1 Seasonal variation in condition and fatty acid composition of 2 coquina clam, Donax trunculus (Linnaeus 1758) (Mollusca, 3 Bivalvia) from the Tunisian coast 4 Variation saisonnière de l’indice de condition et de la 5 composition en acides gras de Donax trunculus (Linnaeus 1758) 6 (Mollusque, Bivalve) des côtes tunisiennes 7
8 Dhouha Boussoufaa,*, Nabila Ghazalia, Elena Vicianob, Juan Carlos Navarrob
9 & Mhamed El Cafsia
10
11 Address:
12 aUnité de Physiologie et d’Ecophysiologie des Organismes Aquatiques,
13 Département des Sciences Biologique, Faculté des Sciences de Tunis,
14 Campus universitaire, 2092 Tunis, Tunisie.
15 bInstituto de Acuicultura de Torre de la Sal (CSIC), 12595 Ribera de
16 Cabanes, Castellón, España.
17 Email: [email protected]
18 *Corresponding author. Dhouha BOUSSOUFA, Faculté des Sciences de
19 Tunis, Département des Sciences Biologiques, Laboratoire de Physiologie
20 et d’Ecophysiologie des Organismes Aquatiques Campus Universitaire,
21 2092 Tunis, Tunisia.
22 Tél. +216-22-870-014; +216-99-880-014; Fax: +216-71-885-480.
23 Email: [email protected]
24 Running title: Fatty acids of D. trunculus 25 Keywords: Bivalves, condition index, Donax trunculus, fatty acids, Gulf of
26 Tunis. 27 Résumé:
28 L’analyse saisonnière de la variation de l’indice de condition, du contenu en
29 lipides et de la composition en acides gras du flion Donax trunculus
30 (Mollusca, bivalvia) récolté dans trois stations de la côte Nord Est de la
31 Tunisie : Borj Cedria (BC), Rades (R) et Kalaat El Andalous (KA) a été
32 réalisée. Les lipides totaux varient entre 4,8 et 7,3%, entre 5,9 et 8,1 % et
33 entre 5,5 et 10,8 % du poids sec respectivement chez D. trunculus de BC, R
34 et KA. L’analyse de la composition en acides gras montre une
35 prédominance des acides palmitique (C16:0), stéarique (C18:0),
36 eicosapentaénoïque (EPA) (C20:5n-3) et docosahéxaénoïque (DHA)
37 (C22:6n-3). Les acides gras polyinsaturés (AGPI) constituent la forme
38 majeure des acides gras chez les D. trunculus des trois stations. Les
39 pourcentages des acides gras saturés (AGS) varient inversement avec la
40 variation des pourcentages des AGPI qui sont plus importants durant l’hiver
41 et le printemps. Deux corrélations ont été signalées avec la variation de la
42 température de l’eau, une négative avec les AGPI de la famille (n-3) et une
43 positive avec les AGPI de la famille (n-6).
44 Abstract
45 The condition index, the lipid content and the fatty acid composition of the
46 coquina clam Donax trunculus (Mollusca, bivalvia) sampled from three
47 stations of the North Eastern coast of Tunisia: Borj Cedria (BC), Rades (R)
48 and Kalaat El Andalous (KA), were analysed. Total lipid content in dry
49 weight ranged from 4.8 to 7.3 %, from 5.9 to 8.1 % and from 5.5 to 10.8 %
50 respectively in BC, R and KA specimens. The analysis of the fatty acid
51 composition showed that the major fatty acids in D. trunculus tissues were
52 palmitic C16:0, stearic C18:0, eicosapentaenoic (EPA) C20:5n-3 and
53 docosahexaenoic (DHA) C22:6n-3 acids. In comparison with saturated fatty 54 acids (SFA) and monounsaturated fatty acids (MUFA), polyunsaturated
55 fatty acids (PUFA) constituted the highest proportion in D. trunculus from
56 different sites. SFA varied during the year inversely to the variation of
57 PUFA which were higher in winter and spring than in summer and autumn.
58 A negative correlation between water temperature and (n-3) PUFA was
59 observed, whereas the (n-6) PUFAs showed a positive correlation with this
60 variable.
61 1-Introduction
62 Donax trunculus (Linnaeus 1758) Mollusca: Bivalvia, is an Atlantic-
63 Mediterranean warm-temperate species inhabiting the Mediterranean and
64 the Black Sea (Bayed & Guillou, 1985), and the coasts from Senegal to the
65 north Atlantic of France (Tebble, 1966). It is commercially of less interest
66 than the oyster Crassostrea gigas (Thunberg, 1793) or the clam Ruditapes
67 decussatus (Linnaeus, 1758) (Ruiz- Azcona et al. (1996)), although there
68 has been a large demand for this species in many countries like Spain (Ruiz-
69 Azcona et al. (1996)), Italy (Manca Zeichen et al. (2002)) and Portugal
70 (Gaspar et al. (1999)). Its ecological preferences have been studied in the
71 Mediterranean Sea, along the Algerian (Degiovanni et al. 1972), and the
72 Israeli (Neuberger et al. 1990) coasts and along the French, Spanish and
73 Moroccan Atlantic coasts (Ansell et al. 1980; Mazé et al. 1988; Guillou et
74 al.1991). In the Mediterranean sea, D. trunculus occupies a relatively
75 narrow zone in the shallow sub littoral in depths of few centimetres to 2 m
76 and its distribution is limited to moderately- exposed beaches of well-graded
77 fine sands (Mazé & Laborda, 1988). The reproductive cycle and population
78 dynamics have been examined by Moueza & Chassel 1976; Guillou 1982;
79 Mazé 1990; Bayed 1998, Gaspar et al. 1999; Manca Zeichen et al. 2002 and
80 Dhaoui Ben Kheder et al. 2003. During the first year of life, at a length of 81 16 mm, the species is already capable of reproducing (Moueza & Frenkiel-
82 Renault, 1973). D.trunculus is gonochoric; no hermaphroditism or sex
83 reversal was encountered (Gaspar et al. 1999). Sex can easily be identified
84 macroscopically when the gonads are fully developed, as there are
85 differences in color: dark blue in females and yellow-white in males (Manca
86 Zeichen et al. 2002). Generally, gametogenesis takes place during winter
87 and spring and gamete emission starts in late spring and continues until
88 summer (Moueza & Frenkiel-Renault, 1973; Manca Zeichen et al. (2002).
89 In a recent study, Dhaoui Benkheder (2003) described the general features
90 of the reproductive cycle of D. trunculus in two sandy beaches located in
91 the Gulf of Tunis: Kalâat El Andalous and Raoued. It was characterised by a
92 large gametogenic cycle comprising two phases: a resting phase (autumn)
93 and a gametogenic one, including ripeness and spawning, during the rest of
94 the year. Life span of D. trunculus decreases with latitude: from 5 years
95 along French Atlantic coast to 2-3 years on the Moroccan coast and
96 Mediterranean (Gaspar et al. 1999). Ruiz-Azcona (1996) has described the
97 larval rearing of D trunculus and showed that growth was greater when
98 larvae were kept at 20°C seawater and fed with multispecies of microalgae
99 including Chaetoceros gracialis.
