Aquaculture 235 (2004) 553 – 567 www.elsevier.com/locate/aqua-online
In vivo incorporation of [U]-14C-amino acids: an alternative protein labelling procedure for use in examining larval digestive physiology Sigurd K. Tonheim a,*, Marit Espe a, Arnt J. Raae b, Maria J. Darias c, Ivar Rønnestad d a
National Institute of Nutrition and Seafood Research, P.O. 176 Sentrum, N-5804 Bergen, Norway b Department of Molecular Biology, University of Bergen, Bergen, Norway c Instituto de Ciencias Marinas Andalucı´a (CSIC), Cadiz, Spain d Department of Zoology, University of Bergen, Bergen, Norway
Received 22 September 2003; received in revised form 17 December 2003; accepted 17 December 2003
Abstract A radioactive soluble model protein for studies of protein digestion, absorption and amino acid (AA) metabolism in larval fish was successfully produced in Atlantic salmon by oral administration of uniformly [U]-14C-labelled amino acids followed by blood sample withdrawal (48 h postadministration) and purification. The salmon serum protein (14C-SSP) was characterised in terms of the protein composition and specific activity of its amino acids. Most radioactivity was found in the three most abundant serum proteins, which had apparent molecular weights of 65, 75 and 120 kDa, respectively, of which labelling was found in all the amino acid residues of the SSP that were analysed. The digestibility of the 14C-SSP was tested by in vivo tube feeding using early stages of Atlantic halibut and was found to be more efficiently digested and utilised than the 14C-methylated bovine serum albumin (14C-BSA) that has been used in previous studies. This supports the notion that proteins labelled by 14C-methylation are not suitable as model proteins in metabolic studies due to modification of their 14C-methylated lysine residues. Further studies on the 14C-SSP demonstrated a digestibility of 59 F 13% in juvenile halibut, while at the pre-metamorphic stage, it was only 25 F 13%. This supports the hypothesis that there is a significant improvement in the ability to digest and utilise dietary proteins as the digestive system becomes fully developed, including a functional (acid-producing) stomach. D 2004 Elsevier B.V. All rights reserved. Keywords: Serum protein; Protein digestion; Amino acid catabolism; Fish larvae; Atlantic halibut; Atlantic salmon
* Corresponding author. Tel.: +47-55-90-51-33; fax: +47-55-99-52-99. E-mail address:
[email protected] (S.K. Tonheim). 0044-8486/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.aquaculture.2003.12.015
554
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
1. Introduction Amino acids (AA) make up a major part of the metabolic fuel in marine teleosts during the egg and yolk-sac stages (Fyhn, 1989; Rønnestad et al., 1992, 1994, 1999; Rønnestad and Fyhn, 1993a; Finn et al., 1995; Seoka et al., 1997; Sivaloganathan et al., 1998) and also represent a major oxidative substrate in the first weeks after the onset of exogenous feeding. In larval Atlantic halibut, AA have been estimated to account for about 60% of the fuel (Rønnestad and Fyhn, 1993b, Rønnestad and Naas, 1993), while in larval Atlantic cod, they represent 70– 95% (Finn et al., 2002). In addition to catabolic requirements for AA, they are also in high demand as anabolic substrates for the synthesis of proteins and other N-containing compounds. In order to supply sufficient quantities of the dietary AA needed by fast-growing larval tissues, there is thus a need for efficient and rapid processing of ingested proteins in the alimentary canal. Knowledge of the digestive capacity for proteins in the developing and pre-gastric larval digestive tract is important for the development of improved feed and feeding protocols for cultured marine fish larvae. Studies of digestive and absorption efficiency in marine fish larvae at the onset of exogenous feeding are seriously limited by a number of factors, including small larval size (often around 3 mm), small feed-particle sizes (50 – 150 Am), feed production technology and lack of acceptance of artificial diets by fish larvae. In addition, the leaching of water-soluble nutrients from such small diet particles potentially makes the interpretation of results from feeding trials difficult, since the nutrient content of the diet at the moment of ingestion becomes uncertain. In many cases, first-feeding marine fish larvae ingest artificial feed but fail to grow. However, recent advances in formulation and production technology have provided promising results (Lazo et al., 2000; Yufera et al., 2000; Cahu and Zambonino Infante, 2001; Hamre et al., 2001; Kolkovski, 2001; Koven et al., 2001). Nevertheless, in most species, problems arising from using formulated feeds from the onset of exogenous feeding persist. An in vivo method for controlled