Fish Physiology and Biochemistry vol. 14 no. 3 pp 179-194 (1995) Kugler Publications, Amsterdam/New York
Dynamics of plasma free amino acids in rainbow trout (Oncorhynchus mykiss under variety of dietary conditions Chutima Tantikittil and B.E. March Department of Animal Science, University of British Columbia, Vancouver BC, Canada V6T 1Z4 Accepted: November 24, 1994 Keywords: rainbow trout, plasma amino acids, dietary protein, dietary free amino acids, feeding frequency
Abstract Plasma amino acid profiles in the systemic circulation of rainbow trout acclimated to diets containing different protein sources, with and without supplementation with free amino acids, were monitored for up to 120 hours after consumption of the diets. Plasma concentrations of lysine, methionine, and isoleucine increased more rapidly after feeding a diet supplemented with these amino acids in free form and, depending upon the increase in plasma concentration, remained at concentrations above those in fish fed the basal diet for at least 24h after feeding. Dietary supplementation with isoleucine increased plasma concentrations of leucine and valine as well as of isoleucine. Maximum plasma concentrations for most amino acids were attained between 12 and 24h postprandial. Dietary inclusion of gelatin caused more rapid elevations in plasma glycine and serine than did free glycine. Feeding at three hour intervals resulted in stable plasma concentrations of individual amino acids in contrast to the fluctuations occurring when fish were fed once daily. Frequent feeding favoured a higher proportion of protein to lipid in body composition.
Introduction There are reports that dietary supplements of crystalline amino acids are not utilized efficiently by fish fed only once daily because they may be absorbed and oxidized before amino acids derived from digestion of the dietary protein are absorbed (Cowey and Walton, 1988; Robinson, 1992). Studies with several species of fish have shown that growth rate and feed efficiency are reduced in fish fed diets supplemented with amino acids (Aoe et al. 1970; Wilson et al. 1978; Walton et al. 1986). Rainbow trout fed amino acid based diets displayed more rapid appearance and higher concentrations 1
of free amino acids in the plasma (Yamada et al. 1981). Similarly, when a mixture of proteins is consumed, not all of the amino acids may be available for absorption at the same time because proteins vary in their rates of digestion. In either case the concentrations of amino acids simultaneously available to body tissues for protein synthesis may not optimal. The implications of the above in the formulation of practical diets for cultured fish are obvious. The present experiments were designed to investigate the degree to which amino acid profiles in the systemic circulation vary following meal consumption depending upon dietary protein source, and
Present address: Department of Aquatic Science, Faculty of Natural Resources, Prince of Songkla University, Hat Yai, Thailand Correspondence to: B.E. March, Department of Animal Science, University of British Columbia, Vancouver BC, Canada V6T IZ4.
180 whether amino acids are supplied entirely from intact protein or whether the diet is supplemented with free amino acids.
Materials and methods Experiment 1: Plasma amino acid patterns, measured over 120h post prandial, were compared in response to diets containing (1) fishmeal as the primary source of protein, (2) a mixture of protein sources with presumablydifferent ratesof digestion and deficient in some essentialamino acids, and (3) the above mixture ofprotein sources supplemented with free amino acids Rainbow trout used in the experiment had a mean weight of 180 ± 16 g (SD). The fish were distributed into fifteen indoor 150 1tanks with 10 fish per tank. The water supply was dechlorinated Vancouver city water. Water temperature was maintained at 13-14°C. Water flow to each tank was 2 1 per min. Photoperiod was 14:10 L:D. The compositions of the three diets formulated for the experiment are shown in Table 1. Diet 1 contained anchovy meal as the principal source of protein. Diet 2 contained a mixture of anchovy meal, soybean oil meal, gelatin, and corn gluten meal. Diet 3 contained the same proportion of protein from the respective sources as were used in Diet 2, but was supplemented with lysine, methionine, isoleucine, and tryptophan to provide dietary concentrations equivalent to those calculated to be present in Diet 1. The diets were analyzed for dry matter, protein and lipid, and were subjected to acid hydrolysis for determination of amino acid concentrations (AOAC 1984). Each experimental treatment was assigned to five replicate tanks of fish. The fish were gradually accustomed to the experimental diets over a five-day period by mixing increasing amounts of the respective diets with the commercial diet that had been fed to the fish during the pre-experimental period. The fish were hand-fed once daily at 09:30h until they appeared satiated. Record was kept of feed consumption. The experimental period was 10 days following acceptance of the diets.
Blood samplingprocedure On the 10th day of the experiment blood samples were drawn from the fish at 3, 12, 24, 36, 48, 72, and 120h after feeding. Six fish were taken from one replicate tank for each dietary treatment at the first four sampling times. Fish for the final sampling were taken randomly from all of the tanks for the respective treatments. The fish that were sampled were anaesthetized with 0.01% MS 222 and blood withdrawn from the caudal blood vessels into heparinized vacutainers. Amino acid concentrations in this and the subsequent two experiments therefore refer to blood in the systemic circulation that would be available to the tissues for protein synthesis. Following centrifugation of the blood; the plasma samples from three fish were pooled so that there were duplicate samples from each dietary treatment at each sampling time. The pooled samples of plasma were kept at -70C pending amino acid analysis.
Amino acid analysis The frozen plasma samples were thawed and deproteinized by the addition of 10% trichloroacetic acid (v:v), followed by vortexing for 2 s, and centrifugation at 10,000 x g for 5 min at -40°C. Lipid was extracted by vortexing with 3 ml of diethyl ether per ml of plasma followed by aspiration of the ether layer. The deproteinized, defatted plasma was filtered through a 0.22 tm polycarbonate filter. The final filtered plasma samples were kept at -70°C until amino acid analysis. Amino acid concentrations were determined by ion exchange chromatography (Beckman Amino Acid Analyzer Model 6300). Identification and quantification were accomplished using external standards. Tryptophan was not detected by the procedure. Threonine peaks were often eluted with serine and the concentrations of threonine in some samples were accordingly calculated from the area under the co-eluted peak (Beckman 1982).
181 Table 1. Composition of diets (g kg-l) Diet 1
2
3
4
5
6
7
8
9
Anchovy meal Herring meal
460.0 -
153.00 -
153.0 -
231.9
155.0
-
-
-
-
215.4
423.5
141.2
141.2
Ground wheat'
300.0
300.0
300.0
300.0
300.0
Ingredient
300.0
300.0
Gelatin
-
102.0
102.0
Corn gluten meal Soybean protein
-
79.0 60.0
79.0 60.0
124.2 90.0
83.0 60.0
115.3 83.4
118.0
118.0
105.5
-114.7
107.5
13.5 64.9 30.0 40.0
15.0 100.3 30.0 40.0
13.5 72.9 30.0 40.0
Sardine oil2 Herring oil2
93.0
Bonemeal Dextrin Binder3 Premix4
9.0 88.0 20.0 30.0
L-lysine DL-methionine L-tryptophan L-isoleucine Glycine
Protein analysis, %
-
-
39.0
-
38.0 100.0 20.0 30.0 -
39.0
-
38.0 83.5 20.0 30.0
-
7.3 4.0 1.7 3.5 -
-
40.3
38.2
102.0
-
-
-
-
-
-
300.0
117.4
117.4
-
77.1 59.1
77.1 59.1
93.0 -
118.3 -
118.3 -
16.0 107.5 20.0 40.0
38.0 88.9 20.0 40.0
38.0 75.9 20.0 40.0
-
-
-
-
-
-
-
--
300.0
-
-
7.5 4.0 1.5 -
22.0
38.5
38.7
36.2
39.0
39.6
'Autoclaved at 121 C for 1.5h; 2 Stabilized with 0.05% ethoxyquin; 3Ca lignosulfonate; 4 vitamin premix supplied per kg of diet: thiamin HCI, 60 mg; riboflavin, 100 mg; niacin, 400 mg; biotin, 5 mg; folic acid, 25 mg; pyridoxine HCl, 50 mg; cyanocobalamine 0.1 mg; D - calcium pantothenate, 200 mg; ascorbic acid, 1500 mg; choline (as choline chloride), 4000 mg; inositol, 2000 mg; vitamin K, 30 mg; vitamin A, 10,000 IU; vitamin D3 300 IU, vitamin E, 1000 IU; mineral premix supplied per kg of diet for diets 1, 2, 3, 7, 8 and 9: Mg (as MgSO 4 ) 380 mg; Mn (as MnSO4 .H20) 17 mg; Zn (as ZnO) 50 mg; Fe (as FeSO 4 .7H 2 0) 85 mg; Cu (as CuSO4 .5H2 0) 2 mg; Co (as CoC12.6H 20) 3 pig; I (as KIO3 ) 5 mg; F(as NaF) 4.5 mg; Se (as Na2 SeO 3), 0.1 mg; K (as K2SO 4) 895 mg; NaCI 1116 mg; mineral premix supplied per kg of diet for diets 4, 5 and 6: As above except Mn (as MnSO 4.5H2 0) 30 mg; Zn (as ZnO) 70 mg; KH 2PO4 2070 mg; no NaCI was added.
Experiment 2: Fishfed the diets containinggelatin in Experiment I showed markedly higher concentrationsof glycine andserinethatpeaked before the concentrationsof these particularamino acidsfrom thefishmeal diet. Experiment 2 was designedto determine whether amino acidsfrom dietary gelatin were absorbedparticularlyrapidly. A comparison was made between plasmapatterns of glycine and serine in fish fed free glycine or an equivalent amount of glycinefrom dietary gelatin. Fish in seawater drink continuously. This experiment was conducted with trout in seawater to determine whether waterconsumption would affect the rate at which amino acids appear in the systemic circulation after meal consumption.
The rainbow trout used in the experiment were acclimated to seawater and were reared in indoor 200 1tanks supplied with seawater at 32 ppt salinity and with a dissolved oxygen concentration of 8.5 ppm. Water temperature ranged from 9.0 to 11.00 C during the experiment. Lighting was according to the natural photoperiod because the aquarium room was not light-tight. Each tank contained 15 or 16 fish with mean weights between 238 and 259 g. The three experimental diets were assigned randomly to the tanks with three replicate tanks per diet. The compositions of the diets are shown in Table 1. The control diet (Diet 4) contained herring meal, corn gluten meal, and soybean protein concentrate as the principal sources of protein. Diet 5 contained these protein concentrates in the same proportions
182 as in Diet 4, but with addition of gelatin. The proportions of herring meal, corn gluten meal, and soybean protein were thus similar in Diets 4, 5, and 6 and the diets were isonitrogenous. The diets were analyzed for protein and amino acid composition (AOAC 1984). The fish were accustomed to the experimental diets by gradual substitution in the commercial diet that had been fed previously. The fish accepted the test diet after three days. Thereafter the fish were fed the experimental diets for 17 days. Feeding was once daily to satiation. The fish were sampled on day 7 for determination of concentrations of free amino acids in the plasma at 3, 9, 15, and 24h after feeding. Four fish from each dietary treatment were anaesthetized as in Experiment 1. To minimize any effects from disturbance of the fish, the fish were taken from only one of the replicate tanks for each treatment at each sampling time. The exception was the last sampling time when the fish were chosen randomly from all of the replicate tanks. Blood samples were taken as before and plasma samples from two fish from each treatment were pooled, i.e., there were two pooled samples per dietary treatment. The plasma sampleswere frozen immediately in liquid nitrogen and held at -70C pending analysis for free amino acids as described above. A second sampling was conducted on day 17. Blood was drawn from four fish per dietary treatment at 26 and 36 hours after feeding and the plasma prepared as described above. The fish were killed with an excess of anaesthetic. A sample of white muscle was excised from the area below the dorsal fin. The time required to take the muscle sample was less than 10 s. Muscle samples were frozen in liquid nitrogen and held at -70°C pending analysis for free amino acids. Muscle samples were deproteinized by homogenization of 5 g of tissue with 15 ml of 10% (v/v) trichloroacetic acid for 10 min using a Virtis No. 45 homogenizer. The homogenizing vessel was immersed in an ice bath. The homogenate was centrifuged at 20,000 x g at 4°C for 20 min. The supernatant was analyzed for amino acids as described for plasma samples.
