Aquaculture nutrition, tập 19, số 2, 2013

calcu-To evaluate the water quality, dissolved oxygen DO wasrespectively determined using a DO meter on a daily basis,and ammonia-N, nitrite-N and phosphate were measuredaccording to the

1 2 2 2 1 1

Department of Aquaculture, National Pingtung University of Science and Technology, Pingtung, Taiwan; 2 Department ofFood Science, National Pingtung University of Science and Technology, Pingtung, Taiwan

To produce a thermostable and neutral phytase (phy) of

Bacillus subtilis E20 in Escherichia coli HMS174 and

evalu-ate its efficiency in improving growth performance The

phy C of B subtilis E20 was expressed in E coli HMS 174,

and then the 42-kDa recombinant phy C was purified by

Ni-NAT and analysed by SDS–PAGE The recombinant

phy C had optimal ranges of pH of 6~ 7 and temperature

of 50 ~ 60 °C A thermostability analysis showed that the

enzyme is a thermostable phytase, and around 33% of

residual activity was detected after being incubated at

90 ~ 100 °C for 10 min The recombinant phy C-pretreated

soybean meal for feed preparation improved white shrimp,

Litopenaeus vannamei, growth and feed efficiency Overall,

the neutral and thermostable phy C is suitable for

aqua-feed, and it is able to improve the nutritional utilization,

resulting in enhanced shrimp growth and reduced feed

costs

Litopenaeus vannamei, phytase, soybean meal

Received 4 August 2011; accepted 8 January 2012

Correspondence: Chun-Hung Liu, Department of Aquaculture, National

Pingtung University of Science and Technology, Pingtung 91201,

Taiwan E-mail: chliu@mail.npust.edu.tw

There is continuing interest in identifying and developing

ingredients from plants as alternatives to the high cost of

fish meal for use in aquafeed (Hardy 1995; Tacon et al.1998) Nowadays, concern is raised about the negativeimpacts on global fish meal production of overfishing andnatural disasters, such as the tsunami in Chile Amongplant feedstuffs being investigated to replace fish meal, soy-bean meal is considered a promising protein source because

of its stable supply, price and nutritional composition.One of the major problems associated with the use ofplant proteins in aquafeed is the presence of anti-nutri-tional factors, such as phytate (myo-inositol-1,2,3,4,5,6-hexakisphosphates), which is the main storage form ofphosphate (P) in plant feedstuff Up to 80% of the total Pcontent in plants may be present in the form of phytate(Lall 1991; Eeckhout & De Paepe 1994) and is not practi-cally available for monogastric or agastric aquatic animals(National Research Council (NRC) 1993) In addition tophytate-P, the absorption and bioavailability of indispens-able minerals such as calcium, zinc, iron and magnesiumcan be negatively affected by forming insoluble chelatedcomplexes with phytase (Papatryphon et al 1999) In addi-tion, phytate can also combine with proteins and vitamins

in insoluble complexes (Liu et al 1998; Sugiura et al.2001), resulting in their decreased utilization, activity anddigestibility because of interference with lipid and starchdigestibility (Cosgrove 1996) Therefore, the issue of phy-tate must be dealt with to improve nutrition utilization ofsoybean meal and increase substitution ratios of fish meal

to bring down the consequent costs of aquafeed

Microbial phytase is widely used in feed for animal duction to minimize the problem of phytate and is included

pro-in aquafeed (Cao et al 2007) to improve the nutritivevalue (Forster et al 1999), the digestibility of nutrients and

.

Aquaculture Nutrition

minerals (Cheng & Hardy 2003; Yoo et al 2005), growth

performance (Sajjadi & Carter 2004) and reduce

phospho-rus discharge (Forster et al 1999; Sajjadi & Carter 2004)

On the contrary, dietary phytase supplementation had no

effect on the growth of tiger shrimp, Penaeus monodon

(Biswas et al 2007) Actually, most commercial phytase,

adapted for digestive systems of livestock animals, may

dif-fer in its efficiency of phytate degradation in fish (Liebert

& Portz 2005) because of the varying physiological

condi-tions in different species because of its great dependence on

the gut pH In addition, native phytase activity may also

be inactivated by heat treatment during aquafeed

process-ing Therefore, developing a neutral phytase with improved

thermal stability is important and suitable for some fish

and/or shrimps with near-neutral digestive tracts Shen

et al (2004) reported that the pH values in the intestine,

liver, and stomach of Litopenaeus vannamei were 5.9~ 6.1

(6.08± 0.04), 6.7 ~ 7.0 (6.82 ± 0.04) and 5.1 ~ 5.3 (5.22 ±

0.04), respectively

Phytase with a beta-propeller structure is mainly isolated

from Bacillus sp The phytase (phy C) of Bacillus is a

non-glycosylated protein with higher thermal stability and an

optimal pH in the neutral range (Gulati et al 2007; Rao

et al 2008; Guerrero-Olazara´n et al 2010), in contrast to

commercial production of phytase, which is currently

focused on fungal histidine acid phytase from Aspergillus

sp (Ha et al 2000; Cao et al 2007) Therefore, we report

the expression of phy C from Bacillus subtilis E20 in

Escherichia coli HMS147, and the biochemical properties

of the recombinant phytase The enzyme was subsequently

used to pretreat soybean meal for experimental diet

prepa-ration to evaluate the improvement in the growth

perfor-mance of white shrimp, L vannamei juveniles

Escherichia coli XL-1 Blue was used in the cloning steps

and routinely grown with shaking in Luria-Bertani (LB)

broth (L3022; Sigma, St Louis, MO, USA) at 37 °C

Bacil-lus subtilis E20 isolated from natto (Liu et al 2009) was

used as a source of the phy C gene and grown in nutrient

broth (NB) (BD, Sparks, MD, USA) at 40°C with

shak-ing Escherichia coli HMS 147 (DE3) (Navagen, Madison,

WI, USA) was used for phytase expression and grown in

LB broth with shaking The LB broth for growing both of

the E coli strains was supplemented with ampicillin

by a polymerase chain reaction (PCR) with the 6XHisphyF (5′-CAACATATGCACCACCACCACCACCAAATCATCAAAAACACTTTTGT-3′) primer with insertions ofthe Nde I site and His-tag encoding sequence, and phy-R(5′-TTGGATCCTTATTTTCCGCTTCTGTC-3′) primerwith insertion of the Bam HI site The following PCRprotocol was used: initial denaturation at 95°C for 5 minand 20 s; followed by 30 cycles of denaturation at 95°Cfor 1 min, annealing at 54°C for 5 min, and extension at

72°C for 1.5 min; followed by an additional incubation at

72°C for 10 min and then 4 °C for 30 s at the end of thefinal cycle After the PCR, samples were analysed by elec-trophoresis on 20lg mL 1 agarose gels, and then theamplified PCR fragment was purified by a Gel ExtractionSystem (GE10200; ALS, Kaohsiung, Taiwan) The purifiedPCR fragments were cloned into the PCRII TOPO vector

of the TOPO TA cloning system (Invitrogen, Carlsbad,

CA, USA) and transferred into E coli XL-1 Blue ing to the manufacturer’s protocol and then subjected to asequence analysis

accord-The PCRII TOPO vector containing the phy C gene wasdigested with Nde I and Bam HI and then analysed by elec-trophoresis on 20lg mL 1

agarose gels, and the phy Cgene was purified using the Gel Extraction System(GE10200; ALS) Thereafter, the purified phy C gene wasligated with the Nde I/Bam HI-digested pET 25b(+) vector

to generate 6XHis-phytase DNA-pET 25b(+) The ligationmixture was used to transform competent E coli HMS 174cells by electroporation (Micro PulserTM; Bio-Rad, CA,USA) using 0.1-cm cuvettes and a 1.8-kV pulse and thenchecked for protein expression

The 6XHis-phytase DNA-pET 25b(+) transformant wasgrown in LB medium containing 100lg mL 1 ampicillinand agitated at 200 rpm at 37°C until the OD600 reached0.5 Isopropyl-b-D-thiogalactopyranoside (IPTG at0.1 mM) was then added, and the culture was shifted to

23°C for 16 h because the production levels of the

.

Aquaculture Nutrition, 19; 117–127 ª 2012 Blackwell Publishing Ltd

recombinant phy C was found to be higher at 23 than at

20 and 28°C After 16 h of incubation, the sample was

collected for electrophoresis and phytase activity analysis

Cells harvested from the above induced culture were

sub-jected to centrifugation at 5000 xg and 4°C for 10 min

The pelleted cells were resuspended in 10 mM Tris buffer

(pH 8) containing 1 mM CaCl2 and sonicated on ice using

MicrosonTM XL 2000 (Misonix, New York, NY, USA)

Subsequently, they were centrifuged at 11 000 g for 30 min

at 4°C The suspension was used for the phytase activity

assay

The enzyme assay method was modified from Shimizu

(1992) Phytase activity was assayed by measuring the rate

of increase in inorganic orthophosphate cleaved from

phy-tate A reaction mixture containing 100lL of the enzyme

preparation and 300lL of 1 mg L 1

phytate in 10 mM

Tris buffer (pH 8) containing 1 mM CaCl2 was incubated

at 50°C for 15 min Then, the reaction was stopped by

adding 400 lL of 50 lg mL 1trichloroacetic acid The

lib-erated phosphate was measured at 700 nm by following the

production of phosphomolybdate with 1.5 mL of colour

reagent (freshly prepared by mixing four volumes of

30 lg mL 1

ammonium molybdate solution in 50lg mL 1sulphuric acid and one volume of 50 lg mL 1

ferrous phate solution) One unit of phytase activity was defined as

sul-the amount of enzyme hydrolysis of 1lmmol

phos-phate min 1 under the assay conditions The protein

con-centration in the enzyme preparations was determined

using the protein assay dye (Bio-Rad, ST, USA)

Samples were fractionated by sodium dodecylsulphate

poly-acrylamide gel electrophoresis (SDS–PAGE) using a

0.12 g mL 1 SDS–PAGE gel Proteins in the resolved gel

were detected by Coomassie Brilliant Blue R250 For the

phytase zymogram analysis, the gel was soaked in

10 lg mL 1

Triton X-100 for a period of 1 h at room perature, and then it was moved to 0.1M sodium acetate

tem-buffer (pH 5) at a temperature of 4°C for 1 h Thereafter,

the gel was changed to 0.1Msodium acetate buffer (pH 5)

containing 4 lg mL 1phytate at 4°C for 16 h for phytase

activity detection Activity bands were visualized after the

gel was immersed in a 20lg mL 1

cobalt chloride solutionfor 5 min at room temperature, and then the cobalt chlo-

ride solution was replaced with a freshly prepared solution

containing equal volumes of a 62.5lg mL ammoniummolybdate solution and 4.2lg mL 1

ammonium vanadatesolution Phytase activity was evident as zones of clearing

on an opaque background (Bae et al 1999)

The recombinant phy C was purified using an acetic acid (NTA) column A sample in 10 mMTris buffercontaining 150 mM NaCl and 1 mM CaCl2 at pH 7.6 wasapplied to the Ni-NTA matrix and then washed with theabove buffer containing 20 mM imidazole and then elutedwith the above buffer except that the imidazole concentra-tions used were 50 and 100 mM

Ni-nitrilotri-The effect of pH on the enzyme activity at 50°C wasdetermined Purified recombinant phytase was replacedwith buffer using Amicon YM-10 (Millipore, Bedford,

MA, USA) with various pH buffers of 200 mMsodium tate (pH 3~ 6) or 100 mM Tris buffer (pH 7~ 10) There-after, enzyme assays were performed as described earlierwith phytate in different buffers at pH values of 3~ 10.Temperature effects on enzyme activity were determined

ace-in this study The enzyme and substrate, phytate, were firstmixed; then the reaction mixtures were placed at varioustemperatures of 20~ 80 °C, and the enzyme activity wasanalysed as described above

Metal requirements for the active conformation of theenzyme were tested by analysing the enzyme in 100 mM

Tris buffer (pH 8.0) only or supplemented with 1 mM

CaCl2 or 1 mM CaCl2 plus 10 mM EDTA Enzyme assayswere performed as described earlier

For analysis of the thermal stability, purified recombinantphytase was incubated at various temperatures of 10, 20,

30, 40, 50, 60, 70, 80, 90 and 100°C for 10 min, followed

by cooling to 28°C Phytase activity was detected by themethod described earlier

Recombinant phy C preparation and use for soybean mealpretreatment Recombinant phy C was produced and

purified as described earlier The purified enzyme was then

diluted by a buffer (10 mM Tris–HCl; pH, 7.6; 150 mM

NaCl; 1 mM CaCl2) to prosperity concentration for

soy-bean meal treatment In addition, the inactive enzyme was

prepared by incubation at 100°C for 1 h, and then the

enzyme activity was detected as described earlier

For recombinant phy C-treated soybean meal

prepara-tion, soybean meal for each experimental diet, preground

and passed through a 60-mesh screen was evenly mixed

with purified recombinant phy C at 50°C for 1 h and then

was dried in an incubator with a dehumidifier at 4°C until

the moisture was <100 g kg 1before diet preparation The

soybean meal treated only with inactive enzyme was used

for control For active enzyme-treated soybean meal, active

enzyme was added to the soybean meal (400 g) at levels of

250, 500, 1000 and 2000 FTU with a corresponding

decrease in the amount of inactive enzyme Amount of

enzyme and inactive enzyme used for soybean meal

treat-ment were listed, respectively, in Tables 1 and 2

Preparation of experimental diets Two batches of

experi-mental diets included the diets containing phosphorus

sup-plement and enzyme-pretreated soybean meal, and the dietswithout phosphorus supplement but with enzyme-pretreat-

ed soybean meal were prepared, respectively The tion of the experimental diets is presented, respectively, inTables 1 and 2 The moisture, ash, crude protein, etherextract and total phosphate of key ingredients includingfishmeal and soybean meal were 100.8, 148.2, 663.2, 80.7and 20.1 g kg 1, and 100.3, 73.2, 421.9, 30 and 6.9 g kg 1,respectively Briefly, the ingredients except enzyme-pretreat-

formula-ed soybean meal were ground up in a Hammer mill to passthrough a 60-mesh screen The experimental diets were pre-pared by mixing the dry ingredients and enzyme-pretreatedsoybean meal, and then adding water until a stiff doughresulted Each diet was then passed through a mincer with

a die, and the resulting spaghetti-like strings were dried in

an incubator with a dehumidifier at 4°C until the moisturewas <100 g kg 1 After drying, the pellets were stored inplastic bins at 4°C until being used Crude protein, crudefat (ether extract), ash and moisture contents of the experi-mental diets were determined according to AOAC (1995)

Samples for total phosphorus (P) were digested in nitricacid and perchloric acid and measured colorimetrically

Table 1 Ingredients of the first batch of experiment diets (g kg 1 ) containing soybean meal pretreated by different levels of phytase for

white shrimp rearing

Vitamin mixture supplied the following [IU (mg) kg 1 mixture]: Vit A = 3 000 000 IU kg 1

; Vit B 1 = 1000 mg; Vit B 2 = 2000 mg; Vit

D 3 = 5 000 000 IU kg 1

; Vit E = 2000 IU kg 1

; Vit B 6 = 1000 mg; Vit K 3 = 500 mg; CuSO 4 = 1000 mg; ZnO = 6000 mg; Methionine = 10 gm; C 18 H 32 CaN 2 O 10 = 3000 mg; MnO = 25 gm; 6000 mg; C 6 H 5 O 2 N = 12 gm; C 19 H 19 N 7 O 6 = 1000 mg; CoCO 3 = 10 mg; Cd = <10 ppm;

Hg = <3 ppm; Pb = <20 ppm; As = <3 ppm; Cr = <10 ppm (Bellcom feed company, Taiwan).

3 Mineral mixture supplied the following (mg g 1 mixture): K 2 HPO 4 = 200; NaH 2 PO 4 = 215; Ca(H 2 PO 4 ) ·H 2 O = 265; CaCO 3 = 105; KCl = 28;

KI = 0.233; MgSO 4 ·7H 2 O = 100; CuCl 2 ·2H 2 O = 0.153; AlCl 3 ·6H 2 O = 0.240; ZnSO 4 ·7H 2 O = 0.240 All ingredients were diluted with

a-cellu-lose to 1 g.

.

