Diets containing1% brewers yeast or 30% DDGS sup-ported the same level of growth and feed e⁄ciency ratio FER as the diet containing 5% ¢sh meal.. How-ever, the weight gain of ¢sh fed the
Trang 2Bioremediation and reuse of shrimp aquaculture
Luis R Mart|¤nez-Co¤rdova1, Jose¤ A Lo¤pez-El|¤as1, Guadalupe Leyva-Miranda2, Luis Armenta-Ayo¤n2&Marcel Martinez-Porchas3
1 Departamento de Investigaciones Cient|¤¢cas y Tecnolo¤gicas de la Universidad de Sonora, Sonora, Me¤xico
2 Maestr|¤a en Biociencias, Universidad de Sonora, Sonora, Me¤xico
3 Centro de Investigacio¤n en Alimentacio¤n y Desarrollo, Sonora, Me¤xico
Correspondence: M Mart|¤nez-Porchas, Centro de Investigacio¤n en Alimentacio¤n y Desarrollo, Km 0.6 Carretera a La Victoria, sillo, Sonora, Me¤xico E-mail: marcel@ciad.mx
Hermo-Abstract
Shrimp aquaculture e¥uents were bioremediated in
a two-phase system (System A) using the black clam
Chione £uctifraga and the benthic microalgae
Navicu-la sp., and then reused to farm whiteleg shrimp
Lito-penaeus vannamei In the experimental design,
Systems B and C had an identical structure as System
A, but no clams or microalgae were added System B
received the same shrimp e¥uents while System C
received only estuarine water Shrimp raw e¥uents
had a poor water quality System A improved
the water quality by decreasing the concentrations
of total nitrogen, total ammonia nitrogen (TAN),
ni-trites, nitrates, phosphates, total suspended solids
(TSS) and organic suspended solids (OSS) System B
also decreased the concentration of TAN, TSS
and OSS via sedimentation, but the e¡ect was less
pronounced than that observed in System A Shrimp
reared in the bioremediated e¥uents (System A) had
better production (3166 kg ha 1) and higher survival
(89.2%) than those reared in e¥uents from Systems B
(2610 kg ha 1,75.1%) and C (2874 kg ha 1, 82.1%) It
is concluded that the bioremediation system was
moderately e⁄cient and the bioremediated e¥uents
were suitable to farm L vannamei
Keywords: bioremediation, shrimp e¥uent, Chione
£uctifraga, Navicula sp
Introduction
Aquaculture has experienced vigorous growth
worldwide in the last two decades Its contribution
to the global production of aquatic organisms grewfrom 3.9% in 1970 to more than 36% in 2006 (FAO2009) The Crustaceans are the group with the high-est growth rate (almost 17% per year from 2000 to2006) and penaeid shrimp are by far the most impor-tant in terms of volume and value of production (FAO2009)
However, the explosive development of shrimpaquaculture has caused some serious problems, such
as competition for water and land (PaŁez-Osuna, cia, Flores-Verdugo, Lyle-Fritch, Alonso-Rodr|¤guez.,Roque & Ruiz-FernaŁndez 2003), environmental im-pacts, including deforestation, eutrophication of re-ceiving ecosystems, modi¢cation of habitat forterrestrial and aquatic animals, modi¢cation of land-scape and hydrological patterns (GonzaŁlez-Ocampo,Morales, CaŁceres-Mart|¤nez, Aguirre, HernaŁndez-VaŁz-quez, Troyo-Dieguez & Ortega-Rubio 2006; Thomas,Courties, El Helwe, Herbland & Lemonnier 2010), thedependence of formulated shrimp feed from ¢sh meal
Gra-as the main protein ingredient (Tacon 2002) and thecontinuous presence of epizooties (SaŁnchez-Mart|¤-nez, Aguirre-GuzmaŁn & Mej|¤a-Ruiz 2007)
As an example of the potential impact of ture e¥uents, Tacon (2002) shows data regardinghow much organic matter (OM), nitrogen (N) andphosphorous (P) is discharged into the environmentfor each tonne of shrimp harvested, depending onthe feed conversion ratio (FCR) In Mexico, shrimpaquaculture operates with a mean FCR of about 1.8;considering the annual production and the data pro-vided by Tacon (2002), it is calculated that 112 mil-lion kg of OM, 7.8 million kg of N and 2.5 million kg of
Trang 3P are discharged into the receiving ecosystems each
year (Mart|¤nez-Co¤rdova, Mart|¤nez-Porchas &
Corte¤s-Jacinto 2009) These are huge amounts and it is
abso-lutely necessary to stop and if possible to reverse this
process if we wish to have a sustainable aquaculture
Some strategies have been used or suggested to
minimize these impacts, including settling lagoons
(Mart|¤nez-Co¤rdova & Enriquez-Ocana 2007), septic
tank treatments (Summerfelt & Penne 2007), low or
zero water exchange (Balasubramanian, Pillai &
Ra-vichandran 2005), recirculation systems (Timmons,
Ebeling, Wheaton, Summerfelt & Vinci 2002;
Leza-ma-Cervantes, Paniagua-Michel & Zamora-Castro
2010), the use of mangrove forests as nutrient sinks
(Rivera-Monroy, Torres, Bahamon, Newmark &
Twil-ley 1999), polyculture practices (Martinez-Cordova &
Martinez-Porchas 2006; Porchas,
Mart|¤nez-Co¤rdova, Porchas-Cornejo & Lo¤pez-El|¤as 2010) and
bioremediation (Paniagua-Michel & Garcia 2003)
Bioremediation is the use of individual or
combined organisms (animal, vegetal, bacteria, etc.)
to minimize the polluting charge of e¥uents from
aquaculture or any other activity, taking advantage
of the natural or modi¢ed abilities of those organisms
to reduce and/or transform waste products
(Chavez-Crooker & Oberque-Contreras 2010) Bioremediation
can be conducted in di¡erent forms: in situ, ex situ,
biostimulation, bioagmentation and others Some
examples of successful bioremediation practices are
the use of plants (phytoremediation), macroalgae,
microalgae, ¢lter feeders, bio¢lters (polymer spheres
with immobilized microorganisms), bio¢lms^bio£ocs
(De Schryver, Crab, Deforidt, Boon & Verstraete 2008;
Kuhn, Boardman, Craig, Flick Jr & Mclean 2009) or
combined systems including two or more of these
practices Although it has been demonstrated that
some bivalves and micro- or macroalgae are capable
of bioremediating e¥uents, many of these studies
have been focused on the bioremediation of ¢sh
e¥u-ents (Hussenot 2003; Zhou,Yang, Hu, Liu, Mao, Zhou,
Xu & Zhang 2006; Liu,Wang & Lin 2010)
The use of endemic species of bivalve and
microal-gae to bioremediate e¥uents would prevent the
intro-duction of exotic species, which may cause other
problems It is important to study di¡erent
combina-tions of these species in order to achieve the greatest
e⁄ciency
The black clam (Chione £uctifraga) inhabits
estu-aries and shallow coastal waters in the Gulf of
Cali-fornia It tolerates high concentrations of OM in the
water column and can withstand a wide range of
temperatures and salinities (Mart|¤nez-Co¤rdova
1988), conditions that are similar to those prevailing
in shrimp farm e¥uent; also, the species has mercial importance in north-western Mexico,mainly as an artisanal ¢shery, but it is beginning to
com-be farmed in the region (Tinoco-Orta & t|¤nez 2003) In the case of microalgae, Navicula sp is adiatom that has been found in shrimp ponds and can
CaŁceres-Mar-be a food source for shrimp; also, it is reported to havethe ability to act as a bioremediator of water (Pania-gua-Michel & Garcia 2003)
Based on the above information, the study wasfocused on evaluating an integrated system usingbenthic microalgae (Navicula sp.) and clam (C £uctifra-ga) to bioremediate shrimp aquaculture e¥uents andreuse the bioremediated e¥uents to farm whitelegshrimp Litopenaeus vannamei at a microcosm level
Material and methodsOrganisms
The shrimp were obtained from a commercial farm(Maricultura del Pac|¤¢co S.A., Mazatlan, Me¤xico)and maintained under the laboratory conditions ofdissolved oxygen (DO) 6.05 mg L 1, temperature
28 1C, total ammonia nitrogen (TAN) 0.01mg L 1,fed ad libitum (35% crude protein; Purina Me¤xicos,Hermosillo, Me¤xico) and a daily water exchange of100% These conditions were maintained until theshrimp achieved an individual biomass of 5 g.Adult black clams (C £uctifraga) with an averagesize of 3.0 0.4 cm were hand collected from the LaCruz estuary (2814805700N, 11115503000O); organisms
of a lower size were discarded, because we aimed toevaluate only the capacity of adult clam, which have
a greater ¢ltration capacity The clam was tained under the above laboratory conditions for 2weeks During this period, the organisms were fedwith the diatom Chaetoceros muelleri
main-The microalga Navicula sp was obtained from anaquaculture laboratory at Centro de Estudios Super-iores del Estado de Sonora (CESUES, Navojoa, Sonora,Me¤xico) The microalgae were scaled up from 10 mL
to 200 L using an F/2 medium with a double tration of silicates Thereafter, the experimental unitswere inoculated with the microalgae 1 week beforebeginning the trial
concen-Shrimp culture systemThe e¥uents used for the study were obtained from asemiintensive culture of white shrimp L vannamei
Trang 4(25 shrimp m 2) reared in tanks with estuarine
water
The shrimp were farmed in six rectangular, plastic
tanks with an area of 6 m2 and ¢lled with water
pumped from the estuary; each tank was provided
with 0.9 m3of sediment to achieve a height of 5^
6 cm The culture conditions were as follows: time of
culture 50 days, initial size 5.0 0.3 g, stocking
den-sity 25 org m 2, feeding supply and frequency twice
a day to satiation (in feeding trays), the feed used was
Camaronina with 35% of crude protein (Purina
Me¤x-icos) and the daily water exchange was 20%
Bioremediation system
One half of the e¥uent was sent to a bioremediation
system and the other half was sent to an identical
physical system but was not treated Both types of
water were then used to cultivate white shrimp
The bioremediation system (System A) consisted of
two phases In Phase I, the e¥uents were equally
dis-tributed and made to £ow through three di¡erent
plastic tanks (1000 L) with black clams (C £uctifraga)distributed on the bottom (35 org m 2) In Phase II,the water treated in each tank with clams £owed intosubsequent tanks (1000 L) containing the benthicmicroalgae Navicula sp at an initial concentration of
50 000 cells mL 1(Fig 1) The microalgae were tached to the walls and to arti¢cial substrates (plasticnets) introduced into the tanks, with a surface area of1.65 m2
at-The untreated e¥uents £owed through the System
B, which had a structure identical to the tion system (System A), but with no clams or micro-algae in the 1000 L tanks (Fig 1)
bioremedia-A third system (System C) was constructed to termine the quality of the inlet estuarine water andwas used as a control of the shrimp rearing in the ef-
de-£uents The structure of the System C was identical tothat of Systems A and B, but shrimp were not reared
in the rectangular tanks that received the estuarinewater; also, no clams or microalgae were introducedinto the 1000 L tanks
Every system was constructed with three cates with a water £ow of 0.56 L min 1for every ex-
repli-intensive culture of white shrimpEffluents
Trang 5perimental unit The £ow rates of the experimental
systems were controlled by water valves The
hydrau-lic retention time (HRT) (yh) was 29.7 h for each
circu-lar tank containing clams or microalgae and 184.5 h
for each complete system
Reuse of treated e¥uents to cultivate white
shrimp
After being treated in Systems A^C, the water was
then reused for rearing shrimp (L vannamei) in
plas-tic tanks (6 m2) identical to those mentioned above
Nine pools were used to farm the shrimp in the
trea-ted water The treatrea-ted water from System A was
dis-tributed into three pools with a shrimp stocking
density of 25 shrimp m 2(Fig.1); the water from
Sys-tem B also £owed into the other three culture pools,
and the same was done for System C Particularly for
System C, the water used was pumped directly from
the estuary, but previously £owed through all the
pools without animals or microalgae The culture
conditions were the same as mentioned above for
the shrimp culture system
Analysis of water quality and production
variables
The environmental and water quality variables were
measured in the estuarine inlet water and e¥uents
(before and after bioremediation) The estuarine inlet
water was sampled in the rectangular tanks of
Sys-tem C, which did not have shrimp and received water
directly from the estuary The raw e¥uents were
measured at the exit of each shrimp culture tank
be-fore the water entered into System A or B The
e¥u-ents from each system were then analysed as they
exited the tanks of Phase II
The variables were monitored periodically for
tem-perature, salinity, DO, pH and chlorophyll a twice a
day, using a multisensor YSI 6600 series (Yellow
Springs, OH, USA) Total nitrogen (TN), N-NO2,
N-NO3, TAN and P-PO4, total suspended solids (TSS)
and organic suspended solids (OSS) were all
deter-mined once a week
The TN was measured using the micro-Kjeldahl
method The concentration of TAN was evaluated
using the ammonia-salicylate method (Bower &
Holm-Hansen 1980) The nitrite concentration was
determined using the colorimetric method described
by Strickland and Parsons (1972, NitriVer Method)
Ni-trate was determined using the cadmium^copper
re-duction method (Wood, Armstrong & Richards 1967),and the orthophosphates were measured using thePhosVer 3 method (HACH, Loveland, CO, USA)
To measure the TSS and OSS, 1 L of the sampledwater was ¢ltered through GFC 47 mm Whatman ¢l-ters, which were then washed and dried at 90 1C for
4 h The di¡erence in weight between the dried ¢ltersbefore and after ¢ltration was estimated as the TSS.The OSS were determined by incineration of the dried
¢lters in a mu¥e furnace at 450 1C for 8 h, and thencooled and weighed The same process was carriedout with clean seawater (previously ¢ltered and ster-ilized) from the estuary, which was a basepoint sub-tracted for the results of TSS and OSS
The concentration of microalgae cells (Navicula sp.)was determined by treating samples of water with aBranson 2210 ultrasonic bath and the concentration
of cells per millilitre was determined in a ometer
hematocyt-The shrimp production variables, growth, val, ¢nal biomass and FCR were evaluated in each ofthe pools The FCR was estimated as the weight offeed administered/weight gain of the shrimps
ResultsBioremediationSigni¢cant di¡erences in some of the water qualityparameters were observed among the raw e¥uents,the di¡erent Systems of e¥uent processing (A, B orC) and the inlet estuarine water (Figs 2 and 3).Most of the parameters monitored for water qual-ity were signi¢cantly higher in the raw e¥uents andthe e¥uents from System B that were not bioreme-diated, while lower concentrations of those para-meters were observed in System A (bioremediatede¥uents), System C (control) and the Estuary.The TN increased with time in all the systems,but signi¢cantly higher values were found in theraw e¥uents and System B, while no di¡erences were
Trang 6observed among the rest of the Systems (A and C) and
the estuarine water (Fig 2)
The concentration of TAN recorded a similar
dynamic in all the treatments through the
experimental period; however, the raw e¥uents
and the e¥uents from System B had higher
concentrations of TAN that those found in System C
and the Estuary In the case of the bioremediated
ef-£uents (System A), the concentrations of TAN were
signi¢cantly lower than the raw e¥uents but no
dif-ferences were found with regard to the rest of the
treatments
For nitrite levels, the raw e¥uents and System B
showed the highest concentrations during the
ex-periment Although System A showed results similar
to those of System B (P 5 0.069), they were also
simi-lar to those observed in System C and Estuary
(P40.7) A similar tendency was observed for
ni-trates as that for TAN and nitrite concentrations
(Fig 2)
The phosphates were also higher in the
nonbiore-mediated e¥uents (raw e¥uents and System B) as
compared with the rest of the treatments System A
had lower values than the nonbioremediated
e¥u-ents, but higher than those of System C and the
Estu-ary (Fig 2)
The TSS were the highest in the raw e¥uents,
fol-lowed by those found in System B Both
nonbioreme-diated e¥uents had higher values of TSS than
Systems A and C and the estuarine water Systems A
and C and the estuarine water showed statistically
si-milar values (Fig 3) The OSS were also the highest in
the raw e¥uents and those from System B No ences were found among the OSS levels of System Aand the rest of the treatments (Fig 3)
1 2 3 4 5 6 7
0 0.1 0.2 0.3 0.4 0.5
1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
0.01 0.02 0.03 0.04 0.05
1 2 3 4 5 6 7 0
0.3 0.6 0.9 1.2
Figure 2 Concentrations of water quality parameters found throughout time in the raw e¥uents, the bioremediatede¥uents with bivalve and microalga (System A), the nonbioremediated e¥uents (System B), the estuarine that £owed into
a similar system of tanks but without animals (System C) and the estuarine water (Estuary) Di¡erent letters on the left ofeach marker in each graphic indicate signi¢cant di¡erences
–1 )
050100150200250300
Total Suspended Solids
Raw Effluents System A System B System C Estuary
010203040
50 Organic Suspended Solids
Time (Weeks)
Raw Effluents System A System B System C Estuary
c ab a a b
a a
a b c
Figure 3 Concentrations of suspended solids found out time in the raw e¥uents, the bioremediated e¥uents withbivalve and microalga (System A), the nonbioremediated e¥u-ents (System B), the estuarine that £owed into a similar system
through-of tanks but without animals (System C) and the estuarinewater (Estuary) Di¡erent letters on the left of each marker ineach graphic indicate signi¢cant di¡erences
Aquaculture Research, 2011, 42, 1415^1423 Bioremediation and reuse of e¥uents L R Mart|¤nez-Co¤rdova et al.