tube feeding (Rust et al., 1993; Rust, 1995; Rønnestad et al., 2000) has the potential to overcome some of these problems and allow quantification of the digestive capacity and assimilation efficiency of marine fish larvae. A recent refinement of the method (Rønnestad et al., 2001a) is based on distribution of the tracer after tube feeding a single pulse of 14C-labelled amino acids solution in combination with sampling at various times, dissection of the gut and scintillation counting. The few studies performed so far suggest that simple forms of AA (free amino acids; FAA) are more rapidly and efficiently assimilated than protein by marine fish larvae (Rust et al., 1993; Rønnestad et al., 2000; Rojas-Garcı´a and Rønnestad, 2003a). In their studies, Rust et al. (1993) used a 35S-labelled protein from bacteria, but 14C-labelled proteins were subsequently chosen since 14C-labelled nutrients, in combination with a CO2 trap, permit labelled evacuated nutrients to be distinguished from label (14CO2) originating from catabolism of the absorbed nutrients (Rønnestad et al., 2001a). However, as discussed by Rojas-Garcı´a and Rønnestad (2003b) and the [methyl 14C]-methylated bovine serum albumin (BSA) tracer used may have led to underestimation of proteolytic capacity. Commercially available 14C-labelled proteins are typically chemically labelled by methylation based on 14C-formaldehyde. This labelling process adds 2 mol formaldehyde to 1 mol lysine (and to some extent arginine). This alters the size of the amino acid residue and
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
555
also eliminates the native positive charge of the amino acid (French and Edsall, 1945). Methylated BSA is more slowly hydrolysed by calpain, a cytosolic enzyme that specifically catalyse the hydrolysis of peptide bonds adjacent to tyrosine, methionine and arginine residues, compared to native BSA (Murakami and Etlinger, 1987). An even stronger negative effect of methylation would thus be expected on the hydrolytic power of trypsin, one important pancreatic protease in vertebrates, which specifically catalyses the hydrolysis of peptide bonds adjacent to both lysine and arginine residues. Labelled lysine and arginine eventually liberated through digestion are also modified and thereby probably discriminated against as substrates by metabolic processes in general, resulting in nonoptimal metabolism. For this reason, the results of assimilation studies of protein and peptides in Atlantic halibut post-larvae need verification using non-methylated compounds. In uniformly [U]labelled amino acids, the substituted 14C atom does not change the chemical or biological properties of the amino acids. Berge et al. (1994) produced [U]-14C-lysine-labelled fish muscle protein by oral administration of [U]-14C-lysine. A similar procedure was chosen in the present experiment, with the exception that a mixture of [U]-14C-labelled amino acids was used and the water-soluble serum proteins were harvested. The aim of this work was to produce a suitable model protein to be used in future studies of protein digestion and utilisation in fish larvae.
2. Material and methods 2.1. In vivo labelling of salmon serum protein with [U]-14C-amino acids [U]-14C-labelled L-amino acid mixtures were obtained from ICN Biomedicals, Irvine, CA, USA (lot no. 306345) and ARC, St. Louis, MO, USA (lot no. 020428). The batches were identical with respect to total isotope activity (37 MBq) and the relative specific activity of each amino acid present (L-alanine 8%, L-arginine 7%, L-aspartic acid 8%, Lglutamic acid 12.5%, glycine 4%, L-histidine 1.5%, L-isoleucine 5% L-leucine 14%, Llysine 6%, L-phenylalanine 8%, L-proline 5%, L-serine 4%, L-threonine 5%, L-tyrosine 4%, L-valine 8%). Frozen, standard dry commercial salmon feed pellets (Atlantic HP50, 6.0 mm; Skretting, Norway) were crushed to a fine powder in a mortar. About 30 g was mixed with 35 ml water into a moist paste. One milliliter of the feed paste was carefully mixed with the [U]-14C-labelled L-amino acid mixture (75 MBq; 10 ml) in a plastic syringe. The volume of the solution containing the labelled amino acids was 11 ml. It thus greatly increased the volume and fluidity of the initial feed paste. Excess water was removed by freeze-drying for 20 h and the dry diet was stored at 20 jC until use. Before administration to the salmon the labelled feed was re-dissolved in 1 ml of water and transferred to a 1-ml syringe. An Atlantic salmon (Salmo salar, 150 g body weight) reared in freshwater at 10.5 jC was starved for 20 h before being tube-fed the labelled diet. In a trial intended to increase labelling efficiency by compensatory elevation of serum protein synthesis, 0.9 ml of blood was drained from the caudal vein by a syringe immediately before tube feeding.
