Dietary fish oil replacement with lard and soybean oil affects triacylglycerol and phospholipid muscle and liver docosahexaenoic acid content but not in the brain and eyes of surubim juveniles Pseudoplatystoma sp

Abstract

Triplicate groups of juvenile suribim were fed for 183 days one of four different isonitrogenous (47.6% crude protein) and isolipidic (18.7% lipid) diets formulated using three different lipid sources: 100% fish oil (FO, diet 1); 100% pig lard (L, diet 2); 100% soybean oil (SO, diet 3), and FO/L/SO (1:1:1, w/w/w; diet 4). The tissue levels of fatty acids 18:2n-6 and 18:3n-3 decreased relative to corresponding dietary fatty acid values. The 20:5n-3 and 22:6n-3 composition of muscle and liver neutral lipids were linearly correlated with corresponding dietary fatty acid composition. In contrast, the 22:6n-3 composition of the brain and eye were similar among treatments. The 22:6n-3 level was enriched in all tissues, particularly in the neural tissues. Similar results were observed for tissue polar lipids: fatty acids content reflected dietary composition, with the exception of the 22:6n-3 level, which showed enrichment and no differences between groups. Given these results, the importance of the biochemical functions (transport and/or metabolism) of 22:6n-3 in the development of the neural system of surubim warrants further investigation.

Full text

Dietary fish oil replacement with lard and soybean oil affects
triacylglycerol and phospholipid muscle and liver
docosahexaenoic acid content but not in the brain
and eyes of surubim juveniles Pseudoplatystoma sp.
M. D. Noffs ÆR. C. Martino ÆL. C. Trugo Æ
E. C. Urbinati ÆJ. B. K. Fernandes Æ
L. S. Takahashi
Received: 11 April 2008 / Accepted: 25 August 2008 / Published online: 7 November 2008
ÓSpringer Science+Business Media B.V. 2008
Abstract Triplicate groups of juvenile suribim
were fed for 183 days one of four different isonitr-
ogenous (47.6% crude protein) and isolipidic (18.7%
lipid) diets formulated using three different lipid
sources: 100% fish oil (FO, diet 1); 100% pig lard
(L, diet 2); 100% soybean oil (SO, diet 3), and FO/L/
SO (1:1:1, w/w/w; diet 4). The tissue levels of fatty
acids 18:2n -6 and 18:3n -3 decreased relative to
corresponding dietary fatty acid values. The 20:5n
-3 and 22:6n -3 composition of muscle and liver
neutral lipids were linearly correlated with corre-
sponding dietary fatty acid composition. In contrast,
the 22:6n -3 composition of the brain and eye were
similar among treatments. The 22:6n -3 level was
enriched in all tissues, particularly in the neural
tissues. Similar results were observed for tissue polar
lipids: fatty acids content reflected dietary composi-
tion, with the exception of the 22:6n -3 level, which
showed enrichment and no differences between
groups. Given these results, the importance of the
biochemical functions (transport and/or metabolism)
of 22:6n -3 in the development of the neural system
of surubim warrants further investigation.
Keywords Fish oil Lipid metabolism 
Phospholipids Pig lard Polyunsaturated
fatty acids Pseudoplatystoma sp. Soybean oil 
Triacylglycerols
Introduction
The established trend in fish feeds is to use higher
levels of dietary lipids, particularly for carnivorous
species. Fish oil is the main dietary lipid source in
aquaculture feeds and is used to increase the energy
content of the feed and to provide essential fatty acids
(Tacon and Jackson 1985). Although increasing
levels of dietary lipids can help reduce the high costs
of diets by partially sparing protein in the feed, there
are a number of problems associated with its use,
such as excessive fat deposition in the liver, which
can lead to decreased growth performance (Caballero
M.D. Noffs and R.C. Martino dedicate this study to the
memory of Prof. Dr. Luiz Carlos Trugo.
M. D. Noffs L. C. Trugo
Instituto de Quı
´mica (CT), Universidade Federal do Rio
de Janeiro–Laborato
´rio de Bioquı
´mica Nutricional e de
Alimentos, Bl. A Lab. 528, Ilha do Funda
˜o,
21949-900 Rio de Janeiro, RJ, Brazil
R. C. Martino (&)
Fundac¸a
˜o Instituto de Pesca do Estado do Rio de Janeiro,
Unidade de Tecnologia do Pescado, Avenida das
Ame
´ricas 31501, Guaratiba, 23032-050 Rio de Janeiro,
RJ, Brazil
e-mail: rmartino.rlk@terra.com.br
E. C. Urbinati J. B. K. Fernandes L. S. Takahashi
Centro de Aqu
¨icultura da UNESP (CAUNESP),
Universidade do Estado de Sa
˜o Paulo, Via de Acesso
Rodovia Carlos Tonanni, Km 5, 14870-000 Jaboticabal,
SP, Brazil
123
Fish Physiol Biochem (2009) 35:399–412
DOI 10.1007/s10695-008-9264-8

et al. 1999;2002) and even alter the market quality of
the fish (O
¨zogul et al. 2005).
The level of the fatty acid 22:6n -3 (docosahex-
aenoic acid, DHA) actively increases in the central
nervous system during the developmental period,
primarily relying on circulating plasma 22:6n -3
derived from the diet or from biosynthesis in the liver.
However, local biosynthesis of 22:6n -3 also occurs
in the brain, thereby providing an alternative source of
22:6n -3 for its accumulation in the brain. It is well
established that 22:6n -3 can be biosynthesized from
a-linolenic acid (18:3n -3), a shorter chain n -3
fatty acid precursor, through chain elongation and
desaturation processes. The 18:3n -3 is desaturated
to 18:4n -3byD6-desaturase, chain-elongated to
20:4n -3, subsequently converted to 20:5n -3by
D5-desaturase, and then further elongated to 22:5n
-3 in the endoplasmic reticulum (ER). In vertebrates,
22:5n -3 is chain-elongated to 24:5n -3 followed
by desaturation through D6-desaturase to 24:6n -3.
Afterwards, 24:6n -3 is transferred to peroxisomes
and converted to 22:6n -3 through the removal of two
carbon chains by b-oxidation (Tocher 2003).
The liver is considered to be the primary site for
biosynthesis of 22:6n -3 (Tocher 2003), which
becomes available for uptake by the brain through
subsequent secretion into the circulating blood
stream. Among neural cells, the capacity to synthe-
size 22:6n -3 has been demonstrated only in
astrocytes (Moore et al. 1991). Despite the fact that
neurons are major targets for 22:6n -3 accumula-
tion, they cannot produce 22:6n -3 because of a
lack of desaturase activity. Cerebromicrovascular
endothelia can also elongate and desaturate shorter
carbon chain fatty acids. However, they cannot
perform the final desaturation step to produce either
22:5n -6 or 22:6n -3 (Moore et al. 1990). The
22:6n -3 synthesis in astrocytes is negatively influ-
enced by the availability of preformed 22:6n -3
(Williard et al. 2001) and thus may represent a
quantitatively minor source for the neural 22:6n -3
accretion when the circulating 22:6n -3 supply is
adequate. The incorporation of circulating 22:6n -3
across the blood brain barrier appears to be an
important route for maintaining adequate levels of
22:6n -3 in the brain.
In agreement with this concept, the lipid class
composition of cultured trout astrocytes is similar to
that of mammals (Tocher and Sargent 1990), as
demonstrated in astrocyte and mammalian brain cell
cultures, with most of the poly-unsaturated fatty acids
(PUFA) being incorporated into polar lipids (Tocher
and Sargent 1990), although fish astrocytes did have
relatively low levels of a4 desaturation activity
(Tocher 1993). In general, it is difficult to deplete
22:6n -3 from the neural membranes of adult
mammals even when the latter are on a diet low in
22:6n -3, presumably because of preferential
uptake of 22:6n -3 into the brain to support the
basal turnover. Although 22:6n -3 can be lost in the
case of an insufficient supply of n -3 fatty acids
during development, this loss is compensated with
22:5n -6, 22:4n -6, and 20:4n -6 through reci-
procal replacement (Greiner et al. 2003), suggesting a
requirement of very long-chain, highly unsaturated
fatty acids in neural membranes.
Many authors (Watanabe 1982; Sargent et al.
1999; Martino et al. 2002a,b,2003) have reported
the importance of utilizing dietary lipids for carniv-
orous fish species. These lipids are the most
appropriate source of energy for these species, owing
to their low ability to digest carbohydrate (NRC
1993). They are also sources of metabolic energy in
gonad formation, particularly in female fish, and
function as structural components for cellular mem-
brane production (Rainuzzo et al. 1997). In addition,
increasing consumer demand for fish products has
increased the utilization of fish oil by the aquaculture
industry, with a predicted rise of 30% between 2002
and 2010 (Barlow 2002). This scenario is providing
researchers with an incentive to find substitutes for
this essential raw material.
The aim of our study was to investigate the
metabolism of dietary fatty acids in the muscle, liver,
brain, and eyes of surubim juveniles, Pseudoplatys-
toma sp., particularly the tissue incorporation of
eicosapentaenoic acid (EPA, 20:5n -3) and DHA,
in response to different fatty acid compositions in
alternative dietary lipid sources.
Material and methods
Experimental fish, husbandry and feeding trial
Juvenile hybrid surubim fish (average weight
5±1.0 g) were obtained from a commercial fish
400 Fish Physiol Biochem (2009) 35:399–412
123

