Aquaculture 261 (2006) 80 – 88
www.elsevier.com/locate/aqua-online
Impact of high water carbon dioxide levels on Atlantic salmon
smolts (Salmo salar L.): Effects on fish performance, vertebrae
composition and structure
Laura Gil Martens a,⁎, P. Eckhard Witten b , Sveinung Fivelstad c , Ann Huysseune d ,
Bjarte Sævareid a , Vibeke Vikeså a , Alex Obach a
a
c
Nutreco Aquaculture Research Centre AS, Sjøhagen 3, P.O. Box 48, N-4001 Stavanger, Norway
b
AKVAFORSK (Institute of Aquaculture Research), N-6600 Sunndalsøra, Norway
Laboratory of Environment, Bergen College of Engineering, Nygårdsgaten 112, P.O. Box 6030, N-5020 Bergen, Norway
d
Ghent University, Vakgroep Biologie, Ledeganckstraat 35,B-9000 Ghent, Belgium
Received 8 July 2005; received in revised form 2 June 2006; accepted 8 June 2006
Abstract
The role of high carbon dioxide (CO2) levels on fish performance, bone structure/composition and as a potential cause of spinal
deformities was studied. Two groups of fish were exposed to a low (control) and a high level of CO2 for 135 days during the
freshwater period. After smoltification, the fish were transferred to seawater and followed up for 517 days until they reached
harvest weight (3.1 kg BW).
Differences in body weight between the control and high CO2 groups were observed. At the end of the freshwater period,
average weight in the group exposed to high CO2 levels was 20.9% lower than in the control group. Specific growth rates (SGR)
from the start of the experiment (10 g BW) to smolt stage were 1.63 ± 0.04 and 1.36 ± 0.01 for the control group and the high CO2
group, respectively. Differences in body weight were maintained during the initial stages of the seawater period, but were not
observed at harvest weight.
Nephrocalcinosis was not observed in any of the experimental groups at the end of the freshwater period and no external signs
of spinal deformities were observed either at smolt stage or at harvest weight. X-rays revealed mild abnormalities in some vertebrae
bodies, which could not be related to any experimental group. Despite the lack of signs of pathological bone alterations, the
histological examination suggested that the exposure to high CO2levels resulted in an increase in trabeculae volume and a higher
rate of bone remodeling at the end of the freshwater period. Furthermore, fish exposed to a high CO2 level presented a higher bone
ash content at the end of the freshwater period. These differences could not be observed at the end of the grow-out period.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Atlantic salmon; Carbon dioxide; Growth; Minerals; Bone histology; Deformities
1. Introduction
⁎ Corresponding author. Tel.: +47 51657994.
E-mail address: [email protected] (L. Gil Martens).
0044-8486/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.aquaculture.2006.06.031
In recent years, due to the fast growth of the industry,
the need for higher numbers of smolts has resulted in a
significant increase in fish densities in the tanks during
L. Gil Martens et al. / Aquaculture 261 (2006) 80–88
the freshwater period. This has been achieved by the
addition of oxygen (O2) to the water, and has resulted in
a significant increase in the concentration of carbon
dioxide (CO2) reaching levels up to 40 mg/l. CO2 in the
water depends on water flow and CO2 excretion rate
(Fivelstad et al., 2004a). High levels of CO2 in the water
will reduce pH and increase the toxicity of aluminum
causing hypertrophy and hyperplasia of the gill
epithelium (Fivelstad et al., 2003). Effects of longterm CO2 exposure in rainbow trout (Eddy et al., 1977;
Smart, 1979, 1981) and Atlantic salmon smolts
(Fivelstad et al., 1999) in freshwater include reduced
feed conversion ratio and reduced growth. When fish are
exposed to high levels of CO2 for long periods, blood
CO2 will increase (hypercapnia) and blood pH will
decrease resulting in respiratory acidosis (Eddy et al.,
1977; Ultsch, 1996). Fish compensate for acidosis by
increasing the plasma bicarbonate levels and excreting
phosphate via the kidney (Lloyd and Swift, 1967) and
might also mobilise ions from the bones (Storset et al.,
1997). Thus, the question remains whether these
compensation mechanisms can induce bone demineralization and thereby cause spinal deformities. Exposing
salmonids to high levels of CO2 is associated with
nephrocalcinosis, a condition characterized by the
presence of mineral deposits in the kidney tubules
(Smart, 1981; Fivelstad et al., 1999). The most abundant
minerals in these deposits are Ca, P and Mg (Smart,
1979). In the trial by Fivelstad et al. (1999), Atlantic
salmon were exposed for 62 days to different concentrations of CO2, during the fresh water phase. The
percentage of fish presenting nephrocalcinosis was
significantly higher in the group exposed to 32 mg/l
CO2 compared to the groups exposed to 7 and 19 mg/l
CO 2 . Likewise, in the study by Smart (1979),
significantly higher Ca and P concentrations were
found in the kidney of fish exposed to high CO2 levels.
