Aquaculture nutrition, tập 18, số 2, 2012

1 1 21 USDA/ARS Catfish Genetics Research Unit, Thad Cochran National Warmwater Aquaculture Center, Stoneville, MS, USA; 2 National Warmwater Aquaculture Center, Mississippi State Univers

1 2 3 4 1

Faculty of Biosciences, Fisheries and Economics, Norwegian College of Fishery Science, University of Tromsø, Tromsø, way;2 Key Laboratory for Feed Biotechnology of the Ministry of Agriculture, Feed Research Institute, Chinese Academy

Nor-of Agricultural Sciences, Beijing, China; 3 Institute of Marine Research, Matre Research Station, Matredal, Norway;4

School of BioScience, Handong University, Pohang, South Korea

Chitin consists ofb-1,4-linked N-acetylglucosamine residues

and is estimated as the second most abundant biomass in

the world after cellulose However, relatively little chitin is

utilized as a material for industrial, agricultural and

medi-cal applications and aquacultural purposes Chitin may be

useful as a constitutive material in formulated fish feed,

and the interesting effects in fish merit further evaluation

There is evidence that fish and aquatic animals harbour a

gut bacterial community that is distinctly different from

that reported in the surrounding habitat or in the diet

Thus, the gut environment provides a specific niche, and

bacterial activity in the gut is not merely a continuum of

that observed in the environment Today, it is well

accepted that the gut microbiota in fish are modulated by

dietary manipulations But to what extent can dietary

chi-tin and krill (chichi-tin-rich) modulate the inteschi-tinal microbiota

of fish and how do these dietary components affect the

immune system? These questions will be discussed in the

present review

KEY WORDS: dietary chitin and krill, digestion, fish, growth,

gut microbiota, immunology

Received 18 March 2011; accepted 6 December 2011

Correspondence: E Ringø, Norwegian College of Fishery Science,

Fac-ulty of Biosciences, Fisheries and Economics, University of Tromsø,

Tromsø, Norway E-mail: Einar.Ringo@uit.no

Chitin was first investigated in 1811 by Professor M Henri

Braconnot, who discovered it in the cell walls of

mush-rooms, but it was first named chitin by Odier (1823) Chitin(C8H13O5N)n consists of b-1,4-linked N-acetylglucosamineresidues (Fig 1), is estimated as the second most abundantbiomass in the world after cellulose and forms a majorstructural component of many organisms, including fungi,crustaceans, molluscs, coelenterates, protozoan and greenalgae (Rinaudo 2006; Khoushab & Yamabhai 2010) Chitinoccurs in two major forms: (i) a-chitin, which has a veryrigid crystalline structure because of intersheet and intra-sheet hydrogen bonding, and (ii)b-chitin, which has a rela-tively weak intermolecular force because of intrasheetinteraction (Jang et al 2004) The degree of acetylationalso varies from 0% to 100% (chitosan) Fungi containonly a-chitin, and a-chitin is the most abundant formfound in animals (Ruiz-Herrera 1992) The annual biosyn-thesis of chitin has been estimated at more than 100 bil-lion tons (Li & Roseman 2004) The chitin content ofvarious copepods that are food organisms for wild juvenilefish has been reported to range from 21 to 95 mg g 1(average of 46 mg g 1) by dry weight (for review, see Ba˚m-stedt 1986) The chitin content of Artemia urminanca cystshells range between 29.3% and 34.5% of the shell’s dryweight (Tajik et al 2008) Furthermore, chitin has beenreported to make up 3.6% (wet weight) of the stomachcontents of 50-day-old juvenile black sea bream (Acantho-pagrus schlegeli) (Om et al 2003), but this value decreaseswith growth

The ability to degrade chitin in vivo is thought to involvethe action of the enzymes chitinase (EC 3.2.1.14) andchitobiase (EC 3.2.1.30) (Jeuniaux 1961) Chitinasedegrades chitin to di- and trisaccharides and chitobiases(N-acetylglucosaminidases), which further degrades these tothe sugar monomer beta-N-acetyl-D-glucosamine (Danulat

.

.

Aquaculture Nutrition

1987) In Atlantic cod (Gadus morhua L.), strong

chitinolyt-ic activities have been measured in the stomach (Danulat &

Kausch 1984; Lindsay & Gooday 1985; Danulat 1986) In

addition, substantial chitinase activity has been reported in

the pyloric caecae and the intestine of the cod (Danulat

1986) Chitinases in the guts of cod hydrolysed chitin both

in vivo and in vitro (Danulat 1987), and only up to 9% of

ingested chitin was recovered from intestinal contents and

the faeces of the fish Based on this finding, the authors

concluded that Atlantic cod is able to digest chitin to a

large extent Feeding Atlantic salmon (Salmo salar L.) diets

where the fishmeal was replaced with chitin-containing krill

meals showed no influence of diet on the apparent

digest-ibility coefficients (ADC) of dry matter and protein, while

chitin was not utilized to any major extent (Olsen et al

2006) They also reported the ADC for chitin ranging from

13% to 40% and that the chitin digestibility tended to

decrease with increasing krill inclusion (Olsen et al 2006)

Moreover, oligomers of N-acetyl-D-glucosamine

pro-duced by chitinases in the digestive tract of fish might

func-tion as a bioactive substance (Tamai et al 2003), although

beneficial effects on fish have not yet been defined Readers

with special interest in chitinases in fish are referred to the

comprehensive review of Krogdahl et al (2005) and Ray

et al (2012) and the research paper of German & Bittong

(2009)

Although numerous investigations have been conducted

to determine the effects of diet on the intestinal microbiota

(for review, see Ringø et al 2011), little information is

available about the effect of chitin and krill (chitin-rich) on

the gut microbiota (Sera & Kimata 1972; Kono et al

1987a; Kumar et al 2006; Ringø et al 2006; Askarian

et al 2012; Z Zhou, S He, R E Olsen, B Yao & E

Ringø, unpublished data) From a microbial point of view,

one of the most important goals has been to obtain a

sta-bile indigenous gut microbiota in fish The practical effect

of this activity is the exclusion of invading populations of

non-indigenous microorganisms, including pathogens thatattempt to colonize the gastrointestinal (GI) tract (Ringø

et al 2005, 2010; Merrifield et al 2010) This fundamentaltopic arises as numerous papers published during the last

25 years have suggested that the alimentary tract isinvolved in Aeromonas and Vibrio infections (e.g Groff &

LaPatra 2000; Birkbeck & Ringø 2005; Harikrishnan &

Balasundaram 2005; Jutfelt et al 2006; Ringø et al 2007,2010; Sugita et al 2008) Therefore, one can hypothesizethat beneficial bacteria colonizing the digestive tract byproducing, for example, bacterocins may offer protectionagainst invading fish pathogens (Ringø et al 2005; Desriac

et al.2010)

The effects of chitin on the fish immune system havebeen relatively well studied One of the major purposes ofthese studies were to investigate the possibility that chitincould be used as an immunostimulatory additive for fishfeed and that chitin-fed fish might be protected against fishpathogenic bacteria In order to determine the optimizedconditions of the chitin additive, multiple conditions havebeen investigated: the mode of chitin administration (Este-ban et al 2001), the size of chitin additive (Cuesta et al

2003) and the dietary chitin supplementation To ourknowledge, a broad range of studies have been conducted,including various fish species and shellfish Although somedifferences exist depending on the conditions of chitintreatment and fish species, generally dietary chitin activatesthe innate immune system of the fish and shellfish tested

Information is also available that chitin-fed organisms wereprotected against pathogens in challenge experiments

Readers with special interest in the applications of chitinare referred to the comprehensive review of Kumar (2000)

The objectives of the present review were to evaluate theavailable information in aquatic animals with regard to theeffect of chitin and krill on growth and feed utilization,bacterial capacity to degrade chitin, modulation of the gutmicrobiota and the effect on the innate immune system anddisease resistance The results cited in the present reviewmight be of importance for future aquaculture, and furtherdirections will be discussed

In order for fish to utilize chitin, fish must be able to digestand utilize it If not, chitin will become an energy sink low-ering the potential energy availability of the fish Further-more, undigested chitin may also affect nutrient utilization

For example, chitin is known to absorb lipid and bile in

Figure 1 Chemical structure of chitin.

.

the GI tract, thus lowering lipid digestion and absorption

(Tharanathan & Kittur 2003) High nutrient utilization is

particularly important in intensive aquaculture practices

with high growth rates as reported for Atlantic salmon and

Atlantic cod

Most fish examined to date do seem to possess some

form of chitin-degrading enzymes such as such as chitinases

and/or chitobiases in their GI tract (Danulat & Kausch

1984; Lindsay 1984, 1985, 1987; Lindsay & Gooday 1985;

Danulat 1986; Rehbein et al 1986; Kono et al 1987b,

1990; Sabapathy & Teo 1993; Moe & Place 1999;

Gut-owska et al 2004; Karasuda et al 2004; Fines & Holt

2010) The location of the enzymes seems to vary between

studies and species, with some finding most activity in the

stomach, while others have more in the intestine There are

also some data indicating that fish feeding on chitin-rich

prey have higher chitinase activity than other fish

(Gut-owska et al 2004) and that feeding chitin-rich diets will

increase enzyme activity (Danulat 1986) However, current

data are rather incomplete and to some extent

contradic-tory

However, possessing chitinase activity does not guarantee

that there will be a complete digestion of chitin At present,

such utilization can only be carried out in real-time

experi-ments In cyprinids, inclusions of small amounts of chitin

will not affect growth rates and feed utilization to any

major extent In golden mahseer (Tor putitora), feeding up

to 2% of diet as chitin has no effect on growth rate,

indi-cating a not-well-developed system for utilization of chitin

(Mohan et al 2009) In other cyprinids, such as snow trout

Schizothorax richardsonii, 2% chitin will actually enhance

growth most likely due to factors other than energy

con-sumption (Mohan et al 2009) In common carp Cyprinus

carpio, chitin has no effect on growth rate as such, but

chitosan, the deacetylated form of chitin, will enhance

growth and feed conversion when included at 1–2% levels

(Victor et al 2004; Gopalakannan & Arul 2006) In hybrid

tilapia (Oreochromis niloticus 9 O aureus), on the other

hand, growth and feed utilization is reduced in a

dose-dependent manner up to at least 10% inclusion (Shiau &

Yu 1999) The same appears to be true for many other

cichlids (Spataru 1978; Spataru & Zorn 1978; Buddington

1980), suggesting a generalized poor ability to utilize prey

chitin In their study with juvenile cobia (Rachycentron

can-adum), Fines & Holt (2010) reported that chitinase activity

was only detected in the stomach As antibiotic treatment

did not reveal differences in chitinase activity, the authors

suggested that the bacterial contribution from chitinolytic

bacteria was not significant

Salmonids also seem to have a low ability to utilize tin In practical experiments, a high rate of replacement offish meal with krill meal containing chitin affects fish per-formance In Atlantic salmon, chitin is not digested to anysignificant extent, and complete replacement of fishmealwith krill meal will reduce growth and lipid utilization (Ol-sen et al 2006) Growth reductions have also been seen inchum salmon (Oncorhynchus keta) fingerlings (Murai et al.1980) and rainbow trout (Oncorhynchus mykiss) fed highlevels of krill meal (Wojno & Dabrowska 1984) It has alsobeen shown that rainbow trout are unable to digest chitin(Buddington 1980) However, 6% chitin supplementationhas been reported to enhance growth in rainbow troutjuveniles (Lellis & Barrows 2000) and may thus indicatethat chitin under some circumstances may be digested andindirectly enhance fish performance

chi-There are few data on the performance of marine fishfed chitin But so far, most data point to good abilitieswith respect to dietary utilization At 10% inclusion level,growth is enhanced in red sea bream (Pagrus major) and to

a lesser extent in Japanese eel (Anguilla japonica) and lowtail (Seriola quinqueradiata) (Kono et al 1987c) Like-wise, chitin is well digested in Atlantic cod (Danulat 1987),and growth studies have shown good growth rates withhigh inclusion levels of chitin-rich krill meal (Moren et al.2006) or up to 5% purified chitin (Ø Karlsen, H Amlund,

yel-E Ringø, A Berg & R E Olsen, unpublished data), cating that chitin is well utilized under intensive conditions

indi-Chitin is completely hydrolysed to its constituent oligomersand monomer by the binary chitinase enzyme system:chitinase (EC 3.2.1.14) and b-N-acetylglucosaminidase (EC3.2.1.30) It can also directly inhibit the growth of a widerange of fungi and bacteria (for review, see Tsai et al.2002)

The observation that chitin does not accumulate in ure has prompted numerous researchers to investigate theprevalence of chitin-degrading bacteria and the rates of chi-tin degradation in the environment To our knowledge, thefirst study reporting chitin ‘destruction’ by bacteria wasconducted by Benecke (1905), who reported the isolation

nat-of Bacillus chitinovorus from the polluted waters nat-of KielHarbour Three years later, Sto¨rmer (1908) reported thatactinomycetes in soil were able to degrade chitin Thebacterial capacity to degrade chitin is widespread amongtaxonomic groups of prokaryotes, including gliding bacte-ria, vibrios, Acinetobacter, Achromobacter, Agarbacterium, .

Chromobacterium, Cytophaga, Photobacterium, Plesiomonas,

Pseudomonas, Flavobacterium, Marinobacter, Micrococcus,

Sulfitobacter, Streptomyces, Listeria, Enterococcus faecalis,

enteric bacteria, actinomycetes, bacilli, carnobacteria,

clo-stridia, marine bacteria and archaea An overview of bacteria

able to digest chitin is presented in Table 1

The general method to evaluate chitinolytic bacteria is

using casein–chitin–agar with the following composition:

0.2% casein, pancreatic digest, 0.8% colloidal chitin and

1.5% technical-grade agar, made up with 75% aged filtered

sea water and pH adjusted to 6.7 Hydrolysis of chitin can

also be measured on ZA agar plates containing 0.5% chitin

and phenolphthalein diphosphate sodium salt (0.01%)

according to the method of Cowan (1974) or using 1/20

PYBG agar plates containing 0.2% colloidal chitin (Itoi

et al.2006)

In the study of Sugita & Ito (2006), the authors reported

that 98.8% of the bacterial strains isolated from the GI

tract of Japanese flounder were chitinolytic, and they were

identified as Vibrio fischeri, Vibrio harveyi and Vibrio

scop-hthalmi–Vibrio ichthyoenteri group The authors also

analy-sed partial sequences of the chiA gene from V harveyi and

V scophthalmi–V ichthyoenteri isolates Sugita & Ito (2006)

did not detect the chiA gene in chitinolytic isolates of

V fischeri,but Ramaiah et al (2000) detected and sequenced

the chiA gene in V fischeri Based on these results, one can

suggest that V fischeri isolates produce various chitinases,

and Sugita & Ito (2006) suggested that the PCR

amplifica-tion technique for the chiA gene might be useful in detecting

chitinolytic bacteria associated with fish GI tract

Studies on fish-associated microorganism carried out in the

1970s, 1980s and 1990s used culture-dependent techniques

of dubious sensitivity, and mostly aerobic and facultative

heterotrophic bacteria were investigated (Cahill 1990;

Ringø et al 1995; Montes et al 1999) At that time,

anaer-obic bacteria and unculturable clones were neglected

(Ka-mei et al 1985; Lee & Lee 1995; Gonzalez et al 1999),

possibly through the shortage of techniques and the

expen-sive nature of the culture-independent techniques available

However, today there is increasing evidence that

micro-organisms occur in large numbers in the alimentary tract of

fresh and marine fish In line with other studies on

micro-bial biodiversity, emphasis has been focussed on

molecular-based culture-independent techniques, which are generating

interesting data and revealed the presence of anaerobes and

uncultured organisms (e.g Moran et al 2005; Clements

et al.2007; He et al 2010; Zhou et al 2011) Furthermore,there are some data available that the gut microbiota offish is more closely related to the food than to the rearingwater (Han et al 2010) Thus, the gut environment pro-vides a specific niche, and bacterial activity in the guts isnot merely a continuum of that observed in the environ-ment This has consequences for the ecosystem as a whole,and the activities and abilities of these distinct bacterialcommunities may contribute uniquely to the nutrientcycling in the system Today, it is well accepted that thegut microbiota in fish are modulated by dietary manipula-tions; to our knowledge, the first study to evaluate theeffect of dietary chitin on gut microbiota was carried out in

1972 (Sera & Kimata 1972) Since then, five studies havebeen carried out investigating the effect of dietary krill andchitin on the culturable heterotrophic gut microbiota ofaquatic organisms, and an overview of these studies is pre-sented in Table 2 One of the studies listed in Table 2 eval-uate the gut microbiota by using denaturing gradient gelelectrophoresis

Atlantic salmon Ringø et al (2006) evaluated the effect ofdietary krill supplementation on epithelium-associated bac-teria in the distal intestine (DI) of Atlantic salmon (Salmosalar L.) In this study, both microbial and electronmicroscopical investigations were included The adherent

DI microbiota was modulated by dietary krill meal as theGram-positive bacteria Carnobacterium maltaromaticum,Microbacterium oxydans, Microbacterium luteolum andStaphylococcus equorumssp linens and the Gram-negativesPsychrobacter spp and Psychrobacter glacincola were notdetected in the DI of fish fed krill meal but were reported

to be present in non-krill-fed fish Transmission electronmicroscopy (TEM) revealed bacteria-like profiles betweenthe microvilli in both rearing groups, indicating an autoch-thonous microbiota TEM evaluations also showed that byfeeding fish a krill meal diet, the DI enterocytes werereplete with numerous irregular vacuoles not detected whenthe fish were fed the control diet Whether the changes ingut microbiota and histology influence fish health was notevaluated and should be investigated in future studies

Red sea bream snapper and crimson sea bream To theauthor’s knowledge, the first study reporting the effect ofchitin on fish gut microbiota was carried out by Sera &

.

Table 1 Bacteria isolated from aquatic organisms and aquatic environment able to digest chitin

Achromobacter labrum (1), A ureasophorum (1), A lipophagum

(1), A hyperopticum (1), Flavobacterium indoltheticum (1),

Pseudomonas cryothasia (1), P subruba (1) and Micrococcus

colpogenes

(1951)

Agarbacterium spp (3), Beneckea lipophaga (1),

Beneckea indolthetica (1) and Pseudomonas sp (1)

Sea bottom deposits of Kisarazu Coast in Tokyo Bay

Kihara & Morooka (1962)

of octopus and squid and intestinal contents of swell fish

Seki & Taga (1963)

bream

Sera & Ishida (1972a)

of red sea bream

Sera & Ishida (1972b)

sea bream

Sera et al (1974)

species

Sakata et al (1978b)

Acinetobacter sp (1), Enterobacteriaceae (4), Flavobacterium sp.

