The inuence of ploidy on saltwater adaptation, acute stress

response and immune function following seawater transfer in

non-smolting rainbow trout
J.F. Taylor
*
, M.P. Needham, B.P. North, A. Morgan, K. Thompson, H. Migaud
Institute of Aquaculture, University of Stirling, Stirling, Scotland FK9 4LA, UK
Received 14 September 2006; revised 23 February 2007; accepted 24 February 2007
Available online 2 March 2007
Abstract
We investigated the eect of ploidy on osmoregulatory, stress and immune responses in non-smolting rainbow trout during saltwater
adaptation. Sibling groups of diploid and triploid trout were acclimated in freshwater (FW) and then subjected to abrupt transfer to full
strength (35 ppt) saltwater (SW) or back to FW. Fish were sampled pre-stress, and 1, 3, 6, 12, 24, 48, 72 and 168 h post-stress. Overall
mortality in SW was less than 5% in either ploidy, with no mortality in FW. Signicant elevations in plasma osmolality and gill ATPase
were observed within 13 h of SW transfer, but retuned to basal levels within 72 h indicative of rapid saltwater adaptation and did not
dier between ploidy. Furthermore, FWSW transfer also caused signicant and sustained elevations in total blood haemoglobin,
plasma IGF-I, cortisol, glucose, total white blood cell counts, increased plasma but decreased mucus lysozyme, and enhanced head kid-
ney macrophage respiratory burst activity. Conversely, FWFW transfer evoked more transient and less elevated responses, more typical
of primary and secondary responses to a single stressor. We conclude that the more elevated levels in these parameters are a function of
saltwater adaptation as well as the generic stress response, and that this did not dier between ploidy. Strong positive correlations were
found between plasma IGF-I and cortisol, and with osmolality, glucose and WBC, while a negative correlation was found with plasma
lysozyme irrespective of ploidy. Overall, the current results suggest that triploidy does not aect the ability of non-smolting trout to adapt
to full strength seawater under optimum conditions, and that the osmotic and stress response to such transfer is similar to diploids.
2007 Elsevier Inc. All rights reserved.
Keywords: Triploid; Sea water challenge; Osmoregulation; Stress; Immune function; Rainbow trout
1. Introduction
Rainbow trout (Oncorhynchus mykiss) is one of the most
important aquaculture species worldwide, with a growing
trend towards the production of large (3 kg+) sea-grown
trout since this improves growth and esh quality. A signif-
icant problem encountered when growing large trout is a
high incidence of pre-harvest maturation, which is associ-
ated with the diversion of energy into gonadal recrudes-
cence, leading to reductions in somatic growth and esh
quality, as well as increased susceptibility to disease
(Bromage et al., 2001). Unlike the Atlantic salmon (Salmo
salar) industry which uses photoperiod application exten-
sively during grow-out to resolve this problem, such tech-
niques have not been successful with trout to date. In this
respect, articial induction of triploidy may oer a poten-
tial solution to the problem.
Morphologically, triploids are similar to diploids but
dier in three fundamental ways; being generally more het-
erozygous, having larger but fewer cells due to the extra
chromosomal set, and gonadal development is generally
disrupted resulting in sterility (Benfey, 1999), the latter
being the trait of greatest interest to industry. However,
there is signicant scepticism within the industry regarding
the use of triploids due to reports on inferior performance,
morphological abnormalities, reduced immune function,
0016-6480/$ - see front matter 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.ygcen.2007.02.029
*
Corresponding author. Fax: +44 01768 472133.
E-mail address: jft2@stir.ac.uk (J.F. Taylor).
www.elsevier.com/locate/ygcen
General and Comparative Endocrinology 152 (2007) 314325
and higher mortalities. Within the literature there is still
debate between sh and ploidy status regarding their phys-
iology, and the mechanisms responsible for the apparent
performance dierences are poorly understood. Further-
more, high rates of mortality have been observed particu-
larly under sub-optimal conditions such as hypoxia or
high temperature (Ojolick et al., 1995; Cotter et al., 2000;
Hyndman et al., 2003b). Impaired ability to maintain su-
cient oxygen supplies during and after stress may relate to
the cellular nature of triploidy and generally lower blood
haemoglobin concentrations (Benfey and Sutterlin, 1984;
Sadler et al., 2000a).
Rainbow trout have been shown to undergo a typical
stress response when transferred from fresh to saltwater
environments (Hegab and Hanke, 1986; Liebert and
Shreck, 2006). It is well known that both acute and chronic
stress can increase susceptibility of sh to disease (Schreck,
1997; Wendelaar Bonga, 1997). A major factor responsible
for this is an alteration in both the number and composi-
tion of circulating leucocytes (Barton and Iwama, 1991).
Acute stress often results in a decrease in circulating total
leukocyte concentrations (Benfey and Biron, 2000). Stress
also aects the phagocytic and/or respiratory burst activity
of spleen, head kidney and plasma leucocytes (Thompson
et al., 1993; Pulsford et al., 1994; Vazzana et al., 2002; Lie-
bert and Shreck, 2006). In addition, the response in plasma
lysozyme activity following acute stress has been shown to
be both enhanced as well as suppressed depending on the
type, intensity and duration of the stressor (Fevolden
et al., 2003). However, despite considerable research into
stress and immune response to environmental challenge,
dierences in salinity tolerance between diploid and trip-
loid trout remain unexplored.
The initial period following seawater transfer is an extre-
mely stressful event for anadromous sh which must com-
pensate for the eects of changing environmental salinity,
and alter the morphology and physiology of the ionic-
osmotic regulatory apparatus in order to maintain homo-
eostasis (McCormick, 2001; McCormick and Bradshaw,
2006). In salmonids, this involves an increase in number
and size of chloride cells in the gill and intestine, both
major sites of active ionic transport, characterised by high
concentrations of mitochondria and Na
+
, K
+
, ATPase
activity (McCormmick, 1995). In sh, marked elevations
in numerous endocrine hormones including cortisol,
growth hormone (GH) and insulin-like growth factor-I
(IGF-I) have been documented which subsequently
increase Na
+
, K
+
, ATPase activity and enhance saltwater
tolerance (McCormick, 2001). This process occurs prior
to seawater transfer during smoltication which pre-adapts
smolts to life in the sea in the case of anadromous salmo-
nids (Handeland et al., 2004). Conversely, non-smolting
strains of rainbow trout are not required to undergo any
pre-adaptation to seawater (Jackson, 1981; Alexis et al.,
1984). The presence of salt alone appears to result in an
increase in chloride cell size as well as Na
+
, K
+
, ATPase
activity (Salaman and Eddy, 1987). In this respect, it has
been suggested that the cellular characteristics of triploids
may result in a reduction in the capacity for osmoregula-
tion, associated with a smaller surface area for active ion
transport (Sadler et al., 2000a). Furthermore, increased
incidence of gill deformity and reduced cell surface
area:volume ratios in triploid sh may signicantly lower
the capacity to utilise oxygen in the marine environment,
and subsequently reduce survival (Sadler et al., 2000b;
Cal et al., 2006).
To date the ability of triploid trout to adapt to seawater
transfer has received very little attention with regards to
ionic-osmotic physiology or the stress response in relation
to their diploid conspecics. Therefore the objectives of
the current study were to (a) characterise the adaptive
response of a non-smolting rainbow trout to abrupt seawa-
ter transfer, (b) to compare this response between sibling
diploids and triploids and (c) use these results to clarify
whether or not the increased mortality of triploids during
seawater transfer is due to an inability to adapt. This was
achieved by studying a number of indicators of osmoregu-
latory ability as well as primary and secondary stress and
immune parameters.
2. Materials and methods
2.1. Fish and facilities
All-female rainbow trout eggs of a non-smolting strain were obtained
from Barony College (Dumfries, Scotland, UK) in January 2005. Both
diploids and triploids originated from the same parent broodstock
(1 male:2 females) to eliminate any genotype interaction. Triploidy was
induced by applying a hydrostatic pressure shock of 9500 psi, for 5 min,
200 C min post-fertilisation. Erythrocyte major axis length was used to
verify triploidy status prior to experimentation. Verication was con-
rmed with a mean axis length of 35.50 0.13 lm in control diploids
and 54.50 0.24 lm in triploids (p < 0.001). Hatching (April 2005) and
pre-experimental rearing took place at the Niall Bromage Freshwater
Research Laboratory (NBFRL, Stirling, Scotland, UK, 57N). During
the rearing period sh (1000 per ploidy) were ongrown in 25 m
3
circular
tanks (one per ploidy) within a ow through system under ambient light
and temperature and were fed on a commercial pelleted feed (Trouw, Elite
45) according to manufacturers feeding tables.
2.2. Experimental design and sampling procedure
Two weeks prior to each experimentation, sh from the common pop-
ulation were anesthetised using 2-phenoxyethanol (1:10,000 Sigma Chem-
icals, UK) and hand graded into two batches of 80 sh (one per ploidy)
and held in separate 1 m
3
FW tanks. Initial mean weight and length were
191.1 9.3 g and 256.4 4.0 mm, respectively, and was maintained for all
trials. Stocking density during acclimation was approximately 16 kg m
3
and tanks received constant aeration with mechanical/biological ltration,
with 20% water changes made every other day using de-chlorinated tap
water. Fish were left to acclimate for a period of 2 weeks prior to experi-
mental saltwater (SW) or freshwater (FW) challenge during which time
they were fed three times weekly to excess on a commercial pelleted feed
(Trouw, Elite 45).
Following the acclimation period, sh were subjected to abrupt trans-
fer from either FW to full strength (35 ppt) SW, or FW to FW (used to
discriminate between handling and SW induced responses). FWSW chal-
lenge was carried out in duplicate with replicates 1 and 2 taking place
between 1017th and 24th April1st May 2006, respectively. Unfortu-
