Diets rich in eicosapentaenoic acid and γ-linolenic acid affect phospholipid fatty acid composition and production of prostaglandins E1, E2 and E3 in turbot (Scophthalmus maximus), a species deficient in Δ5 fatty acid desaturase

PROSTAGLANDINSLEUKOTRIENES AND ESSENTIALFATTYACIDS Prostaglandins Leukotrienes and Essential Fatty Acids (1995) 53, 279-286 © Pearson Professional Ltd 1995

Diets Rich in Eicosapentaenoic Acid and y-Linolenic Acid Affect Phospholipid Fatty Acid Composition and Production of Prostaglandins El, E2 and E3 in Turbot (Scophthalmus maximus), a Species Deficient in A5 Fatty Acid Desaturase J. G. Bell, D. R. Tocher, F. M. MacDonald and J. R. Sargent

NERC Unit of Aquatic Biochemistry, Department of Biological and Molecular Sciences, University of Stirling, Stirling FK9 4LA, UK (Reprint requests to JGB)

ABSTRACT. Duplicate groups of juvenile turbot, (Scophthalmus maximus), were fed diets containing either Marinol K (MO), a marine fish oil rich in eicosapentaenoic acid (EPA; 20:5, n-3) or borage oil (BO), rich in 7linolenic acid (GLA; 18:3, n-6), for a period of 12 weeks. Individual phospholipid fatty acid compositions from hearts of fish fed BO had significantly more 18:2, n-6, GLA, 20:2, n-6, dihomo-7-1inolenic acid (DHGLA; 20:3, n-6) and total n-6 polyunsaturated fatty acids (PUFA), but significantly less arachidonic acid (AA; 20:4, n-6), compared to fish fed MO. In both phosphatidylcholine (PC) and phosphatidylethanolamine (PE) from heart, the DHGLA was increased by over 50-fold in fish fed BO while AA was reduced by over two-thirds, compared to fish fed MO. In brain, EPA was the major C20 PUFA, i.e. potential eicosanoid precursor in all phospholipids from fish fed MO, with the EPA level being twice that of AA in brain phosphatidylinositol (PI). DHGLA was the major C20 PUFA in all phospholipid classes from fish fed BO. In kidney and gill, EPA was the predominant C20 PUFA in all phospholipid classes, except PI, in fish fed MO. In kidney of fish fed BO, DHGLA was the major Cz0 PUFA in all phospholipid classes, except PE. In gill of fish fed BO, DHGLA was the major C~0 PUFA in all phospholipid classes, including P1, where DHGLA was over 2.5-fold greater than AA. In homogenates of heart, kidney and gill from BO-fed fish the prostaglandin E 1 (PGE1) concentration was significantly increased compared to MO-fed fish. In heart and kidney homogenates from fish fed MO the PGE 3 concentration was significantly increased compared to fish fed BO. The ratio of PGEJPGE1 was significantly reduced in brain, heart, kidney and gill homogenates from fish fed BO compared to those fed MO.

derived homologues so that increasing tissue levels of DHGLA and EPA can reduce production and efficacy of 2-series PGs (3, 4). PGs have been found in a wide range of fish species (5) and all major fish tissues possess cyclooxygenase activity (6). AA-derived prostanoids have been synthesized from both exogenous and endogenous precursors in numerous fish species (7) and prostanoids have also been generated from exogenous DHGLA (8). While no evidence exists for prostanoid formation from endogenous DHGLA, prostaglandin E2 (PGE2) production from exogenous AA can be reduced by addition of exogenous DHGLA (9). Production of thromboxane B 3 (TXB3) from exogenous EPA and PGE3 from endogenous EPA has been recorded in rainbow trout and in carp and sheat-fish, respectively (10, 11). In a recent study we demonstrated that production of PGs E and F of the 1-, 2- and 3-series, from endogenous precursors, could be measured in turbot astroglial cells in primary culture after supplementation with DHGLA,

INTRODUCTION The enzymatic pathway involved in eicosanoid production is often called the 'arachidonic acid cascade' since arachidonic acid (AA; 20:4, n-6) is the major eicosanoid precursor in mammalian cells, giving rise to 2-series prostaglandins (PGs) (1). In addition, dihomo7-1inolenic acid, (DHGLA; 20:3, n-6) and eicosapentaenoic acid (EPA; 20:5, n-3) are substrates for eicosanoid production and yield PGs of the 1- and 3series, respectively, although they are generally poorer substrates for prostaglandin synthetase compared to AA (2). However, both DHGLA and EPA can compete for the enzyme binding site and can reduce production of AA-derived PGs (3). The 1- and 3-series PGs generally have lower biological activity compared to their AADate received 7 March 1995 Date accepted 4 April 1995 279