(1), Photobacterium spp (6), Vibrio spp (48) and an unidentified

Gram-negative rod

Rod-shaped, Gram-negative, terminal spherical spores that

swelled the sporangium (8)

Sediment of an estuarine environment

Pel & Gottschal (1986)

bream and Japanese eel

Kono et al (1987a) Vibrio parahaemolyticus (4), V alginolyticus (21), V fluviales (9),

V mimicus (1), Listonella anguillarum (1) and Aeromonas

hydrophila (12)

River and marine water of Tokushima

Osawa & Koga (1995)

Aeromonas caviae, A hydrophila, A jandaei, A sobria and

A veroni

Intestinal digesta of common carp, crucian carp, grey mullet, surface water and sediment samples from the Hikiji River

Sugita et al (1999)

sediment and bottom sediments of lake Jeziorak, Jeziorak Maly and Tynwald

Donderski & Brzezinska (2001)

Cytophaga/Flavobacteria (1), Marinibacter sp (1) and

Sulfitobacter pontiacus (1)

Organic particles of the upper water column of the Equatorial Atlantic

Berkenheger & Fischer (2004)

Enterobacter aerogenes (28%), Aeromonas sp (25%) and

Chromobacterium sp (16%)

Aeromonas hydrophila (46%) and Aeromonas sp (15%)

Oligo-mesotrophic lake Eutrophic lake

Brzezinska & Donderski (2006)

Marinobacter lutaoensis (1), Ferrimonas balearica (1),

Pseudoalteromonas piscicida (1), Enterovibrio norvegicus (13),

Grimontia hollisae (10), Photobacterium damselae spp damselae

(20), P leiognathi (28), P lipolyticum (1), P phosphoreum (39),

P rosenbergii (9), Vibrio campbelli (5), V chagasii (5), V fischeri

(6), V fortis (4), V gallicus (1), V harveyi (77), V natrigenes (15),

V nigripulchritudo (3), V ordalii (18), V parahaemolyticus (33),

V pomeroyi (26), V ponticus (12), V proteolyticus (4), V rumoiensis

(1), V shilonii (2), V tasmaniensis (2) and V tubiashii (24)

Intestinal contents from various Japanese costal fishes

Itoi et al (2006)

.

Kimata (1972) In this study, red sea bream and crimson

sea bream (Evynnis japonica) were fed diets supplemented

with or without 15% chitin The population level of

allo-chthonous heterotrophic and chitin-decomposing bacteria

in the stomach and intestine was investigated after 5, 15

and 25 days of feeding, but the authors did not distinguish

between the different fish species Some variation in the

allochthonous bacterial population level was observed

between the different dietary groups The heterotrophic

population level was generally higher at all sampling points

when fish were fed the chitin-supplemented diet, and the

frequency of gut bacteria able to utilize chitin seemed

higher when the fish were fed chitin compared to those fed

the fish meat paste

Red sea bream and Japanese eel In a study with red sea

bream and Japanese eel (Anguilla japonica), Kono et al

(1987a) compared the effect of a 20% chitin-supplemented

diet with that of fish fed a control diet Red sea bream

were fed the experimental diets for 30 days, and total

via-ble counts and chitin-decomposing bacteria counts were

estimated in the stomach and intestinal contents The

pop-ulation level of chitin-decomposing bacteria in stomach

and intestinal contents of fish fed the control diet was

gen-erally tenfold less than the total viable counts, and this

trend was not changed when chitin was supplemented to

the diet Based on their findings, the authors suggested that

supplementation of chitin had no effect

Crustaceans In a 45-day feeding trial of giant freshwater

prawn (Macrobrachium rosenbergii), Kumar et al (2006)

evaluated the effect of dietary chitin, at 5% and 10%

inclusion, on total viable counts and chitinoclastic

bacte-ria The total viable bacteria count was gradually reduced

as the chitin level increased, but the rate of reduction was

less in natural chitin-fed groups in comparison with the

purified chitin groups The authors suggested that this

may be due to the antimicrobial properties of chitin, but

no evidence was presented No chitinoclastic bacteria were

observed in the gut of any of the treatment groups, which

is in agreement with the findings of Fox (1993), whoreported few chitinoclasts and concluded that the synthesis

of endogenous chitinase in the digestive gland of shrimp(Penaeus monodon) occurs at a slow rate and hence thatjuvenile shrimps are able to digest only small amounts ofdietary chitin in the absence of bacterially producedchitinase

Atlantic salmon Askarian et al (2012) identified 139 chthonous bacterial strains (based on biochemical andphysiological properties) isolated from proximal intestine(PI) and DI of Atlantic salmon fed control or 5% chitindiet Seventy-four of these isolates were further identified

auto-by 16S rRNA gene sequencing The predominant thonous bacteria in the PI and DI of fish fed the controldiet belonged to Staphylococcus, Bacillus and Agrococcus,while Staphylococcus, lactobacilli, bacilli and Acinetobacterwere dominant in the intestine of fish fed the chitin diet Afurther description of the dietary effect of chitin on theautochthonous microbiota is presented in Table 2 Themost promising gut bacteria isolated from the controlgroup, displaying chitinase activity, belonged to the Bacil-lus genera, while gut bacteria showing highest chitinaseactivity when the fish were fed the 5% chitin diet belonged

autoch-to bacilli and Acineautoch-tobacter One surprising observationnoticed in this study was that the lactic acid bacteria(LAB) (only isolated from the GI tract of fish fed the chi-tin-enriched diet) displayed no chitinase activity; however,these LAB were able to inhibit the growth of fish patho-genic bacteria in vitro

Atlantic cod In their study with Atlantic cod (Gadus huaL.), Z Zhou, S He, R E Olsen, B Yao & E Ringø,unpublished data, investigated the autochthonous gutmicrobiota of fish fed diets with or without 5% chitin sup-plementation by using denaturing gradient gel electrophore-sis (DGGE) DGGE was used as this method is reliable,rapid, sensitive and easy to operate From each of the two

mor-Table 1 (Continued)

Atlantic salmon fed ± 5% chitin

Askarian et al (2012)

.

dietary groups (control and chitin-supplemented diet),

sam-ples were taken from the PI and DI of three individual fish

The results clearly displayed that dietary chitin modulated

the intestinal microbiota (Table 2) and that the gut

micro-biota in the dietary chitin treatment was more diverse than

that of the control group Furthermore, the presence of

Escherichia coli-like bacteria, Anaerorhabdus furcosa and an

uncultured bacterium was depressed by chitin in the PI,

while Aliivibrio wodanis was stimulated in the PI by chitin

Analysis of the DI microbiota revealed that uncultured

Spi-rochete-like bacterium, a swine faecal bacterium and an

uncultured bacterium were significantly stimulated by

die-tary chitin

The immunological effects of chitin have been studied in

many different fish and shellfish species However, very

few, if any, reports have been published about the effects

of chitin on the adaptive immune system of fish This may

be due to the inadequate resources to understand the fish

adaptive immune system In mouse systems, however,

chi-tin activities have been studied in much more detail (for

review, see Lee et al 2008) Chitin activates the mouse lung

alveolar macrophages to express cytokines such as IL-12,

tumour necrosis factor-a and IL-18, leading to INF-c

pro-duction mainly by NK cells (Shibata et al 1997) Although

chitin promotes the type I immune responses, it

down-regu-lates type II immune responses of mice (Shibata et al

2001) Chitin induces an immune response characterized by

the infiltration of cells that express IL-4, IL-13, eosinophils

and basophiles, a response that typically has been

associ-ated with allergic and parasitic worm immune responses

(Reese et al 2007)

Recently, chitin receptor molecules on target cells have

been reported A chitin-binding protein was discovered in

rainbow trout (Russell et al 2008) The protein was

homol-ogous to human and murine plasma interlectins and

local-ized in the gill, spleen, hepatic sinus, renal interstitium,

intestine, skin, swim bladder and within the leucocytes of

the rainbow trout Another type of chitin receptor was

identified in mice (Schlosser et al 2009) The chitin

recep-tor is a novel homotetrameric transmembrane protein

encoded by the FIBCD1 gene The receptor protein was

conserved among major vertebrates including fish and was

highly expressed in the GI tract The receptor is involved

in endocytosis through binding acetylated components of

chitin and chitin fragments in a calcium-dependent manner

with a high affinity

The innate immune system is a fundamental defence anism in fish (Magnado´ttir 2006) Innate immunity is theinitial response to microbes that prevent, control or elimi-nate infections of the fish Innate immunity to microbesstimulates adaptive immune responses and can also influ-ence the nature of the adaptive responses to make themoptimally effective against different types of microbes

mech-Components of the innate immune system consist of thelial barriers, phagocytes (neutrals, monocytes, macro-phages and dendritic cells) and circulating proteins (thecomplement system, pentraxins, collectins and ficolins)(Abbas et al 2010) The scales of fish, the mucous surfaces

epi-of skin and gills, and the epidermis act as the primary rier against infection (Ellis 2001) Fish mucus containsimmune components such as lectins, pentraxins, lysozyme,complement proteins, antibacterial peptides and IgM (Alex-ander & Ingram 1992; Rombout et al 1993; Aranishi &

bar-Nakane 1997)

Innate immune cells recognize non-self molecules throughgermline-encoded pattern recognition receptors (PRR) orpattern recognition proteins (PRP) that are present in arelatively low numbers and are vertically transmitted(Kaisho & Akira 2001) The molecules that are recognized

by PRR or PRP of the innate immune cells contain ahighly conserved molecular pattern Two categories of themolecular patterns can induce immune responses byinteracting with PRR or PRP: foreign or pathogen-associ-ated molecular patterns (PAMP) and danger signals thatare released from the damaged host cells and tissues owing

to infection, necrosis and natural cell death PAMP is thecollective term used for highly conserved molecules inmicrobes and is not generally expressed in multicellularorganisms These molecules include polysaccharides (LPS)

in bacterial cell walls, peptidoglycans, bacterial DNA anddouble-stranded viral RNA In contrast, danger signalsinclude the molecules such as the host’s DNA, RNA, heat-shock proteins and other chaperons, and oligomannose ofpresecreted glycoproteins (Matzinger 1998; Elward &

Gasque 2003)

Among the parameters of the innate immune system, thephagocytic, the lysozyme and spontaneous haemolyticactivities and, in some cases, pentraxins have been used todetermine the effects of the inherent or external factors onthe immune system and the disease resistance of fish Suchfactors include immunostimulants, fish diets and feed addi-tives in addition to genetic traits, seasonal factors, environ-mental temperature, pollution, handling and crowding .

stress, probiotics, and the effects of diseases and

vaccina-tion (Magnado´ttir 2006) The activities of phagocytic cells

of the fish treated with immunostimulants such as chitin

can be detected by phagocytosis, killing of pathogens and

cytokine production The killing activity of phagocytic cells

is most important in the elimination of pathogens and can

be determined by measuring the direct cell killing of

macro-phages or the respiratory burst activity, which indicates the

production level of reactive oxygen species, a mediator of

oxygen-dependent killing (Sakai 1999) In this section, the

effects of chitin on the immune system of various species of

fish and shellfish are discussed As several studies have been

carried out on the effects of chitin on the immune system,

growth and survival of various species of fish and shellfish,

an overview of these studies is presented in Table 3

Crayfish Zhu et al (2010) reported the effect of chitin on

the survival and immune reactivity of crayfish

(Procamba-rus clarkia) In general, crayfish fed chitin displayed

ele-vated total hemocyte counts (THCs), prophenol oxidase

and superoxide dismutase activities However, cumulative

mortality of P clarkia challenged with white spot

syn-drome virus was not significantly affected by chitin

Gilthead seabream In a study with gilthead seabream

(Sparus aurata L.), Esteban et al (2000) injected chitin

(0.1 mg of chitin particles< 10 lm in 100 lL PBS)

intrave-nously (i.v.) or with 1 mg of chitin intraperitoneally (i.p.)

and assessed humoral and cellular immune responses at 3,

5 or 10 days postinjection Seabream injected with chitin i

v remained unaffected in their innate immune responses

However, fish i.p injected showed an increase in humoral

and cellular immune responses such as natural haemolytic

complement activity, head–kidney leucocyte respiratory

burst activity, and phagocytic and cytotoxic activities An

assumption as to why chitin i.p injection solely showed an

increase in the fishes’ immune response was suggested to be

of the immunomodulatory activity of chitin on the

macro-phages In fish, similar to mammals, a lectin-type receptor

may be involved in the recognition and attachment of

chi-tin particles to the macrophage membrane, and the

produc-tion of gamma-interferon may trigger the activaproduc-tion of the

macrophage Esteban et al (2001) evaluated the effects of

dietary chitin on the innate immune response of gilthead

seabream Dietary feeding is non-stressful and can act as a

vector to provide the immunostimulant to a larger number

of fish, minimizing cost and effort Seabream were fed 0,

25, 50 or 100 mg chitin kg 1for 2, 4 and 6 weeks and were

analysed for the non-specific modulation of haemolytic

complement activity, leucocyte respiratory burst activityand cytotoxicity Fish fed 25 and 50 mg chitin kg 1showed increased activity of the seabream innate immunesystem Natural haemolytic complement activity and cyto-toxic activity increased after 2 weeks of chitin feeding Therespiratory burst activity and phagocytic activity showed adelayed response with increased activity observed after 4and 6 weeks, respectively However, lysozyme activity andgrowth performance were unaffected by the chitin supple-mentation These studies indicate that the high-chitin dietmay not be beneficial to the innate immunity of seabreamand that the immunological parameters are enhancedsequentially at different time points of post-chitin dietaryprovision Not only the concentration but also the size ofchitin particles has been reported to be important for theactivation of leucocytes (Cuesta et al 2003) Three differentsizes of chitin particles (unfiltered, <10 lm and >10 lm)were prepared and incubated with gilthead seabream head–kidney leucocytes for 1, 4, 24 or 48 h The leucocytes wereable to phagocytose only the chitin particles that weresmaller than 10lm Leucocyte phagocytosis of bacteriawas increased after chitin incubation for 1 or 4 h, while therespiratory burst activity was unaffected Preincubation ofleucocytes with chitin particles enhanced phagocytosis,tumour cell cytotoxicity and respiratory burst

Yellowtail Yellowtail (Seriola quinqueradiata) injectedonly with chitin showed an increase in protection against the

P piscicida challenge until 45 days after the treatment(Kawakami et al 1998) Although chitin increased the non-specific protective immunity in yellowtail, chitin did notshow adjuvant effects, as had been demonstrated in mice andguinea pigs

Rainbow trout In an early study, Sakai et al (1992)reported that rainbow trout injected with chitin showedstimulated macrophage activities and increased resistance

to V anguillarum infection Based on their results, theauthors suggested that chitin has immunostimulating effects

con-of the treated carp were also significantly stimulated Whenthe treated fish were intraperitoneally injected with .

Aeromonas hydrophilaon the 45th day, the relative

percent-age survival was increased up to 40% in the chitin-fed fish

In common carp, dietary intake of chitin enhanced the

innate immune system and resistance against A hydrophila

infection

Rohu Choudhury et al (2005) fed rohu (Labeo rohita L.)

juveniles with a chitin-supplemented diet (25 mg

chi-tin kg 1) for 15 days and recorded the body weights,

respi-ratory burst and resistance to a bacterial pathogen The

chitin-supplemented diet did not influence the body weight

gain and the respiratory burst activity The chitin diet did

not affect the survival of L rohita juveniles against A

hy-drophila infection The authors suspected that the chitin

concentration in the diet might be insufficient to stimulate

the rohu immune system

Tilapia Shiau & Yu (1999) reported chitin depressed

growth of tilapia (Oreochromis niloticus9 O auratus)

regardless of the supplementation level of chitin The group

fed high chitin level showed more weight loss These results

may indicate that the administration of too much of chitin

may cause immunosuppression (Anderson 1992) However,

whether the chitin supplementation in the tilapia diets

tested is optimal is still unclear and merits further

investi-gations

Shore crab The survival rate of shore crabs (Carcinus

ma-enas) that were fed on chitin supplements (a total

concen-tration of 5% or 10% chitin in the crab feed) was greatly

increased (Powell & Rowley 2007) Crabs treated with

chi-tin had more hyaline hemocytes and lower numbers of

cul-tivable bacteria in their hepatopancreas than those on the

basal diet alone However, the enhanced survivability was

not attributable to phagocytic activity of the hemocytes

because the phagocytic activity was unaffected by chitin

supplementation

White shrimp White shrimp (Litopenaeus vannamei)

injec-ted with chitin (4–8 lg g 1

) were challenged with the ogen Vibrio alginolyticus on the second day, and survival

path-was monitored (Wang & Chen 2005) The survival rate of

the chitin-treated shrimp increased up to 75% during the

6 days after the pathogen challenge This resistance against

the pathogen was attributable to enhanced innate immune

activity of the shrimp Total hemocyte count and

respira-tory burst were significantly increased 2 days after chitin

injection Significantly higher phenoloxidase activity was

still maintained after 6 days Phenoloxidase catalyses

phe-nol oxidation, affecting the melanization processes and hostdefence

Artemia nauplii The effect of chitin has also been tested

on Artemia nauplii (Soltanian et al 2007) Chitin particlescombined with a major feed (approximately 2% in dryweight of feed) were fed to Artemia nauplii for 6 days Atday 3, the nauplii were challenged with Vibrio campbellii,and survival was observed for the next 3 days The resultsdisplayed that chitin feeding did not improved survival ofchallenged Artemia

Data on the effect of chitin on performance of fish areinconsistent and to some extent conflicting This may becaused by different origins or treatments of chitin Forexample, there appear to be different effects of chitin andchitosan on fish performance For intensive aquaculture inparticular, performance is one major production parameter

If fish utilize chitin, then inclusions may have interestingadditional effects on fish welfare On the other hand, if chi-tin is not digested, then reduced growth rates or reducedfeed conversion and higher faeces production are expected.These factors may limit the inclusion level into diets unlessoutweighed by other positive effects, for example increasingthe proportion of beneficial gut bacteria involved in welfare

of the host There appear to be major species differences inthe ability to utilize dietary chitin Whether this is an evo-lutionary adaptation to the natural diets remains to beseen In general, cyprinids appear to utilize chitin relativelyeffectively, and in some cases, increased growth has beenreported (Gopalakannan & Arul 2006) Chitin is also wellutilized by many marine fish The latter may be linked totheir natural diet as many fish species, such as Atlanticcod, eat chitin-rich prey like crabs On the other hand, incichlids and pelagic salmonids, chitin is poorly digestedand will normally reduce growth in a dose-dependent man-ner (Spataru 1978; Spataru & Zorn 1978; Buddington1980; Shiau & Yu 1999)

In fish, the effect of krill and chitin as dietary ments on gut microbiota, disease susceptibility and innateimmune parameters has been investigated to a certainextent, but these topics merit further studies One can alsohypothesize that beneficial bacteria colonizing the digestivetract by producing high enzyme activities might contribute

supple-to the fish and aquatic invertebrates nutrition (Harris 1993;Nayak 2010; Ray et al 2012) Whether the gut bacteriashowing high chitinase activity can contribute to nutrition .

is not properly investigated and merits further

investiga-tions However, one has to bear in mind that the presence

of any microorganism within the GI tract does not

necessar-ily signify its functional role Uses of culture-independent

methods are important in gathering information on the

microbial community in the GI tract of fish However,

char-acterization and identification of the microbiota designated

with its functional role should incorporate molecular

meth-ods such as 16S rRNA/26S rDNA sequence analysis

(in case of bacteria and yeasts, respectively) alongside

con-ventional methods as previously suggested by Ray et al

(2012)

Numerous studies of chitin effects on the fish immune

system clearly indicate that chitin can be used as an

immu-nostimulant of many fish species when supplemented in

feed The immune activation induced by chitin can in turn

lead to protection from the infection of pathogens The

studies also suggest that the conditions of chitin

supple-mentation depends on the fish species, and optimization is

very important to maximize its effects and to prevent

possi-ble negative effects such as immune suppression and

reduced growth rates Future studies therefore have to be

addressed for the optimization of the following conditions:

the size of chitin particles, the amount of chitin

supple-mented, the duration of chitin supplementation as well as

standards for the immune parameters that represent the

chitin effect most correctly, and further investigation and

identification of chitin receptors in the target fish As chitin

has been reported to be useful in the control of diseases in

fish culture, we recommend that challenge studies are

included as a gold standard to assess the effect of dietary

chitin on fish health

The composition of the chitin preparation and its route

of administration are important determinants of its effects

in vivo and in vitro Size of the chitin fragment is also a

crucial factor for the effector responses that it elicits In

addition, the type of fish treated with chitin, the innate

immune parameters studied and the pathogenic strains used

in challenge experiments may also contribute to the

vari-ability of the chitin effects

This project was a co-operation project between Norway,

China and South Korea Financial support from Norwegian

Research Council (project no 199812-S40) is gratefully

acknowledged The authors kindly thank Dr Daniel L

Mer-rifield, University of Plymouth, UK, for his constructive

comments on the revised version of the manuscript

Abbas, A.K., Lichtman, A.H & Pillai, S (2010) Cellular and

Saun-ders, Philadelphia, PA.