farm, Projeto Pacu (Campo Grande, MS, Brazil), and
allowed to acclimate for 3 months prior to the
feeding trial in 1000-l tanks equipped with a flow-
through freshwater system. The fish were fed twice
daily with a commercial fish diet (Nutron Alimentos,
Campinas, SP, Brazil) comprising 40% crude protein
and 10% crude lipid.
Prior to sampling, the fish were starved for 24 h and
then sorted based on weight and physical appearance.
In May 2004, surubim juveniles (n=204) with an
initial weight of about 28.0 ±5.3 g were distributed
(17 fish per cage) into 12 net cages (capacity 84 l)
placed in a 3000-l circular fiberglass tanks supplied
with recirculated spring freshwater. The tanks were
located at a facility of the Unesp Aquaculture Center
(CAUNESP) (Jaboticabal, Sa
˜o Paulo, Brazil). Fresh-
water temperatures over the experimental period (May
2004–November 2004) ranged from 23.0 to 28.1°C,
with a mean temperature of 26.6 ±1.0°C. The water
temperature was regulated using an electric heat
system. The fish were manually fed to apparent
satiation twice daily (19:00 and 23:00 h) with a
sinking pelleted feed for 183 days. Three replicates
cages were randomly assigned for each diet. The tanks
and the cages were cleaned every other day. The water
quality parameters of dissolved oxygen, pH, and
conductivity were periodically checked in each tank.
After the first 88 days, the fish were starved for 24 h,
and then seven fish from each net cage were killed with
0.5 g/l of benzocaine and frozen for subsequent
analysis of polar and neutral fatty acid composition
in the muscle, liver, brain, and eyes. The remaining
fish were kept in the cages for another 95 days. At the
end of the experimental period, all fish in each
treatment were weighed, killed, and submitted to the
same analysis as that for the 88-day sample. A
representative edible portion of fish muscle was
minced in a blender (Walita Gama, Sa
˜o Paulo, Brazil),
while the livers, brains, and eyes were minced in a
potter homogenizer unity (Tecnal, Te-099, Piracicaba,
SP, Brazil), placed in plastic bags under liquid
nitrogen, and stored in a freezer at -20°C to prevent
oxidation until subsequent analysis.
Four practical-type pelleted diets were formulated
to provide 47.6% crude protein and 18.7% lipid. The
pellets were prepared by mixing the ingredients with
distilled water until consistent. The mixture was cold
pelleted in a meat processor machine into 3- and
6-mm pellets. The pellets were then dried in an oven
(model 320-SE; FANAN, Sa
˜o Paulo, Brazil) with
circulating air at 40°C for approximately 15 h. In
order to minimize lipid oxidation, pellets were placed
in plastic bags and stored in a freezer at -20°C until
the fish were fed. About 40 g of each diet was ground
in an analytical grinder (model Q-298A21; Quimis,
Diadema, Brazil) for proximate composition and fatty
acid analysis. These diets differed only in their
sources of dietary lipids. The oil added to each diet
was either 100% fish oil (100% FO, control diet),
100% pig lard (100% L, diet 2), 100% soybean oil
(100% SO, diet 3), and FO/L/SO (1:1:1, w/w/w, diet
4) (Table 1). The diets were designed according to
Martino et al. (2002a,b) to satisfy the nutritional
requirements of the fish. The proximate composition of
the experimental diets is shown in Table 1, and the
dietary fatty acid compositions are shown in Table 2.
The substitution of the three dietary lipids was reflected
in the fatty acid compositions of the four diets (Table 2).
Sampling procedure
An initial sample of 40 fish was taken at the start of
the experiment to determine the baseline levels of
lipid and fatty acid compositions in the muscle, liver,
brain, and eyes. After 88 days, similar samples of
seven fish were selected at random from each cage.
The samples from each cage replicate were pooled by
treatment. The pooled tissues were frozen and stored
at -20°C in plastic bags under liquid nitrogen until
analysis: muscle was skinned, deboned, and homog-
enized in a food processor (Walita Gama, Sa
˜o Paulo,
Brazil); liver, brain, and eyes were homogenized in a
potter homogenizer unity (model Te-099; Tecnal,
Piracicaba, Brazil). At the end of the experimental
period (183 days), the remaining fish was taken as
described above. Each fish tissue was minced and
then stored as described previously.
Proximate composition
The composition of the four experimental diets and
muscle samples was determined by proximate anal-
ysis. Moisture, crude protein, and ash contents were
submitted to chemical analysis according to the
procedures of the AOAC (2000) and as described in
a previous publications (Martino et al. 2002a,b,
2003,2005; Campos et al. 2006). The crude fiber
content of the diets was calculated based on
Fish Physiol Biochem (2009) 35:399–412 401
123

published data of the NRC (1993) feed ingredient
compositions.
Lipid extraction, lipid classes, and fatty acid
analysis
Total lipid was extracted from the diets and tissues
(muscle, liver, brain, and eyes) with chloroform/
methanol (2:1, v/v) according to the Folch et al.
(1957) method. The weight of the lipid was determined
gravimetrically after evaporation of the solvent. Fatty
acids of the diets were obtained from the total lipid by
saponification with KOH (50%).
Fatty acid methyl esters (FAME) were obtained by
esterification with boron-trifluoride in methanol (7%)
according to Metcalfe and Schmitz (1961). The
FAME were separated by chromatography on a
Shimadzu GC 15-A (Shimadzu Corp, Kyoto, Japan)
equipped with a flame ionization detector and fitted
with a fused silica capillary column (Omegawax
320 930 m 90.32 mm i.d; Supelco, Bellefonte,
PA). Hydrogen was used as a carrier gas at a flow
Table 1 Formulation and proximate composition (wt %) of the experimental diets
Formulation and proximate composition of diets Lipid sources
100% FO 100% L 100% SO Blend
Ingredients
a
Fish meal 55 55 55 55
Corn gluten meal 8 8 8 8
Soybean meal 4 4 4 4
Wheat flour 14 14 14 14
Corn flour 5 5 5 5
Fish oil 12 4
Lard 12 4
Soybean oil 12 4
Vitamin/mineral mixture
b
1.99 1.99 1.99 1.99
BHT
c
0.01 0.01 0.01 0.01
Proximate composition (g/100 g)
Moisture 4.4 4.2 5.5 4.2
CP 47.5 47.3 47.8 47.6
CL 18.6 18.8 18.7 18.5
Ash 8.0 7.9 8.0 8.1
Crude fiber
d
1.2 1.2 1.2 1.2
NFE
e
21.6 21.8 20.0 21.7
FO, fish oil; L, lard; SO, soybean oil; CP, crude protein; CL, crude lipid; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid
a
Sources: Fish meal-anchovy, from Nutron Alimentos, Brazil (CP: 65.6%, CL: 10.16%; EPA: 7.92%; DHA: 7.52%); Fish oil, from
Mogiana Alimentos, Brazil (EPA: 13.10%; DHA: 7.74%); corn gluten, locally available (CP: 60.9%, CL:8.0%); soybean meal,
locally available (CP: 41.1%, CL: 2.3%); wheat flour, locally available (CP: 11.3%, CL: 3.3%); corn flour, locally available (CP:
7.6%, CL: 2.4%)
b
Vitamin/mineral mixture, from Nutron Alimentos, Brazil, in units/kg of premix: antioxidant, 0.60 g; Vit. A, 1,000,000 IU; Vit. D3,
500,000 IU; Vit. E, 20,000 IU; Vit. K3, 500 mg; thiamin, 1250 mg; riboflavin, 2500 mg; pyridoxine, 2485 mg; pantothenic acid,
5000 mg; niacin, 5000 mg; biotin 125 mg; folic acid, 250 mg; cyanocobalamin, 3750 mg; ascorbic acid, 28,000 mg; cobalt (Co),
25 mg; copper (Cu), 2000 mg; iodine (I), 100 mg; iron (Fe), 13,820 mg; manganese (Mn), 3750 mg; zinc (Zn), 150 mg; selenium
(Se), 75 mg
c
Butilated hydroxy toluene (Inspec Spain, Spain)
d
Calculated based on published data ((NRC 1993)
e
Nitrogen-free extract =100 -(moisture ?crude protein ?crude lipid ?ash ?crude fiber)
402 Fish Physiol Biochem (2009) 35:399–412
123

