Keywords: sperm cryopreservation, straw vo-lume, cryoprotectants, thawing temperature, sperm quality, Piaractus brachypomus Introduction The use of high-quality gametes in aquaculture is
Trang 2Introduction to the special issue on ‘‘Development of native species for aquaculture in Latin America II’’
Gustavo M Somoza1& Lindsay G Ross2
1
IIB-INTECH (CONICET-UNSAM), Chascomu¤s, Argentina
2 Institute of Aquaculture, University of Stirling, Stirling, Scotland, UK
Correspondence: L G Ross, Institute of Aquaculture, University of Stirling, Stirling, FK9 4LA Scotland, UK E-mail: lgr1@stir.ac.uk; G M Somoza, IIB-INTECH (CONICET-UNSAM), Chascomu¤s, Argentina E-mail: somoza@intech.gov.ar
Following the ¢rst special edition of Aquaculture
Research on Development of native species for
aqua-culture in Latin America II, in which selected
presen-tations from the First Latin American Conference on
Culture of Native Fishes were published in 2008, and
in view of the continued growth of research and
de-velopment of aquaculture of native ¢sh species in
La-tin America, an international steering group was
established with the objective of organizing a
meet-ing on this topic on a regular basis The concept for
this series of meetings and publications is the
promo-tion and development of native species culture in
or-der to minimize the introduction, distribution and
transplantation of exotic species throughout and
within the region The conservation of biodiversity is
also taken into account as an important objective
As a consequence, the Second Latin American
Con-ference on Culture of Native Fishes was held in
Chasco-mu¤s, Buenos Aires, Argentina between 3rd and 6th
November 2009 and the ¢rst 11 articles in the present
issue represents a selection of the presented papers
Chascomu¤s is a small city located120 km south of
Bue-nos Aires with a tradition in studies of the biology,
cul-ture and propagation of an emblematic species for that
area, the pejerrey Odontesthes bonariensisValenciennes,
1835, Atherinopsidae This meeting was supported by
CONICET (Argentine National Research Council), CIC
(Committee for Scienti¢c Research of Buenos Aires
Province), CFI (Federal Investment Council), the
Min-istry of Agricultural A¡airs of the Province of Buenos
Aires and the Chascomu¤s Municipality The meeting
was extremely successful and approximately 250
students, technicians and researchers attended the
di¡erent sessions A series of ¢ve lectures on key lected topics were presented by invited speakersfrom Colombia, Spain, United States, France, Chileand Mexico There were also 50 selected oral presen-tations and 159 posters from across the continentcovering a wide range of species and subject areas.This meeting had fundamental importance forgathering together scientists, technicians and localfarmers interested in the biology and culture ofLatin American species A detailed description ofthe meeting is given at http://www.laccnf.com.ar/index.php?idioma=1&laccnf=0
se-We would like to thank the enthusiastic e¡orts of
Dr Luis F Canosa, Dr Dar|¤o Collauti, Mr Manuel ina, Dr Leandro Miranda (IIB-INTECH) and Lic Gus-tavo Berasain and Lic Marcela Velasco (Ministry ofAgricultural A¡airs of the Province of Buenos Aires)
Mar-It is also important to mention the help of the ing persons: Andre¤s Alonso, Leandro Balboni, Danie-
follow-la Campanelfollow-la,TomaŁs Chalde, Mar|¤a Victoria Crivelli,Mariano Elisio, Juan Ignacio Fernandino, Javier Gar-cia de Souza, AŁngela GaŁrriz, Pedro Go¤mez-Requeni,Yamila Grunblat, Mar|¤a Ester Haspert, MauricioKraemer, Facundo Llompart, Gabriela Carina Lo¤pez,TomaŁs Maiztegui, Horacio Onatibia, Mar|¤a Rita Pe¤rez,Mar|¤a Ine¤s Pietrantuono, Maria Emilia Scharrig andPatricio Solimano without whom the symposiumcould not have been possible
We are looking forward to the next symposium to
be held at the University of Lavras, Minas Gerais, zil in 2011, hoping that it will show further substan-tial advances in the culture of native species for LatinAmerica
Bra-Aquaculture Research, 2011, 42, 737 doi:10.1111/j.1365-2109.2011.02881.x
Trang 3Cryopreservation effects on the sperm quality of
Juan Antonio Ramirez-Merlano, Yohana Mar|¤a Velasco-Santamar|¤a, V|¤ctor Mauricio Robles & Pablo Emilio Cruz-Casallas
Medina-Research Group on Reproduction and Toxicology of Aquatic Organisms ^ GRITOX, Aquaculture Institute, University of the Llanos, Meta, Colombia
Correspondence: P E Cruz-Casallas, Research Group on Reproduction and Toxicology of Aquatic Organisms ^ GRITOX, Aquaculture Institute, University of the Llanos, Km 12 via Puerto Lo¤pez,Villavicencio, Meta, Colombia E-mail: pecruzcasallas@unillanos.edu.co
Present address: Y M Velasco-Santamar|¤a, Institute of Biology, University of Southern Denmark, Campusvej 55, DK-5230 Odense M, Denmark.
Abstract
The e¡ects of straws volume, cryoprotectants and
thawing temperatures were evaluated on the sperm
quality of cachama blanca Piaractus brachypomus
(Cuvier), an important Colombian ¢sh species
Sexu-ally mature ¢sh were induced to ovulation or
sper-miation with a carp pituitary extract A pool of
suitable sperm samples was diluted in glucose, egg
yolk, dimethyl sulphoxide (DMSO-10%), methanol
(MET-10%) or ethylene glycol (ETG-5%) and packed
in 0.5, 2.5 or 5.0 mL straws and frozen in nitrogen
va-pour The thawing process was performed in a 35 or
an 80 1C water bath The fertility was evaluated after
6 h post fertilization The highest motility percentage
(33 3%) was observed with sperm cryopreserved
with DMSO, packed in 5 mL straws and thawed at
35 1C The treatments with DMSO and MET packed
in 0.5 and 5.0 mL straws and thawed at 35 1C showed
the highest fertility (higher than 71%) and the lowest
fertility was obtained with MET-2.5 mL (9 5%) In
all the treatments, a signi¢cant decrease in the sperm
quality was observed at 80 1C Sperm cryopreserved
with DMSO-10% or MET-10%, packed in 2.5 or
5.0 mL straws are suitable to achieve acceptable
ferti-lization and to fertilize high amounts of eggs
Keywords: sperm cryopreservation, straw
vo-lume, cryoprotectants, thawing temperature,
sperm quality, Piaractus brachypomus
Introduction
The use of high-quality gametes in aquaculture is an
important factor to achieve the successful
reproduc-tion of ¢sh species under captivity (Rurangwa, Kime,Ollevier & Nash 2004) Cryopreservation is consid-ered to be one of the most e⁄cient strategies to facil-itate the handling and storage of gametes (Sarvi,Niksirat, Mojazi Amiri, Mirtorabi, Ra¢ee & Bakh-tiyari 2006) The use of cryopreserved sperm is a prac-tical alternative to increase the population size and tomaintain the genetic diversity, especially in thosespecies kept in captivity (Piironen 1994) Cachamablanca Piaractus brachypomus (Cuvier1818) is a native
¢sh of South America inhabiting the basins of theOrinoco and Amazon rivers Under farming condi-tions, this species shows an excellent performancedue to its rusticity, omnivorous food habits, docility,meat quality, food conversion and resistance to dis-eases, and therefore is the most widely farmed native
¢sh species in Colombia (Arias, VaŁsquez-Torres,
Orre-go & Isaza 1989; HernaŁndez, Munoz, Ferraz De Lima,
De Santis, VaŁsquez-Torres, GonzaŁlez, Morales, tara, Luna & Kossowski 1992)
Alcan-Despite the successful research on their tive aspects, the current ¢ngerling production of ca-chama blanca is still limited due to the low or evenabsent male gamete availability during most of theyear An alternative to solve this problem is the spermcryopreservation, a biotechnology implemented inabout 200 ¢sh species (Rana 1995); however, the use
reproduc-of this technology in P brachypomus is restricted due
to the need for high amounts of suitable sperm toachieve high fecundity in this species Sperm is tradi-tionally packed in small straws (0.5 mL FrenchStraws), which does not allow the fertilization of ahigh number of oocytes Protocols for sperm cryopre-servation in straws of large volume (44 mL) have
Aquaculture Research, 2011, 42, 738^745 doi:10.1111/j.1365-2109.2011.02835.x
Trang 4been evaluated in some freshwater ¢sh species such
as Rainbow trout Oncorhynchus mykiss (Walbaum
1972) (Wheeler & Thorgaard 1991;
Lahnsteiner,Weis-mann & Patzner 1997), Brown trout Salmo trutta fario
(Linnaeus 1758), Salmo trutta lacustris (Linnaeus
1758), Arctic char Salvelinus alpinus (Linnaeus 1758)
(Lahnsteiner et al 1997; Richardson, Miller &
McNi-ven 2000),Wels cat¢sh Silurus glanis (Linnaeus 1758)
(Bart, Wolfe & Dunham 1998), Prochilodus lineatus
(Valenciennes 1837) (Viveiros, Orfao, Maria &
Alla-man 2009) and Brycon amazonicus (Spix & Agassiz
1829) (Velasco-Santamar|¤a, Medina-Robles &
Cruz-Casallas 2006) These previous studies concluded
that packing sperm in large straws or macrotubes
could reduce the time required to pack and thaw the
sperm and also facilitate the handling during the
fer-tilization process
Another important issue is the use of
cryoprotec-tants, which role is to prevent cell damage during
the freezing and thawing steps during the
cryopre-servation process Several cryoprotectants have been
used for sperm ¢sh cryopreservation, including
methanol, ethylene glycol and dimethyl sulphoxide
(DMSO); however, DMSO is reported to be the most
e⁄cient to cryopreservate ¢sh spermatozoa (Anel &
Cabrita 2000; Cruz 2001) due mainly to its small
mo-lecular size, which allows it to enter and exit the
sper-matic cell easily (Tiersch, Williamson, Carmichael &
Gorman1998;Vincent, Pruliere, Pajot-Augy, Campion
& Douzou 1998)
On the other hand, a wide range of temperatures
used to thawed cryopreserved sperm with
tempera-tures from refrigeration (4 1C) to 80 1C are reported
(Lahnsteiner, Berger, Horvath, Urbanyi & Weismann
2000) A fast thawing temperature decreases the
recrystallization e¡ect in the spermatic cells and
therefore diminishes the membrane damage (Tiersch
et al 1998)
The use of various cryoprotectants in cachama
blanca sperm has been reported previously including
the most widely used DMSO, dissolved in glucose and
egg yolk solution and packed in small straws
(Fresne-da, Lenis, Agudelo & Olivera-AŁngel 2004; Navarro,
Velasco-Santamar|¤a & Cruz-Casallas 2004;
Nasci-mento, Maria, Pessoa, Carvalho & Viveiros 2010)
The main goal of the current study was to develop
an appropriate protocol for P brachypomus sperm
cryopreservation, minimizing the deleterious e¡ects
of the cryopreservation process, therefore
contribut-ing to increase the gamete availability outside the
breeding season The e¡ects of three straw volumes,
three cryoprotectants and two thawing temperatures
on the post-thaw sperm quality of P brachypomuswere assessed
Materials and methodsExperimental animalsSexually mature cachama blanca P brachypomus,5.0 0.6 kg body weight (BW) and 64 3 cm (totallength), were used The ¢sh were reared and kept inponds at a ratio1.5 ¢sh m 2at the Aquaculture Insti-tute of the University of the Llanos,Villavicencio, Co-lombia The reproductive maturity was assessed inmales by the presence of sperm in the urogenital pa-pillae (after a slight massage on the coelomic cavity)and in females by measuring, through an ovarianbiopsy the oocyte diameter (41200 mm) and evaluat-ing the migration of the oocyte germinal vesicle
Gametes extractionMature ¢sh were collected from the earthen pondsand kept in circular ponds with permanent aerationand water recirculation The males were injected in-tramuscularly with a pituitary carp extract (EPC,Stoller Fisheries, Spirit Lake, IA, USA) at 4.0 mg kg 1
BW The females were injected with 5.75 mg kg 1BWdistributed in three applications at 0 h (0.25 mg kg 1BW), 24 h (0.5 mg kg 1BW) and 36 h (5.0 mg kg 1BW) To obtain the gametes, the ¢sh were anaesthe-tized in a 2-phenoxyethanol solution (300 mg L 1,Sigma Chemical Co., St Louis, MO, USA) until loss ofthe swimming axis was observed The sperm was col-lected 18 h after the hormone injection in glass grad-uated tubes (15 mL) Sperm samples contaminatedwith water, urine, faeces or blood were discarded Atthe ovulation time, the oocytes were collected 6^7 hafter the last hormone injection The abdominal re-gion in each ¢sh was carefully dried to avoid contactbetween water and the gametes
Sperm quality evaluationThe sperm collected was kept at room temperature(25 2 1C) The gross motility and duration of moti-lity were evaluated using light microscopy ( 10) byadding 10mL of sperm to a concave slide (1.0^1.2 mmdepth, Micro Slides Premiere, Shanghai, China) andactivating the motility spermatozoa with 190mL ofdistilled water (1:20) Samples witho80% progres-sive motility were not processed for cryopreservation.The duration of motility was evaluated immediately
Aquaculture Research, 2011, 42, 738^745 Cryopreservation of Piaractus brachypomus sperm J A Ramirez-Merlano et al.
Trang 5after the addition of water, and registered until the
sperm movements decreased to ca 5% The sperm
concentration was determined using two methods
The ¢rst one corresponded to an indirect known
technique called spermatocrit, where sperm is placed
in microcapillary tubes (75 mm length and 1.1mm
internal diameter) and centrifuged at 14 000 g for
5 min The direct method was carried out through
the haemocytometer method, diluting (1:1200)
pre-viously the sperm in a saline physiological solution
(0.9% NaCl), kept under a humid atmosphere for
10 min, and subsequently, individual spermatozoa
were counted in a Neubauer chamber (Bright Line,
Optik Labor, Friedrichshofen, Germany) at 40
magni¢cation (Cruz-Casallas 2001)
Sperm cryopreservation process and thawing
conditions
The selected sperm for cryopreservation were pooled
and diluted (1:5) initially in an extender solution
con-taining glucose as an external cell cryoprotectant
(5.5%, Merck, Darmstadt, Germany) and hen’s egg yolk
as both a stabilizer membrane and an internal
cryo-protectant To evaluate the cryoprotectants e¡ects,
10% DMSO (Sigma Chemical Co.), 5% ethylene glycol
(ETG, Sigma Chemical Co.) or 10% methanol (MET,
Sigma Chemical Co.) were used After the sperm
dilu-tion in each extender, no motile spermatozoa were
ob-served in any of the samples Afterwards, the diluted
sperm was packed in di¡erent straw volumes: 0.5 mL
(French Straws,130 3 mm, Instrument de Me¤decine
Ve¤te¤rinaire, Minneapolis, MN, USA), 2.5 (140 5 mm)
and 5.0 mL macrotubes (280 5 mm, Minitub,
Ab-fˇl-und Labortechnik GmbH, Tiefenbach, Germany)
The French straws were sealed with polyvinyl and
the macrotubes were sealed with metallic balls
The straws and macrotubes were placed vertically
in a custom-made support and frozen for 30 min in
nitrogen vapour in a dry shipper (CP100
Taylor-Wharton, Theodore, AL, USA) The decrease in the
temperature was monitored using a thermocouple
(WBrand, Friendswood, TX, USA, precision 0.01 1C
and range 200 to 800 1C) inserted directly into
the ¢lled straws or macrotubes All straws were
sub-merged and maintained in a liquid nitrogen
contain-er at 196 1C (35 HC, Taylor-Wharton) until further
evaluation The thawing process at 35 1C was carried
out by submerging the straws or macrotubes for 90 s
in a water bath The thawing time at 80 1C varied
ac-cording to the straw size as follows: 10 s for 0.5 mL
straws and 25 s for 2.5 and 5.0 mL macrotubes
Experimental design
To evaluate the e¡ect of di¡erent straw volumes (0.5,2.5 or 5.0 mL), cryoprotectant substances (DMSO,MET or ETG) and thawing temperatures (35 or
80 1C), a random experimental design with a ial 3 3 2 was used The factor 1 corresponded tothe three cryoprotectants used, factor 2 to the threestraw size volumes and factor 3 to the two thawingtemperatures, with a total of 18 treatments Thepost-thawing sperm quality was evaluated using thegross motility, duration of motility and fertility test(n 5 6) The gross motility was induced using 1% so-dium bicarbonate (NaHCO3, Merck) and subjectivelyevaluated using a scale from 1% to 100% The dura-tion of motility (s) was recorded using a stopwatch.For the fertility test, 2 g of eggs (1640 eggs g 1) wereseminated with 400mL of post-thawed sperm (ca
factor-296280 motile spermatozoa egg 1ratio) and the tivation was carried out by adding 5 mL of 1% NaH-
ac-CO3 After semination, the hydrated eggs wereplaced in experimental incubators with an ascen-dant £ow (2 L) and permanent water until hatching
As a control, the same amount of eggs was seminatedwith 100mL of fresh sperm (ca 54 649 motilespermatozoa egg 1ratio) The fertility in each treat-ment was evaluated 6 h post semination, evaluatingthe proportion of fertile eggs (translucent aspect andnormal embryo development) in a total knownamount of eggs
Statistical analysisData are expressed as mean standard error of themean (SEM) In order to verify the data normality dis-tribution and homogeneity of variance, Kolmogorov^Smirnov and Bartlett’s tests were carried out In addi-tion, the data were analysed considering the plot resi-dual distributions Data were analysed by a multiplecomparison analysis, followed by a Tukey’s test in or-der to evaluate the e¡ect of straw sizes, cryoprotectantsubstances and thawing temperatures on the spermquality In all cases, Po0.05 was used to show signi¢-cant di¡erences All statistical analysis was conductedusing theSASsystem for Windows software version9.1.3 (2002^2003 by SAS Institute, Cary, NC, USA)
ResultsThe fresh sperm characteristics of the samples suita-ble for cryopreservation are shown in Table 1 In allcases, the males had a seminal volume higher than
Cryopreservation of Piaractus brachypomus sperm J A Ramirez-Merlano et al Aquaculture Research, 2011, 42, 738^745
Trang 610 mL with a sperm motility480% The duration of
motility £uctuated between 36 and 56 s, while the
sperm concentration varied between 2.0 106
The freezing rates were analysed considering three
temperatures ranges from 28 to 20 1C, 20
to 100 1C and from 100 to 196 1C for all three
straw sizes (Table 2) In the range 28 to 20 1C
and 20 to 100 1C, the 0.5 mL straws showed the
fastest freezing curve compared with 2.5 and 5.0 mL
straws Nevertheless, the freezing temperatures in all
straws were considered to be stable from 160 1C
approximately (Fig 2)
Sperm qualityPost-thaw sperm motilityThe highest percentages of post-thaw motility wereobserved in sperm cryopreserved with 10% DMSOand thawed at 35 1C in all straw volumes In general,sperm frozen in 5 mL straws showed the highestpost-thaw motility regardless of both the thawingtemperature and the cryoprotectant (Table 3)
At 35 1C, the lowest sperm post-thaw motility wasobserved with ethylene glycol as a cryoprotectantand frozen in 2.5 mL straws (17%) On the otherhand, thawing temperature at 80 1C substantially de-creased the sperm post-thaw motility, with spermpacked in 0.5 mL straws being the most a¡ected This
¢nding was clearly observed in sperm cryopreserved
in MET-10% and ETG-5% (both with 3% post-thaw
Table 1 Fresh sperm characteristics of cachama blanca
Piaractus brachypomus (Cuvier 1818) broodstock, sampled
18 h after an intramuscular injection of a pituitary carp
extract (4.0 mg kg 1 body weight)
Figure 1 Relationship between spermatocrit (%) and
sperm concentration (spermatozoa 106mL 1) with a
1:1200 dilution (n 5 23; r250.62) of cachama blanca
Piar-actus brachypomus (Cuvier 1818) sperm
Table 2 Freezing rate ( 1C min 1 ) of extended cachama blanca Piaractus brachypomus (Cuvier 1818) sperm packed
in 0.5, 2.5 and 5.0 mL straws (n 5 2)
Temperature range ( 1C)
Straw size (mL) 0.5 2.5 5.0
Figure 2 Freezing curves of cachama blanca Piaractusbrachypomus (Cuvier 1818) sperm packed in 0.5, 2.5 and5.0 mL straws and frozen in nitrogen vapour in a dry ship-per for 30 min (n 5 2)
Aquaculture Research, 2011, 42, 738^745 Cryopreservation of Piaractus brachypomus sperm J A Ramirez-Merlano et al.