nately, due to tank and system limitations the FWFW trial could not
J.F. Taylor et al. / General and Comparative Endocrinology 152 (2007) 314325 315
be replicated, with the trial carried out from 17 to 24th April. For each
trial a baseline sample (0 h) was taken (n = 6) from each ploidy prior to
transfer. During transfer, sh were rapidly netted into the experimental
tanks within 5 min from initial disturbance for each ploidy. Two tanks
(one per ploidy) were assigned to each time point during the trials; 1 h,
3 h, 6 h, 12 h, 24 h, 48 h, 72 h and 7 days post-transfer. Time points were
allocated dierent tanks to ensure the responses measured were due to the
experimental challenge and not the eect of repeated disturbance, in par-
ticular with regards to cortisol. Each tank was stocked with 10 sh so that
10 diploids and 10 triploids were assigned to every time point (stocking
density of approximately 6 kg m
3
). Fish were allocated in block design
so that diploids and triploids for each time point were situated next to
one another. The SW trial was carried out using a fully recirculating par-
allel line tank (0.5 m
3
) system. During acclimation in FW and SW chal-
lenge water temperature was maintained at 10 1 C and total
ammonia (TAN) remained below 3 mg ml
1
with dissolved oxygen
>7 mg l
1
. The FW control trial was carried out using a ow-through tank
(0.5 m
3
) system. During acclimation and throughout the FW trial temper-
atures were maintained at 8 1 C and dissolved oxygen >7 mg l
1
.
At each time point six diploid and six triploid sh were netted from
their respective two tanks and instantly killed with a lethal dose of 2-phen-
oxyethanol (1:100; Sigma Chemicals, UK). Fish were immediately bled
from the caudal vein with heparinised syringes (total volume of approxi-
mately 23 ml) and sub-samples (200 ll) of whole blood were placed aside
on ice for haematological analysis. The remaining blood was centrifuged
(1200g for 15 min) and the plasma was aliquoted for plasma analyses of
cortisol, IGF-I, glucose, osmolality and lysozyme activity and stored at
70 C prior to analysis. In each case, all sh were bled within 35 min
in order to minimise the aects of sampling stress (i.e. disturbance and net-
ting). Cutaneous mucus was collected from the entire body surface of each
using the back of a sterile scalpel blade. Mucus was washed in an ammo-
nium bicarbonate buer (100 mM) and centrifuged (2500g for 5 min) after
which the supernatant was removed and stored at 70 C prior to analysis
for mucus lysozyme activity. Gill biopsies were performed on the second
gill arch and 34 lament tips were stored in a sucrose, sodium EDTA
and imizadole (SEI) buer at 70 C prior to determination of gill Na
+
,
K
+
, ATPase activity. At 0 h, 24 h, 72 h and day 7 the anterior kidney
was aseptically removed and forced through a 100 lm nylon mesh into
heparinised Leibovitz-15 medium (Sigma, Aldrich, UK) and the resultant
macrophage suspensions were immediately assayed for respiratory burst
activity.
2.3. Sample analyses
Total white (WBC) and red blood cell (RBC) counts were taken from
whole blood diluted in phosphate buered saline (WBC, 1:100; RBC,
1:1000). Counts were made under an Olympus CH light microscope using
a Neubauer haemocytometer. Haematocrit (Hct, expressed as % of packed
RBC volume) was measured on a micro-haematocrit reader after centri-
fuging whole blood (5000g for 5 min) in heparinised micro-haematocrit
tubes. Total blood haemoglobin concentration (Hb) was measured by
Drabkins colorimetric assay using a commercially available kit and stan-
dard (Sigma, Aldrich, UK). From the previous parameters mean corpus-
cular volume (MCV), mean corpuscular haemoglobin (MCH) and mean
corpuscular haemoglobin concentration (MCHC) were obtained using
the following formulas (Perruzi et al., 2005): MCV (fL) = Hct/RBC
(10
6
ll
1
); MCH (pg) = [Hb (g dl
1
) 10]/RBC (10
6
ll
1
); MCHC
(g dl
1
) = [Hb (g dl
1
) 10]/Hct.
Plasma osmolality were measured on sub-samples using a micro-
osmometer (Advanced Instruments, Massachusetts, US) in order to deter-
mine the time course of osmotic disturbance. Gill Na
+
, K
+
, ATPase was
determined using the method outlined by McCormick (1993) which is
based on the measurement of NADH oxidation by the ouabain sensitive
hydrolysis of ATP and expressed as lmol mg prot
1
h
1
.
Plasma cortisol concentrations were determined using the radioimmu-
noassay method described by Ellis et al. (2004) and adaptated by North
et al. (2006). The tritiated label was supplied by Amersham Biotech
(UK) and a sheep anti-cortisol antibody from Diagnostic Scotland
(UK). Intra- and inter-coecients of variation were 2% and 11%, respec-
tively. Minimum sensitivity was 12 pg tube
1
. Total plasma IGF-I was
measured using Gropep Ltd. Fish RIA kits following acid:ethanol extrac-
tion. Extraction involved the incubation of 40 ll of plasma with 160 ll
acid:ethanol mix (62.5 ml 2 M HCl:437.5 ml 100% ethanol) for 30 min at
room temperature. The supernatant was neutralised with 80 ll of
0.855 M Tris. Samples were centrifuged at 10,000g for 10 min at 4 C.
The resultant supernatant was collected and 50 ll assayed in triplicate
according to the Gropep kit protocol. The detection limit was
0.15 ng ml
1
, with intra- and inter-assay coecient of variation of 4.4%
(n = 10) and 13.9% (n = 10), respectively. Plasma glucose concentrations
were assayed using a colorometric method (Sigma, Aldrich, UK). Plasma
samples were pipetted into a 96-well plate, in quadruplicate, along with
Trinder reagent. Change in absorbance (505 nm), as a result of the glucose
dependant production of quinoneimine dye, was measured after 5 min in
an ELISA microplate reader (Dynex, UK). Plasma and mucus lysozyme
activities were assayed using a 96-well plate method adapted from Ellis
(1990). Samples of plasma/mucus were added to a suspension of Micro-
coccus lysodeiticus in 0.04 M sodium phosphate buer (pH 5.8). The
change in absorbance (540 nm), dependant on the rate of lysis, was mea-
sured after 1 and 5 min.
Respiratory burst activity was measured by the rate of reduction of
nitro blue tetrazolium (NBT) into blue formazan which is dependant on
the rate of superoxide production during respiratory burst (Secombes,
1990). Samples (200 ll) of head kidney suspension were pipetted into a
96-well plate in six replicates and incubated to form monolayers. Phorbol
myristate (PMA) was used to stimulate respiratory burst in half of these
replicates and the dierence in absorbance (605 nm) between unstimulated
and stimulated cells was used to determine the rate of activity. The mean
number of macrophages per sample was calculated by lysing a sub-sample
of cells and counting the number of nuclei as described for WBC and RBC
counts. Values are expressed as NBT reduction (OD 605 nm) per 10
5
cells.
Finally, cumulative mortality was recorded throughout the period of
the trials.
2.4. Statistical analyses
Data were checked for normality and homogeneity of variance by
KolmogornovSmirnov and Bartletts F test. Where appropriate log
transformations where applied to normalise the data. A three-way nested
ANOVA was also applied in order to calculate the overall eects of time,
ploidy, and treatment (SW or FW) on all parameters measured. Tukeys
post hoc tests were applied to identify where signicant dierences
occurred. Preliminary analyses prior to regression analysis were performed
to ensure normality, linearity and homoscedasticity. Relationships
between IGF-I, cortisol, osmolality, glucose and total white blood cell
counts (WBC) were tested using Pearson product moment correlation
coecient and simple linear regression. All statistical analyses were carried
out on Minitab (v14). Signicant dierences were determined at p 6 0.05.
All results are presented as means SEM with saltwater and freshwater
trials referred to by the abbreviations SW and FW, respectively. No signif-
icant dierences were seen between replicates with the exception of head
kidney macrophage respiratory burst activity.
3. Results
3.1. Growth and mortality
Throughout each trial weight, length and condition
factor did not dier signicantly between ploidy or
between trials. During SW challenge total mortality was
2.5% (n = 2) and 5% (n = 4) in triploids and diploids,
respectively, with mortalities occurring at 72 h post-trans-
fer. No mortalities were recorded throughout the
FWFW control trial.
316 J.F. Taylor et al. / General and Comparative Endocrinology 152 (2007) 314325
3.2. Haematology
Haematocrit (Hct) remained constant over time in both
diploids (31.3 2.0% PCV) and triploids (32.0 1.5%
PCV) and did not dier between FWSW and FWFW.
RBC concentration also remained constant over time in
both diploids and triploids and did not dier between FW
SW and FWFW trials. However, triploids had signi-
cantly lower RBC concentrations than diploids (6.5 0.7
vs. 9.2 0.7 10
9
cells ml
1
).
Triploids had signicantly lower WBC concentrations
throughout the FWSW trial but not during the FWFW
trial except at 1 and 72 h post-transfer (Fig. 1a and b).
Resting WBC concentrations were signicantly elevated
within 1 h of seawater transfer and remained so until 3
and 12 h post-transfer in triploid and diploids, respectively
(Fig. 1a). Thereafter levels declined and remained relatively
similar to pre-transfer concentrations. WBC concentration
remained constant throughout the FW trial (Fig. 1b). In
SW, WBC levels were signicantly higher in diploids than
in FW between 1 and 72 h whereas in triploids this was
only observed at 3, 6 and 72 h.
Ploidy did not aect total blood haemoglobin (Hb) in
either trial (Fig. 1c and d). Hb increased signicantly by
12 h post-SW transfer, remained constant and signicantly
higher than pre-transfer levels until 72 h, before returning
to pre-transfer concentrations by day 7 (Fig. 1c). No signif-
icant dierences in Hb were apparent in FW between
ploidy over time, although levels uctuated (Fig. 1d). Mean
corpuscular volume (MCV) remained constant in both tri-
als irrespective of time or salinity, with triploids having a
signicantly higher MCV than diploids (SW: 2n
34.2 3.2 fL vs. 3n 52.0 2.5 fL; FW: 2n 36.4 1.3 fL
vs. 3n 50.3 1.6 fL, p < 0.0001). Mean corpuscular Hb
(MCH) also remained constant in both trials irrespective
of time or salinity, with diploids having a signicantly
lower MCH than triploids (SW: 2n 11.3 0.4 pg vs. 3n
16.3 0.6 pg; FW: 2n 11.4 0.3 pg vs. 3n 15.7 0.6 pg,
p < 0.0001). Mean corpuscular Hb concentration (MCHC)
increased signicantly post-SW transfer in each ploidy
0 6 12 18 24 30 36 42 48 54 60 66 72 168
0 6 12 18 24 30 36 42 48 54 60 66 72 168
W
B
C

