Aquaculture nutrition, tập 17, số 3, 2011

Aquaculture Nutrition
2011 17; 235–243

..........................................................................................

doi: 10.1111/j.1365-2095.2009.00734.x

DE

School of Marine and Tropical Biology, James Cook University, Townsville, Qld, Australia

Cultured barramundi, Lates calcarifer, suffer from abnormalities affecting the jaw, opercula and spine. The aim of this
study was to quantify for the first time the effects of supplemented dietary vitamin C, vitamin D3 and ultraviolet
(UV) light on the development of jaw, opercula and spinal
deformities. Four diets were formulated to contain (i) no
vitamin C or vitamin D3, (ii) only vitamin D3, (iii) only
vitamin C and (iv) both vitamin C and vitamin D3. In
addition, two commercial diets (diets 5 and 6) were also
tested. These diets were replicated in the presence, and in the
absence, of ultraviolet (UV) light as this may also affect
skeletal development. Diets formulated with 170 ± 1 mg
kg)1 and 195 ± 0.5 mg kg)1 of vitamin C (diets 3 and 4,
respectively) and the commercial diets (diets 5 and 6) had
significantly lower incidences of spinal deformities (0–2.5%;
P < 0.01) and opercula deformities (nil detected). Spinal
deformities were Ôbroken backÕ syndrome, which was found
only in the precaudal vertebrae, and lordosis which was only
in the caudal vertebrae. UV light and vitamin D3 did not
affect spinal or opercula deformities. There was no change in
the occurrence of jaw deformities in vitamin C, vitamin D3 or
UV light treatments.

KEY WORDS: deformities, Lates calcarifer, ultraviolet light,
vitamin C, vitamin D3

Received 5 March 2009, accepted 30 September 2009
Correspondence: Matthew Fraser, School of Marine and Tropical Biology,
James Cook University, Townsville, QLD, Australia, 4811. E-mail:

Morphological deformities pose a problem for many propagated fish species as they affect the survival and commercial
value of the product (Divanach et al. 1997). As such,

..............................................................................................

Ó 2010 Blackwell Publishing Ltd

investigating the development of skeletal deformities in cultured finfish has been the focus of recent studies (Fjelldal
et al. 2007; Sullivan et al. 2007; Kamler et al. 2008). Species
with a documented incidence of abnormal development
include the Red sea bream, Pagrus major (Temminck and
Schlegel) (Kihara et al. 2002; Hattori et al. 2003), European
sea bass, Dicentrarchus labrax (Linnaeus) (Paperna 1978;
Barahona-Fernandes 1982; Daoulas et al. 1991; Koumoundouros et al. 2001), Rainbow trout, Oncorhynchus mykiss
(Walbaum) (Dabrowski et al. 1990), Milkfish, Chanos chanos
(Forsskal) (Gapasin et al. 1998), Tilapia, Oreochromis mossambicus (Peters) (Soliman et al. 1986) and Barramundi,
Lates calcarifer (Block) (Fraser et al. 2004; Fraser & de Nys
2005). The three major regions of deformation are the jaw
(Roberts et al. 2001), opercula (Galeotti et al. 2000) and spine
(Koumoundouros et al. 2001) with environmental, nutrition
and genetic factors proposed to be causal mechanisms
(Andrades et al. 1996). Because deformities affect both bone
and cartilage tissue, investigating the requirements for their

correct development has the potential to determine and mitigate the underlying causes of morphological deformities.
Vitamin C is required in the formation of collagen, a
principle component of bone and cartilage (Horton et al.
1993). Vitamin C is an essential dietary nutrient (Soliman
et al. 1986; Dabrowski et al. 1990; Alexis et al. 1997; Li &
Robinson 2001), because teleosts are unable to produce,
L-gulonolactone, the enzyme required for vitamin C biosynthesis (Dabrowski et al. 1990). Many studies have demonstrated the requirement of dietary vitamin C for correct
skeletal development in teleosts including Coho salmon,
Oncorhynchus kisutch (Walbaum), Rainbow trout, O. mykiss
(Halver et al. 1969), Channel catfish, Ictalurus punctatus
(Rafinesque), (Wilson & Poe 1973), Spotted snakehead,
Channa punctatus (Bloch) (Mahajan & Agrawal 1979) and
Tilapia, O. mossambicus (Soliman et al. 1986).
Vitamin D3 is also suggested to play a role in teleost
skeletal deformation (Galeotti et al. 2000) with fish reportedly having a higher vitamin requirement than terrestrial


animals (NRC 1993). Vitamin D3 regulates calcium absorption and bone mineralization in vertebrates and requires
ultraviolet (UV) light exposure for its formation in most
terrestrial vertebrates (Webb 1993). However, the function of
vitamin D3 in teleost skeletal metabolism and the role of UV
radiation in regulating the synthesis of vitamin D3 from its
precursors are not clearly defined (Graff et al. 2002; Lall &
Lewis-McCrea 2007). Studies on vitamin D3 requirements in
cultured fish have investigated its effects of on growth and
mortality in salmonids (Hilton & Ferguson 1982), but none
have reported a relationship between vitamin D3 and the
development of deformities.
The barramundi, L. calcarifer is a developing species for
finfish aquaculture production in Australia and southeast

Asia (Tucker et al. 2002) and suffers from spinal, jaw and
opercula deformities during the larval and postlarval phases (Fraser et al. 2004; Fraser & de Nys 2005). Furthermore, juvenile Lates calcarifer fed diets devoid of vitamin
C develop skeletal deformities including lordosis, scoliosis
and Ôbroken backÕ syndrome (Phromkunthong et al. 1997).
The minimum requirement of vitamin C in juvenile
L. calcarifer to prevent pathological signs of deficiency is
30 mg kg)1 (as ascorbal phosphate) (Phromkunthong et al.
1997); however, the quantification and specific histological
identification of spinal deformities that result from a
dietary vitamin C deficiency has not been determined.
Similarly, the roles of vitamin D3 and UV light in the
development of morphological deformities in barramundi
have not been addressed. By documenting the nature and
frequency of deformities this study quantifies, for the first
time, the quantitative effects of vitamin C, vitamin D3 and
UV light on the development of morphological deformities
in juvenile L. calcarifer.

Four dietary treatments varying in vitamin C (ascorbic acid)
and vitamin D3 (cholecalciferol) levels (diets 1, 2, 3 and 4),
and two commercial diets (diet 5 and 6), were fed in the
presence of, and in the absence of, ultraviolet (UV) light
(Table 1). Each diet was replicated with five tanks per light
treatment.
Diets 1 to 4 were prepared from individual base ingredients
(Table 2). Each vitamin and mineral was individually
weighed and combined to form a premix used in diet preparation (Tables 3 & 4) with vitamin C and vitamin D3 individually added as required (Table 3). Vitamin C in the form

Table 1 Experimental design used for the manipulation of diet and
ultraviolet light exposure for juvenile Lates calcarifer

Treatment
Diet

UV light

No UV light

1
2
3
4
51
62

No Vit C or D3
Vit D3
Vit C
Vit C and D3
Commercial diet 1
Commercial diet 2

No Vit C or D3
Vit D3
Vit C
Vit C and D3
Commercial diet 1
Commercial diet 2

1
2

Skretting barramundi pellet (3 mm).
Ridley Aqua-feed barramundi pellet (3 mm).

Table 2 The base ingredients used in formulating diets 1 to 4

Ingredient

Quantity
in (g kg)1)

Fish meal
Cellulose
Casein
Fish oil
Gelatine
Starch1

469.8
217.9
175.6
52.2
34.1
16

1

Gelatinized by autoclaving at 110 °C for 20 min.
Ingredients obtained from: Fish meal, Fish oil; Ridleys Aqua-Feed
Company, Brisbane, Qld, Australia. Cellulose; Hahnflock Hahn and

Co., Germany. Casein; Dairy Farmers Association, Malanda, Qld,
Australia. Gelatine; Trumps Pty. Ltd., Brisbane, Qld, Australia.
Starch; Poppy cornflour, Australia.

of free ascorbic acid was used in this study because it has the
highest level of bioavailability (Li & Robinson 2001).
Because free ascorbic acid is readily oxidized during feed
manufacture, a supplement of 2000 mg kg)1 was added to
allow for losses (as quantified below). Previous work has
shown that vitamin C supplements of less than 2000 mg kg)1
are not toxic in L. calcarifer (Boonyaratpalin et al. 1989).
The minimum requirement of vitamin D3 for normal growth
in L. calcarifer has not been determined. Therefore, supplemental vitamin D3 levels were formulated according to
previous work with Atlantic salmon, Salmo salar. The level
of 100 000 IU of vitamin D3 was taken from Graff et al.
(2002) who found this level of vitamin D3 to be non-toxic in
S. salar.
All ingredients were combined in a Hobart mixer (Model
A120, Hobart Corporation, Troy, OH, USA) and mixed for
a minimum of 30 min with distilled water being added to the
mixture until the appropriate consistency was achieved for
pelletization. Each diet was extruded (Model A120) to form a
3-mm pellet for the first 4 weeks of the trial and a 4-mm
pellet for the remaining 5 weeks of culture. All diets were

..............................................................................................

Aquaculture Nutrition 17; 235–243 Ó 2010 Blackwell Publishing Ltd

Table 3 The vitamins combined to form a premix for diets 1 to 4

Vitamin

Quantity
(g kg)1 of diet)

Retinylacetate
Menadione
Alpha-tocopherol
Choline chloride
Myo-inositol
PABA
Thiamin
Riboflavin
Pyridoxine HLC
Pantothenate acid
Nicotine acid
Biotin
Cyanocobalamin
Folic acid
Ethoxyquin
Citric acid
Cellulose

0.012
0.033
0.375
1.2
0.255

0.105
0.018
0.021
195 · 10)4
555 · 10)4
0.075
4 · 10)4
6 · 10)5
45 · 10)4
1275 · 10)4
6
18.174

Vitamin C levels present after processing were 30 ± 0 mg kg)1 for
diet 1; 40 ± 1 mg kg)1 for diet 2; 170 ± 1 mg kg)1 for diet 3;
195 ± 0.5 mg kg)1 for diet 4; 50 ± 1 mg kg)1 for diet 5; 45 ±
0.5 mg kg)1 for diet 6. Vitamin D3 levels present after processing
were 11 200 ± 800 IU kg)1 for diet 1; 164 000 ± 32 000 IU kg)1 for
diet 2; 8600 ± 200 IU kg)1 for diet 3; 82 000 ± 6000 IU kg)1 for diet
4; 8200 ± 200 IU kg)1 for diet 5; 14 200 ± 600 IU kg)1 for diet 6.

Table 4 The minerals combined to form a premix for diets 1 to 4
Mineral

Quantity g kg)1 of diet

Aluminium chloride
Cobalt chloride
Copper sulphate
Magnesium sulphate

Manganese sulphate
Sodium selenate
Zinc sulphate
Cellulose

25 · 10)4
0.001
15 · 10)4
1.5
375 · 10)4
1 · 10)4
0.185
31625 · 10)4

oven dried (50 °C) overnight and broken into designated
sized pellets and stored ()12 °C) between feeding.

3

Two samples of all diets were analysed for vitamin C (free
ascorbate) and vitamin D3 (cholecalciferol) content using
HPLC (2-6-dichloroindopheno; titrimetric method) analysis
(performed by Dairy Technical Services Ltd., Melbourne,
Vic., Australia). Commercial diets (diets 5 & 6) also contain
supplemented vitamin C as a stabilized monophosphate salt
(ascorbal-2-monophosphate 30–40 mg kg)1) which is not
quantifiable using this method (Dairy Technical Services),
but is nutritionally available (Lin & Shiau 2004).

..............................................................................................

Aquaculture Nutrition 17; 235–243 Ó 2010 Blackwell Publishing Ltd

Juvenile L. calcarifer (25–30 mm total length) was supplied
from a commercial hatchery in Mourylian, Queensland and
housed at the Marine and Aquaculture Facilities Unit at
James Cook University. The fish were kept in a freshwater
recirculating system consisting of 60 tanks (70 L in capacity).
Eight fish were weighed (0.01 g) and allocated to a tank
randomly assigned a dietary treatment (n=5). Each tank
received gentle and constant aeration with 100% water
exchange every hour supplied through a biological and sand
filter. Temperature was maintained between 24–30 °C and
photoperiod at 13L:11D. The maximum levels for ammonia,
nitrite and nitrate attained during the trial were 1 · 10)4,
0.001 and 0.08 g L)1, respectively. Water from the recirculating system was exchanged by 25–70% daily as required to
maintain water quality.
Fish were acclimated for 50 days before salt bathing and
anaesthetizing (0.08 g L)1 benzocaine solution; Sigma E
1501). The weight of fish per tank was recorded (0.01 g), and
fish were visually assessed for jaw, opercula and spinal
deformities. Only fish without visually identifiable deformities were used. All fish were fed twice daily to satiety, and
each tank was cleaned daily by siphoning. Mortalities were
removed, weighed and stored at )12 °C.
The visible lighting in both the UV light and non-UV light
treatments was supplied using three cool white fluorescent
tubes (Philips Fluorotone, 36 W, ÔTLÕD 36W/33, wavelength
400–700 nm). Ultraviolet light was supplied in the UV light
treatment using six UV fluorescent tubes (NEC black light,
40 W, wavelength 300–425 nm). The non-UV light treatment

was separated from the UV light treatment using a nontransparent black screen.

The experiment ran for 9 weeks after which time all fish were
euthanized in ice slurry. Each fish was visually assessed for
spinal, jaw and opercula deformities and the total mass of
fish for each tank recorded (0.01 g). Growth was measured as
% mass gain and was determined by dividing the increase in
mass by the initial mass multiplied by 100 (for each tank).
Growth was measured only to indicate fish were fed appropriately and that normal growth was achieved. The fish were
immediately transported on ice to Townsville General Hospital for X-raying (Philips Optimus Digital Radiography
Unit, set at 50 kV and 4.00 mAs, processed on Agfa laser
film). All spinal deformities were recorded for each treatment
including the location of the vertebrae and number of


vertebrae affected. Vertebral numbers are referenced as one
being cranial to 24 being caudal.