280 ProstaglandinsLeukotrienesand EssentialFatty Acids AA and EPA (12). However, the predominant PGs formed when astrocytes were stimulated with the calcium ionophore A23187 were of the 2-series, regardless of polyunsaturated fatty acid (PUFA) supplement. The supply of precursor PUFA for eicosanoid synthesis is directly related to the fatty acid composition of membrane phospholipids, which in turn is influenced by dietary PUFA intake (13). It is now established that the high incidence of atherothrombotic, inflammatory and autoimmune conditions prevalent in Western society is related, in part, to over production of AA-derived eicosanoids and that these conditions can be attenuated by dietary supplementation with fish oils, rich in EPA, or plant oils, rich in 7-1inolenic acid (GLA; 18:3, n-6) the precursor of DHGLA (14, 15). Fish offer a potentially useful model system for the study of eicosanoid interactions since they are naturally abundant in long-chain n-3 PUFA and also have significant levels of AA in their phospholipids, especially PI (7, 16). Turbot are particularly interesting because, in common with all marine fish studied to date, they lack A5-desaturase activity and are therefore unable to convert DHGLA to AA or 20:4, n-3 to EPA (17, 18). However, dietary studies feeding safflower oil and linseed oil and experiments using turbot astrocytes in primary culture supplemented with 18:2, n-6 and 18:3, n-3, suggest that turbot possess an active elongase resulting in accumulation of 20:2, n-6 and 20:3, n-3 into all phospholipid classes (19, 20). In the present study duplicate tanks of juvenile turbot were fed diets containing Marinol K (MO), an EPA-rich fish oil, or borage oil (BO), a 7-1inolenic acid-rich plant oil for 12 weeks. We hoped to exploit the lack of A5desaturase and the high elongase activity to increase incorporation of DHGLA into turbot tissues by feeding BO and to increase tissue EPA using MO. The fatty acid composition of membrane phospholipids was determined in heart, kidney, gill and brain mad the concentrations of PGE1, PGE2 and PGE 3 measured in homogenates from the same tissues.

MATERIALS AND METHODS Animals and diets 252 juvenile turbot obtained from Golden Sea Produce, Hunterston, Scotland, of initial mean weight 1.4 g were distributed randomly into 4 tanks. The circular 1 m diameter tanks contained 500 1 of seawater which was partially recirculated and supplied at a rate of 5 1/min. The tanks were subjected to natural photoperiod and the water temperature was maintained at 17 +_ I°C. Diets were supplied by automatic feeders adjusted to provide 80 g/kg biomass per day initially, and reduced to 60 g/kg per day when average fish weight reached 5 g. Diets were fed for 7 days per week and fish were bulk weighed every 28 days and the ration adjusted accordingly. The diets, utilizing low-temperature fishmeal and casein

as protein sources, contained 50% protein and 20% lipid and have been described in detail previously (19). The diet contained fishmeal (400 g/kg) (LT 94, Ewos Ltd, Westfield, Lothian, Scotland), casein (240 g&g), precooked starch (150 g/kg), a-cellulose (11.5 g/kg), vitamin mix (10 g/kg), mineral mix (24 g/kg) and choline chloride (4 g/kg). The lipid component (160 g/kg) was either MO (Fishing Industry Research Unit, Rosebank, 7700 Republic of South Africa) or BO a gift from Croda Universal Ltd, Hull, UK. The antioxidant ethoxyquin (0.5 g/kg) was mixed with the oil before mixing with the dry ingredients. The fatty acid compositions of the two diets are given in Table 1.

Sampling procedure Fish were killed by severing the spinal column posterior to the brain and samples of heart, gill, brain and kidney, for lipid analysis, were dissected from 9 fish per tank (18 fish per dietary treatment) and frozen immediately in liquid nitrogen. A further 8 fish per tank were killed, to provide samples of heart, gill, brain and kidney for PGE analysis, and these were combined to give 4 samples of 4 fish per dietary treatment. Each sample was immediately homogenized in 4 ml of Hank's balanced salt solution containing 0.6 ml of absolute ethanol and

Table 1 Fattyacid compositionsof diets (values are weight %) Fatty acid

Marinol diet (MO)

Borage oil diet (BO)

14:0 16:0 18:0 Total saturates~ 16:1, n-7 18:1, n-9 18:1, n-7 20:1, n-9 22:1, n-ll 22:1, n-9 24:1 Total monoenes2 16:2 16:3 16:4 Total 16C PUFA 18:2, n-6 18:3, n-6 20:2, n-6 20:3, n-6 20:4, n-6 Total n-6 18:3, n-3 18:4, n-3 20:4, n-3 20:5, n-3 22:5, n-3 22:6, n-3 Total n-3 Total PUFA n-3/n-6

7.8 15.8 2.2 26.7 9.1 7.7 2.6 3.3 3.6 0.4 1.5 28.6 1.2 1.5 3.4 6.1 1.3 0.4 t t 0.6 2.6 0.6 2.9 0.7 17.7 1.1 9.0 32.0 34.6 12.3

1.0 9.9 2.8 14.2 1.5 13.6 1.0 5.0 1.7 1.8 2.1 26.8 0.1 0.1 0.1 0.3 30.5 19.7 0.2 t 0.1 50.7 2.5 0.6 0.1 1.7 0.1 2.4 7.4 58.1 0.2

qncludes 15:0, 17:0, 20:0 and 22:0. ~Includes20:1, n-ll and 20:1, n-7. t = trace value < 0.05%.

Effect of Dietary EicosapentaenoicAcid and 7-LinolenicAcid in the Turbot 281 0.2 ml of 2 M formic acid and the homogenate frozen in liquid nitrogen.