Aquaculture Nutrition, 19; 117–127 ª 2012 Blackwell Publishing Ltd

using ascorbic acid as the reducing agent in developing

molybdenum blue colour (Allen 1989) No significant

dif-ference in proximate analysis was found between the

exper-imental diets, and the range of moisture, ash, crude protein

and crude fat were 86.5–92.5, 99.8–108.6, 348.9–358.1 and

Growth performance trial White shrimp, Litopenaeus

van-namei, was sampled from a farm at the Department of

Aquaculture, National Pingtung University of Science and

Technology, Pingtung County, Taiwan Prior to the

experi-ment, shrimps were acclimatized in a cement tank

(69 291 m) of 12 metric ton with a 5 metric ton of

aer-ated brackish water (20 g L 1) at 28°C for 1 week

Shrimps were fed the control diet at 2% of body weight

twice a day until the experiment began

Two 56-day growth trials including shrimps fed

respec-tively with different batch of diets were conducted in the

sinking cages (1.2 m in diameter and 0.8 m in height)

placed in the outdoor cement tanks (69 291 m) of 10

metric ton containing 5 ton of 20 g L 1water For the first

growth trial, 450 shrimps (with a mean weight of

0.27± 0.01 g) were randomly assigned to five groups,

including shrimps fed individually with the diets (with

phosphorus supplement) containing inactive phy C trol) or phy C-pretreated soybean meal at the levels of 250,

(con-500, 1000 and 2000 FTU Each cage consisted of 30shrimps Tests were carried out in triplicate For the secondgrowth trial, 360 shrimps (with a mean weight of0.61± 0.02 g) were randomly assigned to four groups,including shrimps fed individually with the diets (withoutphosphorus supplement) containing inactive phy C (con-trol) or phy C-pretreated soybean meal at the levels of 500,

1000 and 2000 FTU Each cage consisted of 30 shrimps.Tests were carried out in triplicate

During the experiment, aeration was supplied via a singleair-stone to maintain the dissolved oxygen at  6 mg L 1

.Water temperature was controlled at 28± 1 °C by heater.Shrimps were fed twice daily at a ratio of 10% of theirbody weight at 08 : 00 and 15 : 00 Any uneaten portionwas collected 1 h after feeding and then immediately dried

in an oven at 80°C The amount of all diets fed was lated by subtracting the uneaten portion, and data wererecorded on a daily basis The weight of the shrimp wasmeasured at the start of the experiment and at 2-weekintervals until the end of the experiment During the cul-ture period, ~1/3 water was renewed once every 2 weeks

calcu-To evaluate the water quality, dissolved oxygen (DO) wasrespectively determined using a DO meter on a daily basis,and ammonia-N, nitrite-N and phosphate were measuredaccording to the methods of Bower and Bidwell (1978),

Table 2 Ingredients of the second batch of experiment diets (g kg ) containing soybean meal pretreated by different levels of phytase for white shrimp rearing

1 The description of the source of fish meal and dehulled soybean meal, and composition of vitamin mixture is as Table 1.

2 Mineral mixture supplied the following (mg g 1 mixture): CaCO 3 = 105; KCl = 28; KI = 0.233; MgSO 4 ·7H 2 O = 100; CuCl 2 ·2H 2 O = 0.153; AlCl 3 ·6H 2 O = 0.240; ZnSO 4 ·7H 2 O = 0.240 All ingredients were diluted with a-cellulose to 1 g.

Bendschreider and Robinson (1952) and Murphy & Riley

(1962), respectively, once every 2 weeks The biological

per-formance criteria were the growth rate, percentage weight

gain, feed efficiency and survival rate

Statistical analyses of the data obtained from these

experi-ments were conducted by the one-way analysis of variance

SAScomputer software (SAS Institute, Cary, NC, USA) to

determine whether significant differences (P< 0.05) existed

among treatment means All percentage data were

trans-formed to square-root arcsine values before the statistical

analysis In addition, linear regression of thermostability

analysis of recombinant phytase at different temperature

was analysed using Microsoft Excel (Microsoft

Corpora-tion, Redmond, WA, USA)

Because purification of proteins from the cell content is

very tedious and time consuming, the vector containing the

gene for the target protein used a His6-tag to generate a

recombinant protein with six histidines for a one-step

pro-cedure of protein purification by affinity chromatography

In the cloning procedure, the resulting phytase gene

encoded six additional amino acids of histidines from the

His-tag at the N-terminal of the recombinant phytase in

this study After protein expression, the cell lysis

superna-tant was directly loaded onto an Ni-NTA column Figure 1

shows that 50 mM imidazole was necessary for

non-recom-binant protein washing, and 100 mMimidazole was able to

elute the pure enzyme Molecular weights of the proteins

were ~42 kDa as estimated by SDS–PAGE (Fig 1a) For

the phytase zymogram analysis, only a single 42-kDa band

was visualized in the crude protein and purified protein

from the Ni-NTA column by elution with 100 mM

imidaz-ole (Fig 1b)

The optimal conditions of the pH value and temperature

for recombinant phy C activity are shown in Fig 2 The

pH value for the maximum activity of recombinant phy C

was 6~ 7 (Fig 2a) Activities of the enzyme at pH 4 and 3

decreased by 17.2% and 24.1%, respectively However,activities of the enzyme, respectively, decreased by 52.8%

and 66.5% when the pH values reached 9 and 10 Theenzyme had an optimal reaction temperature of 50~ 60 °C

Figure 1 SDS–PAGE (a) and zymogram (b) analysis of the binant phytase (phy) C at each purification step using Ni-NTA.

recom-M, marker; lanes 1 and 6: crude enzyme; lanes 2 and 7: unbound protein; lanes 3 and 8: protein eluted with 20 mM imidazole buffer;

lanes 4 and 9: protein eluted with 50 mM imidazole buffer; and lanes 5 and 10: purified recombinant phy C eluted with 100 mM imidazole buffer.

0 20 40 60 80 100 120

cd c

d

Figure 2 Effects of pH (a) and temperature (b) on the recombinant phytase (phy) C activity Enzyme activity is expressed as relative values Data (mean ± SE) among different treatments with differ- ent letters significantly differ (n = 3).

.

Aquaculture Nutrition, 19; 117–127 ª 2012 Blackwell Publishing Ltd

for phytate hydrolysis (Fig 2b) The enzyme activity

decreased by 31.5% when the reaction temperature reached

70 °C In addition, enzyme activities significantly decreased

by 38.1% and 51.1% at 40 and 30 °C compared to 60 °C

No enzyme activity was detected in a buffer without

CaCl2 Enzymes in the presence of CaCl2 had significant

activity of hydrolysing phytate compared with that of the

enzyme in Tris buffer only or plus CaCl2and EDTA

Enzyme activities at various temperatures of 10~ 100 °C

for 10 min are shown in Fig 3 Enzyme activity decreased

as the temperature increased to 100 °C for 10 min The

highest enzyme activity was recorded in the treatment with

enzyme incubation at 10°C for 10 min No significant

dif-ferences in enzyme activities were seen at 20~ 50 °C When

the temperature increased to 60°C, the enzyme activity

remained at 50.9% The enzyme activity decreased by

66.1% When the temperature reached 100°C for 10 min,

33% of the enzyme activity remained

During the culture experiment, no water parameters

showed a significant difference between the control and

various treatments in two growth trials, and the water

quality was always within an acceptable range For the first

trial, the range of pH was 7.6~ 8.0 Levels of ammonia-N,

nitrite-N and phosphate were 0.01~ 0.53, 0.06 ~ 3.72 and

0.26~ 0.78 mML 1, respectively For the second trial, the

range of pH was 7.5~ 8.0 Levels of ammonia-N, nitrite-N

and phosphate were 0.02~ 0.65, 0.26 ~ 6.71 and

0.16~ 0.54 mML 1, respectively

In the first growth trial, shrimps have a higher growthperformance when they were fed the diets (with phosphorusaddition) containing recombinant phy C-pretreated soy-bean meal at the levels of 1000 and 2000 FTU (Table 3).Body weights of shrimps fed the diets containing recombi-nant phy C-pretreated soybean meal at the levels of 1000and 2000 FTU were 1.24 and 1.27-folds higher than that ofshrimps fed control diet at the 56 days of rearing No sig-nificant difference was seen in survival rates between thecontrol and various treatments Shrimps fed the diet con-taining recombinant phy C-pretreated soybean meal at thelevels of 1000 and 2000 FTU had a significantly higher per-centage of weight gain compared with that of shrimps fedthe diets of control and 250 FTU (Table 4)

The second growth trial confirms that shrimps showedgood growth even fed with a diet lack of phosphorus whenthe diet containing phy C-pretreated soybean meal at thelevel of 2000 FTU (Table 5) However, the growth perfor-mance was significantly higher in shrimps fed with the dietcontaining the 2000 FTU phy C-pretreated soybean mealthan that of control and 500 FTU at the end of trial Nosignificant difference was found in survival rates betweenthe control and treatments (Table 6) Shrimps fed the diet(without phosphorus supplement) containing 2000 FTUphy C-pretreated had significant better percentage ofweight gain and feed efficiency compared with control and

Y = –0.1652X + 21.592

0 5 10

Figure 3 Thermostability analysis of the recombinant phytase

(phy) C at a temperature range of 10 –100 °C.

1999; Sajjadi & Carter 2004; Yoo et al 2005), which help

shrimps digest nutrients from fish meal and soybean meal

The phytase gene in B subtilis E20 was investigated for

use as an indicator to detect B subtilis E20 colonization in

the shrimp intestine, and it is considered to be an

impor-tant enzyme for improving shrimp growth (Liu et al 2009)

Bacillus sp phytase properties were examined in several

studies through native protein purification (Kerovuo et al

1998; Kim et al 1998; Gulati et al 2007;

Guerrero-Ola-zara´n et al 2010) or recombinant protein expression in

bacterial systems (Kerovuo & Tynkkynen 2000; Oh et al

2001; Rao et al 2008; Guerrero-Olazara´n et al 2010)

Ker-ovuo et al (1998) and KerKer-ovuo & Tynkkynen (2000)

reported that phy C isolated from B subtilis strain VTT

E-68013 or expressed by Lactobacillus plantarum 755 has amolecular mass of 43 kDa, which is similar to the 44-kDaphytase isolated from Bacillus sp DS11 (Kim et al 1998),and 43- and 40-kDa Bacillus sp phytase expressed in

E coliBL21 (Rao et al 2008) In the present study, a 42-kDaphy C of B subtilis E20 expressed in E coli HMS174 wasdetected by the SDS–PAGE analysis, and the enzyme activitywas detected by a zymogram analysis in SDS–PAGE

The phytase of Bacillus sp is a beta-propeller phytaserequiring calcium for activity and stability and exhibiting

an optimum pH at around 6~ 9 (Gulati et al 2007; Rao

et al 2008; Guerrero-Olazara´n et al 2010), which is able for animals’ neutral digestive tracts In addition, thestability of Bacillus phytase at high temperatures is anotherimportant characteristic for use in aquafeed processed athigh temperatures (Kim et al 1998; Oh et al 2004; Gulati

suit-et al 2007; Fu et al 2008; Rao et al 2008; zara´n et al 2010) Kim et al (1998) showed that phytasefrom Bacillus sp DS11 had an optimum temperature at

Guerrero-Ola-70°C and optimal pH of 7, and about 50% of its originalactivity remained after incubation at 90°C for 10 min inthe presence of 5 mMCaCl2 The partially purified phytasefrom Bacillus laevolacticus showed optimal activity at pH7.0 and 70°C, and ~80% of residual activity remained at

60°C after 3 h of incubation at pH 8.0 in the presence/

absence of CaCl2 Phytase was also stable at 80 °C, butthen declined sharply thereafter (Gulati et al 2007) In the

Table 3 Average weight (g±SE) of shrimp juveniles after being fed experimental diets containing soybean meal pretreated by different

levels of phytase during the first growth trial

Data (mean ± SE) with different letters are significant difference (p < 0.05) among different treatments at the same time.

Table 4 Growth parameters and survival of shrimp fed the experimental diets containing soybean meal pretreated by different levels of

phytase in the first growth trial

Data (mean ± SE) in the same row having different letters are significantly different (p < 0.05).

Table 5 Average weight (g±SE) of shrimp juveniles after being

fed experimental diets containing soybean meal pretreated by

dif-ferent levels of phytase during the second growth trial

Data (mean ± SE) with different letters are significant difference

(p < 0.05) among different treatments at the same time.

.

Aquaculture Nutrition, 19; 117–127 ª 2012 Blackwell Publishing Ltd

present study, the recombinant phy C had significantly

higher enzyme activities at the ranges of pH of 6 ~ 7 and

temperature of 50~ 60 °C The recombinant phy C was

thermostable below 50°C and showed 33% residual

activ-ity at 90 °C and 33.3% at 100 °C for 10 min It is thought

that the recombinant phy C may be suitable for use with

aquatic animals with neutral digestive tracts; this is not so

for Aspergillus phytase, the enzyme activity of which is

only found at a low pH range of 2~ 5 because of its acid

phosphatase activity (Cao et al 2007)

Phytase requires divalent metal ions such as Ca2+ for

catalytic activity and thermostability Analysis of the

crys-tal structure showed that a novel scaffolding architecture

of a six-bladed propeller for phosphatase activity and that

a triadic Ca2+ centre at the active site cleft provides a

favourable electrostatic environment for the metal phytate

complex (Ha et al 2000) Ca2+ acts as an essential

activa-tor for calcium phytate hydrolysis by properly binding

sub-strate to the enzyme’s active site (Oh et al 2001) In this

study, the recombinant phy C had Ca2+-dependent

cata-lytic properties, and no activity was detected while the

enzyme was in a buffer without CaCl2 or to which EDTA

had been added

Phytate is a phosphate source and an important

anti-nutritional factor in plant materials because of its low

bioavailability to monogastric or agastric aquatic animal

without phytase, and when combined with other nutrients,

it results in reduced utilization efficiency (Cosgrove 1996;

Liu et al 1998; Papatryphon et al 1999; Sugiura et al

2001; Cao et al 2007) Therefore, phytase supplementation

in aquafeed for aquatic animal to improve nutritional

utili-zation and growth performance was tried (Cao et al 2007)

Phytase that used to evaluate the growth performance of

animals is a commercial product for livestock animals such

as Aspergillus phytase It is more suitable for aquatic

ani-mal with acidic digestive tract such as carnivorous fish A

commercial phytase-incorporating diet containing only

plant protein or a combination of plant and animal protein

sources was able to enhance channel catfish Lctalurus

punctatus growth performance (Jackson et al 1996) Li &Robinson (1997) indicated that channel catfish fed dietscontaining 250 U kg 1of commercial microbial phytase ormore consumed more feed, gained more weight, and had alower feed conversion ratio compared to fish fed the basaldiet containing no commercial microbial phytase Yoo

et al (2005) reported that 1000 U kg 1 of phytase in thediet in Korean rockfish could gain better growth rate thanthe control However, the acidic phytase may be unsuitablefor shrimp that has a neutral digestive tract (Shen et al.2004), such as dietary commercial phytase (Natuphos®5000G; BASF India, Mumbai, India) supplementation had

no effect on the growth of tiger shrimp, Penaeus monodon(Biswas et al 2007) In this study, we developed a thermo-stable and neutral phytase from B subtilis E20 by expres-sion in a bacterial system and use it in shrimp diet toimprove shrimp growth performance and feed efficiency

As the results shown in the first growth trial, the rus supplemented diets containing recombinant phy C-pre-treated soybean meal of 1000 FTU may be enough toimprove shrimp growth performance and feed efficiency.However, 2000 FTU recombinant phy C-pretreated soy-bean meal in the diet lack of phosphorus may be necessaryfor the increase in growth performance and feed efficiency

phospho-It is thought that higher growth performance and feed ciency of shrimp may have resulted from the reduction inphytate chelating with other nutrition and phosphatecaused by recombinant phy C

effi-In conclusion, a thermostable and neutral phytase of

B subtilisE20 of 42 kDa was successfully expressed in the

E coli HMS174 system The recombinant phy C had anoptimal pH range of 6~ 7 and temperature range of

50~ 60 °C to hydrolyse phytate At 90 ~ 100 °C, therecombinant phy C had around 30% residual activity Toevaluate shrimp growth performance, shrimp fed a soybeanmeal (40%)-containing diet containing pretreated soybeanmeal showed improved growth performance and feedefficiency For a diet supplemented phosphorus, 1000 FTUrecombinant phy C-pretreated soybean meal is recommended

Table 6 Growth parameters and survival of shrimp fed the experimental diets containing soybean meal pretreated by different levels of phytase in the second growth trial

For a diet lack of phosphorus, 2000 FTU recombinant phy

C-pretreated soybean meal is needed

This research was supported by a grant from the National

Science Council (NSC 98-2221-E-020-013; NSC

98-2622-B-020-007-CC3), Taiwan

Allen, S.E (1989) Chemical Analysis of Ecological Materials, 2nd

edn Blackwell, Oxford.

AOAC (1995) Official Methods of Analysis of the Association of

Official Analytical Chemists International, 16th edn Association

of Official Analytical Chemists, Arlington, VA.

Bae, H.D., Yanke, L.J., Cheng, K.J & Selinger, L.B (1999) A

novel staining method for detecting phytase activity J

Micro-biol Methods, 39, 17–22.

Bendschneider, K & Robinson, R.J (1952) A new spectrometric

method for the determination of nitrite in the seawater J Mar.

Res., 11, 87–96.

Biswas, P., Pal, A.K., Sahu, N.P., Reddy, A.K., Prusty, A.K &

Misra, S (2007) Lysine and/or phytase supplementation in the

diet of Penaeus monodon (Fabricius) juveniles: effect on growth

body composition and lipid profile Aquaculture, 265, 253–260.

Bower, C.E & Bidwell, J.P (1978) Ionization of ammonia in

sea-water: effects of temperature, pH and salinity J Fish Res.

Board Can., 35, 1012–1016.