Mean incidence of deformities (%)

The effects of UV light and supplimented vitamin C and
vitamin D3 on spinal deformity development were analysed
30
25
20
15
10
5
0
1

2

3

4

5

6

Diet

Mean incidence of deformities (%)

Figure 1 The incidence of spinal deformities (Ôbroken backÕ syndrome and lordosis) (mean ± SE pooled from UV+ and UV)
treatments) in juvenile Lates calcarifer fed formulated diets varying
in vitamin C and vitamin D content (diets 1 to 4) and two commercial diets (diets 5 and 6) over 9 weeks.

by a three-factor ANOVA. Diets 5 and 6 were not incorporated
in the analysis as they contained vitamin C as stabilized
monophosphate salt (see Analysis of vitamin C and D3 in
Methods section) and therefore could not be compared with
diets 1 to 4. However, they provide a useful control for
comparison with standard commercial diets. Log (ln + 1)
transformations of spinal deformity data minimized the heterogeneity of variance but not completely. As a result, a was
set at P = 0.01 (Quinn & Keough 2002). Opercula deformity
data were analysed only for the effects of UV light and dietary vitamin D with two-factor ANOVA because of the
occurrence of zeros in most diet treatments. Scatter plots and
histograms of residuals were used to examine the ANOVA

assumptions of homogeneity of variances and normality,
respectively (Quinn & Keough 2002). The effects of UV light,
vitamin C and vitamin D on growth were analysed by a threefactor ANOVA. Scatter plots and histograms of residuals were
used to examine the ANOVA assumptions of homogeneity of
variances and normality, respectively (Quinn & Keough
2002). Frequency distributions of Ôbroken backÕ syndrome
and lordosis along the vertebral column were compared using
a Kolmogorov-Smirnov test. All analyses were performed
using SPSS (Version11, SPSS Inc., Chicago, IL, USA).

50
40

3

30
20
10
0
1

2

3

4

5

6

Diet

Figure 2 The incidence of opercula deformities (mean ± SE pooled
from UV+ and UV) treatments) in juvenile Lates calcarifer fed
formulated diets varying in vitamin C and vitamin D content (diets 1
to 4) and two commercial diets (diets 5 and 6) over 9 weeks (n = 5).

Only non-supplemented vitamin C diets affected the incidence
of skeletal deformities in barramundi. Supplemented vitamin
C had a significant affect on the development of spinal (threefactor ANOVA, df = 1, P < 0.01) (Fig. 1) and opercula
deformities (Fig. 2) (Table 5). Diets supplemented with vitamin C (diets 3 and 4) had no or very low occurrence of spinal
deformity (diet 3, 0%; diet 4, 1.25 ± 1.25%) (Fig. 1; Table 5)
and there were no deformities of the opercula (Fig. 2;
Table 5). Similarly, both commercial diets (diets 5 and 6) had

Table 5 Three-factor Analysis of Variance for the effects of vitamin C, vitamin D and UV light on the development of spinal deformities and
one-factor Analysis of Variance for the effects of UV light on the development of opercula deformities in juvenile Lates calcarifer (P = 0.01)
F -value

P -value

Degrees of freedom

Mean square

Factor

Spinal

Opercula

Spinal

Opercula

Spinal

Opercula

Spinal

Opercula

UV light
Vitamin C
Vit D3
UV light · Vit C
UV light · Vit D3
Vit C · Vit D3
UV light · Vit C · Vit D3

1
1
1
1
1
1
1

1
–
–
–
–
–
–

1.939
3259.117
7.629
27.324
4.127
43.518
3.252

0.270
–
–
–
–
–
–

0.010
17.603
0.041
0.147
0.022
0.235

0.017

1.522
–
–
–
–
–
–

0.919
<0.001
0.840
0.703
0.882
0.631
0.895

0.223
–
0.714
–
0.781
–
–

..............................................................................................

Aquaculture Nutrition 17; 235–243 Ó 2010 Blackwell Publishing Ltd

a very low incidence of spinal deformities (diet 5 and 6,
2.5 ± 1.6% each) (Fig. 1) and there were no opercula
deformities (Fig. 2). In contrast, fish-fed diets not supplemented with vitamin C (diets 1 and 2) had a significantly
higher occurrence of spinal (diet 1, 20.63 ± 7.15%; diet 2,
17.59 ± 4.94%) (Fig. 1) and opercula deformities (diet 1,
30.15 ± 12.2%; diet 2, 36.24 ± 8.33%) (Fig. 2). The levels of
vitamin D3 did not affect the occurrence of spinal or opercula
deformation (two-factor ANOVA, df = 1, P > 0.05) (Table 5),
and levels of vitamin C and D3 did not affect the formation of
jaw deformities. Similarly, UV light did not affect the incidence of spinal (three-factor ANOVA, df = 1, P > 0.05) or
opercula deformities (two-factor ANOVA, df = 1, P > 0.05)
(Table 5). There was no interaction between UV light, Vitamin C and Vitamin D3 in the development of spinal deformities (three-factor ANOVA, df = 1, P > 0.01) (Table 5).

Table 7 Three-factor Analysis of Variance for the effects of vitamin
C, vitamin D and UV light on growth in juvenile Lates calcarifer
(P = 0.05)

Factor
UV light
Vitamin C
Vit D3
UV light · Vit C
UV light · Vit D3
Vit C · Vit D3
UV light · Vit
C · Vit D3

Degrees of
freedom

Mean
square

F-value

P-value

1
1
1
1
1
1
1

23 925
200 984
33
22 228
34 696
19 530
1359

2.10
17.64
0.002
1.95
3.04
1.71

0.1

0.157
<0.001
0.957
0.172
0.091
0.200
0.73

Supplemented dietary vitamin C significantly increased mean
growth (%) in juvenile L. calcarifer (diet 3 with UV
657.6 ± 39.7, diet 3 without UV 607.33 ± 15.6; diet 4 with
UV 658.9 ± 72.9, diet 4 without UV 511.20 ± 35.5)
(Table 6 & 7) in comparison with diets not supplemented
with dietary vitamin C (diet 1 with UV 405.3 ± 56.8, diet 1
without UV 476.26 ± 32.8; diet 2 with UV 521.7 ± 59.4,
diet 2 without UV 447.20 ± 59.3) (Table 6 & 7). Vitamin D3
and UV light did not affect growth (Table 7) and there was
no significant interaction between UV light, vitamin C and
vitamin D3 (Table 7).

receiving, samples of all diets were analysed after processing.
Diets supplemented with vitamin C contained elevated levels
of free ascorbate (diet 3, 170 ± 1 mg kg)1; diet 4,
195 ± 0.5 mg kg)1) after processing. In contrast, diets not
supplemented with vitamin C and both commercial diets
contained proportionally lower levels of free ascorbate (diet
1, 30 ± 0 mg kg)1; diet 2, 40 ± 1 mg kg)1; diet 5,
50 ± 1 mg kg)1; diet 6, 45 ± 0.5 mg kg)1). The level of

vitamin D3 present in each diet after processing shows
increased levels in the supplemented diets (diet 2,
164 000 ± 32 000 IU kg)1; diet 4, 82 000 ± 6000 IU kg)1);
however, the levels present in the diets not supplemented with
vitamin D3 are proportionally lower and comparable to the
commercial diets (diet 1, 11 200 ± 800 IU kg)1; diet 3,
8600 ± 200 IU kg)1; diet 5, 8200 ± 200 IU kg)1; diet 6,
14 200 ± 600 IU kg)1). Measurable levels of vitamin D3
were present in all diets because of the use of fish meal and
fish oil components.

To quantify changes in vitamin C and vitamin D3 levels
during diet processing, and the exact levels the fish were

Three types of deformities developed in the juvenile
L. calcarifer including the spinal deformities Ôbroken backÕ

3

Table 6 Mean initial weight (g) ± SE (n = 5 tanks), mean final weight (g) ± SE (n = 5 tanks) and weight gain (%) ± SE (n = 5 tanks) for
juvenile Lates calcarifer fed diets 6 experimental diets in the presence (UV+) or absence (UV)) of UV light

Diet
1
2
3
4
5
6

Mean initial weight (g) ± SE

Mean final weight (g) ± SE

Weight gain (%) ± SE

UV+

UV+

UV+

240.5
187.6
205.4
207.6
228.6
212.7

UV)
±
±
±
±
±
±

31.8
28.9
26.0

25.0
14.7
22.0

276.5
254.4
252.1
261.9
267.3
253.4

±
±
±
±
±
±

176.5
22.3
141.3
53.1
95.4
77.3

1137.2
1086.0
1488.6
1434.8
1202.3

947.0

UV)
±
±
±
±
±
±

22.4
12.8
20.9
27.6
17.2
28.1

1486.6
1271.1
1685.2
1478.9
1397.9
953.2

±
±
±
±
±
±

142.4
136.6
68.7
143.8
78.0
36.7

405.3
521.7
657.6
658.9
424.5
368.3

UV)
±
±
±
±
±
±

1

56.8
59.41
39.72
72.92
21.1

14.4

476.26
447.20
607.33
511.20
448.13
330.35

±
±
±
±
±
±

32.81
59.31
15.62
35.52
18.0
64.7

Values sharing the same superscript in the same column are not significantly different. Diets 5 and 6 were not incorporated in statistical
analysis because they may differ in calorific content.

..............................................................................................

Aquaculture Nutrition 17; 235–243 Ó 2010 Blackwell Publishing Ltd

(a)

syndrome involved a fracture or break in either one vertebrae
or between two vertebrae, often resulting in paralysis of the
fish. This occurred predominantly with precaudal vertebrae
(between vertebrae 2 to 17) (P < 0.05) (Table 8). Lordosis
was formed through the distortion of one to four vertebrae
and involved only the caudal vertebrae (P < 0.05) (vertebral
numbers 15 and 22) resulting in an abnormal U-shaped
curvature in the sagittal plane of the spine (Table 8). Lordosis and Ôbroken backÕ syndrome never co-occurred in
individual barramundi. At the termination of the experiment,
opercula deformities were in their early stages of development, characterized by slight inward folding of the preopercula and sub-opercula soft tissue.

(b)

(c)

Figure 3 a, Radiograph of a juvenile Lates calcarifer specimen with
the spinal deformity Ôbroken backÕ syndrome resulting from a diet
deficient in vitamin C; b, Radiograph of a juvenile Lates calcarifer
specimen suffering from lordosis as the result of a diet deficient in
vitamin C; c, Radiograph of a correctly formed juvenile
Lates calcarifer specimen fed sufficient dietary vitamin C.

syndrome (Fig. 3a) and lordosis (Fig. 3b), apparent when
compared with a correctly formed individual (Fig. 3c), and
opercula folding. All fish with spinal deformities had a correctly inflated swim-bladder. Fish fed a diet not supplemented with vitamin C also showed additional signs of ill
health such as haemorrhaging around the mouth, gills, eyes
and opercula, and necrosis of the caudal fin. Several fish-fed

diets not supplemented with vitamin C became paralysed,
characterized by the fish lying still on the bottom of the tank
with opercula and pectoral fin movement. When stimulated,
the paralysed fish reacted with a swim-like spasm arching its
body in the lateral plane. The condition of Ôbroken backÕ

This study is the first to quantify and histologically examine
the effects of supplemented dietary vitamin C on spinal
deformities in juvenile L. calcarifer and test the effects of
vitamin D3 and UV light on spinal deformities. Vitamin C
levels of 170–195 mg kg)1, as free ascorbic acid in the diet,
prevented spinal deformities and folding of the opercula in
juvenile L. calcarifer. In contrast, supplemented vitamin D3
and exposure to UV light had no effect on the development of
deformities. Additionally, the two commercial diets had sufficient levels of vitamin C as ascorbal-2-phosphate to greatly
reduce the rate of deformation in the fish compared to diets not
supplemented with vitamin C. Diets not supplemented with
vitamin C or D3 did not affect the incidence of jaw deformities.
A study examining vitamin C requirements in juvenile
L. calcarifer found vitamin C deficiency signs in fish fed diets
containing 500 mg kg)1 vitamin C, and optimal growth was
achieved with diets supplemented with 1000 mg kg)1
(Boonyaratpalin 1997). However, vitamin C was added as
free ascorbate, and levels of vitamin C postprocessing are not
available. Furthermore, the histological nature of deformities
was not documented preventing specific comparisons. However, the inclusion of lower levels of vitamin C as ascorbyl-2monophosphate (30 mg kg)1) in the diets of juvenile
L. calcarifer was effective in preventing the formation of
deformities when visually assessed (Phromkunthong et al.
1997) and is comparable with levels in commercial diets,
thereby setting a benchmark for minimum vitamin C

requirements. In addition, analysis of diets not supplemented
with vitamin C in this study demonstrates underlying vitamin
C content within the basal ingredients. Therefore, any recommendations on dietary vitamin C requirements for barramundi should account for this content in determining
minimum requirements.

..............................................................................................

Aquaculture Nutrition 17; 235–243 Ó 2010 Blackwell Publishing Ltd


Table 8 Occurrence of lordosis and Ôbroken backÕ syndrome and
their position along the vertebral column

Vertebral
number

Spinal
deformity

Number of
vertebrae
involved

% of all
spinal
deformities

2
3
5

6
7
8
9
10
15

BB
BB
BB
BB
BB
BB
BB
BB
BB
LORD
LORD
BB
LORD
LORD
LORD
LORD
LORD
LORD

2
1–2
2
2

1–2
1
1–2
1
1
3–4
1–4
1
2–5
2–4
4
4
3
1–2

3.2
6.5
3.2
6.5
6.5
3.2
6.5
3.2
6.5
3.2
12.9
3.2
3.2
12.9
3.2

3.2
9.6
3.2

16
17
18
19
20
21
22

BB,Ôbroken backÕ syndrome; LORD, lordosis.

