Fish Physiology and Biochemistry vol. 14 no. 6 pp 431-437 (1995) Kugler Publications, Amsterdam/New York
Maturation of the pancreatic and intestinal digestive functions in sea bass (Dicentrarchus labrax): effect of weaning with different protein sources C.L. Cahu and J.L. Zambonino Infante Unite Mixte de Nutrition des Poissons IFREMER-INRA, IFREMER Centre de Brest, B.P. 70, 29280 Plouzan, France Accepted: June 23, 1995 Keywords: pancreatic enzymes, intestinal brush border enzymes, weaning, maturation, protein, marine fish larvae
Abstract The maturation of the digestive functions in sea bass (Dicentrarchuslabrax) larvae was evaluated by the enzymatic profile of pancreas and intestine brush border membranes. Sea bass larvae were weaned at day 25 with three simplified diets different by their protein nature: 100°70 fish meal (FP), 100% casein mixture (CP) and 50% fish meal-50% casein mixture (CFP). The casein mixture contained 35°70 of hydrolysate. The control group was fed live preys. The specific activity of amylase decreased with age irrespectively of the diets whereas the specific activity of trypsin was enhanced. The casein mixture reduced pancreatic secretion in amylase and trypsin. The CFP group differed from the other groups fed on compound diets, exhibiting as soon as day 32 high activities of brush border enzymes, similar to controls. This sharp increase between day 25 and 32 appeared to be crucial for larval survival. The addition of a protein hydrolysate in a weaning diet seems to facilitate this maturation process.
Introduction
Compound diet substitution for live prey is still crucial in marine fish larvae rearing; this phase, known as weaning, can only be performed after some weeks of life in marine species, whereas it can be done as early as mouth opening in fresh water species. Until now, the protein fraction of weaning diets has been composed of different sources such as fish meal, krill meal, squid meal, mussel, egg, beef liver, milk, casein, wheat, soybean, yeast (Teshima et al. 1982; Charlon and Bergot 1984; Dabrowski 1984; Person Le Ruyet et al. 1993; Zambonino Infante and Cahu 1994; Watanabe and Kiron 1994). The use of mixtures containing a great number of protein sources reveals a certain lack of knowledge of the specific requirements of the young stages. This empirical approach was mainly
based on the results of body composition, growth and survival of larvae (Kanazawa et al. 1989; Fyhn 1989; Walford and Lam 1991), the main objective being the reduction of the growth delay observed in larvae fed on compound diet compared to larvae fed on live prey. These diets have been generally formulated from scanty information concerning the specificities of the larvae, and especially concerning the degree of maturation of their digestive tract. Effectively, the digestive processes appear sequentially during the first thirty days of larval life (Dabrowski 1986; Ferraris et al. 1987; Beccaria et al. 1991; Walford and Lam 1993). We have recently reported that, as in mammals, the secretory function of the exocrine pancreas takes place chronologically after the synthesis function; in the same way, the digestive features of the enterocytes evolve by enhancing the
432 Table 1. Composition of the experimental diets (g 100g dry matter - ') LP a
AGE (days)
Fig. 1. Growth of the four experimental groups of larvae ( O LP, FP, X CFP, CP). Data are given as means + SEM (n = 30).
hydrolytic functions of the brush border membrane. Weaning using a conventional compound diet delays or stops these maturation processes (Cahu and Zambonino Infante 1994, 1995). The use of a simplified diet containing well identified components and a restricted number of protein sources is necessary to understand the nutrientdigestive tract interaction. The aim of this work was to study the effect of the nature of the protein supply on the development of larvae, and particularly their digestive capacity, by simplifying the formulation of a weaning diet.
Materials and methods Animals and diets Eggs of European sea bass (Dicentrarchuslabrax) were obtained from the Ifremer-Station of Palavas. The larvae were reared for 48 days at the IfremerStation de Brest. Newly hatched larvae were transferred from incubators to 20 conical fiber glass tanks (35 1)with black walls (initial stocking density: 80 larvae 1-1). They were supplied with running sea water which had been filtered through a sand filter, then passed successively through a tungsten heater and a degassing column packed with plastic rings. Temperature range was 18-19°C, salinity 34.5 ppt. The oxygen level was maintained above 6 ppm by setting the water exchange up to 30% per hour (flow rate: 0.18 1 min-1). The light
Fish meal' Casein mixture Casein 2 Casein salt 3 Hydrolysate 4 DL methionine L tryptophan Maltose Starch Cellulose Cod liver oil Lecithin Mineral mixture Vitamin mixture Ascorbic acid Choline Inositol N x 6.25 Lipids Ash Energy J 100 g-1
FP
CFP
74
37
35% 28%7 35% 1.5% 0.5%
0
57 10 11.0 1701
0
28.5
57
4.9 8 3 3.7 0.9 2 2.2 0.3 0.5 0.5
4.9 8 6.3 7.8 2 2 2.2 0.3 0.5 0.5
4.9 8 9.6 12 3 2 2.2 0.3 0.5 0.5
57 15 11.7
57 15 7.8
57 15 4.3
1735
'Norseamink, 77% protein, 14% lipid; C8654; 4 Sigma N4642.
CP
2
1735
1735
Merck 2242;
3
Sigma
intensity was 9 Wm - 2 maximum at the surface. The larvae were allocated into 4 groups (5 tanks per group). In the control group (LP group), larvae were fed live prey until the end of the experiment (day 48), following the sequence described in a previous paper (Zambonino Infante and Cahu 1994). Larvae of the 3 other groups were weaned at day 25 with compound diets containing 57% protein, exclusively fish meal or a casein mixture for FP and CP groups respectively; for CFP group, half the nitrogen supply was made up of fish meal, the other of casein mixture (Table 1). The amino acid composition of the 3 diets is given in Table 2. Microparticles of size 200-400 gAm were used. Fish were continuously fed in large excess 18h per day using a belt feeder. Food ingestion was controlled by observing the larvae digestive tract under a binocular microscope.