Trang 7In terms of e⁄ciency, System A with clams and
mi-croalgae removed 17.3% of TN, 24.5% of TAN, 19.2% of
nitrites, 13.5% of nitrates, 21.6% of phosphates, 22.3%
of TSS and 23.2% of OSS from the raw e¥uents System
B showed a removal e⁄ciency ofo4% of TN, nitrites,
nitrates and phosphates, while the removal of TAN,TSS
and OSS was 10.2%, 10.1% and 8.6% respectively
The concentration of the benthonic microalgae
re-mained at levels above the 50 000 cells mL 1, during
the ¢rst 5 weeks of the experiment, but declined
dur-ing the last week (Fig.4) By the last week, the presence
of Navicula sp was replaced in part by an unidenti¢ed
diatom The chlorophyll levels were signi¢cantly
higher in Phase II of System A (23.5 3.8 mg m 3),
which had the benthonic microalgae, followed by
Phase II from Systems B (7.0 2.6 mg m 3) and C
(4.0 2.7 mg m 3) respectively The survival of the
clams was constant during the entire experiment,
with an average survival above 90% in every week
(Fig 4); as adult clams were used for the experiment,
the growth during the experimental period was not
signi¢cant (0.5^1.0 mm)
Shrimp rearing in e¥uents
No signi¢cant di¡erences were detected among the
treatments with respect to the following
environ-mental variables: temperature, salinity, DO, pH and
PO4(Table 1) However, higher values of TAN, NO2and NO3were observed in the tanks where shrimpfrom System B were reared
Regarding the production parameters of theshrimps farmed in di¡erent types of e¥uents, the ¢-nal weight and the growth rate were statistically si-milar among all the treatments (Systems A, B and C).However, the highest survival was found in shrimpsreared in the bioremediated e¥uents (System A), fol-lowed by those reared in estuarine water (System C),while the lowest survival was observed in theshrimps cultivated in nonbioremediated e¥uents(System B) (Table 2)
The highest biomass was also achieved in System
A and the lowest in System B, while no di¡erenceswere observed among System C and the rest of thetreatments The biomass of shrimp from System Awas 21% higher than that of System B (Table 2)
Figure 4 Concentration of the benthic microalga
(Navi-cula sp.) and survival of clams (Chione £uctifraga) during
the experiment
Table 1 Environmental variables in the microcosm culture
of shrimp in the bioremediated e¥uents (System A), the bioremediated e¥uents (System B) and the estuarine water (System C)
non-System A System B System C
Temperature ( 1C) 26.1 1.1a 26.2 1.2a 25.9 0.9a Salinity (%) 41.2 2.1a 41.5 2.0a 41.0 1.9a
DO (mg L 1) 5.6 0.5a 5.7 0.6a 5.9 0.6a
pH 8.2 0.3a 8.2 0.2a 8.3 0.2a
NO3(mg L 1 ) 0.30 0.06a 0.41 0.05b 0.28 0.05a
NO 2 (mg L 1 ) 1.04 0.55a 1.51 0.76b 1.01 0.62a TAN (mg L 1 ) 0.68 0.35ab 1.05 0.6b 0.60 0.32a
PO 4 (mg L 1 ) 0.16 04a 0.21 09a 0.15 0.06a Di¡erent letters in the same row indicate signi¢cant di¡erences (P o0.05).
DO, distilled water; TAN, total ammonia nitrogen.
Table 2 Production parameters of white shrimp farmed at microcosms in the bioremediated e¥uents (System A), the nonbioremediated e¥uents (System B) and the estuarine water (System C)
Final weight (g) 14.2 1.3 a 13.8 1.2 a 14.0 1.6 a
Growth rate (g week 1 )
1.15 0.04 a 1.10 0.05 a 1.12 0.06 a
Survival (%) 89.2 2.1 a 75.1 3.6 c 82.1 3.8 b
Final biomass (kg ha 1 )
Trang 8Finally, FCRs were lower in Systems A and C
com-pared with those obtained for the shrimps reared in
System B (Table 2)
Discussion
Shrimp rearing at a semiintensive scale is activity
practice capable of exerting a signi¢cant impact on
the environment Some of the parameters of water
quality increased by 100% or more from the
estuar-ine water to the raw e¥uents Although this was
de-monstrated at an experimental scale, similar results
have been observed in commercial farms (Jackson,
Preston & Thompson 2004) Jackson et al (2004)
stu-died the discharge nutrient loads at di¡erent shrimp
farms, ¢nding concentrations of TN and TSS as high
as 3 and 200 mg L 1, respectively; similar results
were observed in this experiment
The bioremediation system appeared to have a
moderate e⁄ciency, as nitrogenous compounds and
the phosphates were lower after the e¥uent £owed
through the pools with clams and the benthonic
mi-croalgae Some authors have documented that the
presence of bivalves and microalgae can decrease
the concentration of di¡erent compounds that
con-taminate receiving ecosystems For instance,
Her-nandez, Bashan and Bashan (2006) found a P
removal e⁄ciency of around 25% for Chlorella spp
alone, and up to 72% for Chlorella spp
co-immobi-lized with Azospirillum braziliense In addition, Jones,
Dennison and Preston (2001) evaluated a
multi-phase system (sedimentation, ¢lter feeders and
mi-croalgae) to treat shrimp e¥uents and achieved an
overall improvement in water quality as follows: TSS
(12%), TN (28%), P (14%), NH41(76%), NO3 (30%),
PO4
(35%), bacteria (30%) and chlorophyll a (0.7%);
however, the HRT of this system was lower than that
observed in our experiment, which suggests that the
e⁄ciency of our system was lower that of Jones et al
(2001) Also, the density of bivalves they used was
much higher, although the retention time used in
our system corresponded to a daily water exchange
similar to that used in commercial farms (10^20%)
The bioremediation e⁄ciency in the
above-men-tioned experiments as well as in our experiment was
estimated by measuring the number of nutrients
removed from the raw e¥uents The results suggest
that the quality of water was signi¢cantly improved
after being treated by the bioremediation system (A);
however, System B, which did not have clams or
mi-croalgae, also showed some e⁄ciency in removing
TAN,TSS and OSS These results may be attributed to
the sedimentation in the pools Hence, it can be pothesized that the e⁄ciency of System A in remov-ing nitrogenous metabolites and phosphates can beattributed to the presence of Navicula sp., while theremoval of suspended solids may be caused in part
hy-by the presence of clams and hy-by sedimentation inthe tanks (sediments were observed in the treatmentpools of Systems A and B, although they were notmeasured) Vymazaj (1988) found that microalgaespecies such as Navicula sp were capable of removingnutrients from polluted streams with a maximum ef-
¢ciency of 80% and 70% for ammonium and phosphates respectively Navicula sp has been used aspart of bio¢lms to improve the water quality, due to itsability to remove nitrogenous compounds and phos-phates (Thompson, Abreu & Wasielesky 2002) Re-garding suspended solids, the presence of bivalvesand the use of tanks as sedimentation units havebeen shown to decrease the concentration of sus-pended solids and TN from aquaculture e¥uents(Jones et al 2001; Bernal-Jaspeado 2006; Li, Veilleux
Although the bioremediation system with C fraga and Navicula sp had acceptable e⁄ciency, theconcentrations of the microalgae decreased duringthe last week This decrease may be attributed to thehigh turbidity observed in System A during the last 2weeks, caused by the increase in TSS and OSS It hasbeen reported that the abundance of some microal-gae species such as Navicula sp depends on the tur-bidity and the concentration of suspended solids(Unrein & Vincour 1999) Some alternatives to solvethis problem might be to increase bivalves in Phase I
£ucti-or include a sedimentation tank, as suggested byJones et al (2001) to decrease the concentration ofsuspended solids Although Navicula sp levels de-creased by the last week, the nitrogenous metabolitescontinued decreasing in System A, which may be at-tributed to the remaining concentration of Navicula
sp and the unidenti¢ed diatom that was nant during the last few days
predomi-As a ¢rst approach, it was observed that thebioremediation system had lower e⁄ciency thanthose using macroalgae, in terms of nutrient removaland biomass production (Xu, Fang & Wei 2008;Marinho-Soriano, Nunes, Carneiro & Pereira 2009;Mart|¤nez-Porchas et al 2010) However, it is importantAquaculture Research, 2011, 42, 1415^1423 Bioremediation and reuse of e¥uents L R Mart|¤nez-Co¤rdova et al.
Trang 9to continue studying the ability of C £uctifraga and
Navicula sp as potential bioremediators of shrimp
(or ¢sh) e¥uents, at di¡erent densities and using
dif-ferent system designs, to achieve a better e⁄ciency
Regarding the production parameters of the
shrimp reared on the three systems, it was observed
that the bioremediated e¥uents (System A) were
very suitable for the culture of white shrimp
The growth rates were similar to or higher than the
0.9^1.0 g week 1, reported as commercially feasible
(Mart|¤nez-Co¤rdova 1999) The productive response of
the shrimp reared in the bioremediated e¥uents was
better than that observed in those cultivated in
e¥u-ents from Systems B and C The di¡erence in survival
and biomass among the shrimps reared in
bioreme-diated e¥uents (from System A) and those reared
in the nonbiormediated e¥uents (System B) may be
attributed to the improvement in the water quality
Although none of the water quality parameters
reached the lethal concentration (LC50) in the
non-bioremediated e¥uents, they were almost twofold
higher than those from the bioremediated e¥uents
Di¡erent authors have documented that the chronic
exposure of penaeid shrimps to high concentrations
of nitrogenous compounds and suspended solids can
diminish their growth and food intake
(Frias-Esperi-cueta, Harfush-Melendez & Paez-Osuna 2000; Ray,
Lewis, Browdy & Le¥er 2009) Moreover, the slight
increase in the productive response of shrimps from
System A as compared with those reared in System C
could be attributed to the higher survival and to the
presence of OM in System A, such as bio£ocs and
mi-croalgae (Navicula sp and other diatoms), which
could be an alternative source of food
The biomass obtained in Systems A and C was
higher than the mean reported in most semiintensive
farms of the region The FCRs in the same treatments
are considered to a pro¢table value for commercial
purposes (Juarez 2008) The results suggest that the
shrimp can thrive in bioremediated e¥uents, which
indicates that the treated water (by clams and
micro-algae) can be reused by recirculation This practice
would reduce the environmental impact caused by
the massive discharges of shrimp aquaculture
Furthermore, the black clam have commercial value
and may represent and extra economical income for
farmers (Mart|¤nez-Porchas et al 2010)
It can be concluded that the bioremediation system
was moderately e⁄cient in removing nutrients and
solids (TSS and OSS) from shrimp aquaculture
e¥u-ents Also, the tanks themselves are useful due to
the sedimentation activity
It is necessary, however, to improve System A withsome modi¢cations in the design The productionparameters obtained in the present study stronglysuggest that the bioremediated e¥uents can be usedfor farming white shrimp without a negative e¡ect onits survival and growth
in-Chavez-Crooker P & Oberque-Contreras J (2010) diation of aquaculture wastes Current Opinion in Biotech- nology 21, 313^317.
Bioreme-De Schryver P., Crab R., Bioreme-Deforidt T., Boon N & Verstraete W (2008) The basics of bio-£ocs technology: the added value for aquaculture Aquaculture 277, 125^137.
FAO (2009) The State ofWorld Fisheries and Aquaculture 2008 Food and Agriculture Organization of the United Nations, Rome, Italy, 196pp.
Frias-Espericueta M.G., Harfush-Melendez M & Paez-Osuna
F (2000) E¡ects of ammonia on mortality and feeding of postlarvae shrimp Litopenaeus vannamei Bulletin of Envir- onmental Contamination and Toxicology 65, 98^103 GonzaŁlez-Ocampo H., Morales L., CaŁceres-Mart|¤nez C., Aguirre H., HernaŁndez-VaŁzquez S.,Troyo-Dieguez E & Or- tega-Rubio A (2006) Shrimp aquaculture environmental diagnosis in the semiarid coastal zone in Mexico Frese- nius Environmental Bulletin 15, 659^669.
Hernandez J., Bashan L & Bashan Y (2006) Starvation hanced phosphorous removal from wastewater by the mi- croalgae Chlorella spp co-immobilized with Azospirillum brazilense Enzyme and Microbial Technology 38, 190^198 Hussenot J.M.E (2003) Emerging e¥uent management stra- tegies in marine ¢sh-culture farms located in European coastal wetlands Aquaculture 226, 113^128.
en-Jackson C., Preston N & Thompson P.J (2004) Intake and discharge nutrient loads at three intensive shrimp farms Aquaculture Research 35, 1053^1061.
Jones A.B., Dennison W.C & Preston N.P (2001) Integrated treatment of shrimp e¥uent by sedimentation, oyster ¢l- tration and macroalgal absorption: a laboratory scale study Aquaculture 193, 155^178.
Juarez L.M (2008) Current status of shrimp aquaculture in Mexico Panorama Acu|¤cola 13, 49^53.
Trang 10Kuhn D.D., Boardman G.D., Craig S.R., Flick G.J Jr & Mclean
E (2009) Use of microbial £ocs generated from tilapia
ef-£uent as a nutritional supplement for shrimp, Litopenaeus
vannamei Panorama Acu|¤cola 14, 32^35.
Lezama-Cervantes C., Paniagua-Michel J.J & Zamora-Castro
J (2010) Bioremediacion of e¥uents ones of the culture of
Litopenaeus vannamei (Boone, 1931) using microbial mats
in a recirculating system Latin American Journal of
Aqua-tic Research 38, 129^142.
Li Y.,Veilleux D.J & Wikfors G.H (2009) Particle removal by
Northern bay scallops Argopecten irradians irradians in a
semi-natural setting: application of a £ow-cytometric
technique Aquaculture 296, 237^245.
Liu J.,Wang Z & Lin W (2010) De-eutrophication of e¥uent
wastewater from ¢sh aquaculture by using marine green
alga Ulva pertusa Chinese Journal of Oceanology and
Lim-nology 28, 201^208.
Marinho-Soriano E., Nunes S.O., Carneiro M.A & Pereira
D.C (2009) Nutrients removal from aquaculture
waste-water using the macroalgae Gracilaria birdiae Biomass
and Bioenergy 33, 327^331.
Martinez-Cordova L & Enriquez-Ocana F (2007) Study of
the benthic fauna in a discharge lagoon of a shrimp farm
with special emphasis on polychaeta Online Journal of
Biological Sciences 7, 12^17.
Mart|¤nez-Co¤rdova L.R (1988) Bioecolog|¤a de la almeja negra
Chione £uctifraga (Sowerby,1853) Revista de
Biolog|¤aTropi-cal 36, 213^219.
Mart|¤nez-Co¤rdova L.R (1999) Cultivo de Camarones Peneidos,
Principios y PraŁcticas AGT Editor, M e¤ xico DF, Mexico,298pp.
Martinez-Cordova L.R & Martinez-Porchas M (2006)
Polycul-ture of the Paci¢c white shrimp, Litopenaeus vannamei, giant
oyster, Crassostrea gigas, and black clam, Chione £uctifraga in
ponds in Sonora, Mexico Aquaculture 258, 321^326.
Mart|¤nez Co¤rdova L.R., Mart|¤nez-Porchas M & Corte¤s
Jacin-to E (2009) Camaronicultura Mexicana y Mundial:
„acti-vidad sustentable o industria contaminante? Revista
Internacional de Contaminacio¤n Ambiental 25, 181^196.
Mart|¤nez-Porchas M., Mart|¤nez-Co¤rdova L.R.,
Porchas-Cor-nejo M & Lo¤pez-El|¤as J.A (2010) Shrimp polyculture: a
po-tentially, pro¢table, sustainable but yet uncommon
aquacultural practice Reviews in Aquaculture 2,73^85.
Newell R.I.E & Jordan S.J (1983) Preferential ingestion of
or-ganic material by the American oyster, Crassostrea
virgini-ca Marine Ecology Progress Series 13, 47^53.
PaŁez-Osuna F., Gracia A., Flores-Verdugo F., Lyle-Fritch L.P.,
Alonso-Rodr|¤guez R., Roque A & Ruiz-FernaŁndez A.C.
(2003) Shrimp aquaculture development and the
environ-ment in the Gulf of California ecoregion Marine Pollution
Bulletin 46, 806^815.
Paniagua-Michel J & Garcia O (2003) Ex-situ
bioremedia-tion of shrimp culture e¥uent using constructed
micro-bial mats Aquacultural Engineering 28, 131^139.
Ray A.J., Lewis B.L., Browdy B.L & Le¥er J.W (2009)
Sus-pended solids removal to improve shrimp (Litopenaeus
vannamei) production and an evaluation of a plant-based
feed in minimal-exchange, superintensive culture tems Aquaculture 299, 89^98.
sys-Rivera-Monroy V.H.,Torres L.A., Bahamon L., Newmark F & Twilley R.R (1999) The potential use of mangrove forests
as nitrogen sinks of shrimp aquaculture pond e¥uents: the role of denitri¢cation Journal of theWorld Aquaculture Society 30, 12^25.
SaŁnchez-Mart|¤nez J.G., Aguirre-GuzmaŁn G & Mej|¤a-Ruiz H (2007) White spot syndrome virus in cultured shrimp: a review Aquaculture Research 38, 1339^1354.
Strickland J.D.H & Parsons T.R (1972) A Practical Handbook
of Seawater Analysis, Bulletin 167, 2nd edn Fisheries search Board Canada, Ottawa, ON, Canada, 310pp Summerfelt R.C & Penne C.R (2007) Septic tank treatment
Re-of the e¥uent from a small-scale commercial recycle aquaculture system North American Journal of Aquacul- ture 69, 59^68.
Tacon A.G.J (2002) Thematic review of feeds and feed ment practices in shrimp aquaculture Report prepared un- der the World Bank, NACA, WWF and FAO Consortium Program on Shrimp Farming and the Environment,Work
manage-in Progress for Public Discussion Published by the sortium, 69pp.
Con-Thomas Y., Courties C., El Helwe Y., Herbland A & nier H (2010) Spatial and temporal extension of eutrophi- cation associated with shrimp farm wastewater discharges in the New Caledonia lagoon Marine Pollution Bulletin 61, 387^398.
Lemon-Thompson F.L., Abreu P.C & Wasielesky W (2002) tance of bio¢lm for water quality and nourishment in in- tensive shrimp culture Aquaculture 203, 263^278 Timmons M.B., Ebeling J.M.,Wheaton F.W., Summerfelt S.T & Vinci B (2002) Sistemas de Recirculacio¤n para la Acuicultura Fundacio¤n Chile,Vitacura, Santiago de Chile, Chile,748pp Tinoco-Orta G.D & CaŁceres-Mart|¤nez J (2003) Infestation of the clam Chione £uctifraga by the burrowing worm Poly- dora sp nov in laboratory conditions Journal of Inverte- brate Pathology 83, 196^205.
Impor-Unrein F & Vincour A (1999) Phytoplankton structure and dynamics in a turbid Antarctic lake (Potter Peninsula, King George Island) Polar Biology 22, 93^101.
Vymazaj J (1988) The use of periphyton communities for trient removal from polluted streams Hydrobiologia 166, 225^237.
nu-Wood E.D., Armstrong F.A.G & Richards F.A (1967) mination of nitrate in seawater by cadmium-copper reduction to nitrite Journal of Marine Biology Association (United Kingdom) 47, 23–31.
Deter-Xu Y., Fang J & Wei W (2008) Application of Gracilaria noides (Rhodophyta) for alleviating excess nutrients in aquaculture Journal of Applied Phycology 20, 199^203 Zhou Y., Yang H., Hu H., Liu Y., Mao Y., Zhou H., Xu X & Zhang F (2006) Bioremediation potential of the macroal-
liche-ga Gracilaria lemaneiformis (Rhodophyta) integrated into fed ¢sh culture in coastal waters of north China Aquacul- ture 252, 264^276.
Aquaculture Research, 2011, 42, 1415^1423 Bioremediation and reuse of e¥uents L R Mart|¤nez-Co¤rdova et al.