Similar to the current research, vitamin C is essential for
the correct vertebral development of many teleost species
(Mahajan & Agrawal 1980; Soliman et al. 1986; Dabrowski
et al. 1990; Boonyaratpalin & Phromkunthong 2001). Diets
deficient in vitamin C can specifically induce lordosis, scoliosis (Halver et al. 1969; Dabrowski et al. 1996) and Ôbroken
backÕ syndrome (Wilson & Poe 1973). Furthermore, insufficient vitamin C can cause decalcification of the vertebrate
(Dabrowski et al. 1990) and reduce the collagen within the
bones (Lovell 1973; Wilson & Poe 1973).
The addition of vitamin C to diets also completely prevented the development of opercula and jaw deformities in
barramundi. Notably, the opercula deformities induced by
the deficiency of vitamin C were not acute folds of the entire
opercula but consisted of folding in the soft tissues of the
sub- and preopercula. More severe deformities may have
been observed had the fish been cultured for a longer period.
Alternatively, because younger fish require higher levels of
vitamin C (Dabrowski et al. 1996), deformities may be more

prevalent when conducted with younger fish undergoing
more rapid growth over a longer period (Wang et al. 2003).
Similarly, while vitamin C did not affect the development of
jaw deformities in juvenile L. calcarifer, abnormal jaw formation can occur in very young fish fed a diet lacking vitamin
C (Chavez De Martinez 1990).
From the current findings, it is possible to infer the role of
dietary vitamin C in the ontogeny of skeletal deformities in

..............................................................................................

Aquaculture Nutrition 17; 235–243 Ó 2010 Blackwell Publishing Ltd

larval barramundi. The development of spinal deformities in
commercially reared larval L. calcarifer is initiated from
20 days after hatching (DAH), with the incidence increasing
until 38 days after hatching (Fraser et al. 2004). In the current study, feeding of commercial diets to juvenile L. calcarifer (>30 mm in length, 60 DAH) did not affect the
incidence of spinal deformities, suggesting that they develop
predominantly during the larval phase of growth in fish
reared under standard industry protocols.
In contrast to the developing understanding of the minimum requirements of vitamin C, the requirement and role of
vitamin D3 in fish nutrition remains unclear (Lall & LewisMcCrea 2007) and the outcomes of this study do not advance
the current understanding. A lack of supplemented dietary
vitamin D3 does not influence the rate of deformities in cultured juvenile L. calcarifer. One reason may be the naturally
high levels of vitamin D3 found in the fish meal and oil
components of artificial diets (see Results section).
The results of this study allow for the development of
hypotheses for the physical mechanisms underlying the
development of skeletal deformities of the spine. Lordotic
deformities occur primarily in the caudal vertebrae, while
Ôbroken backÕ syndrome occurs in the precaudal vertebrae.

The underlying cause of this pattern is unknown; however,
analysis of vertebral column function and the mechanics of
swimming may provide some explanation. In scombrids,
swimming motion is achieved through muscular forces
exerted on the caudal vertebrae causing a lever action
between the caudal and precaudal vertebrae (Westneat &
Wainwright 2001). The forces exerted on the caudal vertebrae are distributed across several of the caudal centra at
multiple attachment points causing it to bend while there is
minimal flexing between the precaudal centra. Forces exerted
on the caudal vertebrae, which are developing without the
correct collagen content, may undergo buckling failure of the
caudal centra resulting in a U-shaped curvature. This is
anecdotally supported by changes in the detailed architecture
of lordotic vertebrae of D. labrax which appeared to have
undergone locally increased bending movements (Kranenbarg et al. 2005). Furthermore, Divanach et al. (1997) demonstrated that P. major reared in high water currents to
induce excessive swimming activity induced lordotic deformities in the caudal vertebrae. Additionally, in many teleost
fishes, the neural and haemal spines are bound by collagen
fibres. This forms a vertical septum used for the transmission
of energy when swimming and for the prevention of vertebrae rotating out of the sagittal plane (Westneat & Wainwright 2001). Without sufficient levels of vitamin C, it may be
expected that structural integrity of the collagen fibres could


be compromized. Therefore, forces exerted by the surrounding musculature on precaudal vertebrae without correct structural rigidity supplied by collagen fibres and correct
bone formation may induce Ôbroken backÕ syndrome.
In conclusion, L. calcarifer fed a diet not supplemented
with vitamin C develop high levels of spinal deformities with
lordosis occurring predominantly in caudal vertebrae and
Ôbroken backÕ syndrome in the precaudal vertebrae. The
precise nature of spinal deformities also suggests that biomechanical forces play a key role in their development.
Supplementing diets with vitamin C is suggested as a safeguard mechanism to prevent the development of deformities

in juvenile barramundi, and may also be critical in larval
feeding.

This study was financially supported by Cell Aquaculture
Ltd. We would like to thank Skretting, Ridley Aqua-feed and
Bluewater Barramundi for supporting this project.

Alexis, M.N., Karanikolas, K.K. & Richards, R.H. (1997)
Pathological findings owing to the lack of ascorbic acid in cultured gilthead sea bream (Sparus aurata L.). Aquaculture, 151,
209–218.
Andrades, J.A., Becerra, J. & Fernandez-Llebrez, P. (1996) Skeletal
deformities in larval, juvenile and adult stages of cultured sea
bream (Sparus aurata L.). Aquaculture, 141, 1–11.
Barahona-Fernandes, M.H. (1982) Body deformation in hatchery
reared European sea bass Dicentrarchus labrax (L). Types, prevalence and effect on fish survival. J. Fish Biol., 21, 239–249.
Boonyaratpalin, M. (1997) Nutrient requirements of marine food fish
cultured in Southeast Asia. Aquaculture, 151, 283–313.
Boonyaratpalin, M. & Phromkunthong, W. (2001) Bioavailability of
ascorbyl phosphate calcium in hybrid catfish, Clarias macrocephalus (Gunther) · Clarias gariepinus (Burchell) feed. Aquacult. Res.,
32(Suppl. 1) 126–134.
Boonyaratpalin, M., Unprasert, N. & Buranapanidgit, J. (1989)
Optimal supplementary vitamin C level in seabass fingerling diet.
In: The Current Status of Fish Nutrition in Aquaculture (Takeda,
M. & Wantanabe, T. eds), pp. 149–157. Tokyo University of
Fisheries, Tokyo, Japan.
Chavez De Martinez, M.C. (1990) Vitamin C requirement of the
Mexican Native Cichlid Cichlasoma urophthalmus (Gunther).
Aquaculture, 86, 409–416.
Dabrowski, K., El-Fiky, N., Kock, G., Frigg, M. & Wieser, W.
(1990) Requirement and utilisation of ascorbic acid and ascorbic

sulphate in juvenile rainbow trout. Aquaculture, 91, 317–337.
Dabrowski, K., Moreau, R., El-Saidy, D. & Ebeling, J. (1996)
Ontogenetic sensitivity of channel catfish to ascorbic acid deficiency. J. Aquat. Anim. Health, 8, 22–27.
Daoulas, C., Economou, A.N. & Bantavas, I. (1991) Osteological
abnormalities in laboratory reared seabass (Dicentrarchus labrax)
fingerlings. Aquaculture, 97, 169–180.

Divanach, P., Papandroulakis, N., Anastasiadis, P., Koumoundouros, G. & Kentouri, M. (1997) Effects of skeletal deformities in sea
bass (Dicentrarchus labrax L.) with functional swimbladder during
postlarval and nursery phase. Aquaculture, 156, 145–155.
Fjelldal, P.G., Hansen, T.J. & Berg, A.E. (2007) A radiological study
on the development of vertebral deformities in cultured Atlantic
Salmon (Salmo salar L.). Aquaculture, 273, 721–728.
Fraser, M.R. & de Nys, R. (2005) The morphology an occurrence of
jaw and operculum deformities in cultured barramundi (Lates
calcarifer) larvae. Aquaculture, 250, 496–503.
Fraser, M.R., Anderson, T. & de Nys, R. (2004) Ontogenic development of the spine and spinal deformities in larval barramundi
(Lates calcarifer) culture. Aquaculture, 242, 697–711.
Galeotti, M., Beraldo, P., de Dominis, S., DÕAngelo, L., Ballestrazzi,
R., Musetti, R., Pizzolito, S. & Pinosa, M. (2000) A preliminary
histological and ultrastructural study of opercular anomalies in
gilthead sea bream larvae (Sparus aurata). Fish Physiol. Biochem.,
22, 151–157.
Gapasin, R.S.J., Bombeo, R., Lavens, P., Sorgeloos, P. & Nelis, H.
(1998) Enrichment of live food with essential fatty acids and
vitamin C: effects on milkfish (Chanos chanos) larval performance.
Aquaculture, 162, 269–286.
Graff, I.E., Hoie, S., Totland, G.K. & Lie, O. (2002) Three different
levels of dietary vitamin D3 fed to first-feeding fry of Atlantic
salmon (Salmo salar L.): effect on growth, mortality, calcium

content and bone formation. Aquacult. Nutr., 8, 103–111.
Halver, J.E., Ashley, L.M. & Smith, R.R. (1969) Ascorbic acid
requirements of Coho Salmon and Rainbow Trout. Trans. Am.
Fish. Soc., 98, 762–771.
Hattori, M., Sawada, Y., Takagi, Y., Suzuki, R., Okada, T. &
Kumai, H. (2003) Vertebral deformities in cultured red sea bream,
Pagrus major, Temminck and Schlegel. Aquacult. Res., 34, 1129–
1137.
Hilton, J.W. & Ferguson, H.W. (1982) Effect of excess vitamin D3 on
calcium metabolism in rainbow trout Salmo gairdneri (Richardson). J. Fish Biol., 21, 373–379.
Horton, H., Moran, L.A., Ochs, R.S., Rawn, D.J. & Scrimgeour,
G.K. (1993) Principles of Biochemistry. N. Patterson PublisherÕs,
Eaglewood Cliffs, New Jersey, pp. 412–415.
Kamler, E., Wolnicki, J. & Kaminski, R. (2008) Fatty acid composition, growth and morphological deformities in juvenile cyprinid,
Scardinius erythropthalmus fed formulated diet supplemented with
natural food. Aquaculture, 278, 69–76.
Kihara, M., Ogata, S., Kawano, N., Kubota, I. & Yamaguchi, R.
(2002) Lordosis induction in juvenile red sea bream, Pagarus
major, by high swimming activity. Aquaculture, 212, 149–158.
Koumoundouros, G., Maingot, E., Divanach, P. & Kentouri, M.
(2001) Kyphosis in reared sea bass (Dicentrarchus labrax L.):
ontogeny and effects on mortality. Aquaculture, 209, 49–58.
Kranenbarg, S., Waarsing, J.H., Muller, M., Weinans, H. & van
Leeuwen, J.L. (2005) Lordotic vertebrae in sea bass (Dicentrarchus
labrax L.) are adapted to increased loads. J. Biomech., 38, 1239–
1246.
Lall, S.P. & Lewis-McCrea, L.M. (2007) Role of nutrients in skeletal
metabolism and pathology in fish – An overview. Aquaculture, 267,
3–19.
Li, M.H. & Robinson, E.H. (2001) Dietary ascorbic acid requirement

for growth and health in fish. In: Nutrition and Fish Health (Lim,
C. & Webster, C.D. eds), The Haworth Press, Inc., Binghamto,
New York, pp. 163–187.
Lin, M.F. & Shiau, S.Y. (2004) Requirements of vitamin C (L-ascorbyl-2-monophosphate-Mg and L-ascorbyl-2-monophosphate-Na)

..............................................................................................

Aquaculture Nutrition 17; 235–243 Ó 2010 Blackwell Publishing Ltd


and its effects on immune responses of grouper, Epinephelus malabaricus. Aquacult. Nutr., 10, 327–333.
Lovell, R.T. (1973) Essentiality of vitamin – C in feeds for intensively
fed caged channel catfish. J. Nutr., 103, 134–138.
Mahajan, C.L. & Agrawal, N.K. (1979) Vitamin C deficiency in
Channa punctatus Bloch. J. Fish Biol., 15, 613–622.
Mahajan, C.L. & Agrawal, N.K. (1980) Nutritional requirement of
ascorbic acid by Indian major carp, Cirrhina mrigala, during early
growth. Aquaculture, 19, 37–48.
NRC (National Research Council) (1993) Nutrient Requirements of
Fish. National Academy Press, Washington, DC, USA. 114 p.
Paperna, I. (1978) Swim bladder and skeletal deformations in
hatchery bred Sparus aurata. J. Fish Biol., 12, 109–114.
Phromkunthong, W., Boonyaratpalin, M. & Storch, V. (1997) Different concentrations of ascorbyl-2-monophosphate-magnesium as
dietary sources of vitamin C for seabass, Lates calcarifer. Aquaculture, 151, 225–243.
Quinn, G.P. & Keough, M.J. (2002) Experimental Design and
Data Analysis for Biologists. Cambridge University Press,
Cambridge.
Roberts, R.J., Hardy, R.W. & Sugiura, S.H. (2001) Screamer diseases in Atlantic salmon, Salmo salar L., in Chile. J. Fish Dis., 24,
543–549.
Soliman, A.K., Jauncey, K. & Roberts, R.J. (1986) The effect of

dietary ascorbic-acid supplementation on hatchability, survival

..............................................................................................

Aquaculture Nutrition 17; 235–243 Ó 2010 Blackwell Publishing Ltd

rate and fry performance in Oreochromis-mossambicus (Peters).
Aquaculture, 59, 197–208.
Sullivan, M., Reid, S.W.J., Ternent, H., Manchester, N.J., Roberts,
R.J., Stone, D.A.J. & Hardy, R.W. (2007) The aetiology of spinal
deformity in Atlantic Salmon, Salmo salar L.: influence of different
commercial diets on the incidence and severity of the preclinical
condition in salmon parr under two contrasting husbandry
regimes. J. Fish Dis., 30, 759–767.
Tucker, J.W., Russell, J.D. & Rimmer, M.A. (2002) Barramundi
culture: a success story for aquaculture in Asia and Australia.
World Aquac., 33, 53–59.
Wang, X., Kim, K.-W., Bai, S.C., Huh, M.-D. & Cho, B.-Y. (2003)
Effects of the different levels of dietary vitamin C on growth and
tissue ascorbic acid changes in parrot fish (Oplegnathus fasciatus).
Aquaculture, 215, 203–211.
Webb, A.R. (1993) Vitamin D synthesis under changing UV spectra.
In: Environmental UV Photobiology (Young, A.R., Bjorn, L.O.,
Moan, J. & Nultsch, W. eds), Plenum Press, New York, pp. 185–
202.
Westneat, M.W. & Wainwright, S.A. (2001) Mechanical design for
swimming: muscle, tendon and bone. In: Tuna: Physiology, Ecology and Evolution (Block, B.A. & Stevens, E.D. eds), Academic
Press, San Diego, California, pp. 272–308.
Wilson, R.P. & Poe, W.E. (1973) Impaired collagen formation in the
scorbutic channel catfish. J. Nutr., 103, 1359–1364.