Such a pathological accumulation of minerals in the
kidney clearly reflects a disturbance in mineral homeostasis, which could also affect the bone.
The purpose of the present study was to evaluate the
impact of high CO2 levels in freshwater on growth
performance, bone structure and development of spinal
deformities in Atlantic salmon both during the freshwater
period and at the end of the grow-out period in the sea.
81
high CO2 level) for 134 days during the freshwater
period until they reached smolt stage. Fish were
observed in seawater until an average weight of 3.1 kg
was achieved. Growth performance was assessed at
different intervals throughout the study in freshwater
and at the end of the seawater period. Bone samples
from the vertebral column were analyzed by radiology
and histology and for mineral composition at the end of
the freshwater phase only.
2.2. Rearing conditions
Atlantic salmon with known history and the same
genetic background were distributed in six indoor,
experimental tanks (volume of 125 l) at Lerang Research
Station (Stavanger, Norway), corresponding to 3 tanks
per experimental treatment. Each tank contained 60 fish
of an average weight of 10 g. The two experimental
groups were exposed to the following CO2 conditions:
• Control group (no CO2 addition): CO2 levels
corresponded to the CO2 generated by the fish
(max. 7 mg CO2/l).
• High CO2 group: CO2 was added gradually according to fish size in order to mimic the conditions to
which fish may be exposed to under commercial
conditions. The targeted CO2 levels in the water were
10, 20 and 35 mg/l during periods 1, 2 and 3
respectively (Table 1). In the present experiment,
carbon dioxide of a 99.9% purity was added to the
water from a pressurized bottle in a mixing tank, then
the water was distributed to the experimental tanks.
The water flow was 8 l/min in all the experimental
tanks. Seawater was added to the tanks in order to
neutralize the acidity of the water. This is common
practice in many Norwegian commercial hatcheries to
reduce aluminum toxicity. Seawater addition increased
with fish size in order to adjust salinities from 0.5‰ for
10 g fish to 5‰ for 30 g fish and larger. The light regime
was in accordance to that used for O+ smolts production
recommended by Handeland (1998). The fish were
exposed to a 12 h light and 12 h darkness regime for
6 weeks, then continued with a 24 h light regime for
2. Material and methods
Table 1
Schedule of the trial and targeted CO2 levels (mg/l) in the high CO2
group during the FW stage
2.1. Experimental design
Period
Day
Duration (days)
Targeted CO2 level (mg/l)
1
2
3
1–27
28–92
93–135
27
64
41
10
20
35
Atlantic salmon parrs (average weight of 10 g) were
exposed to two water CO2 regimes (a standard and a
82
L. Gil Martens et al. / Aquaculture 261 (2006) 80–88
3 weeks until smolt stage. During the trial, the fish were
fed commercial extruded diets manufactured by Skretting AS, Stavanger, Norway. The diets were Nutra™
and Nutra Parr™ for the freshwater period and
Atlantic™ for the seawater period. Fish were fed
twice a day to apparent satiation using automatic
feeders. After smoltification, fish were transferred to
larger tanks (circular 1m tanks), supplied with seawater.
Forty-nine days after transfer, 30 fish per tank (90 per
experimental group) were PIT tagged by implanting
micro-transponders into the peritoneal cavity. The other
30 fish in each tank were discarded. Tagged fish were
then transferred to a 5 × 5 m cage in the sea and followed
up during the grow-out period in the sea (519 days) until
they reached an average weight of 3.1 kg.
the dorsal fin (comprising 8 vertebra) was preserved in
10% buffered formalin (formaldehyde solution at 37%)
for histological examination. The following ten vertebrae were sampled for mineral analyses.
2.3. Water parameters
2.6. Histological procedures
Water temperature was recorded daily. The average
temperatures were 13.1 ± 1.7 °C and 9.2 ± 4.2 °C for the
freshwater and seawater periods respectively. During the
freshwater period, dissolved oxygen (Oxyguard Handy
meter, Point Four Systems Inc., Richmond, BC,
Canada) and water pH (pH meter WTW 340, Weilheim,
Germany) were measured daily. Oxygen level was
always higher than 8 mg/l at the outlet of the tank.
Water samples were analyzed for CO2 on days 27, 92
and 134, which correspond to the end of the periods 1, 2,
and 3 respectively, according to the method described by
Fivelstad et al. (2003).
Vertebrae were fixed in 10% neutral buffered
paraformaldehyde for 24 h, rinsed in tap water for
24 h, and decalcified for 72 h in a 10% EDTA solution
buffered with 0.1 M Tris base, pH 7.0. After
decalcification, samples were stepwise dehydrated and
embedded in Paraplast. Ten micrometer (μm) serial
sections were prepared in the sagittal plane of the
vertebrae, from the periphery to the middle plane of the
spine. Sections were stained with Masson's trichrome as
outlined in Witten and Hall (2002). From each
individual four histological sections were examined to
evaluate the bone microstructure and the bone cell
morphology of osteoblasts, osteocytes and osteoclasts.