rate of 1.4 ml/min. Injector and detector temperatures
were programmed to 240 and 250°C, respectively.
The column temperature was programmed to be
isothermal (205°C). Individual FAME were identified
by means of reference standards (Supelco) and
quantified by using a Shimadzu C-R4A integrator
model (Shimadzu).
The total lipid of fish tissues was separated into
polar and neutral fractions using the silica Sep-Pak
cartridges (Waters do Brasil, Sa
˜o Paulo, Brazil)
according to the method of Juaneda and Rocquelin
(1985). The FAME of the polar lipids were prepared
by esterification with boron-trifluoride (7%) plus
10 ml of benzene, reflux at 80°C for 20–30 min, and
then extraction with petroleum ether. The FAME of
neutral lipids were prepared as described for polar
lipids, but using ethyl ether instead of petroleum
ether for extraction. The gas chromatography
conditions for separation, identification, and quanti-
fication of individual FAME, were the same as
described above.
Statistical analysis
Significance of difference (P\0.05) between dietary
treatments was determined by one-way analysis of
variance (ANOVA). Differences between means
were determined by Tukey’s test. The ANOVA was
performed using a GRAPHPAD PRISM ver. 4.0 statistical
package (Graphpad Software, San Diego, CA).
Table 2 Fatty acid composition (expressed as % of total fatty acids) of the experimental diets
Fatty acids Lipid sources
100% FO 100% L 100% SO Blend
14:0 10.8 ±2.6 2.7 ±0.6 1.9 ±0.1 4.5 ±0.1
16:0 20.4 ±2.1 25.1 ±1.3 14.4 ±0.3 19.1 ±1.3
18:0 3.4 ±0.9 9.1 ±1.2 3.2 ±0.1 5.7 ±0.5
RSaturated
a
37.8 ±5.3 a 37.7 ±1.9 a 20.2 ±0.2 b 30.8 ±1.6 a,b
16:1
b
10.4 ±0.1 4.5 ±0.2 2.7 ±0.1 5.4 ±0.0
18:1n -9 11.8 ±3.7 32.0 ±3.5 20.7 ±0.7 22.2 ±0.2
RMonounsaturated
c
24.1 ±2.9 b 37.5 ±3.9 a 24.3 ±0.7 b 28.9 ±0.3 b
18:2n -6 6.1 ±0.3 12.3 ±0.3 39.7 ±0.2 19.8 ±0.1
20:4n -6 1.0 ±0.5 0.5 ±0.0 0.4 ±0.0 0.7 ±0.2
Rn-6 Polyunsaturated
d, e
8.6 ±1.3 d 13.3 ±0.7 c 40.4 ±0.3 a 21.4 ±0.0 b
18:3n -3 1.0 ±0.1 0.8 ±0.1 4.9 ±0.3 2.0 ±0.1
20:5n -3 9.7 ±1.3 1.9 ±0.3 2.0 ±0.2 5.0 ±0.2
22:5n -3 1.1 ±0.4 0.8 ±0.2 0.7 ±0.1 1.0 ±0.2
22:6n -3 6.3 ±3.3 3.4 ±1.1 3.3 ±0.6 5.0 ±1.2
Rn-3 Polyunsaturated
e, f
20.5 ±4.4 a 7.6 ±1.8 b 11.8 ±1.3 a,b 14.3 ±1.5 a,b
Saturated/unsaturated 0.5 ±0.1 a 0.4 ±0.0 a 0.2 ±0.0 a 0.3 ±0.0 a
Rn-6 HUFA
g
1.5 ±0.7 a 0.6 ±0.0 a 0.4 ±0.1 a 1.0 ±0.2 a
Rn-3 HUFA
g
17.5 ±5.1 a 6.3 ±1.6 b 6.3 ±1.0 b 11.4 ±1.7 a,b
n-3/n – 6 2.4 ±0.2 a 0.6 ±0.1 b 0.3 ±0.0 b 0.7 ±0.1 b
DHA/EPA 0.6 ±0.3 c 1.8 ±0.3 a 1.7 ±0.1 a,b 1.0 ±0.2 b,c
Mean values (±standard deviation) followed by the same letter(s) within a row are not significantly different (P[0.05)
a
Includes fatty acids 15:0, 17:0, 20:0, and 22:0
b
Includes fatty acids 16:1 (n -7 and n -9)
c
Includes fatty acids 15:1 and 20:1 (n -11, n -9 and n -7)
d
Includes fatty acids 18:3n -6 and 22:4n -6
e
Polyunsaturated fatty acid (ÓC18:2n-)
f
Includes fatty acids 18:4n -3 and 20:4n -3
g
Highly unsaturated fatty acid (ÓC20:4n-)
Fish Physiol Biochem (2009) 35:399–412 403
123

Results
The experimental diets were well accepted by the
experimental fish. The dissolved oxygen, pH, and
conductivity of the water did not differ during the
experimental period (183 days), with standard values
of 6.8 mg/l, 7.5, and 147.8 ms/cm, respectively.
Natural mortality occurred during the experimental
period. The mean survival rate observed for the
treatments was 97.4% for 100% FO, 99.0% for 100%
L, 98.3% for 100% SO, and 100% for blend lipid
group (Table 3).
By the end of the feed trials, all groups of fish had
increased in weight by approximately eight fold.
Specific growth rates (SGR) varied between 1.10 and
1.19, while feed conversion ratios (FCR) were similar
among treatments, ranging from 0.87 to 0.99
(Table 3).
No differences were observed among the treatment
groups in terms of proximate composition of tissues
at the 88-day time point and at the end of the
experimental period. The proximate composition of
muscle was similar between groups (Table 4); how-
ever, muscle ash levels were greater in fish fed the
blend lipids than in those fed 100% FO. Total lipid in
the liver and eyes was not affected by dietary
treatment (Table 4). In contrast, the total lipid levels
in the brain were statistically greater (P\0.05) in
fish fed 100% L than in those fed 100% FO or 100%
SO.
The levels of phospholipids and the triacylglycerol
(TAG) fatty acid composition of the muscle, liver,
brain, and eye tissue are shown in Tables 5and 6,
respectively. The concentrations of fatty acids in
dietary lipid were related linearly to their concentra-
tions in tissue phospholipids and TAG. The slopes of
Table 3 Performance parameters of surubim juveniles fed experimental diets containing different lipid sources for 183 days
Parameter 100% FO 100% L 100% SO Blend
Initial weight (g) 28.9 ±2.1 25.8 ±1.7 28.3 ±2.4 29.0 ±2.2
Final weight (g) 239.9 ±10.8 226.6 ±26.4 232.5 ±18.4 216.1 ±29.5
Weight gain, fold 8.3 8.8 8.2 7.5
Mortalities, n2110
Survival, % 97.4 ±4.6 99.0 ±1.6 98.3 ±3.0 100
SGR (n=51) 1.16 ±0.02 1.19 ±0.06 1.15 ±0.03 1.10 ±0.07
FCR (n=30) 0.87 ±0.04 0.92 ±0.12 0.90 ±0.07 0.99 ±0.15
Where appropriate, values are given as means ±standard deviation
SGR, Specific growth rate; FCR, feed conversion ratio
Table 4 Proximate composition of muscle and lipid concentrations of liver, brain, and eyes in surubim fed diets containing different
lipid sources for 183 days
a
Parameter Initial 100% FO 100% L 100% SO Blend
Moisture
b
80.07 76.92 ±0.22 76.37 ±0.67 76.17 ±0.34 76.85 ±0.29
Protein
b
18.55 18.60 ±0.35 b 18.73 ±0.29 b 19.60 ±0.22 a 18.84 ±0.33 a,b
Ash
b
0.91 0.52 ±0.07 b 0.69 ±0.22 a,b 0.82 ±0.25 a,b 1.09 ±0.01 a
Muscle lipid
b
1.03 3.88 ±0.49 3.21 ±0.98 3.39 ±0.97 3.15 ±0.33
Liver lipid
b
2.29 3.54 ±1.26 2.92 ±0.16 3.66 ±0.21 3.60 ±1.08
Brain lipid
b
3.65 7.99 ±0.16 b 10.55 ±0.41 a 8.38 ±1.17 b 8.97 ±1.10 b
Eyes lipid
b
0.57 0.90 ±0.39 0.60 ±0.49 0.70 ±0.33 1.10 ±0.13
a
All values are means ±SD (n=30). Values within each row followed by a different letter (s) are significantly different at
P\0.05
b
Values are g/100 g of wet weight
404 Fish Physiol Biochem (2009) 35:399–412
123