Aquaculture Nutrition
doi: 10.1111/j.1365-2095.2009.00736.x

2011 17; 244–247

..............................................................................................

Dependencia de Educacio´n Superior Ciencias Naturales y Exactas, Universidad Auto´noma del Carmen, Ciudad del Carmen,
Campeche, Me´xico

Because of the filter-feeding behavior of shrimp larvae, it is
important to define precisely the size of the particle ingested
in the different stages until postlarval stage where raptorial
habits are more evident than the filter-feeding lifestyle.
Selectivity assays were conducted by using Polystyrene DVB
particles with diameter between 1 and 50 lm as food.
A group of organisms from each stage were put into the
particle suspension for 15 min to let the polystyrene particles
be ingested. The particle distribution in the media and the
content of the gut of the larvae were characterized with
digital image processing analysis. The results were compared
using Ivlev selectivity formula, which compares the frequency
distribution of each size of the particle in the media and in
the gut of larvae. The results of selectivity were adjusted with
a third-order polynomial regression to determine the optimum and preferred size of the food particles for each larval
stage between Zoea I and Postlarva I. It is concluded that the
different larval stages of Litopenaeus vannamei may be
considered as a single group of larvae who ingest foods with

size between 5.71 and 20.33 lm. The optimal size of the food
ingested was 14.42 lm wide.
KEY WORDS:

feeding, larvae, particle size, selectivity, shrimp

Received 23 June 2009, accepted 28 September 2009
Correspondence: R. Gelabert, Dependencia de Educacio´n Superior Ciencias
Naturales y Exactas, Universidad Auto´noma del Carmen, Calle 56 No. 4
Esq. Ave. Concordia. Col. Benito Jua´rez, C.P. 24180, Ciudad del Carmen,
Campeche, Me´xico. E-mail:

Formulated feeds play an important role in semi-intensive
shrimp production, constituting nearly 55% of the total
operating cost (Mohanty 2001), so it is necessary for food

producers to carefully handle not only the ingredients to
support the appropriate nutritional value and stability in the
water, but also the portion size. The food not ingested sinks
to the bottom of the culture tank and affects the water
quality. Furthermore, the cost of larval production is
increased when the food is not utilized by larvae, raising the
total cost of the hatchery.
The nutritional effectiveness of a food organism is in the
first place determined by its ingestibility and as a consequence by its size and configuration (Leger et al. 1986). Food
ration and food selectivity are the basis for intensive rearing
technology for zooplankton-fed fish (Szlaminska et al. 1999).
Partial or total replacement of microalgae by artificial food
in the assays conducted by Robinson et al. (2005) indicated
that further investigations are necessary to determine

whether inert feeds offer a less nutritionally balanced diet, a
less digestible diet, or inadequate particle sizes that are
rejected by shrimp larvae.
Feeding shrimp larviculture has been studied to evaluate
replacement of live food by artificial diets (Brito et al. 2001;
Pedroza- Islas et al. 2004; DÕAbramo et al. 2006; Pin˜a et al.
2006), but artificial diets for the total replacement of live
feeds for rearing marine larvae have not been developed
despite many years of research. The types of microparticles
used to deliver nutrients to larvae need to be carefully
evaluated and improved (Langdon 2003).
Considering the filter-feeding behavior of shrimp larvae as
well as the scarce information about the size of the food
ingested by shrimp larvae and the importance for the larviculture, the present paper evaluated the particle size ingested
by Litopenaeus vannamei larvae from Zoea I to the first
postlarval stage.

The larvae, obtained from a commercial hatchery at the
Nauplius stage, were reared to the postlarva 1 (PL-1) stage

..............................................................................................

Ó 2009 Blackwell Publishing Ltd
No claim to original US government works


Table 1 Feeding schedule for rearing Litopenaeus vannamei larvae. (Z: zoea, M: mysis, PL postlarva)

)1

Chaetoceros gracilis (cel mL )
Tetraselmis chui (cel mL)1)
Artemia franciscana (nauplii mL)1)

NV

ZI

ZII

ZIII

MI

10 000
5 000

40 000
10 000

75 000

85 000
25 000
0–0.2

60 000

in 10 L cylindroconical fiberglass tanks using natural seawater at 35 PSU filtered through sand and cartridge filters
(20 and 5 lm) and sterilized with UV radiation. Oxygen was

maintained at >5 mg L)1 by aeration, temperature and pH
were 28 °C and 8.0, respectively. The stocking density was 50
larvae L-1, and the larvae were maintained in triplicate
cylindroconical fiberglass tanks, with 50% daily natural
seawater (35 PSU) exchanges. The larvae were fed twice a
day with Chaetoceros gracilis, Tetraselmis chui, and Artemia
franciscana nauplii, following the feeding schedule shown in
Table 1.

Selectivity experiments were carried out using larval stages
from Zoea I (ZI) to Postlarva 1 (PL1). A group near of 100
larvae at the same stage was put in one liter Erlenmeyer flasks
with filtered seawater where a suspension of Polystyrene
DVB particles (Duke Scientific Corp., CAT. No. 445) was
added. The size of the particles ranged between 1 and 50 lm,
and the suspension was provided to establish 200 000 particles mL)1 in the culture media. Previous assays indicated
that it was not necessary that the gut of experimental larvae
was empty to allow the polystyrene particles to be ingested,
so larvae did not receive any special treatment before they
were placed in the flask. The vessels were inverted for every
30 s to avoid the fast precipitation of the bigger particles and
to avoid a discriminate selectivity of the small particles with
slow precipitation. Fifteen minutes after the suspension was
supplied, all the larvae were collected with a mesh and were
rinsed with enough filtered seawater to remove the particles
stuck to their bodies. The organisms were measured from the
beginning of the rostrum to the end of the telson to establish
the total length. An Evolution MP camera linked to the
microscope was used to obtain images that were analyzed
with Image Pro Plus software for Windows (V. 4.51) (Media

Cybernetic, Inc.) to do all the measures. Larvae studied were
dissected to recover the particles ingested, which were measured using the same procedure to obtain the frequency
distribution of each large classes observed.
The selectivity of particles to be ingested in each larval
stage was determined through the comparison of the pro-

..............................................................................................

Aquaculture Nutrition 17; 244–247 Ó 2009 Blackwell Publishing Ltd
No claim to original US government works

MII

1

MIII

PLI

50 000
20 000
1.5

2

15 000
3

portion of particles present in the gut and the proportion of
the same large class present in the experimental media. The

Ivlev selectivity index (Salazar & Gonza´lez 1986) was used:
S = G ) E/G + E, where S = selectivity index, G =
percentage of particles of specific size in the gut, and E =
percentage of particles of the same size in suspension. The
values of S range from )1 to+1. When the frequency of a
given size class is greater in the gut than in the medium, the
value of S is positive, and the organism selects those particles
for ingestion.
To reduce the individual variations of the results, a thirdorder polynomial regression of the selectivity index with
respect to the sizes of the particles ingested by the individuals
from different stages was calculated. It showed the ingestion
limits for each of the stages analyzed. The F value and
resulting P value were used as an overall F test of the relation
between the dependent variable and the independent variable. The program Microsoft OfficeÒ ExcelÒ 2007 was used
for the preliminary handling of data and Statistica for
Windows version 8.0 for statistical and regression analyses.

A total of 8675 particles were measured in the experimental
media to establish the frequency distribution for size classes
in the polystyrene product used. The number of organisms
studied as well as their size for each larval stage and the
number of particles analyzed in the gut of the larvae are
shown in Table 2. The values of Determination coefficient
(R2) and P for the polynomial regression model obtained
were used to explain the relation between the size of the
particle ingested and the selectivity index.
The model shows the range of particle size ingested with
positive selectivity and the optimal size ingested for all larval
stages. Minimal and maximal sizes with positive selectivity
found in the regression model include the optimal size found

for each of the different stages (Table 3). A correlation
analysis between the size of the larva grouped by larval stages
and the optimal particle size ingested is shown in Table 4.
The results indicate that there is no significant linear
relationship between the optimal size of the particles ingested
and the size of the larvae in neither the zoea nor the mysis


Larval
size (lm)

N particles
analyzed
in the gut

R2

F value

P value

19.47
F3,29
61.46
F3,27
20.33
F3,46
10.14
F3,31
18.85

F3,28
37.23
F3,33
28.75
F3,26

0.000

Stage

N larvae

ZI

31

945.3

33.4

1 031

0.668

ZII

60

1 344.4

103.2

1 948

0.872

ZIII

32

1 838.4

147.4

2 012

0.542

MI

35

2 730

180.3

1 601

0.495

MII

11

3 154.4

164.9

602

0.623

MIII

27

3 863.5

100.9

1 226

0.751

PL1

32

4 172.7

502.5

255

0.768

SD

Table 3 Range of ingested particle size (lm) for the different larval
stages (Z: zoea, M: mysis, PL: postlarva), ICL (inferior confidence
level) SCL (superior confidence level) from the polynomial regression
Larval
stage
ZI
ZII
ZIII
MI
MII
MIII
PL1
All stages

ICL
1.76
4.80
5.27
4.73
5.17
4.36
5.71

Minimal
size

Optimal
size

Maximal
size

SCL

4.4
4.94
7.04
6.73
6.11
6.51
5.23
7.47

8.02
9.67
9.92
12.86
11.58
12.31
10.15
14.42

12.23
15.70
13.22
20.14
18.42
19.44
16.36
18.23

14.48
16.79
18.02
22.22
20.68
21.21
17.59
20.33

Table 4 Correlation analysis between the larval size and the optimal
particle size ingested. (Z: zoea, M: mysis)
Larval
stage

R2

F value

P value

Regression equation

ZI-ZIII
MI-MIII
All stages

0.80
0.38
0.43

4.00 (1,1)
0.61 (1,1)
3.73 (1,5)

0.295070
0.576298
0.111342

y = 6.363 + 0.0021*x
y = 15.0829)0.0008*x
y = 8.353 + 0.0009*x

stages. With this finding, all the larvae were considered to be
only one group, and a common selectivity polynomial
regression was generated (All stages, Table 3).

Up to the present, it has been essential to provide live feed for
the first feeding of marine species (Lee 2003). The reason is
not clear, but it is supposed that the motility and the palatability of the prey stimulate the predator to catch it, the
exogenous enzymatic battery ported by the live food which

0.000

Table 2 Number of particles analyzed
and quantity of larvae studied (N) for
each larval stage (Z: zoea, M: mysis,
PL: postlarva) (SD-standard deviation, R2-multiple correlation coefficient,
F value degree of freedom in parenthesis

0.000
0.000
0.000
0.000
0.000

contributes to the digestion of the food ingested (Kolkovski
2001), and the minimal leaching of nutrients to the media
could be some of the answers.
Partial and complete replacement of live microalgae and
Artemia nauplii with microalgae pastes and different inert
feeds were used in the study carried out by Robinson et al.
(2005), which focused on larval feeding of Farfantepenaeus
aztecus. The results indicate that it is better to use live food
than an artificial diet. The authors argued that food particle
diameter could have reduced growth and survival of larval
shrimp because food particles must be of an appropriate size
to assure maximum rate of ingestion.
For shrimp larviculture, artificial diets as a complement or
substitute for live food have been promoted by different
producers. Most commercial companies producing feeds for
shrimp larvae (Nippai, Argent Foods, Aquatic Ecosystem,

Zeigler Bros Inc) have a particle size range of around
100 lm, which is considerably larger than the optimum
suggested by the present results.
As well as the fact that the commercial artificial food does
not satisfy the nutritional requirements for species which they
were not designed to, it is important to consider that the
commercial brands are not always going to be of the size of
the food required, not for all larval stages, nor the different
species. Considering the results obtained here and the data
collected from different brands, the food produced for the
shrimp larviculture does not always fill the need of the size
required to promote adequate ingestion for all larvae stages.
The non-lineal correlation found between the optimal size
of the particle ingested and the size of the larvae do not agree
with the results of Cruz & Ortega (1989), who indicated that
the size of the particle ingested increases with the size of the
larvae in L. schmitti larvae. These results could be explained

..............................................................................................

Aquaculture Nutrition 17; 244–247 Ó 2009 Blackwell Publishing Ltd
No claim to original US government works


by the number of animals analyzed in each experiment, by
the precision of the measurement methods employed to
characterize the gut content, or mainly by the morphological
differences between species studied. The results found here
suggested that larger L. vannamei larvae do not ingest by
filtration big particles.

The optimal size of polystyrene particles ingested by the
different stages varies between 8.02 and 12.86 lm, which
could be considered insignificant for the industrial production
of food in aquaculture. The confidence level obtained extends
food producers the possibility of expanding the production of
artificial diets by producing diets from 5.71 to 20.33 lm wide.
Probably, the most efficient means of improving microparticle
feed efficiency lies in reducing variance in particle size and
could ultimately be derived from improvements in feed
particle-manufacturing technology (Robinson et al. 2005).
Research has yielded microbound and microencapsulated
diets that possess the following desirable characteristics:
acceptability, digestibility, adequate particle size, and most
importantly, good water stability (Pedroza- Islas et al. 2004).
Results obtained in the experiments carried out here illustrated the necessity for more research in the field of artificial
food production, not only considering their nutritional value,
acceptability, digestibility, and water stability, but also taking into account the particle size.
A simple way to contribute to the improvement of larviculture is to consider the size of the food supplied, which is
going to lead to less energy used by the larvae in the feeding
process and reduced decomposition in the culture media.

The authors thank Dra. Gabriela Gaxiola, Gabriela Palomino, and Adriana Paredes, members of Unidad Multidisciplinaria de Docencia e Investigacio´n from Universidad
Nacional Auto´noma de Me´xico, for their support given to
the research.

..............................................................................................