Trang 11Evaluation of corn distillers dried grains with solubles and brewers yeast in diets for channel catfish
Menghe H Li, Daniel F Oberle & Penelope M Lucas
Thad Cochran National Warmwater Aquaculture Center, Mississippi State University, Stoneville, MS, USA
Correspondence: M Li,Thad Cochran National Warmwater Aquaculture Center, Mississippi State University, PO Box 197, Stoneville, MS
38776, USA E-mail: mli@drec.msstate.edu
Abstract
A study was conducted to examine the use of
distil-lers grains with solubles (DDGS), ethanol extracted
DDGS (EDDGS), and brewers yeast in channel cat¢sh,
Ictalurus punctatus, diets Diets containing these
in-gredients were compared with all-plant and ¢sh meal
control diets Juvenile channel cat¢sh (initial weight:
9.1 0.2 g ¢sh 1) were stocked in £ow-through
aquaria and fed one of six practical diets for 8 weeks
Diets containing1% brewers yeast or 30% DDGS
sup-ported the same level of growth and feed e⁄ciency
ratio (FER) as the diet containing 5% ¢sh meal
Etha-nol extraction e¡ectively removed most of the fat and
yellow pigments in DDGS The diet containing 30%
EDDGS resulted in signi¢cantly lower growth and
FER compared with the diet containing DDGS
How-ever, the weight gain of ¢sh fed the EDDGS diet was
intermediate compared with ¢sh fed the all-plant
control, ¢sh meal control, and 1% and 2% brewers
yeast diets The EDDGS could potentially be used at
high levels as a substitution for soybean meal
with-out causing yellow pigment deposition in cat¢sh
£esh, provided that the ethanol extraction process is
proven economical Brewers yeast, used at 1^2% of
the diet, appears to be e¡ective at improving weight
gain and FER of channel cat¢sh over the all-plant diet
Keywords: channel cat¢sh, distillers dried grains,
brewers yeast, growth, feed e⁄ciency
Introduction
Distillers dried grains with solubles (DDGS) from
corn is a by-product of ethanol production It is
rela-tively high in protein (27%) and highly palatable to
channel cat¢sh, Ictalurus punctatus With the rapidexpansion of ethanol production in the United States,the prices for DDGS have become more competitivecompared with soybean meal and other plant proteinsources Use of this by-product in cat¢sh feeds wouldreduce feed cost
Several studies have been reported on the use ofDDGS in channel cat¢sh diets Early studies demon-strated that up to 35% DDGS without lysine supple-mentation (Webster, Tidwell & Yancey 1991; Webster,Tidwell, Goodgame, Yancey & Mackey 1992; Webster,Tidwell, Goodgame & Johnsen 1993) and up to 70%with lysine supplementation (Webster et al 1991)could be used to partially replace soybean mealand ¢sh meal in channel cat¢sh diets without a¡ect-ing ¢sh growth Recently, Lim, Yildirim-Aksoy andKlesius (2009) also reported no di¡erences in thegrowth of juvenile channel cat¢sh fed diets contain-ing up to 40% DDGS with supplemental lysine In apond study with channel cat¢sh, Robinson and Li(2008) found that up to 30^40% DDGS with supple-mental lysine could be used in food ¢sh diets Theynoted that feed e⁄ciency ratio (FER) was improved
in ¢sh fed diets containing 30^40% DDGS However,
it was not clear whether the improved FER of ¢sh feddiets containing DDGS was caused by the increaseddietary fat level, because of high levels of fat (about9%) contained in the DDGS, or by other compoundspresent in the product
Li, Robinson, Oberle and Lucas (2010) examinedthe use of several corn distillers by-products includ-ing DDGS, distillers solubles and high-protein DDGS
in diets and the e¡ects of additional dietary fat onjuvenile channel cat¢sh performance They foundthat elevated fat levels in diets containing distillers
Trang 12by-products were only partially responsible for the
improvement in FER of ằsh fed the distillers
by-products The presence of the distillers solubles in
the diet, possibly due to the brewers yeast,
Saccharo-myces cerevisiae, further improved FER, and also
improved weight gain over the control diets with or
without additional fat
Distillers grains with solubles contain up to three
times of the amount yellow pigments lutein and
zeax-anthin found in yellow corn The high level of yellow
pigments may limit its use in catằsh diets because
high dietary yellow pigment levels can result in
pigment deposition in the ặesh, rendering it less
appealing to the general consumer in the United
States (Lee 1987; Li, Robinson & Oberle 2009) If levels
of pigmented compounds in DDGS can be reduced,
more DDGS could be used without adversely
a¡ect-ing marketability of the catằsh product Therefore,
the present study was conducted to examine the
e¡ect of DDGS, de-pigmented DDGS and brewers
yeast in the diet on the growth, FER and body
proxi-mate composition of juvenile channel catằsh
Materials and methods
Six practical diets containing 28% crude protein and
5% crude fat (Table 1) were formulated to meet or
exceed all known nutrient requirements of channel
catằsh (National Research Council 1993) Diet
descrip-tions follow:
Diet 1 ^ all-plant control diet
Diet 2 ^ ằsh meal control diet (similar to Diet 1
ex-cept with 5% menhaden ằsh meal in replacement
of part of soybean meal)
Diet 3 ^ 30% ethanol extracted DDGS (EDDGS)
Diet 4 ^ 30% DDGS
Diet 5 ^ 1% brewers yeast (similar to Diet 1 except
with brewers yeast in replacement of part of
soy-bean meal)
Diet 6 ^ 2% brewers yeast (similar to Diet 1 except
with brewers yeast in replacement of part of
soy-bean meal)
The DDGS was provided by Poet, LLC (Sioux Falls,
SD, USA) and Brewtechsdried brewers yeast was
provided by International Ingredient Corporation
(St Louis, MO, USA) Remaining dietary ingredients
were obtained from the Delta Western Feed Mill
(Indianola, MS, USA) and were from commercial
sources The DDGS was extracted with hexane and
ethanol, respectively, in a manner similar to the ether
extraction method described by Association of
O⁄-cial Analytical Chemists International (AOAC)(2000) using the Soxtec System (Foss North America,Eden Prairie, MN, USA) The resulting material wasdried at 60 1C for 60 min to evaporate the solventresidue The hexane extracted DDGS was not used inthe feeding study because only a small amount ofyellow pigment was removed during the process.The experimental diets were prepared as sinkingpellets according to procedures described previously(Li, Johnson & Robinson 1993)
Juvenile channel catằsh were obtained from theUSDA Agriculture Research ServiceỖs Catằsh Genet-ics Research Unit (Stoneville, MS, USA) Thirty ằshwere stocked into each of thirty 110 L ặow-throughaquaria at the Thad Cochran National WarmwaterAquaculture Center (NWAC), Mississippi StateUniversity (Stoneville, MS, USA) The aquaria weresupplied with well water (ặow rate: approximately
1 L min 1) and continuous aeration Water ture and dissolved oxygen were monitored in thesystem once daily using a YSI oxygen meter (YellowSprings Instruments, Yellow Springs, OH, USA)and averaged at 29.8 0.2 1C and 6.8 0.2 mg L 1respectively A diurnal light:dark cycle was regulated
tempera-at 14:10 h
Before initiation of the experiment, the ằsh wereacclimated for 2 weeks and fed an all-plant con-ditioning diet once daily to apparent satiation at08:00 hours After acclimation, all ằsh were pooledand graded to a uniform size, and 15 ằsh were col-lectively weighed and restocked in each aquarium.Initial ằsh weight was determined and averaged9.1 0.2 g ằsh 1 (mean SD) Fish were fed toapparent satiation (in about 40 min) once daily for
8 weeks Satiation was achieved by ằrst feeding anamount of diet based on the percentage of ằsh bodyweight (less than satiation), followed by feedingseveral times from a pre-weighed diet container.Diet consumption was monitored and recorded ateach feeding Dead ằsh, if any, were removed dailyfrom the aquarium and weighed Aquaria werecleaned weekly
At the end of the feeding period, feed consumptionand weight gain per ằsh, FER and survival were calcu-lated Feed e⁄ciency ratio was determined as follows:
FERỬ đơfinal fish weight; g tank1
ơinitial fish weight; g tank1
ợ ơweight of dead fish; g tank1ỡ=đtotal feed fed; g tank1ỡ
Aquaculture Research, 2011, 42, 1424^1430 Distillers grains, brewers yeast in catằsh diets M H Li et al.
Trang 13After the ¢nal ¢sh number and weight were
deter-mined, ¢ve ¢sh from each aquarium were euthanized
by an overdose (500 mg L 1) of tricaine
methanesul-phonate (MS-222TM, Argent Chemical Laboratories,
Redmond,WA, USA), and ¢llet samples were removed
Digital pictures were taken of one ¢llet from each ¢sh
using an EOS 1D Mark II digital SLR camera (Cannon
USA, Lake Success, NY, USA) The yellow intensity
va-lues [Commission Internationale de I’Eclairage (CIE)
b(negative: blueness; positive: yellowness)] were
de-termined from the digital picture of the ¢llet at three
locations along the dorsal line of the ¢llet using
Adobe Photoshop CS3 image editing software (Adobe
Systems, San Jose, CA, USA) The ¢llets were then
pooled by aquarium, and stored at 80 1C for
subse-quent proximate and pigment analyses
The ¢llet samples were homogenized into a paste
using a Grindomix GM-200 Knife Mill (Retsch
GmbH, Haan, Germany) and part of the sample was
lyophilized with a Freezone Freeze Dry System
(Lab-conco, Kansas City, MO, USA) for 16^18 h for protein
and fat analyses
Proximate analyses were performed in duplicate
on diet and pooled ¢llet samples from each aquarium
with methods described by AOAC (2000) Crude
pro-tein of diet and ¢llet samples was analysed by thecombustion method with the FP-2000 protein deter-minator (Leco, St Joseph, MI, USA), crude fat by etherextraction with the Soxtec System (Foss North Amer-ica, Eden Prairie, MN, USA) and moisture by oven dry-ing with a mechanical convection oven (Precision,Winchester, VA, USA) Diet and ¢llet samples wereanalysed for lutein and zeaxanthin concentrationsusing high-performance liquid chromatography(Moros, Darnoko, Cheryan, Perkins & Jerrell 2002).Data were subjected to one-way analysis of var-iance (ANOVA) and the Fisher’s protected least signi¢-cant di¡erence procedure (Steel, Torrie & Dickey1997) with theSTATISTICAL ANALYSIS SYSTEMversion 9.1software (SAS Institute 2004) Aquaria were the ex-perimental units and variation among aquaria with-
in a treatment was used as the experimental error intests of signi¢cance Ana level of 0.05 was used
ResultsHexane extraction method used in the present studyremoved almost all fat (from 9.49% to 0.06%) inDDGS, but was not e¡ective in removing the yellow
Table 1 Ingredient and proximate compositions of experimental diets (expressed percentage on an as-fed basis)
All-plant control
Fish meal control
Dry matter 87.9 0.02 88.67 0.01 88.30 0.02 87.80 0.02 88.37 0.03 87.64 0.06 Crude proteinz 27.7 0.10 27.59 0.11 28.22 0.10 28.25 0.18 27.38 0.09 27.35 0.18 Crude fatz 4.68 0.02 4.64 0.04 4.50 0.00 5.01 0.01 4.49 0.00 4.49 0.03 Lutein1zeaxanthinz (mg kg 1 ) 3.91 0.00 4.16 0.11 3.54 0.10 9.78 0.48 3.90 0.06 4.00 0.06 DE:P ratio (kcal g 1 protein) k 10.2 10.2 9.2 9.5 10.2 10.2
Ethanol extracted distillers dried grains with solubles.
wDistillers dried grains with solubles.
zIncludes 7.5% cottonseed meal, 1% menhaden ¢sh oil, 2% carboxymethyl cellulose (pellet binder), 0.05% vitamin premix, 0.05% L -ascorbyl monophosphate and 0.1% trace mineral premix.Vitamin and trace mineral premixes were the same as described by Robinson and Li (2007).
‰Values represent mean SD (n 5 2, two batches per diet).
zExpressed as 900 g kg 1 dry matter basis.
kEstimated digestibility to protein ratio.
Trang 14pigments lutein and zeaxanthin (from 30.9 to
26.7 mg kg 1) Ethanol extraction of DDGS removed
most of the fat (from 9.49% to 1.89%) and almost all
yellow pigments (from 30.9 to 0.6 mg kg 1)
The feeding study showed no signi¢cant
di¡er-ences among dietary treatments for feed
consump-tion and survival (Table 2) The overall mortality was
2.5% and the cause was not known Weight gain of
¢sh fed diets containing 30% DDGS, and 1% and 2%
brewers yeast was signi¢cantly higher than that of
¢sh fed the all-plant control diet, but not signi¢cantly
di¡erent from that of ¢sh fed the ¢sh meal control diet
(5% menhaden meal).Weight gain of ¢sh fed the diet
containing 30% EDDGS was signi¢cantly lower than
that of ¢sh fed the diet containing 30% DDGS, but
was intermediate with that of ¢sh fed the all-plant
and ¢sh meal control diets Feed e⁄ciency ratio of ¢sh
fed diets containing 30% DDGS and1% brewers yeast
was signi¢cantly higher than that of ¢sh fed the
all-plant control diet and the diet containing 30%
EDDGS, but not signi¢cantly di¡erent from that of
¢sh fed the ¢sh meal control diet Feed e⁄ciency ratio
of ¢sh fed the diet containing the 2% brewers yeast
diet was intermediate, not signi¢cantly di¡erent from
that of ¢sh fed other diets
Fish fed the ¢sh meal control diet had a
signi¢-cantly higher protein, but had similar levels of fat
and moisture compared with ¢sh fed the all-plant
control diet (Table 3) Fish fed diets containing 30%
EDDGS and 30% DDGS had similar levels of ¢llet
pro-tein, fat and moisture levels Fillet protein levels of
¢sh fed the EDDGS and DDGS diets were similar to
that of ¢sh fed the all-plant control diet, but were
low-er than that of ¢sh fed the ¢sh meal control diet andbrewers yeast diets Fillet fat levels of ¢sh fed theEDDGS diet were similar to that of ¢sh fed the all-plant control and ¢sh meal control diets, but werelower than that of ¢sh fed the brewers yeast diets Fil-let moisture levels of ¢sh fed the EDDGS diet werehigher than that of ¢sh fed the all-plant control, ¢shmeal control and brewers yeast diets Fillet protein,fat and moisture levels of ¢sh fed the 1% and 2%brewers yeast diets were similar
Fish fed the diet containing 30% EDDGS had a ni¢cantly lower CIE bvalue than ¢sh fed the all-plant control diet and the diet containing 30% DDGS(Table 4) The CIE bvalue of ¢sh fed the all-plant con-trol diet was signi¢cantly lower than that of ¢sh fedthe 30% DDGS diet Lutein1zeaxanthin levels in the
sig-£esh of ¢sh fed the all-plant control and the diet taining 30% EDDGS were signi¢cantly lower thanthat of ¢sh fed the 30% DDGS diet
con-Discussion
Results from the present study support the tion by Li et al (2010) that the use of 30% DDGS inthe diet improved weight gain and FER over anall-plant control diet In addition, the present studydemonstrated that the 30% DDGS diet provided thesame level of growth and FER as the ¢sh meal controldiet Li et al (2010) suggests that the improvement ofweight gain and FER by feeding 30% DDGS is likelycaused by the presence of distillers solubles, possiblydue to the brewers yeast In the present study, ¢sh
observa-Table 2 Mean feed consumption, weight gain, feed e⁄ciency ratio and survival of juvenile channel cat¢sh fed various mental diets for 8 weeks
experi-Diet description
Feed consumption (g fish 1)w
Weight gain (g fish 1)z
Feed efficiency
All-plant control 102.1 61.3 c 0.600 b 96.0 Fish meal control 103.3 68.7 ab 0.665 a 98.3 30% EDDGS‰ 106.6 64.8 bc 0.607 b 98.7 30% DDGSz 108.7 71.3 a 0.656 a 98.7 1% brewers yeast 104.9 68.2 ab 0.651 a 100.0 2% brewers yeast 110.1 68.7 ab 0.624 ab 93.3
Means represent average values of ¢ve tanks per diet Means within each column followed by di¡erent letters were di¡erent (P 0.05, the Fisher’s protected least signi¢cant di¡erence procedure).
wBased on 900 g kg 1
dry matter of the diet.
zInitial weight was 9.1 0.2 g ¢sh 1
.
‰Ethanol extracted distillers dried grains with solubles.
zDistillers dried grains with solubles.
Aquaculture Research, 2011, 42, 1424^1430 Distillers grains, brewers yeast in cat¢sh diets M H Li et al.
Trang 15fed 1% and 2% brewers yeast had similar weight
gain and FER as ¢sh fed the 30% DDGS diet This
indirectly con¢rms that brewers yeast in the DDGS
may play a role in the improvement of ¢sh growth
performance
During ethanol production, brewers yeast is
typi-cally used to ferment corn to produce ethanol
Distil-lers dried grains with solubles have been estimated to
contain about 3.9% yeast cells (Ingledew 1999)
Pre-vious studies with sea bass, Dicentrarchus labrax
(Oliva-Teles & Goncalves 2001) and hybrid striped
bass, Morone chrysops Morone saxatilis (Li & Gatlin
III 2005), showed that the use of brewers yeast in the
diet improved ¢sh growth and FER Yeast cells
con-tain 5^12% nucleic acids from which nucleotides
are derived (Tacon & Jackson1985) The brewers yeastused in the present study contained 3.0% nucleotides(analysed by Euro¢ns Scienti¢c, Des Moines, IA, USA).The nucleotides included adenosine-, cytidine-,guanosine-, uridine-50-monophosphate and a traceamount of inosine-50-monophosphate Dietary sup-plementation of a mixture of nucleotides has beenshown to increase the height of villa in the intestine
of rat (Uauy, Stringel, Thomas & Quan 1990) andAtlantic salmon, Salmo salar (Burrells, Williams,Southgate & Wadsworth 2001) As a result, the mu-cosal surface area of the intestine is increased andtherefore nutrients are more e⁄ciently absorbed andutilized Li, Gatlin III and Neil (2007) reported thatdietary supplementation of a mixture of nucleotidesenhanced the growth and FER of red drum, Sciaenopsocellatus during the ¢rst week of feeding, but the im-provement diminished during the following 3 weeks
of feeding Lin,Wang and Shiau (2009) found that theaddition of individual and a mixture of nucleotides inthe diet improved the growth and FER in grouper, Epi-nephelus malabaricus after 8 weeks of feeding.Nucleotides such as adenosine-, inosine- and uri-dine-50-monophosphate have been shown to stimu-late gustatory sensory cells in several ¢sh species(Ishida & Hidaka 1987; Ikeda, Hosokawa, Shimeno &Takeda 1991; Kubitza, Lovshin & Lovell 1997) Li et al.(2010) reported that the use of DDGS or distillerssolubles in the diet increased feed intake in channelcat¢sh, which they attributed to the possible chemo-attractive e¡ects of nucleotides present in the yeastcells However, no signi¢cant di¡erences in feedconsumption were observed among ¢sh fed variousdiets in the present study The discrepancy betweenresponses of the present study and Li et al (2010) can-not be easily explained, but could be due to the largervariation in feed consumption of ¢sh in various repli-cated tanks in the present study
In the present study, ¢sh did not perform well onthe all-plant control diet as compared with ¢sh feddiets containing ¢sh meal, DDGS and 1% brewersyeast Although results from previous studies are in-conclusive, more evidence appears to indicate thatthe inclusion of ¢sh meal in the diet improves thegrowth and FER of juvenile channel cat¢sh (Mohsen
& Lovell 1990; Li, Peterson, Janes & Robinson 2006;
Li, Robinson, Peterson & Bates 2008) Fish muscle isrich in nucleotides (Ikeda, Hosokawa, Shimeno &Takeda 1988), which may contribute to the growth-enhancing e¡ect of ¢sh meal on channel cat¢sh.Several studies have demonstrated that relativelyhigh levels of DDGS can be used in channel cat¢sh
Table 3 Mean ¢llet protein, fat and moisture
concentra-tions of juvenile channel cat¢sh fed various experimental
diets for 8 weeks
Diet description
Fillet protein (%)w
Fillet fat (%)w
Fillet moisture (%)
Means represent average values of ¢ve tanks with ¢ve ¢sh per
tank Means within each column followed by di¡erent letters
were di¡erent (P 0.05, the Fisher’s protected least signi¢cant
di¡erence procedure).
wOn wet-tissue basis.
zEthanol-extracted distillers dried grains with solubles.
