Fish Physiology and Biochemistry vol. 12 no. 4 pp 305-315 (1993) Kugler Publications, Amsterdam/New York
Protein-nitrogen flux and protein growth efficiency of individual Atlantic salmon (Salmo salar L.). C.G. Carter, D.F. Houlihan, B. Buchanan and A.I. Mitchell Department of Zoology, University of Aberdeen, Tillydrone Avenue, Aberdeen, AB9 2TN, Scotland Accepted: July 23, 1993 Keywords: Atlantic salmon; protein growth efficiency; protein-nitrogen flux; individual variation
Abstract Protein-nitrogen flux (the proportions of consumed and absorbed protein-nitrogen partitioned into protein synthesis and growth) was examined in Atlantic salmon, Salmo salar L. Salmon were held in groups and fed high or low rations or starved. Individual food consumption rates were measured using radiography. Fish varied widely in protein growth efficiency (protein growth divided by protein consumption), but this did not correlate with consumption rate, digestive capacity (as measured by absorption efficiency, trypsin levels and pyloric caecal size) or feeding hierarchy rank. Protein synthesis rates, measured in whole-animals, were linearly correlated with protein consumption and assimilation. There was a significant correlation between protein growth efficiency and the efficiency of retention of synthesised proteins. The capacity for protein synthesis and RNA activity were positively correlated with rates of food consumption and growth but were not correlated with protein growth efficiency. It was concluded that individual differences in protein growth efficiency related to differences in synthesis retention efficiency, but not to differences in the capacity for protein synthesis, RNA activity, digestive capacity or feeding hierarchy rank.
Introduction Rates of food consumption of individual fish in groups can vary widely. These differences, which may be the result of socially dominant animals eating more of the available food, are a major influence on growth rates (Jenkins 1969; Jobling et al. 1989; Metcalfe et al. 1989; Carter et al. 1992a; McCarthy et al. 1992). However, inter-individual variability in food consumption does not provide a complete explanation because some fish grow faster than others when consuming identical quantities of food (Carter et al. 1992a, 1993; Houlihan et al. 1993b). This results in there being differences in the efficiency with which food is converted into growth (growth efficiency) between fish. We do not know why fish eating the same
amount of food should have different growth efficiencies, but sources of individual variability in growth efficiency may partly be explained by differences in protein turnover (Hawkins, 1991; Tomas et al. 1991; Carter et al. 1993; Houlihan et al. 1993b), social status (Abbott and Dill 1989) or variability in digestive efficiency (Torrissen and Shearer 1992). Ultimately genetic variation may explain these differences (Bergot et al. 1981; Hawkins 1991; Torrissen and Shearer 1992). Protein turnover could influence growth efficiency through the link between protein synthesis and energy expenditure so that individuals with reduced protein turnover would be expected to have higher growth rates for given levels of food consumption (Hawkins 1991; Houlihan 1991). The aim of this study was to investigate the
306 source(s) of individual differences in protein growth efficiency in Atlantic salmon, Salmo salar L., by quantifying rates of protein-nitrogen flux. Food consumption rates of individual fish were determined by radiography (Talbot and Higgins 1983) and protein intake and deposition determined in order to calculate protein growth efficiency. The "digestive capacity" of individual fish was estimated from apparent absorption efficiencies for nitrogen and carbon (Austreng 1978), trypsin activity in the pancreas (Pringle et al. 1992) and the weight of the pyloric caeca (Bergot et al. 1981; Ulla and Gjedrem 1985). Social/feeding rank was indicated by the variation in the size of meal consumed on different days by each individual fish (McCarthy et al. 1992; Winberg et al. 1993). Rates of protein synthesis (Garlick et al. 1980; Houlihan et al. 1986) were investigated in selected fish at the end of the three month growth period; it being hypothesised that the proportion of synthesised proteins retained as protein growth should be correlated with protein growth efficiency (Tomas et al. 1991; Carter et al. 1993). Recently there have been a number of attempts at correlating RNA concentration with protein growth rate on fish on the assumption that food consumption rates can alter RNA concentration and that the latter largely determine rates of protein synthesis (Mathers et al. 1992; Foster et al. 1993; Houlihan et al. 1993a). In the present study this hypothesis was tested by analyzing the relationships between RNA concentration and food intake, protein synthesis and protein growth. Furthermore, the relationships between growth efficiency and RNA concentration and RNA activity were investigated.
