Effects of different protein hydrolysate products and levels on growth, survival rate and digestive capacity in Asian seabass (Lates calcarifer Bloch) larvae

Aquaculture 428–429 (2014) 195–202

Contents lists available at ScienceDirect

Aquaculture journal homepage: www.elsevier.com/locate/aqua-online

Effects of different protein hydrolysate products and levels on growth, survival rate and digestive capacity in Asian seabass (Lates calcarifer Bloch) larvae Manee Srichanun a, Chutima Tantikitti a,⁎, Trond M. Kortner b, Åshild Krogdahl b, Rutchanee Chotikachinda a a b

Department of Aquatic Science, Faculty of Natural Resources, Prince of Songkla University, Hat Yai, Songkhla 90112, Thailand Aquaculture Protein Centre (a CoE), Department of Basic Sciences and Aquatic Medicine, Norwegian School of Veterinary Science, Oslo, Norway

a r t i c l e

i n f o

Article history: Received 22 October 2013 Received in revised form 6 March 2014 Accepted 7 March 2014 Available online 27 March 2014 Keywords: Asian seabass Protein hydrolysate Larval diet Digestive enzymes Gene expression

a b s t r a c t Larval rearing of Asian seabass is dependent on live food, similar to several other carnivorous fish species. The present study aimed to find suitable predigested protein sources for use in Asian seabass larval diets in order to reduce dependency on live feed. Fish muscle, squid mantle and soybean meal were hydrolyzed by Alcalase (alkaline enzyme), pepsin (acidic enzyme), and the combination of both enzymes producing a total of 9 protein hydrolysate (PH) products. The soybean meal was found least favorable for the larvae and omitted from the in vivo investigation. Fish muscle and squid mantle hydrolysate products were selected based on in vitro digestibility to replace fishmeal protein at 25% and 50% in isonitrogenous (50% protein) and isolipidic (20% lipid) diets. A diet with 0% PH was used as a control diet, and two other diets, one with minced mackerel muscle supplying protein, and a commercial diet were included as reference diets. The experimental diets were fed to larvae four times daily for 30 days from 17 days post hatch. Survival rate and growth performance were measured and enzyme activity and gene expression of stomach, pancreatic and brush border enzymes were determined. Larvae accepted all PH diets and the control diet well but did not accept the two reference diets which caused 100% mortality in the third week of the trial. In general, 25% PH inclusion increased larval growth compared to the 0% control group and increased specific activity of pancreatic and brush border enzymes. At 50% inclusion level, negative effects on growth performance and survival rate were observed. High levels of small peptides in the PH might have influenced digestive enzyme capacity as reflected in induced chymotrypsin activity but reduced mRNA level and specific enzyme activity of the brush border enzyme, leucine aminopeptidase. In conclusion, incorporation of 25% PH from either fish muscle or squid mantle treated with the Alcalase-pepsin enzyme combination improved digestive capacity and growth performance of the larvae with acceptable survival rate. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Early weaning of fish larvae with microbound diets has been studied over the past three decades with the aim of shortening the live prey feeding period. Live feeds tend to vary in nutritional value and often carry high levels of pathogenic bacteria. Additionally, live feeds are associated with high rearing costs and labor requirements. Pre-digested protein, or protein hydrolysate (PH) containing small size peptides has been used as a protein source in fish larval diets with beneficial effects Abbreviations: PH, protein hydrolysate; FM, fish muscle; SM, squid mantle; SB, soybean meal; A, Alcalase; P, pepsin; C, combination of pepsin and Alcalase; DH, degree of hydrolysis; AG, amino acid group; FAA, free amino acid; dph, days post hatch; ef-1α, elongation factor-1α; α-tub, α-tubulin; ubq, ubiquitin; pg, pepsinogen; try, trypsinogen; ctr, chymotrypsinogen; bal, bile salt activated lipase; amy, amylase; lap, leucine aminopeptidase; alp, alkaline phosphatase; Cq, quantification cycle. ⁎ Corresponding author. Tel./fax: +66 74 465102. E-mail address: [email protected] (C. Tantikitti).

http://dx.doi.org/10.1016/j.aquaculture.2014.03.004 0044-8486/© 2014 Elsevier B.V. All rights reserved.

on maturation of digestive organs (Zambonino-Infante et al., 1997) and improved digestion and assimilation of nutrients (Tonheim et al., 2005). Pre-digested protein has been shown to improve growth of many marine fish species (Cahu et al., 1999; Kvåle et al., 2009). Protein hydrolysate has also been shown to improve the immune system in European seabass larvae (Kotzamanis et al., 2007) and large yellow croaker (Tang et al., 2008). According to Cahu et al. (1999), a moderate PH inclusion level (25%) in microbound diets may improve survival rate, growth and the onset of the adult mode of digestion, whereas excessively high inclusion levels (N50%) can lead to reduced larval growth (Cahu et al., 1999; Carvalho et al., 2004; Kolkovski and Tandler, 2000; Kvåle et al., 2002). However, Sovoie et al. (2006) observed no significant effects of 10–20% PH on growth and survival rate in newly-hatched spotted wolffish (Anarhichas minor). The suitability and nutritional value of PH in larval diets depend on certain characteristics of the hydrolysate. Kotzamanis et al. (2007) suggested that the source and characteristics of the protein used, the

196

M. Srichanun et al. / Aquaculture 428–429 (2014) 195–202

hydrolytic method, and the degree of hydrolysis may influence larval growth. A study in Asian seabass showed that the incorporation of acid digested fish meal at 45% into microbound diets improved larval growth (Nankervis and Southgate, 2006). However, effects of other types of protein hydrolysates and the optimum level of incorporation have not been explored. The objective of the present study was to examine the effect of different PH products and incorporation levels on survival rate, growth and digestive capacity in Asian seabass larvae. Fish muscle (FM), squid mantle (SM) and full fat soybean meal (SB) were selected as protein sources for preparing PH using different enzymes: Alcalase, pepsin and a combination of the two. Alcalase is a commercial enzyme that has been widely used in the production of protein hydrolysates (Benjakul and Morrissey, 1997). Pepsin was selected in order to mimic the gastric digestion because most marine fish larvae lack a functional stomach during early larval stages, possibly limiting protein digestion (Rønnestad et al., 2003). In vitro protein digestibility using crude enzyme extracts from juvenile Asian seabass was determined for selection of the potential PHs. Subsequently, the best performing PHs were evaluated in vivo in a feeding trial. End point measurements included larval survival rate, growth, enzyme activity and gene expression of the main digestive enzymes. 2. Materials and methods

hydrolysis was initiated at pH 3, and subsequent alkalinization to pH 8 using 6 N NaOH prior to Alcalase hydrolysis. After hydrolysis, the enzymes were inactivated by incubating the reaction mixture at 95 °C for 20 min. In the case of PH treated with enzyme P, the hydrolysate was neutralized by 6 N NaOH prior to inactivation. After heat treatment the mixtures were centrifuged at 7400 ×g at 4 °C for 20 min and the supernatants were collected. All products were dried at 60 °C overnight, ground thoroughly into powdered hydrolysate and stored at − 20 °C for later incorporation into diets. 2.2. Analysis of PH products 2.2.1. Chemical composition Proximate composition of native protein sources, FM, SM, SB and protein hydrolysate products were measured. Moisture, protein, fat and ash were determined according to AOAC (1990): dry matter by drying in an oven at 105 °C, ash by combustion at 550 °C in a muffle furnace for 3 h, protein content using the Kjeldahl method (Kjeldahl apparatus, Gerhardt, Germany) and fat using dichloromethane extraction (Soxtec System HT6, FOSS TECATOR, Sweden). 2.2.2. Protein solubility Protein solubility was determined with the nitrogen solubility index (NSI) following the procedure of Morr (1985) which is calculated as follows:

