Duan 2013

Journal of Fish Biology (2013) 83, 1183–1196
doi:10.1111/jfb.12213, available online at wileyonlinelibrary.com
Risk-taking behaviour may explain high predation

mortality of GH-transgenic common carp Cyprinus carpio
M. Duan*†, T. Zhang*‡, W. Hu*, S. Xie†, L. F. Sundström§, Z. Li* and

Z. Zhu*
*State Key Laboratory of Freshwater Ecology and Biotechnology, Institute of Hydrobiology,
Chinese Academy of Sciences, Wuhan 430072, P. R. China, †Key Laboratory of Biodiversity
and Conservation of Aquatic Organisms, Institute of Hydrobiology, Chinese Academy of
Sciences, Wuhan 430072, P. R. China and §Department of Animal Ecology, Evolutionary
Biology Centre, Uppsala University, Norbyvägen 18D, SE-752 72 Uppsala, Sweden
(Received 18 February 2013, Accepted 3 July 2013)
The competitive ability and habitat selection of juvenile all-fish GH-transgenic common carp Cypri-
nus carpio and their size-matched non-transgenic conspecifics, in the absence and presence of
predation risk, under different food distributions, were compared. Unequal-competitor ideal-free-
distribution analysis showed that a larger proportion of transgenic C. carpio fed within the system,
although they were not overrepresented at a higher-quantity food source. Moreover, the analysis
showed that transgenic C. carpio maintained a faster growth rate, and were more willing to risk
exposure to a predator when foraging, thereby supporting the hypothesis that predation selects
against maximal growth rates by removing individuals that display increased foraging effort. With-
out compensatory behaviours that could mitigate the effects of predation risk, the escaped or released
transgenic C. carpio with high-gain and high-risk performance would grow well but probably suffer
high predation mortality in nature.
© 2013 The Fisheries Society of the British Isles
Key words: food distribution; foraging; growth hormone; predation risk; transgene.
INTRODUCTION
Numerous varieties of genetically engineered fishes are being developed for aqua-
culture production around the world (Nam et al., 2008). Growth hormone (GH)-
transgenic fishes would not only be adopted for use in commercial aquaculture in
the near future (FDA, 2012), but also provide useful model systems for studying
the consequences of enhancement from genetic, behavioural, physiological, evolu-
tionary and ecological standpoints (Devlin et al., 2006; Nam et al., 2008). Escaped
or released GH-transgenic fishes, however, raise public concerns with regard to the
potential effects on aquatic ecosystems (Kapuscinski et al., 2007; Nam et al., 2008;
Hu & Zhu, 2010).
Growth is of great evolutionary interest (Sibly et al., 1985; Sogard, 1997). In
many fish species, rapid growth should increase fitness as faster growing fishes are
more dominant, stronger competitors, outgrow gape-limited predators faster and can
‡Author to whom correspondence should be addressed. Tel.: +86 27 6878 0369; email: tlzhang@ihb.ac.cn
1183
© 2013 The Fisheries Society of the British Isles
1184 M . D UA N E T A L .
consume a greater range of potential prey (Werner & Gilliam, 1984; Arendt, 1997;
Juanes et al., 2002). They may also have higher reproductive potential (Fleming,
1996) and reach sexual maturity earlier (Alm, 1959). Substantial genetic variation in
growth rates within and among species, however, has been documented, demonstrat-
ing that organisms rarely grow at their physiological maximum and theory suggests
that maximization of growth is unlikely (Arendt, 1997). An increasing amount of
research suggests that faster growth does not always increase fitness, mainly from
various trade-offs that limit the benefit of rapid growth (Arendt, 1997; Mangel &
Stamps, 2001; Morgan & Metcalfe, 2001; Munch & Conover, 2003; Biro et al.,
2005; Devlin et al., 2006; Duan et al., 2010; Dmitriew, 2011). One major trade-
off is between rapid growth and survival, acting through heightened activity during
foraging (Abrahams & Dill, 1989; Lima & Dill, 1990).
Fishes are capable of expressing flexible growth in response to food availabil-
ity (Sebens, 1987; Jobling, 1995). As fast-growing transgenic fishes possess higher
feeding motivation compared to non-transgenic fishes (Devlin et al., 1999; Duan
et al., 2009), they may have different phenotypic and behavioural responses to food
availability. When subjected to low food availability in a microcosm, populations of
coho salmon Oncorhynchus kisutch (Walbaum 1792) experienced population crashes
or extinctions when fast-growing transgenic individuals were present (Devlin et al.,
2004), indicating that food availability could have population-wide consequences.
In rainbow trout Oncorhynchus mykiss (Walbaum 1792), fast-growing juveniles
had to shift the optimal trade-offs between growth and survival to a more ‘high-
gain and high-risk’ phenotype (Johnsson, 1993). This strategy has been interpreted
as boldness, a personality trait, where bold individuals are more risk prone com-
pared with shy individuals (Wilson et al., 1993; Sneddon, 2003; Stamps, 2007).
For instance, variation in personality traits (e.g. boldness) influences risk-taking,
diet and consequently growth and fitness in pumpkinseed Lepomis gibbosus (L.
1758) (Wilson et al., 1993), as well as in the poeciliid Brachyrhaphis episcopi
(Steindachner 1878) (Brown & Braithwaite, 2004; Brown et al., 2007). Similar
results have been found in GH-treated and GH-transgenic salmonids, which were
more willing to take risks that incurred higher predation (Johnsson et al., 1996;
Jönsson et al., 1996; Abrahams & Sutterlin, 1999; Sundström et al., 2004; Tymchuk
et al., 2005).
All-fish GH-transgenic common carp Cyprinus carpio L. 1758 were produced two
decades ago (Zhu et al., 1992). Previous studies have shown that such C. carpio are
more active and aggressive, show higher feeding motivation, achieve higher social
status and elevated individual ability to compete for limited food resources (point and
clumped food) compared with non-transgenic conspecifics (Duan et al., 2009, 2011).
At the same time, they are inferior swimmers (Li et al., 2007, 2009) and suffer higher
predation mortality (Duan et al., 2010) compared with non-transgenics. These effects
are probably mediated by behaviours influencing habitat selection under different
food distributions and the decision to take risks when the reward is food. There have
not been any reports on the competitive ability for food of the transgenic C. carpio
at the population level. Generally, if transgenic C. carpio are to retain their growth
advantage under natural conditions, it is assumed that they would need to be more
effective at competing for food than non-transgenic C. carpio to meet their increased
metabolic requirements (Hu & Zhu, 2010). Ideal free distribution theory can be
used to test the relative competitive abilities of fishes (Grand, 1997; Grand & Dill,
© 2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 1183–1196
R I S K - TA K I N G B E H AV I O U R I N T R A N S G E N I C C Y P R I N U S C A R P I O 1185
1997), which has been used to assess variation in competitive abilities among non-
transgenic and transgenic O. kisutch populations (Tymchuk et al., 2005). In an ideal
free distribution, individuals are distributed so that none would benefit by switching
sites (Fretwell & Lucas, 1970; Fretwell, 1972), and the proportion of individuals at
each site matches the proportion of resources in each site (input matching; Parker,
1974). The ideal-free-distribution theory has been further developed by considering
the situation where not all individuals are equal but instead differ in competitive
ability (Sutherland & Parker, 1985; Parker & Sutherland, 1986).
Here, the competitive ability and habitat selection between the transgenic C. car-
pio and their non-transgenic conspecifics, in the absence and presence of predation
risk, under different food distributions were compared. It is helpful to explain the
behavioural costs and trade-offs that cause fishes to grow below their physiological
maximum, as well as to assess the potential effects on the natural environment of
released or escaped GH-transgenic fishes.
MATERIALS AND METHODS