Aquaculture Nutrition 17; 244–247 Ó 2009 Blackwell Publishing Ltd
No claim to original US government works

Brito, R., Rosas, C., Chimal, M.E. & Gaxiola, G. (2001) Effect of
different diets on growth and digestive enzyme activity in Litopenaeus vannamei (Boone, 1931) early post-larvae. Aquacult. Res., 32,
257–266.
Cruz, S.A. & Ortega, S. (1989) La seleccio´n del taman˜o de partı´ culas
alimenticias por las larvas del camaro´n blanco Penaeus schmitti.
Rev. Investig. Mar., 2, 163–174.
DÕAbramo, L. R., Perez, E. I., Sangha, R. & Puello-Cruz, A. (2006)
Successful culture of larvae of Litopenaeus vannamei fed a microbound formulated diet exclusively from either stage PZ2 or M1 to
PL1. Aquaculture, 261, 1356–1362.
Kolkovski, S. (2001) Digestive enzymes in fish larvae and juveniles—implications and applications to formulated diets. Aquaculture, 200, 181–201.
Langdon, C. (2003) Microparticle types for delivering nutrients to
marine fish larvae. Aquaculture, 227, 259–275.
Lee, C.S. (2003) Biotechnological advances in finfish hatchery production: a review. Aquaculture, 227, 439–458.
Leger, P., Bengtson, D.A., Simpson, K.L. & Sorgeloos, P. (1986) The
use and nutritional value of Artemia as a food source. Oceanogr.
Mar. Biol. Ann. Rev., 24, 521–623.
Mohanty, R.K. (2001) Feeding management and waste production
in semi-intensive farming of Penaeus monodon (Fab.) at different
stocking densities. Aquacult. Int., 9, 345–355.
Pedroza- Islas, R., Gallardo, P., Vernon-Carter, E.J., Garcı´ aGalano, T., Rosas, C., Pascual, C. & Gaxiola, G. (2004) Growth,
survival, quality and digestive enzyme activities of larval shrimp
fed microencapsulated, mixed and live diets. Aquac. Nutr., 10,
167–173.
Pin˜a, P., Voltolina, D., Nieves, M. & Robles, M. (2006) Survival,
development and growth of the Pacific white shrimp Litopenaeus
vannamei protozoea larvae, fed with monoalgal and mixed diets.
Aquaculture, 253, 523–530.
Robinson, C.B., Samocha, T.M., Fox, J.M., Gandy, R.L. &
McKee, D.A. (2005) The use of inert artificial commercial food
sources as replacements of traditional live food items in the culture

of larval shrimp, Farfantepenaeus aztecus. Aquaculture, 245, 135–
147.
Salazar, O. & Gonza´lez, M.A. (1986) Intensidad de la alimentacio´n
diaria en postlarvas de Ictiobus cyprinellus. Boletı´n Te´cnico,
Ministerio de la Industria Pesquera, Empresa Nacional de Acuicultura, Cuba, 31, 1–4.
Szlaminska, M., Zarubov, A., Hamackova, I., Kouril, J., Vachta, R.,
Adamkova, I. & Mun˜oz-Asenjo, C. (1999) Food passage and food
selectivity of tench Tinca tinca (l.) larvae fed zooplankton. Acta
ichthyologica et Piscatoria, XXIX, 41–47.


Aquaculture Nutrition
2011 17; 248–257

doi: 10.1111/j.1365-2095.2009.00740.x

..............................................................................................

1
1

1

2

2

CCMAR, Universidade do Algarve, Campus de Gambelas, Faro, Portugal;
Seafood Research, Nordnes, Bergen, Norway

To study the effect of dietary supplementation of iodine in
Solea senegalensis, larvae were randomly distributed in six
tanks. Larvae in three tanks were given rotifers and Artemia
enriched with iodine in addition to Rich Advance or Super
Selco from 2 days after hatch (DAH) until 31 DAH. Larvae in
a second set of three tanks were fed control rotifers and
Artemia, enriched only with Rich Advance or Super Selco.
Samples were collected at 2, 5, 10, 15 and 31 DAH to determine
dry weight, total length, myotome height and thyroid status.
Larvae fed the iodine-enriched diet had significantly higher
weight at 31 DAH and higher levels of whole body iodine
concentration, compared to control larvae. At 31 DAH, larvae from the control treatment showed typical goitrous thyroid follicles. Thyroid cells of larvae from this treatment
appeared columnar or afollicular, with the colloid partly or
completely depleted, representative of hyperplasia (goitre).
The lower growth rate in fish larvae from the control treatment
was possibly a consequence of the hyperplasia, and the iodine
enrichment prevented Senegalese sole larvae from developing
goitre. This study demonstrates the importance of iodine
enrichment of live feed for fish reared in a recirculation system.
KEY WORDS: Artemia, growth, iodine, outer-ring deiodinase,
rotifers, Solea senegalensis, thyroid follicle

Received 9 March 2009, accepted 22 September 2009
Correspondence: A.R.A. Ribeiro, CCMAR, Universidade do Algarve,
Campus de Gambelas, 8000-117 Faro, Portugal. E-mail: ;

The Senegalese sole (Solea senegalensis) is a fish species
whose biology has been widely studied (Martinez et al. 1999;

1

2

2

NIFES, National Institute of Nutrition and

Ribeiro et al. 1999a; Yu´fera et al. 1999; Conceic¸a˜o et al.
2007) because of its great economic potential in the North
African Atlantic, in Spain and Portugal, where it is cultured
in extensive aquaculture production (Dinis et al. 1999;
Imsland et al. 2003). As a flatfish, Senegalese sole undergoes
metamorphosis, which is a critical period where the symmetrical pelagic larva changes into an asymmetrical benthic
juvenile. Metamorphosis involves eye migration, craniofacial
remodelling and implies important changes in food habits
and in digestive physiology (Tanaka et al. 1996). The thyroid
hormones (TH), thyroxine (T4) and triiodothyronine (T3),
are known to initiate metamorphosis (Schreiber & Specker
1998; Solbakken et al. 1999; Einarsdo´ttir et al. 2006). TH is
synthesized from tyrosine or phenylalanine and iodine in the
thyroid follicles (Solbakken et al. 2002). The powerful and
active T3 is converted from T4 (Eales & Brown 1993), by
outer outer-ring 5¢-monodeiodination (ORD). This event
occurs mainly in peripheral tissues and is catalyzed by the
enzymatic action of iodothyronine deiodinases (Klaren et al.
2005). Metamorphosis is not always successful and problems
such as arrested eye migration or malpigmentation represent
a bottleneck for the aquaculture production of several flatfish
species (Pittman et al. 1998; Power et al. 2001; Sæle et al.
2003). Studies have shown that these problems are owed not
only to broodstock management and the environmental

conditions of the husbandry, but to nutrition during larval
rearing as well (Dinis et al. 1999; Hamre et al. 2005).
Rotifers and Artemia are intensively produced and commonly used as live feed to marine fish larvae in aquaculture
production. Compared to copepods, the natural feed for
marine fish larvae, rotifers and Artemia have different levels
of some essential nutrients (Watanabe 1993; Whyte et al.
1994; Hamre et al. 2008b). Iodine, as other minerals, is very
low in both types of commercial live feed. While copepods
contain between 60 and 350 lg g)1 (dry weight) iodine, the

..............................................................................................

Ó 2009 Blackwell Publishing Ltd


iodine content of Artemia ranges between 0.5 and 4.6 lg g)1
(dry wt.) (Solbakken et al. 2002; Moren et al. 2006) in rotifers
the iodine content ranges between 0.52 and 7.9 lg g)1 (dry
wt.) (Hamre et al. 2008a). Being an essential part of TH,
iodine in the right concentration is crucial to assure sufficient
level of TH, consequently assuring a successful metamorphosis and normal fish development (Leatherland 1994).
Studies by Moren et al. (2006) showed that when iodineenriched Artemia was given to Atlantic halibut larvae, there
was an increase in their iodine content. Knowing this, our
goal was to feed Senegalese sole larvae with iodine-enriched
rotifers and Artemia and evaluate its effect on the development of fish larvae, focusing on growth parameters and
thyroid status.

Initial System
Control

..............................................................................................

Aquaculture Nutrition 17; 248–257 Ó 2009 Blackwell Publishing Ltd

Control

Transfer

Iodine

Iodine

Transfer

0

2
Rotifers
Green water

Eggs of S. senegalensis were obtained by natural spawning
from a broodstock adapted to captivity at the University of
the Algarve (Center of Marine Sciences, Portugal).
After hatching, larvae were transferred and reared in 100-L
conical tanks at a density of 100 larvae L)1, from hatching
(0 days after hatching – DAH) to 17 DAH, (Ribeiro et al.
1999a). At this age, all larvae had adopted a benthic life style
and were transferred to flat-bottom tanks in a recirculation
system (Fig. 1). Each conical tank was harvest and a new
tank was assigned in the recirculation system. Moreover, all

larvae were counted to determine survival (from 0 DAH until
17 DAH). Consequently, 700 larvae from each tank, were
transferred to the recirculation system, with 21-L flatbottomed tanks (width 30 cm · length 70 cm · height
10 cm), and kept there until the end of the experiment
(31 DAH), at a density of 3000 larvae m2 in 8 cm water
column (20 L) (Fig. 1). At the end of the experiment, survival
was determined for the period of 17 DAH until 31 DAH.
During the whole experiment the photoperiod was of
12 : 12 h dark/light, with an intensity of 900 lux at water
surface. Water temperature was 21.1 ± 1.0 °C, salinity was
33.3 ± 1.1 (g L)1), and oxygen saturation was 89.4 ± 9.0%.
Before transfer, each conical tank was an individual
recirculation unit with its own bio-filter (105 L; tank volume plus bio-filter), with a water flow of 280 mL min)1. In
this period (from 0 DAH until 17 DAH), 25% of new
seawater was added to each tank in order to maintain
constant salinity. After transfer, the flat-bottomed tanks
were linked to one common 1700-L recirculation system
with 24 tanks, a bio-filter, a protein skimmer and addi-

Recirculation System

7

14

17

Liver Arternia Liver + Frozen
Arternia

31 DAH
Frozen Arternia

Figure 1 Larvae were given rotifers from 2 days after hatch (DAH)
until 7 DAH, and green water technique was used. From that day
until 14 DAH larvae were given live Artemia, at this point frozen
Artemia was introduced along with live Artemia. From 17 DAH until
31 DAH, only frozen Artemia was given. Senegalese sole larvae were
distributed in six tanks. Larvae in three tanks were given rotifers and
Artemia enriched with iodine and Rich Advance or Super Selco.
Larvae in a second set of three thanks were fed control rotifers and
Artemia, enriched only with Rich Advance or Super Selco. At 17 DAH
larvae were transferred from individual flow-through recirculation
units (105 L) to a common 1700 -L recirculation system.

tional use of ozone injection. The flow for each tank was
1400 mL min)1, and a maximum of 10% of new seawater
was added to the whole system, during experimental time
(from 17 DAH until 31 DAH). Ammonia was monitored
three times a week using a commercial kit – TetratestÒ
Total Ammonia kit (Terta Werke, Melle, Germany).

Rotifers Brachionus plicatilis were reared in 120-L tanks
containing microalgae, Chlorella sp. or Nannochloropsis sp.
After grazing, the microalgae, rotifers were transferred to
another set of rearing tanks and given microalgae on the first
day and Culture Selco (Inve Aquaculture NV, Dendermode,
Belgium) on the following days. Rotifers enrichment was
conducted as described by Ribeiro et al. (2007). Artemia was
hatched according to Van Stappen (1996), and at instar II

they were enriched with Super Selco (Inve Aquaculture NV)
in 60 L conical tanks at a density of 200 nauplii per 1 mL of
seawater (35 g L)1) at 28.8 °C and with strong aeration.
After the enrichment, some of the Artemia was collected in


new seawater, counted and immediately frozen at )20 °C for
later feeding.
For iodine enrichment, a dose of 780 mg of sodium iodide
(NaI; Merck, Darmstadt, Germany) was used per 1 g emulsion: Rich Advance (Rich, Athens, Greece) for rotifers and
Super Selco (Inve Aquaculture NV) for Artemia. The
enrichment was then conducted over a period of 6 h with two
doses, one being given to live feed at 0 h (only emulsion) and
at 3 h after starting the enrichment (emulsion and NaI). The
control treatment was prepared by the same method, but
without NaI. The amount of NaI used and the dose-time was
based on the levels of iodine found in rotifers by Ribeiro
et al. (2007). The amount of total iodine in the enrichment
tanks ranged between 0.1 and 0.2 g I) L)1, depending on the
density of the live prey.

Senegalese sole larvae were randomly distributed in six tanks.
Larvae in three tanks were given rotifers and Artemia
enriched with iodine and Rich Advance or Super Selco.
Larvae in a second set of three thanks were fed control rotifers and Artemia, enriched only with Rich Advance or Super
Selco. Feed was offered to fish larvae four times a day, and
green water technique was used until 7 DAH, using Tetraselmis chuii (Rocha et al. 2008) (Fig. 1).
In order to study the effect of dietary supplementation of
iodine, fish larvae were fed ad libitum based on predicted
maximum growth. Daily adjustments were made based on

visual inspection, in order to avoid an excess of uneaten prey.
The density of life prey was calculated per day (Engrola et al.
2005). Then, from 2 DAH (mouth opening), larvae were given
rotifers, at a starting density of 5 prey mL)1, and their density
was gradually increased until 7 DAH (Ribeiro et al. 1999b).
Afterward, live Artemia metanauplii (EG Inve Aquaculture
NV) was introduced at a density of 8 prey mL)1, and their
density gradually increased until 14 DAH. On this day, live
Artemia was given along with frozen Artemia, which became,
form 17 DAH, the only prey offered until the end of the
experiment (31 DAH) (Ribeiro et al. 1999b).

Individual and pooled samples of fish larvae were collected
before feeding at mouth opening (2 DAH), prior to metamorphosis (5 DAH; 10 DAH), metamorphosis peak
(15 DAH) and after settling (31 DAH).
At all sampling times, 15 larvae (pooled until 10 DAH), in
duplicate, were collected from each tank in order to deter-

mine dry weight (dw). Pictures of 15 larvae were digitally
photographed every day and standard length and myotome
height measured using UTHSCSA Image Tools (University
of Texas).
Pooled samples of larvae with 250 mg wet weight were
collected to determine TH and deiodinase activity and
another pooled sample with 100 mg wet weight was collected
for iodine analyses. All samples collected were thoroughly
washed in distilled water, frozen in liquid nitrogen and stored
at )80 °C until further analyses. In order to determine the
volume of thyroid follicles, 20 specimens from each tank were
randomly collected at all sampling points, anesthetized with

0.1% phenoxyethanol and fixed in Bouin-Holland. Then they
were rinsed and washed in distilled water and preserved in
70% ethanol for later use. Samples of 100 mg wet weight of
feed were taken after enrichment for iodine analysis and kept
at )80 °C for further analysis.