(

1
0
7
m
l
-
1
)
0
1
2
3
4
5
2n FW-SW
3n FW-SW
0 6 12 18 24 30 36 42 48 54 60 66 72 168
0
1
2
3
4
5
2n FW-FW
3n FW-FW
a
b
a
b
a*
a*
a*
a*
a
a
a
b
a
b*
b*
b
b
b
b
b
Time (h)
0 6 12 18 24 30 36 42 48 54 60 66 72 168
M
C
H
C

(
g

d
L
-
1
)
0
20
25
30
35
40
45
50
55
Time (h)
0 6 12 18 24 30 36 42 48 54 60 66 72 168
0 6 12 18 24 30 36 42 48 54 60 66 72 168
H
b

c
o
n
c
.

(
g

d
L
-
1
)
0
8
9
10
11
12
13
14
*
*
*
*
*
*
*
*
*
*
*
*
Fig. 1. Time course changes in total white blood cell concentration (WBC, a and b), total blood haemoglobin (Hb, c and d) and mean corpuscular
haemoglobin concentration (MCHC, e and f) in diploid (black) and triploid (white) rainbow trout transferred from FW to SW (circles) or FW to FW
(squares). Values are expressed as means SEM (n = 6 sh/ploidy/tank). Signicant dierences between resting (0 h) and post-transfer (1168 h) values
are indicated by
*
. Signicant dierences between ploidies at a given time point are indicated by dierent superscripts.
J.F. Taylor et al. / General and Comparative Endocrinology 152 (2007) 314325 317
(Fig. 1e). No signicant dierences between ploidy were
observed. Diploids reached a signicantly higher MCHC
than pre-transfer levels at 12 h, while triploids achieved this
by 72 h. In general both ploidy showed a general increase
in MCHC from 12 h onwards, with highest MCHC
observed on day 7 in both ploidy. Conversely, no signi-
cant dierences in MCHC were observed as a factor of time
or ploidy in FW (Fig. 1f).
3.3. Salt balance
SW transfer resulted in rapid and signicant elevations
in plasma osmolality. However, there were no ploidy dier-
ences in SW at any time point (Fig. 2a). Plasma osmolality
was signicantly elevated above pre-transfer levels from 1
to 3 h in triploids and diploids, respectively. Levels
remained signicantly elevated until 48 h, thereafter return-
ing to basal. In FW, handling and transfer caused a brief
disturbance in plasma osmolality of triploids which was
signicantly elevated over pre-transfer levels, and higher
than diploids but returned to basal by 3 h (Fig. 2b). Dip-
loids did not show change in osmolality at this time point.
SW transfer induced an immediate marked increase in
gill Na
+
, K
+
, ATPase activity in both diploids and triploids
from <5 to 45 lmol mg prot
1
h
1
within 24 h of saltwa-
ter transfer (Fig. 2c). Activity returned to basal levels
within 48 and 72 h in triploids and diploids, respectively.
There was no aect of handling on Na
+
, K
+
, ATPase activ-
ity in FW which remained below 5 lmol mg prot
1
h
1
in
both ploidies throughout (Fig. 2d). Overall, statistical anal-
ysis found no overall aect of ploidy during the FWSW or
the FWFW trials.
3.4. Endocrine, stress and immune responses
Ploidy had no signicant eect on plasma IGF-I levels
in either trial (Fig. 3a and b). SW transfer induced a signif-
icant increase in both ploides 1 h post-transfer, which
Time (h)
0 6 12 18 24 30 36 42 48 54 60 66 72 168
0
5
10
15
20
25
30
35
40
45
50
0 6 12 18 24 30 36 42 48 54 60 66 72 168
0
300
320
340
360
380
400
2n FW-FW
3n FW-FW
b
*
*
0 6 12 18 24 30 36 42 48 54 60 66 72 168
O
s
m
o
l
a
l
i
t
y

(
m
O
s
m
)
0
300
320
340
360
380
400
2n FW-SW
3n FW-SW
Time (h)
0 6 12 18 24 30 36 42 48 54 60 66 72 168
A
T
P
a
s
e

A
c
t
i
v
i
t
y

(

m
o
l

m
g
.

p
r
o
t
.
-
1
h
-
1
)
0
5
10
15
20
25
30
35
40
45
50
a*
b
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
* *
*
Fig. 2. Time course changes in plasma osmolality (a and b) and gill Na
+
, K
+
, ATPase activity (c and d) in diploid (black) and triploid (white) rainbow
trout transferred from FW to SW (circles) or FW to FW (squares). Values are expressed as means SEM (n = 6 sh/ploidy/tank). Signicant dierences
between resting (0 h) and post-transfer (1168 h) values are indicated by
*
. Signicant dierences between ploidies at a given time point are indicated by
dierent superscripts.
318 J.F. Taylor et al. / General and Comparative Endocrinology 152 (2007) 314325
remained elevated above basal levels until 12 and 24 h post-
transfer in diploid and triploid sh, respectively (Fig. 3a).
Thereafter levels declined to basal and remained constant.
FWFW transfer in trial 2 did not induce any change in
plasma IGF-I over time in either ploidy, although both
showed slightly elevated levels 3 h post handling (Fig. 3b).
SW transfer induced a signicant elevation in plasma
cortisol in both diploids and triploids which only returned
to basal levels (<20 ng ml
1
) at the end of the experiment
(Fig. 3c). There were no signicant dierences between
ploidy in pre- or post-transfer plasma cortisol levels,
although triploid levels appeared to recover more rapidly
than diploids by 72 h post-transfer. FWFW transfer in
trial 2 induced a signicant elevation in plasma cortisol lev-
els which subsequently returned to basal levels within 12 h
(Fig. 3d). No signicant dierences in pre-transfer levels
were apparent between SW and FW trials, while maximum
peak cortisol levels (1 h post-stressor) achieved in FW were
signicantly lower than observed in SW for both ploidies.
No signicant eect of ploidy was apparent on plasma
glucose concentrations in the SW trial, while ploidy had
an eect in FW (Fig. 3e and f). SW transfer resulted in a
signicant elevation of plasma glucose concentrations
which were maintained signicantly higher than pre-trans-
fer levels from 1 to 48 h in triploids and 1 to 24 h in dip-
loids (Fig. 3e. Levels in both ploidies returned to basal
levels by the end of the experiment. In the FW trial diploid
levels were elevated above basal 3 h post-transfer, remained
constant until 48 h and then returned to pre-stress levels by
72 h (Fig. 3f). Triploid levels also increased signicantly,
peaking at 6 h, signicantly higher than in diploids. There-
after, levels declined steadily to pre-stress levels by 48 h,
signicantly lower than the diploids. Basal levels were
attained by 72 h and remained constant to day 7.
SW transfer also resulted in elevated plasma lysozyme
activities within 12 and 24 h in triploids and diploids, respec-
tively (Fig. 4a). There after lysozyme activity remained ele-
vated above basal for the remainder of the experiment.
Time (h)
0 6 12 18 24 30 36 42 48 54 60 66 72 168
G
l
u
c
o
s
e

c
o
n
c
.

(
m
g

d
l
-
1
)
0
60
80
100
120
140
160
180
200
0 6 12 18 24 30 36 42 48 54 60 66 72 168
C
o
r
t
i
s
o
l

c
o
n
c
.

(
n
g

m
l
-
1
)
0
20
40
60
80
100
120
140
160
2n FW-SW
3n FW-SW
0 6 12 18 24 30 36 42 48 54 60 66 72 168
I
G
F
-
I

c
o
n
c
.

(
n
g

m
l
-
1
)
0
50
100
150
200
250
300
0 6 12 18 24 30 36 42 48 54 60 66 72 168
0
50
100
150
200
250
300
Time (h)
0 6 12 18 24 30 36 42 48 54 60 66 72 168
0
60
80
100
120
140
160
180
200
0 6 12 18 24 30 36 42 48 54 60 66 72 168
0
20
40
60
80
100
120
140
160
2n FW-FW
3n FW-FW
*
*
*
*
*
*
*
*
* *
*
*
a*
b
a
b
*
*
*
*
*
*
*
*
*
*
*
*
*
* *
*
*
*
*
*
*
*
*
*
*
*
Fig. 3. Time course changes in plasma IGF-I (a and b) cortisol (c and d) and glucose concentration (e and f) in diploid (black) and triploid (white) rainbow
trout transferred from FW to SW (circles) or FW to FW (squares). Values are expressed as means SEM (n = 6 sh/ploidy/tank). Signicant dierences
between resting (0 h) and post-transfer (1168 h) values are indicated by
*
. Signicant dierences between ploidies at a given time point are indicated by
dierent superscripts.
J.F. Taylor et al. / General and Comparative Endocrinology 152 (2007) 314325 319
Triploids had signicantly higher levels of activity than dip-
loids after 12 h in SW, but at no other point were ploidy dif-
ferences observed. Plasma lysozyme activity did not change
during the FW trial in either ploidy (Fig. 4b). Lysozyme
activity in both triploids and diploids in the SWtrial was sig-
nicantly elevated above those in the FW trial from 12 h
onwards in triploids and 24 h onwards in diploids.
There was no signicant eect of ploidy on mucus lyso-
zyme activities in either trial, although triploids generally
had lower resting levels (Fig. 4c and d). A depression in
mucus lysozyme activity was observed within 3 h of trans-
fer to SW in both ploidies although not signicant due to
high variation (Fig. 4c). Activities at the nal time point
were signicantly lower than resting level in diploids.
FWFW transfer did not aect mucus lysozyme activity
(Fig. 4d). No signicant dierences were seen between the
SW and FW trials at any time point.
Replicate dierences for respiratory burst were observed
between the two SW trials which is believed to be due to
poor adhesion of macrophages to the ELISA plate in the
assay of replicate 2 as indicated by low nuclei counts fol-
lowing lysis. As such, replicate 2 was discarded from anal-
ysis (Table 1). However, triploid macrophages generally
displayed a 50% greater respiratory burst activity than dip-
loids in both saltwater and freshwater trials although this
dierence was only signicant at 24 h in replicate 1 (Table
1). Both diploids and triploid macrophages also displayed
an increased activity by approximately 50% within 24 h
post-SW transfer. Activity peaked at 72 h in both ploi-
dies returning to basal levels by the end of the experiment.
Conversely, FWFW transfer did not appear to stimulate
activity.
A strong positive correlation was found between plasma
IGF-I and plasma cortisol levels in the saltwater trials irre-
spective of ploidy (Fig. 5a). Conversely, no such relation-
ship was found in the freshwater trial (Fig. 5b). In both
ploidy, strong positive correlations were found between
plasma cortisol and osmolality, WBC and glucose follow-
0 6 12 18 24 30 36 42 48 54 60 66 72 168
P
l
a
s
m
a

