Lipid classes and fatty acid composition in female gonads of great scallops--a selective field study.

ABSTRACT Fatty acid composition was analyzed in neutral lipids (NL), phosphatidyletanolamine (PE) and phosphatidylcholine (PC) of Pecten maximus (L.) ovaries from Western Norway. Individual ovaries were selected based on macroscopic characterization. Two series were compared; (1) ovary development from recovering to half-full ovaries during the autumnal restoration of the gonads and (2) seasonal variation with late filling ovaries from different pairs of the year. The relative lipid content and the proportion of triacylglycerols increased during development. Changes in fatty acid composition during ovary development were minor compared with the changes observed with seasonal variation. The samples from June differed from the other samples with a large increase in 20:5n-3 on the expense of 22:6n-3 in neutral lipids (NL) and phosphatidylcholine (PC) and a concomitant increase in 20:5n-3 and 22:6n-3 in phosphatidyletanolamine (PE). The changes in NL and PC can probably be attributed to a dietary effect from the diatom dominated spring phytoplankton bloom. Only small changes in the degree of unsaturation in PC were seen that may be due to a temperature effect. The differences observed represent large differences in status of broodstock from field prior to conditioning in season independent hatchery production.

KEY WORDS Fatty acids, lipid composition, ovaries, Pecten maximus, phospholipids, scallops, triacyglycerols

INTRODUCTION

The importance of lipids for the quality of eggs and larvae of bivalves is well established. Broodstock diet in particular influences the lipid composition in the gonads and eggs (Helm et al. 1991, Utting & Doyou 1992, Soudant et al. 1996a, Soudant et al. 1996b, Soudant et al. 1996c, Soudant et al. 1999). Special attention has been given to the polyunsaturated fatty acids 22:6n-3, 20:4n-6 and to a lesser but still important extent 20:5n-3. These are selectively incorporated in the phospholipids during gonad build up, despite deficiencies in diet (Soudant et al. 1996a). Lipid composition of the ovaries and eggs then affects larval performance. 22:6n-3 has been demonstrated as essential for growth and survival of bivalves (Langdon & Waldock 1981). Higher levels of 22:6n-3 and a higher 22:6n-3/20:5n-3 ratio were associated with improved hatching rate and enhanced gametogenesis in Pecten maximus (Soudant et al. 1996a, Soudant et al. 1996b), while growth rate was significantly related to the content of 22:6n-3 in oyster larvae (Thompson & Harrison 1992, Berntsson et al. 1997). Another factor known to influence membrane lipid composition in poikiloterms is temperature. Lipid class and fatty acid composition is regulated according to the temperature to maintain optimal fluidity of the membranes (Williams & Hazel 1994), and dietary input of essential fatty acids may be required for cold acclimation (Farkas et al. 1980). Increased unsaturation of the phospolipids or increase in sterol content at low temperatures is also reported for bivalves (Ueda 1974, Piretti et al. 1988, Chu & Greaves 1991, Napolitano et al. 1992).

Seasonal changes in lipid composition of ovaries have been reported in several studies of scallops (Besnard et al. 1989, Napolitano & Ackman 1992, Pazos et al. 1997). Changes in lipid composition have been related to reproductive activity, but the effects of season and stage of maturity have not been separated. Scallops in the field experience large fluctuations in fond composition from various phytoplankton blooms, known to differ in fatty acid composition (Budge et al. 2001) which gives imprints in filter feeders (Pedersen et al. 1999, Shirai et al. 2002). Temperature fluctuations in the field can also affect lipid composition of bivalves (Ueda 1974, Piretti et al. 1988). In seasonal independent hatchery production of P. maximus in Norway, broodstock animals are taken from the sea for conditioning in the hatchery at different times of the year. Variations in lipid composition of animals from the field will hence represent differences in status prior to conditioning.

Field data are often important basis for evaluating the quality of artificial conditioning and other aquacultural manipulations (e.g. Soudant et al. 1999). A study was performed to describe variations in lipid composition in ovaries from a P. maximus field population. From a parallel study, large variations in ovary developmental stages were seen during the reproductive cycle of this population (Duinker & Nylund 2002). Individual ovaries were therefore selected based on macroscopic characteristics of ovary development, to get a more precise description of lipid composition in the different ovary developmental stages than a population average would have given. Two series of ovaries were selected, one consisting of ovaries in the same stage of development from different times of the year, and one with successive developmental stages during the autumn restoration of the gonads. The aim of this study is to compare seasonal variations in lipid class and fatty acid composition with the variations during development.

MATERIALS AND METHODS

Sampling Procedures

Individual ovaries were sampled monthly in combination with studies of seasonal variation in chemical composition of ovaries and storage organs in Pecten maximus (Strohmeier et al. 2000) and histologic and visual characterizations of the ovaries (Duinker & Nylund 2002). The scallops were collected in Raunefjorden south of Bergen, Norway, and ovary samples were stored at -80[degrees]C. Ovaries were selected based on visual characterization as described by Duinker & Nylund (2002). The characterization is based on the study of Mason (1958), and the stages sampled for ovary development were (numerical values in brackets); "recovering" (3), "filling" (4), "late filling" (4.5), and "half full" (5). These ovaries were sampled on September 15, October 18, December 12, 1997, and February 7, 1998, respectively (Table 1). Filling (4) ovaries from February 15, 1997 and half-full (5) ovaries from March 12. 1997 were sampled as spring controls for comparison with the corresponding stages sampled during autumn. Only changes that were consistent between the samples in both series were considered as related to ovary development. Late filling (4.5) ovaries were selected for seasonal variation from the March 31, June 8, December 12, 1997, and February 7, 1998 samples. The samples for the late filling stage in the ovary development series and the seasonal variation sample from December were identical.