Cao, L., Wang, W., Yang, C., Yang, Y., Diana, J., Yakupitiyage,

A., Luo, Z & Li, D (2007) Application of microbial phytase in

fish feed Enzyme Microbial Technol., 40, 497–507.

Cheng, Z.J & Hardy, R.W (2003) Effects of extrusion and expelling

processing, and microbial phytase supplementation on apparent

digestibility coefficients of nutrients in full-fat soybeans for

rain-bow trout (Oncorhynchus mykiss) Aquaculture, 218, 501–514.

Cosgrove, D.J (1996) The chemistry and biochemistry of inositol

poly phosphates Rev Pure Appl Chem., 16, 209–224.

Eeckhout, W & De Paepe, M (1994) Total phosphorus, phytate

phosphorus and phytase activity in plant feedstuffs Anim Feed

Sci Technol., 47, 19–29.

Forster, I., Higgs, D.A., Dosanjh, B.S., Rowshandeli, M & Parr,

J (1999) Potential for dietary phytase to improve the nutritive

value of canola protein concentrate and decrease phosphorus

output in rainbow trout (Oncorhynchus mykiss) held in 11 °C

fresh water Aquaculture, 179, 109–125.

Fu, S., Sun, J., Qian, L & Li, Z (2008) Bacillus phytases: present

scenario and future perspectives Appl Biochem Biotechnol.,

151, 1–8.

Guerrero-Olazara´n, M., Rodrı´guez-Blanco, L., Carreon-Trevin˜, G.,

Gallegos-Lo´pez, J.A & Viader-Salvado´, J.M (2010) Expression

of a Bacillus phytase C gene in Pichia pastoris and properties of

the recombinant enzyme Appl Environ Microbiol., 76, 5601–

5608.

Gulati, H.K., Chadha, B.S & Saini, H.S (2007) Production and

characterization of thermostable alkaline phytase from Bacillus

laevolacticus isolated from rhizosphere soil J Ind Microbiol.

Biotechnol., 34, 91–98.

Ha, N.C., Oh, B.C., Shin, S., Kim, H.J., Oh, T.K., Kim, Y.O.,

Choi, K.Y & Oh, B.H (2000) Crystal structures of a novel

ther-mostable phytase in partially and fully calcium-loaded states.

Nat Struct Biol., 7, 147–153.

Hardy, R.W (1995) Current issues in salmonid nutrition In:

Nutrition and Utilization Technology in Aquaculture (Lim, C &

Sessa, D.J eds), pp 26 –35 AOAC Press, Champaign, IL.

Jackson, L., Li, M.H & Robinson, E.H (1996) Use of microbial phytase in channel catfish Lctalurus punctatus diets to improve utilization of phytate phosphorus J World Aquacult Soc., 27,

Apa-Appl Environ Microbiol., 64, 2079–2085.

Kim, Y.O., Kim, H.K., Bae, K.S., Yu, J.H & Oh, T.K (1998) Purification and properties of a thermostable phytase from Bacil- lus sp DS11 Enzyme Microbiol Technol., 22, 2–7.

Lall, S.P (1991) Digestibility, metabolism and excretion of dietary phosphorus in fish In: Nutritional Strategies and Aquaculture Waste (Cowey, C.B & Cho, C.Y eds), pp 21 –36 University of Guelph, Guelph, ON, Canada.

Li, M.H & Robinson, E.H (1997) Microbial phytase can replace inorganic phosphorus supplements in channel catfish Lctalurus punctatus diets J World Aquacult Soc., 28, 402–406.

Liebert, F & Portz, L (2005) Nutrient utilization of Nile tilapia Oreochromis niloticus fed plant based low phosphorus diets sup- plemented with graded levels of different sources of microbial phytase Aquaculture, 248, 111–119.

Liu, B.L., Rafing, A., Tzeng, Y.M & Rob, A (1998) The tion and characterization of phytase and beyond Enzyme Mic- rob Technol., 22, 415–424.

induc-Liu, C.H., Chiu, C.S., Ho, P.L & Wang, S.W (2009) ment in the growth performance of white shrimp, Litopenaeus vannamei, by a protease-producing probiotic, Bacillus subtilis E20, from natto J Appl Microbiol., 107, 1031–1041.

Improve-Liu, K.F., Chiu, C.H., Shiu, Y.L., Cheng, W & Improve-Liu, C.H (2010) Effects of the probiotic, Bacillus subtilis E20, on the survival, development, stress tolerance, and immune status of white shrimp, Litopenaeus vannamei larvae Fish Shellfish Immunol., 28,

837 –844.

Murphy, J & Riley, J.P (1962) A modified single solution method for the determination of phosphate in natural waters Anal.

Chim Acta, 27, 31–36.

National Research Council (NRC) (1993) Nutrient Requirements

of Fish p 144 National Academies Press, Washington, DC.

Ochoa-Solano, J.L & Olmos-Soto, J (2006) The functional property of Bacillus for shrimp feeds Food Microbiol., 23, 519–

525.

Oh, B.C., Chang, B.S., Park, K.H., Ha, N.C., Kim, H.K., Oh, B.

H & Oh, T.K (2001) Calcium-dependent catalytic activity of a novel phytase from Bacillus amyloliquefaciens DS11 Biochemis- try, 40, 9669–9676.

Oh, B.C., Choi, W.C., Park, S., Kim, Y.O & Oh, T.K (2004) Biochemical properties and substrate specificities of alkaline and histidine acid phytases Appl Microbiol Biotechnol., 63, 362–

372.

Papatryphon, E., Howell, R.A & Soares, J.H Jr (1999) Growth and mineral absorption by striped bass Morone saxatilis fed a plant feedstuff based diet supplemented with phytase J World Aquacult Soc., 30, 161–173.

.

Aquaculture Nutrition, 19; 117–127 ª 2012 Blackwell Publishing Ltd

Rao, D.E.C.S., Rao, K.V & Reddy, V.D (2008) Cloning and

expression of Bacillus phytase gene (phy) in Escherichia coli and

recovery of active enzyme from the inclusion bodies J Appl.

Microbiol., 105, 1128–1137.

Rengpipat, S., Rukpratanporn, S., Piyatiratitivorakul, S &

Mena-saveta, P (2000) Immunity enhancement in black tiger shrimp

(Penaeus monodon) by a probiont bacterium (Bacillus S11).

Aquaculture, 191, 271–288.

Sajjadi, M & Carter, C.G (2004) Dietary phytase supplementation

and the utilization of phosphorus by Atlantic salmon (Salmo

sa-lar L.) fed a canola-meal-based diet Aquaculture, 240, 417–431.

Shen, W.Y., Hu, H.G & Pan, Y.J (2004) Effects of temperature

and pH on activity of digestive enzymes in shrimp Penaeus

van-namei Oceanol Limnol Sinica, 35, 543–548.

Shimizu, M (1992) Purification and characterization of phytase

from Bacillus subtilis (natto) N-77 Biosci Biotechnol Biochem.,

56, 1266–1269.

Sugiura, S.H., Gabaudan, J., Dong, F.M & Hardy, R.W (2001) Dietary microbial phytase supplementation and the utilization of phosphorus, trace minerals and protein by rainbow trout On- corhynchus mykiss (Walbaum) fed soybean meal-based diets Aquacult Res., 32, 583–592.

Tacon, A.G.J., Dominy, W.G & Pruder, G.D (1998) Global trends and challenges in aquafeeds for marine shrimp Int Aqua- feed, 4, 28–35.

Tseng, D.Y., Ho, P.L., Huang, S.Y., Cheng, S.C., Shiu, Y.L., Chiu, C.S & Liu, C.H (2009) Enhancement of immunity and disease resistance in the white shrimp, Litopenaeus vannamei, by the probiotic, Bacillus subtilis E20 Fish Shellfish Immunol., 26,

339 –343.

Yoo, G.Y., Wang, X., Choi, S., Han, K., Kang, J.C & Bai, S.C (2005) Dietary microbial phytase increased the phosphorus digestibility in juvenile Korean rockfish Sebastes schlegeli fed diets containing soybean meal Aquaculture, 243, 315–322.

Selected (G8) and wild-type (W) genotypes of black tiger

ani-mal) were fed either of two diet types in a clear-water tank

trial to examine the effects of diet type and genetics on

growth and feed utilization parameters Animals were fed

twice daily at one of the five ration levels from starvation

to apparent satiety All uneaten feed was accounted for

and moults removed Starved animals were measured after

3 weeks; those fed were measured at both three and

6 weeks Diet type varied by protein content, raw material

choice and the presence [high-specification diet (HSD)] or

absence [low-specification diet (LSD)] of bioactive

sub-stances At the end of the study, faecal samples were also

collected to determine the digestible protein and energy

content of each diet by each genotype Whole animal

pro-tein and energy content were also assessed from samples

from the initial populations and those from each tank

Growth after 6 weeks of those animals fed to satiety

showed that the G8 animals fed the HSD diet had grown

other treatment Those G8 animals fed the LSD diet

slowest Using the data from the varying ration levels, we

were able to define that the growth gains of the G8 animals

were achieved not only by a greater appetite, but also

through lower maintenance energy costs (29 versus 57 kJ

(19.5% versus 11.6% when fed the HSD diet) Use of a

low-specification diet with the G8 and W shrimps limited

their growth and impaired their potential as demonstrated

by a curvilinear response of growth to intake By son, those shrimp fed the HSD diet had a relatively lineargrowth response to intake

compari-KEY WORDS: energetics, genotype, maintenance

Received 28 October 2011; accepted 11 January 2012 Correspondence: Brett Glencross, CSIRO Food Futures Flagship, GPO Box

2583, Brisbane, Qld 4001, Australia E-mail: Brett.Glencross@csiro.au

There has been a considerable amount of effort in past ades to improve the productivity of aquaculture speciesthrough genetic selection in breeding programmes (e.g

dec-Beattie 1984; Tave 1988; Bentsen et al 1998; Parsons 1998;

Gjedrem 2000; Nell et al 2000; Hulata 2001; Argue et al

2002; Quinton et al 2005, 2007; Kause et al 2006; Preston

black tiger shrimp (Penaeus monodon) has resulted in icant enhancements in the production of this species incommercial ponds (Preston & Coman 2009; Preston et al

signif-2009)

Genetic improvement programmes for high-value water fish species have often focused selections on traitssuch as growth and, in some cases, utilization efficiency inaddition to product qualities (i.e fat content or pigment)(Gjedrem 2000; Quinton et al 2007) However, for manyother aquaculture species, market weight has typically beenthe only growth-related trait defined in the breeding objec-tive of the programmes, and little discrimination has beenmade between component traits contributing to the growthimprovements (e.g Argue et al 2002; Quinton et al 2005)

cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold- cold-.

ª 2012 Blackwell Publishing Ltd

2013 19; 128–138 .doi: 10.1111/j.1365-2095.2012.00941.x

Aquaculture Nutrition

However, for many of the species that have undergone

genetic selection programmes, there appears to be a lack of

any analysis on why those selected animals actually

per-form better

A robust strategy to assess the traits that contribute to

the enhanced performance is to adopt a nutritional

energet-ics evaluation design (reviewed by Bureau et al 2002;

reviewed by Glencross et al 2007) Such a strategy allows

for not only the determination of intake dependent and

independent effects, but also the determination of the

utili-zation efficiencies of energy and key nutrients and also the

metabolic demands of the animals for maintaining vital

processes (Glencross 2008, 2009)

Based on anecdotal observations of the different

genera-tions of selected shrimp, it is hypothesized that the gains

achieved in growth by selected shrimp will be largely

attrib-utable to increases in feed intake (Preston & Coman 2009)

However, additional gains in protein and energy utilization

efficiencies and reduced maintenance demands for protein

and energy cannot be discounted either Therefore, this

study proposes to examine the basis for the gains observed

in a genetically selected genotype of shrimp relative to an

unselected wild-type genotype To do this, we adopted a

nutritional energetics strategy that allowed the evaluation

of variation in each of the parameters of feed intake,

utili-zation and maintenance demands In addition, there is also

value in determining whether the gains generated by genetic

selection can also affect the nutritional demands To do

this, two different diets were fed to the different genotypes

to allow some assessment of diet by genotype interaction

effects

The design of this experiment was based on a nutritional

bioenergetics strategy used to define discrete parameters of

nutrient and energy utilization and partitioning (Glencross

2008, 2009) This was done to allow the determination of

the particular effects of growth, feed intake, protein and

energy utilization efficiencies and also the demands for

pro-tein and energy for maintenance by the two different

shrimp genotypes In addition to this, two distinctly

differ-ent diets were formulated and prepared One diet was a

mimic of the commercial feeds typically used in the

(LSD)] The second diet was formulated to include several

and also had a higher protein content [high-specificationdiet: (HSD)] Both diets contained an inert marker (yttrium

respective digestible protein and energy contents (Table 1).Each diet was prepared by ensuring all ingredients were

upright planetary mixer (Hobart, Sydney, NSW, Australia)

Table 1 Formulations and composition of the experiment diets (all values are g kg
1as fed basis – unless otherwise specified)

Diet high-specification diet

Diet low-specification diet

Digestible energy (MJkg
1DM)

1 Ridley Aquafeeds, Narangba, Qld, Australia.

2 Manildra, Auburn, NSW, Australia.

3 CSIRO, Cleveland, Qld, Australia, PCT Patent AU 2008201886.

4 MP Bio, Aurora, OH, USA.

5 BEC Feed Solutions, Carole Park, Qld, Australia.

6 DSM, Wagga Wagga, NSW, Australia.

7 Rabar, Beaudesert, Qld, Australia; includes (IU kg
1or g kg
1

of premix): Vitamin A, 2.5 MIU; Vitamin D3, 1.25 MIU; Vitamin E,

100 g; Vitamin K3, 10 g; Vitamin B1, 25 g; Vitamin B2, 20 g; min B3, 100 g; Vitamin B5, 100; Vitamin B6, 30 g; Vitamin B9, 5; Vitamin B12, 0.05 g; Biotin, 1 g; Vitamin C, 250 g; Banox-E, 13 g.

Vita-8 Stanford Materials, Aliso Viejo, CA, USA.

DP : DE, digestible protein to digestible energy ratio; DM, dry matter.

Water was then added during the mixing to form dough,

which was subsequently screw-pressed (Dolly, La

Monferri-na, Castell’Alfero, Italy) through a 3 mm die and cut to

pellet lengths of about 6 mm The pellets were then

not being fed

selected genotype of black tiger shrimp (Preston et al

2009) were collected from three separate grow-out ponds

(Gold Coast Marine Aquaculture Pty Ltd, Woongoolba,

Qld, Australia) by cast-net and transferred to a holding

tank (10 000 L) where they were held pending allocation to

non-selected genotype (offspring of wild animals) of Black tiger

shrimp were also collected from three separate grow-out

ponds (Rossman’s Shrimp Farm Pty Ltd, Woongoolba,

Qld, Australia) by cast-net and transferred to a separate

holding tank (10 000 L) where they were held pending

allo-cation to trial tanks Shrimp from each farm were selected

to be about the same initial size

Five shrimp from each genotype were allocated to each

geno-type) The mean initial weights for each genotype were

wild-type genotype (W) To minimize the initial variance

among tanks, the shrimp were initially sorted in groups of

animals in 0.25 g increments/size classes A tight weight

range of animals from around the mode of those size-class

groups were then selected and randomly allocated across

the 72 tank array

Tanks of the shrimp were maintained with flow-through

Shrimp were individually weighed at day 21 and again at

day 42

Shrimp were fed one of five dietary rations Those

shrimp from the satietal treatments were manually fed the

diets twice daily to marginal excess and the amount of feed

remaining the following day scored and that score used to

adjust the next day’s ration up or down according to feed

intake score The initial allocation was close to 2.0 g of

feed per day for each tank Subsequently, the restricted

rations were arbitrarily allocated amounts 1.5, 1.0, 0.5 and

0.0 g per day per tank These restricted ration levels were

maintained at that allocation for the full experiment, while

the satietal ration was modulated daily as described before(Smith et al 2007)

At day 21 and 42, each shrimp within each tank wasindividually weighed and the mean weight of all shrimpwithin each tank determined to calculate the mean weight

per treatment) At day 21, three individuals from each ofthe starved treatments were collected, frozen and kept forbiochemical analysis At day 42, three individuals from all

of the other treatments were collected, frozen and kept forbiochemical analysis

A separate subexperiment was performed on shrimp afterday 42 A digestibility study on the two genotypes anddiets was undertaken whereby the faeces were collectedbased on the methods reported in the study of Smith et al

(2007) At the conclusion of the growth experiment, 12 mals from each genotype and each feed were kept and

24 tanks The temperature in the experimental system was

to provide gentle mixing and aeration without suspendingany faecal material Lighting intensity was kept as low aspossible on a 12 h light : dark cycle, and only red hand-held spotlights used to aid in siphoning and minimize dis-turbance of the animals The animals were fed a ration(approximately 0.5 g) every 6 h and allowed 30 min to con-sume the ration, before all uneaten food was siphoned towaste Three hours after feed was first offered, all faeceswill be siphoned into a labelled bucket and allowed to set-tle briefly before the faeces are transferred to a 10-mL cen-trifuge tube The seawater was then decanted and replacedwith deionized water and the volume made up to 10 mL

then decanted and the tube capped and frozen The frozenfaecal pellet was then transferred to a sample vial for pool-ing (same tank, but pooled through time) and subsequentsample preparation If any animal moulted in any tank, thefaeces and moult were discarded and the tank siphonedthoroughly The regime was maintained until sufficient drymatter of faeces (~1 g) had been collected (~1 week)

Differences in the ratios of the parameters of dry matter,protein or gross energy to yttrium, in the feed and faeces ineach treatment, were calculated to determine the apparentdigestibility (ADdiet) for each of the nutritional parameters .