Material and methods Husbrandy andfeeding Atlantic salmon were supplied by and maintained at Otter Ferry Salmon Ltd, Tighnabruaich, Argyll. At the start of the experiment fish were removed from production tanks, anaesthetized in benzocaine, weighed, individually marked with alcian blue using a Panjet (Hart and Pitcher 1969),
divided between tanks and left to acclimate for a week. The salmon were held in circular tanks containing approximately 4500 1 of water and supplied with filtered, aereated sea water (salinity = 35 ppt: oxygen saturation = 91.5 + 0.5%) from Loch Fyne at a rate of 40 1/min. The fish were fed for five days and then left for a further two days before being reweighed. Twenty-five fish were returned to each tank and feeding started on the next day. At this time ten fish were selected as an initial group, killed, weighed, dissected and the pyloric caeca weighed. The pyloric caeca were returned to the carcass, the carcasses were then stored at - 70°C until further analysis. The mean ( SEM) wet weight of the fish at the start of the experiment was 95.8 + 1.4 g. The experiment lasted for 76 days, from July 18 to October 3. Towards the end of the experiment 19 fish (283.4 + 8.0 g) were taken from stock, deprived of food for 14 days and killed on October 3. The mean weekly water temperatures ranged between 12.1 and 14.1°C, with an overall mean (+ SD) of 12.9 0.6°C. Automatic belt feeders (FIAP, Allersburg) operated continuously from 09:00 to 15:00 each day. Two groups were fed an experimental diet, details of which are given by Carter et al. (1994), at initial rations of either 2.5 %BW/d (percent body weight per day) or 1.25 %BW/d and denoted the high or low ration group, respectively. The fish were weighed during the estimation of individual consumption rates (see below) and the tank rations adjusted accordingly. Elemental carbon (46.9% dry weight) and nitrogen (8.1% dry weight) were measured using a Carlo Erba Elemental Analyzer 1106 (Carter et al. 1992b). Crude protein (47.4% dry weight) was calculated from nitrogen assuming a conversion of 5.85 (Gnaiger and Bitterlich 1984), fat (16.2% dry weight) was measured by Soxhlet extraction, energy content (20.9 kJ/dry g) and ash (16.6% dry weight) were determined with a Gallenkamp Ballistic bomb calorimeter calibrated with benzoic acid. Chromic oxide (2%) was included in the diet fed over the last five days and the apparent absorption efficiency of nitrogen (AEN) and carbon (AEc) calculated according to Fnge and Grove (1979). On the final day of the experiment the fish were
307 first X-rayed (see below) and returned to the tanks to recover. After a further four to six hours the fish were anaesthetized, removed from the tank and stripped of faeces (Austreng 1978). Rates of protein synthesis were determined for 10 fish from the high ration group, 10 from the low ration group and 5 starved fish. These fish were selected to give as wide a range of consumption and growth rates as possible. The remaining fish were killed by a blow to the head and transection of the spinal cord. The pyloric caeca were dissected out and weighed; 500 mg samples were removed for the analysis of trypsin [EC 3.4.21.4] activity (Pringle et al. 1992) and frozen in liquid nitrogen. The pyloric caecal somatic index (PCSI) was calculated as PCSI (o)
= 100 x (weight of pyloric caeca/wet weight of fish)
multiplying the kr by the AEN for individual fish. The effect of feeding rank on protein growth efficiency was assessed using the intra-individual variation in food intake. Intra-individual variation in food consumption was calculated using the coefficient of variation for mean weight specific consumption (CVc) CV (o)
= 100 x (SD/Cm)
where Cm and SD are the mean consumption rates (mg/g/d) and their standard deviations, respectively, for each fish. The inter-individual variation in food consumption was calculated as the proportion of a group meal eaten by each fish on each occasion and averaged to obtain an estimate for the mean share of meal (MSM: %o) eaten by each invididual fish (McCarthy et al. 1992).
Estimation of individual consumption rates
Protein growth rate
The daily consumption rates of individual fish were determined by radiography (Jobling et al. 1989; Carter et al. 1994) on four occasions (days 25, 42, 63, 76). Separate batches of the diet were pelleted to contain 2% Ballotini glass beads (size 9, 290420 gzm, Jencons Scientific Ltd). After the labelled food had been supplied for 6 h the fish were lightly anaesthetized with benzocaine whilst still in the tank before 8 or 9 fish were transferred to a stronger solution of benzocaine (60 ppm) to induce deeper anaesthesia. Radiographs of fish were then taken (Todd Research Machines 80/20 portable X-ray unit; 60 mAs exposure; Kodak Industrex CX Film), the fish identified, weighed and returned to the tank. This procedure took less than five minutes for each batch of fish. Weight specific consumption rates (mg/g/d) were calculated and these data used to calculate a mean weight specific consumption rate (Cm) for each fish for the entire experiment (Jobling et al. 1989; Carter et al. 1992b, 1994). The diet contained 47.4o crude protein and the protein consumption rate (kr: %/d) was calculated as a percent of the initial protein content of fish (see below) consumed per day. The apparent "absorption" rate of dietary protein-nitrogen (ka: %/d) was calculated by
Fractional rates of protein growth (kg) were calculated as kg (%/d) = 100 x (logeP 2 - logeP1)/t where P 1 and P 2 were the initial and final protein contents (g), respectively, after t days (Wootton 1990). The initial protein content was calculated using the initial wet weight of each fish and the average protein content of the initial group. Whole fish used for the determination of rates of protein synthesis were homogenised in 0.5 M PCA (300 ml/100 g wet weight of fish). The homogenate was then spun at 6000 rpm for 5 min at 4°C and the supernatant discarded, the process was repeated four more times (after which no supernatant could be removed). The total precipitate was weighed and triplicate samples of approximately 50 mg used for measurements of protein and RNA and the calculation of whole-animal protein and RNA content and the measurement of protein-bound phenylalaninespecific radioactivities (see below). Protein was measured by the method of Lowry et al. (1951) as modified by Schacterle and Pollock (1973). The final protein contents of the remaining fish (fish not used for protein synthesis measurement) were estimated from the mean final protein content of fish
308 used to measure protein synthesis and from the appropriate ration group. RNA concentrations (lsg/g wet tissue) were calculated from dual wavelength measurements (Ashford and Pain 1986; McMillan and Houlihan 1992) and expressed as the RNA: protein ratio or capacity for protein synthesis (Cs: mg RNA/g protein) (Sugden and Fuller 1991).