2.1. Protein hydrolysate preparation A local marine food fish, the Japanese scad (Decapterus maruadsi), and longfin squid (Loligo pealeii) were purchased from the market, placed on ice and transported to the Department of Aquatic Science, Prince of Songkla University, Hat Yai. Upon arrival, both the FM and SM were filleted, chopped to 3 cm lengths and ground using a meat grinder (Super grinder, National, MK20NR, Japan). The SB (from Lee Feed Mill Public Company Limited) was ground (Super blender, National, MXT 2GN, Taiwan) and autoclaved at 121 °C, 15 psi for 15 min in order to inactivate heat-labile anti-nutritional factors. One hundread grams of each protein source was packed into polyethylene bags and stored at −20 °C until used. As shown in Table 1, the prepared FM, SM and SB were hydrolyzed using two different proteolytic enzymes: (P) pepsin (P7000, from porcine stomach mucosa; Sigma, St. Louis, MO, USA), and (A) alkaline protease (Alcalase 2.4L®, Sigma, Novozymes, Bagsvaerd, Denmark), as well as a combination of pepsin and alkaline protease (C). For PH treated with enzymes P and A, thawed samples were homogenized (Multiquick, Braun, MR400HC, Spain) in 0.1 M HCl (pH 3.0) and 0.2 M Tris–HCl (pH 8.0), respectively at a ratio of 1: 2 (w/v) for 2 min at ambient temperature. The enzyme concentration and hydrolysis time required to obtain the 50% degree of hydrolysis (DH) for each PH product was optimized prior to the processing. For the (C) type PH produced using the combination of both enzymes, pepsin

NSIð% Þ ¼ 100 

Protein content in supernatant : Total protein content in sample

2.2.3. Degree of hydrolysis The DH was calculated as a proportion (%) of free amino groups (AG) released after hydrolysis with respect to total AG in each sample (Adler-Nissen, 1979; Benjakul and Morrissey, 1997). The DH was calculated as follows: DH ¼ ½ðLt –L0 Þ=ðLmax –L0 Þ  100: Lt L0 Lmax

amount of α-AG released at time t amount of α-AG in original homogenate total α-AG in original homogenate obtained after acid hydrolysis (6 N HCl at 110 °C for 24 h)

2.2.4. The molecular weights of soluble protein fractions in hydrolysate products The molecular weights of soluble protein fractions were analyzed by gel filtration chromatography. Ten milligrams of dry hydrolysate powder were dissolved in 0.05 M sodium acetate (Merck, Darmstadt, Germany) buffer (pH 5.0), centrifuged at 7840 ×g (KUBOTA 3500,

Table 1 Conditions for production of different protein hydrolysate products at 50% DH. Protein sources

Hydrolyzed enzyme

Enz. conc.a (units g protein−1)

pH

Temp. (°C)

Time (h)

Fish muscle

Pepsinb (FMP) Alcalasec (FMA) Pepsin/Alcalase (FMC) Pepsin (SMP) Alcalase (SMA) Pepsin/Alcalase (SMC) Pepsin (SBP) Alcalase (SBA) Pepsin/Alcalase (SBC)

6.7 × 104 0.18 1.2 × 104/0.02 8.0 × 104 0.04 1.9 × 104/0.01 11.2 × 104 0.15 1.4 × 104/0.01

3.0 8.0 3.0/8.0 3.0 8.0 3.0/8.0 3.0 8.0 3.0/8.0

37 50 37/50 37 50 37/50 37 50 37/50

10 1 10/1 10 1 10/1 10 2 10/2

Squid mantle

Soybean meal

a b c

Enzyme concentration per reaction mixture containing 30 g of fish muscle and squid mantle and 10 g of soybean meal. Pepsin from porcine stomach mucosa 439 units mg solid−1. Alcalase 2.4 Anson unit g−1.

M. Srichanun et al. / Aquaculture 428–429 (2014) 195–202

Tokyo, Japan) for 10 min and separated on a FPLC column (Superdex™ peptide 10/300 GL, Amersham Biosciences, Uppsala, Sweden). Molecular weights of the peptide fractions were calculated and compared with distributions of the standard compounds: cytochrome c (Mw = 12,384), Aprotinin (Mw = 6512) and Vitamin B12 (Mw = 1355) (Sigma Chemical, MO, USA). The relative areas of each fraction were given as percentage relative to the total area (Šližytė et al., 2005). Four soluble peptide fractions were defined based on their respective retention times and according to the International Union of Pure and Applied Chemistry (IUPAC) peptide nomenclature, b 200 Da (corresponding mainly to free amino acids), 200–500 Da (corresponding mainly to di-and tripeptides), 500–2500 Da (corresponding mainly to oligopeptides), and N2500 Da (corresponding mainly to polypeptides) (Kotzamanis et al., 2007). 2.3. Assay of in vitro protein digestibility 2.3.1. Enzyme preparation A crude enzyme preparation for in vitro protein digestibility was obtained from the intestine of healthy juvenile Asian seabass (Lates calcarifer) (2.15 ± 0.06 g). The dissected intestine was homogenized with 50 mM Tris-buffer saline containing 5 mM CaCl2, at pH 8.0 at 4 °C, and centrifuged at 4 °C and 12,000 ×g for 30 min to obtain the supernatant. Protein content was determined by a modified Lowry method (Lowry et al., 1951). Thereafter, enzyme was diluted to contain 5 mg protein mL−1 and the trypsin activity was determined at pH 8.0. 2.3.2. Protein digestibility The in vitro protein digestibility assay was performed for each sample by a modification of the method of Tonheim et al. (2007). The three untreated protein sources and the nine PHs were suspended in 1: 5 w/v of 50 mM phosphate buffer (pH 8.0) in amounts corresponding to 10 mg of crude protein to make a final volume of 1 mL. Thereafter, 10 μL of 1% chloramphenicol was added to the samples. The crude enzyme containing 8.5 units of trypsin was added to each sample tube. Microtubes were incubated on a rotator holder (SB rotator holder, SB3, STUART, UK) at ambient temperature (25–26 °C) which was similar to the rearing water temperature. At 0, 1, 2, 3, and 5 h after adding the enzyme, the incubation was terminated by adding 250 μL of 40% trichloroacetic acid and samples were collected. The incubation mixtures were then centrifuged at 4 °C and 5000 ×g for 10 min. The digested protein in the supernatant was determined as liberated reactive amino groups of the peptides using the trinitrobenzene sulphonic acid (TNBS) method (modified from Ihekoronye, 1986). A 0.2 mL aliquot of either the undigested control (0 h) or one of the digested mixtures (1, 2, 3 and 5 h) was mixed thoroughly with 1 mL of 0.01% TNBS (P2297, Sigma-Aldrich, MO, USA) in 10 mM phosphate buffer (pH 8.2) and incubated at 60 °C for 1 h in the dark. The reaction was terminated by adding 1 mL of 1 M HCl and cooled to room temperature prior to spectrophotometric measurement at 420 nm. The concentration of the reactive amino groups was calculated using DL-alanine as a standard. The in vitro digestion procedure provided three parameters: enzymatically liberated AG, AG pre-digestion (AG at 0 h) and AG post-digestion. Liberation of AG was calculated as the AG at each time point minus the AG at 0 h. 2.4. Experimental diets and larval rearing Experimental diet compositions are shown in Table 2. The PH products were incorporated into the microbound diets at different ratios of fishmeal to hydrolysate protein: 100: 0 (PH 0), 75: 25 (PH 25), and 50: 50 (PH 50). The diets were formulated to be isonitrogenous (50% protein) and isolipidic (20% lipid). Extra ascorbic acid was added because larval seabass require a higher level of Vitamin C to improve growth, survival rate and reduction of skeletal deformities (Dabrowski, 1992). The ingredients were ground and sieved to less than150 μm size before they were mixed thoroughly, pelleted using a 1 mm sieve and