the plots are different for each fatty acid, which
indicates that the relationship between the concen-
tration of a fatty acid in the diet and the concentration
in tissues is different for each fatty acid. The
correlation results indicate the differences (Dvalues)
between the concentration of individual fatty acids in
Table 5 Composition of the principal fatty acids (% of total fatty acids) of the polar lipid fraction determined in tissues of surubim
fed diets supplemented with different lipid sources
a
Tissue Phospholipids
Lipid sources
Fatty acid Initial
b
100% FO 100% Lard 100% SO Blend
Muscle
c
16:0 18.4 21.2 ±2.8 a 19.1 ±0.9 a 18.7 ±1.0 a 18.9 ±1.3 a
18:0 9.8 8.7 ±1.6 a 9.7 ±1.3 a 8.1 ±1.1 a 8.8 ±1.3 a
18:1n -9 16.8 6.3 ±1.1 b 16.9 ±3.3 a 15.3 ±2.1 a,b 12.8 ±3.1 a,b
18:2n -6 12.4 3.7 ±0.6 c 8.9 ±4.2 b 17.5 ±4.0 a 8.5 ±1.1 b
18:3n -3 0.4 0.4 ±0.3 a 0.4 ±0.2 a 0.6 ±0.3 a 0.4 ±0.1 a
20:4n -6 3.1 2.5 ±0.5 a 2.3 ±0.3 a,b 1.8 ±0.3 b 2.1 ±0.3 a,b
20:5n -3 1.7 6.5 ±0.5 a 2.6 ±0.2 c 1.9 ±0.2 c 4.4 ±0.9 b
22:6n -3 8.3 23.2 ±1.5 a 18.8 ±1.7 b 18.7 ±2.2 b 20.8 ±1.5 a,b
Liver
d
16:0 18.6 26.3 ±1.8 a 23.7 ±0.6 a 24.4 ±2.3 a 23.6 ±2.8 a
18:0 14.0 16.0 ±1.6 a 18.1 ±0.7 a 16.1 ±2.1 a 18.0 ±2.2 a
18:1n -9 12.2 6.2 ±0.6 b 16.4 ±2.3 a 12.3 ±2.9 a,b 11.3 ±1.8 a,b
18:2n -6 9.6 2.9 ±0.3 c 8.0 ±1.2 b 15.0 ±1.9 a 7.6 ±0.7 b
18:3n -3 0.4 0.0 ±0.1 a 0.1 ±0.1 a 0.2 ±0.2 a 0.2 ±0.2 a
20:4n -6 4.3 3.1 ±0.3 a 3.0 ±1.0 a 2.8 ±0.4 a 2.8 ±0.3 a
20:5n -3 1.3 8.3 ±2.2 a 3.3 ±0.5 b,c 2.2 ±0.2 c 5.1 ±0.7 b
22:6n -3 4.2 14.4 ±1.6 a 12.9 ±2.0 a 11.6 ±1.0 a 13.8 ±1.4 a
Brain
d
16:0 18.4 25.8 ±2.6 a 26.9 ±2.5 a 28.0 ±1.5 a 26.6 ±2.3 a
18:0 9.8 11.3 ±1.2 a 11.8 ±0.8 a 11.3 ±0.7 a 11.7 ±0.5 a
18:1n -9 16.6 23.7 ±2.8 b 28.9 ±2.0 a 24.8 ±0.8 b 26.9 ±2.1 a,b
18:2n -6 12.4 0.9 ±0.1 c 1.3 ±0.1 b,c 4.2 ±0.6 a 1.5 ±0.2 b
18:3n -3 0.3 0.3 ±0.1 a,b 0.1 ±0.1 b,c 0.3 ±0.2 a 0.1 ±0.1 c
20:4n -6 3.1 1.6 ±0.3 a 1.7 ±0.2 a 1.5 ±0.2 a 1.6 ±0.1 a
20:5n -3 1.7 0.9 ±0.2 a 0.4 ±0.1 b 0.2 ±0.0 c 0.5 ±0.1 b
22:6n -3 8.3 13.6 ±1.0 a 12.5 ±1.0 a,b 11.7 ±1.8 b 13.1 ±1.0 a,b
Eyes
d
16:0 19.0 17.9 ±1.7 a 17.7 ±1.1 a 16.5 ±0.8 a 17.8 ±2.2 a
18:0 19.0 13.5 ±0.4 a 13.0 ±1.1 a,b 11.8 ±0.0 b 12.4 ±0.6 a,b
18:1n -9 25.7 12.2 ±0.8 b 17.4 ±3.5 a 11.5 ±2.2 b 15.1 ±1.4 a,b
18:2n -6 4.9 2.3 ±0.5 c 5.3 ±1.0 b 8.1 ±2.1 a 3.8 ±0.3 b,c
18:3n -3 – 0.0 ±0.0 b 0.0 ±0.0 b 0.4 ±0.2 a 0.0 ±0.1 b
20:4n -6 2.5 2.4 ±0.1 a,b 2.7 ±0.2 a 2.0 ±0.1 b 2.2 ±0.4 a,b
20:5n -3 – 3.7 ±0.6 a 2.1 ±0.3 b,c 1.4 ±0.1 c 2.6 ±0.8 b
22:6n -3 15.7 33.2 ±9.4 a 29.2 ±4.4 a 29.5 ±0.8 a 32.4 ±6.7 a
Note: Means followed by the same letter in a row are not significantly different at P[0.05
a
Tissues sampled at the end of the experimental period (183 days)
b
Fish at the beginning of the experiment fed on commercial diet
c
Means (n=6)
d
Means (n=4)
Fish Physiol Biochem (2009) 35:399–412 405
123