Analytical Procedures 

The ovary samples were freeze dried and dry weight percentage was determined. After homogenizing the samples, lipids were extracted with chloroform-methanol (2:1, v/v), modified from Bligh and Dyer (1959) according to Ronnestad et al. (1995), and total lipid content was determined gravimetrically. Further, lipid classes were separated by HPLC as described by Lie and Lambertsen (1991) using a Constametric II solvent-delivery system at 205 nm with a variable-wavelength spectrophotometer (LDC Spectromonitor III). The column (25 cm x 0.46 cm I.D.) was packed with silica gel (LiChrosorb 5 [micro]m, Merck). The initial solvent mixture consisted of hexane:2-propanol:acetonitrile (4:3:7, v/v/v). This was added to the protocol of Lie & Lambertsen (1991) and eluted two unidentified peaks (1) whereas NL eluted with the front. The second solvent mixture of hexane:2-propanol:acetonitrile: water (372:496:97:35, by volume) eluted PE. PC was eluted with hexane-2-propanol-water (394:526:80, v/v/v). The identity of the isolated phospholipids was checked by TLC (Kiselgel 60 Merck), using phospholipid standards and a solvent system of ethyl acetate-n propanol-chloroform-methanol-0.25% aqueous KC1 (15:25:25:10:4.5, by volume).

The three tractions were collected manually into vials, saponified, and esterified in 12% BF3 in methanol, and the fatty acid composition of the different lipid classes was determined according to Lie and Lambertsen (1991). The methyl esters were separated using a Carlo Erba gas chromatograph ("cold on column" injection, 60[degrees]C for 1 min at 25[degrees]C [min.sup.-1] 160[degrees]C for 28 min at 25[degrees]C [min.sup.-1] , 190[degrees]C for 17 min at 25[degrees]C [min.sup.-1], 220([[degrees]C for min) equipped with a 50-m CP-sil 88 (Chromopack) fused silica capillary column (i.d. 0.32 mm, 0.20-[micro]m film) (Lie & Lambertsen 1991). The fatty acid composition was calculated using an integrator (Turbochrom Navigator, Version 6.1), connected to the GC and identification ascertained by standard mixtures of methyl esters (Nu-Chek, Elyian, USA). The quantification of lipid classes in recovering (3) and half-full (5) ovaries was performed using high-performance thin-layer chromatography (HPTLC) as described by Bell et al. (1993). Total lipid (10 [micro]g) was applied to a 10 x 20 cm HPTLC plate that had been prerun in hexane:diethyl ether (1:1 v/v) and activated at 110[degrees]C for 30 min. The plates were developed at 5.5 cm in methyl acetate: isopropanol: chloroform: methanol: 0.25% (w/v) aqueous KC1 (25:25:25:10:9, by volume) to separate phospholipid classes with neutral lipids running at the solvent front (Vitello & Zanetta 1978). After drying, the plates were developed fully in hexane: diethyl ether: acetic acid (80:20:2, v/v/v) to separate neutral lipids and cholesterol. Lipid classes were visualized by charring at 160[degrees]C for 15 min after spraying with 3% copper acetate (w/v) in 8% (v/v) phosphoric acid identified by comparison with commercially available standards. Lipid classes were quantified by scanning densiometry using a CAMAG TLC Scanner 3 and calculated using a integrator (WinCATS-Planar Chromatography, Version 1.2.0).

Statistics

The data for fatty acid composition in the lipid classes of recovering (3) and half-full (5) ovaries were compared using the Mann-Whitney test (Zar 1999). With the low n of three individuals per point these statistics are only used as guidance for interpreting the results. The relative fatty acid composition data of the lipid classes was analyzed using SIRIUS for Windows (Pattern recognition systems AS. Norway, Version 6.0). Principal component analysis (PCA) (Wold et al. 1987) was performed in the data matrix of the relative fatty acid compositions. The purpose of PCA is to express the main information in the variables by a lower number of variables, the so-called principal components (PC1, PC2, ...). A high positive or negative loading reveals a significant variable in the actual PCA model. Score plots from the PCA explore the main trends in the data, and their respective loading reveal fatty acids with a significant loading. Basically, samples or variables with a similar angle from origin in the score and loading plots, respectively, are positively correlated. A 90[degrees] angle reflects no correlation and towards 180[degrees] reflects negative correlation. Variables (fatty acids) and samples can be compared in the same way, so that samples are positively influenced by variables with a similar angle from origin and so on.

RESULTS

The relative lipid content of dry weight in the ovaries increased linearly 2.6-fold during development from recovering (3) to half-full (5) ovaries. In contrast, there was no variation in relative lipid content throughout the season (Fig. 1). Total content of lipid per ovary increased 10-fold during development, whereas only some variation due to different sizes of the gonads were seen in the samples for seasonal variation, with no clear trends (see Fig. 1). Triacylglycerols (TAG) increased from 32% in recovering ovaries to 47% in half-full ovaries concomitant with a corresponding decrease in phospholipids (PL) and cholesterol (Fig. 2a). This corresponds to an increase in neutral lipids from 46% to 61% if only fatty acid containing classes are quantified, hence omitting cholesterol and adding up TAG and free fatty acids (FFA). Within the PL, only marginal differences were seen when comparing recovering (3) and half-full (5) ovaries (see Fig. 2b).