556
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
This was based on theoretical considerations only, and was not checked analytically. The fish was anaesthetised with benzocaine (60 mg l 1) before blood sampling and tube feeding. 2.2. Preparation of labelled salmon serum protein After 48 h of incubation, the fish was anaesthetised and a maximum blood sample was collected from the caudal vein. The blood was immediately transferred to 1.5-ml Eppendorf tubes and the plasma collected after centrifugation at 1500 g for 5 min to remove blood cells. Thrombin (4.5 units ml 1, Sigma-Aldrich, St. Louis, MO, USA) was added and the plasma was stored on ice for coagulation (2 – 4 h). The resulting fibrin gel was squeezed with a spatula to increase the serum yield during collection. Serum was stored (1 week) at 20 jC until gel filtration. The serum was gel-filtered on a prepacked (G-25 Sephadex) HiPrepk 26/10 desalting column (Amersham Biosciences, Uppsala, Sweden) and eluted with PBS buffer (pH 7.2) at a flow rate of 6 ml min 1. Fractions, each of approximately 0.8 and 3.8 ml, were collected as eluted between 10 and 26 ml and between 26 and 138 ml, respectively, and the exact volume of each fraction measured. All fractions collected were monitored at A = 278 nm and analysed for their 14C-activity by being counted twice for 5– 10 min in a Tri-Carb 2300 TRk (Packard Instrument, Meriden, CT, USA) liquid scintillation counter after thoroughly mixing with Ultima Gold XRk(Packard Bioscience, Groningen, The Netherlands). Serum fractions that were eluted between 11.6 and 18.0 ml were pooled and concentrated by cut-off centrifugation at 7500 g using 30-kDa cut-off centrifuge tubes (Microsep 30Kk Filtron Technology, Northborough, MA, USA). All steps during rinsing and centrifugation were carried out at 3 F 1 jC. The total protein content of the rinsed 14Csalmon serum protein (14C-SSP) was quantified using the DC Protein Assay Kit II from Bio-Rad Laboratories, Hercules, CA, USA. 2.3. Characterisation of the model protein A small sample of 14C-SSP was diluted to 1.0 mg ml 1, and 1.0 ml was passed through a prepacked Superosek 6 column (Amersham Biosciences). The serum proteins were eluted with PBS buffer (pH 7.2) at a flow rate of 0.5 ml min 1. The eluate was continuously monitored at A = 280 nm. Fractions each of 1.0 ml were collected between 6 and 45 ml and measured for 14C-activity. Subsamples of the fractions containing protein were qualitatively analysed for proteins by sodium dodecyl sulfate – polyacrylamide gel electrophoresis (SDS – PAGE). 2.3.1. SDS – PAGE SDS – PAGE was performed as described by Laemmli (1970) using 4 –20% linear gradient gels. Protein bands were visualised by Bio-Safe Coomassiek. Tris – HCl Ready Gelk, premixed-sample buffer, Tris –glycine– SDS running buffer and Coomassie stain were obtained from Bio-Rad Laboratories. To evaluate the protein sizes, pre-stained Precision Proteink Standards (Bio-Rad Laboratories) were utilised.
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
557
In order to determine the partition of radioactivity for the most abundant proteins, each protein band was sampled from the gel by a scalpel, dissolved in 0.2M NaIO4 during 24 h of incubation at 60 jC and its 14C-activity measured as described above. 2.3.2. Hydrolysis and separation of amino acids A small sample (4.17 Al) of 14C-SSP was hydrolysed in 6 N HCl for 24 h at 110 jC in a nitrogen atmosphere. After the HCl had evaporated, the hydrolysate was diluted in 100 Al sample buffer and analysed by automated cation-exchange chromatography using an LKB 4451 Alpha Plus amino acid analyzer with a 4.6 200 mm Ultropac 8 column in a lithium citrate and ninhydrin post-column detection system (LKB Biochrom, Cambridge, UK). Fractions containing the separated amino acids and intervening fractions were collected in accordance with a standardised sampling procedure and the radioactivity of each fraction determined. Tryptophan and cysteine were not analysed and glutamine and aspargine were co-eluted with the glutamic acid and aspartic acid fractions, respectively. 2.4. Biological test of the model protein The digestibility and utilisation of the 14C-SSP were tested by in vivo tube feeding into pre-metamorphosis larvae and post-metamorphosis juvenile Atlantic halibut (Hippoglosssus hippoglossus) at 36 and 78 days post first feeding (dpff), respectively. Tube feeding and subsequent incubation, including sampling of expired CO2 from the fish, were
Fig. 1. Experimental set up for collecting metabolic 14CO2 produced during post-tube feeding incubation. After sampling the larvae and resealing the incubation vial, CO2 was eliminated from the incubation water (1) due to lowered pH ( < 2.0) by injecting HCl through the rubber seal. CO2 was then led to the CO2 trap by the air flow and finally trapped by bubbling through 0.5 M KOH (2). Radioactivity in the larval faeces remained in the incubation water.
558
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
performed as previously described by Rønnestad et al. (2001a). However, a modified incubator system based on 20-ml scintillation vials was used (Fig. 1). CO2 was trapped with 97 F 2% efficiency. Larvae and juveniles were transported in closed 20-l beakers to the laboratory and acclimatised for 1 day before performing the tube-feeding experiment. The larvae and juveniles were kept at a constant temperature, 13 F 1 jC, from collection from the rearing tanks until they were killed at sampling at the end of the incubation period. The juveniles were kept in filtered (0.2 Am) seawater (34 g l 1) while larvae were kept in filtered and diluted seawater (26 g l 1) to reduce mortality. The model protein was deposited into the mid-gut of the larvae at 36 dpff and into the stomach of the juveniles at 78 dpff after being slightly anaesthetised by a short immersion (2– 3 min) in MS-222 (0.1g l 1, Sigma, St. Louis, MO, USA). Larvae 36 dpff (60 F 12 mg wet weight) and juveniles 78 dpff (250 F 52 mg wet weight) were tube-fed 11.3 Ag (140 nl) and 45.4 Ag (560 nl) of the model protein solutions, respectively, thus maintaining constant ratio between deposited amount and average body weight. The larvae and juveniles were sampled and killed by cutting the spinal cord 16 h after tube feeding. The gut, including the liver, was immediately dissected and the body and gut were extracted separately twice for 24 h in 6% trichloroacetic acid (TCA) before being solubilised in 2.0 and 1.0 ml of Soluene 350k, respectively. The incubation water and contents of the CO2 traps were also sampled, and finally, 14C-activity in all fractions was analysed by LSC as described above. At 78 dpff, the digestion and utilisation of 14C-SSP by the Atlantic halibut juveniles were compared to the utilisation of commercial 14C-BSA obtained from ARC (lot no. 020111).