Experiment3: Plasmaamino acidconcentrationsin response to feeding once andfive times daily with diets containing fishmeal, a mixture of protein sources, and a mixture of protein sources supplemented with free amino acids. The experimental fish were rainbow trout with initial weights ranging from 16 to 31 g. They were placed into three size groups with average weights of 25.9 ± 2.4, 21.5 ± 1.4, and 18.5 ± 1.5 g respectively. Fish from each of the three groups were assigned to six 150 1 tanks with 50 fish per tank. The water supply was dechlorinated Vancouver city water maintained at 13 to 14 0C. Water flow to each tank was 2 1per min and dissolved oxygen concentration was 8 ppm. Lighting was continuous to accommodate the feeding and sampling regimens. The experimental diets (7, 8, and 9) were formulated to essentially the same composition as Diets 1, 2 and 3 respectively, except that herring meal was used in place of anchovy meal in all three diets, and isoleucine was omitted from the supplementary crystalline amino acids added to Diet 1 (Table 1). Each diet was fed once daily and five times daily for a total of six experimental treatments. Each treatment was assigned to a tank of the small, medium, and large fish. The fish that were fed once daily were fed at 09:30h to satiation. The fish fed five times daily were given controlled amounts of feed at 3h intervals between 06:30 and 18:30 h. The quantity of feed given each day was determined by the average consumption of the fish of the respective size fed the diet once daily on the previous 2 days. The daily ration was divided equally among the five feedings. The tanks of fish that were fed once daily were located in a section of the room where they would not be disturbed by the feeding of the fish fed five times daily. The experimental period was 26 days following acceptance of the experimental diets. The total feed consumption of the fish was recorded. Twelve fish from each size group were killed with an excess of anaesthetic at the beginning of the experiment and frozen for subsequent analysis. At the conclusion of the experiment blood samples were taken for determination of plasma amino acids as described previously. Five fish were killed from
183 Table 2. Amino acid composition of diets (mmol 16g- N) Diet 1
2
3
4
5
6
27.6 24.3 27.4 50.3 43.8 15.4 4.2 23.6 14.3 :31.9 :36.7 60.6 61.6 91.0 '75.9 :39.1 35.2
30.4 27.0 22.1 51.1 30.1 10.1 3.7 21.8 12.7 24.3 29.9 67.3 58.6 96.5 126.5 63.4 36.2
28.7 23.4 28.2 49.6 40.4 15.4 3.3 21.8 10.5 24.3 29.0 67.3 57.1 95.8 121.2 63.4 35.2
53.4 26.1 30.5 64.8 36.3 14.7 5.8 29.7 18.8 30.2 40.1 65.1 62.4 119.7 67.9 53.0 43.8
47.6 24.3 24.4 57.2 32.1 11.4 4.6 23.6 13.2 26.9 32.4 74.1 62.4 110.2 138.5 73.0 40.9
47.6 30.6 25.9 64.8 36.3 13.4 5.8 29.1 18.2 28.5 37.6 58.4 57.1 108.8 141.2 46.9 40.9
299.5 363.4
263.2 448.5
274.6 440.1
350.4 411.9
297.8 499.0
0.82
0.59
0.62
0.85
0.60
Amino acid Arg His Isl Leu Lys Met Cys Phe Tyr Thr Val Ala Asp Glu Gly Pro Ser EAA NEAA EAA/NEAA
7
8
9
36.7 19.8 29.0 54.1 40.4 16.8 4.2 22.4 35.3 34.4 39.3 61.7 65.4 99.2 78.6 42.6 39.0
32.7 16.2 22.1 50.3 27.4 10.7 3.7 20.6 30.9 25.2 29.9 71.8 60.9 99.2 134.5 69.5 37.1
36.2 19.8 22.1 51.8 39.0 16.8 3.3 20.6 29.8 25.2 29.9 70.7 60.9 99.9 133.2 68.6 37.1
337.8 453.3
332.3 386.5
269.7 473.0
294.4 470.4
0.75
0.86
0.57
0.63
'Exclusive of tryptophan
each tank for determination of carcass composition (AOAC, 1984).
Results Experiment I Mean daily feed consumption over the 10-day experimental period was 4.1, 3.8, and 3.8 g per fish for Diets 1, 2, and 3 respectively. The amino acid concentrations in the diets are shown in Table 2. Diet 1, in which fishmeal was the principal source of protein, contained total essential (EAA) and total non-essential amino acids. (NEAA) in approximately equal proportions on a weight basis. On a molar basis the EAA/NEAA ratio was 0.82. The EAA were present in relatively lower concentrations in Diets 2 and 3 with EAA/ NEAA ratios of 0.59 and 0.62 respectively.
The changes in plasma concentrations of individual amino acids are depicted in Figure 1. Molar plasma concentrations of total amino acids and the molar ratios of EAA to NEAA are shown in Table 3. The NEAA that were taken into consideration in this and the subsequent experiments were those whose concentrations were determined for the diets. It should be noted that the concentrations of plasma amino acids in the fish at 24h correspond to the concentrations existing at the time of feeding. Plasma EAA in fish fed Diet 1 can be placed in three groups according to the pattern of postprandial plasma concentrations. One group included histidine, lysine, methionine + cystine, and threonine, all of which reached peak plasma concentration at 24h after feeding, following which time concentrations declined to reach minimum concentrations at either 36 or 48 h. The branched-chain amino acids comprised the second group and these amino acids attained their peak concentrations at
184 12h and then declined to minimum concentrations at 36h postprandial. The third group consisted of arginine, tyrosine, and phenylalanine. Concentrations of arginine increased slightly in the first hours after feeding then declined to a minimum value at 48h (not illustrated). Plasma concentrations of tyrosine and phenylalanine varied only slightly after feeding (not illustrated). Tryptophan was not measured.
Plasma concentrations of most EAA in fish fed Diet 2 fluctuated very little during the period of 24h after meal consumption. Supplementation of Diet 2 with lysine, methionine, and isoleucine in Diet 3 promoted increases in the plasma concentration of these amino acids which reached peaks at 12h postprandial. Supplementary isoleucine also promoted increases in the plasma concentrations of leucine and valine. Except for methionine, the plasma con-
1
Histidine
Isoleucine 0.8
-
EO . 6 I
0.o
I0A
OA
o ,*-
1
0.2 -- f
i·t · c\_,
M 3
12
24
36
48
O
72
120
3
12
24
36
4
72
120
80o
72
120
36 4 60so Hours after feeding
72
120
60so
1
Leucine
Lysine
3
0.8
C
0.o
a
I
0.4
4~~X;,
'
+:
0 6 +_
2E
I
+
. 3
12
24
so
48
so
72
120
_
_
3
.
12
.
_
24
3
l
_
_
Threonine
Methionine+Cysthe
E
._
48
0o 0.J oA
'A . 0.4 .o -.
0.4
a E 0.2
-+ 4.
I+
02
~I-
it ---
3
12
24 30 48 60 Hours after feeding
72
120
3
12
i- ·
24
z-
185 not peak until 24 h, thereafter declining to minimum values at 48h after feeding. (Concentrations of proline were too low to be determined accurately.) Fish fed Diets 2 and 3, in contrast to those fed Diet 1, exhibited peak plasma concentrations of aspartic acid, glycine, proline, and serine 12h after feeding. Plasma concentrations of alanine exhibited peaks at 12 and 36h respectively. Plasma concen-
centrations of these amino acids had declined to similar levels to those in the fish fed the unsupplemented basal diet at 24h after feeding. Concentrations of all EAA reached their lowest concentrations at 48 h, after which time concentrations increased until 120h post postprandial when the experiment was terminated. Most of the plasma NEAA in fish fed Diet 1 did
-
1
1
Alanine
Valine 0.8
0.8
.9 0.6
~~~~''4-
A-
0.6o
0.4
+
'
o 0.2 0
0.2 a. 12
3
24
3e
48
o60
72
120
3
12
24
38
48
60
72
120
--
3.5
1
Aspartic acid
Glycine
I I
3 2.6'
8
.4't 2 *8
Q o. A
I , , -* _;
gC
rF1
*
- -_ '
I,
oU
-I
IA
,.
-4
o..2
0.
Wt
3
I
i
12
0.o
~~7 24
./. 36
48
-.
i
60
72
-,y
120
3
1.4 -
4.
36
48
eo
72
120
Proline
.4.
I-;I·
0.8
£-
f.
8..--
--
0.4 0 o4 0.4 0.2
diet 1
+ diet 2 -*- diet 3
o0
0.
o
24
1
Serine
1
al
12
I*
0.2
3
l
12
)
..
24
.
.
36
48
.
0o
Hours after feeding
.
72
120
3
12
24
38
48
so
Hours after feeding
72
120
Fig. 1. Postprandial concentrations of amino acids in plasma of rainbow trout fed different diets (Table 1) in Experiment 1.
186 Table 3. Concentrations (mol ml- 1) of total plasma amino acids (TAA) (mean SEM) and the ratios of plasma essential amino acids/non-essential amino acids (EAA/NEAA) in response to different diets in Experiment 1' Hours after feeding 3
12
24
36
48
Diet 1 TAA
5.29 ± 0.15
6.06 ± 0.11
6.67 ± 0.89
4.66 ± 0.01
EAA/NEAA
0.92
1.08
0.87
0.55
TAA
6.41 ± 1.04
8.60 ± 1.16
7.20 ± 1.74
EAA/NEAA
0.37
0.28
0.30
TAA
6.84 ± 0.70
9.46 ± 0.21
6.84 ± 0.18
EAA/NEAA
0.53
0.43
0.36
72
120
3.31 ± 1.39
4.12 ± 0.76
4.69 ± 0.16
0.70
0.71
1.00
5.87 ± 0.59
4.62 ± 0.39
5.09 ± 0.20
4.72 ± 0.34
0.41
0.40
0.66
0.72
6.61 ± 0.13
4.54 ± 0.52
4.74 ± 0.14
4.83 ± 0.82
0.40
0.44
0.59
0.86
Diet 2
Diet 3
I With consideration of those amino acids listed in Table 2.
trations of phenylalanine and tyrosine were stable and are not illustrated. The concentrations of NEAA in fish fed Diets 2 and 3 were maintained at levels higher than those in fish fed Diet 1 throughout the sampling period, until 120h after meal consumption. The ratio of EAA/NEAA for fish fed Diet 1 was highest (1.08) at 12h and lowest (0.55) at 36h postprandial (Table 3). By 120h the ratio had risen again to 1.0. EAA/NEAA ratios in the plasma of fish fed Diet 2 were lower than with Diet 1 and the ratio was lowest at 12-24h postprandial and highest at 120 h. The ratios for fish fed Diet 3 were initially higher than for Diet 2, but showed a similar trend to those with Diet 2, being lowest at 24-36h and highest at 120 h.
Experiment 2 The amino acid compositions of the experimental diets are shown in Table 2. The analyses confirmed that the glycine concentrations in Diets 5 and 6 were similar. The ratio of essential to non-essential amino acids was lowest in Diet 5 which contained gelatin. Because fish were sampled at intervals throughout the experiment, feed consumption per fish was not calculated. It may be noted, however, that the appetite of fish fed Diet 6 was poor.
The fish in this experiment were reared in seawater. The total plasma concentrations of amino acids in these fish rose to higher levels than in the fish in Experiments 1 and 3 in freshwater (Table 4). This observation is in agreement with the finding of Kaushik and Luquet (1979) who compared the free amino acid concentrations in the blood of rainbow trout in fresh- and seawater. This aspect of the data will not be further discussed, however, because of the confounding effects of water temperature and fish size. The profiles for changes in the plasma concentrations of total essential, total non-essential amino acids, alanine, glycine, proline and serine in the fish fed the respective diets are illustrated in Figure 2. The differences in the profiles induced by diet were greater in the case of the non-essential amino acids and reflected the differences in the relative concentrations of these amino acids in the diets fed to the respective groups of fish. Measured at 7 days, plasma glycine concentration peaked at 9h post prandial in the fish fed the gelatin-supplemented diet and at 15h in fish fed the glycine-supplemented diet. The total amount of glycine under the respective response lines covering the 24h period after meal consumption was, however, greater in the case of the fish given the diet containing gelatin. The concentrations of free amino acids in the plasma of blood samples collected 26 and 36h after
187 Table 4. Concentrations (mol ml- ) of total plasma amino acids (TAA) (mean acids/non-essential amino acids (EAA/NEAA) at day 7 in Experiment 21
SEM) and the ratios of plasma essential amino
Hours postprandial 3
9
15
24
6.65 + 0.10 1.48
8.39 + 0.34 1.80
6.23 2.09
0.56
11.20 0.60
9.15 0.63
0.53
12.0 + 0.08 0.80
8.66 1.24
2.58
Diet 4 TAA EAA/NEAA
5.96 1.61
Diet 5 TAA EAA/NEAA
7.34 ± 0.59 0.51
11.40 + 1.44 0.60
Diet 6 TAA EAA/NEAA
6.36 ± 1.43 0.90
6.42 + 0.26 0.74
0.45
'With consideration of those amino acids listed in Table 2.
feeding the experimental diets for 17 days are shown in Table 5. The total concentration of free amino acids was higher in the muscle, umol gwet weight, than in the plasma, with much of the difference accounted for by the high concentrations of glycine and histidine in the muscle. In contrast to other amino acids, the branched chain amino acids were present at higher concentrations in plasma than in muscle. Plasma concentrations of the total essential and non-essential amino acids declined to a greater extent in plasma than in muscle during the time between 26 and 36h after meal ingestion. Measured at day 17 of the experiment, both muscle and plasma concentrations of glycine at 26 and 36h were distinctly higher in the fish that had been fed Diet 6 that was supplemented with free glycine than in fish fed the Diet 5 which contained the same concentration of glycine supplied to the diet by gelatin.
