Aquaculture nutrition, tập 19, số 4, 2013

Soybean meal is often considered as the most reliable ingredient and cost-effective protein source in shrimp feed because of its high protein content, high digestibility, rela-tively wel

Department of Fisheries and Allied Aquacultures, Auburn University, Auburn, AL, USA

The production of the Pacific white shrimp (Litopenaeus

vannamei) has expanded to the point of being the most

widely cultured species of shrimp One of the advantages of

this species is its acceptance of a wide variety of feed

for-mulations including plant-based feeds Given the increases

in ingredient costs, particularly fish meal, there is

consider-able interest in the use of alternative feed formulations for

cultured species Given that soybean meal is one of the

most widely available protein sources for which production

can be expanded, the move to soy-based diets is inevitable

The successful use of alternative feed ingredients for shrimp

production depends on a number of factors This paper

summarizes studies regarding the move towards high soy

diets concerning manipulation of ingredients and nutrient

profiles to maintain balanced feed formulations

KEY WORDS: alternative feed, practical diets, soybean,

van-namei

Received 9 September 2011, accepted 20 January 2013

Correspondence: D.A Davis, Department of Fisheries and Allied

Aqua-cultures, 203 Swingle Hall, Auburn University, Auburn, AL 36849-5419,

USA E-mail: davisda@auburn.edu

Pacific white shrimp, Litopenaeus vannamei (Boone) is

native to the eastern Pacific Ocean from Sonora, Mexico to

Northern Peru Currently, it is the most popular cultured

shrimp species and has experienced a dramatic increase in

aquaculture production from 186 113 tonnes in 1999 to

2 296 630 tonnes in 2007 (FAO 2009) The industry growth

has been paralleled by an increase in shrimp feed

produc-tion The increases in demand and limitations of supply

have resulted in some ingredients becoming less availableand more costly, especially fish meal and fish oil Fish mealand other marine ingredients are considered desirableingredients in shrimp feed because of their nutrient contentand palatability In commercial feeds, fishmeal typicallyaccounts for 200–300 g kg 1

of the shrimp feed tion (Tacon & Metian 2008) The cost of fish meal and fishoil has generally increased over time as a result of theuncertainty of availability and large fluctuations in theprice Furthermore, there are growing social and environ-ment concerns regarding the long-term sustainability of theuse of marine ingredients In addition to feed pricesincreasing, the market value for shrimp has declinedbecause of increased production and limited demand Thishas resulted in a reduction in the profit margin for shrimpfarmers When margins were good, feed manufacturerscould afford to use expensive ingredients and over formu-late a diet However, as the margin decreased, feeds mustbecome more cost-effective Feed costs can account for asmuch as 40–60% of production costs (Hertrampf & Pie-dad-Pascual 2000) Feed costs and feed management bothinfluence the investment in feeds Reducing or removingcostly protein sources through the use of a combination ofless expensive and more economical protein and lipidsources could result in substantial saving in feed cost Prac-tical diets using plant-based ingredients to replace fish mealand fish oil have become an interesting alternative whichcould reduce these problems

formula-The use of renewable plant protein sources has becomethe focus of protein substitution studies in shrimp feedsaround the world because of their acceptable protein level,suitable amino acid content, economic opportunity andconsistent quality (Watanabe 2002) Formulated dietsare designed to contain sufficient levels of nutrients tomeet requirements using plant-based protein sources forwhich production can be expanded and are often more

cost-effective Feeding plant-based proteins to shrimp

requires that the ingredients possess certain nutritional

characteristics, such as low levels of fibre, starch (especially

insoluble carbohydrates) and antinutrients They must also

contain a relatively high protein content, favourable amino

acid profile, high nutrient digestibility and reasonable

pal-atability (Gatlin et al 2007; Naylor et al 2009)

Ten indispensable amino acids that are required for

growth and maintenance of shrimp are arginine, histidine,

isoleucine, leucine, lysine, methionine, phenylalanine,

threo-nine, tryptophan and valine (Kanazawa 1989; Guillaume

1997) These amino acids should satisfy shrimp

require-ments to support optimum growth performance Fish meal

is considered ideal protein source for fish and shrimp feed

production because of its high level of essential amino

acids On the other hand, plant protein sources contain

lower levels of some essential amino acids (Tacon 1994)

Thus, the balance of essential amino acids must be

consid-ered when diets are formulated to contain plant protein

sources to replace fish meal In general, the amino acid

profile of soybean meal is comparable with that of fish

meal, albeit is lower in sulphur amino acids, that is,

methi-onine and cystine (Peres & Lim 2008)

Soybean meal is often considered as the most reliable

ingredient and cost-effective protein source in shrimp feed

because of its high protein content, high digestibility,

rela-tively well-balanced amino acid profile, reasonable price

and steady supply (Davis & Arnold 2000; Amaya et al

2007a,b) The protein digestibility was found higher in

soybean protein than that in the marine animal meals

(Akiyama 1989) Ezquerra et al (1997, 1998) reported

in vivo and in vitro protein digestibility by pH drop of feed

using the white shrimp hepatopancreas ranged from 64%

to 91% where soybean protein showed greater APD thanthose in fish meal or crab meal (Table 1)

However, the inclusion of soybean meal at high levels or

as a sole protein source has resulted in reduced mance of the shrimp (Lim & Dominy 1990) This could bethe results of imbalanced amino acid profiles or deficiencies

perfor-of other dietary nutrients that were not taken into account

Fish meal is utilized as a protein source but it also provideslipids, essential fatty acids (EFAs), minerals and vitamins

to the diet Consequently, there will be most likely a need

to use a variety of feed ingredients in association with bean meal to provide a better balanced nutrient profile

soy-Utilization of various potential protein sources in shrimpfeeds such as animal by-product and other plant sources(listed in table 2) has been evaluated under different rear-ing conditions (Lim & Dominy 1990; Piedad-Pascual et al

1990; Sudaryono et al 1995; Cruz-Suarez et al 2001;

Amaya et al 2007a,b; Ray et al 2009)

One of those ingredients that are considered a promisingalternative for the substitution of fish meal in shrimp feeds

is poultry by-product meal (Davis & Arnold 2000;

Samocha et al 2004; Amaya et al 2007a; Markey 2007)

Distiller’s dried grains with solubles (DDGS) is also apotential protein source for shrimp feed because of its lowcost and consistent supply as a coproduct of the bio-etha-nol production, which is expected to increase rapidly in thenext decade Several studies reported the successful use ofDDGS as an alternative protein source in fish and crusta-cean feeds without causing negative impact on growth per-formance (Webster et al 1991, 1992; Wu et al 1994;

Cheng & Hardy 2004; Coyle et al 2004; Stone et al 2005;

Lim et al 2007, 2009; Robinson & Li 2008; Thompson

et al 2008) Pea meal is also another widely used feed

Table 1 Chemical composition of the test ingredients and in vivo and in vitro protein digestibility of L vannamei fed different protein

1

et al 1998).

.

ingredient, mostly in livestock because of its high energy,

moderate protein level (220–260 g kg 1 crude protein),

amino acid profile and low cost (Borlongan et al 2003)

Several studies indicated that feed pea is another potential

ingredient in fish and shrimp feeds (Gomes et al 1995;

Bu-rel et al 2000; Carter & Hauler 2000; Gouveia & Davies

2000; Booth et al 2001; Cruz-Suarez et al 2001; Davis

et al 2002; Bautista-Teruel et al 2003; Borlongan et al

2003) Because of the limitation in nutrient component of

most ingredients, more than one ingredient is required for

balanced feed formulations Therefore, shrimp diets

con-taining soybean meal as a main protein source should be

combined with other alternative protein ingredients, that is,

poultry by-product meal, DDGS and pea meal

Soybean and its products are acceptable protein sources

with good digestibility for shrimp However, soybean meal

is deficient in the essential amino acids (EAAs) such as

methionine, lysine and tryptophan as well as essential fatty

acids and minerals (Lim & Dominy 1990) Methionine is

one of the ten essential or indispensable amino acids that

are dietary essential for shrimp (Millamena et al 1996)

Thus, supplementation of sulphur amino acids, that is,

methionine or cystine, in soybean-based diets to meet the

shrimp requirement is recommended to provide a good

growth response (Akiyama 1989) Low levels of methionine

found in soybean meal can also be countered by mixing

with other protein sources and/or the supplementation of

synthetic methionine Several studies had reported

success-fully replacing fish meal with soybean meal with a

methio-nine supplement in Milkfish (Davis et al 1995; Shiau et al

2007) Conversely, a diet containing only soybean protein

with a methionine supplement was poorly utilized by red

drum (Reigh & Ellis 1992) McGoogan & Gatlin (1997)

suggested that diets containing soybean meal with low

lev-els or no fish meal may have palatability problems Thus,

the inclusion of attractants or palatability enhancers, forexample, fish solubles, may be considered A reduction infeed intake was reported in largemouth bass fed diets withincreased soybean meal levels (Cho et al 1974; Kubitza

et al 1997) Similar results were observed in red drum(Reigh & Ellis 1992; Davis et al 1995) and Pacific whiteshrimp (Lim & Dominy 1990)

Along with protein, lipids constitute the major trients that are required to provide the energy and cellularbuilding blocks as well as maintain growth, health, welfareand reproduction in shrimp (Lim et al 1997) Reducing orreplacement of marine ingredients that are good sources ofhigh quality oils from shrimp feed formulation may result inEFAs deficiencies As we replace fish meal with alternativeingredients, for example soybean meal, we must ensure that

macronu-we meet the shrimp EFAs requirements Lipid content andthe associated C18 PUFA (poly unsaturated fatty acids),linoleic (18:2n-6) and linolenic (18:3n-3) acids, as well as n-3and n-6HUFA (highly unsaturated fatty acids), eicosapenta-enoic acid (EPA, 20:5n-3), docosahexaenoic acid (DHA,22:6n-3) and arachidonic acid (ARA, 20:4n-6) are required

in shrimp and other crustacean feeds at levels between 5and 10 g kg 1 (Akiyama et al 1991; Gonzalez-Felix &Perez-Velazquez 2002) Generally, the primary protein andlipid sources used in practical shrimp feeds are fish meal andfish oil (Cheng et al 2002) Lim et al (1997) have reportedthat menhaden oil rich in n-3 HUFA (20:5n-3 and 22:6n-3)was better utilized by Penaeus vannamei than vegetable lipidsources such as linseed, sunflower, corn, soybean and coco-nut oil and stearic acid Samocha et al (2010) also suggestedthat the supplementation of HUFAs is a critical component

to replace marine fish oil in shrimp feed Their results onstrated that the complete replacement of fish meal andfish oil using non-marine ingredients can be accomplishedusing supplementation of plant oils with DHA- and

dem-Table 2 Chemical composition of the test ingredients (as-fed basis) (NRC 2011)

ARA-rich oils from fermented products Other studies have

reported that partial or total placement of fish meal and fish

oil with soybean meal and soy oil had no adverse effect

on shrimp growth performance (Davis & Arnold 2000;

Samocha et al 2004; Gonzalez-Felix et al 2010), but

shrimp body crude fat and cholesterol concentration were

reduced (Cheng & Hardy 2004) According to

Gonzalez-Felix et al (2010), the substitution of fish oil up to 90% by

plant-based oils in diets can be done without a significant

reduction in growth performance, FCR, production yield

and survival in L vannamei Apparently, this 10% of fish

oil remaining in the diet supply enough of the essential fatty

acids ARA, EPA and DHA for the proper development of

this species, although the fatty acid composition of the

mus-cular tissue of the animal reflected the lipid source fatty acid

profile added to the diet followed by a reduction in HUFAs

as fish oil was replaced High levels of n-3 fatty acids can be

obtained with the use of linseed oil, most of it comprised of

the PUFA a-linolenic acid; yet, the levels of the essential

HUFAs in linseed oil are found at low levels

Cholesterol is a vital component of cell membranes It is

the precursor of bile acids, steroids, and moulting

hor-mones (Cheng & Hardy 2004) It is reported to be an

essential nutrient for growth and survival of shrimp

(Kanazawa et al 1971; Gong et al 2000; Morris et al

2011) Gong et al (2000) suggested that dietary cholesterol

requirement of L vannamei juveniles was affected by

die-tary phospholipids such as soybean lecithin and purified

phosphatidylcholine Phospholipids are considered another

dietary necessity for optimum shrimp growth Dietary

cho-lesterol and phospholipids interact to improve growth as

well as affect retention of total lipid and triglycerides in

hepatopancreas and cholesterol in muscle of L vannamei

juveniles Several studies have indicated a clear need for

cholesterol supplementation in plant-based diets (Gong

et al 2000; Morris et al 2011) Gong et al (2000)

sug-gested that optimal growth of L vannamei was obtained

with 3.5 g kg 1, 1.4 g kg 1, 1.3 g kg 1and 0.5 g kg 1

sup-plemental cholesterol at dietary PL levels of 0, 15, 30 and

50 g kg 1, respectively, in dietary treatments containing no

fish meal Similar results reported by Morris et al (2011)

demonstrated that the cholesterol supplements in dietary

treatments formulated with no fish meal and targeted crude

protein levels of 350 g kg 1 for L vannamei were between

0.2 and 0.4 g kg 1, thus containing a cholesterol level

between 0.76 and 1.1 g kg 1of diet

Shrimp are able to assimilate minerals directly from the

aquatic environment (Montoya et al 2000) In shrimp,

minerals serve as structural components of hard and soft

tissues and metalloproteins as well as enzymatic cofactorsand enzymatic activators (Davis & Lawrence 1997) Shrimpcan utilize some soluble minerals such as calcium, copper,iron, magnesium, phosphorus, potassium, selenium, sodiumand zinc from the water through the gill, epidermis orboth Generally, phosphorus is found at low concentration

in natural water relative to its requirement by ton (Boyd 2007) When fish meal is replaced by soybeanmeal, the first limiting mineral in shrimp feed formulation

phytoplank-is phosphorus as only 30–40% of total phosphorus content

in soybean meal is available for L vannamei (Hertrampf &

Piedad-Pascual 2000) Therefore, supplemental phosphorus

is essential for optimal shrimp growth as fishmeal wasremoved The dietary phosphorus requirement for juvenile

L vannamei ranges from 3.4 to 20 g kg 1 (Davis et al

1993) and 20.9 – 22.0 g kg 1

for postlarval L vannamei(Niu et al 2008) The dietary phosphorus requirement forshrimp is dependent on the calcium content in dietalthough a dietary calcium supplement is not required(Davis et al., 1993; Cheng et al 2006)

Compared with fish meal, soybean meal is found to havelow availability of selenium, 48.0% and 17.5%, respectively(Gabrielsen & Opstvedt 1980) Selenium is an essentialtrace element that functions as a component of the enzymeglutathione peroxidase in shrimp, but it can be toxic (Davis

& Gatlin 1996; Wang et al 2006) Glutathione peroxidaseconverts hydrogen peroxide and lipid hydroperoxides intowater and lipid alcohols, respectively, thus protecting thecell from the deleterious effects of peroxides (Davis & Gat-lin 1996) Juvenile P vannaemi was found to grow bestwhen fed semi-purified diets supplemented with 0.2–0.4 mg

Se kg 1diet (Davis & Gatlin 1996) Supplemental selenium

is not required in practical diets containing more than

150 g kg 1fish meal Therefore, selenium supplementationmay be required in diets formulated with predominantlyplant ingredients Due to potentially toxic effects, seleniumsupplementation of 0.1 mg kg 1 is approved to be usedwith fish and crustaceans (Davis & Gatlin 1996)

There are other issues of using soybean meal as an native to fishmeal besides nutritional factors such as thepresence of nutrient inhibitors Raw soybean contains anti-nutritional factors such as trypsin inhibitors, lectins, oligo-saccharides, antigens and saponins that may affect thedigestion and reduce nutrient availability to shrimp(Dersjant-Li 2002) However, the effect of some of theseantinutrients can be reduced by heat process (New 1987)

alter-Clearly, the use of soybean meal in shrimp feed is ble However, there are several other plant protein sourcesthat may be considered as alternative ingredients used in

feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi- feasi-.

association with soybean meal to balance nutritional

com-position in feed formulations when non-fishmeal diets are

formulated It appears that fish meal can be partially or

completely removed from shrimp formulations if suitable

alternative sources of protein and lipids are provided to

meet the nutritional requirements of the animal (Lim &

Dominy 1990, 1992; Piedad-Pascual et al 1990; Sudaryono

et al 1995; Cruz-Suarez et al 2001; Smith et al 2001;

Davis et al., 2004; Samocha et al 2004; Amaya et al

2007a,b; Roy et al., 2009) Recently, National Research

Council (NRC 2011) reported the minimum nutrient

requirements for maximum performance of L vannamei

(Table 3) Yet, there is still limited information available

on amino acid requirement data for L.vannamei, as well as

fatty acids, vitamins and minerals are highly digestible;

therefore, the values presented represent nearly 100%

bio-availability

The use of complementary ingredients is a practice used

to obtain a more balanced nutrient profile in the feeds (i.e

essential amino acids, fatty acids, minerals) and to increase

nutrient utilization and facilitate feed processing (Amaya

et al 2007a) Sookying (2010) reported on a series of

stud-ies that demonstrated and developed a range of soy-based

diets This includes diets that demonstrate that fish meal

(100 g kg diet) could be totally removed from diets for

L vannameiby a combination of plant and animal proteinsources (soybean meal and poultry by-product meal) or allplant protein sources (soybean meal in combination withDDGS or pea meal with the inclusion of corn gluten mealand squid meal) when diets are formulated to containacceptable nutrient levels and proper balanced nutrientswithout any apparent effect on survival, growth and feedpalatability (Sookying & Davis 2011) They also demon-strated that up to 120 g kg 1soy protein concentrate could

be used in a high soy diet under outdoor production tions without an effect on production performance of theshrimp (Sookying & Davis 2012)

condi-Alternative feed formulations for the pacific whiteshrimp seem to work across a number of culture technolo-gies (clear water research systems, outdoor tank systemsand research ponds) as well as across a range of densities

in outdoor ponds (Sookying et al 2011) Given the range

of culture systems and densities, the use of alternative feedformulations for this species is warranted and appropriatefor commercial production

Akiyama, D.M (1989) Soybean meal utilization by marine shrimp In: Proceeding of the World Congress, Vegetable Protein Utiliza- tion in Human Foods and Animal Feedstuffs (Applewhite, T.H.