Alexander, J.B & Ingram, G.A (1992) Noncellular nonspecific

Anderson, D.P (1992) Immunostimulants, adjuvants, and vaccine carriers in fish: applications to aquaculture Ann Rev Fish Dis.,

2, 281–307.

Aranishi, F & Nakane, M (1997) Epidermal proteases of the

Askarian, F., Zhou, Z., Olsen, R.E., Sperstad, S & Ringø, E.

(2012) Culturable autochthonous gut bacteria in Atlantic salmon (Salmo salar L.) fed diets with or without chitin Characteriza- tion by 16S rRNA gene sequencing, ability to produce enzymes and in vitro growth inhibition of four fish pathogens Aquacul-

Ba˚mstedt, U (1986) Chemical composition and energy content In:

The Biological Chemistry of Marine Copepods (Corner, E.D.S.

Berkenheger, I & Fischer, U (2004) Competition for polymers among heterotrophic bacteria, isolated from particles of the

Birkbeck, T.H & Ringø, E (2005) Pathogenesis and the testinal tract of growing fish In: Microbial Ecology in Growing

Else-vier, Edinburgh, UK.

Braconnot, H.M (1811) Recherches analytique sur la nature des

Brzezinska, M.S & Donderski, W (2001) Occurrence and activity

of the chitinolytic bacteria of Aeromonas genus Polish J

Brzezinska, M.S & Donderski, W (2006) Chitinolytic bacteria in

Campbell, L.L & Williams, O.B (1951) A study of

Choudhury, D., Pal, A.K., Sahu, N.P., Kumar, S., Das, S.S &

Mukherjee, S.C (2005) Dietary yeast RNA supplementation reduces mortality by Aeromonas hydrophila in rohu (Labeo rohit-

Clements, K.D., Pasch, I.B.Y., Moran, D & Turner, S.J (2007) Clostridia dominate 16S rRNA gene libraries prepared from the

Cowan, S.T (1974) Cowan and Stee‘s Manual for the tion of Medical Bacteria, 2nd edn Cambridge University Press, London.

Identifica-Cuesta, A., Esteban, M.A & Meseguer, J (2003) In vitro effect of chitin particles on the innate cellular immune system of gilthead

Danulat, E (1986) The effects of various diets on chitinase and beta-glucosidase activities and the condition of cod, Gadus mor-

.

Danulat, E (1987) Digestibility of chitin in cod, Gadus morhua,

Danulat, E & Kausch, H (1984) Chitinase activity in the digestive

Dempsey, A.C & Kitting, C.L (1987) Characteristics of bacteria

Desriac, F., Defer, D., Bourgougnon, N., Brillet, B., le Chevalier,

P & Fleury, Y (2010) Bacteriocin as weapons in the marine

Donderski, W & Brzezinska, M.W (2001) Occurrence of

chitino-lytic bacteria in water and bottom sediment of eutrophic lakes in

Ellis, A.E (2001) Innate host defence mechanism of fish against

Elward, K & Gasque, P (2003) “Eat me” and “don’t eat me”

sig-nals govern the innate immune response and tissue repair in the

CNS: emphasis on the critical role of the complement system.

Esteban, M.A., Mulero, V., Cuesta, A., Ortun˜o, J & Meseguer, J.

(2000) Effects of injecting chitin particles on the innate immune

response of gilthead seabream (Sparus aurata L.) Fish Shellfish

Esteban, M.A., Cuesta, A., Ortun˜o, J & Meseguer, J (2001)

Immunomodulatory effects of dietary intake of chitin on

gilt-head seabream (Sparus aurata L.) innate immune system Fish

Fines, B.C & Holt, G.J (2010) Chitinase and apparent

digestibil-ity of chitin in the digestive tract of juvenile cobia, Rachycentron

Fox, C.J (1993) The effect of dietary chitin on the growth,

sur-vival and chitinase levels in the digestive glands of juvenile

German, D.P & Bittong, R.A (2009) Digestive enzyme activities

Gonzalez, C.-J., Lopez-Diaz, T.-M., Garcia-Lopez, M.-L., Prieto,

M & Otero, A (1999) Bacterial microflora of wild brown trout

(Salmo trutta), wild pike (Esox lucius), and aquacultured

1277.

Gopalakannan, A & Arul, V (2006) Immunostimulatory effects of

dietary intake of chitin, chitosan and levamisole on the immune

system of Cyprinus carpio and control of Aeromonas hydrophila

Groff, J & LaPatra, S (2000) Infectious diseases impacting the

Gutowska, M.A., Drazen, J.C & Robison, B.H (2004) Digestive

chitinolytic activity in marine fishes of Monterey Bay, California.

Han, S., Liu, Y., Zhou, Z., He, S., Cao, Y., Shi, P., Yao, B &

Ringø, E (2010) Analysis of bacterial diversity in the intestine

of grass carp (Ctenopharyngodon idellus) based on 16S rDNA

Harikrishnan, R & Balasundaram, C (2005) Modern trends in

Harris, J.M (1993) The presence, nature, and role of gut

231.

He, S., Zhou, Z., Liu, Y., Cao, Y., Meng, K., Shi, P., Yao, B &

Ringø, E (2010) Effects of the antibiotic growth promoters

flavomycin and florfenicol on the autochthonous intestinal

♂) Arch Microbiol., 192, 985–994.

Itoi, S., Okamura, T., Koyama, Y & Sugita, H (2006)

Chitinolyt-ic bacteria in the intestinal tract of Japanese coastal fishes Can.

Jang, M.-K., Kong, B.-G., Jeong, Y.-I., Lee, C.H & Nah, J.-W (2004) Physiochemical characterization of a-chitin, B-chitin, and y-chitin separated from natural resources J Polym Sci Part A,

42, 3423–3432.

Jeuniaux, C (1961) Chitinases: an addition to the list of hydrolases

Jutfelt, F., Olsen, R.E., Glette, J., Ringø, E & Sundell, K (2006) Translocation of viable Aeromonas salmonicida across the intes- tine of rainbow trout, Oncorhynchus mykiss (Walbaum) J Fish

Kamei, Y., Sakata, T & Katimoto, D (1985) Microflora in the alimentary tract of tilapia: characterization and distribution of

Karasuda, S., Yamamoto, K., Kono, M., Sakuda, S & Koga, D (2004) Kinetic analysis of a chitinase from red sea bream, Pagrus

Kawakami, H., Shinohara, N & Sakai, M (1998) The non-specific immunostimulant and adjuvant effects of Vibrio anguillarum bac- terin, M-glucan, chitin or Freund’s complete adjuvant in yellow tail Seriola quinqueradiata to Pasteurella piscicida infection Fish

Khoushab, F & Yamabhai, M (2010) Chitin research revisited.

Kihara, K & Morooka, N (1962) Studies on marine

Kono, M., Matsui, T & Shimizu, C (1987a) Chitin decomposing bacteria in the digestive tract of red sea bream and Japanese eel.

Kono, M., Matsui, T & Shimizu, C (1987b) Purification and some properties of chitinase from the stomach of Red-Sea bream

Kono, M., Matsui, T & Shimizu, C (1987c) Effect of chitin, chitosan, and cellulose as diet supplements on the growth of cul-

Kono, M., Matsui, T., Shimizu, C & Koga, D (1990) Purification and some properties of chitinase from the stomach of Japanese

Krogdahl, A˚., Hemre, G.-I & Mommsen, T.P (2005) drates in fish nutrition: digestion and absorption in postlarval

Kumar, M.N.V.R (2000) A review of chitin and chitosan

Kumar, P., Sahu, N.P., Saharan, N., Reddy, A.K & Kumar, S (2006) Effect of dietary source and level of chitin on growth and survival of post-larvae Macrobrachium rosenbergii J Appl Ich-

Lee, S & Lee, Y (1995) Identification of intestinal microflora in

Lee, C.G., Silva, C.S., Lee, J.-Y., Hartl, D & Elias, J.A (2008) Chitin regulation of immune responses: an old molecule with

Leisner, J.J., Laursen, B.G., Prevost, H., Drider, D & Dalgaard,

P (2007) Carnobacterium: positive and negative effects in the

Leisner, J.J., Vogensen, F.K., Kollmann, J., Aideh, B., Vandamme,

.

Lellis, W.A & Barrows, F.T (2000) Effect of dietary ingredient

substitution on dorsal fin erosion of steelhead N Am J Aquac.,

62, 135–138.

Li, X & Roseman, S (2004) The chitinolytic cascade in Vibrios is

regulated by chitin oligosaccharides and a two-component chitin

631.

Lindsay, G.J.H (1984) Distribution and function of digestive tract

Lindsay, G.J.H (1985) Development of gastric chitinase activity in

Lindsay, G.J.H (1987) Seasonal activities of chitinase and

chitobi-ase in the digestive tract and serum of Cod, Gadus morhua (L).

Lindsay, G.J.H & Gooday, G.W (1985) Chitinolytic enzymes and

the bacterial microflora in the digestive tract of cod, Gadus

MacDonald, N.L., Stark, J.R & Austin, B (1986) Bacterial

micro-flora in the gastro-intestinal tract of Dover sole (Solea solea L.),

with emphasis on the possible role of bacteria in the nutrient of

Magnado´ttir, B (2006) Innate immunity of fish (overview) Fish

Merrifield, D.L., Dimitroglou, A., Foey, A., Davies, S.J., Baker,

R.T., Bøgwald, J., Castex, M., & Ringø, E (2010) The current

status and future focus of probiotic and prebiotic applications

Moe, C.M & Place, A.R (1999) Characterization of a vertebrate

Mohan, M., Bhanja, S.K & Basade, Y (2009) Performance of

chitin incorporated diet on the indigenous Kumaon Himalayan

fishes: snow trout, Schizothorax richardsonii (Gray) and golden

137.

Montes, M., Perez, M.J & Nieto, T.P (1999) Numerical

taxon-omy of gram-negative facultative anaerobic bacteria isolated

from skin of turbot (Scophthalmus maximus) and surrounding

Moran, D., Turner, S.J & Clements, K.D (2005) Ontogenetic

development of the gastrointestinal microbiota in the marine

597.

Moren, M., Suontama, J., Hemre, G.I., Karlsen, Ø., Olsen, R.E.,

Mundheim, H & Julshamn, K (2006) Element concentrations in

181.

Murai, T., Yagisawa, I., Hirasawa, Y., Akiyama, T & Nose, T.

(1980) Protein, fat and carbohydrate sources of practical diet for

fingerling chum salmon, Oncorhynchus keta Bull Nat Res Inst.

Nayak, S.K (2010) Role of gastrointestinal microbiota in fish.

Odier, A (1823) Me`moire sur la composition chimique des parties

Olsen, M.A., Blix, A.S., Utsi, T.H.A., Sørmo, W & Mathiesen, S.

D (2000) Chitinolytic bacteria in the minke whale forestomach.

Olsen, R.E., Suontama, J., Langmyhr, E., Mundheim, H., Ringø,

E., Melle, W., Malde, M.K & Hemre, G.I (2006) The

replace-ment of fish meal with Antarctic krill, Euphausia superba in diets

Om, A.D., Ji, H., Umino, T., Nakagawa, H., Okada, K., ashi, M & Sakamoto, S (2003) Effect of dietary chitin on lipid metabolism and vitality of black sea bream juvenile Acanthopag-

Osawa, R & Koga, T (1995) An investigation of aquatic bacteria capable of utilizing chitin as the sole source of nutrients Lett.

Ray, A.K., Ghosh, K & Ringø, E (2012) Enzyme-producing teria isolated from fish gut: a review Aquacult Nutr., in press.

bac-Reese, T.A., Liang, H.E., Tager, A.M., Luster, A.D., Van Rooijen, N., Voehringer, D & Locksley, R.M (2007) Chitin induces accumulation in tissue of innate immune cells associated with

Rehbein, H., Danulat, E & Leineman, M (1986) Activities of chitinase and protease and concentration of fluoride in the diges- tive tract of Antarctic fishes feeding on krill (Euphausia superba

Rinaudo, M (2006) Chitin and chitosan: properties and

Ringø, E., Strøm, E & Tabachek, J.-A (1995) Intestinal

Ringø, E., Schillinger, U & Holzapfel, W (2005) Antibacterial abilities of lactic acid bacteria isolated from aquatic animals and the use of lactic acid bacteria in aquaculture In: Microbial Ecol- ogy in Growing Animals (Holzapfel, W & Naughton, P eds),

Ringø, E., Sperstad, S., Myklebust, R., Mayhew, T.M., Mjelde, A., Melle, W & Olsen, R.E (2006) The effect of dietary krill supplementation on epithelium-associated bacteria in the hindgut

of Atlantic salmon (Salmo salar L.) A microbial and electron

Ringø, E., Myklebust, R., Mayhew, T.M & Olsen, R.E (2007) Bacterial translocation and pathogenesis in the digestive tract of

Ringø, E., Løvmo, L., Kristiansen, M., Bakken, Y., Salinas, I., Myklebust, R., Olsen, R.E & Mayhew, T.M (2010) Lactic acid bacteria vs pathogens in the gastrointestinal tract of fish: a

Rivonker, C.U., Abuvarajan, C.R & Sangodkar, U.M.X (1999) Chitin degrading bacteria from the prawn Metapenaeus dobsom

Rombout, J.H.W.M., Traverne, N., van de Kamp, M & Thiele, A.J (1993) Differences in mucus and serum immuno-

Ruiz-Herrera, J (1992) Fungal Cell Wall: Structure, Synthesis, and Assembly CRC Press, Boca Raton, FL.

Russell, S., Young, K.M., Smith, M., Hayes, M.A & Lumsden, J.

S (2008) Identification, cloning and tissue localization of a

.

rainbow trout (Oncorhynchus mykiss) interlectin-like protein that

Sabapathy, U & Teo, L.H (1993) A quantitative study of some

digestive enzymes in the rabbitfish, Siganus canaliculatus and the

Sakai, M (1999) Current research status of fish immunostimulants.

Sakai, M., Kamiya, H., Ishii, S., Atsuta, S & Kobayashi, M.

(1992) The immunostimulating effects of chitin in rainbow trout,

Oncorhynchus mykiss In: Proceedings of the First Symposium

Asian Fisheries Society, Manila, Philippines.

Sakata, T & Koreeda, Y (1986) A numerical taxonomic study of

the dominant bacteria isolated from tilapia intestine Bull Jpn.

Sakata, T., Nakaji, M & Kakimoto, D (1978a) Microflora in the

Sakata, T., Okabayashi, J & Kakimoto, D (1978b) Microflora in

iso-lates and Vibrio standards strains Mem Fac Fish Kagoshima

Sakata, T., Okabayashi, J & Kakimoto, D (1980) Variations in

the intestinal microflora of tilapia reared in fresh and sea water.

Schlosser, A., Thomsen, T., Moeller, J.B., Nielsen, O., Tornøe, I.,

Mollenhauer, J., Moestrup, S.K & Holmskov, U (2009)

Char-acterization of FIBCD1 as an acetyl group-binding receptor that

Seki, H & Taga, N (1963) Microbiological studies on the

Sera, H & Ishida, Y (1972a) Bacterial flora in the digestive tracts

Sera, H & Ishida, Y (1972b) Bacterial flora in the digestive tracts

Sera, H & Kimata, M (1972) Bacterial flora in the digestive tracts

of marine fish I Bacterial flora of fish, red sea bream snapper

and crimson sea bream, fed three kinds of diets Bull Jpn Soc.

Sera, H., Ishida, Y & Kadota, H (1974) Bacterial flora in the

digestive tracts of marine fish In: Effects of the Ocean

Environ-ment on Microbial Activities (Colwell, R.R & Morita, R.Y.

Shiau, S.Y & Yu, Y.P (1999) Dietary supplementation of chitin

and chitosan depresses growth in Tilapia, Oreochromis niloticus

× O auratus Aquaculture, 179, 439–446.

Shibata, Y., Foster, L.A., Metzger, W.J & Myrvik, Q.N (1997)

Alveolar macrophage priming by intravenous administration of

chitin particles, polymers of N-acetyl-D-glucosamine, in mice.

Shibata, Y., Honda, I., Justice, J.P., Van Scott, M.R., Nakamura,

R.M & Myrvik, Q.N (2001) Th1 adjuvant

N-acetyl-D-glucosa-mine polymer upregulates Th1 immunity but down-regulates

Th2 immunity against a mycobacterial protein (MPB-59) in

interleukin-10-knockout and wild-type mice Infect Immunol.,

69, 6123–6130.

Soltanian, S., Franc¸ois, J.-M., Dhont, J., Arnouts, S., Sorgeloos,

P & Bossier, P (2007) Enhanced disease resistance in Artemia

Spataru, P (1978) Food and feeding habits of Tilapia zillii

338.

Spataru, P & Zorn, M (1978) Food and feeding habits of Tilapia

Sugita, H., Yamada, S., Konagaya, Y & Deguchi, Y (1999)

Sugita, H., Mizuki, H & Itoi, S (2008) Prevalence of a fish gen, Listonella anguillarum in the intestinal tract of fish collected

Tajik, H., Moradi, M., Rohani, S.M.R., Erfani, A.M & Jalali, F S.S (2008) Preparation of chitosan from brine shrimp (Artemia urmiana) cysts shells and effects of different processing sequences and the physiological and functional properties of the product.

Tamai, Y., Miyatake, K., Okamoto, Y., Takamori, Y., Sakamoto,

K & Minami, S (2003) Enhanced healing of cartilaginous ries by N-acetyl-D-glucosamine and glucuronic acid Carbohydr.

Wang, S.-H & Chen, J.-C (2005) The protective effect of

204.

Wojno, T & Dabrowska, H (1984) Investigations on the use of meal from krill in feeding the rainbow trout (Salmo gairdneri

Xue, J & Wu, H (2003) On the chitin digestive bacteria from

Zhou, Z., He, S., Liu, Y., Cao, Y., Meng, K., Yao, B., Ringø, E.

& Yoon, I (2011) Gut microbial status induced by antibiotic growth promoter alters the prebiotic effects of dietary DVA-

intestinal bacterial community and non-specific immunity Vet.

Zhu, F.Z., Quan, H., Du, H & Xu, Z (2010) The effect of dietary chitosan and chitin supplementation on the survival and immune reactivity of crayfish, Procambarus clarkii J World Aquac Soc.,

41, 284–290.

.