Experimental animals were maintained at a specially constructed facility at the Fish
Behavioural Ecology Laboratory of the Institute of Hydrobiology (IHB), Chinese Academy
of Sciences (CAS). This facility was designed to minimize the risk of C. carpio escaping by
the use of multiple containment screen systems. All experimental procedures employed were
approved by the Institute of Hydrobiology Institutional Animal Care and Use Committee
(Approval ID: keshuizhuan 08529).
SOURCE OF FISHES
P0 all-fish GH-transgenic Yellow River C. carpio were initially produced by microinjection
of the pcagcgh into the fertilized eggs of C. carpio (Yellow River variety). The all-fish gene
construct pcagcgh was a recombinant construct of grass carp Ctenopharyngodon idella (Valen-
ciennes 1844) GH gene (gcgh), the expression of which is driven by the β-actin gene promoter
of C. carpio (pca) (Wang et al., 2001). F1 C. carpio were produced as follows: P0 transgenic
male (sperm revealed to be pcagcgh positive) × non-transgenic female. The F1 , F2 and F3
generation were, respectively, 1·6 times (Wang et al., 2001), 1·8–2·5 times (W. Hu, unpubl.
data) and 1·4–1·9 times (Li et al., 2007) the body mass of non-transgenic counterparts under
hatchery-reared conditions, showing how the growth enhancement remains relatively stable
across generations. The F4 generation transgenic and non-transgenic C. carpio were produced
on 21 April 2008 from crosses between a wild-type female and an F3 hemizygous transgenic
male of a fast-growing transgenic strain. Frequencies of transgene transmission to F4 progeny
from this line were c. 50% (W. Hu, unpubl. data). Siblings were used to minimize effects of
genetic differences and maternal or paternal effects (Metcalfe et al., 1989). After emergence
(28 April) in Duofu Technology Farm, Wuhan, China, the first-feeding fry containing a mix of
the two genotypes were transferred to an earthen pond (area 667 m2 , water depth 1 m) with a
rearing density of 100 individuals m−3 each. A month later, 2000 individuals of the mixed pop-
ulations were transferred to an indoor recirculating system and reared in two circular fibreglass
tanks (diameter 150 cm, volume 1000 l) at the IHB, CAS. Thereafter, the pcagcgh transgene-
positive C. carpio were identified by PCR following Guan et al. (2010). The two genotypes
were then reared separately in the two fibreglass tanks as described above. Non-transgenics
were fed to satiation with frozen chironomid Chironomus sp. larvae twice daily, whereas the
transgenics were fed roughly the same amount every other day to allow size-matching and
avoid the possible effects of size on competitive ability (Huntingford et al., 1990). This feed
regime was maintained until the C. carpio gape sizes matched the size of the artificial food
pellets (mean ± s.e. mass 5·52 ± 0·16 mg, length 2·89 ± 0·05 mm, diameter 1·54 ± 0·02 mm;
comprising 36·0% protein, 12·1% lipid, 11·3% ash and 18·2 J mg−1 energy). In a feeding
1186 M . D UA N E T A L .
trial, Duan et al. (2011) showed that the transgenic C. carpio used in the present experiments
consumed on average 58% more pellets than did non-transgenics (mean ± s.e. 25·3 ± 1·3 v .
16·0 ± 2·6). The water temperature varied from 25·0 to 27·4◦ C during this period.
The mandarin fish Siniperca chuatsi (Basilewsky 1855) was chosen as the predator. It is
a demersal large-sized piscivorous predator, widely distributed in many rivers and lakes of
China. Siniperca chuatsi consume only live larvae of fishes and shrimp when first-feeding
and refuse to consume dead prey or artificial diets during their entire life-history stages (Liang
et al., 1998). Age 1 year S. chuatsi in Liangzi Lake were captured with trap nets and three
individuals were selected for use in this study (mean ± s.e. total length, LT , 26·3 ± 1·1 cm,
mass 297·4 ± 25·7 g). Prior to the experiment, the predators were given at least 1 week to
acclimate to confined concrete-tank conditions.
E X P E R I M E N TA L P R O C E D U R E S
To test the responses of transgenic and non-transgenic C. carpio to food resources and pre-
dation risk, the experiment was conducted in four tempered glass aquaria (175 × 35 × 35 cm
in size; Fig. 1) with a water depth of 20 cm. Eight fluorescent lights (8 W) were fixed 15 cm
above the water level on each aquarium (a 12L:12D on at 0700 hours and off at 1900 hours).
All aquaria had their four walls covered with an opaque, black plastic screen to minimize
disturbance and maintain similar levels of illumination. Each aquarium was divided into five
sections: two non-predator or predator sections in the two ends (50 cm length each), two safe
or risky sections (25 cm length each) near the two ends and one buffer section (25 cm length
each) in the middle of the aquarium (Fig. 1). There were several holes (2 cm diameter) in
the two screens between non-predator or predator and safe or risky sections through which
fishes could both smell and see each other, but neither the prey nor the predator were able
to go through the holes. In addition, each aquarium had a white gravel substratum and two
white-bottomed Petri dishes (15 cm) located at the centre of each of the two feeding sections
considered as feeding areas (Fig. 1). Two plastic funnels from the outside of one of the lateral
walls were fixed above each feeding area, 5 cm above the water surface, allowing food pellets
to be administered to the feeding area each time, without any disturbance to the fishes (Fig. 1).
The experiment consisted of two main parts: (1) a treatment period (days 1–5) and
(2) an observation period (days 6–7) and was carried out in an indoor recirculating
system. Incoming non-chlorinated water was automatically filtered, sterilized and constantly
aerated to ensure high water quality. Throughout the experiment, the water temperature,
dissolved oxygen and pH were maintained at 26·4 ± 0·3◦ C, 7·1 ± 0·1 mg l−1 and 7·6 ± 0·1
(mean ± s.e.), respectively.
Treatment period
On day 1 (20 August) at 0800 hours, four size-matched groups (eight C. carpio in each
group), with four individuals of each genotype (size-matched as two and two; see Tables I
and II), were anaesthetized with 0·25 ml l−1 eugenol (purity ≥ 99·99%, Sinopharm Chemical
Reagent Co. Ltd; http://en.reagent.com.cn/). Body mass (M , to the nearest 0·01 g) and LT (to
the nearest 0·01 cm) were measured. Individual C. carpio from each genotype were identified
visually by tagging with different coloured beads (mean ± s.e. diameter 2·3 ± 0·2 mm, mass
14·3 ± 0·3 mg) on nylon thread (diameter 0·25 mm) inserted through the dorsal musculature,
anterior to the dorsal fin and then tied in a surgical knot (Duan et al., 2011). Thereafter, the
eight C. carpio were transferred to the buffer section of each aquarium to acclimate overnight.
The following day (day 2) at 0800 hours, two screens of each buffer section were gently
removed and after 3 min the C. carpio were fed pellets as described above. In each aquarium,
one feeder provided 100 food pellets (total of 0·6 g), one pellet every 5 s during a 20 min
feeding session, while the other feeder provided twice that amount by providing two pellets
each time (200 food pellets, 1·2 g). Thereafter, the C. carpio were gently pushed back into
the buffer sections and the screens were put back in place. Uneaten food items and faeces
were removed by a glass siphon. This feeding regime was repeated twice daily (0800–1100
and 1400–1700 hours), for a period of 3 days (days 2–4). In the late afternoon of day 5, a
S. chuatsi was introduced into the predator sections next to the high-food abundance habitat
(a) A1 A2
C2 C1 C1 C2
B1 B2
(b) A1 A2
C2 C1 C1 C2
B1 B2
Fig. 1. Schematic representation of the apparatus used for the experiment. A1 and A2 are the feeding funnels
(food sources); food abundance ratio in B1 :B2 = 1:2 (0·6 g, 100 pellets:1·2 g, 200 pellets); C1 is the
non-porous transparent glass screen and C2 is the porous transparent glass screen (mesh size, 2 cm);
(a) in the absence of a predator, which could test the feeding competition of the two genotypes and (b)
in the presence of a predator, smell and visual contacts between the prey Cyprinus carpio and predator
exist, but none of the C. carpio could pass through the porous screens (C2 ), which could test whether
the transgenic C. carpio would be more risk-taking when foraging.
in any two of the four systems. The prey C. carpio were deprived of food for 1 day (day 5)
before the observations commenced on day 6.
Observations
The same feeding procedures described above were repeated during the two-day
behavioural observations. Behaviours of the prey C. carpio were recorded by a digital
videocassette recorder (Sony, HDR-HC1E Handy Camera; www.sony.net) placed 1 m above
the aquarium.
The order of observations was randomly allocated among the four aquaria so as to remove
any bias and order effects among tanks. For each aquarium, feeding and video recording
started 10 min after the buffer screens were removed. Observations were made twice daily, at
0800–1100 and 1400–1700 hours (i.e. two trials for each group) during the feeding periods.
Upon completion of the second trial on the first day, the location of the predator and the
food supply was reversed, and another two trials were conducted the following day. On the
eighth day morning (27 August), LT and M of the 32 C. carpio were again measured under
a light anaesthesia with 0·25 ml l−1 eugenol (purity ≥ 99·99%). Thereafter, another four size-
matched groups and two predators were randomly selected and the procedures repeated as
described above from 28 August to 4 September and from 4 to 11 September. Consequently,
for each treatment, two size-matched groups were filmed for each trial set, which yielded
six repeated trials for each of the two treatments during the two-day observations, and six
different size-matched groups with the two genotypes were used in total (Tables I and II).
Spatial analysis of the two genotypes was completed from the video. For each trial, location
of each individual in the aquarium was captured every 30 s during the 20 min, which was
used to determine the spatial distribution of competitors at the risky high and safe low-food
1188 M . D UA N E T A L .
Table I. Total length (LT ), body mass (M ), condition factor (K ) and tag colour (TC ) of
transgenic and non-transgenic Cyprinus carpio used in the absence of a predator (mean ± s.e.)
Group Genotype LT (cm) M (g) K (g cm−3 ) TC