197

Table 2 Composition of experimental diets (% as-fed basis) for Asian seabass larvae. Ingredients

Diet PH 0

PH 25

PH 50

Fish meal FMAb FMPc FMCd SMAe SMPf SMCg Gelatin Dextrinh Fish oil Lecithin Vitamini Mineralj

67.20 – – – – – – 3.00 9.77 12.63 1.00 2.2 4.2

50.30 15.86 16.47 16.15 16.57 16.68 16.85 3.00 8.41–9.358 13.41–14.09 1.00 2.2 4.2

33.55 31.72 32.94 32.30 33.18 33.37 33.63 3.00 7.2–8.798 14.74–15.54 1.00 2.2 4.2

Proximate composition Protein Lipid

51.10 18.01

50.09–51.67 17.73–20.33

51.00–52.69 17.32–21.39

a

a

Premium grade fishmeal (70.4% protein). Fish muscle hydrolyzed by Alcalase. c Fish muscle hydrolyzed by pepsin d Fish muscle hydrolyzed by combination of Alcalase and pepsin e Squid mantle hydrolyzed by Alcalase. f Squid mantle hydrolyzed by pepsin. g Squid mantle hydrolyzed by combination of Alcalase and pepsin. h The range of dextrin depends on the amount of fishmeal and PH in order to fulfill to 100. i Vitamin premix donated by Thai Union Feed Mill Co., Ltd., Thailand = 1.2% + Vit. C 1.0%. j Mineral premix (g kg feed−1); NaH2PO4.2H20 15, CaHPO4 8, KCl 5, KH2PO4 10, NaCl 2, Filler 2. b

dried at 60 °C for 10 h. The pellets were then crushed and sieved to obtain 3 different particle sizes of b0.5, N0.5 b 1 and N1 mm. Minced mackerel muscle (fresh from a market) and a commercial larval diet (NRD 5/8, INVE, Belgium) were employed as reference diets. Eleven days post hatch (dph) Asian seabass larvae, reared in 28 ppt seawater, were obtained from the National Institute of Coastal Aquaculture (NICA), Department of Fisheries, Thailand. The larvae were reared in 1 m3 fiberglass tanks containing 500 L of water at the density of 20 fish L−1. They were acclimatized to freshwater by gradually lowering salinity by 5 ppt per day with addition of freshwater to the rearing tanks. During the acclimatization period, the fish larvae were fed with newly hatched Artemia sp. (Petrel Brand® Artemia cyst, Ocean Star International, Inc. The Great Salt Lake, USA) at the density of 10 individuals L−1 four times daily at 7 a.m., 11 a.m., 4 p.m. and 8 p.m. The tanks were cleaned and dead fish collected every day. The growth trial was started when the fish larvae were 17 dph. Two hundred fish were counted and weighed in groups before placement in each glass aquaria (34 L) for each treatment (15 dietary groups with 3 replicates per group — 45 aquaria in total). The larvae were gradually adapted to the experimental diets over 6 days by decreasing the density of Artemia nauplii from, 1 nauplii mL − 1 to 0.8 nauplii mL− 1 , 0.6 nauplii mL − 1 , 0.4 nauplii mL − 1 , 0.2 nauplii mL − 1 and 0 nauplii mL−1, respectively with an increasing amount of the experimental diets. The fish were fed the experimental diets in excess four times daily for 30 days. During the experimental period, the aquaria were cleaned daily, dead larvae were collected and counted, and rearing water was exchanged by 70–80% daily. Water temperature and dissolved oxygen were recorded daily. Ammonia, nitrite and nitrate were recorded every two weeks using a commercial test kit (Aqua-VBC®, Thailand). The maximum level of ammonia, nitrite and nitrate attained during the trial were 1 × 10−4, 0.001 and 0.08 g L−1, respectively. Temperature was between 26 and 28 °C and dissolved O2 was 7.7–7.8 mg mL−1.

198

M. Srichanun et al. / Aquaculture 428–429 (2014) 195–202

2.5. Sampling and dissection At the end of the 30-day rearing period (47 dph), all remaining fish were counted and weighed. Survival rate, weight gain and daily specific growth rate were calculated by the following equations:

Survival rate ¼ 100 

ðInitial No‐Final NoÞ Initial No

digestive enzyme were analyzed by one way ANOVA. Differences between means were determined and compared by Tukey's HSD test. Probability values of p b 0.05 were considered significant. A linear regression equation between specific chymotrypsin activity and level of peptide size was calculated and the corresponding coefficient (R2) of determination was tabulated by SPSS. 3. Results

    −1 −1 Weight gain ¼ final weight g fish ‐initial weight g fish :

3.1. Chemical characterization of hydrolysates and in vitro protein digestibility

Daily specific growth rate (Houde and Schekter, 1981) = (eg − 1 × 100)