Aquaculture Nutrition, 19; 128–138 ª 2012 Blackwell Publishing Ltd

examined in each diet based on the following formula

(Smith & Tabrett 2004):

ADdiet¼ ð1
Ydiet Parameterfaeces

Yfaeces Parameterdiet

Þ  100

where Ydietand Yfaecesrepresent the yttrium content of the

Parame-terfaecesrepresent the nutritional parameter of concern

(pro-tein or energy) content of the diet and faeces, respectively

Parameters for digestible protein and energy for each diet

are presented in Table 1 All calculations were carried out

at the tank/replicate level

Diet, shrimp and faecal samples were analysed for dry

mat-ter, ash, nitrogen, lipid and gross energy content Diet and

faecal samples were in addition analysed for yttrium

con-tent Shrimp samples were minced while frozen, then

refro-zen and freeze-dried A subsample of the original frorefro-zen

mince was analysed for its moisture content by gravimetric

matter of other samples was calculated by gravimetric

24 h Total yttrium concentrations were determined after

mixed acid digestion using inductively coupled plasma

atomic emission spectrophotometry (ICP-AES) Protein

levels were calculated from the determination of total

content was determined gravimetrically following loss of

mass after combustion of a sample in a muffle furnace at

bomb calorimetry All methods were consistent with those

recommended by AOAC (2005)

The utilization of protein (P) and energy (E) was

deter-mined based on the somatic gain in both P and E over the

respective periods of the experiment (21 days or 42 days),

against the respective consumption of digestible P and E

over that period Both gain and intake values were

calcu-lated based on a daily gain amount per unit body weight,

which was done over the 21-day period for the starved

prawns and the 42-day period for the fed prawns To

pro-vide some independence of size effects, modelling of the

protein and energy utilization data was carried out with

respect to known energy and protein body weight

, respectively (Dall1986; Bureau et al 2002)

error of the mean is presented All levels of significancewere determined using a Tukey’s test with critical ranges

tests depending on data set Effects of genotype and diet

energy and protein retention data from each genotype

gen-eralized linear models analysis function within Statisticawith genotype as the fixed effect and intake (digestibleenergy or digestible protein respectively) as the covariate

relationships was undertaken using the data analysistools and graphics elements of Microsoft Excel (Microsoft

(Statsoft, Tulsa, OK, USA)

There were significant effects of both diet and genotype inthis experiment (Tables 2a and 3) Weight gain of satietal

greater than that of the satietal fed wild-type genotype (diet

effect was significant with the use of either diet Under ditions of restricted ration, growth of the selected genotype

con-at corresponding similar rcon-ation levels when fed to the wildtype was also significantly greater, though at the lowerration levels the significance of this difference wanes A cleardiet effect was also seen with both genotypes Diet HSDsupported better growth than diet LSD in both genotypes.Ration level had a significant effect on growth with both

magnitude of the responses varied substantially betweengenotypes and diets There was significant variation inthe feed intake of the satiety fed shrimp (Tables 2a and 3).The selected shrimp fed diet HSD (37.5 g per shrimp) ate

the most amount of feed, which was significantly higher

than that of the selected shrimp fed diet LSD (34.0 g per

shrimp, the next highest amount of feed) Those wild-type

shrimp fed diet HSD or LSD ate about the same (24.3 g

per shrimp or 24.8 g per shrimp, respectively), both of

which were not significantly different from each other, but

both were lower than that obtained by the selected shrimp

There were also significant interaction effects observed in

were no significant effects though on survival at week 3 or

at week 6

Digestibility analysis of the experimental diets determined

that there was a significantly higher level of digestible

for diet LSD There was a smaller, though significant

differ-ent in the digestible energy in diet HSD (16.5± 0.17 kJ g
1)

genotype-based effects on digestibility of either protein or energy wereobserved

Protein utilization by shrimp varied with genotype, dietand even feed intake level (Table 2c,e) As each differenceprovides important information about the nature of the uti-lization of this nutrient, each parameter is assigned anequation to demonstrate each response The intake rangeswere examined on a linear basis based on being either low-range (four lowest ration levels/all restricted ration levels)

or high-range (three highest ration levels) in each case Theoverall effect was examined on a nonlinear basis as perGlencross (2008):

Table 2 MANOVA analysis (a) of diet 9 genotype effects for the satietal fed treatments and ANCOVA analysis (b –e)

of genotype effect on the energy and protein gain of the pair-fed (restric- tively fed) treatments within each diet

(a) 1

(b) E gain – high-specification diet (HSD)

W 9 HSD treatments

+

In each case, the coefficient of the linear regressions (low

and high range) can be used as an estimate of the marginal

efficiency (%) of protein utilization by each treatment

(Glencross 2009)

Energy utilization by shrimp also varied with genotype,

diet and even feed intake level (Table 2b,d) As each

differ-ence provides important information about the nature of

the utilization of energy, each parameter is assigned an

equation to demonstrate each response:

Maintenance demands for protein were estimated in eachcase by deriving the x-intercept value of the linear equationfrom the three lowest protein intake levels (including star-vation) for each genotype and diet treatment (Figs 1 and2) This provides an estimate of the amount of proteinrequired to be consumed for maintenance demands Main-tenance protein (Pm) demands were significantly affected

by genotype, but not by diet

Figure 1 Effect of genotype on partial efficiencies of protein tion within the high-specification diet (HSD) Maintenance protein intake of selected (G8) shrimps was 0.53 g kg
0.7day
1compared with that of wild-type (W) shrimps, which was 0.95 g kg
0.7day
1,

reten-as determined from linear regression of the three lowest feed intake levels for each genotype Using the four restricted feed intake lev- els, the utilization efficiency of the W shrimps was calculated as 19.3% and the G8 shrimps as 21.3%, which was significantly dif- ferent based on an ANCOVA analysis (Table 2d) Over the full data range, energy gain for the G8 shrimps was defined by the equation

y =
0.0036x 2

+ 0.2216x
0.1126, while the energy gain for the

W shrimps was defined by the equation y =
0.0139x 2 + 0.238x
0.1941 Each data point represents the mean of an indi- vidual tank of five shrimp.

.

Aquaculture Nutrition, 19; 128–138 ª 2012 Blackwell Publishing Ltd

W9 LSD: Pm = 0.78 g kg
0.7day
1

Maintenance demands for energy were also estimated in

each case by deriving the x-intercept value of the linear

equation from the three lowest energy intake levels

(includ-ing starvation) for each genotype and diet treatment

(Figs 3 and 4) This gives an estimate of the amount of

energy required to be consumed for maintenance demands

Maintenance energy (HEm) demands were significantly

affected by genotype, but not by diet

This experiment was designed to examine the gains in

growth achieved by selected G8 shrimp relative to a

wild-type genowild-type The design used a bioenergetics approach to

examine gains due to feed intake, feed utilization cies and maintenance demands by the animals In addition,the use of two different diets, one similar to current com-mercial diets (LSD) and the other a high-performance spec-ification (HSD), enabled the examination of diet bygenotype interactions and also the effects of diet on thepotential of a shrimp’s growth potential to be achieved

efficien-In this study, it was observed among the different ments that there were substantial effects of both genotypeand diet on the growth and feed utilization of the shrimp.Growth of the selected genotypes was significantly betterthan those of the wild type with either diet

treat-Although the magnitude of the genotype effect in thisexperiment was greater than that of the diet effect, thegains attributable to diet were also significant The use of aMANOVAanalysis also clearly identified an interaction effect

in this study between genotype and diet This supports thatdifferent responses are seen between the different genotypes

to each of the diets At a more facile level, it can be seenthat the growth potential of the selected genotype is better

Figure 2 Effect of genotype on partial efficiencies of protein

reten-tion within the low-specificareten-tion diet (LSD) Maintenance protein

intake of selected (G8) shrimps was 0.40 g kg
0.7day
1compared

with that of wild-type (W) shrimps, which was 0.78 g kg
0.7day
1,

as determined from linear regression of the three lowest feed intake

levels for each genotype Utilization efficiency of the W shrimps

was calculated as 4.8% and the G8 shrimps as 13.6% based on

lin-ear regression of the three highest feed intake levels Using the four

restricted feed intake levels, the utilization efficiency of the W

shrimps was calculated as 18.9% and the G8 shrimps as 21.1%,

which was significantly different based on an ANCOVA analysis

(Table 2e) Over the full data range, energy gain for the S shrimps

was defined by the equation y =
0.0114x 2

+ 0.243x
0.0869, while the energy gain for the W shrimps was defined by the equa-

realized when a superior diet is provided that allows that

genotype to express its potential

On examination of the satietal fed treatments, it can be

seen that those treatments growing fastest are also those

with the greatest feed intake Therefore, one of the

ele-ments contributing to the faster growth of the selected

genotype in this study is a higher feed intake However, it

can be noted that the efficiency of feed conversion (FCR)

is better at a subsatietal, or marginal intake level, and this

is consistent with observations of other such dietary

restric-tion effects on this species (Glencross et al 1999) When

the effects of marginal intake levels are examined, it also

becomes possible to determine that the selected genotypes

are also converting feed to weight gain more efficiently

than the wild-type genotype Specifically, when the

conver-sion of shrimp fed the 75% ration is compared (held

constant across both diets and genotypes), it can be seen

that the lowest FCR is obtained with the selected shrimp

fed diet HSD (2.32 : 1), whereas the same diet fed to the

wild-type shrimp produced an FCR of 2.89 : 1 An effect

of diet on FCR is also noted (selected: 2.32 cf 2.92; wild

type 2.89 cf 3.52) between diets HSD and LSD within each

When the marginal utilization efficiencies are examinedbetween genotypes, it can be seen that the selected genotypes

utilization near maintenance and about (subject to diet)

examined as the marginal utilization efficiency of protein,the genotype effects can largely be seen to be related tothose effects of protein utilization, which are about (subject

more efficient at high-protein intake levels These resultsdemonstrate that the selected shrimp are more efficient atusing dietary protein for protein synthesis, but that there is

a strong dietary effect in their ability to achieve this

The marginal protein utilization efficiency (ranges from5% to 21%) observed in the present study is consistentwith those results reported by Richard et al (2010), whoshowed a range of utilization values (ranges from 6% to38%) subject to protein intake level Notably, the utiliza-tion efficiency of this shrimp species (of either genotype) inthe present study is considerably lower than that observed

in most fish species (Dumas et al 2010) Typically rous fish, such as rainbow trout (Oncorhynchus mykiss) andbarramundi (Lates calcarifer), have a marginal protein uti-lization efficiency of close to 50% (Glencross 2008, 2009)

carnivo-Omnivorous fish, such as Pangasius catfish (Pangasianodonhypopthalamus), have a lower marginal utilization efficiency

at around 32% (Glencross et al 2011), though this is stillhigher than that seen in the present study with shrimp

Marginal energy utilization (ranges from 4% to 23%)efficiency by this shrimp species (of either genotype) is alsoconsiderably lower than that observed in fish species(Dumas et al 2010) Those carnivorous fish, such as rain-bow trout and barramundi, have a marginal energy utiliza-tion efficiency of close to 66% (Glencross 2008, 2009),while omnivorous fish, such as Pangasius catfish, wasreported to have a lower marginal utilization efficiency ofaround 50% (Glencross et al 2011)

Figure 4 Effect of genotype on partial efficiencies of energy

reten-tion within the low-specificareten-tion diet (LSD) Maintenance energy

intake of selected (G8) shrimps was 28.5 kJ kg
0.8day
1compared

with that of wild-type (W) shrimps, which was 56.7 kJ kg
0.8

day
1, as determined from linear regression of the three lowest

feed intake levels for each genotype Using the four restricted feed

intake levels, the utilization efficiency of the W shrimps was

calcu-lated as 15.7% and the G8 shrimps as 17.7%, which was

signifi-cantly different based on an ANCOVA analysis (Table 2c) Over the

full data range, energy gain for the G8 shrimps was defined by the

equation y =
0.0001x 2

+ 0.2049x
5.3022, while the energy gain for the W shrimps was defined by the equation y =
0.0004x 2 +

0.2421x
10.9 Each data point represents the mean of an

indi-vidual tank of five shrimp.

.

Aquaculture Nutrition, 19; 128–138 ª 2012 Blackwell Publishing Ltd

By calculation of the x-intercept of the regression line

fit-ted to the energy/protein intake and gain data, it becomes

possible to determine the amount of energy/protein required

by the animal to maintain its physiological processes

(Rodehutscord & Pfeffer 1999; Glencross 2008) In energetic

terms, this is often referred to as the heat energy of

mainte-nance (HEm) (Bureau et al 2002) That there are significant

differences between the two shrimp genotypes at the same

temperature and with different diets demonstrates that this

is a genetic effect and most likely one of the facets of why

the animals grow better, due to a lower demand for energy

to maintain physiological processes In comparison with

2008; Glencross et al 2011)

The effects seen on the HEm between the two shrimp

genotypes are largely driven by the maintenance demands

for protein (Pm) The Pm is also distinctly different between

the two genotypes, but not between the two diets The Pm

of the selected shrimp (0.40–0.53 g kg
0.7day
1) is similar

to that seen for fish species, which is often around

considerably higher than most fish species By comparison,

of Richard et al (2010) for the same species, which is less

than the wild-type shrimp from Australia, but higher than

that of the selected genotype These authors also examined

this maintenance demand on an untransformed basis and

reported that protein maintenance was 4.5 g kg
1day
1

The definition of these bioenergetic terms also provides

considerable scope for the development of factorial

bioen-ergetic models and subsequently iterative diet designs and

management strategies (Glencross 2008; Dumas et al 2010;

Glencross et al 2011) In contrast to fish models, shrimp

models will require additional levels of complexity to define

losses caused by the moulting process

The findings of this study demonstrate that the gains

achieved via genetic selection, when comparing a wild type

against a G8 selected genotype, are a trifecta of increased

feed intake, reduced maintenance demands for energy and

improved protein and energy utilization efficiencies It

would be of value to examine whether such gains are

consis-tent with each generation or whether the gains diminish withsuccessive generations as have been noted with other domes-ticated animals (Mulder & Bijma 2005) In addition to theclear effects attributable to genotype observed in this study,clear effects due to diet were also observed The results showthat feeding a lower specification diet significantly restricts ashrimp’s ability to express its full genetic potential Thesedata also lend themselves to the potential development ofnutritional energetics models, which have seen critical appli-cation in a range of fish species but are yet to make muchimpact in shrimp nutrition (Jackson & Wang 1998; Franco

et al.2006; Sara et al 2009; Richard et al 2010)

These findings also clearly show that selection in theabsence of concurrent nutritional research to underpinthose gains will not realize the full potential the animal canachieve By establishing the potential gains achievablethrough various research efforts, it becomes possible to bet-ter appropriate resources such that the greatest gains can

be achieved for the least cost

The provision of selected shrimp by Gold Coast MarineAquaculture Pty Ltd and the non-selected shrimp by Ross-man’s Shrimp Farm Pty Ltd is gratefully acknowledged.The chemical analyses work by Ken Dods of the ChemistryCentre, Western Australia is gratefully acknowledged Spe-cial thanks to Stuart Arnold for constructive comment andeditorial on an early draft Special thanks to Dr Nick Ellis

of CSIRO Mathematical and Information Sciences foradvice on the statistical analysis

AOAC (Association of Official Analytical Chemists) (2005) cial Methods of Analysis of the Association of Official Analyti- cal Chemists, 15th edn Association of Official Analytical Chemists, Washington, DC, USA.

Offi-Argue, B.J., Arce, S.M., Lotz, J.M & Moss, S.M (2002) Selective breeding of Pacific white shrimp (Litopenaeus vannamei) for growth and resistance to Taura syndrome virus Aquaculture,

Dall, W (1986) Estimation of routine metabolic rate in a penaeid prawn, Penaeus esculentus, Haswell J Exp Mar Biol Ecol., 96,

57 –74.

Dumas, A., France, J & Bureau, D.P (2010) Modelling growth

and body composition in fish nutrition: where have we been and

where are we going? Aquacult Res., 41, 161–181.

Franco, A.R., Ferreira, J.G & Nobre, A.M (2006) Development of

a growth model for penaeid shrimp Aquaculture, 259, 268–277.

Gjedrem, B (2000) Genetic improvement of cold-water fish species.

Aquacult Res., 31, 25–33.

Glencross, B.D (2008) A factorial growth and feed utilisation

model for barramundi, Lates calcarifer based on Australian

pro-duction conditions Aquacult Nutr., 14, 360–373.