1 1 2

1 USDA/ARS Catfish Genetics Research Unit, Thad Cochran National Warmwater Aquaculture Center, Stoneville, MS, USA;

2

National Warmwater Aquaculture Center, Mississippi State University, Stoneville, MS, USA

We examined the effects of a yeast-derived protein source

(NuPro) as a replacement for menhaden fish meal on weight

gain, specific growth rate (SGR), food conversion ratio

(FCR), whole-body composition and disease resistance in

juvenile channel catfish (9.9 ± 0.2 g fish)1) NuProreplaced

fish meal at six levels (0, 25, 50, 75, 100 and 125 g kg)1diet)

Catfish were sampled for whole-body composition and then

challenged with the bacterium Edwardsiella ictaluri Growth

performance was negatively affected (P < 0.01) when NuPro

was added at 125 g kg)1diet The amount of whole-body fat

decreased (P < 0.05) when NuProwas added at 75 g kg)1or

more of the diet Regardless of the amount of NuProadded,

survival after challenge with E ictaluri was similar among

treatments Results indicate that up to 100 g kg)1of NuPro

can be added without negatively affecting growth

perfor-mance The yeast-derived protein source used in this study is a

sustainable protein alternative that could be used as a partial

replacement for fish meal in juvenile channel catfish diets

key words: channel catfish, disease, fish meal, growth,

replacement, yeast

Received 16 September 2010, accepted 1 May 2011

Correspondence: Brian C Peterson, USDA/ARS Catfish Genetics

Research Unit, Thad Cochran National Warmwater Aquaculture Center, PO

Box 38, Stoneville, MS 38776, USA E-mail: brian.peterson@ars.usda.gov

The channel catfish Ictalurus punctatus is one of the most

important aquaculture species in the USA Recent difficult

economic times and the price of grains have forced tionists to re-evaluate the formulation of catfish diets Fishmeal was once deemed an essential ingredient for catfish diets

nutri-as well nutri-as other aquaculture species because it wnutri-as highlypalatable and digestible, relatively inexpensive and easy toobtain It is now known that there are drawbacks associatedwith feeding fish meal such as availability, price and fluctu-ations in quality (Borghesi et al 2009) The sustainability offish meal has also come into question with an increase in thegrowth of aquaculture industries worldwide as well as growth

of other competing industries Therefore, efforts havefocused on decreasing the amount of fish meal in the diet andreplacing fish meal with alternative protein sources for bothchannel catfish as well as other aquaculture species

Over the last 35 years, several studies have been conducted

to evaluate the effects of replacing fish meal with plant andanimal protein sources on channel catfish growth Unfortu-nately, the results are not consistent For example, studieswith channel catfish showed that complete removal of fishmeal in the diet reduced fish growth in both fingerlings(Andrews & Page 1974; Mohsen & Lovell 1990; Robinson &

Li 1998; Li et al 2006a,b) and food-size fish (Murray 1982;

Robinson & Li 1998) In contrast, other studies showedthat nutritionally balanced all-plantprotein diets could pro-vide normal growth in pond-raised (food-size) and aquaria-raised catfish (juveniles) (Robinson and Li 1994, 1999;

Webster et al 1992; Reigh 1999; Hedrick et al 2005; Li et al

2010)

Replacements for fish meal protein are often those of plantorigin, especially cereal grains, legumes and oilseeds (Gay-lord et al 2006; Gatlin et al 2007) Soybean meal is one ofthe most widely used alternate protein sources used by

aquaculture, including catfish, because of its broad global

distribution, relative high digestibility, high protein content

and good amino acid profile (Storebakken et al 2000)

Others such as corn gluten meal and cotton seed meal have

been used successfully in replacing fish meal in catfish diets

(Li et al 2010) Although many alternate animal and plant

protein sources are available for use as replacements for fish

meal in catfish diets, for them to benefit aquaculture, they

should be readily available, competitively priced, capable of

being produced in large quantities, contain proper crude

protein levels and balanced amino acid profiles, and not

compromise growth or health of the fish (Hardy & Tacon

2002; Hardy 2004)

Another potential alternative protein source is yeast-based

proteins These proteins have relatively high crude protein

levels and can serve as a source of dietary nucleotides, which

have been shown to promote growth and immune function in

mammals (Uauy 1989; Uauy et al 1990) and fishes (Burrells

et al.2001a,b) Yeast-based proteins have also recently been

used as a partial dietary replacement for fish meal in juvenile

cobia (Rachycentron canadum) (Lunger et al 2006, 2007)

These studies showed that a yeast-based protein could

replace a portion of fish meal protein without detrimental

impacts on weight gain, feed efficiency or fillet composition

(Lunger et al 2006, 2007) Using alternative yeast-based

proteins as a replacement for fish meal protein in juvenile

catfish has not been examined

Growth performance and fish health are predominant

concerns with fish meal replacement The objectives of the

present study were to replace fish meal with a yeast-derived

protein source and examine the effects on growth, body

composition and survival after an immersion challenge with

Edwardsiella ictaluri, the causative agent of enteric

septicae-mia of catfish (ESC)

Juvenile channel catfish (NWAC103 strain) were obtained

from natural pond spawns at the USDA Catfish Genetics

Research Unit, Stoneville, MS, USA Five hundred catfish

(average fish weight, 9.9 ± 0.2 g) were randomly assigned to

six treatments with five replicates each The fish were stocked

into 76-L tanks (25 fish tank)1) and allowed to acclimate for

2 weeks During the acclimation period, the fish were fed the

control diet After the acclimation period, the fish were

anesthetized with 0.1 g L)1tricainemethane sulfonate

(MS-222; Western Chemical Inc., Ferndale, WA, USA) and group

weighed to the nearest 0.1 g The fish were fed their tive diets once per day to apparent satiation Fish weremaintained in 26.9 ± 0.3C flow-through well water and a

respec-14 L:10 D h photoperiod Water quality (pH8.5 and solved oxygen levels >5.0 mg L)1) and flow rates(3.8 L min)1) were similar among tanks The fish were fedexperimental diets for 9 weeks, and the amount of feedprovided was recorded weekly At the end of the 9-weekgrowth study, the fish were weighed as previously described,and two fish per tank were taken and frozen at)20 C forsubsequent proximate analysis Individual fish were minced

dis-in a grdis-inder, freeze-dried (Virtis Unitop 800 L, Garddis-iner,

NY, USA) and re-ground, and crude protein, fat, moistureand ash were determined in duplicate by methods described

by the Association of Official Analytical Chemists tional (2000)

Interna-Two mortalities were recorded (one in a control tank andone in a treated tank) The fish was taken to the FishDiagnostic Laboratory at the Delta Research and ExtensionCenter (Stoneville, MS, USA), and the cause of death wasreported as ÔunknownÕ

Five experimental diets were formulated with the use of ayeast-based protein source called NuPro (Alltech Inc.,Nicholasville, KY, USA) that contained a mixture of nucle-otides, peptides and the contents of the cytoplasm The dietswere isonitrogenous and consisted of a control diet (fish mealadded at 125 g kg)1 diet) and five other diets in whichNuProreplaced fish meal at five levels (25, 50, 75, 100 and

125 g kg)1diet) (Table 1)

The control diet contained high levels of fish meal(125 g kg)1diet) which is not typical of a commercial catfishdiet but was used experimentally to examine the effects ofreplacing fish meal with NuPro The diets were mixed with aHobart mixer and pelleted using a Hobart grinder (Hobart,Troy, OH, USA) The diets were then air-dried, ground byhand, sieved through a screen to obtain the desired size andstored in a freezer ()20 C) prior to feeding

Following 9 weeks of feeding their respective diets, the fish(N = 23 tank)1) were challenged with E ictaluri 1 day afterthe fish were weighed and sampled One control replicate andone of the treated replicates contained only 22 fish per tank

as one fish in each tank died of unknown causes An

E ictaluriisolate from a natural outbreak (confirmed by the

.

Fish Diagnostic Laboratory) was used for the challenge Fish

were challenged with virulent E ictaluri (1.9· 107

cfu mL)1;final concentration) by an in situ bath immersion for 30 min

(Wolters & Johnson 1994) Mortality was recorded daily for

21 days The fish were fed their respective diets during the

challenge Fish were not fed 1 day prior to challenge nor on

the day of the challenge

Studies were conducted in accordance with the principles

and procedures approved by the Institutional Animal Care

and Use Committee, United States Department of

Agricul-ture/Agriculture Research Service Catfish Genetics Research

Unit

Data were subjected toANOVAprocedure utilizing Statistical

Analysis System (SAS) version 9.1 software (SAS Institute

Inc., Cary, NC, USA) Tank served as the experimental unit

for each variable measured When appropriate, data were

also subjected to a Holm-Sidak test for means separation

Differences were considered significant at P < 0.05

At the end of the 9-week growth trial, weight gain, specific

growth rate (SGR) and food conversion ratio (FCR) were

negatively affected (P < 0.01) when NuPro was added at

125 g kg)1diet (Table 2) Weight gain and SGR tended to

decrease, while FCR tended to increase with increasing

amounts of NuPro The amount of whole-body fat

decreased (P < 0.05) when NuProwas added at 75 g kg)1

or more of the diet (Table 3) Protein and ash tended toincrease when NuProwas added at 100 g kg)1 or more ofthe diet but was not statistically different from the othertreatments Regardless of the amount of NuProadded tothe diet, survival after challenge with E ictaluri was similar(P > 0.10) among treatments (Table 2) Survival was 55% inthe control fish, while survival averaged 52.5% in theNuProfed fish

Table 1 Composition of experimental diets

Ingredient (g kg)1)

Control (0)

NuPro  (25)

NuPro  (50)

NuPro  (75)

NuPro  (100)

NuPro  (125)

Omega Proteins, Hammond, LA, USA.

2 Alltech Incorporated, Nicholasville, KY, USA.

3 US Biochemical Corporation, Aurora, IL, USA.

4

Described by Robinson & Li 2007.

5 Provided by L -ascorbyl-2-monophosphate (35% activity).

6 Values are means (dry matter basis, n = 2).

Table 2 Mean weight gain, specific growth rate (SGR), feed version ratio (FCR) and percent survival of channel catfish fed various levels of NuProor control diet

con-Diet (NuPro  g kg)1)

Weight gain (g fish)1) 1 SGR 2 FCR 3

Survival (%) 4

Within columns, values with different letters are different (P < 0.01).

1 Mean initial weight was 9.9 ± 0.2 g fish)1.

2 Specific growth rates were calculated from the formula (ln (BW 2 ) )

ln (BW 1 )/(t)) · 100, where BW 1 and BW 2 are initial and final weights, respectively, and t is feeding period (days).

3 Feed conversion ratios were calculated as ingested food (g)/

As the global demand for fish diet ingredients become limited

in supply and higher priced, sustainable alternatives must be

identified Because of the risks associated with feeding fish

meal to fish (Borghesi et al 2009), much effort has been put

towards finding a plant or animal protein source

replace-ment It is still not clear whether all-plant proteins can

completely replace fish meal and other animal protein

sour-ces in diets of food-size or juvenile channel catfish Several

studies with channel catfish demonstrated that fish meal

could be completely replaced by all plant proteins without

significant reductions in fish growth (Webster et al 1992;

Robinson and Li 1994, 1999; Reigh 1999; Hedrick et al

2005; Li et al 2010) In contrast, other studies have shown

that removal of fish meal in the diet of catfish reduced fish

growth (Andrews & Page 1974; Murray 1982; Mohsen &

Lovell 1990; Robinson & Li 1998; Li et al 2006a)

Differ-ences in results from the aforementioned studies may be

explained by variations in diet composition, amount of feed

provided, fish size or strain, and environmental conditions

(pond versus tank)

The present study demonstrates that fish meal can be

replaced with up to 100 g kg)1 of NuPro without

nega-tively affecting weight gain, FCR or ESC resistance or

susceptibility in juvenile channel catfish The control diet

contained 125 g kg)1diet of fish meal, which is considerably

higher than what is typically contained in commercial

fin-gerling catfish diets In juvenile aquaria-raised catfish,

40 g kg)1 fish meal in the diet is adequate for optimum

growth and feed efficiency (Li et al 2006a,b, 2008) The

intent of the current study was to examine the effects of

replacing fish meal with varying levels of NuPro This is

the first catfish study to show that a yeast protein alternative

could be used as a partial replacement for fish meal in

juvenile catfish diets

Juvenile catfish fed 125 g kg)1 of yeast-based diet wereseverely compromised with respect to weight gain, SGR andFCR One reason for the decreased growth performance inthese fish could be low palatability When all of the fish mealwas replaced with NuPro, feed intake was reducedapproximately 23% compared with control fish In support

of this hypothesis, Lunger et al (2006) also found thatgrowth performance was negatively impacted when juvenilecobia were fed a diet in which all of the fish meal wasreplaced with a yeast-based diet (NuPro) Cobia fed theall-inclusive yeast-based protein diet were observed to havemuch remaining feed in the tank after feeding, suggestingpalatability was poor in that diet (Lunger et al 2006)

In the present study, the amount of whole-body fatdecreased when the yeast-based protein was added at

75 g kg)1 or more of the diet Protein and ash tended toincrease when NuProwas added at 100 g kg)1or more ofthe diet but was not statistically different from the othertreatments In a cobia study, as inclusion rate of NuProincreased, liver lipid and muscle protein decreased, whilemuscle ash increased (Lunger et al 2006) Differences inproximate analysis indices between studies may be related tothe length of the study (6 weeks versus 9 weeks in the currentstudy), tissue differences (muscle and liver versus wholebody) and species differences (carnivore versus omnivore).Another potential attribute of dietary yeasts is theirimmunostimulating properties Many fish studies, includingcatfish, have demonstrated enhanced non-specific immuneactivity feeding dietary yeasts and yeast products (Lara-Flores et al 2002; Olvera-Nova et al 2002; Li & Gatlin 2004;

Li et al 2004a,b; Peterson et al 2010) In the current study,regardless of the amount of yeast-based protein added to thecatfish diets, survival after challenge with E ictaluri wassimilar among treatments It is not known whether NuProhad a direct immunostimulating effect on the immune system

of channel catfish as this was not tested in this study Therewere no apparent negative effects on catfish health as survivalwas 55% in the control fish, while survival averaged 52.5% inthe treated fish In contrast to our study, juvenile cobia fed adiet in which all of the fish meal was replaced by NuProshowed a decrease in survival (Lunger et al 2006) A directimmune response was not tested in the juvenile cobia study,but a heightened level of plasma protein was observed incobia fed the diet containing a portion of dietary proteinfrom NuPro, which may indicate a beneficial immunologi-cal impact on inclusion of NuPro into feeds for cobia(Lunger et al 2006) Differences in survival betweenour catfish study and the cobia study may be explained

by nutritional requirement differences between species

Table 3 Mean whole-body composition indices (g kg)1 dry-weight

basis) of channel catfish fed various levels of NuProor control diet

(omnivore versus carnivore) and the length of the studies.

The carnivorous cobia fed an all-yeast-based protein source

gained 86 g fish)1 over a 15-week growth study, while the

control (100% fish meal diet) gained 512 g fish)1 It is

pos-sible that the severe reduced feed intake and thus growth

decreased the ability of the fish to fight infection and that

resulted in a decrease in survival Weight gain in channel

catfish fed an all yeast-based protein source was 22.5 g fish)1

over a 9-week study, while the control gained 33.9 g fish)1 It

is possible that an additional 6 weeks of feeding an

all-yeast-based protein diet to catfish would have also resulted in a

decrease in survival

Eliminating fish meal, as well as other animal proteins in

catfish diets (both fingerlings and food fish diets), without

any negative consequences to animal performance or health

of the fish would reduce feed cost, increase profitability and

improve sustainability of the catfish industry As fish meal

and the yeast-based protein used in the current study are

similarly priced, it is likely the benefit in the short term would

come from improving sustainability (feeding an ingredient

that is renewable and is thus more environmental friendly)

The organic catfish aquaculture sector is at an infancy

stage as only one study has investigated the effects of organic

diets on the production of catfish (Li et al 2006b) One of the

major impediments to the development of the organic

aquaculture sector is the lack of available organically

certi-fied alternative protein sources (Craig & McLean 2005) The

yeast-based protein used in the current study (NuPro) is an

organically certified alternative protein and could also play a

role in organically grown catfish

In summary, NuPro can be added at 100 g kg)1 diet

without negatively affecting weight gain or FCR in

tank-raised juvenile channel catfish The results also show that

NuPro does not appear to negatively affect disease

sus-ceptibility of catfish challenged with E ictaluri The

yeast-derived protein source used in this study is a sustainable

organic protein alternative that could be used as a partial

replacement for fish meal in juvenile channel catfish diets

The authors thank the assistance of Monica Wood of the

USDA/ARS Catfish Genetics Research Unit for her efforts

in maintaining the fish and in the disease challenge and

Dr Menghe Li for his assistance with the proximate analysis

The authors thank Dr Tyler Bramble and Ginny Stephens of

Alltech, Inc for providing support for this project Mention

of trade names or commercial products is solely for the

purpose of providing specific information and does not imply

recommendation or endorsement by the U.S Department ofAgriculture

Andrews, J.W & Page, J.W (1974) Growth factors in the fish meal component of catfish diets J Nutr., 104, 1091–1096.

Association of Official Analytical Chemists International (2000) Official Methods of Analysis of AOAC International, 17th edn.

AOAC International, Gaithersburg, MD.

Borghesi, R., Dairiki, J.K & Cyrino, J.E.P (2009) Apparent digestibility coefficients of selected feed ingredients for dourado Salminus brasiliensis Aquacult Nutr., 15, 453–458.

Burrells, C., Williams, P.D & Forno, P.F (2001a) Dietary tides: a novel supplement in fish feeds 1 Effects on resistance to disease in salmonids Aquaculture, 199, 159–169.

Burrells, C., Williams, P.D & Forno, P.F (2001b) Dietary tides: a novel supplement in fish feeds 2 Effects on vaccination, salt water transfer, growth rates and physiology of Atlantic salmon (Salmo salar L.) Aquaculture, 199, 171–184.

nucleo-Craig, S.R & McLean, E (2005) The organic movement: a role for NuPro as an alternative protein source In: Nutritional Bio- technology in the Food and Feed Industry (Lyons, T.P & Jacques, K.A eds), pp 285–294 Nottingham University Press, Notting- ham, UK.

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

Aquacult Res., 38, 551–579.

Gaylord, T.G., Teague, A.M & Barrows, F.T (2006) Taurine supplementation of all-plant protein diets for rainbow trout (Oncorhynchus mykiss) J World Aquacult Soc., 37, 509–517.

Hardy, R.W (2004) New developments in alternate proteins Aquac.

Mag., 30, 56–59.

Hardy, R.W & Tacon, A.G.J (2002) Fish meal: historical uses, production trends and future outlook for sustainable supplies In:

Responsible Marine Aquaculture (Stickney, R.R & McVey, J.P.

eds), pp 311–325 CAB International, Oxon, UK.

Hedrick, R.L., Popma, T.J & Davis, D.A (2005) Pond production and fatty acid profiles of the fillets of channel catfish reared on diets with different protein sources NAJA, 67, 304–311.

Lara-Flores, M., Olvera-Novoa, M.A., Guzma´n-Me´ndez, B.E &

Lo´pez-Madrid, W (2002) Use of the bacteria Streptococcus cium and Lactobacillus aciddophilus, and the yeast Saccharomyces cerevisiae as growth promoters in Nile tilapia (Oreochromis nil- oticus) Aquaculture, 216, 193–201.

fae-Li, P & Gatlin, D.M III (2004) Dietary brewers yeast and the probiotic Grobiotic AE influence growth performance, immune response and resistance of hybrid striped bass (Morone chrysops ·

M saxatilis) to Strepococcus iniae infection Aquaculture, 231, 445–456.

Li, P., Wang, X., Hardy, R.W & Gatlin, D.M III (2004a) tional value of fisheries by-catch and by-product meals in the diet

Nutri-of red drum (Sciaenops ocellatus) Aquaculture, 236, 485–496.

Li, P., Lewis, D.H & Gatlin, D.M III (2004b) Dietary nucleotides from yeast RNA influence immune response and resistance of hybrid striped bass Morone chrysops · Morone saxatilis to Strepococcus iniae infection Aquaculture, 231, 445–456.