‰Distillers dried grains with solubles.
Table 4 Mean CIE bvalue and lutein plus zeaxanthin
concentrations of juvenile channel cat¢sh fed experimental
diets containing distillers dried grains with solubles (DDGS)
and ethanol extracted distillers dried grains with soluble
(EDDGS) for 8 weeks
Diet description CIE b
Lutein1zeaxanthin (lg g 1)w
All-plant control 15.1 b 0.83 b
30% EDDGS 13.7 c 0.83 b
30% DDGS 18.2 a 1.03 a
Pooled SEM 0.4 0.01
Means represent average values of ¢ve tanks with ¢ve ¢sh per
tank Means within each column followed by di¡erent letters
were di¡erent (P 0.05, the Fisher’s protected least signi¢cant
di¡erence procedure).
wOn wet-tissue basis.
Trang 16diets without adversely a¡ecting ¢sh performance
(Webster et al 1991; Robinson & Li 2008; Lim et al
2009), but there is a concern about high yellow
pigment levels in DDGS Studies have shown that
feeding diets containing lutein1zeaxanthin levels
higher than 7^10 mg kg 1 can deposit enough
pigment in cat¢sh £esh to be visible (Lee 1987; Li
et al 2009) Fish fed the DDGS diet in the present
study only showed slightly yellow colouration by
visual examination Also, the CIE b values and
lutein1zeaxanthin concentration in ¢sh £esh were
at low levels This is mostly due to the short feeding
period and small ¢sh size used in the present study
However, both the CIE b value and
lutein1zeax-anthin concentration in ¢sh fed EDDGS were
signi¢-cantly lower than in ¢sh fed DDGS This is anticipated
because yellow pigments lutein and zeaxanthin in
the EDDGS were e¡ectively removed by ethanol
extraction Analyses of yellow pigments conducted
at our laboratory on various batches of DDGS showed
that the pigment varied from 22 to 40 mg kg 1
Longer periods of feeding of DDGS containing a high
level of the pigment at 30% would result in yellow
pigment deposition that may reduce marketability of
the cat¢sh product
Ethanol extraction does not appear to extract the
nucleotides from the DDGS Total nucleotide (the same
four main nucleotides found in brewers yeast)
concen-trations were 0.17% for DDGS and 0.25% for EDDGS
However, ¢sh fed the 30% EDDGS diet had lower
weight gain and FER than ¢sh fed the 30% DDGS diet
This response cannot be easily explained with regard
to the nucleotide concentrations of the diets
Signi¢cant di¡erences were observed in ¢llet
prox-imate composition of ¢sh fed various experimental
diets The di¡erences cannot be easily explained
However, these di¡erences were relatively small and
¢llet protein, fat and moisture levels were within the
normal range for this size of ¢sh Although all diets
were formulated to be isonitrogenous and isolipidic,
di¡erences in dietary ingredient composition,
diges-tible energy and other nutrients may have a¡ected
the nutrient retention of these ¢sh
With the method used in the present study, ethanol
extraction e¡ectively removed most of yellow
pig-ments in DDGS, while hexane did not Further
investi-gations are needed to optimize extraction conditions
using hexane as a solvent to remove oil and pigments
because hexane is commonly used as the solvent in
extracting oil from oilseeds and other feedstu¡s
In summary, the present study demonstrates that
diets containing 1% brewers yeast or 30% DDGS
sup-port the same level of growth and FER as a diet taining 5% ¢sh meal for juvenile channel cat¢sh Thediet containing EDDGS resulted in signi¢cantly lowergrowth and FER compared with the diet containingDDGS However, the weight gain of ¢sh fed theEDDGS diet was intermediate compared with ¢shfed the all-plant control, ¢sh meal control and 1^2%brewers yeast diets Ethanol extracted DDGS couldpotentially be used at high levels as a substitution ofsoybean meal without causing yellow pigment de-position in cat¢sh £esh, provided that the ethanol ex-traction process is proven economical Brewers yeast,used at 1^2% of the diet, appears to improve weightgain and FER over the all-plant control diet used inthe present study
con-Acknowledgments
We thank Sandra Philips and Cli¡ Smith for dailymanagement of the experiment Special thanks toPoet, LLC of Sioux Falls (South Dakota, USA), whichprovided the distillers dried grains with solubles andInternational Ingredient Corporation (St Louis, MO,USA) for providing the brewers yeast This manu-script is approved for publication as Journal Article
No J-11855 of the Mississippi Agricultural and estry Experiment Station, Mississippi State Univer-sity This project is supported under Project NumberMIS-371390 through a grant from USDA National In-stitute of Food and Agriculture
For-References
Association of O⁄cial Analytical Chemists International (AOAC) (2000) O⁄cial Methods of Analysis, 17th edn AOAC International, Gaithersburg, MD, USA.
Burrells C., Williams P.D., Southgate P.J & Wadsworth S.L (2001) Dietary nucleotides: a novel supplement in ¢sh feeds 2 E¡ects on vaccination, salt water transfer, growth rates and physiology of Atlantic salmon (Salmo salar L) Aquaculture 199, 171^184.
Ikeda I., Hosokawa H., Shimeno S & Takeda M (1988) ti¢cation of feeding stimulates for jack mackerel in its muscle extract Nippon Suisan Gakkaishi 54, 229^233 Ikeda I., Hosokawa H., Shimeno S & Takeda M (1991) Feed- ing stimulant activity of nucleotides, tryptophan, and their related compounds of jack mackerel Nippon Suisan Gakkaishi 57, 1539^1542.
Iden-Ingledew W.M (1999) Yeast ^ could you base business on this bug? In: Under the Microscope ^ Focal Points for the New Millennium ^ Biotechnology in the Feed Industry, Pro- ceedings of Alltech’s 15th Annual Symposium (ed by T.P Aquaculture Research, 2011, 42, 1424^1430 Distillers grains, brewers yeast in cat¢sh diets M H Li et al.
Trang 17Lyons & K.A Jacques), pp 27^47 Nottingham University
Press, Nottingham, UK.
Ishida Y & Hidaka I (1987) Gustatory responses pro¢les for
amino acids, glycine, betaine, and nucleotides in several
marine teleosts Nippon Suisan Gakkaishi 53, 1391^1398.
Kubitza F., Lovshin L.L & Lovell R.T (1997) Identi¢cation of
feed enhancers for largemouth bass Micropterus
sal-moides Aquaculture 148, 191^200.
Lee P.H (1987) Carotenoids in cultured channel cat¢sh PhD
dissertation Auburn University, AL, USA.
Li M., Robinson E & Oberle D (2009) Yellow pigments in
cat-¢sh Cat¢sh Journal 23, 11^14.
Li M.H., Johnson M.R & Robinson E.H (1993) Elevated
diet-ary vitamin C concentrations did not improve resistance
of channel cat¢sh, Ictalurus punctatus, against
Edwardsiel-la ictaluri infection Aquaculture 117, 303^312.
Li M.H., Peterson B.C., Janes C.L & Robinson E.H (2006)
Comparison of diets containing various ¢sh meal levels
on growth performance, body composition, and
insulin-like growth factor-I of juvenile channel cat¢sh Ictalurus
punctatus of di¡erent strains Aquaculture 253, 628^635.
Li M.H., Robinson E.H., Peterson B.C & Bates T.D (2008)
Growth and feed e⁄ciency of juvenile channel cat¢sh
reared at di¡erent water temperatures and fed diets
con-taining various levels of ¢sh meal North American Journal
of Aquaculture 70, 347^352.
Li M.H., Robinson E.H., Oberle D.F & Lucas P.M (2010)
Ef-fects of various corn distillers by-products on growth
and feed e⁄ciency of channel cat¢sh, Ictalurus punctatus.
Aquaculture Nutrition 16, 188^193.
Li P & Gatlin D.M III (2005) Evaluation of the prebiotic
Gro-Biotics^ A and brewers yeast as dietary supplements for
sub-adult hybrid striped bass (Morone chrysops M
sax-atilis) challenge in situ with Mycobacterium marinum.
Aquaculture 248, 197^205.
Li P., Gatlin D.M III & Neil W.H (2007) Dietary
supplemtation of a puri¢ed nucleotide mixture transiently
en-hanced growth and feed utilization of juvenile red drum,
Sciaenops ocellatus Journal of theWorld Aquaculture Society
38, 281^286.
Lim C., Yildirim-Aksoy M & Klesius P.H (2009) Growth
re-sponse and resistance to Edwardsiella ictaluri of channel
cat¢sh, Ictalurus punctatus, fed diets containing distillers
dried grains with solubles Journal of theWorld Aquaculture
Society 40, 182^193.
LinY.H.,Wang H & Shiau S.Y (2009) Dietary nucleotide
sup-plementation enhances growth and immune responses of
grouper, Epinephelus malabaricus Aquaculture Nutrition
15, 117^122.
Mohsen A.A & Lovell R.T (1990) Partial substitution of bean meal with animal protein sources in diets for chan- nel cat¢sh Aquaculture 90, 303^311.
soy-Moros E.E., Darnoko D., Cheryan M., Perkins E.G & Jerrell J (2002) Analysis of xanthophylls in corn by HPLC Journal
of Agriculture and Food Chemistry 50, 5787^5790 National Research Council (1993) Nutritional Requirements
of Fish National Academy Press,Washington, DC, USA Oliva-Teles A & Goncalves P (2001) Partial replacement of
¢shmeal by brewers yeast (Saccharomyces cerevisiae) in diet for sea bass (Dicentrarchus labrax) juveniles Aquacul- ture 202, 269^278.
Robinson E.H & Li M.H (2007) Cat¢sh Feeds and Feeding Mississippi Agricultural and Forestry Experiment Station Bulletin No 1163 Mississippi Agricultural and Forestry Experiment Station, Mississippi State, MS, USA Robinson E.H & Li M.H (2008) Replacement of soybean meal in channel cat¢sh, Ictalurus punctatus, diets with cottonseed meal and distillers dried grains with solubles Journal of theWorld Aquaculture Society 39, 521^527 SAS Institute (2004) SAS/STAT 9.1 User’s Guide SAS Insti- tute, Cary, NC, USA.
Steel R.G.,Torrie J.H & Dickey D.A (1997) Principles and cedures of Statistics, A Biometric Approach, 3rd edn McGraw-Hill Companies, New York, NY, USA.
Pro-Tacon A.G.J & Jackson A.J (1985) Utilization of conventional and unconventional protein sources in practical ¢sh feeds In: Nutrition and Feeding in Fish (ed by C.B Cowey, A.M Mackie & J.G Bell), pp 119^145 Academic Press, London, UK.
Uauy R., Stringel G., Thomas R & Quan R (1990) E¡ect of dietary nucleotides on growth and maturation of the de- veloping gut in the rat Journal of Pediatric Gastroenterol- ogy and Nutrition 10, 497^503.
Webster C.D.,Tidwell J.H & Yancey D.H (1991) Evaluation of distillers’grains with solubles as a protein source in diets for channel cat¢sh Aquaculture 96, 179^190.
Webster C.D., Tidwell J.H., Goodgame L.S., Yancey D.H & Mackey L (1992) Use of soybean meal and distillers grains with solubles as partial or total replacement of ¢sh meal
in diets for channel cat¢sh, Ictalurus punctatus ture 106, 301^309.
Aquacul-Webster C.D., Tidwell J.H., Goodgame L.S & Johnsen P.B (1993) Growth, body composition, and organolepitic eva- luation of channel cat¢sh fed diets containing di¡erent percentages of distillers’ grains with solubles Progressive Fish-Culturists 55, 95^100.
Trang 18A new system for the culture and stock enhancement
in cofferdams
Libin Zhang1, Hongsheng Yang1, Qiang Xu1, Kun Xing1,3, Peng Zhao1,2& Chenggang Lin1,4
1 Key Laboratory of Marine Ecology and Environmental Sciences, Institute of Oceanology, Chinese Academy of Sciences, Qingdao, China
2 Chinese Academy of Sciences, Graduate University, Beijing, China
3 Dalian Fisheries University, Dalian, China
4 College of Marine Life Sciences, Ocean University of China, Qingdao, China
Correspondence: H Yang, Key Laboratory of Marine Ecology and Environmental Sciences, Institute of Oceanology, Chinese Academy of Sciences, Nanhai Road 7, Qingdao 266071, Shandong Province, China E-mail: hshyang@ms.qdio.ac.cn
Abstract
A new multilayer, plate-type system for the culture
and stock enhancement of sea cucumbers in
co¡er-dam was developed To optimize and evaluate the
sys-tem, four experimental designs were implemented
using polyethylene (PE)-corrugated sheets of various
colours, interval spacing and shapes/styles Results
showed that a system equipped with black
PE-corru-gated sheets attracted more animals than either blue,
green, transparent or a selection of mixed sheets (six
transparent sheets in the upper layer and ¢ve black
sheets in the lower layer) (Po0.05) Also, more
ani-mals gathered in the system with oblique-angled
sheets (301 to the base plate) than either a wavy (the
bottom and every second sheet was at an angle of 101
to the base plate) or parallel arrangement (Po0.05),
and more animals assembled in the system with
2 cm between sheets than spacings of 3, 4 or 5 cm
(Po0.05) As expected, the upper layers of the
sys-tems attracted more animals than lower layers in
most cases except for those with transparent and
mixed oblique-angled sheets with a 3 cm spacing
(Po0.05) Thus, a system with black,
oblique-angled-corrugated sheets and 2 cm spacing is recommended
forApostichopus japonicus (Selenka) culture and stock
enhancement in co¡erdams or ponds
Keywords: sea cucumber, Apostichopus japonicus,
culture system, stock enhancement
IntroductionSince the early1980s, due to its high nutritional valueand the development of successful hatchery techni-ques (Liao 1997), farming of the sea cucumber, Apos-tichopus japonicus Selenka (Liao 1980), has become asigni¢cant mariculture sector in North China (Chen2004; Yuan, Yang, Zhou, Mao, Zhang & Liu 2006).This species of sea cucumber inhabits reefs or grittysubstrates with a gentle current of high-quality sea-water, an abundance of natural food and no fresh-water input (Chang, Ding & Song 2004) Thus, theaddition of structures, such as arti¢cial substrates orreefs, is a simple method of habitat improvement andstock enhancement The functions of such systemsare (1) to protect broodstock and larvae from preda-tors, (2) to increase the availability of benthic algaeand organic debris and (3) to improve the habitat foraestivation and hibernation (Chen 2003)
In China, many materials have been utilized as ti¢cial substrates or reefs for sea cucumber culture,such as stones (Chen 2004, 2007; Li, Hao, Gao & Yin2004; Sun & Chen 2006), tiles (Chen 2003; Wang,Zhang, Zhang, Qu & Song 2004; Qin, Dong, Niu,Tian,Wang, Gao & Dong 2009), concrete structures (Zhao1995; Sun 2004; Qin et al 2009), scallop lantern nets(Li & Huo 2007), woven fabrics (Lin 2007), plasticcomponents (Li & Huo 2007) and even Chinese oakbranches (Yang & Shan 2007) Although thesesubstrates were mainly developed through trial and
Trang 19error, they had no regular structure and could not be
easily removed from ponds, for example, for cleaning
and harvesting
Many factors in£uence habitat selection in
ani-mals, such as habitat features, characteristics of the
animal and the presence/absence of food, predation,
competition, etc (Riechert 1976; Abramsky, Al¢a,
Schachak & Brand 1990; Ward & Porter 1993; Yan &
Chen 1998) Apostichopus japonicus, as a sedimentary
feeder, ingests organic matter, including bacteria,
protozoa, diatoms and detritus from plants and
ani-mals (Zhang, Sun & Wu 1995; Yang, Yuan, Zhou,
Mao, Zhang & Liu 2005) and re-utilizes residual food
and faeces (Yang, Wang, Zhou, Zhang, Wang, He
& Zhang 2000; Yang, Zhou, Wang, Zhang, Wang,
He & Zhang 2000; Yang et al 2005) Thus, arti¢cial
reefs are commonly used for sea cucumber ranching
in China (Chen 2003), as they provide protection
against predators (Ambrose & Anderson 1990),
sup-ply shelter and food, in addition to sites for aestivation
(Chen 2004, 2007; Qin et al 2009)
An increasing demand for beche-de-mer, together
with evidence of a world-wide decline in natural
stocks (Conand 2004), suggests that the development
of e⁄cient enhancement methods, especially new
types of structures, are urgently required This study
describes the development and trials of a new culture
system, a multilayer, plate-type reef, which can be
easily varied in colour, shape/style or spacing, for
pond culture of sea cucumbers
Materials and methodsMultilayer, plate-type sea cucumber reefsystem
The new multilayer, plate-type sea cucumber reef(Fig 1) consists of a base plate (supporting the sys-tem), polyethylene (PE)-corrugated sheet (stacked onthe base plate), polyvinyl chloride (PVC) cannulae(acting as spacers for the PE-corrugated sheet) andtie wires (holding the system in place)
The base plate is a cast-reinforced concrete slab of
75 70 8 cm containing nine holes of10 cm in meter arranged in a regular pattern Three strut bars,made from galvanized tubes, 70 cm in length and2.5 cm external diameter, were cast vertically nearthe centre of the base plate The shade netting systemconsisted of 2.0 mm high-density polyethylene nettingsupported (sown) on a polypropylene U-shaped frame,
dia-70 cm in height,60 cm in width and 2.5 cm in externaldiameter, cast on one side of the base plate
The PE-corrugated sheet was 70 cm2and 2 cm inwave height Circular holes of 4.0 cm in diameterwere punched symmetrically in the troughs Threeother circular holes, 3.0 cm in diameter, werepunched near the centre of the corrugated sheet toaccommodate the strut bars and load the sheet e⁄-ciently The colour of the corrugated sheet wasdecided by experimental design
The cannulae or spacers were made from PVCtubing, 4 cm in external diameter and 0.2 cm in wall
Figure 1 Multilayer, plate-type sea cucumber reef: (a) graphic, (b) lateral view 1, Base plate; 2, polypropylene (PPR) ing; 3, high-density polyethylene (HDPE) netting; 4, shade netting; 5, strut bar; 6, polyethylene (PE)-corrugated sheet; 7,polyvinyl chloride (PVC) cannula; 8, tie wire
Trang 20pip-thickness The length of the cannulae depended on
the spacing required between the corrugated sheets
The corrugated sheets were loaded on the base
plate via the three holes onto the three strut bars
and spaced with the PVC cannulae The shape and
style of the corrugated sheets can also be varied by
using di¡erent-length cannulae Finally, a tie wire
was fastened around the strut bars to secure the
mul-tilayer, plate-type sea cucumber reef system
Deployment of the new system
Forty new structures were grouped into 10 categories
by sheet colours, spacing between corrugated sheets
and shape/styles of the sheets (Table 1) The number
and location of each structure are also listed in
Table 1 Depending on the arrangements of the
corru-gated sheets, each shape/style was classi¢ed as
‘obli-que’, ‘parallel’ and ‘wavy’ ‘Oblique’ means that all
sheets were parallel and at an angle of 301 to the base
plate.‘Parallel’ means that all corrugated sheets were
parallel to the base plate ‘Wavy’ means that every
second sheet was at an angle of 101 and the bottom
sheet was also at an angle of 101 to the base plate
All 40 structures were placed at depths41.8 m in
a 467 000 m2co¡erdam in Haiyang (Yantai Province,
China) between 3 July 2008 and 4 July 2008 (Table1).All assigned sites had near-identical physical habitatcharacteristics During deployment, the shade net-tings in all systems faced south, so that structureswith the ‘oblique’arrangement had more shading
Experimental designExperiments were performed more than 8 monthsafter deployment of the new structures to allow accli-mation in the co¡erdam The ¢rst experiment wascarried out between 18 March 2009 and 4 May
2009 to access which colour was the most attractive
to sea cucumbers among systems BO3, TO3, BLO3,GO3 and MO3 From the result of ¢rst experiment,another experiment was carried out between 20 Oc-tober 2009 and 29 November 2009 to ascertain theoptimum spacing among the BO2, BO3, BO4 andBO5 categories and shape/style among BO3, BP3and BW3 (Table 1) All experiments were conductedduring the high mobility phases of sea cucumbers
Experimental conditionsDuring deployment of the systems in the co¡erdam,seawater temperature ranged 24.5^26.0 1C, dissolved
Table 1 Number and location of each structure
Features of polyethylene-corrugated sheets
Site Number of structures System abbreviation
‘Spacing’ indicates the distance between neighbouring sheets.