433 Table 2. Amino acid composition of the diets (% dry matter) LIP
FP
CFP
CP
3.7 8.2 1.8 2.2 3.4 1.7 2.1 0.2 2.6 0.7 2.1 2.8 1.7 2.1 3.5 1.1 1.9
3.4 10.4 1.8 2.5 4.9 0.9 1.5 0.3 3.0 0.9 2.5 4.0 2.0 2.4 3.6 1.2 1.6
Table 3. Specific activities of trypsin (mU mg protein-') in pancreatic segment* LP
Aspartic acid Glutamic acid Threonine Serine Proline Glycine Alanine Cysteine Valine Methionine Isoleucine Leucine Tyrosine Phenylalanine Lysine Histidine Arginine
4.2 5.5 2.1 2.4 2.7 2.2 2.5 0.7 2.4 1.2 2.3 3.1 2.2 2.2 3.5 1.4 3.0
4.4 6.2 1.9 1.8 1.8 2.9 3.0 0.2 2.4 0.8 2.2 3.4 1.5 2.0 3.6 1.0 2.5
Sampling and dissection In order to monitor growth, 30 larvae were taken from each group every week. They were preserved for one month in 4% formaldehyde sea water prior to weighing (Lockwood 1973). Larval survival rates were determined by counting individually at the end of the experiment. Larvae were collected at day 25, 32, 39 and 48 from each tank before morning food distribution. They were immediately stored at -80 0 C pending dissection and assays. Dissections under microscope were conducted on a glass maintained at 0°C. Animals were cut into 4 parts: head, pancreatic segment (PS), intestinal segment (IS) and tail, in order to assay enzymes only on pancreatic and intestinal segments as described previously (Cahu and Zambonino Infante 1994).
FP
CFP
CP
Day 25
b ab
9+5.0C 33 2.3b
39+5.7 b
*Means SEM with a same superscript letter in a row are not significantly different (p > 0.05).
30 volumes (v/w) of mannitol 50 mM-Tris, 2 mM, pH = 7. Alkaline phosphatase (AP), leucine aminopeptidase N (AN), y-glutamyl transpeptidase (yGT) and maltase were assayed according to Bessey et al. (1946), Maroux et al. (1973), Meister et al. (1981), and Dahlqvist (1970), respectively. Enzyme specific activities were expressed as moles of substrate hydrolysed per minute per mg protein (i.e., U mg prot- 1) at 370 C for AP, AN, yGT and maltase, and at 25°C for trypsin. Amylase activity was expressed as the equivalent enzyme activity which was required to hydrolyse one mg of starch in 30 min at 37C. Purified brush border membranes (BBM) from the homogenate of intestinal segment were obtained according to a method designed for intestinal scraping (Crane et al. 1979). Protein was determined by the Bradford procedure (1976). The amino acid composition of the diets was analysed using a Waters HPLC system. Hydrolysis, derivatization and analysis were made according to the Pico-Tag method (Bidlingmeyer et al. 1984). Statistical analysis Results are given as means SEM (n = 5). Data were compared by one way analysis of variance followed by LSD multiple range test when significant differences were found at a 0.05 level.
Analytical methods
Results Results
Pancreatic segments were homogenized in 5 volumes (v/w) of ice-cold distilled water. Trypsin and amylase activities were assayed according to Holm et al. (1988) and Metais and Bieth (1968), respectively. Intestinal segments were homogenized in
Effect of the diets on larval growth and survival Larvae fed on compound diets had a significantly lower growth than control animals before day 39 (p < 0.05). From this date, the FP group made up
434 Table 4. Specific activities of trypsin (mU mg protein-') in intestinal segment* LP
FP
CFP
20 ± 2.6b
14 + 3.2c
20 ± 2.2ab
16+ 3.lb
27+ 1.5C
28+3.6C
Day 25
2.1+0.75 3.3 0.39a 1.8 0.26a 1.1+0.22a
2.4 0.22b 0.9 +0.36b 0.6 +0.20b
1.8 0.49b 1.0+0.59b 0.5 0.16b
*Means SEM with a same superscript letter in a row are not significantly different (p > 0.05).
*Means SEM with a same superscript letter in a row are not significantly different (p > 0.05).
Table 5. Specific activities of amylase (U mg protein- 1) in pancreatic segment*
Table 7. Specific activities of leucine aminopeptidase (mU mg protein-') in brush border membranes*
LP Day 25 Day 32 Day 39 Day 48
< 0.7 ±0.23a 0.5 ±0.09bc 0.6±0.11 a
FP
CFP
>
Day 25
b
171+26.0 140+48.7a 165 56.6b
*Means + SEM with a same superscript letter in a row are not significantly different (p > 0.05).
*Means + SEM with a same superscript letter in a row are not significantly different (p > 0.05).
for its growth delay; at the end of the experiment, no significant differences appeared between the weight in LP, FP and CFP groups, in spite of a slightly lower growth in the CFP group. On the other hand, significant differences were observed in the survival rates (%): 57 3.9, 21 2.6 and 35 + 3.2 for LP, FP and CFP groups, respectively. No significant growth was obtained in the CP group when survival reached 37 + 3.6.
Amylase activity in the pancreatic segment decreased irrespectively of the diets between day 25 and day 48 and the major decreases were observed in the FP and CFP groups (Table 5). In the intestinal segment, amylase activity reduced with age (Table 6). LP and FP groups exhibited significantly higher activities than CFP and CP groups.
Effect of the diets on enzymes of intestinalBBM Effect of the diets on pancreatic enzyme levels No pronounced differences appeared in trypsin activities assayed in the pancreatic segment of all dietary groups, except for the CP group whose activity was always lower than the FP and CFP groups (Table 3). At day 48, the highest trypsin activities in pancreas were found in FP and CFP groups. Intestinal trypsin activity was always lower in CP than in LP and FP groups (Table 4). The activity detected in the CFP group became as low as in the CP group at day 48. Control animals showed the highest activities in IS.
At day 32, the CFP group presented an AN activity as high as the controls, whereas FP and CP groups were significantly lower (Table 7). From day 32, the AN activity in the FP group was enhanced and reached LP and CFP levels; it still remained low in the CP group. At days 32 and 39, the highest activity of yGT were assayed in the CP group. After this time, its activity dropped (Table 8). In the control group, yGT did not really change between day 32 and 48, whereas it increased significantly in FP and CFP groups. AP activity increased with age, up to 6 times for LP and FP groups, 4 times and 2 times for CFP and
435 Table 8. Specific activities of y-glutamyl transpeptidase (mU mg protein-') in brush border membranes* LP
FP
CFP
Day 25
0.05). Table 9. Specific activities of alkaline phosphatase (mU mg protein - l ) in brush border membranes* LP Day 25 Day 32 Day 39 Day 48
< 539± 75.2a a 784 + 80.5 a 934+ 122.5
FP
CFP
145 ±+23.4 bc
198+35.6 745 39.3a 883+94.0 a
248 + 16 .3b 397+48.4b 537 + 84.9b
Table 10. Specific activities of maltase (mU mg protein-') in brush border membranes*
CP > 182 + 41.3c 250±99.1c 292+ 42.2c
*Means SEM with a same superscript letter in a row are not significantly different (p > 0.05).