dietary lipid and tissue lipid for fish fed the diets
containing 100% FO, 100% L, 100% SO and the
equally blend lipid sources diet.
The 16:0, 18:0, 18:1n -9, 18:2n -6, 18:3n -3,
20:5n -3, and 22:6n -3 fatty acids were the most
prevalent fatty acids in surubim tissues. The TAG
Table 6 Composition of the principal fatty acids (% of total fatty acids) of the neutral lipid fraction determined in tissues of surubim
fed diets supplemented with different lipid sources
a
Tissue Triacylglycerol
Lipid sources
Fatty acid Initial
b
100% FO 100% Lard 100% SO Blend
Muscle
c
16:0 13.7 19.3 ±5.1 a 17.8 ±1.5 a 14.9 ±1.3 a 16.4 ±1.6 a
18:0 8.0 4.6 ±0.7 b 8.8 ±1.5 a 4.4 ±1.0 b 6.0 ±1.0 b
18:1n -9 22.8 15.5 ±2.2 c 37.1 ±2.8 a 22.3 ±1.6 b 24.5 ±1.3 b
18:2n -6 23.3 6.4 ±1.1 d 14.7 ±1.8 c 35.7 ±1.1 a 20.2 ±2.0 b
18:3n -3 1.2 0.7 ±0.4 b 0.6 ±0.1 b 3.0 ±1.7 a 1.8 ±0.4 a,b
20:4n -6 3.6 1.9 ±0.9 a 0.8 ±0.3 b 0.8 ±0.2 b 1.1 ±0.5 a,b
20:5n -3 1.6 12.9 ±4.6 a 2.1 ±0.3 b 2.0 ±0.2 b 5.3 ±0.3 b
22:6n -3 6.1 14.0 ±3.6 a 6.4 ±1.3 b 6.7 ±1.1 b 9.4 ±2.0 b
Liver
d
16:0 14.4 23.0 ±2.2 a 22.0 ±1.2 a,b 15.8 ±0.2 c 19.2 ±1.0 b
18:0 12.3 7.6 ±0.4 b 9.7 ±0.4 a 5.9 ±0.4 c 7.7 ±0.4 b
18:1n -9 16.9 10.7 ±2.5 d 30.6 ±1.0 a 21.7 ±0.9 b 18.2 ±1.9 c
18:2n -6 14.1 5.4 ±0.3 c 11.6 ±2.0 b 29.3 ±1.4 a 14.3 ±1.6 b
18:3n -3 0.3 0.5 ±0.2 a,b 0.2 ±0.1 b 1.1 ±0.5 a 0.5 ±0.3 a,b
20:4n -6 7.6 2.9 ±1.0 a 2.6 ±0.4 a 1.8 ±0.5 a 2.6 ±0.8 a
20:5n -3 1.0 10.1 ±1.2 a 3.1 ±0.4 c 1.8 ±0.3 c 5.6 ±0.8 b
22:6n -3 5.8 12.8 ±2.3 a 8.4 ±0.7 b,c 7.5 ±1.1 c 11.1 ±1.9 a,b
Brain
d
16:0 19.6 13.8 ±2.2 a 16.7 ±3.3 a 14.6 ±3.4 a 14.8 ±2.0 a
18:0 11.4 15.0 ±1.7 a 13.1 ±1.3 a 13.0 ±1.5 a 13.9 ±2.2 a
18:1n -9 20.3 20.1 ±2.9 b 29.7 ±3.7 a 22.6 ±3.5 b 23.5 ±3.1 b
18:2n -6 1.6 2.8 ±1.1 c 7.7 ±2.1 b 10.2 ±1.0 a 6.5 ±1.4 b
18:3n -3 – 0.2 ±0.2 b 0.3 ±0.2 b 0.9 ±0.1 a 0.3 ±0.3 b
20:4n -6 1.4 3.7 ±1.0 a 2.6 ±0.6 b 3.6 ±0.2 a,b 3.1 ±0.6 a,b
20:5n -3 0.9 3.4 ±0.7 a 1.2 ±0.2 b,c 0.6 ±0.1 c 1.7 ±0.7 b
22:6n -3 11.5 19.9 ±1.2 a 12.1 ±2.2 c 16.3 ±0.7 b 16.3 ±2.3 b
Eyes
d
16:0 11.3 15.5 ±1.7 a 16.4 ±2.0 a 11.7 ±0.2 b 14.3 ±1.8 a
18:0 9.2 7.7 ±0.6 b 9.9 ±1.1 a 7.3 ±0.3 b 8.1 ±0.6 b
18:1n -9 12.8 12.5 ±1.1 c 24.0 ±5.3 a 16.0 ±0.2 b,c 17.7 ±2.2 b
18:2n -6 5.6 5.4 ±0.5 d 8.1 ±1.3 c 18.9 ±0.6 a 11.6 ±2.1 b
18:3n -3 – 0.4 ±0.2 a,b 0.1 ±0.2 b 0.8 ±0.6 a 0.7 ±0.3 a,b
20:4n -6 1.8 1.9 ±0.6 a 1.5 ±0.3 a 1.3 ±0.1 a 1.6 ±0.5 a
20:5n -3 2.0 6.7 ±0.5 a 1.5 ±0.3 c 1.5 ±0.0 c 4.1 ±0.3 b
22:6n -3 7.1 31.3 ±5.7 a 20.0 ±4.2 b 23.1 ±1.3 b 28.9 ±7.2 a
Note: Means followed by the same letter are not significantly different at P[0.05
a
Tissues sampled at the end of the experimental period (183 days)
b
Fish at the beginning of the experiment fed on commercial diet
c
Means (n=6)
d
Means (n=4)
406 Fish Physiol Biochem (2009) 35:399–412
123

fatty acids profiles of the liver and muscle were very
similar to the content of dietary fatty acids. In
contrast, the phospholipids profile showed enriched
levels of 20:4n -6 and 22:6n -3, accompanied by
proportionally decreased levels of 18:2n -6 and
18:3n -3. The TAG fatty acids profile in the brain
showed that high quantities of 18:2n -6 had been
lengthened to 20:4n -6, but in the phospholipids, it
had been oxidized, probably due to energy demand.
The TAG and phospholipids profiles of the brain and
eyes showed that 18:3n -3 had been catabolized to
20:5n -3 and 22:6n -3, mainly in the eye tissue.
The concentration of 18:1n -9 was elevated in
TAG and phospholipids in all tissues, but mainly in
TAG in the brain, which showed the highest values
for both lipid classes. Since both 16:0 and 18:0 were
present in the diets at similar quantities, these fatty
acids appear to be the preferred substrate for catab-
olism relative to 18:1n -9.
When dietary 18:2n -6 and 18:3n -3 were
present in similar quantities, relative to dietary 16:0
and 18:0 content, these PUFA were preferably
metabolized in surubim tissues rather than converted
to 20:4n -6 and 22:6n -3, respectively.
Following consumption of the experimental diets
for 183 days, the 18:2n -6 TAG concentration in
muscle showed a slight increase compared to the
concentration in the diet, except in fish fed 100% SO
(diet 3); however, this treatment showed the highest
value between groups. Alternatively, this fatty acid
content in all tissue phospholipids showed smaller
values than the respective dietary content.
The concentration of 18:3n -3 in TAG was
significantly different for each treatment, with the
highest values determined mainly the muscle and
liver samples of fish fed 100% SO. However, this
fatty acid did not differ in tissue phospholipids
between treatments. The content of 18:3n -3in
tissues was smaller than that in the diet, indepen-
dently of the lipid class.
Discussion
In a previous study in which FO was replaced by
alternative lipid sources, Martino et al. (2003)
reported that the replacement of FO occurred without
any detrimental effects on surubim growth and
health. However, this 2003 study lasted for only a
short time, with fish growing to a final weight of
approximately 30 g. In our study, surubim juveniles,
with an initial weight of approximately 30 g, were
grown for a longer time to a final weight of
approximately 230 g using diets in which the added
oil component contained between 33 and 100% L
and/or SO. Neither L and SO contain any n -3
HUFA; only SO contains substantial levels of
18:3n -3, the element proven to have beneficial
effects on human health (Connor 2000).
Some fatty acids, such as 16:0, 18:1n -9,
20:1n -9, and 22:1n -11, are always utilized by
energy metabolism. These fatty acids are heavily
catabolized in fish during the growth of farmed fish
and specifically during roe formation by female fish
(Henderson and Almatar 1989).
Martino et al. (2002b,2003) reported that the fatty
acid compositions of surubim tissue lipids were
closely related to that of dietary fatty acid composi-
tions, especially in the flesh in which TAG are the
predominant lipid class. This relationship is clearly
demonstrated in our study in which linear correlations
were observed between flesh fatty acid concentrations
and their concentrations in dietary lipid.
Bell et al. (2001,2002) observed different slopes
and correlation coefficients among their studies,
suggesting that selective retention or the metabolism
of individual fatty acids may vary in accordance with
the blend of dietary fatty acids and the size and age of
the fish. The correlation coefficients in our study are
particularly high in flesh TAG lipids, with rvalues of
between 0.96 and 0.99 (Table 7), which may reflect
the longer trial period and the larger size of the fish at
sampling.
The linear correlations observed in the Table 7can
be used to predict the flesh fatty acid concentration in
surubim fed different blends of FO, L, and SO during
various growth stages in future studies. The data
observed here reveal how different dietary fatty acids
are selectively retained or metabolized in tissue lipids
depending on their relative concentration in each oil.
In all dietary treatments, the 22:6n -3 was selec-
tively deposited in tissue lipid regardless of its
concentration in the diet, mainly in neural tissues,
where the greatest accumulation occurs relative to
flesh and liver tissues. This preferential retention of
22:6n -3 is clearly demonstrated in Table 7. The
differences (Dvalues) between diet and tissue
22:6n -3 concentration give values of 3.0–7.7
Fish Physiol Biochem (2009) 35:399–412 407
123

