot kidney. Values are expressed as percentage of total weight and were obtained from the analyses of 18 kidneys per dietary treatment.

Fig. 4. Effects of diets with varying DHA/AA ratio on the relative contents of 20:4(n-6), 20:5(n-3) and 22:6(n-3) in the major phospholipids of turbot heart. Values are expressed as percentage of total weight and were obtained from the analyses of 18 hearts per dietary treatment.

*, Diet 1;

*, Diet 1;

, diet 2;

, diet 3; ,

diet 4;

1, diet 5; 0, diet 6.

PGE 2 in gills were not significantly different between dietary treatments (Table 6). However, gill homogenates from fish fed diet 3 had significantly lower 6-ketoPGFia, compared to fish fed diets 4 and 6 (Table 7). In kidney homogenates fish fed diet 3 had significantly lower 6-ketoPGFIa, compared to fish fed diets 4 and 5. In contrast, the concentrations of 6-ketoPGF]a in heart and brain were sig-

, diet 2;

, diet 3; 0, diet 4; [D, diet 5; 0E, diet 6.

nificantly different between dietary treatments (Table 7).

Discussion Previous studies established that the long-chain (n-3) HUFA, EPA and DHA, are essential for

148 1 Table 6. PGE 2 concentrations (pg mg- tissue) in tissue homogenates from turbot fed diets with varying DHA/AA ratio.

Diet Tissue

I (control)

2

3

4

5

6

Heart Brain Gill Kidney

17.4 9.9 2.3 6.8

2.2a 11.3 9.3 0.6a 2.6 + 1.9 1.1 + 0.3b

18.2 + 3.4 ab 10.3 + 2.3a 1.4 0.6 1.5 + 0.6b

21.6 + 1.7ab 8.4 0.6a 4.3 + 2.4 3.0 + .Oab

9.4bC 36.5 8.8 ± 1.5 a 4.0 ± 0.2 4.9 + 0.9ab

29.1 23.5 4.4 6.0

+ + + ±

2.8ab 3.7a 1.1 2.6a

+ + + +

3.8c 1.7b 0.9 2.4a

Values are mean + SD for three samples per treatment; each sample was obtained by homogenising the tissues from four individual fish; values in the same row with different superscript letters are significantly different (p < 0.05). l Table 7. 6-ketoPGF,, concentrations (pg mg- tissue) in tissue homogenates from turbot fed diets with varying DHA/AA ratio.

Diet Tissue

I (control)

2

Heart Brain Gill Kidney

3.2 1.5 2.8 3.2

3.2 0.9 1.0 3.3

± + + +

0.7 0.8 1.6ab 1.6ab

+ + ± +

1.6 0.6 0.3ab 1.3ab

5.3 3.8 0.7 1.8

5

4

3 + + + +

1.4 1.4 0.5b 0.8a

4.3 3.4 3.4 6.8

+ + + ±

3.5 3.1 1.0a 2.0b

5.1 1.9 1.9 6.0

6 + + + +

2.1 0.9 0.9ab 1.6b

5.5 6.6 3.5 5.7

+ + + +

1.8 4.5 0.4a 1.2ab

Footnotes as described in Table 6.

growth and development of turbot (Gatesoupe et al. 1977). In fish of around 85 g the minimum level of (n-3) HUFA was 0.6% of the dry diet (Leger et al. 1979), whereas in larval turbot fed rotifers the requirement was increased to 1.3% of the dry diet (Le Milinaire et al. 1983). In these studies the requirement was established as a mixture of EPA and DHA, but Bell et al. (1985a) showed that turbot grown with EPA alone had higher mortalities than those given supplemental DHA. Although the limited capacity of turbot to elongate and desaturate 18:3(n-3) to EPA has been long established (Cowey et al. 1976), the ability of live turbot and isolated turbot brain cells to convert EPA to DHA has only recently been confirmed (Linares and Henderson 1991; Tocher et al. 1992). However, the inability of dietary EPA to maintain levels of DHA in juvenile turbot suggests that the bioconversion of EPA to DHA may be inadequate and that DHA may have a greater essential fatty acid efficacy than EPA (Mourente and Tocher 1992). For this season the purified diets in this study utilized pure DHA as essential (n-3) HUFA. In liver, heart and kidney, fish fed diets 2 and 3 (containing 0.93 and 0.86% of

dry diet as DHA) had levels of DHA in membrane phospholipids which were greater than or equal to the levels in fish fed fish oil (diet 1). However, in brain and gill the levels of DHA in PC and PE of fish fed diet 2 were considerably less than those in fish fed fish oil which suggests that 0.93% DHA may be insufficient to satisfy requirements for this fatty acid in juvenile turbot. Recent studies have suggested that high levels of the di-DHA molecular species of PE occurring in neural tissues of marine fish may have an important physiological role in membrane function (Bell and Dick 1991). In general, the EPA content of all phospholipid classes from fish fed diets 2-6 are considerably lower than those fed diet 1. This suggests that retroconversion of DHA to EPA may be limited in juvenile turbot. Studies using an established cell line derived from turbot fin suggest that limited retroconversion of DHA to EPA does occur (Tocher et al. 1989; Tocher and MacKinlay 1990). It is possible that dietary DHA levels in the present experiment were not sufficiently elevated to stimulate the retroconversion pathway. While it is now generally accepted that marine