CP groups respectively. This increase appeared sooner in the control (day 32) than in the other groups. Among these last three groups, the activity tended to be the highest in the CFP group, even not significantly. Maltase seemed to follow the same pattern as AP, with more pronounced trends (Table 10): at day 32, maltase activity in the CFP group was significantly higher than in FP and CP groups, but remained inferior to the controls. On the other hand, a sharp increase in maltase activity was observed with age in all groups, except in the CP group whose activity remained at a basic level.
Discussion In this experiment, growth obtained in larvae fed on compound diets, FP and CFP, was similar to control, whereas until now, compound diets always induced lower growth (Person Le Ruyet et al. 1993; Zambonino Infante and Cahu 1994; Cahu and Zambonino Infante 1994). This result does not mean that FP and CFP are as efficient as live prey
FP
CFP
>
99+ 14.7 94 + 33.7c 470 + 84.6a 402 ± 64.3a
16 3 + 14.0b 2 6 6+23.9b 310
CP
+5 4. 3 b
88 20.5C 112+ 12.1 c 108 13.9c
*Means ± SEM with a same superscript letter in a row are not significantly different (p > 0.05).
for larval growth, but it demonstrates that a diet with such a simplified protein source allows an acceptable growth. We can point out that, in previous experiments, extruded microparticles formulated with a mixture of different protein sources did not lead to better growth performances (Zambonino Infante and Cahu 1994; Cahu and Zambonino Infante 1994). On the other hand, the introduction of the casein mixture in a diet, alone or with fish meal, enhanced larval survival significantly. Previous experiments have suggested that this improvement may be due to hydrolysate fraction (Cahu and Zambonino Infante 1995). This beneficial effect of the casein hydrolysate has also been reported in the fresh water fish, Carassius auratus (Szlaminska et al. 1991). However, the growth observed in CP group demonstrates that casein, used as the only nitrogen supply, did not meet larval requirements. Lower activities of pancreatic enzymes were observed in larvae fed with diets containing a casein mixture (CFP and CP). This drop in level was mainly confined to the intestinal segment both for amylase and trypsin, and suggests that the casein mixture particularly reduced the secretion of pancreatic enzymes. We have previously reported that the addition of 10% of casein hydrolysate in a compound diet resulted in a low pancreatic secretion of amylase and trypsin (Cahu and Zambonino Infante 1995). The casein mixture, affecting two enzymes as different as trypsin and amylase, could act mainly by a hormonal mechanism. Owyang's (1994) conclusions, based on data obtained in rat, may explain this drop in pancreatic secretions. According to this author, a feedback regulation of pancreatic secretion by trypsin is mediated by a cholecystokinin-releasing peptide which controls the release of
436 CCK. The inactivation of this releasing factor by trypsin is prevented when dietary proteins compete for this enzyme. Owyang also showed that casein hydrolysate does not inhibit trypsin activity, and then fails to stimulate CCK release and pancreatic secretion. A decrease in amylase activity during larval development irrespectively of the diets has been evidenced in this work. The high amylase activities assayed in young larvae, in this study and in previous ones (Zambonino Infante and Cahu 1994; Cahu and Zambonino Infante 1994) may express the predisposition of sea bass to use glucides during the early stages of life. Buddington (1985) assumed that these changes in enzyme activities during the development of lake sturgeon correspond to differences in nutritional requirements of the distinct life stages. Such a depression in amylase activity concomitant with an enhancement in trypsin activity has been also observed during the maturation of the digestive tract in rats (Harada et al. 1994). The CFP group differed from the other groups fed on compound diets, by exhibiting as soon as day 32 high activities of BBM enzymes, AN, y-GT and maltase, similar to the controls. This phenomenon reveals that the maturation of the intestine has begun in this group, at the same time as the controls. Indeed, the sudden increase in the activities of some BBM enzymes indicates intestine maturation in mammal pups at weaning (Harada et al. 1994; Henning 1985; Henning et al. 1994). We can deduce from these data that intestine maturation in the FP group was delayed; the increase in BBM enzyme activities was observed only at day 39. In the case of AP, this phenomenon also seems to occur at day 32 in the CFP group, even though activities were not significantly higher than in the FP group. The fact that fish meal contains more organic phosphorus than casein, 1.3% vs. 0.8%, can explain that AP activity increased in the FP group. Indeed, it is well known that AP is stimulated by phosphorylated substrates as phosphoproteins and phospholipids (Shirazi et al. 1978; McCarthy et al. 1980) and this effect appeared clearly from day 39. Larvae fed on formulated diets exhibited y-GT activities higher than the control. y-GT is involved in the synthesis and degradation of glutathione
(Orlowski and Meister 1970; Meister et al. 1981) which is considered as a reservoir for mucosal cysteine in mammals. The level of glutathione degradation by BBM y-GT is inversely related to the luminal thiol levels (Dahm and Jones 1994). The three formulated diets used were poor in cysteine compared to live preys (three times less). We can hypothesize that these low cysteine levels stimulated yGT which in return released cysteine-glycine from glutathione. The assay of mucosal glutathione and other enzymes involved in the y-glutamyl cycle appears necessary before any further speculations. The fall in y-GT activity observed in CP groups at the end of the experiment may express a poor nutritional status. This study shows that specific phenomena, involved in the digestive tract maturation of sea bass, occurred between day 25 and 39. A depression in amylase and an enhancement in trypsin were observed in the pancreas, concurrently with a clear increase in BBM enzyme activities, especially maltase, in the intestine. These observations are consistent with the maturation processes described in mammals at the time of weaning (Harada et al. 1994; Henning 1985; Henning et al. 1994). Moreover, a sharp increase in BBM enzyme activities between day 25 and 32 appeared to be crucial for larval survival. The addition of protein hydrolysate in a weaning diet seems to facilitate this maturation process.