100 In Tunisia, D. trunculus shows a large distribution with high density along
101 the sandy beaches especially in the Gulf of Tunis coasts. These are high
102 productive areas of exposed well-graded sandy beaches, which could be
103 representative of the habitat of the genus (Gaspar et al. (1999); Manca
104 Zeichen et al. (2002)).
105 Several authors (Martinez (1991); Orban et al. (2004)) have shown that
106 seasonal metabolic activities in molluscs result from complex interactions
107 among factors like food availability, growth, reproduction and 108 environmental parameters such as temperature. In general, when food is
109 abundant, energy in marine bivalves is stored in the form of glycogen,
110 protein and lipid prior to gametogenesis (Barber & Blake (1985)).
111 To the best of our knowledge, information concerning the quality requisites
112 of D. trunculus in lipid and fatty acid composition barely exists. The aim of
113 this work is to study the natural seasonal variations in condition index, lipids
114 and fatty acid composition of D. trunculus and to examine the potential
115 influence of some environmental parameters along the reproductive cycle.
116 2-Materials and methods
117 2-1 Sample collection and environmental parameters:
118 The specimens of D. trunculus used in this study were collected from three
119 stations BC, R and KA situated in the gulf of Tunis (Fig. 1, tab. 1). The
120 geological parameters of this coastal area are affected by many factors such
121 as the presence of the Medjerda River in Kalâat El Andalous plain, Méliane
122 River in Radès and of other minor fresh water streams. Specimens were
123 collected using a specially designed hand dredge, at approximately monthly
124 intervals between July 2004 and June 2005. Along the sandy beach of KA
125 and R specimens were collected at 30 cm water depth, and from BC at a
126 depth of 50 cm. Data on the surface water temperature and salinity of the
127 areas of collection were measured once per sampling day (between 8 am
128 and 10 am) using an SCT meter (Salinity, Conductivity and Temperature)
129 (YSI-Model 33).
130 2-2 Biometric measurement and condition index:
131 Fifty adults of similar size (anteroposterior length≥ 22 mm) were collected
132 and placed in running filtered sea water (24 hours at least) to empty their
133 guts before being dissected and stored at – 30 °C until the analyses were
134 carried out. Thirty individuals were randomly selected for biometric 135 measurements. Thickness, length and width were measured using a 0.05 mm
136 precision calliper, and reproductive activity was assessed using Walne
137 (1976) maturity index (condition index (CI)), based on the ratio of meat dry
138 weight: shell dry weight X 100). This index is commonly used to evaluate
139 the healthy state of bivalves, and adopted in foreign trade as a quality
140 criterion (Orban et al. 2006). The ratio also reflects the seasonal
141 fluctuations, the gametogenic cycle and the environmental conditions
142 (Lucas and Beninger (1985)). Soft tissues were carefully separated from
143 shells and washed in distilled water to remove dirt. For dry weight
144 determination, both soft tissues and shells were dried in an oven to a
145 constant weight at 60°C.
146 2-3 Lipid content and fatty acid composition:
147 Lipids and fatty acids from coquina clams from each station were sampled
148 seasonally (in July 04 (summer 04), October 04 (autumn 04), January 05
149 (winter 05) and April 05 (spring 05). Total lipid from each sample (six
150 replicates of the whole animal per season and per station) was extracted
151 with chloroform: methanol (2:1, v/v) using the method of Folch et al.
152 (1957). Following solvent evaporation under nitrogen, lipids were
153 transferred into a pre-weighed 2 ml vial. The solvent mixture was again
154 evaporated under nitrogen and the extracts were further dried overnight in a
155 vacuum desiccator. After weighing, lipids were redissolved in chloroform:
156 methanol (2:1, v/v) with 0.01% butylated hydroxy toluene (BHT, Sigma-
157 Aldrich) added as an anti-oxidant. Total lipids were expressed as g/100g dry
158 weight (DW). Fatty acids from total lipids were acid catalysed
159 transmethylated with toluene and 1 % sulphuric acid in Methanol overnight
160 (16 h) at 50 °C (Christie (1982)). Nonadecanoic acid (C19:0) was added to
161 each sample as an internal standard. The resulting fatty acid methyl esters 162 (FAME) were extracted with hexane: diethyl ether (1:1, v/v) and purified by
163 thin layer chromatography (TLC). The FAME were subsequently analysed
164 by high resolution gas chromatography (GC 8000 Series, Fisions
165 Instruments, Rodano, Italy), using a cold on-column injection system, a
166 flame ionisation detector (FID) and a fused silica capillary column (30 m x
167 0.25 mm inner diameter, 0.25 µm film thickness Tracer, TR-WAX,
168 Teknokroma, Barcelona, Spain), with Helium as a carrier gas. Oven
169 temperature was increased from 50 to 180ºC at 40ºC/min and then to 220ºC
170 at 3ºC/min. Peaks were recorded in a personal computer using a software
171 package (version 4.6.0.0. Azur, Datalys, St Martin d’Heres, France).
172 Individual FAME were separated and identified by comparison of their
173 retention times with those of well characterized fish oils and commercial
174 standards (FAME 37, Supelco). The relative amount of each FA was
175 expressed as a percentage of the total amount of FA in the analyzed sample.
176 2-4 Statistical methods:
177 The results of the seasonal variations of condition index, lipids and fatty
178 acid composition of D. trunculus in the three sites of collection are reported
179 as means ± standard deviations. After checking for normality and correcting
180 if necessary transforming data, One-Way analysis of variance (ANOVA)
181 was performed using the STATISTICA 6.0. Tukey Honest Significant
182 Differences (HSD) multiple comparisons test was conducted to determine
183 differences among the means at a significance level of 5 %. The Pearson’s
184 correlation coefficient was used to study relationships among the major fatty
185 acids groups, the reproductive cycle and the environmental parameters.
186 3-Results
187 3-1. Temperature and salinity 188 The mean sea water temperature and salinity are shown in (Fig. 2). The
189 water temperature followed the same trend for the three sampling sites: the
190 highest temperature was recorded during July 2004 in R (30 °C) and KA (27
191 °C), and during August 2004 in BC (27 °C), whereas The minimum
192 temperature of 7°C, 8.5°C and 9.5 °C was measured during January 2005
193 respectively in BC, R and KA sea water.
194 Over the sampling period, salinity depended on the amount of rainfall, and
195 also on the fresh water outflow from the rivers. It ranged from 34 to 39 ‰,
196 from 23 to 38 ‰ and from 32.5 to 38.5 ‰ respectively in BC, R and KA
197 water surfaces. In fact, the Gulf of Tunis is influenced by the discharge of
198 Medjerda and Méliane wadi respectively in KA plain and R beach, this may
199 explain the sharp decrease of the salinity registered in February (23 ‰) and
200 May (28 ‰) in R. During these periods of the year precipitations were
201 abundant and more frequent.