Rotifers, Artemia and larvae were analyzed for total iodine
concentration using inductively coupled plasma mass spectroscopy, after alkaline decomposition as described by
(Julshamn et al. 2001). Total iodine was quantified using an
external standard curve prepared using the same matrix as
the actual samples, possible background was taken into
account by analyzing two blind samples at each run. Detection limit was £0.04 lg iodine g)1 sample.

Analysis of TH Larvae were stored in methanol at )20 °C
until extraction was carried out according to Einarsdo´ttir
et al. (2006). Approximately 250 mg larvae were pooled and
weighed and the hormones extracted by homogenizing in an
equal volume of ice-cold methanol. Homogenates were left
on ice for 30 min and then centrifuged (30 min, 3000 · g,
4 °C). After collecting the supernatant, the resulting pellet
was re-extracted in 0.5 volumes of methanol, and then the
supernatant was collected and added to the first supernatant,
and freeze-dried. The TH, T3 and T4 were analyzed by
radioimmunoassay (Einarsdo´ttir et al. 2006).
Outer-ring deiodinase (ORD) activity The assay was performed as described by Klaren et al. (2005) with some
modifications. The reaction was run without ÔcoldÕ
reverse T3 as in Hamre et al. (2008a). The ORD activity
was measured by incubation of larvae homogenates
containing 0.5 lg of protein for 1 h at 37 °C with 105 cpm

..............................................................................................

Aquaculture Nutrition 17; 248–257 Ó 2009 Blackwell Publishing Ltd


Stereological analysis of the thyroid glands Larvae were
dehydrated in an increasing gradient of ethanol and embedded in Technovit 7100 (Heraeus Kluzer GmbH, Wehrheim,
Germany). Three resin-embedded larvae from each tank were
sectioned at 2 lm thicknesses for 2 and 5 DAH, 3 lm for 10
and 15 DAH larvae and 5 lm for 31 DAH and stained with
toluidine blue. The area of thyroid follicles was measured
using a computer-aided stereology CAST 2 (Olympus, Ballerup, Denmark A/S 2000) according to (Sæle et al. 2003).
Volume was calculated based on the thickness of the slices
and the distance between the sections, measuring the area of
the follicles on a known number of sections.

All statistical analyses were preformed with Statistica software (Ver.7; Statsoft Inc., Tulsa, OK, USA). The Kolmogorov–Smirnov test was used to test for normal distribution
and LeveneÕs test was used to test for variance homogeneity.
Then data were subjected to a factorial ANOVA, to test
potential difference between two groups over a period of time.
Tukey test (or multiple comparison test) was used to test for
significant differences between group means in an analysis of
variance. Also, an Unequal N HSD (honestly significant
differences) post hoc test was conducted to test for significant
differences between group means in analysis of variance when
the number of N was unequal. Effects and differences were
considered significant at P < 0.05 for all tests.

..............................................................................................

Aquaculture Nutrition 17; 248–257 Ó 2009 Blackwell Publishing Ltd

In the first system (from 0 DAH until 17 DAH), survival
rates were 87.2 ± 3.3 (average ± SD) and 86.4.3 ± 2.5
(average ± SD) (P > 0.05) for control treated larvae and
iodine treated larvae, respectively. In the recirculation system
(from 17 DAH until 31 DAH), survival rates were
98.8 ± 0.36 (average ± SD) and 99.1 ± 0.52 (average ± SD) (P > 0.05) for control treated larvae and iodine
treated larvae, respectively.
Larvae from both treatments increased their weights significantly from 10 DAH until the end of the experiment.
From 10 DAH until 15 DAH there are no significant differences in larval dry weight between treatments. However,
after 15 DAH, the enrichment of live feed with iodine caused
a significant increase in growth (Fig. 2).
Data from standard length showed the same growth trend
as dry weight. No significant differences were observed until
11 DAH, where larvae from the control treatment exhibited
a larger standard length compared to larvae from iodine
treatment (5.6 ± 0.4 and 5.1 ± 0.3 mm, respectively,
P < 0.001). At 31 DAH, the larvae from the iodine treatment reached significantly higher average standard length
than larvae from control treatment (13.7 ± 0.9 and

5.0

d

4.5

c

4.0

3.5
Dry weight (mg)

of L-3,3¢,5¢ – [125I]-Triiodothyronine (NEX109; Perkin
Elmer, Boston, MA, USA) in 200 mM PBS (pH 7.0) with
4 mM EDTA (Sigma–Aldrich, Munich, Germany). Incubations were carried out in duplicate together with blanks,
containing no protein, to correct for non-enzymatic degradation of the tracer. The radiotracer was purified on a 10%
(w/v) Sephadex LH-20 (GE Healthcare Bio-Sciences AB,
Uppsala, Sweden) mini-column before use. Deiodination
products were analyzed after precipitation of protein-bound
iodothyronines with trichloroacetic acid (TCA; Merk). The
reaction was stopped by addition of ice-cold 100 lL 5%
(w/v) BSA (Bovine serum albumin, Sigma–Aldrich) and
500 lL 10% (w/v) TCA at 0 °C. After centrifugation
(15 min, 1500 g, 4 °C), 500 lL of the supernatant was
combined with 500 lL of 1.0 M HCl, and liberated iodide
was separated from the native iodothyronine with the use of
Sephadex LH-20 mini-column chromatography pre-equilibrated with 0.1 M HCl. 125I radioactivity was measured in a
Tri-Carb 1900 TR Liquid Scintillation counter (Packard,
Ramsey, MN, USA).

3.0
2.5
2.0
1.5

b

1.0
0.5

–0.5

a

a

0.0

2

5

a

b

a

10
15
Days after hatching (DAH)

31

Figure 2 Larval growth (mg ± SD) of Senegalese sole larvae, fed
either iodine-enriched live feed or control live feed, from 2 days after
hatch (DAH) to 31 DAH. ( ) Control treatment; ( ) Iodine treatment. Values are means of 15 individual larvae at 15 and 31 DAH
and of pooled 15 larvae at 5 and 10 DAH, taken from each tank
(n = 3). a,b,c,dMean values with unlike superscript letters are significantly different (P < 0.05 two-way ANOVA). 2 DAH not included in

the test.


Results showed that both rotifers and Artemia enriched with
iodine had significant higher levels of iodine than control
rotifers and Artemia (Table 1).
At 2 DAH, larvae from both treatments had the same
iodine level. The enrichment of live feed with iodine caused a
significant increase in larval whole body iodine content from
10 DAH until 31 DAH (Fig. 3). At 15 DAH, there was a
peak in the whole body iodine content of larvae from the
iodine treatment. Afterward iodine levels dropped to lower
values than observed at 10 DAH (Fig. 3).

T3 levels showed no differences for either treatment or different ages (Fig. 4).
Because of lack of replicates, statistical analyses could not
be performed on T4 (Fig. 5). However, the level of T4 was not
detected at 2 DAH and at 10 DAH in the control group. At
metamorphosis peak (15 DAH), there was an increase of T4

)1

Table 1 Iodine content of rotifers and Artemia (lg I g
wt
weight ± SD). Values are means of pooled live feed (100 mg wet
weight) from each enrichment tank (n = 2). Both iodine-enriched
rotifers and Artemia and control rotifers and Artemia were sampled
at 09 : 00 AM
Control

Amount of iodine in Senegalsese sole
larvae (µg g–1 wet weight)

0.7

b

bc

0.6
0.5

bd

0.4
0.3
0.2
0.1

a
a
a

a

a

0.0

2

10
15
Days after hatching (DAH)

31

Figure 3 Iodine concentration (lg g)1 wet weight ± SD) in Senegalese sole larvae, fed either iodine-enriched live feed or control live
feed, from 2 days after hatch (DAH) to 31 DAH. ( ) Control
treatment; ( ) Iodine treatment. Values are means of pooled larvae
(100 mg wet weight), taken from each tank (n = 3). a,b,cMean values
with unlike superscript letters are significantly different (P < 0.05
two-way ANOVA).

3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
–0.5

2

10
15
Days after hatching (DAH)

31

Figure 4 Levels of triiodothyronine (T3) (pg mg)1 dry weight ± SD)
in Senegalese sole larvae, fed either iodine-enriched live feed or
control live feed, from 2 days after hatch (DAH) to 31 DAH. ( )
Control treatment; ( ) Iodine treatment. Values are means of pooled
larvae (250 mg wet weight), taken from each tank (n = 3; except
10 DAH n = 2).

Iodine

Live feed

Mean

SD

Mean

SD

Rotifers
Artemia

2.51b
0.73a

0.43
0.10

47.86c
64.22d

0.46
8.11

SD, standard deviations.
Mean values in a within a row with unlike superscript letters
are significantly different (P < 0.05 one-way ANOVA).
a,b,c,d

0.8

T3 (pg mg–1 dry weight)

12.9 ± 0.8 mm, respectively, P < 0.001). Also, myotome
height showed the same tendency as standard length and dry
weight, where no differences were observed between the two
treatments until 11 DAH. At this point, larvae from control
treatment exhibited significantly higher myotome height,
compared to larvae from the iodine treatment (0.7 ± 0.1 and
0.5 ± 0.1 mm, respectively, P < 0.001). At 15 DAH, larvae
from the iodine treatment had significantly larger myotome
compared to control larvae (1.3 ± 0.1 and 1.2 ± 0.1 mm,
respectively, P < 0.003). From this day until the end of the
experiment, myotome height tended to be higher in larvae
form the iodine treatment, although with no significant difference.

in iodine treated larvae. T4 levels in that group decreased
rapidly from 15 to 31 DAH (Fig. 5).

Although no ORD activity was detected at 2 DAH, the
overall activity increased significantly over time. Results also
showed no significant differences in ORD activity between
treatments at 10 DAH. However, at 15 DAH the ORD

..............................................................................................

Aquaculture Nutrition 17; 248–257 Ó 2009 Blackwell Publishing Ltd


0.006
Volume of thyroid follicles epithelia (mm3)

c
0.005
0.004
0.003
0.002

b
ab

0.001

ab
0.000

a

a

–0.001

Figure 5 Levels of thyroxine (T4) (pg mg)1 dry weight ± SD) in
Senegalese sole larvae, fed either iodine-enriched live feed or control
live feed, from 2 days after hatch (DAH) to 31 DAH. ( ) Control
treatment; ( ) Iodine treatment. Values are means of pooled larvae
(250 mg wet weight), taken from each tank (10 DAH n = 1;
15 DAH n = 2; 31 DAH n = 1). N.D. – not detected – values
below the detection level.

5

ab

a

10
15
Days after hatching (DAH)

31

Figure 7 Volume of thyroid follicles epithelia (mm3 ± SD) in
Senegalese sole larvae, fed either iodine-enriched live feed or control
live feed, from 2 days after hatch (DAH) to 31 DAH. ( ) Control
treatment; ( ) Iodine treatment. Values are means of three larvae per
tank (n = 3). a,b,c,dMean values with unlike superscript letters are
significantly different (P < 0.05 two-way ANOVA).

0.0009

12

b
Deiodinase activity (fmol min µg–1)

10

b
b

c

0.0008
Volume of thyroid follicles colloid (mm3)

activity of iodine treated larvae was significantly higher than
larvae from the control treatment. At 31 DAH, the ORD
activity becomes higher, although no significant different is
observed between treatments (Fig. 6).

0.0007
0.0006

b

b

b

0.0005
0.0004
0.0003
0.0002

a

0.0001
0.0000

a

a

a

8

–0.0001

5
6

a
4

a

2

N.D.
2

10
15
Days after hatching (DAH)

31

Figure 8 Volume of thyroid follicles colloid (mm3 ± SD) in Senegalese sole larvae, fed either iodine-enriched live feed or control live
feed, from 2 days after hatch (DAH) to 31 DAH. ( ) Control
treatment; ( ) Iodine treatment. Values are means of three larvae per
tank (n = 3). a,b,cMean values with unlike superscript letters are
significantly different (P < 0.05 two-way ANOVA).

a

0

10
15
Days after hatching (DAH)

31

Figure 6 Outer-ring deiodinase activity (fmol iodide min lg)1 of
protein ± SD) in Senegalese sole larvae, fed either iodine-enriched
live feed or control live feed, from 2 days after hatch (DAH) to
31 DAH. ( ) Control treatment; ( ) Iodine treatment. Values are

means of pooled larvae (250 mg wet weight), taken form each tank
(n = 3). a,b,cMean values with unlike superscript letters are significantly different (P < 0.05 two-way ANOVA). N.D. – not detected –
values below the detection level.

..............................................................................................

Aquaculture Nutrition 17; 248–257 Ó 2009 Blackwell Publishing Ltd

No thyroid follicles were found in 2 DAH larvae for any of
the treatments (Figs 7 & 8). As the figures show, volumes of
both thyroid epithelia and colloid increased significantly as
the fish developed. From 5 DAH until 15 DAH there were no
significant differences between treatments. At the end of the
experiment (31 DAH), larvae from the control treatment
showed a volume of epithelia significantly higher than larvae
from the iodine treatment. Also, larvae from the iodine


(a)

(b)

(c)

treatment had significantly higher volume of thyroid colloid
than larvae from control treatment (Figs 7–9). The histology
of the thyroid follicles from 31 DAH showed that the larvae
from the control treatment had dense epithelia, characteristic
of thyroid hyperplasia and hypertrophy (goitre). These epithelial cells had lost the follicular type of morphology, and the
colloid was partially or completely depleted of thyroglobulin

compared to the follicles in larvae fed the iodine-enriched diet.
These follicles showed epithelial cells that were cuboidal and
colloids were replete with thyroglobulin (Fig. 9).