of protein degradation (kd: %/d) were calculated as the difference between the fractional rates of protein synthesis and protein growth (Millward et al. 1975). The 'anabolic stimulation efficiency' (%o) was calculated as ks/kr (%) = 100 x (ks/kr) (Houlihan et al. 1993a). The 'synthesis retention efficiency' (o) was calculated as kg/ks (%) = 100 x (kg/ks)
Protein synthesis In vivo whole-animal protein synthesis rates were determined for 20 feeding fish, approximately 24 hours after their last meal, and for 5 starving fish. Fish received a single injection of 3 H phenylalanine in the caudal vein without anaesthesia (Houlihan et al. 1986, 1988). The injection solution contained 135 mmol L-phenylalanine and L-[2,6 3 H] phenylalanine (Amersham International) (3.7 MBq/ ml) in Cortland Ringer at pH 7.4 (nominal phenylalanine-specific radioactivity of 1600 dpm/nmol). The dose was 1.0 ml/100 g wet weight (Houlihan et al. 1988). Individually identified fish were injected at a known time, returned to tanks containing aerated seawater at 14°C and then killed after either 81 or 120 min (fed fish) and 150 min (starved fish). The fish were killed by a blow to the head and transection of the spinal cord, weighed and a 300 mg sample of white muscle dissected from below the dorsal fin. White muscle samples were frozen in liquid nitrogen and later used to measure free-pool phenylalanine-specific radioactivities. Protein-bound phenylalanine-specific radioactivities were measured in samples of homogenised whole-animals (see above). Protein-bound and freepool phenylalanine-specific radioactivities were measured as described by Houlihan et al. (1986, 1988). Fractional rates of protein synthesis (k5: %/d) were calculated as
ks = 100 x ((Sb/Sa) x (1440/t)) where Sb (dpm/nmol) is the protein-bound phenylalanine-specific radioactivity at time t and Sa (dpm/nmol) the free-pool phenylalanine-specific radioactivity (Garlick et al. 1983). Fractional rates
(Houlihan et al. 1993a). Protein growth efficiency (%) was calculated as kg/kr (%) = 100 x (kg/kr) (Houlihan et al. 1993a). Translational efficiency of the ribosomes or 'RNA activity' (kRNA: g protein synthesized/g RNA/d) was calculated as kRNA (g protein synthesized/g RNA/d) = 10 x (ks/Cs) (Sugden and Fuller 1991). Statistical analysis Means and their standard errors are used throughout. Significance was accepted at probabilities of 5% or less. Linear regression was by the method of least-squares and significance tested by analysis of variance. Paired t-tests were used to compare pairs of fish with similar consumption rates. Statistical analysis was performed by Statistix (Analytical Software). Results Feeding rank The MSM and CV c of each fish were equated with feeding rank and assumed to be measurable characteristics of that fish (Winberg et al. 1993). Consequently, the data from the two tanks were pooled. The MSM ranged between 2.4 and 8.9% of the food consumed, the mean CVc was 37.8 + 2.9% (n = 47) and varied between 6.7 and 89.0%. There was a significant correlation between MSM
309 l 0
0 S .
-j
0
1
2
3
4
5
6
kr (%/d) Fig. 1. The relationship between fractional rates of protein consumption (kr) and fractional rates of protein growth (kg) for Atlantic salmon. Relationship described by kg = -0.186 + 0.306kr (n = 57; R2 = 0.673; p < 0.0001). Filled circles indicate fish used to measure protein synthesis.
and CV c (n = 47; r = - 0.320; p < 0.05) which indicated that the fish which consumed a greater share of the food had consistently larger meals. The effect of feeding rank on protein growth efficiency was investigated but there was no relationship between protein growth efficiency and CVc.
Digestive capacity Mean values for the pyloric caecal somatic index (PCSI), trypsin activity and apparent absorption efficiencies for nitrogen and carbon were 3.4 + 0.1% (n = 27), 0.072 ± 0.012/Amol pNA/min/g (n = 18), 80.5 1.3% (n = 33) and 61.9 1.5% (n = 33), respectively. There were no significant relationships between protein growth efficiency and these indices of digestive capacity. Nitrogen and carbon absorption efficiencies were not correlated with consumption rate. Individual fish with high nitrogen absorption efficiencies also had high absorption efficiencies for carbon, but absorption efficiency was not significantly correlated with trypsin activity or the weight of the pyloric caeca.
Protein turnover Time course of protein synthesis White muscle free pool concentrations of phe-
0
1
2
3
4
5
6
kr (%/d) Fig. 2. The relationship between fractional rates of protein consumption (kr) and fractional rates of protein synthesis (ks) for Atlantic salmon. Relationship described by k, = 1.082 + 0.513kr (n = 25; R2 = 0.488; p < 0.001).
nylalanine at 81 and 120 min were not significantly different from each other and had a mean value of 386.1 + 18.3 nmol/g white muscle which was approximately 8 times the normal concentration of phenylalanine in Atlantic salmon white muscle (Carter et al., unpublished data). The phenylalanine-specific radioactivity in the white muscle freepools remained stable over the time course of protein synthesis measurements. The free-pool phenylalanine-specific radioactivities were not significantly different at the different times (81 and 120 min) and had an overall mean value of 1323 25 dpm/nmol compared to 1600 dpm/nmol for the injection solution. The mean protein-bound phenylalanine-specific radioactivity was significantly different at 81 (1.74 0.21 dpm/nmol) and 120 (2.94 + 0.27 dpm/nmol) min and indicated linear incorporation with time. Whole-animal protein turnover Protein accretion increased with increasing protein consumption (Fig. 1). The relationship between fractional rates of protein consumption and growth was described by a simple linear model (y = a + bx) from which the maintenance protein ration (i.e. kr at kg = 0) was calculated to be 0.61 %/d (Fig. 1). Fractional rates of protein synthesis increased with increasing protein consumption (Fig. 2) and the fractional rate of protein synthesis at maintenance was predicted to be 1.39 ± 0.26%/d. The
310 6
a. 7
= I.s 0
4
C
.