Based on digitized microphotographs, the two-dimensional bone volume (area covered by bone) and the
trabecular number were determined on vertebrae from five
animals per group (10 animals in total) following the
recommendations of American Society for Bone and
Mineral Research (Parfitt et al., 1987). From each animal,
four consecutive sections were analyzed. Sagittal sections
were taken in a plane, half way between the outer rim and
the midline of the vertebral body (Fig. 1a). The area
covered by bone was determined by measuring the
percentage of surface area occupied by mineralized and
non-mineralized bone matrix bone in a square of
600 × 600 μm. Bone structures of the vertebral end plates
were not measured. The area covered by bone was
manually selected (Fig. 1b). The number of bone
trabeculae within the 600 × 600 μm square was determined by regarding each unbranched leg of a bone
trabecula as an entity (Fig. 1c).
Using the same material as for determining the twodimensional bone volume, osteoclast activity was estimated based on the presence or absence of osteoclasts
2.4. Fish sampling
All the fish in the two experimental groups were
individually weighed at the start of the trial, at the end of
periods 1, 2 and, 3 in freshwater, after 49 days in
seawater, and at the end of the seawater phase. The
specific growth rate (SGR) was calculated according to
the formula:
!
FinalW 1=Days
SGR ¼
−1 d100
%
InitialW
where InitialW and FinalW are the initial and final fish
weights for a given period and Days is the number of
feeding days for the period.
At the end of both the freshwater and the seawater
period, all fish were examined for external signs of
deformities. In addition, at the end of the freshwater
period, five fish per tank were individually measured,
weighed and X-rayed. Fish were examined for visible
signs of nephrocalcinosis in the kidney. The spine below
2.5. Radiology
X-rays from all sampled fish (15 per treatment) were
taken using a portable Mini X-ray HF80+ machine
(Mini X-ray Inc., Northbrock, USA) with Kodak
Industrex M Film Ready Pack II (Kodak Industry,
France) according to the protocol described by Witten
and Hall (2002). Based on enlarged radiographs,
vertebral columns of all sampled individuals were
examined for signs of skeletal deformation.
L. Gil Martens et al. / Aquaculture 261 (2006) 80–88
83
Fig. 1. Diagram showing the location of sections that were taken to determine the area covered by bone and to count the number of bone trabeculae. a:
Scheme of a vertebral body. 1 = vertebral body end plate, 2 = bone spongiosa, 3 = parasagittal section plane, halfway between the centre and the outer
rim of the vertebral body (black arrow). b: Scheme of a parasagittal section. The black square represents the plane that was chosen to determine the
area covered by bone. c: An example of how bone trabeculae were counted within the selected area, here 1–5.
(bone resorbing cells) or resorption lacunae, as traces of
osteoclast activity. Each section was assigned to one of the
following scores: 0 = no resorption, 1 = moderate resorption, 2 = abundant resorption.
package, version 5.5.0). Results from the histological
examinations were analyzed by non-parametric statistical tests (Kruskal–Wallis). The level of significance
was chosen at p < 0.10.
2.7. Vertebrae mineral analyses
3. Results
Pooled samples of vertebrae from five fish per tank
were taken for mineral analysis. The soft tissue was
removed from the spine and the bones were cleaned with
demineralized water. Subsequently lipids were removed
by rinsing the samples twice in a mixture of acetone and
methanol (1:1, v/v). The samples were then dried for
24 h at 105 °C, ashed for 18 h at 550 °C and digested
according to the AOAC method (AOAC, 1995). The
phosphorous content of the ash was determined colorimetrically (Taussky and Shorr, 1953) whereas all other
minerals were quantified by atomic absorption spectroscopy, ICP-AES (Kroglund and Finstad, 2003).
3.1. Water parameters
2.8. Aluminum in gills
At the end of the freshwater period, the second gill arch
the left side from six fish per CO2 level were dissected and
frozen for determination of aluminum concentration.
Aluminum was determined using ICP-AES (Kroglund
and Finstad, 2003). The levels of aluminum in gills were
18.6± 5.0 and 8.7 ± 0.5 μg/g gill in the control and high
CO2 groups respectively.
2.9. Statistics
Growth parameters and bone mineral analyses for the
two experimental groups were statistically evaluated
according to a one-way ANOVA (Unistat® statistical
CO2 levels and water pH were determined on days
27, 92 and 134, which correspond to the end of
freshwater periods 1, 2, and 3, respectively. CO2 levels
in the control group remained low throughout the study,
varying from 5 to 7 mg CO2/l, as compared to increasing
CO2 levels from period to period in the high CO2 group
accomplished by manual addition of CO2 to the water
(Fig. 2a). The pH measurements reflected the CO2
levels. In the control group, pH remained constant,
averaging 6.6, whereas in the high CO2 group, pH
decreased from 6.2 at the end of periods 1 and 2, to 5.8
at the end of period 3 (Fig. 2b).