Trang 7motility) and DMSO-10% (10%), thawed at 80 1C and
packed in 0.5 mL straws (Po0.05)
Post-thaw duration of motility
The highest post-thaw duration of motility was
ob-served with sperm cryopreob-served in DMSO-10% and
thawed at 35 1C (Table 3) Under these conditions,
di-luted sperm frozen in 0.5 and 5.0 mL straws showed
the highest value (62 and 58 s, respectively) On the
other hand, sperm cryopreserved with ETG-5% packed
in 0.5 mL straws and MET-10% frozen in 0.5 and
2.5 mL straws, and thawed at 80 1C showed the lowest
duration of motility (10 and 9 s, respectively) Overall,
frozen sperm thawed at 35 1C was the optimum
condi-tion to obtain a suitable duracondi-tion of motility
Fertility assays
Table 4 shows the e¡ects of cryoprotectants, thawing
temperatures and straw sizes on P brachypomus egg
fertility The fertilization percentage obtained with
fresh sperm was signi¢cantly higher (89%) than all
frozen thawed sperm (Po0.05) Sperm cryopreserved
with DMSO and MET, frozen in 0.5 or 5.0 mL strawsand thawed at 35 1C showed the greatest fertility per-centages (higher than 71%) However, sperm cryo-preserved with ETG, frozen in 2.5 mL straws andthawed at 35 1C obtained the lowest fertilization per-centage (42%), although no signi¢cant di¡erenceswere observed with the above treatments In con-trast, the thawing temperature at 80 1C showed acritical e¡ect on the fertility assay, i.e sperm cryopre-served in MET, frozen in 2.5 mL straws and thawed at
80 1C had the lowest fertility (9%, Po0.05) Sperm luted with ethylene glycol and DMSO as a cryoprotec-tant and thawed at 80 1C had a higher fertility at thistemperature regardless of the straw volume (between27% and 36%) During the entire incubation process,the temperature and pH were maintained constantwith values of 26.9 0.4 1C and pH 6.5 0.4
di-DiscussionThe results obtained in the current study contributesigni¢cantly towards improving the protocol devel-
Table 3 Post-thaw sperm motility and duration of motility
of cachama blanca Piaractus brachypomus (Cuvier 1818)
sperm cryopreserved with DMSO, ETG or MET, frozen in
0.5, 2.5 or 5.0 mL straws and thawed in a 35 or an 80 1C
Motility (%)
Duration of motility (s)
Between rows, means with di¡erent superscripts are
signi¢-cantly di¡erent (P o0.05).
DMSO, dimethyl sulphoxide; ETG, ethylene glycol; MET,
metha-nol; ND, not de¢ned.
Table 4 Fertilization percentages of cachama blanca actus brachypomus (Cuvier 1818) sperm cryopreserved with DMSO, ETG or MET, frozen in 0.5, 2.5 or 5.0 mL straws and thawed in a 35 or an 80 1C water bath
Piar-Cryoprotector
Thawing temperature ( 1C)
Straw size volume (mL) Fertility (%)
Control (fresh sperm) 89 7.3 a
Between rows, means with di¡erent superscripts are cantly di¡erent (P o0.05).
signi¢-Values expressed as mean SEM (n 5 6).
DMSO, dimethyl sulphoxide; ETG, ethylene glycol; MET, methanol.
Cryopreservation of Piaractus brachypomus sperm J A Ramirez-Merlano et al Aquaculture Research, 2011, 42, 738^745
Trang 8opment of sperm cryopreservation in P brachypomus
at large scale
The sperm quality of samples before
cryopreserva-tion had an optimal gross motility ca 86%, which is
in agreement with previous studies on this species
(Fresneda et al 2004; Navarro et al 2004) and in
re-lated species such as B amazonicus (Cruz-Casallas,
Velasco-Santamar|¤a & Medina-Robles 2006)
Simi-larly, the seminal volume obtained was higher than
the reported in other South American characidae
such as Piaractus mesopotamicus (Holmberg 1887)
(5.02 mL) (Fogli Da Silveira, Kavamoto & Narahara
1985) and P lineatus (2.2 mL) (Viveiros et al
2009) The sperm concentration was similar to
the previously study reported by Navarro et al
(2004); however, a higher sperm concentration
(30 spermatozoamL 1 106
) and duration of lity (1006 s) were reported in the same species by
moti-Fresneda et al (2004)
Despite the low number of ¢sh used, a good and
signi¢cant relationship between spermatocrit
and sperm concentration was observed in this study
This value is close to previous studies in B
amazoni-cus (75%) (Cruz-Casallas et al 2006), Cod¢sh Gadus
morhua (Linnaeus1758) (75%) (Rakitin, Moira,
Fergu-son & Trippel 1999) and Rainbow trout (65%)
(Bastardo, Guedez & Leo¤n 2004) These results
con-¢rm the clear relationship between these two
vari-ables and therefore the spermatocrit is an important
alternative to determine indirectly the sperm
con-centration in P brachypomus This result has a high
relevance and practical application mainly when a
high signi¢cantly number of samples have to be
evaluated
The sperm cryopreservation using nitrogen
vapours is the technique most used in ¢sh (Linhart,
Rodina & Cosson 2000; Cruz-Casallas et al 2006)
During this process, the spermatozoa is subject
to drastic physical and chemical changes such
as ice crystals’ formation, mechanical and osmotic
stress and destabilization of the plasmatic membrane
(Lahnsteiner, Patzner & Weismann 1992; Labbe,
Crowe & Crowe 1997) Straws with a small diameter
such as 0.5 mL o¡er a higher freezing index than
the macrotubes (2.5 and 5.0 mL) which have a
smaller index with a longer plateau (Bwanga,
Braganca, Einarsson, & Rodriguez-Martinez 1990;
Bwanga, Einarsson & Rodriguez-Martinez 1991)
Lahnsteiner et al (2000) and Cabrita, Robles, Alvarez
and HerraŁez (2001) showed a higher freezing index in
Cyprinids and Rainbow trout sperm packed in
ac-60 1C for 8 s, which is close to the values obtained inthe current study In addition, higher post-thawsperm motility (77 17%) was observed whenmethylglycol was used as a cryoprotectant; neverthe-less, fertility tests were not carried out (Nascimento
et al 2010)
Compared with the current results, Fresneda et al.(2004) obtained higher sperm motility and a higherduration of motility using DMSO and MET as cryo-protectants (80% and 78%, respectively) Cabrita
et al (2001) reported in Rainbow trout a sperm lity close to 45% when the sperm was cryopreservedwith 7% DMSO and thawed at 25 1C for 30 s Piarac-tus brachypomus cryopreserved sperm thawed at
moti-80 1C showed low post-thaw motility, suggesting terations in the cellular membrane integrity, mainly
al-in sperm cryopreserved al-in 0.5 mL straws In contrast,Velasco-Santamar|¤a et al (2006) reported similarsperm motility (ca 38%) in sperm cryopreservedwith DMSO, packed in 1.8, 2.5 or 4.0 mL straws andthawed at 80 1C In the current study, DMSO andMET showed better results as cryoprotectants prob-ably due to their capacity to enter and exit the spermcell based on the gradient concentration (Tiersch
et al 1998; Vincent et al 1998) and as well as to theability to maintain the membrane integrity andthe mitochondrial function during the cryopreserva-tion process (Ogier De Baulny, Vern, Kerboeuf &Mais 1997)
The signi¢cant decrease in the sperm post-thawmotility and duration of motility when the spermwas thawed at 80 1C, can be explained by the factthat high temperatures are viable within a short timesince fast thawing rates can contribute towards de-creasing the re-crystallization e¡ect observed in thespermatic cells and minimizing the thermal e¡ect
on the membrane (Tiersch et al 1998) Folgi da veira et al (1985) obtained 21% fertility using Rham-dia hilarii frozen sperm packed in 0.5 mL straws and
Sil-Aquaculture Research, 2011, 42, 738^745 Cryopreservation of Piaractus brachypomus sperm J A Ramirez-Merlano et al.
Trang 9thawed at 70^80 1C for 3^4 s Similarly, low fertility
(o45%) was observed in B amazonicus sperm
thawed at 80 1C (Velasco-Santamar|¤a et al 2006) due
to the detrimental e¡ect on ¢sh sperm Thawing
tem-peratures at 80 1C could induce several changes in
the motility, viability and fertility of the sperm cell
probably a¡ecting the spermatozoa fertilizing ability
(Wamecke & Pluta 2003)
The fertility obtained in P brachypomus sperm
cryo-preserved with 10% DMSO (71%), 5% ETG (63%) or
10% MET (71%), packed in 0.5 mL straws and thawed
at 35 1C was higher than that obtained in a previous
study in the same species (Navarro et al 2004)
High fertility using cryopreserved sperm has been
reported in Rainbow trout, with percentages varying
slightly with the straw sizes e.g 81% in sperm packed
in 4.0 and 5.0 mL straws and thawed at 25 1C
(Stein-berg, Hedder, Baulain & Holtz 1995), and 75% in
sperm packed in 5.0 mL straws and thawed at 60 1C
(Cabrita et al 2001) In contrast to the previous
stu-dies, Cat¢sh Ictalurus punctatus (Ra¢nesque 1818)
sperm cryopreserved in methanol, Hanks solution
and power milk and packed in 0.5 and 1.0 mL straws
and thawed at 25 1C showed no fertility (0%) (Bart
et al 1998), con¢rming the relevance to optimize the
sperm cryopreservation process in each species
Straws with a non-traditional volume i.e 2.5 and
5.0 mL showed a signi¢cant similar fertility to those
obtained with fresh sperm, despite both the possible
high cell damage and the low sperm motility (o50%)
reported in other species (Richardson,Wilson, Crim &
Yao 1999; Yao, Crim, Richardson & Emerson 2000)
In this respect, Ciereszko, Drabrowski, Lin, Chris
and Toth (1999) concluded that in Muskellunge Exos
masquinongy (Mitchill 1824), there is no clear
rela-tionship between the spermatic post-thaw motility
and the fertility percentage Our results and those
obtained in other ¢sh species highlight the
impor-tance of carrying out a fertility test to validate sperm
cryopreservation protocols
The signi¢cant results obtained in the present
study demonstrate the practical application of large
volume straws on a commercial scale in ¢sh farms,
as their use will reduce the time spent freezing and
thawing the straws and higher amounts of oocytes
will be fertilized at the same time with fewer straws
Based on the results obtained, it is possible to
sug-gest that sperm cryopreserved in 10% DMSO or 10%
MET, packed in large-volume straws i.e 2.5 or 5.0 mL
and thawed at 35 1C are the most suitable conditions
to retain the sperm quality in P brachypomus having
optimal sperm motility, duration of motility as well as
high fertility percentages close to the values obtainedwith fresh sperm
AcknowledgmentsThis research was supported by Instituto de Investi-gaciones de la Orinoqu|¤a Colombiana (IIOC), Univer-sity of the Llanos,Villavicencio, Meta, Colombia
References
Anel I & Cabrita E (2000) E¡ect of external cryoprotectants
as membrane stabilizers on cryopreserved rainbow trout sperm Theriogenology 13, 623^635.
Arias C.J.A., VaŁsquez-Torres W., Orrego J & Isaza M (1989) Avances en la reproduccio¤n inducida de Piaractus brachypo- mus, cachama blanca Bolet|¤n Red de Acuicultura 3, 9^10 Bart A.N., Wolfe D.F & Dunham R.A (1998) Cryopreserva- tion of blue cat¢sh spermatozoa and subsequent fertiliza- tion of Channel cat¢sh eggs Transactions of the American Fisheries Society 127, 819^824.
Bastardo H., Guedez C & Leo¤n M (2004) Caracter|¤sticas del semen de trucha arco iris de diferentes edades, bajo con- diciones de cultivo en Me¤rida, Venezuela Zootecnia tropi- cal 22, 277^288.
Bwanga C.O., Braganca M.M., Einarsson S & Martinez H (1990) Cryopreservation of boar semen in mini and maxi-straws Zentralblatt fur Veterinaermedizin
Rodriguez-37, 651^658.
Bwanga C.O., Einarsson S & Rodriguez-Martinez H (1991) Cryopreservation of boar semen: II E¡ect of cooling rate and duration of freezing point plateau on boar semen fro- zen in mini and maxi-straws and plastic bags ActaVeteri- naria Scandinavica 32, 455^461.
Cabrita E., Robles V., Alvarez R & HerraŁez M.P (2001) preservation of rainbow trout sperm in large volume straws: application to large scale fertilization Aquaculture
Cruz V.L.B (2001) Criopreservacao do seŒmen de curimbataŁ (Prochilodus lineatus) Masters thesis in Vertebrate Zool- ogy, Pontif|¤cia Universidade Cato¤lica de Minas Gerais, Belo Horizonte, MG, Brazil, 59pp.
Cruz-Casallas P.E (2001) Te¤cnicas de laboratorio para la evaluacio¤n de la calidad seminal en peces Orinoquia 5, 155^163.
Cruz-Casallas P.E., Velasco-Santamar|¤a Y & Medina-Robles V.M (2006) Determinacio¤n del espermatocrito y efecto Cryopreservation of Piaractus brachypomus sperm J A Ramirez-Merlano et al Aquaculture Research, 2011, 42, 738^745
Trang 10del volumen de la dosis seminante sobre la fertilidad en
yamu¤ (Brycon amazonicus) Revista Colombiana de Ciencias
Pecuarias 19, 140^145.
Fogli Da Silveira W., Kavamoto E.T & Narahara M.Y (1985)
Avaliacao da qualidade e criopreservacao em forma de
‘pellets’do se¤men do bagre, Rhamdia hilarii (Valenciennes,
1840) Boletin Instituto de Pesca 12,7^11.
Fresneda A., Lenis G., Agudelo E & Olivera-AŁngel M (2004)
Espermiacio¤n inducida y Crioconservacio¤n de Cachama
blanca (Piaractus brachypomus) Revista Colombiana de
Ciencias Pecuarias Suplemento 17, 46^52.
HernaŁndez A., Munoz D., Ferraz De Lima J.A., De Santis D
F.R., VaŁsquez-Torres W., GonzaŁlez R., Morales R.,
Alcan-tara F., Luna TM & Kossowski C (1992) Estado actual del
cultivo de Colossoma y Piaractus en Brasil, Colombia,
PanamaŁ, Peru¤ yVenezuela Segunda Reunio¤n
Internacio-nal del grupo de trabajo te¤cnico de Colossoma y Piaractus.
Bolet|¤n Red de Acuicultura 6, 3^27.
Labbe C., Crowe L.M & Crowe J.H (1997) Stability of the lipid
component of Trout sperm plasma membrane during
freeze-thawing Cryobiology 34, 176^182.
Lahnsteiner F., Patzner R.A & Weismann T (1992) Energy
metabolism in spermatozoa of the grayling (Thymallus
thymallus) In: Proceedings of the Fourth International
Symposium of the Reproductive Physiology of Fish (ed by
A.P Scott, J.P.K Sumpter, D.E Kime & M.S Rofle),
pp 279 Sheffield, UK: FishSymp.
Lahnsteiner F.,Weismann T & Patzner R.A (1997) Methanol
as cryoprotectant and the suitability of 1.2 mL and 5.0 mL
straws for cryopreservation of semen from salmonid
¢shes Aquaculture Research 28, 471^479.
Lahnsteiner F., Berger B., Horvath A., Urbanyi B &
Weis-mann T (2000) Cryopreservation of spermatozoa in
Cipri-nid ¢shes Theriogenology 54, 1477^1498.
Linhart O., Rodina M & Cosson J (2000) Cryopreservation
of sperm in common carp Cyprinus carpio: sperm
motility and hatching success of embryos Cryobiology
41, 241^250.
Nascimento A.F., Maria A.N., Pessoa N.O., Carvhalo M.A.M.
& Viveiros A.T.M (2010) Out-of-season sperm
cryopre-served in di¡erent media of the Amazonian freshwater
¢sh piraitinga (Piaractus brachypomus) Animal
Reproduc-tion Science 118, 324^329.
Navarro O.J., Velasco-Santamar|¤a Y.M & Cruz-Casallas P.E.