IL, USA.

Akiyama, D.M., Dominy, W.G & Lawrence, A.L (1991) Penaeid

American Soybean Association and Oceanic Institute, analo, USA.

Waim-Amaya, E.A., Davis, D.A & Rouse, D.B (2007a) Replacement of fish meal in practical diets for the Pacific white shrimp (Litope-

Booth, M.A., Allan, G.L., Frances, J & Parkinson, S (2001) Replacement of fish meal in diets for Australian silver perch, Bidyanus bidyanus: IV Effects of dehulling and protein concen-

85.

Borlongan, I.G., Eusebio, P.S & Welsh, T (2003) Potential of feed pea (Pisum sativum) meal as a protein source in practical diets

Boyd, C.E (2007) Phosphorus: key to phytoplankton m

Burel, C., Boujard, T., Tulli, F & Kaushik, S.J (2000) ity of extruded peas, extruded lupin, and rapeseed meal in

(from NRC 2011)

Item

Minimum requirement

Typical energy and protein concentrations

ingredients in which the nutrient composition has been defined.

.

Aquaculture Nutrition 19; 441–448 ª 2013 John Wiley & Sons Ltd

rainbow trout (Oncorhynchus mykiss) and turbot (Psetta

Carter, C.G & Hauler, R.C (2000) Fish meal replacement by

plant meals in extruded feeds for Atlantic salmon, Salmo salar

Cheng, Z.J & Hardy, R.W (2004) Nutritional value of diets

con-taining distiller’s dried grain with solubles for rainbow trout,

Cheng, Z.J., Behnke, K.C & Dominy, W.G (2002) Effects of

poultry by-product meal as a substitute for fish meal in diets on

growth and body composition of juvenile Pacific white shrimp,

Cheng, K., Hu, C., Liu, Y., Zheng, S & Qi, X (2006) Effects of

dietary calcium, phosphorus and calcium/phosphorus ratio on

the growth and tissue mineralization of Litopenaeus vannamei

Cho, C.Y., Bayley, H.S & Slinger, S.J (1974) Partial replacement

of herring meal with soybean meal and other changes in a diet

for rainbow trout (Salmo gairdneri) J Fish Res Board Can.,

31, 1523–1528.

Coyle, S.D., Mengel, G.J., Tidwell, J.H & Webster, C.D (2004)

Evaluation of growth, feed utilization, and economics of

diets containing different protein sources in combination with

370.

Cruz-Suarez, L.E., Ricque-Marie, D., Tapia-Salazar, M.,

McCal-lum, 274.I.M & Hickling, D (2001) Assessment of differently

processed feed pea (Pisum sativum) meals and canola meal

(Bras-sica sp.) in diets for blue shrimp (Litopenaeus stylirostris)

Davis, D.A & Arnold, C.R (2000) Replacement of fish meal in

practical diets for the Pacific white shrimp, Litopenaeus

Davis, D.A & Gatlin, D.M (1996) Dietary mineral requirements

Davis, D.A & Lawrence, A.L (1997) Minerals In: Crustacean

Nutrition (D’Abramo, L.R., Conklin, D.E & Akiyama, D.M.

Rouge, LA, USA.

Davis, D.A., Lawrence, A.L & Gatlin, D.M (1993) Response of

Davis, D.A., Jirsa, D & Arnold, C.R (1995) Evaluation of

soy-bean proteins as replacements for menhaden fish meal in

practi-cal diets for the red drum Sciaenops ocellatus J World

Davis, D.A., Samocha, T.M., Bullis, R.A., Patnaik, S., Browdy,

C.L., Stokes, A.D & Atwood, H.L (2004) Practical diets for

Li-topenaeus vannamei (Boone, 1931): working towards organic

Nutricion Acuıcola (Cruz Suarez, L.E., Ricque Marie, D., Nieto

Lopez, M.G., Villarreal Cavazos, D.A., Scholz, U & Gonzalez

Felix, M.L eds), pp 202–214, 16 al 19 de Noviembre de 2004.

Hermosillo, Sonora, Mexico.

Davis, D.A., Arnold, C.R & McCallum, I (2002) Nutritional

value of feed peas (Pisum sativum) in practical diet formulations

Dersjant-Li, Y (2002) The use of soy protein in aquafeeds In:

Advaces en Nutricion Acuicola VI Memorias del VI Simposium

Internacional de Nutrcion Acuicola, 3 al 6 de Septiembre del

Gaxiola-Cortes, M.G & Simoes, N eds), pp 541–558 Cancun, Quintana Roo, Mexico.

Ezquerra, J.M., Garcia-Carreno, F.L., Civera, R & Haard, N.F.

(1997) pH-stat method to predict digestibility in white shrimp

Ezquerra, J.M., Garcia-Carreno, F.L & Carrillo, O (1998)

In vitro digestibility of protein sources for white shrimp Penaeus

Food and Agriculture Organization of the United Nations (FAO).

(2009) FAO Yearbook 2007 Fishery and Aquaculture Statistics,

FAO, Rome, Italia.

Gabrielsen, B.O & Opstvedt, J (1980) Availability of selenium in fish meal in comparison with soybean meal, corn gluten meal and selenomethionine relative to selenium in sodiumselenite for restoring glutathione peroxidase activity in selenium-depleted

Gatlin, D.M., Barrows, F.T., Brown, P et al (2007) Expanding the utilization of sustainable plant products in aquafeeds: a

Gomes, E.F., Rema, P & Kaushik, S.J (1995) Replacement of fish meal by plant proteins in the diet of rainbow trout (Oncorhyn- chus mykiss): digestibility and growth performance Aquaculture,

130, 177–186.

Gong, H., Lawrence, A.L., Jiang, D.H., Castille, F.L & GatlinIii, D.M (2000) Lipid nutrition of juvenile Litopenaeus vannamei: I.

Dietary cholesterol and de-oiled soy lecithin requirements and

of lipid nutrition of Pacific white shrimp, Litopenaeus vannamei.

Gonzalez-Felix, M.L., da Silva, F.S.D., Davis, D.A., Samocha, T.M., Morris, T.C., Wilkenfeld, J.S & Perez-Velazquez, M.

(2010) Replacement of fish oil in plant based diets for Pacific

Gouveia, A & Davies, S.J (2000) Inclusion of an extruded hulled pea seed meal in diets for juvenile European sea bass

Guillaume, J (1997) Protein and amino acids In: Crustacean tion (D’Abramo, L.R., Conklin, D.E & Akiyama, D.M eds), pp.

Hertrampf, J.W & Piedad-Pascual, F (2000) Handbook on dients for Aquaculture Feeds, pp 573 Kluwer Academic Pub- lishers, Dordrecht, The Netherlands.

Ingre-Kanazawa, A (1989) Protein requirements of Penaeid shrimp In:

Advances in Tropical Aquaculture, Tahiti (French Polynesia), 20

Kanazawa, A., Tanaka, N., Teshima, S & Kashiwada, K (1971) Nutritional requirements of prawn-II Requirement for sterols.

Kubitza, F., Lovshin, L.L & Lovell, R.T (1997) Identification of feed enhancers for juvenile largemouth bass Micropterus salmo-

Lim, C & Dominy, W (1990) Evaluation of soybean meal as a replacement for marine animal protein in diets for shrimp (Pena-

Lim, C & Dominy, W (1992) Substitution of full-fat soybeans for commercial soybean meal in diets for shrimp Penaeus vannamei.

.

Lim, C., Ako, H., Brown, C.L & Hahn, K (1997) Growth

response and fatty acid composition of juvenile Penaeus

153.

Lim, C., Garcia, J.C., Yildirim-Aksoy, M., Klesius, P.H.,

Shoe-maker, C.A & Evans, J.J (2007) Growth response and

resis-tance to Streptococcus iniae of Nile tilapia, Oreochromis

niloticus, fed diets containing distiller’s dried grains with

Lim, C., Yildirim-Aksoy, M & Klesius, P.H (2009) Growth

response and resistance to Edwardsiella ictaluri of channel

–193.

Markey, J.C (2007) Replacement of poultry by-product meal in

production diets for the Pacific white shrimp (Litopenaeus

vanna-mei), M.Sc Thesis Auburn University, Auburn, AL pp 44.

McGoogan, B.B & Gatlin, D.M (1997) Effects of replacing fish

meal with soybean meal in diets for red drum Sciaenops ocellatus

and potential for palatability enhancement J World Aquaculture

Millamena, O.M., Bautista-Teruel, M.N & Kanazawa, A (1996)

Methionine requirement of juvenile tiger shrimp Penaeus

Montoya, R.A., Lawrence, A.L., Grant, W.E & Velasco, M.

(2000) Simulation of phosphorus dynamics in an intensive

shrimp culture system: effects of feed formulations and feeding

Morris, T., Samocha, T.M., Davis, D.A & Fox, J.M (2011)

Cho-lesterol supplements for Litopenaeus vannamei reared on plant

based diets in the presence of natural productivity Aquaculture,

314, 140–144.

National Research Council (NRC) (2011) Nutrient Requirements

of Fish and Shrimp, pp 360 National cademies Press

Washing-ton, DC, USA.

Naylor, R.L., Hardy, R.W., Bureau, D.P., Chiu, A., Elliott, M.,

Farrell, A.P., Forster, I., Gatlin, D.M., Goldburg, R.J & Hua,

K (2009) Feeding aquaculture in an era of finite resources Proc.

New, M.B (1987) Feed and feeding of fish and shrimp A

man-ual on the preparation and presentation of compound feeds

for shrimp and fish in aquaculture United nations

275 pp.

Niu, J., Liu, Y.J., Tian, L.X., Mai, K.S., Yang, H.J., Ye, C.X &

Gao, W (2008) Effect of dietary phosphorus sources and

vary-ing levels of supplemental phosphorus on survival, growth and

body composition of postlarval shrimp (Litopenaeus vannamei).

Peres, H & Lim, C (2008) Utilization of soybean products in diets

of nonsalmonid marine finfish In: Alternative Protein Sources in

Aquaculture Diets (Lim, C., Webster, C.D & Lee, C.S eds), pp.

York, NY, USA.

Piedad-Pascual, F., Cruz, E.M & Sumalangcay, A Jr (1990)

Sup-plemental feeding of Penaeus monodon juveniles with diets

Ray, A.J., Lewis, B.L., Browdy, C.L & Leffler, J.W (2009)

Sus-pended solids removal to improve shrimp (Litopenaeus vannamei)

production and an evaluation of a plant-based feed in

Robinson, E.H & Li, M.H (2008) Replacement of soybean meal

in channel catfish, Ictalurus punctatus, diets with cottonseed meal and distiller’s dried grains with solubles J World Aquaculture

Roy, L.A., Bordinhon, A., Sookying, D., Davis, D.A & Whitis, G.N (2009) Demonstration of alternative feeds for the Pacific white shrimp (Litopenaeus vannamei) reared in low salinity

Samocha, T.M., Davis, D.A., Saoud, I.P & DeBault, K (2004) Substitution of fish meal by co extruded soybean poultry by- product meal in practical diets for the Pacific white shrimp, Li-

Samocha, T.M., Patnaik, S., Davis, D.A., Bullis, R.A & Browdy, C.L (2010) Use of commercial fermentation products as a highly unsaturated fatty acid source in practical diets for the

Shiau, S.Y., Pan, B.S., Chen, S., Yu, H.L & Lin, S.L (2007) cessful use of soybean meal with a methionine supplement to replace fish meal in diets fed to milkfish Chanos chanos Forskal.

Smith, D.M., Allan, G.L., Williams, K.C & Barlow, C.G (2001) Fishmeal replacement research for shrimp feed in Australia In:

Sookying, D (2010) Development and application of soybean based diets for Pacific white shrimp Litopenaeus vannamei, Ph.D Dissertation Auburn University, Auburn, AL pp 141.

Sookying, D & Davis, D.A (2011) Pond production of Pacific

408 white shrimp (Litopenaeus vannamei) fed high levels of

Sookying, D & Davis, D.A (2012) Use of soy protein concentrate

in practical diets for Pacific white shrimp (Litopenaeus vannamei)

Sookying, D., Silva, F.S.D., Davis, D.A & Hanson, T (2011) Effects of stocking density on the performance of Pacific white shrimp Litopenaeus vannamei cultured under pond and outdoor tank conditions using a high soybean meal diet Aquaculture,

319, 231–239.

Stone, D.A.J., Hardy, R.W., Barrows, F.T & Cheng, Z.J (2005) Effects of extrusion on nutritional value of diets containing corn gluten meal and corn distiller’s dried grain for rainbow trout,

Sudaryono, A., Hoxey, M.J., Kailis, S.G & Evans, L.H (1995) Investigation of alternative protein sources in practical diets for

Tacon, A.G.J (1994) Feed Ingredients for Carnivorous Fish cies: Alternatives to Fishmeal and Other Fishery Resources, pp.

Spe-39 FAO, Fisheries Circular No 881, Rome, Italy.

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

Thompson, K.R., Rawles, S.D., Metts, L.S., Smith, R.G., satt, A., Gannam, A.L., Twibell, R.G., Johnson, R.B., Brady, Y.J & Webster, C.D (2008) Digestibility of dry Matter, protein,

Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim- Wim-.

Aquaculture Nutrition 19; 441–448 ª 2013 John Wiley & Sons Ltd

lipid, and organic matter of two fish meals, two poultry

by-prod-uct meals, soybean meal, and distiller’s dried grains with solubles

in practical diets for sunshine bass, Morone chrysops x M

Wang, W.-N., Wang, A.-L & Zhang, Y.-J (2006) Effect of dietary

higher level of selenium and nitrite concentration on the cellular

563.

Watanabe, T (2002) Strategies for further development of aquatic

Webster, C.D., Tidwell, J.H & Yancey, D.H (1991) Evaluation of distillers’ grains with solubles as a protein source in diets for

Webster, D., Tidwell, J.H., Goodgame, L.S., Yancey, D.H &

Mackey, L (1992) Use of soybean meal and distillers grains with solubles as partial or total replacement of fish meal in diets for

Wu, Y.V., Rosati, R., Sessa, D.J & Brown, P (1994) Utilization

of protein-rich ethanol co products from corn in tilapia feed.

.