3.2. Growth
3.2.1. Freshwater period
Statistical differences in weight during the freshwater
period were observed between the fish exposed to the
high CO2 level and the control group (Table 2). The
average weight of fish exposed to the high CO2 levels
was significantly lower by 10.3%, 7.3% and 20.9% for
periods 1, 2 and 3 respectively. Accordingly, the SGR
was also significantly lower in the high CO2 group,
except in period 2. For the whole freshwater period, the
cumulative SGR in the group exposed to high CO2
levels was 16.5% lower than in the control group. No
84
L. Gil Martens et al. / Aquaculture 261 (2006) 80–88
3.2.2. Seawater period
The differences in body weight observed at the end of
the freshwater period persisted during the first 48 days
after transfer to seawater with the SGR being significantly higher in the group exposed to high CO2 levels
(Table 3). At the end of the seawater period, no
significant differences between the high CO2 group and
the control group were observed in terms of weight nor
growth (Table 3).
3.3. External examination
At the end of the freshwater phase, no external signs
of deformities were observed. Condition factor was
slightly lower in the high CO2 group (1.05 ± 0.04)
compared to the control group (1.10 ± 0.04). No signs of
visible nephrocalcinosis were observed in any of the
groups.
At the end of the grow-out period, all fish were
examined externally. Again no visible signs of spinal
deformities or vertebral abnormalities could be detected
in any of the experimental groups.
3.4. Bone structure
Fig. 2. (a) Water CO2 levels measured during the freshwater phase
(mean ± S.D.; n = 3). (b) Variation in water pH during the freshwater
phase (mean ± S.D.; n = 3).
significant differences in mortality rate were found
between the two experimental groups. At smolt stage the
percentage of cumulative mortality was 6.62 ± 3.1 in the
high CO2 group and 6.13 ± 1.5 in the control group.
Table 2
Body weights (mean ± S.D.) and specific growth rate (SGR,%) in the
two experimental groups for periods 1–3 and the cumulative data for
the total freshwater phase
Control
group
Period 1 (days 1–27)
Initial weight (g)
9.2 ± 0.8
Final weight (g)
28.51 ± 0.8
SGR (%)
2.88 ± 0.3
Period 2 (day 28–92)
Final weight (g)
55.91 ± 2.3
SGR (%)
1.58 ± 0.08
Period 3 (days 93–135)
Final weight (g)
82.35 ± 7.3
SGR (%)
0.97 ± 0.1
Cumulative data (days 0–
135)
SGR (%)
1.63 ± 0.04
Mortality (%)
6.13 ± 1.5
ns: not significant (p > 0.10).
High CO2
group
10 mg/l CO2
10.4 ± 0.3
26.76 ± 0.7
1.99 ± 0.1
20 mg/l CO2
51.82 ± 0.1
1.63 ± 0.1
35 mg/l CO2
65.10 ± 2.2
0.63 ± 0.1
1.36 ± 0.01
6.62 ± 3.1
Statistics
ns
p < 0.10
p < 0.05
p < 0.10
ns
p < 0.05
p < 0.05
p < 0.05
ns
An evaluation of the bone structure was conducted
based on radiographs and histological sections. The data
obtained from the analysis of X-rays revealed no
significant differences between fish from the two
experimental groups (Fig. 3a–b). Vertebrae deformities,
that had no visual effect on the fish external shape, were
only observed in one individual out of fifteen individuals for each experimental group (Fig. 3c–d).
Even though the X-rays did not reveal morphological
differences between the two groups, the examination of
histological serial sections suggests that the water CO2
level has an impact on the bone structure (Fig. 4a–c).
Fish exposed to a high CO2 level displayed strong
Table 3
Initial and final body weight (mean ± S.D.) and specific growth rate
(SGR,%) in seawater
Control group
Period days 1–48 in seawater
Initial weight (g)
82.3 ± 7.3
Final weight (g)
144. 8 ± 8.7
SGR (%)
1.28 ± 0.1
Period days 49–519 in seawater
Final weight (g)
3215.6 ± 944.7
SGR (%)
0.86 ± 0.05
ns: not significant (p > 0.10).
High CO2 group
Statistics
65.1 ± 2.2
125.25 ± 5.3
1.41 ± 0.1
p < 0.05
p < 0.05
p < 0.10
3065.0 ± 1094.1
0.85 ± 0.07
ns
ns
L. Gil Martens et al. / Aquaculture 261 (2006) 80–88
85
Fig. 3. (a) Radiograph of the vertebral column from Atlantic salmon raised in 7 mg/l CO2 (smolt stage). Vertebral bodies are regular shaped with no
observed significant variation. Magnification ×6, scale bar = 2 mm. (b) Radiograph of the vertebral column from Atlantic salmon raised in 35 mg/l
CO2 (smolt stage). Vertebral bodies are regular shaped and no significant variations are observed. Magnification ×6, scale bar = 2 mm. (c) Radiograph
of the vertebral column form Atlantic salmon raised under 7 mg/l CO2 (smolt stage), depicting an example of mild deformation of vertebral bodies
consisting of slightly compressed vertebrae (white arrowheads) and enlarged intervertebral spaces (black arrowheads). The deformities observed did
not affect the external shape of the fish. Magnification ×6, scale bar = 2 mm. (d) Radiograph of the vertebral column from Atlantic salmon raised
under 35 mg/l CO2 (smolt stage), showing an example of mild vertebrae deformation consisting of one slightly compressed (white arrowhead) and
two half sided deformed adjunct vertebral bodies (black arrowheads). A similar type of deformity (compression) was observed in the group of animals
that were raised under non-elevated CO2 conditions (see panel c). Deformities had no visible consequences for the animals' external shape.