Table 7 Correlation coefficients of dietary fatty acid concentrations versus tissue phospholipids fatty acid concentrations
including the difference (D) between diet and tissue fatty acid values for surubim fed 100% FO, 100% L, 100% SO, and a
blend of lipid sources
Tissue Fatty acid Correlation
coefficient (r)
Pvalue
a
Accumulation index (D)
b
100% FO
c
100% Lard
c
100% SO
c
Blend
c
Phospholipids (PL)
Muscle 16:0 0.23 0.7717 0.8 -6.1 4.2 -0.2
18:0 0.95 0.0538 5.3 0.5 4.9 3.2
18:1n -9 0.90 0.1016 -5.5 -15.0 -5.4 -9.4
18:2n -6 0.97 0.0316 -2.4 -3.4 -22.2 -11.3
18:3n -3 0.95 0.0519 -0.6 -0.4 -4.3 -1.6
20:4n -6 0.80 0.1973 1.5 1.8 1.4 1.4
20:5n -3 0.98 0.0183 -3.2 0.7 -0.1 -0.6
22:6n -3 0.99 0.0061 16.9 15.3 15.4 15.7
Liver 16:0 -0.11 0.8908 5.9 -1.4 10.0 4.5
18:0 0.87 0.1300 12.6 9.0 12.9 12.4
18:1n -9 0.98 0.0207 -5.6 -15.6 -8.4 -10.9
18:2n -6 0.96 0.0414 -3.3 -4.3 -24.8 -12.2
18:3n -3 0.61 0.3899 -1.0 -0.7 -4.7 -1.7
20:4n -6 0.72 0.2821 2.1 2.5 2.4 2.2
20:5n -3 0.98 0.0164 -1.5 1.4 0.2 0.0
22:6n -3 0.90 0.0977 8.1 9.5 8.2 8.7
Brain 16:0 -0.55 0.4531 5.4 1.7 13.6 7.6
18:0 0.93 0.0728 7.9 2.7 8.1 6.1
18:1n -9 0.95 0.0496 11.9 -3.1 4.1 4.6
18:2n -6 0.97 0.0271 -5.2 -11.1 -35.5 -18.3
18:3n -3 0.53 0.4724 -0.7 -0.7 -4.6 -1.9
20:4n -6 0.21 0.7928 0.7 1.2 1.2 0.9
20:5n -3 0.94 0.0558 -8.9 -1.5 -1.8 -4.6
22:6n -3 0.93 0.0704 7.4 9.0 8.3 8.1
Eyes 16:0 0.76 0.2447 -2.5 -7.4 2.1 -1.3
18:0 0.27 0.7290 10.1 3.9 8.6 6.8
18:1n -9 0.83 0.1735 0.4 -14.6 -9.2 -7.1
18:2n -6 0.89 0.1149 -3.8 -7.0 -31.6 -16.0
18:3n -3 0.98 0.0216 -1.0 -0.8 -4.5 -1.9
20:4n -6 0.23 0.7730 1.5 2.2 1.7 1.5
20:5n -3 0.95 0.0467 -6.0 0.2 -0.6 -2.5
22:6n -3 0.97 0.0251 26.9 25.7 26.1 27.4
Triacylglycerols (TAG)
Muscle 16:0 0.70 0.3011 -1.1 -7.3 0.5 -2.7
18:0 1.00 0.0020 1.2 -0.3 1.2 0.3
18:1n -9 0.99 0.0119 3.6 5.1 1.6 2.3
18:2n -6 0.99 0.0061 0.3 2.4 -4.0 0.4
18:3n -3 0.98 0.0216 -0.2 -0.1 -1.9 -0.2
20:4n -6 0.96 0.0384 1.0 0.3 0.4 0.4
20:5n -3 0.99 0.0059 3.2 0.2 0.0 0.2
22:6n -3 0.98 0.0221 7.7 3.0 3.4 4.3
408 Fish Physiol Biochem (2009) 35:399–412
123

(muscle), 4.2–6.5 (liver), 8.6–13.7 (brain) and 16.6–
25.0 (eyes) for TAG lipids, and values of 15.3–16.9
(muscle), 8.1–9.5 (liver), 7.4–9.0 (brain), and 25.7–
27.4 (eyes) for phospholipids.
Conversely, all other fatty acids described in
Table 7, except for 20:4n -6, were progressively
favored for metabolism, presumably largely being
oxidized for energy, and both 18:3n -3 and
18:2n -6 were substrates for D6-desaturation and
elongation. Bell et al. (1997) described which appar-
ent differences among the mobilization of flesh
18:3n -3, 18:2n -6, and 18:1n -9 may be due
in part to the favoring of the former for conversion to
HUFA as well as for oxidization. The activity of the
desaturation and elongation pathways is significantly
increased following the inclusion of vegetable oils in
the diet (Tocher 2003).
The levels of 20:5n -3 and 22:6n -3 within the
muscle of fish changed in a direct correlation with
dietary levels. Since these fatty acids are considered
to be important for human nutrition, their concentra-
tions within the final aquaculture product should be
maintained at satisfactory levels through dietary
manipulation.
Previous studies demonstrated the presence of
22:6n -3 in the brain and eye tissues of marine fish
during development, but any significant biosynthesis
of 22:6n -3 from dietary 18:3n -3 was not
observed in either liver or neural tissues. Mourente
and Tocher (1998) and Mourente (2003) studied the
Table 7 continued
Tissue Fatty acid Correlation
coefficient (r)
Pvalue
a
Accumulation index (D)
b
100% FO
c
100% Lard
c
100% SO
c
Blend
c
Liver 16:0 0.83 0.1667 2.6 -3.1 1.4 0.2
18:0 0.90 0.0959 4.2 0.6 2.7 2.1
18:1n -9 0.97 0.0312 -1.1 -1.4 1.0 -4.0
18:2n -6 0.99 0.0054 -0.7 -0.8 -10.4 -5.5
18:3n -3 0.96 0.0381 -0.5 -0.5 -3.8 -1.4
20:4n -6 0.84 0.1642 1.9 2.1 1.4 1.9
20:5n -3 0.99 0.0119 0.4 1.2 -0.2 0.6
22:6n -3 0.99 0.0094 6.5 5.0 4.2 6.1
Brain 16:0 0.66 0.3379 -6.6 -8.5 0.2 -4.3
18:0 -0.42 0.5754 11.6 3.9 9.8 8.2
18:1n -9 0.97 0.0281 8.3 -2.3 1.9 1.3
18:2n -6 0.86 0.1431 -3.3 -4.7 -29.6 -13.4
18:3n -3 0.96 0.0371 -0.8 -0.4 -4.0 -1.7
20:4n -6 0.36 0.6443 2.7 2.1 3.2 2.5
20:5n -3 0.97 0.0305 -6.3 -0.7 -1.4 -3.4
22:6n -3 0.81 0.1886 13.7 8.6 12.9 11.3
Eyes 16:0 0.96 0.0383 -4.9 -8.7 -2.7 -4.7
18:0 0.98 0.0172 4.3 0.8 4.1 2.5
18:1n -9 0.99 0.0124 0.7 -8.0 -4.7 -4.6
18:2n -6 1.00 0.0014 -0.7 -4.2 -20.9 -8.3
18:3n -3 0.84 0.1609 -0.6 -0.7 -4.1 -1.3
20:4n -6 0.99 0.0127 0.9 1.0 1.0 0.9
20:5n -3 0.99 0.0057 -3.1 -0.4 -0.5 -1.0
22:6n -3 0.95 0.0531 25.0 16.6 19.8 23.8
a
P\0.05 are significant
b
Difference between concentrations are g fatty acid/100 g total fatty acids in diet vs. tissues
c
Negative Dvalues indicate lower values in tissue compared with diet, whereas positive values indicate accumulation in tissue
relative to diet
Fish Physiol Biochem (2009) 35:399–412 409
123