Glencross, B.D (2009) Reduced water oxygen levels affect

maxi-mal feed intake, but not protein or energy utilisation by rainbow

trout (Oncorhynchus mykiss) Aquacult Nutr., 15, 1–8.

Glencross, B.D., Smith, D.M., Tonks, M.L., Tabrett, S.M &

Wil-liams, K.C (1999) A reference diet for nutritional studies of the

prawn, Penaeus monodon Aquacult Nutr., 5, 33–40.

Glencross, B.D., Booth, M & Allan, G.L (2007) A feed is only as

good as its ingredients – a review of ingredient evaluation for

aquaculture feeds Aquacult Nutr., 13, 17–34.

Glencross, B.D., Phuong, N.T., Hien, T.T.T., Cam Tu, T.T &

Phu, T.M (2011) A factorial growth and feed utilisation model

for Tra catfish, Pangasianodon hypopthalamus Aquacult Nutr.,

17, e396–e405.

Hulata, G (2001) Genetic manipulation in aquaculture: a review

of stock improvement by classical and modern technologies.

Genetica, 111, 155–173.

Jackson, C.J & Wang, Y.G (1998) Modelling growth rate of

Penaeus monodon Fabricius in intensively managed ponds: effects

of temperature, pond age and stocking density Aquacult Res.,

23, 27–36.

Kause, A., Tobin, D., Houlihan, D.F., Martin, S.A.M.,

Ma¨ntysa-ari, E.A., Ritola, O & Ruohonen, K (2006) Feed efficiency of

rainbow trout can be improved through selection: different

genetic potential on alternative diets J Anim Sci., 84, 807–

817.

Mulder, H.A & Bijma, P (2005) Effect of genotype x environment

interaction on genetic gain in breeding programs J Anim Sci.,

83, 49–61.

Nell, J.A., Smith, I.R & McPhee, C.C (2000) The Sydney rock

oyster Saccostrea glomerata (Gould 1850) breeding programme:

progress and goals Aquacult Res., 31, 45–49.

Parsons, J (1998) Status of genetic improvement in commercially reared stocks of rainbow trout World Aquacult., 29, 44–47.

Preston, N.P & Coman, G.J (2009) Black tiger shrimp tion advances – biosecure production, genetic markers assist development Global Aquacult Advocate, 5, 8–59.

domestica-Preston, N.P., Coman, G.J., Sellars, M.J., Cowley, J.A., Dixon, T.J., Li, Y & Murphy, B.S (2009) Advances in Penaeus mon- odon breeding and genetics In: The Rising Tide – Proceedings

of the Special Session on Sustainable Shrimp Farming (Browdy, C.L & Jory, D.E eds) pp 1 –11 World Aquaculture Society, Baton Rouge, LO, USA.

Quinton, C.D., McMillan, I & Glebe, B.D (2005) Development of

an Atlantic salmon (Salmo salar) genetic improvement program:

genetic parameters of harvest body weight and carcass quality traits estimate with animal models Aquaculture, 247, 211–217.

Quinton, C., Kause, A., Koskela, J & Ritola, O (2007) Breeding salmonids for feed efficiency in current fishmeal and future plant-based diet environments Genet Sel Evol., 39, 431–446.

Richard, L., Blanc, P.-P., Rigolet, V., Kaushik, S.J & Geurden, I.

(2010) Maintenance and growth requirements for nitrogen, lysine and methionine and their utilisation efficiencies in juvenile black tiger shrimp, Penaeus monodon, using a factorial approach Br.

J Nutr., 103, 984–995.

Rodehutscord, M & Pfeffer, E (1999) Maintenance requirement for digestible energy and efficiency of utilization of digestible energy for retention in rainbow trout, Oncorhynchus mykiss.

Aquaculture, 179, 95–107.

Sara, J.P., Gous, R.M & Bureau, D.P (2009) Describing growth and predicting feed intake in the marine prawn Fenneropenaeus indicus Part I: theoretical and practical aspects of measuring and predicting genetic parameters Aquaculture, 287, 402–413.

Smith, D.M & Tabrett, S.J (2004) Accurate measurement of in vivo digestibility in shrimp feeds Aquaculture, 232, 563–580.

Smith, D.M., Tabrett, S.J., Glencross, B.D., Irvin, S.J & Barclay, M.C (2007) Digestibility of lupin kernel meals in feeds for the black tiger shrimp, Penaeus monodon Aquaculture, 264, 353–362.

Tave, D (1988) Genetics and breeding of Tilapia: a review In:

Proceedings of the 2nd International Symposium on Tilapia in Aquaculture (Pullin, R.S.V., Bhukaswan, T.Tonguthai, K &

Maclean, J.L eds), pp 285 –294 16–20 March 1987 Bangkok, Thailand.

.

Aquaculture Nutrition, 19; 128–138 ª 2012 Blackwell Publishing Ltd

Apparent digestibility coefficients (ADCs) of dry matter,

crude protein, crude lipid, gross energy, phosphorus and

amino acids in Peruvian fish meal, poultry by-product

meal, meat and bone meal, spray-dried blood meal,

hydro-lysed feather meal, corn gluten meal, soybean meal, peanut

meal, cottonseed meal and rapeseed meal were determined

for juvenile snakehead (Ophiocephalus argus) with initial

mean body weight of 78.1 g A reference diet and test diets

that consisted of a 70 : 30 mixture of the reference diet to

test ingredient were used with 5 g kg 1 Cr2O3as an

exter-nal indicator Fish meal, poultry by-product meal and corn

gluten meal had higher ADCs of dry matter, crude protein,

and gross energy among ingredients tested Dry matter

ADCs ranged 61.9–81.5% for animal ingredients and corn

gluten meal and ranged 52.2–68.0% for soybean meal,

pea-nut meal, cottonseed meal and rapeseed meal Energy

ADCs of ingredients followed similar trends to differences

in dry matter digestibility Protein ADCs of animal and

plant ingredients ranged 73.6–92.8% and 75.3–85.6%,

respectively Amino acid ADCs generally reflected protein

digestibility Lipid ADCs were relatively high for the

dients tested Phosphorus ADCs of animal and plant

ingre-dients ranged 39.5–65.2% and 38.7–57.1%, respectively

Ophiocephalus argus, snakehead

Received 14 September 2011, accepted 6 February 2012

Correspondence: Q Zhang, Key Laboratory of Marine Biotechnology of

Guangxi, Guangxi Institute of Oceanology, Beihai, Guangxi 536000,

China E-mail: celery996@yahoo.com.cn

Snakehead, Ophiocephalus argus, an obligatory air-breather,

is a popular and valuable food fish widely cultured in China,and in most southern and south-eastern Asian countriesbecause of its good taste, fast growth, resistance to disease,handling and extreme tolerance to inferior water quality(Webster & Lim 2002; Hossain et al 2008) Until now,insufficient information is available regarding the nutri-tion of this species to make least-cost feed formulationpractical In culture practices, snakehead is strictly carniv-orous and is generally fed with diets of animal origin,especially trash fish In recent years, however, regionallyunavailable supply and increasing price of trash fish haveresulted in the development and use of formulated diets

as alternative food for sustainable expansion of head production

snake-Fishmeal has traditionally been the main dietary proteinsource used in feeds for carnivorous fish species includingsnakehead Because of its expensive cost and finite supply,replacement of fishmeal by less expensive animal and/orplant protein sources has been the important research area

in aquaculture nutrition (Hardy & Kissil 1997; Webster

et al.2000; Tacon & Metian 2008) Substantial efforts havebeen expended over the past decades in evaluating thepotential of cost-effective alternative protein sources, such

as animal products (Tidwell et al 2005; Ai et al 2006;Wang et al 2008; Saadiah et al 2011) and plant products(Luo et al 2006; Pham et al 2007; Wang et al 2011).Many of these ingredients require thorough evaluation todetermine their nutritional value and appropriate use levels

in prospective diets (Glencross et al 2007)

The digestibility coefficients of feed ingredients provideinsight concerning nutrient utilization and should enablemore accurate ingredient substitutions in diets designed for

.

Aquaculture Nutrition

target fish species (Lee 2002) However, no information is

available on digestion of major nutrients and energy from

various feed ingredients for juvenile snakehead Thus, this

study was conducted to determine apparent digestibility

coefficients (ADCs) for dry matter, crude protein, crude

lipid, phosphorus, energy and amino acids in Peruvian fish

meal, poultry by-product meal, meat and bone meal,

spray-dried blood meal, hydrolysed feather meal, corn gluten

meal, soybean meal, cottonseed meal, cottonseed meal and

rapeseed meal in diets for juvenile snakehead

ADC of the test ingredients was determined following the

method of Cho & Slinger (1979), the test diets consisted of

a 70 : 30 mixture of the reference diet (Table 1) to test

ingredient Chromic oxide (Cr2O3, 5 g kg 1) was added as

external indicator and was incorporated into the reference

and test diets The reference diet was formulated to satisfythe protein and lipid requirements of snakehead (Boonya-ratpalin 1980, 1981; Aliyu-Paiko et al 2010) All the testingredients for ADCs were obtained locally from commer-cial sources, which consisted of Peruvian fishmeal, poultryby-product meal, meat and bone meal, spray-dried bloodmeal, hydrolysed feather meal, corn gluten meal, soybeanmeal, peanut meal, cottonseed meal and rapeseed meal

Proximate and amino acid composition of the test ents and diets are shown in Tables 2 & 3, respectively

ingredi-Ingredients were ground into fine powder through a

198-lm mesh by a feed hammer grinder (Model 830; hou Huaxing Machine Co Ltd., Guangzhou, China) Allmajor dry ingredients were thoroughly mixed for 15 min in

Guangz-a drum mixer (Model SYTH-0.2; JiGuangz-angsu MuyGuangz-ang Group

Co Ltd., Yangzhou, China) The fish oil and soybean oilwere blended in a kitchen aid mixer, then added to themash and mixed for an additional 15 min Water(350 g kg 1 of the feed ingredients mixture) was added toproduce a stiff dough The dough was then passed through

a single screw extruder (Model Q65; Guangdong HuaqiangFeed Extruder Co Ltd., Luoding, China) to make slowsinking pellets with particle size of 3.59 5.5 mm Nosteam was used, and the pellet temperature at diet was ran-ged from 90 to 100°C The feed pellets were then dried in

a current of air at room temperature until the moisturelevel was less than 100 g kg 1and stored in plastic bags at

20 °C until fed The amount of diet needed weekly wasthen kept at room temperature (Bureau et al 1999)

Snakehead fish in this trial were obtained from a localrearing hatchery in Foshan, China Prior to the start ofthe experiment, the juvenile snakehead were reared infloating cages (1.59 1.5 9 2.0 m) in a farming pond(1209 106 9 2.8 m) and fed with a commerciallyextruded diet (contained 440 g kg 1 crude protein and

80 g kg 1 crude lipid) three times daily (7:30, 13:00,17:30) for 3 weeks to acclimate to the experimental condi-tions Afterwards, the fish were fasted for 24 h, and 30fish of each weighting 78.1± 2.52 g were randomly dis-tributed into each of the 33 cages (1.59 1.5 9 2.0 m)with three cages per experimental diet During the 8-weekexperimental period, the fish were hand-fed to apparentsatiation three times daily (7:30, 13:00, 17:30), except days

of storm Water temperature ranged from 26.0 to32.0°C, and dissolved oxygen ranged between 5.0 and8.0 mg L 1

Table 1 Reference diet formulation for the determination of

digestibility coefficients in juvenile snakehead (Ophiocephalus argus)

Ingredient

Concentration (g kg 1 dry matter)

2 Composition (IU or g kg 1 vitamin premix): retinal palmitate,

1 500 000 IU; cholecalciferol, 300 000 IU; DL- a-tocopherol

ace-tate, 20.0 g; menadione, 8.0 g; thiamin–HCl, 5.0 g; riboflavin,

5.0 g; D -calcium pantothenate, 16.0 g; pyridoxine–HCl, 4.0 g;

meso-inositol, 200.0 g; D -biotin, 8.0 g; folic acid, 1.5 g;

para-am-inobenzoic acid, 5.0 g; niacin, 20.0 g; cyanocobalamin, 0.01 g.

3 Composition (g kg 1 mineral premix): CoSO 4 4H 2 O, 0.30;

CuSO 4 5H 2 O, 10.0; FeSO 4 7H 2 O, 100.0; KCl, 100.0; KI, 0.2;

MgSO 4 2H 2 O, 203.4; MnSO 4 4H 2 O, 36.0; NaCl, 160.0; Na 2 SeO 3

H 2 O, 0.1; ZnSO 4 7H 2 O, 40.0.

.

Aquaculture Nutrition, 19; 139–147 ª 2012 Blackwell Publishing Ltd

Five hours (7:30 to 12:30) later after the feeding, samples

of faeces were collected from all fish in each cage by

apply-ing gentle pressure from the ventral fin to the anal region

as described by Austreng (1978) Prior to stripping, the fish

were anaesthetized with eugenol (1 : 10000) and gently

cleaned with a soft tissue Stripping of faeces was repeated

at 5-day interval until sufficient amounts of faeces were

obtained Pooled samples of faeces from each tank were

frozen at 20°C until analysis

Ingredient and feed samples were homogenized in an

all-purpose grinder Faeces were freeze-dried, homogenized

with a pestle and mortar Proximate composition analysis

was performed by standard methods of Association of cial Analytical Chemists (AOAC 1995) The samples ofingredients and diets were dried to a constant weight at

Offi-105°C to determine the dry matter content Crude proteinwas determined by measuring nitrogen (N9 6.25) usingthe Kjeldahl method, crude lipid by ether extraction usingthe Soxhlet method and ash by heating at 550°C for 24 h

in a muffle furnace The gross energy content was mined using an automatic Parr 1281 oxygen bombCalorimeter (Parr, Moline, IL, USA) Chromium andphosphorus in the feed and faeces samples were determined

deter-by inductively coupled plasma-atomic emission tometer (ICP-OES; Vista-MPX, Varian, Palo Alto, CA,USA) after perchloric acid digestion For amino acids com-position, samples were freeze-dried, and then hydrolysedwith 6 N HCl at 110°C for 24 h The chromatographicseparation and analysis of the amino acids were performed

spectropho-Table 2 Proximate and amino acid composition (g kg ) of the test ingredients

1 Purchased from Guangzhou Nongrun Feed Co., Ltd., Guangzhou, China.

2 Obtained from Weifang Meibaole Feed Co., Ltd., Weifang, China.

3 Obtained from Foshan Lanke Feed Co., Ltd., Foshan, China.

4 Carbohydrate was calculated by difference.

5

Tryptophan could not be analysed because of its destruction during acid hydrolysis.

after orthophthaldehyde derivation using reverse-phase

high-performance liquid chromatography (HPLC, HP1100;

GenTech Scientific, Inc., Arcade, NY, USA) followed the

modified procedure of Gardner & Miller (1980)

The following calculations were performed:

The ADCs of the experimental diets were calculated

according to Maynard & Loosli (1969):

ADC of dry matter (%) = (100 (% chromium in

feed/% chromium in faeces)9 100)

ADC of nutrient or energy (%) = (100 (%

chro-mium in feed/% chrochro-mium in faeces)9 (% nutrient or

energy in faeces/% nutrient or energy in feed)9 100)

The ADC in the test ingredient was calculated according to

Cho et al (1982):

ADC of test ingredient (%) = 100/309 (ADC in test

diet 0.7 ADC in reference diet)

All results are given as mean ± SD All percentage datawere arcsine-transformed prior to analysis The data weresubjected to one-way analysis of variance (ANOVA) followedusing Duncan’s multiple range test using the software pro-gram SPSS 13.0 for Windows (SPSS Inc., Chicago, IL,USA) A significance level of 5% was used for all compari-sons

ADCs of dry matter in the test ingredients ranged from52.2% to 81.5% (Table 4) Peruvian fishmeal, poultry by-product meal, spray-dried blood meal and corn gluten mealdid not differ in dry matter ADCs (P> 0.05), and all fourwere significantly higher (P< 0.05) than meat and bonemeal, hydrolysed feather meal, soybean meal, peanut meal,cottonseed meal and rapeseed meal Fish meal and corngluten meal exhibited the highest dry matter ADCs in ani-mal and plant ingredients, while meat and bone meal andrapeseed meal were the lowest, respectively

Table 3 Proximate and amino acid composition (g kg ) of the reference and test diets

Test diets (a 70 : 30 mixture of the reference diet to test ingredient)

PFM, Peruvian fishmeal; PBPM, poultry by-product meal; MBM, meat and bone meal; SDBM, spray-dried blood meal; HFM, hydrolysed

feather meal; CGM, corn gluten meal; SBM, soybean meal; PNM, peanut meal; CSM, cottonseed meal; RSM, rapeseed meal.

1 Tryptophan could not be analysed because of its destruction during acid hydrolysis.

.