The aim of the present study was to determine the

opti-mum dietary levels of krill phospholipids (KPL) for sea

bream (Sparus aurata) larvae, and its influence on larval

development and digestive enzymes activity Larvae were

fed five formulated microdiets with five different levels of

KPL Complete replacement of live preys with the

experi-mental microdiets for seabream larvae produced high

sur-vival and growth rates, particularly in fish fed the highest

levels of KPL In the present study, increase in dietary

KPL up to 120 g kg 1 (100 g kg 1 total PL) significantly

improved larval survival and growth, whereas further

increase did not improve those parameters An increase in

alkaline phosphatase, trypsin and lipase activity with the

elevation of KPL up to 120 g kg 1was also found

denot-ing a better functiondenot-ing of digestive system Besides, there

was a linear substrate stimulatory effect of dietary KPL on

phospholipase A2 activity Finally, increasing dietary KPL

lead to better assimilation of n-3 HUFA especially

eicosa-pentaenoic acid, reflected in the higher content of these

fatty acids in both neutral and polar lipids of the larvae In

summary, KPL were found to be an excellent source of

lip-ids for seabream larvae Optimum inclusion levels of this

ingredient in microdiets to completely substitute live preys

at this larval age were found to be 120 g kg 1KPL

KEY WORDS: alkaline phosphatase, fatty acids, krill

phospho-lipids, phospholipase A2, sea bream larvae, trypsin

Received 30 January 2012; accepted 18 June 2012

Correspondence: R Saleh, Grupo de Investigacio´n en Acuicultura (IUSA

& ICCM), University of Las Palmas de Gran Canaria, Carretera de

Taliarte, s/n, 35200 Telde, Gran Canaria, Espan˜a and Oceanography

Department, Faculty of Science, Alexandria University, 21515

Mo-harram Bek, Alexandria, Egypt E-mail: reda-saleh@hotmail.com

Dietary phospholipids (PL) improve culture performance

of various freshwater and marine fish species (Izquierdo &Koven 2011), enhancing growth and survival, reducingmorphological alterations in larvae (Kanazawa 1993; Salhi

et al 1999; Izquierdo et al 2001; Kjørsvik 2009) and earlyjuveniles (Coutteau et al 1997), and increasing fish resis-tance to stress (Takeuchi et al 1992; Kanazawa 1993).Despite PL metabolic pathways, including those of de novo

PL biosynthesis, are essentially the same in fish as in mals (Caballero et al 2006b), the fish larvae and earlyjuvenile have a limited capacity to synthesize de novo(Coutteau et al 1997; Salhi et al 1999) Thus, addition of

mam-PL to diets for fish larvae contributes to assimilation ofdietary lipid by increasing the enteric lipoprotein synthesis(Liu et al 2002; Hadas et al 2003) and release in laminapropria, significantly reducing lipoprotein size by promot-ing VLDL synthesis (higher in PL), rather than chylomi-cron production (higher in NL) (Liu et al 2002; Caballero

et al 2003, 2006b) Indeed, young fish receive abundant of

PL during embryo and larval development either from yolksac lipids or from wild preys (Rainuzzo et al 1997; VanDer Meeren et al 2008) Therefore, PL seems to be essen-tial for the adequate growth and development of fishlarvae Described phospholipid requirements are around

20–40 g kg 1

DW of diet for juvenile fish, and they may behigher in larval fish Frequently, those requirements haveused plant PL such as soybean lecithin or egg yolk lecithin,whereas marine PL, rich in docosahexaenoic acid (DHA)and eicosapentaenoic acid (EPA), have been more scarcelystudied (Salhi et al 1999; Cahu et al 2003) DHA andEPA present in the PL fraction of larval diets seem to bemore beneficial than in the NL fraction (Salhi et al 1999;Cahu et al 2003; Gisbert et al 2005; Wold et al 2007)

Cahu et al (2003) carried out a dose–response study with

sea bass larvae, using five levels of phospholipids at a

con-stant dietary lipid level (PL, 27–116 g kg 1 dw) They

found that the diet with the highest dietary PL gave the

best larval performance and lower skeletal malformation

rates A similar result was found by Hamza et al (2008)

for pikeperch larvae, which also showed best growth with

the diet highest in PL (90 g kg 1of dry matter)

The digestive system of larvae is not fully developed at

first feeding The digestion of ingested food occurs in the

larval intestine, where the pH remains alkaline and

trypsin-like enzyme activity accounts for the proteolytic activity

(Walford & Lam 1993) At first feeding, the pancreatic and

intestinal enzyme activities are generally low (Cousin et al

1987) Digestive enzyme activity increase during the first

10 dph in Solea senegalensis (Ribeiro et al 1999), whereas

an increase in alkaline phosphatase activity has been found

to reflect the development of the brush border membranes

of enterocytes in Atlantic cod (Gadus morhua) (Wold et al

2007) Moreover, the addition of dietary PL enhanced gut

maturation index in this species, based on the relation

between brush border alkaline phosphatase and cytosolic

leucine–alanine aminopeptidase Enhancement of gut

matu-ration by dietary PL could be related with a higher

intra-cellular availability of PL for cell membrane and cell

organelles formation, as dietary PL promotes re-acylation

of digested lipids, increasing intracellular PL availability

for lipoprotein synthesis in gilthead seabream (Liu et al

2002; Caballero et al 2003)

In sea bass larvae, the response of phospholipase A2 to

dietary phospholipid content was gradual and showed a

great modulation range in expression Also, amylase and

alkaline phosphatase activities revealed a proper

matura-tion of the digestive tract in the larvae fed the highest

die-tary phospholipid levels (Cahu et al 2003)

On the basis of the few studies where single pure

phos-pholipid species have been used, the rank order for efficacy

appears to be phosphatidylcholine, phosphatidylinositol,

phosphatidylethanolamine and phosphatidylserine (Izquierdo

& Koven 2011) Several studies also suggested that the

phospholipid effect was not attributed to a general

enhanced emulsification and digestion of lipids The

evi-dence rather led to the hypothesis that early developing

stages of fish had impaired ability to transport dietary

lip-ids away from the intestine possibly through limitations in

lipoprotein synthesis Thus, dietary PL increases the

effi-ciency of transport of dietary fatty acids and lipids from

the gut to the rest of the body (Coutteau et al 1997;

Fon-tagne´ et al 1998; Salhi et al 1999; Izquierdo et al 2001)

However, despite the many studies available denoting theimportance of dietary PL, few of them have intended todetermine quantitative PL requirements testing diets with

at least five different levels of this nutrient Recently, krill

PL have been found to constitute a suitable phospholipidsource in diets for larval gilthead seabream (Betancor et al

2012) Thus, the aim of the present study was to determinethe optimum requirements of krill PL for gilthead seabream (Sparus aurata) larvae, and its influence on larvalproduction performance and digestive enzymes activity

Gilthead seabream larvae were obtained from naturalspawnings from Instituto Canario de Ciencias Marinas[Grupo de Investigacio´n en Acuicultura (GIA), Las Palmas

de Gran Canaria, Spain] Larvae (5.4 mm total length,

120lg dry body weight) previously fed rotifers (Brachinusplicatilis) enriched with DHA Protein Selco®(INVE, Dend-ermond, Belgium) until 16 dph were randomly distributed

in 15 experimental tanks at a density of 2100 larvae tank 1and fed one of the diets tested in triplicate All tanks(200 L fibreglass cylinder tanks with conical bottom andpainted a light grey colour) were supplied with filtered sea-water (37 g L 1 salinity) at an increasing rate of 0.4–1.0 L min 1to assure good water quality during the entiretrial Water entered from the tank bottom and exited fromthe top to ensure water renewal and maintain high waterquality, which was tested daily and no deterioration wasobserved Water was continuously aerated (125 mL min 1)attaining 6.1± 1 mg L 1dissolved O2 Average water tem-perature and pH along the trial were 19.1± 1 °C and 7.85,respectively Photoperiod was kept at 12 h light/12 h dark,

by fluorescent daylights, and the light intensity was kept at

1700 lux (digital Lux Tester YF-1065, Powertech Rentals,Osborne Park City, WA, Australia)

Five experimental microdiets (pellet size< 250 lm) withincreasing phospholipid contents were formulated usingsardine oil (Agramar S.A.,Las Palmas City, Spain) andkrill oil (Qrill, high PL, Aker BioMarine, Fjordalle´en, Nor-way) as sources of triglycerides and PL, respectively Theirformulation and proximate analysis are shown in Table 1

The fatty acid content is shown in Tables 2, 3 and 4 Thedesired lipid content (about 210 g kg 1 DW) was com-pleted if necessary with a non-essential fatty acid source,oleic acid (Oleic acid; Merck, Darmstadt, Germany) Themicrodiets were prepared by mixing squid powder andwater-soluble components, and then the lipids and fat-solu-ble vitamins and, finally, gelatine dissolved in warm water

.

The paste was compressed pelleted (Severin, Suderm,

Germany) and dried in an oven at 38°C for 24 h (Ako,

Barcelona, Spain) Pellets were ground (Braun, Kronberg,

Germany) and sieved (Filtra, Barcelona, Spain) to obtain a

particle size below 250lm Diets were prepared and

analy-sed for proximate and fatty acid composition (Tables 1 and

2) at GIA laboratories (Las Palmas de Gran Canaria, Spain)

Diets were manually supplied 14 times per day each

45 min from 9:00 to 19:00 for 16 days Non-enriched rotifers

were co-fed during days 16th and 17th (1 rotifer ml 1) To

assure feed availability, daily feed supplied was maintained

at 1.5 and 2.5 g tank 1during the first and second weeks of

feeding, respectively Larvae were observed under the

bin-ocular microscope to determine feed acceptance Before the

end of the experiment, an activity test was conducted by

han-dling 20 larvae tank 1out of the water in a scoop net for

1 min and subsequently allocating them in another tank

sup-plied with clean seawater and aeration, to determine survival

after 24 h Final survival was calculated by individually

counting all the living larvae at the beginning and at the end

of the experiment Growth was determined by measuring dry

body weight (105 °C 24 h) and total length (Profile Projector

V-12A Nikon, Tokyo, Japan) of 30 fish tank 1at the

begin-ning, in the middle and at the end of the trial In addition, at

the end of the trial and after 12 h of starvation, the all larvae

in each tank were washed with distilled water, sampled and

kept at 80°C for biochemical composition

Moisture (A.O.A.C 1995), protein (Kjeldhal) and crude

lipid (Folch et al 1957) contents of larvae and diets were

analysed Fatty acid methyl esters were obtained by

Table 2 Fatty acids (% dry weight) composition in total lipids of diets containing five PL levels

0 KPL

3 KPL

6 KPL

12 KPL

17.5 KPL

Table 1 Formulation and proximate composition of the

experi-mental microdiets containing several levels of marine PL

.

Aquaculture Nutrition 19; 449–460 ª 2012 John Wiley & Sons Ltd

Table 4 Fatty acids (% dry weight) composition in polar lipids of diets containing five PL levels

0 KPL

3 KPL

6 KPL

12 KPL

17.5 KPL

Table 3 Fatty acids (% dry weight) composition in neutral lipids

of diets containing five PL levels

0 KPL

3 KPL

6 KPL

12 KPL

17.5 KPL

transmethylation of crude lipids as described by Christie

(1982) Fatty acid methyl esters were separated by GLC

(GC-14A, Shimadzu, Tokyo, Japan) in a

Supercolvax-10-fused silica capillary column (length, 30 m; internal

diame-ter, 0.32 mm; Supelco, Bellefonte, PA, USA) using helium

as a carrier gas Column temperature was 180°C for the

first 10 min, increasing to 215°C at a rate of 2.5 °C per

min and then held at 215°C for 10 min Fatty acid methyl

esters were quantified by FID (GC-14A, Shimadzu, Tokyo,

Japan) following the conditions described in Izquierdo

et al (1990) and identified by comparison with previously

characterized standards and GLC-MS

For enzymes activity determination, the alkaline

phos-phatase, trypsin and lipase activities are expressed as

rela-tive fluorescence units (RFU), and PLA2 is expressed as

units (U) The larvae were homogenized by a sonicator

ultrasound (Misonix Microson XL2007 Ultrasonic

Homog-enizer) on ice in 110lL of high-purified water, centrifuged

and the supernatant used as the sample stock solution The

alkaline phosphatase activities were quantified by

fluoro-metric assay using a spectrofluorometer (Thermolab

Sys-tems, Helsinki, Finland) at excitation wavelengths of

358 nm and emission wavelengths of 455 nm After

prepa-ration of 140lL of reaction buffer at pH 10.4 (100 mM

Glycine, 1 mM MgCl2, 1 mM ZnCl2) and 50lL of

sub-strate stock solution (200lM of 6,8 Difluoro-4

methyllum-belliferyl phosphate (DiFMUP) in DMSO, the reaction

was started by adding 10lL of the sample stock solution,

and the kinetic curves were recorded for 20 min (Gee

et al.1999)

Trypsin activities were quantified by fluorometric assay

using spectrofluorometer at excitation wavelengths of

380 nm and emission wavelengths of 440 nm After

prepa-ration of 195lL of reaction buffer pH 8.0 (50 mM

Tris-HCl, 10 mM CaCl2) and 5lL of substrate stock solution

(20lM of Boc-Gln-Ala-Arg-7 amido-4 methylcoumarin

hydrochloride in DMSO), the reaction was started by

add-ing 10lL of the sample stock solution, and the kinetic

curves were recorded for 5 min (Rotllant et al 2008)

Neutral lipase activities were quantified by fluorometric

assay using spectrofluorometer at excitation wavelengths of

355 nm and emission wavelengths of 460 nm First,

247.4lL of 0.1 M phosphate reaction buffer pH 7.0

(19.5 mL of 0.2 M phosphate monobasic (NaH2PO4 1

H2O) solution and 30.5 mL of 0.2 M Phosphate dibasic

(NaH2PO4 7 H2O) solution were mixed and completed up

to 100 mL by adding 50 mL high-purified water), then

2.6lL of substrate stock solution [40 mM of

4-Methylum-belliferol butyrate (MUB) in N, N Dimethyl formamide

(DMSO) was added, and then reaction was started by ing 10lL of the sample stock solution] Kinetics curveswere recorded for 5 min (Rotllant et al 2008) The PLA2activities were quantified by fluorometric assay using spec-trofluorometer at excitation wavelengths of 377 nm andemission wavelengths of 450 nm First 160lL of reactionbuffer pH 8 (50 mM Tris-HCl, 100 mM NaCl, 2 mMNaN3, 5lg ml bovine serum albumin, and 10 lm 1-anili-nonaphthalene-8-sulphonate and 20lL of substrate stocksolution (50lL of 1,2-dimyristoyl-sn-glycero-3-phosphoch-oline solution in 40 mM, methanol mixed with 15lL de-oxycholic acid solution in 40 mM, methanol and quicklyinjected into 1 mL high-purified water, stirred for 1 minand sonicated for 2 min) and 10lL of 100 mM CaCl2solution were incubated at 25°C for 10 min The reactionwas started by adding 20lL of the sample stock solution,and the kinetic curves were recorded for 40 min (Huang

add-et al.2006)

All data were tested for normality and homogeneity ofvariances with Levene′s test, not requiring any transforma-tion and were treated using one-way ANOVA Means com-pared by Duncan’s test (P< 0.05) using a SPSS software(SPSS for Windows 11.5; SPSS Inc., Chicago, IL, USA)

All the experimental diets were well accepted by giltheadseabream larvae according to the microscopic observations.Generally, survival was very high for this type of studies,and it was significantly correlated to the increase in dietarymarine PL (Fig 1) Thus, average survivals of larvae feddiet 12 KPL (100 g kg 1dietary PL content) and diet 17.5KPL (110 g kg 1 dietary PL content) were significantlyhigher than the other treatments (Fig 1) The resistance tostress also increased by dietary PL levels, larvae fed the

y = 4.1683x + 25.737

R2 = 0.9085

10 15 20 25 30 35 40 45 50 55

b b

b

Figure 1 Survival rate (% of population) of larvae reared from 16

.

Aquaculture Nutrition 19; 449–460 ª 2012 John Wiley & Sons Ltd

17.5 KPL diet showing higher survival after the activity

test, although not significantly different from that of larvae

fed diets 12 KPL and 6 KPL (90 g kg 1 dietary PL

con-tent) (Fig 2) On the contrary, larvae fed diet 0 KPL

(60 g kg 1dietary PL content) had a marked drop in

sur-vival after stress (Fig 2)

Larval growth was also improved by dietary PL (Figs 3

& 4) and after 15 days of feeding, growth of larvae fed

17.5 KPL(110 g kg 1 dietary PL content) diet had

signifi-cantly higher total length than those fed 0 KPL, 3 KPL

and 6 KPL, but did not differ from those fed 12 KPL

(100 g kg 1 dietary PL content) (Fig 3) In terms of body

weight, larvae fed 17.5 KPL were bigger than larvae fed 0

KPL and 3 KPL but did not differ significantly from larvae

fed 6 KPL or 12 KPL diets (Fig 4)

In general, digestive enzyme activity was increased by the

elevation of dietary marine PL Thus, alkaline phosphatase

was lowest in 0 KPL larvae and significantly increased with

the elevation of dietary PL up to 90 g kg 1 (diet 6 KPL),

whereas trypsin activity was significantly lowest in both 0

KPL and 3 KPL larvae (Figs 5 & 6) Neutral lipase activitywas lowest in 0 KPL and 3 KPL larvae and significantlyhigher in 12 KPL and 17.5 KPL larvae (Fig 7) PLA2activity was significantly higher in larvae fed 17.5 KPLthan in larvae fed 0 KPL, 3 KPL or 6 KPL, but did notdiffer from that of 12 KPL larvae (Fig 8)

Fatty acid composition of total lipids of the microdietsshowed that gradual inclusion of PL lead to an increase inn-3 fatty acids, particularly due to the increase in EPA andDHA, together with 14:0 On the contrary, oleic acid andn-6 fatty acids, particularly 18:2n-6, were reduced by theinclusion of dietary PL (Table 2) However, fatty acid com-position of PL from microdiets was much more stable andonly showed a slight reduction in 20:1n-9+ n-7 andincrease in EPA (Tables 3 and 4)

Analysis of fatty acids composition of larvae showed nificant differences between the five dietary PL treatments

sig-in both neutral and polar lipids (Tables 5 and 6) Neutral

Figure 2 Survival 24 h after activity test of larvae (31 dph) fed five

Figure 3 Total length of larvae (31 dph) fed five dietary PL levels.

y = 48.304x + 403.08

R2 = 0.8738

200 300 400 500 600 700 800

Figure 4 Dry weight of larvae (31 dph) fed five dietary PL levels.

y = –3.0666x 2 + 67.183x – 67026

R2 = 0.9982

150 200 250 300 350

Figure 5 Alkaline phosphatase activity in sea bream larvae

.

lipids were more markedly affected by dietary lipids and

showed increased myrisitc acid (14:0), 18:1n-7, 18:4n-3, n-3

fatty acids, particularly EPA, and reduced stearic (18:0),

oleic (18:1n-9), 20:1n-9 and linoleic (18:2n-6) acids with

increased dietary PL Thus, 17.5 KPL larvae were

signifi-cantly highest in EPA, while the 0 KPL has the lowest

Larvae fed diets 12 KPL and 17.5 KPL diets showed

higher n-3/n-6 and EPA/ARA ratios than 0 KPL, 3 KPL

and 6 KPL larvae Larval polar lipids were more

conserva-tive than the neutral lipids and only showed a marked

increased in n-3 fatty acids and EPA, together with a

reduction in n-6 fatty acids because of the lower linoleic

acid contents On the contrary, regardless dietary levels,

ARA and DHA contents were not significantly different

among larvae fed different levels of dietary PL Larvae fed

12 KPL and 17.5 KPL diets showed the highest total n-3,

n-3HUFA and EPA content of polar lipids

Complete replacement of both rotifers and Artemia withthe experimental microdiets for gilthead seabream larvaeproduced high survival and growth rates, particularly infish fed the two highest levels of krill PL High survivaland fast growth rates are required to be able to determineoptimum ingredient contents and nutrient requirements infish diets In the present study, final average survival (48%)and total length (8.6 mm) in seabream fed diets containing

120 g kg 1 and 175 g kg 1 krill PL was higher than thebest obtained by other authors (30-38% survival) feedingonly microdiets for a similar period in this (Seiliez et al.2006) or other species (Cahu et al 1998; Zambonino-Infante & Cahu 1999) Moreover, it was even higher thanthat obtained feeding live preys under similar experimentalconditions (41%) (Salhi et al 1997) or in commercialhatcheries (about 20%) (Roo et al 2005) However, amuch higher survival (73%) was obtained by Cahu et al.(2003) feeding high PL microdiets to European sea basslarvae (Dicentrarchus labrax), a species with a faster devel-opment of the digestive system than gilthead seabream.Fatty acid composition in larvae reflected that of corre-sponding diet, where the supplementation of dietary PL led

to an increase in larval n-3 HUFA fatty acids Best results

of growth and survival in 12 KPL and 17.5 KPL larvaecould hence be due to higher levels of PUFA, which areessential for marine fish larvae, particularly EPA andDHA, which are increased with Krill PL inclusion in thediets An increase in both dietary DHA and EPA improveslarval performance, in terms of survival, stress resistanceand growth (Liu et al 2002; Izquierdo et al 2005) More-over, a higher n-3/n-6 fatty acid ratio was observed in the

12 KPL and 17.5 KPL diets, which resulted in the bestgrowth and survival in those groups This was alsoobserved by Izquierdo et al 2000, 2001); , Caballero et al

bc

Figure 7 Lipase activity in sea bream larvae (31 dph) fed five

y = –3.8295x 2 + 87.92x – 338.91

R2 = 09227

0 50

b b

Figure 6 Trypsin activity in sea bream larvae (31 dph) fed five

y = 0.7022x 2 – 6.2453x + 60.837

R2 = 0.8976

30 40 50 60 70 80 90 100

Figure 8 Phospholipase A2 activity in sea bream larvae (31 dph)

.