w‘Mixed’ indicates six transparent sheets in the upper layers and ¢ve black sheets in bottom layers.
Aquaculture Research, 2011, 42, 1431^1439 New system for the culture of sea cucumber L Zhang et al.
Trang 21oxygen ranged 5.00^5.34 mg L 1, pH ranged 8.09^8.18
and salinity ranged 28.0^28.4% During the ¢rst and
the second experiments, respectively, seawater
tem-peratures ranged 6.6^18.0 and 3.8^18.8 1C, dissolved
oxygen ranged 4.39^5.89 and 4.29^5.69 mg L 1, pH
ranged 7.72^8.13 and 7.81^8.05, and salinity ranged
29.3^30.0 and 28.5^30.0% Over the experimental
per-iod, dissolved oxygen of surface seawater was 0.20^
0.60 mg L 1higher than that of the bottom
Sample collection
All sea cucumbers in each experimental structure
were separately collected by a SCUBA diver on 18
March, 9 April, 4 May and 20 October, 10 November
and 29 November 2009 Individuals were weighed
and the numbers on each layer were noted Although
numbers on every layer were included in the ¢eld
study, for analysis, the upper six sheets and lower ¢ve
sheets plus the base plate were combined to form an
‘upper’and ‘lower layer’category respectively
As a control, ¢ve samples were collected randomly
on every observation day at a distance from the
ex-perimental systems using a 70 70 cm quadrat
(the same area as the new structure) to assess density
of sea cucumbers All collected animals were
re-placed near the sampling zones
Statistical analysis
Di¡erences in numbers of sea cucumbers between
the ‘upper’ and ‘lower layer’ categories were
ana-lysed by a paired t-test (SPSSV 13.0) Numbers ofsea cucumbers were compared across each obser-vation day The three ¢xed factors were colour
of corrugated sheets (C, within subjects repeatedmeasure factor with ¢ve levels), spacing of corru-gated sheets (I, within subjects repeated measurefactor with four levels) and shape/style of corru-gated sheets (S, within subjects repeated measurefactor with three levels).ANOVAand pair-wise posthoc LSD multiple range tests were performed using
SPSSV 13.0 software Di¡erences were considered to
be signi¢cant at Po0.05
ResultsE¡ect of colourNumbers of A japonicus in systems with di¡erent col-our corrugated sheets (BO3, TO3, BLO3, GO3 andMO3) are shown in Fig 2 Statistical analysis showedthe same pattern on di¡erent observation dates Thenumber of animals in BO3,TO3, BLO3, GO3 and MO3was signi¢cantly higher than the control group(number of sea cucumber sampled by quadrat outsidethe systems; Po0.05).The highest number was found
in BO3 and was signi¢cantly higher than the otherstructures or the control group (Po0.05) The lowestnumber occurred in the TO3 and was signi¢cantlylower than the other systems (Po0.05) No signi¢-cant di¡erences were found between numbers of an-imals in TO3, BLO3 or MO3 (P40.05)
Figure 2 Numbers of Apostichopus japonicus versus colour of polyethylene-corrugated sheets Di¡erent letters indicatesigni¢cant di¡erences (Po0.05) and bars represent standard errors of means
Trang 22E¡ect of spacing
Numbers of A japonicus versus spacing of
PE-corru-gated sheets (BO2, BO3, BO4 and BO5) are presented
in Fig 3 Statistical analysis showed the same trend
on di¡erent observation dates Numbers of animals
in BO2, BO3, BO4 and BO5 were signi¢cantly higher
than in the control group The highest number was
found in system BO2 and was signi¢cantly higher
than in the other structures or control group
(Po0.05) No signi¢cant di¡erences were found
be-tween numbers of animals in BO3, BO4 or BO5
(P40.05)
E¡ect of shape/style
Numbers of A japonicus versus shape/style of
PE-cor-rugated sheets (BO3, BP3 and BW3) are illustrated in
Fig 4 Statistical analysis showed the same pattern on
di¡erent observation dates Numbers of animals in
BO3, BP3 and BW3 were signi¢cantly higher than
in the control group (Po0.05), showing the
relation-ship BO34BW34BP3 There were signi¢cant
di¡er-ences among sea cucumber numbers in BO3, BW3and BP3 (Po0.05)
‘Upper’ versus ‘lower layer’ densityTable 2 showed the numbers of A japonicus in theupper and lower layers with di¡erent coloured sheets(BO3,TO3, BLO3, GO3 and MO3) Numbers of sea cu-cumbers in the upper layers were signi¢cantly higherthan in lower layers for BO3, BLO3 and GO3 on all 3observation days (Po0.05)
However, in MO3, the number of sea cucumbers
in the upper layers was signi¢cantly lower than inthe lower layers on all 3 observation days (Po0.05).Furthermore, in TO3, there was no signi¢cant di¡er-ence in the number of sea cucumbers in upper layersversus the lower layers on either 18 March or 9 April(P40.05) In addition, the number of sea cucumbers
in the upper layers was signi¢cantly lower than inthe lower layers of TO3 on 4 May (Po0.05)
Table 3 showed the numbers of A japonicus in theupper and lower layers with di¡erent sheet spacing
Figure 3 Numbers of Apostichopus japonicus versus
spa-cing of polyethylene-corrugated sheets Di¡erent letters
indicate signi¢cant di¡erences (Po0.05) and bars
repre-sent standard errors of means
Figure 4 Numbers of Apostichopus japonicus versusshape/style of corrugated sheets Di¡erent letters indicatesigni¢cant di¡erences (Po0.05) and bars represent stan-dard errors of means
Table 2 Numbers of Apostichopus japonicus in di¡erent coloured upper and lower layers
Systems
Trang 23(BO2, BO3, BO4 and BO5) and with di¡erent sheet
shape/style (BO3, BP3 and BW3) In all systems and
on all observation days, statistical analysis showed
that the number of sea cucumbers in the upper layers
was signi¢cantly higher than in the lower layers
(Po0.05)
Discussion
Attraction of sea cucumbers to di¡erent
colours
Animals select suitable habitats as protection against
predators and/or competition, better food availability,
etc (Yan & Chen 1998) This includes matching body
colour with the background colour as a protective
de-vice Zhang, Wang, Rong, Cao and Chen (2009) have
demonstrated that black and grey settlement
sub-strates are more desirable to A japonicus than red,
white, green or yellow substrates in laboratory
ex-periments In this study, it showed the similar colour
desirability to A japonicus in the ¢eld, and A
japoni-cus tended to settled on the black material, similar in
body colour, of the ¢ve experimental coloured
materi-als (Table 1) Thus, the number of animmateri-als in BO3 was
signi¢cantly higher than in TO3, BLO3, GO3 or MO3
The number of animals in TO3 was lowest, as the
transparent corrugated sheets provided little shade
and hence were less attractive to sea cucumbers
Attraction of sea cucumbers to di¡erent
spacing
A characteristic of living on rocky reefs (Zhang et al
1995) is that A japonicus tend to inhabit crevices and
holes, giving shade and protection against predation
Apostichopus japonicus cannot tolerate high
tempera-tures It is a temperate water sea cucumber that isknown to aestivate when water temperatures risesabove a threshold level (Yang, Zhou, Zhang,Yuan, Li,Liu & Zhang 2006; An, Dong & Dong 2007; Yuan,Yang, Wang, Zhou, Zhang & Liu 2007; Dong, Dong &
Ji 2008; Ji, Dong & Dong 2008; Wang, Yang, Gabr &Gao 2008; Yuan,Yang,Wang, Zhou & Gabr 2009) andthe animals conceal themselves in crevices duringaestivation In this study, the number of sea cucum-bers in BO2, with the smallest spacing, was signi¢-cantly higher than the other BO3, BO4 and BO5systems
Attraction of sea cucumbers to di¡erentshape/style
Light has a signi¢cant e¡ect on the diurnal rhythm,migration, grouping behaviour, feeding, etc of aqua-tic animals (Zhou, Niu & Li 1999; Chen, Gao, Liu,Shao & Shi 2007) Migration of newly settled juve-niles from sheltered nursery areas to exposed adulthabitats has been demonstrated in the northern seacucumber Cucumaria frondosa (Hamel & Mercier1996) On the other hand, the larvae of Psolus chito-noides were found to settle initially near adults andthen relocate in nearby shaded habitats (Young &Chia 1982) Young and Chia (1982) and Hamel andMercier (1996) found that the main factors regulatingthe post-settlement migration of juvenile holothur-ians was the distribution of shaded substrates andvulnerability to predatory pressure Mercier, Batta-glene and Hamel (2000) found that early-stagepentactula larvae of Holothuria scabra showed anegative phototaxic response and migrated to theshaded side of the substrate Zhang, Chen and Sun(2006) found that when exposure to high light inten-sity, A japonicus migrated to the shade of arti¢cialreefs or onto the arti¢cial reefs
Table 3 Numbers of Apostichopus japonicus in upper and lower layers at di¡erent spacings and shape/style
Systems
Trang 24In this study, varying the angle of the corrugated
sheets gave di¡erent degree of shading Obviously,
more shade is provided by the oblique than either
the parallel or wavy design It might be the reason
why the numbers of A japonicus in the BO3 design
were signi¢cantly higher than in either BW3 or BP3
Attraction of sea cucumbers to ‘upper’ and
‘lower’ layers
Distribution of aquatic animals has a spatial
hetero-geneity (Persson & Svensson 2006) In its natural
en-vironment, A japonicus live in crevices in rocky reefs
close to sea grass (Zostera marina) and low light
inten-sity, or a silty bottom with luxuriant meadows of Z
marina (Zhang et al 1995) However, when bottom
water is low of dissolved oxygen, sea cucumbers
climb up the rocky reefs In this co¡erdam utilized in
this study, the level of dissolved oxygen of bottom
water was lower than that of surface water In
addi-tion, as the study was carried out during the period of
high activity for sea cucumbers, the animals could
climb to the ‘upper layers’ within the experimental
structures Clearly, there was more shade in the
‘low-er’ than in the ‘upp‘low-er’ layer of MO3; thus, sea
cucum-bers prefer the ‘lower’ layer to settle In TO3, with
transparent corrugated sheets, all animals were
ex-posed to the sun equally Consequently, there was no
signi¢cantly di¡erent attraction of sea cucumbers to
the ‘upper’and ‘lower’ layer in 18 March and 19 April
However, from 18 March, 9 April to 4 May, the
sun-light was becoming higher day after day Although
the system TO3 was assembled with transparency,
the illumination of the ‘lower’ layer was a little lower
than that of the‘upper’ layer This might be the reason
why the number of sea cucumbers in the‘upper’ layer
was lower than that in the ‘lower’ layer
The study indicated that the new system provided
sea cucumbers with a safer environment when
sea-water was low in dissolved oxygen in farming ponds,
especially during the summer or in periods of low
seawater quality
Conclusion
The study demonstrated that the numbers of sea
cu-cumbers, A japonicus, in all of the new structural
de-signs investigated was signi¢cantly higher than in
the natural environment The results indicated that
a design of black, corrugated sheets with a 2 cm
spa-cing and oblique orientation was the most attractive
for A japonicus, and most suitable for culture andstock enhancement in co¡erdams or ponds The newmultilayer, plate-type sea cucumber reef can increasethe farming/habitat space Besides, it is not only ea-sily installed and removed, but can also be suppliedwith di¡erent coloured sheets, at di¡erent spacingsand angles to match the pond environment and ani-mal behaviour The new system is less labour inten-sive and more cost e¡ective, and because it reducesthe frequency of feeding, it will improve water quality
in farming areas and protect benthic communitiesand ecosystems
AcknowledgmentsThe authors thank Shandong Oriental Ocean Sci-techHaiyang Branch for providing facilities for the presentstudy, and Edanz Writing for help in editing this manu-script This work was supported by the National MarinePublic Welfare Research Project (No 200805069), theNational High Technology Research and DevelopmentProgram of China (863 Program) (No 2006AA100304)and the National Key Project of Scienti¢c and Technical
of China (No 2006BAD09A02)
References
Abramsky Z., Al¢a H., Schachak M & Brand S (1990) tion by rodents and the distribution and abundance of the snail Trochoidea seetzenii in the Central Negev Desert of Is- rael Oikos 59, 225^234.
Preda-Ambrose R.F & Anderson T.W (1990) In£uence of an cial reef on the surrounding infaunal community Marine Biology 107, 41^52.
arti¢-An Z., Dong Y & Dong S (2007) Temperature e¡ects on growth-ration relationships of juvenile sea cucumber Apostichopus japonicus (Selenka) Aquaculture 272, 644^648.
Chang Y., Ding J & Song J (2004) Biology and Aquaculture of Sea Cucumber and Sea Urchin Ocean Press, Beijing, China Chen J (2003) Overview of sea cucumber farming and sea ranching practices in China SPC Beche-de-mer Informa- tion Bulletin 18, 18^23.
Chen J (2004) Present status and prospects of sea cucumber dustry in China FAO Fisheries Technical Sheet, 26 ChenY (2007) Sea cucumber Apostichopus japonicus culture techniques in pond Journal of Aquaculture 4, 19^20 (in Chinese).
in-Chen Y., Gao F., Liu G., Shao L & Shi G (2007) The e¡ects of temperature, salinity and light cycle on the growth and behavior of Apostichopus japonicus Journal of Fisheries of China 31, 687^691 (in Chinese, with English abstract) Aquaculture Research, 2011, 42, 1431^1439 New system for the culture of sea cucumber L Zhang et al.
Trang 25Conand C (2004) Present status of world sea cucumber
re-sources and utilization: an international overview Advances
in Sea cucumber Aquaculture and Management, FAO
Fisheries Technical Paper No 463, pp 13^23.
Dong Y., Dong S & Ji T (2008) E¡ect of di¡erent thermal
re-gimes on growth and physiological performance of the
sea cucumber Apostichopus japonicus Selenka
Aquacul-ture 275, 329^334.
Hamel J.F & Mercier A (1996) Early development,
settle-ment, growth, and spatial distribution of the sea
cucum-ber Cucumaria frondosa (Echinodermata:Holothuroidea).
Canadian Journal of Fisheries and Aquatic Sciences 53,
253^271.
Ji T., Dong Y & Dong S (2008) Growth and physiological
re-sponses in the sea cucumber, Apostichopus japonicus
Se-lenka: aestivation and temperature Aquaculture 283,
180^187.
Li J., Hao X., Gao W & Yin R (2004) Sea
cucumberAposticho-pus japonicus culture techniques with rocks in shrimp
pond Scienti¢c Fish Farming 4, 33 (in Chinese).
Li L & Huo J (2007) Ecological cultivation development of
sea cucumber Apostichopus japonicus China Fisheries 8,
32^33, 35 (in Chinese).
LiaoY (1980) The aspidochirote holothurians of China with
erection of a new genus In: Echinoderms: Present and Past
(Proceeding of European Colloquium on Echinoderm) (ed by
M Jangoux), pp.115–120 A A Balkema Publishers,
Rotterdam, the Netherlands.
LiaoY (1997) Fauna Sinica, Phylum Echinodermata, Class
Ho-lothuroidea Science Press, Beijing, China, 334pp.
Lin P (2007) Sea cucumber culture techniques with knitted
fabrics in pond China Fisheries 10, 49 (in Chinese).
Mercier A., Battaglene S.C & Hamel J.F (2000) Settlement
preferences and early migration of the tropical sea
cu-cumber Holothuria scabra Journal of Experimental Marine
Biology and Ecology 249, 89^110.
Persson A & Svensson J.M (2006) Vertical distribution of
benthic community responses to ¢sh predators, and
ef-fects on algae and suspended material Aquatic Ecology
40, 85^95.
Qin C., Dong S., Niu Y., Tian X., Wang F., Gao Q & Dong Y.