(2004) Evaluacio¤n de cinco protectores para la
criocon-servacio¤n de semen de cachama blanca (Piaractus
brachy-pomus) Revista Colombiana de Ciencias Pecuarias Supl 17,
53^59.
Ogier De Baulny L.E., Vern Y., Kerboeuf D & Mais G (1997)
Flow cytometric evaluation of mitochondrial activity
and membrane integrity in fresh and cryopreserved
Rain-bow trout (Oncorhynchus mykiss) spermatozoa
Cryobiol-ogy 34, 141^149.
Piironen J (1994) Composition and cryopreservation of
sperm from some Finnish freshwater teleost ¢sh Finnish
Fisheries Research 15, 27^48.
Rakitin A., Moira M., Ferguson E & Trippel A (1999) matocrit and spermatozoa density in Atlantic cod (Gadus morhua): correlation and variation during the spawning season Aquaculture 170, 349^358.
Sper-Rana K (1995) Cryopreservation of aquatic gametes and bryos: recent advances and applications In: Proceedings
em-of the Fifth International Symposium on the Reproductive Physiology of Fish (ed by F.W Goetz & P Thomas), pp 85–89 Austin, TX, USA: University of Texas at Austin Richardson G.F., Wilson C.E., Crim L.W & Yao X.Z (1999) Cryopreservation of yellowtail £ounder (Pleuronectes ferrugineus) semen in large straws Aquaculture 174, 89–94.
Richardson G.F., Miller T.L & McNiven M.A (2000) preservation of Arctic charr Salvelinus alpinus semen in various extenders and three sizes of straw Aquaculture Research 31, 307^315.
Cryo-Rurangwa E., Kime D.E., Ollevier F & Nash J.P (2004) The measurement of sperm motility and factors a¡ecting sperm quality in cultured ¢sh Aquaculture 234, 1^28 Sarvi K., Niksirat H., Mojazi Amiri B., Mirtorabi S.M., Ra¢ee G.R & Bakhtiyari M (2006) Cryopreservation of semen from the endangered Caspian brown trout (Salmo trutta caspius) Aquaculture 256, 564^569.
Steinberg H.A., Hedder R., BaulainY & HoltzW (1995) preservation of rainbow trout (Oncorhynchus mykiss) se- men in straws In: Proceedings of the Fifth International Symposium on the Reproductive Physiology of Fish (ed by F.W Goetz & P Thomas), pp 146 Austin, TX, USA: University of Texas at Austin.
Cryo-Tiersch T.R.,Williamson J.H., Carmichael G.J & Gorman O.T (1998) Cryopreservation of sperm of the endangered ra- zorback sucker Transactions of the American Fisheries So- ciety 127, 95^104.
Velasco-Santamar|¤aY.M., Medina-Robles V.M & las P.E (2006) Cryopreservation of yamu¤ (Brycon amazoni- cus) sperm for large scale fertilization Aquaculture 256, 264^271.
Cruz-Casal-Vincent C., Pruliere G., Pajot-Augy E., Campion E & Douzou
P (1998) Biophysical chemical aspects of cellular havior Biophysical Chemistry 29, 161^169.
cryobe-Viveiros A.T.M., Orfao L.H., Maria A.N & Allaman I.B (2009) A simple, inexpensive and successful freezing method for curimba Prochilodus lineatus (Characiformes) semen Animal Reproduction Science 112, 293^300 Wamecke D & Pluta H.J (2003) Motility and fertilizing capa- city of frozen/thawed common carp (Cyprinus carpio L.) sperm using dimethyl-acetamide as the main cryoprotec- tant Aquaculture 187, 361^375.
Wheeler P & Thorgaard G (1991) Cryopreservation of rainbow trout semen in large straws Aquaculture 93, 95^100.
Yao Z., Crim L.W., Richardson G.F & Emerson C.J (2000) tility, fertility and ultrastructural changes o¡ ocean put (Macrozoarces americanus L.) sperm after cryopreserva- tion Aquaculture 181, 361^375.
Mo-Aquaculture Research, 2011, 42, 738^745 Cryopreservation of Piaractus brachypomus sperm J A Ramirez-Merlano et al.
Trang 11Temperature effects on sex differentiation of the
Junpei Inazawa, Ricardo Shohei Hattori, Miho Oura, Masashi Yokota & Carlos Augusto Strˇssmann
Graduate School of Marine Science and Technology,Tokyo University of Marine Science and Technology,Tokyo, Japan
Correspondence: C A Strˇssmann, Department of Marine Biosciences, Faculty of Marine Science, Tokyo University of Marine Science and Technology, Konan 4-5-7 Minato,Tokyo 108 8477, Japan E-mail: carlos@kaiyodai.ac.jp
Abstract
The pejerrey Odontesthes bonariensis (Valenciennes
1835) and the Patagonian pejerrey Odontesthes
hatch-eri (Eigenmann 1909) are Athhatch-erinopsid species with
commercial importance and potential for
aquacul-ture The hybrids of the two species are viable but
their mode of sex determination is unknown This
study examined the gonadal histology and sex ratios
of reciprocal hybrids that were reared at 15, 17, 21, 25
or 29 1C during the sex di¡erentiation period The
genetic sex of hybrids from O hatcheri fathers was
inferred from a sex-linked SNP marker Both hybrids
showed female-biased sex ratios at the lowest
temperature, female-biased to balanced sex ratios
at intermediate temperatures and balanced or
male-biased sex ratios at 29 1C, but unlike in purebred
O bonariensis, the lowest and highest temperatures
did not yield monosex populations The proportion
of females in the o¡spring was a¡ected more by
par-ental genome than by hybrid combination Female
hy-brids bearing the O hatcheri Y chromosome showed
temporary arrest of ovarian development that was
rescued in adults These results reveal strong
interac-tions between genotype and temperature for sex
de-termination and di¡erentiation of the hybrids and
provide important clues to understand the
mechan-isms of sex determination in these species
Keywords: hybrids, sex determination,
Odontes-thes bonariensis, OdontesOdontes-thes hatcheri,TSD, GSD
Introduction
The control of primary gonadal sex di¡erentiation
and gonadal maturation are of great interest for the
aquaculture industry The former allows the tion of monosex ¢sh of the sex that has a highergrowth rate, £esh or roe value or as in the case of or-namental ¢sh, which has a more attractive appear-ance (Devlin & Nagahama 2002; Strˇssmann,Karube, Miranda, Patino, Somoza, Uchida & Yama-shita 2004; Cnaani & Levavi-Sivan 2009) The lattercan be used for delaying puberty in order to increasethe growth and/or the feed conversion rates (Taran-ger, Carrillo, Schulz, Fontaine, Zanuy, Felip, Weltzien,Dufour, Karlsen, Norberg, Andersson & Hansen2010) One strategy explored to control the sex offarmed ¢sh consists in interspeci¢c hybridization,which can result in sterile animals that divert more
produc-of the ingested nutrients into somatic growth andtherefore show higher growth rates (Wohlfarth2008) Likewise, hybrids occasionally show heterosisand higher tolerance to extreme environmental con-ditions than the original species (Chevassus 1983),and therefore, may be able to adapt to a broader range
of environments
The pejerrey Odontesthes bonariensis (Valenciennes1835) and the Patagonian pejerrey Odontesthes hatch-eri (Eigenmann 1909) are two native Atherinopsidspecies from South America that have generated in-terest for aquaculture both domestically and abroad(Strˇssmann & Yasuda 2005; Somoza, Miranda, Be-rasain, Colautti, Lenicov & Strˇssmann 2008) How-ever, these two species show relatively slow growthrates, which may be due to early sexual maturationand reproductive activity before attaining marketsize (Strˇssmann, Ng, Oshiro & Takashima 1993; So-moza et al 2008) Viable reciprocal hybrids betweenpejerrey and Patagonian pejerrey have been docu-mented in captive-reared and natural populations(Strˇssmann, Ijima,Yamaguchi,Yoshizaki & Takashima
Aquaculture Research, 2011, 42, 746^753 doi:10.1111/j.1365-2109.2010.02757.x
Trang 121997), the latter likely being the result of antropogenic
translocations (Somoza et al 2008) However, little is
known about the gonadal development of reciprocal
hybrids, particularly their sex ratio, sexual maturation
and fertility
These two species have been of interest recently for
studies on sex determination and di¡erentiation
Odontesthes bonariensis presents marked
tempera-ture-dependent sex determination (TSD), whereby
all-male or all-female populations can be reliably
pro-duced by manipulation of the rearing temperature
during early development (Strˇssmann, Moriyama,
Hanke, Calsina Cota & Takashima 1996; Strˇssmann,
Saito, Usui, Yamada & Takashima 1997) Recently,
Hattori, Fernandino, Kishii, Kimura, Kinno, Oura,
Somoza, Yokota, Strˇssmann and Watanabe (2009)
showed that the process of
high-temperature-depdent masculinization in this species involves
en-hanced stress hormone (cortisol) production On the
other hand, rearing temperature a¡ects sex
determi-nation only near the extremes of the thermal range
and gonadal fate is strongly determined by the
geno-type (XX^XYgenotypic sex determination, GSD) in O
hatcheri (Strˇssmann, Calsina Cota, Pronlor, Higuchi
& Takashima 1996; Strˇssmann, Saito et al 1997) In
this species, a sex-linked SNP marker (Koshimizu,
Strˇssmann, Okamoto, Fukuda & Sakamoto 2010)
and a strain (Ehi-M13) carrying this marker (Hattori,
Oura, Yokota, Strˇssmann & Watanabe 2010) have
been developed for use in sex determination studies
The presence of such contrasting sex-determining
mechanisms in these closely related species
(Strˇss-mann, Ijima et al 1997) makes their hybrids an
inter-esting material to study the interactions between
genotype and temperature on sex determination in
¢sh In this context, this study was designed to clarify
how sex is determined in reciprocal hybrids of
pejer-rey and Patagonian pejerpejer-rey reared at ¢ve di¡erent
temperatures and to examine the e¡ects of genotype
and temperature on gonadal development
Materials and methods
Crosses, thermal treatments and rearing
conditions
Fertilized eggs were obtained by arti¢cial
insemina-tion using manually stripped gametes from O
bonar-iensis (Obo) and O hatcheri (Oha) broodstock reared
at the aquatic animal rearing facilities of the Tokyo
University of Marine Science and Technology A total
of three experiments were conducted: two crosses
between Obo, and Oha< (using di¡erent parents)and one cross between Oha, and Obo< The O hatch-eri females and males used in this study were F2 ani-mals of the Ehi-M13 strain and were con¢rmed to be
XX and XYanimals respectively Fertilized eggs wereincubated at 19 1C in £ow-through water until hatch-ing (about 10 days after fertilization) Immediatelyafter hatching, approximately 150^200 larvae werestocked in each of 60 L tanks for rearing at constanttemperatures of 15, 17, 21, 25 and 29 1C ( 0.5 1C).These thermal regimes have been selected based onprevious studies on thermolabile sex determination
in these two species (Strˇssmann, Moriyama et al.1996; Strˇssmann, Calsina Cota et al 1996; Strˇss-mann, Saito et al 1997) Fish were reared in £owingbrackish water (2^5 g L 1NaCl) under a constantphotoperiod (16:00 hours light:8:00 hours dark) andwere fed live (Artemia nauplii) and powdered (Tetra-Min £akes) food to satiation twice daily Because his-tological examination of the juveniles from theObo, Oha< crosses at 3 months revealed twotypes of ovaries, one histologically normal and an-other showing relative atrophy and few perinucleolaroocytes, the group with more females (15 1C in the
¢rst cross) was reared for an additional period of 15months and subjected again to histological analysiswhen they became sub-adults For this purpose, ¢shwere transferred at 3 months to 480 L tanks andreared until 18 months at 21 1C in order to promotegrowth and accelerate development
Sampling and histological analyses of the gonadsJuveniles were sampled at the end of 3 months, whenthey reached about 30^40 mm in body length, forhistological determination of the phenotypic (gona-dal) sex ratio and for DNA analysis of the genotypicsex of each individual In all cases, ¢sh were sacri-
¢ced in ice water, ¢xed in Bouin’s solution and cessed according to standard protocols for thepreparation of haematoxylin^eosin-stained histolo-gical sections Histological preparations were exam-ined under a microscope and juveniles were sexedfollowing the criteria of Ito, Yamashita, Takashimaand Strˇssmann (2005) To quantify the histologicaldi¡erences between normal and atrophic ovaries,the number of perinucleolar oocytes per cross sec-tion of the gonad was counted in representative sec-tions of individuals of known genotype from the ¢rstObo, Oha< cross Gonads of sub-adults from thiscross were also examined by histology as describedabove
pro-Aquaculture Research, 2011, 42, 746^753 Sex determination in Odontesthes hybrids J Inazawa et al.
Trang 13DNA extraction and polymerase chain
reaction (PCR) analysis of genotypic sex
A sex-linked SNP marker (Koshimizu et al 2010) was
used to distinguish o¡spring from the two Obo,
Oha< crosses that inherited the Y and X
chromo-somes of O hatcheri Genomic DNA was extracted
from the caudal ¢n using a standard
phenol:chloro-form protocol and PCR conditions followed the
proce-dures described by Hattori et al (2010)
Statistical analysis
The chi-square test was applied to determine whether
sex ratios in the progeny deviated signi¢cantly
from 1:1 Student’s t-test was used to determine
whether di¡erences in the number of perinucleolar
oocytes per cross section of the ovary between female
hybrids carrying Oha X and Ychromosomes were
sta-tistically signi¢cant All statistical analyses were
per-formed using GRAPHPAD software Version 5.02 (San
Diego, CA, USA) and signi¢cance was de¢ned as
Po0.05
ResultsSex ratios and gonadal development in theOha, Obo< cross
The phenotypic sex ratios of Oha, Obo< hybridswere female-biased at 15 and 17 1C and balanced(about 1:1) between 21 and 29 1C; no group had 100%males or females (Table 1) Gonads of ¢sh kept at high-
er temperatures were generally more developed thanthose reared at lower temperatures but all ovariesand testes within a given temperature group had si-milar histological appearances
Sex ratios and gonadal development in theObo, Oha< crosses
The phenotypic sex ratios of Obo, Oha< hybrids
in the ¢rst trial with this cross were female-biased attemperatures below 25 1C and nearly balanced at
29 1C (Table 2) The analysis of the relationship tween phenotypic sex and the presence of the X orthe Y chromosome of O hatcheri showed that most
be-¢sh with the X chromosome were females (89.7%).Surprisingly, in this cross, the majority (84.9%) ofthe animals bearing the Ychromosome were also fe-males, and these were detected even at 29 1C (Table2) In the second trial, phenotypic sex ratios wereslightly female-biased at 15 1C, balanced between 17and 25 1C and slightly male-biased at 29 1C (Table 3)
In this trial, most of the animals with the O hatcheri
X chromosome were females (88.3%) as in the ¢rsttrial, but those with theYchromosome were predomi-nantly male (85.3%)
In the two crosses of Obo, with Oha<, two types ofovaries could be clearly distinguished In the ¢rsttype, considered to be normal compared with exten-
Table 1 Sex ratios and survival rates of Oha, Obo<
hy-brids reared at di¡erent temperatures during gonadal sex
Signi¢cant deviation from 1:1 (Po0.05).
Table 2 Sex ratios, allocation of Odontesthes hatcheri X and Y chromosomes and survival rates of Obo, Oha< hybrids reared at di¡erent temperatures during gonadal sex determination (¢rst trial)
Signi¢cant deviation from 1:1 (Po0.05).
wValues of X and Y indicate the number of individuals bearing the X and Ychromosome of O hatcheri respectively Note: not all samples available were sex genotyped.
ND, not determined.
Sex determination in Odontesthes hybrids J Inazawa et al Aquaculture Research, 2011, 42, 746^753
Trang 14sive information reported for purebreds of O
bonar-iensis and O hatcheri (Strˇssmann, Moriyama et al
1996; Strˇssmann, Calsina Cota et al 1996;
Strˇss-mann, Takashima & Toda 1996;StrˇssStrˇss-mann, Saito
et al 1997), the ovaries were well developed and
pos-sessed abundant germ cells at various stages from
oo-gonia to perinucleolar oocytes (Figs 1a and 2a) In
contrast, ovaries of the second type were
compara-tively undeveloped and most of the germ cells were
oogonia, with very few in the chromatin nucleolus
and perinucleolar stages (Figs 1a and 2b) A
signi¢-cant di¡erence in the number of perinucleolar
oo-cytes per cross section of the ovaries was observed
between the two types of ovaries When compared
with the genotype of the animals, most ovaries
clas-si¢ed as of the ¢rst type were found to come from
an-imals carrying the X chromosome of O hatcheri
(98.4%; n 5121) whereas those of the second type
were mostly from animals carrying the Y some (96.3%; n 5 51) (Fig 1b) No histological di¡er-ences of any kind were observed between testes ofmales bearing the X and Ychromosomes
chromo-Sub-adult females showed mature ovaries withgerm cells at various developmental stages from oogo-nia to cortical alveoli oocytes regardless of the pre-sence of the X (n 5 2; Fig 2c) and Y chromosomes(n 510; Fig 2d) The males, all with theYchromosome,presented mature testes with a similar histological ap-pearance including germ cells at all developmentalstages (n 5 5; data not shown)
DiscussionInterspeci¢c hybridization among closely related ¢shspecies sometimes produces highly biased sex ratios,
Table 3 Sex ratios, allocation of Odontesthes hatcheri X and Y chromosomes and survival rates of Obo, Oha< hybrids reared at di¡erent temperatures during gonadal sex determination (second trial)
Signi¢cant deviation from 1:1 (Po0.05).
wValues of X and Y indicate the number of individuals bearing the X and Ychromosome of O hatcheri respectively Note: not all samples available were sex genotyped.
0 20 40 60 80 100
Figure 1 (a) Average number of previtelogenic oocytes per cross section of ovaries considered to be normal and abnormal
in female o¡spring from the Obo, Oha< crosses Dotted line indicates a statistically signi¢cant di¡erence (Student’st-test, Po0.05) (b) Relationship between the type of ovary and the presence of the X andYchromosomes from Oha fathers
in female o¡spring from the Obo, Oha< crosses
Aquaculture Research, 2011, 42, 746^753 Sex determination in Odontesthes hybrids J Inazawa et al.
