Efficacy of an equal blend of canola oil and poultry fat
as an alternate dietary lipid source for Atlantic salmon
(Salmo salar L.) in seawater. II: effects on haematology
and immunocompetence
Shannon K Balfry
1
, Janice Oakes
2
, Mahmoud Rowshandeli
2
, Greg Deacon
3
, Brent J Skura
1
&David
A Higgs
2
1
Centre forAquaculture and Environmental Research, Faculty of Land and Food Systems,The University of British Columbia,
WestVancouver, BC, Canada
2
Department of Fisheries and Oceans, Centre forAquaculture and Environmental Research, University of British Columbia,
WestVancouver, BC, Canada
3
Skretting Canada Ltd.,Vancouver, BC, Canada
Correspondence: S K Balfry, Centre forAquaculture and Environmental Research, Faculty of Land and Food Systems,The University of
British Columbia, 4160 Marine Drive,West Vancouver, BC, CanadaV7V1N6. E-mail: balfry@interchange.ubc.ca
Abstract
This research examined the haematological and im-
munological responses of quadruplicate groups of ju-
venile (
400 g initial weight) Atlantic salmon
(Salmo salar L.) that had each been fed daily to satia-
tion for 12 weeks one of three high-energy extruded
diets of identical composition except for the supple-
mental dietary lipid (234.7 g kg
1
) source. The three
experimental diets varied in the composition of sup-
plement lipid; diet 1 contained 100% anchovy oil
(AO), while diets 2 and 3 replaced 29.8% and 59.7%
of the AO (respectively) with a 1:1 blend of canola oil
(CO) and poultry fat (PF). Immediately following the
feeding trial, a random sample of ¢sh from each diet
was sampled for determination of baseline levels of
various haematological and immunological para-
meters. Thereafter, duplicate diet groups were vacci-
nated (against Listonella anguillarum) and reared on
their respective experimental diets for an additional
4 weeks. At that time, the remaining ¢sh were
sampled similarly, and the di¡erent parameters were
measured again. Comparisons between the di¡erent
diet treatment groups were made before and after
vaccination. There were no signi¢cant diet treatment
e¡ects at either sample time, for haematocrit, di¡er-
ential leucocyte counts, erythrocyte counts, serum
hemolytic activity or head kidney leucocyte respira-
tory burst activity. The ¢sh fed diet 1 however, did
show signi¢cantly higher post-vaccination levels of
peripheral blood leucocyte respiratory burst activity
and higher serumantibody titres against L. anguillar-
um.The results suggest that the relatively lown-6/n-3
fatty acid ratios in the muscle and presumably other
tissues of ¢sh fed diet 1, may have resulted in a re-
duced production of immunocompromising eicosa-
noids than were produced in ¢sh ingesting the other
two diets that were based in part on the di¡erent
amounts of the CO and PF blend. Long-term studies
are required to con¢rm this possibility.
Keywords: Atlantic salmon, canola oil, poultry
fat, immunocompetence, haematology, fatty acids
Introduction
Because of the increasing demands for marine ¢sh
oils (MFO) for the manufacture of aquafeeds, it is
expected that the cost of this limited resource will
correspondingly increase in the future (Hardy, Higgs,
Lall & Tacon 2001). Therefore, in an attempt to o¡set
the anticipated increase in the cost of manufacturing
high-energy grower diets for Altantic salmon (Salmo
salar) based on MFO, research has been underway
to address the need to ¢nd suitable alternate dietary
lipid sources of plant and/or animal origin that pro-
duce high ¢sh growth and optimal health as well as
a marketable product for the consumer. Dietary lipids
are used primarily as sources of energyand essential
fatty acids, but they also play vital roles in many as-
Aquaculture Research, 2006, 37, 192^199 doi:10.1111/j.1365-2109.2005.01421.x
192 r 2006 TheAuthors. Journal Compilationr 2005 Blackwell Publishing Ltd

pects of the physiology of Atlantic salmon. Indeed,
dietary lipids of optimal fattyacid compositionare re-
quired for the normal functioning of the immune, re-
productive and nervous systems (Higgs & Dong
2000; Balfry & Higgs 2001). Some lipid sources such
as MFO are characteristically rich in the n-3 highly
unsaturated fatty acids (n-3 HUFAs) namely, eicosa-
pentaenoic acid (EPA; 20:5 (n-3)) and docosahexae-
noic acid (DHA; 22:6 (n-3)) and are known essential
dietary components.
Juvenile Atlantic salmon in sea water have very
high dietary lipid requirements (
330 g lipid kg
1
diet) to achieve maximum growth and feed e⁄ciency
(Higgs & Dong 2000).The use of some alternate lipids
(i.e., alternatives to MFO) such as rendered avian and
mammalian lipids (e.g., poultry fat (PF) and pork lard
singly or in combination with vegetable oils) at this
stage of growth could represent signi¢cant cost sav-
ings to the ¢sh producers and feed manufacturers.