202 3-2. Biometrics and condition index
203 The biometrics of D. trunculus harvested along the gulf of Tunisia during
204 the experimental period, showed that the length of specimens ranged from
205 23 mm to 32 mm (table 2). The evolution of the reproductive activity was
206 followed through CI in mature individuals (Shell length > 16 mm) (Fig.3a).
207 A steady increase in the index was observed since late summer. This
208 increasing trend continued until winter for R specimens and until autumn for
209 BC and KA specimens. During winter, the CI rose progressively until
210 reaching the spring values for the sexually ripe BC and KA animals. Since
211 autumn, CI remained more or less constant throughout the season for R D.
212 trunculus. The change in percentage dry weight followed closely the
213 seasonal changes in CI. In fact, the decrease in the CI values in summer and
214 late autumn corresponded to the loss of dry weight (Fig. 3b). 215 3-3. Lipid content
216 Seasonal total lipid content (TLC) in D. trunculus from the three sampling
217 sites ranged from 4.8 to 7.3, from 5.9 to 8.1 and from 5.6 to 10.8 g/100g
218 DW respectively in BC, R and KA (fig. 3c). The highest TLC was
219 registered during the summer in R and KA and during the winter in BC.
220 This winter value coincides with the gametogenesis period of D. trunculus
221 (reserve accumulation). The summer lipid content in KA specimens was
222 significantly higher than those of autumn, winter and spring. In contrast
223 there was no variation in the TLC of the Donax sampled in R and BC
224 throughout the season.
225 3.4. Fatty acid composition
226 PUFA of D. trunculus harvested from the three sites were generally the
227 major forms of fatty acids during the year followed by SFA (Table 3). The
228 PUFA showed a similar pattern in the individuals harvested from the
229 different sites. They were generally low during summer (31.6 ± 2 % in R,
230 32.9 ± 1.8 % in KA and 38.7 ± 2.2 % in BC) and increased continuously
231 until reaching significantly higher values during winter in BC (46 ± 2.2 %)
232 and during spring in R (39.5 ± 2 %) and KA (38 ± 2 %). No differences
233 were obtained between winter and spring PUFA percentages except for the
234 KA clams that showed autumn (33 ± 2%) and winter (34.7 ± 5.7%) PUFA
235 values more similar than winter and spring ones. In view of the high
236 standard deviations registered in winter it could be argued that PUFA from
237 KA clams followed a slightly different trend than those from BC and R.
238 SFA had an irregular seasonal trend in the individuals from the three
239 sampling stations. They varied seasonally in BC and R specimens and
240 remained more or less stable throughout the season in KA. The significantly
241 lowest values were registered during winter (25.8 ± 1.3 %, 28.8 ± 1.9 % for 242 R and BC respectively), whereas the highest ones were measured in autumn
243 and summer in R (32.5 ± 1.1 %) and BC (34.7 ± 2.4 %) specimens
244 respectively.
245 MUFA represent the lowest proportion of total fatty acids. No significant
246 seasonal changes were observed in BC and KA clams, whereas in R, MUFA
247 registered in autumn were significantly lower than those of winter and
248 spring.
249 SFA, MUFA and PUFA differed significantly among stations. In fact, SFA
250 and MUFA obtained in R differed significantly from SFA and MUFA from
251 KA clams, whereas PUFA from BC clams were significantly higher than
252 those of R and KA.
253 The fatty acid analyses showed that among SFA, C14:0, C16:0 and C18:0
254 were present at the highest levels through all seasons. The level of MUFA
255 was highly dependent of the levels of C16:1n-7, C18:1n-9; C18:1n-7 and
256 C20:1n-9. The most abundant PUFA were C20:5n-3 (EPA), C22:5n-3
257 (DPA) and C22:6n-3 (DHA) followed by the arachidonic acid (C20:4n-6),
258 precursor of eicosaenoids. The seasonal profile of C18:3n-3, C20:5n-3 and
259 C22:6n-3 reflected that of total PUFA, which varied significantly during the
260 seasons in BC and R specimens, DHA being the major PUFA. A
261 comparison of n-3 PUFA and n-6 PUFA suggests that n-3 PUFA were in
262 significantly higher proportion, and n-6 PUFA contained higher percentages
263 of C20:4n-6 than C18:2n-6 and C20:2n-6. The correlations between (n-3)
264 PUFA and temperature were negative (r = -0.86, p = 0.0000003 in R, r = -
265 0.68, p = 0.0006 in BC, r = -0.47, p = 0.02 in KA). However, (n-6) PUFA
266 showed a positive correlation with this variable (r = +0.68, p = 0.0004 in
267 KA), (r = +0.73, p = 0.0001 in R) and (r = +0.527, p = 0.014 in BC). The
268 total (n-6) PUFA and (n-3) PUFA from the individuals collected at the three 269 sites showed a positive correlation with C18:2n-6 and C18:3n-3
270 respectively. No correlation was found between C20:5n-3 and C22:6n-3
271 fatty acids.
272 Two C22 dienoic, C22:2i and C22: 2j, acids appeared successively between
273 the C22:1n-9 and C22:5n-6. They were probably non-methylene interrupted
274 dienoic (NMID) FA with two double bounds. These FA are seemingly
275 ubiquitous lipid mollusc components (Ojea et al., 2004). They were present
276 at low percentages: between 0.4 % and 4.3 % in all samples. The increase in
277 C22:2j percentages coincided with a decrease in the EPA of D. trunculus
278 from the three stations (r = - 0.734, -0.553, -0.526 for BC, KA and R)
279 whereas an increase in C22:2i coincided with a decrease in the levels of
280 DHA only in R clams (r = - 0.657).
281 4-Discussion:
282 4-1. Biometric characteristics and condition index
283 It is well known that CI is influenced by seasonal factors, including sexual
284 maturity and food availability (Matozzo et al. 2005). At the beginning of the
285 spawning period when temperature was highest (during summer), the values
286 of CI were lowest and coincided with the minimum values of dry weight in
287 the different sites. This decrease in CI values may reflect the major
288 biological effort expended in the production and release of gametes. The
289 tissue dry weight increased gradually during spring (gametogenesis period)
290 in the three sites, while the CI remained more or less constant in R and KA
291 coquina clams because the shell and body growth occurred throughout this
292 period (Gaspar et al. 1999). At the beginning of autumn, the abundance of
293 food available in the sampling area allowed recovery of the tissue dry
294 weight which was accompanied by an increase in the values of CI.