[FIGURES 1-2 OMITTED]

Multivariate principal component analysis (PCA) of the relative fatty acid composition of the neutral lipids (NL), phosphatidylcholine (PC) and phosphatidyletanolamine (PE) showed that each of the lipid classes analyzed has its characteristic fatty acid composition and are separated in the score plot (Fig. 3). The variables 22:6n-3 and 16:0 had high positive loadings along PC2, whereas 18:0 and 20:5n-3 had high negative loadings along PC1. Further, 22:6n-3 and 18:0 had high negative loadings along PC1, and 16:0 and 18:4n-3 had high positive loadings along PC1. These loadings correspond to the grouping of the PC, PE, and NL samples with PC containing high levels of 22:6n-3 followed by PE and NL. Furthermore, NL contained high levels of 18:4n-3 and lower levels of 22:6n-3 compared with the phospholipid classes. The level of 20:5n-3 was highest in PE; 16:0 was the dominant saturated fatty acid in PC whereas 18:0 was dominant in PE (see Fig. 3). During development, 20:5n-3 increased 1.6 fold, 1.5-fold and 1.9-fold in NL PE and PC, respectively (Fig. 4 and Table 2). The main increase was between the recovering (3) and filling (4) stages. During development, 22:6n-3 decreased slightly in NL, increased 1.3 fold in PE and showed no variance of significance in PC. The largest differences in fatty acid composition, however, were seen in ovaries from June compared with the other samples for seasonal variation. In June 20:5n-3 was markedly higher on the expense of 22:6n 3 in NL and PC (see Fig. 4), while both 20:5n-3 and 22:6n 3 were higher in PE. This shift resulted in a decrease in the 22:6n-3/20:5n-3 ratio between the other samples and June from 2.7 to 1 in PC and 1.2 to 0.5 in NL, while the ratio was relatively stable in PE.

[FIGURES 3-4 OMITTED]

The ratio of polyunsaturated/saturated fatty acids showed some differences with season in PC. though values from December overlapped with values from February. The ratio was higher in March (3.1) and February (2.8), compared with June (2.6) and December (2.5). The concurrent increase in 20:5n-3 and 22:6n-3 in June lead to higher ratio of polyunsaturated/saturated fatty acids in PE. Some minor changes in this ratio in NL were not consistent with the changes in PC or PE. In NL, the ratio of 16:1n-7/16:0 was higher in June (0.38) compared with the other samples for seasonal variation (0.15-0.25), and also 16:4n-3 was highest in June.

Dimetylacetals (DMA) from PE plasmalogens in the GC analyses were not identified. However, possible candidates with retention times slightly shorter than 18:0 (pers. comm. Y. Marty), never exceeded 5% of total fatty acids. Analysis of selected PE methyl ester fractions from the present study by Y. Marty (University of Brest, France) showed similar low content of DMA. Discrepancy with Soudant et al. (1995) and later works reporting higher levels of DMA may be attributed to incomplete derivation of alkenyl chains into DMA or differences between the populations.

DISCUSSION

Small changes were seen in the relative amounts of lipid classes and their respective fatty acid composition during development from recovering (3) to half-full (5) ovaries. This is consistent with the cellular changes during ovary development in P. maximus. In developing ovaries there is a discontinuous but ongoing recruitment of new cohorts of oocytes. A gradual accumulation of ripe oocytes is seen, with multiple cohorts of different development stages present at the same time (Paulet & Boucher 1991). Hence, changes in composition daring ovary development have limited relation to individual oocyte development, particularly during the earliest stages. In empty ovaries (stage 2), the connective tissue may have a substantial contribution to the lipid content relative to the young oocytes. Recovering (3) ovaries may have different cellular composition according to their history (Duinker & Nylund 2002). The recovering ovaries in the present study, more correctly termed as "almost spent", contained relatively few young oocytes in a mixture with developing and fully-grown oocytes (Duinker unpublished data). "Truly" recovering ovaries with numerous early vitellogenic oocytes in active development would better represent early oocyte development, but such ovaries were not found among the samples from September and October. The further stages of ovary development from filling (4) to half-full (5) also contain mixtures of several cohorts developing at the same time, but with increased accumulation of ripe oocytes. Hence, the changes in lipid composition during this development reflect a gradual increase in the proportion of ripe oocytes. If oocyte development had been a primary interest of the study, a size grading of oocytes after separation from the connective tissue would be a better approach. However, in this study the ovary development itself is of primary interest in comparison with seasonal changes.

During development from recovering (3) to half full (5) ovaries, both relative lipid composition and the total content per ovary increased concomitant with a corresponding increase in all lipid classes during the 10-fold increase in total lipid content. Triacylglycerols (TAG) was the dominant lipid class and increased from about 32% to 47% of total lipids between recovering (3) and half full (5) ovaries. The levels are comparable with other studies of scallops (Napolitano & Ackman 1992, Soudant et al. 1996c, Pazos et al. 1997). Increase in neutral lipids (NL) is found both during conditioning (Soudant et al. 1996a) and with maximum maturity in other field studies (Besnard et al. 1989, Napolitano & Ackman 1992, Pazos et al. 1997). This probably reflects the increased relative incorporation of NL in the later phases of oocyte development. NL are generally used as energy reserve in all life forms (Sargent 1976), and Gallager et al. (1986) found a strong decline in NL during embryogenesis prior to the onset of feeding in the larvae of Crassostrea virginica, Ostrea edulis, and Mercenaria mercenaria. However, also phospholipids (PL) were utilized in the latter study with some difference among the bivalve species, and Delaunay et al. (1992) suggested an energetic role of PL in P. maximus larvae. Within the PL, the small changes in relative amounts of the different classes during ovary development largely agree with Soudant et al. (1996c). However, these authors found an increase in PC from 29% to 43% on the expense of plasmalogens during conditioning of P. maximus, mainly from start to first sampling after 5 weeks of conditioning. These gonads had been emptied by induced spawning at start of the experiment, however, and were hence in an earlier stage at the start of the experiment than the recovering ovaries in the present study. Also in the fatty acid composition small changes were seen during ovary development, in agreement with Soudant et al. (1996c).