Aquaculture Nutrition 19; 449–460 ª 2012 John Wiley & Sons Ltd

Table 5 Fatty acids (% total identified fatty acids) composition in neutral lipids of larvae fed diets containing five PL levels

Table 6 Fatty acids (% total identified fatty acids) composition in polar lipids of larvae fed diets containing five PL levels

(2002, 2006a) who showed that a well-balanced n-3/n-6

fatty acid ratio in the diet improved lipid metabolism such

as digestion, absorption and transport

Despite the large amount of bibliography regarding the

effect of dietary PL, few studies have been conducted to

determine optimum dietary levels with a wide range ( 5)

of dietary PL contents In the present study, increase in

die-tary krill PL up to 120 g kg 1(about 100 g kg 1 total PL

in diet) significantly improved larval survival and promoted

growth, leading to high final total length and body weight,

whereas further increase did not affect those parameters

These levels are lower than those recommended for

Euro-pean sea bass (120 g kg 1 total PL) (Cahu et al 2003) for

maximum survival and growth when soybean lecithin was

used as a source of PL These results can be related to a

higher effectiveness of marine PL in the present study in

comparison with soybean lecithin (Salhi et al 1999;

Iz-quierdo et al 2001; Wold et al 2007) or to different

requirements among species, as they are higher than those

recommended for larval rock bream (Oplegnathus fasciatus)

(50 g kg 1 total PL) (Kanazawa et al 1983a) Larval

growth and survival have been found to be increased by

die-tary PL in several fish species (Kanazawa et al 1983b;

Iz-quierdo et al 2001; Cahu et al 2003; Gisbert et al 2005)

On the contrary, survival was not improved by elevation of

dietary PL in other species such as pikeperch (Sander

luciop-erca) larvae (Hamza et al 2008) The promoting effect of

PL on larval growth is probably related to their importance

as components of biomembranes (Tocher 2003), and it has

been mainly associated with their content in PC (Takeuchi

et al.1992; Kanazawa 1993), the main PL class in cell

mem-brane In relation to improved growth, other authors also

found an increase in feed intake that has been associated to

the stimulation of gustatory response by the trimethyl group

of the choline base of PC (Izquierdo & Koven 2011)

How-ever, in the present study, no differences were found in feed

intake, and all diets were supplemented with the same

amount of choline in the vitamin mixture

The growth improvement obtained with increased dietary

PL contents could be at least partly related to the higher

maturation of the gut suggested by the significant increase in

alkaline phosphatase activity This enzyme, an important

component of brush border membrane enzymes, increases

during development of marine fish larvae in association with

enterocyte development and denoting intestine maturation

In the present study, increased dietary krill PL up to

120 g kg 1 (about 100 g kg 1 total PL) significantly

increased alkaline phosphatase activity, whereas higher PL

levels did not further enhanced it Similar PL values

(90 g kg ) were required to increase alkaline phosphataseactivity in European seabass when soybean lecithin was used

as the PL source (Cahu et al 2003) and cod larvae (Wold

et al.2007) Enhanced enterocyte maturation by PL could berelated with to a higher intracellular availability of PL for cellmembrane and cell organelles formation, as dietary PL pro-motes re-acylation of digested lipids increasing intracellular

PL availability for lipoprotein synthesis in gilthead seabream(Liu et al 2002; Caballero et al 2003) In fact, membranecell organelles such as mitochondria have been found to beaffected by dietary PL (MacQueenLeifson et al 2003)

On the contrary to the saturation effect of dietary PLlevels on alkaline phosphatase activity, there was a linearsubstrate stimulatory effect of dietary PL on PLA2 activity

PLA2 activity has been found in larvae of several speciessuch as striped bass (Morone saxatilis) (Ozkizilcik et al

1996) or European seabass (D labrax) (Cahu & no-Infante 2001) The steady increase in PLA2 activity bydietary PL is in agreement with the transcriptional regula-tion of PLA2 by dietary PL found in European seabass(Zambonino-Infante & Cahu 1999) These authors alsosuggested a posttranscriptional regulation of endocrine ori-gin In concordance with this hypothesis, in the presentstudy, the increase trypsin and lipase activity by elevation

Zamboni-of krill dietary contents up to 60 g kg 1, regardless dietaryprotein and lipid content, denotes a general effect of die-tary PL on pancreatic enzymes Endocrine modulation ofthe pancreatic digestive function in fish is regulated bycholecystokinine (CCK) whose secretion in turn is stimu-lated by the presence of several nutrients such as freeamino acids (Rojas-Garcı´a & Rønnestad 2002) However,free fatty acids, and particularly n-3 polyunsaturated fattyacids, are also potent stimulators of CCK secretion (Little

et al.2007) by mediation of the G-protein-coupled receptorGPR40 (Liou et al 2011) In the present study, increasedPLA2 activity (as well as lipase activity) by elevation ofdietary PL would be responsible for a more effective diges-tion of dietary lipids and, therefore, higher concentrations

of free polyunsaturated fatty acids in intestine lumen thatcould stimulate CCK secretion as it occurs in mammals

Moreover, increase in dietary PL increased the rated fatty acid content in biomembranes of larval tissues,

polyunsatu-as denoted by the fatty acid composition of larval PL,could modulate CCK function responsible for the higherpancreatic enzymes activity In mammals, fatty acid com-position of cell membranes has been found to be determi-nant for specific recognition of CCK receptor (Romano

et al.2003) and modulation of CCK function in pancreatictissue (Chang et al 1984) In the present study, increasing

.

dietary PL lead to better assimilation of n-3HUFA

espe-cially EPA, reflected in the higher content of these fatty

acids in both neutral and polar lipids of the larvae This

could be partly due to a better digestion efficiency of

die-tary lipids by enhanced PLA2 and lipase activity, but also

to an improvement in lipid transport Izquierdo et al

(2001) found a better incorporation of fatty acids from

die-tary polar lipids in larval seabream fed diedie-tary PL in

com-parison with larvae fed triglycerides

In summary, the results of the present study have shown

that dietary krill PL are an excellent source of lipids for

gilthead seabream larvae Optimum inclusion levels of this

ingredient in microdiets to completely substitute live preys

at this larval age were found to be 120 g kg 1 krill PL,

providing about 100 g kg 1 total PL Lower levels of this

ingredient markedly reduced culture performance, gut

development, enzymes activity, dietary lipid utilization,

growth, survival and stress resistance in sea bream larvae

Further studies are required to determine the optimum

content of other phospholipid sources such as soybean

leci-thin in early weaning diets for gilthead seabream

This study was partially supported by a grant from the

Spanish Agency of International Cooperation and

Devel-opment (AECID) to Reda Saleh Mohamed Ibrahim The

present study was funded by the Spanish Ministry of

Sci-ences and Education (AGL2009-14661) This work has

been partly funded under the EU seventh Framework

Pro-gramme by the ARRAINA project N288925: Advanced

Research Initiatives for Nutrition & Aquaculture

A.O.A.C (1995) Official Method of Analysis Association Official

Analytical Chemistry, Washington, DC, 1018 pp.

Betancor, M.B., Nordrum, S., Atalah, E., Caballero, M.J.,

Benı´-tez-Santana, T., Roo, J., Robaina, L & Izquierdo, M.S (2012)

Potential of three new krill products for seabream larval

Caballero, M.J., Obach, A., Rosenlund, G., Montero, D., Gisvold,

M & Izquierdo, M.S (2002) Impact of different dietary lipid

sources on growth, lipid digestibility, tissue fatty acid

composi-tion and histology of rainbow trout, Oncorhynchus mykiss

Caballero, M.J., Izquierdo, M.S., Kjørsvik, E., Montero, D.,

So-corro, J., Ferna´ndez, A.J & Rosenlund, G (2003)

Morphologi-cal aspects of intestinal cells from gilthead seabream (Sparus

aurata) fed diets containing different lipid sources Aquaculture,

225, 325–340.

Caballero, M.J., Gallardo, G., Robaina, L., Montero, D.,

Fernan-dez, A & Izquierdo, M.S (2006a) Vegetable lipid sources affect

Caballero, M.J., Torstensen, B.E., Robaina, L., Montero, D & quierdo, M.S (2006b) Vegetable oils affect the composition of lipo-

Cahu, C & Zambonino-Infante, J (2001) Substitution of live food

180.

Cahu, C.L., Zambonino-Infante, J.L., Escaffre, A.M., Bergot, P & Kaushik, S (1998) Preliminary results on sea bass Dicentrarchus

–7.

Cahu, C.L., Zambonino-Infante, J.L & Barbosa, V (2003) Effect

of dietary phospholipid level and phospholipid:neutral lipid value on the development of sea bass (Dicentrarchuslabrax) lar-

Chang, R.S.L., Lotti, V.J & Chen, T.B (1984) Cholecystokinin receptor mediated hydrolysis of inositol phospholipids in guinea

Christie, W.W (1982) Lipid Analysis Pergamon, Oxford.

Cousin, J.C.B., Baudin-Laurencin, F & Gabaudan, J (1987) Ontogeny of enzymatic activities in fed and fasting turbot,

Coutteau, P., Geurden, I., Camara, M.R., Bergot, P & Sorgeloos,

P (1997) Review on the dietary effects of phospholipids in fish

Folch, J., Lees, M & Stanley, G.H.S (1957) A simple method for the isolation and purification of total lipids from animal tissues.

Fontagne´, S., Geurden, I., Escaffre, A.M & Bergot, P (1998) tological changes induced by dietary phospholipids in intestine and liver of common carp (Cyprinus carpio L.) larvae Aquacul-

Gee, K.R., Sun, W.-C., Bhalgat Mahesh, K., Upson Rosalyn, H., Klaubert Dieter, H., Latham Katherine, A & Haugland Richard, P (1999) Fluorogenic substrates based on fluorinated umbelliferones for continuous assays of phosphatases and b-

Gisbert, E., Villeneuve, L., Zambonino-Infante, J.L., Quazuguel,

P & Cahu, C.L (2005) Dietary phospholipids are more efficient than neutral lipids for long chain polyunsaturated fatty acid sup- ply in European sea bass Dicentrarchus labrax development Lip-

Hadas, E., Koven, W.M., Sklan, D & Tandler, A (2003) The effect of dietary phosphatidylcholine on the assimilation and dis- tribution of ingested free oleic acid (18:1n-9) in gilthead sea-

Hamza, N., Mhetli, M., Khemis, I.B., Cahu, C & Kestemont, P (2008) Effect of dietary phospholipid levels on performance, enzyme activities and fatty acid composition of pikeperch (San-

Huang, C., Lu, Z., Ying, L & Luhua, L (2006) A continuous orescence assay for phospholipase A2 with nontagged lipid.

Izquierdo, M.S & Koven, W.M (2011) Lipids In: Larval Fish

and Sons Publisher Editor, Oxford, United Kingdom.

Izquierdo, M.S., Socorro, J., Arantzamendi, L & Herna´ndez-Cruz, C.M (2000) Recent advances in lipid nutrition in fish larvae.

Izquierdo, M.S., Watanabe, T., Takeuchi, T., Arakawa, T & Kitajima, C (1990) Optimum EFA levels in Artemia to meet the

.

Aquaculture Nutrition 19; 449–460 ª 2012 John Wiley & Sons Ltd

EFA requirements of red sea bream (Pagrus major) In: The

Cur-rent Status of Fish Nutrition in Aquaculture (Takeda, M &

Tokyo.

Izquierdo, M.S., Tandler, A., Salhi, M & Kolkovski, S (2001)

Influence of dietary polar lipids quantity and quality on

inges-tion and assimilainges-tion of labelled fatty acids by larval gilthead

Izquierdo, M.S., Montero, D., Robaina, L., Caballero, M.J.,

Rosenlund, G & Gine´s, R (2005) Alterations in fillet fatty acid

profile and flesh quality in gilthead seabrem (Sparus aurata) fed

vegetable oils for a long term period Recovery of fatty acid

Kanazawa, A (1993) Essential phospholipids of fish and

crusta-ceans In: Fish Nutrition in Practice (Kausbik, S.J & Luquet,

Kanazawa, A., Teshima, S., Inamori, S & Matsubara, H (1983a)

Effect of dietary phospholipids on growth of the larval red sea

Kanazawa, A., Teshima, S., Kobayashi, T., Takae, M., Iwashita,

T & Uehara, R (1983b) Necessity of dietary phospholipids for

growth of the larval ayu Mem Fac Fish Kagoshima University,

32, 115–120.

Kjørsvik, E (2009) Phospholipids vs neutral lipids: effects on

digestive enzymes in Atlantic cod (Gadus morhua) larvae

Liou, A.P., Lu, X., Sei, Y., Zhao, X., Pechhold, S., Carrero, R.J.,

Raybould, H.E & Wank, S (2011) The G-Protein-Coupled

Receptor GPR40 directly mediates long chain fatty acid induced

Little, T.J., Russo, A., Horowitz, M., Meyer, J.H., Smyth, D., Jones,

K.L., Wishart, J., Bellon, M & Feinle, C (2007) The effects of free

fatty acids on gastric emptying, plasma cholecystokinin (CCK)

and peptide YY (PYY), appetite and energy intake in humans are

Liu, J., Caballero, M.J., El-Sayed Ali, T., Izquierdo, M.S.,

Her-nandez Cruz, C.M., Valencia, A & FerHer-nandez-Palacios, H.

(2002) Effect of dietary lecithin and eicosapentaenoic acid on

growth, survival, stress resistance and lipid transport in gilthead

MacQueenLeifson, R., Homme, J.M., Lie, Ø., Myklebust, R &

Strøm, T (2003) Three different lipid sources in formulated start

feeds for turbot (Scophthalmus maximus, L.) larvae, effect on

growth and mitochondrial alteration in enterocytes Aquaculture,

1, 33–42.

Ozkizilcik, S., Chu Fu-Lin, E & Place, A.-R (1996) Ontogenetic

changes of lipolytic enzymes in striped bass (Morone saxatilis).

Rainuzzo, J.R., Reitan, K.I & Olsen, Y (1997) The significance of

Ribeiro, L., Zambonino-Infante, J.L., Cahu, C & Dinis, M.T.

(1999) Development of digestive enzymes in larvae of Solea

Rojas-Garcı´a, C.R & Rønnestad, I (2002) Cholecystokinin and tryptic activity in the gut and body of developing Atlantic hali- but larvae: evidence for participation in the regulation of protein

Romano, R., Bayer, T.M & Moroder, L (2003) Lipophilic tization and its effect on the interaction of cholecystokinin (CCK) nonapeptide with phospholipids Biochim Biophys Acta,

deriva-11, 1–119.

Roo, J., Herna´ndez-Cruz, C.M., Ferna´ndez-Palacios, H & quierdo, M.S (2005) Development of skeletal deformities in gilt- head seabream (Sparus aurata) reared under different larval cultural and dietary conditions In: Larvi’05-Fish and Shellfish Larviculture Symposium (Hendry, C.I., Van Stappen, G., Wille,

Pub 36 Oostende, Belgium.