295 According to Laruelle et al. 1994, who worked on R. decussatus, 296 temperature has a positive effect on gametogenesis. It may have a direct
297 effect on the metabolic rate of the animal, but also an indirect effect through
298 the availability of food, since the maximum rate of increase in weight occurs
299 during spring, when both the sea temperature and food supply increase
300 rapidly. In the work of Ruiz Azcona, 1996, D. trunculus was thermally
301 induced to spawn at 25°C which corresponds to the temperature reached in
302 summer in the three sampling areas of our study. Similar observations of
303 temperature effect on gametogenesis have been made by Lubet (1976) for
304 populations of Ostrea edulis (Linnaeus, 1758) from the Atlantic coast of
305 France and by Mann (1979) for O. edulis under laboratory conditions.
306 Salinity in the Gulf of Tunis was relatively stable throughout the year. It
307 only declines in winter especially in R coinciding with especially rainy
308 periods. Apparently, the fluctuations observed did not have an important
309 effect on D. trunculus reproduction. Similar results have been reported for
310 oysters from San Cibran (Galicia, Spain) (Ruiz et al. 1992).
311 Several authors showed that gametogenesis and spawning of D. trunculus
312 may all have proceeded together (Ansell et al. 1980, Bayed 1990 and
313 Gaspar et al. 1999). During most of the year, mature individuals in
314 gametogenic phase can be present together with individuals in resting phase.
315 This could explain the high levels of standard deviations registered in the
316 percentages of dry weight during the year. Similar observations on the
317 reproductive cycle of D. trunculus were made by Moueza and Frenkiel-
318 Renault (1973) in the Mediterranean sea (Azur Plage, the Algerian coast),
319 by Gaspar et al. (1999) in the Atlantic (off Faro, Southern Portugal), and
320 Bayed et al. (1990) in the Moroccan Atlantic coast. It is generally accepted
321 that, aside from gametogenic cycle influences, the blooming of
322 phytoplankton during the year may affect the CI and the biochemical 323 composition of bivalves relaying on plankton as the main food source
324 (Pollero and Brenner (1979)). For this reason the availability of food
325 represents a very important factor to take into account.
326 4-2. Lipid content
327 The comparatively high lipid levels of D. trunculus from the three locations
328 during summer (spawning season) indicate that lipids may play some role in
329 the maturation of gametes (Awaji and Suzuki (1999)), similar to that in
330 finfishes (Henderson et al. (1984)). Besides, the higher lipid contents in
331 summer suggest that the warm seawater might have promoted accumulation
332 of lipids, perhaps related to a high metabolic activity in this season. In fact,
333 Mackie and Ansell (1993) reported that environmental factors apparently
334 play a dominant role in determining events in the storage cycles of bivalves.
335 Similar variations were found by Ansell et al. 1980 in D. trunculus from
336 Azur Plage (Algerian coast) and indicated that the highest values of the lipid
337 content increased during summer in sexually ripe animals, and decreased in
338 autumn and winter. In this study, the mean lipid contents were low, similar
339 to those observed also for D. trunculus by Ansell et al. (1980) and for other
340 bivalves like Pinctada fucata (Gould 1850) by Saito (2004) and for C.
341 gigas by Dunstan et al. (1993).
342 The variation of lipid content could be explained by the abundance and
343 quality of the available food in the environment at a period of
344 phytoplanktonic bloom, and by the accumulation or utilization of lipid
345 reserves prior to gametogenesis. It should be emphasized that the period of
346 lipid increase in specimens from BC (winter), coincides generally with the
347 gonadal maturation, whereas the decrease in lipid content during autumn
348 and spring could also be associated to its utilization for metabolic energy
349 purposes. 350 4-3. Fatty acids:
351 The results of the variations in fatty acids composition during this study
352 clearly indicate that PUFA, especially DHA were generally the predominant
353 forms of fatty acids measured in D. trunculus from the three stations. This
354 has also been reported for other bivalves, such as R. decussatus (Ojea et al.
355 (2004)) and oyster C. gigas (Dridi et al. (2007)).
356 The seasonal variations of PUFA could hardly be correlated with the
357 different phases of the annual reproductive cycle of the animals and with
358 temperature, although the highest values registered in winter (for BC
359 specimens) and spring (for R and KA) just preceded the more regular
360 spawning period which occurred in late spring and during summer. This
361 may explain the main function of these fatty acids in gamete formation.
362 During the period under study, D. trunculus was characterised by high
363 levels of n-3 PUFA (24.5 - 34.9, 19.5 - 31.2, 21,1 - 29.7 % of total fatty
364 acids for the individuals from BC, R and KA respectively). These are
365 important FA in the human diet, especially for their prevention of
366 cardiovascular diseases. Low levels of n-6 PUFA (6.4 - 9.7, 5.6 - 7.8, 6 -7
367 .7% of total fatty acids in specimens from BC, R and KA respectively)
368 together with relatively high n-3/n-6 PUFA ratios (3.5 - 5.6 in BC, 2.8 - 7.4
369 in R and 3.3 - 5.6 in KA) contribute to a positive bromatological evaluation
370 of the lipid quality of D. trunculus from the South coast of the
371 Mediterranean Sea (Gulf of Tunis).
372 Taking into account the low lipid content of the three D. trunculus
373 populations, it may be assumed that the high percentages of EPA and
374 especially DHA in their total lipid fraction reflect the preferential
375 association of these fatty acids with phospholipids in cell membranes, as
376 reported in the bivalve Chamelea gallina (Linnaeus, 1758) by Orban et al. 377 (2006). These FA are implicated in the structure and function of cell
378 membranes (Pazos et al. (2003)).
379 The DHA and EPA peaked in spring and winter, coinciding with the period
380 of gametogenesis. The same result was reported by Abad et al. (1995) and
381 Dridi et al. (2007) for the oyster.
382 The two C22 NMID were present at low percentages (between 0.4 % and
383 4.3 %) in all samples. The role of 22:2 NMID is unknown, but the
384 association with membrane lipids and the selective retention in starved
385 animals suggests a structural/metabolic function and/or a strong resistance
386 against degradation (Klingensmith (1982)). Molluscs have generally active
387 FA elongation and desaturation systems permitting the de novo synthesis of
388 NMID FA. These are the only PUFA that are synthesized by marine
389 molluscs (Abad et al. (1995); Ojea et al. (2004)). Zhukova (1991) proposed
390 that the biosynthesis of these FA occurs via the n-7 MUFA pathway until
391 C20:1n-7, which is desaturated by ∆5 desaturase to C20:2n-5 and then
392 elongated to C22: 2n-7. It is possible that facing a deficiency of dietary
393 unsaturated FA, including the n-3 and n-6 PUFA, the molluscs synthetize
394 NMID FA to support the necessary fluidity of the cell membrane (Abad et
395 al. (1995)). In fact, Ojea et al. (2004) reported that 22:2i and 22:2j in the
396 polar lipids of R. decussatus were negatively correlated with EPA, and 22:2j
397 was also negatively correlated with DHA. In other study, Klingensmith
398 (1982) found an inverse relationship between n-3 PUFA, especially EPA
399 and DHA, and NMID FA in the clam Mercenaria mercenaria (Linnaeus,
400 1758). Similarly, in the present work, the increase in C22:2j and C22:2i
401 coincides with low levels of EPA and DHA respectively.