The changes in fatty acid composition during ovary development were minor, however, compared with the differences seen in the samples for seasonal variation. The samples from June differed from the samples in February, March, and December, with a marked increase in 20:5n-3 on the expense of 22:6n-3 in PC and NL One of the major events between March and June was the spring bloom, followed by the spring increase in temperature about a month later (Strohmeier et al. 2000). The spring bloom in the study area is dominated by the diatom Skeletomena costatum, whereas different dinoflagellates and coccolithophorids dominate the rest of the year with only short incidents of diatom blooms (Erga & Heimdal 1984, B.R. Heimdal and A.B. Reisegg pers. comm. 1996). From analyses of algae cultures it is seen that diatoms tend to be rich in 20:5n-3 and relatively deficient in 22:6n-3, whereas various flagellates tend to be rich in 22:6n-3 and relatively deficient in 20:5n-3 (Ackman & Tocher 1968, Chuecas & Riley 1969, Volkman et al. 1981, Ackman 1983, Servel et al. 1994. Dunstan et al. 1994, Berge et al. 1995, Bell & Pond 1996, Soudant et al. 1996a). Large differences are seen among species, but the generalization is supported by analyses of plankton communities and in imprints in herbivores. With changes in dominance of diatoms and flagellates, Budge et al. (2001) found highest levels of 22:6n-3 in seston associated with flagellate dominance, and imprints of the two different types of phytoplankton communities have been reported in zooplankton (Pedersen et al. 1999, Shirai et al. 2002). It is hence likely that the spring bloom in March to April in the present study represented a major input of 20:5n-3 in the scallops with a lack of 22:6n-3, compared with the rest of the year with more abundance of 22:6n-3 in the phytoplankton. The change was probably facilitated by an exchange of gonadal material between March and June, as partial spawnings during spring and early summer were followed by gonad growth (Strohmeier et al. 2000). The major increase in the 20:5n-3/22:6n-3 ratio in the samples from June can hence be attributed to a diet effect. Influence of diatoms in the June samples is supported by the observation of increased ratio of 16:1n-7/16:0 and higher proportion of 16:4n-3, known as indicators of diatoms (Ackman & Tocher 1968. Chuecas & Riley 1969, Napolitano & Ackman 1992), in NL. The ratio and level are not as high, though, as diatom imprints reported by Pedersen et al. (1999) and Napolitano et al. (1997). Similar changes are reported in other pectinids, with strong increase of 20:5n-3 in various tissues during diatom dominated spring blooms (Pollero et al. 1979, Napolitano et al. 1997). Pollero et al. (1979) found a marked increase in 22:6n-3, particularly in the gonads, associated with the dominance of a dinoflagellate in the stomachs of Chlamys tehuelca. The importance of diet for the lipid composition of ovaries has also been emphasized in several studies of hatchery conditioning of bivalves (Helm et al. 1991, Utting & Doyou 1992, Soudant et al. 1996a, Soudant et al. 1996b, Soudant et al. 1996c, Soudant et al. 1999).

Changes in the composition of phospholipids due to different temperatures could be expected in the present study, according to mechanisms of temperature adaptation of membrane lipids to maintain a constant or optimal fluidity of the membranes (Williams & Hazel 1994). The temperatures of 8[degrees]C to 10[degrees]C prior to the sampling in June and the 8[degrees]C to 9[degrees]C in December were higher than the winter minima for the March 1997 and February 1998 samples of 5[degrees]C to 6[degrees]C (Strohmeier et al. 2000). Different strategies for temperature adaptation are described, but usually the degree of unsaturation increases with lower temperatures (Williams & Hazel 1994), which is also seen in bivalves (Ueda 1974, Piretti et al. 1988, Chu & Grooves 1991). In Placopecten magellanicus, higher levels of 22:6n-3 in the phospholipids were found in the gonads at lower temperatures and higher depths, whereas higher content of a phytosterol was found in the adductor muscles (Napolitano et al. 1992). These sterols were not measured in the present study. The increased ratio of 20:5n-3/22:6n-3 in PC in June did not give a particular decrease in the degree of unsaturation itself. The fact that the same changes were seen in neutral lipids suggests that these changes were caused mainly by a diet effect as discussed earlier. However, the tendencies to elevated ratio of polyunsaturated to saturated fatty acids in PC in the colder months, relative to both the June and the December samples, may be related to temperature adaptation. This event was probably separate from the shifts in 20:5n-3 and 22:6n-3. In PE, there was an increase in the degree of unsaturation due to the coinciding increases in 20:5n-3 and 22:6n-3 in June (i.e., opposite to a possible effect on unsaturation due to temperature). It seems that the 22:6n-3/20:5n-3 ratio is more regulated in PE compared with PC.

Soudant et al. (1996a, 1996b) reported increased larvae survival under hatchery conditions associated with higher concentrations of 22:6n-3 and a higher 22:6n-3/20:5n-3 ratio in ovary and egg phospholipids, and similar conditions give better growth in oyster larvae (Thompson & Harrison 1992). Soudant et al. (1996b) further suggested that a 22:6n-3120:5n-3 ratio in diet greater than I was required to avoid deficiency, whereas a ratio greater than 2 is recommended for marine fish (Sargent 1995). Soudant et al. (1996b) also discussed the possibility that elevated levels of 20: 5n-3 can have a negative effect on eicosanoid production from 20:4n-6 as seen in fishes (Bell et al. 1994). The differences in 22:6n 3/20:5n-3 ratio that gave impact on egg quality in the study of Soudant et al. (1996c) ranged between 5 and 1 in PC, which was more than the corresponding range from 2.7 to 1 in the present study. The order of magnitude of the changes in the present study may hence not be large enough to affect egg quality. However, the changes probably represent an annual change in the lipid composition status of broodstock from the field. Andersen and Ringvold (2000) found seasonal differences in effect of different broodstock diets on spawning success. This may be related to similar changes in lipid composition status as seen in the present study.

TABLE 1.
Visualisation of the sampling dates for the ovary development
series and the samples for seasonal variation.

                              Ovary
              Ovary        development      Seasonal
           development   (spring control)   variation

Jan                             4
Feb                             5
Mar                                            4.5
Apr
May
Jun                                            4.5
Jul
Aug
Sep            3
Oct            4
Nov
1997 Dec       4.5                             4.5
1998 Jan
Feb            5                               4.5

The numbers represent the ovary development stages recovering,
(3); filling, (4); late filling, (4.5); and half full, (5).