Trang 15including all-male or all-female hybrids, as a result of
di¡erences in sex-factor number, strength and their
location in the genome between the species
(re-viewed by Devlin & Nagahama 2002) In this study,
the sex ratios of the reciprocal hybrids of O
bonarien-sis and O hatcheri varied considerably even between
di¡erent pairs of parents for the same cross but, with
the exception of one group with 100% females at
25 1C in the ¢rst trial of Obo, Oha<, did not result
in all-males or all-females Odontesthes hatcheri has
been shown to have the XX^XY sex chromosome
sys-tem (Hattori et al 2010) Interestingly, most but not all
individuals carrying the O hatcheri X and
Ychromo-somes were females and males, respectively, as would
be expected if the O hatcheri ‘genotypic’ sex-factor
was stronger than the O bonariensis ‘environmental’
sex-factor and the progeny followed the O hatcheri
GSD pathway at intermediate temperatures As
re-gards TSD, the hybrids seemed to also conserve part
of the temperature sensitivity that is common to both
species, although it was not as clear as in O
bonarien-sis, where monosex populations are consistently
formed below 17 1C and above 29 1C (Strˇssmann,
Moriyama et al 1996; Strˇssmann, Saito et al 1997)
In fact, the sensitivity to low temperature was
con-¢rmed in all three crosses while that to high
tempera-ture was apparently lost in two of them These
¢ndings suggest the possibility of complex tions between the putative genetic (Strˇssmann, Cal-sina Cota et al 1996; Strˇssmann, Saito et al 1997;Hattori et al 2010; Koshimizu et al 2010) and environ-mental factors (Strˇssmann, Moriyama et al 1996;Strˇssmann, Saito et al.1997) controlling sex in thesetwo species They might also be an indication thatproper function of the putative genetic sex-factors inthe X and Y chromosomes of O hatcheri may requireother genes in the homologous (sex) chromosomes.Strong parental e¡ects have been reported duringexperimental hybridization in salmonids (Chevassus1983), cyprinids (Pala, Klˇver, Thorsteinsdo¤ttir,Schartl & Coelho 2008; Pala, Schartl, Thorsteinsdo¤t-tir & Coelho 2009), cichlids (Wohlfarth 2008) andmedaka (Hamaguchi & Sakaizumi 1992; Shinomiya,Kato, Yaezawa, Sakaizumi & Hamaguchi 2006; Kato,Takehana, Sakaizumi & Hamaguchi 2010) Amongthese groups, the hybrids of medaka have been themost thoroughly investigated Crosses between O la-tipes and O curvinotus, two species with the testis-de-termining gene DMY (Matsuda, Nagahama,Shinomiya, Sato, Matsuda, Kobayashi, Morrey, Shibata,Asakawa, Shimizu, Hori, Hamaguchi & Sakaizumi2002; Matsuda, Sato,Toyazaki, Nagahama, Hamaguchi
Figure 2 Histological appearance of ovaries from female o¡spring of similar size that inherited the X (a and c, 39.4 and134.5 mm body length respectively) and Y (b and d, 40.8 and 132.5 mm body length respectively) chromosomes of Odon-testhes hatcheri at 3 months (a and b) and 18 months (c and d) after hatching Bars indicate 50mm (a and b) and 100 mm (cand d)
Sex determination in Odontesthes hybrids J Inazawa et al Aquaculture Research, 2011, 42, 746^753
Trang 16& Sakaizumi 2003), gave rise to sex-reversed XY
fe-males with depressed DMY expression in both
reci-procal hybrids (Shinomiya et al 2006) However,
these sex-reversed XY females were found only
among hybrids with the Hd-rR strain of O latipes
and not in those with the HNI strain (Kato et al
2010) Similar results have been observed in hybrid
mice, which showed reduced Sry expression and
male-to-female sex reversal
(Eicher,Washburn,Whit-ney & Morrow 1982) Although we have not
exam-ined it speci¢cally, depressed gene expression may
be the cause of the high incidence of females bearing
the Oha Y chromosome among the progeny from the
¢rst cross of Obo, and Oha< Another interesting
and common feature between our results and those
for medaka (Kato et al 2010) and mice (Eicher and
Washburn 2001) hybrids is that sex reversion in
hy-brids with a female genotype is rare, suggesting the
presence of a conserved regulatory mechanism of
hy-brid sex determination
Among the temperatures tested in our
experi-ments, it became clear that low temperatures had
stronger e¡ects on gonadal fate than the higher ones
This was more evident in the cross between Oha,
and Obo< followed by the ¢rst cross between Obo,
and Oha<, which produced highly female-biased sex
ratios In a recent review of thermolabile sex
determi-nation in teleosts, Ospina-AŁlvarez & Piferrer (2008)
pointed out that our knowledge on the process of
feminization through temperature manipulation is
limited, in part because it is rarely achieved
com-pared with masculinization (see also Hattori, Gould,
Fujioka, Saito, Kurita, Strˇssmann, Yokota &
Wata-nabe 2007; Hattori, et al 2009; Yamaguchi,
Yamagu-chi, Hirai & Kitano 2007; Fernandino, Hattori,
Kimura, Strˇssmann & Somoza 2008) A notable
ex-ception seems to be one of the two species used in this
study, O bonariensis, where all-female populations
can be obtained easily, but its usefulness as an
experi-mental material is limited due to the absence of a
marker for distinguishing genetic sex In this
con-text, the hybrids of these two species could provide
useful materials to study the process of feminization
under low temperatures
This study was not able to ¢nd suitable thermal
treatments for the production of female and
all-male hybrid o¡spring However, we suggest that
ther-mal treatments below or above the range applied in
this study could be attempted to increase the sex
re-versal rates because the survival rates were not
sig-ni¢cantly a¡ected at the extreme temperatures
Furthermore, the absence of body deformations in
hybrids at temperatures of 25 and 29 1C, which arecommon in purebred Patagonian pejerrey and pejer-rey larvae, is an indication that hybrids tolerate ahigher thermal range compared with the parentalspecies Alternatively, treatments starting beforehatching may also be tested because recent studies
in medaka (Hattori et al 2007), Nile tilapia (Rougeota,Prignon,Valence, Kengneb & Me¤lard 2008) and in theparental species O hatcheri (unpublished results)show that such treatments can lead to increasedrates of sex reversal
Although the current histological examination didnot include O bonariensis and O hatcheri controls(purebreds), their histology is known from the avail-able literature (Strˇssmann, Moriyama et al 1996;Strˇssmann, Calsina Cota et al.1996; Strˇssmann,Ta-kashima et al 1996; Strˇssmann, Saito et al 1997).The ovaries and testes of the hybrids, with the nota-ble exception of the Y-bearing Obo, Oha< femalesexamined 3 months after hatching, were considered
to be normal The arrested ovarian development served in the latter resembled histologically the ster-ile gonads of triploid ¢sh (Strˇssmann et al 1993).However, the histological analysis conducted afterthe animals became sub-adults could no longer re-veal any di¡erences between the ovaries of femalescarrying the Oha X and Y chromosomes and thepurebred controls These ¢ndings and the presence
ob-of milt in testes ob-of the sub-adults suggest that bothreciprocal hybrids are fertile but progeny tests should
be performed to con¢rm their capacity to generate able o¡spring Nevertheless, they also indicate that agene/factor present on the Oha Y chromosome andthat con£icts with ovarian di¡erentiation is tempora-rily expressed during early gonadogenesis in thehybrids
vi-In conclusion, the present study revealed tant interactions between temperature and genotypeduring gonadal sex determination of the interspeci¢chybrids Experiments exploring di¡erent treatmenttimings and temperatures may lead to increased sexreversal rates The ¢tness of hybrids, especially the fe-males with arrested ovarian di¡erentiation, shouldalso be examined and could contribute to the devel-opment of atheriniculture
impor-AcknowledgmentsThis work was supported in part by Grants-in-Aidfrom the Ministry of Education, Culture, Sports,Science and Technology of Japan and Tokyo Univer-sity of Marine Science and Technology to CAS
Aquaculture Research, 2011, 42, 746^753 Sex determination in Odontesthes hybrids J Inazawa et al.
Trang 17Chevassus B (1983) Hybridization in ¢sh Aquaculture 33,
245^262.
Cnaani A & Levavi-Sivan B (2009) Sexual development in
¢sh, practical applications for aquaculture Sexual
Devel-opment 3, 164^175.
Devlin R.H & Nagahama Y (2002) Sex determination and
sex di¡erentiation in ¢sh: an overview of genetic,
physio-logical, and environmental in£uences Aquaculture 208,
191^364.
Eicher E.M & Washburn L.L (2001) Does one gene
deter-mine whether a C57BL/6J-YPOS mouse will develop as a
female or as a hermaphrodite? Journal of Experimental
Zo-ology 290, 322^326.
Eicher E.M., Washburn L.L., Whitney J.B & Morrow K.E.
(1982) Mus poschiavinus Y chromosome in the C57BL/6J
murine genome causes sex reversal Science 217,535^537.
Fernandino J.I., Hattori R.S., Kimura H., Strˇssmann C.A &
Somoza G.M (2008) Dimorphic expression of dmrt1 and
cyp19a1 (ovarian aromatase) during early gonadal
devel-opment in pejerrey, Odontesthes bonariensis Sexual
Devel-opment 2, 316^324.
Hamaguchi S & Sakaizumi M (1992) Sexually di¡erentiated
mechanisms of sterility in interspeci¢c hybrids between
Oryzias latipes and O curvinotus Journal of Experimental
Zoology 263, 323^329.
Hattori R.S., Gould R.J., Fujioka T., Saito T., Kurita J.,
Strˇss-mann C.A.,Yokota M & Watanabe S (2007) Thermolabile
sex determination in Hd-rR medaka Oryzias latipes:
gen-der sensitivity, thermal threshold, critical period, and
DMRT1 expression pro¢le Sexual Development 1,138^146.
Hattori R.S., Fernandino J.I., Kishii A., Kimura H., Kinno T.,
Oura M., Somoza G.M,Yokota M., Strˇssmann C.A &
Wa-tanabe S (2009) Cortisol-induced masculinization: does
thermal stress a¡ect Gonadal Fate in Pejerrey, a Teleost
¢sh with temperature-dependent sex determination?
PLoS ONE 4, e6548.
Hattori R.S., Oura M., Yokota M., Strˇssmann C.A &
Wata-nabe S (2010) Establishment of a strain with a sex-linked
SNP marker in Odontesthes hatcheri, a species with
ther-molabile sex determination Animal Genetics 41, 81^84.
Ito L.S., Yamashita M., Takashima F & Strˇssmann C.A.
(2005) Dynamics and histological characteristics of
gona-dal sex di¡erentiation in pejerrey (Odontesthes bonariensis)
at feminizing and masculinizing temperatures Journal of
Experimental Zoology A (Comparative and Experimental
Biology) 303, 504^514.
Kato M., Takehana Y., Sakaizumi M & Hamaguchi S (2010)
A sex-determining region on the Y chromosome controls
the sex-reversal ratio in interespeci¢c hybrids between
Oryzias curvinotus females and Oryzias latipes males
Her-edity 104, 191^195.
Koshimizu E., Strˇssmann C.A., Okamoto N., Fukuda H &
Sakamoto T (2010) Construction of a genetic map and
development of DNA markers linked to the sex
determin-ing locus in the Patagonian pejerrey (Odontesthes eri) Marine Biotechnology 12, 8^13.
hatch-Matsuda M., Nagahama Y., Shinomiya A., Sato T., hatch-Matsuda C., Kobayashi T., Morrey C.E., Shibata N., Asakawa S., Shi- mizu N., Hori H., Hamaguchi S & Sakaizumi M (2002) DMY is aY-speci¢c DM-domain gene required for male de- velopment in the medaka ¢sh Nature 417, 559^563 Matsuda M., SatoT.,ToyazakiY., NagahamaY., Hamaguchi S.
& Sakaizumi M (2003) Oryzias curvinotus has DMY, a gene that is required for male development in the medaka,
O latipes Zoological Science 20, 159^161.
Ospina-AŁlvarez N & Piferrer F (2008) dent in ¢sh revisited: prevalence, a single sex ratio re- sponse pattern, and possible e¡ects of climate change PLoS ONE 3, e2837.
Temperature-depen-Pala I., Klˇver N., Thorsteinsdo¤ttir S., Schartl M & Coelho M.M (2008) Expression pattern of anti-Mˇllerian hor- mone (amh) in the hybrid ¢sh complex of Squalius albur- noides Gene 410, 249^258.
Pala I., Schartl M.,Thorsteinsdo¤ttir S & Coelho M.M (2009) Sex determination in the Squalius alburnoides complex: an initial characterization of sex cascade elements in the context of a hybrid polyploid genome PLoS ONE 4, e6401 Rougeota C., Prignon C.,Valence C., Kengneb N & Me¤lard C (2008) E¡ect of high temperature during embryogenesis
on the sex di¡erentiation process in the Nile tilapia, chromis niloticus Aquaculture 276, 205^208.
Oreo-Shinomiya A., Kato M.,Yaezawa M., Sakaizumi M & guchi S (2006) Interespeci¢c hybridization between Ory- zias latipes and Oryzias curvinotus causes XY sex reversal Journal of Experimental Zoology 305A, 890^896 Somoza G.M., Miranda L.A., Berasain G.E., Colautti D., Leni- cov M.R & Strˇssmann C.A (2008) Historical aspects, current status and prospects of pejerrey aquaculture in South America Aquaculture Research 39,784^793 Strˇssmann C.A & Yasuda N (2005) Biology and aquacul- ture of pejerrey In: Freshwater Aquaculture Systems (ed.
Hama-by F Takashima & M Murai), pp 259^270 Koseisha kaku,Tokyo, Japan (in Japanese).
Kosei-Strˇssmann C.A., Ng B.C., Oshiro T & Takashima F (1993) Triploidy induction in an atherinid ¢sh, the pejerrey Odontesthes bonariensis The Progressive Fish-Culturist 55, 83^89.
Strˇssmann C.A., Calsina Cota J.C., Pronlor G., Higuchi H & Takashima F (1996) Temperature e¡ects on sex di¡eren- tiation of two South American atherinids, Odontesthes ar- gentinensis and Patagonina hatcheri Environmental Biology of Fishes 47, 143^154.
Strˇssmann C.A., Moriyama S., Hanke E.F., Calsina Cota J.C.
& Takashima F (1996) Evidence of thermolabile sex mination in pejerrey Journal of Fish Biology 48, 643^651 Strˇssmann C.A., Takashima F & Toda K (1996) Sex di¡er- entiation and hormonal feminization in pejerrey Odon- testhes bonariensis Aquaculture 139, 31^45.
deter-Strˇssmann C.A., Saito T., Usui M.,Yamada H & Takashima
F (1997) Thermal thresholds and critical period of Sex determination in Odontesthes hybrids J Inazawa et al Aquaculture Research, 2011, 42, 746^753
Trang 18ther-molabile sex determination in two atherinid ¢shes,
Odon-testhes bonariensis and Patagonina hatcheri Journal of
Ex-perimental Zoology 278, 167^177.
Strˇssmann C.A., Ijima K.,Yamaguchi K.,Yoshizaki G &
Ta-kashima F (1997) Spontaneous hybridization in the
la-boratory and genetic markers for the identi¢cation of
hybrids between two atherinid species, Odontesthes
bonar-iensis (Valenciennes,1835) and Patagonina hatcheri
(Eigen-mann, 1909) Aquaculture Research 28, 291^300.
Strˇssmann C.A., Karube M., Miranda L.A., Patino R.,
So-moza G.M., Uchida D & Yamashita M (2004) Methods of
sex control in ¢shes and an overview of novel hypotheses
concerning the mechanisms of sex di¡erentiation In:
Fish Genetics and Aquaculture Biotechnology (ed by T.J.
Pandian, C.A Strˇssmann & M.P Marian), pp 65^79 Science Publishers, New Hampshire, USA.
Taranger G.L., Carrillo M., Schulz R.W., Fontaine P., Zanuy S., Felip A.,Weltzien F., Dufour S., Karlsen ., Norberg B., An- dersson E & Hansen T (2010) Control of puberty in farmed
¢sh General and Comparative Endocrinology 165, 483^515 Wohlfarth G.W (2008) The unexploited potential of tilapia hybrids in aquaculture Aquaculture Research 25,781^788 Yamaguchi T.,Yamaguchi S., Hirai T & Kitano T (2007) Fol- licle-stimulating hormone signaling and Foxl2 are in- volved in transcriptional regulation of aromatase gene during gonadal sex di¡erentiation in Japanese £ounder, Paralichthys olivaceus Biochemical and Biophysical Re- search Communications 359, 935^940.
Aquaculture Research, 2011, 42, 746^753 Sex determination in Odontesthes hybrids J Inazawa et al.