402 The increased PUFA, at colder temperature during winter and spring is
403 mainly correlated with the high n-3 PUFA concentration especially EPA 404 and DHA. It has been reported that PUFA content in fish and shellfish
405 varies inversely with water temperature (Chu and Greaves (1991);
406 Ingemansson et al. (1993)). Several authors (Sargent (1976); Holland
407 (1978); Dunstan et al. (1999)) pointed out that a decrease in temperature
408 produces generally an increase in PUFA and a decrease of SFA in order to
409 maintain the fluidity of the cell membranes.
410 In conclusion, the seasonal changes observed in lipids and FA composition
411 of D. trunculus from the Gulf of Tunis (the Mediterranean coast of Tunisia)
412 follows a trend typical of bivalve molluscs. In fact, apart from influences of
413 the reproductive cycle, the coquina clam D. trunculus, like the filter-feeders
414 on phytoplankton and suspended particulate organic matter are highly
415 dependent on the dietary resources available in their immediate habitat, and
416 therefore on the water temperature. Due to their low lipid and high
417 percentage of healthy n-3 PUFA especially in winter and spring (periods of
418 reserve accumulation), D. trunculus can be considered as a food item with
419 interesting dietetic properties.
420 Acknowledgments
421 This work was supported by a financial aid (grant) from the “Agencia Española
422 de Cooperación Internacional (AECI)”. The authors thank the members of the
423 Ecophysiology Laboratory in Tunis and the Artemia group in Spain for their help.
424
425 References
426 Abad M., Ruiz C., Martinez D., Mosquera G. & Sanchez J.L. 1995.
427 Seasonal variations of lipid classes and fatty acids in flat oyster, Ostrea
428 edulis, from San Cibran (Galicia, Spain). Comparative Biochemistry and
429 Physiology, 110 C: 109-118. 430 Ansell A.D., Frenkiel L., Moueza M. 1980. Seasonal changes in tissue
431 weight and biochemical composition for the Bivalve Donax trunculus L. On
432 the Algerian Coast. Journal of Experimental Marine Biology and Ecology,
433 45: 105-116.
434 Awaji M. & Suzuki T. 1999. The pattern of cell proliferation during pearl
435 sac formation in the Pearl Oyster. Fisheries Science, 61:745–751.
436 Barber B.J. & Blake N.J. 1985. Intra-organ biochemical transformations
437 associated with oogenesis in the Bay scallop, Agropecten irradians
438 concentricus (Say), as indicated by 14C incorporation. Biological Bulletin,
439 168: 39-49.
440 Bayed A. & Guillou J. 1985. Contribution à l’étude des populations du
441 genre Donax : la population de D.trunculus L. (Mollusca, bivalvia) de
442 Mehdia (Maroc). Annales de l'Institut Oceanographique, Paris, 61 (2) :
443 139-147.
444 Bayed A. 1990. Reproduction de Donax trunculus sur la côte atlantique
445 marocaine. Cahier de Biologie Marine, 31 : 159-169.
446 Bayed A. 1998. Variabilité de la croissance de D. trunculus sur le littoral
447 marocain. Centre International de Hautes Etudes Agronomique
448 Méditerranéennes. CIHEM, Cahiers Options Méditerranéennes, 35 : 11-23.
449 Christie W.W. 1982. Lipid Analyses, 2nd Ed. Oxford, PA: Pergamon Press,
450 207pp.
451 Chu E.L.E. & Greaves J. 1991. Metabolism of palmitic, linoleic and
452 linolenic acids in adult oysters, Crassostrea virginica. Marine Biology,110 :
453 229-236. 454 Degiovanni C. & Moueza M. 1972. Contribution à l’étude de la biologie de
455 Donax trunculus L. (Mollusque, Lamellibranche) dans l’Algérois. Ecologie
456 en baie de Bou Ismaïl. Téthys, 4: 729-743.
457 Dhaoui Ben Kheder R., Aloui Bejaoui N. & Le Pennec G. 2003. Cycle
458 sexuel de Donax trunculus (Mollusque, bivalve) du Golfe de Tunis. Bulletin
459 de la Société Zoologique de France, 128 (1-2): 9-20.
460 Dridi S., Romdhane M. S. & El Cafsi M. 2007. Seasonal variation in
461 weight and biochemical composition of the Pacific oyster, Crassostrea
462 gigas in relation to the gametogenic cycle and environmental conditions of
463 the Bizert lagoon, Tunisia. Aquaculture, 263: 238-248.
464 Dunstan G.A., Olley J. & Ratkowsky D.A. 1999. Major environmental
465 and biological factors influencing the fatty acid composition of seafood
466 from indo-Pacific to Antarctic waters. Recent Res. Devel. Lipid Research,
467 3: 63-86.
468 Dunstan G.A., Volkman J.K. & Barrett S.M. 1993. The Effect of
469 Lyophilization on the Solvent Extraction of Lipid Classes,Fatty Acids, and
470 Sterols from the Oyster Crassostrea gigas. Lipids, 28: 937–944.
471 Folch J., Less M. & Sloane-Stanley C.H. 1957. A simple method for the
472 isolation and purification of total lipids from animal tissues. Journal of
473 Biological Chemistry, 266: 497-509.
474 Gaspar M.B., Ferreiro R. & Monteiro C.C. 1999. Growth and
475 reproductive cycle of Donax trunculus ( Mollusca, Bivalvia) off Faro,
476 southern Portugal. Fisheries Research, 41: 309-316.
477 Guillou J. & Bayed A. 1991. Contraintes du milieu sur les populations de
478 Donax trunculus L. et Donax venustus Poli du littoral Atlantique Marocain.
479 Oceanologica Acta, 14 (3) :291-298. 480 Guillou J. 1982. Variabilité des populations de Donax trunculus et Donax
481 vittatus en Baie de Douarnez. Netherlands Journal of Sea Research,16 : 88-
482 95.
483 Henderson R.J., Sargent J.R. & Hopkins C.C.E. 1984. Changes in the
484 Content and Fatty Acid Composition of Lipid in an Isolated Population of
485 the Capelin Mallotus villosus During Sexual Maturation and Spawning,
486 Marine Biology, 78: 255–263.
487 Holland D.L. 1978. Lipid reserves and energy metabolism in the larvae of
488 benthic marine invertebrates. In: Malins, D.C., Sargent, J.R. (Eds),
489 Biochemical and Biophysical Perspectives in Marine Biology, vol.4, pp 85-
490 123. Academic Press, New York.