TABLE 2.
Relative composition of fatty acids

                                 NL

                   Recovering            Half-full

14:0             4.5 [+ or -] 0.3    3.3 [+ or -] 0.2 *
15:0             1.0 [+ or -] 0.1    0.8 [+ or -] 0.1 *
16:0            18.2 [+ or -] 0.4   16.1 [+ or -] 2.0
17:0             3.1 [+ or -] 0.2    1.7 [+ or -] 0.2 *
18:0             3.1 [+ or -] 0.4    2.2 [+ or -] 0.3 *
Sum saturated   30.0 [+ or -] 0.1   24.1 [+ or -] 2.2 *
16:1n-7          2.5 [+ or -] 1.3    4.7 [+ or -] 0.3 *
16:1n-9          1.8 [+ or -] 1.5    0.9 [+ or -] 0.1
18:1n-7          3.6 [+ or -] 0.3    3.9 [+ or -] 0.4
18:1n-9          3.2 [+ or -] 0.2    5.3 [+ or -] 0.2 *
20:1n-7          0.5 [+ or -] 0.1    0.4 [+ or -] --
20:1n-9          0.9 [+ or -] 0.2    0.8 [+ or -] 0.1
20:1n-11         0.2 [+ or -] 0.1    0.3 [+ or -] --
22:1n-9           -- [+ or -] --      -- [+ or -] --
22:1n-11          -- [+ or -] --      -- [+ or -] --
Sum monoenes    12.6 [+ or -] 1.0   16.3 [+ or -] 0.9 *
18:2n-6          1.8 [+ or -] 0.1    2.2 [+ or -] 0.2 *
20:2n-6          0.6 [+ or -] 0.1    0.8 [+ or -] 0.2
20:4n-6          1.3 [+ or -] 0.4    1.3 [+ or -] 0.2
Sum n-6          3.8 [+ or -] 0.4    4.3 [+ or -] 0.2
16:3n-3          0.3 [+ or -] 0.1    0.1 [+ or -] 0.1
18:3n-3          2.9 [+ or -] 0.3    2.6 [+ or -] 0.3
18:4n-3          9.2 [+ or -] 1.3    8.4 [+ or -] 0.1
20:4n-3          0.6 [+ or -] 0.1    0.7 [+ or -] 0.1
20:5n-3         11.0 [+ or -] 0.7   17.3 [+ or -] 2.2 *
22:5n-3          0.6 [+ or -] --     0.7 [+ or -] --
22:6n-3         19.4 [+ or -] 0.3   16.3 [+ or -] 2.6 *
Sum n-3         44.0 [+ or -] 1.2   46.3 [+ or -] 4.3
n-3/n-6         11.0 [+ or -] 1.3   10.2 [+ or -] 1.4
Sum 16:1         4.3 [+ or -] 0.3    5.6 [+ or -] 0.3
Sum 18:1         6.7 [+ or -] 0.4    9.2 [+ or -] 0.6
Sum 20:1         1.6 [+ or -] 0.4    1.5 [+ or -] 0.1
Sum 22:1         --  [+ or -] --      -- [+ or -] --
Sum mettede     30.2 [+ or -] 0.1   24.4 [+ or -] 2.2
Sam monoener    13.3 [+ or -] 1.0   17.2 [+ or -] 1.1
Sum n-3         44.0 [+ or -] 1.2   46.3 [+ or -] 4.3
Sum n-6          4.0 [+ or -] 0.4    4.5 [+ or -] 0.2
Sum polyener    48.0 [+ or -] 0.9   51.4 [+ or -] 4.3

                                 PE

                   Recovering            Half-full

14:0             0.4 [+ or -] 0.1     -- [+ or -] -- *
15:0             2.3 [+ or -] 2.0    0.2 [+ or -] --
16:0             4.5 [+ or -] 1.1    4.7 [+ or -] 0 .5
17:0             1.8 [+ or -] 0.5    1.7 [+ or -] 0.1
18:0            12.8 [+ or -] 2.7   14.1 [+ or -] 1.8
Sum saturated   21.9 [+ or -] 4.4   20.8 [+ or -] 2.2
16:1n-7          0.3 [+ or -] --     0.4 [+ or -] -- *
16:1n-9           -- [+ or -] --      -- [+ or -] --
18:1n-7          0.7 [+ or -] --     0.5 [+ or -] -- *
18:1n-9          1.1 [+ or -] 0.5    0.6 [+ or -] 0.3
20:1n-7           -- [+ or -] --      -- [+ or -] --
20:1n-9          3.2 [+ or -] 0.6    0.7 [+ or -] 0.1 *
20:1n-11          -- [+ or -] --      -- [+ or -] --
22:1n-9          1.3 [+ or -] 2.0     -- [+ or -] --
22:1n-11         0.3 [+ or -] 0.5    0.1 [+ or -] 0.2
Sum monoenes     6.8 [+ or -] 2.6    2.3 [+ or -] 0.4 *
18:2n-6          0.2 [+ or -] 0.2    0.1 [+ or -] 0.2
20:2n-6           -- [+ or -] --     0.1 [+ or -] 0.1
20:4n-6          3.3 [+ or -] 0.9    2.7 [+ or -] 0.5
Sum n-6          3.5 [+ or -] 0.9    2.8 [+ or -] 0.3
16:3n-3          1.2 [+ or -] 1.1    0.6 [+ or -] 0.2
18:3n-3           -- [+ or -] --      -- [+ or -] --
18:4n-3          0.2 [+ or -] 0.1     -- [+ or -] --
20:4n-3           -- [+ or -] --      -- [+ or -] --
20:5n-3         14.8 [+ or -] 2.3   22.0 [+ or -] 2.6 *
22:5n-3           -- [+ or -] --     0.4 [+ or -] 0.1 *
22:6n-3         27.1 [+ or -] 3.5   34.0 [+ or -] 1.1 *
Sum n-3         43.2 [+ or -] 4.6   57.0 [+ or -] 1.7 *
n-3/n-6          6.6 [+ or -] 2.1    6.2 [+ or -] 0.3 *
Sum 16:1         2.6 [+ or -] 2.2    1.0 [+ or -] 0.3 *
Sum 18:1         1.1 [+ or -] 1.0    0.9 [+ or -] 1.1
Sum 20:1         2.2 [+ or -] 0.4    3.2 [+ or -] 0.6
Sum 22:1        21.9 [+ or -] 4.4   21.2 [+ or -] 2.6
Sum mettede      0.1 [+ or -] 0.1    0.3 [+ or -] 0.1
Sam monoener     1.8 [+ or -] 0.5    1.1 [+ or -] 0.3
Sum n-3          3.2 [+ or -] 0.6    3.6 [+ or -] 0.6 *
Sum n-6          1.3 [+ or -] 2.0    0.1 [+ or -] 0.2
Sum polyener      -- [+ or -] --      -- [+ or -] -- *