1 Transgenic 6·62 ± 0·98 3·40 ± 0·18 1·17 ± 0·04 Blue
1 Non-transgenic 6·42 ± 0·13 3·35 ± 0·27 1·26 ± 0·01 Yellow
2 Transgenic 6·49 ± 0·14 3·17 ± 0·13 1·16 ± 0·03 Blue
3 Transgenic 6·61 ± 0·13 3·32 ± 0·19 1·15 ± 0·01 Yellow
3 Non-transgenic 6·60 ± 0·11 3·27 ± 0·18 1·35 ± 0·03 Red
5 Non-transgenic 6·23 ± 0·13 2·99 ± 0·12 1·23 ± 0·04 Blue
sources. Only C. carpio that were feeding at the safe or risky sections were counted in the
spatial distribution. Individuals that were feeding could be identified by their movement,
which included fast darts across the tank to capture a food item or a stationary position
under the food supply with short darts to capture the food. Non-feeding C. carpio generally
maintained a stationary position in the tank unless chased by another individual.
D ATA A N A LY S I S
The specific growth rate was calculated as GM = 100 (ln M 2 − ln M 1 ) t −1 (Ricker, 1979),
where M 2 and M 1 were the final and initial M (g) and t is the experimental period (7 days,
consisting of the 5 day treatment period and the 2 day observations). The condition factor (K )
was calculated as K = 100 M LT −3 .
To test the effects of genotype and predation risk on proportion of individuals feeding,
habitat choice, and the specific growth rate of the transgenic and non-transgenic C. carpio,
Table II. Total length (LT ), body mass (M ), condition factor (K ) and tag colour (TC ) of
transgenic and non-transgenic Cyprinus carpio used in the presence of a predator (mean ± s.e.)
Group Genotype LT (cm) M (g) K (g cm−3 ) TC