Li, M.H., Peterson, B.C., Janes, C.L & Robinson, E.H (2006a) Comparison of diets containing various fish meal levels on growth performance, body composition, and insulin-like growth factor-I

of juvenile channel catfish Ictalurus punctatus of different strains.

Aquaculture, 253, 628–635.

.

Li, M.H., Robinson, E.H., Mischke, C.C., Torrans, E.L & Bosworth,

B.B (2006b) Effects of organic fertilization and organic diets on

production of channel catfish in earthen ponds NAJA, 68, 53–62.

Li, M.H., Robinson, E.H., Peterson, B.C & Bates, T.D (2008)

Growth and feed efficiency of juvenile channel catfish reared at

different water temperatures and fed diets containing various levels

of fish meal NAJA, 70, 347–352.

Li, M.H., Robinson, E.H., Bosworth, B.G., Oberle, B.F & Lucus,

P.M (2010) Effects of varying dietary compositions using

com-mon feed ingredients on growth and feed efficiency of pond-raised

channel catfish Ictalurus punctatus (Rafinesque) Aquacult Res, 41,

1133–1139.

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

meal in cobia (Rachycentron canadum) diets using an organically

certified protein Aquaculture, 257, 393–399.

Lunger, A.N., McLean, E & Craig, S.R (2007) The effects of

organic protein supplementation upon growth, feed conversion

and texture quality parameters of juvenile cobia (Rachycentron

canadum) Aquaculture, 264, 342–352.

Mohsen, A.A & Lovell, R.T (1990) Partial substitution of soybean

meal with animal protein sources in diets for channel catfish.

Aquaculture, 90, 303–311.

Murray, M B (1982) Replacement of fish meal with soybean meal in

diets fed to channel catfish at high stocking densities in earthen

ponds MasterÕs Thesis Auburn University, Auburn, AL.

Olvera-Nova, M., Martinez-Palacious, C & Olivera-Castillo, L.

(2002) Utilization of torula yeast (Candida utilis) as a protein

source in diets for tilapia (Oreochromis mossambicus Peters) fry.

Aquacult Nutr., 8, 257–264.

Peterson, B.C., Bramble, T.C & Manning, B.B (2010) Effects of

Bio-Moson growth and survival of channel catfish challenged

with Edwardsiella ictaluri J World Aquacult Soc., 41, 149–155.

Reigh, R.C (1999) Production characteristics of pond-raised channel

catfish Ictalurus punctatus fed diets with and without animal protein

for three growing seasons J World Aquacult Soc., 30, 154–160.

Robinson, E.H & Li, M.H (1994) Use of plant proteins in catfish feeds: replacement of soybean meal with cottonseed meal and replacement of fish meal with soybean meal and cottonseed meal.

J World Aquacult Soc., 25, 271–276.

Robinson, E.H & Li, M.H (1998) Comparison of practical diets with and without animal protein at various concentrations of dietary protein for performance of channel catfish Ictalurus punctatus raised in earthen ponds J World Aquacult Soc., 29, 273– 280.

Robinson, E.H & Li, M.H (1999) Evaluation of practical diets with various levels of dietary protein and animal protein for pond- raised channel catfish, Ictalurus punctatus J World Aquacult Soc.,

30, 147–153.

Robinson, E.H & Li, M.H (2007) Catfish feeds and feeding sissippi Agricultural and Forestry Experiment Station Bulletin No.

Mis-1163, Mississippi State, Mississippi.

Storebakken, T., Refstie, S & Ruyter, B (2000) Soy products a fat and protein sources in fish feeds for intensive aquaculture In: Soy

in Animal Nutrition (Drackley, J.K ed.), pp 120–170 Fed Animal Sci Soc, Savoy, IL.

Uauy, R (1989) Dietary nucleotides and requirements in early life In: Textbook of Gastroenterology and Nutrition in Infancy, 2nd edn (Lebenthal, E ed.), pp 265–280 Raven Press, New York, NY Uauy, R., Stringel, G., Thomas, R & Quan, R (1990) Effect of dietary nucleotides on growth and maturation of the developing gut in the rat J Pediatr Gastroenterol Nutr., 10, 497–503 Webster, C D., Tidwell, J.H., Goodgame, L.S., D 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 channel catfish, Ictalurus punctatus Aquaculture, 106, 301–309 Wolters, W.R & Johnson, M.R (1994) Enteric septicemia resistance

in blue catfish and three channel catfish strains J Aquat Anim Health, 6, 329–334.

.

1 1 1 2

1

Faculty of Sciences, Department of Biochemistry and Molecular Biology I, University of Granada, Granada, Spain;2 Faculty of

Experimental Sciences, Biochemistry and Molecular Biology Section, Department of Experimental Biology, University of Jae´n,

Jae´n, Spain;3 Faculty of Sciences, Department of Animal Biology, University of Granada, Granada, Spain

Maslinic acid (MA) is a natural triterpene that can be used as

an additive in the diet of trout We investigated the effects of

feeding with MA and a fixed ration (10 g kg)1body weight)

with respect to growth, protein-turnover rates and

nucleic-acid concentration in the liver of gilthead sea bream grown

under fish-farm conditions Five groups of 12 g of a mean

body mass were fed for 210 days with diets containing 0

(control), 0.05 and 0.1 g of MA per kg of diet Two groups

were fed ad libitum (control AL and MA100AL), and three

with a fixed ration (control R, MA50R and MA100R) At the

end of the experiment, higher body weights, liver weight,

feed-efficiency and PER were found in MA100AL and

MA100R fish Fractional and absolute protein-synthesis rates

in liver of MA100R fish were higher than in the control,

resulting in a higher absolute protein-accumulation rate and

tissue growth Total DNA content in MA100AL and MA100R

was higher than in control Studies of light and electron

microscopy corroborated these results These findings

indi-cate that MA added to the diet can stimulate growth, hepatic

protein-turnover rates and tissue hyperplasia in gilthead sea

bream

key words: ad libitum and restricted feeding, fish farm,

gilthead sea bream, growth, liver, Maslinic acid, nucleic-acid

levels, protein-turnover rates

Received 2 December 2010, accepted 26 April 2011

Correspondence: Prof Dr Jose´ A Lupia´n˜ez and Dr Eva E

Rufino-Palo-mares, Departamento de Bioquı´mica y Biologı´a Molecular I, Facultad de

Ciencias, Universidad de Granada, Avda Fuentenueva, s/n, 18071

Granada, Spain E-mail: jlcara@ugr.es; evaevae@ugr.es

Maslinic acid (2-a, 3-b-dihydroxyolean-12-en-28-oic acid,MA) is a natural pentacyclic triterpene isolated and purifiedmainly from the surface wax that coats the leaves and fruits

of the olive tree (Olea europaea; Bianchi et al 1994) MA is acompound of 30 carbon atoms grouped in five cycles thathave several chemical substitutes MA presents two hydroxylgroups bound to the carbons 2 and 3, one carboxyl groupbound to the carbon 17, and a double bound between thecarbons 12 and 13 (Fig 1)

This compound has attracted broad interest because of itsproven pharmacological properties and its participation inmany biological properties such as its anti-inflammatorycapacity (Ma´rquez Martı´n et al 2006; Aladedunye et al

2008), anti-viral activity against HIV (Xu et al 1996;

Garcı´a-Granados 1998a; Vlietinck et al 1998), rial effect in relation to Streptococcus (Scalon Cunha et al

anti-bacte-2007) and anti-parasitic activity against Cryptosporidium(Garcı´a-Granados 1998b) Furthermore, MA offers resis-tance to oxidative stress (Montilla et al 2003; Sultana &

Lee 2007; Aladedunye et al 2008), has anti-diabetogenicactivity (Liu et al 2007) and can induce aorta vasorelax-ation in hypertensive rats through an effect involvingendothelial nitric oxide synthase (Rodrı´guez-Rodrı´guez

et al 2006) Recently, some studies have shown that MAinduces selective apoptosis in tumour-cell lines HT-29 andCaco-2 and could be used as a tumour suppressant in coloncancer (Reyes et al 2006; Reyes-Zurita et al 2009) Also, it

is credited with acting on astrocytoma cells (Martı´n et al

2007) This triterpene also influences certain enzymaticactivities, reportedly inhibiting, for example: glycogenphosphorylase (Wen et al 2005, 2006), rising plasma glu-cose induced in diabetic mice by adrenaline (Wen et al

2005), elastase (Sultana & Lee 2007), acyl cholesterol

transferase (Kim et al 2005) and DNA topoisomerase and

RNA polymerase in rabbit muscle (Pungitore et al 2007)

In addition, previous studies have shown that MA

stimu-lates the growth rate in the whole animal, liver and white

muscle when added to the diet of rainbow trout, as it is

capable of stimulating the protein-turnover rates, resulting in

faster protein accumulation (Ferna´ndez-Navarro et al 2006,

2008) However, when MA was added to the Dentex dentex

diet at concentrations of 20, 40 and 80 mg kg)1for 49 days,

no significant differences were found in the

whole-body-specific growth rate (Hidalgo et al 2006)

Gilthead sea bream, Sparus aurata, is currently one of the

most extensively cultured marine fish species in the

Medi-terranean countries (FAO 2007), but is facing an increasingly

competitive market, caused both by falling prices and by the

boom in production of new species Thus, research in this

sector is being directed towards reducing production costs,

by growing this fish in the shortest possible time, enhancing

the quality of the final product and developing technologies

to minimize the environmental impact of fish farms In this

scenario, the most suitable approach for culturing fish would

be the use of appropriate feeding standards aimed not only to

improve economic returns, but also to develop a lasting

cohabitation of sustainable aquaculture and a cleaner

envi-ronment In this sense, available data indicate that, in

gilt-head sea bream, the potential benefits of high-energy diets

can be compromised by unrestricted feeding A diet

provid-ing all the nutritional requirements of the fish and

adminis-tered at lower quantities without affecting growth is one of

the key goals of aquaculture

Considering the above, this work analyses the effect of MA

on the behaviour and nature of liver growth in juvenile

gilthead sea bream The hypothesis of our work is that MA

regulates the protein-accumulation rate of liver by changingprotein-turnover rates (synthesis and degradation) andnucleic-acid concentration of the tissue These effects wouldinfluence the growth dynamics of this tissue and even of thewhole body of sea bream This study was conducted underfarming conditions, so that the results would be extrapolated

to the prevailing conditions of industrial production of thisfish In addition, we have studied the effect of two differentfeeding regimes, to satiation or ad libitum (AL) and the effect

of a fixed ration to 1.0% by fish weight (R) For that pose, we used three diets with different amounts of MA: 0,0.05 and 0.1 g kg)1 diet This indicated, first, the optimal

pur-MA dosage for the gilthead sea bream Secondly, thisinformation was compared with that used in the rainbowtrout, to evaluate the effects of feeding to satiation versusfixed ration Finally, we studied whether the combining ofthe ration restriction and the addition of MA might coun-teract the negative effects on growth that could result from afixed ration

L-[2, 6-3H] Phenylalanine (37 MBq mL)1) was supplied byAmershan Biosciences (Buckinghamshire, UK) L-Phenylal-anine (L-Phe),L-tyrosine decarboxylase, b-phenylethylamine(b-PEA), leucyl-alanine, pyridoxal phosphate and Ho¨echst

33258 came from Sigma (St Louis, MO, USA) All otherchemical compounds were bought from Fluka (Buchs SG,Switzerland) and were of analytical grade MA was provided

by Biomaslinic S.L., University of Granada (Granada,Spain)

This study was conducted according to the national andinternational guidelines on animal experimentation, and theprotocol was approved by the Ethics Committee on AnimalExperimentation of the University of Granada (Spain) Theexperiments were conducted at a local fish farm ƠAzucareradel GuadalfeoÕ located in Salobren˜a, Granada, Spain This

is located at 3644¢N335¢O coordinates, the area has a 95msn altitude (minimum: 0; maximum: 160), and the UTM(Universal Transverse Mercator) coordinate system are

4065449447916 30S The gilthead sea bream coming fromindustrial production tanks was assigned to 15 experimentalpolyester tanks of 1 m3at a rate of 150 fish per tank The fishhad a mean initial weight of 12 g After 2 weeks of adapting

Figure 1 Molecular structure of maslinic acid (2-a

3-b-dihydroxy-olean-12-en-28-oic acid, C 30 H 48 O 4 , molecular mass 472.7 Da).

.

the fish to the experimental conditions, the treatments, run in

triplicates, lasted for 210 days A continuous circuit of

fil-tered seawater (36 g L)1) supply was maintained at a flow of

5 L min)1 The photoperiod was natural (from December to

June), and temperatures ranged between 15 and 21C The

mean oxygen concentration was 6.9 mg L)1

Three different diets were used in the experiment (Table 1)

All diets were formulated from a standard commercial diet

containing 490 g kg)1protein, 150 g kg)1fat and 210 g kg)1

carbohydrate, as described elsewhere (Vergara et al 1996)

The amount of gross energy was calculated assuming an

energy value for protein, lipids and carbohydrates of 19.6,

39.5 and 17.2 kJ g)1, respectively The only difference

between these diets was the MA added The amount of MA

was 0 mg (control diet), 0.05 (MA50diet) and 0.1 g (MA100diet) per kg of diet Crude protein, total lipids, ash andmoisture were analysed by using an Association OfficialAnalytical Chemist method (AOAC 2000) MA was analysedusing the method described elsewhere (Romero et al 2010)

The inclusion of MA did not change the gross energy of thediet (Table 1)

The fish were distributed into five experimental groups asfollows: Two groups were fed to satiety, these being AL,which was control, and MA100 Three groups were fed with afixed ration equivalent to 1.0% of fish weight, these beingrestricted ration (R), which was control, MA50, and MA100.The fish were hand fed twice daily, 6 days a week, and thefood ration was prepared and checked daily too, with specialcare to remove any uneaten food left in the tank

Food intake was recorded daily, and the overall fish weightgain was determined by weighing individual fish at thebeginning and the end of the experimental period Addi-tionally, every 2 weeks, the entire group in a tank wasweighed to verify the steadiness of the growth ratethroughout the experiment The relative daily intake wascalculated by dividing the absolute daily diet intake by themean body weight plotted

The following nutritional indexes were determined: feedefficiency that indicates the food quantity used to growth(FE, g wet weight gain g dry diet intake)1) and protein-effi-ciency ratio that indicates the protein used for growth (PER,

g wet weight gain g crude protein intake)1)

The mortality ratio in the different groups was <5% spect to the initial pool of fish No significant differences werefound among the different treatments

re-At the beginning of the experiment, all the fish were weighedindividually and separated into different tanks, 150 fish pertank In addition, eight fish (different from those used for theexperiment, but from the same initial stock) were initiallysampled, after which their livers were weighed and the liver-protein content was measured The protein concentrationwas determined following the methods of Lowry et al (1951)and Bradford (1976) The initial mean protein values wereused as a reference for the determination of whole-body andliver growth rate (GR) as well as the liver-protein-accumu-lation rate (KG) throughout the experiment The GR wascalculated as the percentage of whole-body and liver weightincrease per day The equation was defined as:

GRð% day1Þ ¼ 100 ðln final weight  ln initial weightÞ days1

Table 1 Composition and analysis of the experimental diets

Control MA 50 MA 100 Ingredients (g kg)1)

Control, MA 50 and MA 100 correspond to the diets without maslinic

acid (MA) and with 50 and 100 mg kg)1 diet, respectively The

vitamin supplement contained (g 300 kg)1): thiamine 0.6,

ribofla-vin 0.9, pyridoxine 0.45, calcium pantothenate 2.25, nicotinic acid

3.75, folic acid 0.225, inositol 15, choline 75, biotin 0.045,

cyano-cobalamin 0.10, ascorbic acid 15, vitamin A 0.00225, 0.001125

vitamin D, vitamin E 3.75, vitamin K 0.375 and sucrose to 300 g of

mixture The mineral supplement contained (g 800 kg)1mixture):

calcium phosphate monobasic 480, calcium carbonate 104,

potas-sium chloride 40, sodium chloride 64, magnepotas-sium sulphate 3.2,

ferric sulphate 24, magnesium chloride 73.6, potassium iodide 0,

32, copper sulphate 0.8, zinc sulphate 3.2, cobalt sulphate 0.8,

sodium selenite 0.0348, 0.16 and sucrose aluminium sulphate to

800 g of mixture MA content was expressed as means ± SEM of 12

samples It was assumed that the energy value of protein, fat and

carbohydrates was 19.6, 39.5 and 17.2 kJ g)1, respectively.

P/E, Protein/Energy; EP/ENP, Energy Protein/Energy Non-Protein.

.

Nine fish per treatment were used Liver KGwas calculated as

a percentage increase in liver protein per day, using the

fol-lowing equation:

KGð% day1Þ ¼ 100ðlnP2 lnP1Þðt2 t1Þ1where P1and P2represent the total tissue-protein content at

times t1and t2, respectively The tissue-protein concentration

was determined according to Lowry et al (1951) and

Brad-ford (1976)

The fractional liver-protein-synthesis rate (KS) was

deter-mined as described by Ferna´ndez-Navarro et al (2008) at

195 days after the beginning of the experiment, in which the

fish were fed 0 and 0.1 g kg)1diets Both in AL as well as R,

MA100 was the MA concentration at which the greatest

effects on growth were detected and thus in these situations

protein-turnover rates and nucleic-acid concentrations were

determined The caudal–vein injection solution contained

150 mM L-Phe including L-[2, 6 3H] L-Phe at 37.0 MBq

mL)1 (100 lCi mL)1) and a specific radioactivity of 1640

dpm nmol)1 The dosage 50 lCi 100 g)1 body weight per

dosage of 0.5 mL 100 g)1body weight represented 5–12.5%

of the blood volume taking into account a blood volume of

4–10% (Jones & Randall 1978)

Two fish were killed 2 min after the injection and seven

were killed after 45 min Immediately afterwards, a median

ventral incision was made to remove visceral mass, after the

liver was separated and weighed, and their protein content

was determined The liver-protein increase was calculated in

each fish by subtracting the average initial values (of a

ref-erence group) from the final protein content Liver samples

were extracted in the cold and then freeze-clamped in liquid

nitrogen The tissue was homogenized (1 : 10 w/v) with cold

0.2 M HClO4(v/v) to precipitate the protein and then

sep-arated into two equal aliquots, one to determine the

protein-synthesis rate and the other to quantify the DNA and RNA

contents After the centrifugation of one fraction for 15 min

at 2800 g for 15 min at 4C, saturated tripotassium citrate

was added to the acid-soluble fraction (SA), and KClO4was

later precipitated at pH 6.3 by centrifugation at 2800 g for

15 min The insoluble protein fraction (SB) was washed twice

with 96% ethanol and once with ether, and the pellet was

hydrolysed in 6 M HCl for 24 h at 110C The HCl was

removed by evaporation, and the amino acids were

resus-pended in saturated sodium citrate, pH 6.3

Amino acids of both fractions, SAand SB, were incubated

with L-tyrosine decarboxylase, at optimal pH 6.3 Thus,

L-Phe was converted to b-PEA, which was determined by

spectrofluorescence using a standard curve of b-PEA (Suzuki

& Yagi 1976) The radioactivity of the fractions was mined by liquid scintillation The results are expressed asspecific radioactivity in dpm nmol)1

deter-KS, expressed as % protein synthesized day)1, was lated as:

calcu-KSð% day1Þ ¼ ½ðSBt2 SBt1Þ=SAðt2  t1Þ½1440=ðt2 t1Þ  100where SBt1 and SBt2 are the protein-bound specific radio-activity at 2 and 45 min, respectively, of the experimentaltime after injection; SA(t2)t1) is the average free pool ofspecific radioactivity over the period; 1440 is the number ofmin in a day