                                 PC

                   Recovering            Half-full

14:0             3.5 [+ or -] 0.5    1.1 [+ or -] 0.3 *
15:0             1.4 [+ or -] 0.2    0.8 [+ or -] 0.1 *
16:0            19.1 [+ or -] 0.8   15.1 [+ or -] 1.1 *
17:0             1.6 [+ or -] 0.1    1.1 [+ or -] -- *
18:0             3.5 [+ or -] 0.8    3.7 [+ or -] 0.2
Sum saturated   29.2 [+ or -] 1.3   21.8 [+ or -] 1.6 *
16:1n-7          1.6 [+ or -] 0.2    1.2 [+ or -] 0.2
16:1n-9          0.8 [+ or -] --     0.3 [+ or -] --
18:1n-7          2.8 [+ or -] 0.4    2.1 [+ or -] 0.2 *
18:1n-9          2.1 [+ or -] 0.4    2.1 [+ or -] 0.1
20:1n-7          0.1 [+ or -] 0.1     -- [+ or -] --
20:1n-9          0.5 [+ or -] 0.1    0.6 [+ or -] -- *
20:1n-11          -- [+ or -] --      -- [+ or -] --
22:1n-9           -- [+ or -] --      -- [+ or -] --
22:1n-11          -- [+ or -] --      -- [+ or -] --
Sum monoenes     7.1 [+ or -] 1.1    6.3 [+ or -] 0.3
18:2n-6          1.0 [+ or -] 0.1    0.9 [+ or -] --
20:2n-6          0.4 [+ or -] 0.1    0.5 [+ or -] 0.1
20:4n-6          1.3 [+ or -] 0.4    1.3 [+ or -] 0.2
Sum n-6          2.8 [+ or -] 0.4    2.7 [+ or -] 0.1
16:3n-3           -- [+ or -] --     0.2 [+ or -] 0.1
18:3n-3          1.5 [+ or -] 0.2    1.1 [+ or -] 0.1 *
18:4n-3          4.3 [+ or -] 0.8    3.4 [+ or -] 0.1
20:4n-3          0.4 [+ or -] --     0.5 [+ or -] 0.1 *
20:5n-3          9.4 [+ or -] 0.2   17.5 [+ or -] 1.8 *
22:5n-3          1.7 [+ or -] 0.3    2.4 [+ or -] 0.1 *
22:6n-3         35.5 [+ or -] 1.8   39.2 [+ or -] 2.8 *
Sum n-3         52.8 [+ or -] 2.4   64.3 [+ or -] 2.0 *
n-3/n-6         19.6 [+ or -] 3.7   23.7 [+ or -] 0.8 *
Sum 16:1         1.7 [+ or -] 0.3    1.6 [+ or -] 0.2
Sum 18:1         4.8 [+ or -] 0.7    4.1 [+ or -] 0.2
Sum 20:1         0.6 [+ or -] 0.2    0.6 [+ or -] 0.0
Sum 22:1          -- [+ or -] --      -- [+ or -] --
Sum mettede     29.3 [+ or -] 1.0   21.8 [+ or -] 1.6 *
Sam monoener     7.9 [+ or -] 1.1    7.8 [+ or -] 0.4
Sum n-3         52.8 [+ or -] 2.4   64.3 [+ or -] 2.0 *
Sum n-6          2.8 [+ or -] 0.4    2.7 [+ or -] 0.1
Sum polyener    55.6 [+ or -] 2.1   67.0 [+ or -] 2.0 *

NL, neutral lipids; PE, phosphatidyletanolamine; and PC,
phosphatidylcholine.

* indicate significant differences between the recovering (3) and
half-full (5) stages (Mann-Whitney U-test) sampled 15 September
1997 and 7 February 1998, respectively. Values are given as mean
and standard deviation, n = 3.

ACKNOWLEDGMENT

The authors thank Kjersti Ask and Thu Thao Nguyen for training and valuable help in the laboratory. The authors also thank Yanic Marty for inspiring methodological discussions. This work was supported by the Norwegian Research Council, project 111388/100 and National Institute of Nutrition and Seafood Research.

(1) One peak was rich in 22:6n-3 and the other was rich in 20:4n-6. These were eluted similarly as the fraction earlier assumed to be a glycolipid (Soudant et al. 1995), now identified as cardiolipin (Kraffe et al. 2002). The two peaks were only separable using new columns in the present set-up. They were not identified and are thus not reported here.

LITERATURE CITED

Ackman, R. G. 1983. Fatty acid metabolism of bivalves. In: G. D. Pruder & C. Langdon, eds. Biochemical and physiological approaches to shellfish nutrition. Louisiana: Louisiana State University. pp. 358-376.