Lipid extraction and analysis Extraction of total lipid from tissue samples and diets was performed by the method of Folch et al (21). Separation of total lipid into phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS) and phosphatidylinositol (PI) fractions by thin-layer chromatography (TLC) and subsequent preparation of fatty acid methyl esters has been described in detail previously (22). The fatty acid methyl esters were separated and quantified by gas-liquid chromatography (GLC) (Carlo Erba Vega 6000, Fisons Ltd, Crawley, UK) using a 50 m x 0.22 mm capillary column (Free fatty acid phase, S.G.E. Ltd, Milton Keynes, UK). Hydrogen was used as carrier gas and temperature programming was from 50°C to 160°C at 35°C/min and then to 225°C at 1.8°C/min. Individual methyl esters were identified by comparison with known standards and by reference to published data (23, 24).

Extraction, separation and enzyme immunoassay of PGE isomers The frozen tissue homogenates for eicosanoid analysis were thawed and centrifuged at 1000 g for 5 rain to precipitate cell debris. The supernatants were extracted using octadecyl silyl (C18) 'Sep-Pak' minicolumns (Millipore (UK), Watford) by the method of Powell (25) and as described in detail by Bell et al (19). The final extract was redissolved in 100 gl of methanol prior to high-performance liquid chromatography (HPLC). The isomers of PGE were separated by reverse-phase HPLC largely as described in Bell et al (12). An isocratic solvent system was employed containing 17 mM phosphoric acid/acetonitrile (70/30, v/v) at a flow rate of 1 ml/min. The elution times of the three PGE standards were determined by UV detection at 196 nm using a Pye-Unicam LC-UV detector. 50 g l of the eicosanoid homogenate extracts was injected onto the column and 1 ml fractions were collected using an LKB 2112 'Redirac'. Fractions corresponding to each PGE isomer were pooled and extracted as follows. The pooled fractions were applied to a C18 'Sep-Pak' which had been pre-washed with 5 ml methanol and 10ml distilled water. The column was then washed with a further 10 ml of distilled water and the PGE eluted in 5 ml of ethyl acetate. Samples were dried under nitrogen and redissolved in immunoassay buffer. Measurement of PGE isomers was performed using an enzyme immunoassay (EIA) kit for PGE 2 according to the manufacturers' protocol (SPI-Bio, Gif sur Yvette, France).

Statistical analysis Significance of difference between dietary treatments

(p <0.05) was determined by analysis of variance (ANOVA). Analyses were performed using a Stargraphics (system 3.0) computer package. Data which were identified as non-homogeneous (using Bartlett's test) were subjected to either arcsin square root or log transformation before analysis. Differences between means were determined by Tukeys' test.

Materials TLC plates (20 cm x 20 cm x 0.25 mm), pre-coated with silica gel 60 (without fluorescent indicator), were obtained from Merck (Darmstadt, Germany). PGE~, PGE~ and PGE 3 (all > 98%) were obtained from Cascade Biochemicals Ltd (Reading, UK). All solvents were of HPLC grade and were obtained from Rathburn Chemicals Ltd (Walkerburn, Scotland).

RESULTS The juvenile turbot grew well on both experimental diets and average weights increased by over 8-fold in the 12-week trial period. The fatty acid compositions of PC and PE from heart are shown in Table 2. In heart PC total monoenes were significantly increased in fish fed MO compared to those fed BO. Levels of 18:2, n-6, GLA, 20:2, n-6, DHGLA and total n-6 PUFA in heart PC were all significantly higher, but AA and 22:5, n-6 were significantly lower in fish fed BO, compared to those fed MO. The fish fed MO contained significantly higher levels of all n-3 PUFAs, except 18:3, n-3, in heart PC compared to fish fed BO. EPA was the major n-3 PUFA in heart PC of fish fed MO. The total PUFA and ratio of AA/EPA in heart PC were not significantly different between dietary treatments whereas the n-3/n-6 and the AA/DHGLA ratios were both significantly greater in fish fed MO compared to those fed BO. The dietary induced changes in fatty acid compositions of heart PE were broadly similar to those in PC except that the AA/EPA ratio was significantly greater while the percentage of dimethyl acetals, derived from plasmalogens, was significantly lower in fish fed BO compared to fish fed MO. In both heart PC and PE the DHGLA level was increased by over 50-fold in fish fed BO while the level of AA was reduced by over two-thirds, compared to fish fed MO. The fatty acid compositions of PS and PI from heart are shown in Table 3. In heart PS the amount of total saturated fatty acids was significantly greater in BO-fed fish compared to MO-fed fish. The levels of 18:2, n-6, GLA, 20:2, n-6, DHGLA and total n-6 PUFA were all significantly greater in heart PS of BO-fed fish while 22:5, n-6 was significantly reduced and AA was unaltered compared to MO-fed fish. As with PC and PE, the levels of all n-3 PUFA in heart PS, except 18:3, n-3, were significantly lower in BO-fed compared to MO-fed fish. The ratio of n-3/n-6 PUFA and AA/DHGLA were

282

Prostaglandins Leukotrienes and Essential Fatty Acids Table 2 Fatty acid compositions (weight %) of phosphatidylcholine (PC) and phosphatidylethanolamine (PE) from hearts of turbot fed different dietary oils PC