References cited Beccaria, C., Diaz, J.P., Connes, R. and Chatain, B. 1991. Organogenesis of the exocrine pancreas in the sea bass, Dicentrarchuslabrax L., reared extensively and intensively. Aquaculture 99: 339-354. Bessey, O.A., Lowry, O.H. and Brock, M.J. 1946. Rapid coloric method for determination of alcaline phosphatase in five cubic millimeters of serum. J. Biol. Chem. 164: 321-329. Bidlingmeyer, B.A., Cohen, S.A. and Tarvin, T.L. 1984. Rapid analysis of amino acids using pre-column derivatization. J. Chromatogr. 336: 93-104. Bradford, M.M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248254. Buddington, R.K. 1985. Digestive secretions of lake sturgeon,
437 Acipenserfulvescens, during early development. J. Fish Biol. 26: 715-723. Cahu, C.L. and Zambonino Infante, J.L. 1994. Early weaning of sea bass (Dicentrarchuslabrax) larvae with a compound diet: effect on digestive enzymes. Comp. Biochem. Physiol. 109A: 213-222. Cahu, C.L. and Zambonino Infante, J.L. 1995. Effect of molecular form of nitrogen supply in sea bass larvae: response of pancreatic enzymes and intestinal peptidases. Fish Physiol. Biochem. in press. Charlon, N. and Bergot, P. 1984. Rearing systems for feeding fish larvae on dry diet. Trial with carp Cyprinus carpio L. larvae. Aquaculture 41: 1-9. Crane, R.K., Boge, G. and Rigal, A. 1979. Isolation of brush border membranes in vesicular form from the intestinal spiral valve of the small dogfish (Scyliorhinus canicula). Biochim. Biophys. Acta 554: 264-267. Dabrowski, K. 1984. The feeding of fish larvae: Present "state of the art" and perspectives. Reprod. Nutr. Dev. 24: 807833. Dabrowski, K. 1986. Ontogenical aspects of nutritional requirements in fish. Comp. Biochem. Physiol. 85A: 639-655. Dahlqvist, A. 1970. Assay of intestinal disaccharidase. Enzym. Biol. Clin. 11: 52-56. Dahm, L.J. and Jones, D.P. 1994. Secretion of cysteine and glutathione from mucosa to lumen in rat small intestine. Am. J. Physiol. 267: G292-G300. Ferraris, R.P., Tan, J.D. and De la Cruz, M.C. 1987. Development of the digestive tract of milk fish, Chanos chanos (Forsskal): histology and histochemistry. Aquaculture 61: 241257. Fyhn, H.J. 1989. First feeding of marine fish larvae: are free amino acids the source of energy? Aquaculture 80: 111-120. Harada, E., Hashimoto, Y. and Syuto, B. 1994. Precocious cessation of intestinal macromolecular transport and digestive enzymes development by prostaglandin E(2) in suckling rats. Comp. Biochem. Physiol. 109A: 245-253. Heintges, T., Luthen, R. and Niederau, C. 1994. Inhibition of exocrine pancreatic secretion by somatostatin and its analogues. Digestion 55: 1-9. Henning, S.J. 1985. Ontogeny of enzymes in the small intestine. Ann. Rev. Physiol. 47: 231-245. Henning, S.J., Rubin, D.C. and Shulman, R.J. 1994. Ontogeny of the intestinal mucosa. In Physiology of the GastrointestinalTract. pp. 571-610. Edited by L.R. Johnson. Raven Press, New York. Holm, H., Hanssen, L.E., Krogdahl, A. and Florholmen, J. 1988. High and low inhibitor soybean meals affect human duodenal proteinase activity differently: in vivo comparison with bovine serum albumin. J. Nutr. 118: 515-520. Kanazawa, A., Koshio, S. and Teshima, S. 1989. Growth and survival of larval red sea bream Pagrusmajor and Japanese
flounder Paralichthys olivaceus fed microbound diets. J. World Aquacult. Soc. 20: 31-37. Lockwood, S.J. 1973. Weight and length of O-group plaice (Pleuronectes platessa L.) after preservation in 4% neutral formalin. J. Cons. Int. Explor. Mer. 35: 100-101. Maroux, S., Louvard, D. and Baratti, J. 1973. The aminopeptidase from hog-intestinal brush border. Biochim. Biophys. Acta 321: 282-295. Meister, A., Tate, S.S. and Griffith, O.W. 1981. y-glutamyl transpeptidase. In Methods in Enzymology. Vol. 77, pp. 237253. Edited by W.B. Jakoby. Academic Press, New York. Metais, P. and Bieth, J. 1968. Determination de l'a-amylase par une microtechnique. Ann. Biol. Clin. 26: 133-142. McCarthy, D.M., Nicholson, J.A. and Kim, Y.S. 1980. Intestinal enzyme adaptation to normal diets of different composition. Am. J. Physiol. 239: G445-G451. Orlowski, M. and Meister, A. 1970. The y-glutamyl cycle: a possible transport system for amino acids. Proc. Nat. Acad. Sci. U.S.A. 67: 1248-1255. Owyang, C. 1994. Negative feedback control of exocrine pancreatic secretion: role of cholecystokinin and cholinergic pathway. J. Nutr. 124: 1321S-1326S. Person Le Ruyet, J., Alexandre, J.C., Thebaud, L. and Mugnier, C. 1993. Marine fish larvae feeding: formulated diets or live preys? J. World Aquac. Soc. 24: 211-224. Shirazi, S.P., Colston, K.W. and Butterworth, P.J. 1978. Alkaline phosphatase: a possible transport for inorganic phosphase. Biochem. Soc. Trans. 6: 933-935. Szlaminska, M., Escaffre, A.M., Charlon, N. and Bergot, P. 1991. Preliminary data on semisynthetic diets for goldfish (Carassiusauratus L.) larvae. In Fish Nutrition in Practice. Les Colloques 61, pp. 607-612. Edited by S.J. Kaushik and P. Luquet. INRA, Paris. Teshima, S., Kanazawa, A. and Sakamoto, M. 1982. Microparticulate diets for the larvae of aquatic animals. Min. Rev. Data File Fish. Res. 2: 67-86. Walford, J. and Lam, T.J. 1991. Growth and survival of sea bass (Lates calcarifer) larvae fed microencapsulated diets alone or together with live food. In Larvi'91-Fish and crustacean larviculture symposium. pp. 184-187. Edited by P. Lavens, P. Sorgeloos, E. Jasper and F. Ollevier. European Aquaculture Society, Special Publication No. 15, Gent, Belgium. Walford, J. and Lam, T.J. 1993. Development of digestive tract and proteolyc enzyme activity in sea bass (Lates calcarifer) larvae and juvenile. Aquaculture 109: 187-205. Watanabe, T. and Kiron, V. 1994. Prospects in larval fish dietetics. Aquaculture 124: 223-251. Zambonino Infante, J.L. and Cahu, C.L. 1994. Development and response to a diet change of some digestive enzymes in sea bass (Dicentrarchus labrax) larvae. Fish Physiol. Biochem. 12: 399-408.