Ackman, R. G. & C. S. Tocher. 1968. Marine phytoplankter fatty acids. J. Fish. Res. Bd. Canada 25:1603-1620.

Andersen, S. & H. Ringvold. 2000. Seasonal differences in effect of broodstock diet on spawning success. Aquaculture Int. 8:259-265.

Bell, J. G., J. Dick, A. H. McVicar, J. R. Sargent & K. D. Thompson. 1993. Dietary sunflower, linseed and fish oils affect phospholipid fatty acid composition, development of cardiac lesions, phospholipase activity and eicosanoid production in Atlantic salmon (Salmo salar). Prostaglandines. Leukot. Essent. Fatty Acids 49:665-673.

Bell, J. G., D. R. Tocher, F. M. MacDonald & J. R. Sargent. 1994. Effects of diets rich in linoleic (18:2n-6) and alpha linolenic (18:3n-3) acids on the growth, lipid class and fatty acid compositions and eicosanoid production in juvenile turbot (Scophthalmus maximus L.). Fish Physiol. Biochem. 13:105-118.

Bell, M. & D. Pond. 1996. Lipid composition during growth of motile and coccolith forms of Emillania huxleyi. Phytochemistry 41:465-471.

Barge, J.-P., J. P. Gouygou, J.-P. Dubacq & P. Durand. 1995. Reassessment of lipid composition of the diatom. Skeletonema costatum. Phytochemistry 39:1017-1021.

Berntsson, K. M., P. R. Jonsson, S. A. Wangberg & A. S. Carlsson. 1997. Effects of broodstock diets on fatty acid composition, survival and growth rates in larvae of the European flat oyster, Ostrea edulis. Aquaculture 154:139-153.

Besnard, J. Y., P. Lubet & A. Nouvelot. 1989. Seasonal variations of the fatty acid content of the neutral lipids and phospholipids in the female gonad of Pecten maximus L. Comp. Biochem. Physiol. 93B:21-26.

Bligh, E. G. & W. J. Dyer, 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Phys. 32:911-926.

Budge, S. M., C. C. Parrish & C. H. Mckenzie. 2001. Fatty acid composition of phytoplankton, settling particulate matter and sediments at a sheltered bivalve aquaculture site. Mar. Chem. 76:285-303.

Chu, F. L. E. & J. Greaves. 1991. Metabolism of palmitic, linoleic, and linolenic acids in adult oysters, Crassostrea virginica. Mar. Biol. 110: 229-236.

Chuecas, L. & J. Riley. 1969. Component fatty acids of the total lipids of some marine phytoplankton. J. Mar. Biol. Assoc. U.K. 49:97-116.

Delaunay, F., Y. Mary, J. Moal & J.-F. Samain. 1992. Growth and lipid class composition of Pecten maximus (L.) larvae grown under hatchery conditions. J. Exp. Mar. Biol. Ecol. 163:209-219.

Duinker, A. & A. Nylund. 2002. Seasonal variations in the ovaries of the great scallop (Pecten maximus) from western Norway. J. Mar. Biol. Assoc. U.K. 82:477-482.

Dunstan, G. A., J. K. Volkman, S. M. Barrett, J. M. Leroi & S. W. Jeffrey. 1994. Essential polyunsaturated fatty-acids from 14 species of diatom (Bacillariophyceae). Phytochemistry 35:155-161.

Erga, S. R. & B. R. Heimdal. 1984. Ecological studies on the phytoplankton of Korsfjorden, Western Norway. The dynamics of a spring bloom seen in relation to hydrographical conditions and light regime. J. Plankton Res. 6:67-90.

Farkas, T., I. Csengeri, F. Majoros & J. Olah. 1980. Metabolism of fatty acids in fish. 3. Combined effect of environmental temperature and diet on formation and deposition of fatty acids in the carp, Cyprinus carpio Linnaeus 1758. Aquaculture 20:29-40.

Gallager, S., R. Maan & G. Sasaki. 1986. Lipid as an index of growth and viability in three species of bivalve larvae. Aquaculture 56:81-103.

Helm, M. M., D. L. Holland, S. D. Utting & J. East. 1991. Fatty-acid composition of early nonfeeding larvae of the European flat oyster. Ostrea edulis. J. Mar. Biol. Assoc. U.K. 71:691-705.

Kraffe, E., P. Soudant, Y. Marty, N. Kervarec & P. Jehan. 2002. Evidence of a Tetradocosahexaenoic Cardiolipin in Stone Marine Bivalves. Lipids 37:507-514.

Langdon, C. J. & M. J. Waldock. 1981. The effect of algal and artificial diets on the growth and fatty acid composition of Crassostrea gigas spat. J. Mar. Biol, Assoc. U.K. 61:431-48.

Lie, O. & G. Lambertsen. 1991. Fatty-acid composition of glycerophospholipids in 7 tissues of cod (Gadus morhua), determined by combined high performance liquid chromatography and gas-chromatography. J. Chromatogr. 565:119-129.

Mason, J. 1958. The breeding of the scallop, Pecten maximus (L.), in Manx waters. J. Mar. Biol. Assoc. U.K. 37:653-671.

Napolitano, G. E. & R. G. Ackman. 1992. Anatomical distributions and temporal variations of lipid classes in sea scallops Placopecten magellanicus (Gmelin) from Georges Bank (Nova Scotia). Comp. Biochem. Physiol. 103B:645-650.

Napolitano, G. E., B. A. Macdonald, R. J. Thompson & R. G. Ackman. 1992. Lipid-composition of eggs and adductor muscle in giant scallops (Placopecten magellanicus) from different habitats. Mar. Biol. 113:71-76.

Napolitano, G. E., R. J. Pollero, A. M. Gayoso, B. A. Macdonald & R. J. Thompson. 1997. Fatty acids as trophic markers of phytoplankton blooms in the Bahia Blanca estuary (Buenos Aires, Argentina) and in Trinity Bay (Newfoundland, Canada). Biochem. Syst. Ecol 25:739-755.