Experiment 3 The amino acid composition of the diets (7, 8, and 9) in this experiment and the EAA/NEAA ratios are given in Table 2. Feed consumption, carcass concentrations of protein and lipid at the end of the experiment, productive protein values, and lipid gain per fish are shown in Table 6. The mean carcass protein and lipid concentrations were higher
and lower respectively, in the fish that were fed five times daily compared with fish fed only once each day. The mean PPV was slightly, although significantly, higher in the fish fed five times daily. The gain in carcass lipid per fish was, on the other hand, higher in the fish fed only once daily. Plasma concentrations of total EAA, total NEAA, lysine, methionine + cystine, and isoleucine are graphed in Figure 3. Fish fed at three h intervals from 6:30 to 18:30h showed little fluctuation in concentrations of plasma amino acids compared with the fish fed only once daily. The plasma concentrations of amino acids in the fish fed five times daily did, however, reflect the differences in the amino acid composition of the respective diets. The fish fed Diet 1, which contained herring meal as the principal source of protein showed marked fluctuations in the plasma concentrations of branched-chain amino acids when the fish were fed once daily but not when feeding was five times daily.
Discussion The response of plasma concentrations of amino acids to provision of a single meal daily varied depending upon the sources of amino acids in the diet. Diets 1 and 7 contained fishmeals as the principal
188 2
4
1.2
Alanine
*, 2
09 E 8' 0.8
A-.
+
Glycine
4-
:
2
/}
--·--------
*
0 0 1
5 0.2 la 3
9
24
16
3
1A
1.6
Proline
E 1.2
*
1.4
4,
1.2
1
o
24
16
9
c~~~~~
. - . 1
4-
.6
Serine
0.8 '.4
0I6 0
8
0A
J
0.2
O.6
*')
4__ _' ----- --
0. 3
9
0.4 0.2 24
15
3
14
9
14
24
Total non-essential amino acids
Total essential amino acids l
16
12
12
-I 10 C
-B-
det 4
+
diet 6
-+
o
in _s,
1~ -------IE
B0 E
L S'r_ -4-
+';------
10
diet
.~~~~~~~~~~~4
4
4
*
-
-
t-
-* .4'--~ _
2
V. 3
9 16 Hours after feeding
24
3
9 16 Hours after feeding
24
Fig. 2. Postprandial concentrations of amino acids in plasma of rainbow trout fed different diets (Table 1) in Experiment 2.
source of protein (the amount of wheat protein was constant in all diets in the experiments) and were characterized by the high proportion of EAA to NEAA. A certain amount of variation in the rates of digestion and amino acid absorption may also be expected among these dietsbecause of differences from one sample of fishmeal to another in the proportions of different proteins in the raw materi-
al and in digestibility depending upon processing conditions. The remaining diets contained a greater variety of protein sources and the EAA/NEAA ratios in these diets depended upon whether or not gelatin was included in the diets and whether or not the diets were supplemented with amino acids. Differences in the plasma concentrations of total amino acids in fish fed the various diets resulted, in
189 Table 5. Total amino acid concentrations (TAA) in plasma (plmol ml- ) and muscle (Imol g- ) (mean ± SEM) at day 17 and the ratios of plasma and muscle essential/non-essential amino acids (EAA/NEAA) in Experiment 21 Hours postprandial 26
36
26
36
Plasma
Muscle
Diet 4 TAA EAA/NEAA
5.08 1.73
0.55
3.53 + 0.08 0.95
35.75 0.30
2.63
34.32 0.33
1.86
Diet 5 TAA EAA/NEAA
8.36 ± 0.66 0.71
4.12 ± 0.76 0.86
32.38 ± 1.82 .35
32.23 0.38
1.48
Diet 6 TAA EAA/NEAA
7.30 ± 1.11 0.80
4.30 ± 0.83 1.08
41.49 ± 4.13 0.26
40.28 ± 2.22 0.29
i With consideration of those amino acids listed in Table 2. Table 6. Protein utilization in rainbow trout fed once and five times daily Feeding regime
Diet
Feed consumption (g fish- d-')
Carcass protein (% dry wt.)
Carcass lipid (% dry wt.)
PPV*
Lipid gain (g fish- )
Once daily
7 8 9 mean
1.10 1.08 1.09
50.4 50.6 50.5 50.5
36.8 35.4 35.8 36.0
0.474 0.367 0.351 0.397
3.64 2.91 2.84 3.13
Five times daily
7 8 9 mean
1.08 1.06 1.06
51.3 51.2 52.1 51.5**
35.7 32.9 32.0 33.5**
0.470 0.377 0.377 0.408**
3.35 2.57 2.35 2.76**
* Productive protein value:gain in body N/N intake; **these values are significantly different than the corresponding values for the fish fed once daily, Tukey HSD test (Zar 1984).
considerable measure, from differences in the concentrations of plasma NEAA. This effect was most pronounced in comparison of the "steady state" plasma amino acid profiles in fish fed the different diets in Experiment 3 (Fig. 3). Dietary differences remained evident in the plasma amino acid profiles of fish, presumably in the post-absorptive state, up to 72 hours after having been fed diets containing different protein sources in Experiment 1 (Fig. 1, Table 3). Appetite suppression as a result of high plasma concentrations of amino acids has been directly
demonstrated in rats (Anderson et al. 1969) and may be deduced from the results of studies with chicks (March and Walker 1970). A reciprocal relationship in human subjects has been reported between serum amino acid concentrations and appetite (Mellinkoff et al. 1956). The use of alternate (to fishmeal) sources of protein in practical diets for salmonid fish has had limited success to date (Cho et al. 1976; Spinelli et al. 1979; March et al. 1985; Hajen et al. 1993). A frequent observation has been a reduction in appetite when other sources of protein have been included in diets. The results of all
190 12
Total essential amino acids
E
1.
-
diet 7
+
6:-
I 0
Total non-essential amino acids
10
A\
diet 8
+
4
-....
- *- diet 9 6
d
.
a
+
4
-- _-__-
+
1
4
fed once daily
3
6
9
12
fed five times daily 15
3
6
9
12
fed once daily 3
16
6
9
1
Lysine
-l
2
fed five time daily
12
16
3
6
9
12
16
Methionine*Cystine
0.6
fed once dally
fed five times daily
fed five times daily
o0.
0.8 I
l
fed once daily
-... -.
*
OA --4
0.2
-
-
I
-
l 4-
I
3
6
9
12
16
3
6
9
12
16
3
6
9
12
1
3
Hours after feeding
90
12
16
1
Isoleucine
i-
fed once daily
fed five times daily
.OA
aa
OA
0
O
I
AA
A
3
6
9
l
12
II&Jl
16
3
e
Hours after feeding
9
12
16
Fig. 3. Postprandial concentrations of amino acids in plasma of rainbow trout fed different diets (Table 1) in Experiment 3.
three experiments in the present study suggest that a low dietary ratio of EAA/NEAA, leading to similarly low EAA/NEAA ratios and high concentrations of total amino acids in the plasma, was partly responsible for the poor appetite of fish consuming diets containing high proportions of protein from sources that are poorer in essential amino acids than fishmeal.
When the dietary concentrations of EAA were increased by supplementation with free amino acids, the resultant increase in plasma concentrations of EAA, following consumption of the supplemented diet, was followed by a decrease in plasma EAA in advance of the decrease in plasma EAA in fish that had received the EAA in protein form (Fig. 1). The length of time required before an
191
elevated concentration of plasma amino acid declined depended upon the peak concentration reached. In the case of elevated plasma amino acids entering plasma from protein digestion, ingress of amino acids continues over a longer period of time thereby extending the period over which the amino acids are available for protein synthesis. In Experiment 1, Diet 3 was supplemented with methionine, lysine, isoleucine and tryptophan to the levels present in Diet 1. Tryptophan was added as one of the supplementary amino acids to Diet 3 because it was considered to be deficient on the basis of tabled analytical values. It is not, however, discussed here because it was not detected by the analytical method. The plasma concentrations of the other three supplementary amino acids in fish fed Diet 3 were approximately double those in fish fed Diet 1, indicating that the supplementary amino acids were effectively absorbed.Plasma methionine in fish fed Diet 3 was maintained at a level higher than in fish fed Diets 1 and 2 until 36h after feeding. Although the concentrations of plasma lysine in fish fed Diet 3 dropped abruptly at 24h after feeding, the fact that plasma lysine concentration was maintained at a higher level than in fish fed diet 2 until 48h aftr feeding indicates the efficacy of supplementary lysine as a supplement to lysine-deficient diets. The patterns of change in the concentrations of plasma branched-chain amino acids are interesting. Not only was the level of plasma isoleucine in fish fed the diet supplemented with free isoleucine (Diet 3) elevated over that occurring with the unsupplemented diet, but the concentrations of leucine and valine were also increased. Plasma concentrations of leucine and valine have similarly been reported to increase in channel catfish when the diet was supplemented with free isoleucine (Wilson et al. 1980). Evidence of interaction among these amino acids based on growth data was observed by Chance et al. (1964) in chinook salmon, and by Wilson et al. 1980 in channel catfish. Choo et al.