Results from the 31 -day trial indicate good growth performance and good survival rates of fish larvae from both
treatments when compared to other experiments conducted
with the same species (Can˜avate & Ferna´ndez-Dı´ az 1999;
Dinis et al. 1999).
At 31 DAH, the morphology of the follicles in fish fed the
control diet were representative of hyperplasia and hypertrophy (goitre) (Leatherland 1994) and typical of a hypothyroid condition (Leatherland et al. 1978; Moccia et al.
1981; Leatherland & Down 2001). In general, individual
variation in thyroid follicle morphology is normal and may
occur even within the same animal (Fig. 8) and depends on
pituitary stimulation of the thyroid gland (Leatherland
1994). The follicles from fish fed the iodine-enriched diet

Figure 9 Thyroid follicles in Senegalese
Sole. (a) Thyroid follicles (fo) from
control treatment at 31 days after hatch
(DAH). Thyroid tissue is very dense,
characteristic of thyroid hyperplasia and
hypertrophy (goitre); the proximity to
ventral aorta (va) is evident. Follicle
epithelial cells (ep – arrow) have lost the
follicular type of morphology and
colloid (co – arrow) is absent (magnification 400·); (b) Thyroid follicles (fo)
from control treatment at 15 DAH;
(c) Thyroid follicle (fo) from iodine
treatment at 31 DAH. In both figures (b
and c), the thyroid tissue is less dense

compared to (a). Follicles are adjacent
to the ventral aorta and bulbus arteriosus. Follicle epithelial cells are cuboidal
and colloids are replete with thyroglobulin (magnification 640·).

always appeared normal (Fig. 9) when compared to previous
descriptions from Senegalese sole larvae and juveniles
(Delgado et al. 2006; Klaren et al. 2008).
It is known that iodine deficiency can create hypothyroidism and goitre in fish (Sonstegard & Leatherland 1976;
Moccia et al. 1981; Leatherland 1994). Both diet and larvae
from control treatment had significantly lower levels of
iodine compared to the iodine group, and it is therefore likely
that iodine deficiency was the main cause of the observed
hypothyroid condition. This condition is generally associated
with low T4 and T3 levels, and high ORD activity (Burel
et al. 2000). The expected low levels of TH in the control
group (Leatherland & Down 2001; Van Der Geyten et al.
2001) could not be completely documented because of the
low number of samples and large variation of data (Figs 4 &
5). Nonetheless, the T4 levels at different ages in the iodine
treated fish displays a similar pattern as demonstrated in an
earlier study (Klaren et al. 2008).
The higher ORD activity of iodine-treated larvae on
15 DAH may reflect a better iodine status. The increase in
ORD activity in the control groups between 15 and 31 DAH
is twice as large as the increase seen in fish fed iodine, which
may be a response to the hypothyroid condition seen in
histology sections from larvae fed control diet on 31 DAH.
The overall increase in the activity of ORD until 31 DAH,
in both groups (Fig. 6), can be related to the development
in general. To our knowledge, larval ORD activity from

..............................................................................................

Aquaculture Nutrition 17; 248–257 Ó 2009 Blackwell Publishing Ltd


Senegalese sole is not published, but from studies with
Atlantic halibut, it has been shown that the activity rises
through the larval period (Moren et al. unpublished data).
It is known that fish growth is partially regulated by TH
(Leatherland 1994), and it is documented that hypothyroidism can lead to poor growth (Boyages et al. 1989; Van Der
Geyten et al. 2001). The larvae fed the control diet were
smaller than those fed the iodine-enriched diet. Consequently, because the only difference between the two groups
was the iodine level in the diet, it seems likely that the extra
iodine prevented thyroid hyperplasia (goitre), and therefore
growth was not compromised.
Hyperplasia, as other types of thyroid lesions, has been
described in bony and cartilaginous fishes (Leatherland &
Down 2001), and a number of causes have been determined
(Moccia et al. 1981; Burel et al. 2000; Sherrill et al. 2004). In
our experiment, the thyroid follicle hyperplasia became
evident only after the larvae had been transferred to a recirculation system, implicating that something in that system
may have triggered or strengthened their hypothyroid status.
Some studies show that bacteria are introduced in fish
tanks by live feed such as frozen Artemia (Olsen et al. 1999;
Olafsen 2001). Several disinfectants, used in hatcheries, are
iodine-based (Smail et al. 2004; Verner–Jeffreys et al. 2008),
which could lead to the conclusion that the iodine in the
enriched Artemia worked as a disinfectant and not as a
nutritional factor. But the largest amount of frozen Artemia

given to fish larvae in each tank at one time was 0.7 g (wet
wt), which means that the maximum amount of I) added to
the water would be 2.3 lg, assuming 100% leaking of I) from
the Artemia, which is far too little to give an antiseptic effect
(Katharios et al. 2007). Further, the water flow in the tanks
was 1.4 L min)1 meaning that the dissolved I) from the
Artemia would be flushed away within 15 min.
Recirculation systems, as in the case of our experiment,
usually use protein skimmer and ozone contact chambers
with a powerful oxidation–reduction potential to minimize
the level of potential pathogens (Sherrill et al. 2004). Previous studies show that an increasing exposure to ozone alters
the concentration of iodide (I)), dissolves organic iodine by
oxidation and converts these into iodate (IO3)) (Bichsel &
Von Gunten 1999; Sherrill et al. 2004). Knowing that iodate
is probably not biologically available and that the iodine
available from the water decreases or is absent in such
a recirculation system, this might lead to thyroid lesions as
seen before (Sherrill et al. 2004). A study by Moren et al.
(2008) with Atlantic halibut larvae showed that their capacity
to absorb iodide exceeds the sea water concentrations of
iodide by at least 200-folds. The authors suggested that this

..............................................................................................

Aquaculture Nutrition 17; 248–257 Ó 2009 Blackwell Publishing Ltd

might be because of the larvae having an iodine uptake
capacity related to feed rather than seawater. Hence, the
consequences of rearing larvae in a recirculation system
with a diet deficient of iodine might be more severe than

seen in a flow-through system. Atlantic halibut, reared in a
flow-through system and fed iodine-enriched Artemia had
indications of better thyroid follicle status (lower epithelial
volume) when compared to larvae fed control Artemia, but no
goitre was observed in this study (Moren et al. unpublished
data). Nevertheless, the fact that larvae from the control
treatment failed to produce a peak of T4 at 15 DAH prior to
transfer to the recirculation system, while larvae from the
iodine treated group did suggests that iodine might be a necessary supplement even when fish are reared in an open
system. The present study emphasizes that marine fish larvae
cannot sustain their iodine requirement only from water in a
recirculation system when fed either rotifers or Artemia.

At 31 DAH, larvae from the control treatment suffered from
hyperplasia of the thyroid follicles (goitre), whereas iodine
treated larvae did not. Lower growth rate in fish larvae from
the control treatment was probably a consequence of the
hyperplasia. Iodine enrichment prevented Senegalese sole
larvae form developing hyperplasia. The enrichment of live
feed with iodine is crucial for fish reared in a recirculation
system using ozone injection, as it seems to sustain normal
larval development, although the iodine requirement is still
to be determined.

We would like to thank Professor Deborah Power for invaluable help with thyroid hormone extractions and RIAs.
We are indebted to the highly skilled technicians at CCMAR
and NIFES. This work was supported by the DIGFISH
project, POCI/CVT/58790/2004 (FCT, Portugal). Ribeiro
A.R.A and Ribeiro L benefit from grants SFRH/BD/24803/
2005 and SFRH/BPD/7148/2001 (FCT, Portugal), respectively. (Correction added on 29 March 2010, after first online

publication: An acknowledgement to Professor Deborah
Power was added.)

Bichsel, Y. & Von Gunten, U. (1999) Determination of iodide and
iodate by ion chromatography with postcolumn reaction and UV/
visible detection. Anal. Chem., 71, 34–38.


Boyages, S.C., Collins, J.K., Maberly, G.F., Jupp, J.J., Morris, J. &
Eastman, C.J. (1989) Iodine deficiency impairs intellectual and
neuromotor development in apparently-normal persons. A study
of rural inhabitants of north-central China. Med. J. Aust., 150,
676–682.
Burel, C., Boujard, T., Escaffre, M., Kaushik, S.J., Boeuf, G., Mol,
K., Van Der Geyten, S. & Kuhn, E.R. (2000) Dietary lowglucosinolate rapeseed meal affects thyroid status and nutrient
utilization in rainbow trout (Oncorhynchus mykiss). Br. J. Nutr.,
83, 653–664.
Can˜avate, J.P. & Ferna´ndez-Dı´ az, C. (1999) Influence of co-feeding
larvae with live and inert diets on weaning the sole Solea senegalensis onto commercial dry feeds. Aquacuture, 174, 255–263.
Conceic¸a˜o, L.E.C., Ribeiro, L., Engrola, S., Araga˜o, C., Morais, S.,
Lacuisse, M., Soares, F. & Dinis, M.T. (2007) Nutritional physiology during development of Senegalese sole (Solea senegalensis).
Aquaculture, 268, 64–81.
Delgado, J.B.O., Ruane, N.M., Pousa˜o-Ferreira, P., Dinis, M.T. &
Sarasquete, C. (2006) Thyroid gland development in Senegalese
sole (Solea senegalensis Kaup 1858) during early life stages: a
histochemical and immunohistochemical approach. Aquaculture,
260, 346–356.
Dinis, M.T., Ribeiro, L., Soares, F. & Sarasquete, C. (1999) A review
on the cultivation potential of Solea senegalensis in Spain and in
Portugal. Aquaculture, 176, 27–38.

Eales, J.G. & Brown, S.B. (1993) Measurement and regulation of
thyroidal status in teleost fish. Rev. Fish Biol. Fish., 3, 299–347.
Einarsdo´ttir, I., Silva, N., Power, D., Sma´rado´ttir, H. & Bjo¨rnsson,
B. (2006) Thyroid and pituitary gland development from hatching
through metamorphosis of a teleost flatfish, the Atlantic halibut.
Anat. Embryol., 211, 47–60.
Engrola, S., Conceic¸a˜o, L.E.C., Gavaia, P.J., Cancela, M.L. & Dinis,
M.T. (2005) Effects of pre-weaning feeding frequency on growth,
survival, and deformation of Senegalese sole, Solea senegalensis
(Kaup, 1858). Isr. J. Aquac.-Bamidgeh., 57, 10–18.
Hamre, K., Moren, M., Solbakken, J., Opstad, I. & Pittman, K.
(2005) The impact of nutrition on metamorphosis in Atlantic
halibut (Hippoglossus hippoglossus L.). Aquaculture, 250, 555–565.
Hamre, K., Mollan, T.A., Sæle, Ø. & Erstad, B. (2008a) Rotifers
enriched with iodine and selenium increase survival in Atlantic cod
(Gadus morhua) larvae. Aquaculture, 284, 190–195.
Hamre, K., Srivastava, A., Ronnestad, I., Mangor-Jensen, A. &
Stoss, J. (2008b) Several micronutrients in the rotifer Brachionus sp
may not fulfill the nutritional requirements of marine fish larvae.
Aquac. Nutr., 14, 51–60.
Imsland, A.K., Foss, A., Conceica˜o, L.E.C., Dinis, M.T., Delbare,
D., Schram, E., Kamstra, A., Rema, P. & White, P. (2003) A
review of the culture potential of Solea solea and S. senegalensis.
Rev. Fish Biol. Fish., 13, 379–408.
Julshamn, K., Dahl, L. & Eckhoff, K. (2001) Determination of
iodine in seafood by inductively coupled plasma/mass spectrometry. J. AOAC Int., 84, 1976–1982.
Katharios, P., Agathaggelou, A., Paraskevopoulos, S. & Mylonas,
C.C. (2007) Comparison of iodine and glutaraldehyde as surface
disinfectants for red porgy (Pagrus pagrus) and white sea bream
(Diplodus sargus sargus) eggs. Aquac. Res., 38, 527–536.

Klaren, P.H.M., Haasdijk, R., Metz, J.R., Nitsch, L.M.C., Darras,
V.M., Van der Geyten, S. & Flik, G. (2005) Characterization of an
iodothyronine 5¢-deiodinase in gilthead Seabream (Sparus auratus)
that is inhibited by dithiothreitol. Endocrinology, 146, 5621–5630.
Klaren, P.H.M., Wunderink, Y.S., Yufera, M., Mancera, J.M. &
Flik, G. (2008) The thyroid gland and thyroid hormones in

Senegalese sole (Solea senegalensis) during early development and
metamorphosis. Gen. Comp. Endocrinol., 155, 686–694.
Leatherland, J.F. (1994) Reflections on the thyroidology of fishes:
from molecules to humankind. Guelph Ichthyol. Rev., 2, 1–67.
Leatherland, J.F. & Down, N.E. (2001) Tumours and related lesions
of the endocrine system of bony and cartilaginous fishes. Fish
Fish., 2, 59–77.
Leatherland, J.F., Moccia, R. & Sonstegard, R. (1978) Ultrastructure of the thyroid gland in goitered coho salmon (Oncorhynchus
kisutch). Cancer Res., 38, 149–158.
Martinez, I., Moyano, F.J., Fernandez-Diaz, C. & Yufera, M.
(1999) Digestive enzyme activity during larval development of
the Senegal sole (Solea senegalensis). Fish Physiol. Biochem., 21,
317–323.
Moccia, R.D., Leatherland, J.F. & Sonstegard, R.A. (1981) Quantitative interlake comparison of thyroid pathology in great lakes
coho (Oncorhynchus kisutch) and chinook (Oncorhynchus
tschawytscha) salmon. Cancer Res., 41, 2200–2210.
Moren, M., Opstad, I., Van Der Meeren, T. & Hamre, K. (2006)
Iodine enrichment of Artemia and enhanced levels of iodine in
Atlantic halibut larvae (Hippoglossus hippoglossus L.) fed the
enriched Artemia. Aquac. Nutr., 12, 97–102.
Moren, M., Sloth, J.J. & Hamre, K. (2008) Uptake of iodide from
water in Atlantic halibut larvae (Hippoglossus hippoglossus L.).
Aquaculture, 285, 174–178.