Proximate compositions (dry matter) and protein solubility of the protein sources and PH products are presented in Table 3. The three FM hydrolysates had the highest protein content followed by SM hydrolysate and the SB hydrolysates. FMP and SMA contained the highest lipid content. Generally, hydrolysates produced with enzyme treatments P and C showed the highest ash content, while those treated with enzyme A had the lowest content. All PH products showed high protein solubility ranging from 95.2 to 99.2% but SB hydrolysates, particularly SBA and SBC, had significantly lower solubility values compared to the FMP hydrolysate. The molecular weight distribution of the soluble peptides and amino acids differed among the PH products (Fig. 1). In the FM hydrolysates, more than half of the amino acids were bound in polypeptides, somewhat less in oligopeptides, while the free amino acid (FAA) fraction was small and the tri- and di-peptide fractions were very small. The SB hydrolysates showed a dominance of polypeptide fractions. In the squid mantle hydrolysates SMP and SMC, oligopeptides dominated over polypeptides and both the FAA and tri- and di-peptide fractions were larger than those observed for the FM and SB products. The SMA hydrolysate, however, contained about equal amounts of the three amino acid fractions (poly, oligo and FAA), with very small amounts of di- and tri-peptides. Enzyme treatment improved the protein digestibility from all protein sources. Among the untreated native proteins and hydrolysates, SB and SB products showed the lowest in vitro digestibility values (i.e. the slowest release of amino acids) during the incubation which was observed at 1 and 5 h (Fig. 2). Based on these results, the SB hydrolysates were considered unsuitable as protein sources for the fish larvae, and were therefore omitted from the following growth trial.

where g ¼

ð ln final weightðgÞ– ln initial weightðgÞÞ : Experimental days

From each aquarium, fish were randomly collected, eight to ten for enzymatic analysis, and five fish for gene expression. In low survival rate groups, the numbers of selected fish were reduced. The fish were then euthanized in iced water, and dissected to separate the stomach and intestinal segments (pyloric caeca and intestine) for determination of enzyme activity. Dissected samples were rinsed in distilled water and subsequently transferred to 1.8 mL cryotubes before being immediately frozen in liquid nitrogen and stored at −80 °C. For gene expression, a segment of digestive organ was dissected into small pieces (b0.5 cm in any dimension) and placed immediately in RNAlater stabilization reagent (Ambion, Life Technologies, Carlsbad, CA, USA). Samples were stored at 4 °C overnight to allow thorough penetration of the solution into the tissue. Thereafter, the supernatant was removed and samples were stored at −20 °C until further processing. 2.6. Enzymatic measurements Tissue extracts of stomach and intestine were used for analyses of enzymatic activity of pepsin, trypsin, chymotrypsin, lipase, α-amylase, leucine aminopeptidase (lap) and alkaline phosphatase (alp) as previously described (Srichanun et al., 2013). 2.7. Quantitative real time PCR (qPCR) of digestive enzyme expression Messenger RNA (mRNA) expression of genes encoding pepsinogen (pg), trypsinogen (try), chymotrypsinogen (ctr), bile salt-activated lipase (bal), α-amylase (amy), leucine aminopeptidase (lap) and alkaline phosphatase (alp) were measured by quantitative real time PCR (qPCR). Primer design and optimization, RNA extraction, cDNA synthesis and qPCR assays were performed as described in detail previously (Srichanun et al., 2013). The geometric average expression of elongation factor 1α (ef-1α), α-tubulin (α-tub) and ubiquitin (ubq) was used as normalization factor. Mean normalized expression of the target genes was calculated from raw Cq values by relative quantification (Muller et al., 2002). 2.8. Statistical analysis Data were tested for normality and homogeneity of variance by the Shapiro–Wilk test and Levene's test, respectively, using SPSS 13 for Windows. To achieve normality, data on survival were log10 transformed, and daily specific growth rate were square-root transformed. Three-way ANOVA was used for evaluation of main effects of the class variables, i.e. protein source (FM and SM), enzyme treatment (A, P and C) and inclusion level of PH (25 and 50%), as well as interaction effects between protein sources, proteolytic method and inclusion level of PH on survival rate and growth performance. Proximate composition and protein solubility in PH products, specific activity and mRNA level of

3.2. Survival rate and growth performance The larvae easily accepted the experimental diets but did not accept the reference diets (minced mackerel muscle and commercial diet) which caused 100% mortality in these groups after three weeks of the Table 3 Proximate composition (% dry matter) and protein solubility (%) of different protein sources and protein hydrolysate products. Data are mean ± SEM (n = 3). Means with different superscripts in the same row are significantly different (p b 0.05). Protein

Lipid

Ash

Native protein FM 79.2 ± 2.65 SM 84.5 ± 1.52 SB 38.1 ± 0.10

1.6 ± 0.29 1.8 ± 0.14 23.9 ± 0.30

Hydrolysate products FMA 79.9 ± 0.57a FMP 76.8 ± 0.24ab FMC 78.2 ± 0.46ab SMA 77.1 ± 0.08abc SMP 76.6 ± 1.19ab SMC 75.3 ± 1.05bcd SBA 72.1 ± 0.38cde SBP 70.2 ± 0.68ce SBC 71.5 ± 0.40cde

2.5 6.0 2.4 4.5 2.5 3.3 3.8 2.75 3.5

± ± ± ± ± ± ± ± ±

0.29c 0.30a 0.04c 0.13ab 0.57c 0.45bc 0.29bc 0.12bc 0.41bc

Protein solubility

5.7 ± 0.13 6.6 ± 0.41 5.8 ± 0.09

6.1 16.2 17.3 10.3 19.8 20.5 6.1 17.9 19.5

± ± ± ± ± ± ± ± ±

0.13f 0.05d 0.09c 0.00e 0.09ab 0.17a 0.29f 0.30c 0.14b

26.5 ± 1.50 26.1 ± 0.47 11.4 ± 1.80

98.8 99.2 96.6 97.6 97.1 96.1 96.1 96.7 95.2

± ± ± ± ± ± ± ± ±

0.08ab 0.09a 0.14ab 0.98ab 0.09ab 0.21ab 0.52b 0.06ab 0.23b

M. Srichanun et al. / Aquaculture 428–429 (2014) 195–202

199

Table 4 Weight gain, daily specific growth rate and survival rate of Asian seabass larvae after 30 days rearing from 17 dph. Data are mean ± SEM (n = 3). Means with different superscripts in the same column are significantly different (p b 0.05). p values for the three way ANOVA analysis are given. Treatment

Fig. 1. Distribution of molecular weight peptide in protein hydrolysate products. FMA: fish muscle hydrolyzed with Alcalase, FMP: fish muscle hydrolyzed with pepsin, FMC: fish muscle hydrolyzed with combination of pepsin and Alcalase, SMA: squid mantle hydrolyzed with Alcalase, SMP: squid mantle hydrolyzed with pepsin, SMC: squid mantle hydrolyzed with combination of pepsin and Alcalase, SBA: soybean hydrolyzed with Alcalase, SBP: soybean hydrolyzed with pepsin, SBC: soybean hydrolyzed with combination of pepsin and Alcalase.