Magnification ×6, scale bar = 2 mm.
internal bone structures (bone trabeculae), reflected by
the increased trabeculae volume (Table 4, Fig. 4b). The
analysis of histological sections also suggests that
increased CO2 levels result in a high rate of bone
remodeling as indicated by the frequent occurrence of
osteoclasts (bone resorbing cells) and signs of osteoclast
activity (resorption lacunae), as shown in Fig. 4c. This
suggestion is based on higher osteoclast index in the
high CO2 group when compared to the control group
(Table 4).
3.5. Mineral composition of the bone
At the end of the freshwater phase, ash content was
significantly higher (p < 0.05) in the high CO2 group
than in the control group (Table 5). In general all the
minerals analyzed were higher in this group compared
to the control fish; however these differences were only
significant for Ca, P, Cu, and Fe. A slight increase in the
Ca:P ratio was also observed in fish exposed to high
CO2 levels.
4. Discussion
4.1. Growth
Fish exposed to high water CO2 level (10–35 mg/l
CO2) during the freshwater period presented a significant lower growth and body weight compared to fish
exposed to 7 mg/l CO2. A significant reduction in
growth parameters (weight, condition factor) has been
observed in earlier investigations (Smart, 1981; Fivelstad et al., 1998, 1999; Graff et al., 2002). Results from
experiments performed with Atlantic salmon in freshwater (pH 6.5) indicated that safe levels of carbon
dioxide range between 15 and 20 mg/l CO2 (Fivelstad et
86
L. Gil Martens et al. / Aquaculture 261 (2006) 80–88
Fig. 4. (a) Parasagittal section through a vertebral body from Atlantic salmon raised in 7 mg/l CO2 (smolt stage). Bone trabeculae are regularly formed
and display no signs of structural changes that could indicate pathological alterations such as micro fractures or enhanced bone remodeling. Masson's
trichrome staining, ×63, scale bar = 200 μm. (b) Parasagittal section through vertebral body from Atlantic salmon raised in 35 mg/l CO2 (smolt stage).
Similar to the bone trabeculae shown in panel a. Trabeculae are regularly formed and display no sign of pathological alterations. In comparison to
panel a the trabeculae are, however, wider and less numerous (see also Table 4). Parts of bone surface are eroded (black arrowheads) indicating
abundant bone resorption. Masson's trichrome staining, ×63, scale bar = 200 μm. (c) Higher magnification of an area of intensive bone resorption at
the bone trabeculae from a fish raised in 35 mg/l CO2 (smolt stage). Black arrowheads point to typical resorption lacunae, white arrowheads point to
multinucleated osteoclasts. Masson's trichrome staining, ×160, scale bar = 80 μm.
al., 1999, 2003). In the present experiment a 10%
reduction in body weight was observed at 10 mg/l CO2.
There is limited knowledge available concerning the
effects of carbon dioxide on salmon growth (Fivelstad et
al., 1998). One hypothesis is that the reduced growth
during the freshwater period might be due to a lower
food intake (Smart, 1981) and increased use of energy
during stressful conditions (Wendelaar Bonga, 1997;
Graff et al., 2002). In addition, it is also known that
stress inhibits growth by affecting hormonal pathways,
which regulates growth. These effects have been
reported in relation to the growth hormone (Pickering,
1993), thyroidal hormones T3 and T4 (Brown et al.,
1991) and insulin-like growth factor (IGF-1) levels
(Plisetskaya and Duan, 1994). A second hypothesis is
based on the aluminum toxicity on the gills. In the
present study, water pH was not low enough to increase
free aluminum levels (Fivelstad et al., 2004b), as shown
by the low levels of aluminum in the gills of the fish
from the two experimental groups. Unexpectedly the
Table 4
Trabeculae number (mean ± S.D.), area covered by bone (mean ± S.D.)
and osteoclast activity in smolts raised under two different CO2
regimes (n = 5)
Group
Trabeculae
number
Area covered by
bone (%)
Osteoclast
activity (0–2)
Control group
High CO2 group
38.9 ± 12.81
35.4 ± 2.98
36.2 ± 2.83a
41.5 ± 3.60b
1.25
2.00
a,b
Different superscripts indicate a statistical difference (p < 0.05).
level of aluminum was higher in the control group than
in the high CO2 group.