6 0 0
5 0
z 2
E U W) -2
-1
0
1
2
4' 0 0
3 24 1
3
0
S
0
i
i
1
2
3
kg (%/d) Fig. 3. The relationship between fractional rates of protein growth (kg) and fractional rates of protein synthesis (ks) for starved and feeding Atlantic salmon. The dotted line indicates the mean rate of protein synthesis for starving salmon. The full line is weighted to take account of the maintenance rate of protein synthesis indicated by an open circle.
4
5
6
kr (%/d)
b. 7
2 0 0
0
5
~
-~
slope of the line indicated that after accounting for the maintenance requirements approximately 51 %, on average, of the consumed protein ration was partitioned into synthesis. Individual nitrogen absorption efficiencies were measured for 33 of the feeding fish including 14 fish used to measure protein synthesis. Despite individual values for the nitrogen absorption efficiency ranging between 65.4 and 87.4% significant relationships between nitrogen absorption and protein growth and synthesis were found. Thus, kg = -0.202 + 0.365ka (n = 43; R 2 = 0.674; p < 0.0001) and k = 1.148 + 0.598k a (n = 19; R 2 = 0.467; p < 0.001). The mean rate of synthesis in the starved fish was 0.96 0.19%/d and showed no relationship with the rate of loss of body protein. In growing fish synthesis increased with increasing growth rate. After accounting for the predicted rate of synthesis at maintenance (1.39%/d) there was a linear relationship between fractional rates of protein growth and synthesis described by k = 2.04kg (n = 20; R 2 = 0.900; p < 0.001). Incorporation of the predicted value for synthesis at maintenance suggested the relationship between fractional rates of protein growth and synthesis was described by ks = 1.39 + 2.0 4 kg (Fig. 3). The concentration of RNA in the whole-animal
i
i
i
i
z
4
U S
3
~ 0~~~~~
0
_
0_~
1
Ix
v, U
1
0 0
2 1
.
0
1
2
~~~~~~~~~~~~~~~~~~~~~~~ 3
ks
4
5
6
(%/d)
C 7 S
0
6 . 0 o.
5
0 0 *
_,M
z
4
0
0 :0
3
M
bf U
0
.
I
0
s oe
2 _
-2
I
I
I
I
I
-1
o
1
2
3
kg (%/d) Fig 4. The relationships between the capacity for protein synthesis (Cs) and a, the fractional rates of protein consumption (kr) b, the fractional rates of protein synthesis (k,) and c, the fractional rates of protein growth (kg) of seawater Atlantic salmon.
was positively correlated with both rates of protein consumption and protein growth. The increase in RNA resulted in a positive correlation between protein consumption and the capacity for protein syn-
311 Individual variation in protein growth efficiency
100 80 S
60 f1z
0
U, _V
40
0 0 -
20
S W:
0
* 0 . u
0
20
.
i.
40
.
.
60
.
80
.
.
100
kg/kr (%) Fig. 5. The relationship between protein growth efficiency (kg/kr) and synthesis retention efficiency (kg/ks) for feeding seawater Atlantic salmon. thesis (Cs ) described by C s = 2.89 + 0.40kc (n = 25; R2 = 0.357; p < 0.002) Fig. 4a). There were also positive correlations between protein synthesis and C s described by C s = 2.82 + 0.44k5 (n = 25; R 2 = 0.238; p < 0.02) and between protein growth and Cs described by C s = 3.55 + 0. 6 2kg (n = 25; R 2 = 0.234; p < 0.02). This was expected since fish eating more synthesised more protein and grew at higher rates (Fig. 4b and c). The positive correlation (p < 0.03) between protein consumption and RNA activity (kRNA) was weaker than between protein consumption and Cs . This suggested that increased protein synthesis, driven by increased consumption rate, was due primarily to an increased capacity for synthesis (increased RNA concentration) but also to an increase in the RNA activity. The mean protein growth efficiency (kg/kc) was 28.3 + 2.8% and varied between 13.8 and 61.7% for individual fish. There were no significant correlations between protein growth efficiency and rates of protein intake, synthesis or degradation, the capacity for protein synthesis or RNA activity. The mean synthesis retention efficiency was 32.1 + 3.3% and also varied considerably between individuals with values ranging between 16.2 and 65.6%. There was a positive correlation between protein growth efficiency and synthesis retention efficiency (n = 20; r = + 0.500; p < 0.05) (Fig. 5).