trial. Survival rates and growth performance of the larvae after 30 days are shown in Table 4. Three-way ANOVA showed that all class variables (i.e. protein source, enzyme treatment and hydrolysate level) affected survival rate and weight gain significantly (p b 0.05). Generally, survival rates were higher in the groups of fish fed diets containing PHs at the 25% inclusion level than those fed PH at 50% levels, with the exception of the 25% SMA treatment. At 25% PH inclusion level, there was a significant difference in survival rate between the FMA and SMA, whereas this difference was not significant at the 50% PH inclusion level. This indicated that survival rate in fish fed diets with the SMA was not affected by inclusion rate, whereas survival rate of fish fed FMA was lower at the higher inclusion level. Otherwise the differences caused by different protein sources and enzyme treatment were small. Weight gain and daily SGR were also significantly affected by dietary PH inclusion levels (p b 0.05), shown in Fig. 3. Generally, 25% dietary inclusion increased growth compared to the 0% control group, whereas 50% inclusion level led to reduce growth. However, the effect depended on both protein source and enzyme treatment. The FM hydrolysate outperformed the SM hydrolysate at 50% inclusion level. It is notable that the 25% SMA treatment, which had a very low survival compared to other 25% PH groups, gained nearly double the weight of the other 25% PH groups. However, at the 50% level, with a similarly low survival, the SMA group had the lowest SGR and weight gain.

Weight gain (g/fish) bc

Daily SGRa (%/day)

Survival rate (%)

19.48 ± 0.53

34 ± 7a

Control (0% PH)

0.795 ± 0.020

25% FMA FMP FMC SMA SMP SMC

0.978 0.955 0.969 1.834 0.850 0.683

± ± ± ± ± ±

0.145ab 0.049ab 0.017ab 0.019a 0.070 b 0.052bc

20.63 20.66 20.72 23.87 20.12 19.16

± ± ± ± ± ±

1.01 0.58 0.63 0.35 0.65 0.57

19 ± 3abc 22 ± 2ab 33 ± 1a 6 ± 1def 31 ± 2a 32 ± 2a

50% FMA FMP FMC SMA SMP SMC

0.978 0.768 0.901 0.167 0.707 0.417

± ± ± ± ± ±

0.444 b 0.047 bc 0.057 b 0.075 c 0.106bc 0.079bc

18.78 19.79 20.35 12.81 19.30 17.14

± ± ± ± ± ±

3.55 0.46 0.34 2.41 0.78 0.93

4 ± 1f 9 ± 2def 9 ± 1cde 5 ± 1ef 6 ± 1ef 12 ± 2bcd

ANOVA (p) Protein source (P) Hydrolysis enz. (E) Inclusion level (L) P∗E P∗L E∗L P∗E∗L

0.055 0.308 0.000 0.239 0.003 0.006 0.003

a

0.087 0.620 0.002 0.636 0.033 0.013 0.059

0.024 0.000 0.000 0.003 0.130 0.042 0.000

Interaction of factor P ∗ L and E ∗ L are presented in Fig. 3.

and the 50% SMC was highest for all enzymes except pepsin, lap and alp. Protein source obviously triggered more pronounced effect than enzyme treatments. Chymotrypsin and lipase activity in fish fed SM hydrolysate products at both inclusion levels were significantly higher than those of the control diet fed group. However, the activity of trypsin was not significantly different from the control group. In contrast, the,

3.3. Enzyme activity There was less difference in specific pepsin activity between fish fed PH containing diets and the control diet. On the other hand, most of the fish fed PH containing diets showed increased activity of pancreatic enzyme, trypsin, chymotrypsin, and lipase, whereas α-amylase activity was not significantly different (Table 5). This effect seemed to be mainly due to protein sources and enzyme treatments. For example, the 25% SMC fed group showed the highest levels of all enzymes except lap,

Fig. 2. Amino acid liberation (mM leucine) from in vitro protein digestion of native protein and hydrolysate protein at 1 and 5 h.

Fig. 3. Daily specific growth rate of fish fed protein hydrolysate with different protein sources and hydrolysis enzymes at the different inclusion levels (significant P ∗ L and E ∗ L interaction).

200

M. Srichanun et al. / Aquaculture 428–429 (2014) 195–202

Table 5 Specific enzyme activity (units mg protein−1): pepsin, trypsin, chymotrypsin, lipase, α-amylase, leucine aminopeptidase (lap) and alkaline phosphatase (alp). Data are mean ± SEM (n = 3). Means with different superscripts in the same row are significantly different (p b 0.05). Treatment

Pepsin

Trypsin

Chymotrypsin

Lipase

α-amylase

lap

alp

Control (0% PH)

596 ± 25.50ab

2.8 ± 0.63ab

16 ± 3.64ef

0.2 ± 0.02def

34 ± 12.20

13 ± 1.84ab

606 ± 67abc

25% FMA FMP FMC SMA SMP SMC

633 619 627 545 594 607

± ± ± ± ± ±

1.6 ± 0.17bc 2.4 ± 0.92abc 2.3 ± 0.35abc 3.8 ± 1.70abc 6.8 ± 1.78ab 11.4 ± 1.13a

50 ± 4.41def 42 ± 12.13ef 65 ± 25.80cf 207 ± 45.99a 157 ± 25.06abc 222 ± 18.16a

0.2 0.3 0.2 0.2 0.4 0.6

± ± ± ± ± ±

0.02bcdef 0.04bcdef 0.03cdef 0.03ef 0.00abc 0.04a

24 31 45 41 58 82

± ± ± ± ± ±

1.6 11.00 3.38 2.22 6.45 8.96

28 ± 7.54a 18 ± 3.18a 14 ± 2.56ab 3.6 ± 0.27 c 1.9 ± 0.25c 2.9 ± 0.10c

248 ± 18cd 621 ± 156abc 1002 ± 97a 61 ± 7d 856 ± 78ab 959 ± 49a

50% FMA FMP FMC SMA SMP SMC

649 597 513 −a 690 559

± 57.14ab ± 5.32ab ± 20.05b

13 ± 0.96ef 16 ± 3.17ef 136 ± 20.81abcd – 82 ± 29.68bcde 166 ± 2.04ab

0.1 0.3 0.3 – 0.3 0.4

± 0.01f ± 0.07bcde ± 0.02bcdef

28 41 55 – 62 92

± 1.19 ± 5.28 ± 13.99

22 ± 0.56a 15 ± 1.67ab 16 ± 1.04ab – 6 ± 2.13bc 1.9 ± 0.48c

598 534 819 – 349 450

a

14.84ab 12.38ab 22.10ab 35.89ab 38.63ab 22.94ab

± 33.63a ± 25.17ab

0.5 3.0 2.8 – 5.1 9.2

± 0.03c ± 0.16ab ± 1.01abc ± 1.48ab ± 1.88ab

± 0.07abcd ± 0.04ab

± 4.57 ± 38.56

± 88abc ± 114bc ± 17ab ± 65cd ± 83bcd

No sample.