In other studies, there was no effect of high carbon
dioxide levels on fish performance in freshwater (Smart,
1979; Fivelstad et al., 2004b). The possibility that fish
can develop a physiological tolerance towards certain
chronic sub-lethal environmental stressors, like carbon
dioxide, should be considered (Lloyd and Swift, 1967;
Wedemeyer, 1980).
Differences in body weight were maintained during
the initial stages of the seawater period. SGR was
significantly higher in the high CO2 group immediately
after transfer likely due to compensatory growth or
Table 5
Bone mineral composition at smolt stage (mean ± S.D., n = 3; pooled
samples from 5 fish per tank)
Ash (%)
Ca (%)
P (%)
Ca:P
Cu (mg/kg)
Mg (%)
Mn (mg/kg)
K (%)
Na (%)
Fe (mg/kg)
Zn (mg/kg)
Al (mg/kg)
Control group
High CO2 group
Statistics
32.1 ± 1.9
11.1 ± 0.8
6.2 ± 0.2
1.8 ± 0.1
4.2 ± 0.4
0.2 ± 0.0
23.3 ± 2.0
0.6 ± 0.0
0.2 ± 0.0
15.5 ± 3.5
409.2 ± 39.5
<0.2
34.2 ± 1.2
12.9 ± 1.4
6.9 ± 0.8
1.9 ± 0.0
5.0 ± 0.8
0.3 ± 0.1
24.3 ± 3.5
0.6 ± 0.1
0.2 ± 0.1
30.5 ± 17.1
400.8 ± 46.1
<0.2
p < 0.05
p < 0.10
p < 0.05
p < 0.10
p < 0.05
Ns
Ns
Ns
Ns
p < 0.10
Ns
–
ns: not significant (p > 0.10).
L. Gil Martens et al. / Aquaculture 261 (2006) 80–88
because the fish were smaller than in the control group.
This is in accordance with a study by Fivelstad et al.
(1999) and Graff et al. (2002), where post-smolts
seemed to completely recover from a weight gain
depression caused by high dissolved CO2 levels in
freshwater.
However, no significant differences in weight and
condition factor were observed between groups at the
end of the seawater period. The duration of the present
experiment was longer than previous CO2 experiments
(Fivelstad et al., 1999, 2003).
4.2. Bone and spinal deformities
Fish subjected to high water CO2 level displayed a
significantly elevated bone ash and bone mineral
content. The fish appeared externally normal and the
condition factor was relatively similar in both groups.
Furthermore, no external spinal deformities were
observed in any of the two groups. The X-rays revealed
mild abnormalities of vertebral bodies in two animals
(out of 30), which could not be related to any
experimental group. Severe deformities such as lordotic,
kyphotic and skoliotic curvatures, compressed or
fractured vertebral bodies and chordal tissue alterations
(Vågsholm and Djupvik, 1998; Gavaia et al., 2002;
Witten et al., 2005, Gil Martens et al., 2005) involving
large parts of the spine, were not observed. The
histological analysis of the spine indicated that fish
exposed to high CO2 have an increased 2D bone volume
(bone trabeculae) and a high rate of bone remodeling as
indicated by increased bone deposition (osteoblast
activity) and resorption (osteoclast activity).
Since a CO2 related acidosis leads to the mobilisation
of minerals from the skeleton and causes elevated renal
mineral discharge (Storset et al., 1997; Weger et al., 1999),
findings about a higher bone mineral content and a higher
2D bone volume in fish exposed to increased ambient
CO2 levels are surprising. Indeed, also the observed
stimulation of bone resorption should lead to a decrease of
mineral content and volume in the bone. However, similar
to mammals, resorption of the skeleton in salmonids and
in related teleosts groups is coupled with bone formation
(Witten et al., 2000, 2001; Witten and Hall, 2002). As
observed in mammals (Manolagas, 2000), the stimulation
of bone resorption can concurrently stimulate bone
formation, enhance the total bone turnover and result in
the increase of the bone volume. To satisfy extra mineral
demands associated with the elevation of the bone
metabolism, salmon and other teleosts can follow
different pathways such as the increased intake of calcium
via the gills, the mobilisation of minerals from other parts
87
of the skeleton, and the increased intestinal absorption of
calcium and phosphate (Vielma and Lall, 1998; Persson et
al., 2000; Witten and Hall, 2002). Since intestinal
absorption is the main source for phosphate and because
bone is mineralized by calcium phosphate (Vielma and
Lall, 1998; Witten and Huysseune, in press), a CO2
triggered elevated bone turnover likely requires a
sufficient nutritional phosphate supply in order to prevent
under-mineralized and malformed bones (Baeverfjord et
al., 1998; Helland et al., 2005).