Fig. 2. Protein content (A278) and radioactivity (Bq) in fractions from gel-filtered (G-25 Sephadex) Atlantic salmon serum after preparative labelling by in vivo incorporation of 14C-labelled amino acids. Fractions eluted between 11.6 and 18 ml (indicated by the grey box) were pooled and concentrated.
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
559
2.5. Statistics Homogeneity of the variance and differences between groups were tested by Levene’s test and Student’s t test, respectively, performed by Stat Soft Statisticak version 6.0. The significant level was set at 5%. All data concerning larval and juvenile Atlantic halibut are presented as means (n = 6) F S.D.
3. Results 3.1. In vivo labelling and preparation of
14
C-SSP
On the second day after administration of the mixture of radioactive amino acids, 2.5 ml of blood was sampled. The plasma yield was high, 1.9 ml, and reflected the low hematocrit value, 23%, of the blood sample. The blood drained from the fish 2 days earlier, however, had a hematocrit value of 35%, which was considered to be normal (Conroy, 1972). The serum (1.6 ml) was gel-filtered on a desalting column and the fractions eluted between 12.4 and 18 ml were selected and pooled. The pooled fractions contained 81% of
Fig. 3. Proteins constituting the salmon serum protein concentrate (14C-SSP), as revealed by gel filtration (Superosek 6) and SDS – PAGE analysis. Only fractions containing proteins (5 – 14) were analysed. Surrounded area designated A, B and C indicates bands analysed for radioactivity by LSC. Radioactivity was 5, 9 and 3 Bq, respectively. Content of protein is shown by on-line UV absorption (280 nm) and total content of radioactivity of the identical fractions 5 – 14 are shown below.
560
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
the radioactivity in the serum (Fig. 2), while the concentrated 14C-SSP contained 65% of the radioactivity in the serum. The filtrate contained 5.3%; some radioactivity had thus been absorbed by the filter. The final protein concentration and 14C-activity of the 14C-SSP were 81 F 1 mg ml 1 and 247 F 13 kBq ml 1, respectively, equivalent to a specific activity of 3.6 MBq g 1 and a labelling efficiency of 7.4 Bq (g serum protein (g BW fish (wet weight)) 1). 3.2. Protein characterisation A sample of the 14C-SSP was analysed by native (gel-filtration) and reducing (SDS – PAGE) conditions. Most of the salmon serum protein was eluted in fractions 9– 12, as shown by the elution profile (A = 280 nm) and the SDS – PAGE (Fig. 2). Although 13 different polypeptides ranging in size from approximately 15 kDa to above 250 kDa were revealed in these fractions, three polypeptides, designated A, B and C (Fig. 3), were dominant. Their apparent molecular weights were 65, 75 and 120 kDa, respectively. As 85% of the eluted radioactivity was in the same fractions, these three polypeptides in sum carried most of the radioactive 14C-labelled amino acids of the 14C-SSP. The summed levels of radioactivity of the bands A, B and C that were dissected from the gel, as indicated in Fig. 3, were 5, 9 and 3 Bq, respectively. A small amount of protein eluted in the void volume was collected in fractions 5 and 6 (Fig. 3). Denaturing SDS – PAGE revealed two polypeptides with apparent molecular weights of 25 and 75 kDa in these fractions. These polypeptides could be largely ignored as carriers of 14C-labelled amino acids, as only 2.5% of the eluted radioactivity was present in these fractions. This indicated that the distribution of 14CTable 1 Amino acid profile and amino acid specific radioactivity of the salmon serum protein concentrate (14C-SSP) after hydrolysis in 6N HCl and subsequent separation by HPLC Amino acid
Amounts (nmol)
Radioactivity (Bq)
Specific activity (Bq Amol 1)
Asp + Asn Ser Gly Pro Glu + Gln Ala Tyr Thr Val Met Ile Leu Phe Lys His Arg
47 9 45 32 38 67 19 24 37 14 19 43 16 41 12 20
5.3 1.1 5.6 7.6 3.5 5.2 55 5.2 7.8 2.3 5.1 8.7 36 12 1.8 10
113 122 124 238 92 78 2895 217 211 164 268 202 2250 293 150 500
Values are per microliter of
14
C-SSP. Cys and Trp were not analysed.
