Aquaculture Research, 2010, 41, 1553^1573 doi:10.1111/j.1365-2109.2010.02546.x REVIEW ARTICLE Role of gastrointestinal microbiota in fish Sukanta K Nayak Laboratory of Fish Pathology, Department of Veterinary Medicine, College of Bioresorece Sciences, Nihon University, Kanagawa 252-8510, Japan Correspondence: S K Nayak, Laboratory of Fish Pathology, Department of Veterinary Medicine, College of Bioresorece Sciences, Nihon University, 1866 Kameino, Fujisawa, Kanagawa 252-8510, Japan E Mail: sukantanayak@redi¡mail.com Abstract The gastrointestinal (GI) tract of an animal consists of a very complex and dynamic microbial ecosystem that is very important from a nutritional, physiological and pathological point of view A wide range of microbes derived from the surrounding aquatic environment, soil/sediment and feed are found to colonize in the GI tract of ¢sh Among the microbial groups, bacteria (aerobic, facultative anaerobic and obligate aneraobic forms) are the principal colonizers in the GI tract of ¢sh, and in some ¢sh, yeasts are also reported The common bacterial colonizers in the GI tract of freshwater and marine ¢sh include Vibrio, Aeromonas, Flavobacterium, Plesiomonas, Pseudomonas, Enterobacteriaceae, Micrococcus, Acinetobacter, Clostridium, Fusarium and Bacteroides, which may vary from species to species as well as environmental conditions Besides, several unknown bacteria belonging to Mycoplasma, Arthrobacter, Brochothrix, Jeotgailbacillus, Ochrobactrum, Psychrobacter and Sejongia species in the GI tract of di¡erent ¢sh species have now been identi¢ed successfully using culture-independent techniques Gnotobiotic and conventional studies indicate the involvement of GI microbiota in ¢sh nutrition, epithelial development, immunity as well as disease outbreak This review also highlights the need for manipulating the gut microbiota with useful bene¢cial microbes through probiotic, prebiotic and synbiotic concepts for better ¢sh health management Keywords: ¢sh, gastrointestinal tract, gnotobiotic, microbiota, probiotics, prebiotics Introduction The ¢rst observation on the occurrence of a gastrointestinal (GI) microorganism in any host was made r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd by Leewenhock in 1674 (DoBell 1932) However, the comprehensive study of intestinal bacteria was initiated only after the discovery of Escherichia in the human GI tract, which laid the foundation of GI microbiota in other species With the advent of the 20th century, the distribution of GI microbiota has been studied extensively in various animals for di¡erent purposes The GI ‘microbiota’ usually refers to a very complex and dynamic microbial ecosystem that colonizes the GI tract of an animal Furthermore, with the developments in molecular and biotechnological tools, this complex ecosystem is beginning to be unravelled in many species (Rastall 2004) The mammalian GI tract contains an enormous variety of aerobic and anaerobic microbes that interact in its complex ecosystem (Eckburg, Bik, Bernstein, Purdom, Dethlefsen, Sargent, Gill, Nelson & Relman 2005; Nicholson, Holmes & Wilson 2005), but that of ¢sh is believed to be simpler and less in number than that of endothermic animals (Trust & Sparrow 1974; Horsley1977; Finegold, Suher & Mathisen1983; Sakata 1990) Until the 1970s, a concrete report on the existence of a stable indigenous microbiota in many aquatic animals was not available (Savage 1977;Yoshimizu, Kimura & Sakai1980; RingÖ, Olsen, Mayhew & Myklebust 2003), but during the past few decades, substantial research has been carried out to characterize the GI microbiota in a wide range of ¢sh species (RingƯ, Strom & Tabachek 1995; RingÖ & Gatesoupe 1998; RingÖ, Lodemel, Myklebust, Kaino, Mayhew & Olsen 2001; Ward, Steven, Penn, Methe & Detrich III 2009) However, most of the earlier GI microbiota studies in ¢sh have emphasized the microbial spoilage, environmental relationship (Horsley 1973), their enzymatic ability (Shcherbina & Kazlawlene 1971; Lindsay & Harris 1980), studies on nutritional aspects (Moriarty 1990), monitoring of the changes in farms (Allen, 1553 Role of gastrointestinal microbiota in ¢sh S K Nayak Austin & Colwell 1983) and antibiotic resistance (Ogbondeminu & Olayemi 1993) The GI microbiota serve a variety of functions in the host and their importance in the nutrition and health of the host by promoting nutrient supply, preventing the colonization of infectious agents, energy homeostasis and maintenance of normal mucosal immunity is well documented in mammals (Xu, Bjursell, Himrod, Deng, Carmichael, Chiang, Hooper & Gordon 2003; Nicholson et al 2005; Delzenne & Cani 2008) Although the presence of native GI microbiota in ¢sh has been recognized, little is known about the bacterial communities and their establishment, diversity and most importantly their role in ¢sh nutrition and health Therefore, an attempt has been made to review the information available for a better understanding of the GI microbes and their possible functional roles in ¢sh Development and establishment of GI microbiota of fish The microbial colonization, establishment, composition and diversity in the GI tract of ¢sh is a complex process and believed to be a re£ection of the microbial composition of the rearing water, diet and their environment (Liston 1957; Geldreich & Clarke 1966; Nieto,Toranzo & Barja1984; Buras, Duek, Niv, Hepher & Sandbank 1987; Ogbondeminu 1993; Korsnes, Nicolaisen, Skar, Nerland & Bergh 2006; RingÖ, Sperstad, Myklebust, Refstie & Krogdahl 2006; RingÖ, Sperstad, Myklebust, Mayhew & Olsen 2006; Fjellheim, Playfoot, Skjermo & Vadstein 2007) Two distinct groups, i.e either allochthonous (transient) and autochthonous (adherent), are usually found in the GI tract of ¢sh The latter group of bacteria, by virtue of their ability to tolerate the low pH in gastric juices and resistance to the actions of bile acids, only succeeded in colonizing in the epithelial surface of the stomach, small and large intestine (Savage1989) These bacteria can ¢rmly attach to the intestinal mucosa to become the autochthonous microbiota of the host (Yoshimizu, Kimura & Sakai 1976; Onarheim & Raa 1990; Sakata1990; Onarheim,Wiik, Burghardt & Stackebradt 1994) The other group of bacteria is present transiently in the GI tract because they are not able to colonize the mucus layer and/or the epithelial surface (RingÖ & Birkbeck 1999) They either lack this ability entirely or are so ine¡ective at it that they are outcompeted by other bacteria in the mucus/epithelium The initial colonization process is very complex at larval and fry and mostly depend on the ¢sh type, 1554 Aquaculture Research, 2010, 41, 1553–1573 nutrients/food and surrounding conditions (Bignell 1984; Voveriene, Mickeniene & Syvokiene 2002) The total bacterial load at the larval stage is low (approximately 102 CFU larva À 1) before active feeding (Munro, Barbour & Birkbeck 1994; Verner-Je¡reys, Shields, Bricknell & Birkbeck 2003; Reid, Treasurer, Adam & Birkbeck 2009), and this initial load is mostly derived from the water by larvae to maintain osmotic balance (Tytler & Blaxter 1988; Reitan, Natvik & Vadstein 1998) However, the number increases rapidly (4105 CFU larva À 1) once the larvae start to feed (Munro, Birkbeck & Barbour 1993; Munro et al.1994) Furthermore, the microbial composition and density also vary in di¡erent regions of the GI tract of ¢sh depending on the physico-chemical conditions of gut In ¢sh, a progressive increase in culturable bacterial levels from the stomach to the posterior intestine is often reported (Trust & Sparrow 1974; MacDonald, Stark & Austin 1986; Molinari, Scoaris, Pedroso, Bittencourt, Nakamura, Ueda-Nakamura, Abreu Filho & Dias Filho 2003) Molinari et al (2003) recorded higher viable bacteria in both the anterior and the posterior gut than in the stomach of semi-intensively cultured Oreochromis niloticus However, with the successful application of modern non-culturable techniques, the composition of bacteria in the GI tract and their percentage may vary with respect to ¢sh Recently, Navarrete, Espejo and Romero (2009) recorded the average bacterial density to be  107,  106 and  107 bacteria g À in the stomach, pyloric caeca and intestine, respectively, in Salmo salar using epi£uorescence microscopy Factors a¡ecting the establishment of GI microbiota A series of exogenous and endogenous factors can a¡ect the establishment and nature of the microbial composition in the GI tract of ¢sh The developmental stage of ¢sh (Bell, Hoskins & Hodgkiss 1971; Sugita, Enomoto & Deguchi 1982; Sugita,Tokuyama & Deguchi 1985), gut structure (Sera, Ishida & Kadota 1974; Sugita et al 1985), the surrounding environment like ambient water temperature (Lesel & Peringer 1981; Sugita, Oshima, Tamuar & Deguchi 1989), rearing and farming conditions (Trust & Sparrow 1974; Trust 1975;Yoshimizu & Kimura 1976; Horsley1977; Sugita, Isida, Deguchi & Kadota 1982; RingÖ & Strom 1994) are very critical factors that a¡ect the initial colonization and the subsequent establishment process Besides, stress factors can signi¢cantly a¡ect the GI microbiota (Lesel & Sechet 1982) When di¡erent r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, 1553^1573 Aquaculture Research, 2010, 41, 1553^1573 types of chemicals, antibiotics, pollutants like pesticides, herbicides and insecticides enter into the digestive tract of an aquatic animal, they can drastically a¡ect the composition of dominant GI microbiota and may lead to the elimination of individual species from the whole microbial community (Austin & Al-Zahrani 1988; Sugita, Fukumoto, Koyama & Deguchi 1988; Sugita et al 1989; Syvokiene & Mickeniene 2002; Bakke-McKellep, Koppang, Gunnes, Sanden, Hemre, Landsverk & Krogdahl 2007; Mickeniene & Syvokiene 2008; Navarrete, Mardones, Opazo, Espejo & Romero 2008) Feed and feeding conditions considerably in£uence the composition of GI microbiota of ¢sh (Campbell & Buswell 1983; Sugita, Tsunohara, Fukumoto & Deguchi1987; Syvokiene1989; Onarheim & Raa1990; RingÖ 1993; RingÖ & Olsen 1999; Uchii, Matsui, Yonekura, Tani, Kenzaka, Nasu & Kawabata 2006; RingÖ, Sperstad, Myklebust, Refstie et al 2006; Martin-Antonio, Manchado, Infante, Zerolo, Labella, Alonso & Borrego 2007), and during the larval stage, the gut microbial £ora has been found to change rapidly with respect to feed (Brunvold, Sandaa, Mikkelsen, Welde, Bleie & Bergh 2007; Reid et al 2009) A positive correlation of capelin roe diet with Enterobacteriaceae in the GI microbiota of wild charr irrespective of freshwater and seawater maintenance was recorded by RingÖ and Strom (1994) However, they have recorded the predominance of Aeromonas in freshwater and theVibrio species under marine conditions from wild catch charr fed with a commercial diet Recently, RingÖ, Sperstad, Myklebust, Mayhew et al (2006) recorded the variation in GI microbiota with respect to the diet RingÖ and colleagues observed the dominance of Gram-positive bacteria belonging to Brochothrix and Carnobacterium species in the GI tract of Gadus morhua fed with ¢sh meal, while Psychrobacter species and Psychrobacter glacincola, Chryseobacterium and Carnobacterium species dominated the GI tract when fed with bio-processed soybean meal and standard soybean respectively Furthermore, the seasonal and day-to-day £uctuations in GI bacteria in di¡erent ¢sh species have also been recorded (Sugita et al.1987; MacMillan & Santucci 1990; Spanggaard, Huber, Nielsen, Nielsen, Appel & Gram 2000; Al-Harbi & Uddin 2004; Hagi, Tanaka, Iwamura & Hoshino 2004) Variations in the total viable GI bacterial counts from 1.6  106 to 5.1  107 CFU g À intestine in summer, 3.1  108 to 1.3  109 CFU g À intestine in autumn and 8.9  105 to1.3  107 CFU g À intestine in winter were recorded in hybrid tilapia (O niloticus  Oreochromis aureus) Role of gastrointestinal microbiota in ¢sh S K Nayak (Al-Harbi & Uddin 2004) Similarly, MacMillan and Santucci (1990) reported seasonal variations among the bacterial species belonging to Escherichia coli, Klebsiella, Pseudomonas, Plesiomonas, Shigelloides, Streptococcus and Moraxella species in farm-raised Ictalurus punctatus In a yearlong study on the changes in lactic acid bacteria (LAB) composition, Hagi et al (2004) found the predominance of Lactococcus lactis in summer (water temperature 420 1C) and Lactococcus ra⁄nolactis in winter (water temperature range 4^10 1C) in the GI tract of Cyprinus carpio Inter-individual variation and highest daily £uctuation of Bacteroides species in the GI tract of ¢sh like C carpio have also been observed (As¢e,Yoshijima & Sugita 2003) Microbial composition in the GI tract of ¢sh The GI tract is a favourable ecologic niche for microorganisms, and like any other animals, a wide range of microbes colonize in the GI tract of ¢sh (Skrodenyte-Arbaciauskiene 2007) However, information