149 fish have an absolute requirement for (n-3) HUFA no requirement for the (n-6) series HUFA has yet been established. However, analyses of turbot phospholipids have established that AA is present and, as in mammals, is especially abundant in the PI fraction (Bell et al. 1985b). AA is the major precursor of eicosanoids in mammalian tissues (Johnston et al. 1983) and although EPA is also a precursor for eicosanoid production, the EPAderived species generally have a much lower biological activity, than the AA homologue (Lands 1989). AA-derived eicosanoids have been detected in a wide range of fish species and tissues (Mustafa and Srivastava 1989) and indeed AA was found to be the prefered substrate for eicosanoid production in plaice neutrophils, despite EPA being more abundant in membrane phospholipids (Tocher and Sargent 1987). These facts suggest that marine fish should have a specific requirement for dietary AA in addition to that for (n-3) HUFA. In the present study, diet 2 was formulated to be depleted in AA and analysis confirmed that the concentration of AA was 0.02% of the dry diet, compared with the fish oil-containing control diet (diet 1) which had 9 times as much AA. Diets 3-6 had increasing amounts of pure AA added giving a range of concentrations from 0.06 to 0.78% by weight of the dry diet. The incorporation of AA increased in all phospholipid classes in line with dietary levels, with the largest percentage increases in PE and PC, although the highest levels were in PI. In general, PS was less affected by dietary AA than the other phospholipid classes. Fish fed diets 2, 3 and 4 generally had substantial reductions in the percentages of AA in tissue phospholipids, especially PI, compared to fish fed fish oil. Fish fed the two highest dietary levels of AA also accumulated the elongation product, 22:4(n-6), in tissue phospholipids, particularly PS and PE. Fish fed diets 2-6 also had increased levels of 20:2(n-6), compared to fish fed diet 1. The 20:2(n-6) is normally regarded as a "dead end" product of 18:2(n-6) elongation and that is almost certainly the case in this experiment. Diets 2-6 did not contain more than 0.2% 20:2(n-6), the same as in diet 1, indicating that the 20:2(n-6) was produced in the turbot tissues by elongation of the dietary

18:2(n-6). The greater percentages of the elongated product of 18:2(n-6), 20:2(n-6), found in this study compared to the much lower levels of the desaturated products, 18:3(n-6) and 20:3(n-6), are consistent with the existing data showing elognation generally predominates over desaturation in turbot (Tocher and McKinlay 1990; Tocher 1993). However, even in fish fed the highest dietary AA level, no increase in the A4-desaturation product, 22:5(n-6), was observed. These results support the recent observation that elongation activity predominates over desaturase activity in turbot (Tocher 1993). Fish brain PI is unusual in containing a greater concentration of EPA relative to AA when compared to PI in other tissues (Tocher and Harvie 1988; Bell and Dick 1991). In the present study, although brain phospholipid fatty acid composition was generally less altered than other tissues by dietary fatty acid input, brain PI showed increased AA incorporation and concomitantly decreased levels of EPA, with increasing dietary AA. A similar inverse relationship between AA and EPA incorporation was observed in PI from liver, gill and heart. The tendency for PI to accumulate high concentrations of C 20 PUFA has implicated this phospholipid class with a role in supplying precursors for eicosanoid synthesis (Bell et al. 1983). Thus the changes in phospholipid AA and EPA in the present experiment could alter the spectrum and amount of eicosanoids produced. PGE 2 concentrations in tissue homogenates were generally reduced (although not significantly in all tissues) in fish fed diets 2 and/or 3, and increased in fish fed diet 6 (and often diet 5). In brain, only fish given the highest dietary AA had significantly different concentrations of PGE 2 compared to the other dietary treatments. The relative lack of effect of the other dietary treatments on PG production may be partly explained by the specificity of the antibody used in the immunoassay, which is equally active with PGE 3 as with PGE2 . Although not so statistically significant, generally similar trends were apparent in the concentrations of 6ketoPGFla, the stable metabolite of prostacyclin, in tissue homogenates. In general, diets containing depleted levels of AA reduced tissue concentrations