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
561
labelled amino acids among the various proteins in the salmon serum was not homogeneous. Radioactivity was detected in all amino acids present in the salmon serum protein preparation (Table 1). The specific activity varied among the amino acids. Most striking were the high specific activities of the aromatic amino acids tyrosine and phenylalanine, with 2895 and 2250 Bq Amol 1, respectively. This was in contrast to the specific activity of the other amino acids, which varied from 78 to 500 Bq Amol 1 (Table 1). The amino acids quantified constituted 64% of the hydrolysed protein used in the HPLC analysis, while the summed radioactivity of the eluted amino acids accounted for 70% of the radioactivity utilised. 3.3. Utilisation of
14
C-SSP
The 14C-SSP protein was digested, absorbed and utilised by both larval and juvenile Atlantic halibut. However, there were considerable differences between the pre-metamorphosed larvae and the post-metamorphosed juveniles with respect to both the absorption and utilisation of the labelled amino acids (Fig. 4). Significantly ( P < 0.05) smaller
Fig. 4. Compartmental distribution of radioactivity given as percent of administered radioactivity in post and pre metamorphosed Atlantic halibut at 36 and 78 days post first feeding (dpff), respectively, 16 h after tube feeding with 14C-SSP or 14C-BSA. Values are given as mean (n = 6) F S.D.
562
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
fractions of the labelled amino acids were absorbed (25 F 13% versus 59 F 13%), assimilated into body protein (10 F 5% versus 29 F 8%) or catabolised (12 F 7% versus 23 F 8%) by the larvae than by the juveniles, and correspondingly higher fractions were lost through the faeces (72 F 15% versus 22 F 19%). The relative distribution of the absorbed amino acids (amino acid in the free pool and assimilated and catabolised amino acids), however, was almost identical. Low radioactivity in the larval gut (4 F 2%) showed that the digestive and absorptive processes had taken place at the time of sampling. The juveniles, on the other hand, still had 18 F 7% of the administered radioactivity left in the gut 16 h post tube feeding. Digestion and utilisation of the 14C-BSA by the juveniles differed from the 14C-SSP. The assimilation of 14C-SSP into body TCA precipitate was more than twice as high as 14 C-BSA ( P < 0.05; 29 F 8% versus 12 F 3%) and the faecal loss was less than half ( P < 0.05; 22 F 19% versus 48 F 10%). The catabolism of the radioactive amino acids was not significantly different ( P>0.05). Variable amounts of radioactivity in the washing water revealed that vomiting of the administered protein solutions by larvae and juveniles occurred to varying extent, inducing variation between individuals in the amount of protein retained in the gut when installed in the incubators. The average amount of retained protein at the start of the incubation period, and the variation, was however similar for all larval groups. Retained doses (mean F S.D.) were 0.15 F 0.04 Ag protein (mg fish (wet body weight)) 1 for the larvae and juveniles administered 14C-SSP, and 0.16 F 0.04 Ag protein (mg fish (wet body weight)) 1 for the juveniles administered the 14C-BSA.
4. Discussion A mixture of [U]-14C-labelled amino acids orally administered to the Atlantic salmon was successfully incorporated into serum protein during a period of 48 h of incubation. Most of the incorporated radioactive amino acids were found in the three most abundant polypeptides of the salmon serum protein. The molecular weight of salmon serum albumin is about 65 kDa (ExPASy, Expert Protein Analysis System, proteomics server of the Swiss Institute of Bioinformatics). This corresponds well to band A in Fig. 3, which was the second most abundant protein of the present protein preparation in terms of share of radioactive-labelled amino acids. The other highly abundant protein in the same fractions represented by band B had an apparent molecular mass of 75 kDa. Another relatively abundant polypeptide appearing in fractions 5 and 6, however, was not quantitatively important, as it contained only an insignificant fraction of the radioactive amino acids (2.5%). The inhomogeneous distribution of radioactive amino acids between different serum proteins was not surprising, as variations in turnover between the different serum proteins are to be expected. More surprising was the fact that the relatively small polypeptide (about 75 kDa) and an even smaller polypeptide (about 25 kDa) appeared in the void volume collected in fractions 5 and 6 (Fig. 3), as these fractions had been assumed to consist exclusively of molecules larger than 200 kDa. This indicated that these proteins persist in complex structures in their native forms, and are thus eluted in
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
563
the void volume due to their large molecular size. In preparation for SDS – PAGE, however, such quaternary structures are decomposed into their constituting subunits by boiling in h-mercaptoethanol and SDS. All the amino acids constituents of serum proteins were labelled, including methionine, which was not present in the mixture of radioactive amino acids that were administered to the salmon. This was unexpected, since methionine is an indispensable amino acid for most species including fish. Radioactive methyl groups, on the other hand, could be available in the salmon body as a result of catabolic breakdown of the radioactive amino acids offered. Methionine is an important methyl donor and homocysteine, the product of demethylation of methionine, is a potent methyl acceptor. The radioactivity in methionine may therefore be explained by the reversible exchange of methyl groups between methionine and homocysteine (Fruton and Simmonds, 1953). Tyrosine and phenylalanine, which are linked in that phenylalanine is an indispensable amino acid and tyrosine can only be made from phenylalanine, had a substantially higher specific radioactivity than the other amino acids present in the serum proteins. This indicated that these particular amino acids were available at correspondingly higher specific