on the type of bacterial composition in the GI tract of ¢sh is often controversial (Izvekova & Lapteva 2004) Most of the earlier studies of ¢sh GI microbiota have been derived from the homogenates of intestinal content and/or faecal materials using culture-based techniques using selective or non-selective isolation media, followed by phenotypic characterization using a series of conventional morphological and biochemical assays (Horsley 1977; Sakata, Sugita, Mitsuoka, Kakimoto & Kadota 1981; Sugita, Oshima, Tamura & Deguchi 1983; Sakata 1989, 1990; Sugita et al.1989; Cahill 1990; Onarheim & Raa1990; Zaman & Leong 1994; RingÖ et al 1995; Sivakami, Premkishore & Chandran 1996) However, conventional methods are often time consuming and lack accuracy (As¢e et al 2003) and sensitivity in characterizing certain fastidious and obligate anaerobes that require special culture conditions Therefore, culture-based study of GI microbiota of any animal often leads to a very uncertain picture of the total microbial community residing inside the tract Nowadays, several novel molecular technologies are being increasingly used for the analysis of microbes present in the complex GI ecosystem of animals Molecular techniques based on genotypic ¢ngerprinting techniques such as colony hybridization with nucleic acid probes, pulsed ¢eld gel electrophoresis, ribotyping, polymerase chain reaction (PCR), random ampli¢ed polymorphic DNA, multiplex-PCR, arbitrary primed-PCR and triplet arbitrary r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, 1553^1573 1555 Role of gastrointestinal microbiota in ¢sh S K Nayak primed-PCR, denaturing gradient gel electrophoresis (DGGE), temporal temperature gradient gel electrophoresis, £uorescence in situ hybridization (FISH) and also electron microscopy have been used to study and characterize the microbes in the GI tracts of many animals including ¢sh (Spanggaard et al 2000;Walter, Hertel,Tannock, Lis, Munro & Hammes 2001; Holben,Williams, Saarinen, SÌrkilahti & Apajalahti 2002; Temmerman, Huys & Swings 2004; Kim, Brunt & Austin 2007; Peter & Sommaruga 2008) Most of the recent studies have successfully used these techniques alone or in combination with conventional methods to characterize both culturable and unculturable microbiota in the GI tract of ¢sh (Huber, Spanggaard, Appel, Rossen, Nielsen & Gram 2004; Pond, Stone & Alderman 2006; Seppola, Olsen, Sandaker, Kanapathippillai, Holzapfel & Ringo 2006; Shiina, Itoi,Washio & Sugita 2006; Skrodenyte-Arbaciauskiene, Sruoga & Butkauskas 2006; Hovda, Lunestad, Fontanillas & Rosnes 2007; Namba, Mano & Hirose 2007; Liu, Zhou,Yao, Shi, He, Holvold & Ringo 2008; Skrodenyte-Arbaciauskiene, Sruoga, Butkauskas & Skrupskelis 2008; Merri¢eld, Burnard, Bradley, Davies & Baker 2009), and are presented in Table The culture-dependent and -independent studies indicate that bacteria are the major microbial colonizer in the GI tract of ¢sh (MacDonald et al 1986; Spanggaard et al 2000; Molinari et al 2003; Pond et al 2006) Besides, yeast is also reported to colonize in the GI tract of some ¢sh (Andlid,Vazquez-Juarez & Gustafsson 1998; Gatesoupe 2007) Yeasts belonging to Rhodotorula species are frequently found in the GI tract of both marine and freshwater ¢sh while Metschnikowia zobelii, Trichosporon cutaneum and Candida tropicalisare are the dominant GI yeast species in marine ¢sh (Gatesoupe 2007) The GI microbiota of ¢sh mainly consists of aerobic or facultative anaerobic microorganisms (Clements 1997; Bairagi, Ghosh, Sen & Ray 2002; Saha, Roy, Sen & Ray 2006), facultative as well as obligate anaerobes, especially Cetobacterium somerae (previously classi¢ed as Bacteroides type A), Bacteroidaceae and Clostridium species (Cahill 1990; Sugita, Miyajima & Deguchi 1990; Pond et al 2006; Tsuchiya, Sakata & Sugita 2008) The predominance of anaerobes in the GI tract of ¢sh like gold ¢sh (Carrrasius auratus), Oncorhynchus mykiss and O niloticus has been recorded (Sakata, Okabayashi & Kakimoto 1980; Sugita et al 1989; Spanggaard et al 2000) Earlier, Trust, Bull, Currie and Buckley (1979) reported equal numbers (107 cells g À of gut content) of anaerobic bacteria (Bacteroides and Fusabacterium) and aerobic bacteria 1556 Aquaculture Research, 2010, 41, 1553–1573 (Aeromonas and Pseudomonas) in the GI tract of grass carp (Ctenopharyngodon idella) The total bacterial load in the GI tract of ¢sh is low in comparison with warm-blooded animals and their number often varies with age, nutrition and environment (RingƯ et al 2003; Gomez & Balcazar 2008) In ¢sh, the approximate viable aerobic and anaerobic bacteria usually vary from 104^109 to 6.6  104^ 1.6  109 CFU g À intestinal content respectively (Skrodenyte-Arbaciauskiene 2007) Earlier culturedependent studies in faecal samples of ¢sh indicate the aerobic and anaerobic bacterial load to be 108 and 105 bacteria g À of faeces respectively (Trust 1975; Trust et al 1979; Trust & Sparrow 1974; Austin & Al-Zahrani 1988) Recently, As¢e et al (2003) reported the variation in the total microbial cells in ¢ve gold¢sh specimens to range from 9.6  108 to 6.5  1010 cells g À of faeces by FISH Considering the fact that a large population of GI bacteria in ¢sh is unculturable (Romero & Navarrete 2006; Navarrete et al 2009), the total bacterial load is higher as recorded from the total culturable heterotrophic bacteria For instance, Shiina et al (2006), using direct microscopic enumeration of bacteria with 0,6-diamidino-2-phenylindole, reported the total bacterial count to vary from 1.0  104 to 1.4  109 CFU g À intestinal content in contrast to 4.7  1010^1.9  1011 cells g À intestinal content in coastal ¢sh like Ditrema temmincki, Girella punctata, Pseudolabrus japonicas, Sebastes pachycephalus,Takifugu niphobles and Thalassoma cupido Similarly, Sugita, Kurosaki, Okamura, Yamamoto and Tsuchiya (2005) also reported that the total bacterial load of each coastal ¢sh varied from 2.9  109 to 3.0  1010 cells g À intestinal content through direct counts regardless of the ¢sh species and their feeding habitat, while the viable counts of intestinal bacteria of these species ranged from 1.9  103 to 4.2  109 CFU g À intestinal content Recently, Navarrete et al (2009) also recorded such type of di¡erences in the total bacterial density in di¡erent regions of the GI tract such as stomach, pyloric caeca and intestine of S salar using epi£uorescence microscopy as compared with a culture-dependent technique The bacterial composition in the GI tract varies from freshwater to marine water ¢sh, with the predominance of Gram-negative bacteria over Gram-positive bacteria in the intestine of several ¢sh species (Sakata, Uno & Kakimoto 1984; RingÖ 1993; Hatha, Kuruvilla & Cheriyan 2000) Aeromonads are mostly associated with the GI tract of freshwater ¢sh (Sugita et al 1983; Sugita, Nakamura, Tanaka & Deguchi 1994; Wang, He, Live, Hu & Chen 1994; As¢e et al r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, 1553^1573 Aquaculture Research, 2010, 41, 1553^1573 Role of gastrointestinal microbiota in ¢sh S K Nayak Table Di¡erent culture-dependent and -independent methods used to study the whole and/or the speci¢c gastrointestinal microorganisms of ¢sh Sl No Fish Methods of study References Rainbow trout (Oncorhynchus mykiss) Merrifield et al (2009) Atlantic salmon (Salmo salar) Atlantic Cod (Gadus morhua) Grouper (Epinephelus coioides) Antarctic notothenioid fish (Notothenia coriiceps, Chaenocephalus aceratus) S salar Culture-based isolation, followed by partial 16S rRNA gene sequencing Culture-independent analysis of 16S rRNA gene sequence by denaturing gradient gel electrophoresis (DGGE) Scanning electron microscopic study Culture-based isolation and characterization by RFLP analysis of 16S rRNA gene and intergenic spacer region profiles Culture-independent analysis by temporal temperature gradient gel electrophoresis of the 16S rRNA gene and intergenic spacer region profiles Culture-based isolation and biochemical, physiological characterization and partial sequence analysis of the rpoB and 16S rRNA genes by DGGE Culture-based isolation and biochemical and physiological characterization, followed by 16S rRNA gene analysis Culture-independent analysis of the 16S rRNA gene Sea trout (Salmo trutta trutta), S salar S salar 10 Goldfish (Carassius auratus), Common carp (Cyprinus carpio), Mozambique tilapia (Oreochromis mossambicus), Japanese catfish (Silurus asotus), grass carp (Ctenopharyngodon idella) G morhua 11 G morhua 12 S salar 13 O mykiss 14 Senegalese sole (Solea senegalensis) 15 C carpio 16 O mykiss 17 G morhua Navarrete et al (2009) Reid et al (2009) Sun et al 2009 Ward et al (2009) Culture-independent analysis of the 16S rRNA gene by DGGE Culture-based isolation and characterization by 16S rRNA gene analysis Culture-based isolation and biochemical and physiological characterization, followed by 16S rRNA gene analysis Transmission electron microscopic study Culture-based isolation and characterization by biochemical reactions including API ZYM and API 20A systems as well as by 16S rRNA gene analysis Liu et al (2008) Culture-independent analysis of 16S rRNA gene analysis by DGGE Culture-based isolation and phenotypic characterization along with 16S rRNA gene analysis Culture-based isolation and characterization using API 20E and API 20NE systems along with 16S rRNA gene analysis by DGGE Culture-independent analysis of 16S rRNA gene by DGGE Culture-based isolation and characterization by 16S rRNA gene analysis Culture-independent analysis of 16S rRNA gene by DGGE and 16S rDNA clone library techniques Culture-based isolation and biochemical characterization using the API 20NE system and also by 16S rRNA gene analysis Culture-based isolation and phenotypical characterization along with 16S rRNA gene analysis Culture-based isolation and characterization using the Biolog system, API strips and 16S rRNA gene analysis Culture-independent analysis of 16S rRNA gene by restriction fragment length polymorphism (RFLP) Culture-based isolation and phenotypic characterization and also by 16S rRNA gene analysis Electron microscopic study Brunvold et al (2007) r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, 1553^1573 Skrodenyte-Arbaciauskiene et al (2008) Ringø, Sperstad, Kraugerud and Krogdahl (2008) Tsuchiya et al (2008) Fjellheim et al (2007) Hovda et al (2007) Kim et al (2007) Martin-Antonio et al 2007 Namba et al (2007) Pond et al 2006 Ringø et al 2006a 1557 Role of gastrointestinal microbiota in ¢sh S K Nayak Aquaculture Research, 2010, 41, 1553–1573 Table1 Continued Sl No Fish 18 Coho Salmon (Oncorhynchus kisutch) 19 20 21 22 23 24 25 26 27 28 29 30 Methods of study Culture-based isolation and characterization by RFLP analysis of the 16S rRNA gene Culture-independent analysis of the 16S rRNA gene by DGGE G morhua Culture-based isolation and phenotypic characterization in combination with random amplification of polymorphic – DNA (RAPD) analysis Takifugu niphobles Culture-based quantitative study, Culture-independent characterization of 16S rRNA gene analysis using the clonal library method River trout (Salmo trutta fario) Culture-based isolation and characterization by partial 16S rRNA gene sequence analysis Japanese flounder Culture-based isolation and characterization by (Paralichthys olivaceus) biochemical and physiological assays, followed by 16S rRNA gene analysis Bluegill (Lepomis macrochirus) Culture-independent study by community-level physiological profile analysis using Biolog microplates and analysis of the 16S rRNA gene by TGGE Culture-based isolation and characterization by RAPD Deepbodied crucian carp (Carassius uvieri), Channel catfish analysis of the 16S rRNA gene (Ictalurus punctatus), Silver carp (Hypophthalmichthys molitrix), C carpio O mykiss Culture-based isolation and characterization by RAPD analysis of the 16S rRNA gene Culture-independent study by fluorescent in situ hybridization (FISH) and DGGE C auratus, C carpio Culture-based isolation and characterization using the O mossambicus whole-cell hybridization technique using rRNA-targeted oligonucleotide probes Culture-independent analysis of fecal samples by FISH Atlantic halibut Culture-based isolation and characterization using (Hippoglossus hippoglossus) biochemical and the Biolog GN bacterial identification system as well as RFLP analysis of the 16S rRNA gene and the partial 16S rDNA gene S salar Culture-independent study by partial 16S rRNA gene sequence analysis Arctic charr (Salvelinus alpines) Electron microscopic study O mykiss Culture-based isolation and characterization using biochemical nature as well as RAPD analysis of the 16S rRNA gene 2003; Skrodenyte-Arbaciauskiene et al 2008) In freshwater ¢sh, Aeromonas, Pseudomonas and Bacteroides type A are major colonizers in the GI tract, followed by