Trang 19Growth and feeding of Patagonian pejerrey
Juan Pablo Hualde1, Walter DamiaŁn Ceferino Torres1, Pablo Moreno1, Mirna Ferrada2, Mariela AnaDemicheli1, Leonardo Javier Molinari1and Carlos Marcelo Luquet3
1 Centro de Ecolog|¤a Aplicada de Neuque¤n, Jun|¤n de los Andes, Neuque¤n, Argentina
2 Centro Regional Universitario Bariloche, Universidad Nacional del Comahue, Bariloche, Argentina
3 Instituto de Investigaciones en Biodiversidad y Medio Ambiente, Consejo Nacional de Investigacio¤n Cient|¤¢ca y Tecnolo¤gica, Universidad Nacional del Comahue, Bariloche, Argentina
Correspondence: J P Hualde, Centro de Ecolog|¤a Aplicada del Neuque¤n, PO Box 7, Jun|¤n de los Andes (8371), Neuque¤n, Argentina E-mail: pablohualde@yahoo.com.ar
Abstract
An experiment was performed to evaluate the
perfor-mance of Patagonian pejerrey during net cage
rear-ing in the oligomesotrophic reservoir Exequiel
Ramos Mex|¤a Survival, growth, nutrition and
repro-duction were evaluated for two lots of pejerrey, initial
weights 4.0 and 2.5 g, reared in net cages for 22 and
14 months respectively Fish were stocked at 29 and
48 individuals m 3densities and fed with an
experi-mental pejerrey feed During the experiment,
tem-perature £uctuated between 6.6 and 19.6 1C The two
lots did not show substantial di¡erences in survival
(pooled survival at the end of the experiment
480%) and the thermal-unit growth coe⁄cient
(TGC), daily feed intake and feeding e⁄ciency were
0.43 0.19, 1.70 0.80 and 53.6 9.9 respectively
Digestive tract analysis showed that caged pejerrey
can consume substantial quantities of natural food,
taking advantage of its planktivorous condition
Pe-jerrey showed high percentage survival, slow growth
and early sexual maturation in captivity The use of
the TGC is proposed as a model for describing the
growth pattern of this species and other pejerrey
un-der culture conditions
Keywords
Patagonian pejerrey, net cage, Odontesthes hatcheri,
aquaculture, atheriniculture
Introduction
Aquaculture is a constantly growing activity, which
constitutes an important source of food of excellent
quality A signi¢cant part of this production relies
on carnivorous ¢sh and crustaceans, making ¢shmeal and ¢sh oil dominant ingredients in the com-pound feeds utilized As these components aremainly obtained from marine ¢sheries, with most re-sources being over-exploited, alternative productionsystems and/or alternative sources of suitable nutri-ents must be developed Thus, the farming of lowtrophic-level ¢sh, which use a higher proportion ofvegetal nutrients sources, is recognized as a goal forsustainable expansion of world aquaculture (Naylor,Goldburg, Primavera, Kautzky, Beveridge, Clay, Folke,Lubchenco, Mooney & Troell 2000) The use of nativespecies in aquaculture is a worldwide trend in thesearch of alternatives for diversi¢cation of regionalproductions and more e⁄cient utilization of availableresources From a biodiversity conservation view-point, farming native species could also help avoidnegative impacts from introductions of exotic species(Ross, Martinez Palacios & Morales 2008)
Patagonian pejerrey belongs to a group of nopsid ¢sh (silversides) widely spread in inland andmarine waters of the Americas Among the fresh-water native ¢sh of Argentina, the Patagonian pejer-rey (Odontesthes hatcheri) and the more popularOdontesthes bonariensis are two related species con-sidered to be candidates for aquaculture Both spe-cies provide a tasty £esh of excellent quality, which
atheri-is highly appreciated both in the local and in the ternational markets (Somoza, Miranda, Berasain, Co-lautti, Remes Lenicov & Strˇssmann 2008) In fact, O.bonariensis, which is naturally abundant in riversand shallow lakes of the Pampasic region (mainly in
in-Aquaculture Research, 2011, 42, 754^763 doi:10.1111/j.1365-2109.2011.02827.x
Trang 20Buenos Aires), has been successfully introduced for
recreational ¢shing in hydroelectric reservoirs and
lakes of the center, west and north-west of the
coun-try It has also been introduced to other South
Amer-ican countries like Uruguay, the South of Brazil,
Chile, Bolivia and Peru (see Somoza et al 2008, for a
review) Odontesthes hatcheri inhabits Patagonian
rivers and lakes from the R|¤o Colorado basin,
3815903600S; 6410504100W to R|¤o Senguer 451300S;
601W (Cussac, Cervellini & Battini 1992; Dyer 1993)
The extensive rearing of O bonariensis has been
widely practiced in Argentina, Uruguay and South
of Brazil, since the beginning of the 20th century,
while intensive culture of pejerrey has been
devel-oped successfully in Japan during the second half of
the century (see Somoza et al 2008 for a review) In
recent years, many researchers have worked on the
development of culture methods for several
Atheri-nopsidae species (Reartes 1995; Berasain, Colautti &
Velasco 2000; Mart|¤nez Palacios, Racotta, R|¤os
Dur-aŁn, Palacios, Toledo Cuevas & Ross 2006; Miranda,
Berasain, Velasco, Shirojo & Somoza 2006; Orellana
& Toledo 2007) In contrast, the intensive culture of
the Patagonian pejerrey has not yet been evaluated
The Comahue region in north-west Patagonia
pos-sesses seven hydroelectric reservoirs, located on the
rivers Limay and Neuque¤n, covering a surface area
of 1220 km2 The calculated carrying capacity for
aquaculture of the three largest reservoirs of the
Li-may River is 14 900 tonnes year 1(Wicki & Lucchini
2002) However, of these reservoirs, only AlicuraŁ, a
67 km2hydroelectric reservoir on the Limay River, is
exploited at present, with eight ¢sh farms producing
1400 tonnes year 1of rainbow trout in net cages
In an important part of the seven reservoirs
men-tioned above, the summer water temperatures
ex-ceed 18 1C, the upper limit of the optimal growth
temperature range for rainbow trout (Goddard
1996) O hatcheri is a good candidate for aquaculture
in this region; because it is abundant in all the
avail-able reservoirs, and it has great acceptance in
regio-nal and internatioregio-nal markets Because pejerrey are
planktivorous ¢sh, they are able to feed on natural
re-sources when con¢ned in net cages (Colautti, Garcia
de Souza, Balboni & Baigu¤n 2010) This could be an
advantageous feature considering that natural food
can complement the arti¢cial feed in quantity and in
nutrient quality
The aim of this study was to generate information
on the suitability of Patagonian pejerrey for net cage
rearing in a typical Patagonian lentic environment,
the Ramos Mex|¤a reservoir For this purpose, we
examined the survival, growth, nutrition and duction of ¢sh kept for a period of 14^22 months innet cages, where they were fed an arti¢cial diet butalso had access to natural food sources The diet ofreared pejerrey was compared with that of conspeci-
repro-¢c wild ¢sh captured in the same reservoir The use ofthe thermal-unit growth coe⁄cient (TGC), as a modelfor describing the growth pattern of O hatcheri andother pejerrey species, is proposed in this work in-stead of the speci¢c growth rate (SGR)
Materials and methodsStudy site, environmental conditions andrearing cages
Ramos Mexia reservoir is situated in the north of thePatagonian steppe, covering an area of about of
816 km2in the Limay River Valley, with a total watervolume of 20150 hm3 The maximum depth of thisreservoir is 64 m (average 24 m) and it is classi¢ed asoligomesotrophic (Wicki & Lucchini 2002) Duringthis study, the temperature, dissolved oxygen andtransparency were measured daily using a multi-function oxygen metre and a Secchi disc Total phos-phorus and nitrogen were measured seasonallyusing spectrophotometric methods (Greenberg,Clesceri & Eaton 1992)
The net cages had 6 m length 3 m width 4 mheight, holding approximately 50 m3of water, andwere constructed with a 5 mm mesh net The twocages were supported by a 6 6 m raft moored
100 m o¡shore in a bay 17 m deep The nets werecleaned weekly in fall, spring and summer andmonthly in winter
Source of ¢sh and rearing conditionsPatagonian pejerrey were obtained from a 5000 m2arti¢cial pond located in the CEAN facility (Jun|¤n delos Andes Neuque¤n, Argentina) on two occasions andstocked temporarily in a circular 1000 L tank sup-plied with abundant water from the Chimehuin Riverfor about 2 months, for observation before transfer tothe cages During this period, the ¢sh were fed an ex-perimental pejerrey feed formulated and prelimina-rily elaborated in our laboratory; see the descriptionbelow
A ¢rst lot of 2400 juveniles weighing 4.14 1.3 g(mean SD) (lot 1) was transferred to one rearingcage on March 2006 A second lot of 4500 juveniles
Aquaculture Research, 2011, 42, 754^763 Growth of Patagonian pejerrey in net cages J P Hualde et al.
Trang 21weighing 2.4 1.3 g (lot 2) was transferred to the
other cage on May 2006 Fish were transported in a
500 L tank with a constant O2bubbling and a
sali-nity of 5 g L 1NaCl (Tsuzuki, Ogawa, Strussmann,
Maita & Takashima 2001) Lots 1 and 2 were reared
in the experimental cages for 22 and 14 months
re-spectively During this period, the ¢sh were hand fed
commercially produced experimental pejerrey feed
(see the following section) twice a day to satiation, 6
days a week Occasionally, one or both daily meals
had to be cancelled due to strong winds Fish in lot 1
were fed a commercial feed for trout (see below)
dur-ing the ¢rst 50 days of the experiment due to a delay
in the production of the experimental feed for pejerrey
Both lots of ¢sh also had access to naturally available
food (plankton) that passed through the net cages
Preparation of experimental feed
An experimental feed for pejerrey was formulated in
our laboratory and order made by a commercial feed
manufacturer (Molino Don Antonio SA, General
Pico, La Pampa, Argentina) Feed was steamed at
80 1C, pelletized and then dried in a hot air column
Pelletized commercial trout feed was purchased from
the same company
The composition of the experimental feed was
de-termined as follows: total nitrogen (TN) was
deter-mined using the Semi-micro Kjeldahl method (AOAC
1990); crude protein was estimated as 6.25 TN
Crude lipid was measured gravimetrically following
extraction of 1g samples in sulphuric ether, using a
Soxhlet apparatus Moisture was measured
gravime-trically, after drying in an oven at 105 1C for 3 h, and
ash by combustion in a mu¥e at 550 1C for 6 h Total
phosphorous was assayed by wet digestion with
HNO31HClO4and reaction with ascorbic acid
Solu-ble phosphorous (fractionated with deionized water)
was determined according to Satoh,Viyakarn,
Yama-zaky, Takeuchi and Watanabe (1992) Nitrogen-free
extract was calculated by di¡erence (100 crude
protein crude lipid ash moisture content)
Gross energy was calculated at 23.6, 39.5 and
17.2 kJ g 1of protein, lipid and carbohydrate
respec-tively (NRC 1993) The composition of the two feeds
utilized in the experiment is shown in Table 1
Estimation of growth and survival
During the experimental rearing, both groups of ¢sh
were monitored and sampled at 29^120-day
inter-vals Groups of 90^120 ¢sh were anaesthetized with
100 ppm benzocaine and weighed with an electronicbalance to the nearest 0.1g Feeding e⁄ciency (FE)was calculated as 100 weight gain (kg)/feed intake(kg) Daily feed intake (DFI) was calculated as
100 feed intake (kg)/(average body weight day).Speci¢c growth rate was (lnW2^lnW1) day 1 Ther-mal-unit growth coe⁄cient was calculated accord-ing to the following formula (Iwama & Tautz 1981;Cho 1992):
TGC¼ 1000ðW21=3 W11=3Þ=ðT DÞwhere W2 is the ¢nal weight (g), W1is the initialweight (g), T is the mean temperature for the period( 1C) and D is the number of days between measure-ments
Mortality was monitored daily, at the time of ing, and data were grouped to ¢t in the same schedule
feed-of sampling as for growth
Determination of experimental feed ingestionand natural diet
Caged pejerrey were sampled starting at spring 2006,
on several occasions throughout the year, except inwinter, in which only one sampling could be made
Table 1 Ingredients and proximate composition of the diets used in this study
Ingredients (g kg 1) Pejerrey feed Trout feed
^, not indicated in the commercial formula.
Growth of Patagonian pejerrey in net cages J P Hualde et al Aquaculture Research, 2011, 42, 754^763
Trang 22(see Table 3) for determination of the amount of
ex-perimental feed and natural prey ingested Fish were
sampled at noon, about 3 h after a meal of the
experi-mental diet, and kept in ice until analysis Digestive
tracts were dissected and cut open Gastric content
analysis was performed under a dissection
micro-scope or a light micromicro-scope Alimentary components
were classi¢ed at the lowest possible taxonomic level
Individuals corresponding to each component were
counted and weighed to the nearest 0.1mg Chitin
re-mains, which could not be identi¢ed, were classi¢ed
as unidenti¢ed Arthropoda (UA) Commercial feed
was counted as an alimentary component and
trea-ted in the same fashion as the natural components
Algae and Cyanobacteria were classi¢ed and
de-scribed but were not considered for the calculationsdescribed below
Percentage observed frequency: FO% 5 number ofgastric ducts in which a component was present/total number of gastric ducts analysed
Alimentary index (AI) (Lauzanne 1975; Rosecchi &Nouaze 1987)
AI¼ ðFO% FW%Þ=100where FW% 5 (fresh weight of a component /totalfresh weight of the corresponding gastric con-tent) 100
Assessment of the reproductive statusSamples to determine the reproductive status of thecaged ¢sh were taken on a monthly basis, from Sep-tember 2006 to July 2007, except in June The stage ofgonadal development of females was evaluatedmacroscopically according to the criteria for thisgroup of ¢sh (Grosman 1995) Brie£y, seven cate-gories (I^VII) were considered: I (virginal): the go-nads show no evidence of past or present activity; II(preparation): gonads show functionality but oocytesare not macroscopically seen; III (maturation): gonadsize increased, oocytes are evident; IV (pre-spawn-ing): gonads size highly increased with granulose as-pect but spawning cannot be induced by hand; V(spawning) oocytes are clustered around chorionic ¢-laments and are readily released at touch; VI (post-spawning): gonads show a haemorrhagic aspect,isolated and voluminous oocytes and also immatureoocytes; and VII (regression): evidence of past spawn-
Table 3 Diet composition of Odontesthes hatcheri reared in net cages in Ramos Mex|¤a reservoir
Season Dietary item Alimentary index Observed frequency (%)
Number of samplings per season is shown between parentheses.
UA, unidenti¢ed arthropods.
Table 2 Growth and survival of two lots of Patagonian
pe-jerrey, Odontesthes hatcheri, reared in net cages in Ramos
TGC, thermal-unit growth coe⁄cient; SGR, speci¢c growth rate;
FE, feeding e⁄ciency; DFI, daily feed intake; ND, not determined.
Aquaculture Research, 2011, 42, 754^763 Growth of Patagonian pejerrey in net cages J P Hualde et al.
Trang 23ing, gonads dark and reduced The weight increment
in male and female gonads was evaluated through
the gonadosomatic index (GSI), de¢ned as gonad
fresh weight/body fresh weight 100
Statistical analyses
Statistical analyses were performed using BIOSTAT
program The Mann^Whitney non-parametric test
was used to compare TGC and SGR, between
treat-ment groups Results are expressed as mean SD
Results
Environmental conditions
Transparency, measured as Secchi depth, varied
be-tween 2.7 and 9 m according mostly to the
precipita-tion regime and in a lesser degree to plankton
productivity The mean total phosphorous and
nitro-gen were 18.3 6.1 and 67.8 5.7 mg L 1
respec-tively The mean pH, conductivity and hardness were
7.0 0.3, 60 4.2 mS and 48.2 9.0 mg L 1
(CaCO3) Figure 1 shows the average monthly
tem-perature during the study It can be noticed that
be-tween January and March, the water temperature in
Ramos Mex|¤a reservoir remains close to 20 1C
Growth and survival
Patagonian pejerrey lots 1 and 2 were farmed for 22
and 14 months, attaining ¢nal mean weights of
120.0 39 and 37.3 14.1g respectively These ¢sh
showed rapid adaptation to the culture conditions,
readily accepting arti¢cial diets, and showing good
tolerance to disturbing factors such as boat engine
noise or movement of workers around the pen
How-ever, interventions in the culture facility, e.g net
cleaning or removal of dead individuals, producedvisible changes in ¢sh behaviour such as rapid swim-ming, sudden movements and loss of appetite Whenfeed was o¡ered at a low rate, concentrated on a spe-ci¢c point near the centre of the cage, the ¢sh tended
to cluster, forming a dense column from 30^50 cmbelow the surface to near the bottom, showing thisbehaviour as long as feed was delivered Satiationwas presumed when ¢sh were scattered and swamtowards the deepest part of the cage Whenever theweather allowed us to feed the ¢sh in the describedmanner, there was almost no feed loss
The number of ¢sh utilized, the initial and ¢nalweights and the duration of the experiment are sum-marized in Table 2 Mann^Whitney analysis indi-cated that TGC calculated for pejerrey lots 1 and 2did not di¡er signi¢cantly (P 5 0.2781, U 510.847).The pooled mean TGC was 0.43 0.19, coe⁄cient ofvariation (CV) 5 44% Speci¢c growth rate did notdi¡er signi¢cantly between lots either (P 5 0.1407,
U 51.4731) but showed a higher CV of 65% Thepooled SGR was 1.34 0.87 Figure 2 shows thegrowth of pejerrey plotted against time In Fig 3, theTGC and SGR values obtained for ¢sh of both groupsare plotted against body weight
The DFI of Patagonian pejerrey was 1.70 0.80%and FE was 53.6 9.9%
Lot 1 pejerrey was fed trout diet during a short
peri-od (50 days) at the start of the experiment in cages.Mortality during this period gradually increased up to0.41% day 1, and decreased to 0.024% day 1whenfeed was replaced by the pejerrey experimental diet.All dead ¢sh showed deformations in the spine Lot 2,which was fed only with the pejerrey experimentaldiet, showed 0.01% mortality day 1and did not showany sign of malformation or change in the mortalityrate over time The percentage survival of both lots forthe ¢rst 240 days of the study is shown in Fig 4 It can
be noticed that after the ¢rst 50 days, when the trout
reservoir during the experiment
Lot 1
Lot 2
0 20 40 60 80 100 120 140
DateFigure 2 Growth of two groups of Patagonian pejerrey
in net cages in the Ramos Mex|¤a reservoir
Growth of Patagonian pejerrey in net cages J P Hualde et al Aquaculture Research, 2011, 42, 754^763
Trang 24feed supplied to lot 1 was replaced for the experimental
pejerrey feed, mortality was similar between both lots
Since day 240, mortality data from both lots were
pooled and revealed a high percentage survival
throughout the experiment The survival rate, pooling
both lots, at the end of the experiment was 87.81%
Experimental feed ingestion and natural diet
Fish actively ingested the arti¢cial feed throughout
the year but the proportion it represented in the diet
decreased sharply in fall and winter The digestive
tract content of caged pejerrey was heterogeneous
Besides the commercial feed, planktonic crustaceans
accounted for four alimentary components and
in-sects for 6 Additionally, ¢lamentous algae of the
genus Spirogyra sp and phytoplankton were
com-mon in ¢sh samples In summer, Spirogyra sp was
present in 76% of the digestive tracts In fall (March),
other algal genera such as Ulotrix sp., Gonium sp and
Coenococcus sp were also identi¢ed In April, 60% of
the samples contained Cyanobacteria of the genus
Anabaena sp
Because of calculation problems, the AI refers only
to natural components of animal origin plus mercial feed In spring and summer, the dominantcomponent was the commercial feed, with AI of68.86 and 52.11 respectively Another importantcomponent in this period was UA (chitin plaques,whose origin could not be identi¢ed) In fall, the high-est AI (36.99) corresponded to UA Crustacea was thesecond most abundant component, with Cladocera
com-as the most representative group (AI 510.16) In ter, there was only one sampling event in July, inwhich no commercial feed was detected in the diges-tive tracts It must be noticed that this sampling wasperformed after 5 days during which it was not pos-sible to bring feed to the ¢sh UA (AI 5 68.89) and Cla-docera (AI 519.06) were the most importantcomponents in this season Insects were also present
win-in the diet but win-in a smaller proportion Particularly,Diptera larvae were part of the diet in every season,except in fall but always as a rare (AIo10) compo-nent See Table 3 for detailed results
The trophic diversity index (Shannon^Wienner, H)was 0.91 and de 0.96 in spring and summer respec-tively Fall and winter showed much smaller H values,0.16 and 0.41 respectively For comparison, a singlesample of wild Patagonian pejerrey was taken withtrawl nets in the coastal zone of the reservoir, nearthe net cages, in May 2007 The weight of the cap-tured ¢sh ranged from 1.81 to 5.60 g Cladocera werethe dominant component in the diet (AI 5 56), whileDiptera larvae were present in 75% of the digestivetracts, with an AI of 9 The Shannon^Wiener index
of diversity of this sample of wild pejerrey was 0.98
Reproductive statusThe male:female ratio was about1in both lot1and lot
2 of Patagonian pejerrey In spring, 71% of the males reached the pre-spawning stage (IV), with amean body weight of 18.54 g, while in the followingwinter, 100% of the females reached the same stage,weighing 47.07 g The smallest female from lot 2reaching stage IV had 11.5 g body weight and a GSI
fe-of 6.92 The mean body weight recorded for stage IVmales in spring 2006 and winter 2007 was 15.60 and40.70 g respectively Figures 5 and 6 shows the GSI ofmale and female O hatcheri, respectively, from Sep-tember 2006 (age 01) to July 2007 (age 11) In the
¢rst year, peak GSI values were reached in September
by males of both lots and in September and ber by females of lot 1 and lot 2 respectively In the
Figure 3 Relationship between body weight and two
growth descriptors, SGR and TGC, in Patagonian pejerrey
reared in net cages SGR, speci¢c growth rate; TGC,
ther-mal-unit growth coe⁄cient
Lot 1 Lot 2
net cages in the Ramos Mex|¤a reservoir
Aquaculture Research, 2011, 42, 754^763 Growth of Patagonian pejerrey in net cages J P Hualde et al.