radioactivity levels during incubation. A possible explanation was that the fish were low in these amino acids during the experiment. However, this was not verified as the salmon’s free amino acid pool was not analysed. In any case, this non-uniform distribution of radioactivity between the different amino acids will have to be considered in the evaluation of studies based on the 14C-SSP as a model protein. The choice of Atlantic salmon for in vivo incorporation of radioactive amino acids reflects its good availability in laboratories rather than any expected biological advantages. However, for a standardised production of labelled serum proteins, species and stages performing variable blood physiology should be avoided, including vitellogenesis (Waagboe and Sandnes, 1988; Waagbø et al., 1989), infections (Møyner, 1993) and smoltification of salmonids (Melingen et al., 1995). Despite the low labelling efficiency, the specific radioactivity of the purified 14C-SSP protein was sufficient for the tube-feeding experiments with halibut larvae. Administration of 140 nl to the larvae containing 41 Bq (2460 dpm), was sufficient for precise fractionation of radioactivity in the different larval compartments by LSC. Experiments on smaller larvae that only allow injection of much smaller volumes due to their limited gut volume can be performed by further concentrating the solutions of 14C-SSP. The tube-feeding experiments on juvenile Atlantic halibut confirmed that 14C-SSP was digested and well utilised in fish. The poor utilisation of the 14C-SSP at 36 dpff was therefore put down to poor larval capability to digest dietary protein in general, and was supported by the large faecal loss from the larvae. Such a difference may be explicable by the lack of a functional stomach in the larvae. For this reason, the protein solution was deposited into the mid-gut of the larvae. The juveniles on the other hand had developed a functional stomach and pyloric sphincter (Luizi et al., 1999), and it was not possible to deposit the protein solutions into their mid-gut due to the curve of the gut at this stage. However, the difference in depositing location reflects a real difference between larvae and juveniles in how they handle swallowed feed. The early larval stomach is just an elongation of the oesophagus without any storage capacity (Luizi et al., 1999). Thus, swallowed feed bypasses the rudimentary larval stomach, and digestion
564
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
relies solely on the enzymes associated with the pancreas and gut. The juveniles on the other hand are able to retain and acidify swallowed feed in their stomach (own observations). Physiological effects of anaesthetisation with MS-222 upon fish, as well as handling itself, have previously been described by Houston et al. (1971), and the study suggested that MS-222 depresses central autonomic functions. Anaesthetisation with MS-222 has therefore, potentially, an impact on both the release of digestive enzymes and on absorption and post absorptive metabolism of absorbed nutrients. However, Di Benedetto and Farmanfarmaian (1975) have demonstrated that intestinal absorption of glucose and galactose in rainbow trout analysed at anaesthetic stage 4 (based on Massee et al., 1995), with MS-222 was not significantly different from rates analysed in fish that had recovered from anesthesia with MS-222. And previous studies have shown that the absorption of amino acids is nearly 100% efficient in larvae and juveniles following anaesthetisation procedures identical to those used in the present study (Conceicßa˜o et al., 2002; Rønnestad et al., 2001b; Applebaum and Rønnestad, 2004). Previous tests failed to reveal any effect of the present anaesthetisation procedure or the gentle handling itself on the oxygen consumption rates (Rønnestad et al., 2001a). Thus, since the fish in the present experiment received only a mild immersion anaesthetisation (to stages 2– 3, according to Massee et al., 1995), lasting only 2 – 3 min (followed by transfer to clean water and recovery, until the larvae were sampled 16 h later), the anaesthetisation procedure in the present experiment was assumed not to have an important impact on digestion, the intestinal absorption or the post absorptive metabolism, and comparisons between groups were assumed safe since identical procedures were followed for all larvae and juveniles. The differences in the pattern of digestion and utilisation between the 14C-BSA and 14 C-SSP supports the need for an alternative to the 14C-methylated model protein, even though the differences in relative distribution of the labelled amino acids were less than what might have been expected. In comparison with the 14C-SSP, significantly ( P < 0.05) higher fractions of the 14C-BSA were lost in the faeces. This was most probably due to reduced digestibility and thus to the methylation of the BSA lysine residues (French and Edsall, 1945). The high catabolic rate of the radioactive methylated lysine from the 14CBSA, and the correspondingly low degree of assimilation into body protein, was in contrast to the efficient assimilation of non-methylated lysine in herring larvae (Conceicßa˜o et al., 2002) and post-larval Senegalese sole (Rønnestad et al., 2001b). This supports the idea that the methylation of lysine restricts its ability to serve as a substrate in protein synthesis, as non-methylated lysine would be expected to be more efficiently assimilated than the mixture of radioactive dispensable and indispensable amino acids in the 14C-SSP. It may be surprising that methylated lysine is accepted by the protein synthesis process at all. However, the transfer of 14C from 14C-BSA to body TCA precipitate, which includes the protein, does not necessarily prove that this actually occurs. Labelled 14C-methyl groups are likely to occur in the organism after metabolic degradation of di-14Cmethylated lysine. These 14C-methyl groups may be shuttled into amino acid biosynthesis and the formation of methionine by methylation of homocysteine (Fruton and Simmonds, 1953). Hence, the radioactivity in body protein may be in amino acid residues other than that of lysine.