Plesiomonas, Enterobacteriaceae, Micrococcus, Acinetobacter, Clostridium, Bacteroides type B and Fusarium species (Trust et al 1979; Lesel 1981; Sugita, Sakata, Ishida, Deguihi & Kadota 1981; Sugita et al 1985) In contrast to freshwater ¢sh, Vibrio, Pseudomonas, Achromobacter, Corynebacterium, Alteromonas, Flavobacterium and Micrococcus species are predominant in the GI tract of most of the marine ¢sh (Cahill 1990; Onarheim et al 1994; 1558 References Romero and Navarrete (2006) Seppola et al (2006) Shiina et al (2006) Skrodenyte-Arbaciauskiene et al (2006) Sugita and Ito (2006) Uchii et al (2006) Hagi et al (2004) Huber et al (2004) Asfie et al (2003) Verner-Jeffreys et al (2003) Holben et al (2002) Ringø et al (2001) Spanggaard et al (2000) Blanch, Alsina, Simon & Jofre 1997; Verner-Je¡reys et al 2003) Among other bacterial groups that colonize in the GI tract of both freshwater and marine ¢sh are LAB (Strom 1988; Strom & RingÖ 1993; Pilet, Dousset, Barre, Novel, Des mazeaud & Piard 1995; Balcazar, de Blas, Ruiz-Zarzuela,Vendrell, Girones & Muzquiz 2007) However, they are usually not the dominant component of the GI microbiota (RingÖ et al 1995; Jankauskiene 2000a, b), but under certain conditions like a pond culture system, they can dominate, with an abundance as high as 1.1  106 cells g À ¢sh body weight, and can form a r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, 1553^1573 Aquaculture Research, 2010, 41, 1553^1573 stable component of the GI tract of ¢sh (Syvokien 1989) A recent study on GI bacteria of ¢sh like Notothenia coriiceps and Chaenocephalus aceratus, which di¡er in their pelagic distribution and feeding strategies, indicated the dominance of Vibrionanceae (g-proteobacteria) like that in temperate teleost species (Ward et al 2009) Application of a recent molecular technology has provided a major breakthrough in the detection and identi¢cation of the microbial composition in the GI ecosystem of many animals including ¢sh Among the recent technologies, DGGE, a‘genetic ¢ngerprint’ method based on PCR ampli¢cation of 16S rDNA, has been successfully used to study the dynamic behaviour of the dominant microbes in di¡erent environments (Gri⁄ths, Melville, Cook & Vincent 2001; Long & Azam 2001; Sandaa, Magnesen,Torkildsen & Bergh 2003) PCR- and DGGE-based identi¢cation and characterization provides an accurate picture of the complexity of the GI microbiota of ¢sh (Simpson, McCracken, White, Gaskins & Mackie 1999; Gri⁄ths et al 2001; Huber et al 2004; Vanhoutte, Huys, De Brandt, Fahey & Swings 2005; Brunvold et al 2007; Kim et al 2007) Apart from culturableVibrio, Pseudomonas, Janthinobacterium and Acinetobacter species, Hovda et al (2007) successfully characterized predominant but slow-growing culturable bacteria such as Lactobacillus fermentum, Photobacterium phosphoreum, Lactococcus and Bacillus species in the GI tract of Atlantic salmon (S salar) using PCR- and DGGEbased techniques Similarly, Pond et al (2006) succeeded in characterizing the presence of a number of bacterial species like Stenotrophomonas maltophilia, Pseudomonas picketti, Ralstonia eutrophia and b-Proteobacterium in the GI tract of O mykiss using 16S rRNA technology Nowadays, many new uncultuturable bacteria are being identi¢ed using molecular biological tools from the GI tract of ¢sh from freshwater to marine type Several unknown bacteria like Gram-negative Acinetobacter (A johnsoni), Chryseobacterium, Ochrobactrum, Psychrobacter (P luti, P fozii, P glacincola, P psychrophilus and P cibarius) and Sejongia species (S antarctica) and Gram-positive bacteria like Arthrobacter (A agilis and A psychrolactophilus), Brochothrix (B thermosphacta), Jeotgailbacillus (J psychrophilus), Microbacterium and Staphylococcus species (S equorum spp linens) in S salar (RingÖ, Sperstad, Myklebust, Mayhew et al 2006); Mycoplasma and Acinetobacter species (A junii) in S salar (Holben et al 2002); a and b subclass of Proteobacteria in C auratus (As¢e et al 2003); Clostridium species (C gasigenes) in O mykiss Role of gastrointestinal microbiota in ¢sh S K Nayak (Pond et al 2006);Tiedjeia arctica in wild river trout Salmo trutta fario (Skrodenyte-Arbaciauskiene et al 2008), Psychrobacter species; Delftia acidovorans, Burkholderia cepacia and Erwinia carotovora in grouper (Epinephelus coioides) (Sun, Yang, Ling, Chang & Ye 2009); and A aurescens and Janibacter species in O mykiss (Merri¢eld et al 2009) are now found to be part of the normal microbiota in the GI tract of those ¢sh There still remain doubts about the complete microbial composition and load in the GI of majority ¢sh species, and in the near future, culture-independent molecular tools may be able to provide a more detailed picture of the true complexity in the GI tract of di¡erent ¢sh species Role of GI microbiota in fish: gnotobiotic approaches Gnotobiotic models (animals cultured under axenic conditions or with a known microbiota) are excellent tools to study the role GI microbes in host (Marques, Ollevier,Verstraete, Sorgeloos & Bossier 2006; Dierckens, Rekecki, Laureau, Sorgeloos, Boon, Van den Broeck & Bossier 2009) Di¡erent gnotobiotic model studies reveal the importance of GI microbiota in nutrient metabolism and absorption, xenobiotic metabolism, regulation of energy balance, epithelial renewal, angiogenesis and development and maturation of the mucosal immune system (Falk, Hooper, Midtvedt & Gordon 1998; Cebra 1999) Like other animals, gnotobiotic/germ-free technology is now developed in ¢sh (Dahm & Geisler 2006; Pham, Kanther, Semova & Rawls 2008) and is also used to study evolutionarily conserved microbiota among vertebrates, to monitor the microbial behaviour, interaction and localization of microbes in the gut, as well as their role in nutrition, epithelial development and immunity (Rawls, Mahowald, Ley & Gordon 2006; Rawls, Mahowald, Goodman,Trent & Gordon 2007) Gnotobiotic studies in ¢sh indicate the involvement of GI microbiota in epithelial di¡erentiation and maturation Bates, Mittg, Kuhlman, Baden, Cheesman and Guillemin (2006) observed that the di¡erentiation of gut epithelium is arrested by the lack of brushborder intestinal alkaline phosphatase activity and the maintenance of immature patterns of glycan expression in the absence of the microbiota Alkaline phosphatase activity, which is a marker of epithelial maturation, as well as mucous-secreting goblet cells and hormone-secreting enteroendocrine cells, the secretory cell lineages, are found to increase r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, 1553^1573 1559 Role of gastrointestinal microbiota in ¢sh S K Nayak signi¢cantly in the digestive tract of conventional Danio rerio larvae compared with their gnotobiotic counterparts (Bates et al 2006) Furthermore, microbes are found to up-regulate the expression of 15 genes involved in DNA replication and cell division for epithelial proliferation (Rawls, Samuel & Gordon 2004) Similarly, a marked di¡erence in the enterocytes has also been observed in germ-free and conventional ¢sh Rawls et al (2004) observed consistent morphologic enterocytes in gnotobiotic D rerio, with the large supranuclear vacuoles ¢lled with clear electron-lucent material and that of conventional ¢sh ¢lled with eosinophilic and electron-dense material In another study, Rekecki, Dierckens, Laureau, Boon, Bossier and Van den Broeck (2009) observed variations with a slightly higher intestinal epithelium in the midgut consisting of a cuboidal to columnar epithelium in conventional larvae compared with cuboidal to squamous epithelium in the midgut of germ-free larvae of Dicentrarchus labrax after the ninth day post hatching The GI microbiota plays a crucial role in the nutrition of the host and in ¢sh, they are involved in nutrient metabolism, especially in cholesterol metabolism and tra⁄cking It has been reported that gnotobiotic D rerio larvae failed in the uptake of protein macromolecules, with a signi¢cant di¡erence in the levels of farnesyldiphosphate synthetase and apolipoprotein B as compared with conventional larvae (Bates et al 2006) Furthermore, the microbial upregulation of apolipoprotein B, which plays a pivotal role in intra- and extracellular cholesterol tra⁄cking, and downregulation of the liver-speci¢c cholesterol 7a-hydrolase, which catalyses the ¢rst step in cholesterol catabolism and bile acid biosynthesis, indicate the microbial modulation of cholesterol metabolism and tra⁄cking (Rawls et al 2004; Bates et al 2006) Besides these, gnotobiotic studies also indicate the involvement of GI microbes in ¢sh immunity and also in xenobitic metabolism Microbial regulation of glycoprotein production in the GI tract is reported in D labrax (Rekecki et al 2009) Furthermore, microbiota are found to up-regulate the genes involved in innate immunity parameters such as serum amyloid A1, C-reactive protein, complement component 3, angiogenin 4, glutathione peroxidase and myeloperoxidase (Rawls et al 2004; Rawls et al 2007) Although gnotobiotic studies reveal some of the functional roles of GI microbiota in ¢sh, the full extent to which the microbiota in£uences gut development, local as well as systemic immunity and homeostasis at the cellular and molecular levels remains to be explored 1560 Aquaculture Research, 2010, 41, 1553–1573 Role of GI microbiota in immunity The gut immune system, which is known as gut-associated lymphoid tissues (GALT), not only provides defence against infectious agents but also regulates immunity in the alimentary tract The GI microbes play a critical role in the development and maturation of GALT, which in turn mediate a variety of host immune functions (Rhee, Sethupathi, Driks, Lanning & Knight 2004) A complex and integrated interaction between the epithelium, immune components in the mucosa and microbes is responsible for the development and maturation of the gut-associated immune system of the host Gnotobiotic studies in di¡erent animal models also support this notion (Umesaki & Setoyama 2000; Peterson, McNulty, Guruge & Gordon 2007) Several mechanisms are proposed for the involvement of GI bacteria in the development of GALT Bacteria could stimulate B cell proliferation in GALT through a classical antigen-speci¢c immune response like protein A of Staphylococcus aureus and protein L of Peptostreptococcus magnus (Nilson, Solomon, Bjorck & Akerstrom 1992; Silverman & Goodyear 2002) or by directly stimulating the innate immune system (Medzhitov & Janeway 1997; Leadbetter, Rifkin, Hohlbaum, Beaudette, Shlomchik & Marshak-Rothstein 2002) In ¢sh, GALT consists principally of lymphocytes, eosinophil granular cells, several types of granulocytes and plasma cells (Zapata & Amemiya 2000; Zapata, Diez, Cejalvo, Gutierrezde & Cortes 2006) The involvement of GI microbes in the epithelial proliferation, maturation and immunity of ¢sh has already been discussed in the gnotobiotic studies (Rawls et al 2004; Rekecki et al 2009) Similarly, the endocytosis of bacteria by epithelial cells in the hindgut of immature larvae (Hansen, Strom & Olafsen1992) as well as intact uptake of bacterial antigens in columnar epithelial cells in the foregut, followed by their penetration into the gut epithelium, have been recorded in ¢sh (Olafsen & Hansen 1992) All these factors may directly or indirectly contribute to the development and stimulation of the immune system The early exposure of the intestine to live bacteria and subsequent colonization is very important for the development of gut barrier In ¢sh, dietary supplementation of useful microbes (probiotics) at early developmental stages can be helpful in increasing the subpopulations of speci¢c acidophilic granulocytes (Picchietti, Mazzini, Taddei, Renna, Fausto, Mulero, Carnevali, Cresci & Abelli 2007) Although probiotics are often found in a transient state and persist for a r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, 1553^1573 Aquaculture Research, 2010, 41, 1553^1573 certain period in the GI tract after the withdrawal of feed in ¢sh, it is worth noting that dietary supplementation of probiotics can enhance both local as well as systemic immunity in a wide range of ¢sh species (Panigrahi, Kiron, Puangkaew, Kobayashi, Satoh & Sugita 2005; Nayak, Swain & Mukherjee 2007; Panigrahi, Kiron, Satoh, Hirono, Kobayashi, Sugita, Puangkaew & Aoki 2007; Picchietti et al 2007; Aly, Ahmed, Ghareeb & Mohamed 2008; Picchietti, Fausto, Randelli, Carnevali, Taddei, Buonocore, Scapigliati & Abelli 2009; Sharifuzzaman & Austin 2009; Son, Changa,Wu, Guu, Chiu & Cheng 2009) Role of GI microbiota in nutrition The importance of intestinal bacteria in the nutrition and well-being of their hosts has been established in several animals (Floch, Gorbach & Lucky 1970) The ability to synthesize vitamins and essential growth factors and digestive enzymes by GI microorganisms has been demonstrated (Teply, Krehi & Elvehjem1947; Uphill, Jalob & Lall 1977; Drasar & Barrow 1985; Brock, Madigan, Martinko & Parker 1997) The GI microbiota of hydrobionts has been reported to contribute to the nutrition and physiological processes of the host by producing vitamins, amino acids, digestive enzymes and metabolites, similar to that of mammals (Syvokiene 1989; Cahill 1990; Sugita et al.1990; Sugita, Matsuo, Hirose, Iwato & Deguchi 1997; Mickeniene 1999; SkrodenyteArbaeiauskiene 2000; Skrodenyte-Arbaeiauskiene et al 2006) Nevertheless, a wide range of enzymes like carbohydrases, phosphatases, esterases, lipases and peptidases, cellulase, lipase and proteases (Bairagi et al 2002; Ramirez & Dixon 2003; Izvekova & Lapteva 2004) produced by GI bacteria could be a contributory source to digestive enzymes in ¢sh The presence of a high concentration of Aeromonas in the GI tract can play an important role in digestion as Aeromonas species secrete several proteases (Pemberton, Kidd & Schmidt 1997) Similarly, the p-nitrophenyl-b-n-acetylglucosaminide-, chitin-, cellulose- and collagen-degrading ability of gut bacteria indicates their possible involvement in the nutrition of ¢sh (Shcherbina & Kazlawlene 1971; Lindsay & Harris 1980; Lesel, Fromageot & Lesel 1986; Macdonald et al 1986; Das & Tripathi 1991; Kar & Ghosh 2008) Recent studies indicate that anaerobic bacteria might play a role in the digestion and absorption of nutrients (Ramirez & Dixon 2003) Anaerobic bacteria can contribute to ¢sh nutrition by supplying it with Role of gastrointestinal microbiota in ¢sh S K Nayak volatile fatty acids (Clements 1997) This is due to the fact that volatile fatty acids, end products of anaerobic fermentation, are often reported in the intestines of carp (C carpio), shad (Dorosoma cepedianum) and largemouth bass (Micropterus salmoides) (Smith,Wahl & Mackie1996) Nevertheless, the ability of GI aerobic, anaerobic and facultative aerobic bacteria to synthesize di¡erent vitamins and amino acids in ¢sh like C carpio, C auratus, I punctatus and O nilotica is noteworthy (Kashiwada & Teshima 1966; Teshima & Kashiwada 1967; Limsuwan & Lovell 1981; Sugita et al 1989; Sugita, Miyajima & Deguchi 1991a; Sugita, Takahashi, Miyajima & Deguchi 1991b) Among the vitamins, the production of vitamin B12 by GI bacteria is well documented in ¢sh (Sugita et al 1991a,b; Sugita, Takahashi, Miyajima & Deguchi 1992) The production of vitamin B12 di¡ers from species to species and is correlated with the abundance of more anaerobes as compared with aerobes in the GI tract Fish like O nilotica produce more vitamin B12 as compared with I punctatus due to the presence of more anaerobic bacteria in the gut of former ¢sh than the latter (Sugita et al 1990) Similarly, a signi¢cant di¡erence in daily vitamin B12 synthesis in O nilotica (11.2 ng kg À body weight) and I punctatus (1.4 ng kg À body weight) has also been recorded by Lovell and Limsuwan (1982) In contrast to endothermic animals, the exact role of gut microbiota in ¢sh nutrition is di⁄cult to conclude because of the complex and variable ecology of the GI tract of ¢sh Despite recent conventional and gnotobiotic studies that indicate the possible involvement of GI bacteria in several physiological and nutritional functions in ¢sh, more emphasis and/or thorough research is required in order to establish the nutritional importance of the gut microbiota Role of GI microbiota in disease outbreak The gut of an organism usually harbours a diverse population of non-pathogenic, pathogenic and commensal bacteria, which can contribute signi¢cantly to the overall health and disease outbreak in a host In a healthy animal, some microbiota are established and others are transient in the intestine There occurs a proper balance between the endogenous microbiota of the intestine and the host’s control mechanism However, if this balance is disturbed, several pathogens present in the transient state can establish lethal infections (Sekirov & Finlay 2009) r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, 1553^1573 1561 Aquaculture Research, 2010, 41, e862^e871 Cloning of myostatin-like genes from Pandalopsis japonica K S Kim et al Lachaise F., Roux A.L., Hubert M & Lafont R (1993) The molting gland of crustaceans: localization, activity, and endocrine control (a review) Journal of Crustacean Biology 13, 198^234 Lee S.J (2004) Regulation of muscle mass by myostatin Annual Review of Cell and Developmental Biology 20, 61^86 Lee S.J & McPherron A.C (2001) Regulation of myostatin activity and muscle growth Proceedings of the National Academy of Sciences of the United States of America 98, 9306^9311 Lee S.J., Reed L.A., Davies M.V., Girgenrath S., Goad M.E., Tomkinson K.N., Wright J.F., Barker C., Ehrmantraut G., Holmstrom J., Trowell B., Gertz B., Jiang M.S., Sebald S.M., Matzuk M., Li E., Liang L.F., Quattlebaum E., Stotish R.L & Wolfman N.M (2005) Regulation of muscle growth by multiple ligands signaling through activin type II receptors Proceedings of the National Academy of Sciences of the United States of America 102, 18117^18122 Lin J., Arnold H.B., Della-Fera M.A., Azain M.J., Hartzell D.L & Baile C.A (2002) Myostatin knockout in mice increases myogenesis and decreases adipogenesis Biochemical and Biophysical Research Communication 291,701^706 McPherron A.C & Lee S.J (1997) Double muscling in cattle due to mutations in the myostatin gene Proceedings of the National Academy of Sciences of the United States of America 94,12457^12461 McPherron A.C & Lee S.J (2002) Suppression of body fat accumulation in myostatin-de¢cient mice Journal of Clinical Investigation 109, 595^601 McPherron A.C., Lawler A.M & Lee S.J (1997) Regulation of skeletal muscle mass in mice by a new TGF-beta superfamily member Nature 387, 83^90 McPherron A.C., Lawler A.M & Lee S.J (1999) Regulation of anterior/posterior patterning of the axial skeleton by growth/di¡erentiation factor 11 Nature Genetics 22, 260^264 Medler S., Brown K.J., Chang E.S & Mykles D.L (2005) Eyestalk ablation has little e¡ect on actin and myosin heavy chain gene expression in adult lobster skeletal muscles Biological Bulletin 208, 127^137 Mykles D.L (2000) Puri¢cation and characterization of crustacean calpain-like proteinases Methods in Molecular Biology 144, 55^66 Mykles D.L & Skinner D.M (1985) The role of calcium-dependent proteinase in molt-induced claw muscle atrophy Progress in Clinical and Biological Research 180, 141^150 Nakashima M.,Toyono T., Akamine A & Joyner A (1999) Expression of growth/di¡erentiation factor 11, a new member of the BMP/TGFbeta superfamily during mouse embryogenesis Mechanisms of Development 80, 185^189 Ostbye T.K., Wetten O.F., Tooming-Klunderud A., Jakobsen K.S., Yafe A., Etzioni S., Moen T & Andersen O (2007) Myostatin (MSTN) gene duplications in Atlantic salmon (Salmo salar): evidence for di¡erent selective pressure on teleost MSTN-1 and -2 Gene 403, 159^169 Radaelli G., Rowlerson A., Mascarello F., Patruno M & Funkenstein B (2003) Myostatin precursor is present in several tissues in teleost ¢sh: a comparative immunolocalization study Cell andTissue Research 311, 239^250 Rebhan Y & Funkenstein B (2008) Inhibition of ¢sh myostatin activity by recombinant ¢sh follistatin and myostatin prodomain: potential implications for enhancing muscle growth in farmed ¢sh Aquaculture 284, 231^238 Rescan P.Y., Jutel I & Ralliere C (2001) Two myostatin genes are di¡erentially expressed in myotomal muscles of the trout (Oncorhynchus mykiss) Journal of Experimental Biology 204, 3523^3529 Rodgers B.D & Weber G.M (2001) Sequence conservation among ¢sh myostatin orthologues and the characterization of two additional cDNA clones from Morone saxatilis and Morone americana Comparative Biochemistry and Physiology Part B, Biochemistry and Molecular Biology 129, 597^603 Rodgers B.D., Roalson E.H.,Weber G.M., Roberts S.B & Goetz F.W (2007) A proposed nomenclature consensus for the myostatin gene family AmericanJournal of Physiology, Endocrinology and Metabolism 292, E371^E372 Scamu¡a N., Calvo F., Chretien M., Seidah N.G & Khatib A.M (2006) Proprotein convertases: lessons from knockouts FASEB Journal 20,1954^1963 Sharma M., Kambadur R., Matthews K.G., Somers W.G., Devlin G.P., Conaglen J.V., Fowke P.J & Bass J.J (1999) Myostatin, a transforming growth factor-beta superfamily member, is expressed in heart muscle and is upregulated in cardiomyocytes after infarct Journal of Cellular Physiology 180,1^9 Snyder M.J & Chang E.S (1991) Ecdysteroids in relation to the molt cycle of the American lobster, Homarus americanus I Hemolymph titers and metabolites General and Comparative Endocrinology 81,133^145 Tamura K., Dudley J., Nei M & Kumar S (2007) MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0 Molecular Biology and Evolution 24, 1596^1599 Thomas G (2002) Furin at the cutting edge: from protein tra⁄c to embryogenesis and disease Nature Reviews Molecular Cell Biology 3,753^766 Vianello S., Brazzoduro L., Dalla Valle L., Belvedere P & Colombo L (2003) Myostatin expression during development and chronic stress in zebra¢sh (Danio rerio) Journal of Endocrinology 176, 47^59 Wicks S.J., Grocott T., Haros K., Maillard M., ten Dijke P & Chantry A (2006) Reversible ubiquitination regulates the Smad/TGF-beta signalling pathway Biochemical SocietyTransactions 34,761^763 Xu C.,Wu G., ZoharY & Du S.J (2003) Analysis of myostatin gene structure, expression and function in zebra¢sh Journal of Experimental Biology 206, 4067^4079 Zhu X.,Topouzis S., Liang L.F & Stotish R.L (2004) Myostatin signaling through Smad2, Smad3 and Smad4 is regulated by the inhibitory Smad7 by a negative feedback mechanism Cytokine 26, 262^272 r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e862^e871 e871 Aquaculture Research, 2010, 41, e872^e877 doi:10.1111/j.1365-2109.2010.02611.x Limited evidence for genetic variation for resistance to the white spot syndrome virus in Indian populations of Penaeus monodon Ben J Hayes1,2,Thomas Gitterle3,4, Gopalapillay Gopikrishna5, Chavali Gopal5, Gopal Krishna6, Shrivinas Jahageerdar6, Carlos Lozano4, Shankar Alavandi5, Sivagnanam Paulpandi5, Pitchaiyappan Ravichandran5 & Morten Rye4 Biosciences Research Division, Department of Primary Industries Victoria, Bundoora,Vic., Australia No¢ma, —s, Norway CENIACUA, BogotaŁ, Colombia, South America Akvaforsk Genetics Center AS, SunndalsÖra, Norway Central Institute for Brackish Water Aquaculture, Chennai,Tamil Nadu, India Central Institute for Fisheries Education,Versova, Mumbai, India Correspondence: B J Hayes, Biosciences Research Division, Department of Primary Industries Victoria, La Trobe R&D Park,1 Park Drive, Bundoora,Vic 3083, Australia E-mail: ben.hayes@dpi.vic.gov.au Abstract There has been a highly detrimental impact of the white spot syndrome virus (WSSV) on black tiger shrimp (Penaeus monodon) aquaculture in India Currently, no cost-e¡ective measures are available for controlling the disease One alternative is to improve WSSV resistance through a selective breeding programme for disease-resistant shrimp, provided that genetic variation exists for this trait The aim of this study was to evaluate the evidence for genetic variation in resistance to WSSV in P monodon sourced from Indian populations Post-larval shrimp (n 51950) from 54 full-sibling families were challenged with WSSV using WSSV-infected mince meat The heritability was estimated using four di¡erent statistical models ¢tted to the resulting time to death data, including two linear models and two Weibull proportional hazard frailty models None of the estimated heritabilities were signi¢cantly di¡erent from zero.We suggest three possible explanations for these results: there actually is very little variation between P monodon inWSSVresistance and all individuals are highly susceptible to the disease; there is genetic variation in resistance to WSSV in P monodon but we did not ¢nd it in our experiment because the level of challenge in the experiment was too high to allow genetic di¡erences to be expressed; the variation is due to e872 mutations conferring resistance, which are at a low frequency in the population, and we did not sample a broad enough genetic base to capture these mutations Keywords: Penaeus monodon, white spot syndrome virus, genetic variation Introduction The impact of the white spot syndrome virus (WSSV) on black tiger shrimp (Penaeus monodon) aquaculture in India has been highly detrimental since it was ¢rst reported in 1994 (Karunasagar, Otta & Karunasagar 1997, for a review see Escobedo-bonilla, Alday-sanz, Wille, Sorgeloos, Pensaert & Nauwynck 2008) All age groups and sizes of shrimp are a¡ected by WSSV, and in most kinds of production systems (Karunasagar et al 1997) Transmission of WSSVcan occur vertically, from infected broodstock to larvae, or horizontally, through the water column, or from animal^animal