1 Transgenic 6·54 ± 0·14 3·26 ± 0·26 1·16 ± 0·02 Blue
2 Transgenic 6·49 ± 0·11 3·22 ± 0·12 1·18 ± 0·02 Blue
two-way ANCOVA were used with M 1 as a covariate. In the absence of a predator, the
difference between the proportion of C. carpio feeding at the high-food source that were
transgenic and the proportion of transgenic C. carpio in the entire feeding population was
obtained for each group, and then compared to an expected difference of zero with a one-
sample t-test. If the transgene had no effect on feeding behaviour, it was expected that
the difference between the proportion of C. carpio feeding at the high-food source that
were transgenic and proportion of transgenic C. carpio in the feeding population would be
zero (Sutherland & Parker, 1985; Parker & Sutherland, 1986; Tymchuk et al., 2005). The
differences between the proportion of transgenic and non-transgenic C. carpio feeding were
tested by one-way ANCOVA with M 1 as a covariate.
All the data were normally distributed and had homogeneous variance. Differences
were regarded as significant when P < 0·05. All the data were analysed with SPSS 15.0
(http://www-01.ibm.com/software/analytics/spss/) and described as mean ± s.e.
RESULTS
P O P U L AT I O N C O M P E T I T I V E A B I L I T Y A N D P R E D AT I O N
AV O I D A N C E
Overall, there was a significant difference between the proportion of the two
genotypes that was feeding (ANCOVA, F 1,19 = 5·561, P < 0·05), however, a smaller
proportion of both genotypes fed in the presence of the predator (F 1,19 = 11·843,
P < 0·01; Fig. 2). Initial M had no effect on the proportional feeding of the two
genotypes (F 1,19 = 2·021, P > 0·05), nor was the interaction between genotype
and predation risk significant (F 1,19 = 0·009, P > 0·05). In the absence of the
predator, the per cent feeding in the transgenic C. carpio averaged 79·0 ± 5·7%,
while the per cent feeding in the non-transgenic C. carpio was 68·7 ± 7·9%, and
these values were not significantly different (ANCOVA, F 1,9 = 0·119, P > 0·05).
In the presence of the predator, the per cent feeding in the transgenic C. carpio
averaged 59·0 ± 6·2%, which was higher than that in the non-transgenic C. carpio
(42·2 ± 6·6%, F 1,9 = 7·545, P < 0·05; Fig. 2).
In the absence of the predator, there was 79·9 ± 2·8% C. carpio feeding at the
high-food supply habitat [Fig. 3(a)]. If the GH transgene had no effect on feeding
behaviour that is dependent on food supply level, it would be expected that the
difference between the proportion of C. carpio feeding at the high-food source that
were transgenic and the proportion of transgenic C. carpio in the entire feeding
population would be zero. The per cent of C. carpio feeding at the high-food source
that were transgenic (53·7 ± 3·5%) was not significantly higher than the per cent of
transgenic C. carpio in the feeding population (54·3 ± 4·4%, t-test, t 5 = −1·178,
P > 0·05), indicating no strong influence of the transgene on access to the higher-
quantity food source [Fig. 3(a)].
In the presence of the predator, fewer C. carpio of both genotypes fed compared to
when the predator was absent [F 1,19 = 119·004, P < 0·001; Fig. 3(b)]. The number
of transgenic C. carpio feeding either at low-food and no-risk or high-food and
high-risk habitat was much higher than that of the non-transgenic C. carpio feeding
in the same habitat [ANCOVA, both of P < 0·05; Fig.3(b)]. In the high-food and
high-risk habitat, the per cent feeding in the transgenic C. carpio (5·4 ± 0·7%) was
much higher than that in the non-transgenic individuals [1·1 ± 0·4%; Fig. 3(b)],
indicating that transgenic C. carpio were less risk-sensitive at the high-food resource.
1190 M . D UA N E T A L .
1·0
*
0·8
Proportion feeding
*
0·6
0·4
0·2
0
No predation Predation
Fig. 2. Proportion of transgenic (n = 4) ( ) and non-transgenic (n = 4) ( ) Cyprinus carpio feeding in the