accumulation of 22:6n -3 in the brain and eyes
during fish development and did not observe any
significant biosynthesis of 22:6n -3 from dietary
18:3n -3, either in the liver or neural tissues. They
reported that it was not clear whether liver or neural
tissues themselves were of greater importance in the
biosynthesis of 22:6n -3 from dietary 18:3n -3. In
a study with sea bass, Navarro et al (1997) observed
that the different dietary 22:6n -3 levels markedly
reflected the composition of the lipid classes in fish
eyes.
In the tissue of fish eyes, 22:6n -3 is concentrated
as di-22:6n -3, mainly in the retina phospholipids
(Bell and Dick 1993). Brodtkorb et al. (1997), in a
study on Atlantic salmon, reported that the diet-
dependent levels of 22:5n -3 found in the eyes in
the final sampling is probably due to the chain
elongation of 20:5n -3, as in the brain. Our results
showed that 22:6n -3 enrichment in the phospholip-
ids and TAG of the eyes were similar between groups,
independent of the dietary lipid source. Dietary
22:6n -3 content presented values varying between
3.3 and 6.3% among treatments. This difference was
due to the higher levels of fish meal in the diets, which
was around 55%. Based on our results, we suggest a
selective uptake of 22:6n -3 fatty acid, which has
already been demonstrated by (Brodtkorb et al. 1997)
for Atlantic salmon. These authors reported that high
levels of 22:6n -3 in the eye are necessar yto maintain
good vision, which is particularly important for
hunting and capturing live prey.
Because 22:6n -3 is an essential fatty acid, it
must be either provided to the brain preformed or
synthesized in the brain from other n -3 PUFA.
Small amounts of 22:6n -3 and several of its n -3
PUFA precursors are normally present in the plasma
(Lands et al. 1992; Conquer and Holub 1998), and
there is evidence that the brain can utilize all of these
substrates. A number of studies of experimental
animals have shown that plasma 22:6n -3, either
obtained directly from the diet or synthesized in the
liver from n -3 PUFA precursors, is the main source
of 22:6n -3 for the brain (Sinclair 1975; Nouvelot
et al. 1986; Scott and Bazan 1989; Sheaff Greiner
et al. 1996; Pawlosky et al. 1996; Su et al. 1999).
This is consistent with the observation that fatty acid
D6-desaturase, the rate-limiting enzyme in the con-
version of n -3 PUFA precursors to 22:6n -3in
vertebrates (Hastings et al. 2001), decreases in the
brain soon after birth (Bourre et al. 1990). However,
other studies have shown that the brain can synthesize
22:6n -3 from 18:3n -3 and 22:5n -3 (Dhopesh-
warkar and Subramanian 1976; Cook 1978; Pawlosky
et al. 1994). The conversion of 18:3n -3to
22:6n -3 increases during essential fatty acid defi-
ciency (Dwyer and Bernsohn 1979), suggesting that
the amount of 22:6n -3 synthesis from n -3 PUFA
precursors in the brain may be regulated by the
availability of 22:6n -3 or other PUFA in the brain
tissue or cerebral circulation.
For the health-conscious consumer, the findings of
our study suggest that cultured surubim is a healthy
food option, with high levels of 20:5n -3 and
22:6n -3 in the flesh tissue. The substitution of FO
with L and/or SO higher than 66% of the total added
lipid led to a moderate reduction of n -3 HUFA
compared to fish fed 100% FO. These data are similar
to those reported in a previous study using smaller
surubim in which diets containing similar lipid levels,
but variations in the lipid source, were fed for 9 weeks
(Martino et al. 2003). This study showed that the
presence of the n -3 PUFA precursor in the diet
increased 20:5n -3 and 22:6n -3 content in fish
tissue, similar to how n -6 PUFA, increases 20:4n
-6 levels in fish tissues (Martino et al. 2002b).
In conclusion, it was not the objective of this study
to establish the synthesis rate of 20:5n -3 and
22:6n -3 in the tissues from EFA, but these findings
do correspond to the in vivo activity of the elonga-
tion/desaturation pathway, as suggested by Tocher
(2003).
Acknowledgments The authors are grateful to ‘‘Projeto
Pacu’’, Brazil for providing the experimental fish, to Nutron
Alimentos and Mogiana Alimentos, Brazil for providing some
of the diet ingredients, and to Dr. Juliette Delabbio of
Aquaculture Research Station of Northwestern State
University, Louisiana, USA for the English review of this
manuscript. The authors are also grateful to the Rio de Janeiro
State Research Support Foundation (FAPERJ) (Grant: E/26-
170.987-03) as a partial sponsor of this work and to
Coordenac¸a
˜o de Aperfeic¸oamento de Pessoal de Nı
´vel
Superior (CAPES) for the doctorate scholarship of Mr. M. D.
Noffs.
References
AOAC (Association of Official Analytical Chemists) (2000).
Official methods of analysis,17th edn. Association of
Official Analytical Chemists, Gaithersburg, MD
410 Fish Physiol Biochem (2009) 35:399–412
123

Barlow S (2002) Resources and markets: The world market
overview of fish meal and fish oil. Paper presented to the
2nd Seafood By-products Conf. Anchorage, Alaska
Bell MV, Dick JR (1993) The appearance of rods in the eyes of
herring and increased didocosahexaenoyl molecular spe-
cies of phospholipids. J Mar Biol Assoc UK 73:679–688
Bell JG, Tocher DR, Farndale BM, Cox DI, McKinney RW,
Sargent JR (1997) The effect of dietary lipid on polyun-
saturated fatty acid metabolism in Atlantic salmon
(Salmo salar) undergoing parr-smolt transformation. Lip-
ids 32:515–525. doi:10.1007/s11745-997-0066-4
Bell JG, McEvoy J, Tocher DR, McGhee F, Campbell PJ,
Sargent JR (2001) Replacement of fish oil with rapeseed
oil in diets of Atlantic salmon (Salmo salar) affects tissue
lipid composition and hepatocyte fatty acid metabolism.
J Nutr 131(5):1535–1543
Bell JG, Henderson RJ, Tocher DR, McGhee F, Dick JR, Porter
A et al (2002) Substituting fish oil with crude palm oil in
the diet of Atlantic salmon (Salmo salar) affects tissue
fatty acid compositions and hepatic fatty acid metabolism.
J Nutr 132(2):222–230
Bourre JM, Piciotti M, Dumont O (1990) 6-Desaturase in brain
and liver during development and aging. Lipids 25:354–
356. doi:10.1007/BF02544347
Brodtkorb T, Rosenlund G, Lie Ø (1997) Effects of dietary
levels of 20:5n – 3 and 22:6n – 3 on tissue lipid compo-
sition in juvenile Atlantic salmon, Salmo salar, with
emphasis on brain and eye. Aquacult Nutr 3:175–187. doi:
10.1046/j.1365-2095.1997.00085.x
Caballero MJ, Lo
´pez-Calero G, Socorro J, Ro
ˆo FJ, Izquierdo
MS, Fe
´rnandez AJ (1999) Combined effect of lipid level
and fish meal quality on liver histology of gilthead sea-
bream (Sparus aurata). Aquaculture 179:277–290. doi:
10.1016/S0044-8486(99)00165-9
Caballero MJ, Obach A, Rosenlund G, Montero D, Gisvold M,
Izquierdo MS (2002) Impact of different dietary lipid
sources on growth, lipid digestibility, tissue fatty acid
composition and histology of rainbow trout, Oncorhyn-
chus mykiss. Aquaculture 214:253–271. doi:10.1016/
S0044-8486(01)00852-3
Campos P, Martino RC, Trugo LC (2006) Amino acid composition
of Brazilian surubim fish (Pseudoplatystoma coruscans)fed
diets with different levels and sources of fat. Food Chem
96:126–130. doi:10.1016/j.foodchem.2005.02.017
Connor WE (2000) Importance of n -3 fatty acids in health
and disease. Am J Clin Nutr 71(1):171S–175S
Conquer JA, Holub BJ (1998) Effect of supplementation with
different doses of DHA on the levels of circulating DHA
as non-esterified fatty acid in subjects of Asian Indian
background. J Lipid Res 39:286–292
Cook HW (1978) In vitro formationof polyunsaturatedfatty acids
by desaturation in the rat brain: some properties of enzymes
in developingbrain and comparison with liver. J Neurochem
30:1327–1334. doi:10.1111/j.1471-4159.1978.tb10463.x
Dhopeshwarkar GA, Subramanian C (1976) Biosynthesis of
polyunsaturated fatty acids in the developing brain. I.
Metabolic transformations of intracranially administered
1-14C linolenic acid. Lipids 11:67–71. doi:10.1007/BF0
2532586
Dwyer B, Bernsohn J (1979) Incorporation of [1–14C] linol-
enate into lipids of developing rat brain during essential
fatty acid deficiency. J Neurochem 32:833–838. doi:
10.1111/j.1471-4159.1979.tb04567.x
Folch JM, Lees M, Sloane-Stanley GH (1957) A simple
method for the isolation and purification of total lipids
from animal tissues. J Biol Chem 226:497–507
Greiner RS, Catalan JN, Moriguchi T, Salem N Jr (2003)
Docosapentaenoic acid does not completely replace DHA
in n -3 FA-deficient rats during early development.
Lipids 38:431–435. doi:10.1007/s11745-003-1080-2
Hastings N, Agaba M, Tocher DR, Leaver MJ, Dick JR, Sar-
gent JR et al (2001) A vertebrate fatty acid desaturase
with delta 5 and delta 6 activities. Proc Natl Acad Sci
USA 98:14304–14309. doi:10.1073/pnas.251516598
Henderson RJ, Almatar S (1989) Seasonal changes in the lipid
composition of herring Clupea harengus in relation to
gonad maturation. J Mar Biol Assoc UK 69:323–334
Juaneda P, Rocquelin G (1985) Rapid and convenient separa-
tion of phospholipids and non phosphorus lipids from rat
heart using silica cartridges. Lipids 20:40–41. doi:
10.1007/BF02534360
Lands WEM, Libelt B, Morris A, Kramer NC, Prewitt TE,
Bowen P et al (1992) Maintenance of lower proportions of
(n -6) eicosanoid precursors in phospholipids of human
plasma in response to added dietary (n -3) fatty acids.
Biochim Biophys Acta 1180:147–162
Martino RC, Cyrino JEP, Portz L, Trugo LC (2002a) Effect of
dietary lipid level on nutritional performance of the sur-
ubim (Pseudoplatystoma coruscans). Aquaculture
209:209–218. doi:10.1016/S0044-8486(01)00738-4
Martino RC,Cyrino JEP, Portz L, TrugoLC (2002b) Performance
and fatty acid composition of surubim (Pseudoplatystom-
acoruscans) fed diets with animal and plant lipids.
Aquaculture 209:233–246. doi:10.1016/S0044-8486(01)00
847-X
Martino RC, Trugo LC, Cyrino JEP, Portz L (2003) Use of
white fat as a replacement for squid liver oil in practical
diets for surubim (Pseudoplatystoma coruscans). J World
Aquacult Soc 34:192–202. doi:10.1111/j.1749-7345.2003.
tb00056.x
Martino RC, Cyrino JEP, Portz L, Trugo LC (2005) Perfor-
mance, carcass, composition and nutrient utilization of
surubim (Pseudoplatystoma coruscans, Agassiz) fed diets
with varying carbohydrate and lipid levels. Aquacult Nutr
11:131–137. doi:10.1111/j.1365-2095.2004.00332.x
Metcalfe AP, Schmitz AA (1961) The rapid preparation of
fatty acids for gas chromatographic analysis. Anal Chem
33:363–364
Moore SA, Yoder E, Spector AA (1990) Role of the blood-brain
barrier in theformation of long-chain omega-3 and omega-6
fatty acids fromessential fatty acid precursors. J Neurochem
55:391–402. doi:10.1111/j.1471-4159.1990.tb04150.x
Moore SA, Yoder E, Murphy S, Dutton GR, Spector AA
(1991) Astrocytes, not neurons, produce docosahexaenoic
acid (22:6 omega-3) and arachidonic acid (20:4 omega-6).
J Neurochem 56:518–524. doi:10.1111/j.1471-4159.1991.
tb08180.x
Mourente G (2003) Accumulation of DHA (docosahexaenoic
acid; 22:6n -3) in larval and juvenile fish brain. The big
fish bang. In: Browman HI, Skiftesvik AB (eds) Proc 26th
Ann Larval Fish Conf 2003. Institute of Marine Research,
Bergen
Fish Physiol Biochem (2009) 35:399–412 411
123