contact The disease is prevalent in commercial hatcheries in India Uma, Koteeswaran, Karunasagar and Karunsangar (2005) randomly sampled broodstock used in commercial hatcheries along the southeast Indian coast, and found that 39.4% tested positive for WSSV using a PCR test r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd Aquaculture Research, 2010, 41, e872^e877 Genetic variation for WWSV resistance in P monodon B J Hayes et al Despite a large research e¡ort, currently, there are no cost-e¡ective measures for controlling WSSV The development of vaccines for example is thought to be prevented by the absence of an acquired immunity system in crustaceans (however, for an alternative viewpoint, see Witteveldt 2006; Johnson, Van Hulten & Barnes 2008) An alternative is to improve WSSV resistance through a selective breeding programme for disease-resistant shrimp, provided that genetic variation exists for this trait Limited genetic variation in WSSV resistance has been demonstrated in Penaeus vannamei under controlled challenge testing conditions (Gitterle, Salte, Gjerde, Cock, Johansen, Salazar, Lozano & Rye 2005; Gitterle, Gjerde, Cock, Salazar, Rye,Vidal, Lozano, Erazo & Salte 2006) The heritability of WSSV resistance, calculated from survival across full-sibling families under challenge test conditions, ranged from 0.00 to 0.07, depending on whether the shrimp in the challenge test were infected as a result of consuming minced muscle tissue infected with WSSV, given an individual oral infection or infected through waterborne virus particles Gitterle et al (2006) concluded that the dosage of WSSV was better controlled with oral infection than with other methods, as all animals were exposed to approximately the same risk of infection at the same time, and that this should im- 80°E prove the accuracy of estimating the genetic variance and hence the accuracy of breeding values (BV) for use in the selection programme Genetic variation for WSSV resistance in P monodon has not been investigated previously The aim of this study was to evaluate the evidence for genetic variation in resistance to WSSV in P monodon sourced from Indian populations Material and methods Collection of broodstock and establishment of shrimp families With the goal of establishing a breeding programme in P monodon mainly focused on growth, survival under commercial conditions and resistance to WSSV, gravid females were collected from three di¡erent Indian states (Tamil Nadu, Andhra Pradesh and Andaman and Nicobar Islands) to ensure genetic variability of the base population (Fig 1) These females were spawned in commercial hatcheries and families were reared from postlarvae (PL) to tagging size in individual tanks placed at two di¡erent research stations: CIBA’s research station in Muttukadu and CIFE’s station in Kakinada A total of 54 full-sibling families were produced, and 10 of these families (four from Tamil 90°E 20°N 20°N Andhra Pradesh 10°N Andaman and Nicobar Islands Tamil Nadu 80°E 180 360 720 km 90°E r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e872^e877 10°N Figure Location of the sampled populations are indicated in red e873 Genetic variation for WWSV resistance in P monodon B J Hayes et al Nadu and six from Andhra Pradesh) were reared in both research stations so that family and location effects could be jointly estimated At Muttudaku, a total of 37 individually tagged families were reared (26 from Tamil Nadu, 10 from Andhra Pradesh and one from the Andaman Islands) These families were produced between 1st of July and 13th of August 2006 in commercial hatcheries and were stocked in CIBA’s station at an average stage of PL 15 over August and September 2006 At the CIFE station in Kakinada, a total of 27 families (23 from Andhra Pradesh and four fromTamil Nadu) were reared Families were produced in August and stocked in individual tanks on September 2006 in order to reach the tagging weight On average, 219 individuals from each of the families reared at CIFE and 198 animals from the families reared at CIBA were individually tagged using visible implant £uorescent elastomers (Godin, Carr, Hagino, Segura, Sweeney & Blankenship 1995) Challenge test for resistance against the WSSV In December 2006, approximately 30 tagged individuals per family were transferred to CIBA, Santhome Challenge test facilities in order to evaluate the family resistance to WSSV Families were placed in two tonne tanks for experimental infections (15 animals per tank per family) WSSV-infected minced muscle tissue was administered once at 16:00 hours on 20 December 2006 A total of 1950 animals were infected The test was terminated when all animals died Genetic analysis The ¢xed e¡ects used in all the models were origin [geographical origin of the family, three levels: Tamil Nadu Coast, Andhra Pradesh Coast and Andaman Islands (only one family)], the rearing place (two levels: CIBA research station or CIFE research station) and the infection tank (two levels: tank and tank 2) Because no half-sibling families were produced, only one random e¡ect was considered (actually a family e¡ect) Note that this means that family and tank of rearing are completely confounded in all models We used four di¡erent approaches to estimate the heritability of white spot resistance These were: (a) A linear animal model where disease resistance was de¢ned based on whether or not the animal was alive when the population reached 50% total mortality (model LAM) The ‘animals’ here are the challenged individuals e874 Aquaculture Research, 2010, 41, e872–e877 (b) A linear animal model based on time to death [hours post-infection (pi)] This model does not take censored observations into account; however, as no animal survived the challenge, the model is appropriate (LTM) (c) A Weibull proportional hazard frailty model based on time until death (days pi) and taking censored observations into account (WHD) (d) A Weibull proportional hazard frailty model based on time until death as above, but with hours pi rather than days (WHH) This model was ¢tted to determine whether there was an advantage in assessing survival hourly or whether daily records were su⁄cient Fixed e¡ects were equal in all models, while random e¡ects varied depending on whether it was an animal model (LAM and LTM) or a family model (WHD and WHH) A preliminary analysis was performed to estimate the relative e¡ect of the di¡erent ¢xed e¡ects over the mortality for the linear models and to check the assumption on proportionality in the hazards models The four models are now described in more detail LAM: A linear model was applied to the observed binary variable y (0 dead, 5alive) truncated at 50% overall mortality: yj ẳ Fi ỵ aj ỵ ej 1ị where Fi is the ¢xed e¡ect of the ith origin by rearing place by tank, aj is the random e¡ect of animal j assumed to be multivariate normal distributed with mean vector and covariance matrix Asa2, where A is the additive genetic relationship matrix For example, for a pair of full-siblings k and l, Akl will be 0.5 ej is the random residual for animal j, where ej is assumed to be normally distributed with the covariance matrix Ise2 LTM: A linear model was applied to the observed death times (hours of the animals), where y is now the time to death Fixed and random e¡ects are as in the previous model WHD: A sire^dam proportional hazard model was assumed for days to death (t): hijl tị ẳ h0 tị expFi ỵ fj ị where hijl(t) is the hazard function for animal l at time t, h0(t) is the baseline hazard function that follows a Weibull distribution (i.e lr(lt)r À 1), where r and l are the parameters of the Weibull distribution, Fi is the ¢xed e¡ects of the ith origin by rearing place by tank and fj is the random e¡ect of family j, assumed to be multivariate normal distributed with mean vec- r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e872^e877 Aquaculture Research, 2010, 41, e872^e877 Genetic variation for WWSV resistance in P monodon B J Hayes et al tor and covariance matrix Asf2, where A is the additive genetic relationship matrix Traditionally, f has been assumed to follow a loggamma distribution because of its £exibility and mathematical convenience The gamma distribution tends to show a log-normal distribution as the parameters of the gamma distribution of random e¡ects become larger (Kalb£eisch & Prentice 1980, p 26) and then f can be regarded as (at least approximately) normally distributed Therefore, it has been suggested to account for the genetic relationship between animals by assuming a multivariate normal distribution (Ducrocq 1987) WHH: A proportional hazard model was assumed for hours to death (t): The model and parameters are the same as in the WHD model For the linear models, the heritability for disease resistance was calculated as Results On day11of the challenge test, mortality reached100% (Fig 2) Mortality data were registered only for1555 animals, which corresponds to 80% of the infected animals Thus, 395 animals died, and information could not be retrieved from them (date and hour of death) This was probably due to cannibalism of the recently diseased animals In our experiment, animals started dying at day pi before reaching 100% mortality at day 11 pi The high and rapid mortality re£ects a very strong infection As shown in Fig 3a, no di¡erences in the mortality rate were observed among the tanks, and the risk of dying increased over time (Fig 3b) This is an atypical situation in experimental infections (Gitterle et al 2006) and also re£ects possible cannibalism leading to increased dosages over time In order to ful¢l the proportional hazard assumption in the Weibull model, the hazard ratio of the h2 ẳ s2a s2a ỵ s2e ị s2f þ s2e Accumulated mortality where sf2 is the sire^dam variance, se2 is the residual variance and is p2/6 in the Weibull frailty models (Ducrocq & Casella 1996) Correlations between estimated breeding values (EBVs) of full-sibling families from each model were calculated to assess the agreement between genetic predictions of the di¡erent methods Then, to evaluate the accuracy of each method, family full-sibling BV were independently predicted for the replicated tanks using the variance components estimated from all data as input parameters The Pearson correlation coe⁄cients between the resulting EBV (rEBV) from each tank are closely related to the accuracy of selection (rt) (Gitterle et al 2006) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 10 11 Day Figure Cumulative proportion of shrimp dead versus days after challenge Survival distribution function h ¼ 2s2f (b) Hazard function (a) Tank 0.8 Tank 0.6 0.4 0.2 10 11 12 Days 2.5 1.5 0.5 10 11 Day (c) Log negative Log SDF where sa2 is the additive genetic variance and se2 is the residual variance and for the sire^dam proportional hazard models as: –1 –2 –3 –4 –5 0.4 0.5 0.6 0.7 0.8 0.9 Log (Days) 1.1 1.2 Figure (a) Survival distribution function (SDF) against time from tanks and 2, (b) hazard function from tanks ^0;n ðtÞÞÞ from tanks1and against and and c) logðÀ logðS log t in days r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e872^e877 e875 Genetic variation for WWSV resistance in P monodon B J Hayes et al di¡erent levels (or strata) from the covariates in the model has to be constant over time This can be checked by plotting the values of the ^0;n ðtÞÞÞ against log t, where ðS ^0;n ðtÞÞ is logðÀ logðS the baseline survival function from each stratum Parallel lines indicate that the proportional hazard assumption holds Moreover, if the lines are straight, the baseline hazard function is assumed to follow a Weibull distribution Figure 3c shows the plot of the ^0;n ðtÞÞÞ against log t, from tanks and logðÀ logðS We can see straight, parallel lines, indicating that the proportional hazard assumption holds Similar results were obtained when the proportional hazard assumption was checked for origin and rearing place Therefore, it was not necessary to stratify for any ¢xed e¡ect in either of the two hazard models The heritability of resistance to white spot infection was not signi¢cantly di¡erent from zero for any of the statistical models used (Table1) In all cases, the lack of heritability was due to the lack of genetic variance rather than due to a high environment variance When the correlation between full-sibling mean BV was investigated, there were higher correlations among the BV from models that used time to death (LTM, WHH and WHD) than between LAM and the other models (Table 2) Table Heritabilities and their standard errors from four statistical models used to analyse data from a white spot challenge test Model h2 SE LAM LTM WHD WHH 0.000010 0.000798 0.000079 0.000231 0.0129 1.1187 LAM, linear model with (died) or (survive) as the independent variable; LTM, linear model with time to death (hours) as the independent variable; WHD, a