Rotllant, G., Moyano, F.J., Andre´s, M., Dı´az, M., Este´vez, A &

Gisbert, E (2008) Evaluation of fluorogenic substrates in the assessment of digestive enzymes in a decapod crustacean Ma-

Salhi, M., Izquierdo, M.S., Herna´ndez-Cruz, C.M., Socorro, J &

Ferna´ndez-Palacios, H (1997) The improved incorporation of polyunsaturated fatty acids and changes in liver structure in

–879.

Salhi, M., Herna´ndez-Cruz, C.M., Bessonart, M., Izquierdo, M.S.

& Ferna´ndez-Palacios, H (1999) Effect of different dietary polar lipid levels and different n-3 HUFA content in polar lipids on gut and liver histological structure of gilthead seabream (Sparus

Seiliez, I., Bruant, S., Zamboino-Infante, J.L., Kaushik, S & got, P (2006) Effect of dietary phospholipid level on the devel- opment of gilthead seabream (Sparus aurata) larvae fed a

Takeuchi, T., Arakawa, T., Satoh, S & Watanabe, T (1992) plemental effect of phospholipids and requirement of eicosapen- taenoic acid and docosahexaenoic acid of juvenile striped jack.

Tocher, D.R (2003) Metabolism and functions of lipids and fatty

Van Der Meeren, T., Olsen, R.E., Hamre, K & Fyhn, H.J (2008) Biochemical composition of copepods for evaluation of feed

Walford, J & Lam, T.J (1993) Development of digestive tract and proteolytic enzyme activity in sea bass Lates calcarifer larvae

Wold, P.A., Hoehne-Reitan, K., Cahu, C.L., ZamboninoInfante, J.L., Rainuzzo, J & Kjørsvik, E (2007) Phospholipids vs neu- tral lipids: effects on digestive enzymes in Atlantic cod (Gadus

Zambonino-Infante, J.L & Cahu, C.L (1999) High dietary lipid levels enhance digestive tract maturation and improve

1200.

.

1,2 1 1 1 2 2 1

The Key Laboratory of Mariculture, Ministry of Education, Ocean University of China, Qingdao, China; 2 GuangdongYuehai Feed Group Co Ltd., Zhanjiang, China

A 10-week feeding trial was conducted to estimate the

opti-mum dietary manganese requirement for juvenile cobia,

Rachycentron canadumL The basal diet was formulated to

contain 501 g kg
1crude protein from vitamin-free casein,

gelatin and fish protein concentrate Manganese sulphate

was added to the basal diet at 0 (control group), 6, 12, 18,

24 and 36 mg Mn kg
1 diet providing 5.98, 7.23, 16.05,

23.87, 28.87 and 41.29 mg Mn kg
1diet, respectively Each

diet was randomly fed to three replicate groups of cobia

for 10 weeks, and each tank was stocked with 30 fish

(ini-tial weight, 6.27 ± 0.03 g) The manganese concentration in

rearing water was monitored during the feeding period and

was< 0.01 mg L
1 Dietary manganese level significantly

influenced survival ratio (SR), specific growth ratio (SGR),

feed efficiency ratio (FER) and the manganese

concentra-tions in the whole body, vertebra and liver of cobia When

the dietary manganese level rose from 5.98 mg kg
1 to

23.87 mg kg
1, the superoxide dismutase (SOD; EC 1.15

1.1) activities in liver also increased (P< 0.05) But there

was no significant change in SOD activities for the groups

fed with diets containing manganese level higher than

23.87 mg kg
1 On the basis of broken-line regression of

SGR, manganese concentration in whole body and vertebra

the manganese requirements of juvenile cobia were

21.72 mg kg
1, 22.38 mg kg
1 and 24.93 mg kg
1 diet in

the form of manganese sulphate, respectively

KEY WORDS: fish nutrition, growth, manganese,

Rachycen-tron canadumL., requirement

Received 21 March 2012; accepted 12 June 2012

Correspondence: Quinghui Ai, The Key Laboratory of Mariculture

(Education Ministry of China), Ocean University of China, Qingdao

266003, China.

E-mail: qhai@ouc.edu.cn

Cobia is a carnivorous pelagic fish that can grow fromfingerling to 4–6 kg in 1 year in offshore net cage systems(Liao et al 2004) The white flesh of this fish, which is suit-able for sashimi consumption, is highly prized (Chou et al.2004) Excellent flesh quality, rapid growth and adaptabil-ity to culture conditions confer highly desirable characteris-tics for global commercial aquaculture on cobia (Holt

et al 2007) Following successful aquaculture development

in Taiwan of China (Liao et al 2004), cobia is extensivelyfarmed in cages in China, Vietnam and Philippines.Recently, its production has been initiated in EU, Brazil,etc Among the technical limitations in global cobia farm-ing, development of sustainable, high-quality feeds forcobia is one of the major objectives identified by Interna-tional Initiative of Sustainable and Biosecure Aquafarming,established in 2005 to accelerate commercial viability ofcobia culture through international collaborations (Holt

et al 2007) The nutritional value of several plant proteinsources has been evaluated for potential use in cobia for-mulated feeds (Chou et al 2004; Lunger et al 2006) Die-tary requirements for some nutriments including crudeprotein (Chou et al 2001; Craig et al 2006), lipid (Wang

et al 2005), methionine (Zhou et al 2006), lysine (Zhou

et al 2007) and choline (Mai et al 2009) have beenreported However, the dietary requirements for mineralshave only been reported for a few elements, such as zinc,iron, copper (Qiao 2007) and selenium (Liu et al 2010).Manganese is known to be an essential trace element forgrowth, reproduction and prevention of skeletal abnormali-ties in terrestrial animals and fish It serves as a cofactor formany enzymes (e.g glycosyl transferases) and as an integralconstituent of certain metalloenzymes such as arginase, pyru-vate carboxylase and Mn superoxide dismutase Mn defi-ciency has been shown to produce impaired growth, skeletalabnormalities and reduced reproduction together with

defects in lipid and carbohydrate metabolism in mammals

and chicken (Lall 2002; Leach 1976)

The requirement of Mn has been quantified in some fish

species, such as common carp, Cyprinus carpio L and

rain-bow trout, Salmo gairdneri (Ogino & Yang 1980), channel

catfish, Ictalurus punctatus fingerling (Gatlin & Wilson

1984), Atlantic salmon, Salmo salar (Maage et al 2000),

juvenile gible carp, Carassius auratus gibelio (Pan et al

2008), juvenile tilapia, Oreochromis niloticus9 O aureus

(Lin et al 2008), grass carp, Ctenopharyngodon idella

fin-gerling (Wang & Zhao 1994) and grouper, Epinephelus

mal-abaricus(Ye et al 2008) at levels of 12–13, 12–13, 2.4, 7.5,

13.77, 7, 15 and 19 mg kg
1diet, respectively The optimum

dietary Mn requirement for juvenile cobia, however, has

not been reported Therefore, this study was designed to

determine the requirement of dietary Mn in the form of

manganese sulphate for juvenile cobia, to determine the

optimal Mn levels in commercial diets

The basal diet, using casein, gelatin and fish protein

concen-trate as protein sources, fish oil and lecithin as lipid sources,

was formulated to contain 501 g kg
1 crude protein and

112 g kg
1crude lipid (Table 1), which satisfied the protein

and lipid requirements of this fish (Chou et al 2001; Craig

et al.2006) The experimental diets were formulated with a

manganese-free mineral premix Graded levels (0.0, 6.0, 12.0,

18.0, 24.0 and 36.0 mg manganese (Mn) kg
1diet) of

man-ganese sulphate (purity 990 g kg
1; Shanghai Reagent

Corp., Shanghai, China) were supplemented to the basal

diet The actual levels of dietary manganese in experimental

diet, analysed by ICP atomic emission spectrophotometry

(Vista-mpx, Varian, USA), were 5.98, 7.23, 16.05, 23.87,

28.87 and 41.29 mg Mn kg
1, respectively (Table 1)

Ingre-dients were ground to a fine powder and sieved through a

246-lm screen All ingredients were thoroughly mixed with

menhaden fish oil, and deionized water was added to

pro-duce stiff dough The dough was then pelleted with an

exper-imental feed mill and dried for 24 h in a ventilated oven at

38 °C The diets were then broken up and sieved into proper

pellet size (2.59 5.0 mm) and stored at
20 °C until used

The experimental fish were obtained from a commercial

farm in Sanya, Hainan, China The cobia was reared in

flow-through plastic tanks Initially, all fish were daily fedthe basal diet to apparent satiation twice (08:00 and17:00 h) After 21 days, all fish readily accepted the starterdiets, and then were converted to the experimental diets

Water flow rate was maintained at approximately

2 L min
1 to maintain optimal water quality throughoutthe study Before commencing the feeding trial, fish werefasted for 24 h, and then weighed after being anesthetizedwith eugenol (1 : 10 000) (Shanghai Reagent Corp., Shang-hai, China) Fish of similar size (6.27± 0.03 g) were dis-tributed into 18 tanks at a density of 30 fish per tank Eachexperimental diet was randomly assigned to three tanks

Fish were hand-fed to apparent satiation twice (08:00 and17:00 h) daily for 10 weeks The remaining feed and faeceswere removed by siphoning immediately after feeding Dur-ing the trial period, water temperature ranged from 27.5 to

31°C and the salinity of seawater ranged from 24 to

26 g L
1 Air stone in each tank maintained dissolved gen concentration was 7 mg L
1or more Fish were rearedunder natural light The manganese concentration in

oxy-Table 1 Formulation and proximate composition of the basal diet

10 mg; inositol, 800 mg; pantothenic acid, 60 mg; folic acid,

20 mg; niacin acid, 200 mg; biotin, 1.20 mg; retinol acetate,

32 mg; cholecalciferol, 5 mg; alpha-tocopherol, 120 mg; quin, 150 mg; microcrystalline cellulose, 135 117 mg].

ethoxy-3

cellu-lose, 30.4 g].

.

rearing water was monitored during the feeding period and

was< 0.01 mg L
1 Cobia finished their ration within

1–2 min after feeding, thus manganese sulphate leached off

into water was very low and negligible

At the termination of the feeding trial, five fish randomly

selected from the sampled fish of each tank were used to

remove liver and vertebra Livers were obtained by

surgi-cally After heating the fish in a microwave oven for 50 s,

vertebrae were easily removed from fish, then lightly

scrubbed, and finally washed with distilled water to

remove flesh The vertebrae were dried for 2 h at 100°C,

ether extracted in a Soxhlet apparatus for 12 h (AOAC

1995) to remove lipid, and dried again (Mai et al 2006)

The vertebra and a part of liver samples were used for

manganese analysis by ICP atomic emission spectrometer

(Vista-mpx, Varian, USA), and the rest part of liver

sam-ples was used for the superoxide dismutase (SOD)

activi-ties analysis by the method of Knox et al (1981) The

rest of the trial samples were pooled for individual

proxi-mate composition analysis Fish body and diet

composi-tion were performed by standard methods (AOAC 1995)

Dry matter was determined by drying at 105 °C for 24 h,

crude protein by the Kjeldahl method, crude fat after

extraction with ether by the Soxhlet method and ash by

combustion at 550°C

The following variables were calculated:

Survival ratioðSR; %Þ ¼ 100

final number of fish/initial number of fish

Feed efficiency ratio (FER)= wet weight gain g/dry feed

fed g (Hardy & Barrows 2002)

Specific growth ratioðSGR; %Þ ¼ ½ðLn final weight - Ln

initial weight) /Rearing period (days)  100

All data from the feeding trial were subjected to Levene’s

test of equality of error variances and one-way ANOVA

fol-lowed by Tukey’s test using SPSS® (SPSS, Inc, Chicago,

IL, USA) All treatment effects were considered significant

at a P value of 0.05 or less Response indices that were

sig-nificantly influenced by dietary manganese level also were

subjected to linear regression analysis against dietary

man-ganese Broken-line regression analysis was performed on

SGR, manganese concentrations in whole body and

verte-bra to establish the dietary requirement for manganese(Robbins et al 1979) The equation used in the model is

Y¼ L þ UðR
XLRÞwhere Y is the parameter (SGR, manganese concentra-tion in liver or vertebra) chosen to estimate the require-ment, L is the ordinate and R is the abscissa of thebreakpoint R is taken as the estimated requirement XLRmeans X< R, and U is the slope of the line for XLR Bydefinition R
XLR= 0 when X > R

In this study, both SGR and FER showed a significantlypositive correlation with dietary manganese levels below23.87 mg kg
1and reached a plateau when dietary manga-nese levels were higher than 23.87 mg kg
1 (Table 2) Theminimum dietary requirement for manganese is estimated

to be 21.72 mg kg
1 by broken-line regression analysis onthe basis of SGR for juvenile cobia under the experimentalconditions (Fig 1) Fish fed the basal diet had significantlylower SGR compared to those fed the other experimentaldiets (P< 0.05) After the 10-week feeding trial, survivalratio of cobia fed the basal diet averaged 79.97%, whichwas lower than those fed the manganese-supplementeddiets (P< 0.05) Similarly, FER of cobia fed the basal dietwere significantly lower than those fed diets containingadded manganese (P< 0.05)

The carcass crude protein content of cobia increased (from160.7 g kg
1 to 171.2 g kg
1) with an increase in dietarymanganese level from 5.98 to 28.87 mg kg
1(P< 0.05) andthen slightly decreased for the groups fed diets containinghigher levels of manganese The crude lipid content ofcobia also increased (from 46.9 g kg
1to 60.9 g kg
1) with

an increase in dietary manganese level from 5.98 to23.87 mg kg
1(P < 0.05) The carcass ash content of cobiafed basal diets was higher than other groups (Table 3)

The manganese concentration in cobia fresh liver was gressively increased with increasing concentration of dietarymanganese within the range of 0.76–3.92 mg kg
1

pro-Aquaculture Nutrition 19; 461–467 ª 2012 John Wiley & Sons Ltd

(Table 4) The manganese concentration of vertebrae and

whole body also significantly increased with increasing

die-tary manganese (P< 0.05) On the basis of linear

regres-sion of whole-body manganese concentration (P< 0.05,

R2= 0.896), a minimum dietary requirement for manganese

in the form of manganese sulphate was estimated to be

22.38 mg kg
1 diet (Fig 2) Similarly, manganese

concen-tration in cobia vertebra reached a plateau when dietary

manganese was higher than 24.93 mg kg
1diet (Fig 3)

A significant increase in total SOD activity has been

observed in the liver of cobia fed diets containing with

5.98, 7.23 and 16.05 mg Mn kg
1, and the total SOD

activity reached a plateau in fish fed diets with 23.87, 28.87

and 41.29 mg Mn kg
1 Cu–Zn SOD and Mn SOD

activi-ties in cobia liver showed the same trend and has the

low-est enzyme activity in fish fed diet with 5.98 mg Mn kg
1(Table 5)

Fish readily accepted the experimental diet from the ning of the experiment and maintained normal behaviourthroughout the experimental period In this study, thegrowth response of juvenile cobia was significantly affected

begin-by the supplementation of dietary manganese, and a tive relationship was found between the growth and thedietary manganese levels Weight gain was lower in fish fedthe basal diet owing to insufficient manganese This result

posi-is similar to the reports on rainbow trout (Ogino & Yang1980), grass carp fingerling (Wang & Zhao 1994) and juve-nile gible carp (Pan et al 2008) Fish fed insufficient man-ganese diet had typical manganese-deficient symptoms such

as cataracts and skeletal abnormalities (dwarfism) in somestudies (Lall 2002) In this study, cobia did not have thesetypical manganese-deficient symptoms The manganese-deficient symptom has also not been observed in study ofAtlantic salmon (Lorentzen et al 1996), grouper (Ye et al

2008) and tilapia (Lin et al 2008)

In the present study, broken-line analysis was employed

to establish the relationship between SGR and dietarymanganese Based on the SGR, the minimum requirement

of dietary manganese for the optimal growth of juvenilecobia was 21.72 mg Mn kg
1(Fig 1) This result is muchhigher than that reports for channel catfish fingerling(2.4 mg Mn kg
1, Gatlin & Wilson 1984) Atlantic salmon(7.5 mg Mn kg
1, Maage et al 2000), juvenile giblecarp (13.77 mg Mn kg
1, Pan et al 2008), juvenile tilapia(7 mg Mn kg
1,Lin et al 2008), grass carp fingerling(15 mg Mn kg
1, Wang & Zhao 1994) common carp and

Figure 1 Relationship between dietary manganese level and SGR

of cobia fed the six diets for10 weeks Each point represents the

mean of three groups of fish within a treatment with 6 fish per

group Requirements derived with the broken-line method for

Table 2 Effects of different dietary manganese level on specific growth ratio (SGR), feed efficiency ratio (FER) and survival ratio (SR) of

ANOVA , analysis of variance.