(2009) E¡ects of shelter type on the growth and survival
rate of sea cucumber, Apostichopus japonicus Selenka, in
earthen pond Periodical of Ocean University of China 39,
392^396 (in Chinese, with English abstract).
Riechert S.E (1976) Web-site selection in the desert spider
Agelenopsis aperta Oikos 27, 311^315.
Sun D & Chen A (2006) Development of sea cucumber
bot-tom-culture techniques with rocks in shallow sea
Shan-dong Fisheries 7, 3^4 (in Chinese).
Sun Z (2004) The third of aquaculture
techniques-discus-sion of several questions in sea cucumber culture in pond.
China Fisheries 5, 56^58 (in Chinese).
Wang F.,Yang H., Gabr H.R & Gao F (2008) Immune
condi-tion of Apostichopus japonicus during aestivacondi-tion
Aqua-culture 285, 238^243.
WangY., Zhang S., Zhang X., Qu J & Song Z (2004) tion of movable and portable shelters in sea cucumber cultivation in pond Shandong Fisheries 21, 9^10 (in Chi- nese).
Applica-Ward P & Porter A.H (1993) The relative roles of habitat structure and male^male competition in the mating sys- tem of Gammarus pulex (Crustacea:Amphipoda): a simu- lation study Animal Behaviour 45, 119^133.
Yan Z & ChenY (1998) Habitat selection in animals Chinese Journal of Ecology 17, 43^49 (in Chinese, with English abstract).
Yang H & Shan X (2007) Developing sea cucumber reefs artfully Shandong Fisheries 24, 29 (in Chinese) Yang H.,Wang J., ZhouY., Zhang T.,Wang P., HeY & Zhang F (2000) Comparison of e⁄ciencies of di¡erent culture sys- tems in the shallow sea alongYantai Journal of Fisheries of China 24, 140^145 (in Chinese, with English abstract) Yang H., Yuan X., Zhou Y., Mao Y., Zhang T & Liu Y (2005) E¡ects of body size and water temperature on food con- sumption and growth in the sea cucumber Apostichopus japonicus (Selenka) with special reference to aestivation Aquaculture Research 36, 1085^1092.
Yang H., ZhouY.,Wang J., Zhang T.,Wang P., HeY & Zhang F (2000) A modelling estimation of carrying capacities for Chlamys farreri, Laminaria japonica and Apostichopus japo- nicus in Sishiliwan Bay,Yantai, China Journal of Fisheries
of China 7, 27^31 (in Chinese, with English abstract) Yang H., Zhou Y., Zhang T., Yuan X., Li X., Liu Y & Zhang F (2006) Metabolic characteristics of sea cucumber Aposti- chopus japonicus (Selenka) during aestivation Journal of Experimental Marine Biology and Ecology 330, 505^510 Young C.M & Chia F.S (1982) Factors controlling spatial dis- tribution of the sea cucumber Psolus chitonoides: settling and post-settling behavior Marine Biology 69, 195^205 Yuan X., Yang H., Zhou Y., Mao Y., Zhang T & Liu Y (2006) The in£uence of diets containing dried bivalve feces and/
or powdered algae on growth and energy distribution in sea cucumber Apostichopus japonicus (Selenka) (Echino- dermata: Holothuroidea) Aquaculture 256, 457^467 Yuan X.,Yang H.,Wang L., Zhou Y., Zhang T & Liu Y (2007) E¡ects of aestivation on the energy budget of sea cucum- ber Apostichopus japonicus (Selenka) (Echinodermata:Ho- lothuroidea) Acta Ecologica Sinica 27, 3155^3161 Yuan X.,Yang H.,Wang L., ZhouY & Gabr H.R (2009) Bioe- nergetic responses of sub-adult sea cucumber Aposticho- pus japonicus (Selenka) (Echinodermata:Holothuroidea) to temperature with special discussion regarding its south- ernmost distribution limit in China Journal of Thermal Biology 34, 315^319.
Zhang B., Sun D & WuY (1995) Preliminary analysis on the feeding habit of Apostichopus japonicus in the rocky coast waters o¡ Lingshan Island Marine Sciences 3, 11^13 (in Chinese, with English abstract).
Zhang H., Wang Y., Rong X., Cao S & Chen X (2009) vioral responses of sea cucumber (Apostichopus japonicus)
Beha-to di¡erent light intensity and settlement substratum
Trang 26color Chinese Journal of Ecology 28, 477^482 (in Chinese,
with English abstract).
Zhang S., Chen Y & Sun M (2006) Behavior characteristics
of Apostichopus japonicus and attractive e¡ects of arti¢cial
reef models under di¡erent light intensities Journal of
Fishery Sciences of China 13, 20^26 (in Chinese, with
Eng-lish abstract).
Zhao Z (1995) Status and problems involved in management
of arti¢cial reefs for ¢shes and sea cucumbers o¡ china coast Marine Science Bulletin 14,79^84 (in Chinese, with English abstract).
Zhou X., Niu C & Li Q (1999) The e¡ects of light on aquatic animals Chinese Journal of Zoology 34, 45^48 (in Chinese).
Aquaculture Research, 2011, 42, 1431^1439 New system for the culture of sea cucumber L Zhang et al.
Trang 27Genetic divergence among broodstocks of Arctic charr
same founding populations
Craig T Blackie, Michael B Morrisseyw, Roy G Danzmann & Moira M Ferguson
Department of Integrative Biology, University of Guelph, Guelph, Ontario, Canada
Correspondence: M M Ferguson, Department of Integrative Biology, University of Guelph, Guelph, Ontario, Canada N1G 2W1 E-mail: mmfergus@uoguelph.ca
Present address: Craig T Blackie, Biology Department, Dalhousie University, Halifax, Nova Scotia, Canada B3H 4J1.
w
Present address: Michael B Morrissey, Institute of Evolutionary Biology, University of Edinburgh, Scotland.
Abstract
We examined the amount and distribution of
mole-cular variation at microsatellite loci in 21
brood-stocks of Arctic charr Salvelinus alpinus derived from
the Fraser River, Labrador and the Nauyuk Lake,
Nunavut, Canada Our goal was to assess the amount
of genetic diversity and di¡erentiation as broodstocks
are subdivided and propagated in di¡erent
hatch-eries and grow-out facilities We observed signi¢cant
heterogeneity across pairs of ancestral and
descen-dant broodstocks in the mean numbers of alleles at
microsatellite loci.We detected a signi¢cant decrease
in the observed heterozygosity between ancestors
and descendants but the amount of decrease did not
depend on either the degree of removal from the wild
(number of sequential transfers) or the strain (Fraser
vs Nauyuk) Based on allele frequency distributions,
there was little genetic evidence of bottlenecks
during the creation of subsequent broodstock
popu-lations after the initial founding events All
brood-stock samples were signi¢cantly di¡erentiated from
each other but those within the same strain were
more similar to each other than to broodstocks
from di¡erent strains Broodstocks from the Nauyuk
Lake broodstocks showed greater di¡erentiation from
each other than did Fraser River broodstocks, which
could be attributed to di¡erences in the number
of founders
Keywords: Arctic charr, Salvelinus alpinus,
popu-lation di¡erentiation, genetic variation, popupopu-lation
bottlenecks, cultured ¢sh
IntroductionManagement of genetic resources has become a majorconcern in aquaculture populations (Bentsen & Ole-sen 2002; Fjalestad, Moen & Gomez-Raya 2003) It iscrucial to maintain high levels of genetic variation inbroodstocks because the rate of response to arti¢cialselection increases with the amount of genetic varia-tion present (Gjedrem 2000) Small e¡ective popula-tion sizes during the founding (Luikart & Cornuet1998; Waples 2002) and propagation (Barker 2001)can lead to a loss of genetic variation through theaction of random genetic drift (Falconer & MacKay1996; Woodworth, Montgomery, Briscoe & Frankham2002) As well, the risk of close inbreeding may be ele-vated in small e¡ective populations (Bentsen & Olesen2002; Lind, Evans, Knauer,Taylor & Jerry 2009).The genetic variation within and among di¡erentcultured strains relative to wild strains has been stu-died in many species (Evans, Bartlett, Sweijd, Cook &Elliott 2004; Skaala,Taggart & Gunnes 2005; Brown,Wang, Li, Givens & Wallat 2007; Porta, Porta, Cana-vate, Martinez-Rodriguez & Alvarez 2007; Chen, Li &Yang 2008) Several studies have reported substantiallosses of genetic variation in broodstocks in only afew generations after founding from wild populations(e.g., Porta et al 2007; Lind et al 2009) In addition tofounder e¡ects, genetic variation can be lost as cul-tured populations are subdivided and propagated indi¡erent facilities (Norris, Bradley & Cunningham1999; Saisa, Koljonen & Tahtinen 2003; Romana-Eguia, Ikeda, Basiao & Taniguchi 2004) Given thatpopulations arising from the same founder event are
Trang 28often considered to be as the same strain from a
per-formance perspective, it is important to genetically
characterize such populations as they represent the
fundamental resource for breeding programmes
The farming of Arctic charr Salvelinus alpinus in
Canada is a relatively recent industry and strains
are largely unimproved genetically (Lundrigan, Reist
& Ferguson 2005) Two natural populations from
geographically distant locations and phylogenetic
lineages (Wilson, Hebert, Reist & Dempson 1996;
Brunner, Douglas, Osinov, Wilson & Bernatchez
2001) have been the primary contributors to the
Arctic charr being produced in the Canadian
indus-try: the Nauyuk Lake system in Nunavut, and the
Fraser River in Labrador These strains were
estab-lished from limited numbers of founders in the 1970s
and 1980s (see Lundrigan et al 2005 for details, Table
1) Brie£y, Fraser and Nauyuk strain embryos were
shipped to the Rockwood Aquaculture Research
Cen-tre (RARC) from the ¢eld and distributed to other
cul-ture facilities throughout Canada Unfortunately, the
timing and the numbers of ¢sh involved in these
trans-fers are mostly unknown A subsample of Fraser
em-bryos was also shipped to the Huntsman Marine
Sciences Centre, New Brunswick where they were
combined with progeny derived from additional ¢eld
collections Lundrigan et al (2005) showed that recent
descendants (two to four generations) of the original
founders of the Fraser and Nauyuk strains at the
RARC had less allelic diversity but not observed
het-erozygosity than their wild source populations
Transfers of ¢sh from the RARC to other facilities
throughout Canada might have induced additional
demographic bottlenecks and further reduced
gene-tic variation Levels of genegene-tic variation in S alpinus
populations more distantly derived from the wild are
mostly unknown (Rogers & Davidson 2001) However,
Ditlecadet, Dufresne, Le Francois and Blier (2006)
showed that two domesticated strains from Quebec
(one derived from the Fraser broodstocks at RARC)
showed lower genetic variability than the RARC
brood-stock samples analysed by Lundrigan et al (2005)
We conducted a genetic analysis of 17 broodstock
samples of Fraser River and Nauyuk Lake Arctic
charr produced at 11 aquaculture facilities across
eastern Canada Our objective was to compare
the amount and distribution of genetic variation
within and among contemporary broodstocks
rela-tive with those that were only two to four generations
removed from the original founding event at the
RARC (Lundrigan et al 2005), using a common set of
microsatellite loci We predicted that descendant
broodstocks would show reduced levels of geneticvariation compared with their ancestors We alsosearched for evidence of bottlenecks in the found-ing of contemporary hatchery broodstocks using avariety of approaches
Materials and methods
We sampled 17 contemporary broodstocks of Arcticcharr in 2004 and 2005 from culture facilities in east-ern Canada (Table 1) We de¢ned a broodstock as apopulation of interbreeding ¢sh that was not crossedwith other such groups in the same or di¡erent culturefacilities All of the broodstocks we sampled belong
to either the Fraser or Nauyuk strains dependingupon their ancestry and natural population of origin(Fig 1) Several culture facilities rear and propagateboth Fraser and Nauyuk broodstocks and sometimesmultiple broodstocks of the same strain (e.g., residentand anadromous forms of the Nauyuk strain) Indivi-dual broodstocks are often reared in multiple tankswithin a facility and we attempted to obtain represen-tative samples of entire broodstocks
We non-lethally sampled forty-eight individualsfrom each broodstock The ¢sh were anaesthetizedwith MS-222 and a portion of the ventral lobe of thecaudal ¢n was removed and preserved in 95% etha-nol for subsequent genetic analysis Sex informationwas unobtainable because we avoided sampling dur-ing the spawning season
Four broodstocks from the RARC (from Lundrigan
et al 2005) were included (Table 1) so that the initiallevel of genetic variation and population di¡erentia-tion in the broodstocks shortly after founding could
be evaluated and compared with subsequent tured generations
cul-We isolated total genomic DNA from tissue samplespreserved in ethanol using a modi¢ed phenol/chloro-form/isoamyl alcohol extraction method (Bardacki &Skibinski 1994) We genotyped all samples at the mi-crosatellite loci Sco-19 (Taylor, Redenbach, Costello,Pollard & Pacas 2001), Omy-301 (Jackson, Ferguson,Danzmann, Fishback, Ihssen, O’Connell & Crease1998), MST-85 (Presa & Guyomard 1996), Sfo-8 (An-gers, Bernatchez, Angers & Desgroseillers 1995), Sfo-
23 (Angers et al 1995) and Ogo-8 (Olsen, Bentzen &Seeb 1998) With the exception of Sfo-8, which hasnot been mapped, these loci occur on di¡erent link-age groups and none are linked to the sex-determin-ing locus (Danzmann, Cairney, Davidson, Ferguson,Gharbi, Guyomard, Leder, Okamoto, Ozaki, Rexroad,Aquaculture Research, 2011, 42, 1440^1452 Genetic divergence among broodstocks of Arctic charr C T Blackie et al.
Trang 29Sakamoto, Taggart & Woram 2005) We performed
polymerase chain reactions (PCRs) for each locus
in-dividually We used the PCR conditions described by
Lundrigan et al (2005) for all loci except Omy-301
for which we followed Wilson, Gislason, Skulason,
Snorrason, Adams, Alexander, Danzmann and
Ferguson (2004) In each PCR reaction, one primer
contained a £uorescent dye that allowed
visualiza-tion on a scanner after separavisualiza-tion in 7%
polyacryla-mide gels We sized bands relative to a TAMRA-350
lane standard
We calculated observed allele frequencies, number
of alleles (A), observed heterozygosity (HO), expected
heterozygosity (He), deviations from
Hardy^Wein-berg equilibrium (HWE) and conducted tests for
link-age disequilibrium for each locus in each broodstock
sample using GENEPOP version 3.3 (Raymond &
Rousset 2001).We expressed deviations from HWE as
FISand quanti¢ed their statistical signi¢cance with
exact probability tests.We tested for linkage
disequili-brium between all pairs of loci within each stock population using exact probability tests Weimplemented all tests usingGENEPOPv3.3 (Raymond
brood-& Rousset 2001) with default parameters.We appliedsequential Bonferroni’s corrections to all P values
in tests for linkage disequilibrium to reduce Type Istatistical error (Rice 1989)
In order to test for changes in genetic variation sociated with the subdivision, founding and propaga-tion of broodstocks in di¡erent facilities (transfers),
as-we calculated the change in estimated A, Heand Ho
at every locus for each of13 pairs of dant broodstock populations (Fig.1) In cases where adescendant broodstock could have arisen from either
ancestor^descen-or both of two ancestral broodstocks (e.g., Fr1P01/Fr1P02 giving rise to Fr2P08), we averaged per-locussummary statistics for the two possible ancestralpopulations We then ¢tted linear mixed models tothe changes in genetic parameters with strain andnumber of transfers relative to the original wild stock
Table 1 Details of the 21 Arctic charr broodstocks examined in this study including the strain of origin, location of the stock and broodstock designation (N 5 48 per broodstock)
Fraser Rockwood Aquaculture Research Centre (RARC) Gunton, Manitoba Fr1P01
Fraser Rockwood Aquaculture Research Centre Gunton, Manitoba Fr1P02w
Fraser Carpenter’s Aquaculture (CA) St John, New Brunswick Fr2P06
Fraser Cardigan Fish Hatchery (CFH) Cardigan, Prince Edward Island Fr2P07
Fraser Coastal Zones Research Institute (CZRI) Shippagan, New Brunswick Fr2P08
Fraser Char-Tec (CT) Sussex, New Brunswick Fr2P09
Fraser Pisciculture des Alleghanys (PA) Ste Emelie-de-l’Energie, Quebec Fr2P10
Fraser Nova Scotia Agricultural College (NSAC) Truro, Nova Scotia Fr3P14
Fraser Darren Cameron (DC) Truro, Nova Scotia Fr3P15
Fraser Cooke Aquaculture (COA) St John, New Brunswick Fr3P16
Fraser Moose Mountain Fisheries (MMF) Falconbridge, Ontario Fr3P17
Fraser Coldwater Fisheries (CF) Coldwater, Ontario Fr3P18
Fraser Coldwater Fisheries Coldwater, Ontario Fr3P20
Nauyuk (Anadromous) Rockwood Aquaculture Research Centre Gunton, Manitoba Na1P03z
Nauyuk (Resident) Rockwood Aquaculture Research Centre Gunton, Manitoba Na1P04‰
Nauyuk Agazziz Hatchery (AH) (formerly Rockwood
Aquaculture Research Centre)
Gunton, Manitoba Na1P05 Nauyuk (Anadromous) Cardigan Fish Hatchery Cardigan, Prince Edward Island Na2P11
Nauyuk (Resident) Cardigan Fish Hatchery Cardigan, Prince Edward Island Na2P12
Nauyuk Pisciculture des Alleghanys Ste Emelie-de-l’Energie, Quebec Na2P13
Nauyuk Moose Mountain Fisheries Falconbridge, Ontario Na3P19
Nauyuk Coldwater Fisheries Coldwater, Ontario Na3P21
See Fig 1 for relationships among broodstocks.
Designated as LRoX in Lundrigan et al (2005): third-generation descendants from at least 40 founders collected from the Fraser River, Labrador in 1981.
wDesignated as LRoW in Lundrigan et al (2005): second-generation descendants from four females and four males collected from the Fraser River, Labrador in 1984.
zDesignated as NRoLx in Lundrigan et al (2005): third-generation descendants from three females and one male (all anadromous) collected from Willow Lake, Nunavut (continuous with Nauyuk Lake) in 1978.
‰Designated as NRoX in Lundrigan et al (2005): fourth-generation descendants from two females and three males (all mous) collected from Willow Lake, Nunavut (continuous with Nauyuk Lake) in 1978.