In order to further investigate the link between protein turnover and protein growth efficiency 6 pairs of fish were selected from the fish on which protein synthesis had been measured. Both fish in each pair were from the same tank and had similar (within 5°70 of each other) protein consumption rates but different growth rates (greater than 5% difference). The pairs were divided into low and high (protein growth) efficiency groups which, as a consequence of the selection procedure, had significantly different rates of protein growth and protein growth efficiency (Table 1). There were no significant differences between the two groups in the remaining parameters (Table 1) although the differences between synthesis retention efficiency was close to significance (p = 0.074). However, rates of synthesis and degradation, synthesis retention efficiency, anabolic stimulation efficiency and RNA activity were higher in the high efficiency fish in 4 or more of the pairs (Table 1). These data suggested (together with the previously demonstrated relationship between efficiencies of synthesis retention and protein growth) that small differences in protein turnover which did not attain the level of statistical significance lead to the difference in protein growth efficiency between the two groups. Degradation rates were almost the same for the two groups but the influx of dietary protein-nitrogen caused a greater anabolic stimulation of synthesis of which more was retained in the high efficiency fish. As a proportion of the absorbed protein-nitrogen the mean anabolic stimulation efficiency was higher (109% compared to 977o) for the high efficiency fish. Thus, a greater proportion of the degradation products were recycled into synthesis in the high efficiency fish whereas they were lost (oxidised) in the low efficiency fish. Discussion Methodology This study utilized a variety of techniques to identify possible correlates of protein growth efficiency.
312 Table 1. Comparison between measures of protein-nitrogen flux and protein metabolism of low and high efficiency groups (6 pairs) Parameter
"Selected" parameters kr (%/d) kg (o/d) kg/kr (%) "Non-selected" parameters k5 (%/d) kd (%/d) kg/ks (%) ks/kr (%) Cs (RNA/protein) kRNA g (protein synthesized/g RNA/d)
Low efficiency (mean SEM)
High efficiency (mean + SEM)
2.72 + 0.37 0.64 + 0.06 21.34 + 1.54
2.70 + 0.38 0.88 + 0.06 34.03 + 3.77
2.47 1.89 24.41 97.06 5.05 5.43
+ 0.37 + 0.33 + 4.13 + 14.67 + 0.83 + 1.05
2.73 1.86 31.90 109.12 4.15 6.76
+ + + + + +
0.22 0.16 1.96 14.25 0.28 0.71
p
Trend
0.615 0.0001 0.016
1/6 6/6 6/6
0.510 0.932 0.074 0.416 0.279 0.301
4/6 4/6 5/6 4/6 1/6 4/6
p, significance of paired t-test; trend, indicates the proportion of pairs in which the value was higher in the high efficiency fish.
The measurement of individual consumption rates using radiography has been critically discussed by McCarthy et al. (1993). There is no ideal method for the measurement of absorption efficiency in fish (reviewed by Talbot 1985). Stripping faeces had the advantage of collecting faeces from individual fish as well as preventing leaching and was selected as the most suitable method. In the present study the relationships between protein consumption, growth and synthesis were described using simple linear models in accordance with the scheme proposed by Houlihan et al. (1993a). Linear models relating protein or nitrogen consumption and growth provide good descriptors of this relationship in mammals (reviewed by Boorman 1980) and fish (Carter et al. 1992b; Houlihan et al. 1993a). The linear model was a more appropriate descriptor of the relationship between protein consumption and growth than a curvi-linear model, since the latter model also indicated a slope of close to one for the Atlantic salmon data. In all previous studies the relationships between protein consumption and synthesis and between protein growth and synthesis were linear (e.g., Houlihan et al. 1988, 1989; Carter et al. 1993). The validity of the flooding dose method to measure in vivo rates of protein synthesis in fish has been discussed previously (e.g., Houlihan et al. 1986, 1988, 1992). Protein synthesis was measured
once at the end of the study and over a few hours. It is possible that measured rates of synthesis might not reflect daily protein synthesis or changes in synthesis over time. Rates of protein synthesis remain relatively stable over 24 h in rat white muscle (Reeds et al. 1986). Since fish white muscle is the largest tissue and accounts for the majority of wholeanimal synthesis it was assumed that whole-animal rates of synthesis demonstrated this same stability (Houlihan et al. 1993b).