fish fed diets containing FM hydrolysates showed a higher specific activity of the brush border enzyme lap than those fed SM hydrolysates at both inclusion levels. On the other hand, with alp activity there was no significant differences among all treatments, except for fish fed 25% SMA diets, which were significantly lower. Moreover, a correlation between peptide size and specific chymotrypsin activity was observed (Fig. 4). Fish fed diets containing high amounts of small peptides (b500 Da, composed of free amino acids and di/tri peptides) showed increased chymotrypsin activity. 3.4. mRNA levels Transcriptional levels of try, ctr, amy and alp were not significantly different among the groups (Table 6), however differences of pg, bal and lap mRNA levels were observed. The fish fed 25% SMA diet showed the lowest level of pg which was significantly different from those fed SMP at the same inclusion level and SMC at the 50% inclusion level. Similarly, bal mRNA level was high in fish fed SMP at 25% but decreased to the lowest level at higher inclusion levels. pg and bal mRNA levels of all treatments were not significantly different from those of the control group (p N 0.05) while the mRNA levels of the brush border membrane, lap, was reduced in fish fed the diets incorporated with SM hydrolysates in particular SMA and SMC at 25% inclusion.

Fig. 4. Correlation between contents of small peptide size (b500 Da) distributed in protein hydrolysate on specific chymotrypsin activity.

4. Discussion The use of in vitro protein digestibility methods to screen highly digestible ingredients in formulated diets for fish larvae has been applied successfully in several studies (Lindner et al., 1995; Perera et al., 2010; Tibbetts et al., 2011). Application of such technique to estimate the efficiency of feeds at the post larval stage can be performed by the proteolytic system of larger juveniles (Lindner et al., 1995). In this study, FM and SM hydrolysates were well digested by an enzyme mixture extracted from the intestine of juvenile Asian seabass, while SB hydrolysates were less well digested. This was not unexpected, because the Asian seabass is a strictly carnivorous fish. The proteins of fish muscle and squid mantle, containing higher amounts of arginine and lysine compared to soy protein, were more compatible with the specificity of the extracted seabass digestive enzymes. This study showed there was a positive correlation between high values of the sum of protein lysine and arginine and the degree of proteolysis achieved, which had been reported previously (Lindner et al., 1995). Moreover, the protein solubility of SB hydrolysate products was lower than those of FM and SM, probably a result of lower protein digestibility, which has been reported by others (Carvalho et al., 2004). Incorporation of PHs in the microbound diet for 17 dph Asian seabass larvae did not improve survival rate in comparison with the control diet. This observation is in line with the results of the study on larval development by Khochchawech (2013) showing that larvae at this stage have developed a complete digestive apparatus enabling effective digestion of both native and pre-digested protein. On the other hand, 25% PH inclusion generally improved the growth performance, reflected in the highest weight gain and specific growth rate in fish fed FMA, FMP, FMC and SMA. At 50% PH inclusion level, reduced growth and survival rate were observed. Similar phenomena were also reported by Cahu et al. (1999), Cahu et al. (2004), Carvalho et al. (2004), Kotzamanis et al. (2007) and Zambonino-Infante et al. (1997). The taste of the diets may have affected the diet acceptability. According to the human sense of taste, protein hydrolysate often has a bitter taste (Daukšas et al., 2004). The visual observation that fish fed 50% hydrolysate diets tended to spit out their feed implies that they may have found the taste unpleasant, and the lowered feed intake eventually resulted in low survival rates. In addition, physical properties of high PH incorporated diets may also have contributed to the low diet acceptability. During preparation of the diets with high PH inclusion, there seemed to be an interaction between the high moisture content and lipid causing the distribution

M. Srichanun et al. / Aquaculture 428–429 (2014) 195–202

201

Table 6 mRNA levels of digestive enzyme genes; pepsinogen (pg), trypsinogen (try), chymotrypsinogen (ctr), bile salt activated lipase (bal), α-amylase (amy), leucine aminopeptidase (lap) and alkaline phosphatase (alp). Data are mean ± SEM (n = 5). Means with different superscripts in the same row are significantly different (p b 0.05). Treatment

pg

try

ctr

bal (×10−3)

amy (×10−3)

lap (×10−3)

alp (×10−3)

Control

9 ± 1.90ab

3.43 ± 0.83

0.7 ± 0.04

14.17 ± 1.27ab

0.93 ± 0.19

7.7 ± 0.56ab

0.13 ± 0.02

25% FMA FMP FMC SMA SMP SMC

7 ± 1.07ab 6 ± 0.81ab 11 ± 2.11ab 3 ± 0.82b 11 ± 1.51a 9 ± 1.21ab

5.26 ± 1.18 8.06 ± 0.91 5.55 ± 1.36 3.25 ± 1.58 10.26 ± 2.70 2.98 ± 0.66

0.5 1.0 1.1 0.8 1.1 0.7

17.98 ± 2.34ab 17.93 ± 2.67ab 15.68 ± 2.98ab 10.37 ± 3.46ab 29.05 ± 9.59a 5.59 ± 0.99b

0.89 1.57 1.15 0.68 2.82 0.74

9.1 5.9 5.8 4.9 5.0 4.8

50% FMAa FMP FMC SMAa SMP SMC

3.09 5 ± 0.67ab 6 ± 0.87ab 8.01 6 ± 0.81ab 14 ± 3.47a

6.57 4.41 ± 3.00 ± 10.91 2.29 ± 2.87 ±

1.76 0.7 ± 0.6 ± 0.86 0.8 ± 0.7 ±

30.54 13.22 ± 3.20ab 9.82 ± 1.12ab 47.18 4.67 ± 0.14b 10.07 ± 2.10ab

1.51 1.00 0.76 3.85 0.43 0.65

a

1.57 0.42 0.49 0.53

± ± ± ± ± ±

0.05 0.27 0.47 0.34 0.41 0.24

0.08 0.12 0.01 0.13

± ± ± ± ± ±

0.19 0.12 0.23 0.26 1.10 0.19

± 0.24 ± 0.15 ± 0.02 ± 0.16

± ± ± ± ± ±

12.69 5.7 ± 5.5 ± 6.37 5.8 ± 5.0 ±

0.60a 0.42abc 0.36abc 0.57c 0.50bc 0.40c

0.23bc 0.22bc 0.65abc 0.57bc

0.14 0.13 0.12 0.17 0.10 0.17

0.18 0.13 0.15 0.18 0.16 0.20

± ± ± ± ± ±

0.02 0.03 0.02 0.03 0.01 0.06

± 0.00 ± 0.02 ± 0.01 ± 0.06

One replicate sample.