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
565
5. Conclusions Salmon serum protein was successfully labelled in vivo after oral administration of [U]-14C-labelled amino acids. Most of the labelled amino acids were incorporated into three proteins with apparent molecular weights of 65, 75 and 110 kDa. The 14C-SSP was digested and utilised in a different way, and more efficiently, than 14C-methylated BSA. Post-metamorphic juvenile Atlantic halibut were better able to digest dietary proteins than pre-metamorphic larvae.
Acknowledgements We would like to thank Harald B. Jensen of the Department of Molecular Biology, University of Bergen, Norway, for useful discussions concerning protein and amino acid chemistry. We would also like to thank Pa˚l Falkenberg at the Norwegian College of Fishery Science, University of Tromsø, Norway, and Dmitri Svistounov at the Department of Experimental Pathology, University of Tromsø, for performing amino acid analysis. The Institute of Marine Research, represented by Tom Hansen and Ingjerd Opstad, is also sincerely thanked for supplying fish and tank facilities. This work was supported by grant no. 130195/130 from the Research Council of Norway. Marı´a J. Darias was supported by a short-stay grant of the Ministerio de Ciencia y Tecnologı´a, Spain.
References Applebaum, S.L., Rønnestad, I., 2004. Absorption, assimilation and catabolism of individual free amino acids by larval Atlantic halibut (Hippoglossus hippoglossus). Aquaculture 230, 313 – 322. Berge, G.E., Lied, E., Espe, M., 1994. Absorption and incorporation of dietary free and protein bound (U14C)lysine in Atlantic cod (Gadus morhua). Comp. Biochem. Physiol. 109A, 681 – 688. Cahu, C.L., Zambonino Infante, J.L., 2001. Substitution of live food by formulated diets in marine fish larvae. Aquaculture 200, 161 – 180. Conceicß a˜o, L.E.C., Rønnestad, I., Tonheim, S.K., 2002. Metabolic budgets for lysine and glutamate in unfed herring (Clupea harengus). Aquaculture 206, 305 – 312. Conroy, D.A., 1972. Studies on the haematology of the Atlantic salmon (Salmon salar L.). Symp. Zool. Soc. Lond. 30, 101 – 127. Di Benedetto, F.E., Farmanfarmaian, A., 1975. Intestinal absorption of naturally occurring sugars in the rainbow trout, Salmo gairdneri, under free-swimming conditions. Comp. Biochem. Physiol. 50A, 555 – 559. Finn, R.N., Rønnestad, I., Fyhn, H.J., 1995. Respiration, nitrogen and energy metabolism of developing yolk-sac larvae of Atlantic halibut (Hippoglossus hippoglossus L.). Comp. Biochem. Physiol. 111A, 647 – 671. Finn, R.N., Rønnestad, I., van der Meeren, T., Fyhn, H.J., 2002. Fuel and metabolic scaling during the early life stages of Atlantic cod Gadus morhua. Mar. Ecol., Prog. Ser. 243, 217 – 234. French, D., Edsall, J.T., 1945. The reactions of formaldehydes with amino acids and proteins. Adv. Protein Chem. 2, 277 – 335. Fruton, J.S., Simmonds, S., 1953. General Biochemistry. Wiley, New York. 1077pp. Fyhn, H.J., 1989. First feeding of marine fish larvae: are free amino acids the source of energy? Aquaculture 80, 111 – 120. Hamre, K., Næss, T., Espe, M., Holm, J.C., Lie, Ø., 2001. A formulated diet for Atlantic halibut (Hippoglossus hippoglossus L.) larvae. Aquac. Nutr. 7, 123 – 132.
566
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
Houston, A.H., Madden, J.A., Woods, J.R., Miles, H.M., 1971. Some physiological effects of handling and tricaine methane-sulphonate anesthetization upon the brook trout, Salvelinus fontinalis. J. Fish. Res. Board Can. 28, 625 – 633. Kolkovski, S., 2001. Digestive enzymes in fish larvae and juveniles—implications and applications to formulated diets. Aquaculture 200, 181 – 200. Koven, W., Kolkovski, S., Hadas, E., Gamsiz, K.T.A., 2001. Advances in the development of microdiets for gilthead seabream, Sparus aurata: a review. Aquaculture 194, 107 – 121. Laemmli, U.K., 1970. Cleavage of structural proteins during assembly of head of bacteriophage-T4. Nature 227, 680 – 685. Lazo, J.P., Dinis, M.T., Holt, J.G., Faulk, C., Arnold, C.R., 2000. Co-feeding microparticulate diets with algae: towards eliminating the need of zooplankton at first feeding red drum (Sciaenops ocellatus). Aquaculture 188, 339 – 351. Luizi, F.S., Gara, B., Shields, R., Bromage, N.R., 1999. Further description of the development of the digestive organs in Atlantic halibut (Hippoglossus hippoglossus) larvae, with notes on differential absorption of copepod and Artemia prey. Aquaculture 176, 101 – 116. Massee, K.C., Rust, B.M., Hardy, R.W., Stickney, R.R., 1995. The effectiveness of tricaine, quinaldine sulfate and metomidate as anesthetics for larval fish. Aquaculture 134, 351 – 359. Melingen, G.O., Stefansson, S.O., Berg, A., Wergeland, H.I., 1995. Changes in serum protein and IgM concentration during smolting and early post-smolt period in vaccinated and unvaccinated Atlantic salmon (Salmo salar L.). Fish Shellfish Immunol. 