PE

Fatty acid Marinol diet

Borage oil diet

Marinol diet

Borage oil diet

Total saturates Total monoenes

35.7 + 1.1 a 18.1 + 2.1

30.5 + 2.1 b 18.0 + 2.5

19.4 + 2.4 13.5 + 1.0

22.0 + 1.0 13.3 + 2.0

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

0.7 0.2 ta ta 2.2 0.2 3.5

+ 0.1 ~ + 0.0 a

0.4 a + 0.1 ~ + 0.5 a

24.5 7.1 1.2 2.7 0.4 tb 35.9

+ + + + +

0.5 ta 0.3 t~ 1.9 0.5 3.3

+ 0.2 ~ + 0.1 + 0.3 ~

19.4 3.2 1.1 2.4 0.6 0.2 27.0

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

0.2 0.3 0.3 19.6 1.7 19.0 41.0

_+0.1 ~ + 0.1 ~ + 0.1 a + 2.1 ~ + 0.2 a _ 2.0 a + 3.7 a

1.0 tb tb 2.9 0,4 9.3 13.5

+ 0.2 b

+ + + +

0.8 b 0.1 b 2.6 b 3.6 b

+ + + + +

0.1 ~ 0.5 ~ 0.4 ~ 3.2 ~ 3.8"

0.6 + 0.1 b 0.6 _+0.2 b tb 2.7 + 0.3 b 1.0 + 0.1 b 29.2 + 4.3 b 34.3 + 4.8 b

Total PUFA n-3/n-6 20:4/20:5 20:4/20:3 Dimethyl acetals

44.4 12.1 0.11 39.7 -

+ + + +

49.5 0.4 0.14 0.15 -

+ + + +

5.9 0.1 b 0.01 0.03 b

3.8 2.0 a 0.02 ~ 17.8 ~ 0.4 ~

61.4 1.3 0.22 0.24 2.8

-

3.4 2.7" 0.02 25.9 a

1.2b 1.3b 0.2 b 0.2 b 0.1 b

+ 2.5 b

t~ t~ 0.3 11.7 4.1 41.0 57.2

+ 0.1 a + 0.1 ~

60.5 + 17.6 + 0.16 + 37.7 + 5.1 +

+ + + + + + +

3.2b 0.1 b 0.2 b 0.1 b 0.0 b 0.2 3.26

+ 2.4 _+0.3 b + 0.02 b + 0.01 b + 0.1 b

Values are means + SD from three samples per dietary treatment. Each sample comprised 6 hearts, t, trace value (< 0.05%) ; -, not detected; PUFA, polyunsaturated fatty acids. Values in the same row, for each phospholipid, having a different superscript letter are significantly different (p < 0.05).

Table 3 Fatty acid compositions of phosphatidylserine (PS) and phosphatidylinositol (PI) from hearts of turbot fed different dietary oils PS

PI

Fatty acid Marinol diet

Borage oil diet

Total saturates Total monoenes

29.9 + 1.3" 17.4 + 4.2

35.2 + 0.5 b 16.0 + 0.4

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

0.7 ta 0.5 ta 0.7 0.8 2.7

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

t

Total PUFA n-3/n-6 20:4/20:5 20:4/20:3

Marinol diet

Borage oil diet

Weight%

± 0.2 a

+ 0.1 + 0.1 a + 0.3 a

10.9 3.1 1,3 4.4 0.7 0.2 20.4

+ + + + + + +

1.8b 0.4 b 0.1 b 0.2 b 0.1 0.1 b 2.7 b

3.3 4.3 38.1 45.7

_+ 1.2 a + 0.5 a + 3.0 a + 4.7 ~

0.3 + 1.0 + 1.3 + 25.4 + 27.9 +

48.4 17.3 0.23 20.0

+ + + +

48.3 1.4 0.70 0.2

+ 0.2 a

4.5 3.1 ~ 0.10 ~ 5.4 a

+ + + +

3.4 + 1.4 13.5 + 1.2 0.5 + 0.1 a

34.5 + 6.1 14.1 + 1.1

0.5 + 0.3 ta 18.6 + 2.0 a t 19.6 +_2.2 a

14.6 2.5 1.2 5.9 9.3 t 33.4

+ 3.4 b

0.2 0.3 b 0.3 b 4.3 b 4.2 b

t" 19.3 0.9 8.7 28.9

+ 2.1" +_0.1" _+4.3 +_2.4 ~

1.0 7.9 0.3 6.4 15.6

-+ 0.6 b + 2.2 b + 0.2 b + 1.6 + 3.1 b

1.9 0.4 b 0.30 b 0.1 b

48.5 1.5 1.0 500.0

_+0.8 +_0.3 a _+0.0 + 50.0 ~

49.0 0.5 1.2 1.6

+ + + +

-a

+ 5.3 b _+0.4 b + 0.4 + 1.5b + 2.0 b

5.7 0.1 b 0.1 0.5 b

Footnotes as described in Table 2.

significantly higher while the AA/EPA

ratio was signifi-

cantly lower in fish fed MO compared

to those fed BO.