150 of prostaglandins, while diets high in AA increased prostaglandin concentrations, compared to 'control' fish fed fish oil. Although many of the physiological functions of prostaglandins are as yet unknown, a number of important roles have been assigned. In rainbow trout, PGE 2 affects fluid and electrolyte balance in the kidney by decreasing glomerular filtration rate and increasing renal water reabsorption (Brown and Bucknall 1986). The tubular antidiuretic action probably reflects a direct effect of PGE 2 on renal tubule adenylate cyclase resulting in increased water permeability (Brown and Bucknall 1986). PGE 2 and prostacyclin are generally vasodilatory in mammals and the former prostaglandin has been shown to be hypotensive in rainbow trout. PGE2 injection also causes bradycardia in the same species (Brown and Bucknall 1986). PGE 2 is inhibitory to muscle fibre formation and also stimulates protein degradation in mammals, while PGF 2a appears to have the opposite effect (Palmer 1990). PGE 2 is also involved in reproductive functions and regulation of the immune response in fish (Mustafa and Srivastava 1989). In conclusion, the present study has demonstrated that phospholipid AA contents can be substantially influenced by dietary AA levels in juvenile turbot. The altered tissue phospholipid AA levels are reflected in changes in prostaglandin production which could influence a wide range of physiological processes important to the growth and general health of the fish.

References cited Ackman, R.G. 1980. Fish Lipids, part 1. In Advances in Fish Science and Technology. pp. 86-103. Edited by J.J. Connell. Fishing News Books, Farnham. Bell, J.G., McVicar, A.H., Park, M.T. and Sargent, J.R. 1991. High dietary linoleic acid affects the fatty acid compositions of individual phospholipids from tissues of Atlantic salmon (Salmo salar): association with stress susceptibility and cardiac lesion. J. Nutr. 121: 1163-1172. Bell, J.G. and Sargent, J.R. 1992. The incorporation and metabolism of polyunsaturated fatty acids in phospholipids of cultured cells from chum salmon (Oncorhynchus keta). Fish Physiol. Biochem. 10: 99-109. Bell. M.V., Simpson, C.M.F. and Sargent, J.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. 1985a. Effects of dietary polyunsaturated fatty acid deficiences on mortality, growth and gill structure in the turbot (Scophthalmus maximus, Linnaeus). J. Fish Biol. 26: 181191. Bell, M.V., Henderson, R.J. and Sargent, J.R. 1985b. Changes in the fatty acid composition of turbot (Scophthalmus maximus) in relation to dietary polyunsaturated fatty acid deficiences. Comp. Biochem. Physiol. 81B: 193-198. Bell, M.V. and Dick, J.R. 1991. Molecular species composition of the major diacyl glycerophospholipids from muscle, liver, retina and brain of cod (Gadus morhua). Lipids 26: 565-573. Brown, J.A. and Bucknall, R.M. 1986. Antiduiretic and cardiovascular actions of prostaglandin E2 in the rainbow trout Salmo gairdneri. Gen. Comp. Endocrinol. 61: 330-337. Castell, J.D., Bell, J.G., Tocher, D.R. and Sargent, J.R. 1994. Effects of purified diets containing different combinations of arachidonic and docosahexaenoic acid on survival, growth and fatty acid composition of juvenile turbot (Scophthalmus maximus). Aquaculture 128: 315-333. Christie, W.W. 1982. Lipid Analyses, 2nd Edition, Pergamon Press, Oxford. Cowey, C.B., Owen, J.M., Adron, J.W. and Middleton, C. 1976. Studies on the nutrition of marine flatfish. The effect of different fatty acids on the growth and fatty acid composition of turbot (Scophthalmus maximus). Br. J. Nutr. 36: 479-486. Folch, J., Lees, M. and Sloane-Stanley, G.H. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226: 497-509. Gatesoupe, F.J., Leger, C., Metailler, R. and Luquet, P. 1977. Alimentation lipidique du turbot (Scophthalmus maximus L.) I. Influence de la longeur de chaine des acides gras de la serie wo3. Ann. Hydrobiol. 8: 89-97. Henderson, R.J. and Tocher, D.R. 1987. The lipid composition and biochemistry of freshwater fish. Prog. Lipid Res. 26: 281-347. Johnston, M., Carey, F. and McMillan, R.M. 1983. Alternative pathways of arachidonate metabolism: prostaglandins, thromboxanes and leukotrienes. In Essays in Biochemistry. Vol. 19, pp. 40-141. Edited by P.N. Campbell and R.D. Marshall. Academic Press, London. Kanazawa, A. 1985. Essential fatty acid and lipid requirement of the fish In Nutrition and Feeding in Fish. pp. 281-298. Edited by C.B. Cowey, A.M. Mackie and J.G. Bell. Academic Press, London. Lands, W.E.M. 1989. Differences in n-3 and n-6 eicosanoid precursors. In Advances in Prostaglandin, Thromboxane and Leukotriene Research. Vol. 19, pp. 602-605. Edited by B. Samuelsson, Wong, P.Y.K. and F.F. Sun. Raven Press, New York. Leger, C., Gatesoupe, F.J., Metailler, R., Luquet, P. and Fremont, L. 1979. Effect of fatty acids differing by chain