5, 211 – 221. Møyner, K., 1993. Changes in serum protein composition occur in Atlantic salmon, Salmo salar L. during Aeromonas salmonicida infection. J. Fish. Dis. 16, 601 – 604. Murakami, K., Etlinger, J.D., 1987. Degradation of proteins with blocked amino groups by cytoplasmic proteases. Biochem. Biophys. Res. Commun. 146 (3), 1249 – 1255. Rojas-Garcı´a, C.R., Rønnestad, I., 2003a. Assimilation of dietary free amino acids, peptides and protein in postlarval Atlantic halibut (Hippoglossus hippoglossus). Mar. Biol. 142, 801 – 808. Rojas-Garcı´a, C.R., Rønnestad, I., 2003b. Comment on the article ‘‘Assimilation of dietary free amino acids, peptides and protein in post-larval Atlantic halibut (Hippoglossus hippoglossus)’’. Mar. Biol. 142, 799. Rønnestad, I., Fyhn, H.J., 1993a. Metabolic aspects of free amino acids in developing marine fish eggs and larvae. Rev. Fish. Sci. 1, 239 – 259. Rønnestad, I., Fyhn, H.J., 1993b. Importance of free amino acids in embryonic energy production of three marine flatfishes as revealed by measurements of oxygen consumption and ammonia production. In: Walther, B.T., Fyhn, H.J. (Eds.), Fish Larval Physiology and Biochemistry. University of Bergen, Bergen, pp. 284 – 289. Rønnestad, I., Naas, K.E., 1993. Oxygen consumption and ammonia excretion in larval Atlantic halibut (Hippooglossus hippoglossus L.) at first feeding: a first step towards an energetic model. In: Walther, B.T., Fyhn, H.J. (Eds.), Fish Larval Physiology and Biochemistry. University of Bergen, Bergen, pp. 279 – 284. Rønnestad, I., Finn, R.N., Groot, E.P., Fyhn, H.J., 1992. Utilization of free amino acids related to energy metabolism of developing eggs and larvae of lemon sole Microstomus kitt reared in the laboratory. Mar. Ecol., Prog. Ser. 88, 195 – 205. Rønnestad, I., Koven, W.M., Tandler, A., Harel, M., Fyhn, H.J., 1994. Energy metabolism during development of eggs and larvae of gilthead sea bream (Sparus aurata). Mar. Biol. 120, 187 – 196. Rønnestad, I., Thorsen, A., Finn, R.N., 1999. Fish larval nutrition: a review of recent advances in the roles of amino acids. Aquaculture 177, 201 – 216. Rønnestad, I., Conceicßa˜o, L.E.C., Araga˜o, 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., Rojas-Garcı´a, C.R., Tonheim, S.K., Conceicß a˜o, L.E.C., 2001a. In vivo studies of digestion and nutrient assimilation in marine fish larvae. Aquaculture 201, 161 – 175. Rønnestad, I., Conceicßa˜o, L.E.C., Araga˜o, C., Dinis, M.T., 2001b. Assimilation and catabolism of dispensable and indispensable free amino acids in post-larval Senegal sole (Solea senegalensis). Comp. Biochem. Physiol., C, Comp. Pharmacol. 130, 461 – 466.
S.K. Tonheim et al. / Aquaculture 235 (2004) 553–567
567
Rust, M.B., 1995. Qantitative aspects of nutrient assimilation in six species of fish larvae. PhD thesis. School of Fisheries: University of Washington, Seattle, USA. Rust, M.B., Hardy, R.W., Stickney, R.R., 1993. A new method for force-feeding larval fish. Aquaculture 116, 341 – 352. Seoka, M., Takii, K., Takaoka, O., Nakamura, M., Kumai, H., 1997. Biochemical phases in embryonic red sea bream development. Fish. Sci. 63, 122 – 127. Sivaloganathan, B., Walford, J., Ip, Y.K., Lam, T.J., 1998. Free amino acids and energy metabolism in eggs and larvae of seabass, Lates calcarifer. Mar. Biol. 131, 695 – 702. Waagbø, R., Thorsen, T., Sandnes, K., 1989. Role of dietary ascorbic acid in vitellogenesis in rainbow trout (Salmo gairdneri). Aquaculture 80, 301 – 314. Waagboe, R., Sandnes, K., 1988. Determination of vitellogenin in serum of rainbow trout (Salmo gairdneri) by high-performance gel permeation chromatography. J. Chromatogr., Biomed. 427, 138 – 143. Yufera, M., Fernandez-Diaz, C., Pascual, E., Sarasquete, M.C., Moyano, F.J., Diaz, M., Alarcon, F.J., GarciaGallego, M., Parra, G., 2000. Towards an inert diet for first-feeding gilthead seabream Sparus aurata L. larvae. Aquac. Nutr. 6, 143 – 152.