The percentages

of DHGLA,

AA

and EPA

in indi-

vidual phospholipids of brain, kidney and gill are repre-

The fatty acid compositions of heart PI characteristically

s e n t e d i n F i g u r e s 1, 2 a n d 3, r e s p e c t i v e l y . I n b r a i n , E P A

showed

w a s t h e m a j o r C20 P U F A

the highest levels of AA for any phospholipid

and potential eicosanoid pre-

class. However, in fish fed MO the level of EPA was in

cursor in all phospholipids

excess of AA while all three potential eicosanoid precur-

E P A l e v e l b e i n g t w i c e t h a t o f A A i n b r a i n P I ( F i g . 1).

sors, DHGLA,

EPA

and AA, were present in PI in the

ratio 1:1.3:1.6, respectively.

from fish fed MO, with the

In contrast, DHGLA

was the major eicosanoid precursor

in all phospholipid

classes from brain of fish fed BO,

Effect of Dietary Eicosapentaenoic Acid and "~-Linolenic Acid in the Turbot

283

20

ca

16 12

II 20:3n-6 [] 20:4n-6 I"1 20:5n-3

4 0

PC

PE

PS

PI

Marinol K diet

PC

PE

PS

PI

Borage oil diet

Fig. 1 DHGLA, AA and EPA compositions (weight %) of individual phospholipid classes from brain of turbot fed either MO or BO. 20:3, n-6, I ; 20:4, n-6, Ill; 20:5, n-3, [Z.

with AA being the lowest eicosanoid precursor, after DHGLA and EPA, in fish fed BO. In kidney, EPA was the major eicosanoid precursor in all phospholipid classes of fish fed MO, except in PI, where AA was over twice the value of EPA (Fig. 2). In kidney of fish fed BO, DHGLA was the predominant eicosanoid precursor in all phospholipid classes, except PE, where EPA was the largest value. In gill, EPA was the predominant eicosanoid precursor in all phospholipid classes of fish fed MO, except in PI, where the AA level was slightly higher than EPA (Fig. 3). In gill of fish fed BO, DHGLA was the major eicosanoid precursor in all phospholipid classes, including PI, where the DHGLA level was over 2.5-fold greater than AA.

The concentrations of PGE 1, PGE2 and PGE3 in homogenates of heart, brain, kidney and gill are shown in Tables 4-7, respectively. In heart homogenates the concentration of PGE 1 was significantly greater in fish fed BO than in fish fed MO (Table 4). However, the converse was true for PGE s, with the fish fed MO having a higher concentration compared to those fed BO. The ratio of PGEJPGE 1 was significantly reduced in BO-fed fish compared to MO-fed fish. In brain homogenates only the ratio of PGEJPGE1 was significantly different between dietary treatments (Table 5). In kidney homogenates the concentration of PGE 1 was significantly higher and PGE 3 significantly lower in fish fed BO compared to those fed MO (Table 6). The ratio

24 20 16 ~D ¢D ~0

12 [] 20:3n-6 [] 20:4n-6 [] 20:5n-3

8

4 1 0 l_ PC

_

PE

PS

M a r i n o l K diet

PI

PC

PE

PS

PI

Borage oil diet

Fig. 2 DHGLA, AA and EPA compositions (weight %) of individual phospholipid classes from kidney of turbot fed either MO or BO. 20:3, n-6, I ; 20:4, n-6, I ; 20:5, n-3, 7q.

284

Prostaglandins Leukotrienes and Essential Fatty Acids

14 !,

1211 0~

0~

&

4~[; PC

[] 20:3n-6 [] 20:4n-6 [] 20:5n-3

[ L t PE

PS

PI

MarinolK diet

PC

PE

PS

PI

Borageoil diet

Fig. 3 DHGLA, AA and EPA compositions (weight %) of individual phospholipid classes from gill of turbot fed either MO or BO. 20:3, n-6, II; 20:4, n-6, II; 20:5, n-3, D.

Table4

Concentrations of PGE l, PGE2 and PGE3 in homogenates of heart from turbot fed diets containing Marinol or borage oil

Table6

Concentrations of PGE 1, PGE2 and PGE3 in homogenates of kidney from juvenile turbot fed diets containing Marinol or borage oil

Diet Prostaglandin PGE1 PGE2 PGE3 PGE]PGE3 PGEz/PGE1

Marinol diet

Borage oil diet

1.00 +_0.10" 3.22 _+ 1.21 1.20 + 0.12a 2.7 +_0.7 3.4 _+ 1.5"

2.33 + 0.84u 1.78 ± 0.15 0.39 +_0.18b 5.2 ± 2.2 0.8 ! 0.2b

Values are pg/mg tissue and are means +_SD from 4 samples per dietary treatment. Each sample is composed of tissues from 6 individual turbot. Values in the same row having different superscript letters are significantly different (p < 0.05).

Table5

Concentrations of PGE 1, PGE2 and PGE3 in homogenates of brain from juvenile turbot fed diets containing Marinot or borage oil

Diet Prostaglandin

Marinol diet

Borage oil diet

PGE1 PGE2 PGE3 PGEz/PGE3 PGEz/PGE1

0.06 _+0.02a 0.41 + 0.09 0.29 -+ 0.01 a 1.4 + 0.3 6.7 ± 0.4a

0.48 _+0.02b 0.32 _+0.05 0.16 ± 0.02b 2.0 -+ 0.3 0.7 -+ 0 . 1 b

Values are pg/mg tissue and are means +_SD from 4 samples per dietary treatment. Each sample is composed of tissues from 6 individual turbot. Values in the same row having different superscript letters are significantly different (p < 0.05).