An additional argument to explain the findings that
animals exposed to elevated ambient CO2 levels display
a high bone mineral content relates to the fact that the
fish of the experimental group display a reduced specific
growth rate. Bone formation consists of two steps. The
first step is the production of the non-mineralized bone
matrix by bone osteoblasts and the second step is the
slow mineralization of the bone matrix (Witten and
Huysseune, in press). Fast bone growth, as it is generally
the case under farming condition compared to the
normal growth of wild animals, can thus lead to the
production of large amounts of non-mineralized bone
matrix, that only later (e.g. during the salmon sea life
phase) mineralizes sufficiently (Witten and Hall, 2002;
Fjelldal et al., 2004). In turn, the slower growth of
animals in the experimental group could have allowed a
more complete mineralization of the bone matrix and
thus contributed to the elevated bone mineral content.
5. Conclusions
In the present study levels of water carbon dioxide
from 10 to 35 mg/l have been shown to be detrimental to
Atlantic salmon growth in the freshwater period.
Differences in weight were maintained during the initial
stages of the seawater period but were not observed at
harvest weight.
No spinal deformities were observed throughout the
study. Based on our observations we hypothesize that
CO2 stimulates the amount of bone remodeling, which
under conditions of sufficient mineral supply increases
the bone volume. The physio-pathological significance
of these changes requires further investigation.
References
AOAC, 1995. Official Methods of Analysis, 16th ed. Association of
Official Analytical Chemists, Arlington, VA, p. 1094.
Baeverfjord, G., Åsgaård, T., Shearer, K.D., 1998. Development and
detection of phosphorous deficiency in Atlantic salmon (Salmo
salar L.) parr and post-smolts. Aquac. Nutr. 4, 1–11.
Brown, S.B., McLatchy, D.L., Hara, T.J., Eales, J.C., 1991. Effect of
cortisol on aspects of 3, 5, 5′-triido-L-thyronine metabolism in
88
L. Gil Martens et al. / Aquaculture 261 (2006) 80–88
rainbow trout (Oncorhynchus mykiss). Gen. Comp. Endocrinol.
81, 207–216.
Eddy, F.B., Lomholt, J.P., Weber, R.E., Johansen, K., 1977. Blood
respiratory properties of rainbow trout kept in water of high CO2
tension. J. Exp. Biol. 67, 37–47.
Fivelstad, S., Olsen, A., Zeitz, S., Diesen Hosfeldt, C., Kløften, H., Ski,
H., Rasmussen, T., Vindheim, T., Steffanson, S., 1998. Effekter av
karbondioxid på laksesmolt. Norskfiskeoppdrett 16, 40–41.
Fivelstad, S., Olsen, A., Kløften, H., Ski, H., Steffanson, S., 1999.
Effects of carbon dioxide on Atlantic salmon (Salmo salar L.)
smolts at constant pH in bicarbonate rich freshwater. Aquaculture
178, 171–187.
Fivelstad, S., Olsen, A., Åsgård, T., Bæverfjord, G., Rasmussen, T.,
Vindheim, T., Stefansson, S.O., 2003. Long-term sub-lethal effects
of carbon dioxide on Atlantic salmon smolts: ion regulation,
haematology, element composition, nephrocalcinosis and growth
parameters. Aquaculture 215, 301–319.
Fivelstad, S., Bergheim, A., Hølland, P.M., Fjermedal, A.B., 2004a.
Water flow requirements in the intensive production of Atlantic
salmon (Salmo salar L.) parr–smolt at two salinity levels.
Aquaculture 231, 263–277.
Fivelstad, S., Olsen, A.B., Stefansson, S., Handeland, S., Waagbø, R.,
Kroglund, F., Colt, J., 2004b. Lack of long-term sublethal effects
of reduced freshwater pH alone on Atlantic salmon (Salmo salar)
smolts subsequently transferred to seawater. Can. J. Fish. Aquat.
Sci. 61, 511–518.
Fjelldal, P.G., Grotmol, S., Kryvi, H., Gjerde, N.R., Taranger, G.L.,
Hansen, T., Porter, M.J.R., Totland, G.K., 2004. Pinealectomy
induces malformation of the spine and reduces the mechanical
strength of the vertebrae in Atlantic salmon, Salmo salar. J. Pineal
Res. 36, 132–139.
Gavaia, P.J., Dinis, M.T., Cancela, M.L., 2002. Osteological development and abnormalities of the vertebral column and caudal
skeleton in larval and juvenile stages of hatchery-raced Senegale
sole (Solea senegalensis). Aquaculture 211, 305–323.
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, 72–80.
Graff, I.E, Waagbø, R., Fivelstad, S., Vermeer, C., Lie, Ø., Lundebye,
A.K., 2002. Multivariate study on the effects of dietary Vitamin K;
D3, calcium and dissolved CO2 on bone, health and fish
performance in Atlantic salmon. J. Fish Dis. 25, 599–614.
Handeland, S., 1998. Osmoregulation and growth in Atlantic salmon
smolts during the early seawater phase. Dr. Scient. Thesis.
Department of Fisheries and Marine Biology. University of
Bergen, Norway.