Individual variation in protein growth efficiency The present study has revealed inter-individual variation in consumption, growth, apparent absorption efficiency, protein turnover and protein growth efficiency in Atlantic salmon. Although consumption rate was the major determinant of growth rate, individuals differed in their ability to utilize food for growth, irrespective of ration. A major aim of the present study was to ascertain whether differences in feeding rank, digestive capacity, RNA concentration/activity and/or protein turnover were associated with these differences in protein growth efficiency. Some studies have shown that growth efficiency is decreased due to the "physiological stress" of subordination (Koebele 1985; Jobling and Reinsnes 1986; Abbott and Dill 1989). Alternatively a high
313 social position can result in decreased growth efficiency due to increased activity involved in maintaining dominance (Yamagishi et al. 1974). In the present study there was no indication that fish ranked higher in the feeding hierarchy had higher or lower protein growth efficiencies than fish with lower ranks. It was therefore concluded that feeding rank, principally the variation in consumption rate, did not influence protein growth efficiency in this experiment. Absorption efficiency has been demonstrated to decrease with increasing rations of both natural food and formulated pelleted diets (e.g., Elliott 1976; Henken et al. 1985). However, no relationship was observed in some studies (e.g., Birkett 1969; Itawa 1970; Cui and Wootton 1988). In Atlantic salmon individual differences in protein growth efficiency were not correlated with apparent absorption efficiency, trypsin activity or the weight of the pyloric caeca. Studies on Atlantic cod, Gadus morhua L., have indicated that growth rate is positively correlated with the capacity for protein synthesis (RNA/protein ratio) measured in whole-animals (Houlihan et al. 1989) and in white muscle (Foster et al. 1993). In the present study we were able to demonstrate positive correlations between the capacity for protein synthesis and protein consumption, protein synthesis and protein growth for individual fish. These findings supported the hypothesis that increased food consumption rates resulted in an increase in the RNA concentration which provided a mechanism for the resultant increase in synthesis and growth (Mathers et al. 1992; Houlihan et al. 1993a). There were however, no correlations between protein growth efficiency and the capacity for protein synthesis or RNA activity. It is well established that the amount of protein synthesised exceeds the amount of protein retained as growth (Houlihan 1991). The positive correlation between protein growth efficiency and the retention of synthesised protein suggests that this provides a mechanism through which increased protein growth efficiency is achieved. This finding is in agreement with studies on grass carp, Ctenopharyngodon idella (Val.) (Carter et al. 1993) and rainbow trout, Oncorhynchus mykiss (Walbaum)
(McCarthy 1993). Since protein synthesis accounts for a considerable part of the total energy expenditure in fish (Houlihan et al. 1993a, 1993b) costs of growth will therefore be lower in fish which retain a greater proportion of the synthesised protein as growth. Therefore, this study supported the hypothesis that the efficiency with which synthesised protein is retained, irrespective of the amount synthesised, is the important determinant of protein growth efficiency. The partitioning of dietary protein nitrogen between synthesis and oxidation has direct bearing on protein growth efficiency and pairs of salmon with similar consumption rates, but different protein growth efficiencies were selected to investigate this. The high efficiency salmon tended to retain more of the synthesised protein than the low efficiency salmon. The anabolic stimulation efficiency showed that in the high efficiency salmon the rate of synthesis exceeded the amount of protein-nitrogen absorbed by about 10%. This level of anabolic stimulation was not measured in the low efficiency fish. We propose that the greater use of recycled amino acids for protein synthesis and growth, rather than oxidation and excretion, enabled some salmon to retain protein more efficiently. Rainbow trout strains with different nitrogen retention efficiencies can be identified and rates of ammonia excretion were higher in strains with lower nitrogen retention efficiencies (Ming 1985). This suggests differences occur in the partitioning of protein nitrogen between oxidation and growth between fish of the same species and that it has a genetic component. Furthermore, it has been hypothesized that a major influence on individual differences in performance may be due to individual differences in genotype-dependent protein turnover (Hawkins 1991; Tomas et al. 1991). Acknowledgements Our thanks go to Dr. I.D. McCarthy for his critical reading of the manuscript and to David Patterson and Mark Russel (Otter Ferry Salmon Ltd) for feeding and maintenance of the fish. Financial support was from the Ministry of Agriculture, Fisheries and Food.
314