Maturation of the pancreatic and intestinal digestive functions in sea bass (Dicentrarchus labrax): effect of weaning with different protein sources C.L. Cahu and J.L. Zambonino Infante Unite Mixte de Nutrition des Poissons IFREMER-INRA, IFREMER Centre de Brest, B.P. 70, 29280 Plouzan, France Accepted: June 23, 1995 Keywords: pancreatic enzymes, intestinal brush border enzymes, weaning, maturation, protein, marine fish larvae
Abstract The maturation of the digestive functions in sea bass (Dicentrarchuslabrax) larvae was evaluated by the enzymatic profile of pancreas and intestine brush border membranes. Sea bass larvae were weaned at day 25 with three simplified diets different by their protein nature: 100°70 fish meal (FP), 100% casein mixture (CP) and 50% fish meal-50% casein mixture (CFP). The casein mixture contained 35°70 of hydrolysate. The control group was fed live preys. The specific activity of amylase decreased with age irrespectively of the diets whereas the specific activity of trypsin was enhanced. The casein mixture reduced pancreatic secretion in amylase and trypsin. The CFP group differed from the other groups fed on compound diets, exhibiting as soon as day 32 high activities of brush border enzymes, similar to controls. This sharp increase between day 25 and 32 appeared to be crucial for larval survival. The addition of a protein hydrolysate in a weaning diet seems to facilitate this maturation process.
Introduction
Compound diet substitution for live prey is still crucial in marine fish larvae rearing; this phase, known as weaning, can only be performed after some weeks of life in marine species, whereas it can be done as early as mouth opening in fresh water species. Until now, the protein fraction of weaning diets has been composed of different sources such as fish meal, krill meal, squid meal, mussel, egg, beef liver, milk, casein, wheat, soybean, yeast (Teshima et al. 1982; Charlon and Bergot 1984; Dabrowski 1984; Person Le Ruyet et al. 1993; Zambonino Infante and Cahu 1994; Watanabe and Kiron 1994). The use of mixtures containing a great number of protein sources reveals a certain lack of knowledge of the specific requirements of the young stages. This empirical approach was mainly
based on the results of body composition, growth and survival of larvae (Kanazawa et al. 1989; Fyhn 1989; Walford and Lam 1991), the main objective being the reduction of the growth delay observed in larvae fed on compound diet compared to larvae fed on live prey. These diets have been generally formulated from scanty information concerning the specificities of the larvae, and especially concerning the degree of maturation of their digestive tract. Effectively, the digestive processes appear sequentially during the first thirty days of larval life (Dabrowski 1986; Ferraris et al. 1987; Beccaria et al. 1991; Walford and Lam 1993). We have recently reported that, as in mammals, the secretory function of the exocrine pancreas takes place chronologically after the synthesis function; in the same way, the digestive features of the enterocytes evolve by enhancing the
432 Table 1. Composition of the experimental diets (g 100g dry matter - ') LP a
AGE (days)
Fig. 1. Growth of the four experimental groups of larvae ( O LP, FP, X CFP, CP). Data are given as means + SEM (n = 30).
hydrolytic functions of the brush border membrane. Weaning using a conventional compound diet delays or stops these maturation processes (Cahu and Zambonino Infante 1994, 1995). The use of a simplified diet containing well identified components and a restricted number of protein sources is necessary to understand the nutrientdigestive tract interaction. The aim of this work was to study the effect of the nature of the protein supply on the development of larvae, and particularly their digestive capacity, by simplifying the formulation of a weaning diet.
Materials and methods Animals and diets Eggs of European sea bass (Dicentrarchuslabrax) were obtained from the Ifremer-Station of Palavas. The larvae were reared for 48 days at the IfremerStation de Brest. Newly hatched larvae were transferred from incubators to 20 conical fiber glass tanks (35 1)with black walls (initial stocking density: 80 larvae 1-1). They were supplied with running sea water which had been filtered through a sand filter, then passed successively through a tungsten heater and a degassing column packed with plastic rings. Temperature range was 18-19°C, salinity 34.5 ppt. The oxygen level was maintained above 6 ppm by setting the water exchange up to 30% per hour (flow rate: 0.18 1 min-1). The light
Fish meal' Casein mixture Casein 2 Casein salt 3 Hydrolysate 4 DL methionine L tryptophan Maltose Starch Cellulose Cod liver oil Lecithin Mineral mixture Vitamin mixture Ascorbic acid Choline Inositol N x 6.25 Lipids Ash Energy J 100 g-1
FP
CFP
74
37
35% 28%7 35% 1.5% 0.5%
0
57 10 11.0 1701
0
28.5
57
4.9 8 3 3.7 0.9 2 2.2 0.3 0.5 0.5
4.9 8 6.3 7.8 2 2 2.2 0.3 0.5 0.5
4.9 8 9.6 12 3 2 2.2 0.3 0.5 0.5
57 15 11.7
57 15 7.8
57 15 4.3
1735
'Norseamink, 77% protein, 14% lipid; C8654; 4 Sigma N4642.
CP
2
1735
1735
Merck 2242;
3
Sigma
intensity was 9 Wm - 2 maximum at the surface. The larvae were allocated into 4 groups (5 tanks per group). In the control group (LP group), larvae were fed live prey until the end of the experiment (day 48), following the sequence described in a previous paper (Zambonino Infante and Cahu 1994). Larvae of the 3 other groups were weaned at day 25 with compound diets containing 57% protein, exclusively fish meal or a casein mixture for FP and CP groups respectively; for CFP group, half the nitrogen supply was made up of fish meal, the other of casein mixture (Table 1). The amino acid composition of the 3 diets is given in Table 2. Microparticles of size 200-400 gAm were used. Fish were continuously fed in large excess 18h per day using a belt feeder. Food ingestion was controlled by observing the larvae digestive tract under a binocular microscope.