Table7

Concentrations of PGE 1, PGE2 and PGE3 in homogenates of gill from turbot fed diets containing Marinol or borage oil

Diet

Diet

Prostaglandin

Marinol diet

Borage oil diet

Prostaglandin

Marinol diet

Borage oil diet

PGE I PGE2 PGE3 PGEJPGE3 PGEz/PGEa

0.33 +_0.06 3.67 -+ 0.99 0.29 -+ 0.03 12.7 + 2.1 11.3 + 3.9"

0.87 +_0.38 2.28 -+ 1.11 0.26 -+ 0.13 9.3 -+ 1.5 2.7 +- 0.9b

PGE1 PGE2 PGE3 PGEz/PGE3 PGEz/PGE1

0.04 _+0.01 ~ 0.98 -+ 0.25 0.21 ! 0.09 5.2 ± 2.2 26.7 _-.2.4a

0.21 ~ 0.08 b 0.85 _+0.08 0.12 -+ 0.01 7.5 + 1.3 4.7 + 2.1b

Values are pg/mg tissue and are means +_SD from 4 samples per dietary treatment. Each sample is composed of tissues from 6 individual turbot. Values in the same row having different superscript letters are significantly different (p < 0.05).

Values are pg/mg tissue and are means _+SD from 4 samples per dietary treatment. Each sample is composed of tissues from 6 individual turbot. Values in the same row having different superscript letters are significantly different (p < 0.05).

o f P G E J P G E 1 w a s s i g n i f i c a n t l y r e d u c e d in f i s h f e d B O c o m p a r e d to f i s h f e d M O . In gilt h o m o g e n a t e s the

the results o f p r e v i o u s studies that w h i l e turbot are defic i e n t in A 5 - d e s a t u r a s e activity t h e y p o s s e s s an active

c o n c e n t r a t i o n o f P G E 1 w a s s i g n i f i c a n t l y g r e a t e r in f i s h

e l o n g a s e (17, 19). F e e d i n g B O r e s u l t e d in a c c u m u l a t i o n

f e d B O a n d c o n s e q u e n t l y the ratio o f P G E a / P G E 1 w a s s i g n i f i c a n t l y r e d u c e d in B O - f e d fish c o m p a r e d to f i s h

n o t o n l y o f 18:2, n-6 and G L A f r o m the diet, but o f

f e d M O (Table 7).

the e l o n g a t i o n p r o d u c t D H G L A . A t the s a m e t i m e the d e f i c i e n c y o f A5 d e s a t u r a s e in this f i s h s p e c i e s e n a b l e d high levels of D H G L A

to a c c u m u l a t e in m e m b r a n e

DISCUSSION

phospholipids while the concentration of A A remained low. In f i s h f e d M O , the l e v e l o f E P A in tissue p h o s p h o l i p i d s w a s i n c r e a s e d c o m p a r e d to fish f e d B O

F e e d i n g diets c o n t a i n i n g B O , r i c h in G L A , c o n f i r m e d

a l t h o u g h the l e v e l s o f A A w e r e also g r e a t e r c o m p a r e d

Effect of Dietary EicosapentaenoicAcid and y-LinolenicAcid in the Turbot 285 to BO-fed fish, reflecting the difference in dietary intake. In heart, gill and kidney phospholipids from fish fed MO, the highest levels of EPA were found in PC, with lower levels in PE and PI. However, in brain of fish fed MO EPA levels were highest in PI, being double the value of AA while in heart PI, EPA was also the predominant C20 PUFA. The occurrence of EPA-rich PI in brain has been observed previously in a number of fish species (26, 27) but in general fish tissues contain AA-rich PI, similar to mammals, which contrasts with the high EPA/AA ratio found in other phospholipids (7, 16). In the tissue phospholipids of fish fed BO, GLA was highest in PC, whereas its elongation product, DHGLA, was incorporated predominantly into PI. In brain, kidney and especially gill, DHGLA was the most abundant C20 PUFA, being more than 2-fold in excess of AA in the latter tissue. The DHGLA-rich P1 in tissues of turbot fed BO has not been observed previously in whole fish but has been observed in turbot fin cells in culture which were supplemented with GLA and DHGLA (28). The ability of PI to selectively accumulate C20 PUFA suggests that this phospholipid class may be a source of precursors for the synthesis of eicosanoids (23). The prostaglandins PGE1, PGE2 and PGE3, derived from DHGLA, AA and EPA, respectively, were measured in homogenates of heart, brain, kidney and gill. In all tissues the increase in PGE1 concentration (up to 8-fold) in fish fed BO compared to fish fed MO was greater than the increase in PGE 3 (up to 3-fold) in fish fed MO compared to those fed BO. This may be partly explained by the very low levels of DHGLA in the MO-fed fish while the BO-fed group still contained appreciable levels of EPA but may also reflect the ability of prostaglandin synthetase to utilize fatty acids other than AA. While the ability of both DHGLA and EPA to competitively inhibit the binding of AA at the enzyme binding site has been long established (3), it is also known that both DHGLA and EPA are inferior substrates for PG production compared to AA (2). In fish, which naturally have an abundance of EPA compared to AA in their membrane phospholipids, the preferred substrate for eicosanoid production is still AA (29, 30). In a recent study using turbot astrocytes in culture, which had been supplemented with either DHGLA, AA or EPA, the predominant PG produced on stimulation with calcium ionophore was PGE2, regardless of the supplemented PUFA (12). However, supplementation with either DHGLA or EPA significantly inhibited PGE z production in turbot astrocytes compared to unsupplemented or AA-supplemented astrocytes (12). The ability of both DHGLA and EPA to attenuate the production of AAderived eicosanoids is fundamental in the control of pathophysiological processes prevalent in numerous inflammatory conditions occurring in human populations (4, 15). In the present study, feeding BO reduced the mean concentration of PGE2 in all tissues examined. While the reductions were not statistically significant, the fact that levels of series 2 PG were reduced by feed-