References cited Abbott, J.C. and Dill, L.M. 1989. The relative growth of dominant and subordinate juvenile steelhead trout (Salmo gairdneri) fed equal rations. Behaviour 108: 104-113. Ashford, A.J. and Pain, V.M. 1986. Effect of diabetes on the rates of synthesis and degradation of ribosomes in rat muscle and liver in vivo. J. Biol. Chem. 261: 4059-4065. Austreng, E. 1978. Digestibility determination in fish using chromic oxide marking and analysis of contents from different segments of the gastrointestinal tract. Aquaculture 13: 265-272. Bergot, P., Blanc, J.M. and Escaftre, A.M. 1981. Relationship between number of pyloric caeca and growth in rainbow trout (Salmo gairdneri Richardson). Aquaculture 22: 81-96. Birkett, L. 1969. The nitrogen balance of plaice, sole and perch. J. Exp. Biol. 50: 373-386. Boorman, K.N. 1980. Dietary constraints on nitrogen retention. In Protein Deposition in Animals. pp. 147-166. Edited by P.J. Buttery and D.B. Lindsay. Butterworths, London. Carter, C.G., Houlihan, D.F., McCarthy, I.D. and Brafield, A.E. 1992a. Variation in the food intake of grass carp, Ctenopharyngodon idella (Val.), fed singly or in groups. Aquat. Living Resour. 5: 225-228. Carter, C.G., Houlihan, D.F. and McCarthy, I.D. 1992b. Feed utilization efficiencies of Atlantic salmon (Salmo salar L.) parr: effect of a single supplementary enzyme. Comp. Biochem. Physiol. 101A: 369-374. Carter, C.G., Houlihan, D.F., Brechin, J. and McCarthy, I.D. 1993. The relationships between protein intake and protein accretion, synthesis and retention efficiency for individual grass carp, Ctenopharyngodon idella (Val.). Can. J. Zool. 71: 392-400. Carter, C.G., Houlihan, D.F., Buchanan, B. and Mitchell, A.I. 1994. Growth and feed utilization efficiencies of Atlantic salmon, (Salmo salarL.), fed a diet containing supplementary enzymes. Aquacult. Fish. Manage. (In press) Cui, Y. and Wootton, R.J. 1988. Bioenergetics of growth of a cyprinid, Phoxinusphoxinus: the effect of ration, temperature and body size on food consumption, faecal production and nitrogenous excretion. J. Fish Biol. 33: 431-443. Elliott, J.M. 1976. Energy losses in the waste products of brown trout (Salmo trutta L). J. Anim. Ecol. 45: 561-580. Foster, A.R., Houlihan, D.F. and Hall, S.J. 1993. Effects of nutritional regime on correlates of growth rate in juvenile Atlantic cod (Gadus morhua): Comparison of morphological and biochemical measurements. Can. J. Fish. Aquat. Sci. 50: 502-512. Fange, R. and Grove, D.J. 1979. Fish Digestion. In Fish Physiology Vol. 8, pp. 161-260. Edited by W.S. Hoar, D.J. Randall and J.R. Brett. Academic Press, New York. Garlick, P.J., McNurlan, M.A. and Preedy, V.R. 1980. A rapid and convenient technique for measurin the rate of protein 3 synthesis in tissues by the injection of H phenylalanine. Biochem. J. 217: 507-516. Garlick, P.J., Fern, M. and Preedy, V.R. 1983. The effect of
insulin infusion and food intake on muscle protein synthesis in postabsorptive rats. Biochem. J. 210: 669-676. Gnaiger, E. and Bitterlich, G. 1984. Proximate biochemical composition and caloric content calculated from elemental CHN analysis: a stoichiometric concept. Oecologia (Berlin) 62: 289-298. Hart, P.J.B. and Pitcher, T.J. 1969. Field trials of fish marking using a jet inoculator. J. Fish Biol. 1: 383-385. Hawkins, A.J.S. 1991. Protein turnover: a functional appraisal. Funct. Ecol. 5: 222-233. Henken, A.M., Kleingeld, D.W. and Tijssen, P.A.T. 1985. The effect of feeding level on apparent digestibility of dietary dry matter, crude protein and gross energy in the African catfish, Clariasgariepinus (Burchell, 1822). Aquaculture 51: 1-11. Houlihan, D.F. 1991. Protein turnover in ectotherms and its relationship to energetics. Adv. Comp. Env. Physiol. 7: 143. Houlihan, D.F., McMillan, D.N. and Laurent, P. 1986. Growth rates, protein synthesis, and protein degradation rates in rainbow trout: Effects of body size. Physiol. Zool. 59: 482-493. Houlihan, D.F., Hall, S.J., Gray, C. and Noble, B.S. 1988. Growth rates and protein turnover in cod, Gadus morhua. Can. J. Fish. Aquat. Sci. 45: 951-964. Houlihan, D.F., Hall, S.J. and Gray, C. 1989. Effects of ration on protein turnover in cod. Aquaculture 79: 103-110. Houlihan, D.F., Wieser, W., Foster, A.R. and Brechin, J. 1992. In vivo protein synthesis rates in larval nase, Chondrostoma nasus L. Can. J. Zool. 70: 2436-2440. Houlihan, D.F., Mathers, E. and Foster, A. 1993a. Biochemical correlates of growth rate in fish. In Fish Ecophysiology. pp. 45-71. Edited by J.C. Rankin and F.B. Jensen. Chapman and Hall, London. Houlihan, D.F., Carter, C.G. and McCarthy, I.D. 1993b. Protein synthesis in fish. (in preparation) In Fish Molecular Biology and Biochemistry. Edited by P. Hochachka and T. Mommsen. (In press). Iwata, K. 1970. Relationship between food and growth in young crucian carps, Carassius auratus cuvieri, as determined by nitrogen balance. Jap. J. Lim. 31: 129-151. Jenkins, T.M. 1969. Social structure, position and microdistribution of two trout species (Salmo trutta and Salmo gairdneri)resident in mountain streams. An. Behav. Monog. 2: 56-123. Jobling, M. and Reinsnes, T.G. 1986. Physiological and social constraints on growth of Arctic charr, Salvelinus alpinus: and investigation of factors leading to stunting. J. Fish Biol. 28: 379-384. Jobling, M., Baardvik, B.M. and Jorgensen, E.H., 1989. Investigation of food-growth relationships of arctic charr, Salvelinus alpinusL., using radiography. Aquaculture 81: 367-372. Koebele, B.P. 1985. Growth and the size hierarchy effect: an experimental assessment of three proposed mechanisms; activity differences, disproportional food acquisition, physiological stress. Env. Biol. Fish. 12: 181-188. Lowry, H.O., Rosebrough, N.J., Farr, A.L., and Randall, R.J. 1951. Protein measurement with Folin phenol reagent. J.