absence or presence of a predator during the experiment. *, a significant difference between the two
genotypes or between the treatments (P < 0·05). The whisker on each bar indicates s.e.
Initial body M had no significant effects on the habitat choice of the two genotypes
(F 1,19 = 0·148, P > 0·05), nor was interaction between genotype and predation risk
significant (F 1,19 = 1·778, P > 0·05), indicating that transgenic C. carpio were not
more likely to spend time in the risky zone.
G ROW T H
At the beginning of the experiment, there was no significant difference in LT , M
and K between the two genotypes in six different size-matched groups (ANOVA, all
P > 0·05; Tables I and II).
Genotype had significant effects on the growth of the experimental C. carpio
(ANCOVA, F 1,19 = 46·122, P < 0·001; Fig. 4). During the seven-day experiment,
the G W of transgenic C. carpio (3·24 ± 0·09% day−1 ) was 1·62 times that of the
non-transgenic C. carpio (2·00 ± 0·18% day−1 ) in the absence of the predator. In
the presence of the predator, the GM of transgenic C. carpio (3·27 ± 0·15% day−1 )
was 3·41 times higher than that of non-transgenic C. carpio (0·96 ± 0·45% day−1 ).
Initial M and predation risk had no effects on GM of the two genotypes, nor was the
interaction between genotype and predation risk significant (all P > 0·05; Fig. 4).
DISCUSSION
In this study, transgenic C. carpio were competitively equal to their non-transgenic
conspecifics as there was no difference in the proportion of transgenic C. carpio
feeding at the high-food source compared to the proportion of transgenic C. carpio
within the group. Furthermore, there was no difference between the proportion of
the two genotypes that fed at the lower or higher quantity food resources without
predation. The results were similar to the previous study on O. kisutch at the fry stage
in that they were not overrepresented at a higher-quantity food resource and thus
3·0 (a) *
2·5
2·0
1·5
1·0
0·5
Number feeding
3·0 (b)
*
2·5 *
2·0
1·5
1·0
0·5
*
0
Low food High food
Fig. 3. Mean + s.e. number of transgenic (n = 4) ( ) and non-transgenic (n = 4) ( ) Cyprinus carpio feeding at
both the low and high-food abundance location (a) in the absence and (b) in the presence of a predator.
*, a significant difference between the two genotypes or between the two feeding areas in the same
treatment (P < 0·05).
exhibited equal competitive ability to controls (Tymchuk et al., 2005). Moreover, in