.

rainbow trout (12–13 mg kg
1, Ogino & Yang 1980), and

somewhat higher than the report of grouper

(19 mg Mn kg
1, Ye et al 2008) These differences in the

estimated manganese requirements of different species are

probably real species specific, variations in intestinal

man-ganese absorption rate and feed efficiency (Shearer 1995)

Moreover, the disparity probably comes from the methods

of data analysis, manganese forms and their availability

There are also some special reasons for the different

results In the present study, the water temperature varied

from 27.5 to 31°C and in the range of 25 to 32 °C, which

is optimal water temperature for cobia (Guo et al 2007)

However, this temperature is higher than optimal water

temperature of other fish (except tilapia and grouper’s),

such as rainbow trout, carp, Atlantic salmon and juvenile

gible carp Such high temperature increased the oxidationpressure to cobia, and produced some antioxidant response(Parihar et al 1997) This is probably one possibility whymanganese requirement of cobia and grouper is higher thanrainbow trout, carp, Atlantic salmon and juvenile giblecarp After 10-week experiment, the final body weight offish fed the diet with 23.87 mg Mn kg
1 was 7.6 times ofits initial body weight In the present study, the SGR ofjuvenile cobia is higher than that in other fish, such asgrouper (Ye et al 2008) and juvenile tilapia (Lin et al.2008) The juvenile cobia grows so fast that it needs moremineral nutrients (Qiao 2007; Liu et al 2010) includingmanganese

The requirements of dietary manganese based on thewhole-body manganese and vertebrae manganese were22.38 (Fig 2) and 24.93 mg Mn kg
1(Fig 3), respectively

Table 3 Effect of different dietary manganese level on whole-body composition of cobia

Dietary manganese (mg.kg-1)

ANOVA , analysis of variances.

of variances.

Table 4 Manganese concentrations in liver, vertebra and whole

ANOVA , analysis of variances.

Values in a column that do not have the same superscript are

analysis of variances.

9 10 11 12 13 14 15 16 17

6 fish per group Requirements derived with the broken-line

.

Aquaculture Nutrition 19; 461–467 ª 2012 John Wiley & Sons Ltd

These results were higher than the requirement value

(21.72 mg Mn kg
1) based on SGR This finding suggested

that vertebrae, liver and other body tissues have a capacity

to buffer changes in manganese supply, and manganese

deposition need not be at its maximum for the highest

weight gain, as what was found in phosphorus requirement

of juvenile large yellow croaker, Pseudosciaena crocea R

(Mai et al 2006) and juvenile Japanese seabass,

Lateolab-rax japonicus(Zhang et al 2006)

Hepatic total SOD activity has been found to decrease

when dietary manganese was deficient (NRC 1993) Similar

results have also been reported in rainbow trout (Knox

et al.1981), Atlantic salmon (Maage et al 2000) and nile gible carp (Pan et al 2008) And the present study alsosupports this point The activities of Cu–Zn SOD and MnSOD were all suppressed at low dietary manganese level,and this had been confirmed by the findings of Knox et al

juve-(1981) who reported that hepatic Cu–Zn SOD activity wassuppressed in the Mn-deficient trout because the Cu and

Zn concentration in Mn-deficient trout liver was lower

Results of the present study showed that manganese ciency suppressed cobia growth and reduced SOD activi-ties This clearly indicated that cobia have a requirementfor Mn that cannot be met by Mn in the unsupplementeddiet, thus dietary supplementation is necessary On thebasis of broken-line regression of SGR, manganese concen-tration in whole body and vertebra and the manganeserequirements of juvenile cobia were 21.72 mg kg
1,22.38 mg kg
1and 24.93 mg kg
1diet in the form of man-ganese sulphate, respectively

defi-This study was financially supported by National KeyTechnology R&D Program for the 11th Five-year Plan ofChina (Grant no.: 2006BAD03B03) The author thankMingchun Ren for his helping in feeding and sampling

Thanks are also due to Xingwang Liu, Lindong Xiao andXiaoru Chen for their assistances in the study

Association of Official Analytical Chemists (AOAC) (1995) cial Method Analysis, 16th edn Association of Official Analytical Chemists, Arlington, VA, USA pp 1141.

Offi-Chou, R.L., Su, M.S & Chen, H.Y (2001) Optimal dietary tein and lipid levels for juvenile cobia Rachycentron canadum.

Chou, R.L., Her, B.Y., Su, M.S., Wang, G., Wu, Y.H & Chen, H.Y (2004) Substituting fishmeal with soybean meal in diets of

Craig, S.R., Schwarz, M.H & McLean, E (2006) Juvenile cobia (Rachycentron canadum) can utilize a wide range of protein and lipid levels without impacts on production characteristics Aqua-

Gatlin, D.M & Wilson, R.P (1984) Studies on the manganese

Guo, X.W., Ou, Y.J & Liao, R (2007) Present status on studies

Hardy, R.W & Barrows, F.T (2002) Diet formulation and facture In: Fish Nutrition, 3rd edn (Halver, J.E & Hardy, R.W.

Holt, G.J., Faulk, C.K & Swarchz, M.H (2007) A review on val culture of cobia, Rachycentron canadum, a warm water mar-

Y = 54.4899–0.8270(24.9314-X)

X 24.9314 R 2 = 0.920

Figure 3 Relationship between dietary manganese level and

verte-bra manganese of cobia fed the six diets for 10 weeks Each point

represents the mean of three groups of fish within a treatment with

6 fish per group Requirements derived with the broken-line

Table 5 Total superoxide dismutase (T-SOD), Cu–Zn superoxide

ANOVA , analysis of variances.

Values in a column that do not have the same superscript are

analysis of variances.

.

Knox, D., Cowey, C.B & Adron, J.W (1981) The effect of low

dietary manganese intake on rainbow trout (Salmo gairdneri).

Leach, R.M (1976) Metabolism and function of manganese In:

Essential and Toxic Elements, Vol 2, Trace elements in human

Press, New York.

Liao, I.C., Huang, T.S., Tsai, W.S., Hsueh, C.M., Chang, S.L &

Leano, E.M (2004) Cobia culture in Taiwan: current status and

Lin, Y.H., Lin, S.M & Shiau, S.Y (2008) Dietary manganese

Liu, K., Wang, X.J., Ai, Q.H., Mai, K.S & Zhang, W.B (2010)

Dietary selenium requirement for juvenile cobia, Rachycentron

Lorentzen, M., Maage, A & Julshamn, K (1996) Manganese

sup-plementation of a practical, fish meal based diet for Atlantic

Lunger, A.N., Craig, S.R & McLean, E (2006) Replacement of

fish meal in cobia (Rachycentron canadum) diets using an

Maage, A., Lygren, B & El-Mowafi, A.F.A (2000) Manganese

requirement of Atlantic salmon (Salmo salar) fry Fisheries Sci.,

66, 1–8.

Mai, K.S., Zhang, C.X., Ai, Q.H., Duan, Q.Y., Xu, W., Zhang,

L., Liufu, Z.G & Tan, B.P (2006) Dietary phosphorus

require-ment of large yellow croaker, Pseudosciaena crocea Aquaculture,

251, 346–353.

Mai, K.S., Xiao, L.D., Ai, Q.H., Wang, X.J., XU, W., Zhang,

W.B., Liufu, Z.G & Ren, M.C (2009) Dietary choline

require-ment for juvenile cobia, Rachycentron canadum Aquaculture,

289, 124–128.

NRC (National Research Council) (1993) Nutrient Requirements

of Fish National Academy Press, Washington, DC.

Ogino, C & Yang, G.Y (1980) Requirement of carp and rainbow

trout for dietary manganese and copper Bull Jpn Soc Sci.

Pan, L., Zhu, X., Xie, S., Lei, W., Han, D & Yang, Y (2008)

Effects of dietary manganese on growth and tissue manganese

concentrations of juvenile gible carp, Carassius auratus gibelio.

Parihar, M.S., Javeri, T., Hemnani, T., Dubey, A.K & Prakash,

P (1997) Responses of superoxide dismutase, glutathione dase and reduced glutathione antioxidant defenses in gills of the freshwater catfish (heteropneustes fossills) to short-term elevated

Qiao, Y.G (2007) Study on nutrition physiology of zinc, iron and copper in cobia (Rachycentron canadum) MA.Sc dissertation Ocean University of China, China.

Robbins, K.R., Norton, H.W & Baker, D.H (1979) Estimation of

1714.

Lall, S.P (2002) The minerals In: Fish Nutrition, 3rd edn (Halver,

Diego, CA, USA.

Shearer, K.D (1995) The use of factorial modeling to determine the dietary requirements for essential elements in fishes Aquacul-

Wang, D.Z & Zhao, L (1994) Requirement of fingerling grass carp (Ctenopharyngodon idellus) for manganese J Shanghai Fish.

Wang, J.T., Liu, Y.J., Tian, L.X., Mai, K.S., Du, Z.Y., Wang, Y.

& Yang, H.J (2005) Effect of dietary lipid level on growth formance, lipid deposition, hepatic lipogenesis in juvenile cobia

Ye, C.X., Tian, L.X., Yang, H.J., Liang, J.J., Niu, J & Liu, Y.J (2008) Growth performance and tissue mineral content of juvenile grouper(Epinephelus coioides) fed diets supplemented

614.

Zhang, C.X., Mai, K.S., Ai, Q.H., Zhang, W.B., Duan, Q.Y., Tan, B.P., Ma, H.M., Xu, W., Liufu, Z.G & Wang, X.J (2006) Die- tary phosphorus requirement of juvenile Japanese seebass, Lateo-

Zhou, Q.C., Wu, Z.H., Tan, B.P., Chi, S.Y & Yang, Q.H (2006) Optimal dietary methionine requirement for juvenile cobia

Zhou, Q.C., Wu, Z.H., Chi, S.Y & Yang, Q.H (2007) Dietary lysine requirement of juvenile cobia (Rachycentron canadum).

.

Aquaculture Nutrition 19; 461–467 ª 2012 John Wiley & Sons Ltd

Institute of Aquaculture, College of Fisheries and Ocean Sciences, University of the Philippines Visayas, Miagao, Iloilo,

Philip-pines

Common carp were fed diets containing various levels of

Quillaja saponins (QS) (0, 150, 300 and 450 mg kg 1 dry

diet) in a completely computerized respirometric system

for 4 weeks Fish fed diets containing QS exhibited

sig-nificantly higher ABW and specific growth rate than did

those fed the control diet; those fed diets containing QS

150 grew fastest but were not significantly different from

those fed diets with QS 300 and QS 00450 All the

utili-zation efficiency indices, namely food conversion

effi-ciency (FCE), protein productive value and PG were

increased by QS supplementation There were no

signifi-cant differences in the average routine metabolic rate

between treatments, indicating that dietary QS at the

lev-els tested were not toxic to the carp Increases in

amy-lase and trypsin specific activities were observed at QS

300 and QS 450 Enzymes of carbohydrate metabolism

such as G6PDH, 6-phosphogluconate dehydrogenase and

pyruvate kinase were not significantly affected by dietary

QS Activities of the aerobic enzyme Cox and to a

lim-ited extent that of the anaerobic enzyme lactate

dehydro-genase were significantly increased by dietary QS but the

net effect was a shift towards aerobic metabolism,

indi-cating absence of stress and favouring the anabolic

pro-cesses Thus, Quillaja saponin was beneficial as a feed

supplement in the common carp

KEY WORDS: carp, enzymes, growth, metabolism, Quillaja,

saponins

Received 1 April 2012; accepted 25 June 2012

Correspondence: Augusto E Serrano, Jr., Institute of Aquaculture,

Col-lege of Fisheries and Ocean Sciences, University of the Philippines

Visa-yas, Miagao, 5023 Iloilo, Philippines E-mail: serrano.gus@gmail.com

The cost of feeds in aquaculture is a major factor in theexpense of fish farmers Thus, the search for micronutrientsthat will boost the palatability and efficiency of aquafeedsand at the same time lower the cost seems to be very relevant

One such compound is saponin, although results of feedingtrials in farm and aquatic animals are mixed Saponins havebeen documented to lower nutrient availability in higher ani-mals (West et al 1978) and decrease enzyme activity contrib-uting to a growth-retarding effect in animals (Cheek 1971)

On the other hand, there are also reports that saponin lowersblood cholesterol, preventing cancer and improving theimmune system (Rao & Sung 1995) Beneficial effects of sup-plementing animal feed with extracts from Yucca and Quil-laja saponins (QS) on performance and health on variouslivestock species are also well documented (Price et al 1987;

Rao & Sung 1995; Makkar & Becker 1995)

Saponins consist of a sugar moiety linked with ahydrophobic aglycone (sapogenin), which may be triterpe-noid or steroid in nature QS contain predominantly trit-erpenoid aglycone produced from Quillaja saponaria, atree native to the Andes region; it possesses strong bio-logical activity partly owing to the presence of Quillaicacid Its inclusion in the diet at 150 mg kg 1 of commoncarp increases growth and metabolic efficiency (Francis

et al 2002a,b) In the Nile tilapia, dietary QS at

300 mg kg 1 significantly increase average values forenergy retention, apparent lipid conversion, carcass fat,energy and significantly decrease average values forapparently unutilized energy and carcass ash contentcompared with the group fed diet without saponins(Francis et al 2001)

In fisheries and aquaculture, saponin-containing plantshave been used for catching fish or in the eradication ofunwanted organisms in earthen fishponds Thus, saponins

.

doi: 10.1111/j.1365-2095.2012.00980.x .2013 19; 468–474

Aquaculture Nutrition

have been known for their high toxicity to fish, while the

fish can be eaten by man without any risk of intoxication

The aim of the present study is to determine the effects of

Quillaja saponin on the growth, feed efficiency, digestive

enzyme activities and metabolism of the common carp,

Cyprinus carpio L Evaluation of whether it is a toxicant

was measured at the organismic as well as tissue levels

Common carp (C carpio) with initial average body weight

of 6.6 ± 2.5 SD (range 3.5–11.2 g) were used in this study

Before the feeding trials, the fish were fed at maintenance

level for 1 week in the respiration boxes The fish were then

randomly divided into 12 respiration boxes with three fish

per box keeping the total weight almost similar in all

boxes

Respiration during the whole feeding trials was measured

in an open-flow recirculation system (800 L total capacity,

as described by Focken et al (1994)) converted to a

flow-through system for purposes of this study Water was

pumped from a lower reservoir to an overhead reservoir

where it was heated and temperature thermostatically

maintained at 27°C Water was partly diverted to a

bio-logical filter column (250 L Norpac medium; Norddeutsche

Seekabelwerke, Nordenham, FRG) for nitrification of urea

and ammonia Water flow rate to each box was maintained

at 0.25–0.3 L min 1

, monitored continuously by electronicpaddlewheel flowmeters and were adjusted by dosage

valves Oximeter readings as well as flow rates in the

differ-ent respiration boxes were transmitted continuously to the

central computer as digital signals The system ran

continu-ously, except during sampling in which fish were weighed,

the system cleaned, the flowmeters and oxygen probe

checked and calibrated

The composition of the basal diet and the proximate

analy-sis is shown in Table 1 Four diets containing 0, 150, 300

and 450 mg kg 1of QS (No 2149; Sigma, St Louis, MO,

USA) were prepared Previously, a concentration range test

was run with 150 mg kg 1 as the maximum level but the

effects did not reach a plateau Thus, multiples of the

previ-ous highest level was now going to be tested for possibleclear cut differences in the response of the common carp.All diets were prepared from one batch of basal diet whichwas repulverized, to which QS solution was added from aprepared stock solution, made into 2-mm pellets and thenfreeze-dried Each experimental diet was fed to triplicaterespiration boxes of fish Vitamin and mineral supplementswere formulated according to Focken et al (1997) Sapo-nins were not measured in the prepared experimental feeds

Fish were fed five times at maintenance level(3.2 g kg 0.8days 1at 23°C; Focken et al 1994) in equalmeals using automatic feeders The ration for each fish wasadjusted weekly according to its body mass until the4-week feeding trial was completed According to Lazo &Davis (2000), a growth trial should be of sufficient duration

to produce relatively large increases in growth and cally significant differences between some of the dietarytreatments The 4-week feeding trial has sufficiently satis-fied these two requirements for feeding trials

statisti-Food conversion efficiency, specific growth rate (SGR),protein gained (PG) and the protein productive value(PPV) were calculated on a per fish basis as follows:

Table 1 Composition of the basal diet and proximate analysis of the experimental diets (dry weight basis)

25 mg; Panthothenic acid, 10 mg; Choline chloride, 100 mg;

.