l
y
s
o
z
y
m
e

a
c
t
i
v
i
t
y

(
u
n
i
t
s

m
i
n
-
1

m
l
-
1
)
0
500
1000
1500
2000
2500
3000
2n FW-SW
3n FW-SW
0 6 12 18 24 30 36 42 48 54 60 66 72 168
0
500
1000
1500
2000
2500
3000
2n FW-FW
3n FW-FW
Time (h)
0 6 12 18 24 30 36 42 48 54 60 66 72 168
M
u
c
u
s

l
y
s
o
z
y
m
e

a
c
t
i
v
i
t
y

(
u
n
i
t
s

m
i
n
-
1

m
l
-
1
)
0
100
200
300
400
500
600
Time (h)
0 6 12 18 24 30 36 42 48 54 60 66 72 168
0
100
200
300
400
500
600
*
*
*
* *
*
*
*
a*
b
*
Fig. 4. Time course changes in plasma lysozyme (a and b) and mucus lysozyme activity (c and d) in diploid (black) and triploid (white) rainbow trout
transferred from FW to SW (circles) or FW to FW (squares). Values are expressed as means SEM (n = 6 sh/ploidy/tank). Signicant dierences
between resting (0 h) and post-transfer (1168 h) values are indicated by
*
. Signicant dierences between ploidies at a given time point are indicated by
dierent superscripts.
320 J.F. Taylor et al. / General and Comparative Endocrinology 152 (2007) 314325
ing FWSW transfer (Table 2). However, this relationship
was only found between cortisol and osmolality following
FWFW transfer. In contrast, a strong negative relation-
ship was found between cortisol and plasma lysozyme in
both ploidies following FWSW transfer, but was not
found in the FWFW trial. Similarly, strong positive corre-
lations were found between plasma IGF-I and osmolality,
WBC and glucose in both FWSW and FWFW trials.
A strong negative relationship was also found between
IGF-I and plasma lysozyme in both ploidies following
FWSW transfer, but not in the FWFW trial.
4. Discussion
Few studies have investigated osmoregulation and stress
responses in non-smolting salmonids transferred to seawa-
ter (Jackson, 1981; Alexis et al., 1984; Hegab and Hanke,
1986; Besner and Pelletier, 1991; Almendras et al., 1993;
Table 1
Respiratory burst activity (per 10
5
cells) determined by NBT reduction in the presence of PMA by head kidney leukocytes prior- (0 h) and 24, 72, and
168 h post-transfer to seawater and freshwater in diploid and triploid rainbow trout
Treatment Ploidy Time (h)
0 24 72 168
FWSW Diploid 0.54 0.13
a
1.04 0.25
a
1.57 0.44
ab
0.47 0.14
b
Triploid 1.97 0.81
a
3.31 0.97
b
3.92 1.31
b
1.95 0.67
b
FWFW Diploid 0.27 0.08
a
0.22 0.04
a
0.16 0.02
a
0.25 0.04
a
Triploid 0.43 0.09
a
0.97 0.13
a
0.50 0.10
a
0.27 0.06
a
Values given are means SEM (n = 6 sh/ploidy/tank). Signicant dierences between ploidies at a given time point are indicated by dierent letters.
Cortisol (ng ml
-1
) Cortisol (ng ml
-1
)
0 20 40 60 80 100 120 140 160
I
G
F
-
I

(
n
g

m
l
-
1
)
I
G
F
-
I

(
n
g

m
l
-
1
)
0
50
100
150
200
250
300
350
y = 1.95x - 7.78
r
2
= 0.71
p<0.0001
0 20 40 60 80 100 120 140 160
0
50
100
150
200
250
300
350
Fig. 5. Linear correlation between plasma IGF-I and cortisol in diploid (black) and triploid (white) rainbow trout transferred from (a) FWSW (circles) or
(b) FWFW (squares). Values are expressed as mean of each time point SEM (n = 6) per ploidy.
Table 2
Linear correlations between mean plasma IGF-I or cortisol and dierent variables sampled throughout FWSW and FWFW trials in diploid (2n) and
triploid (3n) rainbow trout
Parameter IGF-I Cortisol
FWSW FWFW FWSW FWFW
2n 3n 2n 3n 2n 3n 2n 3n
Osmolality 0.22 0.62 0.37 ns 0.39 0.76 0.23 0.56
+ + + + + + +
WBC 0.55 0.50 0.70 0.30 0.65 0.64 ns ns
+ + + + + +
Glucose 0.21 0.72 0.24 0.42 0.27 0.50 ns ns
+ + + + + +
Plasma lysozyme 0.88 0.80 ns ns 0.43 0.57 ns ns