The absolute protein-synthesis rate (AS) was calculated asthe product of KS/100 and the total liver-protein content andexpressed as mg protein synthesized per day The fractionalprotein-degradation rate (KD) was taken as the differencebetween the protein-synthesis (KS) and protein-accumulation(KG) rates, calculated for a period of 43 min and expressed as

a percentage per day Absolute protein-degradation (AD)and absolute protein-accumulation (AG) rates were calcu-lated in a way similar to that used for AS The efficiency ofprotein deposition was calculated as the ratio of the proteinsynthesized versus that retained as growth (KGKS )1· 100)

The RNA concentration was determined by the methoddescribed by Munro & Fleck (1966) and modified by Fern-a´ndez-Navarro et al (2008) This method is based on initialprecipitation of nucleic acids with HClO4 followed byselective basic hydrolysis of the RNA with KOH In the samesamples used for protein-turnover determination, RNA wasseparated, purified and quantified from the second fraction

of liver precipitated with 0.2 M HClO4 The pellet waswashed twice with 0.2 M HClO4where upon the RNA wasseparated from the DNA and protein by basic hydrolysiswith 0.3 M KOH at 37C for 1 h followed by acidification

in 1.2 M HClO4 The supernatant was diluted to 0.6 MHClO4, and the RNA concentration was determined byspectrophotometry at 260 nm, comparing with a RNAstandard curve processed in the same way as the experi-mental samples

The DNA concentration was determined fluorometricallyusing the method described by Labarca & Paigen (1980).The method is based on the enhancement of fluorescenceseen when bisbenzimidazole (Ho¨echst 33258) binds to DNA

In the same fish used for protein-turnover determination,

.

liver homogenates were made in saline phosphate buffer at

1 : 10 w/v A standard curve of salmon-liver DNA was also

made Aliquots of 10 lL of both types of samples were

incubated in darkness with 0.2 mL of Ho¨echst 33258 After

10 min, fluorescence was measured at 450 nm after exciting

at 350 nm

The RNA and DNA concentrations were expressed as

mg g tissue)1 The protein: DNA and RNA : DNA ratios

were calculated together with the total protein, RNA and

DNA contents Because values of the protein-synthesis rate

are largely proportional to RNA concentrations, the

protein-synthesis capacity (CS) can also be defined as RNA : protein

ratio and expressed as mg g)1 Protein-synthesis efficiency

(KRNA) was defined as the amount (g) of protein synthesized

per day, and the RNA unit (g) was calculated as [(KS/

CS)· 10] The protein-synthesis rate DNA unit)1 (KDNA)

was defined as the amount (g) of protein synthesized per day

and DNA unit and was calculated as [(KS/100)· protein/

DNA] (Sudgen & Fuller 1991)

After 195 days of experimental time, livers from four fish per

experimental group were dissected out, cut into 1–2 mm

blocks and fixed for 48 h in BouinÕs fluid (15 : 5 : 1, picric

acid, formalin neutralized with MgCO3 and glacial acetic

acid, respectively) Afterwards, the tissue samples were

embedded in paraffin and cut into serial sections 5 mm thick

These sections were stained with Harris haematoxylin and

1 g kg)1aqueous eosin solution, by Alcian Blue pH 2.5 (AB),

periodic acid-Schiff (PAS; Lillie 1954) and by its combination

(PAS + AB) These stained samples were used to determine

the nature of the filling material of the lumen of the

organelles

For the TEM ultrastructural study, fish-liver sections were

fixed with a solution of 4% glutaraldehyde in 0.1 M

caco-dylate buffer pH 7.4 (Watson 1958) Next, a new section was

made and was again fixed in a solution of 1.5% osmium

tetraoxide in cacodylate buffer Next, it was dehydrated with

acetone and finally embedded in Epon 812 (Burke &

Geis-elman 1971) Ultrathin sections 700 A˚ thick, cut with a

Reichert-Jung ultramicrotome, were contrasted with uranyl

acetate and lead citrate (Reynolds 1963) The selection of

these ultrathin sections from different regions of liver was

based on previous observations of semithin sections 1 lm

thick stained with toluidine blue

Data are shown as mean ± the standard error of the mean(SEM) The effects of the experimental diets on the differentparameters have been examined using the two-way analysis

of variance (ANOVA) The statistical significance of differentialfindings between experimental groups and controls wasdetermined by StudentÕs t-test using SPSSversion 15.0 (IBMCorporation, Somer, NY, USA) for Windows softwarepackage P-values smaller than 0.05 were considered statis-tically significant

Whole-body weight, final liver weight, relative daily intakeand feed-conversion parameters were monitored for 210 days

in five groups of gilthead sea bream fed with 0, 0.05 and 0.1 g

of MA per kg of diet The results are shown in Tables 2 and 3

The fish fed with MA increased in body weight with respect tocontrol A significant effect on growth was observed in the fishfed with the MA100diet, as the whole-body weight was some15% higher in the MA100AL group and 5% in the MA100Rgroup In addition, a significant increase in total liver weightwas found in the fish fed with MA100diet In the case of theliver, fish of MA100AL increased a 21% compared to control

AL, and a 32% in the group MA100R with respect to itscontrol R Notwithstanding, in fish fed the restricted diets, thedifferences in liver growth rate (GR) were found to be only a13% higher in MA100AL versus control AL (Table 2) Theaddition of MA to the diet showed no significant changes in therelative food intake of the different dietary nutrients betweenthe AL and R groups The values of the FE and protein-efficiency ratio (PER) increased slightly in the fish fed with

MA The rise, in the case of FE, was 11% for the group of fishfed with MA100AL diet with respect to control AL, and 33%

for the group fed MA100R over control R With regard to thenutritional index, PER, the increase was 19% for the groupfed MA100AL in relation to control AL, without changes inrestricted conditions (Table 3)

The concentration of total protein and total RNA in theliver of gilthead sea bream of the different experimentalgroups is shown in Fig 2 The values of total protein

.

showed significant changes between the experimental

groups fed AL and fixed ration; the value of MA100AL was

21% higher than control AL and the value of the group

MA100R 30% being higher than the control R MA added

to the diet boosted the total RNA content (mg), this rise

being 46% in fish fed MA100AL diet AL and a 78% in the

group MA100R, compared with its respective control

groups

Tissue growth involves two kinds of processes,

hyper-trophy or larger cell size, and hyperplasia or greater cell

number Liver growth, expressed as total DNA content andprotein: DNA ratio, is shown in Fig 3 After 195 days ofexperiment, the total DNA content, which indicates thenumber of total cells, increased by 45% and 82% for fishfed with MA100AL and MA100R, respectively, compared tovalues of the control AL and R groups The pro-tein : DNA ratio, indicating cell size, declined in the groups

of animals fed with MA added to the diet, under bothconditions, AL and R These results indicate that MAstimulated DNA synthesis and, therefore, the formation ofthe new cells

Table 2 Body and liver growth in gilthead sea bream fed with

maslinic acid (MA) and different rations

Ad libitum feeding (AL) n

MA 50 and MA 100 , group of fish fed with 0.05 and 0.1 g MA kg)1

diet, respectively The results are expressed as the mean ± SEM of n

data The experimental time in all cases was 210 days The terms

ÔinitialÕ and ÔfinalÕ indicate the average initial and final mass of all

experimental fish The results were analysed by a two-way ANOVA

followed by StudentÕs t-test Probabilities of P < 0.05 or less were

considered statistically significant For each parameter, for

com-parison between different feeding rations (AL versus R), the data in

each row followed by different superscript (a, b) are statistically

different For comparison between different feeding diets (Control

versus MA 100 ), the data in each column followed by different

subscript (x, y) are statistically different HSI, ratio of liver weight to

body weight.

Table 3 Food intake and nutritional indexes for gilthead sea bream fed with maslinic acid (MA) and different rations

Ad libitum feeding (AL) n

Fixed ration (R) n Relative daily ingestion

Diet (g kg fish)1) Control 1.21 ± 0.05 a

.

The effects of the different experimental conditions on

protein-turnover rates in the liver of gilthead sea bream are

shown in Table 4 In the two experimental groups fed MA,

KSincreased at the end of experimental period The rise was

81% and 191% in the groups MA100AL and MA100R with

their respective control groups The pattern of KD showed

the same trend for KS The value of control R was the lowest

of all the groups On the other hand, KGdid not significantly

change between groups after 210 days

The values of CS, KRNA, KDNA and PRE are detailed in

Table 3 CSvalues were significantly higher by the addition of

MA to the diet This increase was 34.7% for the MA100AL and42.3% for MA100R compared with their respective controls

The KRNA also increased in animals fed MA The valuereached in the MA100AL was 34.4% higher than control AL,while the MA100R was 105% higher than for the control Rgroup Moreover, the value of the MA100R group was signif-icantly (33%) higher than the corresponding MA100AL

Also, the values of KDNA were higher for two groups fed

MA The values of MA100AL and MA100R were higher by36% and 99%, respectively, compared to their controls Inturn, animals fed MA100R showed an increase of 34%

compared to MA100AL group No differences were foundbetween the control AL and control R groups In the case ofPRE, the control values were higher than the MA group

01234

0102030405060

Data were treated with a two-way ANOVA followed by StudentÕs t-test Probabilities of P < 0.05 or less were considered statistically significant Letters a, b above the bars indicate statistical differences between different feeding ration (ad libitum versus fixed ration), and letters x, y above the bars indicate statistical differences between treatments (control versus MA 100 ).

Figure 2 Total liver protein and RNA contents in gilthead sea bream

fed different doses of maslinic acid (MA) and ration The results are

expressed as the means ± SEM of eight observations Data were

treated with a two-way ANOVA followed by StudentÕs t-test

Proba-bilities of P < 0.05 or less were considered statistically significant.

Letters a, b above the bars indicate statistical differences between

different feeding ration (ad libitum versus fixed ration), and letters

x, y above the bars indicate statistical differences between treatments

(control versus MA 100 ).

.

In addition, the absolute protein-turnover (AG, AS, AD)

rates were also determined in the liver of these fish These

parameters are shown in Fig 4 The value of AS for the

MA100AL group was 97% higher than control AL and a

257% higher in the MA100R than control R The values of

ADshowed the same trend as the relative values, reflecting a

higher protein-degradation rate for the groups fed with MA

The MA100AL group increased 109.7% compared with the

control AL group and the MA100R 289% respect to control

R In turn, the result of the control AL group was 29%

higher than the control R group MA increased the values of

AG too Values of MA100AL group increased 24% with

respect to the control AL group, and values of MA100R

group were 44% higher than control R

The hepatic structure of the gilthead sea bream from different

experimental groups was analysed using both optical and

transmission-electron microscopy The results are shown inFigs 5–7 In the images from optical microscopy in all exper-imental groups, liver tissue showed no signs of tissue damage

0.0 0.5 1.0 1.5 2.0 2.5 0 10 20 30 40 50 60 0 10 20 30 40 50 60

Table 4 Hepatic protein-turnover parameters and protein-retention

efficiency of gilthead sea bream fed with maslinic acid (MA) and

different rations

Ad libitum feeding (AL) Fixed ration (R)

K S , protein-synthesis rate (% day)1)

K DNA , protein-synthesis rate/DNA unit, g protein

synthesized day)1g DNA)1

MA 100 , group of fish fed with 0.1 g MA kg)1 diet Results are

means ± SEM of seven data The results were analysed by a

two-way ANOVA followed by StudentÕs t-test Probabilities of P < 0.05 or

less were considered statistically significant For each parameter,

for comparison between different feeding rations (AL versus R), the

data in each row followed by different superscript (a, b) are

sta-tistically different For comparison between different feeding diets

(Control versus MA 100 ), the data in each column followed by

dif-ferent subscript (x, y) are statistically difdif-ferent.

.

Evidence of hepatopancreas in the liver of these fish was

examined in both control and in the groups fed with MA

Structural details of the hepatopancreas found in the liver all

experimental groups of sea bream are shown in Fig 5 The

exocrine pancreas or hepatopancreas were arranged by

pancreatic acini and separated from the hepatic parenchyma

by a thin layer of connective tissue Pancreatic cells were

arrayed around of a branch of the portal vein, separated by

a basal membrane and reticular fibres Hepatopancreas cells

had a special organization in which opposing layers of cells

were divided by a canaliculus – that is, cells of the inner zone

were in contact with the basement membrane of the vein, and

cells of the outer zone were in contact with the outer layer

connective tissue Exocrine cells were tall and columnar, with

a spherical nucleus (Fig 5) Panel A (10·) shows the liverparenchyma with infiltrations of the exocrine pancreas, withpancreatic cells infiltrating between the hepatic sinusoids

Panel B (40·) shows in detail the structure of the pancreas, and Panel C (40·) presents a cross-section of thesame structure, displaying the two cell layers that constitute

hepato-it The thin layer of cells rests on the basal membrane of thecentral vein (V) Also visible between cells is an arteriole (A),and a mass of endocrine cells that might be similar to the islet

of Langerhans in mammals Moreover, a secretion duct (SD)related to the elimination of secretion products

The effect of MA and restricted food on liver structure ofgilthead sea bream is shown in Fig 6 This figure shows aneosin–haematoxylin staining section of hepatic parenchyma

SD A

CT V EC H

H

Figure 5 Liver general structure of liver

of Sparus aurata by optical microscopy.

Photomicrograph of liver parenchyma with sinusoid hepatic and pancreas infiltrations (panel a and panel b).

Cross-section of the exocrine pancreas displaying two cell layers (EC: exocrine cell) surrounded by connective tissue (CT: connective tissue), central vein (V:

vein), arteriole (A) and peripheral secretion duct (SD: secretion duct) (panel C).

of MA treatment is characterized by higher numbers of hepatocytes in

MA 100 AL and MA 100 R treatments than

in the respective controls V: vein.

.

in the four experimental groups The most notable difference

between groups is the number of liver cells In the MA100AL

and MA100R groups (panel B and D, respectively),

hepato-cytes were more numerous and showed a greater degree of

compaction of the parenchyma than those observed in the

control AL and R groups (panel A and C, respectively) The

sizes of the cells and nuclei were also the same in all

treatments

The hepatic ultrastructure was studied by TEM The

results are shown in Fig 7 In the control AL and MA100AL

groups, the number and size of individual organelles showed

no significant changes The most distinguished difference

between groups was the morphology of the secondary

lyso-somes Fish fed with MA presented lysosomes with the inside

membranes occupying nearly the entire interior, while these

membranes did not appear in the control groups

The aim of this study was to determine whether MA could

stimulate or regulate liver protein-turnover and growth rates

in young gilthead sea bream maintained under similar

con-ditions to a fish farm For this, we used two different food

regimens, AL and fixed ration, to establish which of them

could improve growth

One factor that affects growth in fish is the amount of

food ingested (in terms of energy and nutrients) In practice,

aquaculture normally selects individuals with a high intake

capacity, but not necessarily with high feed-conversionefficiency, and it is well known that the latter is essential forgood growth of this fish Although we might infer that if thefish eats more, it should grow more, we also need to takeinto account the quality of this growth That is, we need todetermine whether different kinds of diets can improve thefish-culture efficiency According to some authors, thefeeding of gilthead sea bream to satiety does not necessarilyensure the best conditions for growth (Robaina et al 1999;Mingarro et al 2002; Sitja-Bodabilla & Alva´rez-Pellitero2003) These latter authors investigated two differentstocking times and the effect of the long-term food restric-tion on the pathological and immunological status of fishreared under laboratory conditions, and their results re-vealed that by manipulating ration size, they could vary theingested amount of some ingredients Nevertheless, theyfailed to demonstrate significant differences between feedingregimes However, other works have shown the effect of dietcomposition on fish immune system and health status(Robaina et al 1999; Montero et al 2001; Ortun˜o et al.2001)

The gilthead sea bream in its natural state has developedcompensatory growth, which, as the term implies, involvescompensating periods of food with the restriction stages.Under culture conditions, the natural growth has alteredthose feeding cycles (feeding and restriction) can be masked,because food is ensured in excess to decrease the fatteningtime (Pe´rez-Sa´nchez 2000) Under fish-farming conditions,

Figure 7 Effect of maslinic acid (MA) and a fixed ration on liver by ultrastructure microscopy Number and size of individual organelles was the same in both treatments ad libitum (AL) (control: panel a, b and c; MA 100 : panel d, e and f ) Morphology of lysosomes in control AL and

MA 100 AL groups (panel c and f, respectively).

.

the feeding-optimization patterns need to be beneficial to the

animals, which grow better with a less risk of disease, and

also beneficial to the producers and the consumer, to ensure

that the final product has optimum quality

Some authors suggest that a diet to improve the growth of

gilthead sea bream under farming conditions needs more

precise food-intake regulation, and in this sense, a viable

solution might be a restricted diet (Mingarro et al 2002;

Go´mez-Requeni et al 2004) These researchers compare the

effect of feeding to satiety in these fish with a restricted diet

administered at three different developmental stages, and the

results found were that the animals under restricted feeding

had, in general, a better immune status, lower mortality rate,

fewer infections and fewer pathogenic alterations that those

fed to satiation (Mingarro et al 2002) Our results suggest, in

general, that food restriction at 1.0% of animal weight

pre-sents better values in all parameters of protein-turnover rates

and nutritional indexes studied than those found in groups

under AL feeding

Besides examining the effect of a restricted ration of food

on growth and related parameters in farmed gilthead sea

bream, we also studied the consequences of MA added to the

standard diet of these fish This natural compound has been

used as an additive in other fish species such as rainbow trout

(Ferna´ndez-Navarro et al 2006, 2008) and D dentex

(Hi-dalgo et al 2006)

Maslinic acid acts as a factor that stimulates the growth

rate in the liver and white muscle of trout under laboratory

conditions The results of these studies indicate that the

intake of diet containing MA stimulated the growth rate

(GR), increased the liver and white-muscle weight, while the

protein-efficiency ratio (PER) indicated that protein was best

used for growth Furthermore, these studies showed that MA

stimulated DNA synthesis, total RNA and protein content in

both tissues In addition, the intake a diet with 0.025 and

0.25 mg kg)1 of MA stimulated the protein-accumulation

rate in liver and white muscle mainly as result of regulating

protein-synthesis rates (Ferna´ndez-Navarro et al 2006,

2008)

In our study, on the species gilthead sea bream, the fish

differed in biology, nutritional habitats and culture

charac-teristic with respect to trout rainbow In this work, we

checked whether MA stimulated both the growth and the

protein-turnover rates of this fish species under these

exper-imental conditions, making the results much more applicable

to the large-scale farming of these fish

The major gain in growth under these conditions would

represent a noteworthy boost in fish production and could

justify its use as a feed additive in other fish species, or other

animal species, including humans (Ferna´ndez-Navarro et al

2010) If these results were reproducible in humans, dietarysupplementation with MA might improve the protein meta-bolic state of patients affected with pathologies, for examplecachexia, which causes a serious loss of muscle mass Also,

we studied and compared the effect of MA addition to thestandard diet of these fish under two feeding conditions, ALversus food restriction

It is clear that when the fish fed with a diet that providesnutrients and energy below the optimal requirements, the use

of food under these conditions has strong interest for severalreasons First, for the use of a restricted ration, it is appro-priate to evaluate the regulation of different metabolicpathways, so that during a situation where the nutrientsupply is slightly below the requirements for optimal growth,certain pathways will be activated and others inhibited tosecure basic cellular functions Secondly, the effects of acompound capable of influencing cell growth will be moreapparent in restriction situations Thirdly, the combinedeffects of the ration restriction and the addition of MA couldcounteract the effects of food restriction

The results of our study show that MA induced a significantgain in final weight, both in absolute terms (g) as well as

g day)1 Moreover, MA intake significantly boosted liverweight, the total RNA content and total protein content ofthe tissue With respect to the nature of growth, the intake of

MA increased the total DNA-content index, augmenting thenumber of cells without prompting significant changes in cellsize These effects were most evident in the case of restrictedfeeding ration The results are consistent with those reportedpreviously in the liver of trout (Ferna´ndez-Navarro et al

2008), suggesting that MA intake accelerated the rate of celldivision, producing a larger number of detected cells and thus

a weight gain This increase in cell division would be tained by a stimulation of DNA synthesis, RNA and proteins

main-The values found for fractional rates of protein lation (KG, 1.09 ± 0.08% day)1), synthesis (KS, 14.91 ±1.00% day)1) and degradation (KD, 13, 82 ± 1.11% day)1)

accumu-in the liver of control gilthead sea bream fed AL are similar

to those reported for other marine species such as codfish(Houlihan et al 1988), or freshwater species such as trout(Perago´n et al 1999) at different developmental stages

The protein-turnover rates, synthesis and degradationfound in S aurata reached levels higher than those of troutand codfish at certain stages of development In the case ofjuvenile or adult trout, the KSvalue reaches about 8% day)1and KD of 4–6% day)1(Perago´n et al 1998) The high KSand KD values found in our fish can be explained by con-sidering the developmental stage The effect of development

.

on protein-turnover rates in the liver of trout (Perago´n et al.