ing BO in comparison to a diet containing very high levels of EPA, which itself would tend to reduce PGE2 concentrations, is worthy of note. Although the changes in fatty acid composition resulting from both dietary treatments resulted in extensive alterations in the profile of eicosanoid precursors in the membrane phospholipids, the effect on PGE concentrations was much less dramatic. Only in heart and kidney did the levels of PGE 1 exceed those of PGE 2 and despite the high EPA concentrations in all membrane phospholipids the PGE 3 concentration was never in excess of the AA-derived product. This clearly emphasizes the functional importance of series 2 PGs in the overall eicosanoid hierarchy but also the tissue differences observed may indicate the physiological roles of PGE analogues within different cell systems. The ability of fish kidney and gill to synthesize PGs has been established for some time, although their functions were largely unknown (8, 9). However PGE2 and PGE1 are both active antidiuretic agents in trout and gill kidney (31, 32) while PGE2 inhibits ion transport in gill epithelia (33). Thus, the tendency for a particular tissue to produce different PG analogues may reflect the efficacy of those analogues within the physiological process over which they have regulatory control. In turbot brain, in comparison to the other tissues examined, PGE2 is by far the major PGE analogue produced, despite the fact that EPA is the major eicosanoid precursor in phospholipids from fish fed MO and DHGLA the major precursor in phospholipids from fish fed BO. This supports work done in this laboratory using turbot astrocytes in culture, which demonstrated the predominance of AA-derived PGs, even when cellular lipids contained over 6-fold more DHGLA or EPA than AA (12). Studies on mammalian brain confirm the presence of specific PGE2 receptors, in a number of regions, which suggest that this PG is involved in modulation of neural transmission, nociception and hypothalmic function (34, 35). The present study has demonstrated that a marine teleost fish, like the turbot, which has a low A5 desaturase activity, and high elongase activity, can be exploited to produce membrane phospholipid fatty acid compositions which are rich in DHGLA or EPA, yet low in AA. This creates a particularly interesting and potentially useful model system in which to study the production and interaction of eicosanoids and their precursors.

Acknowledgements This work was supportedby a grant (AQ.2.590) from the Commission of the European Communities,Directorate of Fisheries (DG XIV), as part of the community research programme in the fisheries sector (FAR). We would like to thank all S.O.A.F.D. personnel involvedin running the Fish CultivationUnit, Aultbea, Scotlandfor their expertise and assistance with fish husbandry.