(1991), on the other hand, found no changes in the plasma concentrations of isoleucine and valine in rainbow trout fed diets containing excess leucine. There have been studies, both in fish and in other species, showing that amino acid oxidation is in-
fluenced by the concentration of amino acids in the plasma and that an excess of amino acids not used for protein synthesis will be oxidized rapidly (Brett and Zala 1975; Walton et al. 1982; Harper 1983; Young et al. 1985; Kim et al. 1992). Thebault
(1985), for example, found in sea-bass that the level of plasma methionine in response to dietary supplementation reached a peak and then returned to the pre-feeding level sooner than was the case with other essential amino acids. He concluded, therefore, that supplementary free methionine was absorbed and degraded faster than methionine originating from dietary protein. The continued elevation of plasma methionine and lysine until 36h after feeding in the present study is not necessarily contradictory to the above observation. The rate of disappearance from the plasma may be the same as that from amino acids absorbed from dietary protein. In fact, there is no reason to suppose that, once absorbed, utilization of an amino acid would be affected by its source. It is possible that the enzymes responsible for oxidation of these amino acids were overloaded with substrate (Harper et al. 1970) so that high plasma concentrations of lysine and methionine were maintained for relatively long periods. The ratio of EAA to NEAA, in either the diets or the plasma, has been shown to be important in relation to protein quality and utilization in mammals (Young and Zamora 1968; Hartog and Pol 1972; Fujita et al. 1981; Mercer et al. 1989). During the
absorptive period in Experiment 1, fish fed Diet 1 had a plasma ratio of 1, whereas plasma ratios of EAA to NEAA in fish fed Diets 2 and 3 were considerably lower, reflecting the amino acid composition of the diets. The increase in the concentrations of most essential amino acids in the plasma by 120h after feeding indicated input of free amino acids from catabolism of tissue protein. The increased ratios of plasma EAA/NEAA in the fish from the different dietary treatment groups at 120h confirmed the contribution of amino acids from the catabolism of the body protein. Another significant phenomenon regarding the changes in the plasma concentrations of histidine, lysine, and alanine in fish fed Diets 2 and 3 was observed in Experiment 1. The concentrations of these
192 amino acids exhibited similar patterns of change in that plasma concentrations either remained constant or peaked during 3- 12h postprandial and then rose at 36 h. The surges in the concentrations at 36h were interpreted as the result of delayed absorption of these amino acids following slow digestion of some of the proteins in the mixture of protein sources employed in these two diets. As with other studies in rainbow trout (Nose 1972; Gras et al. 1982; Walton and Wilson 1986), glycine was found in highest concentration among plasma non-essential amino acids. In all three experiments, fish that were fed the diets containing either gelatin or free glycine had plasma concentrations of glycine that rose to as high as three times the concentration of glycine found in fish fed the control diets. Plasma concentrations of other NEAAs were also higher in fish fed diets containing a mixture of protein sources other than fishmeal, in keeping with the higher concentrations of NEAAs in the former. Dietary concentrations of serine were similar within the three experiments. The higher concentrations of plasma serine in fish fed the mixed protein diets are therefore assumed to have derived from plasma glycine since glycine and serine are readily interconvertible (Bender 1985; Schepartz 1973; Walton and Cowey 1981, 1982). Dietary supplementation with crystalline essential amino acids accelerated the increase in the plasma concentrations of those particular amino acids following meal consumption. Diets that contained gelatin (Diets 2 and 3 in Experiment 1, Diet 5 in Experiment 2, and Diets 8 and 9 in Experiment 3) all resulted in rapid uptake of glycine as well as of some other NEAA. This fact is evidence for the rapid digestion of gelatin yielding the component amino acids for absorption in advance of amino acids from other protein sources in the diet. What is particularly interesting in this regard is the fact that the appearance of glycine in th systemic circulation, when glycine was added in free form to Diet 6 in Experiment 2, was slower than from gelatin in Diet 5 (Figure 2). It may be inferred from the rapid increases in other plasma amino acids following consumption of Diet 2 that dietary gelatin was rapidly hydrolyzed releasing the amino acids for absorption. A possibly related observation is that
of Bog6 et al. (1981), who reported that glycine uptake by rainbow trout was more rapid from glycylglycine than from an equivalent amount of free glycine, especially at high dietary concentrations. A similar phenomenon has been observed in mammals (Matthews 1973). The high plasma concentration of glycine coupled with the poorer feed consumption noted for the diets that were high in glycine, whether from gelatin or free glycine, are of interest. These data confirm the findings of Hughes (1985) that both lake trout (Salvelinus namaycush) and rainbow trout grew more slowly and had poorer feed efficiency when glycine was substituted for glutamic acid in the diet as a source of NEAA. The rapid increases in plasma serine and alanine in fish fed Diet 3 over those that occurred with Diet 1 cannot be attributed to the dietary concentration of these amino acids since the only dietary variable was supplementary crystalline glycine. Plasma proline concentration, on the other hand, showed an increase following consumption of Diet 2, in accord with the proline concentration in that diet, which did not occur with Diet 3 in which glycine rather than gelatin was the supplement. It is assumed, therefore, that the increases in serine and alanine concentrations derived from the elevated concentrations of plasma glycine. There were differences between the concentrations of individual amino acids present in free form between plasma and muscle. In plasma, the branched-chain amino acids and glycine were respectively the predominant essential and nonessential amino acids. In muscle, histidine characterized the pool of essential amino acids, constituting 15 to 24% of the total amino acids, and glycine dominated the nonessential amino acid pools, representing 46 to 60% of the total amino acids. The concentration in the plasma of an amino acid at any given time represents the balance between the rates of ingress and removal of that amino acid, whether for protein synthesis or oxidation. Consumption of one meal per day would therefore be expected to result in marked fluctuations in plasma amino acid concentrations. Consumption of the same daily ration provided in a number of smaller meals given at intervals during the day would be expected to result in a more steady state of plasma
193 concentrations of amino acids. This supposition was borne out by the data of Experiment 3. A consequence of the steady state was the higher PPV obtained in response to more frequent feeding. The more efficient use of plasma amino acids for protein synthesis was further exemplified by the higher proportion of protein to lipid in the body composition of fish fed five meals daily compared with fish given the equivalent daily ration as a single meal. A similar finding regarding the effects of feeding a similar ration in the form of one or five meals daily has been made with rats (Friend 1967). Rats fed once daily contained a higher proportion of fat in their bodies and demonstrated greater retention of absorbed nitrogen. Feeding frequency of cultured fish merits further study for its practical consequences for body fat accumulation and for efficiency of protein utilization.
Acknowledgements Financial assistance from the Science Council of British Columbia for support of the research is acknowledged. The authors thank Carol MacMillan for technical assistance during the study and in the preparation of the manuscript. References cited Aoe, H., Masuda, I., Abe, I., Saito, T., Toyoda, T. and Kitamura, S. 1970. Nutrition of protein in young carp. I. Nutritive value of free amino acids. Bull. Jap. Soc. Sci. Fish. 36: 407-413. Anderson, H.L., Benevenga, N.J. and Harper, A.E. 1969. Effect of prior high protein intake on food intake, serine dehydratase activity and plasma amino acids of rats fed amino acid-imbalanced diets. J. Nutr. 97: 463-474. AOAC 1984. Official Methods of Analysis. Association of Official Analytical Chemists, Washington, D.C. Beckman 1982. The System 6300 High Performance Amino Acid Analyzer: Instruction Manual. Beckman Instruments Inc., USA. Bender, D.A. 1985. Amino Acid Metabolism. John Wiley & Sons Ltd., Chichester. Bog6, G., Rigal, A. and Per6s, G. 1981. Rates of in vivo intestinal absorption of glycine and glycylglycine by rainbow trout (Salmo gairdneri R). Comp. Biochem. Physiol. 69A: 455459.
Brett, J.R. and Zala, C.A. 1975. Daily pattern of nitrogen excretion and oxygen consumption of sockeye salmon (Oncorhynchus nerka) under controlled conditions. J. Fish. Res. B. Can. 32: 2479-2486. Chance, R.E., Mertz, E.T. and Halver, J.E. 1964. Isoleucine, leucine, valine and phenylalanine requirements of chinook salmon and interrelationship between isoleucine and leucine for growth. J. Nutr. 83: 177-185. Cho, C.V., Slinger, S.J. and Bayley, H.S. 1976. Influence of level and type of dietary protein and of level of feeding on feed utilization by rainbow trout. J. Nutr. 106: 1547-1556. Choo, P., Smith, T.K., Cho, C.Y. and Ferguson, H.W. 1991. Dietary excesses of leucine influence growth and body composition of rainbow trout. J. Nutr. 121: 1932-1939. Cowey, C.B. and Walton, M.J. 1988. Studies on the uptake of ('4C) amino acids derived from both dietary (14C) protein and dietary (14C) amino acids by rainbow trout, Salmo gairdneri Richardson. J. Fish Biol. 33: 293-305. Cowey, C.B. and Walton, M.J. 1989. Intermediary metabolism. In: Fish Nutrition, pp. 260-330. Edited by J.E. Halver. Academic Press, San Diego. Friend, D.W. 1967. Weight gains, nitrogen metabolism, and body composition of rats fed one or five meals daily. Can. J. Physiol. Pharm. 45: 367-373. Fujita, Y., Yamamoto, T., Rikimaru, T., Ebisawa, H. and Indue, G. 1981. Effect of quality and quantity of dietary protein on free amino acids in plasma and tissues of adult rats. J. Nutr. Sci. Vitaminol. 27: 129-147. Gras, J ., Gudefin, Y., Chagny, F. and Perrier, H. 1982. Free amino acids and ninhydrin-positive substances in fish-II. Cardio-respiratory system: plasma, erythrocytes, heart and gills of the rainbow trout (Salmo gairdneri Richardson). Comp. Biochem. Physiol. 73B: 845-847. Hajen, W.E., Higgs, D.A., Beames, R.M. and Dosanjh, B.S. 1993. Digestibility of various feedstuffs by post-juvenile chinook salmon. (Oncorhynchus tshawytsha) in sea water. 2. Measurement of digestibility. Aquaculture 112: 333-348. Harper, A.E. 1983. Some recent development in the study of amino acid metabolism. Proc. Nutr. Soc. 42: 437-449. Harper, A.E., Benevenga, N.J. and Wohlheuter, R.M. 1970. Effects of ingestion of disproportionate amounts of amino acids. Physiol. Rev. 50: 428-558. Hartog, C.D. and Pol, G. 1972. D. Assays based on measurements of plasma-free amino acids. In Protein and Amino Acid Functions. pp. 353-361. Edited by E.J. Bigwood. Pergamon Press, New York. Hughes, S.G. 1985. Evaluation of glutamic acid and glycine as sources of non-essential amino acids for lake trout (Salvelinus namaycush) and rainbow trout (Salmo gairdneri). Comp. Biochem. Physiol. 81A: 669-671. Kaushik, S.J. and Luquet, P. 1979. Influence of dietary amino acid patterns in the free amino acid contents of blood and muscle of rainbow trout (Salmo gairdneri R.) Comp. Biochem. Physiol. 64B: 175-180. Kim, K.I., Grimshaw, T.W., Kayes, T.B. and Amundson, C.H. 1992. Effect of fasting or feeding diets containing different
194 levels of protein or amino acids on the activities of the liver amino acid-degrading enzymes and amino acid oxidation in rainbow trout (Oncorhynchus mykiss). Aquaculture 107: 89-105. Marcha, B.E., MacMillan, C. and Ming, F.W. 1985. Techniques for evaluation of dietary protein quality for rainbow trout (Salmo gairdneri). Aquaculture 47: 275-292. Marcha, B.E. and Walker, R.O. 1970. The effect of shifts in dietary amino acid pattern on feed consumption in the chick. Can. J. Physiol. Pharm. 48: 265-268. Matthews, D.M. 1973. Protein absorption. In The Gut and Nutrition. Proc. 12th Symp. Group of European Nutritionists, Cambridge. pp. 28-41. Edited by J.C. Somogyi and I. MacDonald. Mellinkoff, S.M., Frankland, M., Boyle, D. and Greipel, M. 1956. Relationship between serum amino acid concentration and fluctuations in appetite. J. Appl. Physiol. 8: 535-538. Mercer, L.P., Dodds, S.J. and Smith, D.I. 1989. Dispensable, indispensable, and conditionally indispensable amino acid ratios in the diet. In Absorption and Utilization of Amino Acids. pp. 1-14. Edited by M. Friedman. CRC Press Inc., Boca Raton. Nose, T. 1972. Changes in pattern of free amino acids in rainbow trout after feeding. Bull. Freshw. Fish. Res. Lab. 22: 137-144. Robinson, E.H. 1992. Use of supplemental lysine in catfish feeds. Aquaculture Magazine. May/June: 94-96. Schepartz, B. 1973. Regulation of Amino Acid Metabolism in Mammals. Saunder, Philadelphia. Spinelli, J., Mahnken, C. and Steinberg, M. 1979. Alternate sources of proteins for fishmeal in salmonid diets. Proc. World Symp. in Finfish Nutrition and Fishfeed Technology. 2: 131-142. Thebault, H. 1985. Plasma essential amino acid changes in seabass (Dicentrarchuslabrax) after feeding diets deficient and supplemented in L-methionine. Comp. Biochem. Physiol. 82A: 233-237. Walton, M.J. and Cowey, C.B. 1981. Distribution and some
kinetic properties of serine catabolizing enzymes in rainbow trout Salmo gairdneri.Comp. Biochem. Physiol. 68B: 147150. Walton, M.J. and Cowey, C.B. 1982. Aspects of intermediary metabolism in salmonid fish. Comp. Biochem. Physiol. 73B: 59-79. Walton, M.J., Cowey, C.B. and Adron, J.W. 1982. Methionine metabolism in rainbow trout fed diets of differing methionine and cystine content. J. Nutr. 112: 1525-1535. Walton, M.J. and Wilson, R.P. 1986. Postprandial changes in plasma and liver free amino acids of rainbow trout fed diets containing casein. Aquaculture 51: 105-115. Walton, M.J., Adron, J.W., Coloso, R.M. and Cowey, C.B. 1986. Dietary requirements of rainbow trout for tryptophan, lysine and arginine as determined by growth and biochemical measurements. Fish Physiol. Biochem. 2: 161-169. Wilson, R.P., Poe, W.E. and Robinson, E.H. 1980. Leucine, isoleucine, valine and histidine requirements of fingerling channel catfish. J. Nutr. 110: 627-633. Wilson, R.P., Allen, O.W., Robinson, E.H. and Poe, W.E. 1978. Tryptophan and threonine requirements of fingerling channel catfish. J. Nutr. 108: 1595-1599. Yamada, S., Simpson, K.L., Tanaka, Y. and Katayama, T. 1981. Plasma amino acid changes in rainbow trout Salmo gairdneri force-fed casein and a corresponding amino acid mixture. Bull. Jap. Soc. Sci. Fish. 47: 1035-1040. Young, V.R. and Zamora, J. 1968. Effects of altering the proportion of essential to nonessential amino acids on growth and plasma amino acid levels in the rat. J. Nutr. 96: 21-27, Young, V.R., Meredith, C., Hoerr, R., Bier, D.M. and Matthews, D.E. 1985. Amino acid kinetics in relation to protein and amino acid requirement: The primary importance of amino acid oxidation. In Substrate and Energy Metabolism in Man. pp. 119-134. Edited by J.S. Garrow and D. Holliday. John Libbey, London. Zar, J.H. 1984. Biostatistical Analysis. Prentice Hall, Englewood Cliffs.