of lipid outside the pellet and harder pellet texture than those of the lower PH level and control diets. The peptide size distribution in the PH may play important roles in regulating enzyme activity. In the present study, high amounts of small peptides in the PH induced chymotrypsin activity, particularly in the SM hydrolysate fed groups. Zambonino-Infante et al. (1997) also found that chymotrypsin activity was induced by higher proportions of small peptides (di/tri peptide), whereas trypsin activity was enhanced by native protein. This is in contrast to brush border enzyme activity, where low specific activity and mRNA level of lap was observed in fish fed SM hydrolysate. The high proportion of small peptides in SM hydrolysate may not need further digestion before being transported across the brush border membrane, resulting in decreased gene expression and activity of lap, which are normally responsible for the final stages of protein digestion and assimilation (Klein et al., 1998). This indicated that the high proportion of small peptides had a negative effect on Asian seabass performance, by reducing the production of brush border enzymes and decreasing the survival rate, irrespective of SMA inclusion level. However, short chain peptides may influence other brush border enzymes which have not been determined in this study. For instance, Zambonino-Infante et al. (1997) found that γ-glutamyl transpeptidase (γGT) production was stimulated in fish fed diet containing short chain peptides in European seabass larvae. Moreover, fish fed the 50% SMA-PH diet containing high level of free amino acids showed the lowest specific growth rate. Similar results of reduced performance were found in European seabass larvae fed the PH containing high level of short chain peptides (Kotzamanis et al., 2007). The poor growth performance may be the result of a faster rate of absorption through the gut wall which caused premature absorption of essential amino acid resulting in the imbalance of amino acid absorption (Hardy, 1991) which consequently decreased protein synthesis (Rønnestad et al., 2000). In addition, the high amount of free amino acids may have resulted in increased catabolism of amino acids, something which has been observed to have a negative effect on the growth performance of pre-smolt Atlantic salmon (Espe and Lied, 1994). However, the specific enzyme and gene expression in the other groups of fish fed the 50% PH were not different from the control group and showed the similar growth rate with those fed 25% PH and the control diet. Low survival rate as well as low larval performance in the 50% SMA-PH fed group may also have been caused by low diet acceptability as observed visually. In this study, the Asian seabass larvae were successfully weaned onto formulated diets from 17 dph with an acceptable survival rate. Curnow et al. (2006) reported that co-feeding (Gemma Micro) of Asian seabass could be initiated as early as 13 dph, but the survival

rate was rather low (13.5%). Interestingly, the experimental diets in the present study performed better than the reference diets, commercial diet and minced mackerel muscle, which caused 100% mortality in the third week of the experiment. This suggests that the experimental diets were of good quality. Based on visual observations during the feeding trial, the fish showed low acceptance of the commercial diet by spitting out the food after ingestion. This may have been due to a harder texture of the commercial diet compared to the experimental diets. A combination of inappropriate particle size compared to fish mouth size, high nutrient leaching and diet palatability were most likely the reasons for the total mortality in the group fed minced mackerel muscle. The low survival rate and the highest specific growth rate especially in fish fed 25% SMA may be due to the cannibalism that has been observed during the rearing period. Size grading of fish every 5–7 days to reduce size variation and the reduction of larval rearing density should be done to improve the survival rate, because Asian seabass are by nature a highly predatory fish (Kailasam et al., 2002), and a large fish can swallow smaller fish which are less than 70% of its body size (Parazo et al., 1991). The present study showed that the PHs produced using the combination of acidic and alkaline enzyme incorporated in the diets at low level improved growth with acceptable larval survival rate. The PH products induced pancreatic and brush border section of intestine enzyme activity. The chain length of the peptides distributed in the protein hydrolysate influenced the larval performance in that the short chain peptide PHs induced the specific activity at the pancreatic enzyme, chymotrypsin, while the long chain peptide PHs were superior for the enzymatic digestion compared with those containing high portions of short peptides. Our results indicated that early weaning of Asian seabass can start at 17 dph which would help in shortening the Artemia feeding period that consequently will reduce the labor requirement for the daily Artemia preparation. We suggest that FMC at a 25% inclusion level would be the preferable protein hydrolysate to be incorporated in microbound diets for Asian seabass larvae, because of the lower expense and higher growth rate compared to SM hydrolysate. Acknowledgment The authors would like to express their most sincere gratitude and appreciation to the Thailand Research Fund (PHD/0169/2549) for its financial support, the National Institute of Coastal Aquaculture (NICA), Department of Fisheries, Songkhla, Thailand for generously providing the experimental larvae, Pattani Fishmeal Company for providing the premium grade fishmeal, and the Kidchakan Supamattaya Aquatic Animal Health Research Center, Department of Aquatic Science, Faculty

202

M. Srichanun et al. / Aquaculture 428–429 (2014) 195–202

of Natural Resources, Prince of Songkla University, Thailand for allowing the investigators to make use of the research facilities, Finally, we acknowledge the staff at the Department of Biotechnology, Norwegian University of Science and Technology, Trondheim, Norway for their contribution to FPLC analysis. We highly appreciated Dr. Thomas Wilson for doing the final proof reading of the manuscript. References Adler-Nissen, J., 1979. Determination of the degree of hydrolysis of food feed protein hydrolysates by trinitrobenzene sulfonic acid. J. Agric. Food Chem. 27, 1256–1262. AOAC, 1990. Official Methods of Analysis, 15th ed. The Association of Official Analytical Chemists, Washington, D.C. Benjakul, S., Morrissey, M., 1997. Protein hydrolysates from Pacific whiting solid wastes. J. Agric. Food Chem. 45, 3423–3430. Cahu, C.L., Zambonino-Infante, J.L., Quazuguel, P., Le Gall, M.M., 1999. Protein hydrolysate vs. fish meal in compound diets for 10-day old seabass Dicentrarchus labrax larvae. Aquaculture 171, 109–119. Cahu, C., Rønnestad, I., Grangier, V., Zambonino Infante, J.L., 2004. Expression and activities of pancreatic enzymes in developing sea bass larvae (Dicentrarchus labrax) in relation to intact and hydrolyzed dietary protein; involvement of cholecystokinin. Aquaculture 238, 295–308. Carvalho, A.P., Sá, R., Oliva Teles, A., Bergot, P., 2004. Solubility and peptide profile affect the utilization of dietary protein by common carp (Cyprinus carpio) during early larval stages. Aquaculture 234, 319–333. Curnow, J., King, J., Partridge, G., Kolkovski, S., 2006. Effects of two commercial microdiets on growth and survival of barramundi (Lates calcarifer Bloch) larvae within various early weaning protocols. Aquac. Nutr. 12, 247–255. Dabrowski, K., 1992. Ascorbate concentration in fish ontogeny. J. Fish Biol. 40, 273–279. Daukšas, E., Šližyte, R., Rustard, T., Storrø1, I., 2004. Bitterness in fish protein hydrolysates and methods for removal. J. Aquat. Food Prod. Technol. 13, 101–114. Espe, M., Lied, E., 1994. Do Atlantic salmon (Salmo salar) utilize mixtures of free amino acids to the same extent as intact protein sources for muscle protein synthesis? Comp. Biochem. Physiol. 107A, 249–254. Hardy, R.W., 1991. Fish hydrolysate: production and use in aquaculture feeds. In: Akiyama, D.M., Tan, R.K.H. (Eds.), Proceedings of the Aquaculture Feed Processing and Nutrition Workshops. American Soybean, Singapore, pp. 109–115 (September 19–25). Houde, D.E., Schekter, R.C., 1981. Growth rates, rations and cohort consumption of marine fish larvae in relation to prey concentrations. Rapports et Procès-verbaux des Réunions du Conseil International pour l'Exploration de la Mer, 178, pp. 441–453. Ihekoronye, I.A., 1986. Rapid in vitro enzymatic predictive model for the in vivo digestibility of food protein. Int. J. Food Sci. Technol. 21, 81–87. Kailasam, M., Thirunavukkarasu, A.R., Abraham, Mathew, Kishore Chandra, P., Subburaj, R. , 2002. Influence of size variation and feeding on cannibalism of Asian seabass Lates calcarifer (Bloch) during hatchery rearing. Indian J. Fish. 49, 107–113. Khochchawech, J., 2013. Histological changes of digestive tract and development of microparticulate diets for Asian seabass (Lates calcarifer Bloch) larvae. (Master's Thesis in Aquatic Science) Prince of Songkla University, Songkhla, Thailand. Klein, S., Cohn, S.M., Alpers, D.H., 1998. The alimentary tract in nutrition, In: Shile, M.E., Olson, A.J., Shike, M., Ross, A.C. (Eds.), Modern Nutrition in Health and Disease, ninth edition. Williams and Wilkins, Baltimore, MD, pp. 605–630. Kolkovski, S., Tandler, A., 2000. The use of squid protein hydrolysate as a protein source in microdiets for gilthead seabream Sparus aurata larvae. Aquac. Nutr. 6, 11–15. Kotzamanis, Y.P., Gisbert, E., Gatesoupe, F.J., Zambonino Infante, J., Cahu, C., 2007. Effects of different dietary levels of fish protein hydrolysates on growth, digestive enzymes,