r
2
values are given, with + or symbol indicating direction of relationship. +, positive correlation; , negative correlation; ns, non-signicant.
J.F. Taylor et al. / General and Comparative Endocrinology 152 (2007) 314325 321
Liebert and Shreck, 2006). Furthermore, comparisons of
such studies are made dicult by the diversity of species,
salinities and speed of transfer tested. A limited amount
of work has focused on triploids (Biron and Benfey,
1994; Benfey and Biron, 2000; Sadler et al., 2000a,2000b;
Hyndman et al., 2003a,2003b) and to our knowledge no
studies regarding saltwater adaptation and immune func-
tion in triploid rainbow trout have been published to date.
This work is thus the rst to address within the same study
the inuence of ploidy on osmoregulatory, stress and
immune responses in a non-smolting rainbow trout strain
following salt water challenge.
Our ndings clearly show non-smolting rainbow trout
were able to adapt to full strength seawater within 1 week
regardless of ploidy. SW transfer in the current study
resulted in an immediate elevation in plasma osmolality.
Since these values returned to basal within 1 week, did
not exceed the lethal limit for trout (>420 mOsm kg
1
,
Alexis et al., 1984) and were accompanied by virtually no
mortality, it can be concluded that sh had successfully
adapted to seawater and entered a regulatory period. The
timeframe to regain osmotic balance described in this
experiment is similar to that of earlier studies with smaller
rainbow trout (1330 g) albeit they were exposed to lower
salinities (2330 ppt) (Bath and Eddy, 1979; Jackson,
1981; Alexis et al., 1984). Specically, the current study
has shown that abrupt transfer to SW transfer without
the use of intermediate salinities as previously studied in
brook trout (Salvelinus fontinalis) ploidy variants (Dumas
et al., 1995) is tolerated both in triploid and diploid rain-
bow trout.
Our study also induced a marked elevation in gill Na
+
,
K
+
, ATPase activity from <5 to 45 lmol mg prot
1
h
1
within 24 h of SW transfer. This elevation was preceded
by a parallel increase in both plasma IGF-I and cortisol,
which also correlated positively with plasma osmolality,
and concurs with other studies reporting their importance
in regulating hydromineral balance (McCormick, 1996;
Liebert and Shreck, 2006). Since osmolality at the end of
the experiment indicated that ionic-osmotic balance had
been resumed it was presumed that this pattern in ATPase
activity was a normal function of SW adaptation in rain-
bow trout. The current ndings make an interesting com-
parison with smolting salmonids. In Atlantic salmon
smolts transferred to full strength seawater, ATPase activ-
ity took approximately 8 days to increase by
10 lmol mg prot
1
h
1
, and basal levels were resumed
within 14 days (Handeland et al., 1998). Therefore it
appears that the response in ATPase activity exhibited by
rainbow trout in seawater is compressed and amplied
(23 times higher) when compared to that of salmon prob-
ably reecting the greater severity of the challenge and the
need for immediate adaptation in our study. Conversely,
other studies have shown ATPase activity to follow a more
similar pattern to that of salmon in a number of trout spe-
cies (Besner and Pelletier, 1991; Almendras et al., 1993).
However, direct comparisons between studies are dicult
due to species and strain dierences, in addition to the
inuence of diering salinities, temperatures and body sizes
tested on saltwater acclimation (Hegab and Hanke, 1986;
Bonnet et al., 1999; Handeland et al., 2000, 2004).
The major changes associated with saltwater adapta-
tion, such as increased ATPase activity, result in increased
energy and oxygen demands by osmoregulatory tissues
(Sangiao-Alvarellos et al., 2005). This would certainly
explain the rapid elevation of total blood Hb within 6 h
of SW transfer in our study, and would also compensate
for the lower oxygen anity of the saltwater environment.
Our ndings showed no ploidy dierences with regards to
Hb or MCHC although MCH was as expected signicantly
greater in triploids. Furthermore, we saw no major acute
changes in basic haematological parameters (RBC and
Hct) following FWSW transfer. This lends support to ear-
lier work which has shown that the eciency of oxygen
uptake during exercise and after stress is not aected by
ploidy (Sadler et al., 2000a; Hyndman et al., 2003b).
In our study both FWSW and FWFW transfer
resulted in a typical acute stress response characterised by
a marked increase in plasma cortisol concentration. Pre-
(14.58 2.37 ng ml
1
) and post-SW stress (130
14.7 ng ml
1
) cortisol levels were within the ranges previ-
ously described for rainbow trout with regards to acute
stressors (Barton and Iwama, 1991). However, acute stress
is usually characterised by an immediate elevation in corti-
sol which returns to resting levels within a few hours
(Wendelaar Bonga, 1997) as observed in the FWFW trial.
In contrast, following FWSW transfer, cortisol levels
remained elevated up to 72 h. Since this response was sim-
ilar between diploid and triploids, it can be assumed that
both experienced the same degree of physiological stress
during SW adaptation as suggested in brook trout (Biron
and Benfey, 1994), rainbow trout (Benfey and Biron,
2000) and Atlantic salmon (Sadler et al., 2000a). However,
the current ndings expand upon these earlier studies since
both the acute time course as well as the eect of seawater
as a stressor were examined. The degree of post-stress
hyperglycemia, has also been found to be similar between
ploidies in rainbow and brook trout (Biron and Benfey,
1994; Benfey and Biron, 2000). Although the initial
response in hyperglycemia was greater in triploids follow-
ing SW transfer in the current study there was no dierence
between ploidies at any given time point. Furthermore,
handling stress in the FWFW trial also induced a typical
secondary stress response. However, the greater elevation
and more rapid return to basal levels in FW may be indic-
ative of a quicker exhaustion of energy reserves following a
single stress event as previously postulated (Hyndman
et al., 2003b). On the other hand, our results following
FWSW transfer suggest that although the adaptive phase
was a signicantly stressful event, it cannot be assumed
that this was the case throughout the experimental period,
as levels remained elevated signicantly longer than in the
FWFW trial. In this respect, cortisol, as previously
alluded to, is also recognised as a major ionic-osmotic
322 J.F. Taylor et al. / General and Comparative Endocrinology 152 (2007) 314325
regulatory hormone in teleost sh associated with increas-
ing ATPase activity and plasma glucose, through increased
liver gluconeogenesis, which subsequently provides a sub-
strate for the greater energetic demands of SW adaptation
(McCormick, 1996, 2001; Liebert and Shreck, 2006). This
would certainly explain the positive correlation between
glucose and cortisol found in the FWSW trial but not in
the FWFW trial. Therefore, in the current study cortisol
and glucose are undoubtedly performing a dual role during
the adaptive phase following SW transfer.
Seawater transfer also resulted in an immediate dou-
bling in total WBC concentrations of both diploid and trip-
loid trout thus inferring a potentially enhanced cellular
immune capacity. This is in contrast to the generally
accepted view that acute stressors result in a depression
in WBC concentration (Schreck, 1997; Benfey and Biron,
2000) due to the immunosuppressive eects of cortisol.
However, Pulsford et al. (1994) also reported increased
WBC concentrations in dab (Limanda limanda) following
an acute stressor, which was largely due to increases in
peripheral phagocyte concentration. In this respect, it has
been suggested that cortisol protects trout leucocytes
against the immediate aects of stress by maintaining phag-
ocytic index (Narnaware and Baker, 1996). Thus, the
strong positive correlation we observed between cortisol