Dynamics of plasma free amino acids in rainbow trout (Oncorhynchus mykiss under variety of dietary conditions Chutima Tantikittil and B.E. March Department of Animal Science, University of British Columbia, Vancouver BC, Canada V6T 1Z4 Accepted: November 24, 1994 Keywords: rainbow trout, plasma amino acids, dietary protein, dietary free amino acids, feeding frequency
Abstract Plasma amino acid profiles in the systemic circulation of rainbow trout acclimated to diets containing different protein sources, with and without supplementation with free amino acids, were monitored for up to 120 hours after consumption of the diets. Plasma concentrations of lysine, methionine, and isoleucine increased more rapidly after feeding a diet supplemented with these amino acids in free form and, depending upon the increase in plasma concentration, remained at concentrations above those in fish fed the basal diet for at least 24h after feeding. Dietary supplementation with isoleucine increased plasma concentrations of leucine and valine as well as of isoleucine. Maximum plasma concentrations for most amino acids were attained between 12 and 24h postprandial. Dietary inclusion of gelatin caused more rapid elevations in plasma glycine and serine than did free glycine. Feeding at three hour intervals resulted in stable plasma concentrations of individual amino acids in contrast to the fluctuations occurring when fish were fed once daily. Frequent feeding favoured a higher proportion of protein to lipid in body composition.
Introduction There are reports that dietary supplements of crystalline amino acids are not utilized efficiently by fish fed only once daily because they may be absorbed and oxidized before amino acids derived from digestion of the dietary protein are absorbed (Cowey and Walton, 1988; Robinson, 1992). Studies with several species of fish have shown that growth rate and feed efficiency are reduced in fish fed diets supplemented with amino acids (Aoe et al. 1970; Wilson et al. 1978; Walton et al. 1986). Rainbow trout fed amino acid based diets displayed more rapid appearance and higher concentrations 1
of free amino acids in the plasma (Yamada et al. 1981). Similarly, when a mixture of proteins is consumed, not all of the amino acids may be available for absorption at the same time because proteins vary in their rates of digestion. In either case the concentrations of amino acids simultaneously available to body tissues for protein synthesis may not optimal. The implications of the above in the formulation of practical diets for cultured fish are obvious. The present experiments were designed to investigate the degree to which amino acid profiles in the systemic circulation vary following meal consumption depending upon dietary protein source, and
Present address: Department of Aquatic Science, Faculty of Natural Resources, Prince of Songkla University, Hat Yai, Thailand Correspondence to: B.E. March, Department of Animal Science, University of British Columbia, Vancouver BC, Canada V6T IZ4.
180 whether amino acids are supplied entirely from intact protein or whether the diet is supplemented with free amino acids.
Materials and methods Experiment 1: Plasma amino acid patterns, measured over 120h post prandial, were compared in response to diets containing (1) fishmeal as the primary source of protein, (2) a mixture of protein sources with presumablydifferent ratesof digestion and deficient in some essentialamino acids, and (3) the above mixture ofprotein sources supplemented with free amino acids Rainbow trout used in the experiment had a mean weight of 180 ± 16 g (SD). The fish were distributed into fifteen indoor 150 1tanks with 10 fish per tank. The water supply was dechlorinated Vancouver city water. Water temperature was maintained at 13-14°C. Water flow to each tank was 2 1 per min. Photoperiod was 14:10 L:D. The compositions of the three diets formulated for the experiment are shown in Table 1. Diet 1 contained anchovy meal as the principal source of protein. Diet 2 contained a mixture of anchovy meal, soybean oil meal, gelatin, and corn gluten meal. Diet 3 contained the same proportion of protein from the respective sources as were used in Diet 2, but was supplemented with lysine, methionine, isoleucine, and tryptophan to provide dietary concentrations equivalent to those calculated to be present in Diet 1. The diets were analyzed for dry matter, protein and lipid, and were subjected to acid hydrolysis for determination of amino acid concentrations (AOAC 1984). Each experimental treatment was assigned to five replicate tanks of fish. The fish were gradually accustomed to the experimental diets over a five-day period by mixing increasing amounts of the respective diets with the commercial diet that had been fed to the fish during the pre-experimental period. The fish were hand-fed once daily at 09:30h until they appeared satiated. Record was kept of feed consumption. The experimental period was 10 days following acceptance of the diets.
Blood samplingprocedure On the 10th day of the experiment blood samples were drawn from the fish at 3, 12, 24, 36, 48, 72, and 120h after feeding. Six fish were taken from one replicate tank for each dietary treatment at the first four sampling times. Fish for the final sampling were taken randomly from all of the tanks for the respective treatments. The fish that were sampled were anaesthetized with 0.01% MS 222 and blood withdrawn from the caudal blood vessels into heparinized vacutainers. Amino acid concentrations in this and the subsequent two experiments therefore refer to blood in the systemic circulation that would be available to the tissues for protein synthesis. Following centrifugation of the blood; the plasma samples from three fish were pooled so that there were duplicate samples from each dietary treatment at each sampling time. The pooled samples of plasma were kept at -70C pending amino acid analysis.
Amino acid analysis The frozen plasma samples were thawed and deproteinized by the addition of 10% trichloroacetic acid (v:v), followed by vortexing for 2 s, and centrifugation at 10,000 x g for 5 min at -40°C. Lipid was extracted by vortexing with 3 ml of diethyl ether per ml of plasma followed by aspiration of the ether layer. The deproteinized, defatted plasma was filtered through a 0.22 tm polycarbonate filter. The final filtered plasma samples were kept at -70°C until amino acid analysis. Amino acid concentrations were determined by ion exchange chromatography (Beckman Amino Acid Analyzer Model 6300). Identification and quantification were accomplished using external standards. Tryptophan was not detected by the procedure. Threonine peaks were often eluted with serine and the concentrations of threonine in some samples were accordingly calculated from the area under the co-eluted peak (Beckman 1982).
181 Table 1. Composition of diets (g kg-l) Diet 1
2
3
4
5
6
7
8
9
Anchovy meal Herring meal
460.0 -
153.00 -
153.0 -
231.9
155.0
-
-
-
-
215.4
423.5
141.2
141.2
Ground wheat'
300.0
300.0
300.0
300.0
300.0
Ingredient
300.0
300.0
Gelatin
-
102.0
102.0
Corn gluten meal Soybean protein
-
79.0 60.0
79.0 60.0
124.2 90.0
83.0 60.0
115.3 83.4
118.0
118.0
105.5
-114.7
107.5
13.5 64.9 30.0 40.0
15.0 100.3 30.0 40.0
13.5 72.9 30.0 40.0
Sardine oil2 Herring oil2
93.0
Bonemeal Dextrin Binder3 Premix4
9.0 88.0 20.0 30.0
L-lysine DL-methionine L-tryptophan L-isoleucine Glycine
Protein analysis, %
-
-
39.0
-
38.0 100.0 20.0 30.0 -
39.0
-
38.0 83.5 20.0 30.0
-
7.3 4.0 1.7 3.5 -
-
40.3
38.2
102.0
-
-
-
-
-
-
300.0
117.4
117.4
-
77.1 59.1
77.1 59.1
93.0 -
118.3 -
118.3 -
16.0 107.5 20.0 40.0
38.0 88.9 20.0 40.0
38.0 75.9 20.0 40.0
-
-
-
-
-
-
-
--
300.0
-
-
7.5 4.0 1.5 -
22.0
38.5
38.7
36.2
39.0
39.6
'Autoclaved at 121 C for 1.5h; 2 Stabilized with 0.05% ethoxyquin; 3Ca lignosulfonate; 4 vitamin premix supplied per kg of diet: thiamin HCI, 60 mg; riboflavin, 100 mg; niacin, 400 mg; biotin, 5 mg; folic acid, 25 mg; pyridoxine HCl, 50 mg; cyanocobalamine 0.1 mg; D - calcium pantothenate, 200 mg; ascorbic acid, 1500 mg; choline (as choline chloride), 4000 mg; inositol, 2000 mg; vitamin K, 30 mg; vitamin A, 10,000 IU; vitamin D3 300 IU, vitamin E, 1000 IU; mineral premix supplied per kg of diet for diets 1, 2, 3, 7, 8 and 9: Mg (as MgSO 4 ) 380 mg; Mn (as MnSO4 .H20) 17 mg; Zn (as ZnO) 50 mg; Fe (as FeSO 4 .7H 2 0) 85 mg; Cu (as CuSO4 .5H2 0) 2 mg; Co (as CoC12.6H 20) 3 pig; I (as KIO3 ) 5 mg; F(as NaF) 4.5 mg; Se (as Na2 SeO 3), 0.1 mg; K (as K2SO 4) 895 mg; NaCI 1116 mg; mineral premix supplied per kg of diet for diets 4, 5 and 6: As above except Mn (as MnSO 4.5H2 0) 30 mg; Zn (as ZnO) 70 mg; KH 2PO4 2070 mg; no NaCI was added.
Experiment 2: Fishfed the diets containinggelatin in Experiment I showed markedly higher concentrationsof glycine andserinethatpeaked before the concentrationsof these particularamino acidsfrom thefishmeal diet. Experiment 2 was designedto determine whether amino acidsfrom dietary gelatin were absorbedparticularlyrapidly. A comparison was made between plasmapatterns of glycine and serine in fish fed free glycine or an equivalent amount of glycinefrom dietary gelatin. Fish in seawater drink continuously. This experiment was conducted with trout in seawater to determine whether waterconsumption would affect the rate at which amino acids appear in the systemic circulation after meal consumption.
The rainbow trout used in the experiment were acclimated to seawater and were reared in indoor 200 1tanks supplied with seawater at 32 ppt salinity and with a dissolved oxygen concentration of 8.5 ppm. Water temperature ranged from 9.0 to 11.00 C during the experiment. Lighting was according to the natural photoperiod because the aquarium room was not light-tight. Each tank contained 15 or 16 fish with mean weights between 238 and 259 g. The three experimental diets were assigned randomly to the tanks with three replicate tanks per diet. The compositions of the diets are shown in Table 1. The control diet (Diet 4) contained herring meal, corn gluten meal, and soybean protein concentrate as the principal sources of protein. Diet 5 contained these protein concentrates in the same proportions
182 as in Diet 4, but with addition of gelatin. The proportions of herring meal, corn gluten meal, and soybean protein were thus similar in Diets 4, 5, and 6 and the diets were isonitrogenous. The diets were analyzed for protein and amino acid composition (AOAC 1984). The fish were accustomed to the experimental diets by gradual substitution in the commercial diet that had been fed previously. The fish accepted the test diet after three days. Thereafter the fish were fed the experimental diets for 17 days. Feeding was once daily to satiation. The fish were sampled on day 7 for determination of concentrations of free amino acids in the plasma at 3, 9, 15, and 24h after feeding. Four fish from each dietary treatment were anaesthetized as in Experiment 1. To minimize any effects from disturbance of the fish, the fish were taken from only one of the replicate tanks for each treatment at each sampling time. The exception was the last sampling time when the fish were chosen randomly from all of the replicate tanks. Blood samples were taken as before and plasma samples from two fish from each treatment were pooled, i.e., there were two pooled samples per dietary treatment. The plasma sampleswere frozen immediately in liquid nitrogen and held at -70C pending analysis for free amino acids as described above. A second sampling was conducted on day 17. Blood was drawn from four fish per dietary treatment at 26 and 36 hours after feeding and the plasma prepared as described above. The fish were killed with an excess of anaesthetic. A sample of white muscle was excised from the area below the dorsal fin. The time required to take the muscle sample was less than 10 s. Muscle samples were frozen in liquid nitrogen and held at -70°C pending analysis for free amino acids. Muscle samples were deproteinized by homogenization of 5 g of tissue with 15 ml of 10% (v/v) trichloroacetic acid for 10 min using a Virtis No. 45 homogenizer. The homogenizing vessel was immersed in an ice bath. The homogenate was centrifuged at 20,000 x g at 4°C for 20 min. The supernatant was analyzed for amino acids as described for plasma samples.