Weibull proportional hazard frailty model based on time until death (days post-infection) and taking censored observations into account; WHH, a Weibull proportional hazard frailty model based on time until death (hours post-infection) and taking censored observations into account Table Rank correlations among full-sibling mean breeding values among the di¡erent models (see text or Table1for an explanation of the models) LTM WHH WHD LAM LTM WHH 0.71 0.59 0.56 0.88 0.86 0.99 All correlations were signi¢cant Po0.0001 e876 Aquaculture Research, 2010, 41, e872–e877 Table Pearson correlations of full-sibling family mean breeding values among tanks Model rEBV LAM LTM WHH WHD 0.093 0.089 0.026 0.023 LAM, linear model with (died) or (survive) as the independent variable; LTM, linear model with time to death (hours) as the independent variable; WHD, a Weibull proportional hazard frailty model based on time until death (days post-infection) and taking censored observations into account; WHH, a Weibull proportional hazard frailty model based on time until death (hours post-infection) As expected, with the lack of genetic variance, the correlations between full-sibling mean BV between tanks were very low in all the models and none of them was signi¢cant (Table 3) Discussion The major limitation of the current study is the inability to separate PL into tagging tank e¡ects and genetic e¡ects, as each full-sibling family was reared in a separate tank at this stage Thus, the PL to tagging tank variation will be included in our full-sibling means, biasing the estimates of genetic variation and heritability upwards In spite of this, we found little or no evidence for genetic variation in resistance to WSSV infection There are two possible explanations for this One possibility is that there actually is very little variation between P monodon inWSSVresistance and all individuals are highly susceptible to the disease As the virus was only ¢rst documented in 1992 (Chou, Huang, Wang, Chiang & Lo 1995), it is possible that it has become highly pathogenic to shrimp only very recently If this is the case, there would not have su⁄cient time for genetic mutations to confer resistance either to emerge in the population, or if such mutations exist, to reach moderate frequencies in the population Gitterle et al (2005, 2006) also found either very low or zero heritabilities for WSSV resistance in P vannamei The second possibility is that there is genetic variation in resistance to WSSV in P monodon, but we have not demonstrated such a variation in our experiment This may occur if either we have not sampled a wide enough genetic base with 54 full-sibling families or the level of challenge in the experiment was too high r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e872^e877 Aquaculture Research, 2010, 41, e872^e877 Genetic variation for WWSV resistance in P monodon B J Hayes et al to allow genetic di¡erences to be expressed The stocking densities used in this experiment were high, and higher mortalities are found at higher densities (Wu, Namikoshi1, Nishizawa1, Mushiake,Teruya & Muroga 2001) Another di⁄culty with the challenge test is that P monodon is highly cannibalistic, such that shrimp that survive beyond a few hours are likely to consume the dead shrimp, thus exposing themselves to much higher virus loads (Fig.3) Re¢nement of the challenge test protocol and constant removal of dead shrimp may result in more observed genetic variation for WSSV resistance Further, infection protocols other than the feeding with WSSV-infected meat used here could allow better control of the infection Gitterle et al (2006) suggested that the dosage of WSSV in a challenge test could be better controlled with oral infection than with other methods, as all animals were exposed to approximately the same risk on infection at the same time, and that this should improve the accuracy of estimating the genetic variance and hence the accuracy of BV for use in the selection programme Despite this, the levels of genetic variation forWSSVresistance in their study were low If genetic mutations conferring resistance to WSSV exist, but their frequencies are very low, then a very broad genetic base would have to be sampled in order to capture these mutations in the breeding programme One option would be to screen an extremely large number of individuals from a large number of subpopulations, either by breeding from the survivors of naturalWSSVoutbreaks or by performing a natural challenge, as suggested by Cock, Gitterle, Salazari and Rye (2009) The founder population for the breeding programme could then include these individuals Acknowledgments The authors would like to acknowledge the Norwegian Agency for Development Cooperation and The Indian Council of Agricultural Research for funding through the project ‘Genetic improvement of tiger shrimp through selective breeding for growth and white spot disease resistance’ References Chou H.Y., Huang C.V., Wang C.H., Chiang C.H & Lo C.F (1995) Pathogenicity of a baculovirus infection causing white spot syndrome in cultured penaeid shrimp in Taiwan Diseases of Aquatic Organisms 23,165^173 Cock J., Gitterle T., Salazari M & Rye M (2009) Breeding for disease resistance of Penaeid shrimps Aquaculture, 286, 1–11 Ducrocq V (1987) An analysis of length of productive life in dairy cattle PhD dissertation, Cornell University, Ithaca, NY, USA Ducrocq V & Casella G (1996) A Bayesian analysis of mixed survival models Genetics Selection Evolution 28, 505^529 Escobedo-Bonilla C.M., Alday-Sanz V.,Wille M., Sorgeloos P., Pensaert M.B & Nauwynck H.J (2008) A review on the morphology, molecular characterization, morphogenesis and pathogenesis of white spot syndrome Journal of Fish Diseases 31, 1^18 Gitterle T., Salte R., Gjerde B., Cock J., Johansen H., Salazar M., Lozano C & Rye M (2005) Genetic (co)variation in resistance to white spot syndrome virus (WSSV) and harvest weight in Penaeus (Litopenaeus) vannamei Aquaculture 246, 139^149 Gitterle T., Gjerde B., Cock J., Salazar M., Rye M.,Vidal O., Lozano C., Erazo C & Salte R (2006) Optimization of experimental infection protocols for the estimation of genetic parameters of resistance to white spot syndrome virus (WSSV) in Penaeus (Litopenaeus) vannamei Aquaculture 261, 501^509 Godin D.M., Carr W.H., Hagino G., Segura F., Sweeney J.N & Blankenship L (1995) Evaluation of a £uorescent elastomer internal tag in juvenile and adult shrimp Penaeus vannamei Aquaculture 139, 243^248 Johnson K.N., van Hulten M.C.W & Barnes A.C (2008) ‘Vaccination’of shrimp against viral pathogens: phenomenology and underlying mechanisms.Vaccine 26, 4885^4892 Kalb£eisch J.D & Prentice R.L (1980) The Statistical Analysis of Failure Time Data John Wiley and Sons, New York, NY, USA Karunasagar I., Otta S.K & Karunasagar I (1997) Histopathological and bacteriological study of white spot syndrome of Penaeus monodon along the west coast of India Aquaculture 153, 9^13 Uma A., Koteeswaran A., Karunasagar I & Karunsangar I (2005) Prevalence of white spot syndrome virus and monodon baculovirus in Penaeus monodon broodstock and postlarvae from hatcheries in southeast cost of India Current Science 89, 1619^1622 Witteveldt J (2006) On the vaccination of shrimp against white spot syndrome virus Wageningen University dissertation no 3882,Wageningen, the Netherlands Wu J.L., Namikoshi A., Nishizawa T., Mushiake K.,Teruya K & Muroga K (2001) E¡ects of shrimp density on transmission of penaeid acute viremia in Penaeus japonicus by cannibalism and the waterborne route Diseases of Aquatic Organisms 47, 129^135 r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e872^e877 e877 Aquaculture Research, 2010, 41, e878^e880 doi:10.1111/j.1365-2109.2009.02448.x SHORT COMMUNICATION Side effects of sexual maturation on heritability estimates in rainbow trout (Oncorhynchus mykiss) Mathilde Dupont-Nivet1, Bernard Chevassus1, Stephane Mauger1, Pierrick Ha¡ray2 & Marc Vandeputte1,3 INRA, UMR1313 UniteÔ de GeÔneÔtique Animale et Biologie InteÔgrative, Jouy-en-Josas, France Sysaaf, Section aquacole, Station SCRIBE, Campus de Beaulieu, Rennes, France Ifremer, Chemin de Maguelone, Palavas-les-Flots, France Correspondence: M Dupont-Nivet, INRA ^ UniteÔ de GeÔneÔtique Animale et Biologie InteÔgrative, Equipe GenAqua, 78352 Jouy en Josas Cedex, France E-mail: mathilde.dupont-nivet@jouy.inra.fr In ¢sh, sexual maturation is often considered to be a problem because it perturbs growth and product quality Therefore, it is common to select against early-maturing males in commercial breeding programmes (Gjedrem 2000) Genetic determinism of age at sexual maturation has been extensively studied in many species, and even some QTLs have been found (e.g Gjerde 1986; Hankin, Nicholas & Downey 1993; Longalong, Eknath & Bentsen 1999; Kause, Ritola, Paananen, Mantysaari & Eskelinen 2003; Haidle, Janssen, Gharbi, Moghadam, Ferguson & Danzmann 2008) However, the e¡ect of sexual maturation on heritability estimates for other traits is not documented In this paper, we present data collected during the beginning of sexual maturation of a group of rainbow trout and show how it biases heritability estimations The ¢sh studied were issued from a full-factorial mating between two dams and 45 sires Fish were all reared in the same tank since the eyed stage under a natural photoperiod and pedigrees were redrawn using 10 microsatellites Fish were harvested in April at 17 months of age They were killed on ice and several growth and quality traits were determined Body weight was the main trait studied in this paper The sex of the ¢sh was recorded by visual inspection of the gonads For females, none of the individuals showed signs of maturation For males, observers attempted to di¡erentiate non-maturing and maturing males but it turned out to be di⁄cult because we were at the very beginning of the maturation (this e878 rainbow trout strain usually spawns in November^ December) However, we recorded gonads weight to enable calculation of the gonado-somatic index (GSI 5100  gonads weight/body weight), which was used to determine which males were maturing or not.We studied several GSI thresholds: 0.1, 0.2, 0.3, 0.4 and 0.5 For each threshold, males with GSI above the threshold were considered to be maturing males We studied sex e¡ect and heritability estimates for each threshold Heritability was estimated using VCE5 (Groeneveld & Kovac 1990) with a sire model and sex and dam as ¢xed e¡ects Dam was set as a ¢xed e¡ect because there were only two dams in our mating design, and therefore between-dams variance is of no interest Two dams were used instead of one in order to avoid confusion of dominance e¡ects with additive genetic e¡ects in the between-sires variance Moreover, this design was demonstrated to be e⁄cient for estimating heritabilities when the total number of o¡spring analysed is ¢xed (Dupont-Nivet, Vandeputte & Chevassus 2002) First, we analysed all the datasets with a sex e¡ect comprising three levels (female, male and maturing male) Second, maturing males were removed from the dataset and the sex effect was set to two levels (male and female) Using ‘FAP’ software (Taggart 2007), 87% of ¢sh could be attributed to their parents O¡spring from two sires with less than three o¡spring were removed from the data set Four hundred and thirty-three animals were ¢nally analysed The sex ratio was well equilibrated with 217 females, i.e., 50.1% In Table 1, r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd Aquaculture Research, 2010, 41, e878^e880 Sexual maturation and heritability estimates M Dupont-Nivet et al Table Data description: ¢sh number and sex e¡ect on mean body weight Fish number Mean body weight (g) GSI threshold for maturing males Non-maturing male Maturing male Femaleà Non-maturing male Maturing male Femaleà Sex effect GSI40.5 GSI40.4 GSI40.3 GSI40.2 GSI40.1 Observed maturity status 178 148 113 67 23 146 38 68 103 149 193 70 217 217 217 217 217 217 439.47b 430.82b 430.9b 412.7b 349.74b 433.0b 537.8a 513.26a 485.2a 476.6a 469.5a 506.3a 449.5b 449.5b 449.5b 449.5a 449.5a 449.5b o0.0001 o0.0001 0.0015 0.0017 o0.0001 o0.0001 ÃNone of the females were identi¢ed as maturing Means with di¡erent letters are signi¢cantly di¡erent (Po0.05) GSI, gonado-somatic index Table Heritability (Ỉ SE) estimates of body weight GSI threshold h for all animals GSI40.5 GSI40.4 GSI40.3 GSI40.2 GSI40.1 Females only 0.14 0.18 0.16 0.14 0.18 – Ỉ Ỉ Ỉ Ỉ Ỉ 0.08 0.09 0.09 0.08 0.09 h2 when maturing males are excluded 0.16 0.21 0.33 0.36 0.60 0.61 Ỉ Ỉ Ỉ Ỉ Ỉ Ỉ 0.09 0.11 0.15 0.15 0.20 0.21 GSI, gonado-somatic index the proportions of maturing and non-maturing males are presented depending on the GSI value chosen for the threshold Sex e¡ects on body weight are also reported Maturing males were signi¢cantly larger than non-maturing males The number