Aquaculture Nutrition, 19; 139–147 ª 2012 Blackwell Publishing Ltd

ADCs of protein and lipid were relatively high for all the

test ingredients Protein digestibilities of the animal and

plant ingredients tested were ranged 73.6–92.8% and 75.3–

85.6%, respectively Peruvian fishmeal, poultry by-product

meal, corn gluten meal, soybean meal and peanut meal did

not differ in protein ADCs (P> 0.05) ADCs of protein

were significantly higher (P< 0.05) in Peruvian fishmeal

(92.8%) and poultry by-product meal (87.2%) than those

in meat and bone meal (80.1%), hydrolysed feather meal

(73.6%), cottonseed meal (78.1%) and rapeseed meal

(75.3%) Lipid ADCs for poultry by-product meal (77.2%)

and meat and bone meal (75.5%) were significantly lower

(P < 0.05) than those for other ingredients (87.1–93.9%)

ADCs of phosphorus were highly variable among the

test ingredients Peruvian fishmeal had significantly higher

(P < 0.05) phosphorus digestibility than meat and bone

meal, hydrolysed feather meal, soybean meal, peanut meal,

cottonseed meal and rapeseed meal The phosphorus

digestibility of corn gluten meal (57.1%) was inferior to

fish meal (65.2%) and spray-dried blood meal (60.1%),

showed no significant difference (P> 0.05) from the animal

ingredients with the exception of meat and bone meal

(39.5%) The phosphorus digestibility of rapeseed meal

(38.7%) was the lowest among the ingredients tested

ADCs of energy in the test ingredients ranged from

57.2% to 86.3% There was no significant difference

(P > 0.05) among ADCs of energy in fish meal, poultry

by-product meal, spray-dried blood meal and corn gluten

meal, which were significantly higher (P< 0.05) than those

of the other ingredients tested Meat and bone meal,

hydrolysed feather meal, soybean meal and peanut meal

had the intermediate ADCs of energy, while the rapeseed

was the lowest

Amino acid ADC generally reflected protein digestibility

Peruvian fishmeal, poultry by-product meal, corn gluten

meal, soybean meal and peanut meal had higher amino

acid ADCs than other ingredients tested The isoleucine

availability of spray-dried blood meal (67.6%) was the

low-est among the ingredients tlow-ested

The protein quality of dietary ingredients is usually the

leading factor affecting fish performance, and protein

digestibility is the first measure of its availability by fish

(Ko¨pru¨cu¨ & O¨zdemir 2005) In the present study, fish meal

protein was well digested by juvenile snakehead, which is

similar to those reported for most carnivorous fish (Willson

& Poe 1985; Sugiura et al 1998; Lee 2002; Zhou et al

2004; Tibbetts et al 2006; Li et al 2007; Luo et al 2009)

Moreover, poultry by-product meal, corn gluten meal, bean meal and peanut meal had relatively high proteinADCs in this study, indicating that these ingredients can beused efficiently as partial protein sources in snakehead diet

soy-Protein ADC of meat and bone meal in snakehead(80.05%) was slightly lower than those in rainbow trout(Oncorhynchus mykiss) at 83–89% (Bureau et al 1999)

Studies with various fish species showed a wide range ofprotein ADC for spray-dried blood meal, which was higher

in snakehead at 77.4% than that in seabass (Lateolabraxjaponicas) at 62.9% (Han et al 2011), but lower than those

in hybrid striped bass (Morone saxatilis♀ 9 Morone ops ♂), rainbow trout and rockfish (Sebastes schlegeli) at86–87% (Sullivan & Reigh 1995; Bureau et al 1999; Lee2002) This could be due to differences in fish species, mealprocessing condition and meal quality Laining et al (2003)found that for humpback grouper (Cromileptes altivelis),the protein ADC of blood meal increased significantly from55.2% up to 84.2% and 87.5% after propionic and formicacids processing, respectively Similar to those reported forrockfish (Lee 2002) and seabass at 63–79% (Han et al

chrys-2011), protein ADC of hydrolysed feather meal was thelowest for snakehead at 73.6%, which is likely due to indi-gestible keratin protein component in the ingredient Whenselected these animal products as feed ingredients, specialattention should be paid to the origins and processing con-ditions The protein ADCs of cottonseed meal and rape-seed meal were relatively low for snakehead in this study,indicating that these two meals could not be used at highlevels in snakehead feeds

In fish as in other animals, protein quality of dietaryprotein sources depends on the amino acid profile and theirdigestibility and availability (Rollin et al 2003) Deficiency

of an essential amino acid leads to poor utilization of thedietary protein and consequently reduces growth anddecreases feed efficiency (Lee 2002) In this study, aminoacid availability generally followed similar trend to proteindigestion Notably, blood meal had the lowest isoleucineADC among the ingredients tested Similar findings havebeen reported for rockfish (Lee 2002) and hybrid stripedbass (Gaylord et al 2004) Knowledge of individual aminoacid availabilities obtained from this study should allow formore accurate and economical snakehead feed formulation

on an ideal protein basis

Carnivorous fish tend to utilize the dry matter andenergy in animal products better than that in plantproducts (Cho et al 1982; Sullivan & Reigh 1995) Inthe present study, dry matter and energy ADCs were

.

Aquaculture Nutrition, 19; 139–147 ª 2012 Blackwell Publishing Ltd

lower in plant ingredients than those in animal

ingredi-ents except for corn gluten meal Similar trends were

observed with red drum (Sciaenops ocellatus; Gaylord &

Gatlin 1996; McGoogan & Reigh 1996), rockfish (Lee

2002) and cobia (Rachycentron canadum; Zhou et al

2004) The dry matter and energy digestibility appear to

be related to the quantity and chemical composition of

the ingredients Dry matter ADC of meat and bone meal

was markedly lower than that of protein and energy as a

result of the large amount of poorly digested ash

con-tained Dry matter and energy ADCs of snakehead were

negatively correlated with carbohydrate content of

ingre-dients tested (r= 0.56 and r = 0.69, respectively)

Corn gluten meal had intermediate level of carbohydrate

and was efficiently digested by the fish, presumed that a

proper ratio of protein to carbohydrate should be

selected to maximize the availability of carbohydrate

energy in snakehead diets

Dietary lipids are a major provider of energy in all fish,

especially carnivorous species (Sargent et al 2002) Lipid

when administered either alone or in a mixed diet routinely

gives digestibility values range from 85% to 95% for fish

(Cho & Slinger 1979; Aksnes & Opstvedt 1998) In this

study, the lipid was utilized efficiently by the snakehead as

indicated by the high lipid ADCs of the test ingredients

with the exception of poultry by-product meal and meat

and bone meal This was partly explained by the

peroxida-tion of lipid during high-temperatured processing and

sub-sequent storage Moreover, the high saturated fatty acid

content of the lipids of animal protein ingredients could

result in low lipid digestibility values (Cho & Kaushik

1990; Ai et al 2006; Martins et al 2009) By comparison,

spray-dried blood meal and hydrolysed feather meal had

higher lipid ADCs, which may be correlated with the low

lipid contents in the meals

The phosphorus ADC of fish meal for snakehead

(65.2%) in this study was higher than value for large

yel-low croaker (Pseudosciaena crocea) at 53.2% (Li et al

2007) while it was similar to those reported for cobia at

71.2% (Zhou et al 2004) and goby (Synechogobius hasta)

at 64.0% (Luo et al 2009) Although meat and bone meal

had the highest total phosphorus, its phosphorus ADC

(39.5%) was the lowest among all animal ingredients

Inversely, higher ADCs were observed for corn gluten meal

with lower phosphorus content The lower ADC values of

phosphorus from rapeseed meal, compared with those of

the other plant meals, may be due to its higher

glucosino-late content Similar trends were observed for rainbow

trout, turbot (Psetta maxima), cobia and goby (Vielma &

Lall 1997; Burel et al 2000; Zhou et al 2004; Luo et al.2009)

In conclusion, this study has shown that for snakehead,the highest protein and energy ADCs were observed inPeruvian fish meal, poultry by-product meal and corn glu-ten meal among the ingredients tested Of the ingredientstested, poultry by-product meal and meat and bone mealhad the lowest lipid ADCs, presumed correlation with theperoxidation of lipid during high-temperatured processingand high saturated fatty acid content Digestibility ofenergy tends to be negatively related to the carbohydratecontent, suggested that carbohydrate-riched oilseed mealssuch as cottonseed meal and rapeseed meal are not promis-ing substitutes for fish meal in snakehead diets The digest-ibility information may be helpful in the formulation ofnutrition-balanced and cost-effective diets for snakehead.Additional research is needed to evaluate the effects ofreplacing fish meal by single or mixed animal and plantprotein ingredients

The study was supported by the Key Laboratory Program

of Marine Biotechnology of Guangxi (Project 201103) and the Xinghuo Program of Shandong Province(Project 2010XH0632) The authors thank Yan-Song Zhengand Pin-Yi Luo for their help in purchasing juvenile snake-head Thanks are also due to Tao Hu, Jun-Fei Yan, ShuaiChen, Yuan-Yuan Jia, Xiao-Ting Liu, Cai-Hong Bi andShu-Juan Wang for their assistance in this study

GLMBT-Ai, Q.H., Mai, K.S., Tan, B.P., Xu, W., Duan, Q.Y., Ma, H.M & Zhang, L (2006) Replacement of fish meal by meat and bone meal in diets for large yellow croaker, Pseudosciaena crocea Aquaculture, 260, 255–263.

Aksnes, A & Opstvedt, J (1998) Content of digestible energy in fish feed ingredients determined by the ingredient-substitution method Aquaculture, 161, 45–53.

Aliyu-Paiko, M., Hashim, R & Shu-Chien, A.C (2010) Influence

of dietary lipid/protein ratio on survival, growth, body indices and digestive lipase activity in Snakehead (Channa striatus, Bloch 1793) fry reared in re-circulating water system Aquacult Nutr., 16, 466–474.

Association of Official Analytical Chemists (AOAC) (1995) Official methods of analysis, 16th edn AOAC, Arlington, VA.

Austreng, E (1978) Digestibility determination in fish using mic oxide marking and analysis of contents from different seg- ments of the gastrointestinal tract Aquaculture, 13, 265–272 Boonyaratpalin, M (1980) Protein requirement of Ophicephalus striatus In: National Inland Fisheries Institute Annual Report,

chro-pp 37 –38 Department of Fisheries, Bangkok, Thailand (in Thai).

Boonyaratpalin, M (1981) Lipid requirement of snakehead

finger-lings Progress Report of the Regional Project RAS/76/003, pp.

30 Network of Aquaculture Centers in Asia, Bangkok, Thailand.

Bureau, D.P., Harris, A.M & Cho, C.Y (1999) Apparent

digest-ibility of rendered animal protein ingredients for rainbow trout

(Oncorhynchus mykiss) Aquaculture, 180, 345–358.

Burel, C., Boujard, T., Tulli, F & Kaushik, S.J (2000)

Digestibil-ity of extruded peas, extruded lupin, and rapeseed meal in

rain-bow trout (Oncorhynchus mykiss) and turbot (Psetta maxima).

Aquaculture, 188, 285–298.

Cho, C.Y & Kaushik, S.J (1990) Nutritional energetics in fish:

energy and protein utilization in rainbow trout (Salmo gairdneri).

In: Aspects of Food Production, Consumption and Energy

Val-ues (Bourne, G.H ed) World Review of Nutrition and Dietetics,

61, 132–172.

Cho, C.Y & Slinger, S.J (1979) Apparent digestibility

measure-ment in feedstuff for rainbow trout In: Finfish Nutrition and

Fish feed Technology, Vol II (Halver, J.E & Tiews, K eds),

pp 239 –248 Heeneman, Berlin, Germany.

Cho, C.Y., Slinger, S.J & Bayley, H.S (1982) Bioenergetics of

sal-monid fishes: energy intake, expenditure and productivity Comp.

Biochem Phys., 73(B), 25–41.

Gardner, W.S & Miller, W.H (1980) Reverse-phase liquid

chro-matography of amino acids after reaction with Ophthalaadehde.

Anal Biochem., 101, 61–70.

Gaylord, T.G & Gatlin, D.M III (1996) Determination of

digest-ibility coefficients of various feedstuffs for red drum (Sciaenops

ocellatus) Aquaculture, 139, 303–314.

Gaylord, T.G., Rawles, S.D & Gatlin, D.M III (2004) Amino

acid availability from animal, blended, and plant feedstuffs for

hybrid striped bass (Morone chrysops 9 M saxatilis) Aquacult.

Nutr., 10, 345–352.

Glencross, B.D., Booth, M & Allan, G.L (2007) A feed is only as

good as its ingredients – a review of ingredient evaluation

strate-gies for aquaculture feeds Aquacult Nutr., 13, 17–34.

Han, Q.W., Liang, M.Q., Yao, H.B., Chang, Q & Wu, L.X.

(2011) Effects of seven feed ingredients on growth

perfor-mance, and liver and intestine histology of Lateolabrax

japoni-cas Prog Fishery Sci., 32, 32–39 (In Chinese with English

abstract).

Hardy, R.W & Kissil, G.W.M (1997) Trends in aquaculture

feed-ing Feed Mix, 5, 31–34.

Hossain, M.K., Latifa, G.A & Rahman, M.M (2008)

Observa-tions on induced breeding of Snakehead murrel, Channa striatus

(Bloch, 1793) Int J Sust Crop Prod., 3, 65–68.

Ko¨pru¨cu¨, K & O¨zdemir, Y (2005) Apparent digestibility of

selected feed ingredients for Nile tilapia (Oreochromis niloticus).

Aquaculture, 250, 308–316.

Laining, A., Rachmansyah, Ahmad, T & Williams, K (2003)

Apparent digestibility of selected feed ingredients for humpback

grouper, Cromileptes altivelis Aquaculture, 218, 529–538.

Lee, S.M (2002) Apparent digestibility coefficients of various feed

ingredients for juvenile and grower rockfish (Sebastes schlegeli).

Aquaculture, 207, 79–95.

Li, H.T., Mai, K.S., Ai, Q.H., Zhang, L., Zhang, C.X., Zhang, W.

B & Liufu, Z.G (2007) Apparent digestibility of selected protein

ingredients for large yellow croaker Pseudosciaena crocea Acta

Hydrobiol Sin., 35, 105–114 (In Chinese with English abstract).

Luo, L., Xue, M., Wu, X., Cai, X., Cao, H & Liang, Y (2006)

Partial or total replacement of fishmeal by solvent-extracted

cot-tonseed meal in diets for juvenile rainbow trout (Oncorhynchus

mykiss) Aquacult Nutr., 12, 418–424.

Luo, Z., Li, X.D., Gong, S.Y & Xi, W.Q (2009) Apparent ibility coefficients of four feed ingredients for Synechogobius has-

digest-ta Aquacult Res., 40, 558–565.

Martins, D.A., Valente, L.M.P & Lall, S.P (2009) Apparent digestibility of lipid and fatty acids in fish oil, poultry fat and vegetable oil diets by Atlantic halibut, Hippoglossus hippoglossus

L Aquaculture, 294, 132–137.

Maynard, L.A & Loosli, J.K (1969) Animal Nutrition, 6th edn,

pp 613 McGraw-Hill Book Company, London.

McGoogan, B.B & Reigh, R.C (1996) Apparent digestibility of selected ingredients in red drum (Sciaenops ocellatus) diets.

Aquaculture, 141, 233–244.

Pham, M.A., Lee, K.J., Lim, S.J & Park, K.H (2007) Evaluation

of cottonseed and soybean meal as partial replacement for meal in diets for juvenile Japanese flounder Paralichthys olivac- eus Fish Sci., 73, 760–769.

fish-Rollin, X., Mambrini, M., Abboudi, T., Larondelle, Y & Kaushik, S.J (2003) The optimum dietary indispensable amino acid pattern for growing Atlantic salmon (Salmo salar L.) fry Br.

J Nutr., 90, 865–876.

Saadiah, I., Abol-Munafi, A.M & Che Utama, C.M (2011) Replacement of fishmeal in cobia (Rachycentron canadum) diets using poultry by-product meal Aquacult Int., 19, 637–648.

Sargent, J., Tocher, D.R & Bell, J.G (2002) The lipids In: Fish Nutrition (Halver, J.E & Hardy, R.W eds), 3rd edn, pp 181 –

258 Academic Press, San Diego, California, USA.

Sugiura, S.H., Dong, F.M., Rathbone, C.K & Hardy, R.W.

(1998) Apparent protein digestibility and mineral availabilities in various feed ingredients for salmonids feeds Aquaculture, 159,

177 –202.

Sullivan, J.A & Reigh, R.C (1995) Apparent digestibility of selected feedstuffs in diets for hybrid striped bass (Morone saxa- tilis ♀ 9 Morone chrysops ♂) Aquaculture, 138, 313–322.

Tacon, A.G.J & Metian, M (2008) Global overview on the use of fish meal and fish oil in industrially compounded aquafeeds:

trends and future prospects Aquaculture, 285, 146–158.

Tibbetts, S.M., Milley, J.E & Lall, S.P (2006) Apparent protein and energy digestibility of common and alternative feed ingredi- ents by Atlantic cod, Gadus morhua (Linnaeus, 1758) Aquacul- ture, 261, 1314–1327.