491 Ingemansson T., Olsson N.U. & Kaufmann P. 1993. Lipid composition of
492 light and dark muscle of rainbow trout (Oncorhynchus mykiss) after thermal
493 acclimation: A multivariate approach. Aquaculture, 113: 153-165.
494 Klingensmith J.S. 1982. Disrtibution of methylene and nonmethylene-
495 interrupted dienoic fatty acids in polar lipids and triacylglycerols of selected
496 tissue of the hardshell clam (Mercenaria mercenaria). Lipids, 17: 976-981.
497 Laruelle F., Guillou J., Paulet Y. M. 1994. Reproductive pattern of the
498 clams, Ruditapes decussatus and Ruditapes philippinarium on intertidal flats
499 in Brittany. Journal of the Marine Biological Association of the United
500 Kingdom, 74: 351-366.
501 Lubet P. 1976. Ecophysiolohie de la reproduction chez les mollusques
502 lamellibranches. Haliotis, Paris, 7: 49-55
503 Lucas A. & Beninger P.G. 1985. The use of physiological condition
504 indices in marine bivalve aquaculture. Aquaculture, 44: 187-200.
505 Mackie L.A. & Ansell A.D. 1993. Differences in reproductive ecology in
506 natural and transplanted populations of Pecten maximus: evidence for the 507 existence of separate stocks. Journal of Experimental Marine Biology and
508 Ecology, 169: 57-75.
509 Manca Zeichen M., Agnesi S., Maccaroni A. & Ardizzone G.D. 2002.
510 Biology and population dynamics of Donax trunculus L. (Bivalvia:
511 Donacidae) in the South Adriatic Coast (Italy). Estuarine, Coastal and Shelf
512 Science, 54: 971-982.
513 Mann R. 1979. Some biochemical and physiological aspects of growth and
514 gametogenesis in Crassostrea gigas and Ostrea edulis grown at sustained
515 elevated temperatures. Journal of the Marine Biological Association of the
516 United Kingdom, 59: 95-100.
517 Martinez G. 1991. Seasonal variation in biochemical composition of three
518 size classes of Chilean scallop Agropecten purpuratus Lamarck, 1819.
519 Veliger, 34: 335-343.
520 Matozzo V., Tomei A. & Marin M.G. 2005. Acetylcholinesterase as a
521 biomarker of exposure to neurotoxic compounds in the clam Tapes
522 philippinarium from the Lagoon of Venice. Marine Pollution Bulletin, 50:
523 1686-1693.
524 Mazé R.A. 1990. Estimacion de la produccion del bivalva marino Donax
525 trunculus L. en la ria de El Baquero (Galicia). Iberus ,9(1-2) : 155-160.
526 Mazé R.A. & Laborda A.J. 1988. Aspectos de la dinamica de poblacion de
527 Donax trunculus L. (Bivalvia: Donacidae) en la ria de El Barquero (Lugo,
528 NW Espaňa). Investigacion Pesquera Barcelona, 52: 299-312.
529 Moueza M. & Chassel D. 1976. Contribution à l’étude de la biologie de
530 Donax trunculus L. dans l’Algerois : analyse statistique de la dispersion le 531 long d’une plage en baie de Bou Ismaïl. Journal of Experimental Marine
532 Biology and Ecology, 21: 211-221.
533 Mouêza M. & Frenkiel-Renault L. 1973. Contribution à l’étude de la
534 biologie de Donax trunculus (Mollusque, lamellibranches) dans l’Algérois :
535 La reproduction. Cahier de biologie Marine, 14: 261-283.
536 Neuberger-Cywiak L., Achituv Y., Mizrahi L. 1990. The ecology of
537 Donax trunculus Linnaeus and Donax semistriatus Poli from the
538 Mediterranean coast of Israel. Journal of Experimental Marine Biology and
539 Ecology, 134: 203-220.
540 Ojea J., Pazos A.J., Martinez D., Novoa Sanchez S.J.L. & Abad M.
541 2004. Seasonal variation in weight and biochemical composition of the
542 tissues of Ruditapes decussatus in relation to the gametogenic cycle.
543 Aquaculture, 238: 451-468.
544 Orban E., Di Lena G., Masci M., Nevigato T., Casini I. & Caproni R.
545 2004. Growth, nutritional quality and safety of oysters (Crassostrea gigas)
546 cultured in the lagoon of Venice (Italy). Journal of the science of food and
547 agriculture, 84:1929-1938.
548 Orban E., Di Lena G., Nevigato T., Casini I., Caproni R., Santaroni G.
549 & Guilini G. 2006. Nutritional and commercial quality of the striped Venus
550 clam, Chamelea gallina, from the Adriatic Sea. Food Chemistry, 101: 1063-
551 1070.
552 Pazos Antonio J., Sanchez José L., Roman Guillermo, Perez-Parallé M
553 Luz. & Abad Marcelina 2003. Seasonal changes in lipid classes and fatty
554 acids composition in the digestive gland of Pecten maximus. Comparative
555 Biochemistry and Physiology, 134(B): 367-380. 556 Pollero R.J., Ré M.E. & Brenner R.R. 1979. Seasonal changes of the
557 lipids of the mollusc Chlamys telmelcha. Comparative Biochemistry and
558 Physiology, 64 (A): 257-263.
559 Ruiz-Azcona P., Rodríguez- Sierra R. & Martin J.B. 1996. Culture of
560 coquina clam, Donax trunculus, larvae. Aquaculture, 139: 151-155.
561 Ruiz C., Martinez D., Mosquera G., Abad M. & Sàanchez J. L. 1992.
562 Seasonal variations in condition, reproductive activity and biochemical
563 composition of the flat oyster, Ostrea edulis, from San Cibran (Galicia,
564 Spain). Marine Biology, 112: 67-74.
565 Saito H. 2004. Lipid and Fatty acid composition of the Pearl Oyster
566 Pinctada fucata martensii: Influence of Season and Maturation. Lipids, 39:
567 997- 1005.
568 Sargent, J.R. 1976. The structure, metabolism and function of lipids in
569 marine organisms. In: Malin, D.C., Sargent, J.R. (Eds), Biochemical and
570 Biophysical Perspective in Marine Biology Vol.3. pp. 149-212. Academic
571 Press, London.
572 Tebble, N. 1966. British bivalve seashells. A Handbook for identification.
573 British Museum, Edinburgh. p.212.
574 Walne P.R. 1976. Experiments on the culture in the sea of the butterfish
575 Venerupis decussata L. Aquaculture, 8: 371-381.
576 Zhukova N.V. 1991. The pathway of the biosynthesis of, non-methylene-
577 interrupted dienoic fatty acids in molluscs. Comparative Biochemestry and
578 Physiology, 110(B): 801-804.
579
580
581 582 Figure legends:
583 Figure 1: Localisation of the three sampling sites in the Gulf of Tunis (North
584 Tunisia).
585 Figure 1: Localisation des trois stations d’étude dans le Golfe de Tunis
586 (Nord de la Tunisie)
587 Figure 2: Monthly variation of water temperature and salinity in Borj Cedria
588 (BC), Radès (R) and Kalâat El Andalous (KA) during the year 2004-2005.