stream microcosms, first-feeding GH-transgenic and non-transgenic Atlantic salmon
Salmo salar L. 1758 fry were equally likely to be dominant (Moreau et al., 2011).
In contrast, juvenile GH-transgenic C. carpio and GH-transgenic O. kisutch both
display elevated ability to compete for pointed and clumped food resources (Devlin
et al., 1999; Duan et al., 2009, 2011). These previous results were obtained under
limited food supply in single paired tests (one transgenic v . one non-transgenic fish),
which did not consider variable food distribution and population effects. In this
study, conditions were more complex with four transgenic and four non-transgenic
C. carpio in each group, which probably mimicked the foraging-associated
behaviours such as schooling behaviour (Huntingford et al., 2010) and visual inter-
ference (Grosenick et al., 2007) seen in nature. Besides, in this feeding system, food
resources in the two habitats were supplied simultaneously over a 20 min time period,
which formed a temporal clumping of resources that may reduce dominant defensibil-
ity. According to resource defence theory ‘synchrony hypothesis’ (Grant & Kramer,
1992), when a large number of prey items are encountered simultaneously, less
1192 M . D UA N E T A L .
* *
3
GM (% day–1)
2
0
No predation Predation
Fig. 4. Mean + s.e. specific growth rates of body mass (GM ) for the transgenic ( ) and non-transgenic ( )
Cyprinus carpio in the absence and in the presence of a predator. *, a significant difference between the
two genotypes (P < 0·05).
competitive ability group members are able to exploit the remaining resources for the
period of time the dominant individual is incapacitated whilst handling prey, and in
this way benefit from group membership more than when resources are encountered
sequentially and are therefore more defensible. In previous work, GH-transgenic
C. carpio were better at using low-quality food and had higher feed-conversion effi-
ciency (Fu et al., 1998), which could partly explain why transgenic C. carpio grew
much faster than their non-transgenic equal competitors during the study period.
Transgenic C. carpio were significantly more willing to risk exposure to a predator
when foraging for food, and realized their rapid growth potential as a consequence in
this study. This may provide direct behavioural evidence for a trade-off between rapid
growth and high predation mortality in transgenic C. carpio, mediated by their anti-
predator behaviour when foraging, indicating that the genetic capacity for growth
evolves towards an optimum rather than an absolute maximum. Previous studies
on the same fast-growing transgenic strain showed that they increased movements
(Duan et al., 2009, 2011), reduced swimming speed (Li et al., 2009) and pos-
sessed lower anti-predator ability (higher predation mortality) when directly facing
predators (Duan et al., 2010). Similar results were found for fast-growing domestic
O. mykiss that were more willing to forage in productive and risky habitats, which
resulted in higher predation mortality compared to wild individuals, suggesting a
trade-off between growth rate and survival such that rapid growth entailed a cost in
terms of mortality (Biro et al., 2006). One possible explanation was that O. mykiss
had to be more active in order to gain access to more food to meet their higher
metabolic requirement which increased risk of predation. Indeed, although preda-
tion risk resulted in decreased feeding movements in the two genotypes, there were
more transgenic C. carpio feeding than non-transgenics at the high-food and high-risk
habitat, which was consistent with theoretical expectations about how animals should
integrate the risk of predation into their foraging decisions (Werner & Gilliam, 1984).
The results, however, parallel studies on GH-treated and GH-transgenic salmonids
that were bolder towards a predator when foraging for food (Johnsson et al., 1996;
Jönsson et al., 1996; Abrahams & Sutterlin, 1999), indicating that levels of GH
represent a balance between the positive effects of growth and the increased mortal-
ity associated with changes in anti-predator behaviour. Based on a field study, Biro
et al. (2006) suggested that one of the personalities, boldness, may be the major
determinant of predation mortality in O. mykiss, and therefore the major mechanism
for selection against maximum growth rates. In this study, however, the two geno-
types did not have direct contact with the predator, indicating that these individuals
should be less willing to forfeit profitable activities in order to reduce the probability
of death from predation. As a consequence, the GH-transgenic C. carpio should be
less sensitive to the indirect effects of predator intimidation.
It seems that escaped or released transgenic C. carpio with high-gain and high-
risk performance would grow well but suffer high predation mortality in nature.
It is worth noting that hungry animals are more risk-prone and willing to pay the
costs of predation with greater motivation to take the benefits (e.g. food) of risk-
taking behaviour (Lima & Dill, 1990). In this study, transgenic C. carpio were fed
with a restricted ration before the acclimation period (i.e. feeding every other day),
which may also be a factor that could affect the transgene effects on the risk-taking
behaviour due to hunger. Under natural conditions, however, food availability would
be less predictable and predation risk may influence feeding decisions so that the
advantage of transgenic fishes could be lessened or increased depending on their
response under such complex environments. Several recent field studies suggest that
the relation between boldness, growth and mortality can be quite variable in natural
populations (Adriaenssens & Johnsson, 2013). The effects found on domesticated O.
mykiss may be specifically influenced by a history of directional selection for growth
in captivity, dragging boldness and aggression along, whereas natural selection is
often stabilizing with fluctuating selection pressures. Arendt (1997) argued that there
were many functions that may be compromised when intrinsic growth rates were
accelerated to meet evolutionary challenges. Conversely, if selection acts to improve
any of these functions, intrinsic growth rates might be reduced.
Overall, the results clearly showed that the juvenile all-fish GH-transgenic
C. carpio maintained a faster growth rate, and were more willing to risk exposure
to a predator when foraging, as well as were competitively equal to their non-
transgenic conspecifics. Short-term, small-scale experiments, however, are likely