Aquaculture Nutrition 19; 468–474 ª 2012 John Wiley & Sons Ltd

SGR (% day )= (lnWf lnWo)9 100/(tf to)

FCE= wet weight gain (g)/feed consumed (g)

PG (g)= (final initial) whole body protein

PPV= PG (g)/crude protein fed (g)

Where Wf and Woare the final and initial weight of a

fish in a respiration box at the end of a 7-days period,

respectively

Representative samples at the start and all fish at the

end of the experiment were killed; intestines and livers were

excised and immediately deep frozen Prior to analysis, the

carcasses were autoclaved for 2 h at 120°C, homogenized,

refrozen and freeze-dried

Samples of feed and proximate analysis were analysed

according to official methods (AOAC 1980), that is, dry

matter by drying to a constant weight at 105°C, crude

protein as macro-Kjeldahl Nx6.25, lipids by extraction with

petroleum ether and gross energy by bomb calorimeter

(IKA C700; IKA, Guangzhou, China) with benzoic acid as

standard

Intestines and livers were homogenized in 20 volumes of

ice-cold salt solution (10 g kg-1) NaCl+ 3 g kg-1

CaCl2)and 0.10 M sodium-phosphate buffer (pH 7.4), respec-

tively, using ultra turrax for 20 s The homogenate was

centrifuged at 8949 g at 4°C for 15 min The supernatant

was used for enzyme assays and protein determination

(Lowry et al 1951)

Amylase Activity was assayed as described by Bernfeld

(1955) in which the increase in reducing power of a

buf-fered starch solution was measured with 3,5-dinitrosalicylic

acid (DNS, Chemos GmBH, Regenstauf, FRG) at 540 nm

The assay mixture consisted of 1.0 mL soluble starch,

0.5 mL enzyme preparation and 0.5 mL salt solution (i.e

the homogenizing solution) The reaction was carried out

at 25°C and was stopped by adding the DNS solution

The mixture was then heated for 5 min in boiling water,

cooled in running tap water and absorbance read at

546 nm Amylase activity was expressed in terms of mg

maltose liberated from starch

Trypsin (E.C 3.4.21.4) Tryptic activity was determined

according to Geiger & Fritz (1984) using the specific

sub-strate Benzoyl-arginine-p-nitroanilide (BAPNA;

Appli-Chem GmbH, Darmstadt, Germany) The assay mixture

consisted of 1.25 mL of the substrate solution, 0.1 mL of

purified trypsin solution and buffer in a final volume of2.25 mL The reaction was carried out at 25°C and wasstarted by adding BAPNA solution for 5 min andstopped by adding 0.25 mL of 30% acetic acid Theabsorbance of the supernatant was read at 405 nm, andthe enzyme activity was expressed as micromoles ofproduct formed min 1L 1 of enzyme preparation Theactivity of the purified enzyme preparation was sub-tracted from the total trypsin activity

With the exception of Cox, all liver enzymes weremeasured at 340 nm in a recording spectrophotometerusing 3.0 mL of assay mixture in 1-cm cuvettes at 22°C

The activity of cytochrome c-oxidase was determined at

550 nm The enzyme activities were calculated using e(millimolar absorption coefficient) of 19.1 for Cox(Chance 1952) and 6.229 103

cm M 1 for NADH pliChem GmbH; Horecker & Kornberg 1948) The assaymixtures for each of the measured enzymes were asfollows:

(Ap-Glucose-6-phosphate dehydrogenase (G6PDH; E.C 1.1.1

49) Of 33 mM sodium-potassium phosphate buffer pH7.4, 0.53 mM MgCl 6H2O, 0.033 mM NADP (ABCRGmbH, Karlsruhe, Germany), 0.1 mM glucose-6-phos-phate (No G5758; Sigma-Aldrich, St Loius, MO, USA),0.067 mM 6-phosphogluconic acid pH 7.6 (No G5758;

Sigma-Aldrich), 0.2 mL of enzyme preparation

6-Phosphogluconate dehydrogenase (E.C 1.1.1.44) Of

67 mM sodium-potassium phosphate buffer pH 7.4, 8 mMMgCl, 0.5 mM NADP, 1 mM 6-phosphogluconic acid pH8.7, 0.2 mL of enzyme preparation

Pyruvate kinase (E.C 2.7.1.40) Of 33 mM sium phosphate buffer pH 7.4, 8 mM MgKl, 75 mM KCl,

sodium-potas-2 mM ADP, 0.15 mM NADH, 1 mM uvate, 58 U (No 79418; Sigma), lactate dehydrogenase(LDH) pH 7.4 (Roche Applied Science, Mannheim, Ger-many), 0.1 mL enzyme preparation

phosphoenolpyr-Lactate dehydrogenase (E.C 1.1.1.27) Of 33 mM potassium phosphate buffer pH 7.4, 0.27 mM NADH(Sigma-Aldrich), 1.33 mM pyruvate pH 7.4 (AppliChemGmbH), 0.2 mL enzyme preparation

sodium-Cytochrome c-oxidase (Cox; E.C 1.9.3.1) Of 33 mMsodium-potassium phosphate buffer, 8 mM cytochrome

c pH 7.4 (Sigma-Aldrich), 0.1 mL enzyme preparation,deionized water in a final volume of 2.5 mL

.

Protein determination Protein concentration was estimated

by the method of Lowry et al (1951) using bovine serum

albumin (Sigma-Aldrich) as standard

Routine metabolic rate The routine metabolic rate (RMR)

was calculated as mg oxygen consumed h 1kg 2(0.8); the

relevant oxygen consumption values for calculating RMR

were continuously recorded Thirty-two measurements of

oxygen consumption per chamber were taken every 24 h

and recorded on the hard disk of the computer, which

con-trols the system

Statistical methods Standard error of the mean (SEM)

was calculated for all mean values Data were subjected to

analysis of variance (ANOVA) and Tukey–Kramer Range

Test to determine differences in means (P < 0.05) ABW

and SGR were subjected to analysis of covariance (ANCOVA)

with initial weight as co-variate (Dowdy et al 2004)

There were no significant differences in average body

weight (ABW) among treatments for the first 2 weeks of

the feeding trial (Table 2) During the third and fourth

week, diets with QS 150 resulted in significantly the highest

ABW, followed by QS 300 and QS 450 which were not

sig-nificantly different from each other; the control diet

resulted in significantly the lowest ABW

Unlike the ABW, differences in SGR were already

observed in the second week of the feeding trial (Table 3)

Fish fed diets with QS 150 mg kg 1 exhibited the fastest

growth rate from the second week until the final week, but

this was not significantly different from those fed diets

con-taining QS 300 and QS 450 while those fed the control diet

(QS 0) displayed the lowest SGR

Fish fed diet with QS 450 exhibited significantly the highestvalues for FCE and PPV (Table 2) while those fed diets with

QS 0 exhibited the lowest PG, on the other hand, was icantly the highest in fish fed QS 150, higher than in fish fed

signif-QS 300 or signif-QS 450, and lowest in those fed signif-QS 0

Adding QS to the diet resulted in no significant ence in the average RMR from those fed the control diet(QS 0)

differ-Amylase specific activity was significantly increased only at

QS 450, while that of trypsin increased significantly at both

QS 300 and QS 450 and was not significantly differentfrom each other (Table 4)

Among the liver carbohydrate-metabolizing enzymes,Cox activities were significantly increased as a result of theaddition of QS to the diet at all concentrations LDH, incontrast, was significantly increased only at QS 150; furtherincreases in the QS concentration resulted in LDH activitylevels not significantly different from those of the control.G6PDH, 6-phosphogluconate dehydrogenase (6-PGDH)and pyruvate kinase (PK) activities were not significantlyaffected by the dietary QS (Table 4) The ratio of Cox toLDH increased upon QS supplementation

Results of the present study showed that adding saponin tothe diet significantly increased the final ABW by a range of37.5–73.2% over those fed the control diet; the ABW val-ues tested at various QS concentrations tested were not sig-nificantly different The same was observed when thegrowth rate was expressed as SGR, those fed diets with QSsupplement showed an increase of 0.7–1.18% per day over

Table 2 Periodic average body weight (ABW) of carp fed the experimental diets containing various levels of Quillaja saponins for 4 weeks

in respirometric chamber Means with different letters are significantly different (P<0.05)

those fed the control diet; the SGRs of fish fed diet with

various QS concentrations were not significantly different

Francis et al (2002a,b) did not find any significant

differ-ence between carps fed the control and those supplemented

with QS Bureau et al (1998) have fed Chinook salmon

and rainbow trout diets with QS at 0.15 and 0.30% (=1500

and 3000 mg kg 1, respectively); these levels represent a

tenfold increase over the concentrations in the present

study In both Chinook salmon and rainbow trout, fishes

fed the QS 1500 diet had a growth performance similar to

the fishes fed the control diet whereas those fed the QS

3000 diet with the higher QS level had significantly lower

feed intake, growth and feed efficiency As the maximum

concentration of the QS in the present study was just

450 g kg 1, this observation on common carp seemed not

far fetched However, upon microscopic examination of the

intestinal tissues of both fishes by Bureau et al (1998),

there were extensive damages In contrast, purified soy

saponin did not cause soybean-induced enteritis in Atlantic

salmon (Krogdahl et al 1995) No histologic examination

of the tissues was done, thus was not confirmed in the

pres-ent study Siddhuraju & Becker (2003) did not observe any

effect on the growth of common carp fed diets containing

autoclaved Mucuna seed meal containing about 0.58%

saponin (=5750 mg kg 1

)

The overall positive effects of dietary QS in the commoncarp could have stemmed from the potential of saponins toalter the permeability of the gut membrane, resulting in theuptake of nutrients, which are usually excluded by the gutmembrane, and interfering with the absorption of essentialnutrients (Petterson et al 1999) An in vitro study has shownthat oat saponins increased permeability for the macromole-cules such as ovalbumin (Onning et al 1996) and that thismay be accomplished by combining irreversibly with discretesites within the plasma membrane (Price et al 1987) In fish,macromolecules such as intact proteins have been docu-mented to penetrate both intercellular spaces and subepithe-lial connective tissue and plasma (Ash 1985) of the tench andcarp, respectively Thus, the effect of QS on the increasingpermeability of macromolecules in the intestine of higheranimals might also be at work in the common carp in thepresent study The increased growth rate and efficiency infeed utilization may have stemmed from an increasedefficiency of absorption of the digested food in the intestine

of the fish Also, the increased rate of dietary protein andcarbohydrate digestion presumably brought about by the

Table 3 Growth, efficiency and routine metabolic parameters of carp fed diets containing various levels of saponin for 28 days in

respiro-metric chamber Means with different letters are significantly different (P<0.05)

Values are means of ± SEM.

Table 4 Effects of various concentrations of QS on the activities of selected digestive and metabolic enzymes (amylase = mg

Values are means of ±SEM.

LDH, lactate dehydrogenase; Cox, cytochrome c-oxidase; G6PDH, glucose-6-phosphate dehydrogenase; 6-PGDH, 6-phosphogluconate

dehydrogenase; PK, pyruvate kinase.

.

increased specific activities of amylase and trypsin in the

present study could have hastened the absorption of these

nutrients across the gut that could have resulted in an

increased efficiency of nutrient utilization

Toxicants are known to influence physiological and

bio-chemical state of aquatic organisms by exhibiting marked

changes in the activities of several enzymes of carbohydrate

metabolism (Omoregie 2002) It was our aim to determine

whether or not QS in the diet caused some organismic and

tissue level indications of stress in the common carp

paral-lel to the effects of a toxicant At the organismic level, one

sign of stress is visible hyperventilation or an increased in

oxygen uptake if exacting measurement is required In the

present study, there were no significant differences in the

oxygen uptake between dietary treatments were observed

in terms of RMR and thus were not under stress even

when QS was in the diet At the tissue level, one sign of

stress is a shift towards anaerobic metabolism especially in

carbohydrate metabolic pathways This is emphasized by

Matthews & Phillip (2006) who hypothesize that the

impact of some toxicants on different tissues of fish

sug-gests the tendency of shift in the metabolism of

carbohy-drates more towards anaerobic dependence than aerobic

oxidation through Krebs cycle In the present study,

activi-ties of carbohydrate metabolic enzymes 6-PGDH, G6PDH

and PK and to a limited extent LDH were unaffected by

the dietary QS Only Cox, an enzyme of the oxidative

phosphorylation pathway, was positively affected The

ratios of the activities of the aerobic enzyme Cox to those

of the anaerobic enzyme LDH were increased by QS

sup-plementation indicating that stress was not detected at the

tissue level and instead the rate of anabolic processes were

higher than those of catabolic processes

In conclusion, dietary QS promoted growth (ABW, SGR)

starting on the third week of feeding at all three

concentra-tions tested (150–450 mg kg 1

diet) better than did the trol diet Indices of efficiency of feed utilization (FCE, PG

con-and PPV) were also improved by the QS supplementation

but the highest improvement was at 450 mg kg 1 Amylase

specific activity was elevated at QS 450 while that of trypsin

at QS 300 demonstrating the the QS could increase

effi-ciency of digestion of carbohydrates and protein at these

levels The activity of the metabolic enzyme Cox was

increased by the QS supplementation while RMR and the

activities of carbohydrate metabolic enzymes- 6-PGDH,

G6PDH and PK and to a limited extent LDH were

unaf-fected; these were indications that the anabolic processes

dominated over catabolic processes resulting in better

growth and feed utilization efficiencies

The author wishes to express his gratitude to DeutscherAkademischer Austausch Dienst for the short-term fellow-ship it provided to the author, to Dr Ulfert Focken for theinvitation and technical advice, Prof Klaus Becker forallowing the use of facilities at the Hohenheim University,and to George Francis for the assistance

Ash, R (1985) Protein digestion and absorption In: Nutrition and Feeding in Fish (Cowey, C.B., Mackie, A.M & Bell, J.G eds),

AOAC (Association of Official Analytical Chemists) (1980)

Washington, DC, pp 1018

In: Methods in Enzymology (Colowick, S.P & Kaplan, N.O.

Bureau, D.P., Harris, A.M & Cho, C.Y (1998) The effects of purified alcohol extracts from soy products on feed intake and growth of chinook salmon (Oncorhynchus tshawytscha) and rain-

Chance, P.R (1952) Spectra and reaction kinetics of respiratory

221.

Cheek, P.R (1971) Nutrition and physiological implications of

Dowdy, S., Weardon, S & Chilko, D (2004) Statistics for Research, 3rd edn John Wiley & Sons, Inc., New Jersey, NY Focken, U., Schiller, M & Becker, K (1994) A computer-con- trolled system for the continuous determination of metabolic rates of fish In: Measures for Success: Metrology and Instru- mentation in Aquaculture Management (Kestemont, P., Muir, J., Sevilla, F & Wilot, P eds), Paper presented at the Inter- national Conference Bordeaux Aquaculture 1994, Antony,

Focken, U., Becker, K & Lawrence, P (1997) A note on the effects of L-carnitine on the energy metabolism of individ-

264.

Francis, G., Makkar, H.P.S & Becker, K (2001) Effects of

cholesterol in individually reared Nile tilapia (Oreochromis

Francis, G., Makkar, H.P.S & Becker, K (2002a) Effects of cyclic and regular feeding of a Quillaja saponin supplemented diet on growth and metabolism of common carp (Cyprinus carpio L.).

Francis, G., Makkar, H.P.S & Becker, K (2002b) Dietary mentation with a Quillaja saponin mixture improve growth per- formance and metabolic efficiency in common carp (Cyprinus

Geiger, R & Fritz, H (1984) Trypsin In: Methods of Enzymatic Analysis Vol 5 (Bergmeyer, H.U., Bergmeyer, J & Grabl, M.

Horecker, B.L & Kornberg, A (1948) The extinction coefficients

.

Aquaculture Nutrition 19; 468–474 ª 2012 John Wiley & Sons Ltd

Krogdahl, A., Roem, A & Baeverfjord, G (1995) Effects of

soy-bean saponin, raffinose and soysoy-bean alcohol extract on nutrient

digestibilities, growth and intestinal morphology in Atlantic

sal-mon In: Proceedings of International Conference in

Aquacul-ture ‘95 and the Satellite Meeting of the Nutrition and Feeding

of Cold Water Species (Svennevig, N & Krogdahl, A eds), pp.

Spec Publ No 23, Gent, Belgium.

Lazo, J.P & Davis, D.A (2000) Ingredient and feed evaluation.

In: Encyclopedia of Aquaculture (Stickney, R.R ed.), pp.

Lowry, O.h., Rosebrough, N.J., Farr, A.L & Randall, R.J (1951)

Protein measurement with the Folin phenol reagent J Biol.

Makkar, H.P.S & Becker, K (1995) Are Quillaja saponins

antinu-tritive to ruminants? 210th National Meeting of the American

of Abstracts.

Matthews, J.P & Phillip, B (2006) Metabolic responses in

Omoregie, E (2002) Acute toxicity of water soluble fractions of

crude oil to the Nile Tilapia, Oreochromis niloticus (L.) Bull.

Onning, g., Wang, Q., Westrom, B.R., Asp, N.-G & Karlsson, B.W (1996) Influence of oat saponins on intestinal permeability

Petterson, D.S., Harris, D.J., Rayner, C.J., Blakeney, A.B &

Choct, M (1999) Methods for analysis of premium livestock

Price, K.R., Johnson, I.T & Fenwick, G.R (1987) The chemistry and biological significance of saponins in foods and feedstuffs.

Rao, A.V & Sung, M.-K (1995) Saponins as anticarcinogens.

Siddhuraju, P & Becker, K (2003) Comparative nutritional evaluation of differentially processed [Mucuna seeds (Mucuna

on growth performance, feed utilization and body composition

West, L.G., Greger, J.L., White, A & Nonnamaker, B.J (1978) In

.

1 1 1 1 1 2 1

1

Departamento de Produccio´n Animal, Facultad de Veterinaria, Universidad Complutense de Madrid, Madrid, Spain;2 DibaqAcuicultura, Fuentepelayo, Segovia, Spain

The objectives of the present study were to investigate

the effects of inulin and fructooligosaccharides (FOS) on

growth performance, whole body and fillet chemical

com-position and intestinal microbiota of rainbow trouts

reared under fish farming conditions Trouts fed

inulin-or FOS-containing diets (5 and 10 g kg 1) exhibited

significant (P = 0.030) body weight gain improvements

compared with controls An increase in gross energy

(P = 0.044) and Ca content (P = 0.034) in the whole

body of trouts was observed for prebiotic treatments A

decrease in crude protein content (P = 0.009) and a

ten-dency to increase total lipid and gross energy contents

(P = 0.090 and P = 0.069, respectively) were detected in

the fillet tissue for prebiotic treatments These results

clearly indicate that inulin and FOS improved the

intesti-nal absorption of Ca and that the increased amount was

predominantly incorporated into bone tissue Inulin

reduced (P= 0.027) the intestinal population of Vibrio

spp in the distal region to such an extent that no viable

counts were detected The presence of Flavobacterium

spp was not detected in any group, and the numbers of

Aeromonas spp., Pseudomonas spp and Gram-positive

bacteria were not affected (P> 0.05)

KEY WORDS: chemical composition, intestinal microbiota,

performance, prebiotics, rainbow trout

Received 1 February 2012; accepted 25 June 2012

Correspondence: L.T Ortiz, Departamento de Produccio´n Animal,

Fac-ultad de Veterinaria, Universidad Complutense de Madrid, Avda Puerta

de Hierro s/n, 28040 Madrid, Spain.