Trang 25second year, both sexes had increasing GSI values
al-ready in July when the experiment was terminated
Discussion
The adaptation of the Patagonian pejerrey to cage
farming in Ramos Mex|¤a reservoir was tested
through di¡erent indicators of growth, feed use,
sur-vival and reproductive status
To describe the growth pattern of O hatcheri, two
mathematical models were used in this work The
most widely used model in ¢sh studies is the
instan-taneous growth rate or SGR, based on the natural
logarithm of body weight Despite its wide use, SGR
is recognized as an inappropriate growth model for
¢sh because it decreases with ¢sh size and with thelength of the time interval used in the calculation(Iwama & Tautz 1981; Cho 1992) The TGC model wasproposed by Iwama and Tautz (1981), and has beenshown to well represent the growth curves of severalsalmonid species Subsequently, the equation hasalso been used to describe growth in non-salmonidspecies, such as common carp, Nile tilapia and mar-ine ¢sh (Kaushik 1998)
One major advantage of TGC is that at a given perature, it is independent of body weight In thiswork, TGC and SGR for Patagonian pejerrey havebeen calculated for di¡erent size ranges, tempera-
Figure 5 Gonadosomatic index of male Patagonian pejerrey reared in net cages in Ramos Mex|¤a reservoir
Figure 6 Gonadosomatic index of female Patagonian pejerrey reared in net cages in Ramos Mex|¤a reservoir
Growth of Patagonian pejerrey in net cages J P Hualde et al Aquaculture Research, 2011, 42, 754^763
Trang 26tures and time periods Figure 4 shows the
relation-ship between body weight and TGC and SGR;
whereas SGR decreases with increasing body weight,
TGC remains almost constant These results suggest
that TGC rather than SGR should be used to describe
the growth pattern of Patagonian pejerrey
For comparison, we have calculated, from
pub-lished growth data (Orellana & Toledo 2007; Velasco,
Berasain & Ohashi 2008), the TGCs for two other
pe-jerrey species: O regia and O bonariensis (Table 4)
Even when the three species have been reared at
dif-ferent temperatures, salinities and culture
condi-tions, the TGC values are similar The growth
coe⁄cients of these pejerrey species are three to four
fold smaller than those of rainbow trout and other
salmonid species, but fairly close to those reported
for marine species used in aquaculture, which have
TGC values ranging from 0.6 to 1.0 (Kaushik 1998)
The growth rates of Patagonian pejerrey are low
The TGC model predicts that, under the experimental
conditions of this work, Patagonian pejerrey reach
harvest size (220^250 g) after 28 months Poor
growth rate has been considered to be an important
biological constraint to cultivation of the pejerrey O
bonariensis (Somoza et al 2008) These authors report
that the time needed for O bonariensis to reach
har-vest size ranges from 1.5 to 2.5 years
Sexual precocity may also have adversely a¡ected
both the growth rate and the feed e⁄ciency in the
present study, because 70% of females matured
sexu-ally at the age of 01 with about 18 g weight Slow
growth and early maturation are common
character-istics of pejerrey, and constrain their suitability for
farming (Strussmann, Choon, Takashima & Oshiro
1993) Accordingly,Toda,Toshinami,Yasuda and
Sus-uki (1998) have reported that O bonariensis reaches
sexual maturity at the second year of age and
excep-tionally at the ¢rst year
Another constraint for pejerrey aquaculture
re-ported by several authors in early works reviewed by
Somoza et al (2008) is the low survival rate of this
¢sh This problem was solved for O bonariensis
dur-ing the last decades of the 20th century in part by
adding 3^5 g L 1NaCl to the water during ¢sh nipulations and acclimatization (Tsuzuki et al 2001).Following this recommendation, we have trans-ported the ¢sh 300 km to the aquaculture facilitywith negligible mortality and the overall survivalrate was high
ma-The observed mortality in lot 1 at the beginning ofthe cage farming cycle could be associated with thelow availability of phosphorus in the trout diet usedduring this period Preliminary results from our la-boratory show that diets poor in soluble phosphorusproduce de¢ciency signs, such as spine demineraliza-tion and malformation in Patagonian pejerrey, sug-gesting a poor absorption capacity for this mineral.The natural diet study has shown that cage-reared
O hatcheri is able to obtain food from the ment, consuming in the highest proportion the princi-pal diet components consumed by wild individuals ofthe same species A similar preference for cladoceransand copepods, supplemented by insects when thesecrustaceans groups were not abundant, was reportedfor O bonariensis in Japanese lakes (Toda et al 1998).Colautti et al (2010) have demonstrated that juvenile
environ-O bonariensis can be reared successfully in net cages
in a eutrophic lake without the addition of arti¢cialfeed The plankton productivity of the oligomeso-trophic north Patagonian reservoirs is unlikely to sup-port such extensive ¢sh production However, we haveshown that natural diet can replace part of the com-mercial feed in the intensive system, which may be ad-vantageous for economic and environmental reasons.The reduction and even the absence of commercialfeed in the gastric ducts of caged pejerrey sampled infall and winter reveals a failure in the feeding method,associated with adverse climatic conditions Manualfeeding was certainly complicated or even impossibleduring the windiest days Under such conditions,when little or no arti¢cial feed could be delivered, theuptake of natural food probably supported ¢shgrowth but at a low rate Additionally, these di⁄cul-ties in delivering feed in the slow but steady mannerrequired by this species probably led to feed wastage.This, together with the lack of nutritional information
to produce an adequate feed, can probably explain thelow FE recorded in this work (about 53%)
ConclusionsThe Patagonian pejerrey shows good adaptation tonet cage farming conditions, su¡ering low mortality.Like other Atherinopsids, O hatcheri exhibits low
Table 4 Thermal-unit growth coe⁄cient (TGC values
cal-culated) for di¡erent Odontesthes species
Species TGC Reference
O hatcheri 0.43 Present study
O bonariensis 0.60 Velasco et al (2008)
O regia 0.54 Orellana and Toledo (2007)
Aquaculture Research, 2011, 42, 754^763 Growth of Patagonian pejerrey in net cages J P Hualde et al.
Trang 27growth rates and early sexual maturation, which
constitute the main drawbacks to its commercial
cul-ture However, the growth rate can be improved by
the development of better feeds and feeding methods
and by selective breeding and genetic techniques
Our results show that the Patagonian pejerrey may
be a good alternative to rainbow trout in reservoirs
of Northern Patagonia, where temperatures in
sum-mer exceed 20 1C The ability of pejerrey to e⁄ciently
utilize natural food makes it a good candidate for
breeding in semi-intensive or intensive, open systems
Acknowledgments
This study was funded by grants from Ministerio de
Ciencia y Tecnolog|¤a (COFECIT, SECTIP Res 1028/
04; CONICET PIP 6244), Municipalidad de Villa El
Choco¤n and Ministerio de Desarrollo Territorial,
Neu-que¤n province, Argentina The authors are indebted
to Mr Miguel Leo¤n for his support in the operation of
facilities and Julio Palacios Foundation for the
admin-istration of the COFECIT, SECTIP grant.We thank two
anonymous reviewers for greatly improving the
manuscript
References
AOAC (1990) O⁄cial methods of analysis,Vol 1, 15th edn (ed.
by K Helrich), Association of O⁄cial Analytical Chemists
Inc, Arlington,VA, USA.
Berasain G., Colautti D & Velasco C.A.M (2000)
Experien-cias de cr|¤a de pejerrey, Odontesthes bonariensis, durante
su primer ano de vida Revista de Ictiolog|¤a 8,1^7.
Cho C.Y (1992) Feeding systems for rainbow trout and other
salmonids with reference to current estimates of energy
and protein requirements Aquaculture 100, 107^123.
Colautti D.C., Garcia de Souza J.R., Balboni L & Baigu¤n
C.R.M (2010) Extensive cage culture of pejerrey
(Odon-testhes bonariensis) in a shallow pampean lake in
Argenti-na Aquaculture Research 41, 376^384.
Cussac V., Cervellini P & Battini M (1992) Intralacustrine
movements of Galaxias maculatus and Odontesthes
hatch-eri during their early life history Environmental Biology of
Fishes 35, 141^148.
Dyer B.S (1993) A Phylogenetic study of Ateriniform ¢shes
with a systematic revision of the South American silversides
(Atherinomorpha, Atherinopsinae, Sorgentinini) PhD
the-sis, University of Michigan, 596pp.
Goddard S (1996) Feed Management in Intensive Aquaculture.
Chapman & Hall, NewYork, 194 pp.
Greenberg A.E., Clesceri L.S & Eaton A.D (1992) Standard
Methods for the Examination of Water and Wastewater,
18thhed APHA,Washington, DC, USA.
Grosman F (1995) El pejerrey Ecolog|¤a, cultivo, pesca y tacio¤n Editorial Astyanax, Buenos Aires, Argentina, 132pp (in Spanish).
explo-Iwama G.K & Tautz A.F (1981) A simple growth model for salmonids in hatcheries Canadian Journal of Fisheries and Aquatic Sciences 38, 649^656.
Kaushik S.J (1998) Nutritional bioenergetics and estimation
of waste production in non-salmonids Aquatic Living sources 11, 211^217.
Re-Lauzanne L (1975) Re¤gime alimentaire d’Hidrocyon forskali (Pisces:Characidae) dans le lac Tchad et ses tributaries Cahiers ORSTOM Se¤rie Hydrobiologie 9, 105^121 Mart|¤nez Palacios C.A., Racotta I.S., R|¤os DuraŁn M.G., Pala- cios E., Toledo Cuevas M & Ross L.G (2006) Advances in applied research for the culture of Mexican silversides (Chirostoma, Atherinopsidae) Biocell 30, 137^148 Miranda L.A., Berasain G.E.,Velasco C.A.M., ShirojoY & So- moza G.M (2006) Natural spawning and intensive cul- ture of pejerrey Odontesthes bonariensis juveniles Biocell
30, 157^162.
Naylor R., Goldburg R., Primavera J., Kautzky N., Beveridge M., Clay J., Folke C., Lubchenco J., Mooney H & Troell M (2000) E¡ect of aquaculture on world ¢sh supplies Nature
Reartes J (1995) El pejerrey (Odontesthes bonariensis): me¤todos de cr|¤a y cultivo masivo COPESCAL (FAO) Docu- mento Ocasional 9, 1^35.
Rosecchi E & NouazeY (1987) Comparaison de cinq indices alimentaires utilise¤s dans l’analyse des contenus stoma- caux Revues des Travaux de I’Institut des PeŒches Maritimes
49, 11^123.
Ross L.G., Martinez Palacios C.A & Morales E.J (2008) Developing native ¢sh species for aquaculture: the interacting demands of biodiversity, sustainable aqua- culture and livelihoods Aquaculture Research 39, 675^683.
Satoh S.,ViyakarnV.,YamazakyY.,Takeuchi T & WatanabeT (1992) A simple method for determination of available phosphorous content in ¢sh diet Nippon Suisan Gak- kaishii 58, 2095^2100.
Somoza G.M., Miranda L.A., Berasain G.E., Colautti D., Remes Lenicov M & Strˇssmann C.A (2008) Historical as- pects, current status, and prospects of pejerrey aquacul- ture in South America Aquaculture Research 39,784^793 Strussmann C.A., Choon N.B., Takashima F & Oshiro T (1993) Triploidy induction in an Atherinid ¢sh, the pejer- rey (Odontesthes bonariensis) The progressive Fish Culturist
55, 83^89.
Toda K., Toshinami K., Yasuda N & Susuki S (1998) Cultivo del Pejerrey en Japo¤n Asociacio¤n Argentino-Japonesa del Pejerrey, Buenos Aires, Argentina, 69pp.
Growth of Patagonian pejerrey in net cages J P Hualde et al Aquaculture Research, 2011, 42, 754^763
Trang 28Tsuzuki M.Y., Ogawa K., Strussmann C.A., Maita M &
Taka-shima F (2001) Physiological responses during stress and
subsequent recovery at di¡erent salinities in adult
pejer-rey Odontesthes bonariensis Aquaculture 200, 349^362.
Velasco C.A., Berasain G.E & Ohashi M (2008) Produccio¤n
intensiva de juveniles de pejerrey (Odontesthes
bonarien-sis) Instituto de Limnolog|¤a ‘‘Dr R.A Ringuelet’’ Biolog|¤a AcuaŁtica 24, 53^58.
Wicki G.A & Lucchini L (2002) Evaluacio¤n del potencial para acuicultura en la regio¤n del Comahue (Provincias de Neuque¤n
y R|¤o Negro) Informacio¤n BaŁsica Secretar|¤a de tura, Ganader|¤a, Pesca yAlimentos (SAGPyA), 52pp Aquaculture Research, 2011, 42, 754^763 Growth of Patagonian pejerrey in net cages J P Hualde et al.