and WBC in the FWSW trials could support such a the-
ory. Unfortunately, as no dierential leukocyte counts
were made within our study we cannot derive which cell
type may have been responsible for this increase and this
certainly warrants further study.
An increase in peripheral and head kidney WBC concen-
trations has also been observed within 2 weeks of SW
transfer in diploid Atlantic salmon smolts (Pettersen
et al., 2003). It may be possible that increased WBC con-
centrations are a specic reaction to SW transfer which is
independent of the generic stress response. This may also
be due to increases in circulating GH observed during sea-
water transfer (McCormick, 2001) since this hormone has
been suggested to prevent the immunosuppressive activity
of cortisol (Yada et al., 2004). Certainly since we demon-
strated a signicant positive correlation between plasma
IGF-I and total WBC, we could postulate that IGF-I
may be involved in promoting some of the eects of GH
at the cellular level, although this remains to be investi-
gated. Clearly, the high degree of positive correlation
between both plasma cortisol and IGF-I in the FWSW tri-
als, and subsequently with WBC counts, necessitates the
individual role of each hormone in these various responses
to be elucidated. Furthermore the interaction between
stress and IGF-I should be given greater consideration.
We clearly showed an elevation in IGF-I following FW
SW transfer but also a minor elevation within 1 h of
FWFW transfer. Similarly, McCormick et al. (1998)
showed elevated IGF-I within 4 h of an acute stress, while
Liebert and Shreck (2006) observed no change in IGF-I
after an acute stressor and saltwater (25 ppt) transfer which
may in part relate to the role of IGFBPs. Of note, the lack
of correlation between IGF-I and cortisol in the FWFW
trials suggest that, although handling clearly evoked a
typical stress elevation in cortisol, it is not in itself
directly driving the increase in IGF-I.
At present, although both ploidies increased their WBC
concentrations this was most pronounced in diploids.
However, given the morphological dierences between dip-
loid and triploid cells it is impossible to accurately infer a
greater cellular immune capacity in diploids at this time.
One of the major eects of ploidy seen in the current study
was the apparent greater respiratory burst activity (approx-
imately 50% higher) of head kidney macrophages in trip-
loids. Although phagocytic activity was examined during
this study, results were not reported due to a problem of
poor macrophage adhesion to the slides with the assay.
However, this should undoubtedly be addressed in further
studies as this is a major indicator of innate immunity
which may be inuenced by ploidy status. It may therefore
be suggested that the greater respiratory burst observed in
triploid head kidney macrophages, and possibly peripheral
leucocytes, may allow triploid sh to compensate for lower
WBC concentrations by having a higher killing capacity.
This is clearly an interesting avenue for further study and
has only recently been investigated in turbot (Budino
et al., 2006).
In the current experiment increased levels in plasma
lysozyme activity stimulated by seawater are in agreement
with previous studies in rainbow trout exposed to dilute
(12 ppt) seawater (Yada et al., 2001). Notably, we also
found a strong negative correlation between lysozyme
activity and cortisol or IGF-I which contradicts current
opinion (Fevolden et al., 2003) but are in accordance with
North et al. (2006). To date few reports exist, although it is
widely accepted that corticosteroids are known to be
potent immunosuppressants. In terms of mucus lysozyme,
SW transfer did not appear to statistically aect activity
when compared to controls in freshwater. However, a more
pronounced general depression in activity was seen when
compared to basal levels before transfer, which concurs
with observations of Fast et al. (2002) in rainbow trout,
coho salmon (Oncorhynchus kisutch) and Atlantic salmon.
In summary, on transfer to full strength seawater, non-
smolting rainbow trout undergo an initial adaptive phase
lasting up to 72 h before ionic-osmotic balance is re-estab-
lished. This is accompanied with marked increases in ATP-
ase activity which is more intense and short lived than those
exhibited by salmon smolts. This process appears to be
accompanied by a certain amount of physiological stress
as indicated by elevated cortisol and glucose, although the
level of stress can not be directly inferred due to the dual
role of these factors in ionic-osmotic regulation. Seawater
transfer also increased blood leukocyte concentrations,
respiratory burst and plasma lysozyme activity but
appeared to decrease mucus lysozyme activity. This might
suggest a generally improved defence mechanism against
pathogens during this physiologically stressful time. We
also showed these responses do not dier markedly between
J.F. Taylor et al. / General and Comparative Endocrinology 152 (2007) 314325 323
diploid and triploid trout of the same strain and parental
origin. On the basis of our results it may be concluded that
an inability to adapt to changing salinity is not the sole rea-
son behind the high rates of triploid mortalities reported
during seawater transfer. It is more likely the result of multi-
ple stressors (handling, temperature, pollution etc.) occur-
ring alongside changing salinity.
Acknowledgments
The authors like to extend their thanks to all the techni-
cal sta at the NBFRL and Institute of Aquaculture for
Fish Maintenance during these trials. A special thanks goes
to Dr. Iain Berrill for assistance with the ATPase assay.
This project was funded by the Institute of Aquaculture,
University of Stirling.
References
Alexis, M.N., Papaparaskeva-Papoutsoglou, E., Papoutsoglou, S., 1984.
Inuence of acclimation temperature on the osmotic regulation and
survival of rainbow trout (Salmo gairdneri) rapidly transferred from
fresh water to sea water. Aquaculture 40, 333341.
Almendras, J.M.E., Prunet, P., Buf, G., 1993. Responses of a non-
migratory stock of brown trout (Salmo trutta) to ovine growth
hormone treatment and seawater exposure. Aquaculture 114, 169179.
Barton, B., Iwama, G.K., 1991. Physiological changes in sh from stress in
aquaculture with emphasis on the response and eects of corticoste-
roids. Ann. Rev. Fish Dis., 326.
Bath, R.N., Eddy, F.B., 1979. Salt and water balance in rainbow trout
(Salmo gairderi R.) rapidly transferred from freshwater to seawater. J.
Exp. Biol. 83, 193202.
Benfey, T.J., Sutterlin, A.M., 1984. The haematology of triploid
landlocked salmon (Salmo salar L.). J. Fish Biol. 24, 3338.
Benfey, T.J., 1999. The physiology and behaviour of triploid shes. Rev.
Fish. Sci. 7, 3867.
Benfey, T.J., Biron, M., 2000. Acute stress response in triploid rainbow
trout (Oncohynchus mykiss) and brook trout (Salvelinus fontinalis).
Aquaculture 184, 167176.
Besner, M., Pelletier, D., 1991. Adaptation of the brook trout (Salvelinus
fontinalis) to direct transfer to sea water in spring and summer.
Aquaculture 97, 217230.
Biron, M., Benfey, T.J., 1994. Cortisol, glucose and hematocrit changes
during acute stress, cohort sampling, and the diel cycle in diploid and
triploid brook trout (Salvelinus fontinalis M.). Fish Physiol. Biochem.
13, 153160.
Bonnet, S., Haray, P., Blanc, J.M., Vallee, F., Vauchez, C., Faure, A.,
Fauconneau, B., 1999. Genetic variation in growth parameters until
commercial size in diploid and triploid freshwater rainbow trout
(Oncorhynchus mykiss) and seawater brown trout (Salmo trutta).
Aquaculture 173, 359375.
Bromage, N., Porter, M., Randall, C., 2001. The environmental regulation
of maturation in farmed nsh with special reference to the role of
photoperiod and melatonin. Aquaculture 197, 6398.
Budino, B., Cal, R.M., Piazzon, M.C., Lamas, J., 2006. The activity of
several components of the innate immune system in diploid and
triploid turbot. Comp. Biochem. Physiol. (A) 130, 411423.
Cal, R.M., Vidal, S., Go mez, C., A