Trang 30non-anadro-as ¢xed e¡ects and with locus and individual transfer
as random e¡ects We ¢tted reduced models from
which strain, number of transfers removed from the
wild and individual transfer e¡ects had been
elimi-nated We did not ¢t reduced models from which the
random e¡ect for locus had been eliminated, because
of the potential for non-independence among
trans-fers to be biased, i.e it is reasonable to expect that
reduction in number of alleles should be more severe
at loci with more alleles in ancestor populations We
¢t and obtained 1000 samples of the posterior
distri-butions of all models using the Gibbs sampling
algorithm implemented in the R PACKAGE MCMCglmm
(Had¢eld 2010) The most parsimonious models were
chosen based on the lowest deviance information
cri-terion value, which is appropriate for testing both
¢xed and random e¡ects (Bolker, Brooks, Clark,
Geange, Poulsen, Stevens & White 2009) We
esti-mated the average e¡ects of transfers on A, Hoand
He, and their statistical support by examining the
posterior distribution of the intercepts of the mixed
models Speci¢cally, we calculated the mode of the
samples of the posterior distributions to obtain an
estimate of the mean e¡ect of transfers on each
mea-sure of genetic diversity, and we determined whether
or not the 95% highest region of posterior densityoverlapped zero in order to obtain a measure of thestatistical support for the e¡ect of transfers
We tested for the genetic signatures of bottlenecksresulting from the founding of contemporary brood-stocks using a variety of approaches Populations thathave undergone a recent bottleneck tend to show areduction in allelic diversity but not necessarily a de-cline in observed heterozygosity (Ramstad, Woody,Sage & Allendorf 2004) As well, bottlenecked popu-lations can exhibit linkage disequilibrium (Waples2002), elevated heterozygosity levels from thoseexpected at mutation-drift equilibrium (Maruyama
& Fuerst 1985; Luikart & Cornuet 1998) and a modeshift in allele frequency distributions associated with
a reduction in low-frequency alleles (Luikart, dorf, Cornuet & Sherwin 1998)
Allen-In order to test for the genetic signatures of recentpopulation bottlenecks, we tested for (a) an increasedheterozygosity relative to that expected at mutation-drift equilibrium and (b) mode shifts in allele frequen-cies We compared heterozygosities of contemporarybroodstock populations with numerically derived dis-
Figure 1 Ancestor^descendant relationships among brookstocks of (a) Fraser River strain and (b) Nauyuk Lake strainArctic charr (Salvelinus alpinus) from eastern Canada The numbers in parentheses are the average numbers of alleles acrosssix microsatellite loci and observed heterozygosity detected in a particular broodstock sample Refer to Table 1 for details
on identi¢cation codes and locations Populations that do not have an identi¢cation number included were not sampled
Aquaculture Research, 2011, 42, 1440^1452 Genetic divergence among broodstocks of Arctic charr C T Blackie et al.
Trang 31tributions of heterozygosity at mutation-drift
equili-brium using Wilcoxon’s tests as implemented in the
programBOTTLENECKversion 1.2.02 (Cornuet & Luikart
1996) with subsequent sequential Bonferroni’s tests
We used the mode shift method in BOTTLENECK to
graphically investigate decreases in low-frequency
al-leles in each population This method plots the allele
frequency distribution and contrasts this with the null
hypothesis of an L-shaped distribution (described in
Luikart et al 1998) To provide a quantitative test
of changes in allele frequency distributions that might
be indicative of bottlenecks during the process
of founding new broodstocks, we appliedw2tests to
the proportions of alleles with frequencies below 0.1
(a threshold suggested by Allendorf 1986) in each pair
of ancestor^descendant broodstock populations
We tested for genotypic di¡erentiation between all
pairs of broodstock samples with Fisher’s exact tests
as implemented inGENEPOPwith default parameters
Given that broodstock samples exhibited several
devia-tions from HWE, this was a more suitable approach
than using methods involving allele frequencies
(Gou-det, Raymond, De Meeus & Rousset 1996) We
calcu-lated pairwise estimates of FST using Weir and
Cockerham’s (1984) estimator y using FSTAT v2.9.3.2
(Goudet 2001) with 10,000 permutations of
multi-lo-cus genotypes and subsequent Bonferroni’s
correc-tions for multiple tests To make inferences of the
extent of genetic di¡erentiation among broodstocks
within strains that would not be biased by di¡erences
in overall levels of diversity between strains (i.e., to
avoid the confounding e¡ect of non-independence of
FSTand within-population heterozygosity; Jost 2008),
we calculated Nei’s genetic distance for all pairs of
po-pulations using theADE4 PACKAGE(Dray & Dufour 2007)
We partitioned genetic variance among strains and
among broodstocks within strains (Fraser or Nauyuk)
using hierarchical analysis of molecular variance
(AMOVA) (Exco⁄er, Smouse & Quattro 1992) with
ARLE-QUINversion 3.11 (Exco⁄er, Laval & Schneider 2005)
As such, the components of variance, or ¢xation
indices (FCT, FSCand FSTrespectively) were calculated
and their signi¢cance was tested using1000
permuta-tions (Exco⁄er et al 1992) The FCTreferred to the
proportion of the variance that is due to di¡erences
between the Fraser River and Nauyuk Lake strains
The FSC was measured as the proportion of the
variance that could be ascribed among broodstocks
within strains regardless of ancestral/descendant
relationships Lastly, FSTwas calculated as the
propor-tion of the variance among broodstocks relative to the
total variance for all broodstocks
We used programs from the suite PHYLIP v3.5c(Felsenstein 1993) to produce a neighbour-joiningphenogram from a matrix of Nei’s genetic distance(DA) estimates (Nei, Tajima & Tateno 1983) We usedSEQBOOT in conjunction with GENDIST to generate
1000 bootstrap replicates of the matrix of DA, andgenerated a majority rule consensus phenogramusing the modules NEIGHBOR and CONDENSE
ResultsThe level of polymorphism in Canadian broodstockpopulations of Arctic charr ranges between 11 and
26 alleles per locus across all samples, with meannumbers of observed alleles within broodstock popu-lations between 4.17 and 9.83 Fraser River brood-stock samples have an average within-stock alleliccount of 6.59 while Nauyuk Lake broodstock sampleshave a mean of 5.65 alleles Mean observed hetero-zygosity (Ho) across loci in the samples varies be-tween 0.44 and 0.73 Fraser River broodstocksamples have a higher average within-population Hothan Nauyuk Lake charr (0.60 and 0.53 respectively).Deviations from HWE are common in our samples(Table 2) Following Bonferroni’s correction, 55 out of
a possible 126 cases deviate signi¢cantly from HWE.The number of deviations per locus varies from 4(MST-85) to 17 (Sfo-8) for the 21 broodstock samples.Departures from HWE are not restricted to speci¢cbroodstock samples Of the signi¢cant per-locus HWdeviations, the vast majority are due to heterozygotede¢cits We reran the major analyses excluding Sfo-8and although there were minor changes in absolutevalues, the identical qualitative results were obtained.Neither the strain of origin nor the degree ofremoval from the wild is a signi¢cant predictor ofmean changes in genetic diversity associated withthe subdivision and propagation of broodstock popu-lations in di¡erent facilities (transfers) There is sig-ni¢cant heterogeneity among transfers with regard
to changes in number of observed alleles betweenancestral and descendant broodstocks (Table 3) Theregions of 95% highest posterior density of the inter-cepts of the most justi¢ed models overlap zero for allthree metrics of genetic variation The prediction thatgenetic variation is reduced with the establishmentand propagation of descendant broodstocks is sup-ported with respect to Ho, with only 3.3% of samples
of the posterior distribution of the intercept for themodel of Hohaving values above zero However, theamount of decrease does not depend on either the de-gree of removal from the wild or the strain
Trang 34We detect little evidence of bottlenecks following
the initial founding of the strains from wild
progeni-tors There are only four cases of signi¢cant linkage
disequilibrium out of nearly 300 possible cases, which
is far fewer than would be expected by chance alone
Following Bonferroni’s correction, no statistically
sig-ni¢cant heterozygote excess relative to that expected
based on the observed levels allelic diversity exists
Qualitatively, L-shaped distributions of allele
frequen-cies exist in all populations, and alleles segregating at
frequencies below 0.1 are more numerous in all
popu-lations than alleles of any other such frequency
inter-vals (results not shown) Only one broodstock
population, Fr2P07 has a signi¢cantly smaller portion
of alleles with frequencies below the proportion in a
potential ancestor, and this is one of the broodstocks
with an uncertain immediate ancestor
All broodstock samples are signi¢cantly
di¡er-entiated from each other (all Fisher’s exact tests
Po0.001) The FSTestimates range from 0.043 to
0.405 and all signi¢cantly di¡er from zero (all
Po0.001) As expected, broodstock populations
with-in the same strawith-in (Fraser River or Nauyuk Lake) are
more similar to each other than to populations from
di¡erent strains based on pairwise FST(not shown)
va-lues and Nei’s distance metric (Fig 2) within and
be-tween strains Global FSTcomparisons indicate that
the magnitude of di¡erentiation among broodstocks
within each strain is statistically signi¢cant (Fraser
FST5 0.122, Nauyuk FST5 0.234, both Po0.001)
The three hierarchical AMOVAs indicate that a
highly signi¢cant proportion of genetic variation
(Po0.0001) is attributable to each level of
partition-ing (Table 4) For allAMOVAs, the greatest amount ofvariation occurs within broodstock samples, eventhough the strains were highly di¡erentiated fromone another The individualAMOVAs for each strain in-dicate that the Nauyuk Lake broodstocks had twicethe proportion of variation partitioned among sam-ples within the strain than did the Fraser Riverbroodstock samples Our ¢nding of greater subdivi-sion within the Nauyuk Lake strain compared withthe Fraser River strain is supported by the pattern inNei’s distance where mean pairwise DAis 0.65 amongNauyuk Lake and 0.39 among Fraser River strains.The topology of the neighbor-joining phenogrambased on Nei’s DAdistances is congruent with the
AMOVAanalyses (Fig 3) The existence of two strains
is a robust feature of the phenogram (93% bootstrapsupport) Poor bootstrap support exists for speci¢c re-lationships among broodstocks from the same strainand not surprisingly the clustering patterns depicted
in the phenogram do not necessarily correspond toknown ancestral/descendant relationships (Fig 1).Branches between Nauyuk Lake samples are bettersupported than those within the Fraser River clade,although this support is still weak
DiscussionThe microsatellite data for the cultured Arctic charr
in this study reveal that the subdivision, foundingand propagation of broodstocks in di¡erent culturefacilities are associated with lower observed hetero-zygosity in descendant broodstocks compared withtheir ancestors This result is generally consistentamong transfers, and is not a¡ected by the degree ofremoval from the wild (number of sequential trans-fers in the history of the broodstock) and the strain
of origin (Fraser vs Nauyuk) Although there is ni¢cant heterogeneity in the change in allelic diver-sity between pairs of ancestral and descendantbroodstocks, there was no consistent directionaltrend This heterogeneity without a consistent de-cline may indicate that practices concerning themaintenance of broodstocks have had in£uences ongenetic diversity that are at least as important aspractices used in their establishment Deviations fromHWE occur in approximately half of the loci withinsamples, with no consistent trend either among loci
sig-or populations, and heterozygote de¢cits were sible for480% of these deviations Reductions in ge-netic variation and the high proportion of deviationsfrom HWE are common attributes of ¢sh in closed
respon-Figure 2 Graphical representation of population
subdi-vision among Arctic charr (Salvelinus alpinus) broodstock
populations as measured by Nei’s distance metric
Aquaculture Research, 2011, 42, 1440^1452 Genetic divergence among broodstocks of Arctic charr C T Blackie et al.
Trang 35Fr2P10Fr2P06
27424516
3249
Fr3P16
Fr2P07
2453Fr3P17
44
72
Fr3P18
Fr2P09Fr3P15
Na1P04Na2P11
Na2P12Na2P13Na3P19
Na3P21
Figure 3 Unrooted neighbour-joining phenogram based on Nei’s distances (DA) for 21 Arctic charr (Salvelinus alpinus)broodstock populations of both Fraser River and Nauyuk Lake origin See Table 1 for population codes Numbers alongbranches indicate bootstrap support based on 1000 replicates
Table 4 Hierarchical analysis of molecular variance ( AMOVA ) based on F ST measures for 21 broodstock populations of culture Arctic charr (Salvelinus alpinus) at six microsatellite loci
Sum of squares
Variance components
Fixation indices
Percentage
of variation
Analysis 1
Between strains 1 436.76 0.42 Va F CT 5 0.16 16.38 Among populations within strains 19 675.04 0.35 Vb F SC 5 0.16 13.65 Within populations 1991 3593.10 1.80 Vc F ST 5 0.30 69.97 Analysis 2
Among populations within the Fraser River strain 12 330.27 0.27 Va 12.25 Within Fraser River populations 1231 2359.61 1.92 Vb FST5 0.12 87.75 Analysis 3
Among populations within the Nauyuk Lake strain 7 344.77 0.50 Va 23.41 Within Nauyuk Lake population 760 1178.49 1.62 Vb F ST 5 0.23 76.59 Broodstock populations are categorized into either Fraser River or Nauyuk Lake strains Statistically signi¢cant variation was explained
at all levels of partitioning (P o0.0001).
Trang 36hatchery systems where drift, arti¢cial selection and
non-random mating occur in the absence of gene £ow
(Taniguchi, Perez-Enriquez & Estu 2003)
We found little evidence that further population
bottlenecks have occurred in these hatchery
brood-stocks following the collections from the wild The
lack of signi¢cant heterozygote excess and reduction
in allelic diversity as populations become further
re-moved from the wild, and the low number of cases of
linkage disequilibrium indicate that bottlenecks
dur-ing the foundation of each stock have not been major
determinants of levels of genetic variation This
sug-gests that adequate numbers of parental ¢sh have
been used to create subsequent hatchery populations
after founding from the wild
Numerous deviations from HWE exist in the
brood-stock samples and heterozygote de¢cits were
responsi-ble for the majority of the deviations This is not
surprising given that these broodstocks are
propa-gated with ¢nite numbers of breeders and that the
choice of particular males and females for breeding is
not random Several factors can cause populations to
deviate from HWE (Hartl & Clark 1989) Small e¡ective
population size (through drift), inbreeding, population
subdivision, null alleles or sampling error (Hartl 1988)
could be responsible for the observed heterozygote
de¢cits Null alleles, particularly at Sfo-8 could have
caused deviations in Hardy^Weinberg proportions,
although other studies with Arctic charr did not
iden-tify a null allele problem with this locus (Angers &
Bernatchez 1998; Brunner, Douglas & Bernatchez
1998; Lundrigan et al 2005) Inferring the presence of
null alleles through the proportion of non-amplifying
individuals (mean across loci: 2.2% Sfo-8, 4.7% Sfo-23,
2.1% Sco-19, 3.0% Ogo-8, 3.2% Omy-301 and 2.9%
MST-85 respectively) (Wilson et al 2004) suggests that
they would be in a relatively low frequency and thus
contribute in a minor way to the deviations observed
No population was out of HWE across all loci,
suggest-ing that these populations are not inbred Instead, the
trend across loci suggests that other forms of
non-ran-dom mating are responsible (Frankham, Ballou &
Briscoe 2002) For example, producers tend to select
broodstock with similar phenotypic characters (such
as fast growth or £esh quality)
All broodstock samples are signi¢cantly
di¡eren-tiated from each other These results, based on exact
tests and pairwise FSTestimates, included
compari-sons within and between strains as well as
brood-stocks that appear to be two to four generations
removed from their source hatchery population The
pairwise FSTvalues are much higher for populations
from di¡erent strains compared with those withinstrains, which is consistent with the geographicorigin of the two strains (Bernatchez, Dempson &Martin 1998; Primmer, Aho, Piironen, Estoup, Cor-nuet & Ranta 1999) This result is supported by thetopology of the neighbour-joining phenogram whererobust bootstrap support (93%) existed for thebranch that divided the two strains Similar levelsand patterns of di¡erentiation have been observed instudies on Atlantic salmon Salmo salar (Norris et al.1999), rainbow trout Oncorhynchus mykiss (Butler &Cross 1996) and Arctic charr (Primmer et al 1999).The magnitude and the distribution of genetic var-iation within and among broodstock populations dif-fer between strains Twice as much of the totalvariation was partitioned among populations in theNauyuk Lake strain compared with the Fraser Riverstrain Moreover, the Nauyuk Lake strain has asigni¢cantly greater global FSTand Nei’s D than theFraser River strain as well as greater bootstrap supportfor branches that delineated di¡erent populationswithin the neighbour-joining phenogram The con-trasting pattern of within strain di¡erentiation mayre£ect the smaller founding populations from the
1978 Nauyuk Lake wild collection as opposed to largercollections over numerous years from the Fraser Riverwild population (Lundrigan et al 2005) In the NauyukLake strain, the elevated levels of di¡erentiation com-bined with the allelic diversity and allele frequencydata indicate a pattern of reduced variation withinand increased partitioning of variation among popula-tions, a trend consistent with other small cultured po-pulations (Fiumera, Parker & Fuerst 2000; Taniguchi
& Perez-Enriquiez 2000; Frankham et al 2002)
Analytical considerationsCertain factors may have confounded the various ana-lyses we conducted to test for bottlenecks Recently,bottlenecked populations will show transient symp-toms such as heterozygote excess, reduced allelic di-versity and a greater proportion of high-frequencyalleles (Luikart et al.1998; Ramstad et al 2004) Failure
to detect symptoms may be related to the features ofthe test rather than whether bottlenecks have oc-curred For example, it has been proposed that multi-ple bottleneck events can mask transient symptomssuch as heterozygote excess (Ramstad et al 2004) Inthis instance, it would be expected that other symp-toms such as allelic diversity and increased linkagedisequilibrium would still be evident, which is not thecase in our results The lack of evidence for bottlenecksAquaculture Research, 2011, 42, 1440^1452 Genetic divergence among broodstocks of Arctic charr C T Blackie et al.