Trang 29Agricul-Hatching rate and larval growth variations in
paternal effects
Jesu¤s Nu¤nez1,2, Diana Castro3, Christian FernaŁndez2, Re¤mi Dugue¤1, Fred Chu-Koo4,
Fabrice Duponchelle1,2, Carmen Garc|¤a3& Jean-Francois Renno1,3
1 IRD UR 175 CAVIAR, BP 5095, Cedex, France
2 UNFV-FOPCA, Lima, Peru¤
3 Laboratorio de Biolog|¤a Molecular y Biotecnolog|¤a, Instituto de Investigaciones de la Amazon|¤a Peruana ^ IIAP, Iquitos, Peru¤
4 Instituto de Investigaciones de la Amazon|¤a Peruana ^ IIAP, Programa para el Uso y Conservacio¤n del Agua y sus Recursos (AQUAREC), Iquitos, Peru¤
Correspondence: J Nunez, 351 RUE J.F BRETON, BP 5095, Montpellier Cedex 05, France E-mail: jesus.nunez@ird.fr
Abstract
In Pseudoplatystoma punctifer (e.g Pseudoplatystoma
fasciatum) larvae, parental e¡ects on hatching,
growth of initial stages and dry feed adaptation were
evaluated as they could in£uence fry heterogeneity,
which is responsible for the enhancement of
canni-balism, and which remains one of the main factors
of mortality during larval stages A full factorial
ex-periment was carried out with 3 females 3 males
producing nine families of full siblings, raised
sepa-rately in triplicates into 30 L tanks at 28 0.5 1C in
a water recirculating system Paternal and maternal
e¡ects were observed on hatching success, yolk
utili-zation e⁄ciency and growth until 26 days post
ferti-lization Hatching success was generally over 80%
except for one male female combination (25%)
To-tal length (TL) at hatching and during the ¢rst 4
weeks of exogenous feeding on live Artemia nauplii
and dry feeds was determined in each family using
digital photographs of larvae andNIH IMAGE Janalysis
freeware Mean TL was calculated for each family at
each sampling time and analysed using
multifactor-ial analysis of variance tests These results indicate
not only dam but also sire e¡ects at very early
devel-opmental stages as well as in subsequent stages of P
punctifer
Keywords: Pseudoplatystoma punctifer,
Pseudopla-tystoma fasciatum, Doncella, reproduction, Peru,
¢sh culture, parental e¡ects
IntroductionThe Doncella, known previously as Pseudoplatystomafasciatum, has been renamed Pseudoplatystoma punc-tifer (Buitrago-SuaŁrez & Burr 2007); however, recentgenetic characterization of this species indicated thatthere was no genetic di¡erentiation between the twodescribed species in the Bolivian and PeruvianAmazon (Torrico, Hubert, Desmarais, Duponchelle,Nunez Rodriguez, Montoya-Burgos, Garcia Davila,Carvajal-Vallejos, Grajales, Bonhomme & Renno 2009).Nevertheless, P punctifer will be used in the text.Pseudoplatystoma punctifer is a large cat¢sh widelydistributed in the Amazon basin and one of the mostcommercially appreciated species (Goulding, Smith &Mahar1996) As a piscivorous species (Barbarino Du-que & Winemiller 2003), its meat is very popular inAmazonian markets because of its taste and the lack
of intra-muscular spines It reaches sizes of morethan 100 cm, but its young are also praised on theornamental market All these traits have stronglyincreased the ¢shing pressure on this species.The Doncella has been regarded as a potential can-didate for the diversi¢cation of South Americanaquaculture for a while (Kossowski 1996), but the de-velopment of its aquaculture has been hampered byhigh mortality rates during the larval and early juve-nile stages, essentially because of intense cannibalis-tic behaviour (Kossowski & Madrid1985; Kossowski &Madrid 1991; Padilla Pe¤rez, AlcaŁntara Bocanegra &Ismino Orbe 2001) This has fostered hybridization
Aquaculture Research, 2011, 42, 764^775 doi:10.1111/j.1365-2109.2011.02803.x
Trang 30attempts with other cat¢sh species, mostly from the
Pimelodidae family, with the objective of producing
fry with reduced cannibalistic behaviour (Kossowski
& Madrid 1991; Kossowski 1996) Cannibalism is
de-scribed as the main cause of death in many ¢sh
spe-cies when it appears during early stages of fry
production (Qin & Fast 1996; Kestemont, Xu, Hamza,
Maboudou & Toko 2007; Arslan, Dabrowski &
Portel-la 2009) Besides the cannibalistic behaviour in this
species, the transition from live prey to inert food
consumption in young stages was found equally
di⁄-cult and this aspect continues to hamper the
produc-tion of large cat¢sh (Pimelodidae) in Latin America
(Kossowski 1996; Nunez 2009) In the last decade,
several studies have provided signi¢cant
contribu-tions for ¢ngerling production (Romagosa, Paiva,
Godinho & Andrade-Talmelli 2003; GervaŁsio
Leonar-do, Romagosa, Borella & Batlouni 2004; Nunez,
Du-gue¤, Corcuy Arana, Duponchelle, Renno, Raynaud &
Legendre 2008; Diaz-Olarte, Cruz-Casallas,
Mar-ciales-Caro, Medina-Robles & Cruz-Casallas 2009)
There are similar di⁄culties in Pseudoplatystoma
cor-uscans, but signi¢cant progresses have been made
especially with regard to weaning schedules
(Marti-no, Cyri(Marti-no, Portz & Trugo 2002; Segura, Hayashi, De
Souza & Soares 2004) Nevertheless, as for P
puncti-fer, recent intra-generic hybrid attempts have been
made to obtain faster growth or less cannibalistic
be-haviour (Faustino, Nakaghi, Marques, Makino &
Sen-horini 2007) Selection programmes are still largely
undeveloped except for some species traditionally
reared for centuries (carps) or more recently due to
strong growth in aquaculture production (trout,
tila-pia) It has been shown that the gain obtained by
se-lection after three generations could allow to nearly
double the growth (Chevassus, Quillet, Krieg,
Holle-becq, Mambrini, Faure, Labbe, Hiseux & Vandeputte
2004) The study of the variability of certain traits,
such as growth, in larval or juvenile stage, has been
initiated in several species, sole, Solea solea and
her-ring, Clupea harengus (Panagiotaki & Ge¡en 1992);
sea bass, Dicentrarchus labrax (Saillant, Chatain,
Fos-tier, Przybyla & Fauvel 2001); haddock,
Melanogram-mus aegle¢nus (Probst, Kraus, Rideout & Trippel
2006); winter £ounder, Pseudopleuronectes
ameri-canus (Butts & Litvak 2007); atlantic halibut,
Hippo-glossus hippoHippo-glossus (Ottesen & Babiak 2007) and a
coral reef ¢sh, Acanthochromis polyacanthus
(Donel-son, Munday & McCormick 2009) In most cases,
these studies have shown that the parentage of the
o¡spring produced from factorial male^female
crosses had a clear impact on larval growth
perfor-mance or size heterogeneity, which is a very tant e¡ect in the culture of ¢sh having cannibalisticbehaviour that induce the necessity of periodicallygrading juveniles
impor-The aim of this study was to evaluate in P punctiferlarvae, the paternal or maternal e¡ects on hatchingand growth on initial stages and during an earlydry feed adaptation period as they could in£uencefry heterogeneity, which is responsible for the en-hancement of cannibalism in P punctifer and re-mains one of the main factors of mortality in earlylarval stages
Material and methodsThe breeding stock consisted of 52 wild ¢sh (28 fe-males and 24 males) maintained for more than 3years in the Instituto de Investigaciones de la Amazo-n|¤a Peruana (IIAPs) Quistococha Research Stationfacilities The ¢sh were kept in a 2000 m2earthenpond and fed with live forage ¢sh supplemented withbeef liver distributed twice a day
Males and females were checked periodically to termine their reproductive status Males were testedfor their capability of semen emission after a slightabdominal pressure, and for females, an intra-ovar-ian biopsy was performed Females were selectedbased on oocyte modal diameter, coe⁄cient of varia-tion (CV%) [CV% 5 (SD/mean) 100] of oocytediameter distributions and percentage of atretic oo-cytes For the three females used, the mean oocytemodal diameter, CV% and percentage of atresia ran-ged from 0.72 to 0.76 mm, 7.2^9.9% and 3.2^6.5%respectively
de-Gamete characterizationFor each female, a sample of stripped eggs was col-lected, then divided in three replicates and digital pic-tures were taken immediately in 9% saline solution
in a Petri dish with a size reference The photographswere then analysed withIMAGE Jfree software pack-age (http://rsb.info.nih.gov/nih-image/)
Sperm was collected by gentle stripping with 5 mLsyringes ¢lled with 4 mL of 9% saline solution, toavoid sperm activation (1:5 sperm dilution) The col-lected sperm was checked for activation under micro-scope The motility was evaluated by a 10 dilution
of the collected sperm with distilled water depositeddirectly on the microscope plate Collected spermmust be completely immotile before activation with
Aquaculture Research, 2011, 42, 764^775 Parental e¡ects on larval growth in P punctifer J Nu¤nez et al.
Trang 31distilled water, otherwise the sample was discarded.
Spermatozoa counts were performed on a Thoma cell
counting chamber using 100^500 diluted sperm
in 9% saline
Spawning induction and gamete collection
Arti¢cial reproduction of P punctifer was performed
according to (Nunez et al 2008) Brie£y, females
in-jected with Ovaprims (Syndel Laboratories,
Quali-cum Beach, BC, Canada), received a total dose of
0.5 mL kg 1body weight, administered as two
injec-tions; a priming one at 10% of the total dose, and 12 h
later, a resolving one at 90% of the total dose
Strip-ping occurred between 8 and 10 h after the second
injection, depending on the average water
tempera-ture (28^27 1C)
Sperm was collected as described previously and
stored at 4 1C until use
Ovules are collected by gentle stripping of the
fe-males in dry 5 L plastic cups and used immediately
Fertilization and incubation
For factorial crosses, 25 g of eggs (approximately
55 000 eggs) of each female were fertilized with 5
diluted semen ( 10 000 spermatozoa egg 1)
Sper-matozoa concentration was adjusted with 9% saline
solution for the three males Eggs and semen were
gently mixed for 30 s and activation was performed
with 25 mL of water Fertilization was achieved after
1min under gentle agitation and fertilized eggs were
rinsed 3 with100 mL of water and distributed into
nine individual 40 L incubators with a constant
water £ow of 1 L min 1at 27.5 0.5 1C
Hatching
Larvae were not collected from the incubators before
the age of 24 hours post fertilization (hpf), to ensure
that hatching had ended in all crosses Larvae were
collected and concentrated into a 7 L aerated
con-tainer and the total amount of hatched larvae was
determined by a volumetric method Fifty millilitres
sample of concentrated larvae was counted in
tripli-cate and the proportions of normal and deformed
hatched larvae were calculated For the experimental
design, the desired number of larvae was sampled by
a similar volumetric method Then, larvae were
quickly transferred to 1m3tanks with aeration and
£ow-through water circulation (1.5 L min 1)
Experimental design
At 1 days post fertilization (dpf), 3000 larvae metric estimate) of each family were assigned ran-domly to 30 L tanks in an indoor recirculation watersystem (three replicate groups per cross) Water tem-perature (28.0 0.5 1C) was kept constant duringthe experiment and all tanks were maintained incomplete darkness (o0.001Lx during the day andnight)
(volu-From the age of 3 dpf onwards, larvae were o¡eredfreshly hatched Artemia nauplii ¢ve times a day every
4 h from 06:00 to 22:00 hours The feeding level at
3 dpf was ¢ve nauplii per larvae per meal and was creased by 25% every day thereafter Feeding was
in-ad libitum in slight excess, on the basis of the artemiafeeding chart modi¢ed from previous work (Nunez
et al 2008), and by controlling 30 min after artemiadistribution that few Artemia nauplii were stillpresent in the water tank, and that larvae stomachswere coloured with the typical orange artemiacolour From 9 dpf, artemia were progressively re-placed by dry feeds (microparticulate shrimp feedand then trout pellets) For microparticulate andtrout pellets, distributed ad libitum, nonconsumedfeed was removed with a siphon 45 min after distri-bution Initial size, yolk and corporal area were deter-mined on 10 larvae at 1 and 3 dpf on triplicatesamples Larvae were randomly taken from eachtank, anaesthetized with clove oil and photographedwith a calibrated size marker Pictures were analysedusing theNIH IMAGEfree software as described pre-viously Corporal (Ca) and yolk sac (Va) areas werecontoured with the hand-drawn closed polygon tool
inNIH IMAGE J As the image was calibrated with thesize marker, the surface was automatically calculated
by the freeware
Yolk utilization e⁄ciency (YUE) was calculated asfollows:
where Carepresents the corporal area in mm2,Vapresents the yolk area in mm2
re-Thereafter, samples of at least 15 larvae were lected from each tank at 5,9,18 and 26 dpf Their totallength (TL) was measured on digital photographs asdescribed previously
col-Parental e¡ects on larval growth in P punctifer J Nu¤nez et al Aquaculture Research, 2011, 42, 764^775
Trang 32At 5 and 19 dpf, densities and artemia or food
rations were adjusted following the established
protocol (Table 1) Three di¡erent successive
decreasing densities were used, 3000, 1500 and
90 larvae tank 1, corresponding to 100, 50 and
3 larvae L 1 At the end of each of the three rearing
periods, survival was determined by a total count of
remaining larvae
Results were analysed using one-way and
multi-factor analysis of variance (ANOVA) for experimental
design procedure (Statgraphics Plus, StatPoint
Tech-nologies,Washington, DC, USA)
Results
Male, female and gamete characteristics
Male and female characteristics are summarized in
Table 2 Body mass ranged from 4.1 to 5.4 kg for
fe-males and from 1.4 to 2.5 kg for fe-males Fork length
varied from 76 to 78 cm in females and from 57 to
66 cm in males Mean egg diameter, determined in
triplicate samples for each female, revealed no
signif-icant di¡erences among females (P 5 0.85) Males
have been chosen for similar sperm concentration
(Table 2), determined in triplicates, and this meter revealed no signi¢cant di¡erences among thethree males (P 5 0.79) used in the 3 3 factorialcrosses with females
para-HatchingTotal hatching rates varied between 65% and 95%,except for one family whose hatching rate dropped
to 25% (Fig 1) Females 1 and 2 (crosses C1^C6) hadsimilar hatching success (around 90%), whereas fe-male 3 (families C7^C9) had a much more variablehatching rate TheANOVAanalysis (Table 3) indicatedsigni¢cant male (P 5 0.0001) and female (Po0.0001)e¡ects, and a signi¢cant male^female interaction(P 5 0.0001)
Over the three females tested (Fig 1), two of themgave similar low rates of abnormal larvae (females 2and 3) but female 1 gave signi¢cantly higher de-formed larvae (Po0.05) than the two other females.The proportion of abnormal larvae varied from 0.4%for family 9 to 17% for family 1 Analysis of variance(Table 4) indicated a strong female e¡ect (Po0.01) ondeformed larvae at 24 hpf, but no male or male^
Table 1 Density and feeding protocol of Pseudoplatystoma fasciatum larvae from 3 to 26 days post fertilization (dpf)
Days of breeding 3^5 dpf 6^9 dpf 10^13 dpf 14^18 dpf 19^26 dpfInitial larvae density per tank 3000 1500 90
Mean egg diameter (mm)
Spermatozoa concentration (spermatozoa mL 1) SD N
Identical superscript letters indicate no signi¢cant variations among oocyte diameter or spermatozoa concentration means at P 5 0.05.
N, the total number of measured eggs, and the number of semen concentration determinations.
Aquaculture Research, 2011, 42, 764^775 Parental e¡ects on larval growth in P punctifer J Nu¤nez et al.
Trang 33female interaction e¡ect was found (P 5 0.893 and0.945 respectively).
Results of initial (24 hpf), yolk area, ¢nal (96 hpf)yolk sac remaining area and YUE are summarized inFig 2 Female 1 (families 1^3) had larger initial and
¢nal yolk sac area, whereas females 2 and 3 had lar values Yolk utilization e⁄ciency tended to varymore homogeneously among females TheANOVAindi-cated a signi¢cant female e¡ect (Po0.05) on the initi-
simi-al yolk area at hatching (24 hpf) but not on the ¢nsimi-alyolk area (P 5 0.900) or onYUE (P 5 0.106) at 3 dpf Asigni¢cant male e¡ect was found on the ¢nal yolkarea (Po0.05) but not on the initial yolk area(P 5 0.67) or onYUE (P 5 0.88)
No male^female interaction was observed withany of the three variables (initial yolk area, ¢nal yolkarea and YUE)
GrowthTheTL was determined at 3 dpf, close to the end of yolkresorption process, and at 5 dpf, when all larvae hadstarted exogenous feeding At 3 dpf, the mean TL ofthe nine families varied from 3.18 to 4.34 mm (Fig 3)and signi¢cant di¡erences were found among them:the factorial ANOVA (Table 5) indicated both male(Po0.0001) and female (Po0.0001) e¡ects as well as
a signi¢cant male^female interaction (Po0.0001)
Figure 1 Percentage of total hatching (dark grey) and percentage of abnormal larvae (light grey) among total hatched ofthe nine crosses of Pseudoplatystoma punctifer (3 males 3 females) Values represent the mean of three samples taken
20 h after fertilization Error bars represent standard deviation C1^C3, female 1 and males 1^3; C4^C6, female 2 andmales 1^3; C7^C9, female 3 and males 1^3
Table 3 Multivariate analysis for hatching ^ Type III sums
of squares ^ F-ratios are based on the residual mean square
Table 4 Multivariate analysis for abnormal larvae ^ Type
III sums of squares ^ F-ratios are based on the residual mean
Trang 34At 5 dpf, mean TL varied from 4.70 to 5.88 mm (Fig.
3) and as for 3 dpf,ANOVA(Table 6) indicated a
signi¢-cant male, female and male^female interaction e¡ect
(Po0.0001)
At the end of the experiment (26 dpf), the mean TLs
of the nine families di¡ered substantially; they varied
from 24.54 to 40.46 mm (Fig 4), with a pronounced
female (Po0.001) e¡ect, but no male e¡ect or male^
female interaction (Table 7) was observed The
growth di¡erential at 26 dpf between the highest
mean size (C8) and the lowest mean size (C5) sented 64%.When pooled per female, the growth dif-ference between the progenies presented signi¢cantvariations (Po0.01) The growth di¡erence was 47%(Po0.01) between the progeny of females 3 (C7^C9)and 2 (C4^C6),34% (Po0.01) between the progeny offemales 1 (C1^C3) and 2 (C4^C6), whereas the size(TL) di¡erence between females 3 and 1 progenies re-presented only 9% and was not signi¢cant (P 5 0.21).The mean size of pooled males progenies repre-sented only a maximum of about 9% total variation,and these di¡erences were not signi¢cant (P 5 0.84).Results from individual family growth over the 3^
repre-26 dpf period are summarized in Fig 5 From 3 to
9 dpf, growth is quite similar among the nine milies, but important di¡erences are observed on
fa-18 dpf and these di¡erences increase substantially
26 dpf The most notable change concerned C5,
Figure 2 Variations of yolk area during yolk sac
resorp-tion in Pseudoplatystoma punctifer Yolk utilizaresorp-tion
e⁄-ciency (YUE) is calculated between 1 and 3 days post
fertilization Values represent the mean of triplicates of 10
larvae per family C1^C3, female 1 and males 1^3; C4^C6,
female 2 and males 1^3; C7^C9, female 3 and males 1^3
Figure 3 Mean total length of the nine families of
Pseu-doplatystoma punctifer larvae, at 3 and 5 days post
fertili-zation (dpf) Values represent the mean and standard
deviation of triplicate samples for each family (10 larvae
per sample, n 5 270) C1^C3, female 1 and males 1^3;
C4^C6, female 2 and males 1^3; C7^C9, female 3 and
males 1^3
Table 6 Multivariate analysis for total length at 5 days post fertilization (dpf) ^ Type III sums of squares ^ F-ratios are based on the residual mean square error
Source
Sum of squares d.f.
Mean square F-ratio P-value
Main effects A: female 1.033 2 0.516 171.7 0.0000 B: male 0.290 2 0.145 48 2 0.0000 Interactions
Source
Sum of squares d.f.
Mean square F-ratio P-value
Main effects A: female 1.107 2 0.553 276.9 0.0000
Trang 35which was the one that showed the best growth on
18 dpf but the worst on 26 dpf (Fig 5) On the
con-trary, C7 had among the worst growth at 18 dpf but
one of the highest at 26 dpf
Survival
The overall survival rates have been calculated at the
end of each rearing density period (100, 50 and
3 larvae L 1) Results of family survival are
summar-Figure 4 Mean total length of the nine families of Pseudoplatystoma punctifer larvae at 26 days post fertilization.Valuesrepresent the mean and standard deviation of triplicate samples for each family (minimum of 15 larvae per sample,
n 5 717) C1^C3, female1and males1^3; C4^C6, female 2 and males1^3; C7^C9, female 3 and males1^3, similar shading
of the bars correspond to the same male (solid black: male 1, dark grey: male 2, light grey: male 3)
Table 7 Multivariate analysis for total length at 26 days
post fertilization (dpf) ^ Type III sums of squares ^ F-ratios
are based on the residual mean square error
Source
Sum of
squares d.f.