lvarez-Blazquez, B., Mart nez, P.,

Piferrer, F., 2006. Growth and gonadal development in diploid and
triploid turbot (Scophthalmus maximus). Aquaculture 251, 99108.
Cotter, D., ODonovon, V., OMaoileidigh, N., Rogan, G., Roche, N.,
Wilkins, N.P., 2000. An evaluation of the use of triploid Atlantic
salmon (Salmo salar) in minimizing the impact of escaped farmed
salmon on wild populations. Aquaculture 186, 6175.
Dumas, S., Audet, C., Blanc, J.M., De la Nou e, J., 1995. Seawater
acclimation of diploid and triploid brook charr (Salvelinus fontinalis),
diploid Arctic charr (Salvelinus alpinus), and their diploid and triploid
hybrids. J. Fish Biol. 46, 302316.
Ellis, A.E., 1990. Lysozyme assays. In: Stolen, J., Fletcher, T., Anderson,
D., Robertson, B., Van Muiswinkel, W. (Eds.), Techniques in Fish
Immunology. SOS Publications, Fairhaven, USA.
Ellis, T., James, J.D., Stewart, C., Scott, A.P., 2004. A non-invasive stress
assay based upon measurement of free cortisol released into the water
by rainbow trout. J. Fish Biol. 65, 12331252.
Fast, M.D., Sims, D.E., Burka, J.F., Mustafa, A., Ross, N.W., 2002. Skin
morphology and humoral non-specic defence parameters of mucus
and plasma in rainbow trout, coho and Atlantic salmon. Comp.
Biochem. Physiol. (A) 132, 645657.
Fevolden, S.-E., Red, K.H., Fjalestad, K., 2003. A combined salt and
connement stress enhances mortality in rainbow trout (Oncorhynchus
mykiss) selected for high stress responsiveness. Aquaculture 216, 67
76.
Handeland, S.O., Berge, A

., Bjo rnsson, B.Th., Stefansson, S.O., 1998.