433 Table 2. Amino acid composition of the diets (% dry matter) LIP
FP
CFP
CP
3.7 8.2 1.8 2.2 3.4 1.7 2.1 0.2 2.6 0.7 2.1 2.8 1.7 2.1 3.5 1.1 1.9
3.4 10.4 1.8 2.5 4.9 0.9 1.5 0.3 3.0 0.9 2.5 4.0 2.0 2.4 3.6 1.2 1.6
Table 3. Specific activities of trypsin (mU mg protein-') in pancreatic segment* LP
Aspartic acid Glutamic acid Threonine Serine Proline Glycine Alanine Cysteine Valine Methionine Isoleucine Leucine Tyrosine Phenylalanine Lysine Histidine Arginine
4.2 5.5 2.1 2.4 2.7 2.2 2.5 0.7 2.4 1.2 2.3 3.1 2.2 2.2 3.5 1.4 3.0
4.4 6.2 1.9 1.8 1.8 2.9 3.0 0.2 2.4 0.8 2.2 3.4 1.5 2.0 3.6 1.0 2.5
Sampling and dissection In order to monitor growth, 30 larvae were taken from each group every week. They were preserved for one month in 4% formaldehyde sea water prior to weighing (Lockwood 1973). Larval survival rates were determined by counting individually at the end of the experiment. Larvae were collected at day 25, 32, 39 and 48 from each tank before morning food distribution. They were immediately stored at -80 0 C pending dissection and assays. Dissections under microscope were conducted on a glass maintained at 0°C. Animals were cut into 4 parts: head, pancreatic segment (PS), intestinal segment (IS) and tail, in order to assay enzymes only on pancreatic and intestinal segments as described previously (Cahu and Zambonino Infante 1994).
FP
CFP
CP
Day 25
b ab
9+5.0C 33 2.3b
39+5.7 b
*Means SEM with a same superscript letter in a row are not significantly different (p > 0.05).
30 volumes (v/w) of mannitol 50 mM-Tris, 2 mM, pH = 7. Alkaline phosphatase (AP), leucine aminopeptidase N (AN), y-glutamyl transpeptidase (yGT) and maltase were assayed according to Bessey et al. (1946), Maroux et al. (1973), Meister et al. (1981), and Dahlqvist (1970), respectively. Enzyme specific activities were expressed as moles of substrate hydrolysed per minute per mg protein (i.e., U mg prot- 1) at 370 C for AP, AN, yGT and maltase, and at 25°C for trypsin. Amylase activity was expressed as the equivalent enzyme activity which was required to hydrolyse one mg of starch in 30 min at 37C. Purified brush border membranes (BBM) from the homogenate of intestinal segment were obtained according to a method designed for intestinal scraping (Crane et al. 1979). Protein was determined by the Bradford procedure (1976). The amino acid composition of the diets was analysed using a Waters HPLC system. Hydrolysis, derivatization and analysis were made according to the Pico-Tag method (Bidlingmeyer et al. 1984). Statistical analysis Results are given as means SEM (n = 5). Data were compared by one way analysis of variance followed by LSD multiple range test when significant differences were found at a 0.05 level.
Analytical methods
Results Results
Pancreatic segments were homogenized in 5 volumes (v/w) of ice-cold distilled water. Trypsin and amylase activities were assayed according to Holm et al. (1988) and Metais and Bieth (1968), respectively. Intestinal segments were homogenized in
Effect of the diets on larval growth and survival Larvae fed on compound diets had a significantly lower growth than control animals before day 39 (p < 0.05). From this date, the FP group made up
434 Table 4. Specific activities of trypsin (mU mg protein-') in intestinal segment* LP
FP
CFP
20 ± 2.6b
14 + 3.2c
20 ± 2.2ab
16+ 3.lb
27+ 1.5C
28+3.6C
Day 25
2.1+0.75 3.3 0.39a 1.8 0.26a 1.1+0.22a
2.4 0.22b 0.9 +0.36b 0.6 +0.20b
1.8 0.49b 1.0+0.59b 0.5 0.16b
*Means SEM with a same superscript letter in a row are not significantly different (p > 0.05).
*Means SEM with a same superscript letter in a row are not significantly different (p > 0.05).
Table 5. Specific activities of amylase (U mg protein- 1) in pancreatic segment*
Table 7. Specific activities of leucine aminopeptidase (mU mg protein-') in brush border membranes*
LP Day 25 Day 32 Day 39 Day 48
< 0.7 ±0.23a 0.5 ±0.09bc 0.6±0.11 a
FP
CFP
>
Day 25
b
171+26.0 140+48.7a 165 56.6b
*Means + SEM with a same superscript letter in a row are not significantly different (p > 0.05).
*Means + SEM with a same superscript letter in a row are not significantly different (p > 0.05).
for its growth delay; at the end of the experiment, no significant differences appeared between the weight in LP, FP and CFP groups, in spite of a slightly lower growth in the CFP group. On the other hand, significant differences were observed in the survival rates (%): 57 3.9, 21 2.6 and 35 + 3.2 for LP, FP and CFP groups, respectively. No significant growth was obtained in the CP group when survival reached 37 + 3.6.
Amylase activity in the pancreatic segment decreased irrespectively of the diets between day 25 and day 48 and the major decreases were observed in the FP and CFP groups (Table 5). In the intestinal segment, amylase activity reduced with age (Table 6). LP and FP groups exhibited significantly higher activities than CFP and CP groups.
Effect of the diets on enzymes of intestinalBBM Effect of the diets on pancreatic enzyme levels No pronounced differences appeared in trypsin activities assayed in the pancreatic segment of all dietary groups, except for the CP group whose activity was always lower than the FP and CFP groups (Table 3). At day 48, the highest trypsin activities in pancreas were found in FP and CFP groups. Intestinal trypsin activity was always lower in CP than in LP and FP groups (Table 4). The activity detected in the CFP group became as low as in the CP group at day 48. Control animals showed the highest activities in IS.
At day 32, the CFP group presented an AN activity as high as the controls, whereas FP and CP groups were significantly lower (Table 7). From day 32, the AN activity in the FP group was enhanced and reached LP and CFP levels; it still remained low in the CP group. At days 32 and 39, the highest activity of yGT were assayed in the CP group. After this time, its activity dropped (Table 8). In the control group, yGT did not really change between day 32 and 48, whereas it increased significantly in FP and CFP groups. AP activity increased with age, up to 6 times for LP and FP groups, 4 times and 2 times for CFP and
435 Table 8. Specific activities of y-glutamyl transpeptidase (mU mg protein-') in brush border membranes* LP
FP
CFP
Day 25
0.05). Table 9. Specific activities of alkaline phosphatase (mU mg protein - l ) in brush border membranes* LP Day 25 Day 32 Day 39 Day 48
< 539± 75.2a a 784 + 80.5 a 934+ 122.5
FP
CFP
145 ±+23.4 bc
198+35.6 745 39.3a 883+94.0 a
248 + 16 .3b 397+48.4b 537 + 84.9b
Table 10. Specific activities of maltase (mU mg protein-') in brush border membranes*
CP > 182 + 41.3c 250±99.1c 292+ 42.2c
*Means SEM with a same superscript letter in a row are not significantly different (p > 0.05).