to overestimate the importance of behavioural variation (Sutherland, 1996; Lima,
1998; Schmitz, 2001). Because these experiments did not have sufficient realism
to allow results to be extrapolated to natural situations (Sutherland, 1996; Lima,
1998), or allow expression of compensatory behaviours that could mitigate effects
of predation risk (Lind & Cresswell, 2005), the present results could not make
convincing connections between behaviour, growth and survival at a population
or community level in nature. Further work on the behavioural and developmental
alterations of transgenic fishes in complex environments is urgently needed.
The study was financially supported by the National Natural Science Foundation of China
(Grant Nos 31200423 and 30970553) and the Development Plan of the State Key Fundamental
Research of China (Grant No. 2009CB119205).
References
Abrahams, M. V. & Dill, L. M. (1989). A determination of the energetic equivalence of the
risk of predation. Ecology 70, 999–1007.
1194 M . D UA N E T A L .
Abrahams, M. V. & Sutterlin, A. (1999). The foraging and antipredator behaviour of growth-
enhanced transgenic Atlantic salmon. Animal Behaviour 58, 933–942.
Adriaenssens, B. & Johnsson, J. I. (2013). Natural selection, plasticity and the emergence of
a behavioural syndrome in the wild. Ecology Letters 16, 47–55.
Alm, G. (1959). Connection between maturity, size and age in fishes. Report of the Institute
of Freshwater Research Drottningholm 40, 4–145.
Arendt, J. D. (1997). Adaptive intrinsic growth rates: an integration across taxa. Quarterly
Review of Biology 72, 149–177.
Biro, P. A., Post, J. R. & Abrahams, M. V. (2005). Ontogeny of energy allocation reveals
selective pressure promoting risk-taking behaviour in young fish cohorts. Proceedings
of the Royal Society B 272, 1443–1448.
Biro, P. A., Abrahams, M. V., Post, J. R. & Parkinson, E. A. (2006). Behavioural trade-offs
between growth and mortality explain evolution of submaximal growth rates. Journal
of Ecology 75, 1165–1171.
Brown, C. & Braithwaite, V. A. (2004). Size matters: a test of boldness in eight populations
of the poeciliid Brachyrhaphis episcopi . Animal Behaviour 68, 1325–1329.
Brown, C., Jones, F. & Braithwaite, V. A. (2007). Correlation between boldness and body
mass in natural populations of the poeciliid Brachyrhaphis episcopi . Journal of Fish
Biology 71, 1590–1601.
Devlin, R. H., Johnsson, J. I., Smailus, D. E., Biagi, C. A., Jönsson, E. & Björnsson, B.
T. (1999). Increased ability to compete for food by growth hormone-transgenic coho
salmon Oncorhynchus kisutch (Walbaum). Aquaculture Research 30, 479–482.
Devlin, R. H., D’Andrade, M., Uh, M. & Biagi, C. A. (2004). Population effects of growth
hormone transgenic coho salmon depend on food availability and genotype by envi-
ronment interactions. Proceedings of the National Academy of Sciences of the United
States of America 101, 9303–9308.
Devlin, R. H., Sundström, L. F. & Muir, W. M. (2006). Interface of biotechnology and
ecology for environmental risk assessments of transgenic fish. Trends in Biotechnology
24, 89–97.
Dmitriew, C. M. (2011). The evolution of growth trajectories: what limits growth rate? Bio-
logical Reviews 86, 97–116.
Duan, M., Zhang, T., Hu, W., Sundström, L. F., Wang, Y., Li, Z. & Zhu, Z. (2009). Elevated
ability to compete for limited food resources by ‘all-fish’ growth hormone transgenic
common carp Cyprinus carpio. Journal of Fish Biology 75, 1459–1472.
Duan, M., Zhang, T., Hu, W., Guan, B., Wang, Y., Li, Z. & Zhu, Z. (2010). Increased mortality
of growth-enhanced transgenic common carp (Cyprinus carpio L.) under short-term
predation risk. Journal of Applied Ichthyology 26, 908–912.
Duan, M., Zhang, T., Hu, W., Li, Z., Sundström, L. F., Zhu, T., Zhong, C. & Zhu, Z.
(2011). Behavioral alterations in GH transgenic common carp may explain enhanced
competitive feeding ability. Aquaculture 317, 175–181.
Fleming, I. A. (1996). Reproductive strategies of Atlantic salmon: ecology and evolution.
Reviews in Fish Biology and Fisheries 6, 379–416.
Fretwell, S. D. (1972). Populations in a Seasonal Environment. Princeton, NJ: Princeton
University Press.
Fretwell, S. D. & Lucas, H. L. Jr. (1970). On territorial behavior and other factors influencing
habitat distribution in birds. I. Theoretical development. Acta Biotheoretica 19, 16–36.
Fu, C., Cui, Y., Hung, S. S. O. & Zhu, Z. (1998). Growth and feed utilization by F4 human
growth hormone transgenic carp fed diets with different protein levels. Journal of Fish
Biology 53, 115–129.
Grand, T. C. (1997). Foraging site selection by juvenile coho salmon: ideal free distributions
of unequal competitors. Animal Behaviour 53, 185–196.
Grand, T. C. & Dill, L. M. (1997). The energetic equivalence of cover to juvenile coho salmon
(Oncorhynchus kisutch): ideal free distribution theory applied. Behavioral Ecology 8,
437–447.
Grant, J. W. A. & Kramer, D. L. (1992). Temporal clumping of food arrival reduces its
monopolization and defence by zebrafish, Brachydanio rerio. Animal Behaviour 44,
101–110.
Grosenick, L., Clement, T. S. & Fernald, R. D. (2007). Fish can infer social rank by obser-
vation alone. Nature 445, 429–432.
Guan, B., Hu, W., Zhang, T., Duan, M., Li, D., Wang, Y. & Zhu, Z. (2010). Acute and
chronic un-ionized ammonia toxicity to ‘all-fish’ growth hormone transgenic common
carp (Cyprinus carpio L.). Chinese Science Bulletin 55, 4032–4036.
Hu, W. & Zhu, Z. (2010). Integration mechanisms of transgenes and population fitness of
GH transgenic fish. Science China Life Sciences 53, 401–408.
Huntingford, F. A., Metcalfe, N. B., Thorpe, J. E., Graham, W. D. & Adams, C. E. (1990).
Social dominance and body size in Atlantic salmon parr, Salmo salar L. Journal of
Fish Biology 36, 877–881.
Huntingford, F. A., Andrew, G., Machenzie, S., Morera, D., Coyle, S. M., Pilarczyk, M.
& Kadri, S. (2010). Coping strategies in a strongly schooling fish, the common carp
Cyprinus carpio. Journal of Fish Biology 76, 1576–1591.
Jobling, M. (1995). Environmental Biology of Fishes. London: Chapman & Hall.
Johnsson, J. I. (1993). Big and brave: size selection affects foraging under risk of predation
in juvenile rainbow trout, Oncorhynchus mykiss. Animal Behaviour 45, 1219–1225.
Johnsson, J. I., Petersson, E., Jönsson, E., Björnsson, B. T. & Järvi, T. (1996). Domestica-
tion and growth hormone alter antipredator behaviour and growth patterns in juvenile
brown trout, Salmo trutta. Canadian Journal of Fisheries and Aquatic Sciences 53,
1546–1554.
Jönsson, E., Johnsson, J. I. & Björnsson, B. T. (1996). Growth hormone increases predation
exposure of rainbow trout. Proceedings of the Royal Society B 263, 647–651.
Juanes, F., Buckel, J. A. & Scharf, F. S. (2002). Feeding ecology of piscivorous fishes. In