Tidwell, J.H., Coyle, S.D., Bright, L.A & Yasharian, D (2005) Evaluation of plant and animal source proteins for replacement

of fish meal in practical Diets for the largemouth bass rus salmoides J World Aquacult Soc., 36, 454–463.

Micropte-Vielma, J & Lall, S.P (1997) Dietary formic acid enhances ent digestibility of minerals in rainbow trout Oncorhynchus my- kiss (Walbaum) Aquacult Nutr., 3, 265–268.

appar-Wang, Y., Li, K., Han, H., Zheng, Z.X & Bureau, D.P (2008) Potential of using a blend of rendered animal protein ingredients

to replace fish meal in practical diets for malabar grouper nephelus malabricus) Aquaculture, 281, 113–117.

(Epi-Wang, Y.Y., Yang, Y.H., (Epi-Wang, F., (Epi-Wang, L., (Epi-Wang, R.X & Lu,

Y (2011) Effects of replacing fish meal with soybean protein late on growth performance and nitrogen and phosphorus excre- tion of rainbow trout (Oncorhynchus mykiss) Acta Hydrobiol.

iso-Sin., 35, 105–114.

Webster, C.D & Lim, C (eds) (2002) Nutrient Requirements and Feeding of Finfish for Aquaculture, pp 397 CABI Publishing, Cambridge, MA, USA.

Webster, C.D., Thompson, K.R., Morgan, A.M., Grisby, E.J &

Gannam, A.L (2000) Use of hempseed meal, poultry by-product meal, and canola meal in practical diets without fish meal for

.

Aquaculture Nutrition, 19; 139–147 ª 2012 Blackwell Publishing Ltd

sunshine bass (Mornone chrysops 9 M saxatilis) Aquaculture,

188, 299–309.

Willson, R.P & Poe, W.E (1985) Apparent digestibility of protein

and energy in feed ingredients for channel catfish Prog

Fish-Cult., 47, 154–158.

Zhou, Q.C., Tan, B.P., Mai, K.S & Liu, Y.J (2004) Apparent digestibility of selected feed ingredients for juvenile cobia (Rachycentron canadum) Aquaculture, 241, 441–451.

School of Animal Production Technology, Institute of Agricultural Technology, Suranaree University of Technology, Muang,

Thailand

This study investigated the effects of the

co-supplementa-tion of vitamins C (0, 500, and 1000 mg kg 1) and E (0,

62.5, and 125 mg kg 1) on the growth performance,

hae-matology and the modulation of blood stress indicators

and immune parameters in hybrid catfish (Clarias

macro-cephalus 9 Clarias gariepinus) under combinations of

ther-mal and acidic stress Supplementation of vitamins C and

E influenced the growth, haematological indices, serum

chloride, plasma protein and immune parameters

(lyso-zyme, total immunoglobulin and alternative complement

haemolytic assay) (P< 0.05) Although vitamins C and E

did not prevent a significant reduction in serum chloride,

they minimized not only the modulation of blood glucose

and plasma protein, but also the reduction in immune

parameters (P< 0.05) owing to stress Our results

demon-strated that co-supplementation of 500 mg kg 1vitamin C

and 125 mg kg 1 vitamin E, or 1000 mg kg 1 vitamin C

alone, for four weeks and co-supplementation of both

vita-mins at low levels (vitavita-mins C at 500 mg kg 1and E at 62.5

mg kg 1) for eight weeks had beneficial effects on the

growth, amelioration of stress-mediated adverse changes in

the physiological and immunosuppressive responses of

hybrid catfish under stressful conditions

haematol-ogy, hybrid catfish, immune system, thermal stress, vitamin

C, vitamin E

Received 21 July 2011, accepted 6 February 2012

Correspondence: S Boonanuntanasarn, School of Animal Production

Technology, Institute of Agricultural Technology, Suranaree University

of Technology, 111 University Avenue, Muang, Nakhon Ratchasima

30000, Thailand E-mail: surinton@sut.ac.th

Fluctuations in water quality are a common daily and sonal occurrence in fish culture systems The water qualityparameters that vary daily include temperature, pH and dis-solved oxygen The seasonal climate affects the degree offluctuation Moreover, climate change has recently affectedthe changes in the water quality (Subehi & Fakhrudin 2011;

sea-Wagner et al 2011) Fish are ectothermic animals whosephysiological processes of growth and reproduction depend

on the water temperature Therefore, fish are intolerant ofsudden changes in water temperature (Cataldi et al 1998;

Das et al 2005; Whiteley et al 2006) Acidification, ing anthropogenic and/or episodic acidic conditions, hashad detrimental effects not only on freshwater ecosystemsbut also on fish cultures Acidic water has an impact on thereproductive system of fish, affecting gametogenesis, embry-ological development and the hatching process Therefore,anthropogenic acid conditions lead to the reduction of bio-diversity in freshwater ecology (Geffen 1990; Punzo &

includ-Thompson 1990; Zelennikov 1997; Felten et al 2006) sodic acid stress conditions that result from heavy rain orsnow melts have been reported to impair homeostasis mech-anisms that involve water and the ions in body fluids (Heis-ler 1989) and, consequently, cause acid stress in fish

Epi-Water quality fluctuations are important stresses thatcause a series of physiological responses The primaryresponses to stress are mediated by the neuroendocrine sys-tem through the release of stress hormones In turn, thestress hormones cause changes in the physiological pro-cesses that drive the secondary responses Finally, thechanges in physiological processes affect growth and immu-nocompetence as well as the reproductive system(Wedemeyer 1996) Hybrid catfish (Clarias macrocepha-lus9 Clarias gariepinus) have been cultured intensively intropical Asia, where the weather is ruled by monsoons,

.

ª 2012 Blackwell Publishing Ltddoi: 10.1111/j.1365-2095.2012.00950.x .2013 19; 148–162

Aquaculture Nutrition

which cause rapid fluctuations in the level of dissolved

oxy-gen, temperature and pH A hybrid catfish has an

air-breathing organ (the dendrite), which allows it to live in

turbid and very poorly oxygenated water The water

tem-perature at which hybrid catfish is cultured is

approxi-mately 27–32 °C Like other tropical fish species, the

hybrid catfish is thermosensitive to low temperatures and

any rapid changes in temperature (Fedoruk 1981) The

water temperature recording information from the main

river (Chao Phraya river) [http://www.wqmonline.com],

which is the source of water for inland aquaculture in

Thailand, is provided (Figure S1) as an example While

rapid changes in temperature impair the physiological

pro-cesses of fish metabolism and immunity (Langston et al

2002; Dominguez et al 2005; Kumari & Sahoo 2006), the

temperature range in the tropical zone is within the optimal

range for pathogen and parasite virulence Moreover, a

combination of the stresses that are caused by temperature

changes and acidic water has detrimental effects on the

growth and health of fish Although it is impractical to

control the fluctuations in the temperature and pH of water

in outdoor ponds, dietary manipulation by

supplementa-tion of essential micronutrients that minimize the stressful

responses of fish is a practical way to address this problem

Among nutritional antioxidants, it has been reported

that vitamins C (ascorbic acid) and E (a-tocopherol) play a

beneficial role in ameliorating stressful conditions in fish

(Montero et al 2001) Both vitamins are required for

nor-mal physiological functions Consequently, an appropriate

amount of these vitamins, which will vary depending on

the fish species, age or stage and the environmental

condi-tions, must be added to fish diets to ensure their normal

growth and health Vitamin C is involved in collagen

bio-synthesis, absorption of iron and the metabolism of stored

iron Further, it is a strong biological reducing agent that

provides electrons to free radicals in the aqueous phase in

vivo (Liposchitz et al 1971; Chaterjee 1978) Vitamin E

functions as a lipid component of cellular membranes and

also as an antioxidant by quenching singlet oxygen,

react-ing with OH radical and interacting with free radicals

Vitamin E eliminates lipid reactive oxygen species (ROS)

and, therefore, prevents lipid ROS from attacking new

unsaturated fatty acids Consequently, vitamin E is a major

chain-breaking antioxidant (Frei & Ames 1993) Vitamins

C and E synergistically support the overall cellular

antioxi-dant functions (Barclay et al 1983)

In this study, the quantitative level of co-supplemental

vitamins C and E on tissue accumulation, growth

perfor-mance, haematological indices, blood stress indicators and

humoral immune parameters were investigated more, the ability of these vitamins to minimize the adverseeffects of a combination of thermal and acidic stress wasinvestigated by means of blood stress indicators and non-specific immune responses

Further-A 3*3 factorial design with three levels of vitamin C (0, 500,and 1000 mg kg 1) and 3 levels of vitamin E (0, 62.5 and

125 mg kg 1) was employed in a randomized completedesign with five replicates (aquaria) Table 1 shows the basaldietary ingredients and the proximate composition of theexperimental diet The diet was analysed for moisture, crudeprotein, crude fat and ash content according to the standardmethod of Association of Official Analytical Chemists(AOAC) (1990) Nine experimental diets were designed toco-supplement vitamins C and E (Table 2) The experimen-tal diets were prepared individually to ensure that theycontained the desired levels of ascorbic acid and vitamin E(Stay C-35 and Rovimix E 50, respectively, Hofmann-LaRoche, Basel, Switzerland) and the vitamin equivalent kg 1

of dry food The vitamins were applied to the pellets in a1% soybean oil spray The L-ascorbyl-2-monophosphate

Table 1 Ingredients and chemical composition (g kg 1 ) of the basal diet

1 Vitamin and trace mineral mix provided the following (IU kg 1

or g kg 1 diet): biotin, 0.125 g; folic acid, 0.0015 g; inositol, 0.125 mg; niacin, 0.0158 g; pantothenic acid, 0.015 g; vitamin A,

2500 IU; vitamin B1, 0.0013 g; vitamin B2, 0.0006 g; vitamin B6, 0.0038 g; vitamin B12 0.00003 mg; vitamin D3, 500 IU; vitamin K, 0.004 g; copper, 0.01 g; iron, 0.1 g; selenium, 0.15 mg; zinc, 0.16 g.

2 Nitrogen-free extract = 1000 (crude protein + crude lipid + crude ash).

and the a-tocopheryl acetate concentrations of each

experimental diet were analysed by high-performance liquid

chromatography (HPLC)

Hybrid catfish were obtained from the Suranaree University

of Technology Farm (SUT Farm) Before beginning the

feeding experiment, 18 hybrid catfish (21–27 g) were

ran-domly selected and stocked in each aquarium (80 L) under

continuous aeration To acclimatize the hybrid catfish to the

experimental conditions, the fish were fed the basal diet for

two weeks Five aquaria (replications) were randomly

assigned to each dietary treatment The experimental hybrid

catfish were fed to apparent satiation twice a day for eight

weeks A flow-through water change system was implemented

by replacing one half of the water in each aquarium with

de-chlorinated water every three days During the experimental

period, the air and water temperatures were measured daily,

and the dissolved oxygen and pH were determined weekly by

means of a DO meter and a pH meter, respectively The air

and water temperatures (means± SD) were 31.9 ± 1.9 °C

and 29.9± 1.4 °C, respectively The pH and dissolved oxygen

were within acceptable ranges, that is, a pH of 8.14± 0.46

and a dissolved oxygen level of 6.76 ± 0.40 mg L 1

During the experiment, any dead fish were recorded and

removed daily The growth performance and feed

utiliza-tion were evaluated at the end of week 6 The content ofvitamins C and E was determined after feeding periods

of 3 and 6 weeks The fish were not fed for 18 h prior tosampling Two representative fish from each diet replica-tion were selected and anaesthetized with 2-phenoxyetha-nol (0.35 mL L 1) A blood sample was taken from thecaudal vein of each hybrid catfish with a 21-gauge needleand transferred to a tube in the presence of 1.0% (v/v) of15% EDTA After bleeding, the liver and kidney weredissected

After weeks 4 and 8, blood sampling was conducted toassess the effects of dietary vitamins C and E on haemato-logical parameters In addition, the blood stress indicatorsand humoral immune parameters of the fish (normalgroup) were examined Two fish from each diet replicationwere not fed for a period of 18 h and then anesthetizedwith phenoxyethanol A blood sample was taken from thecaudal vein of each hybrid catfish by means of a 21-gaugeneedle and divided into two aliquots One of the sets ofblood samples was transferred to tubes that contained1.0% (v/v) of 15% EDTA for the haematological assay

The other blood set was allowed to clot at room ture for 1 h The plasma was collected by centrifugation ofthe EDTA blood at 9000 g for 10 min at 4°C The serumwas collected by centrifuging the clotted blood at 9000 gfor 10 min at room temperature

tempera-Thermal and acidic stresses were assessed after week 4and week 8 of the experimental period Fish were not fedfor 18 h before exposing them to stress On the day of fishsampling as described earlier, two fish per replicationaquarium were randomly selected and subjected to stress

A combination of cold temperature and acidic stress wasachieved by exposing the sampled fish to acidic water (pH5.5) at a temperature of 19°C for 24 h (a cooling rate of0.1°C min 1

) The fish were then anesthetized with oxyethanol for blood collection

phen-Vitamin C extraction in the diet The dietary vitamin Clevel was determined according to the Stay C-35 (Roche)instructions, with some modifications Briefly, 5 g of eachexperimental diet was ground and extracted in 50 mL of

388 mM phosphate buffer, pH 3.0, for 15 min at roomtemperature The sample suspension was centrifuged at

15 000 g for 10 min at 4°C The clear supernatant was tered through a 0.45-lm-pore-size syringe filter Tenmicrolitres of the extracted diet was analysed to determinethe ascorbic acid levels

fil-Table 2 Supplementation of vitamins C and E and analysed levels

in nine experimental diets

L -AA, L -ascorbic acid; 2-AMP, L -ascorbyl-2-monophosphate; T,

a-tocopherol; a-T-Ac, a-tocopheryl acetate; ND, not detected.

.

Aquaculture Nutrition, 19; 148–162 ª 2012 Blackwell Publishing Ltd

Vitamin C extraction in tissues After feeding periods of 3

and 6 weeks, two fish per aquaria were randomly sampled

to determine the content of vitamin C in the plasma, liver

and kidney Each liver or kidney was weighed and then

homogenized in ice-cold 388 mM phosphate buffer (pH

3.0) that was equal in volume (mL) to 10 times the liver or

kidney weight (in grams) This step was followed by

sonica-tion in an ice bath for 3 min; 100lL of plasma was then

mixed with 900 lL of ice-cold phosphate buffer The

extract was then centrifuged at 15 000 g for 10 min The

clear supernatant was filtered through a 0.45-lm-pore-size

syringe filter Ten microlitres of clear filtrate was injected

into the HPLC system

Quantification of vitamin C The extracted vitamin C was

quantified by a reverse-phase HPLC (HP 1100) with a C18

column (4.0 mm ID9 250 mm), according to the Stay

C-35 (Roche) instructions, with some modifications The

mobile phase (flow rate of 0.8 mL min 1) was a solution of

Eluent1/Eluent2 (3 : 7) (Eluent 1: 78 mM potassium

dihy-drogen phosphate, 0.2% 1,5-dimethylhexylamine, pH3;

Eluent 2: 900 mL Eluent1, 98 mL acetonitrile, 42 mL

etha-nol) The effluent was monitored by a UV detector (wave

length of 254 nm) Peak identification was performed with

retention times that were compared to standard L-ascorbic

acid and L-ascorbic acid-2-phosphate magnesium salt

(Sigma, St Louis, MO, USA)

Vitamin E extraction in diet The dietary vitamin E was

determined according to Rupe´rez et al (1999), with some

modifications Briefly, 2 g of each experimental diet was

ground Then, 20 mL of 0.01% BHT (in methanol),

5 mL of 0.1 mM of EDTA, and 10 mL of hexane were

added The mixture was shaken for 5 min The hexane

extract was transferred to a new tube Five millilitres

more of hexane was added and shaken for 5 min The

hexane extract was again transferred and pooled into the

same tube, followed by drying under N gas The residue

was redissolved in 600lL of chloroform/methanol (1 : 1;

v/v)

Vitamin E extraction in tissues Tissue ∞-tocopherol was

analysed according to Schweigert et al (2002), with slight

modifications After a feeding period of 3 and 6 weeks,

two fish per aquaria were randomly sampled to determine

the vitamin E content of the plasma, liver and kidney

After weighing, each liver or kidney (approximately

0.3 g) was homogenized in a mixture reagent that

con-sisted of 1 mL of 0.05% BHT (in hexane/isopropanol,

3 : 2) and 1 mL of 0.1 M NaCl The extract was left atroom temperature for 30 min The hexane was trans-ferred to a new tube and dried under N gas The residuewas dissolved in 200lL of chloroform/methanol (1 : 1;v/v) Plasma (200lL) was mixed with 200 lL of ethanoland 1 mL of 0.05% BHT (in hexane) The organicextract was transferred to a new tube One millilitre more

of 0.05% BHT was added, and the mixture was shaken.The organic extract was transferred and pooled into thesame tube, followed by drying under N gas The residuewas redissolved in 200lL of chloroform/methanol (1 : 1;v/v)