of maturing males visually identi¢ed at slaughter was similar to that obtained from setting a GSI threshold of 0.4 In Table 2, we present heritability estimates for body weight either when all animals were retained for analysis or when maturing males were excluded from the dataset Heritability estimates were much larger when maturing males were removed, and the lower the threshold, the higher the heritability estimate Moreover, it appears that even if maturing males were recorded, addition of a maturing male level in the model was not enough to correct estimates Thus, sexual maturation had a huge impact on the additive genetic variability of weight, which could not be accounted for as a ¢xed e¡ect in classical quantitative genetics models Larger datasets were needed to precisely estimate the magnitude of the effect of sexual maturation Probably, physiological changes due to maturation highly perturb growth so that growth before and growth during maturation are two di¡erent traits, with di¡erent genetic determinisms This could partly explain the range of estimates found in the same species for the same trait, for example 0.1^0.6 for growth in rainbow trout (Gjerde 1986; Martyniuk, Perry, Moghadam, Ferguson & Danzmann 2003) For breeding values, the consequences may also be important, leading to less e⁄cient selection However, this should be investigated further For other traits, for example fatmeter data (indirect measure of fat level) for which there is no signi¢cant sex e¡ect (P40.2), no e¡ect on heritability was observed: estimates (0.72^0.82) were similar whether maturing males were included in the dataset or not In most species, it is easy to ¢nd mature males but it is not possible to record which males are maturing if gonads are not observed or/and weighed For traits in£uenced by sex, special care should be taken in new species for which the physiology and consequences are not as well known as in salmonids Even in salmonids, the question is not simple as maturing males cannot be externally identi¢ed in the early stages of maturation that we described here The best solution is of course to measure ¢sh before the onset of maturation, but this is not always possible if commercial sizes are targeted Many breeding programmes eliminate early-maturing ¢sh (Longalong et al 1999; Gjedrem 2000) Another solution used in rainbow trout is to work with triploid ¢sh but this does not solve the problem for large sizes because male maturation is generally not totally suppressed in triploid ¢sh It is also possible to work with all-female populations, in which male maturation is, of course, not a problem anymore (Chourrout & Quillet 1982; Galbreath 1994) r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e878^e880 e879 Sexual maturation and heritability estimates M Dupont-Nivet et al References Chourrout D & Quillet E (1982) Induced gynogenesis in the rainbow trout: sex and survival of progenies of all-triploid populations Theoretical and Applied Genetics 63, 201^205 Dupont-Nivet M.,Vandeputte M & Chevassus B (2002) Optimization of mating designs for inference on heritability in ¢sh species Aquaculture 204, 361^370 Galbreath P.F (1994) Freshwater performance of all-female diploid and triploid Atlantic salmon Aquaculture 128, 41^49 Gjedrem T (2000) Genetic improvement of cold-water ¢sh species Aquaculture Research 31, 25^33 Gjerde B (1986) Growth and reproduction in ¢sh and shell¢sh Aquaculture 57, 37^55 Groeneveld E & Kovac M (1990) A generalized computing procedure for setting up and solving mixed linear models Journal of Dairy Science 73, 513^531 Haidle L., Janssen J.E., Gharbi K., Moghadam H.K., Ferguson M.M & Danzmann R.G (2008) Determination of Quantitative Trait Loci (QTL) for Early Maturation in Rainbow Trout (Oncorhynchus mykiss) Marine Biotechnology 10, 579^592 Hankin D.G., Nicholas J.W & Downey T.W (1993) Evidence for inheritance of age of maturity in Chinook salmon (On- e880 Aquaculture Research, 2010, 41, e878–e880 corhynchus tshawytscha) Canadian Journal of Fisheries and Aquatic Science 50, 347^358 Kause A., Ritola O., Paananen T., Mantysaari E & Eskelinen U (2003) Selection against early maturity in large rainbow trout Oncorhynchus mykiss: the quantitative genetics of sexual dimorphism and genotype-by-environment interactions Aquaculture 228, 53^68 Longalong F.M., Eknath A.E & Bentsen H.B (1999) Response to bi-directional selection for frequency of early maturing females in Nile tilapia (Oreochromis niloticus) Aquaculture 178,13^25 Martyniuk C., Perry G.M.L., Moghadam H.K., Ferguson M.M & Danzmann R.G (2003) The genetic architecture of correlation among growth-related traits and male age at maturation in rainbow trout Journal of Fish Biology 63,746^764 Taggart J (2007) FAP: an exclusion-based parental assignment program with enhanced predictive functions Molecular Ecology Notes 7, 412^415 Keywords: ¢sh, genetic parameters, sex e¡ect, growth r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e878^e880 Aquaculture Research, 2010, 41, e881^e885 doi:10.1111/j.1365-2109.2010.02582.x SHORT COMMUNICATION The effect of different macroalgae on the growth of sea cucumbers (Apostichopus japonicus Selenka) Ying Liu, Shuanglin Dong, Xiangli Tian, Fang Wang & Qinfeng Gao The Key Laboratory of Mariculture, Ministry of Education, Ocean University of China, Qingdao, China Correspondence: S Dong,The Key Laboratory of Mariculture, Ministry of Education, Ocean University of China, Qingdao 266003, China E-mail: dongsl@ouc.edu.cn Sea cucumberApostichopus japonicus is an important aquaculture species, especially in northern China (Chen1990), and ingests sediment containing organic matter (Moriarty1982; Zhang, Sun & Wu1995) In a semi-intensive culture, sea cucumbers are fed formulated diets, which are commonly made of macroalgal powder and sea mud (Chang,Yi & Mu 2003) The algae Sargassum thunbergii and Sargassum polycystum are the most commonly used and highquality food for sea cucumbers With the expansion of sea cucumber farming, however, the resources of S thunbergii and S polycystum have decreased dramatically Their price has increased in recent years (Yuan 2005) In order to meet the demand of sea cucumber farming in northern China, it is necessary to ¢nd other local substitutes for these macroalgae The alga Laminaria japonica, with its extensive sources and low price in China (Yuan 2005), is widely used in the aquaculture of abalone (Qi, Liu, Li, Mao, Jiang, Zhang & Fang 2010), sea urchin (Chang & Wang1997; Agatsuma 2000) Zhu, Liu, Leng, Qu,Wang, Xue and Sun (2007) also reported that fresh kelp L japonica could be a part of the diet of sea cucumber Laminaria japonica holdfast is a by-product of kelp processing, but it contains many bioactive substances (Bu, Chen, Wang, Zhou, Zheng & Lin 2009) The feasibility of L japonica for the diet of sea cucumbers needs to be investigated In order to ¢nd a new substitute for Sargassum as a diet in a semi-intensive culture of the sea cucumber, we determined the growth rate, ingestion rate, faeces production rate and feed e⁄ciency of sea cucumbers fed diets containing powders of algae of S thunbergii, r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd S polycystum, L japonica thallus and L japonica holdfast Eight compound diets were formulated, and every compound contained 80% algal powder (dry weight) and 20% sea mud or yellow soil (dry weight) SST ^ sea mud and S thunbergii, YST ^ yellow soil and S thunbergii, SSP ^ sea mud and S polycystum,YSP ^ yellow soil and S polycystum, SLT ^ sea mud and L japonica thallus, YLT ^ yellow soil and L japonica thallus, SLH ^ sea mud and L japonica holdfast,YLH ^ yellow soil and L japonica holdfast Sea mud (Sea mud was sediment collected from the intertidal zone, near Qingdao, China); yellow soil (collected from farmland 1m underground, near Qingdao, China) and macroalgae (Shandong Liuhe Group, Qingdao City, Shandong Province, China) were dried at 65 1C to a constant weight, ground and sieved using a 0.15 mm mesh The proximate composition of sea mud and yellow soil is shown in Table Powdered algae were well mixed with mud or soil in designated proportions The slurry was slightly stirred, extruded by a meat grinder to a cylindrical form (diameter: 0.5^0.8 mm, length: 1.0^1.5 mm), then dried at 65 1C for 36 h and stored at 1C for use The proximate composition of the diets was analysed as follows: the protein content was estimated using a Vario ELIII Elemental Analyzer (Elementar, Dortmund, Germany), lipid was estimated using the Soxhlet method and ash was determined using a mu¥e furnace (SXL-1030, Jinghong Experimental Instrument, Shanghai, China) at 550 1C The proximate compositions of the eight diets are listed in Table e881 The e¡ect of di¡erent macroalgae on growth of A japonicus Y Liu et al Aquaculture Research, 2010, 41, e881–e885 Table Nutrient content of eight test diets Nutrient contentà Diet treatments Protein (%) Lipid (%) Ash (%) Energy (kJ g À 1) SST YST SSP YSP SLT YLT SLH YLH Yellow soil composition Sea mud composition 13.00 12.73 11.74 11.76 6.96 6.84 7.64 7.57 0.13 0.22 7.00 7.03 7.39 6.72 6.75 6.53 5.46 5.71 0.57 0.26 48.56 48.44 48.66 48.65 59.71 59.35 61.99 61.77 91.27 92.70 7.72 7.67 7.44 7.55 5.69 5.62 5.47 5.58 0.19 0.27 ÃData were calculated on a dry matter basis SST, sea mud and Sargassum thunbergii; YST, yellow soil and S thunbergii; SSP, sea mud and Sargassum polycystum; YSP, yellow soil and S polycystum; SLT, sea mud and Laminaria japonica thallus; YLT, yellow soil and L japonica thallus; SLH, sea mud and L japonica holdfast; YLH, yellow soil and L japonica holdfast Sea cucumbers were collected from Jiaonan, Qingdao, China, and were acclimated for 15 days before the start of the experiment During acclimation and the experiment, seawater temperature was set at 17 Ỉ 0.5 1C; dissolved oxygen was maintained above 5.0 mg L À 1; the level of ammonia was o0.25 mg L À 1; salinity ranged from 28 to 30 g L À 1; and a photoperiod of 13:11h (light:dark) was used During the acclimation period, A japonicus were fed once daily to excess After 48 h of starvation, 128 sea cucumbers were measured individually The wet weight measurements were taken within one of removal from seawater External water was removed from specimens by drying them on sterile gauzes (Dong, Dong, Tian, Wang & Zhang 2006) Sea cucumbers were allocated in equal numbers into 32 glass tanks (55  30  35 cm,40 L) to yield eight treatments and four replicates A complete randomized block design was used in the experiment The trial lasted for 55 days Twenty sea cucumbers were sampled before the experiment to determine the initial water content During the experiment, the animals were fed once daily to excess at 17:00 hours Faeces and uneaten feed were collected by siphon after 20 h of feeding, and then dried at 65 1C to a constant weight Potential loss of uneaten feed was determined by placing feed in seawater for 20 h and then collecting, drying and weighing The proportion remaining was calculated and this value was used to adjust the amount of the feed intake The animals were starved for 48 h at the end of the experiment, weighed individually and then dried at 65 1C to a constant weight The e882 tank’s mean average values were used for statistical analysis Speci¢c growth rate (SGR), ingestion rate (IR), faeces production rate (FPR) and feed e⁄ciency (FE) were calculated as follows: SGR ð% day1 ị ẳ 100 ln DW2 ln DW1 ị T ; IR g g1 day1 ị ẳ C=ẵT DW2 ỵ DW1 ị=2; FPR g g1 day1 ị ẳ F=ẵT DW2 ỵ DW1 ị=2; FE %ị ẳ 100 DW2 À DW1 Þ=C where DW1 and DW2 are the initial and the ¢nal dry body weight of sea cucumbers;T is the duration of the experiment; C is the dry weight of feed consumed and F is the dry weight of faeces Two-way analysis of variance (ANOVA) was used to compare the e¡ects of the two experimental treatments, i.e sediment types (sea mud or yellow soil) and algal species, on the body weight, SGR, IR, FPR and FE, followed by post hoc Duncan multiple range tests (Zar 1999) Before multiple comparisons, the interaction of the two factors was analysed For the data with a non-signi¢cant interaction (P40.05), the interaction item was eliminated from the ANOVA model and the main e¡ects of sediment types (sea mud or yellow soil) and algal species were further compared As for the data with signi¢cant interaction (Po0.05), the e¡ects of each treatment were analysed within the respective level of another treatment with Bonferroni adjustment to correct the magni¢cation of type I error (Corston & Colman 2000) Before ANOVA comparisons, raw data were diagnosed for normality of distribution and homogeneity of variance using r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e881^e885 Aquaculture Research, 2010, 41, e881^e885 The e¡ect of di¡erent macroalgae on growth of A japonicus Y Liu et al The algae S thunbergii and S polycystum are