589 Figure 2: Variation mensuelle de la température et de la salinité de l’eau
590 dans les stations de Borj Cedria (BC), Radès (R) et Kalâat El Andalous
591 (KA).
592 Figure 3: Seasonal variations in Condition Index (a), percentage of dry
593 weight (b) and Lipids (g/100g DW) (c) in D. trunculus from Borj Cedria
594 (BC), Radès (R) and Kalâat El Andalous (KA) during the year 2004-2005.
595 Figure 3: variation saisonnière de l’indice de condition (a), des pourcentages
596 de la matière sèche (b) et du contenu en lipides (g/100g PS) (c) chez D.
597 trunculus de Borj Cedria (BC), Rades (R) et Kalâat El Andalous (KA).
598
599
600
601
602
603
604
605
606
607
608 609 Table legends:
610 Table 1: Geographic position of Borj Cedria (BC), Radès (R) and Kalâat El
611 Andalous (KA) in the Gulf of Tunis.
612 Tableau 1: Position géographique de Borj Cedria (BC), Radès (R) et Kalaat
613 El Andalous (KA) dans le Golfe de Tunis.
614 Table 2: Seasonal biometric characteristics of the shells of coquina clam Donax
615 trunculus harvested from Borj Cedria (BC), Radès (R) and Kalâat El Andalous
616 (KA) during the year 2004-2005.
617 Means of 30 individuals ± standard deviation.
618 Tableau 2: Caractéristiques saisonnières de la biométrie de la coquille de D.
619 trunculus dans les trois stations d’étude : Borj Cedria (BC), Rades (R) et Kalaat
620 El Andalous (KA) durant l’année 2004-2005.
621 Moyenne de 30 individus ± écartypes
622 Table 3: Seasonal variations of the fatty acid composition (% of total fatty acids,
623 means of six replicates ± standard deviation) in the total mass of D. trunculus
624 harvested from Radès (R), Kalâat El Andalous (KA) and Borj Cedria (BC) during
625 2004-2005.
626 UK: Unknown Fatty Acid, nd: Not detected, 22:2i and 22:2j are possibly Non-
627 Methylene- Interrupted- Dionic fatty acids (NMID).
628 Tableau 3: Variations saisonnières de la composition en acides gras (% des acides
629 gras totaux, la moyenne de six réplications± écartypes) dans la masse totale de D.
630 trunculus récolté dans les stations de Radès (R), Kalâat El Andalous (KA) et Borj
631 Cedria (BC) durant l’année 2004-2005.
632 UK : Acide gras inconnu, nd : valeur non détectée, 22:2i et 22:2j sont
633 probablement les Non-Methylene- Interrupted- Dionic fatty acids (NMID).
634 635
636 Figure 1
637
638
639 640 Figure 2
641
642
643
644 645
646 Figure 3
647
648
649 650 Table 1
R KA BC North 36°46 37°03 36°43 East 10°17 10°10 10°26 651
652 Table 2
Length (mm) Height (mm) Thickness (mm) Summer 22.8 ±2.0 11.8 ±0.9 7.1±0.6 R Autumn 25.0 ±1.1 12.9 ±0.5 7.8±0.5 Winter 28.1 ±1.7 14.6 ±0.7 9.0±0.6 Spring 31.6 ±1.6 16.5 ±0.7 10.3±0.8 Summer 27.9 ±1.7 15.2 ±0.8 8.8±0.5 KA Autumn 27.7 ±1.3 15.1 ±0.7 8.8±0.5 Winter 28.3 ±1.2 15.0 ±0.5 8.7±0.3 Spring 30.0 ±1.0 16.1±0.5 9.5±0.4 Summer 28.4 ±1.2 14.9±0.5 9.0±0.4 BC Autumn 30.0 ±2.0 15.8±1.1 9.7±0.7 Winter 31.0 ±1.3 16.3±0.4 9.8±0.4 Spring 32.1 ±2.2 16.9±1.1 10.3±0.7 653
654
655
656
657
658
659
660
661
662
663
664
665
666
667 668 Table 3