CP groups respectively. This increase appeared sooner in the control (day 32) than in the other groups. Among these last three groups, the activity tended to be the highest in the CFP group, even not significantly. Maltase seemed to follow the same pattern as AP, with more pronounced trends (Table 10): at day 32, maltase activity in the CFP group was significantly higher than in FP and CP groups, but remained inferior to the controls. On the other hand, a sharp increase in maltase activity was observed with age in all groups, except in the CP group whose activity remained at a basic level.
Discussion In this experiment, growth obtained in larvae fed on compound diets, FP and CFP, was similar to control, whereas until now, compound diets always induced lower growth (Person Le Ruyet et al. 1993; Zambonino Infante and Cahu 1994; Cahu and Zambonino Infante 1994). This result does not mean that FP and CFP are as efficient as live prey
FP
CFP
>
99+ 14.7 94 + 33.7c 470 + 84.6a 402 ± 64.3a
16 3 + 14.0b 2 6 6+23.9b 310
CP
+5 4. 3 b
88 20.5C 112+ 12.1 c 108 13.9c
*Means ± SEM with a same superscript letter in a row are not significantly different (p > 0.05).
for larval growth, but it demonstrates that a diet with such a simplified protein source allows an acceptable growth. We can point out that, in previous experiments, extruded microparticles formulated with a mixture of different protein sources did not lead to better growth performances (Zambonino Infante and Cahu 1994; Cahu and Zambonino Infante 1994). On the other hand, the introduction of the casein mixture in a diet, alone or with fish meal, enhanced larval survival significantly. Previous experiments have suggested that this improvement may be due to hydrolysate fraction (Cahu and Zambonino Infante 1995). This beneficial effect of the casein hydrolysate has also been reported in the fresh water fish, Carassius auratus (Szlaminska et al. 1991). However, the growth observed in CP group demonstrates that casein, used as the only nitrogen supply, did not meet larval requirements. Lower activities of pancreatic enzymes were observed in larvae fed with diets containing a casein mixture (CFP and CP). This drop in level was mainly confined to the intestinal segment both for amylase and trypsin, and suggests that the casein mixture particularly reduced the secretion of pancreatic enzymes. We have previously reported that the addition of 10% of casein hydrolysate in a compound diet resulted in a low pancreatic secretion of amylase and trypsin (Cahu and Zambonino Infante 1995). The casein mixture, affecting two enzymes as different as trypsin and amylase, could act mainly by a hormonal mechanism. Owyang's (1994) conclusions, based on data obtained in rat, may explain this drop in pancreatic secretions. According to this author, a feedback regulation of pancreatic secretion by trypsin is mediated by a cholecystokinin-releasing peptide which controls the release of
436 CCK. The inactivation of this releasing factor by trypsin is prevented when dietary proteins compete for this enzyme. Owyang also showed that casein hydrolysate does not inhibit trypsin activity, and then fails to stimulate CCK release and pancreatic secretion. A decrease in amylase activity during larval development irrespectively of the diets has been evidenced in this work. The high amylase activities assayed in young larvae, in this study and in previous ones (Zambonino Infante and Cahu 1994; Cahu and Zambonino Infante 1994) may express the predisposition of sea bass to use glucides during the early stages of life. Buddington (1985) assumed that these changes in enzyme activities during the development of lake sturgeon correspond to differences in nutritional requirements of the distinct life stages. Such a depression in amylase activity concomitant with an enhancement in trypsin activity has been also observed during the maturation of the digestive tract in rats (Harada et al. 1994). The CFP group differed from the other groups fed on compound diets, by exhibiting as soon as day 32 high activities of BBM enzymes, AN, y-GT and maltase, similar to the controls. This phenomenon reveals that the maturation of the intestine has begun in this group, at the same time as the controls. Indeed, the sudden increase in the activities of some BBM enzymes indicates intestine maturation in mammal pups at weaning (Harada et al. 1994; Henning 1985; Henning et al. 1994). We can deduce from these data that intestine maturation in the FP group was delayed; the increase in BBM enzyme activities was observed only at day 39. In the case of AP, this phenomenon also seems to occur at day 32 in the CFP group, even though activities were not significantly higher than in the FP group. The fact that fish meal contains more organic phosphorus than casein, 1.3% vs. 0.8%, can explain that AP activity increased in the FP group. Indeed, it is well known that AP is stimulated by phosphorylated substrates as phosphoproteins and phospholipids (Shirazi et al. 1978; McCarthy et al. 1980) and this effect appeared clearly from day 39. Larvae fed on formulated diets exhibited y-GT activities higher than the control. y-GT is involved in the synthesis and degradation of glutathione
(Orlowski and Meister 1970; Meister et al. 1981) which is considered as a reservoir for mucosal cysteine in mammals. The level of glutathione degradation by BBM y-GT is inversely related to the luminal thiol levels (Dahm and Jones 1994). The three formulated diets used were poor in cysteine compared to live preys (three times less). We can hypothesize that these low cysteine levels stimulated yGT which in return released cysteine-glycine from glutathione. The assay of mucosal glutathione and other enzymes involved in the y-glutamyl cycle appears necessary before any further speculations. The fall in y-GT activity observed in CP groups at the end of the experiment may express a poor nutritional status. This study shows that specific phenomena, involved in the digestive tract maturation of sea bass, occurred between day 25 and 39. A depression in amylase and an enhancement in trypsin were observed in the pancreas, concurrently with a clear increase in BBM enzyme activities, especially maltase, in the intestine. These observations are consistent with the maturation processes described in mammals at the time of weaning (Harada et al. 1994; Henning 1985; Henning et al. 1994). Moreover, a sharp increase in BBM enzyme activities between day 25 and 32 appeared to be crucial for larval survival. The addition of protein hydrolysate in a weaning diet seems to facilitate this maturation process.