1998) reveals high KSand KD values during early

develop-mental stages, particularly in trout of 7.4 and 43.5 g of body

weight In the case of the trout of 43.5 g in weight, the KSand

KDvalues were 15.94 and 12.26% day)1, respectively, these

values being similar to those described in the present work

for gilthead sea bream of 76.31 g in weight

Growth study is important in gilthead sea bream in early

developmental stages when the growth is higher, because, at

this time, all factors or compounds that influence the body

growth may have a greater effect This work together with

that of Sierra (1995) is the only ones to date that determine

the protein-turnover rates in gilthead sea bream liver The

values of the parameters in both works are very similar, and

in both cases, the protein-turnover rates in the liver are high

We found high values in the liver protein-turnover rates

(KS and KD) in both food regimens studied, AL and R,

although, in general, the values found in the control group R

were lower than the values found in the control AL; however,

the addition of MA increased the values in the restricted

situation even above of that reached in the fish fed AL with

MA In addition, the values of other parameters related to

the protein-turnover rates (CS, KRNA, KDNAand ERP) in the

liver of gilthead sea bream were relatively high CS and

KDNA, as well as KS, were significantly higher in the sea

bream than those found in trout fed with different quantities

of MA (Ferna´ndez-Navarro et al 2006), whereas KRNA

values were similar in both species (Ferna´ndez-Navarro et al

2006) This result indicates that high protein-synthesis rates

in the liver of gilthead sea bream correlated with high

capacity for protein synthesis, with a change also in

trans-lation efficiency The protein-retention efficiency (PRE) was

significantly lower than reported for trout

(Ferna´ndez-Navarro et al 2006), a finding that agrees with the relatively

low values of KG

Gilthead sea bream weighing about 66 g of weight, fed AL

with a control diet, have a liver with high protein-turnover

rates, a relatively low protein-retention and -accumulation

efficiency On the other hand, feeding a restricted diet

produced a low turnover rate, whereas the

protein-retention and -accumulation efficiency were similar to those

observed with feeding AL

Diets with added MA throughout the experimental time

produced a significant increase in the values of KS, KD, CS

and KRNA, KDNA, while KGvalues remained unchanged and

PRE decreased In short, MA stimulated the

protein-turn-over rates (synthesis and degradation) without altering the

protein-accumulation rate The RNA concentration and

protein-translation efficiency also rose The increase described

in the protein-synthesis rate reached values of 26.98 ±1.22 and 31.09 ± 3.66% day)1, respectively All resultsfound in relation to liver growth and protein-turnover ratesare consistent with those observed in the microscopy studies,even though the animals fed with MA showed a bettermetabolic status and a higher number cells

On the other hand, the behaviour of the KG and ERPshows that the increase in protein turnover is not allocated toaccelerate the growth rates or protein accumulation in thistissue, but rather to increase protein synthesis and exportfrom the liver to the rest of the fish (Ferna´ndez-Navarro

et al.2006) The liver is a regulatory organ that maintains themetabolic homoeostasis of many compounds in the synthesis

of exportable proteins that perform their function in otherplaces, and responsible for the regulation of fuel levels andcell debris (Henderson & Sargent 1981)

The high values of RNA and DNA concentration indicatethe existence of an active metabolism of these two biomole-cules MA stimulates protein turnover in this tissue andhigher levels of RNA and DNA In the case of the restrictedfeeding ration, the increases observed in the gilthead seabream fed the diet with MA are even greater, reaching KSand KDvalues almost triple those of control, thus indicatingthat the liver-protein metabolism is remarkably stimulated

by MA intake

In conclusion, our results shown that MA can be used as afeed additive to stimulate growth and protein-turnover ratesboth in freshwater fish as well as marine fish grown either inthe laboratory or in semi-industrial production This com-pound can stimulate protein synthesis and hyperplasia pro-cesses without damaging hepatic tissue; nevertheless, thesecells become biosynthetically more active This propriety of

MA may have applications in other fish species, and bly in other animal species, including human beings (Fern-a´ndez-Navarro et al 2008) Currently, we are investigatingthe effect of MA on the Jun N-terminal kinases pathway,which is involved in the molecular mechanism of cell growth

proba-This work was supported by the research grants BIO-157from the Plan Andaluz de Investigacio´n, Junta de Andalucı´a,Spain and a research contract the OTRI (University ofGranada), ƠAzucarera del GuadalfeoÕ S.L & ƠBiomaslinicÕS.A The authors thank Drs Ramo´n Carmona and SusanaCamacho, from Department of Cellular Biology, University

of Granada and the Scientific Instrumentation Centre of theUniversity of Granada (SIC-UGR) for his technical assis-tance in the microscopic studies

.

Aladedunye, F.A., Okorie, D.A & Ighodaro, O.M (2008)

Anti-inflammatory and antioxidant activities and constituents of

Platostoma africanum P Beauv Nat Prod Res., 22, 1067–1073.

AOAC (2000) Official Methods of Analysis Association of Official

Analytical chemist, Gaithersburg, Maryland, USA, p 1018.

Bianchi, G., Vlahov, G & Pozzi, N (1994) The lipids of Olea europaea.

Pentacyclic triterpene acids in olives Phytochemistry, 37, 205–207.

Bradford, M (1976) A rapid and sensitive method for the

quanti-tation of microgram quantities of protein using the principle of

protein dye-binding Anal Biochem., 72, 248–254.

Burke, C.N & Geiselman, C.W (1971) Exact anhydride epoxy

percentages for electron microscopy embedding (Rpon) J

Ultra-struct Res., 36, 119–126.

FAO (2007) FAO Yearbook Fishery Statistics Aquaculture

Pro-duction 90, 2, Rome, Italy.

Ferna´ndez-Navarro, M., Perago´n, J., Esteban, F.J., de la Higuera,

M & Lupia´n˜ez, J.A (2006) Maslinic acid as a feed additive to

stimulate growth and hepatic protein-turnover rates in rainbow

trout (Oncorhynchus mykiss) Comp Biochem Physiol C Toxicol.

Pharmacol., 144, 130–140.

Ferna´ndez-Navarro, M., Perago´n, J., Amores, V., de la Higuera, M.

& Lupia´n˜ez, J.A (2008) Maslinic acid added to the diet increases

growth and protein-turnover rates in the white muscle of rainbow

trout (Oncorhynchus mykiss) Comp Biochem Physiol C Toxicol.

Pharmacol., 147, 158–167.

Ferna´ndez-Navarro, M., Perago´n, J., Esteban, F.J., Amores, V., de

la Higuera, M & Lupia´n˜ez, J.A (2010) Maslinic acid: a

compo-nent of Olive oil on growth and protein-turnover rates In: Olives

and Olive Oil in Health and Disease Prevention (Preedy, V.R &

Watson, R.R., eds), 157, pp 1415–1421 Elsevier Limited, Oxford:

Academic Press, United Kingdom.

Garcı´a-Granados, A (1998a) Use of maslinic acid as serı´n proteases

inhibitor in the treatment of the disease produced by parasite of

Cryptosporidium genus Spanish Patent P9701029.

Garcı´a-Granados, A (1998b) Use of maslinic acid as serı´n proteases

inhibitor in the treatment of the disease produced by the acquired

immunodeficiency viruses Spanish Patent P9702528.

Go´mez-Requeni, P., Mingarro, M., Calduch-Giner, J.A., Me´dale, F.,

Martin, S.A.M., Houlihan, D.F., Kaushik, S & Pe´rez-Sa´nchez, J.

(2004) Protein growth performance, amino acid utilization and

somatotropic axis responsiveness to fish meal replacement by plant

protein sources in gilthead sea bream (Sparus aurata) Aquaculture,

232, 1–4, 5, 93–510.

Henderson, R.J & Sargent, J.R (1981) Lipid biosynthesis in

rain-bow trout fed diets of different fat content Comp Biochem.

Physiol C, 69, 31–37.

Hidalgo, M.C., Skalli, A., Abella´n, E., Arizcun, M & Cardenete, G.

(2006) Dietary intake of probiotics and maslinic acid in juvenile

dentex (Dentex dentex L.): effects on growth performance, survival

and liver proteolytic activities Aquacult Nutr., 12, 256–266.

Houlihan, D.F., Hall, S.J., Gray, C & Noble, B.S (1988) Growth

rates and protein turnover in Atlantic cod Gadus morhua Can J.

Fish Aquat Sci., 55, 649–657.

Jones, D.R & Randall, D.J (1978) The respiratory and circulatory

systems during exercises In: Fish Physiol., 7, pp 425–492.

Kim, D.H., Han, K.M., Chung, I.S., Kim, S.H., Kwon, B.M., Jeong,

T.S., Park, M.H., Ahn, E.M & Baek, N.I (2005) Triterpenoids from

the flower of Campsis grandiflora K schum As human acyl-CoA:

cholesterol acyltransferase inhibitors Arch Pharm Res., 8, 550–556.

Labarca, C & Paigen, K (1980) A simple, rapid, and sensitive DNA

assay procedure Anal Biochem., 102, 344–352.

Lillie, R.D (1954) Argentaffin and Schiff reactions after periodic acid oxidation and aldehyde blocking reactions J Histochem Cyto- chem., 2, 127–136.

Liu, J., Sun, H., Duan, W., Mu, D & Zhang, L (2007) Maslinic acid reduces blood glucose in KK-Ay mice Biol Pharm Bull., 30, 2075–2078.

Lowry, O.H., Rosebrough, N.J., Farr, A.L & Randall, R.J (1951) Protein measurement with the Folin phenol reagent J Biol.

Chem., 193, 265–275.

Ma´rquez Martı´n, A., de la Puerta Va´zquez, R., Ferna´ndez-Arche, A.

& Ruiz-Gutie´rrez, V (2006) Suppressive effect of maslinic acid from pomace olive oil on oxidative stress and cytokine pro- duction in stimulated murine macrophages Free Radic Res., 40, 295–302.

Martı´n, R., Carvalho-Tavares, J., Ibeas, E., Herna´ndez, M., Gutie´rrez, V & Nieto, M.L (2007) Acidic triterpenes compromise growth and survival of astrocytoma cell lines by regulating reactive oxygen species accumulation Cancer Res., 67, 3741–3751.

Ruiz-Mingarro, M., Vega-Rubin de Celis, S., Astola, A., Pendon, C., Valdivia, M.M & Pe´rez-Sa´nchez, J (2002) Endocrine mediators of seasonal growth in gilthead sea bream (Sparus aurata): the growth hormone and somatolactin paradigm Gen Comp Endocrinol.,

128, 102–111.

Montero, D., Tort, L., Robaina, L., Vergara, J.M & Izquierdo, M.

(2001) Low vitamin E in diet reduces stress resistance of gilthead seabream (Sparus aurata) juveniles Fish Shellfish Immunol., 11, 473–490.

Montilla, M.P., Agil, A., Navarro, M.C., Jime´nez, M.I., Granados, A., Parra, A & Cabo, M.M (2003) Antioxidant activity of maslinic acid, a triterpene derivative obtained from Olea europaea Planta Med., 69, 472–474.

Garcı´a-Munro, H.N & Fleck, A (1966) The determination of nucleic acids.

In: Methods Biochem Anal., Vol XIV (Glick, D ed.), pp 113–

176 Interscience Publishers, John Wiley and sons, New York, London, Sidney.

Ortun˜o, J., Cuesta, A., Esteban, M.A & Meseguer, J (2001) Effect

of oral administration of high vitamin C and E dosages on the gilthead sea bream (Sparus aurata L.) innate immune system Vet.

Immunol Immunopathol., 79, 167–180.

Perago´n, J., Barroso, J.B., de la Higuera, M & Lupia´n˜ez, J.A (1998) Relationship between growth and protein turnover rates and nucleic acids in the liver of rainbow trout (Oncorhynchus mykiss) during development Can J Fish Aquat Sci., 55, 649–657.

Perago´n, J., Barroso, J.B., de la Higuera, M & Lupia´n˜ez, J.A (1999) Selective changes in the protein-turnover rates and nature of growth induced in trout liver by long-term starvation followed by re-feeding Mol Cell Biochem., 201, 1–10.

Pe´rez-Sa´nchez, J (2000) The involvement of growth hormone in growth regulation, energy homeostasis and immune function in the gilthead sea bream (Sparus aurata): a short review Fish Physiol.

Biochem., 22, 135–144.

Pungitore, C.R., Padron, J.M., Leon, L.G., Garcı´a, C., Ciuffo, G.M., Martin, V.S & Tonn, C.E (2007) Inhibition of DNA topoisomerase I and growth inhibition of human cancer cell lines

by an oleanane from Junellia aspera (Verbenaceae) Cell Mol Biol.

(Noisy-le-grand), 53, 13–17.

Reyes, F.J., Centelles, J.J., Lupia´n˜ez, J.A & Cascante, M (2006) (2Alpha, 3beta)-2, 3-dihydroxyolean-12-en-28-oic acid, a new natural triterpene from Olea europaea, induces caspase dependent apoptosis selectively in colon adenocarcinoma cells FEBS Lett.,

580, 6302–6310.

Reyes-Zurita, F.J., Rufino-Palomares, E.E., Lupia´n˜ez, J.A & ante, M (2009) Maslinic acid, a natural triterpene from Olea

Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc- Casc-.

europaea L., induces apoptosis in HT29 human colon-cancer cells

via the mitochondrial apoptotic pathway Cancer Lett., 273, 44–54.

Reynolds, E.S (1963) The use of lead citrate at high pH as an

electron-opaque stain in electron microscopy J Cell Biol., 17,

208–212.

Robaina, L., Corraze, G., Aguirre, P., Blanc, D., Melcion, J.P &

Kaushik, S (1999) Digestibility, postprandial excretion and

selected plasma metabolites in European sea bass (Dicentrarchus

labrax) fed pelleted or extruded diets with or without wheat gluten.

Aquaculture, 179, 45–56.

Rodrı´guez-Rodrı´guez, R., Perona, J.S., Herrera, M.D &

Ruiz-Gutie´rrez, V (2006) Triterpenic compounds from ‘‘orujo’’ olive oil

elicit vasorelaxation in aorta from spontaneously hypertensive

rats J Agric Food Chem., 54, 2096–2102.

Romero, C., Garcı´a, A., Medina, E., Ruı´z-Me´ndez, M.V., de Castro,

A & Brenes, M (2010) Triterpenic acids in table olives Food

Chem., 118, 670–674.

Scalon Cunha, L.C., Andrade e Silva, M.L., Cardoso Furtado, N.A.,

Vinholis, A.H., Gomes Martins, C.H., da Silva Filho, A.A &

Cunha, W.R (2007) Antibacterial activity of triterpene acids and

semi-synthetic derivatives against oral pathogens Z Naturforsch.

C, 62, 668–672.

Sierra, M.A (1995) La encapsulacio´n como estrategia para

estable-cer las necesidades de metionina y la suplementacio´n de proteı´na

de soja Consecuencias sobre el recambio proteico tisular y el

crecimiento de la dorada (Sparus aurata) Ph.D Thesis University

of Granada, Granada, Spain.

Sitja-Bodabilla, A & Alva´rez-Pellitero, P (2003) Experimental

transmission of Cryptosporidium molnari (Apicomplexa: Coccidia)

to gilthead sea bream (Sparus aurata L.) and European sea bass (Dicentrarchus labrax L.) Parasitol Res., 91, 209–214.

Sudgen, P.H & Fuller, S.J (1991) Regulation protein turnover im skeletaland cardiac muscle Biochemistry, 75, 192–200.

Sultana, M & Lee, N.H (2007) Antielastase and free radical enging activities of compounds from the stems of Cornus Kousa Phytother Res., 21, 1171–1176.

scav-Suzuki, O & Yagi, K (1976) A fluorimetric assay for ylamine in rat brain Anal Biochem., 75, 192–200.

b-phenyleth-Vergara, J.M., Robaina, L., Izquierdo, M & de la Higuera, M (1996) Protein sparing effect of lipids in diets for fingerlings of gilthead sea bream Fish Sci., 62, 62–628.

Vlietinck, A.J., De Bruyne, T., Apers, S & Pieters, L.A (1998) derived leading compounds for chemotherapy of human immu- nodeficiency virus (HIV) infection Planta Med., 64, 97–109 Watson, J.H (1958) Electron microscopy, iron crystals and medi- cine Henry Ford Hosp Med Bull., 6, 111–119.

Plant-Wen, X., Sun, H., Liu, J., Wu, G., Zhang, L., Wu, X & Ni, P (2005) Pentacyclic triterpenes Part 1: the first examples of naturally occurring pentacyclic triterpenes as a new class of inhibitors of glycogen phosphorylases Bioorg Med Chem Lett., 15, 4944– 4948.

Wen, X., Zhang, P., Liu, J., Zhang, L., Wu, X., Ni, P & Sun, H (2006) Pentacyclic triterpenes Part 2: synthesis and biological evaluation of maslinic acid derivatives as glycogen phosphorylase inhibitors Bioorg Med Chem Lett., 16, 722–726.

Xu, H.X., Zeng, P.Q., Wan, M & Sim, K.Y (1996) Anti-HIV triterpene acids from Geum japonicum J Nat Prod., 59, 643– 645.

.

1 2,3 1 2 1

School of Applied Sciences, Auckland University of Technology (AUT), Auckland, New Zealand;2 IRTA, Centre de Sant

Carles de la Ra`pita (IRTA-SCR), Unitat de Cultius Experimentals Crta del Poble Nou s/n, Sant Carles de la Ra`pita, Spain;

3 TECNOVIT, Tecnologia & Vitaminas, Polı´gono Industrial Les Sorts, Parcela, Alforja, Tarragona, Spain

Elvers and glass eels of Anguilla anguilla were fed diets

con-taining two types of feeding stimulants (FS) that were based on

processed marine (MBFS) and yeast proteins (YBFS) Elvers

(1.5 ± 0.3 g) were fed seven diets (MBFS and YBFS diets at

20 g kg)1, 40 g kg)1and 60 g kg)1plus control) for 60 days

Glass eels (250 ± 100 mg) were weaned to 60 g kg)1MBFS,

60 g kg)1YBFS and control diets for 30 days Diets

contain-ing 60 g kg)1FS had a beneficial effect in terms of growth,

homogenous size distribution and feed intake in elvers Elvers

fed 60 g kg)1MBFS and YBFS diets grew 11.9% and 5.6%

faster than the control group No differences in growth and size

distribution were detected in glass eels fed 60 g kg)1MBFS

and YBFS diets However, FS affected the digestive system

maturation; fish fed the 60 g kg)1 MBFS and YBFS diets

showed higher and intermediate values in the specific enzyme

activities in comparison with the control group This study

revealed that the incorporation of FS into a pelleted diet was

beneficial on the overall performance of European glass eels

and elvers However, the observed results were different

depending on the eelÕs stage of development, as well as the type

and inclusion level of the FS

key words: Anguilla anguilla, elver, European eel, feed

stimulant, glass eel, histology, weaning

Received 4 February 2011, accepted 12 April 2011

Correspondence: Jorge A Hirt-Chabbert, School of Applied Sciences,

Auckland University of Technology (AUT), Auckland, New Zealand.