Experiment3: Plasmaamino acidconcentrationsin response to feeding once andfive times daily with diets containing fishmeal, a mixture of protein sources, and a mixture of protein sources supplemented with free amino acids. The experimental fish were rainbow trout with initial weights ranging from 16 to 31 g. They were placed into three size groups with average weights of 25.9 ± 2.4, 21.5 ± 1.4, and 18.5 ± 1.5 g respectively. Fish from each of the three groups were assigned to six 150 1 tanks with 50 fish per tank. The water supply was dechlorinated Vancouver city water maintained at 13 to 14 0C. Water flow to each tank was 2 1per min and dissolved oxygen concentration was 8 ppm. Lighting was continuous to accommodate the feeding and sampling regimens. The experimental diets (7, 8, and 9) were formulated to essentially the same composition as Diets 1, 2 and 3 respectively, except that herring meal was used in place of anchovy meal in all three diets, and isoleucine was omitted from the supplementary crystalline amino acids added to Diet 1 (Table 1). Each diet was fed once daily and five times daily for a total of six experimental treatments. Each treatment was assigned to a tank of the small, medium, and large fish. The fish that were fed once daily were fed at 09:30h to satiation. The fish fed five times daily were given controlled amounts of feed at 3h intervals between 06:30 and 18:30 h. The quantity of feed given each day was determined by the average consumption of the fish of the respective size fed the diet once daily on the previous 2 days. The daily ration was divided equally among the five feedings. The tanks of fish that were fed once daily were located in a section of the room where they would not be disturbed by the feeding of the fish fed five times daily. The experimental period was 26 days following acceptance of the experimental diets. The total feed consumption of the fish was recorded. Twelve fish from each size group were killed with an excess of anaesthetic at the beginning of the experiment and frozen for subsequent analysis. At the conclusion of the experiment blood samples were taken for determination of plasma amino acids as described previously. Five fish were killed from
183 Table 2. Amino acid composition of diets (mmol 16g- N) Diet 1
2
3
4
5
6
27.6 24.3 27.4 50.3 43.8 15.4 4.2 23.6 14.3 :31.9 :36.7 60.6 61.6 91.0 '75.9 :39.1 35.2
30.4 27.0 22.1 51.1 30.1 10.1 3.7 21.8 12.7 24.3 29.9 67.3 58.6 96.5 126.5 63.4 36.2
28.7 23.4 28.2 49.6 40.4 15.4 3.3 21.8 10.5 24.3 29.0 67.3 57.1 95.8 121.2 63.4 35.2
53.4 26.1 30.5 64.8 36.3 14.7 5.8 29.7 18.8 30.2 40.1 65.1 62.4 119.7 67.9 53.0 43.8
47.6 24.3 24.4 57.2 32.1 11.4 4.6 23.6 13.2 26.9 32.4 74.1 62.4 110.2 138.5 73.0 40.9
47.6 30.6 25.9 64.8 36.3 13.4 5.8 29.1 18.2 28.5 37.6 58.4 57.1 108.8 141.2 46.9 40.9
299.5 363.4
263.2 448.5
274.6 440.1
350.4 411.9
297.8 499.0
0.82
0.59
0.62
0.85
0.60
Amino acid Arg His Isl Leu Lys Met Cys Phe Tyr Thr Val Ala Asp Glu Gly Pro Ser EAA NEAA EAA/NEAA
7
8
9
36.7 19.8 29.0 54.1 40.4 16.8 4.2 22.4 35.3 34.4 39.3 61.7 65.4 99.2 78.6 42.6 39.0
32.7 16.2 22.1 50.3 27.4 10.7 3.7 20.6 30.9 25.2 29.9 71.8 60.9 99.2 134.5 69.5 37.1
36.2 19.8 22.1 51.8 39.0 16.8 3.3 20.6 29.8 25.2 29.9 70.7 60.9 99.9 133.2 68.6 37.1
337.8 453.3
332.3 386.5
269.7 473.0
294.4 470.4
0.75
0.86
0.57
0.63
'Exclusive of tryptophan
each tank for determination of carcass composition (AOAC, 1984).
Results Experiment I Mean daily feed consumption over the 10-day experimental period was 4.1, 3.8, and 3.8 g per fish for Diets 1, 2, and 3 respectively. The amino acid concentrations in the diets are shown in Table 2. Diet 1, in which fishmeal was the principal source of protein, contained total essential (EAA) and total non-essential amino acids. (NEAA) in approximately equal proportions on a weight basis. On a molar basis the EAA/NEAA ratio was 0.82. The EAA were present in relatively lower concentrations in Diets 2 and 3 with EAA/ NEAA ratios of 0.59 and 0.62 respectively.
The changes in plasma concentrations of individual amino acids are depicted in Figure 1. Molar plasma concentrations of total amino acids and the molar ratios of EAA to NEAA are shown in Table 3. The NEAA that were taken into consideration in this and the subsequent experiments were those whose concentrations were determined for the diets. It should be noted that the concentrations of plasma amino acids in the fish at 24h correspond to the concentrations existing at the time of feeding. Plasma EAA in fish fed Diet 1 can be placed in three groups according to the pattern of postprandial plasma concentrations. One group included histidine, lysine, methionine + cystine, and threonine, all of which reached peak plasma concentration at 24h after feeding, following which time concentrations declined to reach minimum concentrations at either 36 or 48 h. The branched-chain amino acids comprised the second group and these amino acids attained their peak concentrations at
184 12h and then declined to minimum concentrations at 36h postprandial. The third group consisted of arginine, tyrosine, and phenylalanine. Concentrations of arginine increased slightly in the first hours after feeding then declined to a minimum value at 48h (not illustrated). Plasma concentrations of tyrosine and phenylalanine varied only slightly after feeding (not illustrated). Tryptophan was not measured.
Plasma concentrations of most EAA in fish fed Diet 2 fluctuated very little during the period of 24h after meal consumption. Supplementation of Diet 2 with lysine, methionine, and isoleucine in Diet 3 promoted increases in the plasma concentration of these amino acids which reached peaks at 12h postprandial. Supplementary isoleucine also promoted increases in the plasma concentrations of leucine and valine. Except for methionine, the plasma con-
1
Histidine
Isoleucine 0.8
-
EO . 6 I
0.o
I0A
OA
o ,*-
1
0.2 -- f
i·t · c\_,
M 3
12
24
36
48
O
72
120
3
12
24
36
4
72
120
80o
72
120
36 4 60so Hours after feeding
72
120
60so
1
Leucine
Lysine
3
0.8
C
0.o
a
I
0.4
4~~X;,
'
+:
0 6 +_
2E
I
+
. 3
12
24
so
48
so
72
120
_
_
3
.
12
.
_
24
3
l
_
_
Threonine
Methionine+Cysthe
E
._
48
0o 0.J oA
'A . 0.4 .o -.
0.4
a E 0.2
-+ 4.
I+
02
~I-
it ---
3
12
24 30 48 60 Hours after feeding
72
120
3
12
i- ·
24
z-
185 not peak until 24 h, thereafter declining to minimum values at 48h after feeding. (Concentrations of proline were too low to be determined accurately.) Fish fed Diets 2 and 3, in contrast to those fed Diet 1, exhibited peak plasma concentrations of aspartic acid, glycine, proline, and serine 12h after feeding. Plasma concentrations of alanine exhibited peaks at 12 and 36h respectively. Plasma concen-
centrations of these amino acids had declined to similar levels to those in the fish fed the unsupplemented basal diet at 24h after feeding. Concentrations of all EAA reached their lowest concentrations at 48 h, after which time concentrations increased until 120h post postprandial when the experiment was terminated. Most of the plasma NEAA in fish fed Diet 1 did
-
1
1
Alanine
Valine 0.8
0.8
.9 0.6
~~~~''4-
A-
0.6o
0.4
+
'
o 0.2 0
0.2 a. 12
3
24
3e
48
o60
72
120
3
12
24
38
48
60
72
120
--
3.5
1
Aspartic acid
Glycine
I I
3 2.6'
8
.4't 2 *8
Q o. A
I , , -* _;
gC
rF1
*
- -_ '
I,
oU
-I
IA
,.
-4
o..2
0.
Wt
3
I
i
12
0.o
~~7 24
./. 36
48
-.
i
60
72
-,y
120
3
1.4 -
4.
36
48
eo
72
120
Proline
.4.
I-;I·
0.8
£-
f.
8..--
--
0.4 0 o4 0.4 0.2
diet 1
+ diet 2 -*- diet 3
o0
0.
o
24
1
Serine
1
al
12
I*
0.2
3
l
12
)
..
24
.
.
36
48
.
0o
Hours after feeding
.
72
120
3
12
24
38
48
so
Hours after feeding
72
120
Fig. 1. Postprandial concentrations of amino acids in plasma of rainbow trout fed different diets (Table 1) in Experiment 1.
186 Table 3. Concentrations (mol ml- 1) of total plasma amino acids (TAA) (mean SEM) and the ratios of plasma essential amino acids/non-essential amino acids (EAA/NEAA) in response to different diets in Experiment 1' Hours after feeding 3
12
24
36
48
Diet 1 TAA
5.29 ± 0.15
6.06 ± 0.11
6.67 ± 0.89
4.66 ± 0.01
EAA/NEAA
0.92
1.08
0.87
0.55
TAA
6.41 ± 1.04
8.60 ± 1.16
7.20 ± 1.74
EAA/NEAA
0.37
0.28
0.30
TAA
6.84 ± 0.70
9.46 ± 0.21
6.84 ± 0.18
EAA/NEAA
0.53
0.43
0.36
72
120
3.31 ± 1.39
4.12 ± 0.76
4.69 ± 0.16
0.70
0.71
1.00
5.87 ± 0.59
4.62 ± 0.39
5.09 ± 0.20
4.72 ± 0.34
0.41
0.40
0.66
0.72
6.61 ± 0.13
4.54 ± 0.52
4.74 ± 0.14
4.83 ± 0.82
0.40
0.44
0.59
0.86
Diet 2
Diet 3
I With consideration of those amino acids listed in Table 2.
trations of phenylalanine and tyrosine were stable and are not illustrated. The concentrations of NEAA in fish fed Diets 2 and 3 were maintained at levels higher than those in fish fed Diet 1 throughout the sampling period, until 120h after meal consumption. The ratio of EAA/NEAA for fish fed Diet 1 was highest (1.08) at 12h and lowest (0.55) at 36h postprandial (Table 3). By 120h the ratio had risen again to 1.0. EAA/NEAA ratios in the plasma of fish fed Diet 2 were lower than with Diet 1 and the ratio was lowest at 12-24h postprandial and highest at 120 h. The ratios for fish fed Diet 3 were initially higher than for Diet 2, but showed a similar trend to those with Diet 2, being lowest at 24-36h and highest at 120 h.
Experiment 2 The amino acid compositions of the experimental diets are shown in Table 2. The analyses confirmed that the glycine concentrations in Diets 5 and 6 were similar. The ratio of essential to non-essential amino acids was lowest in Diet 5 which contained gelatin. Because fish were sampled at intervals throughout the experiment, feed consumption per fish was not calculated. It may be noted, however, that the appetite of fish fed Diet 6 was poor.
The fish in this experiment were reared in seawater. The total plasma concentrations of amino acids in these fish rose to higher levels than in the fish in Experiments 1 and 3 in freshwater (Table 4). This observation is in agreement with the finding of Kaushik and Luquet (1979) who compared the free amino acid concentrations in the blood of rainbow trout in fresh- and seawater. This aspect of the data will not be further discussed, however, because of the confounding effects of water temperature and fish size. The profiles for changes in the plasma concentrations of total essential, total non-essential amino acids, alanine, glycine, proline and serine in the fish fed the respective diets are illustrated in Figure 2. The differences in the profiles induced by diet were greater in the case of the non-essential amino acids and reflected the differences in the relative concentrations of these amino acids in the diets fed to the respective groups of fish. Measured at 7 days, plasma glycine concentration peaked at 9h post prandial in the fish fed the gelatin-supplemented diet and at 15h in fish fed the glycine-supplemented diet. The total amount of glycine under the respective response lines covering the 24h period after meal consumption was, however, greater in the case of the fish given the diet containing gelatin. The concentrations of free amino acids in the plasma of blood samples collected 26 and 36h after
187 Table 4. Concentrations (mol ml- ) of total plasma amino acids (TAA) (mean acids/non-essential amino acids (EAA/NEAA) at day 7 in Experiment 21
SEM) and the ratios of plasma essential amino
Hours postprandial 3
9
15
24
6.65 + 0.10 1.48
8.39 + 0.34 1.80
6.23 2.09
0.56
11.20 0.60
9.15 0.63
0.53
12.0 + 0.08 0.80
8.66 1.24
2.58
Diet 4 TAA EAA/NEAA
5.96 1.61
Diet 5 TAA EAA/NEAA
7.34 ± 0.59 0.51
11.40 + 1.44 0.60
Diet 6 TAA EAA/NEAA
6.36 ± 1.43 0.90
6.42 + 0.26 0.74
0.45
'With consideration of those amino acids listed in Table 2.
feeding the experimental diets for 17 days are shown in Table 5. The total concentration of free amino acids was higher in the muscle, umol gwet weight, than in the plasma, with much of the difference accounted for by the high concentrations of glycine and histidine in the muscle. In contrast to other amino acids, the branched chain amino acids were present at higher concentrations in plasma than in muscle. Plasma concentrations of the total essential and non-essential amino acids declined to a greater extent in plasma than in muscle during the time between 26 and 36h after meal ingestion. Measured at day 17 of the experiment, both muscle and plasma concentrations of glycine at 26 and 36h were distinctly higher in the fish that had been fed Diet 6 that was supplemented with free glycine than in fish fed the Diet 5 which contained the same concentration of glycine supplied to the diet by gelatin.