Handbook of Fish Biology and Fisheries, Vol. I (Hart, P. J. B. & Reynolds, J. D., eds),
pp. 267–283. Oxford: Blackwell Science.
Kapuscinski, A. R., Hayes, K. R., Li, S. & Dana, G. (2007). Environmental Risk Assessment
of Genetically Modified Organisms, Vol. III. Wallingford: CABI Publishers.
Li, D., Fu, C., Hu, W., Wang, Y. & Zhu, Z. (2007). Rapid growth cost in ‘all-fish’ growth hor-
mone gene transgenic carp: reduced critical swimming speed. Chinese Science Bulletin
52, 1501–1506.
Li, D., Hu, W., Wang, Y., Zhu, Z. & Fu, C. (2009). Reduced swimming abilities in fast-
growing transgenic common carp Cyprinus carpio associated with their morphological
variations. Journal of Fish Biology 74, 186–197.
Liang, X., Liu, J. & Huang, B. (1998). The role of sense organs in the feeding behaviour of
Chinese perch. Journal of Fish Biology 52, 1058–1067.
Lima, S. L. (1998). Stress and decision making under the risk of predation: recent devel-
opments from behavioral, reproductive, and ecological perspectives. Advances in the
Study of Behavior 27, 215–290.
Lima, S. L. & Dill, L. M. (1990). Behavioral decisions made under the risk of predation: a
review and prospectus. Canadian Journal of Zoology 68, 619–640.
Lind, J. & Cresswell, W. (2005). Determining the fitness consequences of antipredation behav-
ior. Behavioral Ecology 16, 945–956.
Mangel, M. & Stamps, J. (2001). Trade-offs between growth and mortality and the mainte-
nance of individual variation in growth. Evolutionary Ecology Research 3, 583–593.
Metcalfe, N. B., Huntingford, F. A., Graham, W. D. & Thorpe, J. E. (1989). Early social
status and the development of life-history strategies in Atlantic salmon. Proceedings
of the Royal Society B 236, 7–19.
Moreau, D. T. R., Fleming, I. A., Fletcher, G. L. & Brown, J. A. (2011). Growth hormone
transgenesis does not influence territorial dominance or growth and survival of first-
feeding Atlantic salmon Salmo salar in food-limited stream microcosms. Journal of
Fish Biology 78, 726–740.
Morgan, I. J. & Metcalfe, N. B. (2001). Deferred costs of compensatory growth after autumnal
food shortage in juvenile salmon. Proceedings of the Royal Society B 268, 295–301.
Munch, S. B. & Conover, D. O. (2003). Rapid growth results in increased susceptibility to
predation in Menidia menidia. Evolution 57, 2119–2127.
Nam, Y. K., Maclean, N., Hwang, G. & Kim, D. S. (2008). Autotransgenic and allotransgenic
manipulation of growth traits in fish for aquaculture: a review. Journal of Fish Biology
72, 1–26.
1196 M . D UA N E T A L .
Parker, G. A. (1974). The reproductive behaviour and the nature of sexual selection in
Scatophaga stercoraria L. (Diptera: Scatophagidae). IX. Spatial distribution of fer-
tilisation rates and evolution of male search strategy within the reproductive area.
Evolution 28, 93–108.
Parker, G. A. & Sutherland, W. J. (1986). Ideal free distributions when individuals dif-
fer in competitive ability: phenotype-limited ideal free model. Animal Behaviour 34,
1222–1242.
Ricker, W. E. (1979). Growth rates and models. In Fish Physiology, Vol. VIII (Hoar, W. S.,
Randall, D. J. & Brett, J. R., eds), pp. 677–743. New York, NY: Academic Press.
Schmitz, O. J. (2001). From interesting details to dynamical relevance: toward more effective
use of empirical insights in theory construction. Oikos 94, 39–50.
Sebens, K. (1987). The ecology of indeterminate growth in animals. Annual Review of Ecology
and Systematics 18, 371–407.
Sibly, R., Calow, P. & Nichols, N. (1985). Are patterns of growth adaptive? Journal of
Theoretical Biology 112, 553–574.
Sneddon, L. U. (2003). The bold and the shy: individual differences in rainbow trout
(Oncorhynchus mykiss). Journal of Fish Biology 62, 971–975.
Sogard, S. M. (1997). Size-selective mortality in the juvenile stage of teleost fishes: a review.
Bulletin of Marine Science 60, 1129–1157.
Stamps, J. A. (2007). Growth-mortality tradeoffs and ‘personality traits’ in animals. Ecology
Letters 10, 355–363.
Sundström, L. F., Lõhmus, M., Johnsson, J. I. & Devlin, R. H. (2004). Growth hormone trans-
genic salmon pay for growth potential with increased predation mortality. Proceedings
of the Royal Society B 271, S350–S352.
Sutherland, W. J. (1996). From Individual Behaviour to Population Ecology. Oxford: Oxford
University Press.
Sutherland, W. J. & Parker, G. A. (1985). Distribution of unequal competitors. In Behavioural
Ecology: Ecological Consequences of Adaptive Behaviour (Sibly, R. M. & Smith, R.
H., eds), pp. 255–275. Oxford: Blackwell Scientific Publications.
Tymchuk, W. E., Abrahams, M. V. & Devlin, R. H. (2005). Competitive ability and mortality
of growth-enhanced transgenic coho salmon fry and parr when foraging for food.
Transactions of the American Fisheries Society 134, 381–389.
Wang, Y., Hu, W., Wu, G., Sun, Y., Chen, S., Zhang, F., Zhu, Z., Feng, J. & Zhang, X.
(2001). Genetic analysis of ‘all-fish’ growth hormone gene transferred carp (Cyprinus
carpio L.) and its F1 generation. Chinese Science Bulletin 46, 1174–1177.
Werner, E. E. & Gilliam, J. F. (1984). The ontogenetic niche and species interactions in
size-structured populations. Annual Review of Ecology and Systematics 15, 393–425.
Wilson, D. S., Coleman, K., Clark, A. B. & Biederman, L. (1993). Shy-bold continuum in
pumpkinseed sunfish (Lepomis gibbosus): an ecological study of a psychological trait.
Journal of Comparative Psychology 107, 205–260.
Zhu, Z., He, L. & Chen, T. T. (1992). Primary-structural and evolutionary analyses of
growth-hormone gene from grass carp (Ctenopharyngodon idellus). European Journal
of Biochemistry 207, 643–648.
Electronic Reference
FDA (2012). Veterinary Medicine Advisory Committee; notice of meeting. Federal Reg-
ister 77, 247. Available at http://www.gpo.gov/fdsys/pkg/FR-2012-12-26/html/2012-
31118.htm/ (last accessed 26 December 2012).

Duan 2013

Uploaded by

Copyright:

Available Formats

Duan 2013

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Duan 2013

Uploaded by

Copyright:

Available Formats

Journal of Fish Biology (2013) 83, 1183–1196

doi:10.1111/jfb.12213, available online at wileyonlinelibrary.com

Risk-taking behaviour may explain high predation

M. Duan†, T. Zhang‡, W. Hu, S. Xie†, L. F. Sundström§, Z. Li and

(Received 18 February 2013, Accepted 3 July 2013)

MATERIALS AND METHODS

Group Genotype LT (cm) M (g) K (g cm−3 ) TC

Group Genotype LT (cm) M (g) K (g cm−3 ) TC

Fig. 2. Proportion of transgenic (n = 4) ( ) and non-transgenic (n = 4) ( ) Cyprinus carpio feeding in the

exhibited equal competitive ability to controls (Tymchuk et al., 2005). Moreover, in

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.