highquality diets for sea cucumbers with higher protein and lower algin (Zhu et al 2007; Slater, Je¡s & Carton 2009) Previous studies found that proteases from the digestive tract of sea cucumbers showed high activity (Fu, Xue, Miao, Li, Gao & Yang 2005; Jiang, Yang, Li, Wang,Wang & Lu 2007), which indicated that sea cucumbers had the ability to digest protein (Wang, Tang, Xu & Cheng 2007) Zhu, Mai, Zhang, Wang and Xu (2005) reported that sea cucumbers showed better growth when fed diets containing 18.21^ 24.18% protein Sun, Liang, Yan and Chen (2004) found that the highest growth rate occurred when food protein reached 21.5% In the present study, S thunbergii and S polycystum diets contained a higher protein level, and contributed to the higher SGR in these treatments The alga L japonica contains a signi¢cant amount of polysaccharides such as cellulose, alginate and laminarin These polysaccharides are an important energy source for abalone, which has high speci¢c enzyme activities against these polysaccharides the Kolmogorov^Smirnov test and the Levene test respectively Analysis of variance was conducted using SPSS for Windows, release 11.5 (SPSS 2002) Two-way ANOVA suggested that the interaction effects of two kinds of soil (sea mud and yellow soil) and algae were not signi¢cant in terms of body weight, SGR, IR, FPR or FE (P40.05), and so the main e¡ects of two kinds of soil (sea mud and yellow soil) and algae were compared The results showed that there was no di¡erence in the body weight, SGR, IR, FPR or FE between groups fed diets containing sea mud and yellow soil (P40.05) Sea cucumbers fed with L japonica thallus had the lowest dry body weights (Po0.05) (Table 2) The SGR of the tested sea cucumbers are shown in Fig The lowest SGR (1.91^1.94% day À 1) was observed in the treatments fed with L japonica thallus (Po0.05) The IR (Fig 2) and FPR (Fig 3) of sea cucumbers fed with L japonica holdfast diets were 0.78^0.79 and 0.55^0.56 g g À day À 1, respectively, which were both signi¢cantly higher than those of the other treatments (Po0.05) The IR and FPR of test animals fed with L japonica thallus were 0.65^0.66 and 0.45^0.46 g g À day À respectively The lowest IR and FPR were found in treatments fed with S thunbergii and S polycystum No signi¢cant di¡erence in FE was found between treatments fed diets containing S thunbergii and S polycystum or between L japonica thallus and L japonica holdfast (P40.05) (Fig 4) However, the FEs of sea cucumbers fed diets containing L japonica thallus and L japonica holdfast (2.60^2.78%) were signi¢cantly lower than those fed diets containing S thunbergii and S polycystum (4.61^4.86%) (Po0.05) In the present study, no signi¢cant di¡erence was found in the growth of sea cucumber fed diets containing sea mud and yellow soil, which con¢rms Liu’s ¢nding (Liu, Dong, Tian, Wang & Gao 2009) that yellow soil is a suitable substitute for sea mud in the diet of sea cucumbers SGR (% d–1) 2.5 1.5 0.5 SST SSP SLT SLH YST YSP YLT YLH Diet treatments Figure Speci¢c growth rate (SGR) of Apostichopus japonicus (n 4) SST, sea mud and Sargassum thunbergii; YST, yellow soil and S thunbergii; SSP, sea mud and Sargassum polycystum; YSP, yellow soil and S polycystum; SLT, sea mud and Laminaria japonica thallus; YLT, yellow soil and L japonica thallus; SLH, sea mud and L japonica holdfast; YLH, yellow soil and L japonica holdfast Table Initial and ¢nal wet weight (WW) and dry weight (DW) of Apostichopus japonicus fed eight test diets (Mean Ỉ S.E.) SST Initial WW Initial DW Final WW Final DW 6.72 0.47 23.10 6.19 YST Ỉ Ỉ Ỉ Ỉ 0.04 0.01 0.71 0.04 6.71 0.47 21.66 6.07 SSP Ỉ Ỉ Ỉ Ỉ 0.01 0.00 1.06 0.09 6.70 0.47 25.46 6.62 YSP Ỉ Ỉ Ỉ Æ 0.03 0.01 1.03 0.10 6.71 0.47 23.07 6.10 SLT Æ Æ Æ Æ 0.03 0.01 0.63 0.02 6.71 0.47 20.19 4.96 YLT Ỉ Ỉ Ỉ Ỉ 0.0 0.01 0.90 0.03 6.69 0.47 18.34 4.79 SLH Ỉ Ỉ Ỉ Ỉ 0.03 0.01 0.78 0.06 6.70 0.47 23.49 6.16 YLH Ỉ Æ Æ Æ 0.02 0.01 1.20 0.06 6.72 0.47 21.74 5.91 Ỉ Ỉ Ỉ Ỉ 0.04 0.01 1.26 0.08 WW, wet weight; DW, dry weight; SST, sea mud and Sargassum thunbergii; YST, yellow soil and S thunbergii; SSP, sea mud and Sargassum polycystum; YSP, yellow soil and S polycystum; SLT, sea mud and Laminaria japonica thallus; YLT, yellow soil and L japonica thallus; SLH, sea mud and L japonica holdfast; YLH, yellow soil and L japonica holdfast r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e881^e885 e883 The e¡ect of di¡erent macroalgae on growth of A japonicus Y Liu et al IR (g g–1 d–1) 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 SST SSP SLT SLH YST YSP YLT YLH Diet treatments Figure Ingestion rate (IR) of Apostichopus japonicus (n 4) SST, sea mud and Sargassum thunbergii; YST, yellow soil and S thunbergii; SSP, sea mud and Sargassum polycystum; YSP, yellow soil and S polycystum; SLT, sea mud and Laminaria japonica thallus; YLT, yellow soil and L japonica thallus; SLH, sea mud and L japonica holdfast; YLH, yellow soil and L japonica holdfast FE (%) (Onishi, Suzuki & Kikuchi1985; Anzai, Enami, Chida, Okoshi, Omuro, Uchida & Nishiide 1991) However, sea cucumbers have lower speci¢c enzyme activities and may not digest macroalgae such as L japonica and Undaria pinnati¢da e⁄ciently (Wang et al 2007) Sun and Chen (1989) reported that there were high bacteria activities on degradation of alginate in the gut of sea cucumbers It is the function of bacteria to digest alginate in the gut of sea cucumber Zhu et al (2007) also found that young sea cucumbers had a low growth rate in the early stage of feeding L japonica thallus In the present study, sea cucumbers fed a diet containing L japonica thallus had the lowest SGR among all the treatments Laminaria japonica holdfast is the by-product of L japonica processing and is not widely used in aqua- Aquaculture Research, 2010, 41, e881–e885 SST SSP SLT SLH YST YSP YLT YLH Diet treatments Figure Feed e⁄ciency (FE) of Apostichopus japonicus (n 4) SST, sea mud and Sargassum thunbergii; YST, yellow soil and S thunbergii; SSP, sea mud and Sargassum polycystum; YSP, yellow soil and S polycystum; SLT, sea mud and Laminaria japonica thallus; YLT, yellow soil and L japonica thallus; SLH, sea mud and L japonica holdfast; YLH, yellow soil and L japonica holdfast culture nowadays In the present study, the protein level of L japonica holdfast was lower than those of S thunbergii and S polycystum (Table 1) However, sea cucumbers preferred to feed L japonica holdfast and had the highest IR among all treatments (Fig 2) Therefore, the SGR of sea cucumber fed the diet containing L japonica holdfast was higher than that of sea cucumbers fed the diet containing L japonica thallus treatment (Po0.05) Because of the higher IR of the animal for L japonica holdfast but a higher protein level of S thunbergii and S polycystum, there were no di¡erences among treatments of L japonica holdfast, S thunbergii and S polycystum (Fig 1) (P40.05) With a huge production (Zemke-White & Ohno 1999), L japonica holdfast can be used as a substitute for S thunbergii and S polycystum in sea cucumber farming FPR (g.g–1 d–1) 0.6 0.5 Acknowledgments 0.4 This work was supported by the National Natural Science Foundation of China (No 30871931) and the National Key Project of Scienti¢c and Technical of China (2006BAD09A01 and 200905020) and 111 Project of China (B08049) 0.3 0.2 0.1 SST SSP SLT SLH YST YSP YLT YLH Diet treatments Figure Faeces production rate (FPR) of Apostichopus japonicus (n 4) SST, sea mud and Sargassum thunbergii; YST, yellow soil and S thunbergii; SSP, sea mud and Sargassum polycystum; YSP, yellow soil and S polycystum; SLT, sea mud and Laminaria japonica thallus; YLT, yellow soil and L japonica thallus; SLH, sea mud and L japonica holdfast;YLH, yellow soil and L japonica holdfast e884 References Agatsuma Y (2000) Food consumption and growth of the juvenile sea urchin Strongylocentrotus intermedius Fisheries Science 66, 467^472 Anzai H., Enami Y., Chida T., Okoshi A., Omuro T., Uchida N & Nishiide E (1991) Polysaccharide digestive enzymes from midgut gland of abalone Bulletin of the College of r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e881^e885 Aquaculture Research, 2010, 41, e881^e885 The e¡ect of di¡erent macroalgae on growth of A japonicus Y Liu et al Agriculture and Veterinary Medicine Nihon University 48, 119^127 (in Japanese, with English abstract) Bu T., Chen L., Wang C.C., Zhou T., Zheng L.H & Lin X.Q (2009) Isolation of alpha-glucosidase inhibitors from Laminaria japonica root and analysis of its enzymatic property Chinese Journal of Marine Drugs 28, 13^16 (in Chinese, with English abstract) ChangY.Q & Wang Z (1997) The raft culture of the sea urchin Strongylocentrotus intermedius Journal of the Dalian Fisheries University 12,7^14 (in Chinese, with English abstract) Chang Z.Y., Yi J.L & Mu K.Q (2003) Factors of in£uence on growth and survival of Apostichopus japonicus Modern Fisheries Information 18, 24^26 (in Chinese, with English abstract) Chen J.X (1990) Brief introduction to mariculture of ¢ve selected species in China UNDP/FAO Regional Seafarming Development and Demonstration Project National Inland Fisheries Institute Kasetsart University Campus, Bangkok,Thailand, p 16 Corston R & Colman A (2000) A Crash Course in SPSS for Windows Blackwell, Oxford, UK DongY.W., Dong S.L.,Tian X.L.,Wang F & Zhang M.Z (2006) E¡ects of diel temperature £uctuations on growth, oxygen consumption and proximate body composition in the sea cucumber Apostichopus japonicus Selenka Aquaculture 255, 514^521 Fu X.Y., Xue C.H., Miao B., Li Z.J., Gao X & Yang W.G (2005) Characterization of proteases from the digestive tract of sea cucumber (Stichopus japonicus): high alkaline protease activity Aquaculture 246, 321^329 Jiang L.X.,Yang N., Li J.,Wang W.Q.,Wang R.J & Lu J (2007) E¡ects of temperature and PH on the activities of digestive enzymes in Apostichopus japonicus Oceanologia et Limnologia Sinica 38, 476^480 (in Chinese, with English abstract) LiuY., Dong S.L.,Tian X.L.,Wang F & Gao Q.F (2009) E¡ects of dietary sea mud and yellow soil on growth and energy budget of the sea cucumberApostichopus japonicus (Selenka) Aquaculture 286, 266^270 Moriarty D.J.W (1982) Feeding of Holothuria atra and Stichopus chloronotus on bacteria, organic carbon and organic nitrogen in sediments of the Great Barrier Reef Australian Journal of Marine and Freshwater Research 33, 255^ 263 Onishi T., Suzuki M & Kikuchi R (1985) The distribution of polysaccharide hydrolase activity in gastropods and bivalves Bulletin of the Japanese Society of Scienti¢c Fisheries 51, 301^308 Qi Z.H., Liu H.M., Li B., Mao Y.Z., Jiang Z.J., Zhang J.H & Fang J.G (2010) Suitability of two seaweeds, Gracilaria lemaneiformis and Sargassum pallidum, as feed for the abalone Haliotis discus hannai ino Aquaculture 300, 189^193 Slater M.J., Je¡s A.G & Carton A.G (2009) The use of the waste from green-lipped mussels as a food source for juvenile sea cucumber, Australostichopus mollis Aquaculture 292, 219^224 SPSS (2002) SPSS Base11.5 User’s Guide SPSS, Chicago, IL, USA Sun H., Liang M., Yan J & Chen B (2004) Nutrient requirements and growth of the sea cucumber, Apostichopus japonicus In: Advances in Sea Cucumber Aquaculture and Management (ed by A Lovatelli, C Conand, S Purcell, S Uthicke, J.-F Hamel & A Mercier), pp.327^331 FAO, Rome, Italy SunY & Chen D (1989) The microbial composition of Stichopus japonicus and its physiological property Oceanlogia et Limnologia Sinica 20, 300^307 (in Chinese, with English abstract) Wang J.Q.,Tang L., Xu C & Cheng J.C (2007) Histological observation of alimentary tract and annual changes of four digestive enzymes in sea cucumber (Apostichopus japonicus) Fisheries Science 26, 481^484 (in Chinese, with English abstract) Yuan C.Y (2005) Current status and development of feed in sea cucumber Fisheries Science 24, 54^56 (in Chinese) Zar J.H (1999) Biostatistical Analysis, Fourth edn PrenticeHall, Englewood Cli¡s, NJ, USA Zemke-White W.L & Ohno M (1999) World seaweed utilization: an end-of-century summary Journal of Applied Phycology 11, 369^376 Zhang B.L., Sun D.L & Wu Y.Q (1995) Preliminary analysis on the feeding habit of Apostichopus japonicus in the rocky coast waters o¡ Lingshan Island Marine Sciences 3,11^13 (in Chinese, with English abstract) Zhu J.X., Liu H., Leng K.L., Qu K.M., Wang S.H., Xue Z.N & SunY (2007) Studies on the e¡ects of some common diets on the growth of Apostichopus japonicus Marine Fisheries Research 28, 48^53 (in Chinese, with English abstract) Zhu W., Mai K.S., Zhang B.G., Wang F.Z & Xu G.Y (2005) Study on dietary protein and lipid requirement for sea cucumber, Stichopus japonicus Marine Sciences 29,54^58 (in Chinese, with English abstract) Keywords: sea cucumber, Apostichopus japonicus, growth, macroalgae, diets r 2010 The Authors Aquaculture Research r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, e881^e885 e885