Mourente G, Tocher DR (1998) The in vivo incorporation and
metabolism of [1–14C] linolenate (18:3n–3) in liver, brain
and eyes of juveniles of rainbow trout Oncorhyn-
chus mykiss L., and gilthead sea bream Sparus aurata L.
Fish Physiol Biochem 18:149–165. doi:10.1023/A:10077
17312480
Navarro JC, McEvoy LA, Bell MV, Amat F, Hontoria F,
Sargent JR (1997) Effect of different dietary levels of
docosahexaenoic acid (DHA, 22:6w–3) on the DHA
composition of lipid classes in sea bass larvae eyes.
Aquacult Int 5:509–516. doi:10.1023/A:1018301215592
Nouvelot A, Delbart C, Bourre JM (1986) Hepatic metabolism
of alpha-linolenic acid in suckling rats, and its possible
importance in polyunsaturated fatty acid uptake by the
brain. Ann Nutr Metab 30:316–323
NRC (National Research Council) (1993) Nutrient requirement
of warmwater fishes and shellfishes. National Academy
Press, Washington, DC, p 114
O
¨zogul Y, O
¨zyurt G, O
¨zogul F, Kuley E, Polat A (2005)
Freshness assessment of European eel (Anguilla anguilla)
by sensory, chemical and microbiological methods. Food
Chem 92:745–751. doi:10.1016/j.foodchem.2004.08.035
Pawlosky RJ, Barnes A, Salem N Jr (1994) Essential fatty acid
metabolism in the feline: relationship between liver and
brain production of long-chain polyunsaturated fatty
acids. J Lipid Res 35:2032–2040
Pawlosky RJ, Ward G, Salem N Jr (1996) Essential fatty acid
uptake and metabolism in the developing rodent brain.
Lipids 31:S-103–S-107
Rainuzzo JR, Reitan KI, Olsen Y (1997) The significance of
lipids at early stages of marine fish: a review. Aquaculture
155:103–115. doi:10.1016/S0044-8486(97)00121-X
Sargent JR, Bell G, McEvoy L, Tocher DR, Estevez A (1999)
Recent developments in the essential fatty acid nutrition
in fish. Aquaculture 177:191–199. doi:10.1016/S0044-
8486(99)00083-6
Scott BL, Bazan NG (1989) Membrane docosahexaenoate is
supplied to the developing brain and retina by the liver.
Proc Natl Acad Sci USA 86:2903–2907. doi:10.1073/
pnas.86.8.2903
Sheaff Greiner RC, Zhang Q, Goodman KJ, Giussani DA,
Nathanielsz PW, Brenna JT (1996) Linoleate, alpha-lin-
olenate, and docosahexaenoate recycling into saturated
and monounsaturated fatty acids is a major pathway in
pregnant or lactating adults and fetal or infant rhesus
monkeys. J Lipid Res 37:2675–2686
Sinclair AJ (1975) Incorporation of radioactive polyunsatu-
rated fatty acids into liver and brain of developing rats.
Lipids 10:175–184. doi:10.1007/BF02534156
Su HM, Bernardo L, Mirmiran M, Ma XH, Corso TN, Na-
thanielsz PW et al (1999) Bioequivalence of dietary-
linolenic and docosahexaenoic acids as sources of doco-
sahexaenoate accretion in brain and associated organs of
neonatal baboons. Pediatr Res 45:87–93. doi:10.1203/
00006450-199901000-00015
Tacon AGJ, Jackson AJ (1985) Utilization of conventional and
unconventional protein sources in practical fish feeds. In:
Cowey CB, Mackie A, Bell J (eds) Nutrition and feeding
in fish. Academic Press, London, pp 119–145
Tocher DR (1993) Elongation predominates over desaturation
in the metabolism of 18:3n -3 and 20:5n -3 in turbot
(Scophthalmus maximus) brain astroglial cells in primary
culture. Lipids 28:267–272. doi:10.1007/BF02536309
Tocher DR (2003) Metabolism and functions of lipids and fatty
acids in teleost fish. Rev Fish Sci 11:107–184. doi:10.1080/
713610925
Tocher DR, Sargent JR (1990) Incorporation into phospholipid
classes and metabolism via desaturation and elongation of
various 14C-labelled (n -3) and (n -6) polyunsaturated
fatty acids in trout astrocytes in primary culture. J Neu-
rochem 54:2118–2124. doi:10.1111/j.1471-4159.1990.
tb04918.x
Watanabe T (1982) Lipid nutrition in fish. Comp Biochem
Physiol B 73:3–15
Williard DE, Harmon SD, Kaduce TL, Preuss M, Moore SA,
Robbins ME et al (2001) Docosahexaenoic acid synthesis
from n -3 polyunsaturated fatty acids in differentiated
rat brain astrocytes. J Lipid Res 42:1368–1376
412 Fish Physiol Biochem (2009) 35:399–412
123