Experiment 3 The amino acid composition of the diets (7, 8, and 9) in this experiment and the EAA/NEAA ratios are given in Table 2. Feed consumption, carcass concentrations of protein and lipid at the end of the experiment, productive protein values, and lipid gain per fish are shown in Table 6. The mean carcass protein and lipid concentrations were higher
and lower respectively, in the fish that were fed five times daily compared with fish fed only once each day. The mean PPV was slightly, although significantly, higher in the fish fed five times daily. The gain in carcass lipid per fish was, on the other hand, higher in the fish fed only once daily. Plasma concentrations of total EAA, total NEAA, lysine, methionine + cystine, and isoleucine are graphed in Figure 3. Fish fed at three h intervals from 6:30 to 18:30h showed little fluctuation in concentrations of plasma amino acids compared with the fish fed only once daily. The plasma concentrations of amino acids in the fish fed five times daily did, however, reflect the differences in the amino acid composition of the respective diets. The fish fed Diet 1, which contained herring meal as the principal source of protein showed marked fluctuations in the plasma concentrations of branched-chain amino acids when the fish were fed once daily but not when feeding was five times daily.
Discussion The response of plasma concentrations of amino acids to provision of a single meal daily varied depending upon the sources of amino acids in the diet. Diets 1 and 7 contained fishmeals as the principal
188 2
4
1.2
Alanine
*, 2
09 E 8' 0.8
A-.
+
Glycine
4-
:
2
/}
--·--------
*
0 0 1
5 0.2 la 3
9
24
16
3
1A
1.6
Proline
E 1.2
*
1.4
4,
1.2
1
o
24
16
9
c~~~~~
. - . 1
4-
.6
Serine
0.8 '.4
0I6 0
8
0A
J
0.2
O.6
*')
4__ _' ----- --
0. 3
9
0.4 0.2 24
15
3
14
9
14
24
Total non-essential amino acids
Total essential amino acids l
16
12
12
-I 10 C
-B-
det 4
+
diet 6
-+
o
in _s,
1~ -------IE
B0 E
L S'r_ -4-
+';------
10
diet
.~~~~~~~~~~~4
4
4
*
-
-
t-
-* .4'--~ _
2
V. 3
9 16 Hours after feeding
24
3
9 16 Hours after feeding
24
Fig. 2. Postprandial concentrations of amino acids in plasma of rainbow trout fed different diets (Table 1) in Experiment 2.
source of protein (the amount of wheat protein was constant in all diets in the experiments) and were characterized by the high proportion of EAA to NEAA. A certain amount of variation in the rates of digestion and amino acid absorption may also be expected among these dietsbecause of differences from one sample of fishmeal to another in the proportions of different proteins in the raw materi-
al and in digestibility depending upon processing conditions. The remaining diets contained a greater variety of protein sources and the EAA/NEAA ratios in these diets depended upon whether or not gelatin was included in the diets and whether or not the diets were supplemented with amino acids. Differences in the plasma concentrations of total amino acids in fish fed the various diets resulted, in
189 Table 5. Total amino acid concentrations (TAA) in plasma (plmol ml- ) and muscle (Imol g- ) (mean ± SEM) at day 17 and the ratios of plasma and muscle essential/non-essential amino acids (EAA/NEAA) in Experiment 21 Hours postprandial 26
36
26
36
Plasma
Muscle
Diet 4 TAA EAA/NEAA
5.08 1.73
0.55
3.53 + 0.08 0.95
35.75 0.30
2.63
34.32 0.33
1.86
Diet 5 TAA EAA/NEAA
8.36 ± 0.66 0.71
4.12 ± 0.76 0.86
32.38 ± 1.82 .35
32.23 0.38
1.48
Diet 6 TAA EAA/NEAA
7.30 ± 1.11 0.80
4.30 ± 0.83 1.08
41.49 ± 4.13 0.26
40.28 ± 2.22 0.29
i With consideration of those amino acids listed in Table 2. Table 6. Protein utilization in rainbow trout fed once and five times daily Feeding regime
Diet
Feed consumption (g fish- d-')
Carcass protein (% dry wt.)
Carcass lipid (% dry wt.)
PPV*
Lipid gain (g fish- )
Once daily
7 8 9 mean
1.10 1.08 1.09
50.4 50.6 50.5 50.5
36.8 35.4 35.8 36.0
0.474 0.367 0.351 0.397
3.64 2.91 2.84 3.13
Five times daily
7 8 9 mean
1.08 1.06 1.06
51.3 51.2 52.1 51.5**
35.7 32.9 32.0 33.5**
0.470 0.377 0.377 0.408**
3.35 2.57 2.35 2.76**
* Productive protein value:gain in body N/N intake; **these values are significantly different than the corresponding values for the fish fed once daily, Tukey HSD test (Zar 1984).
considerable measure, from differences in the concentrations of plasma NEAA. This effect was most pronounced in comparison of the "steady state" plasma amino acid profiles in fish fed the different diets in Experiment 3 (Fig. 3). Dietary differences remained evident in the plasma amino acid profiles of fish, presumably in the post-absorptive state, up to 72 hours after having been fed diets containing different protein sources in Experiment 1 (Fig. 1, Table 3). Appetite suppression as a result of high plasma concentrations of amino acids has been directly
demonstrated in rats (Anderson et al. 1969) and may be deduced from the results of studies with chicks (March and Walker 1970). A reciprocal relationship in human subjects has been reported between serum amino acid concentrations and appetite (Mellinkoff et al. 1956). The use of alternate (to fishmeal) sources of protein in practical diets for salmonid fish has had limited success to date (Cho et al. 1976; Spinelli et al. 1979; March et al. 1985; Hajen et al. 1993). A frequent observation has been a reduction in appetite when other sources of protein have been included in diets. The results of all
190 12
Total essential amino acids
E
1.
-
diet 7
+
6:-
I 0
Total non-essential amino acids
10
A\
diet 8
+
4
-....
- *- diet 9 6
d
.
a
+
4
-- _-__-
+
1
4
fed once daily
3
6
9
12
fed five times daily 15
3
6
9
12
fed once daily 3
16
6
9
1
Lysine
-l
2
fed five time daily
12
16
3
6
9
12
16
Methionine*Cystine
0.6
fed once dally
fed five times daily
fed five times daily
o0.
0.8 I
l
fed once daily
-... -.
*
OA --4
0.2
-
-
I
-
l 4-
I
3
6
9
12
16
3
6
9
12
16
3
6
9
12
1
3
Hours after feeding
90
12
16
1
Isoleucine
i-
fed once daily
fed five times daily
.OA
aa
OA
0
O
I
AA
A
3
6
9
l
12
II&Jl
16
3
e
Hours after feeding
9
12
16
Fig. 3. Postprandial concentrations of amino acids in plasma of rainbow trout fed different diets (Table 1) in Experiment 3.
three experiments in the present study suggest that a low dietary ratio of EAA/NEAA, leading to similarly low EAA/NEAA ratios and high concentrations of total amino acids in the plasma, was partly responsible for the poor appetite of fish consuming diets containing high proportions of protein from sources that are poorer in essential amino acids than fishmeal.
When the dietary concentrations of EAA were increased by supplementation with free amino acids, the resultant increase in plasma concentrations of EAA, following consumption of the supplemented diet, was followed by a decrease in plasma EAA in advance of the decrease in plasma EAA in fish that had received the EAA in protein form (Fig. 1). The length of time required before an
191
elevated concentration of plasma amino acid declined depended upon the peak concentration reached. In the case of elevated plasma amino acids entering plasma from protein digestion, ingress of amino acids continues over a longer period of time thereby extending the period over which the amino acids are available for protein synthesis. In Experiment 1, Diet 3 was supplemented with methionine, lysine, isoleucine and tryptophan to the levels present in Diet 1. Tryptophan was added as one of the supplementary amino acids to Diet 3 because it was considered to be deficient on the basis of tabled analytical values. It is not, however, discussed here because it was not detected by the analytical method. The plasma concentrations of the other three supplementary amino acids in fish fed Diet 3 were approximately double those in fish fed Diet 1, indicating that the supplementary amino acids were effectively absorbed.Plasma methionine in fish fed Diet 3 was maintained at a level higher than in fish fed Diets 1 and 2 until 36h after feeding. Although the concentrations of plasma lysine in fish fed Diet 3 dropped abruptly at 24h after feeding, the fact that plasma lysine concentration was maintained at a higher level than in fish fed diet 2 until 48h aftr feeding indicates the efficacy of supplementary lysine as a supplement to lysine-deficient diets. The patterns of change in the concentrations of plasma branched-chain amino acids are interesting. Not only was the level of plasma isoleucine in fish fed the diet supplemented with free isoleucine (Diet 3) elevated over that occurring with the unsupplemented diet, but the concentrations of leucine and valine were also increased. Plasma concentrations of leucine and valine have similarly been reported to increase in channel catfish when the diet was supplemented with free isoleucine (Wilson et al. 1980). Evidence of interaction among these amino acids based on growth data was observed by Chance et al. (1964) in chinook salmon, and by Wilson et al. 1980 in channel catfish. Choo et al.