Mean square F-ratio P-value
Pseu-Parental e¡ects on larval growth in P punctifer J Nu¤nez et al Aquaculture Research, 2011, 42, 764^775
Trang 36ized in Fig 6 and analysed using multifactorialANOVA
and one-wayANOVA
Family survival during the end of yolk resorption
and the initiation of external feeding (3^5 dpf) varied
from 52% to 91%, and there was no maternal or
pa-ternal e¡ect (P 5 0.45 and 0.946 respectively) or
male^female interaction (P 5 0.632) During this
short period (3 days), larvae were kept at
100 larvae L 1 and were exclusively fed with live
freshly hatched Artemia nauplii One-way ANOVA
indicated signi¢cant variations of mean survival
among families (P 5 0.024), but post hoc tests (Sche¡e¤
or Bonferroni) failed to identify signi¢cant di¡erence
between families There was no signi¢cant
correla-tion (P 5 0.528) between survival and the mean TL
(Fig 7)
From 5 to 18 dpf, initial larvae density was set to
50 larvae L 1and this period corresponded to the
adaptation to microparticulate and trout pelleted
feed Survival varied from 22% to 43% and there
was no paternal, maternal or male^female
interac-tion e¡ects (P 5 0.961; 0.740 and 0.929 respectively)
There was no signi¢cant variations of mean survivalamong families (P 5 0.08), or signi¢cant correlation(P 5 0.686) between survival and the mean TL (Fig.7).From 19 to 26 dpf, initial larvae density was set to
3 larvae L 1and corresponded to 100% trout pelletsfeeding Survival varied from 17% to 40% and asthere was no paternal or maternal or male^femaleinteraction previously (P 5 0.816; 0.758 and 0.11 re-spectively) One-wayANOVAindicated signi¢cant var-iations of mean survival among families (P 5 0.01),but post hoc tests (Sche¡e¤ or Bonferroni) failed toidentify signi¢cant di¡erence between families.There was no signi¢cant correlation (P 5 0.102)between survival and the mean TL (Fig.7)
Discussion
It is accepted that the characteristics of the embryoand the larva depend on genetic and nongenetic(phenotype-based) factors like egg size or quality(Saillant et al 2001; Rideout, Trippel & Litvak 2004;Ottesen & Babiak 2007; Donelson et al 2009) Thenongenetic e¡ects (environmental e¡ects) are oftenassociated with maternal factors, because the female
is responsible for 100% of the yolk characteristics ofthe egg However, the contribution of males in theearly stages of embryonic and larval developmentare sometimes highlighted (Rideout et al 2004;Probst et al 2006), although it may be suspected thatthe male e¡ect is more related to genetic than to en-vironmental factors (Saillant et al 2001) Someauthors, however, indicate that the maternal or pa-ternal e¡ect can be largely o¡set by factors of the
Table 8 Coe⁄cient of variation (CV%) of Pseudoplatystoma
punctifer larvae total length of the nine crosses (C1^C9)
dur-ing the reardur-ing period from 3 to 26 days post fertilization
puncti-fer larvae survival, during
the three breeding
peri-ods [3^5; 6^18 and 19^26
days post fertilization
(dpf)] Values represent
the mean and standard
deviation of triplicates for
each family Survival was
calculated at the end of
each breeding period See
‘Material and methods’for
other details
Aquaculture Research, 2011, 42, 764^775 Parental e¡ects on larval growth in P punctifer J Nu¤nez et al.
Trang 37rearing environment, such as the amount of food
available (Donelson et al 2009), and in some cases,
di¡erences that may occur in early breeding period
can then disappear in weeks or months after birth
(Ottesen, Babiak & Dahle 2009)
In this study, as in previous ones, genetic and
en-vironmental e¡ects cannot be completely
disen-tangled, as the male or female e¡ect observed
cannot be attributed solely to genetic or
environmen-tal factors acting on parenenvironmen-tal characteristics Egg size
is one of the main life-history trait involved with
early embryo development and fry size variability at
hatching (Donelson et al 2009) As egg size did not
di¡er signi¢cantly between the three females used, it
is likely that the observed di¡erences in hatching,
growth and survival are rather linked to the genetic
characteristics of parents; nevertheless, nongenetic
e¡ects related to yolk characteristics (i.e protein or
li-pid content) cannot be totally excluded
The variations in the observed variables on early
larvae stages gave a good picture of larval
develop-ment, growth and progressive early adaptation to
dry feed characteristics for the nine families between
3 and 26 dpf
Hatching
Even though total hatching success was very similar
for the ¢rst six families (C1^C6), it was much lower
for the last three (C7^C9) and dramatically low for
family 9, suggesting some type of incompatibility
be-tween male 3 and female 3 and an overall lower
per-formance at hatching for female 3 Nevertheless, this
family had ¢nally the best growth performance at
26 dpf
The percentages of deformed larvae were low forfemales 2 and 3 but were substantially higher for fe-male 1 However, this characteristic did not a¡ect theoverall growth performance of female 1 families Fe-male 2 had very high hatching rates, and very lowproportions of abnormal larvae, yet the global perfor-mances of its progenies were the worst of the threefemales tested These results strongly suggest that to-tal hatching, or abnormal larvae percentages maynot be good indicators of progeny growth perfor-mance in P punctifer
A weak but signi¢cant female e¡ect was observed
on the initial yolk area, but not on the ¢nal yolk area
As there was no signi¢cant di¡erences of egg meters among females, the di¡erence of yolk area be-tween female 1 and females 2 and 3, probablyrepresents a di¡erent hydration process after fertili-zation and embryogenesis or di¡erences in yolk qual-ity among females, as already suggested for rainbowtrout, Oncorhynchus mykiss (KristjaŁnsson & Vllestad1996) The observed male e¡ect on the ¢nal yolk areamay indicate that the male genome in£uences veryearly larvae metabolism as observed in sea bass, D.labrax (Saillant et al 2001), winter £ounder, P ameri-canus (Butts & Litvak 2007) and in Atlantic halibut,
dia-H hippoglossus (Ottesen & Babiak 2007)
GrowthTotal length varied slightly but signi¢cantly amongfamilies, at 3 and 5 dpf Those variations are gener-ally linked to female, but also to male and male^fe-male interactions (Saillant et al 2001) At 3 or 5 dpf,female 3 had the smallest progenies whereas female
1 had the largest ones As mentioned previously,
Figure 7 Linear sions of total length (TL)
regres-vs survival rate for thethree successive breedingperiods [3^5 days postfertilization (dpf), initialdensity: 100 larvae L 1;6^18 dpf, initial density:
50 larvae L 1; 19^26 dpf,initial density 3 larvae L 1)
Parental e¡ects on larval growth in P punctifer J Nu¤nez et al Aquaculture Research, 2011, 42, 764^775
Trang 38however, the initial size does not appear to be
corre-lated to growth performance later on (at 26 dpf)
After yolk resorption (4 dpf), exogenous feeding
with artemia allowed homogenous size increase for
all progenies with a relatively low size variability as
con¢rmed by the CV% until 9 dpf Between 9 and
14 dpf, artemia were progressively reduced to 0%
while microparticulate food was increased in parallel
from 0% to 100% during the same period This period
corresponds to the observed substantial increase in
size variability among families as well as intra-family
variability, probably because of di¡erent adaptation
capabilities of the individuals at the intra- and
inter-family levels This variability increase was even
higher between 19 and 26 dpf corresponding to the
period of dry food pellets adaptation During this
per-iod, the inter-family variability not only increased
but the hierarchy in family sizes was also modi¢ed
and family size di¡erences were ampli¢ed during this
period Family 5 had the best growth at 18 dpf but
be-came the worst on 26 dpf, suggesting poor adaptation
capabilities to dry pelleted feed It will be interesting
to further study this trend to determine if the high
growth potential can be recovered after the dry feed
adaptation period; otherwise, the initial fast-growing
advantage until 18 or 26 dpf cannot be used as an
ac-curate indicator for selection purposes
Survival
The survival observed during the three consecutive
rearing periods were lower than those observed in
previous experiments (Nunez et al 2008) This
indi-cates that the adaptation time to microparticulate
and to pelleted food was probably too short and that
partial starvation induced lower growth rates and
possibly higher mortality rates Such high mortalities
are probably enhanced by a pronounced
cannibalis-tic behaviour, as described previously in P punctifer
(Kossowski 1996) and in other piscivorous ¢sh
spe-cies, like Perca £uviatilis or Brycon spp (Baras &
Jo-bling 2002) It is indeed likely that the size
di¡erential between individuals adapting at di¡erent
rates to the new food probably reinforce, during the
adaptation period, the natural cannibalistic
ten-dency in this species From our observations, even if
at19 dpf larvae had the possibility to absorb1mm
pel-lets, in many cases, after partial ingestion, pellets
were expulsed probably because of pellet hardness,
because after sinking and hydration, humidi¢ed
pel-lets were already partially consumed by some larvae
at the beginning of the adaptation period This viour possibly decreased feeding rates at the begin-ning of feed replacement and probably induced thehigh mortalities observed in all crosses The mortal-ity variations among the nine families during the lasttwo periods of rearing may have in£uenced to someextent the growth performances because larvaedensity were not exactly the same in all families In-tra-cohort cannibalism may have also played an im-portant role on growth performance because it mayhave contributed to a better food supply of the canni-bals and disappearance of the smallest individuals.However, the linear correlation analysis between size(TL) and survival during the three rearing periods(3^5, 6^18 and 19^26 dpf) revealed no signi¢cantcorrelations between ¢nal density and TL, even if inthe last period (19^26 dpf) there was a slight ten-dency suggesting that lower densities were asso-ciated with higher TL
beha-ConclusionsFinally, the results show that the parental origin sig-ni¢cantly in£uences growth and probably dry feedadaptation capabilities during the ¢rst 4 weeks of life,generating an additional heterogeneity factor inmultiple-family rearing groups This heterogeneityhas been shown to enhance aggressive and canniba-listic behaviour and this aspect has to be consideredfor multiple family communal rearing during earlylarval stages to avoid massive mortalities as men-tioned previously for this species
Family mean TL showed important di¡erences after
1 month of rearing and assuming that at least part ofthe size di¡erential originates from parental genomes,
if growth performance is maintained or even
ampli-¢ed during further breeding months, selection of tential families with the highest growth rates mayallow the improvement of species growth perfor-mance in culture conditions However, further experi-ments should focus on studying the nature of thisimproved performance by some families as these high-
po-er TLs may be at least partly due to othpo-er factors likequicker adaptation to dry food consumption, highercannibalism rates, variable aggressiveness or ¢sh in-teractions In addition, further studies are needed tofully document the persistence of such growth advan-tages over an extended period of time, ideally untilmarketable size, as some recent studies reported thatinitial growth advantage disappeared 50 days afterhatching in a marine ¢sh species, A polyacanthus (Do-nelson et al 2009) This approach has to be continued
Aquaculture Research, 2011, 42, 764^775 Parental e¡ects on larval growth in P punctifer J Nu¤nez et al.
Trang 39until pre-adult stage with large numbers of
‘fast’-growing families, authorizing conservation of
breed-ing stock genetic variability in order to avoid possible
‘bottleneck’e¡ects induced by inappropriate selection
protocols or insu⁄cient number of breeders in the
process of domestication of P punctifer
Acknowledgments
This study was carried out as part of the scienti¢c
collaborations between the Instituto de
Investiga-ciones de la Amazon|¤a Peruana (IIAP) and the
Insti-tut de Recherche pour le De¤veloppement (IRD), both
part of the research network known as Red de
Inves-tigaciones sobre la Ictiofauna Amazo¤nica (RIIA)
(http://www.riiaamazonia.org)
This is a publication IRD-DIVA-ISEM 2011-009
The authors wish to thank Dr Etienne Baras
(hon-orary research associate of the Belgian FNRS) for
constructive comments on a preliminary draft of this
article
This work has been funded in part by the Peruvian
research programme INCAGRO
References
Arslan M., Dabrowski K & Portella M.C (2009) Growth, fat
content and fatty acid pro¢le of South American cat¢sh,
surubim (Pseudoplatystoma fasciatum) juveniles fed live,
commercial and formulated diets Journal of Applied
Ichthyology 25,73^78.
Baras E & Jobling M (2002) Dynamics of intracohort
canni-balism in cultured ¢sh Aquaculture Research 33, 461^479.
Barbarino Duque A & Winemiller K.O (2003) Dietary
segre-gation among large cat¢shes of the Apure and Arauca
Rivers,Venezuela Journal of Fish Biology 63, 410^427.
Buitrago-SuaŁrez U.A & Burr B.M (2007) Taxonomy of the
cat¢sh genus Pseudoplatystoma Bleeker (Siluriformes:
Pi-melodidae) with recognition of eight species Zootaxa
1512, 1^38.
Butts I.A.E & Litvak M.K (2007) Parental and stock e¡ects
on larval growth and survival to metamorphosis in
win-ter £ounder (Pseudopleuronectes americanus) Aquaculture
269, 339^348.
Chevassus B., Quillet E., Krieg F., Hollebecq M-G., Mambrini
M., Faure A., Labbe L., Hiseux J-P & Vandeputte M (2004)
Enhanced individual selection for selecting fast growing
¢sh: the ‘‘PROSPER’’ method, with application on brown
trout (Salmo trutta fario) Genetic Selection Evolution 36,
643^661.
Diaz-Olarte J.J., Cruz-Casallas N.E., Marciales-Caro L.J.,
Med-ina-Robles V.M & Cruz-Casallas P.E (2009) Efectos de la
densidad de siembra y disponibilidad de alimento sobre
el desarrollo y sobrevivencia de larvas de
Pseudoplatysto-ma fasciatum Orinoquia 13, 21^30.
Donelson J.M., Munday P.L & McCormick M.I (2009) tal e¡ects on o¡spring life histories: when are they impor- tant? Biological Letters 5, 262^265.
Paren-Faustino F., Nakaghi L.S.O., Marques C., Makino L.C & horini J.A (2007) Fertilizacao e desenvolv embr.: morfo- metria e anaŁlise estereomicrosco¤pica dos ovos dos h|¤bridos de surubins (pintado, Pseudoplatystoma corrus- cans x cachara, Pseudoplatystoma fasciatum) Acta Scien- tiarum Biological Science 29, 49^55.
Sen-GervaŁsio Leonardo A.F., Romagosa E., Borella M.I &
Batlou-ni S.R (2004) Induced spawBatlou-ning of hatchery-raised zilian cat¢sh, cachara Pseudoplatystoma fasciatum (Linnaeus, 1766) Aquaculture 240, 451^461.
Bra-Goulding M, Smith N.J.H & Mahar J.D (1996) Floods of tune: Ecology & EconomyAlong the Amazon Columbia Uni- versity Press, New York, NY, USA.
For-Kestemont P., Xu X.L., Hamza N., Maboudou J & Toko I.I (2007) E¡ect of weaning age and diet on pikeperch larvi- culture Aquaculture 264, 197^204.
Kossowski C (1996) Perspectives de l’e¤levage des chats (Siluroidei) en Ame¤rique du Sud Aquatic Living Re- sources 9, 189^195.
poissons-Kossowski C & Madrid F (1985) Ensayo de la reproduccio¤n inducida en el bagre rayado cabezo¤n Pseudoplatystoma fasciatum (Linnaeus 1766) (Pisces, Siluriformes) Acta Cienti¢caVenezolana 36, 284^285.
Kossowski C & Madrid F (1991) Observaciones de los dios embrionario y Iarval del bagre rayado cabezo¤n Pseu- doplatystoma fasciatum (Linnaeus, 1766) (Pisces, Siluriformes) BioLlania 8, 9^15.
esta-KristjaŁnsson L.T & Vllestad L.A (1996) Individual tion in progeny size and quality in rainbow trout, Oncor- hynchus mykiss (Walbaum) Aquaculture Research 27,335^ 343.
varia-Martino R.C., Cyrino J.E.P., Portz L & Trugo L.C (2002) formance and fatty acid composition of surubim (Pseudo- platystoma coruscans) fed diets with animal and plant lipids Aquaculture 209, 233^246.
Per-Nunez J (2009) Domestication de nouvelles espe¤ces d’inte¤reŒt aquacole en Amazonie Cahiers Agricultures 18, 136^143 Nunez J., Dugue¤ R., CorcuyArana N., Duponchelle F., Renno J.F., Raynaud T.N.H & Legendre M (2008) Induced breed- ing and larval rearing of Surub|¤, Pseudoplatystoma fascia- tum (Linnaeus, 1766), from the Bolivian Amazon Aquaculture Research 39,764^776.
Ottesen O.H & Babiak I (2007) Parental e¡ects on tion and hatching success and development of Atlantic halibut (Hippoglossus hippoglossus L.) embryos and larvae Theriogenology 68, 1219^1227.
fertiliza-Ottesen O.H., Babiak I & Dahle G (2009) Sperm competition and fertilization success of Atlantic halibut (Hippoglossus hippoglossus L.) Aquaculture 286, 240^245.
Padilla Pe¤rez P.P., AlcaŁntara Bocanegra F & Ismino Orbe R (2001) Reproduccio¤n inducida de la doncella Pseudopla- Parental e¡ects on larval growth in P punctifer J Nu¤nez et al Aquaculture Research, 2011, 42, 764^775
Trang 40tystoma fasciatum y desarrollo embrionario ^ larval Folia
Amazonica 12, 141^154.
Panagiotaki P & Ge¡en A.J (1992) Parental e¡ects on size
variation in ¢sh larvae Journal of Fish Biology 41, 37^42.
Probst W.N., Kraus G., Rideout R.M & Trippel E.A (2006)
Parental e¡ects on early life history traits of haddock
Mel-anogrammus aegle¢nus ICES Journal of Marine Science 63,
224^234.
Qin J & Fast A.W (1996) Size and feed dependent
cannibal-ism with juvenile snakehead Channa striatus Aquaculture
144, 313^320.
Rideout R.M.,Trippel E.A & Litvak M.K (2004) Paternal
ef-fects on haddock early life history traits Journal of Fish
Biology 64, 695^701.
Romagosa E., Paiva P.d., Godinho H.M & Andrade-Talmelli
E.f.d (2003) Caracter|¤Łsticas morfome¤Łtricas e crescimento
do cachara, Pseudoplatystoma fasciatum (Linnaeus, 1766),
em cativeiro Acta Scientiarum Animal Science MaringaŁ
25, 277–283.
Saillant E., Chatain B., Fostier A., Przybyla C & Fauvel C (2001) Parental in£uence on early development in the European sea bass Journal of Fish Biology 58, 1585^ 1600.
Segura L., Hayashi C., De Souza S & Soares C (2004) balismo entre larvas de pintado, Pseudoplatystoma corrus- cans, cultivadas sob diferentes densidades de estocagem Acta Scientiarum Biological Science 26, 299^302 Torrico J.P., Hubert N., Desmarais E., Duponchelle F., Nunez Rodriguez J., Montoya-Burgos J., Garcia Davila C., Carvajal-Vallejos F.M., Grajales A.A., Bonhomme F & Renno J.F (2009) Molecular phylogeny of the genus Pseudoplatystoma: biogeographic and evolutionary implications Molecular Phylogenetics and Evolution 51, 588^594.
Cani-Aquaculture Research, 2011, 42, 764^775 Parental e¡ects on larval growth in P punctifer J Nu¤nez et al.