Trang 37test based on the observed levels of heterozygosity
re-lative to allelic diversity may also result from a
poten-tial lack of statistical power in being able to detect such
events Although the test is recommended to be used
with 20 microsatellite loci, power analyses have
deter-mined that as few as four loci can be used successfully
to identify bottlenecks (Cornuet & Luikart 1996) The
lack of evidence for reductions in e¡ective population
size could also be interpreted as a recovery from earlier
bottlenecks (Luikart et al 1998) This scenario is not
likely because the transient symptoms would not have
had enough time to recover (it should take 0.2^4 Ne
generations; Luikart & Cornuet 1998) However, it is
not clear to what degree heterozygote excess following
a bottleneck event would be counteracted by the loss of
heterozygosity caused by drift, non-random mating
and selection (Kim,Withler, Ritland & Cheng 2004)
The use of six microsatellite loci in combination
with distance-based analytical approaches may have
potentially confounded our results in terms of
popu-lation di¡erentiation and the repopu-lationships among
populations For example, Barker, Bradley, Fries, Hill,
Nei andWayne (1993) recommended that 25
microsa-tellites be used to distinguish cultured breeds This
suggestion was, however, based on information
collected for 25 individuals at loci that had between
four and 10 alleles Our study was based on twice the
recommended number of individuals with allelic
counts between 11 and 26 This almost certainly has
reduced the uncertainty in using fewer loci
Never-theless, it is important to consider the potential
impli-cations of using six loci for distance-based methods
of phylogenetic reconstruction Takezaki and Nei
(1996) have indicated that sample sizes, number of
loci, genetic distance estimation parameters and tree
construction method are important factors when
con-sidering results in terms of the relationships among
populations Our study was based on relatively large
sample sizes with genetic distance estimates and tree
construction methods that are conservative to many
of the biases imposed using other approaches
(Takeza-ki & Nei1996), particularly with a small number of loci
(Angers & Bernatchez1998) These factors in
combina-tion with the high levels of di¡erentiacombina-tion detected
among broodstocks suggest that our conclusions
would not have been altered qualitatively if we had
used greater numbers of loci
Conclusions
This study indicates that all hatchery populations of
Arctic charr from a speci¢c strain are not genetically
homogeneous Although the existence of tion at neutral marker loci does not necessarilyequate to di¡erentiation at genes controlling perfor-mance traits, it does suggest the potential for suchdi¡erences exists Similarly, although direct parallelsbetween patterns of discrete molecular variation andgenetically based variation in important quantitativetraits are tenuous (Reed & Frankham 2001), the sub-stantial genetic variation we report at microsatelliteloci indicates that the recent demographics of Arcticcharr broodstock populations has been generallyamenable to the maintenance of genetic variation.Even though much of the literature on domesticatedanimals only recognizes variation at the strain levelfor enhancement and conservation (Barker 2001;Caballero & Toro 2002), it will be important to con-sider intra-strain variation in Canadian aquacultureArctic charr given the di¡erences that have beendetected in the current study
di¡erentia-AcknowledgmentsSeveral individuals assisted in the collection of ¢eldand genetic data, analyses and manuscript prepara-tion A Ferguson helped to coordinate ¢eld sampling
by initiating contact with several producers The tance in the collection of broodstock provided fromnumerous aquaculture producers is greatly appre-ciated M Robinson and J Poissant o¡ered helpful sug-gestions regarding data analyses This research wasfunded through a Natural Sciences and EngineeringResearch Council of Canada Strategic Project Grant
Angers B.L., Bernatchez L., Angers A & Desgroseillers L (1995) Speci¢c microsatellite loci for brook charr reveal strong population subdivision on a microgeographic scale Journal of Fish Biology 47(Suppl A),77^185 Bardacki F & Skibinski D.O.F (1994) Application of RAPD technique in tilapia ¢sh: species and subspecies identi¢- cation Heredity 73, 117^123.
Barker J.S.F (2001) Conservation and management of
genet-ic diversity: a domestgenet-ic animal perspective nal of Forest Research 31, 588^595.
Trang 38CanadianJour-Barker J.S.F., Bradley D.G., Fries R., HillW.G., Nei M & Wayne
R.K (1993) An Integrated Global Programme to Establish
Relationships Among the Breeds of Each Animal Species.
FAO Animal Production and Health Paper FAO, Rome,
Italy.
Bentsen H.B & Olesen I (2002) Designing aquaculture mass
selection programs to avoid high inbreeding rates
Aqua-culture 204, 349^359.
Bernatchez L., Dempson J.M & Martin S (1998)
Microsatel-lite gene diversity analysis in anadromous arctic charr,
Salvelinus alpinus, from Labrador, Canada Canadian
Jour-nal of Fisheries and Aquatic Sciences 55, 1264^1272.
Bolker B.M., Brooks M.E., Clark C.J., Geange S.W., Poulsen
J.R., Stevens M.H.H & White J.-S.S (2009) Generalized
linear mixed models: a practical guide for ecology and
evolution Trends in Ecology and Evolution 24, 127^135.
Brown B.,Wang H.P., Li L., Givens C & Wallat G (2007)
Yel-low perch strain evaluation I: genetic variance of six
broodstock populations Aquaculture 271, 142^151.
Brunner P.C., Douglas M.R & Bernatchez L (1998)
Microsa-tellite and mitochondrial DNA assessment of population
structure and stocking e¡ects in Arctic charr Salvelinus
alpinus (Teleostei: Salmonidae) from central Alpine Lakes.
Molecular Ecology 7, 209^223.
Brunner P.C., Douglas M., OsinovA.,Wilson C & Bernatchez
L (2001) Holarctic phylogeography of Arctic charr
(Salve-linus alpinus L.) inferred from mitochondrial DNA
se-quences Evolution 55, 573^586.
Butler A & Cross T.F (1996) Genetic di¡erences between
successive year classes of two strains of reared rainbow
trout, Oncorhynchus mykiss (Walbaum) Aquaculture
Re-search 27, 643^649.
Caballero A & Toro M.A (2002) Analysis of genetic diversity
for the management of conserved subdivided
popula-tions Conservation Genetics 3, 289^299.
Chen L., Li Q & Yang J (2008) Microsatellite genetic
varia-tion in wild and hatchery populavaria-tions of the sea
cucum-ber (Apostichopus japonicus Selenka) from northern China.
Aquaculture Research 39, 1541^1549.
Cornuet J.M & Luikart G (1996) Description and power
analy-sis of two tests for detecting recent population bottlenecks
from allele frequency data Genetics 144, 2001^2014.
Danzmann R.G., Cairney M., Davidson W.S., Ferguson M.M.,
Gharbi K., Guyomard R., Leder E., Okamoto N., Ozaki O.,
Rexroad C., Sakamoto T.,Taggart J.B & Woram R.A (2005)
A comparative analysis of the rainbow trout genome with
two other species of ¢sh (Arctic charr and Atlantic salmon)
within the tetraploid derivative Salmonidae family
(sub-family: Salmoninae) Genome 48, 1037^1051.
Ditlecadet D., Dufresne F., Le Francois L.R & Blier P.U (2006)
Applying microsatellites in two commercial strains of
Arctic charr (Salvelinus alpinus): potential for a selective
breeding program Aquaculture 257, 37^43.
Dray S & Dufour A.B (2007) The ade4 package:
implement-ing the duality diagram for ecologists Journal of Statistical
Software 22, 1^20.
Evans B., Bartlett J., Sweijd N., Cook P & Elliott N.G (2004) Loss of genetic variation at microsatellite loci in hatchery produced abalone in Australia (Haliotis rubra) and South Africa (Haliotis midae) Aquaculture 233, 109^127 Exco⁄er L., Smouse P.E & Quattro J.M (1992) Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochon- drial DNA restriction data Genetics 131, 479^491 Exco⁄er L., Laval G & Schneider S (2005) Arlequin (version 3.0): an integrated software package for population genet- ics data analysis Evolutionary Bioinformatics 1, 47^50 Falconer D.S & MacKay T.F.C (1996) Introduction to Quanti- tative Genetics Longman Group, Essex, UK , 464pp Felsenstein J (1993) PHYLIP ^ Phylogeny Inference Package Computer Software (version 3.5c) University of Washing- ton, Seattle,WA, USA.
Fiumera A.C., Parker P.G & Fuerst P.A (2000) E¡ective pulation size and maintenance of genetic diversity in cap- tive-bred populations of a Lake Victoria cichlid Conservation Biology 14, 886^892.
po-Fjalestad K.T., Moen T & Gomez-Raya L (2003) Prospects for genetic technology in salmon breeding programmes Aquaculture Research 34, 397^406.
Frankham R., Ballou J.D & Briscoe D.A (2002) Introduction
to Conservation Genetics Cambridge University Press, Cambridge, UK, 617pp.
Gjedrem T (2000) Genetic improvement of cold-water ¢sh species Aquaculture Research 31, 25^33.
Goudet J (2001) FSTAT, a program to estimate and test gene diversities and ¢xation indices (version 2.9.1) Updated from Goudet (1995) Available at http://www.unil.ch/ izea/softwares/fstat.htm
Goudet J., Raymond M., De Meeus T & Rousset F (1996) ing di¡erentiation in diploid populations Genetics 144, 1933^1940.
Test-Had¢eld J.D (2010) MCMC methods for multi-response eralised linear mixed models: the MCMCglmm R package Journal of Statistical Software 33, 1^22.
gen-Hartl D.L (1988) A Primer of Population Genetics, 2nd edn nauer Associates, Sunderland, MA, USA.
Si-Hartl D.L & Clark A.G (1989) Principles of Population ics, 2nd edn Sinauer Associates, Sunderland, MA, USA Jackson T.R., Ferguson M.M., Danzmann R.G., Fishback A.G., Ihssen P.E., O’Connell M & Crease T.J (1998) Identi-
Genet-¢cation of two QTL-in£uencing upper temperature ance in three rainbow trout (Oncorhychus mykiss) half-sib families Heredity 80, 143^151.
toler-Jost L (2008) G ST and its relatives do not measure tion Molecular Ecology 17, 4015^4026.
di¡erentia-Kim J.E.,Withler R.E., Ritland C & Cheng K.M (2004)
Genet-ic variation within and between domestGenet-icated chinook salmon, Oncorhychus tshawytscha, strains and their progenitor populations Environmental Biology of Fishes
69, 371^378.
Lind C.E., Evans B.S., Knauer J., Taylor J.J.U & Jerry D.R (2009) Decreased genetic diversity and a reduced e¡ective Aquaculture Research, 2011, 42, 1440^1452 Genetic divergence among broodstocks of Arctic charr C T Blackie et al.
Trang 39population size in cultured silver-lipped pearl oysters
(Pinctada maxima) Aquaculture 286, 12^19.
Luikart G & Cornuet J.M (1998) Empirical evaluation of a test
for identifying recently bottlenecked populations from
al-lele frequency data Conservation Biology 12, 228^233.
Luikart G., Allendorf F.W., Cornuet J.M & Sherwin W.B.
(1998) Distortion of allele frequency distributions
pro-vides a test for recent population bottlenecks Journal of
Heredity 89, 238^247.
Lundrigan T.A., Reist J.D & Ferguson M.M (2005)
Microsa-tellite genetic variation within and among Arctic charr
(Salvelinus alpinus) from aquaculture and natural
popula-tions in North America Aquaculture 244, 63^75.
Maruyama T & Fuerst P.A (1985) Population bottlenecks
and nonequilibrium models in population genetics
Ge-netics 111, 675^689.
Nei M., Tajima F & Tateno Y (1983) Accuracy of estimating
phylogenetic trees from molecular data II Gene
fre-quency data Journal of Molecular Evolution 19, 153^170.
Norris A.T., Bradley D.G & Cunningham E.P (1999)
Micro-satellite genetic variation between and within farmed
and wild Atlantic salmon (Salmo salar) populations
Aqua-culture 180, 247^264.
Olsen J.B., Bentzen P & Seeb J.E (1998) Characterization of
seven microsatellite loci derived from pink salmon
Mole-cular Ecology 7, 1087^1088.
Porta J., Porta J.M., Canavate P., Martinez-Rodriguez G &
Al-varez M.C (2007) Substantial loss of genetic variation in a
single generation of Senegalese sole (Solea senegalensis)
culture: implications in the domestication process
Jour-nal of Fish Biology 71, 223^234.
Presa P & Guyomard R (1996) Conservation of
microsatel-lites in three species of salmonids Journal of Fish Biology
49, 1326^1329.
Primmer C.R., Aho T., Piironen J., Estoup A., Cornuet J.M &
Ranta E (1999) Microsatellite analysis of hatchery stocks
and natural populations of Arctic charr, Salvelinus
alpi-nus, from the Nordic region: implications for
conserva-tion Hereditas 130, 277^289.
Ramstad K.M., Woody C.A., Sage G.K & Allendorf F.W.
(2004) Founding events in£uence genetic population
structure of sockeye salmon (Oncorhynchus nerka) in Lake
Clark, Alaska Molecular Ecology 13, 277^290.
Raymond M & Rousset F (2001) Genepop 3.3 (updated from
version 1.2 1995): population software for exact tests and
ecumenicism Journal of Heredity 86, 248^249.
Reed D.H & Frankham R (2001) How closely correlated
are molecular and quantitative measures of genetic
variation? A meta-analysis Evolution 55, 1095^1103.
Rice W.R (1989) Analyzing tables of statistical tests.
Evolution 43, 223^225.
Rogers B & Davidson W.S (2001) Arctic charr aquaculture
review Report to PEI Aquaculture Alliance and
Aquacul-ture Association of Nova Scotia 19pp.
Romana-Eguia R.R., Ikeda M., Basiao Z.U & Taniguchi N (2004) Genetic diversity in farmed Asian Nile and red hybrid tilapia stocks evaluated from microsatellite and mitochondrial DNA analysis Aquaculture 236, 131^150.
Saisa M., Koljonen M.L & Tahtinen J (2003) Genetic changes in Atlantic salmon stocks since historical times and the e¡ective population size of a long-term captive breeding program Conservation Genetics 4, 613^627 Skaala O., Taggart J.B & Gunnes K (2005) Genetic di¡er- ences between ¢ve major domesticated strains of Atlantic salmon and wild salmon Journal of Fish Biology 67, 118^128.
Takezaki N & Nei M (1996) Genetic distances and struction of phylogenetic trees from microsatellite DNA Genetics 144, 389^399.
recon-Taniguchi N & Perez-Enriquiez R (2000) Genetic tion of brood stock for aquaculture of red sea bream by DNA markers In: Recent Advances in Marine Biotechnol- ogy Aquaculture Part B Fishes,Vol 4 (ed by M Fingerman
evalua-& R Nagabhushanam), pp 1^16 Department of Ecology, Evolution and Organismal Biology, Tulane University, Science Publishers, En¢eld, NH, USA.
Taniguchi N., Perez-Enriquez R & Estu N (2003) DNA kers as a tool for genetic management of brood stock for aquaculture In: Aquatic Genomics: Steps Towards a Great Future (ed by N Shimizu, T Aoki, I Hirono & F Takashima), pp 417–429 Springer-Verlag, Tokyo Taylor E.B., Redenbach Z., Costello A.B., Pollard S.M & Pacas C.J (2001) Nested analysis of genetic diversity in north- western North American Char, Dolly Varden (Salvelinus malma) and bull trout (S con£uentus) Canadian Journal of Fisheries and Aquatic Sciences 58, 406^420.
mar-Waples R.S (2002) De¢nition and estimation of e¡ective population size in the conservation of endangered spe- cies In: PopulationViabilityAnalysis (ed by S.R Beissinger
& D.R McCullough), pp 147^168 University of Chicago Press, Chicago, IL, USA.
Weir B.S & Cockerham C.C (1984) Estimating F-statistics for the analysis of population structure Evolution 38, 1358^1370.
Wilson A.J., Gislason D., Skulason S., Snorrason S., Adams C.E., Alexander G., Danzmann R.G & Ferguson M.M (2004) Population genetic structure of Arctic charr, Salve- linus alpinus from northwest Europe on large and small spatial scales Molecular Ecology 13, 1129^1142 Wilson C.C., Hebert P.D.N., Reist J.D & Dempson J.B (1996) Phylogeography and postglacial dispersal of Arctic charr Salvelinus alpinus in North America Molecular Ecology 5, 187^197.
Woodworth L.M., Montgomery M.E., Briscoe D.A & ham R (2002) Rapid genetic deterioration in captive po- pulations: causes and conservation implications Conservation Genetics 3, 277^288.
Trang 40Frank-The effects of different Ca21 concentration fluctuation
on the moulting, growth and energy budget of juvenile
Chunqiang Hou, Fang Wang, Shuanglin Dong & Yujie Zhu
The key laboratory of Mariculture, Ministry of Education, Ocean University of China, Qingdao, China
Correspondence: F Wang,The key laboratory of Mariculture, Ministry of Education, Ocean University of China, Qingdao 266003,China E-mail: wangfang249@ouc.edu.cn
Abstract
A 40-day experiment was conducted to investigate
the e¡ects of di¡erent Ca21concentration
£uctua-tion on the moulting, growth and energy budget
of juvenile Litopenaeus vannamei with an initial
wet body weight of (1.20 0.01) g The Ca21
tration of the control group C385 was 385 mg L 1
concen-throughout the experiment, while the Ca21
con-centration of the C591, C803, C1155 and C2380
groups periodically £uctuated from 385 to 591, 803,
1155 and 2380 mg L 1 respectively The moulting
frequency (MF) of the shrimp in the Ca21
concen-tration £uctuating groups was signi¢cantly higher
than those in the control group (Po0.05).The speci¢c
growth rates (SGRd) and feed intake of the shrimp
in the C591 and C803 groups were signi¢cantly
higher than those in other groups (Po0.05), but
no signi¢cant di¡erences in feed e⁄ciency were
found among all groups (P40.05) The shrimp in
C591 and C803 groups spent signi¢cantly less energy
in respiration, while depositing signi¢cantly more
energy for growth than those in other three groups
(Po0.05) These results showed that proper Ca21
concentration £uctuation could increase the MF
and growth rate of the juvenile L vannamei, and
according to the regression formula made using
SGRd and range of Ca21 concentration
£uctuat-ion, periodically enhanced Ca21 concentration of
295 mg L 1in the seawater was suggested to be used
in shrimp culture
Keywords: Litopenaeus vannamei, Ca21
concen-tration £uctuation, moulting, growth, energy
budget
IntroductionGrowth in crustaceans is con¢ned by the exoskeleton
In order to increase in size, they must moult Penaeids,like other crustaceans, moult at intervals of a few days
or weeks, and increase in size in a series of steps: a pid enlargement at ecdysis is followed by a period oflittle or no increase until the next ecdysis (Dall, Hill &Sharples 1990; Zhang, Zhang & Li 2006) Moultingactivity persists in the entire life of shrimp Penaeidsneed calcium to harden the new exoskeleton after ec-dysis, but they store little calcium and must obtain thebulk of their requirement from the water (Peter 1985;Wang 1997) In addition to constituting the exoskele-ton, calcium is also involved in muscle activity, nervetransmission and osmoregulation (Li 1994) Therefore,calcium plays an important role in the moulting andgrowth of penaeids
ra-Salinity is one of the most important abiotic factors
in aquaculture, and it may rapidly or slowly alter cause of rainfall and evaporation or water exchangesduring the cultivation period Recent researchshowed that optimal £uctuating salinity could bemore useful than constant salinity for the growth ofshrimp (Mu, Wang, Dong, Dong & Zhu 2005; Ding,Wang, Guo & Li 2008; Feng, Tian, Dong, Su, Wang &
be-Ma 2008) There are many ions in seawater, calcium
is one of the most abundant cations in seawater,and salinity £uctuation could result in Ca21concen-tration £uctuation Although a few investigations onthe e¡ects of calcium concentration on shrimpgrowth and survival have been conducted, most ofthem focused on the e¡ects of constant calcium con-centration (Dong, Du & Lai 1994; Dong, Dong,Wang,
Mu, Zhu & Huang 2005), and no work on the e¡ects of