151 lengths and o series on the growth and lipid composition of turbot, Scophthalmus maximus L. Comp. Biochem. Physiol. 64B: 345-350. Le Milinaire, C., Gatesoupe, F.J. and Stephen, G. 1983. Approche du besoin quantitatif en acides gras longs polyinsaturates de la serie n-3 chez la larve de turbot (Scophthalmus maximus) C.R. Acad. Sci., Paris 296: 917-920. Linares, F. and Henderson, R.J. 1991. Incorporation of 14 Clabelled polyunsaturated fatty acids by juvenile turbot, Scophthalmus maximus (L.) in vivo. J. Fish Biol. 38: 335347. Moon, T.W., Walsh, P.J. and Mommsen, T.P. 1985. Fish hepatocytes: a model metabolic system. Can. J. Fish Aquat. Sci. 42: 1772-1782. Mourente, G. and Tocher, D.R. 1992. Effect of weaning onto a pelleted diet on docosahexaenoic acid (22:6n-3) levels in brain of developing turbot (Scophthalmus maximus L.). Aquaculture 105: 363-377. Mustafa, T. and Srivastava, K.C. 1989. Prostaglandins (eicosanoids) and their role in ectothermic organisms. Adv. Comp. Env. Physiol. 5: 157-207. Owen, J.M., Adron, J.W., Middleton, C. and Cowey, C.B. 1975. Elongation and desaturation of dietary fatty acids in turbot Scophthalmus maximus and rainbow trout Salmo gairdneri. Lipids 10: 528-531. Palmer, R.M. 1990. Prostaglandins and the control of muscle protein synthesis and degradation. Prostaglandins, Leukotrienes and Essent. Fatty Acids 39: 95-104. Powell, W.S. 1982. Rapid extraction of arachidonic acid metabolites from biological samples using octadecyl silica. Methods Enzymol. 86: 467-477. Sargent, J.R., Henderson, R.J. and Tocher, D.R. 1989. The Lipids In Fish Nutrition 2nd Edition. pp. 153-218. Edited by J.E. Halver, Academic Press, San Diego.

Terano, T., Salmon, J.A., Higge, G.A. and Moncada, S. 1986. Eicosapentaenoic acid as a modulator of inflammation. Effect on prostaglandin and leukotriene synthesis. Biochem. Pharmacol. 35: 779-785. Tocher, D.R. 1993. Elongation predominates over desaturation in the metabolism of 18:3n-3 and 20:5n-3 in turbot (Scophthalmus maximus) brain astroglial cells in primary culture. Lipids 28: 267-272. Tocher, D.R. and Sargent, J.R. 1987. The effect of calcium ionophore A23187 on the metabolism of arachidonic and eicosapentaenoic acids in neutrophils from a marine teleost fish rich in (n-3) polyunsaturated fatty acids. Comp. Biochem. Physiol. 67B: 733-739. Tocher, D.R. and Harvie, D.G. 1988. Fatty acid composition of the major phosphoglycerides from fish neural tissues; (n-3) and (n-6) polyunsaturated fatty acids in rainbow trout (Salmo gairdneri) and cod (Gadus morhua) brains and retinas. Fish Physiol. Biochem. 5: 229-239. Tocher, D.R., Carr, J. and Sargent, J.R. 1989. Polyunsaturated fatty acid metabolism in fish cells: Differential metabolism of (n-3) and (n-6) series acids by cultured cells originating from a freshwater teleost fish and from a marine teleost fish. Comp. Biochem. Physiol. 94B: 367-374. Tocher, D.R. and MacKinlay, E.E. 1990. Incorporation and metabolism of (n-3) and (n-6) polyunsaturated fatty acids in phospholipid classes in cultured turbot (Scophthalmus maximus) cells. Fish Physiol. Biochem. 8: 251-260. Tocher, D.R., Mourente, G. and Sargent, J.R. 1992. Metabolism of [1-1 4C] docosahexaenoate (22:6n-3). [1- 14C] eicosapentaenoate (20:5n-3) and [1-1 4 C] linolenate (18:3n-3) in brain cells from juvenile turbot Scophthalmus maximus. Lipids 27: 494-499. Vitiello, F. and Zanetta, J.-P. 1978. Thin layer chromatography of phospholipids. J. Chromatogr. 166: 637-640.