Olafsen, J.A. (2001) Interactions between fish larvae and bacteria in
marine aquaculture. Aquaculture, 200, 223–247.
Olsen, A.I., Attramadal, Y., Jensen, A. & Olsen, Y. (1999) Influence
of size and nutritional value of Artemia franciscana on growth and
quality of halibut larvae (Hippoglossus hippoglossus) during the
live feed period. Aquaculture, 179, 475–487.
Pittman, K., Jelmert, A., Næss, T., Harboe, T. & Watanabe, K.
(1998) Plasticity of viable postmetamorphic forms of farmed
Atlantic halibut, Hippoglossus hippoglossus L. Aquac. Res., 29,
949–954.
Power, D.M., Llewellyn, L., Faustino, M., Nowell, M.A., Bjornsson,
B.T., Einarsdottir, I.E., Canario, A.V.M. & Sweeney, G.E. (2001)
Thyroid hormones in growth and development of fish. Comp.
Biochem. Physiol. C, 130, 447–459.
Ribeiro, L., Sarasquete, C. & Dinis, M.T. (1999a) Histological and
histochemical development of the digestive system of Solea senegalensis (Kaup, 1858) larvae. Aquaculture, 171, 293–308.
Ribeiro, L., Zambonino-Infante, J.L., Cahu, C. & Dinis, M.T.
(1999b) Development of digestive enzymes in larvae of Solea
senegalensis, Kaup 1858. Aquaculture, 179, 465–473.
Ribeiro, A.R.A., Moren, M., Ribeiro, L., Dinis, M.T. & Hamre, K.
(2007) Iodine Enrichment of Rotifers Brachionus plicatilis. Aquaculture Europe 07, October 2007, Istanbul, Turkey. Poster, 300
Rocha, J.R., Ribeiro, L., Calado, R. & Dinis, M.T. (2008) Does the
presence of microalgae influence fish larvae prey capture? Aquac.
Res., 39, 362–369.
Sæle, O., Solbakken, J.S., Watanabe, K., Hamre, K. & Pittman, K.
(2003) The effect of diet on ossification and eye migration in
Atlantic halibut larvae (Hippoglossus hippoglossus L.). Aquaculture, 220, 683–696.
Schreiber, A.M. & Specker, J.L. (1998) Metamorphosis in the summer flounder (Paralichthys dentatus): stage-specific developmental
response to altered thyroid status. Gen. Comp. Endocrinol., 111,
156–166.

Sherrill, J., Whitaker, B.R. & Wong, G.T.F. (2004) Effects of
ozonation on the speciation of dissolved iodine in artificial seawater. J. Zoo Wildl. Med., 35, 347–355.

..............................................................................................

Aquaculture Nutrition 17; 248–257 Ó 2009 Blackwell Publishing Ltd


Smail, D.A., Grant, R., Simpson, D., Bain, N. & Hastings, T.S.
(2004) Disinfectants against cultured Infectious Salmon Anaemia
(ISA) virus: the virucidal effect of three iodophors, chloramine T,
chlorine dioxide and peracetic acid/hydrogen peroxide/acetic acid
mixture. Aquaculture, 240, 29–38.
Solbakken, J.S., Norberg, B., Watanabe, K. & Pittman, K.
(1999) Thyroxine as a mediator of metamorphosis of Atlantic
Halibut, Hippoglossus Hippoglossus. Environ. Biol. Fishes, 56,
53–65.
Solbakken, J.S., Berntessen, M.H.G., Norberg, B., Pittman, K. &
Hamre, K. (2002) Different iodine and thyroid hormone levels
between Atlantic halibut larvae fed wild zooplankton or Artemia
from first exogenous feeding until post metamorphosis. J. Fish
Biol., 61, 1345–1362.
Sonstegard, R. & Leatherland, J.F. (1976) The epizootiology and
pathogenesis of thyroid hyperplasia in coho salmon (Oncorhynchus
kisutch) in Lake Ontario. Cancer Res., 36, 4467–4475.
Tanaka, M., Kawai, S., Seikai, T. & Burke, J.S. (1996) Development
of the digestive organ system in japanese flounder in relation to
metamorphosis and settlement. Mar. Freshw. Behav. Physiol., 28,
19–31.

..............................................................................................

Aquaculture Nutrition 17; 248–257 Ó 2009 Blackwell Publishing Ltd

Van Der Geyten, S., Toguyeni, A., Baroiller, J.F., Fauconneau, B.,
Fostier, A., Sanders, J.P., Visser, T.J., Ku¨hn, E.R. & Darras, V.M.
(2001) Hypothyroidism induces type I iodothyronine deiodinase
expression in tilapia liver. Gen. Comp. Endocrinol., 124, 333–342.
Van Stappen, G. (1996) Artemia. In: Manual on the Production and
Use of Live Food for Aquaculture (Lavens, P. & Sorgeloos, P. eds),
1–295. FAO, Rome.
Verner–Jeffreys, D.W., Joiner, C.L., Bagwell, N.J., Reese, R.A.,
Husby, A. & Dixon, P.F. (2008) Development of bactericidal and
virucidal testing standards for aquaculture disinfectants. Aquaculture, 286, 190–197.
Watanabe, T. (1993) Importance of docosahexaenoic acid in marine
larval fish. J. World Aquac. Soc., 24, 152–161.
Whyte, J.N.C., Clarke, W.C., Ginther, N.G., Jensen, J.O.T. &
Townsend, L.D. (1994) Influence of composition of Brachionus
plicatilis and Artemia on growth of larval sablefish (Anoplopoma
fimbria Pallas). Aquaculture, 119, 47–61.
Yu´fera, M., Parra, G., Santiago, R. & Carrascosa, M. (1999)
Growth, carbon, nitrogen and caloric content of Solea senegalensis
(Pisces: Soleidae) from egg fertilization to metamorphosis. Mar.
Biol., 134, 43–49.


Aquaculture Nutrition
2011 17; 258–266

doi: 10.1111/j.1365-2095.2009.00747.x

..............................................................................................

1
1

2

2

Department of Aquaculture, National Kaohsiung Marine University, Kaohsiung, Taiwan;
National Taiwan Ocean University, Keelung, Taiwan

This study aimed to find out whether dietary carotenoid
(CD) supplement could influence the resistance of characins
(Hyphessobrycon eques Steindachner) to ammonia stress.
Two types of CD and its combination [astaxanthin (AX),
b-carotene (BC), 1 : 1 combination of AX and BC (MX)] at
three concentrations (10, 20 and 40 mg kg)1) were used
resulting in nine pigmented diets. No differences in growth
and survival of the fish among treatments were found after
8-week rearing. Experimental and control fish were then
exposed to 15 mg total ammonia nitrogen L)1 (stress group)
and 0.15 mg total ammonia nitrogen L)1 (normal group) for
72 h, and their blood was withdrawn. No mortality resulted
under such TAN concentrations. Serum total antioxidant
status (TAS), serum antioxidant enzymes [superoxide
dismutase (SOD), glutathione peroxidases (GPx)] and serum
transaminases [alanine aminotransferase (ALT), aspartate
aminotransferase (AST)] were chosen as indices of fish antioxidant capacity or stress resistance. SOD, GPx and AST

were affected by the interactions of dietary CD and ammonia
stress. The activities of TAS, SOD, GPx and AST increased
under the stress. Dietary CD reduced serum SOD, GPx, ALT
and AST activities. In conclusion, dietary CD increased the
resistance of characins to ammonia stress.
KEY WORDS:

ammonia stress, antioxidant capacity, astaxanthin, b-carotene, Hyphessobrycon eques, superoxide
dismutase
Received 27 August 2009, accepted 23 October 2009
Correspondence: Y.-H. Chien, Department of Aquaculture, National Taiwan Ocean University, Keelung 202, Taiwan. E-mail: yhchien@mail.
ntou.edu.tw

2

Department of Aquaculture,

Various synthetic and natural carotenoids (CD) have been
used as dietary supplement to improve pigmentation of
aquaculture animals (Storebakken & No 1992; Gouveia et al.
2003; Chien & Shiau 2005; Kalinowski et al. 2005) for higher
market value. On the other hand, dietary CD were also found
to increase cultured animalsÕ resistance to various stressors.
Dietary astaxanthin (AX), a naturally occurring CD pigment,
supplementation in penaeid postlarvae not only increased
body AX, but also enhanced the resistance to hypoxia (Chien
et al. 1999), salinity (Darachai et al. 1998; Merchie et al. 1998;
Chien et al. 2003), temperature (Chien et al. 2003), ammonia
(Pan et al. 2003a) and pathological (Pan et al. 2003b) stressors. The reason for the development of resistance to stress
from CD can be attributed to their antioxidant properties

(Torrissen & Christiansen 1995; Shimidzu et al. 1996). This is
to inactivate the free radicals produced from normal cellular
activity and various stresses so that the oxidative damage is
eliminated (Halliwell & Gutteridge 1989; Chew 1995).
Various types of dietary CD can affect deposition and
conversion of body CD in fish, their pigmentation (Torrissen
et al. 1989; Pan & Chien 2009) and possibly their antioxidant
capacity and stress resistance. b-carotene (BC) is recognized
as a lipid antioxidant, i.e. a free radical trap and quencher of
singlet oxygen (Bohm et al. 1997). AX contains a long conjugated double bond system with relatively unstable electron
orbitals; it may scavenge oxygen radicals in cells (Stanier
et al. 1971). The antioxidant activity of AX was found to be
approximately 10· stronger than BC (Shimidzu et al. 1996).
Besides type of dietary CD, various stresses can also result in
different responses in antioxidant capacity. It was shown that
the activities of hemolymph antioxidation enzymes of juvenile
tiger prawn varied not only with dietary AX types but also

..............................................................................................

Ó 2010 Blackwell Publishing Ltd


with the stressors including temperature and salinity drop
(Chien et al. 2003) and ammonia exposure (Pan et al. 2003b).
Ammonia is the most common toxicant in culture (Colt &
Armstrong 1981) and live-transportation systems (Taylor &
Solomon 1979). It comes from excretion of aquatic animals as
the end product of protein metabolism (Walsh & Wright 1995)
and mineralization of organic nitrogen in faeces, uneaten feed

and other organic matters (Avnimelech & Ritvo 2003). Ammonia can be toxic at low concentrations in the aquatic environment. Unionized ammonia concentration of 0.5 mg L)1
can be harmful to finfish and crustaceans (Van Rijn et al. 1990;
Tomasso 1994; Wasielesky et al. 1994). Short-term exposure
of fish and crustacean to high concentrations of ammonia
causes increased gill ventilation, hyperexcitability, loss of
equilibrium, convulsions and then death (Thurston et al. 1981;
Maltby 1995). Such stress also resulted in changes in haemolymph total antioxidant status (TAS), superoxide dismutase
(SOD), aspartate aminotransferase (AST) and alanine aminotransferase (ALT) in prawn (Pan et al. 2003a) and in AST and
ALT activity in fish (Jeney et al. 1992; Vedel et al. 1998).
Characins (Hyphessobrycon eques) is one of the most
important cultured and exported ornamental fishes in Taiwan. Routine feeding of CD-supplemented diets to this fish
in grow-out farms or aquaria is a common practice. This
study was undertaken to find out whether synthetic AX and/
or BC at various dietary concentrations would influence the
antioxidant capacity of characins, and concomitantly, their
resistance to high ammonia stress.

Control diet was composed of white fishmeal 500 g kg)1,
wheat flour 150 g kg)1, dextrin 270 g kg)1, fish oil 30 g kg)1,

vitamin mix 20 g kg)1 and mineral mix 30 g kg)1. Diets supplemented with CD have the same composition as the control
diet (except for dextrin which was adjusted depending on CD
levels used) but supplemented with either synthetic AX (8%
AX) or BC (10%BC) or both. Water was added to the ingredients to form a dough, which was then extruded through a 2mm-diameter die press. The extruded feed was air dried in the
dark to prevent the degradation of CD. The feed was then
crushed, sieved to attain particle size of 0.9–1.2 mm and stored
at )20 °C to avoid oxidation of the CD. There were nine CD
diets composed of 3 · 3 factorial combinations of CD type
[AX, BC and 1 : 1 mixture of AX and BC (MX)] and CD
concentrations (10, 20, and 40 mg kg)1). Proximate analyses

of these diets are listed in Table 1.

Experimental fish were bought from an ornamental fish farm.
During acclimatization in the laboratory in a 0.5-ton tank, fish
were fed control diet for 2 weeks to equalize their body CD
content. Fish were then transferred to 30 aquaria
(44 cm · 33 cm · 21.5 cm) to receive their respective treatments (three replicates per treatment) at stocking density of 30
fish aquarium)1. Fish size was 0.41 ± 0.09 g. Culture water
was passed through a 1-lm filter and sterilized by ultraviolet
light to eliminate microalgae, a possible source of CD. Moreover, all aquaria were covered with black screen to discourage
algal growth for the same precaution. Fish were fed twice daily
at 0800 and 1500 at 5% body weight. Dissolved oxygen (DO)
was maintained at 6–7 mg L)1 by constant aeration, temperature at 26–28 °C, pH of 7.5–8 and NH3 of 0.1–0.2 mg L)1.
Faeces and uneaten feeds were siphoned out daily, and onethird of the water was exchanged. The fish were reared for
8 weeks. No mortality was observed throughout the experiment. Final overall average fish size was 0.89 ± 0.18 g.

Table 1 Proximate analysis of the experimental diets
Carotenoid type

None (C0)

Astaxanthin (AX)

Carotenoid concentration(mg kg)1)

0

10

20

40

10

20

40

10

20

40

Diet notation

C0

AX-10

AX-20

AX-40

BC-10

BC-20

BC-40

MX-10

MX-20

MX-40

Proximate analysis
Crude protein (g kg)1)
Crude fat (g kg)1)
Ash (g kg)1)
Moisture (g kg)1)
NFE + CF1 (g kg)1)
Astaxanthin (mg kg)1)
b-carotene(mg kg)1)

303.5
54
94.8
132.8
409.3
16
22

307.6
56.8
92.8
134.7
398.5
108.6

22

304.4
54.4
93.5
132.0
397.5
225.5
22

308.1
55.4
92.6
134.3
402.5
416.7
22

303.8
55.8
93.4
135.0
399.5
16
120.4

304.4
57.4
93.5
134.7

405.1
16
234.2

304.5
58.0
92.7
134.0
401.4
16
422.2

305.6
52.3
93.4
134.7
403.0
48
63

301.2
50.4
92.5
133.0
404.7
12.7
10.8

305.8
54.7

93.6
134.7
403.5
217.4
20.8

1

Nitrogen-free extracts and crude fibre.

..............................................................................................

Aquaculture Nutrition 17; 258–266 Ó 2010 Blackwell Publishing Ltd

b-carotene (BC)

1/2 AX + 1/2 BC (MX)