gut microbiota, and resistance to Vibrio anguillarum in European seabass (Dicentrarchus labrax) larvae. Comp. Biochem. Physiol. 147A, 205–214. Kvåle, A., Harboe, T., Espe, M., Næss, T., Hamre, K., 2002. Effect of protein hydrolysate on growth and survival of Atlantic halibut larvae (Hippoglossus hippoglossus L.). Aquac. Res. 33, 311–321. Kvåle, A., Harboe, T., Mangor-Jensen, A., Hamre, K., 2009. Effects of protein hydrolysate in weaning diets for Atlantic cod (Gadus morhua L.) and Atlantic halibut (Hippoglossus hippoglossus L.). Aquac. Nutr. 15, 218–227. Lindner, P., Eshel, A., Kolkovski, S., Tandler, A., Harpaz, S., 1995. Proteolysis by juvenile sea bass Dicentrarchus labrax gastrointestinal enzymes as a method for the evaluation of feed proteins. Fish Physiol. Biochem. 14, 399–407. Lowry, O.H., Rosebrough, N.J., Farr, J.L., Randall, R.J., 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265–275. Morr, C.V., 1985. Composition, physicochemical and functional properties of reference whey protein concentrates. J. Food Sci. 50, 1406–1411. Muller, P.Y., Janovjak, H., Miserez, A.R., Dobbie, Z., 2002. Processing of gene expression data generated by quantitative real-time RT-PCR. Biotechniques 32, 1372–1379. Nankervis, L., Southgate, P.C., 2006. An integrated assessment of gross marine protein sources used in formulated microbound diets for barramundi (Lates calcarifer) larvae. Aquaculture 257, 453–464. Parazo, M.M., Avila, E.M., Reyes Jr., D.M., 1991. Size and weight dependent cannibalism in hatchery-bred sea bass (Lates calcarifer Bloch). J. Appl. Ichthyol. 7, 1–7. Perera, E., Moyano, F.J., Rodriguez-Viera, L., Cervantes, A., Martinez-Rodriguez, G., Mancera, F.M., 2010. In vitro digestion of protein sources by crude enzyme extracts of the spiny lobster Panulirus argus (Latreille, 1804) hepatopancreas with different trypsin. Aquaculture 310, 178–185. Rønnestad, I., Conceição, L.E.C., Aragão, C., Dinis, M.T., 2000. Free amino acids are absorbed faster and assimilated more efficiently than protein in postlarval Senegal sole (Solea senegalensis). J. Nutr. 130, 2809–2812. Rønnestad, I., Tonheim, S.K., Fyhn, H.J., Rojas-García, C.R. Kamisaka, Koven, Y.W., Finn, R.N. , Terjesen, B.F., Barr, Y., Conceição, L.E.C., 2003. The supply of amino acids during early feeding stages of marine fish larvae: a review of recent findings. Aquaculture 227, 147–164. Šližytė, R., Daukšas, E., Falch, E., Storrø, I., Rustad, T., 2005. Characteristics of protein fractions generated from cod (Gadus morhua) by-products. Process Biochem. 40, 2021–2033. Sovoie, A., Le François, N.R., Cahu, C., Blier, P.U., Andreassen, I., 2006. Do protein hydrolysates improve survival and growth of newly-hatched spotted wolffish (Anarhichas minor), a non-metamorphic aquaculture fish species? Aquaculture 261, 782–788. Srichanun, M., Tantikitti, C., Utarabhand, P., Kortner, T.M., 2013. Gene expression and activity of digestive enzymes during the larval development of Asian seabass (Lates calcarifer). Comp. Biochem. Physiol. 165B, 1–9. Tang, H.-G., Wu, T.-X., Zhao, Z.-y., Pan, X.-D., 2008. Effects of fish protein hydrolysate on growth performance and humoral immune response in large yellow croaker (Pseudosciaena crocea R.). J. Zhejiang Univ. (Sci.) B9, 684–690. Tibbetts, S.M., Milley, J.E., Ross, N.W., Verreth, J.A.J., Lall, S.P., 2011. In vitro pH-Stat protein hydrolysis of feed ingredients for Atlantic cod, Gadus morhua, 1. Development of the method. Aquaculture 319, 398–406. Tonheim, S.K., Espe, M., Hamre, K., Ronnestad, I., 2005. Pre-hydrolysis improves utilisation of dietary protein in the larval teleost Atlantic halibut (Hippoglossus hippoglossus L.). J. Exp. Mar. Biol. Ecol. 321, 19–34. Tonheim, S.K., Nordgreen, A., Høgøy, I., Hamre, K., Rønnestad, I., 2007. In vitro digestibility of water soluble and water insoluble protein fractions of some commonly used fish larval feeds and feed ingredients. Aquaculture 262, 426–435. Zambonino-Infante, J.L., Cahu, C.L., Péres, A., 1997. Partial substitution of di and tripeptides for native proteins in seabass diet improves Dicentrarchus labrax larval development. J. Nutr. 127, 608–614.