Paulet. Y. M. & J. Boucher. 1991. Is reproduction mainly regulated by temperature or photoperiod in Pecten maximus? Inv. Repr. Dev. 19: 61-70.

Pazos, A. J., G. Roman, C. P. Acosta, J. L. Sanchez & M. Abad. 1997. Lipid classes and fatty acid composition in the female gonad of Pecten maximus in relation to reproductive cycle and environmental variables. Comp. Biochem. Physiol. 117B:393-402.

Pedersen, L., H. M. Jensen, A. D. Burmeister & B. W. Hansen. 1999. The significance of food web structure for the condition and tracer lipid content of juvenile snail fish (Pisces: Liparis spp.) along 65-72 degree N off West Greenland. J. Plankton Res. 21:1593-1611.

Piretti, M. V., F. Zuppa, G. Pagliuca & F. Taioli. 1988. Investigation of the seasonal-variations of fatty-acid constituents in selected tissues of the bivalve mollusc Scapharca inaequivalvis (Bruguiere). Comp. Biochem. Physiol. 89B:183-187.

Pollero, R. J., M. E. Re & R. R. Brenner, 1979. Seasonal changes of the lipids of the mollusc Chlamys tehuelcha. Comp. Biochem. Physiol. 64A:257-263.

Ronnestad, I., R. Finn, I. Lein & O. Lie. 1995. Compartmental changes in the contents of total lipid, lipid classes and their associated fatty acids in developing yolk-sac larvae of Atlantic halibut, Hippoglossus hippoglossus (L.). Aquacult. Nutr. 1:119-130.

Sargent, J. 1976. The structure, metabolism and function of lipids in marine organisms. In: D. C. Malins & J. R. Sargent, eds. Biochem. Biophys. Perspect. Mar. Biol. London: Academic Press. pp. 149-212.

Sargent, J. 1995. Origins and functions of egg lipids: Nutritional implications. In: N. R. Bromage & R. J. Roberts, eds. Broodstock management and egg and larval quality. Oxford: Blackwell Science Ltd.

Servel. M. O., C. Claire, A. Derrien, L. Coiffard & Y. Deroeckholtzhauer. 1994. Fatty acid composition of some marine microalgae. Phytochemistry 36:691-693.

Shirai, N., M. Terayama & H. Takeda. 2002. Effect of season on the fatty acid composition and free amino acid content of the sardine Sardinops melanostictus. Comp. Biochem. Physiol. 131B:387-393.

Soudant, P., Y. Marry, J. Moal, R. Robert, C. Quere, J. R. Lecoz & J. F. Samain. 1996a. Effect of food fatty acid and sterol quality on Pecten maximus gonad composition and reproduction process. Aquaculture 143:361-378.

Soudant, P., Y. Marry, J. Moal & J. F. Samain. 1995. Separation of major polar lipids in Pecten maximus by high-performance liquid-chromatography and subsequent determination of their fatty-acids using gas-chromatography, J. Chromatogr. 673B:15-26.

Soudant, P., Y. Marty, J. Moal & J. F. Samain. 1996b. Fatty acids and egg quality in great scallop. Aquaculture Int. 4:191-200,

Soudant, P., J. Moal, Y. Marty & J. F. Samain. 1996c, Impact of the quality of dietary fatty acids on metabolism and the composition of polar lipid classes in female gonads of Pecten maximus (L). J. Exp. Mar. Biol. Ecol. 205: 149-163.

Soudant, P., K. Vanryckeghem, Y. Marty, J. Moal, J. F. Samain & P. Sorgeloos. 1999. Comparison of the lipid class and fatty acid composition between a reproductive cycle in nature and a standard hatchery conditioning of the pacific oyster Crassostrea gigas. Comp. Biochem. Physiol. 123B:209-222.

Strohmeier, T., A. Duinker & O. Lie. 2000. Seasonal variations in chemical composition of the female gonad and storage organs in Pecten maximus (L.) suggesting that somatic and reproductive growth are separated in time. J. Shellfish Res. 19:741-747.

Thompson. P. A. & P. J. Harrison. 1992. Effects of monospecific algal diets of varying biochemical composition on the growth and survival of Pacific oyster (Crassostrea gigas) larvae. Mar. Biol. 113:645-654.

Ueda, T. 1974. Changes in the fatty acid composition of short neck clam with reference to environmental mud temperature. Bull. Jap. Soc. Sci. Fish. 40:949-957.

Utting, S. D. & J. Doyou. 1992. The increased utilization of egg lipid reserves following induction of triploidy in the manila clam (Tapes philippinarum). Aquaculture 103:17-28.

Vitello, F. & J.-P. Zanetta. 1978. Thin-layer chromatography of phospholipids. J. Chromatogr. 166:637-640.

Volkman. J. K., D. J. Smith, G. Eglington, T. E. V. Forsberg & E. D. S. Corner. 1981. Sterol and fatty acid composition of tour marine haptophycean algae. Comp. Biochem. Physiol. 61:509-527.

Williams, E. E. & J. R. Hazel. 1994. Thermal adaptation in fish membranes: temporal resolution of adaptive mechanisms. In: A. R. Cossins, ed. Temperature adaptation of biological membranes. London: Portland Press. pp. 91-106.

Wold, S., K. Esbensen & P. Geladi. 1987. Principal Component Analysis. Chemometrics Int. Lab. Systems 2:37-52.

Zar, J. H. 1999. Biostatistical analysis. 4th ed. New Jersey: Prentice-Hall, 123 pp.

ARNE DUINKER, * BENTE E. TORSTENSEN, AND OYVIND LIE

National Institute of Nutrition and Seafood Research, PO Box 176 Sentrum, 5804 Bergen, Norway

* Corresponding author. Email: [e-mail]