Quantification of vitamin E The extracted vitamin E wasquantified by a reverse-phase HPLC (HP 1100) with a C 18column (4.0 mm ID9 250 mm) The mobile phase wasmethanol, with water (98 : 2; v/v) as eluent, at a flow rate

of 1.5 mL min 1 Fluorescence detection was performedwith excitation at 292 nm and emission at 327 nm Peakidentification was performed with retention times that werecompared to standard DL-a-tocopherol and a-tocopherylacetate (Sigma)

Immediately following the blood sampling, EDTA bloodwas used to analyse haematological parameters The redblood cell number (RBC) was measured in duplicate foreach sample using a Neubauer haemocytometer after dilu-tion with Grower’s solution (Voigt 2000) Haematocrit val-ues (Ht) were measured in duplicate by placing fresh bloodinto glass capillary tubes and centrifuging them for 5 min

by microhaematocrit centrifugation The haemoglobin (Hb)content was determined using the photometrical cyanohae-moglobin method The white blood cell number (WBC)was measured in duplicate for each sample using a Neu-bauer haemocytometer after dilution with Dacie’s solution(Dacie & Lewis 2001)

Immediately following the blood sampling, the EDTAblood was used to determine blood glucose levels in wholeblood that was drawn from the caudal vein by the use of ahand-held glucometer (AccuTrend; Roche) The serumchloride was determined by the thiocyanate method (Swain1956) Analysis of the total plasma protein content wasbased on measurements using a microprotein determinationmethod (C-690; Sigma)

The lysozyme activity in the serum was measured using a

turbidimetric assay The lysozyme activity was evaluated

from a standard curve that indicates the level of lysis of

Gram-positive bacterium Micrococcus lysodeikticus by a

known concentration of lysozyme standard This evaluation

was conducted as described elsewhere (Vechklang et al

2011) with some modifications The standard lysozyme was

hen egg white lysozyme (Sigma) Ten microlitres of standard

lysozyme ranging from 0 to 15lg mL 1(in a 0.06 M

phos-phate citrate buffer with pH of 6.0 and 0.09% of NaCl) or

serum was placed into a 96-well plate in duplicate The

sus-pension of M lysodeikticus (0.3 mg mL 1) was prepared in

a 0.06 M phosphate citrate buffer with a pH of 6.0 and

0.09% of NaCl, and 190lL of M lysodeikticus suspension

was added to each well The decrease in absorbance at

450 nm was recorded at 15-min intervals The calculation of

lysozyme activities was based on the lysozyme concentration

in comparison with that of the activities of a known

concen-tration of standard lysozyme

The total immunoglobulin was measured using the method

described in the study by Siwicki et al (1994) It was

deter-mined from the difference in the total plasma protein

before and after precipitation of the plasma

immunoglobu-lin, employing a 12% polyethylene glycol

The activity of the alternative complement pathway was

assayed using goat red blood cells (GRBC) as targets, as

described in the study by Sunyer & Tort (1995), with

modi-fication Briefly, the GRBC were washed three times in

GVB-EGTA (Gelatin Veronol Buffer; 10 mM barbital,

145 mM NaCl, 0.1% gelatin, 0.5 mM MgCl2, 10 mM

EGTA, pH 7.3–7.4) and resuspended to 5 9 107cells mL 1

in the same buffer As a complementary source, serial

dou-bling dilutions (20%, 10%, 5%, 2.5%, 1.25%, 0.625%,

0.313% and 0.157%) of the test serum in GVB-EGTA in a

final volume of 250lL were obtained Fifty microlitres of

GRBC was added and incubated for 90 min at room

tem-perature The control consisted of maximum haemolysis

(250lL of distilled water and 50 lL of GRBC) A

sponta-neous haemolysis (250 lL GVB-EGTA and 50 lL of

GRBC) was used as blank After incubation, the sample

was centrifuged at 14 000 g for 10 min to avoid unlysed

erythrocytes The relative haemoglobin content of the pernatants was assessed by reading their optical density at

su-415 nm The haemolytic activity as a percentage of themaximum was estimated as the percentage of haemolysis(Y) The lysis curve for each sample was constructed byplotting Y (1 Y) 1 against the volume of serum added(mL) on a log10–log10scaled graph The alternative comple-mentary pathway haemolytic activity (ACH50 inunits mL 1) was determined as the volume of serumcausing 50% lysis of the GRBC All of the assays wereconducted in duplicate

The analysis was performed using SPSS for Windows, sion 10 (SPSS Inc., Chicago, IL, USA) The statisticalmodel consisted of the effects of dietary vitamin E, vitamin

ver-C and their interactions A two-way factorial analysis ofvariance (ANOVA) was performed When significant differ-ences were found among factors, a one-wayANOVA follow-ing Duncan’s multiple range test was conducted to rankthe treatment combination groups

Regression analysis and goodness of fit (R2) were mined for the accumulation of vitamin C in tested tissuewhen there were no interaction effects of dietary vitamins

deter-C and E A regression analysis of the tissue vitamin deter-C (Y)and the level of vitamn C supplementation parameters (x)was conducted A t-test analysis was conducted to evaluatethe difference between the normal and stress challenges ofeach parameter within each treatment diet Throughout theexperiment, effects and differences were declared to be sig-nificant when P< 0.05

Analysed vitamins C and E and their derivatives in mental diets are shown in Table 2 The level of detectedascorbic 2-monophosphate and L-ascorbic acid responded

experi-to the supplementation of L-ascorbyl-2-monophosphate inall of the treatment diets The amount ofa-tocopherol wasmeasured in all treatment diets that might contain it in thefeed ingredients The analysed a-tocopheryl acetate in thetreatment diets was related to the supplementation levels

The determination of ascorbic acid in the plasma, liverand kidney was conducted after the 3- and 6-week feedingperiods (Fig 1) By the end of 3 weeks, increased supple-mentation of vitamin C in diets had significantly increasedthe level of ascorbic acid in the plasma (Fig 1a), liver(Fig 1b) and kidney (Fig 1c) Neither the vitamin E effect,

.

Aquaculture Nutrition, 19; 148–162 ª 2012 Blackwell Publishing Ltd

nor the effect of the interaction between vitamins C and E,

was observed for ascorbic acid in the plasma and tissues

that were tested Significant nonlinear relationships were

observed between dietary vitamin C (x) and ascorbic acid

content in plasma, liver or kidney (y) [plasma (R2= 0.819),

y= 1.546E-5x2+ 0.007x + 26.23; liver (R2= 0.995), y =

0.001x2 0.029x+ 139.08; kidney (R2= 0.982), y =

6.426E-5x2+ 0.171x + 68.64] Although there was no

sig-nificant interaction between vitamin C and E

supplementa-tion for plasma ascorbic acid, with dietary vitamin C at

1000 mg kg 1, increased vitamin E tended to decrease

plasma ascorbic acid For instance, fish that were fed

die-tary vitamin C at 1000 mg kg 1 and vitamin E at 62.5–

125 mg kg 1 had significantly lower plasma total ascorbic

acid than fish that were fed vitamin C at 1000 mg kg 1

alone By the end of 6 weeks, the renal total ascorbic acid

also had increased in response to supplementation levels of

vitamin C (Fig 1c) A significant linear relationship

(y= 0.215x + 92.005, R2= 0.989) was observed between

the dietary vitamin C (x) and the renal ascorbic acid (y)

No significant effects of dietary vitamin E or the tion between vitamins C and E on renal ascorbic acid weredetected However, supplementing with vitamins C and Ehad significant effects on plasma and hepatic ascorbic acid(Fig 1a,b) In addition, there were effects of interaction ofsupplemental vitamins C and E on the ascorbic acid con-tent in liver and plasma (P< 0.05) With dietary vitamin C

interac-at 500 mg kg 1, an increase in dietary vitamin E wasattributed to elevated hepatic and plasma ascorbic acidcontent Compared to fish who were fed a diet supple-mented with 1000 mg kg 1vitamin C alone, hepatic ascor-bic acid was higher in fish that were fed a dietsupplemented with vitamin C at 1000 mg kg 1and vitamin

E at 62.5 mg kg 1, but lower in fish on a diet that wassupplemented with vitamin C at 1000 mg kg 1and vitamin

E at 125 mg kg 1.Figure 1 shows thea-tocopherol content of plasma, liverand kidney of fish that were fed experimental diets for 3and 6 weeks By the end of week 3, the supplementation ofvitamins C and E had a significant effect on the plasma

Figure 1 Plasma (a), hepatic (b) and renal (c) vitamins C and E accumulation in hybrid catfish that were fed with experimental diets mins C and E presented as L-ascorbic acid and a-tocopherol, respectively Accumulations of L-ascorbic acid and a-tocopherol were analysed

Vita-at the end of week 3 and week 6 of the feeding period See Table 2 for analysed vitamins C and E levels in experimental diets The dVita-ata are the mean ± standard deviation from at least five samples Significant differences (P < 0.05) of vitamins C and E are denoted by different letters and numbers, respectively.

(Fig 1a), liver (Fig 1b) and kidney (Fig 1c) a-tocopherol

content Additionally, significant interaction effects of

vita-mins C and E were found on the a-tocopherol content of

the liver and kidney An increase in vitamin E significantly

elevated the a-tocopherol content, and with similar

amounts of vitamin E, an increase in vitamin C was

attrib-uted to the increased a-tocopherol content Moreover,

vitamin C at high levels clearly led to an increase in the

a-tocopherol content of the liver Nevertheless, fish that

were fed a diet that was supplemented with vitamin E at

125 mg kg 1and vitamin C at 1000 mg kg 1had a slightly

lower renal a-tocopherol content than fish that were fed a

diet that had been supplemented with vitamin E at

125 mg kg 1and vitamin C at 500 mg kg 1(P < 0.05) By

the end of week 6, there were significant effects of dietary

vitamins C and E as well as their interaction on the

a-tocopherol content of the plasma, liver and kidney

(P < 0.05) (Fig 1a–c) Increased supplementation of

vita-min E caused an increase in the a-tocopherol content of

both tissues and plasma (P < 0.05) With a similar

amount of dietary vitamin E, elevated dietary vitamin C

tended to increase the a-tocopherol content Moreover, the

ability of vitamin C to increase vitamin E levels was clear

in the plasma and kidney of fish that were fed high levels

of vitamin C (1000 mg kg 1)

Growth performance and survival rates of experimental

fish were evaluated at week 6 (Table 3) There were no

significant effects of vitamins C or E or their interaction

on the feed conversion ratio (FCR) Dietary vitamin Cdid not significantly influence the growth response repre-sented by weight gain (WG) and the specific growth rate(SGR), whereas dietary vitamin E significantly affectedthose growth characteristics Moreover, the interactioneffects of dietary vitamin C and E were observed in WGand SGR Fish that were fed a diet that was deficient in

C and E (control diet) had the lowest WG and SGR

Fish that were fed a diet that had been supplementedwith either vitamin C at 1000 mg kg 1 or vitamin E at62.5–125 mg kg 1 had a significantly better WG and SGRthan fish on the control diet With the exception of fishthat were on a diet that was supplemented with the high-est levels of both vitamins C and E, fish that were fed adiet that was supplemented with a combination of vitamin

C at 500–1000 mg kg 1 and vitamin E at 62.5–125 mg

kg 1 had a significantly greater growth response than fish

on the control diet There were no significant differences

in survival rate among the treatment groups (Table 3)

The survival rate of each treatment group ranged between73% and 85%, and the lowest survival rate was found infish that were fed the control diet No vitamins C and Edeficiency symptoms were observed in fish that were fedthe control diet However, fish that were fed a diet thatwas supplemented with vitamin E at 125 mg kg 1 for along period (8 weeks) showed abnormalities, such as

Table 3 Growth performance of hybrid catfish that were fed experimental diets for 6 weeks 1

Statistically significant levels are *P < 0.05, **P < 0.01, and ns P > 0.05.

1 Values are means ± SD of five replicates Means with a different superscript in each column differed significantly from each other

(P < 0.05).

2 Weight gain = 100 9 (final mean body weight initial mean body weight) 9 initial mean body weight 1

3 Specific growth rate (SGR) = 100 9 [(ln final mean body weight ln initial mean body weight) 9 experimental day 1 ]

4 Feed conversion ratio (FCR) = dry feed fed 9 wet weight gain 1

.

Aquaculture Nutrition, 19; 148–162 ª 2012 Blackwell Publishing Ltd

internal bleeding These abnormal symptoms were

proba-bly caused by hypervitaminosis of vitamin E and

hypovi-taminosis of vitamin C

The effects of dietary vitamins C and E on

haematologi-cal parameters were measured at weeks 4 and 8, as seen in

Table 4 At week 4, dietary vitamins C and E had

signifi-cant effects on RBC, Hb and Ht Moreover, interaction

effects of supplementing with vitamins C and E on the

RBC and Hb were detected (P< 0.05) Dietary vitamin C

or E increased the RBC, Hb and Ht in fish that were fed a

diet that was not supplemented with vitamin E or C,

respectively In addition, diets that were supplemented by a

combination of vitamins C and E increased the RBC, Hb

and Ht Supplementation with vitamin C significantly

influ-enced the WBC count, while supplementation with vitamin

E did not Significant interaction effects of vitamin C and

E appear to be present in the WBC numbers Fish that

were fed the control diet had the lowest WBC counts At

week 8, significant effects of vitamins C and E were found,

as well as their interaction on tested haematological

param-eters In addition, there was an interaction effect of dietary

vitamins C and E on WBC counts (Table 4)

Figure 2 shows the effects of a combination of thermal

and acidic stress on some blood stress indicators, including

blood glucose, plasma protein and serum chloride of fish

that were fed the experimental diets Dietary vitamins C

and E and their interactions had no effect on blood glucose

among the experimental groups for both of the analysed

periods (Fig 2a) Throughout the experimental period,

die-tary vitamin E did not significantly affect the serum

chlo-ride levels, whereas dietary vitamin C did so and

significantly (Fig 2b) In addition, the effect of interaction

between dietary vitamins C and E was detectable (P <

0.05) At week 4, the plasma protein was affected by

dietary vitamin C or E or their interaction (Fig 2c) Fish

that were fed a diet that was co-supplemented with high

concentrations of vitamin C (1000 mg kg 1) and low

concentrations of vitamin E (62.5 mg kg 1) or high

con-centrations of vitamin E (125 mg kg 1) and low

concentra-tions of vitamin C (500 mg kg 1) had the highest plasma

protein levels When the experimental period was continued

to 8 weeks, only the interaction of the effects of dietary

vitamin C and E on the plasma protein was detectable At

week 4, a combination of thermal and acidic stressors

caused an increase in the blood glucose of fish on the

control diet and the diets that were supplemented with low

levels of vitamin E alone (62.5 mg kg 1) At week 8, the

effects of stressors on increments of blood glucose levels

were observed in fish that were fed the experimental diet, Table

except diets that had been supplemented with vitamin C

alone at 500–1000 mg kg 1

or diets that had been mented with vitamin C 500 mg kg 1 and vitamin E

supple-62.5 mg kg 1 A combination of stressors appeared to

cause a reduction in serum chloride in fish that were fed all

of the experimental diets (P< 0.05) These stressors had

the effect of decreasing the plasma protein in fish that were

fed the control diet

Figure 3 shows the effects of dietary supplementation of

vitamins C and E on the immune responses of hybrid

catfish that were subjected to a combination of thermal

and acidic stress By the end of the 4 weeks of the feeding

period, the supplementation of vitamin C had significantly

enhanced lysozyme (Lz) (Fig 3a) and the total

immuno-globulin (Ig) (Fig 3b), but did not produce the same

effects for the alternative complement pathway (ACH50)

(Fig 3c) When the feeding period was extended to

8 weeks, the dietary vitamins C and E and their interaction

had significant effects on Lz, Ig and ACH 50 (Fig 3a–c)

Fish that were fed a diet containing 500 or 1000 mg kg 1

of vitamin C, regardless of the vitamin E supplemented

level, had higher levels of Lz and Ig than fish that were fedthe control diet Vitamin E at 125 mg kg 1 alone alsoincreased Lz and Ig compared to fish of the control diet

With each supplemental vitamin C level, an increase indietary vitamin E led to an increase in ACH 50 (P< 0.05)

At week 4, when fish were subjected to thermal and acidicstress, the Lz and Ig of the fish that were fed diets deficient

of vitamin C and diet supplemented with vitamin C at

500 mg kg 1 alone declined (P< 0.05) (Fig 3a,b) Thestressful conditions also had the effect of reducing the Ig infish that were fed a diet supplemented with vitamin C at

This stress also affected the reduction in Lz in fish that werefed a diet supplemented with vitamin E alone at62.5 mg kg 1(P < 0.05)

Figure 2 Effects of dietary vitamins C and E on blood stress indicators under normal and stressed conditions Blood stress indicators

including plasma glucose (a), serum chloride (b) and plasma protein (c) were assessed at the end of 4 weeks and 8 weeks of the feeding

per-iod The data are the mean ± standard deviation from at least five samples The different letters denote significant differences in blood

stress indicators among experimental diets at P < 0.05 Asterisks indicate the alteration of blood stress indicators between normal and stress

fish within each experimental diet group.

.

Aquaculture Nutrition, 19; 148–162 ª 2012 Blackwell Publishing Ltd