315 Biol. Chem. 193: 265-275. Mathers, E.M., Houlihan, D.F. and Cunningham, M.J. 1992. Nucleic acid concentrations and enzyme activities as correlates of growth rate of the saithe Pollachius virens: growth rate estimates of open-sea fish. Mar. Biol. 112: 363-369. McCarthy, I.D. 1993. Feeding behaviour and protein turnover in fish. Ph.D Thesis, University of Aberdeen. McCarthy, I.D., Carter, C.G. and Houlihan, D.F. 1992. The effect of feeding hierarchy on individual variability in daily feeding of rainbow trout, Oncorhynchus mykiss (Walbaum). J. Fish Biol. 41: 257-263. McCarthy, I.D., Houlihan, D.F., Carter, C.G. and Moutou, K. 1993. Variation in individual food consumption rates of fish and its implications to the study of fish nutrition and physiology. Proc. Nutr. Soc. 52: 427-436. McMillan, D.N. and Houlihan, D.F. 1992. Protein synthesis in trout liver is stimulated by both feeding and fasting. Fish Physiol. Biochem. 10: 23-34. Metcalfe, N.B., Huntingford, F.A., Graham, W.D. and Thorpe, J.E. 1989. Early social status and the development of life-history strategies in Atlantic salmon. Proc. Roy. Soc. Lond. B236: 7-19. Millward, D.J., Garlick, P.J., Stewart, J.C., Nnanyelugo, D.O. and Waterlow, J.C. 1975. Skeletal-muscle growth and protein turnover. Biochem. J. 150: 235-243. Ming, F.W. 1985. Ammonia excretion rate as an index for comparing efficiency of dietary protein utilization among rainbow trout (Salmo gairdneri)of different strains. Aquaculture 46: 27-35. Pringle, G.M., Houlihan, D.F., Callanan, K.R., Mitchell, A.I., Raynard, R.S. and Houghton, G.H. 1992. Digestive enzyme levels and histopathology of pancreas disease in farmed Atlantic salmon (Salmo salar). Comp. Biochem. Physiol. 102A: 759-768. Schacterle, G.R. and Pollock, R.L. 1973. A simplified method for the quantitative assay of small amounts of protein in bio-
logic material. Anal. Biochem. 51: 654-655. Reeds, P.J., Palmer, R.M., Hay, S.M. and McMillan, D.N. 1986. Protein synthesis in skeletal muscle at different times during a 24 hour period. Biosci. Rep. 6: 209-213. Sugden, P.H. and Fuller, S.J. 1991. Regulation of protein turnover in skeletal and cardiac muscle. Biochem. J. 273: 21-37. Talbot, C. 1985. Laboratory methods in fish feeding and nutritional studies. In Fish Energetics - New Perspectives. pp. 211-220. Edited by P. Tyler and P. Calow. Croom Helm, London. Talbot, C. and Higgins, P.J. 1983. A radiographic method for feeding studies in fish using metallic iron powder as a marker. J. Fish Biol. 23: 211-220. Tomas, F.M., Pym, R.A. and Johnson, R.J. 1991. Muscle protein turnover in chickens selected for increased growth rates, food consumption or efficiency of food utilization: effects of genotype and relationship to plasma IGF-I and growth hormone. Brit. Poul. Sci. 32: 363-376. Torrissen, K.R. and Shearer, K.D. 1992. Protein digestion, growth and food conversion efficiency in Atlantic salmon and Arctic charr with different trypsin-like isozyme patterns. J. Fish Biol. 41: 409-415. Ulla, 0. and Gjedrem, T. 1985. Number and length of pyloric caeca and their relationship to fat and protein digestibility in rainbow trout. Aquaculture 47: 105-111. Winberg, S., Carter, C.G., McCarthy, I.D., He, Z.-Y., Nilsson, G.E. and Houlihan, D.F. 1993. Feeding rank and brain serotonergic activity in rainbow trout, Oncorhynchus mykiss (Walbaum). J. Exp. Biol. 179: 197-211. Wootton, R.J. 1990. The Ecology of Teleost Fishes. Chapman and Hall, London. Yamagishi, H., Maruyama, T. and Mashiko, K. 1974. Social relation in a small experimental population of Odontobutisobscurus (Temminck et Schelgel) as related to individual growth and food intake. Oecologia (Berlin) 17: 187-202.
Protein-nitrogen flux and protein growth efficiency of individual Atlantic salmon (Salmo salar L.). C.G. Carter, D.F. Houlihan, B. Buchanan and A.I. Mitchell Department of Zoology, University of Aberdeen, Tillydrone Avenue, Aberdeen, AB9 2TN, Scotland Accepted: July 23, 1993 Keywords: Atlantic salmon; protein growth efficiency; protein-nitrogen flux; individual variation
Abstract Protein-nitrogen flux (the proportions of consumed and absorbed protein-nitrogen partitioned into protein synthesis and growth) was examined in Atlantic salmon, Salmo salar L. Salmon were held in groups and fed high or low rations or starved. Individual food consumption rates were measured using radiography. Fish varied widely in protein growth efficiency (protein growth divided by protein consumption), but this did not correlate with consumption rate, digestive capacity (as measured by absorption efficiency, trypsin levels and pyloric caecal size) or feeding hierarchy rank. Protein synthesis rates, measured in whole-animals, were linearly correlated with protein consumption and assimilation. There was a significant correlation between protein growth efficiency and the efficiency of retention of synthesised proteins. The capacity for protein synthesis and RNA activity were positively correlated with rates of food consumption and growth but were not correlated with protein growth efficiency. It was concluded that individual differences in protein growth efficiency related to differences in synthesis retention efficiency, but not to differences in the capacity for protein synthesis, RNA activity, digestive capacity or feeding hierarchy rank.
Introduction Rates of food consumption of individual fish in groups can vary widely. These differences, which may be the result of socially dominant animals eating more of the available food, are a major influence on growth rates (Jenkins 1969; Jobling et al. 1989; Metcalfe et al. 1989; Carter et al. 1992a; McCarthy et al. 1992). However, inter-individual variability in food consumption does not provide a complete explanation because some fish grow faster than others when consuming identical quantities of food (Carter et al. 1992a, 1993; Houlihan et al. 1993b). This results in there being differences in the efficiency with which food is converted into growth (growth efficiency) between fish. We do not know why fish eating the same
amount of food should have different growth efficiencies, but sources of individual variability in growth efficiency may partly be explained by differences in protein turnover (Hawkins, 1991; Tomas et al. 1991; