Aquaculture nutrition, tập 18, số 1, 2012

Aquaculture Nutrition
doi: 10.1111/j.1365-2095.2011.00866.x

2012 18; 1–11

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

1

2

3

1

4

5
1

Aquaculture Division, CIFE, Versova, Mumbai, India; 2 National Institute of Basic Biology, NIBB, Okazaki, Japan; 3 Riverine
Fisheries Ecology Division, CIFRI, Vadodara, Gujarat, India; 4 Aquatic Environment and Health Management Division,
CIFE, Versova, Mumbai, India; 5 Fish Nutrition and Physiology Division, CIFA, Bhubaneswar, India

Six iso-nitrogenous (350 g protein kg)1) and iso-caloric
(4100 kcal kg)1) diets with or without probiotics supplementation namely T1 (Basal feed (BF) without probiotics;
control), T2 (BF + Bacillus subtilis and Lactococcus lactis),
T3 (BF + L. lactis and Saccharomyces cerevisiae), T4
(BF + B. subtilis and S. cerevisiae), T5 (BF + B. subtilis,
L. lactis and S. cerevisiae) and T6 (BF + heat-killed bacteria
of B. subtilis, L. lactis and S. cerevisiae) were fed to Labeo

rohita fingerlings (6.0 ± 0.06 g) for 60 days in triplicate
tanks (30 fish per tank). In all probiotic-supplemented diets,
the probiotic concentration was maintained at 1011 cfu kg)1
feed. After 60 days of culture, the fish fed combination of
three probiotics at equal proportion (T5) had higher
(P < 0.05) growth, protein efficiency ratio, nutrient retention
and digestibility and lower (P > 0.05) feed conversion ratio
over other treatment groups. Total heterotrophic bacterial
population in intestine was drastically reduced on 15th and
30th days of sampling than the initial value (0 day of sampling) for T3, T4 and T5 groups. Except T6, the gut colonization of respective probiotics, which were supplemented
through the diets, was also increased up to 30 days of culture
of fish and thereafter remained constant.
KEY WORDS:

Bacillus, digestibility, growth, Labeo rohita,
Lactococcus, nutrient, probiotics, Saccharomyces

Received 26 August 2010, accepted 7 February 2011
Correspondence: Kedar Nath Mohanta, Fish Nutrition and Physiology
Division, Central Institute of Freshwater Aquaculture, Bhubaneswar
751002, India. E-mail:

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

Ó 2011 Blackwell Publishing Ltd

Sipra Mohapatra, Aquaculture Division, Central Institute of Fisheries
Education, Versova, Mumbai 400061, India. E-mail: mohapatra_
sipra@rediffmail.com

With ever increasing demand for animal protein, aquaculture is emerging as one of the most viable and promising
enterprises for providing nutritional and food security to
the human. But, most of the intensive aquaculture farms
have been facing major hindrance because of disease outbreak, poor growth and low survival of fish. To mitigate
these problems, fish vaccine and probiotics have been used
in aquaculture practices. The use of probiotics in aquaculture has not only resulted the reduction of use of
harmful antimicrobial compounds, particularly the antibiotics, but also improved the appetite and/or biogrowth
performance of the farmed species in an eco-friendly and
sustainable manner (Gatesoupe 1999; Naik et al. 1999;
Robertson et al. 2000; Wang et al. 2005). One obvious
reason for the use of probiotics is that it can be supplemented even at larval and early fry stages (Muroga et al.
1987). Several reports suggest that probiotics supplementation can reduce the cost of culture by improving the
growth and feed utilization efficiency of fish (Swain et al.
1996; Bogut et al. 1998; Ghosh et al. 2003; Carnevali et al.
2006; Wang & Xu 2006; Mazurkiewicz et al. 2007;
Kesarcodi-Watson et al. 2008). Among the different
probiotics that are being used in aquaculture, lactic acid
bacteria (LAB) is found to be the most prominent one
as it is a part of the natural intestinal microflora of a
healthy fish (Noh et al. 1994; Brunt & Austin 2005;


Va¢zquez-Jua¢rez et al. 2005; Nayak et al. 2007; Wang
2007, 2011; Yin et al. 2007; Ramakrishnan et al. 2008). It
is well known that LAB often produces bacteriocins, which
may inhibit the growth of Gram-negative fish pathogens.
In addition, the use of other microbial groups such as
Lactococcus (Hagi et al. 2004; Sugita et al. 2009), Bacillus
sp. (Kumar et al. 2006, 2008), Lactobacillus (Ramakrishnan
et al. 2008) and Saccharomyces cerevisiae (Pal et al. 2007;

Ramakrishnan et al. 2008) as probiotics has been reported
in carp culture. Majority of these probiotics are nonpathogenic and non-toxic and can survive in the gut and
remain potent for long period under storage and field
condition (Ramakrishnan et al. 2008). Studies carried out
by Mohanty et al. (1996) indicated that a combination of
bacteria and yeasts as probiotics resulted higher survival
and better body weight gain and nutrient utilization in
catla (Catla catla). Yanbo & Zirong (2006) reported that
the survivability, weight gain and nutrient utilization of
fish fed probiotic-supplemented diet depend on the type of
probiotics used.
Although several authors have reported that the use of
probiotics has great potential for higher fish production
especially in carp culture, in most of their studies, the
probiotic organisms were used as a single species, either
through the diet or through culture environment (Kumar
et al. 2006). The literature on the use of combinations of
two or more probiotic species at the same time in the diet
or culture environment and their effect on growth, nutrient
utilization and gut microbial population of fish are very
limited. Therefore, in this study, an attempt has been made
to use the three probiotic microorganisms, namely B. subtilis, Lactococcus lactis and S. cerevisiae at different combinations in the diet of Labeo rohita and to evaluate the
growth, nutrient utilization, digestive enzymes activities
and gastrointestinal colonization of the supplemented
probiotics along with the total heterotrophic bacterial
(THB) population in the gut.

Three thousand healthy fingerlings of L. rohita (average
weight; 4.54 g) procured from Palghar fish farm, Maharashtra,
India, were transported to the wet laboratory of Central

Institute of Fisheries Education (CIFE), Mumbai. As a prophylactic measure, the fish were given a salt treatment (1%)
for 5 min and then acclimatized to the laboratory conditions

for 15 days using five 500 L capacity flow-through fibre-reinforced plastic tanks with the provision of continuous aeration.
During acclimatization, the fish were fed with a fish meal-based
formulated diet (350 g protein kg)1 diet and 17.2 MJ kg)1
dietary gross energy) other than the experimental diets. Five
hundred and forty uniform-sized healthy fish (average weight;
6.00 ± 0.06 g) were equally distributed in six dietary treatment groups [T1 (Basal feed without probiotics), T2 (Basal
feed + B. subtilis and L. lactis), T3 (Basal feed + L. lactis
and S. cerevisiae), T4 (Basal feed + B. subtilis and S. cerevisiae), T5 (Basal feed + B. subtilis, L. lactis and S. cerevisiae)
and T6 (Basal feed + heat-killed bacteria of B. subtilis,
L. lactis and S. cerevisiae)] with three replicates each (stocking
density of 30 fish per tank in 300 L of rearing water), following
a completely randomized design. The 500 L capacity flowthrough fibre-reinforced tanks (300 L water) with a flow rate of
0.5 L min)1 were used for rearing the fish. Seasoned groundwater was used for rearing the fish. The natural light cycle was
12 h light/12 h darkness (12 L : 12D) for the entire experimental period.

Pure strains of B. subtilis, L. lactis and S. cerevisiae were
procured from Microbial Type Culture Collection and Gene
Bank, Chandigarh, India, and was maintained at 4 °C in the
laboratory. Subsequently, the microbes were inoculated into
test tube containing brain heart infusion (BHI), de Man
Rogosa and Sharpe (MRS) and yeast extract peptone dextrose (YEPD) broths (Himedia) for B. subtilis, L. lactis and
S. cerevisiae, respectively, and kept in incubator for 24 h at
30 °C. After that, a loopful of microbial culture was streaked
on the respective agar media. The colonies were confirmed as
pure isolates of B. subtilis, L. lactis and S. cerevisiae by
performing the essential biochemical tests, and the cultures
were used for mass culture for subsequent use in the experiment.

For mass culture, freshly grown pure inocula of
B. subtilis, L. lactis and S. cerevisiae were added to 100 mL
of BHI, MRS and YEPD medium, respectively, in a 500mL conical flask and incubated at 30 °C for 24 h in a
shaking incubator. The cultures were centrifuged at 800 g
for 15 min at 4 °C. The supernatant was discarded, while
the pellets were resuspended in phosphate buffer saline (pH
7.2). The microbial pellets were washed and centrifuged
similarly and then mixed in phosphate buffer saline at
different concentrations as required and added to 100 g of
feed.

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

Aquaculture Nutrition 18; 1–11 Ó 2011 Blackwell Publishing Ltd


Table 1 Composition of the ingredients used for formulating the
experimental diets

To determine the concentrations of the microbial probiotic
inoculums to be added into the feed for the experiments, all the
three probiotic microorganisms, namely B. subtilis, L. lactis
and S. cerevisiae, were streaked on BHI, MRS, YEPD plates,
respectively, and incubated for 12 h at 30 °C. One freshly
grown colony was picked up and transferred into 50 mL of
respective broth and incubated under the same conditions for
4 h. A third transfer for each bacteria and yeast was carried
out into 100 mL under same conditions. Then, optical density
(O.D) of the microbial samples was recorded at 600 nm.
Simultaneously, the serial dilutions were performed for each

hour. The dilutions were plated onto the respective agar by
spread plate technique. After 12 h of incubation at 28 °C, the
colonies were counted. The data were related in graphs,
obtaining the relationship cfu versus OD600 versus time. Based
on this, the required probiotic microorganisms were added to
the feed at different concentrations.

To prepare the experimental feeds, the required ingredients
such as fish meal (sun-dried miscellaneous marine trash fish,
mainly the lesser sardines of family Engrolidae and ribbon
fish of family Trichuridae available in the locality, neither
solvent extracted nor dehulled), soybean oil cake (deoiled
soybean; mechanical extraction of oil), rice polish (obtained
from the local rice mill), wheat bran (the bran obtained from
the wheat meal) and corn flour (commercially available) were
purchased from the local market (Mumbai, India). The
ingredients were dried overnight at 80 °C in a hot air oven
and powdered by means of a California feed mill. The
powdered ingredients were sieved through a fine-meshed
screen (0.5 mm diameter). Before formulating the experimental diets, the proximate compositions of the feed ingredients were determined (AOAC 1990) and presented in
Table 1.
Six iso-nitrogenous (350 g protein kg)1 diet) and isocaloric (17.2–17.3 MJ kg)1 diet) experimental diets were
prepared incorporating various combinations of probiotics
added at equal proportions (either 1 : 1 or 1 : 1 : 1 depending on two or three numbers of probiotics used at a time) to
make the final concentration of 1011 cfu kg)1 diet except for
the control (Table 2). The required feed ingredients apart
from probiotic microbes, vitamins and minerals were mixed
with carboxymethyl cellulose, water and oil to make a dough
and steam cooked for 20 min in an autoclave at 15 psi. After

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

Aquaculture Nutrition 18; 1–11 Ó 2011 Blackwell Publishing Ltd

Chemical composition (g kg)1 on dry matter basis)

Ingredients
Fish meal
Soybean oil
cake
Rice polish
Wheat bran
Corn flour

Crude
Fibre

Gross
energy
(MJ kg)1)

Dry
matter

Crude
protein

Ether
extract

926
918

640.0
520.0

80.0
62.0

9.2
72.0

86.0
64.0

24.45
19.57

921
928
932

130.0
124.0
100.0

12.0
19.0
6.0

119.0
106.0
62.0

121.0
72.0
59.0

14.16
15.34
15.88

Ash

cooling, the vitamins and minerals and the respective probiotic microorganisms were added. Finally, the dough was
pressed through a hand pelletizer to get uniform size pellets
(2 mm) and kept overnight in a hot air oven (45 °C). The
commercially available kitchen type hand-operated pelletizer
was used for preparing the experimental diets. The prepared
feeds were stored at 4 °C until used for better shelf life. Fresh
feeds were prepared in every 15 days interval to maintain the
bacterial count at desired level. All the treatment groups of
fish were fed ad libitum to a level close to apparent satiation
at 08:00, 12:00, 15:00 and 18:00 h (Mohanta et al. 2009;
Mohapatra et al. 2010). The experiment lasted for 60 days.

Apparent nutrient digestibility coefficients of the diets were
determined by indirect method using 10 g chromic oxide
(Cr2O3) kg)1 of diet (Gomes et al. 1995) in the expense of
wheat bran during the last 30 days of the experiment. The

unconsumed feed and faeces were removed 1 h after the first
two feedings of the day at 09:00 and 13:00 h, and then, the
freshly voided faeces were collected after 1 h. The faeces of
the initial 5 days were discarded, and the next 25-day samples
were collected following the immediate pipetting method
outlined by Spyridakis et al. (1989). Pooled faecal samples of
each treatment were dried at 55 °C and stored at )20 °C for
subsequent analysis (Mohanta et al. 2009).
The digestibility of dry matter (DM) expressed as a
percentage was calculated using the following formula:
Apparent digestibility coefficient of dry matter (ADCDM Þ ¼
100 À ð%marker in feed=% marker in faeces  100Þ
The apparent nutrient (protein and lipid) digestibility ¼
100 À ½ð%marker in feed % marker infaeces)Â
ð%nutrients in faeces=%nutrient in feedÞ Â 100Š:


Table 2 Composition of experimental diets (g kg)1 on dry matter basis)
Dietary treatments
Ingredients

T1

T2

T3

T4

T5

T6

)1

Ingredients (g kg )
Fish meal1
114
114
114
114
114
114
Soybean oil cake1
420
420
420
420
420
420
Rice polish1
130
130
130
130
130
130
Wheat bran1
110
110

110
110
110
110
Corn flour (Himedia)
100
100
100
100
100
100
Carboxymethyl Cellulose (Himedia)
10
10
10
10
10
10
Sunflower oil2 : Cod liver oil (1 : 1)3
80
80
80
80
80
80
Vitamin mineral mix4
30
30
30
30

30
30
Vitamin C (Sd-fine Chem., Mumbai, India)
1
1
1
1
1
1
Vitamin B complex (Sd-fine Chem., Mumbai, India)
1
1
1
1
1
1
BHT (Himedia)
2
2
2
2
2
2
Glycine (Himedia)
2
2
2
2
2
2

Proximate composition (g kg)1)
Moisture
56.8 ± 0.3
61.8 ± 0.3
57.7 ± 0.4
61.5 ± 0.1
59.5 ± 0.4
58.2 ± 0.1
Crude protein
351 ± 0.2
351 ± 1.1
349 ± 2.4
348 ± 1.2
349 ± 1.9
352 ± 3.9
Total carbohydrate
466.1 ± 4.4
466.4 ± 6.8
470.5 ± 3.4
472.8 ± 1.3
471.6 ± 8.5
468.8 ± 1.3
Ether extract
96.4 ± 1.2
95.9 ± 1.6
92.9 ± 1.0
91.7 ± 6.0
92.3 ± 8.0
91.6 ± 4.0
Total ash

86.5 ± 3.4
86.7 ± 4.2
87.6 ± 1.2
87.5 ± 5.0
87.1 ± 2.1
87.6 ± 2.6
Energy (MJ kg)1)
17.3 ± 0.02
17.3 ± 0.03
17.3 ± 0.02
17.2 ± 0.01
17.3 ± 0.04
17.3 ± 0.01
1

Purchased from local dealers, Mumbai, India.
Marico Industries Limited, Mumbai, India.
3
Universal Medicare Private Limited, Mumbai, India.
4
Vitamin mineral mix (EMIX PLUS, Mumbai, India) (Quantity per kg).
Vitamin A: 22 00 000 IU; Vitamin D3: 4 40 000 IU; Vitamin B2: 800 mg; Vitamin E: 300 mg; Vitamin K: 400 mg; Vitamin B6: 400 mg; Vitamin
B12: 2.4 mg; Calcium Pantothenate: 1000 mg; Nicotinamide: 4 g; Choline Chloride: 60 g; Mn: 10 800 mg; I: 400 mg; Fe: 3000 mg; Zn:
2000 mg; Cu: 800 mg; Co: 180 mg; Ca: 200 g; P: 120 g; L-lysine: 4 g; DL-Methionine: 4 g; Selenium: 20 ppm.
2

The proximate composition of experimental diets, faecal
samples and whole body was analysed in triplicates (AOAC
1990). DM was estimated by oven drying the samples at
105 °C till a constant weight and crude protein per cent were

calculated by estimating nitrogen content by micro-Kjeldahl
method and multiplying with a factor 6.25. Ether extract was
determined by solvent extraction with petroleum ether,
boiling point 40–60 °C, for 10–12 h. Total ash content was
determined by incinerating the sample at 650 °C for 6 h and
crude fibre by acid digestion (1.25%) followed by alkali
digestion (1.25%). Gross energy in diets, faecal samples and
fish body was calculated by using Bomb Calorimeter (Parr,
model 1341; Parr Instrument Company, Moline, IL, USA).
The Cr2O3 content of the feed and faecal samples was
determined as described by Furukawa & Tsukahara (1966).
The total carbohydrate (%) of the feed and whole body fish
was calculated as: 100 ) (Crude protein + Ether extract + Total ash).

The water quality parameters were found to be in the range of
temperature (25.6–26.4 °C), pH (7.4–7.6), dissolved oxygen
(5.8–6.9 mg L)1), free carbon dioxide (1.9–2.7 mg L)1), total
hardness (156–185 mg L)1), ammonia-N (0.04–0.07 mg L)1),
nitrite-N (0.06–0.1 mg L)1) and nitrate-N (0.03–0.14 mg l)1)
throughout the experiment period (APHA-AWWA-WEF
1998). While the water temperature was recorded twice daily at
0600 and 1430 h, the other parameters were analysed in every
15-day interval. All the water quality parameters during the
entire experiment period were found to be in the optimum
range of fish rearing (Debnath et al. 2007; Kumar et al. 2010;
Mohapatra et al. 2010).

The growth parameters of the L. rohita fingerlings were
assessed in terms of weight gain per cent, specific growth rate
(SGR), feed conversion ratio (FCR), protein efficiency ratio

(PER), protein retention efficiency, lipid productive value

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

Aquaculture Nutrition 18; 1–11 Ó 2011 Blackwell Publishing Ltd


(LPV) and energy productive value (EPV) at the end of the
experiment. The experimental fish with respect to each replicate was batch weighed in every 15 days to know the body
weight of fish. The weight gain (%), SGR, FCR, PER, PPV,
LPV and EPV were evaluated based on standard formulae as
follows:
Weight gain% ¼ ðFinal weight À Initial weightÞ=
ðInitial weight) Â 100
Specific growth rate (SGR) ¼ 100ðloge average
Final weight À loge average Initial weightÞ=
Number of culture days
Feed conversion ratio (FCR) ¼ Total dry feed intake (g)=
Wet weight gain (g)
Protein efficiency ratio (PER) ¼ Total wet weight gain (g)=
Protein fed (g)
Nutrients (protein and lipid) and energy productive
values (PPV, LPV and EPVÞ ¼ Nutrients or
energy gain in body=Nutrients or energy intake

Tissue homogenates of intestine were prepared with chilled
sucrose solution (0.25 M), and the different digestive enzymes
were analysed using standard procedures as described below.
The intestines used for digestive enzyme assay with respect to
each dietary treatment were in fresh condition.

Protease activity was determined by the casein digestion
method as Drapeau (1974). One unit of enzyme activity was
defined as the amount of enzyme need to release acid-soluble
fragments equivalent to 0.001A280 per minute at 37 °C and
pH 7.8.
The lipase activity was assayed as described by Cherry &
Crandall (1932) with suitable modification (Debnath et al.
2007). The volume (mL) of N/20 NaOH solution required for
100 mg intestinal tissue in the experimental tube minus the
volume (mL) of N/20 NaOH solution required for the same
amount of intestinal tissue in the control tube represented the
units of intestinal lipase activity per g tissue. One unit will
hydrolyse 1.0 microequivalent of fatty acid from a triglyceride in 24 h at pH 7.7 at 37 °C.

The gastrointestinal microflora analysis was carried out on 0,
15th, 30th, 45th and 60th day of feeding trial. Three fishes

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

Aquaculture Nutrition 18; 1–11 Ó 2011 Blackwell Publishing Ltd

were randomly selected from each treatment (i.e. one fish
from each replicate group) and collected in sterile plastic
bags. The fishes were starved for 20 h before the sampling.
The fresh intestine was aseptically taken out from each of
these fish. The intestine of all replicates of a treatment was
weighed equally to make 1.0 g of sample. Sample was
transferred into tubes containing 9.0 mL sterile 0.85% NaCl
and crushed in a homogenizer. The homogenates were serially diluted up to 10)6.
Nutrient agar was used for THB count. Nutrient agar

media, MRS media and YEPD media were used for B. subtilis, L. lactis and S. cerevisiae, respectively.
In spread plate technique, 0.1 mL from highest dilution
(10)6) was properly spread on the media, and the plates were
incubated at 28 °C for 24 h in Biological Oxygen Demand
(BOD) incubator. On the following day, colonies were counted
and were isolated for characterization. Numbers of colonies
reported in the present study are an average of three replicate
plates. Isolated colonies (at least 10 per plate) were used for
identification by morphological and biochemical tests (Bergy
1986; Guimaraes et al. 2006). Each probiotic colony was
counted separately and deduced from initial count to get per
cent increase in colonization.

Comparison among all the treatments was carried out by oneway ANOVA followed by DuncanÕs multiple range test. Comparison was made at the 5% probability levels. The data were
statistically analysed by statistical package SPSS, version 16.0.

In our study, we fed the fishes with different combinations of
three probiotics, and the various growth parameters, i.e.
weight gain, SGR, FCR, PER, PPV, LPV and EPV, were
determined. The growth data were substantiated by analysis of
different digestive enzyme activities. The colonization of
respective probiotics in the gastrointestinal tract was analysed
so as to study the effect of probiotics on growth and nutrient
utilization.

Growth and nutrient utilization of fish are the two basic criteria that determine the production, productivity and profitability of the fish culture operation. The growth and nutrient
utilization obtained from this study are presented in Table 3.
At the end of the experiment, it was found that the weight gain

(%) was increased significantly (P < 0.05) in all probioticsupplemented groups (except T6) than that of control with a
highest value recorded in T5 (232.16) group followed by T4
(171.3), T3 (167.3) and T2 (166.3) groups. Similar results were
also obtained for SGR and PER. The T5 group had significantly higher (P < 0.05) SGR (0.87) and PER (1.52) values
than the other groups (0.61–0.72 and 0.80–1.13, respectively).
The PPV (0.15), LPV (0.28) and EPV (0.61) of T5 group of fish
were also significantly higher (P < 0.05) than the other
probiotic-fed groups (0.12–0.13, 0.13–0.19, 0.34–0.41, respectively). Significantly lower (P < 0.5) FCR (1.93) was also
recorded in T5 group of fish as compared to other groups (2.52–
3.18). However, the fishes fed heat-inactivated probiotics did
not show any remarkable growth and nutrient gain over the
other probiotic-supplemented groups.

groups. Similar observations were also recorded for lipase
activity (Table 3).

The intestinal microflora of fish reflects the bacterial content
of ingested food and of the environment. It also speculates
the competitive exclusion of harmful gut microflora. To
emphasize the probiotic colonization leading to unwanted
microbial exclusion, we analysed the THB population in
experimental fish gut. The THB population in the intestine of
L. rohita fingerlings at different sampling days is presented in
Table 4. The THB population in the intestine was found to
be significantly higher (P < 0.05) in all sampling days (15, 30,
45 and 60th day) in T1 (control) and T6 groups compared to
other treatment groups with a minimum value (2.39 cfu
· 1010 kg)1) observed in T3 on 30th day. For the T3, T4 and
T5 groups, on 15th and 30th day, the THB population was
also drastically reduced from the initial value (0 day of

sampling) and thereafter remained almost similar on 45 and
60th day of sampling. But for T1 (control) and T6 groups, the
THB population remained almost same for all sampling
days.

The apparent digestibility coefficient of DM (ADCDM) and
protein (ADCProtein) was found to be significantly different
(P < 0.05) among treatment groups with highest being in T5
(52.3% for DM and 86.2% for protein). However, there was
no significant (P > 0.05) effect of probiotic feeding on
apparent lipid digestibility (ADC Lipid; 73.13–75.31%) in
L. rohita fingerlings (Table 3).

To substantiate the fact that the difference in THB is related
to probiotic colonization in gut, we determined the per cent
prevalence of different administered probiotics in the gut of
L. rohita fingerlings and presented in Table 5. On feeding

The protease activity of T2, T3, T4 and T5 groups was significantly higher (P < 0.05) than the control (T1) and T6

Table 3 Effect of different dietary probiotic supplementation on growth parameters, apparent digestibility coefficients (ADC) and digestive
enzyme activities in Labeo rohita fingerlings
Dietary treatment
Parameter studied

T1

Weight gain (%)
SGR
FCR

PER
PPV
LPV
EPV
ADCDM
ADCProtein
ADCLipid
Protease1
Lipase2

131.27d
0.61c
3.18a
0.80c
0.12c
0.17b
0.37b
43.62c
79.76c
73.13
0.12b
0.25b

T2
±
±
±
±
±
±

±
±
±
±
±
±

7.19
0.02
0.25
0.05
0.002
0.01
0.07
1.95
0.28
0.15
0.01
0.02

166.30bc
0.71b
2.52ab
1.13b
0.13bc
0.15bc
0.40b
50.68ab
82.98b
74.92

0.20a
0.62a

T3
±
±
±
±
±
±
±
±
±
±
±
±

9.39
0.03
0.08
0.04
0.01
0.00
0.05
1.27
0.46
0.39
0.03
0.17

167.35bc
0.71b
2.89a
1.01bc
0.13b
0.19b
0.41b
48.6b
83.67b
73.89
0.22a
0.68a

T4
±
±
±
±
±
±
±
±
±
±
±
±

9.98
0.03
0.30

0.11
0.00
0.02
0.08
1.75
0.46
1.45
0.01
0.12

171.32b
0.72b
2.63a
1.13b
0.12bc
0.16bc
0.41b
47.22b
85.72a
74.83
0.21a
0.51a

T5
±
±
±
±
±
±

±
±
±
±
±
±

6.20
0.02
0.20
0.08
0.00
0.01
0.07
1.47
0.30
1.38
0.02
0.07

232.16a
0.87a
1.93b
1.52a
0.15a
0.28a
0.61a
52.26a
86.21a
75.31

0.21a
0.66a

P value

T6
±
±
±
±
±
±
±
±
±
±
±
±

15.31
0.03
0.19
0.14
0.01
0.02
0.10
1.91
1.07
0.60
0.02

0.11

138.69cd
0.63c
2.96a
0.98bc
0.13bc
0.13c
0.34b
46.98b
80.67c
74.48
0.11b
0.29b

±
±
±
±
±
±
±
±
±
±
±
±

4.81
0.01

0.10
0.03
0.01
0.01
0.05
1.14
0.21
0.77
0.04
0.06

0.000
0.000
0.012
0.012
0.008
0.000
0.000
0.020
0.000
0.321
0.017
0.036

Data expressed as Mean ± SE (n = 3).
Mean values in same row with different superscripts vary significantly (P < 0.05).
1
Protease activity expressed as units per mg protein per min.
2
Lipase activity expressed as units per mg protein per hour.

EPV, energy productive value; FCR, feed conversion ratio; LPV, lipid productive value; PER, protein efficiency ratio; SGR, specific growth rate.

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

Aquaculture Nutrition 18; 1–11 Ó 2011 Blackwell Publishing Ltd


Table 4 Total heterotrophic bacteria
(THB) population (cfu · 1010 kg)1) in
the intestine of Labeo rohita fingerlings
at various sampling days in different
dietary probiotic-supplemented groups

Sampling days
Treatment

0th day

15th day

30th day

45th day

60th day

T1
T2
T3
T4

T5
T6
P value

4.54
4.44
4.36
4.63
4.61
4.49
0.06

4.37a ±
3.59b ±
3.40b ±
3.51b ±
3.37b ±
4.04a ±
0.004

4.29a ±
3.40c ±
2.39d ±
2.54d ±
2.44d ±
3.83b ±
0.000

4.25a ±
3.28b ±

2.71c ±
2.53d ±
2.42d ±
4.16a ±
0.026

4.59a ±
3.70c ±
2.83d ±
2.42e ±
2.48e ±
4.27b ±
0.000

±
±
±
±
±
±

0.19
0.06
0.11
0.09
0.08
0.04

0.16
0.12

0.14
0.09
0.09
0.06

0.17
0.17
0.11
0.08
0.06
0.07

0.06
0.01
0.09
0.08
0.07
0.04

0.12
0.10
0.08
0.07
0.10
0.06

Data expressed as Mean ± SE (n = 3), Unit- cfu · 1010 kg)1.
Mean values in same column with different superscripts vary significantly (P < 0.05).

Table 5 Percentage prevalence of Bacillus, Lactococcus and Saccharomyces species isolates found in the intestine of Labeo rohita

fingerlings at various sampling days in different dietary probioticsupplemented groups
Sampling days
0th
day

15th
day

30th
day

45th
day

60th
day

Bacillus sp.
Lactococcus sp.
Saccharomyces sp.

1.9
3.0
ND

2.0
2.8
ND

2.2

3.3
ND

2.1
3.4
ND

2.2
3.5
ND

Bacillus sp.
Lactococcus sp.
Saccharomyces sp.

2.1
2.7
ND

16.9
18.9
ND

24.2
27.0
ND

24.2
26.9
ND

24.4
27.1
ND

Bacillus sp.
Lactococcus sp.
Saccharomyces sp.

2.3
3.2
ND

2.3
19.4
17.9

2.5
27.0
24.9

2.5
26.1
25.9

2.6
27.0
26

Bacillus sp.

Lactococcus sp.
Saccharomyces sp.

2.2
3.1
ND

19.4
3.0
18.4

27.2
3.1
25.3

27
3.2
25.9

27.3
3.4
26.0

Bacillus sp.
Lactococcus sp.
Saccharomyces sp.

2.1
2.8
ND

18.9
17.2
18.1

28.1
25.3
26.7

28.8
25.9
27

29.1
26.2
27.2

Bacillus sp.
Lactococcus sp.
Saccharomyces sp.

2.2
3.3
ND

5.4
4.9
5.1

6.7

6.1
6.3

6.6
6.1
6.3

6.7
6.4
6.6

Treatment
T1

T2

T3

T4

T5

T6

The percentage prevalence of probiotic microbial species in the
intestine is calculated by taking three fish per treatment (one fish
per replicate) in each sampling days.

different probiotics, it was observed that the specific
microbial probiotics that supplemented through diets were

established at much higher percentage in the gut of fish in all
probiotic-fed groups (except T6 group) (Table 5). Although
the increase was observed from the 15th day of feeding, the
maximum gut colonization was observed on the 30th day,
and thereafter, it almost remained constant.

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

Aquaculture Nutrition 18; 1–11 Ó 2011 Blackwell Publishing Ltd

In the present study, except T6, we obtained 150–200%
increase in weight of fish in different probiotic-fed groups,
which was significantly higher (P < 0.05) than the fish fed
diets without probiotic supplementation (T1). This is in
agreement with the earlier findings by different authors
(Swain et al. 1996; Ghosh et al. 2003; Carnevali et al. 2006).
Jafaryan et al. (2008) also reported that the probiotic
(Bacillus)-supplemented diet significantly increased the
weight, length and SGR of fish than the control diet without
probiotic supplementation. Although we obtained higher
SGR in probiotic-supplemented groups (T2, T3, T4 and T5),
not much literature is available regarding the effect of different combinations of probiotics on SGR to compare our
results. According to Noh et al. (1994) and Bogut et al.
(1998), different probiotics act differently to enhance the
growth and nutrient utilization of various fish species. We
also obtained similar observations in our study. Noh et al.
(1994) studied the effect of supplementing yeast (S. cerevisiae) and bacteria (Streptococcus faecium) in the diet of Israeli
carp and reported the better growth response of fish fed
probiotic-supplemented diets than the diet without probiotic
supplementation. But they found better growth and nutrient

utilization in a bacterium-supplemented diet than yeast. The
better food conversion (FCR), nutrient utilization (PER) and
protein gain (PPV) obtained in our study in the probiotic-fed
groups (except T6) as compared to control are in agreement
with Bagheri et al. (2008) reported for rainbow trout (Onchorhynchus mykiss) fed diet supplemented with probiotics.
The higher PPV, LPV and EPV values indicate better
nutrient and energy retention in T5 group than the other
groups (Tables 3 and 6). This may be because of an active
probiotic supplementation effect in T5 as reported by ElDakar et al. (2007). Therefore, supplementation of three live
probiotic microorganisms in the diet at equal proportions
(1 : 1 : 1) improves food conversion and nutrient retention,


Table 6 Whole body chemical composition (g 100 g)1 on dry matter basis) of fish in different probiotic-supplemented groups
Dietary treatments
Parameters

T1

Moisture
Crude protein
Ether extract
Total ash
Total carbohydrate
Energy (MJ 100g)1)

75.6
56.60b
13.54ab
19.06a

10.80c
18.91a

T2
±
±
±
±
±
±

0.4
0.74
1.81
1.20
0.74
0.33

76.4
61.71a
9.04b
16.02b
13.23a
17.86b

T3
±
±
±
±

±
±

0.1
0.40
1.31
0.86
0.34
0.13

T4

75.9
58.86b
11.41b
18.75a
10.98c
18.49ab

±
±
±
±
±
±

0.1
0.85
2.00
1.29

0.24
0.29

76.3
60.57a
11.03b
16.14b
12.26b
18.06b

T5
±
±
±
±
±
±

0.1
0.57
0.37
0.90
0.12
0.19

76.2
56.73b
11.96b
18.91a
12.40b

18.62a

T6
±
±
±
±
±
±

0.2
0.86
1.92
1.24
0.71
0.32

76.1
57.42b
15.60a
16.08b
10.90c
18.28b

±
±
±
±
±
±

0.1
1.32
1.64
0.88
0.27
0.28

Data expressed as Mean ± SE (n = 3).
Mean values in same row with different superscripts vary significantly (P < 0.05).

hence leading to higher growth of fish. However, unlike
Salinas et al. (2008), we could not observe any growth and
nutritional gain in fish by supplementing heat-inactivated
probiotics in feed (T6).
It is reported that the digestive organs are very sensitive to
food composition and cause immediate changes in activities
of the digestive enzymes (Bolasina et al. 2006; Shan et al.
2008), which is finally reflected in fish health and growth.
Moreover, bacteria also secrete proteases to digest the peptide bonds in proteins and therefore break down the proteins
into their constituent monomers and free amino acids, which
can benefit the nutritional status of the animal (MacFarlane
& Cummings 1991). In the present work, except T6 (supplementation of heat-killed probiotics of B. subtilis, L. lactis
and S. cerevisiae), we observed a high protease activity in
other probiotic-fed groups in relation to control (T1). Bacterial enzymatic hydrolysis has been shown to enhance the
bioavailability of protein and fat (Ling & Hanninen 1992),
which may result in higher growth and nutrient utilization as
observed in the present study. Amylase and lipase are the
major enzymes related to carbohydrate and fat digestion,
respectively. From our results, it was observed that the

lipase activity was much higher in fish fed diets supplemented with live probiotics (T2, T3, T4 and T5) than that of
control (T1, diet without probiotic supplementation) and fish
fed diet supplemented with heat-killed probiotics (T6), which
might have resulted comparatively better growth and other
growth-related parameters in all live probiotic-supplemented
groups. It is reported by the earlier workers that the increase
in nutrient digestibility may be because of better availability
of exoenzymes produced by probiotics (Vine et al. 2006) or
better health condition (Yanbo & Zirong 2006) when probiotic-supplemented diets are fed to the fish. Bairagi et al.
(2002) reported that the Bacillus species isolated from the
gut of Cyprinus carpio were found to have high amount of
extracellular amylolytic, proteolytic and lipolytic activity.
Majority of probiotics are capable of secreting lipase, which

triggers production and assimilation of essential fatty acids
resulting higher growth and immunity in fish. Feed supplementation of essential fatty acid not only boosts the
immunity but also triggers the growth (Sharma et al. 2009).
The present data also confirm similar hypothesis. The data
clearly indicate that addition of probiotics significantly increases lipase activity irrespective of species or combination
of probiotics, which corroborates the findings of Yanbo &
Zirong (2006). But the exogenous enzymes produced by the
probiotics represent only a small contribution to the total
enzyme activity of the gut (Ding et al. 2004; Ziaei-Nejad
et al. 2006; Zhang et al. 2010). It suggests that the higher
digestive enzyme activities (protease and lipase) obtained in
the probiotic-supplemented fish (except T6) are mainly the
outcome of stimulation by probiotic itself or exogenous
enzyme produced to synthesize endogenous digestive enzyme
which might have improved nutrient digestibility leading to
better growth performance and feed efficiency in fish. Similar

observations have also been reported for other fishes in
which the nutrient digestibility increased considerably with
the use of probiotic-supplemented diet (Lara-Flores et al.
2003; Yanbo & Zirong 2006). Although we did not find any
significant variation in apparent lipid digestibility (ADCLipid), there was significant increase in the apparent protein
digestibility (ADCProtein) in probiotic fed groups. The increased protease activity in probiotic-supplemented diet
groups might have resulted better protein digestion and
hence better growth and protein gain in fish.
Similar to Bagheri et al. (2008), we also observed significant reduction in THB counts when probiotic-supplemented
diets were fed to the fish, which might have resulted better
health and immunity of fish and hence more growth. It is
reported that the colonization rate of bacteria in the digestive
tracts depends on the dietary bacteria level (Bagheri et al.
2008). In our study, the higher degree of adhesion of specific
microbes that are supplemented through diets may be the
reason for enhanced growth and nutrient utilization of fish.

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

Aquaculture Nutrition 18; 1–11 Ó 2011 Blackwell Publishing Ltd


Several reports suggest that most of probiotics exert their
effect through colonization in host and excretion of several
growth-enhancing nutrients (Ahilan et al. 2004; Bagheri
et al. 2008). The lower colonization in the heat-killed probiotic treatment group can be attributed to the loss of useful
characteristics particularly the colonization capacity of the
microbes because of the higher temperature at which it was
killed. This may resulted the low dietary performance of fishfed heat-killed microbes as probiotics.
In the present work, diet supplemented with two species of

bacteria (L. lactis and B. subtilis) and one species of yeast
(S. cerevisiae) in equal proportion (T5) as probiotics at the
concentration of 1011 cfu kg)1 of feed showed maximum
growth and dietary performance in L. rohita fingerlings than
the other probiotic combinations tested in the diets. This
indicated that combination of more probiotic organisms in
the diet results better performance in fish. It may be attributed to better food conversion (FCR), higher protein digestion (ADC for protein), utilization (PER) and gain (PPV),
significant reduction in THB counts in the intestine and
better adhesion/establishment of probiotic microflora in the
gut. Similar results were also obtained from Tilapia nilotica
(Lara-Flores et al. 2003), L. rohita (Ghosh et al. 2003),
C. carpio (Yanbo & Zirong 2006; Ramakrishnan et al. 2008)
and Fenneropenaeus indicus (Ziaei-Nejad et al. 2006). However, we observed that the use of same combinations of heatinactivated probiotics (T6) in the diet failed to enhance
growth, nutrient digestion and utilization and digestive
enzyme activity. The study results suggest that the live
probiotic microorganisms may be incorporated while formulating the cost-effective nutritionally balanced diet of carp
for its better growth performance and nutrient utilization.

The authors acknowledge Indian Council of Agricultural
Research (ICAR), New Delhi for financial support and
Director, CIFE, Mumbai for providing all the necessary
facilities to carry out this research programme.

Ahilan, B., Shine, G. & Santhanam, R. (2004) Influence of probiotics
on the growth and gut microbial load of juvenile goldfish (Carassius auratus). Asian Fish Sci., 17, 271–278.
AOAC (1990) Official Methods of Analysis of the Association of
Official Analytical Chemists, Vol. 1, 14th edn. 1102 p. Association
of Official Analytical Chemists, Arlington, VA, USA.
APHA-AWWA-WEF (1998) Standard Methods for the Examination of Water and Wastewater, 20th edn pp, 413–426. American

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

Aquaculture Nutrition 18; 1–11 Ó 2011 Blackwell Publishing Ltd

Public Health Association, American Water Works Association,
Water Environment Federation, Washington DC.
Bagheri, T., Hedayati, S.A., Yavari, V., Alizade, M. & Farzanfar, A.
(2008) Growth, survival and gut microbial load of rainbow trout
(Onchorhynchus mykiss) fry given diet supplemented with probiotic
during the two months of first feeding. Turk. J. Fish. Aquat. Sci., 8,
43–48.
Bairagi, A., Sarkar Ghosh, K., Sen, S.K. & Ray, A.K. (2002)
Duckweed (Lemna polyrhiza) leaf meal as a source of feedstuff in
formulated diets for rohu (Labeo rohita Ham.) fingerlings after
fermentation with a fish intestinal bacterium. Bioresour. Technol.,
85, 17–24.
Bergy, D.H. (1986) BergeyÕs manual of systemic bacteriology. Vol. 2
(Holt, J.H., Sneath, P.H. and Krieg, N.R. eds) pp. 636. Lipponcott
Williams and Wilkins, ISBN 0683078933.
Bogut, I., Milakovic, Z., Bukvic, Z., Brkic, S. & Zimmer, R. (1998)
Influence of probiotic Streptococcus faecium M74 on growth and
content of intestinal micro-flora in carp Cyprinus carpio. Czech J.
Anim. Sci., 43, 231–235.
Bolasina, S., Pe´rez, A. & Yamashita, Y. (2006) Digestive enzymes
activity during ontogenetic development and effect of starvation in
Japanese flounder, Paralichthys olivaceus. Aquaculture, 252, 503–
515.
Brunt, J. & Austin, B. (2005) Use of a probiotic to control lactococcosis and streptococcosis in rainbow trout Oncorhynchus
mykiss (Walbaum). J. Fish Dis., 28, 693–701.
Carnevali, O.D., Sulpizio, L., Gioacchini, R., Olivotto, I.G. & Silvi,

S. (2006) Growth improvement by probiotic in European sea bass
juveniles (Dicentrarchus labrax, L.), with particular attention to
IGF-1, myostatin and cortisol gene expression. Aquaculture, 258,
430–438.
Cherry, I.S. & Crandall, L.A. Jr (1932) The specificity of pancreatic
lipase: its appearance in blood after pancreatic injury. Am. J.
Physiol., 100, 266–273.
Debnath, D., Pal, A.K., Sahu, N.P., Yengkokpam, S., Baruah, K.,
Choudhury, D. & Venkateshwarlu, G. (2007) Digestive enzymes
and metabolic profile of Labeo rohita fingerlings fed diets with
different crude protein levels. Comp Biochem Physiol B Biochem
Mol Biol., 146, 107–114.
Ding, X., Li, Z.J., Chen, Y.Q., Lin, H.Z., Yang, Y.Y. & Yang,
K. (2004) Effects of probiotics on growth and activities of
digestive enzymes of Penaeus vannamei. J. Fish. Sci. China, 11,
580–584.
Drapeau, G. (1974) Protease from Staphylococcus aureus. In:
Methods in Enzymology (Lorand, B.L. ed.) Vol. 45, pp. 469.
Academic Press, NY, USA.
El-Dakar, A.Y., Shalaby, S.M. & Saoud, I.P. (2007) Assessing the
use of a dietary probiotic/prebiotic as an enhancer of spinefoot
rabbitfish Siganus rivulatus survival and growth. Aquacult. Nutr.,
13, 407–412.
Furukawa, A. & Tsukahara, H. (1966) On the acid digestion method
for determination of chromic oxide as an index substances in the
study of digestibility of fish feed. Bull. Jpn. Soc. Sci. Fish., 32, 502–
506.
Gatesoupe, F.J. (1999) The use of probiotics in aquaculture. Aquaculture, 180, 147–165.
Ghosh, K., Sen, S.K. & Ray, A.K. (2003) Supplementation of an
isolated fish gut bacterium, Bacillus circulans, in formulated

diets for rohu, Labeo rohita, fingerlings. Israeli J. Aquacult., 55,
13–21.
Gomes, D.M.S., Mantovani, A. & Vieira, R.C. (1995) Anatomia
foliar de Psychotria ternuinerves Mu¨ll. Arg. e Psychotria stenoc-


alix Mu¨ll. Arg. (Rubiaceae). Arquivos de Biologia e Tecnologia, 38,
15–33.
Guimaraes, T.M., Moriel, D.G., Machado, I.P., Fadel Picheth,
C.M.T. & Bonfim, T.M.B. (2006) Isolation and characterization of
Saccharomyces cerevisiae strain of winery interest. Brazilian J.
Pharm. Sci., 42, 119–126.
Hagi, T., Tanaka, D., Iwamura, Y. & Hoshino, T. (2004) Diversity
and seasonal changes in lactic acid bacteria in the intestinal tract of
cultured freshwater fish. Aquaculture, 234, 335–346.
Jafaryan, H., Asadi, R. & Bagheri, A. (2008) The promotion of
growth parameters and feeding efficiency of Acipenser nudiventris
larvae by using of probiotic Bacillus via bioencapsulation of Artemia urmiana. In: Resource management, natural, human and
material resources for the sustainable development of aquaculture.
Aquacult. Europe 08.
Kesarcodi-Watson, A., Kaspar, H., Lategan, M.J. & Gibson, L.
(2008) Probiotics in aquaculture: the need, principles and mechanisms of action and screening process. Aquaculture, 274, 1–14.
Kumar, R., Mukherjee, S.C., Paniprasad, K. & Pal, A.K. (2006)
Evaluation of Bacillus subtilis as a probiotic to Indian major carp
Labeo rohita (ham.). Aquacult. Res., 37, 1215–1221.
Kumar, R., Mukherjee, S.C., Ranjan, R. & Nayak, S.K. (2008)
Enhanced innate immune parameters in Labeo rohita (Ham.) following oral administration of Bacillus subtilis. Fish Shellfish
Immunol., 24, 168–172.
Kumar, V., Sahu, N.P., Pal, A.K., Kumar, S., Sinha, A.K., Ranjan,
J. & Baruah, K. (2010) Modulation of key enzymes of glycolysis,

gluconeogenesis, amino acid catabolism, and TCA cycle of the
tropical freshwater fish Labeo rohita fed gelatinized and nongelatinized starch diet. Fish Physiol. Biochem., 36, 491–499.
Lara-Flores, M., Olvera-Novoa, M.A. & Guzma´n-Me´ndez, B.E.
(2003) Use of the bacteria Streptococcus faecium and Lactobacillus
acidophilus, and the yeast Saccharomyces cerevisiae as growth
promoters in Nile tilapia (Oreochromis niloticus). Aquaculture, 216,
193–201.
Ling, W.H. & Hanninen, O. (1992) Shifting from a conventional diet
to an uncooked vegan diet reversibly alters faecal hydrolytic
activities in humans. J. Nutr., 122, 924–930.
MacFarlane, G.T. & Cummings, J.H. (1991) The colonic flora, fermentation and large bowel digestive function. In: The Large
Intestine: Physiology, Pathophysiology and Disease (Phillips, S.F.,
Pemberton, J.H. & Shorter, R.G. eds), pp. 51–92. Raven Press
Ltd, New York, NY, USA.
Mazurkiewicz, J., Przybyl, A., Sip, A. & Grajek, W. (2007) Effect of
Carnobacterium divergens and Enterococcus hirae as probiotic
bacteria in feed for common carp, Cyprinus carpio. L. Arch. Polish
Fish., 15, 93–102.
Mohanta, K.N., Mohanty, S.N., Jena, J.K. & Sahu, N.P. (2009) A
dietary energy level of 14.6 MJ kg-1 and protein-to-energy ratio of
20.2 g MJ-1 results in best growth performance and nutrient
accretion in silver barb, Puntius gonionotus fingerlings. Aquacult.
Nutr., 15, 627–637.
Mohanty, S.N., Swain, S.K. & Tripathi, S.D. (1996) Rearing of catla
(Catla catla Ham) spawn on formulated diets. J. Aquac. Trop., 11,
253–258.
Mohapatra, S., Sahu, N.P., Pal, A.K., Prusty, A.K., Kumar, V. &
Kumar, S. (2010) Haemato-immunology and histo-architectural
changes in Labeo rohita fingerlings: effect of dietary aflatoxin and
mould inhibitor. Fish Physiol. Biochem., doi: 10.1007/s10695-0109428-1.

Muroga, K., Higashi, M. & Keitoku, H. (1987) The isolation of
intestinal micro-flora of farmed red seabream (Acanthopagrus
schlegeli) at larval and juvenile stages. Aquaculture, 65, 79–88.

Naik, A.T.R., Murthy, H.S. & Ramesha, T.J. (1999) Effect of
graded levels of G-probiotic on growth, survival and feed
conversion of tilapia, Oreochromis mossambicus. Fish. Technol.,
36, 63–66.
Nayak, S.K., Swain, P. & Mukherjee, S.C. (2007) Effect of dietary
supplementation of probiotic and vitamin C on the immune response of Indian major carp, Labeo rohita (Ham.). Fish Shellfish
Immunol., 23, 892–896.
Noh, H., Han, K.I., Won, T.H. & Choi, Y.J. (1994) Effect of antibiotics, enzymes, yeast culture and probiotics on growth performance of Israeli carp. Kor. J. Anim. Sci., 36, 480–486.
Pal, D., Joardar, S.N. & Roy, B. (2007) Immunostimulatory effects
of a yeast (Saccharomyces cerevisiae) cell wall feed supplement on
rohu (Labeo rohita), an Indian major carp. Israeli J. Aquac.
Bamigeh, 59, 175–181.
Ramakrishnan, C.M., Haniffa, M.A., Manohar, M., Dhanaraj, M.,
Jesu Arockiaraj, A., Seetharaman, S. & Arunsingh, S.V. (2008)
Effects of probiotics and spirulina on survival and growth
of juvenile common carp (Cyprinus carpio). Isr. J. Aquacult., 60,
128–133.
Robertson, P.A.W., OÕDowd, C., Burrells, C., Williams, P. & Austin,
B. (2000) Use of Carnobacterium sp. as a probiotic for Atlantic
salmon (Salmo salar L.) and rainbow trout (Oncorhynchus mykiss,
Walbaum). Aquaculture, 185, 235–243.
Salinas, I., Abelli, L., Bertoni, F., Picchietti, S., Roque, A., Furones,
D., Cuesta, A., Meseguer, J. & Esteban, M.A. (2008) Monospecies
and multispecies probiotic formulations produce different systemic
and local immunostimulatory effects in the gilthead seabream
(Sparus aurata L.). Fish Shellfish Immunol., 25, 114–123.

Shan, S., Xiao, Z., Huang, W. & Dou, S. (2008) Effect of photoperiod on growth, mortality and digestive enzymes in miiuy croaker
larvae and juveniles. Aquaculture, 281, 70–76.
Sharma, P., Kumar, V., Sinha, A.K., Ranjan, J., Kithsiri, H.M.P. &
Venkateshwarlu, G. (2009) Comparative fatty acid profiles of wild
and farmed tropical freshwater fish rohu (Labeo rohita). Fish
Physiol. Biochem., 5, 229–239.
Spyridakis, P., Metailler, R. & Gabaudan, J. (1989) Studies on
nutrient digestibility in European sea bass (Dicentrarchus labrax):
2. Effect of sodium alginate on protein and lipid digestibility.
Aquaculture, 77, 71–73.
Sugita, H., Fujie, T., Sagesaka, T. & Itoi, S. (2009) The effect of
Lactococcus lactis on the abundance of aeromonads in the rearing
water of the goldfish, Carassius auratus (Linnaeus). Aquacult. Res.,
41, 153–156.
Swain, S.K., Rangacharyulu, P.V., Sarkar, S. & Das, K.M. (1996)
Effect of a probiotic supplement on growth, nutrient utilization
and carcass composition in mrigal fry. J. Aquac., 4, 29–35.
Va¢zquez-Jua¢rez, A., Gonza¢lez, M.P. & Murado, M.A. (2005)
Effects of lactic acid bacteria cultures on pathogenic microbiota
from fish. Aquaculture, 245, 149–161.
Vine, N.G., Leukes, W.D. & Kaiser, H. (2006) Probiotics in marine
larviculture. FEMS Microbiol. Rev., 30, 404–427.
Wang, Y. (2007) Effect of probiotics on growth performance and
digestive enzyme activity of the shrimp Penaeus vannamei. Aquaculture, 269, 259–264.
Wang, Y. (2011) Use of probiotics Bacillus coagulans, Rhodopseudomonas palustris and Lactobacillus acidophilus as growth promoters in grass carp (Ctenopharyngodon idella) fingerlings.
Aquacult. Nutr., 17, 372–378.
Wang, Y. & Xu, Z.R. (2006) Effect of probiotics for common
carp (Cyprinus carpio) based on growth performance and digestive
enzyme activities. Anim. Feed Sci. Technol., 127, 283–292.

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

Aquaculture Nutrition 18; 1–11 Ó 2011 Blackwell Publishing Ltd


Wang, Y.B., Xu, Z.R. & Xia, M.S. (2005) The effectiveness of
commercial probiotics in Northern White Shrimp (Penaeus vannamei L.) ponds. Fish. Sci., 71, 1034–1039.
Yanbo, W. & Zirong, X. (2006) Effect of probiotics for common carp
(Cyprinus carpio) based on growth performance and digestive enzyme activities. Anim. Feed Sci. Technol., 127, 283–292.
Yin, J.X., Chen, Y. & Meng, L.L. (2007) The influences of probiotics
on intestinal microflora in crucian carp (Carassius auratus). Fish.
Sci., 26, 610–612.
Zhang, Q., Ma, H., Mai, K., Zhang, W., Liufu, Z. & Xu, W. (2010)
Interaction of dietary Bacillus subtilis and fructooligosaccharide

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

Aquaculture Nutrition 18; 1–11 Ó 2011 Blackwell Publishing Ltd

on the growth performance, nonspecific immunity of sea
cucumber, Apostichopus japonicus. Fish Shellfish Immunol., 29,
204–211.
Ziaei-Nejad, S., Rezaei, M.H., Takami, G.A., Lovett, D.L., Mirvaghefi, A.R. & Shakouri, M. (2006) The effect of Bacillus spp.
bacteria used as probiotics on digestive enzyme activity, survival
and growth in the Indian white shrimp Fenneropenaeus indicus.
Aquaculture, 252, 516–524.


Aquaculture Nutrition
2012 18; 12–20

doi: 10.1111/j.1365-2095.2011.00871.x

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

1,2
5
1

3
6,7

1
3

4
1

1

2

NIFES, National Institute of Nutrition and Seafood Research, Bergen, Norway; MarinPet AS, Stavanger, Norway;
3
Institute of Marine Research, Matre Aquaculture Research Station, Matredal, Norway; 4 Institute of Marine Research,
Bergen, Norway; 5 Bergen High-Technology Center, Department of Biology, University of Bergen, Bergen, Norway;
6
Department of Biology, Ghent University, Gent, Belgium; 7 Skretting Aquaculture Research Centre, Stavanger, Norway

Correspondence: Laura Gil Martens, MarinPet AS, Sandvikveien 21, 4016

Stavanger, Norway. E-mail:

Inflammation is a non-specific protective mechanism towards
injury known to affect bone remodelling. This study aimed to
investigate the effect of FreundÕs complete adjuvant (FCA)
induced-inflammation on the prevalence of spinal deformities
of Atlantic salmon postsmolts fed with two different dietary
P levels. Sextuple groups of salmon postsmolts were fed with
either a low-phosphorous (6 g kg)1 available P, LP) or a
high-phosphorous (9 g kg)1 available P) diet for a period of
101 days. On Day 102, individually tagged fish were subjected to (i) single injection with FCA (0.125 mg kg)1 BW)
dissolved in phosphate-buffered saline (PBS) (ii) placebo
injection with PBS or (iii) sham injection (insertion of needle
only) or (iv) remained untreated. On Day 103, fish were given
a common diet for 174 days in seawater. No significant differences in body weight were observed. Injected fish, particularly the FCA group, had more compressions in the
injection site than untreated fish. No effect of diet and no
interaction between treatment and diet were observed. Severe
scoliosis was observed in 7% of FCA-injected individuals,
corresponding to a mixture of bone malformations in the tail
region. In conclusion, experimentally induced inflammation
may be an independent risk factor for bone deformities in
Atlantic salmon.
key words: Atlantic salmon, bone, FreundÕs complete
adjuvant, Freund adjuvant, inflammation, phosphorus,
platyspondyly,
spinal deformities
Received 12 November 2010, accepted 22 March 2011

Spinal deformities in farmed salmon are often a productionrelated disorder that represents an ethical issue for fish welfare. Additionally, economical losses mainly associated with
the downgrading of the fish at slaughterhouses (Michie 2001;

Sullivan et al. 2007) and loss of growth potential of affected
fish have also been reported (Hansen et al. 2009). Twenty
categories of spinal deformities have been classified based on
radiological findings in farmed salmon (Witten et al. 2009),
among them are compression and fusion-related deformities,
changes in bone radio-density, spinal curvatures, symmetry
deviations and displacement of vertebral bodies. The most
common phenotype in farming conditions is known as
platyspondyly (Kvellestad et al. 2000) consisting of vertebral
compression and development of heterotopic cartilage, which
replaces notochord tissue in the intervertebral space (Gil
Martens et al. 2005; Witten et al. 2005).
Many types of spinal deformities likely have a multifactorial causation, and several risk factors have been
described in the literature (Va˚gsholm & Djupvik 1998;
Waagbø et al. 2005; Gil Martens 2010). Among those are
genetics (Gjerde et al. 2005; Fjelldal & Hansen 2010), dietary
minerals and vitamins (Vielma & Lall 1998; Ornsrud et al.
2004; Lall & Lewis-McCrea 2007; Waagbø 2008; Fjelldal
et al. 2009), fast growth (Fjelldal et al. 2005, 2006, 2007),
vaccination (Berg et al. 2006; Aunsmo et al. 2008, 2009;
Koppang et al. 2008), incubation temperature (Ytteborg
et al. 2010) and water quality (Baeverfjord & Wibe 2003;
Wargelius et al. 2005; Gil Martens et al. 2006). Mineral

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

Ó 2011 Blackwell Publishing Ltd


malnutrition is among the risk factors that received attention

more than 30 years ago (Lall & Bishop 1977) and continues
to be investigated (Helland et al. 2005; Fjelldal et al. 2006,
2009; Lall & Lewis-McCrea 2007; Waagbø 2008). The dietary phosphorus concentration has been identified as a
critical factor in the development of bone deformities.
Fjelldal et al. (2009) showed that extra mineral supplementation in the early sea water phase reduced the prevalence of
vertebral deformities in fast-growing underyearling (0+)
smolts. Bæverfjord et al. (1998) and Helland et al. (2005)
showed that insufficient dietary phosphorus to Atlantic salmon postsmolt and juveniles resulted in undermineralized
vertebra but not in the morphological changes that characterize short-tailed salmon. Thus, although mineral supplementation may reduce the prevalence of deformities, it is not
necessarily a causal factor. Indeed, Fjelldal, P.G. (unpublished data) found that continuous light exposure in combination with insufficient P nutrition caused vertebral
compression in salmon smolts, whereas either treatment
alone did not.
Inflammation as a risk factor for spinal deformities in
Atlantic salmon was suggested for the first time by
Kvellestad et al. (2000) who observed the presence of
inflammatory cells in meningeal tissue from short-tailed fish.
Kent et al. (2004) also reported a link between chronic
inflammation of the spine and bone deformities in fish as a
result of a parasitic infection caused by metarcercaria and
myxozoans in cyprinid fishes. It is known from biomedical
research that inflammation can alter the normal pattern of
bone growth (Raisz 1999, 2005; Hughes et al. 2006). Bone
remodelling is regulated by multiple mechanisms that involve osteoblasts, osteocytes and osteoclasts and the interaction of systemic hormones and local factors such as local
cytokines, prostaglandins, growth factors and transcription
factors (Gil Martens 2010). It has been suggested that in
humans, an inflammation-related imbalance between these
factors and their inhibitors could alter the normal pattern
of bone remodelling and cause ankylosing spondylitis
(Gratacos et al. 1994). In salmon aquaculture, vaccination
is a common prophylactic procedure. Vaccines usually

contain one or more adjuvants to boost the immune response. Adjuvant compounds (from the Latin ÔAdjuvareÕ,
help) increase and prolong the immune responses towards a
given antigen and are also being used as vaccine delivery
systems (Singh & Hagan 2002). FreundÕs complete adjuvant
(FCA) is a water-in-oil emulsion with inactivated mycobacteria organisms that has been extensively used in
experimental animal models for inducing inflammation and
autoimmune diseases, particularly in models for adjuvant-

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

Aquaculture Nutrition 18; 12–20 Ó 2011 Blackwell Publishing Ltd

induced arthritis (AIA) in rats in which it induces an acute
periarticular inflammation followed by a phase of bone
involvement (Holmdahl et al. 2001).
In salmon aquaculture, fish are subjected to handling
(Poppe et al. 2007) which can cause extreme mechanical
loading to the spine resulting from, e.g., transfer of fish using
fish pumps, manual grading and vaccination procedures.
These handling procedures may lead to micro-fractures and
inflammation in the vertebrae (Robling et al. 2006; Witten &
Huysseune 2009). Rough handling conditions alone or in
combination with a poor mineralization of the skeleton may
lead to changes in the normal pattern of growth and
remodelling of the spine.
The present study was designed to investigate the combined effects of FCA-induced inflammation close to the spine
and insufficient dietary phosphorus on the potential development of spinal deformities in farmed salmon.

Individually tagged Atlantic salmon postsmolts (n = 600,
initial weight 200 g, T-bar tags; Hallprint Pty Ltd, Hindmarsh

Valley, SA, Australia), Aquagen strain, were distributed randomly among twelve seawater tanks (1 · 1 · 0.43 m diameter
fibreglass tanks) at the Institute of Marine Research facilities
at Matre, Hordaland, Norway. The water temperature was
kept stable at 12 °C (Day 1–45) and 16 °C (Day 46–101).
After a 14-day acclimatization period, sextuple groups of
fish were fed with experimental diets containing either 6 (low
phosphorus, LP) or 9 g kg)1 (high phosphorus, HP) available P for 101 days.
The fish were reared under a simulated natural photoperiod. All tanks were fed continuously during the light period
of the light dark cycle, by automatic feeders (ARVO-TEC T
Drum 2000; Arvotec, Huutokoski, Finland). For illumination, two 18-W fluorescent daylight tubes (OSRAM L 18W/
840 LUMILUX; OSRAM GmbH, Ausburg, Germany) were
used to produce 960 lx measured under water in the centre
of the tank. Photoperiod and feeding were controlled
automatically by a PC-operated system (Normatic AS,
Norfjordeid, Norway).
At Day 102, fish (500 g BW) from each dietary group were
anaesthetized with a solution containing 40 mg L)1 of benzocaine (BenzoakÒ Vet; A.C.D. SA, Braine-LÕalleud, Belgium) and subjected to one of the following treatments: (i)
single injection with FCA (Sigma-Aldrich, Buchs, Switzerland) dissolved in phosphate-buffered saline (PBS), (ii) sham


injection (insertion of needle only), (iii) placebo injection with
PBS or (iv) remained untreated. FreundÕs adjuvant (FCA)
is a water-in-oil emulsion containing dead mycobacterial
organisms and was injected at a dose of 0.125 mg FCA kg)1
BW. Individual sterile syringes (1 mL) with a detachable
sterile needle (27 G) were used for all fish. The total volume
of injected solution in PBS and FCA groups was 0.2 mL. All
injections were performed intramuscularly in a standardized
area close to the spine at the intersection between the lateral
line and a vertical line marked from the anterior edge of the

anal fin. This area corresponds to vertebrae 39–41. In addition, untreated fish within each dietary group remained as
negative control, leaving a total of eight groups. On Day 103,
all groups of fish were transferred to one common 5 · 5 m
sea cage, reared under natural light regime, and fed with one
common commercial diet for 174 days. The trial had a total
duration of 275 days. The fish were fed to satiation using
automatic feeders.

Two sets of diets containing 480/250 and 440/300 g kg)1
protein and g kg)1 lipids respectively were formulated based
on the same basal mix of raw materials and differed only in P
content (6 g kg)1 and 9 g kg)1 available P respectively).
These non-commercial test diets were produced in 3 and
4.5 mm pellet size at Skretting ARC facilities (Stavanger,
Norway). Diets were analysed for P content by the feed
supplier. The feed composition of the diets is indicated in
Table 1.
Table 1 Feed formulation (g kg)1): low phosphorus (6 g kg)1
available P) and high P (9 g kg)1 available P)
Low-P diet, High-P diet, Low-P diet, High-P diet,
3 mm
3 mm
4.5 mm
4.5 mm
Fish meal1
403
Soy protein
212
concentrate
Fish oil2

84
Rapeseed oil
109
Wheat grain
87
Wheat gluten
53
Vitamin mix
2
Mineral mix
1
Available P (g kg)1) 6
Water
50
Crude protein3
478
Fat3
247
1

403
212

414
178

414
178

84

109
87
53
2
1
9
50
478
247

96
127
92
40
2
1
6
50
441
294

96
127
92
40
2
1
9
50
441

294

Mix of Scandinavian, Mackerel and South American fishmeal.
Mix of Nordic and South American fish oil.
3
The theoretical values for protein and fat were 480/250 g kg)1
(3 mm) and 440/300 g kg)1 (4.5 mm), respectively.
2

Three hundred individuals per dietary P level (50 fish per
tank) were individually weighed at the start of the trial (Day
0), 1 day before injection (Day 101), and all remaining
individuals were weighed at the of the end seawater period
(Day 275).
FultonÕs condition factor (CF) was calculated as
100 W L)3, where W is live body weight (g) and L is fork
length (cm) (Busacker et al. 1990). Specific growth rate (SGR,
% day)1) was calculated as (eq ) 1)100 (Houde & Schekter
1981), where q = [ln (W2) ) ln (W1)] (t2 ) t1))1 (Bagenal &
Tesch 1978) and W2 and W1 are average live body weight at
times t1 and t2 respectively. Length growth (LG, mm day)1)
was calculated as (L2 ) L1) (t2 ) t1))1, where L2 and L1 are
average fork length at times t1 and t2 respectively.
Sampled fish from Day 101 and 275 were sedated with a
solution containing 40 mg L)1 of benzocaine and killed by
a blow to the head. The rest of the fish were examined for
external signs of deformities and individually measured,
weighed and radiographed before injection (Day 101,
n = 72 fish) and at the end of seawater period (n = 331,
Day 275).

Vertebral columns were radiographed by using a portable
X-ray apparatus (HI-Ray 100; Eickenmeyer Medizintechnik
fu¨r Tierarzte e.K., Tuttlingen, Germany) and 30 · 40 cm film
(Fujifilm IX 50; Fujifilm Corp.,Tokyo, Japan) following the
protocol described in Fjelldal et al. (2009). The number of
sampled fish per group varied between 35 and 45 individuals
per experimental group. The vertebral malformations were
diagnosed visually following the classification provided by
Witten et al. (2009). Vertebrae 36–46 were selected as the
most likely region where effects of treatment could be
expected (around the area of injection). Morphology of
the vertebrae was used as an indicator for mineral (P) status
in the present study (Day 101) based on previous results
(Fjelldal et al. 2007) The prevalence of spinal deformities was
measured as the ratio of the number of fish with compression
in this region and the severity of bone deformity as number
of compressed vertebra to the total number of vertebra per
fish. The dorso-ventral diameter of the vertebrae, used as an
early marker of incipient compression, was measured in
vertebra 36–46 according to the procedure described by
Fjelldal et al. 2005 and by means of image analysing software
(Image-J 1.43u, Wayne Rasband, National Institute of
Health, USA).

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

Aquaculture Nutrition 18; 12–20 Ó 2011 Blackwell Publishing Ltd

Somatic growth and bone morphometrics were statistically
evaluated according to either one-way (period days 1–101,
tank as experimental unit) or two-way ANOVA (period days
102–275, individual as experimental unit) followed by Tukeys
HSD test. Contingency table analysis using FisherÕs exact test
was performed to evaluate differences in the frequency of
deformities.

Table 2 Body weight, specific growth rate (SGR %), body length
and condition factor (CF) for Atlantic salmon (Salmo salar L.) at
start and after 101 days of feeding a low-phosphorus (6 g kg)1
available P) and high-phosphorus (9 g kg)1 available P) diet

Body weight (g)
Day 0
Day 101
SGR Day 0–101
Body length (cm)
Day 0
Day 101
CF
Day 0
Day 101

Low
phosphorus
(LP)

High
phosphorus

(HP)

One-way
P-value

260 ± 26
445 ± 42
0.53 ± 0.07

261 ± 27
491 ± 74
0.63 ± 0.10

0.93
0.15
0.06*

28.3 ± 1.0
33.2 ± 0.8

28.7 ± 1.0
33.2 ± 0.9

0.80
0.78

1.12 ± 0.07
1.20 ± 0.01

1.13 ± 0.07

1.21 ± 0.04

0.18
0.35

ANOVA

Values presented as mean ± SD (n = 6 tanks).
Significant effects determined by one-way ANOVA are indicated by
P-value.
*P < 0.1.

The data was analysed with the program GraphPad Prism
(version 5.0; GraphPad Software, La Jolla, CA, USA). The
significance level was set to P < 0.05, and the data are presented as mean ± SEM.

Specific growth rate during the feeding period (Day 101) was
0.53 ± 0.07 and 0.63 ± 0.10 in the low- and high-P group,
respectively (P-value 0.06) (Table 2: somatic growth Day
1–101). No significant differences in body weight (443 ± 38
and 448 ± 50), length (33.2 ± 0.8 and 33.2 ± 0.9 cm) or
CF (1.12 ± 0.07 and 1.13 ± 0.07) were observed between
dietary groups.
The fish grew from 450 to 2000 g (overall mean), but no
differences in weight, SGR or CF were observed in the period
after injection (Days 102–275). Fish fed with high P and
subjected to FCA injection were significantly shorter than the
other groups (Table 3); but these differences were already
observed on Day 101 (before treatment) (Table 3: somatic
growth Days 102–275).

All fish were examined externally at the end of the grow-out
period (Day 275; mean body weight, 2 kg). The overall
appearance of the fish was healthy with a CF of 1.4 ± 0.2,

Table 3 Body weight, specific growth rate (SGR %), body length and condition factor (CF) for Atlantic salmon (Salmo salar L.) at Day 102
and 275 of feeding in salmon fed with a low-phosphorus (LP) and high-phosphorus (HP) diet for 101 days and subjected to different treatments
on Day 102: (i) intramuscular injections of Freunds adjuvant (LP- and HP-Freund), (ii) sham-injection (LP- and HP-sham), (iii) placeboinjection (LP- and HP-placebo) or (iv) no treatment (LP- and HP-control)
Two-way

LP control
Body weight (g)
Day 102
457
Day 275
2044
SGR (%)
0.8
Day 102–275
Length (cm)
Day 102
33.6
Day 275
51.9
CF
Day 102
1.2
Day 275
1.4

LP sham

LP placebo

LP Freund

HP control

HP sham

HP placebo HP Freund

Diet

± 68
434 ± 54
465 ± 52
425 ± 48
442 ± 71
447 ± 55
466 ± 74
410 ± 43 0.27
± 250 2098 ± 220 2029 ± 165 2092 ± 255 2034 ± 214 2182 ± 250 2054 ± 245 1798 ± 159
± 0.1
0.9 ± 0.1
0.8 ± 0.1
0.9 ± 0.1
0.8 ± 0.1
0.9 ± 0.1
0.8 ± 0.1

0.8 ± 0.1 0.17

± 1.5
± 1.8
± 0.04
± 0.06

32.7 ± 0.9
52.3 ± 1.3
1.2 ± 0.02
1.4 ± 0.07

33.7 ± 1.1
51.6 ± 1.0
1.2 ± 0.05
1.4 ± 0.09

32.7 ± 0.7
52.0 ± 1.5
1.2 ± 0.07
1.5 ± 0.08

33.2 ± 1.5
50.9 ± 2.0
1.2 ± 0.04
1.4 ± 0.05

33.2 ± 1.1
52.3 ± 2.5
1.2 ± 0.05

1.5 ± 0.09

34.0 ± 1.0
52.0 ± 1.8
1.2 ± 0.01
1.4 ± 0.05

32.1 ± 1.4
48.6 ± 3.1

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

Aquaculture Nutrition 18; 12–20 Ó 2011 Blackwell Publishing Ltd

Treat

Diet ·
treat

0.48

0.23

0.31

0.95

0.09* 0.07* 0.11

1.1 ± 0.08 0.20

1.4 ± 0.12

Values presented as mean ± SD (n = 90 fish per treatment, Day 102 and n = 30 fish per treatment on Day 275).
Significant effects determined by two-way ANOVA are indicated by P-value.
*P < 0.1.

ANOVA

0.76

0.19


(a)

(b)

(c)

(d)

Figure 1 Type of spinal deformities
observed in Atlantic salmon (Salmo
salar) fed a low phosphorus (LP) and
high phosphorus (HP) diet for 101 days
and subjected to different treatments on
Day 275. (a) One-sided vertebral compression; (b) multiple compression; (c)
multiple fusion; (d) vertebrae shift, body
vertebrae abnormalities, small size; (e)
scoliotic spine with compressed and

under-mineralized vertebrae.

8

Untreated control
15

Sham injection
Placebo injection
FCA injection

6

4

2

Severity of deformity
(% deformed vertebra
in deformed fish)

Prevalence of compression in v36-46
(No. of fish with compressions)

(e)

Untreated control
Sham injection
Placebo injection
FCA injection

10

5

0
Low P

High P

0
Low P

Figure 2 Prevalence of spinal deformations in Atlantic salmon
(Salmo salar) fed a low phosphorus (LP) and high phosphorus (HP)
diet for 101 days and subjected to different treatments on Day 102:
(i) intramuscular injections of Freunds adjuvant (LP- and HP-Freund), (ii) sham-injection (LP- and HP-sham), (iii) placebo-injection
(LP- and HP-placebo) or (iv) no treatment (LP- and HP-control)
calculated based on X-rays on Day 275. Untreated fish from both
dietary groups presented almost no vertebral compressions in vertebral area v36–46.

but a few individuals (overall prevalence 3%) were visibly
deformed, independent of diet and injection treatment. By
radiology, the overall prevalence of fish with vertebral malformations (injected and non-injected groups together) was
2.5% and 16.4% at the end of feeding period (Day 101) and
end of the experiment (Day 275), respectively.
The vertebral malformations consisted mainly of compressions and fusions (type 6) and vertebral compressions (type 3–5)
as categorized by Witten et al. (2009) (Fig. 1). However, other
abnormalities in the vertebrae body shape and on bone density
(radio-translucent vertebrae and biconcavity, type 10 and 11)

were observed in few individuals but not quantified.

High P

Figure 3 Severity of bone deformity defined as % deformed vertebrae in deformed fish in Atlantic salmon (Salmo salar) on Day 275.

No effect of diet on the total prevalence of deformities
(P = 0.36) was observed, corresponding to 13% and 17% in
the low- and high-P group, respectively. However, deformities were mainly seen in the vicinity of the injection site
(vertebrae 36–46). The vertebral compressions observed in
this region on Day 275 were mainly found in ÔtreatedÕ fish
(sham-, placebo and FCA-injected fish) showing a local effect
of injection on the prevalence of vertebral compressions
(P = 0.008) (Fig. 2). The severity of deformity (Fig. 3)
ranged between 4% and 12%, and there were no effects of
diet or treatment.
Bone morphometrics, using ratio of the length to the
dorso-ventral vertebral body, showed no significant differences (P = 0.90) between dietary groups after 101 feeding
days. The average dorso-ventral diameter of vertebrae 36–48
was 0.99 ± 0.015 and 0.98 ± 0.013, in the low-P and high-P
groups, respectively. At the end of the trial (Day 275), fish

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

Aquaculture Nutrition 18; 12–20 Ó 2011 Blackwell Publishing Ltd


Ratio length/dorso-ventral
diameter of vertebra 39

1.02

1.00

(a)

Untreated control
Sham injection
Placebo injection
FCA injection

1.04

a
ab

a
ab

ab

ab
b

b

0.98

(b)
0.96

Low P

High P

Figure 4 Bone morphometrics of vertebra 39 (injection point) in
Atlantic salmon (Salmo salar) on Day 275. Different superscript
indicate significant differences (P < 0.05).

(c)

subjected to injection had a significant lower dorso-ventral
diameter of vertebra 39 (P = 0.02), compared with
untreated fish (Fig. 4). In addition, fish immunostimulated
with FreundÕs adjuvant had a lower dorso-ventral diameter
than the LP-Control group (P = 0.001). No interaction
between diet and treatment was observed (P = 0.51). There
was no effect of treatment or diet on any other vertebra in the
area of injection.
Six FCA-injected fish (Fig. 5) presented a severe degree of
scoliosis in the tail region (the prevalence of scoliosis in the
whole fish population was 1.8%), irrespectively of diet. The
prevalence of scoliosis within FCA-injected group, including
high- and low-phosphorus groups, was 7.3%.
These six scoliotic individuals displayed severe multiple
vertebral malformations in the tail region (region 3 and 4),
affecting the shape of the body vertebra, consisting of heterogeneous vertebral size, vertebral dislocation, hypertranslucidy and biconcavity and abnormal inter-vertebral
space (mixed type of bone malformations described as categories 10–16 in Witten et al. (2009). This deformity was not
seen in fish from any other experimental group.

The major findings of this experiment are that the amount of

dietary phosphorus, per se, given to postsmolt Atlantic salmon for a period of 101 days had no effect on the prevalence
of spinal deformities; moreover, there was no interaction
between dietary phosphorus and induced inflammation.
However, the increase of deformed fish in groups that had
been injected strongly suggests that injection alone (needle,
placebo and FCA) represented a risk for a vertebral deformity.

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

Aquaculture Nutrition 18; 12–20 Ó 2011 Blackwell Publishing Ltd

(d)

(e)

Figure 5 Type of spinal deformities observed in Atlantic salmon
(Salmo salar) injected with Freund adjuvant solution in vertebral
region 3 on Day 275. (a) Non- deformed fish; (b) deformed fish
presented scoliosis; (c) vertebral column of scoliotic fish; (d) region 3
of scoliotic fish; (e) five scoliotic individuals observed in FreundÕs
complete adjuvant-injected group.

Mineral nutrition has been highlighted as relevant field in
the pathogenia of bone malformations in farmed Atlantic
salmon. Special emphasis has been put on dietary phosphorus requirements of fast-growing underyearlings (Helland
et al. 2005; Waagbø et al. 2005; Lall & Lewis-McCrea 2007),
because fish reared under intensive farming conditions may
be more prone to developing bone deformities. Fjelldal et al.
(2006) found that underyearling (0+) Atlantic salmon (Salmo salar L.) smolts had a lower vertebral mineral content, a
reduced mechanical strength and a higher prevalence of

vertebral deformities than 1+ smolt during the early sea-


water (SW) phase. Indeed, increased mineral supplementation during the early seawater phase reduced the prevalence
of vertebral deformities in fast growing 0+ salmon smolts
(Fjelldal et al. 2009). So far, no studies have established a
direct relationship between suboptimal dietary phosphorus
supply and vertebral malformations in Atlantic salmon.
Fjelldal, P.G. (unpublished data) found that 24-h light
exposure in combination with insufficient dietary phosphorus
resulted in more compressed vertebral bodies, measured as a
lower length/dorso-ventral diameter ratio, while either
treatment alone did not. These observations imply that bone
under-mineralization as such may not be the only factor
involved in the development of spinal deformities and that
bone malformations could result from the interaction of
different risk factors. The dietary phosphorus level of the
low-P group (6 g kg)1 available P) used in the present study
was similar to the P level used in the study by Fjelldal et al.
(2009), but no observations from radiology and length/
dorso-ventral diameter showed any indication of bone
deformities. However, the fish used in Fjelldal et al. (2009)
were transferred to sea in mid August at higher water temperature and day-length, and with much higher SGR than in
the present study. Indeed, the 0.5–0.6% SGR of the fish in
this study was rather low compared with an SGR of 2% used
in the study by Fjelldal et al. (2009), suggesting that an
interaction between high growth rate and insufficient mineralization may induce bone deformities.
In the present study, no interaction between diet and
injection was observed in connection with the prevalence of
spinal deformities or to the dorso-ventral diameter of the

vertebrae. However, fish given injections, particularly FCA
injected fish, had a more compressed vertebral phenotype in
the area of injection than untreated fish, possibly a result of
inflammation affecting bone tissue, or other tissues in vicinity
to the injection site (e.g. connective and muscle tissue, blood
supply and/or nervous tissue).
The activation of osteoclasts under inflammatory conditions in bone has been widely investigated in mammals (Raisz
1999, 2005). Less information is available for teleost fish, but
reviewing the available literature, Witten & Huysseune (2009)
suggest that activity of teleost osteoclasts is likely triggered
by the same inflammatory factors that also trigger the
activity of mammalian osteoclasts. Indeed, Tagaki & Kaneko
(1995) and Ostland et al. (1997) observed the activation of
multinucleated osteoclasts in salmonids connected to an
inflammatory process. FCA is known to be a potent proinflammatory adjuvant, which is commonly used in a model
of adjuvant induced-arthritis (AIA) and other autoimmune
diseases in mammals (Oliviera et al. 2007). Farmed salmon

are immunized with oil-adjuvanted vaccines, which are
known to cause inflammation in areas adjacent to the injection site (Midtlyng 1996); this may explain the link between
vaccination and deformities reported by Berg et al. (2006)
and Aunsmo et al. (2008). AIA is a chronic condition that
develops in two phases in rats: an acute periarticular
inflammation, followed by a phase of bone involvement. The
mycobacterium component of FCA promotes the differentiation of T cells into a Th1 sub-population, which leads to a
retarded hypersensibility reaction. In the absence of mycobacterium [incomplete Freund adjuvant (IFA)], T cells differentiate towards a Th2 population promoting antibody
production. In our study, a local inflammation and cytokine
activation resulting from FCA injection could have affected
the normal pattern of bone growth as indicated by the
measured alterations of the length/dorso-ventral diameter,

and caused the severe bone deformations observed in the six
scoliotic individuals injected with FCA. However, vertebral
compressions were also found in sham- and placebo-injected
fish indicating that the vertebral compressions may be associated with the injection, possibly because of an inflammatory response caused by penetrating the dermal barrier.
Similarly, Gil Martens et al. (2010) observed vertebral compressions and ankylosis in connection with the injection point
in both LPS-immunostimulated (lipopolysaccaride), and
sham-injected fish, suggesting that a localized inflammationrelated process could have triggered the development of
vertebral body malformations. It remains to be clarified if
inflammatory processes in the vicinity of the vertebral column in salmon directly affect the bone tissue or if the bone is
indirectly affected through changes in associated tissues such
connective tissue, blood vessels, nerves, adipose tissue or
notochord. Unfortunately, bone samples taken for assessment of inflammatory cytokines could not be analysed
because of force majeure.
Alternatively, the effect of injection observed in injected
fish on vertebra 39, and as reported in Gil Martens et al.
(2010) could have been caused by mechanical damage that
might have altered the normal pattern of bone growth. As
reviewed by Hall (2005), tension modulates the transformation of cartilage to fibrous tissue and then to bone
under the influence of IL-1b. Likewise other cytokines and
growth factors such as interleukin-6 (IL-6), IL-8, IL-15,
IL-17, tumour growth factor beta (TGF-b), granulocytemacrophage colony-stimulating factor (GM-CSF), granulocyte colony-stimulating factor and monocyte chemotactic
protein (MCP) are involved in mechanical transduction
and could induce neutrophil chemotaxis after mechanical
strain (Peterson & Pizza 2009). Wargelius et al. (2009,

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

Aquaculture Nutrition 18; 12–20 Ó 2011 Blackwell Publishing Ltd

2010) observed an up-regulation of MMP-13 (metalloproteinase 13) in deformed vertebrae of Atlantic salmon.
Matrix metalloproteinases are involved in the degradation
of extracellular matrix (ECM) and have been associated
with chronic inflammatory processes affecting bone tissue
(Takaishi et al. 2008).
However, the effect of injection was more severe in FCA
injected groups, suggesting that inflammation was indeed the
causal agent for the deformities observed. Inflammation
markers were not assessed in this trial, but previous studies
suggest that inflammation could be a causal agent for
changes in bone remodelling (Witten & Huysseune 2009; Gil
Martens 2010; Gil Martens et al. 2010).
In conclusion, insufficient dietary P was not an independent risk factor for development of spinal deformities in
postsmolt salmon, nor was any interaction between dietary
minerals and experimentally induced inflammation found.
Enhanced levels of dietary phosphorous could not prevent
the injection-related development of spinal deformities.
However, injection alone triggered the development of spinal
deformities. An inflammatory process resulting from the
mechanical stimulation of the spine might have induced
changes in bone remodelling that ended in a spinal deformity;
thus, inflammation may be an independent risk factor for
bone deformities in Atlantic salmon. Further research using
reliable markers for inflammation must be performed to
establish inflammation as a risk factor for spinal deformity in
salmon.

The project has been funded by the Norwegian research
council (NFR) 172483/S40.

Aunsmo, A., Guttvik, A., Midtlyng, P.J., Larssen, R.B., Evensen, Ø.
& Skjerve, E. (2008) Association of spinal deformity and vaccineinduced abdominal lesions in harvest-sized Atlantic salmon, Salmo
salar L. J. Fish Dis., 31, 515–524.
Aunsmo, A., Øvretveit, S., Breck, O., Valle, P.S., Larssen, R.B. &
Sandberg, M. (2009) Modelling sources of variation and risk
factors for spinal deformity in farmed Atlantic salmon using
hierarchical- and cross-classified multilevel models. Prev. Vet.
Med., 90, 137–145.
Baeverfjord, G. & Wibe, A˚. (2003) Shorttail deformities in Atlantic
salmon-effect of freshwater production temperature. In: Beyond
Monoculture. Abstracts at Aquaculture Europe 2003, Trondheim.
EAS Special Publ. no. 27, pp. 3–4.
Bæverfjord, G., A˚sga˚rd, G. & Shearer, K.D. (1998) Development
and detection of phosphorus deficiency in Atlantic salmon, Salmo
salar L., parr and post-smolts. Aquac. Nutr., 4, 1–11.

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

Aquaculture Nutrition 18; 12–20 Ó 2011 Blackwell Publishing Ltd

Bagenal, T.B. & Tesch, F.W. (1978) Age and growth. In: Methods for
Assessment of Fish Production in Fresh Waters (Begenal, T. ed.),
3rd edn. IBP, Handbook No. 3, pp. 101–136. Blackwell Science
Publications, Oxford, UK.
Berg, A., Rødseth, O.M., Tangera˚s, A. & Hansen, T. (2006) Time of
vaccination influences development of adhesions, growth and
spinal deformities in Atlantic salmon Salmo salar. Dis. Aquat.
Organ., 69, 239–248.
Busacker, G.P., Adelaman, I.R. & Goolish, E.M. (1990) Growth.
In: Methods for fish Biology (Schreck, C.B. & Boyle, P.B. eds),

pp. 363–387. American Fisheries Society, Bethesda, MD.
Fjelldal, P.G. & Hansen, T. (2010) Vertebral deformities in triploid
Atlantic salmon (Salmo salar L.) underyearling smolts. Aquaculture, 309, 131–136.
Fjelldal, P.G., Nordgarden, U., Berg, A., Grotmol, S., Totland, G.,
Wargelius, A. & Hansen, T. (2005) Vertebrae of the trunk and tail
display different growth rates in response to photoperiod in
Atlantic salmon, Salmo salar L., post-smolts. Aquaculture, 250,
516–524.
Fjelldal, P.G., Lock, E.J., Grotmol, S., Totland, G.K., Nordgarden,
U., Flik, G. & Hansen, T. (2006) Impact of smolt production
strategy on vertebral growth and mineralisation during smoltification and the early seawater phase in Atlantic salmon (Salmo
salar, L.). Aquaculture, 261, 715–728.
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.
Fjelldal, P.G., Hansen, T., Breck, O., Sandvik, R., Waagbø, R.,
Berg, A. & Ørnsrud, R. (2009) Supplementation of dietary minerals during the early seawater phase increase vertebral strength
and reduce the prevalence of vertebral deformities in fast-growing
under-yearling Atlantic salmon (Salmo salar L.) smolt. Aquac.
Nutr., 15, 366–378.
Gil Martens, L. (2010) Mini-review: inflammation as a risk factor for
spinal deformities in farmed Atlantic salmon (Salmo salar L.).
J. Appl. Icthyol., 26, 350–354.
Gil Martens, L., Obach, A., Ritchie, G. & Witten, P.E. (2005)
Analysis of a short tail phenotype in farmed Atlantic salmon
(Salmo salar). Fish Vet. J., 8, 71–79.
Gil Martens, L., Witten, P.E., Fivelstad, S., Huysseune, A., Sævareid, B., Vikesa˚, V. & Obach, A. (2006) Impact of high water carbon dioxide levels on Atlantic salmon (Salmo salar L.): effects on
fish performance, vertebrae composition and structure. Aquaculture, 261, 80–88.
Gil Martens, L., Lock, E.J., Fjelldal, P.G., Wargelius, A., Araujo, P.,
Torstensen, B.E., Witten, P.E., Hansen, T., Waagbø, R. & Ørnsrud, R. (2010) Dietary fatty acids and inflammation in the vertebral column of Atlantic salmon (Salmo salar) smolts: a possible

link to spinal deformities. J. Fish Dis., 33, 957–972.
Gjerde, B., Pante, M. & Bæverfjord, G. (2005) Genetic variation
for a vertebral deformity in Atlantic salmon. Aquaculture, 244,
77–87.
Gratacos, J., Collado, A., Fillella, X., Sanmarti, R., Llena, J.,
Molina, R., Ballesta, A. & Munoz-Gomez, J. (1994) Serum cytokines (IL-6, TNFa, IL-b and IFN-c) in ankylosis spondylitis: a
close correlation between serum IL-6 and disease activity and
severity. Br. J. Rheumatol., 33, 927–931.
Hall, B.K. (2005) Initiating skeletal growth. In: Bones and Cartilage:
Developmental and Evolutionary Skeletal Biology (Hall, B.K. ed.),
pp. 385–394. Elsevier Academic Press, San Diego CA.
Hansen, T., Fjelldal, P.G., Yurtseva, A. & Berg, A. (2009) The effect of
vertebral deformities on growth in Atlantic salmon (Salmo salar L.).


Abstract presented at the Workshop Interdisciplinary Approaches
in Fish Skeletal Biology, 27–29 April 2009, Tavira, Portugal.
Helland, S., Refstie, S., Espmark, A˚., Hjelkde, K. & Baeverfjord, G.
(2005) Mineral balance and bone formation in fast-growing
Atlantic salmon parr (Salmo salar) in response to dissolved metabolic carbon dioxide and restricted dietary phosphorus supply.
Aquaculture, 250, 364–376.
Holmdahl, R., Lorentzen, J.C., Lu, S., Olofson, P., Wester, L.,
Holmerg, J. & Petterson, U. (2001) Arthritis induced in rats with
nonimmunogenic adjuvants as models for rheumatoid arthritis.
Immunol. Rev., 184, 184–202.
Houde, E.D. & Schekter, R.C. (1981) Growth rates, rations and
cohort consumption of marine larvae in relation to prey concentrations. Raap PV Reun.-Cons.Int. Explor. Mer., 178, 441–443.
Hughes, F.J., Turner, W., Belibasakis, G. & Martuscelli, G. (2006)
Effect of growth factors and cytokines on osteobast differentiation.
Periodontol 2000, 41, 48–72.

Kent, M.L., Watral, V.G., Whipps, C.M., Cunningham, M.E., Criscione, C.D., Heidel, J.R., Curtis, L.R., Spitsbergen, J. & Markle,
D.F. (2004) A digenean metacercaria (Apophallus sp.) and a
myxozoan (Myxobolus sp.) associated with vertebral deformities in
cyprinid fishes from the Willamette River, Oregon. J. Aquat. Anim.
Health, 16, 116–129.
Koppang, E.O., Bjerkas, I., Haugarvoll, E. et al. (2008) Vaccinationinduced systemic autoimmunity in farmed Atlantic salmon.
J. Immunol., 181, 4807–4814.
Kvellestad, A., Høie, S., Thorud, K., Tørud, B. & Lingøy, A. (2000)
Platyspondily and shortness of vertebral column in farmed
Atlantic salmon, Salmo salar, in Norway – description and interpretation of pathological changes. Dis. Aquat. Organ., 39, 97–108.
Lall, S.P. & Bishop, F.J. (1977) Studies on mineral and protein utilization by Atlantic salmon (Salmo salar) grown in sea water. Fish.
Mar. Serv. Res. Dev. Tech. Rep., 688, 16.
Lall, S.P. & Lewis-McCrea, L.M. (2007) Role of nutritents in skeletal
metabolism and pathology in fish – an overview. Aquaculture, 267,
3–19.
Michie, I. (2001) Causes of downgrading in the salmon farming
industry. In: Farmed Fish Quality (Kestin, S.C. & Warriss, P.D.
eds), pp. 405–406. Blackwell, Oxford, UK.
Midtlyng, P.J. (1996) A field study on intraperitoneal vaccination of
Atlantic salmon (Salmo salar L.) against furunculosis. Fish Shellfish Immunol., 6, 553–565.
Oliviera, P.G., Brenol, C.V., Edelweiss, M.I., Meurer, L., Brenol,
J.C.T. & Xavier, R.M. (2007) Subcutaneous inflammation
(panniculitis) in tibio-tarsal joint of rats inoculated with complete
FreundÕs adjuvant. Clin. Exp. Med., 7, 184–187.
Ornsrud, R., Gil, L. & Waagbø, R. (2004) Teratogenicity of elevated
egg incubation temperature and egg vitamin A status in Atlantic
salmon, Salmo salar L. J. Fish Dis., 27, 213–223.
Ostland, V.E., McGrogan, D.G. & Ferguson, H.W. (1997) Cephalic
osteochondritis and necrotic scleritis in intensively reared salmonids
associated with Flexibacter psychrophilus. J. Fish Dis., 20, 443–451.

Peterson, J.M. & Pizza, F.X. (2009) Cytokines derived from cultured
skeletal muscle cells after mechanical strain promote neutrophile
chemotaxis in vitro. J. Appl. Physiol., 106, 130–137.
Poppe, T., Bæverfjord, G. & Hansen, T. (2007) Effects of intensive
production with emphasis on on-growing production: fast growth,
deformities and production-related diseases. In: Aquaculture
Research: From Cage to Consumption (Thomassen, M., Gudding,
R., Norberg, B. & Jørgensen, L. eds), pp. 120–135. The Research
Council of Norway, NO-131 Oslo, Norway.
Raisz, L.G. (1999) Physiology and pathophysiology of bone
remodelling. Clin. Chem., 45, 1353–1358.

Raisz, L.G. (2005) Pathogenesis of osteoporosis: concepts, conflicts,
and prospects. J. Clin. Invest., 115, 3318–3325.
Robling, A., Castillo, A.B. & Turner, C.H. (2006) Biomechanical
and molecular regulation of bone remodelling. Annu. Rev. Biomed.
Eng., 8, 455–498.
Singh, M. & Hagan, D.T. (2002) Recent advances in vaccine adjuvants. Minireview. Pharm. Res., 19, 715–728.
Sullivan, M., Hammond, G., Roberts, R.J. & Manchester, R.J.
(2007) Spinal deformation in commercially cultured Atlantic salmon, Salmo salar L.: a clinical and radiological study. J. Fish Dis.,
30, 745–752.
Tagaki, Y. & Kaneko, T. (1995) Developmental sequence of boneresorbing cells induced by intramuscular implantation of mineralcontaining bone particles into rainbow trout, Oncorhynchus
mykiss. Cell Tissue Res., 280, 153–158.
Takaishi, H., Matsuki, T., Nakazawa, A., et al. (2008) Imbalance in
intestinal microflora constitution could be involved in the pathogenesis of inflammatory bowel disease. Int. J Med Microbiol, 298,
463–472.
Va˚gsholm, I. & Djupvik, H. (1998) Risk factors for spinal
deformities in Atlantic salmon (Salmo salar L.). J. Fish Dis., 21,
47–53.
Vielma, J. & Lall, S.P. (1998) Phosphorus utilization by Atlantic

salmon (Salmo salar) reared in freshwater is not influenced by
higher dietary calcium intake. Aquaculture, 160, 117–128.
Waagbø, R. (2008) Reducing production-related diseases in
farmed fish. In: Improving Farmed Fish Quality and Safety (Lie, Ø.
ed.), pp. 363–398. Woodhead Publishing Limited, Cambridge, UK.
Waagbø, R., Kryvi, H., Breck, O. & Ørnsrud, R. (2005) Workshop on
Bone Disorders in Intensive Aquaculture of Salmon and Cod. Final
report. NIFES, Bergen, Norway.
Wargelius, A., Fjelldal, P.G. & Hansen, T. (2005) Heat shock
during early somitogenesis induces caudal vertebral column defects
in Atlantic salmon (Salmo salar). Dev. Genes Evol., 215, 350–357.
Wargelius, A., Fjelldal, P.G., Grini, A., Gil Martens, L., Kvamme,
B.O. & Hansen, T. (2009) MMP-13 (matrix metalloproteinase
13) may be a potential indicator of ECM remodeling in the
etiology of compressed vertebrae of Atlantic salmon (Salmo
salar L.). Abstract presented at the Workshop Interdisciplinary
Approaches in Fish Skeletal Biology, 27–29, April 2009, Tavira,
Portugal.
Wargelius, A., Fjelldal, P.G., Grini, A., Gil Martens, L., Kvamme,
B.O. & Hansen, T. (2010) MMP-13 (Matrix MetalloProteinase 13)
expression might be an indicator for increased ECM remodelling
and early signs of vertebral compression in farmed Atlantic salmon
(Salmo salar L.). J. Appl. Ichthyol., 26, 366–371.
Witten, P.E. & Huysseune, A. (2009) A comparative view on
mechanisms and functions of skeletal remodeling in teleost fish,
with emphasis on osteoclasts and their function. Biol. Rev. Camb.
Philos. Soc., 84, 315–346.
Witten, P.E., Gil Martens, L., Hall, B., Huysseune, A. & Obach, A.
(2005) Compressed vertebrae in Atlantic salmon (Salmo salar):
evidence for metaplastic chondrogenesis as a skeletogenic late in

ontogeny. Dis. Aquat. Organ., 64, 181–191.
Witten, P.E., Gil Martens, L., Huysseune, A., Takle, H. & Hjelde, K.
(2009) Towards a classification and an understanding of developmental relationships of vertebral body malformations in Atlantic
salmon (Salmo salar L.). Aquaculture, 295, 6–14.
Ytteborg, E., Bæverfjord, G., Torgersen, J., Hjelde, K. & Takle,
H. (2010) Molecular pathology of vertebral deformities in
hyperthermic Atlantic salmon (Salmo salar). BMC Physiol., 10,
12.

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

Aquaculture Nutrition 18; 12–20 Ó 2011 Blackwell Publishing Ltd


Aquaculture Nutrition
doi: 10.1111/j.1365-2095.2011.00872.x

2012 18; 21–34

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

1

1
4

1

1,2

1

3

1

Aquaculture Protein Centre, CoE, Department of Animal and Aquacultural Sciences, Norwegian University of Life Sciences, A˚s;
2
Nofima AS, A˚s; 3 BioMar AS, Trondheim; 4 Addcon Nordic AS, Porsgrunn, Norway

Two experiments were conducted to investigate effects of feed
processing conditions and potassium diformate (KDF) supplementation on apparent digestibility of nutrients in Atlantic
salmon (Salmo salar) and physical quality of extruded feed.
In Exp. 1, diets with raw or expander pretreated full-fat soybean meal (FFSBM) at 100 or 120 °C were extruded at 110 °C.
Expander pretreatment significantly (P < 0.05) improved the
digestibility of arginine, glutamine and tyrosine in Atlantic
salmon. The higher digestibility of expander pretreated
FFSBM was confirmed in mink (Mustela vison). In Exp. 2,
diets with defatted soybean meal (SBM) were extruded at 110,
130 or 150 °C. The results showed that increasing extrusion
temperatures significantly (P < 0.05) improved the digestibility of most major nutrients and amino acids in Atlantic
salmon. In general, KDF supplementation to FFSBM and
SBM diets did not affect digestibility of nutrients in Atlantic
salmon or mink. Expander pretreatment and increasing
extrusion temperatures increased pellet expansion, while KDF
supplementation reduced pellet expansion.
key words: digestibility, expander, extruder, physical quality, potassium diformate, soybeans
Received 16 November 2010, accepted 29 March 2011
Correspondence: Margareth Øverland, Aquaculture Protein Centre, CoE,
Department of Animal and Aquacultural Sciences, Norwegian University of

Life Sciences, P.O. Box 5003, NO-1432 A˚s, Norway. E-mail: margareth.


Soybeans are used as protein source in salmonid diets
because of its high crude protein content, reasonably

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

Ó 2011 Blackwell Publishing Ltd

balanced amino acid profile, abundance and low price per
protein unit (Hertrampf & Piedad-Pascual 2000; Storebakken et al. 2000; Gatlin et al. 2007). The potential of soybean
meal (SBM) for salmonids is, however, limited by the presence of several antinutrients (Storebakken et al. 2000;
Francis et al. 2001), of which one of the most important is
the heat-labile trypsin inhibitors (TI) (Liener 1994). Full-fat
soybean meal (FFSBM) has a higher content of TI than
conventional defatted SBM, which has been toasted following oil extraction (Hardy & Barrows 2002). In addition to the
temperature itself, the duration of heating and moisture
content are also important factors for reducing the TI
activity (TIA) and thereby improving the nutritional value of
soybeans (Zarkadas & Wiseman 2005). Leeson & Atteh
(1996) reported that the heat generated during extrusion was
sufficient to reduce TIA activity in FFSBM, and Romarheim
et al. (2006) showed that extrusion of feed was sufficient to
reduce TIA in flaked, defatted, untoasted soybeans (whiteflakes, WF) to less than 5 g kg)1 feed when dietary WF
inclusion was 250 g kg)1. Previous results have shown that a
dietary TIA level at 5 g kg)1 may be considered acceptable in
diets for Atlantic salmon (Olli et al. 1994). Heating aimed at
completely inactivate TI in the feed is not an option, as
overheating may reduce protein solubility and nutritional

value without further lowering the TI content of the feed
(Arndt et al. 1999; Iwe et al. 2001).
The FFSBM contains approximately 370 g kg)1 protein
and 190 g kg)1 fat, whereas defatted SBM contains on
average 460 g kg)1 protein and low levels of fat (£10 g kg)1)
(Hertrampf & Piedad-Pascual 2000). The fat content of
FFSBM may cause challenges during feed processing when
the total lipid level of the diet exceeds 70 g kg)1 (Rokey
2007). Moreover, high fat contents are known to negatively
influence physical quality of extruded feed (Sørensen et al.


2009). During extrusion processing, raw materials are cooked
by applying moisture, pressure, temperature and shear forces. Cooking initiates a series of physiochemical reactions,
including starch gelatinization (Colonna et al. 1989) and
protein unfolding (Phillips 1989), which affects the binding
between feed particles and hence physical quality (Thomas &
van der Poel 1996). Fat interferes with the binding between
particles, first by reducing the degree of cooking by lubricating the interface between the dough and the extruder
elements and secondly by acting as an insulating agent and
thus preventing water from being absorbed by the particles
(Lin et al. 1997). In intensive aquaculture production, it is
important that feed pellets are of high physical quality to
avoid feed losses and water emissions (Aarseth 2004; Aarseth
et al. 2006a; Sørensen et al. 2011). Additionally, physical
quality may also affect feed intake and nutrient utilization of
fish (Hilton et al. 1981; Baeverfjord et al. 2006; Venou et al.
2009) and should therefore be taken into consideration when
formulating fish feeds.
Organic acids salts have been reported to promote growth

and health performance in warm-blooded monogastric animals and fish because of their antimicrobial effects (Øverland
et al. 2000; Mikkelsen et al. 2009; Zhou et al. 2009). A recent
study with Atlantic salmon has shown that the addition of
potassium diformate (KDF) to a plant-based diet improved
protein digestibility (Storebakken et al. 2010). Furthermore,
interactive effects between KDF supplementation and heat
treatment showed that the digestibility of individual amino
acids in diets with KDF increased with increasing temperatures, while the digestibility decreased in diets without KDF.
Research investigating the effect of acid salts on digestibility
of plant-based diets for salmonids is, however, limited.
Therefore, the objective of this study was to investigate the
effects of ingredients and processing conditions on apparent
digestibility in Atlantic salmon and physical quality of feed
when (i) dietary fish meal was partially replaced with FFSBM
or SBM, (ii) without or with KDF supplemented to the diets
and (iii) pretreating FFSBM with an expander prior to
extrusion or extrusion of SBM at increasing temperatures.

Two experiments were carried out using either FFSBM
(Exp. 1) or defatted, solvent-extracted SBM (Exp. 2) as a
partial replacement for fish meal in the diets (Table 1). Each of
the two test ingredients made up approximately 260 g kg)1 of
the total crude protein content of the diet. FFSBM was either

added directly to the diets prior to extrusion or pretreated with
an expander. In Exp. 1, the dietary treatments were organized
according to a 3 · 2 factorial design. Three different feed
processing conditions were used [(i) extrusion at 110 °C without pretreatment; (ii) expander pretreatment at 100 °C prior to
extrusion at 110 °C; (iii) expander pretreatment at 122 °C prior
to extrusion at 110 °C], and two dietary treatments () or +

KDF) were applied. The effect of feed processing conditions of
FFSBM diets will be referred to as the effect of pretreatment
hereafter. In Exp. 2, the treatments were organized according
to a 3 · 2 factorial design with three extrusion temperatures
(110; 130; 150 °C) and two dietary treatments () or + KDF).

The FFSBM and SBM used in this study were derived from
the same batch of soybeans (Deno-SoyÒ; Denofa AS, Fredrikstad, Norway). From this batch, a subsample from whole
full-fat soybeans with hulls was taken out before the soybeans were processed into hexane-extracted and toasted
SBM with hulls. The dietary macroingredients (fish meal,
wheat, FFSBM and SBM) were thoroughly ground using a
hammer mill (Mu¨nch, HM 21.115, Wuppertal, Germany) to
pass a 3-mm screen. Diet formulations, ingredients and
analysed chemical composition of the diets are presented in
Table 1.
In Exp. 1, the FFSBM was mixed (Dinnissen Pegasus
Menger 400 l, Sevenum, The Netherlands) with wheat in a 2 : 1
ratio, preconditioned with steam to 91 °C and pretreated using
a 75-kWh OE15.2 Kahl expander (Amandus Kahl GmbH &
Co. KG, Reinbek, Germany) with cone pressures of 5 and
11 bar (bar; 1 bar = 105 Pa) to raise the temperature to 100
and 122 °C, respectively. The expandate was pelleted through
3-mm dies (Mu¨nch MDZAD8, MD 350 RPM, Wuppertal,
Germany) and cooled in a counter flow drier (Miltenz VD010
gas; Millbank Technology Ltd, Auckland, New Zealand) prior
to grinding through a 3-mm screen.
For both the experiments, dietary macroingredients (fish
meal, wheat and either FFSBM [untreated; pretreated at
100 °C; pretreated at 122 °C] or SBM) were thoroughly
mixed as 200 kg batches and ground using a hammer mill

with a 1-mm screen. The microingredients (vitamin and
mineral premix, yttrium oxide, and KDF) were manually
added to the mixture during a second mixing cycle. Following mixing, the diets were preconditioned for 62 s and
extruded through a 6-mm die (Sørensen et al. 2011). Two
different screw configurations were used to reach the targeted
extrusion temperatures. Screw configuration 1 (SCF1) was
built up from the following elements, from inlet to outlet:

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

Aquaculture Nutrition 18; 21–34 Ó 2011 Blackwell Publishing Ltd


Table 1 Diet formulation and analyzed contents of main nutrients, amino acids and TIA in the experimental diets
Experiment 1 – FFSBM-based diets
Extruded
110 °C
)

Treatment1

Expanded
100 °C
+

)

Experiment 2 – SBM-based diets
Expanded
120 °C

)

+

Extruded
110 °C

+

)

Extruded
130 °C
+

)

Extruded
150 °C
+

)

+

)1

Feed formulation (g kg )
Fish meal2
383

Fish oil3
202
Wheat starch4
126
FFSBM5
286
SBM6
0
Mineral and vitamin
3.8
premix7
Astaxanthin8
0.5
Yttrium oxide9
0.1
KDF10
0
Dry matter (DM) (g kg)1)
952
Chemical composition (g kg)1 DM)
Lipid
304
Ash
71
Starch
80
Crude protein
425
Total amino acids
408

Essential amino acids
Arginine
26.3
Histidine
9.9
Isoleucine
20.2
Leucine
33.1
Lysine
33.0
Methionine
11.0
Phenylalanine
19.1
Threonine
18.7
Valine
22.0
Non-essential amino acids
Alanine
23.0
Aspargine
43.8
Cysteine11
4.7
Glutamine
69.7
Glycine
21.3

Proline
19.0
Serine
20.1
Tyrosine
13.5
TIA12 (mg g)1 diet)
3.0

377
202
124
281
0
3.8

379
210
124
283
0
3.8

374
209
123
279
0
3.8

379
210
124
283
0
3.8

381
194
125
284
0
3.8

388
252
130
0
226
3.9

385
252
126
0
221
3.8

388
252

130
0
226
3.9

385
252
126
0
221
3.8

388
252
130
0
226
3.9

385
252
126
0
221
3.8

0.5
0.1
12
943

0.5
0.1
0
963

0.5
0.1
12
952

0.5
0.1
0
954

0.5
0.1
12
949

0.5
0.1
0
944

0.5
0.1
12
943

0.5
0.1
0
948

0.5
0.1
12
952

0.5
0.1
0
966

0.5
0.1
12
952

298
75
88
411
391

301
69
94

405
424

314
73
90
414
423

317
69
90
409
428

290
75
86
424
419

283
72
102
423
433

280
76
78

420
448

299
69
88
417
428

280
77
80
411
436

293
70
91
406
436

296
78
82
417
426

25.3
9.5
18.9

31.6
31.3
10.4
18.3
17.9
20.7

28.5
10.2
20.5
34.1
33.5
11.2
19.9
19.3
22.1

28.9
10.0
20.6
34.1
33.9
11.4
19.8
19.2
22.2

29.5
10.0
20.7

34.5
34.0
11.4
20.1
19.5
22.4

28.2
10.0
20.3
33.8
33.7
11.2
19.6
19.1
21.9

28.1
10.5
21.1
35.0
34.5
11.5
20.3
19.8
23.0

30.6
10.8
22.1

36.2
35.4
11.8
21.0
20.3
24.1

27.9
10.4
21.1
34.6
34.1
11.4
20.1
19.5
23.0

29.8
10.4
21.2
35.2
34.5
11.6
20.5
19.9
23.1

28.5
10.5
21.5

35.3
34.7
11.6
20.6
19.8
23.4

29.5
10.2
20.6
34.4
33.9
11.2
19.9
19.4
22.4

22.0
42.0
4.6
67.1
20.4
18.4
19.6
12.9
3.0

23.8
45.1
5.0

73.3
21.9
19.6
21.3
14.5
1.6

23.9
44.8
5.2
72.6
22.0
19.5
20.8
14.4
1.9

24.1
45.4
5.2
73.4
22.3
19.6
21.2
14.6
1.6

23.6
44.6
4.9

71.9
21.7
19.7
20.8
14.2
1.9

24.3
46.4
5.0
74.4
22.5
20.5
21.7
14.3
Na13

24.9
48.0
5.2
76.4
23.1
21.2
22.0
14.8
Na

24.1
45.9
5.0

73.4
22.3
20.5
21.2
14.1
Na

24.4
46.7
4.9
74.4
22.6
20.6
21.7
14.8
Na

24.5
46.7
5.0
74.8
22.7
20.7
21.6
14.5
Na

23.8
45.6
4.9

72.9
22.1
20.2
21.3
14.0
Na

FFSBM, full-fat soybean meal; KDF, potassium diformate; TIA, trypsin inhibitor activity.
Treatments without KDF are denoted Ô)Õ and treatments with KDF are denoted Ô+Õ.
2
NorsECO-LT, Egersund Sildoljefabrikk AS, Egersund, Norway.
3
NorSalmOil, Egersund Sildoljefabrikk AS, Egersund, Norway.
4
Whole wheat, Felleskjøpet Agri, Ski, Norway.
5
Deno-Soyâ, whole full-fat soybeans with hulls, Denofa AS, Fredrikstad, Norway.
6
Deno-Soyâ, hexane extracted and toasted soybean meal with hulls, Denofa AS, Fredrikstad, Norway.
7
Mineral and vitamin premix for fish, Normin AS, Hønefoss, Norway.
8
Carophyllâ Pink, DSM, JH Heerlen, The Netherlands.
9
Yttrium Oxide (Y2O3), Metal Rare Earth Limited, Shenzhen, China.
10
Formiâ, Potassium diformate, Addcon Nordic AS, Porsgrunn, Norway.
11
Cystine and cysteine.
12

Inhibited bovine trypsin.
13
Not analyzed.
1

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

Aquaculture Nutrition 18; 21–34 Ó 2011 Blackwell Publishing Ltd


feed production (Table 2). Fish oil was added to the dried
pellets in amounts of 200–260 g kg)1 using a vacuum coater
(Dinnissen, Sevenum, The Netherlands). The feed was stored
at 4 °C until analyses of physical feed quality, and the
digestibility trials were conducted.

80R80-80R80-60R60-60R60-60R60-60R60-60R60-80R80100R100-P120-60L20-(90° twist off)-60L20-100R100-80R8080R80-80R80-60R60-60R60. The first number gives the flight
length, ÔRÕ denotes a forward conveying element, ÔLÕ indicate
a backward conveying element and the last number gives the
screw element length (mm). The element ÔP120Õ is a polygon
block (kneading element). Screw configuration 2 (SCF2) was
similar to SCF1, except that the two Ô60L20Õ elements were
continuous to each other. In Exp. 1, the untreated FFSBM
was extruded using SCF1, whereas the pretreated FFSBM
were extruded using SCF2. In Exp. 2, SBM-based diets were
extruded using SCF1, and the extruder temperature was
increased from 110 to 130 and 150 °C by increasing the screw
revolutions per minute (rpm). Samples of feed were collected
at steady-state conditions in the extruder and dried in a fluidized bed dryer (Bu¨hler OTW-50, Uzwil, Switzerland).
According to the method described by Sørensen et al. (2011),

the final water content in the feeds was adjusted in small
dryers to approximately 930 g kg)1 dry matter (DM). Extruder parameters were continuously recorded during the

The salmon experiment was carried out at Helgeland
Forsøksstasjon (Dønna, Norway) during the period December 2008–February 2009. Apparent digestibility was measured
in the six FFSBM-based diets from Exp. 1 and the six SBMbased diets from Exp. 2. All experimental diets contained
0.1 g kg)1 yttrium oxide as an indigestible marker (Austreng
et al. 2000). A total of 1644 Atlantic salmon with mean initial
weight of 1500 ± 200 g were distributed to twelve sea pens,
5.5 · 5.5 m wide and 7 m deep. The sea temperatures ranged
from 4.4 to 6.8 °C during the experimental period. The fish
had been fed a commercial diet (BioMar AS, Trondheim,
Norway) prior to the experiment.

Table 2 Processing parameters during extrusion of the experimental diets
Experiment 1 – FFSBM-based diets
Extruded
110 °C
Treatment1

)

Preconditioning
Temperature (°C)
88
Moisture addition (kg h)1)
Water
26
Steam
13

Extrusion
Throughput2 (kg h)1)
226
Temperature3 (°C)
Section 1
70
Section 3
115
Section 5
111
Die
101
Screw speed (rpm)
274
Screw configuration4
SCF1
Die pressure (bar)
21
Torque (Nm)
204
Cutter speed (rpm)
1801
SME5 (Wh kg)1)
26
Vacuum coating
Oil addition6 (g kg)1)
210

Expanded
100 °C

)

+

Experiment 2 – SBM-based diets
Expanded
120 °C
)

+

Extruded
110 °C
)

+

Extruded
130 °C
)

+

Extruded
150 °C
)

+

+

85

87

85

84

86

87

88

100

100

100

100

27
12

29
11

29

11

28
11

28
10

42
15

41
15

35
12

41
12

38
12

36
13

227

223

230

228

228

242

241

235

291

244

296

64
115
113
98
328
SCF1
17
192
1801
29

68

114
114
100
275
SCF2
19
216
1802
35

71
115
116
105
287
SCF2
17
194
1803
25

59
115
110
94
340
SCF2
18
218
1801

34

73
115
110
106
336
SCF2
18
203
1801
31

69
114
117
101
166
SCF1
30
346
2004
25

71
116
115
97
202
SCF1

26
341
2004
29

98
132
125
132
589
SCF1
40
304
2004
80

99
126
124
136
666
SCF1
40
313
2204
75

99
140
133

148
877
SCF1
42
301
2004
114

99
145
134
150
818
SCF1
47
326
2205
95

210

220

220

220

200

260

260

260

260

260

260

FFSBM, full-fat soybean meal; KDF, potassium diformate.
Treatments without KDF are denoted Ô)Õ and treatments with KDF are denoted Ô+Õ.
2
Dry feed mash and water added in the preconditioner.
3
Temperature is measured by sensors mounted on the extruder barrel.
4
Screw configuration (SCF1 = screw configuration 1; SCF2 = screw configuration 2).
5
Specific mechanical energy.
6
Diets were coated with different levels of fish oil in order to obtain sinking pellets.
1

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

Aquaculture Nutrition 18; 21–34 Ó 2011 Blackwell Publishing Ltd

Each of the 12 diets was fed to one group of fish over
the course of three feeding periods to obtain three replicates. Each feeding period was initiated by randomly
switching the 12 diets to the 12 sea cages, ensuring that no
fish were fed the same diet twice throughout the experimental period. The fish were manually fed to satiation
once per day during the first feeding period because of
limited number of daylight hours and twice per day during
the second and third feeding periods. Uneaten feed was
collected by a lift-up system (LiftUP Akva AS, Fusa,
Norway), and feed intake was recorded each day during
the experiment. The three feeding periods were targeted to
last for 10 days each; however, harsh weather conditions at
the research station affected the number of feeding days
for each period. To obtain 10 feeding days in each of the
three feeding periods, each period lasted for 28, 19 and
11 days, respectively.
At termination of a feeding period, fish were collected
using a net and anesthetized with 60 mg L)1 tricane methanesulfonate (MS-222) dissolved in salt water. Approximately 40 fish from each cage were manually stripped for
faeces by gently applying pressure to the lower abdominal
region to express faecal material in a tray and carefully
ensuring that urine was excluded from the sample (Austreng
1978). No fish was stripped more than once in the course of
the experimental period. Stripped fish were marked by either removing the adipose fin (1st stripping) or making a
mark in the gill cover (2nd stripping) before releasing the
fish back into the sea cage. Faecal samples were immediately frozen after the stripping, freeze-dried and ground
prior to analyses.

Because the digestibility trial with salmon had high withintreatment variability, it was decided to verify the digestibility
results with mink, using the six FFSBM-based diets from
Exp. 1. Mink was chosen as a test species because of the close
relationships in digestibility capacities between mink and

salmonids (Skrede et al. 1980, 1998; Romero et al. 1994;
Øverland et al. 2006). The mink experiment was carried out
at the Department of Animal and Aquacultural Sciences
(UMB, A˚s, Norway). A total of 24 adult male mink of the
genotype Standard dark with a weight ranging from 2022 to
2755 g (x = 2391, r = 210) were used in the experiment.
The feeding trial lasted for 9 days and consisted of a 5-day
preliminary period and a 4-day faecal collection period. The
experiment was otherwise conducted as described by Aslaksen et al. (2006).

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

Aquaculture Nutrition 18; 21–34 Ó 2011 Blackwell Publishing Ltd

All physical quality measurements were undertaken at room
temperature and performed in triplicate on uncoated pellet
samples unless otherwise stated. Hardness, diameter, expansion ratio and Holmen durability index (HDI) were measured
according to Sørensen et al. (2011) and reported as the
average of 3 · 30 pellets, except for HDI. Doris durability
was analysed on coated pellet samples, using a Doris tester
(AKVAsmart, Bryne, Norway). Sifted pellets (350 g) were
loaded into the Doris tester, conveyed through the apparatus, loaded onto a set of 5.6-, 3.55-, 2.36- and 0-mm screens
and sieved for 60 s at amplitude 0.5 in a sieving machine
(Retsch AS 200 Control, Haan, Germany). Doris durability
was calculated as the percentage of whole pellets remaining
on the 5.6-mm screen. Water stability index (WSI) was
measured on coated pellets incubated for 120-min intervals in
a shaking water bath at 23 °C, according to Baeverfjord
et al. (2006).

Diets and freeze-dried faecal samples were analysed for DM
(Commission dir. 71/393/EEC), ash (Commission dir. 71/
250/EEC), crude lipid by HCl hydrolysis followed by diethylether extraction (Commission dir. 98/64/EC) and starch
(AOAC enzymatic method 996.11). The diets were defatted
by acetone prior to the starch analysis. The nitrogen content
in diets was determined by the Kjeldahl method (Commission dir. 93/28/EEC). As a result of low amount of faecal
material, the nitrogen content in faeces was determined by
the Dumas method as described by Denstadli et al. (2006).
Crude protein was calculated from the nitrogen content in
the material (N · 6.25). Amino acids in diets and faeces
(Commission dir. 98/64/EC) were analysed on a Biochrom 30
amino acid analyzer (Biochrom Ltd., Cambridge, UK).
Samples for analysis of yttrium oxide were prepared as
described by Denstadli et al. (2006) and analysed with an
ICP-AES Perkin Elmer Optima 5300 DV (Perkin Elmer Inc.,
Shelton, CT, USA). The FFSBM diets from Exp. 1 were
analysed for TIA in accordance with the method of
Hamerstrand et al. (1981). The pH value of the diets was
measured as described by Pandey & Satoh (2008).

Apparent digestibility of main nutrients and individual
amino acids in diets fed to Atlantic salmon and mink was
calculated as: 100 ) (100 MD (MF))1 NF (ND))1), where M