Helland, S., Refstie, S., Espmark, Å., Hjelde, K., Bæverfjord, 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.
Kroglund, F., Finstad, B., 2003. Low concentrations of inorganic
monomeric aluminium impair physiological status and marine
survival of Atlantic salmon. Aquaculture 222, 119–133.
Lloyd, R., Swift, W.R., 1967. Effect of high concentration of carbon
dioxide on the ionic composition of rainbow trout blood. Nature
216, 1341–1342.
Manolagas, S.C., 2000. Birth and death of bone cells: basic regulatory
mechanisms and implications for the pathogenesis and treatment of
osteoporosis. Endocr. Rev. 21, 115–137.
Parfitt, A.M., Drezner, M.K., Glorieux, F.H., Kanis, J.A., Malluche,
H., Meunier, P.J., Ott, S.M., Recker, R.R., 1987. Bone histomor-
phometry: standardisation of nomenclature, symbols and unit. J.
Bone Miner. Res. 2, 595–610.
Persson, P., Shrimpton, J.M., McCormick, S.D., Björnsson, B.T.,
2000. The presence of high-affinity, low-capacity estradiol-17β
binding in rainbow trout scale indicates a possible endocrine route
for the regulation of scale resorption. Gen. Comp. Endocrinol. 120,
35–43.
Pickering, A.D., 1993. Growth and stress in fish production. Genetics
in Aquaculture IV. Aquaculture 111, 253–258.
Plisetskaya, E.M., Duan, C., 1994. Insulin and insulin-like growth
factor I in streptozotocin-injected coho salmon, Oncorhynchus
kisutch. Am. J. Physiol. 267, R1408–R1412.
Smart, G.R., 1979. Nephrocalcinosis in rainbow trout Salmo gairdneri
Richardson; the effect of exposure to elevated CO2 concentration.
J. Fish Dis. 2, 279–289.
Smart, G.R., 1981. Aspects of water quality producing stress in
intensive fish culture. In: Pickering, A.D. (Ed.), Stress and Fish.
Academic Press, London, pp. 277–289.
Storset, A., Åsgård, T., Bæverfjord, G., 1997. Can high free levels of
CO2 in the water damage fish? Nor. Fiskeoppdrett 1, 68–69 (In
Norwegian).
Taussky, H.H., Shorr, E., 1953. A microcolorimetric method for the
determination of inorganic phosphate. J. Biol. Chem. 202,
675–685.
Ultsch, G.R., 1996. Gas exchange, hypercarpnia, and acid–base
balance, palaeochology and the evolutionary transition from waterbreathing to air-breathing among vertebrates. Palaeogeogr. Palaeoclimatol. Palaeoecol. 123, 1–23.
Vielma, J., Lall, S.P., 1998. Control of phosphorus homeostasis of
Atlantic salmon (Salmo salar) in fresh water. Fish Physiol.
Biochem. 19, 83–93.
Vågsholm, I., Djupvik, H.O., 1998. Risk factors for spinal deformities
in Atlantic salmon, Salmo salar L. J. Fish Dis. 21, 47–53.
Wedemeyer, G., 1980. Effects of environmental stressors in aquacultural systems on quality, smoltification and early marine
survival of anadromous fish. Proceedings of the North Pacific
Aquaculture Symposium. Alaska Sea Grant. Alaska University,
pp. 155–169.
Weger, M., Deutschmann, H., Weger, W., Kotanko, P., Skrabal, F.,
1999. Incomplete renal tubular acidosis in ‘primary’ osteoporosis.
Osteoporos. Int. 10, 325–329.
Wendelaar Bonga, S., 1997. The stress response in fish. Physiol. Rev.
77, 591–625.
Witten, P.E., Hall, B.K., 2002. Differentiation and growth of kype
skeletal tissues in anadromous male Atlantic salmon (Salmo salar).
Int. J. Dev. Biol. 46, 719–730.
Witten, P.E, Villwock, W., Peters, N., Hall, B.K., 2000. Bone
resorption and bone remodeling in juvenile carp (Cyprinus carpio).
J. Appl. Ichthyol. 16, 254–261.
Witten, P.E., Hansen, A., Hall, B.K., 2001. Features of mono and
multinucleated bone resorbing cells of the zebrafish Danio rerio
and their contribution to skeletal development, remodeling and
growth. J. Morph. 250, 197–207.
Witten, P.E., Gil-Martens, L., Hall, B.K., Huysseune, A., Obach, A.,
2005. Compressed vertebrae in Atlantic salmon (Salmo salar):
evidence for metaplastic chondrogenesis as a skeletogenic
response late in ontogeny. Dis. Aquat. Org. 64, 237–246.
Witten, P.E., Huysseune, A., in press. Mechanisms of chondrogenesis
and osteogenesis in fins. In: Hall, B.K. (Ed.), Fins and Limbs:
Development, Evolution and Transformation. Chicago Univ.
Press, Chicago.