(1991), on the other hand, found no changes in the plasma concentrations of isoleucine and valine in rainbow trout fed diets containing excess leucine. There have been studies, both in fish and in other species, showing that amino acid oxidation is in-
fluenced by the concentration of amino acids in the plasma and that an excess of amino acids not used for protein synthesis will be oxidized rapidly (Brett and Zala 1975; Walton et al. 1982; Harper 1983; Young et al. 1985; Kim et al. 1992). Thebault
(1985), for example, found in sea-bass that the level of plasma methionine in response to dietary supplementation reached a peak and then returned to the pre-feeding level sooner than was the case with other essential amino acids. He concluded, therefore, that supplementary free methionine was absorbed and degraded faster than methionine originating from dietary protein. The continued elevation of plasma methionine and lysine until 36h after feeding in the present study is not necessarily contradictory to the above observation. The rate of disappearance from the plasma may be the same as that from amino acids absorbed from dietary protein. In fact, there is no reason to suppose that, once absorbed, utilization of an amino acid would be affected by its source. It is possible that the enzymes responsible for oxidation of these amino acids were overloaded with substrate (Harper et al. 1970) so that high plasma concentrations of lysine and methionine were maintained for relatively long periods. The ratio of EAA to NEAA, in either the diets or the plasma, has been shown to be important in relation to protein quality and utilization in mammals (Young and Zamora 1968; Hartog and Pol 1972; Fujita et al. 1981; Mercer et al. 1989). During the
absorptive period in Experiment 1, fish fed Diet 1 had a plasma ratio of 1, whereas plasma ratios of EAA to NEAA in fish fed Diets 2 and 3 were considerably lower, reflecting the amino acid composition of the diets. The increase in the concentrations of most essential amino acids in the plasma by 120h after feeding indicated input of free amino acids from catabolism of tissue protein. The increased ratios of plasma EAA/NEAA in the fish from the different dietary treatment groups at 120h confirmed the contribution of amino acids from the catabolism of the body protein. Another significant phenomenon regarding the changes in the plasma concentrations of histidine, lysine, and alanine in fish fed Diets 2 and 3 was observed in Experiment 1. The concentrations of these
192 amino acids exhibited similar patterns of change in that plasma concentrations either remained constant or peaked during 3- 12h postprandial and then rose at 36 h. The surges in the concentrations at 36h were interpreted as the result of delayed absorption of these amino acids following slow digestion of some of the proteins in the mixture of protein sources employed in these two diets. As with other studies in rainbow trout (Nose 1972; Gras et al. 1982; Walton and Wilson 1986), glycine was found in highest concentration among plasma non-essential amino acids. In all three experiments, fish that were fed the diets containing either gelatin or free glycine had plasma concentrations of glycine that rose to as high as three times the concentration of glycine found in fish fed the control diets. Plasma concentrations of other NEAAs were also higher in fish fed diets containing a mixture of protein sources other than fishmeal, in keeping with the higher concentrations of NEAAs in the former. Dietary concentrations of serine were similar within the three experiments. The higher concentrations of plasma serine in fish fed the mixed protein diets are therefore assumed to have derived from plasma glycine since glycine and serine are readily interconvertible (Bender 1985; Schepartz 1973; Walton and Cowey 1981, 1982). Dietary supplementation with crystalline essential amino acids accelerated the increase in the plasma concentrations of those particular amino acids following meal consumption. Diets that contained gelatin (Diets 2 and 3 in Experiment 1, Diet 5 in Experiment 2, and Diets 8 and 9 in Experiment 3) all resulted in rapid uptake of glycine as well as of some other NEAA. This fact is evidence for the rapid digestion of gelatin yielding the component amino acids for absorption in advance of amino acids from other protein sources in the diet. What is particularly interesting in this regard is the fact that the appearance of glycine in th systemic circulation, when glycine was added in free form to Diet 6 in Experiment 2, was slower than from gelatin in Diet 5 (Figure 2). It may be inferred from the rapid increases in other plasma amino acids following consumption of Diet 2 that dietary gelatin was rapidly hydrolyzed releasing the amino acids for absorption. A possibly related observation is that
of Bog6 et al. (1981), who reported that glycine uptake by rainbow trout was more rapid from glycylglycine than from an equivalent amount of free glycine, especially at high dietary concentrations. A similar phenomenon has been observed in mammals (Matthews 1973). The high plasma concentration of glycine coupled with the poorer feed consumption noted for the diets that were high in glycine, whether from gelatin or free glycine, are of interest. These data confirm the findings of Hughes (1985) that both lake trout (Salvelinus namaycush) and rainbow trout grew more slowly and had poorer feed efficiency when glycine was substituted for glutamic acid in the diet as a source of NEAA. The rapid increases in plasma serine and alanine in fish fed Diet 3 over those that occurred with Diet 1 cannot be attributed to the dietary concentration of these amino acids since the only dietary variable was supplementary crystalline glycine. Plasma proline concentration, on the other hand, showed an increase following consumption of Diet 2, in accord with the proline concentration in that diet, which did not occur with Diet 3 in which glycine rather than gelatin was the supplement. It is assumed, therefore, that the increases in serine and alanine concentrations derived from the elevated concentrations of plasma glycine. There were differences between the concentrations of individual amino acids present in free form between plasma and muscle. In plasma, the branched-chain amino acids and glycine were respectively the predominant essential and nonessential amino acids. In muscle, histidine characterized the pool of essential amino acids, constituting 15 to 24% of the total amino acids, and glycine dominated the nonessential amino acid pools, representing 46 to 60% of the total amino acids. The concentration in the plasma of an amino acid at any given time represents the balance between the rates of ingress and removal of that amino acid, whether for protein synthesis or oxidation. Consumption of one meal per day would therefore be expected to result in marked fluctuations in plasma amino acid concentrations. Consumption of the same daily ration provided in a number of smaller meals given at intervals during the day would be expected to result in a more steady state of plasma
193 concentrations of amino acids. This supposition was borne out by the data of Experiment 3. A consequence of the steady state was the higher PPV obtained in response to more frequent feeding. The more efficient use of plasma amino acids for protein synthesis was further exemplified by the higher proportion of protein to lipid in the body composition of fish fed five meals daily compared with fish given the equivalent daily ration as a single meal. A similar finding regarding the effects of feeding a similar ration in the form of one or five meals daily has been made with rats (Friend 1967). Rats fed once daily contained a higher proportion of fat in their bodies and demonstrated greater retention of absorbed nitrogen. Feeding frequency of cultured fish merits further study for its practical consequences for body fat accumulation and for efficiency of protein utilization.
Acknowledgements Financial assistance from the Science Council of British Columbia for support of the research is acknowledged. The authors thank Carol MacMillan for technical assistance during the study and in the preparation of the manuscript. References cited Aoe, H., Masuda, I., Abe, I., Saito, T., Toyoda, T. and Kitamura, S. 1970. Nutrition of protein in young carp. I. Nutritive value of free amino acids. Bull. Jap. Soc. Sci. Fish. 36: 407-413. Anderson, H.L., Benevenga, N.J. and Harper, A.E. 1969. Effect of prior high protein intake on food intake, serine dehydratase activity and plasma amino acids of rats fed amino acid-imbalanced diets. J. Nutr. 97: 463-474. AOAC 1984. Official Methods of Analysis. Association of Official Analytical Chemists, Washington, D.C. Beckman 1982. The System 6300 High Performance Amino Acid Analyzer: Instruction Manual. Beckman Instruments Inc., USA. Bender, D.A. 1985. Amino Acid Metabolism. John Wiley & Sons Ltd., Chichester. Bog6, G., Rigal, A. and Per6s, G. 1981. Rates of in vivo intestinal absorption of glycine and glycylglycine by rainbow trout (Salmo gairdneri R). Comp. Biochem. Physiol. 69A: 455459.
Brett, J.R. and Zala, C.A. 1975. Daily pattern of nitrogen excretion and oxygen consumption of sockeye salmon (Oncorhynchus nerka) under controlled conditions. J. Fish. Res. B. Can. 32: 2479-2486. Chance, R.E., Mertz, E.T. and Halver, J.E. 1964. Isoleucine, leucine, valine and phenylalanine requirements of chinook salmon and interrelationship between isoleucine and leucine for growth. J. Nutr. 83: 177-185. Cho, C.V., Slinger, S.J. and Bayley, H.S. 1976. Influence of level and type of dietary protein and of level of feeding on feed utilization by rainbow trout. J. Nutr. 106: 1547-1556. Choo, P., Smith, T.K., Cho, C.Y. and Ferguson, H.W. 1991. Dietary excesses of leucine influence growth and body composition of rainbow trout. J. Nutr. 121: 1932-1939. Cowey, C.B. and Walton, M.J. 1988. Studies on the uptake of ('4C) amino acids derived from both dietary (14C) protein and dietary (14C) amino acids by rainbow trout, Salmo gairdneri Richardson. J. Fish Biol. 33: 293-305. Cowey, C.B. and Walton, M.J. 1989. Intermediary metabolism. In: Fish Nutrition, pp. 260-330. Edited by J.E. Halver. Academic Press, San Diego. Friend, D.W. 1967. Weight gains, nitrogen metabolism, and body composition of rats fed one or five meals daily. Can. J. Physiol. Pharm. 45: 367-373. Fujita, Y., Yamamoto, T., Rikimaru, T., Ebisawa, H. and Indue, G. 1981. Effect of quality and quantity of dietary protein on free amino acids in plasma and tissues of adult rats. J. Nutr. Sci. Vitaminol. 27: 129-147. Gras, J ., Gudefin, Y., Chagny, F. and Perrier, H. 1982. Free amino acids and ninhydrin-positive substances in fish-II. Cardio-respiratory system: plasma, erythrocytes, heart and gills of the rainbow trout (Salmo gairdneri Richardson). Comp. Biochem. Physiol. 73B: 845-847. Hajen, W.E., Higgs, D.A., Beames, R.M. and Dosanjh, B.S. 1993. Digestibility of various feedstuffs by post-juvenile chinook salmon. (Oncorhynchus tshawytsha) in sea water. 2. Measurement of digestibility. Aquaculture 112: 333-348. Harper, A.E. 1983. Some recent development in the study of amino acid metabolism. Proc. Nutr. Soc. 42: 437-449. Harper, A.E., Benevenga, N.J. and Wohlheuter, R.M. 1970. Effects of ingestion of disproportionate amounts of amino acids. Physiol. Rev. 50: 428-558. Hartog, C.D. and Pol, G. 1972. D. Assays based on measurements of plasma-free amino acids. In Protein and Amino Acid Functions. pp. 353-361. Edited by E.J. Bigwood. Pergamon Press, New York. Hughes, S.G. 1985. Evaluation of glutamic acid and glycine as sources of non-essential amino acids for lake trout (Salvelinus namaycush) and rainbow trout (Salmo gairdneri). Comp. Biochem. Physiol. 81A: 669-671. Kaushik, S.J. and Luquet, P. 1979. Influence of dietary amino acid patterns in the free amino acid contents of blood and muscle of rainbow trout (Salmo gairdneri R.) Comp. Biochem. Physiol. 64B: 175-180. Kim, K.I., Grimshaw, T.W., Kayes, T.B. and Amundson, C.H. 1992. Effect of fasting or feeding diets containing different
194 levels of protein or amino acids on the activities of the liver amino acid-degrading enzymes and amino acid oxidation in rainbow trout (Oncorhynchus mykiss). Aquaculture 107: 89-105. Marcha, B.E., MacMillan, C. and Ming, F.W. 1985. Techniques for evaluation of dietary protein quality for rainbow trout (Salmo gairdneri). Aquaculture 47: 275-292. Marcha, B.E. and Walker, R.O. 1970. The effect of shifts in dietary amino acid pattern on feed consumption in the chick. Can. J. Physiol. Pharm. 48: 265-268. Matthews, D.M. 1973. Protein absorption. In The Gut and Nutrition. Proc. 12th Symp. Group of European Nutritionists, Cambridge. pp. 28-41. Edited by J.C. Somogyi and I. MacDonald. Mellinkoff, S.M., Frankland, M., Boyle, D. and Greipel, M. 1956. Relationship between serum amino acid concentration and fluctuations in appetite. J. Appl. Physiol. 8: 535-538. Mercer, L.P., Dodds, S.J. and Smith, D.I. 1989. Dispensable, indispensable, and conditionally indispensable amino acid ratios in the diet. In Absorption and Utilization of Amino Acids. pp. 1-14. Edited by M. Friedman. CRC Press Inc., Boca Raton. Nose, T. 1972. Changes in pattern of free amino acids in rainbow trout after feeding. Bull. Freshw. Fish. Res. Lab. 22: 137-144. Robinson, E.H. 1992. Use of supplemental lysine in catfish feeds. Aquaculture Magazine. May/June: 94-96. Schepartz, B. 1973. Regulation of Amino Acid Metabolism in Mammals. Saunder, Philadelphia. Spinelli, J., Mahnken, C. and Steinberg, M. 1979. Alternate sources of proteins for fishmeal in salmonid diets. Proc. World Symp. in Finfish Nutrition and Fishfeed Technology. 2: 131-142. Thebault, H. 1985. Plasma essential amino acid changes in seabass (Dicentrarchuslabrax) after feeding diets deficient and supplemented in L-methionine. Comp. Biochem. Physiol. 82A: 233-237. Walton, M.J. and Cowey, C.B. 1981. Distribution and some
kinetic properties of serine catabolizing enzymes in rainbow trout Salmo gairdneri.Comp. Biochem. Physiol. 68B: 147150. Walton, M.J. and Cowey, C.B. 1982. Aspects of intermediary metabolism in salmonid fish. Comp. Biochem. Physiol. 73B: 59-79. Walton, M.J., Cowey, C.B. and Adron, J.W. 1982. Methionine metabolism in rainbow trout fed diets of differing methionine and cystine content. J. Nutr. 112: 1525-1535. Walton, M.J. and Wilson, R.P. 1986. Postprandial changes in plasma and liver free amino acids of rainbow trout fed diets containing casein. Aquaculture 51: 105-115. Walton, M.J., Adron, J.W., Coloso, R.M. and Cowey, C.B. 1986. Dietary requirements of rainbow trout for tryptophan, lysine and arginine as determined by growth and biochemical measurements. Fish Physiol. Biochem. 2: 161-169. Wilson, R.P., Poe, W.E. and Robinson, E.H. 1980. Leucine, isoleucine, valine and histidine requirements of fingerling channel catfish. J. Nutr. 110: 627-633. Wilson, R.P., Allen, O.W., Robinson, E.H. and Poe, W.E. 1978. Tryptophan and threonine requirements of fingerling channel catfish. J. Nutr. 108: 1595-1599. Yamada, S., Simpson, K.L., Tanaka, Y. and Katayama, T. 1981. Plasma amino acid changes in rainbow trout Salmo gairdneri force-fed casein and a corresponding amino acid mixture. Bull. Jap. Soc. Sci. Fish. 47: 1035-1040. Young, V.R. and Zamora, J. 1968. Effects of altering the proportion of essential to nonessential amino acids on growth and plasma amino acid levels in the rat. J. Nutr. 96: 21-27, Young, V.R., Meredith, C., Hoerr, R., Bier, D.M. and Matthews, D.E. 1985. Amino acid kinetics in relation to protein and amino acid requirement: The primary importance of amino acid oxidation. In Substrate and Energy Metabolism in Man. pp. 119-134. Edited by J.S. Garrow and D. Holliday. John Libbey, London. Zar, J.H. 1984. Biostatistical Analysis. Prentice Hall, Englewood Cliffs.