Eects of temperature and salinity on osmoregulation and growth of
Atlantic salmon (Salmo salar L.) smolts in seawater. Aquaculture 168,
289302.
Handeland, S.O., Berge, A

., Bjo rnsson, B.Th., Lie, ., Stefansson, S.O.,

2000. Seawater adaptation by out-of-season Atlantic salmon (Salmo
salar L.) smolts at dierent temperatures. Aquaculture 181,
377396.
Handeland, S.O., Wilkinson, E., Sveinsb, B., McCormick, S.D.,
Stefansson, S.O., 2004. Temperature inuence on the development
and loss of seawater tolerance in two fast-growing strains of Atlantic
salmon. Aquaculture 233, 513529.
Hegab, S.A., Hanke, W., 1986. Electrolyte changes, cell volume regulation
and hormonal inuences during acclimation of rainbow trout (Salmo
gairdneri) to salt water. Comp. Biochem. Physiol. 83, 4752.
Hyndman, C.A., Kieer, J.D., Benfey, T.J., 2003a. The physiological
response of diploid and triploid brook trout to exhaustive exercise.
Comp. Biochem. Physiol. (A) 134, 167179.
Hyndman, C.A., Kieer, J.D., Benfey, T.J., 2003b. Physiology and
survival of triploid brook trout following exhaustive exercise in warm
water. Aquaculture 221, 629643.
Jackson, A.J., 1981. Osmotic regulation in rainbow trout (Salmo gairdneri)
following transfer to sea water. Aquaculture 24, 143151.
Liebert, A.M., Shreck, C.B., 2006. Eect of acute stess on osmoregulation,
feed intake, IGF-I, and cortisol in yearling steelhead trout (Oncorhyn-
chus mykiss) during seawater adaptation. Gen. Comp. Endocrinol.
148, 195202.
McCormick, S.D., 1993. Methods for nonlethal gill biopsy and measure-
ment of Na
+
, K
+
-ATPase activity. Can. J. Fish. Aquat. Sci. 50,
656658.
McCormick, S.D., 1996. Eects of growth hormone and insulin-like
growth factor-I on salinity tolerance and gill Na
+
, K
+
-ATPase in
Atlantic salmon (Salmo salar): interaction with cortisol. Gen. Comp.
Endocrinol. 101, 311.
McCormick, S.D., 2001. Endocrine control of osmoregulation in teleost
sh. Am. Zool. 41, 781794.
McCormick, S.D., Bradshaw, D., 2006. Hormonal control of salt and
water balance in vertebrates. Gen. Comp. Endocrinol. 147, 38.
McCormick, S.D., Shrimpton, J.M., Carey, J.B., ODea, M.F., Sloan,
K.E., Moriyama, S., Bjornsson, B.T., 1998. Repeated acute stress
reduces growth rate of Atlantic salmon parr and alters plasma levels of
growth hormone, insulin-like growth factor-I and cortisol. Aquacul-
ture 168, 221235.
McCormmick, S.D., 1995. Hormonal control of gill Na
+
, K
+
-ATPase
and chloride cell function. In: Wood, C.M., Shuttleworth, T.J. (Eds.),
Cellular and Molecular Approaches to Fish Ionic Regulation, Vol. 14.
Academic Press, San Diego, pp. 285315.
Narnaware, Y.K., Baker, B.I., 1996. Evidence that cortisol may protect
against the immediate eects of stress on circulating leukocytes in the
trout. Gen. Comp. Endocrinol. 103, 359366.
324 J.F. Taylor et al. / General and Comparative Endocrinology 152 (2007) 314325
North, B.P., Turnbull, J.F., Ellis, T., Porter, M.J., Migaud, H., Bron, J.,
Bromage, N.R., 2006. The impact of stocking density on the welfare of
rainbow trout (Oncorhynchus mykiss). Aquaculture 255, 466479.
Ojolick, E.J., Cusack, R., Benfey, T.J., Kerr, S.R., 1995. Survival and
growth of all-female diploid and triploid rainbow trout (Oncorhynchus
mykiss) reared at chronic high temperature. Aquaculture 131, 177187.
Perruzi, S., Varsamos, S., Chatain, B., Fauvel, C., Menu, B., Falguiere,
J.-C., Severe, A., Flik, G., 2005. Haematological and physiological
characteristics of diploid and triploid sea bass, Dicentrarchus labrazx
L. Aquaculture 244, 359367.
Pettersen, E.F., Ulvenes, M., Melingen, G.0., Wergeland, H.I., 2003.
Peripheral blood and head kidney leucocyte populations during out-
of-season (0+) parr-smolt transformation and seawater transfer of
Atlantic salmon (Salmo salar L.). Fish Shellsh Immunol. 15, 373385.
Pulsford, A.L., Lemaire-Gony, S., Tomlinson, M., Collingwood, N.,
Glynn, P.J., 1994. Eects of acute stress on the immune system of the
dab (Limanda limanda). Comp. Biochem. Physiol. (C) 109, 129139.
Sadler, J., Wells, R.N.G., Pankhurst, P.M., Pankhurst, N.W., 2000a.
Blood oxygen transport, rheology and haematological responses to
connement stress in diploid and triploid Atlantic salmon (Salmo
salar). Aquaculture 184, 349361.
Sadler, J., Pankhurst, P.M., King, H.R., 2000b. High prevalence of
skeletal deformity and reduced gill surface area in triploid Atlantic
salmon (Salmo salar L.). Aquaculture 198, 369386.
Salaman, N.A., Eddy, F.B., 1987. Response of chloride cell numbers and
gill Na
+
/K
+
ATPase activity of freshwater trout (Salmo gairneri R.) to
salt feeding. Aquaculture 61, 4148.
Sangiao-Alvarellos, S., Arjona, F.J., Mart n del R o, M.P., M guez, J.M.,
Mancera, J.M., Soengas, J.L., 2005. Time course of osmoregulatory
and metabolic changes during osmotic acclimation in Sparus auratus.
J. Exp. Biol. 208, 42914304.
Schreck, C.B., 1997. Immunomodulation: endogenous factors. In: Iwama,
G.K., Nakanishi, T. (Eds.), In The sh Immune System; Organism,
Pathogen and Environment. Academic Press, San Diego, pp. 311317.
Secombes, C.J., 1990. Isolation of salmonid macrophages and analysis of
their killing activity. In: Stolen, J., Fletcher, T., Anderson, D.,
Robertson, B., Van Muiswinkel, W. (Eds.), Techniques in Fish
Immunology. SOS Publications, Fairhaven, USA.
Thompson, I., White, A., Fletcher, T.C., Houlihan, D.F., Secombes, C.J.,
1993. The eect of stress on the immune response of Atlantic salmon
(Salmo salar L.) fed diets containing dierent amounts of vitamin C.
Aquaculture 114, 118.
Vazzana, M., Cammarata, M., Cooper, E.L., Parrinello, N., 2002.
Connement stress in sea bass (Dicentrarchus labrax) depresses
peritoneal leukocyte cytotoxicity. Aquaculture 210, 231243.
Wendelaar Bonga, S.E., 1997. The stress response in sh. Physiol. Rev. 77,
591625.
Yada, T., Azuma, T., Takagi, Y., 2001. Stimulation of non-specic
immune functions in seawater-acclimated rainbow trout (Oncorhyn-
chus mykiss) with reference to the role of growth hormone. Comp.
Biochem. Physiol. 129, 695701.
Yada, T., Misumi, I., Muto, K., Azuma, T., Schreck, C.B., 2004. Eects of
prolactin and growth hormone on proliferation and survival of
cultured trout leucocytes. Gen. Comp. Endocrinol. 136, 298306.
J.F. Taylor et al. / General and Comparative Endocrinology 152 (2007) 314325 325