References 1. HorrobinD F. The regulation of prostaglandin

286

Prostaglandins Leukotrienes and Essential Fatty Acids

biosynthesis by the manipulation of essential fatty acid metabolism. Rev Pure Appl Sci 1983; 4: 339-383. 2. Crawford M A. A background to essential fatty acids and their prostanoid derivatives. Br Med Bull 1983; 39: 210-213. 3. Willis A L. Nutritional and pharmacological factors in eicosanoid biology. Nutr Rev 1981; 39: 289-301. 4. Weber P C. The modification of the arachidonic acid cascade by n-3 fatty acids. In: Samuelsson B, Dahlen S -E, Fritsch J, Hedqvist P, eds. Advances in prostaglandin, thromboxane and leukotriene research, vol. 20. New York: Raven Press, 1990: 233-240. 5. Mustafa T, Srivastava K C. Prostaglandins (eicosanoids) and their role in ectothen-nic organisms. Adv Comp Env Physiol 1989; 5: 157-207. 6. Tocher D R. Glycerophospholipid metabohsm. In: Hochachka P W, Mommsen T P, eds. Biochemistry and molecular biology of fishes; vol. 4. Metabolic and adaptational biochemistry. Amsterdam: Elsevier, 1995: 119-157. 7. Henderson R J, Tocher D R. The lipid composition and biochemistry of freshwater fish. Prog Lipid Res 1987; 26:281-347. 8. Ogata H, Nomura T, Hata M. Prostaglandin biosynthesis in the tissue homogenates of marine animals. Bull Jpn Soc Sci Fish 1978; 44: 1367-1370. 9. Christ E J, van Dorp D A. Comparative aspects of prostaglandin biosynthesis in animal tissues. Biochim Biophys Acta 1972; 270: 537-545. 10. Kayama M, Sado T, Iijima N. The prostaglandin synthesis in rainbow trout thrombocyte. Bull Jpn Soc Sci Fish 1986; 52: 925. 11. Nomura T, Ogata H. Distribution of prostaglandins in the animal kingdom. Biochim Biophys Acta 1976; 431: 127-131. 12. Bell J G, Tocher D R, Sargent J R. Effect of supplementation with 20:3(n-6), 20:4(n-6) and 20:5(n-3) on the production of prostaglandins E and F of the 1-, 2and 3-series in turbot (Scophthalmus maximus) brain astroglial cells in primary culture. Biochim Biophys Acta 1994; 1211: 335-342. 13. Lands W E M. Differences in n-3 and n-6 eicosanoid precursors. In: Samuelsson B, Wong P Y K, Sun F F, eds. Prostaglandin, thromboxane and leukotriene research; vol. 19. New York: Raven Press, 1989: 602~505. 14. Simopoulos A P, Kifer R R, Wykes A A. co 3 fatty acids: Research advances in support of the field since June 1985. In: Simopoulos A P, Kifer R R, Martin R E, Barlow S E, eds. Health effects of ~0 3 fatty acids in seafoods. Basle: Karger Press, 1991: 51-73. 15. Horrobin D F. Nutritional and medical importance of gamma-linolenic acid. Prog Lipid Res 1992; 31: 163-194. 16. Sargent J R, Henderson R J, Tocher D R. The lipids. In: Halver J E, ed. Fish nutrition, 2nd edn. San Diego: Academic Press, 1989: 153-218. 17. Owen J M, Adron J A, Middleton C, Cowey C B. Elongation and desaturation of dietary fatty acids in turbot Scophthalmus maximus and rainbow trout Salmo gairdneri. Lipids 1975; 10: 528-531. 18. Bell M V, Henderson R J, Sargent J R. Changes in the fatty acid composition of turbot (Scophthalraus maximus) in relation to dietary polyunsaturated fatty acid deficiencies. Comp Biochem Physiol 1985; 81B: 193-198. 19. Bell J G, Tocher D R, MacDonald F M, Sargent J R. Effects of diets rich in linoleic (18:2n-6) and c~-linolenic (18:3n-3) acids on the growth, lipid class mad fatty acid compositions and eicosanoid production in juvenile turbot

(Scophthalmus maximus L.) Fish Physiol Biochem 1994; 13: 105-118. 20. Tocher D R. 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 1993; 28: 267-272. 21. Folch J, Lees M, Sloane-Stanley G H. A simple method for the isolation and purification of total lipid from animal tissues. J Biol Chem 1957; 226: 497-509. 22. Bell J G, McVicar A H, Park M T, Sargent J R. 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 1991; 121: 1163-1172. 23. Bell M V, Simpson C M F, Sargent J R. (n-3) and(n-6) polyunsaturated fatty acids in the phospholglycerides of salt-secreting epithelia from two marine fish species. Lipids 1983; 18: 720-726. 24. Ackman R G. Fish lipids, part 1. In: Connell J J, ed. Advances in fish science and technology. Famham: Fishing News Books, 1980: 86-103. 25. Powell W S. Rapid extraction of arachidonic acid metabolites from biological samples using octadecyl silica. Methods Enzymol 1982; 86: 467-477. 26. Tocher D R, Harvie D G. 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 1988; 5: 229-239. 27. Mourente G, Tocher D R. Effects of weaning onto a pelleted diet on docosahexaenoic acid (22:6n-3) levels in brain of developing turbot (Scophthalmus maximus L.) Aquaculture 1992; 105: 363-377. 28. Tocher D R, MacKinlay E E. Incorporation and metabolism of (n-3) and (n-6) polyunsaturated fatty acids in phospholipid classes in cultured turbot (Scophthalmus maximus) cells. Fish Physiol Biochem 1990; 8: 251-260. 29. Anderson A A, Fletcher T C, Smith G M. Prostaglandin biosynthesis in the skin of the plaice, Pleuronectes platessa L. Comp Biochem Physiol 1981; 70C: 195-199. 30. Tocher D R, Sargent J R. 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 1987; 87B: 733-739. 31. Wales N A M, Gaunt T. Haemodynamic, renal and steroidogenic actions of prostaglandins E b 132,A2, and F2~ in European eels. Gen Comp Endocrinol 1986; 62: 327-334. 32. Brown J A, Gray C J, Hattersley G, Robinson J. Prostaglandins in the kidney, urinary bladder and gills of the rainbow trout and European eel adapted to freshwater and seawater. Gen Comp Endocrinol 1991; 84: 328-335. 33. Van Pragg D, Farber S J, Minkin E, Primor N. Production of eicosanoids by the killifish gills and opercular epithelia and their effect on active transport of ions. Gen Comp Endocrinol 1987; 67: 50-57. 34. Galli C, Petroni A. Eicosanoids and the central nervous system. Upsala J Med Sci 1990; 48: 133-144. 35. Marriot D, Wilkin G P, Coote P R, Wood J N. Eicosanoid synthesis by spinal cord astrocytes is evoked by substance P; possible implications for nociception and pain. In: Samuelsson B, Ramwell P W, Paoletti R, Folco G, Granstrom E, eds. Advances in prostaglandin, thromboxane, and leukotriene research; vol. 21. New York: Raven Press, 1~990:739-741.