E-mail: hirt.chabbert@gmail.com

It is well established that the financial success of a fish

farm is intimately related to the feed consumption of the

cultured species (Seymour 1989; Heinsbroek et al 2007) Inmost fish culture operations, maximum profitability isreached at the highest fish growth rates (Heinsbroek et al

2008), which is generally connected to the highest feedconsumption rate and lowest food conversion rates (Forbes2000; Oliveira & de Cyrino 2004) There are several pro-cedures used by aquaculturists to increase feed consump-tion One is the incorporation of feeding stimulants (FS)into the diets

Feeding stimulants are specific compounds or ingredientsadded to the feed to enhance the diet palatability and,consequently, its acceptability by the cultured fish As aresult of the improvement in the diet acceptability, the fishcan adapt earlier to artificial dry diet during the weaningperiod and attain a higher overall feed consumption andgrowth rate (Tandler et al 1982; Nakajima et al 1990;

Kolkovski et al 1997; Oliveira & de Cyrino 2004; Gaber2005) Moreover, the use of FS promotes quicker feedintake, minimizing the time that the feed remains in waterand so preventing the deterioration of the water quality(Shankar et al 2008)

Neurophysiologic and behavioural studies on eels indicatethat FS have potential to be used as an enhancer of pro-ductivity in commercial eel culture Studies on the neuro-physiology of eels show that they have high olfactory andgustatory sensitivity (Marui & Caprio 1992; Hara 1994), andseveral behavioural studies have indicated a positive response

of eels to different FS (Sola et al 1993; Knights 1996; Sola &

Tongiorgi 1998)

Laboratory studies on the effect of stimulants on feedintake and/or growth rate have showed positive results witheels Kamstra & Heinsbroek (1991) and Heinsbroek &

Krueger (1992) established that cod roe extract, bovinespleen extract or a mixture ofL-amino acids (alanine, glycine,proline and histidine) improved acceptance and growth rate

of a formulated trout fry diet in European glass eels (Anguilla

anguilla) Other feeding stimulant studies on European eel

(e.g a mixture ofL-amino acids, chicken spleen, and chicken

blood) added to moist-paste diets showed also a positive

increment in growth rate (Mackie & Mitchell 1983; Degani &

Levanon 1986) Experiments with Japanese eel, Anguilla

japonica, concluded that paste diets supplemented with

L-amino acids yield better performance than the plain diets

(Takii et al 1984; Takeda & Takii 1992) However, all these

studies were based on a moist-paste diet or a dry diet not

specifically formulated for eels

Currently in Europe, eel farmers have access to pelleted

feeds formulated specifically for the European eel species;

nevertheless, they still face the problem of the limited feed

acceptance by this species (Barrera, VALAQUA SA, Puc¸ol,

Spain, personal communication) Problems with the initial

phase of adaptation to artificial diets during the glass eel and

elver stages are the most serious Some eels do not become

accustomed to the pelleted feed, so they lose weight or grow

too slowly to have any economic value for the farmers The

use of FS may facilitate the acceptance of these artificial diets

currently used by eel farmers At the same time, it is

neces-sary to evaluate not only the effect but also the minimum

effective level The cost of inclusion must be minimized

without compromising the potential benefits from increases

in feed intake

The aim of this study was to evaluate the effects of FS at

different concentrations on the feed consumption, growth

rate and survival of cultured European eels (A anguilla)

during the glass eel and elver stages

Glass eels and elvers of the European eel used in this study

were obtained from a commercial fish farm Base Viva

lo-cated in Sant Pere Pescador, Spain These eels were

origi-nally collected as glass eels from the Daro´ River, Catalun˜a,

Spain, during the 2008–2009 fishing season and maintained

in outdoor flow-through 500-L tanks at Base Viva In

March 2009, 2 kg of glass eels that were fed only with

Artemianauplii and 4 kg of elvers already weaned onto an

artificial pelleted diet (Microbaq 8; Dibaq Acuicultura,

Fuentepelayo, Spain) were transported by road (3.5 h) to

the Institut de Recerca i Tecnologia Agroalimentaries

(IRTA) research facilities in Sant Carles de la Ra`pita

(IRTA-SCR), Spain On arrival, glass eels and elvers were

stocked separately in two holding tanks of 1500 L that were

connected to IRTAmar 5000 L freshwater recirculationunit (Carbo´ et al 2002)

Fish were maintained under a 12-h light/12-h dark regime,and each tank was covered with a black plastic cloth toreduce the light intensity (80.1 ± 10.5 lux at water surface;Lx-101 Lux Meter; Lutron Electronic Enterprise Co Ltd.,Taipei, Taiwan) The recirculation unit was provided withconstant aeration and a flow of water at 40 L min)1 Watertemperature, conductivity, pH (pHmeter 507; CrisonInstruments SA, Barcelona, Spain) and oxygen (OXI330;Crison Instruments SA) were kept at 22–23C, 2100 ±

200 lS cm)1, 7.5 ± 0.5 and 8–9 mg L)1 (92–100% tion), respectively The concentrations of ammonia (NHþ

satura-40.1

± 0.04 mg L)1), nitrite (NO2 0.1 ± 0.09 mg L)1) and trate (NO3 1.4 ± 0.5 mg L)1) were tested weekly by aHACH DR/870 colorimeter using HACH Standard Col-orimetric Test kits (HACH, Loveland, CO, USA)

ni-Glass eels were fed with a mixture of Artemia nauplii and cod roe (Gadus morhua) to apparent satiationtwice a day, whereas elvers were fed (4% dry weight perfresh body weight day)1) with the commercial pelleted feedMicrobaq 8 (Dibaq, Spain) The proximal biochemicalcomposition (g kg)1) of the pelleted feed was as follows:

meta-500 protein, 200 fat, 100 ash, 5 fibre and 50 moisture (dataprovided by the feed manufacturer) Fish were held for

6 weeks until the start of the experiment During thisperiod of acclimation, tanks were inspected daily and deadfish and debris removed All animal experimental proce-dures were conducted in compliance with the experimentalresearch protocol approved by the Committee of EthicAnimal and Animal Experimentation of the IRTA (refer-ence number 6213033898-3898-4-8), which followed theprinciples of replacement, reduction and refinement for theuse of animals in research

Experimental diets Six diets were prepared at AquativFrance (Elven, France) to contain two types of potentiallystimulatory FS at different concentrations (Table 1) Onetype of stimulant was based on processed marine proteins(MBFS) and the other on yeast proteins (YBFS) (Table 2).Feeding stimulants as powder were applied by top-coatingthe pelleted feed Microbaq8 by spraying with 20 g kg)1fishoil The concentrations achieved were 20 g kg)1, 40 g kg)1and 60 g kg)1 Top-coating was performed in a 7-kg Forbergmixer (Forberg International, Larvik, Norway) with a speed

of 20 rpm Fish oil was applied for 30 s, the FS in powderwere for 60 s, and the retention time was 60 s

.

Experiment 1 – Elvers A total of 1134 elvers were selected

for this experiment and distributed uniformly among the 21

35-L cylindrical 150-lm mesh baskets located in three 1500-L

holding tanks that were connected to a recirculation system

(IRTAmar, Carbo´ et al 2002) Each basket was stocked

with 54 elvers (1.5 ± 0.3 g, mean ± SD) at a density of 2

fish per litre In addition, six cylindrical PVC tubes (100 mm

long, 15 mm inner diameter) were placed in each basket to

provide refuge and protection from aggressive behaviour and

cannibalism (Rodrı´guez et al 2009)

Elvers were maintained under the same environmental

conditions as in the acclimation tanks and fed to apparent

satiation twice a day with different diets for 60 days Seven

different dietary treatments were evaluated: the commercial

pelleted feed Microbaq 8 as control diet and the MBFS

and YBFS diets at 20, 40 and 60 g kg)1 Each treatment was

randomly assigned to three experimental baskets Feeds

provided to the elvers were deposited in feeding stations

(trays; 10· 10 cm) To assure the minimum nutrient

leach-ing from the FS, the uneaten feed was removed from the

trays after 45 min and kept at 4C for posterior calculation

of the feed consumption rates If the first ration was totally

consumed, a second ration was provided During the course

of the experiment, the baskets were cleaned by siphoning

three times a week and checked daily for dead or moribund

fish If any were found, they were removed and recorded as

a death

Experiment 2 – Glass eels A total of 1170 glass eels were

selected for the experiment and distributed uniformly among

nine cylindrical 150-lm mesh baskets located in one 1500-L

holding tank that was connected to a recirculation system

(IRTAmar, Carbo´ et al 2002) Each basket was stocked

with 130 glass eels (250 ± 100 mg) at an initial density of

four glass eels per litre The fish were provided with shelter by

a rectangular meshed basket (100· 60 · 30 mm) suspended

in the middle of the water column of each basket The fishwere kept under the same environmental conditions as in theacclimation tanks

Based on the experimental results obtained from ment 1, three different diets were evaluated to test the effects

Experi-of feeding stimulant on glass eel weaning: the commercialpelleted feed Microbaq 8 (control diet) and the two FSadded to the control diet at the inclusion level of 60 g kg)1(60 g kg)1MBFS and 60 g kg)1 YBFS) (Table 1) Consid-ering the difficulties in rearing European eel during the glasseel stage, diets were only tested for 30 days In particular,during the first 15 days, the fish of each tank were weanedgradually from cod roe to one of the three feeds (15% dailysubstitution) After that, they were fed only with the corre-sponding extruded pellets, twice per day to apparent satia-tion Each treatment was randomly assigned to threeexperimental baskets The feeding procedures, cleaning andremoving of dead or dying fish were as for the Experiment 1

Sampling schedule and growth parameters In the Experiment

1, the wet body weight (BW, g) of each elver was ually measured on days 0, 30 and 60 of the experiment

individ-Prior to their handling, fish were anaesthetized with tricainemethanesulfonate (MS222; Sigma, Barcelona, Spain) at afinal concentration of 50 mg L)1 (Chiba et al 2006) Fishwere not fed the day before and after handling Mortalitieswere recorded daily For analytical purposes, 15 (1.5 ±0.3 g) elvers were sampled from the acclimation tank at theday 0, and 10 elvers were sampled from each basket at day

60 (30 fish per treatment) Specimens were sampled early inthe morning before feed was offered and sacrificed by anoverdose of MS222 The 15 elvers from day 0 werekept frozen at )20 C for later proximate compositionanalysis The 10 elvers collected from each basket (30 per

Table 1 Proximate composition on dry weight basis (g kg ) of experimental diets containing different levels of feeding stimulants

Each value is the mean ± SD of triplicate analysis (n = 3) Gross energy was calculated based on known energetic values of protein, fat and

carbohydrate [NRC (National Research Council) 1993].

.

experimental condition) at day 60 were split in three pools:

(i) three fish for proximate composition analysis were kept

at )20 C, (ii) two fish were dissected and the liver and

intestine were fixed in 4% formaldehyde, dehydrated in a

graded series of ethanol for posterior histological analysis,

and (iii) five fish were kept frozen at)80 C for posterior

digestive enzyme analyses

In the Experiment 2, BW (mg) of each glass eels was

determined at the start (day 0) and at the end of the

exper-iment (day 30) Similar to Experexper-iment 1, fish were

anaesthe-tized prior their handling for BW measurement At the start

of the experiment, 36 glass eels (0.25 ± 0.10 g) were taken

from the acclimation tank, sacrificed by an overdose of

MS222 and kept for later body composition analyses at)20 C At the end of the weaning period (day 15) and at theend of the experiment trial (day 30), 26 glass eels were ran-domly taken each time from each basket (78 for eachexperimental condition): 12 fish for proximate composition,

10 fish for digestive enzyme analyses and four fish for tology The glass eels were sampled early in the morningbefore feed was offered, sacrificed and kept for posterior lateranalyses as previously indicated Mortalities were recordeddaily in each basket

his-For both experiments, data obtained were analysed forfish growth and feed utilization, and the following indiceswere used: SGRW(Specific growth rate, % day)1) = 100 (ln

BWf– ln BWi) days)1; S (Survival, %) = 100 (Initial basketstock numbers – dead fish) (Initial basket stock numbers))1;

BG (Biomass gain, % Bi) = 100 (Bf – Bi) Bi )1; FI (Feed

intake, % Bi) = 100 (basket total feed weight consumed, g)

Bi )1; FCR (Feed conversion ratio) = (basket total feed

weight consumed, g) (Bf– Bi))1, where BWiand BWfare theinitial and final mean BW (mg or g) per experimental basketand Biand Bf the initial and final basket stocked biomass(g)

Chemical analyses of diets for moisture, protein, fat, ash andcarbohydrate were performed according to AOAC (1990)methods The body composition was determined individuallyfor elvers (day 0, n = 15; day 60, n = 9 by treatment) and as

a pool of four fish for glass eels (day 0, n = 9; days 15 and

30, n = 9 by treatment) Specimens for body analysis wereground, and small aliquots were dried (120C, 24 h) toestimate water content The total fat content was quantifiedgravimetrically after extraction in chloroform/methanol(2 : 1) and evaporation of the solvent under a stream ofnitrogen followed by vacuum desiccation overnight (Folch

et al 1957) Protein and carbohydrate contents were mined according to Lowry et al (1951) and Dubois et al.(1956), respectively All chemical analyses were performed intriplicate

deter-The fish were individually dissected to separate pancreaticand intestinal segments, under a dissecting microscope on aprechilled glass plated maintained at 0C as previouslydescribed (Gisbert et al 2009) The pancreatic segment washomogenized (Ultra-Turrax T25 basic; IKA-Werke, Stau-fen, Germany) in five volumes (v/w) of ice-cold Milli-Q

Table 2 Proximate composition and amino acid profile (g kg ),

soluble protein and per cent molecular weight protein profile of the

two feeding stimulants Data provided by the manufacturer of the

feeding stimulants

Yeast-based feeding stimulant

Marine-based feeding stimulant Proximate analysis

Soluble protein ratios (%)

Soluble protein/total protein 67 89

Soluble protein/total stimulant 32 48

Insoluble protein/total stimulant 16 6

Molecular weight (MW) distribution of protein compounds (%)

water, centrifuged at 3300 g for 3 min at 4C, sonicated for

1 min and the supernatant frozen at)20 C for subsequent

enzyme quantification (trypsin, lipase and a-amylase) For

the determination of intestinal brush border membrane

en-zymes (leucine aminopeptidase and alkaline phosphatase),

dissected samples were homogenized in cold 50 mM

manni-tol and 2 mM Tris–HCl buffer (pH 7.0) Intestinal brush

border membranes were purified according to Crane et al

(1979) and stored at)20 C

Trypsin (E.C 3.4.21.4) activity was assayed at 25C using

BAPNA (N-a-benzoyl-dl-arginine p-nitroanilide) as the

substrate One unit of trypsin per mL (U) was defined as

1 lmol BAPNA hydrolysed per min per mL of enzyme

extract at 407 nm (Holm et al 1988) Bile salt-activated

lipase (E.C 3.1.1) activity was assayed for 30 min at 30C

using p-nitrophenyl myristate as substrate dissolved in

0.25 mM Tris–HCl, pH 9.0, 0.25 mM 2-methoxyethanol and

5 mM sodium cholate buffer Lipase activity (U mL)1) was

defined as the lmol of substrate hydrolysed per min per mL

of enzyme extract (Iijima et al 1998) a-Amylase (E.C

3.2.1.1) was measured according to Me´tais & Bieth (1968),

using 0.3% soluble starch dissolved in Na2HPO4buffer (pH

7.4) as substrate a-Amylase activity (U) was defined as the

mg of starch hydrolysed during 30 min mL)1 of tissue

homogenate at 37C at 580 nm

Intestinal alkaline phosphatase (E.C 3.1.3.1) was

quanti-fied at 37C using 4-nitrophenyl phosphate (PNPP) One

unit (U) was defined as 1 lg PNPP released per min per mL

of homogenate at 407 nm (Bessey et al 1946)

Aminopepti-dase-N (E.C.3.4.11.2) was determined at 25C according to

Maroux et al (1973), using L-leucine p-nitroanilide as

sub-strate (in 0.1 mM DMSO) One unit of enzyme activity (U)

was defined as 1 lg nitroanilide released per min per mL of

homogenate at 410 nm

Soluble protein from crude enzyme extracts was quantified

as described by Bradford (1976), using bovine serum albumin

as standard Enzyme activity was expressed as specific

activity (activity units per milligram of protein, U mg

pro-tein)1) For each basket, all enzymatic assays were run

individually for elvers (n = 5; Experiment 1) and from five

pools of two glass eels (Experiment 2)

For histological purposes, six elvers and twelve glass eels

from each experimental condition in each experiment were

fixed in 4% formaldehyde, dehydrated in a graded series

of ethanol, embedded in paraffin and cut in serial sections

(3–5 lm thick) as previously described in Rodrı´guez et al

(2005) Sections were stained by HarrisÕ haematoxylin andeosin for general histomorphological observations, whileperiodic acid-Schiff (PAS) was used to detect glycogendeposits in the liver (Pearse 1985)

The mean values of BW were expressed as mean ± SD Thecalculation was based on the values of the individual BW ofall the eels belonging to the same treatment (fish from thethree baskets/replicates per treatment analysed together),and consequently, the SD describes the dispersion ofthe individual eel values The mean values of specific growthrate (SGRW), survival, BG, feed intake (FI) and conse-quently the food conversion ratio (FCR) were expressed asmean ± standard error of the mean (SEM) In contrast to

BW, these parameters were calculated using the values of thebaskets (n = 3 for each treatment), they cannot be calcu-lated for individual eels The SEM quantifies the error incalculating the mean of the population from the basketvalues

Data were analysed for one-factor variance with Minitabstatistical software 16.1.0 (Minitab Statistical Software,State College, PA, USA) Before analysis, homogeneity ofvariance was confirmed using Bartlett test (Snedecor &

Cochran 1989) When a significant treatment effect wasobserved, individual means were compared with Tukey–

Kramer HSD multiple comparison test

At the beginning of the Experiment 1, elvers from differentexperimental groups were homogeneous in BWi(1.5 ± 0.3 g,

P >0.05) No statistically significant differences amongdietary treatments were observed at day 30 (P > 0.05)

However, at the end of the experimental period (day 60),

BWfwas higher in those groups fed the diet containing 6%

marine-based and yeast FS Elvers fed 60 g kg)1 MBFSshowed a significantly higher BWf than animals fed thecontrol diet, they were 13.5% heavier Likewise, eels fed

60 g kg)1YBFS were 5.7% heavier than the control group(Table 3) Size dispersion in BWf of elvers was not signifi-cantly affected by the different dietary treatments (Fig 1a)

However, the BWf dispersion in the group fed 60 g kg)1MBFS tended to be lower than that observed with the otherdiets (coefficient of variation 35.1 versus 38.9 and 43.2%;

.