Alternate dietary lipids can, however, in£uence the
fatty acid composition of the ¢sh, and in doing so,
could have repercussions on ¢sh health because of
alterations in the types and levels of eicosanoid com-
pounds that are elaborated from fattyacid precursors
present within cell membranes. In general, eicosa-
noids with immunostimulatory properties are de-
rived from the fatty acid arachidonic acid (AA; 20:4
(n-6)), which is more prevalent when there is a high
dietary intake of n-6 fatty acids. Alternatively, eicosa-
noids with a more suppressive in£uence on the im-
mune system are derived from the n-3 HUFAs, which
are more prevalent in diets containing high concen-
trations of linolenic acid and the n-3 HUFAs (Higgs,
Macdonald, Levings & Dosanjh1995).
Researchers have found that feeding ¢sh diets con-
taining alternate plant and/or animal lipid sources
results in changes in the fatty acid composition of
the ¢sh, which in turn, can lead to di¡erences in ei-
cosanoid production and immune responses (Balfry
& Higgs 2001). In this regard, Bell, Dick, McVicar, Sar-
gent andThompson (1993) reported that Atlantic sal-
mon fed a diet containing dietary sun£ower oil (high
in n-6 fatty acids) produced moreAA-derived eicosa-
noids comparedwith ¢sh fed diets containing linseed
or ¢sh oil (lower in n-6 and higher in n-3 fatty acids
than sun£ower oil). Moreover, relationships have
been found between dietary fatty acid composition
and functioning of the immune system in other stu-
dies on ¢sh (Bell, Raynard & Sargent 1991; Erdal,
Evensen, Kaurstad, Lillehaug, Solbakken & Thorud
1991; Sheldon Jr. & Blazer 1991; Fracalossi & Lovell
1994). However, the underlying functional relation-
ship between di¡erent eicosanoids remains unclear.
Therefore, some researchers such as Ashton, Clem-
ents, Barrow, Secombes and Rowley (1994) have cau-
tioned that it may be the relative ratios of fatty acids
and eicosanoid precursors that are the key to under-
standing the e¡ect of dietary lipids on the immune
system.
This researchwas undertaken to examine the hae-
matological and immunological responses of juvenile
Atlantic salmon fed high-energy extruded diets con-
taining di¡erent concentrations of a 1:1 (w/w) blend
of canola oil (CO) and PF through partial replace-
ment of supplemental anchovy oil (AO) over a 12-
week period. Immediately following the feeding trial,
a random sample of ¢sh from eachdiet treatment was
used to measure baseline levels of the various para-
meters. Duplicate diet groups were subsequently vac-
cinated and reared on the various experimental diets
for anadditional 4weeks. At that time, the remaining
¢shwere sampled similarly and levels of the di¡erent
parameters were measured. Comparisons between
the di¡erent diet treatment groups were made prior
to and following vaccination.
Materials and methods
Experimental design
Quadruplicate groups of 25 juvenile Atlantic salmon
(initial mean mass, 370.2^416.2 g) were held in out-
door 4000 L circular ¢breglass tanks supplied with
aerated, £ow-through, ambient seawater. The water
quality conditions throughout the feeding trial were
as follows: dissolved oxygen, 7.88^10.4mg L
1
; £ow
rate, 35^40 L min
1
; temperature, 8.6^10.9 1C; sali-
nity,26^35 g L
1
. Fishwere hand fed one of three ex-
perimental diets twice daily to satiation for12 weeks.
The diets di¡ered in supplemental lipid (234.7 g kg
1
on an air-dry basis) source, but otherwise were iden-
tical in ingredient composition and contained
equivalent protein (488^493 g kg
1
) and lipid (267^
274 g kg
1
) on a dry weight basis (refer to Higgs, Bal-
fry, Oakes, Rowshandeli, Skura & Deacon 2006 for
additional information). An equal blend (1:1 w/w) of
CO (crude super-degummed product supplied by Ca-
nAmera Foods, Wainright, AB, Canada; presently
Bunge Canada, Oakville, ON, Canada) and PF (West
Coast Reduction,Vancouver, BC, Canada) was tested
as an alternate dietary lipid source toAO.The percent
composition of the supplemental lipid in the three ex-
perimental diets was as follows: diet1,100% AO; diet
2,70.2% AO and 29.8% CO and PF (1:1); diet 3, 40.3%
Aquaculture Research, 2006, 37, 192^199 Alternate dietary lipids and immunity in Atlantic salmon SKBalfryet al.
r 2006 TheAuthors. Journal Compilationr 2005 Blackwell Publishing Ltd, Aquaculture Research, 37, 192^ 199 193