E-mail: ltortiz@vet.ucm.es

During the last decade, the use of dietary compounds withpotential prebiotic effects is being considered as a possiblemean of improving gut health and growth performance offarmed animals in the absence of antibiotic growth promot-ers (Verdonk et al 2005) Most of the nutritional andhealth benefits attributable to prebiotics are direct or indi-rectly linked with their selectively stimulating effect ongrowth and/or activity of the beneficial resident intestinalbacteria (Gibson & Roberfroid 1995; Gibson 2000; Patter-son & Burkholder 2003; Merrifield et al 2010; Ringø et al.2010)

The effects of inulin-type fructans [inulin and gosaccharides (FOS)], among other types of prebiotics, ongrowth performance and health of terrestrial farmed ani-mals have been widely studied (Xu et al 2003; Kocher2006; Rebole´ et al 2010) The results are inconclusive,however, and may be affected by many factors such as pre-biotic type, inclusion level, diet form and composition, ani-mal characteristics, husbandry hygiene or environmentalstress conditions (Patterson & Burkholder 2003; Verdonk

fructooli-et al.2005)

In aquaculture, the potential effect of inulin-type tans has been evaluated to a limited extent and variably indifferent fish species (Ringø et al 2010) Reports from vari-ous studies have revealed that inulin and/or FOS mayimprove the growth rate of Siberian sturgeon (Acipenserbaerii) (Mahious et al 2006a) and hybrid tilapia (Oreochr-omis aureus ♂ 9 O niloticus ♀) (Hui-Yuan et al 2007) aswell as modify the gastrointestinal tract microbiota of Arc-tic charr (Salvelinus alpinus) (Ringø et al 2006) and Atlan-tic salmon (Salmo salar) (Bakke-McKellep et al 2007) Toour knowledge, there are no data available about theresponse of rainbow trout (Oncorhynchus mykiss) to dietary

incorporation of inulin or FOS For this reason, the

cur-rent study was designed to examine the possible effects of

inulin and FOS on the growth performance, whole body

and fillet chemical composition and intestinal microbiota of

rainbow trouts reared under farming conditions

Concern-ing intestinal microbiota, the emphasis was on Aeromonas,

Pseudomonas, Vibrio and Flavobacterium as they are the

typical genera, among others, detected in fresh water fish

species (Sugita et al 1996; Huber et al 2004; Kim et al

2007)

Two types of fructan compounds, inulin and FOS, were

used as prebiotics Inulin was a commercial product

(PRE-BIOFEED 88; Qualivet, Las Rozas, Spain) obtained from

chicory (Cichorium intybus) roots Fructooligosaccharides

were a commercial product (OLIGOFRUCTOSE BENEO

P95; Beneo-Orafti Espan˜a SL, Barcelona, Spain) obtained

by partial enzymatic hydrolysis of inulin The analytical

results obtained for both products in our laboratory appear

in Table 1

All diets were manufactured by Dibaq Acuicultura

(Fuen-tepelayo, Segovia, Spain) The five treatments consisted of

a commercial diet (DIBAQ TROUT EVOLUTION)

sup-plemented with three levels of inulin or FOS (0, 5 and

10 g kg 1diet) To obtain these amounts of inulin or FOS

from commercial PREBIOFEED 88 and

OLIGOFRUC-TOSE BENEO P95, 0, 5.9 and 11.9 g kg 1PREBIOFEED

88 or 5.7 and 11.4 g kg 1 OLIGOFRUCTOSE BENEO

P95 were, respectively, used according to their fructan

con-tent as shown in Table 1 To obtain an uniform

distribu-tion in the experimental diets, the prebiotic product was

mixed with the vitamin–mineral premix prior to ing to the other ground ingredients All diets were extrudedand pelleted (5 mm diameter) The analytical composition

incorporat-of the control diet appears in Table 2

The experimental setup was approved by the Animal Careand Ethics Committee of the Universidad Complutense deMadrid (Spain) The feeding experiment was carried out atthe ‘Truchas del Segre’ fish farm by the Segre river (Pera-mola, Le´rida, Spain) A total of 415270 trouts with a meanweight of c 150 g were randomly distributed in 15 race-ways (25.09 2.0 9 1.2 m) The experimental diets wererandomly assigned to the aquaculture units in triplicate

No adaptation period was used, and the experimental triallasted for 49 days from 25 May to 13 July 2009 Watertemperature ranged between 11.8 and 17.3°C and dis-solved oxygen was kept above 9.0 mg L 1 All fish groupswere fed their respective diets at the same rate three timesdaily (09:00, 13:00 and 17:00 h), and this rate was periodi-cally adjusted to apparent satiation with no waste At thestart and end of the feeding trial, a sample of 30–40 troutswas collected by a wire mesh collector from each replicateraceway, bulk-weighed and returned to respective raceway

Mean individual body weight (BW) gain and feed sion ratio were calculated

(PRE-BIOFEED 88) and fructooligosaccharides (OLIGOFRUCTOSE

Twenty-five fish were randomly taken from each of the 15

raceways (three replicate raceways per treatment) and killed

by a blow to the head The fish were packaged individually

in sterile plastic bags and then transported on ice to our

laboratory in <5 h Immediately upon arrival, the content

of distal intestine from six trouts (subset of the 25 fish)

ran-domly selected was obtained by manual restrained and

pressured applied to the abdomen (Glencross et al 2005)

and pooled in aseptic conditions These samples were used

for microbial and pH analyses Additionally, six fish

(sub-set of the 25 fish) selected at random were freeze-dried,

ground (1 mm screen) and stored at 25 °C Other six fish

(subset of the 25 fish) selected at random were filleted, and

the fillets trimmed (no bones, skin on), freeze-dried, ground

(1 mm screen) and stored at 25 °C These samples of

whole body and fillet tissue were analysed for moisture,

crude protein, total lipids, ash, gross energy and mineral

elements as described in the next subheading

All analyses were carried out in duplicate Inulin-type

fruc-tans and soluble sugars were determined in commercial

pre-biotic products using an HPLC system (Hewlett-Packard

1100; Agilent Technologies GmbH, Walbronn, Germany)

with a refraction index detector and an Agilent Technologies

Zorbax carbohydrate column, following the methodology

described by Quemener et al (1994) Moisture (930.15),

crude protein (954.01) and ash (942.05) were analysed in

diets and whole body and fillet samples according to the

Association of Official Analytical Chemists (AOAC 1995)

Total lipids were determined by extraction (Soxhlet) with

chloroform–methanol (2 : 1 vol/vol) Amino acid analysis

was performed by HPLC after 22 h hydrolysis with 6 M

HCl at 110°C in sealed evacuated tubes Protein

hydroly-sates and amino acid calibration mixture were derivatized by

o-phtaldialdehyde and amino acids were measured in a

Hew-lett-Packard 1100 HPLC system fitted with a fluorescence

detector and an Agilent Technologies AminoQuant column,

following the procedure described by Jones et al (1981)

Gross energy was measured using an adiabatic bomb

calo-rimeter (IKA calocalo-rimeter C-4000; Janke & Kunkel GmbH,

Staufen, Germany) Mineral elements (Ca, P, Mg, Fe) were

determined using an inductively coupled plasma emission

spectrometer (Centro de Espectrometrıı´a Ato´mica,

Universi-dad Complutense, Madrid, Spain) after ashing the samples

and treating the ashes with nitric acid (AOAC 1995)

For bacteriological analysis, 1 g of intestinal pooled ples was homogenized each in 9 mL of sterilized physiolog-ical saline solution using a bag mixer (IUL Instruments

sam-SA, Barcelona, Spain) These suspensions were then seriallydiluted with the same diluent and aliquots of 100lL wereplated on selective media Aeromonas Medium Base (Ryan)(Oxoid, Basingstoke, UK) and TCBS media (Difco, Frank-lin Lakes, NJ, USA) were used to isolate Aeromonas spp.and Vibrio spp., respectively, the plates being incubated at

22°C for 72 h Gram-positive bacteria and Pseudomonasspp were grown on Columbia CNA agar 50 g kg 1blood(Bio-Merieous SA, Marcy L‘E´toile, France) and on Pseudo-monasmedium ISO 13720 (Pronadin, Laboratorios Conda,Torrejo´n de Ardoz, Spain), respectively, and the plateswere incubated at 22°C for 48 h Flavobacterium spp werecounted after being plated on TGE agar (Difco) and incu-bated at 22°C for 15 days All the plates were incubated

in aerobic conditions After incubation, the total numbers

of colony-forming units from duplicate plates per samplewere averaged, and the results were expressed as log col-ony-forming units g 1fresh distal intestinal content.The pH of distal intestinal content was measured onfresh homogenized pooled samples (0.5 g) diluted with

5 mL of deionized water and using a combined ence microelectrode (Crison Instruments SA, Barcelona,Spain)

glass-refer-The effects of dietary inclusion of inulin or FOS on the ferent variables were analysed using orthogonal contrasts.Contrasts tested included (i) the control diet versus inulin-

dif-or FOS-containing diets, and (ii) inulin-containing diet sus FOS-containing diet Data from treatments were alsotested for linear and quadratic effects of inclusion level ofprebiotics The computation was done by using the GeneralLinear Model procedure of Statistical Analysis System(SAS Institute 2002) Differences among means were con-sidered to be significant at P< 0.05

ver-Mortality of fish during the feeding experimental periodwas low (1.2%) and no significant (P> 0.05) differenceswere observed among treatment groups

.

Aquaculture Nutrition 19; 475–482 ª 2012 John Wiley & Sons Ltd

In general, performance data showed appreciable

individ-ual variability within treatment groups as evidenced by the

SEM values (Table 3) The addition of inulin or FOS to

the control diet had a positive effect (P= 0.030) on the

growth of trouts Thus, the mean BW gain of fish fed the

diets containing inulin or FOS (5 and 10 g kg 1) was

27.1% and 45.1% respectively higher than that for fish fed

the control diet The effect of dietary FOS level was

qua-dratic (P= 0.011) Feed conversion ratios were numerically

the lowest in fish fed the prebiotic-containing diets

How-ever, differences with respect to the values of control fish

were not significant (P= 0.454) because feed intakes also

were numerically the highest (P= 0.105) in those fish

Improvements in BW gain by the effect of inulin or FOS

have been reported for various fish species Mahious et al

(2006a) observed a significant increase in the growth rate of

Siberian sturgeon (Acipenser baerii) fed 20 g kg 1inulin for

30 days Mahious et al (2006b) and Hui-Yuan et al (2007)

using FOS (10 g kg 1 and 0.8 or 1.2 g kg 1, respectively)

observed improvements in the BW gain of weaning turbot

(Psetta maxima) and hybrid tilapia (Oreochromis aureus

♂ 9 O niloticus ♀), respectively In contrast with these

find-ings, other published data showed non-growth-promoting

effect of inulin (Mahious et al 2006b) or FOS (He et al

2003; Grisdale-Helland et al 2008) on the growth of

wean-ing turbot, hybrid tilapia and Atlantic salmon (Salmo salar),respectively

The addition of inulin or FOS to the control diet did notaffect (P> 0.05) the levels of moisture, crude protein, totallipids and ash in the whole body of trouts (Table 4) Signif-icant differences (P= 0.044) were only noted for grossenergy, higher values being detected in trouts consumingthe prebiotic-containing diets compared with the controlgroup This gross energy increment by the effect of inulinand FOS might be ascribed to the numerically increasedlipid content also observed in the fish fed both prebiotics

In fillet tissue, the results from orthogonal contrast analysis(Table 4) indicated that the dietary inclusion of inulin orFOS reduced significantly (P = 0.025 and P = 0.029,respectively) the content of crude protein, the effect beinglinear Moreover, it was detected that total lipid and grossenergy contents tended (P= 0.090 and P = 0.069, respec-tively) to be higher in the fillet of fish fed the prebiotic-con-taining diets

The changes observed in the current study for the cal composition of trouts might be due, at least in part, tothe greater BW achieved by the fish fed inulin or FOScompared with the control fish group As far as we areaware, there is no available published data about theeffects of inulin and FOS on the whole body or fillet chem-ical composition of rainbow trouts Concerning the use ofother prebiotics in aquaculture, Yilmaz et al (2007)observed that mannanoligosaccharides (MOS) at level of

chemi-45 g kg 1 diet increased (P< 0.05) the content of protein

in the whole body of rainbow trouts and did not affect that

of lipids Torrecillas et al (2007) reported that MOS at 10and 20 g kg 1produced no significant changes in the bodycomposition of European sea bass (Dicentrarchus labrax)

Grisdale-Helland et al (2008) observed that saccharides at level of 10 g kg 1 reduced by 6% proteincontent in the whole body of Atlantic salmon

galactooligo-Mineral content of Ca, P, Mg and Fe in the whole bodyand fillet tissue appears in Table 5 In whole body, dietaryinulin and FOS had a significant positive linear effect(P= 0.047 and 0.036, respectively) on the Ca content,mean values increasing by 14.1% with the inclusion of inu-lin and by 3.4% with the inclusion of FOS compared withthose recorded for the control group Contents of P, Mgand Fe were not affected (P> 0.05) by prebiotic supple-mentation In fillet tissue, inulin or FOS at inclusion level

of 5 g kg 1had no effect (P> 0.05) on the contents of Ca,

FOS on the growth performance of rainbow trout (Oncorhynchus

Treatment

BW gain (g)

Feed intake (g)

Feed conversion (g:g) Prebiotic

BW, body weight; FOS, fructooligosaccharides.

.

P, Mg and Fe, but at a higher level (10 g kg 1) decreased

significantly (P< 0.05) the contents of P and Mg The

find-ing observed in the present study, which showed that inulin

and FOS increased Ca content in the whole body but not

in the fillet tissue, suggests that both prebiotics improved

the intestinal absorption of Ca and that the increased

amount absorbed was predominantly incorporated intobone tissue

Studies performed in laboratory rodents (Lopez et al.2000; Roberfroid et al 2002) and, to lesser extent in humans(Grifting et al 2003), have shown that inulin-type fructanscan improve intestinal mineral absorption, particularly that

Effect of inulin level

Effect of inulin level

of Ca and Mg In poultry, it has been reported (Ortiz et al.

2009) that inulin supplementation of broiler diets increased

the apparent retention of Ca (up to 18%), Zn (up to 35%)

and Cu (up to 456%) The mechanism by which inulin-type

fructans can stimulate the absorption of several minerals and

Ca in particular is not well known The hypothesis more

widely accepted is that the microbial fermentation of inulin

and FOS in the large intestine lowers the pH through

forma-tion of short-chain fatty acids and lactic acid The lower

intestinal pH increases Ca solubility that leads to enhanced

Ca absorption (Levrat et al 1991; Trinidad et al 1993) In

addition, non-digestible fermentable carbohydrates may

increase mineral absorption through increasing the exchange

surface area of the large intestine or through improved

per-meability (Trinidad et al 1993; Kishi et al 1996)

Means of the viable counts of Aeromonas spp.,

Pseudomo-nasspp., Vibrio spp and Gram-positive bacteria in the

dis-tal intestinal content of rainbow trouts are given in

Table 6 The presence of Flavobacterium spp was not

detected in any treatment groups In general, the bacterial

counts were highly variable among fish in the same

treat-ment group, which has been also reported by other

researchers (Ringø et al 1995; Spanggaard et al 2000)

Neither inulin nor FOS added to the control diet had

sig-nificant effect (P< 0.05) on the numbers of Aeromonas

spp., Pseudomonas spp and Gram-positive bacteria

How-ever, the population of Vibrio spp was significantlyaffected when the fish were fed inulin Thus, at level of 5and 10 g kg 1, inulin reduced (P= 0.027) the counts ofVibriospp to such an extent that no counts for these bac-teria were detected in the distal intestinal content of trouts

This suggests that the use of inulin as a prebiotic may duce changes in the intestinal microbiota of rainbow trouts,which agrees with previous observations from studies con-ducted in various fish species Mahious et al (2006b)reported changes in the counts of Vibrio spp and Bacillusspp in the gut microbiota of weaning turbot fed inulin-containing diets Ringø et al (2006) detected that the num-bers of aerobic and anaerobic facultative bacteria werereduced in the distal intestine of Arctic charr (Salvelinusalpinus) when inulin was included in the diet Similarly,Bakke-McKellep et al (2007) observed that inulin reducedthe diversity of intestinal microbiota in Atlantic salmon

pro-There were no differences (P= 0.748) in the pH value ofdistal intestinal content among fish receiving the inulin- orFOS-containing diets and fish fed the control diet (Table 6)

This seems to indicate that inulin and FOS fermentation inthe gut of trouts did not affect short-chain fatty acid and/orlactic acid contents Nevertheless, with respect to this, itshould be kept in mind that the lack of an apparent decrease

in the intestinal pH might be due to the buffering capacity ofgut content and/or to the presence of some dietary ingredi-ents (Younes et al 1996) In this connection, for instance,Rebole´ et al (2010) reported that inulin supplementation tobroiler diets significantly increased the concentration of

Effect of inulin level

FOS, fructooligosaccharides; ND, not detected.

.