R KA BC Fatty acids Summer Autumn Winter Spring Summer Autumn Winter Spring Summer Autumn Winter Spring C14:0 3.1±0.5 2.5±0.5 2.1±0.8 3.5±1.9 3.1±1.2 3.0±0.5 2.2±0.8 4.5±2.1 3.8±0.9 2.3±0.4 2.8±0.4 3.5±0.7 C15:0 0.1±0 0.1±0 0.1±0 0.1±0 0.1±0 0.1±0 0.1±0 0.1±0 0.2±0 0.4±0.2 0.3±0.1 0.2±0 C15:1 0.2±0 0.2±0 0.2±0 0.2±0 nd 0.2±0 0.2±0.1 0.1±0 0.3±0 0.3±0.1 0.3±0.1 0.3±0.1 C16:0 18.1±0.6a 22.0±1.1b 16.8±1.1a 17.3±0.9a 19.0±2b 20.9±2b 21.4±1b 20.3±0.7b 20.5±1.9b 19.5±1.3b 17.9±1a 18.1±0.7a C16:1n-9 3.9±1.0 4.7±1.4 6.6±1.5 6.2±1.6 4.1±2.5 4.2±1.3 3.2±1.8 5.1±1.1 4.7±1.7 2.4±0.6 3.2±0.7 5.0±0.7 C17:0 0.1±0 0.2±0.1 0.1±0 0.3±0.2 0.2±0.1 0.2±0.1 0.1±0.1 0.1±0 0.3±0.1 0.2±0.1 0.2±0.1 0.4±0.1 C16:3 0.3±0.2 0.4±0.2 0.3±0.1 0.4±0.2 0.5±0.2 0.4±0.2 0.3±0.2 0.2±0.1 0.8±0.3 0.5±0.2 0.8±0.3 0.7±0.3 UK 0.3±0.1 0.3±0 0.2±0.1 0.3±0.1 0.3±0 0.3±0 0.2±0.2 0.3±0 0.3±0.1 0.4±0.1 0.5±0.2 0.4±0.1 C16:4 nd 0.1±0.1 0.1±0 0.3±0.3 nd 0.1±0 0.1±0 0.1±0.1 1.8±1.2 2.9±0.8 2.0±0.4 1.5±0.5 C18:0 10.6±0.9a 7.5±0.7b 6.5±0.9b 7.1±0.8b 9.7±1.2a 9.2±2a 9.3±1.5a 7.4±0.7b 8.8±1.1a 9.1±0.8a 7.8±1.3b 7.5±0.4b C18:1n-9 6.3±0.7b 6.3±0.3b 4.7±0.8a 4.4±0.7a 5.6±1b 5.5±0.4b 5.1±1a 4.2±0.6a 6.3±1.2b 7.3±0.5b 4.1±1.7a 3.5±1.5a C18:1n-7 3.1±0.6 2.6±0.3 5.1±0.6 4.2±0.8 2.5±0.8 2.6±0.3 2.4±0.5 3.2±0.7 3.1±0 7.4± 3.1±0 3.0±0.4 C18:2n-6 1.1±0.3 2.1±0.3 2.1±0.3 1.5±0.3 1.2±0.2 1.4±0.1 1.1±0.4 1.1±0.2 1.5±0.2 1.5±0.2 1.0±0.2 1.1±0.2 C18:3n-6 0.3±0.1 0.2±0.1 0.2±0 0.2±0.1 0.4±0.1 0.3±0 0.2±0 0.3±0.1 0.2±0 0.2±0 0.2±0 0.3±0.1 C18:3n-3 0.6±0.3 0.7±0.1 3.5±0.2 2.0±0.5 0.8±0.2 1.3±0.6 1.2±0.6 1.9±0.3 1.0±0.3 0.7±0.3 0.9±0.2 1.7±0.3 C18:4n-3 0.7±0.3 1.3±0.2 3.5±0.5 2.8±0.5 0.9±0.4 1.2±0.7 1.8±0.5 2.0±0.3 1.3±0. 0.5±0.2 1.9±0.4 2.9±0.5 C20:0 0.2±0 0.2±0 0.2±0 0.2±0.1 0.2±0.1 0.2±0.1 0.2±0.1 0.1±0.1 0.3±0. 0.2±0 0.2±0 0.2±0 C20:1n-9 2.0±0.1 1.5±0.3 2.6±0.3 2.6±0.6 1.9±0.3 2.1±0.7 2.0±0.4 2.2±0.2 4.1±1.1 4.7±0.5 3.6±0.6 3.3±0.7 C20:2n-6 2.5±0.1 2.0±0.4 1.5±0.5 1.9±0.4 2.0±0.2 1.7±0.5 1.3±0.2 1.6±0.2 2.7±0.9 2.7±0.7 1.8±0.2 1.7±0.2 C20:3n-6 0.5±0.1 0.4±0 0.2±0 0.2±0 0.3±0 0.3±0 0.2±0 0.2±0 0.7±0.1 0.8±0.2 0.9±0.1 0.6±0 C20:4n-6 2.4±0.2b 2.5±0.2b 1.3±0.2a 1.4±0.2a 2.7±0.5b 3.3±0.5c 2.8±0.4c 2.1±0.3b 2.3±0.3b 3.6±0.2c 2.7±0.2b 2.2±0.2b C20:3n-3 0.1±0.1 0.1±0 0.7±0.2 0.5±0.1 0.3±0.1 0.2±0.1 0.3±0.2 0.4±0.1 0.3±0.1 0.2±0.1 0.5±0.2 0.5±0.2 C20:4n-3 0.4±0.1 0.5±0.1 1.0±0.2 1.1±0.1 0.6±0.2 0.5±1.7 0.8±0.3 0.8±0.1 0.6±0 0.4±0.1 0.9±0.1 1.2±0.3 C20:5n-3 6.4±1.1a 10.8±1.4c 8.0±0.5b 10.2±1.1c 6.9±1.1a 7.7±1.7a 7.7±0.5a 9.2±0.3b 7.6±1.4a 6.4±0.9a 9.3±0.9b 11.4±0.8c C22:0 0.2±0 0.04±0 0.04±0 0.1±0 0.1±0 0.1±0 0.2±0.2 nd 0.1±0 nd nd nd C22:1n-9 0.2±0 0.1±0 0.1±0 0.3±0.3 0.1±0 0.2±0.2 0.2±0 0.3±0 0.6±0.5 0.3±0.2 nd 0.2±0.2 C22 :2i 3.5±0.6 1.9±0.4 1.4±0.2 1.7±0.4 3.1±1 2.2±0.3 2.4±0.4 1.7±0.9 2.0±0.4 3.3±0.5 2.3±0.6 1.6±0.3 C22 :2j 1.0±0.1 0.5±0.1 0.3±0.1 0.4±0.1 0.7±0.3 0.4±0.1 0.4±0.2 0.4±0.2 0.7±0.3 0.9±0.1 0.7±0.2 0.4±0.2 C22:3n-3 0.4±0 0.7±0 0.3±0 0.4±0.3 0.8±0.2 0.7±0.2 0.7±0.1 0.6±0.1 0.7±0.1 1.1±0.1 0.8±0.1 0.7±0.1 C22:5n-3 2.3±0.4 2.1±0.4 2.1±0.3 2.1±2 2.2±0.5 1.9±0.2 2.2±0.3 2.0±0.4 2.2±0.5 2.8±0.3 2.6±0.4 2.6±0.3 C22:6n-3 9±0.7a 10.6±0.8a 12.6±1.1a 12.3±1.4a 9.1±1.7a 9.2±1a 11.1±5.4a 13.2±1.5a 12.6±1.1a 13.1±1.2a 16.5±0.9a 14.3±1.4a SFA 32.2±1.2c 32.5±1.1c 25.8±1.3ab 28.6±1a 32.3±2c 33.6±1.2c 33.3±1.5c 32.6±1.2c 34.7±2.4c 31.6±1.1c 28.8±1.9a 29.6±0.6a MUFA 15.6±1b 14.0±2.8b 19.2±2.2a 17.9±2a 14.6±2.3b 14.8±0.5b 12.6±2.2b 14.9±1.9b 16.8±2.3b 17.2±5.2b 13±2.2b 15.5±2.1b PUFA 31.6±2a 37.1±1.2b 39.4±1.4b 39.5±2b 32.9±1.8a 33±2a 34.7±5.7a 38±2b 38.7±2.2b 41.8±1b 46±2.2c 46.1±1.2c PUFA n-3 19.5±2.3a 26.4±1.1c 31.6±1.6b 31.2±2.2b 21.1±2.3a 22.1±2.1a 25.2±5.8a 29.7±1.7c 25.9±0.7c 24.5±1.5c 33±1.5b 34.9±1.4b PUFA n-6 7.2±0.3b 7.8±0.4b 5.6±0.5c 5.7±0.8c 7.5±0.6b 7.7±0.1b 6.4±0.5c 6.0±0.7c 8.0±1.2b 9.7±0.6a 7.1±0.5b 6.4±0.4c Lipids 8.1±3.9a 6±0.6b 7.3±1b 5.9±1.7b 10.8±1.6a 5.6±0.9b 5.9±0.7b 6.06±1b 5.9±1.4b 4.8±0.9b 7.3±2.1b 4.9±1.9b n- 3/n-6 2.8±0.8 4.4±0.5 7.4±1.3 6.8±1.1 3.3±0.6 3.3±0.2 4.4±1.2 5.6±0.3 3.5±0.7 3.7±1 4.8±1.5 5.6±0.6 669