References cited Beccaria, C., Diaz, J.P., Connes, R. and Chatain, B. 1991. Organogenesis of the exocrine pancreas in the sea bass, Dicentrarchuslabrax L., reared extensively and intensively. Aquaculture 99: 339-354. Bessey, O.A., Lowry, O.H. and Brock, M.J. 1946. Rapid coloric method for determination of alcaline phosphatase in five cubic millimeters of serum. J. Biol. Chem. 164: 321-329. Bidlingmeyer, B.A., Cohen, S.A. and Tarvin, T.L. 1984. Rapid analysis of amino acids using pre-column derivatization. J. Chromatogr. 336: 93-104. Bradford, M.M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248254. Buddington, R.K. 1985. Digestive secretions of lake sturgeon,
437 Acipenserfulvescens, during early development. J. Fish Biol. 26: 715-723. Cahu, C.L. and Zambonino Infante, J.L. 1994. Early weaning of sea bass (Dicentrarchuslabrax) larvae with a compound diet: effect on digestive enzymes. Comp. Biochem. Physiol. 109A: 213-222. Cahu, C.L. and Zambonino Infante, J.L. 1995. Effect of molecular form of nitrogen supply in sea bass larvae: response of pancreatic enzymes and intestinal peptidases. Fish Physiol. Biochem. in press. Charlon, N. and Bergot, P. 1984. Rearing systems for feeding fish larvae on dry diet. Trial with carp Cyprinus carpio L. larvae. Aquaculture 41: 1-9. Crane, R.K., Boge, G. and Rigal, A. 1979. Isolation of brush border membranes in vesicular form from the intestinal spiral valve of the small dogfish (Scyliorhinus canicula). Biochim. Biophys. Acta 554: 264-267. Dabrowski, K. 1984. The feeding of fish larvae: Present "state of the art" and perspectives. Reprod. Nutr. Dev. 24: 807833. Dabrowski, K. 1986. Ontogenical aspects of nutritional requirements in fish. Comp. Biochem. Physiol. 85A: 639-655. Dahlqvist, A. 1970. Assay of intestinal disaccharidase. Enzym. Biol. Clin. 11: 52-56. Dahm, L.J. and Jones, D.P. 1994. Secretion of cysteine and glutathione from mucosa to lumen in rat small intestine. Am. J. Physiol. 267: G292-G300. Ferraris, R.P., Tan, J.D. and De la Cruz, M.C. 1987. Development of the digestive tract of milk fish, Chanos chanos (Forsskal): histology and histochemistry. Aquaculture 61: 241257. Fyhn, H.J. 1989. First feeding of marine fish larvae: are free amino acids the source of energy? Aquaculture 80: 111-120. Harada, E., Hashimoto, Y. and Syuto, B. 1994. Precocious cessation of intestinal macromolecular transport and digestive enzymes development by prostaglandin E(2) in suckling rats. Comp. Biochem. Physiol. 109A: 245-253. Heintges, T., Luthen, R. and Niederau, C. 1994. Inhibition of exocrine pancreatic secretion by somatostatin and its analogues. Digestion 55: 1-9. Henning, S.J. 1985. Ontogeny of enzymes in the small intestine. Ann. Rev. Physiol. 47: 231-245. Henning, S.J., Rubin, D.C. and Shulman, R.J. 1994. Ontogeny of the intestinal mucosa. In Physiology of the GastrointestinalTract. pp. 571-610. Edited by L.R. Johnson. Raven Press, New York. Holm, H., Hanssen, L.E., Krogdahl, A. and Florholmen, J. 1988. High and low inhibitor soybean meals affect human duodenal proteinase activity differently: in vivo comparison with bovine serum albumin. J. Nutr. 118: 515-520. Kanazawa, A., Koshio, S. and Teshima, S. 1989. Growth and survival of larval red sea bream Pagrusmajor and Japanese
flounder Paralichthys olivaceus fed microbound diets. J. World Aquacult. Soc. 20: 31-37. Lockwood, S.J. 1973. Weight and length of O-group plaice (Pleuronectes platessa L.) after preservation in 4% neutral formalin. J. Cons. Int. Explor. Mer. 35: 100-101. Maroux, S., Louvard, D. and Baratti, J. 1973. The aminopeptidase from hog-intestinal brush border. Biochim. Biophys. Acta 321: 282-295. Meister, A., Tate, S.S. and Griffith, O.W. 1981. y-glutamyl transpeptidase. In Methods in Enzymology. Vol. 77, pp. 237253. Edited by W.B. Jakoby. Academic Press, New York. Metais, P. and Bieth, J. 1968. Determination de l'a-amylase par une microtechnique. Ann. Biol. Clin. 26: 133-142. McCarthy, D.M., Nicholson, J.A. and Kim, Y.S. 1980. Intestinal enzyme adaptation to normal diets of different composition. Am. J. Physiol. 239: G445-G451. Orlowski, M. and Meister, A. 1970. The y-glutamyl cycle: a possible transport system for amino acids. Proc. Nat. Acad. Sci. U.S.A. 67: 1248-1255. Owyang, C. 1994. Negative feedback control of exocrine pancreatic secretion: role of cholecystokinin and cholinergic pathway. J. Nutr. 124: 1321S-1326S. Person Le Ruyet, J., Alexandre, J.C., Thebaud, L. and Mugnier, C. 1993. Marine fish larvae feeding: formulated diets or live preys? J. World Aquac. Soc. 24: 211-224. Shirazi, S.P., Colston, K.W. and Butterworth, P.J. 1978. Alkaline phosphatase: a possible transport for inorganic phosphase. Biochem. Soc. Trans. 6: 933-935. Szlaminska, M., Escaffre, A.M., Charlon, N. and Bergot, P. 1991. Preliminary data on semisynthetic diets for goldfish (Carassiusauratus L.) larvae. In Fish Nutrition in Practice. Les Colloques 61, pp. 607-612. Edited by S.J. Kaushik and P. Luquet. INRA, Paris. Teshima, S., Kanazawa, A. and Sakamoto, M. 1982. Microparticulate diets for the larvae of aquatic animals. Min. Rev. Data File Fish. Res. 2: 67-86. Walford, J. and Lam, T.J. 1991. Growth and survival of sea bass (Lates calcarifer) larvae fed microencapsulated diets alone or together with live food. In Larvi'91-Fish and crustacean larviculture symposium. pp. 184-187. Edited by P. Lavens, P. Sorgeloos, E. Jasper and F. Ollevier. European Aquaculture Society, Special Publication No. 15, Gent, Belgium. Walford, J. and Lam, T.J. 1993. Development of digestive tract and proteolyc enzyme activity in sea bass (Lates calcarifer) larvae and juvenile. Aquaculture 109: 187-205. Watanabe, T. and Kiron, V. 1994. Prospects in larval fish dietetics. Aquaculture 124: 223-251. Zambonino Infante, J.L. and Cahu, C.L. 1994. Development and response to a diet change of some digestive enzymes in sea bass (Dicentrarchus labrax) larvae. Fish Physiol. Biochem. 12: 399-408.
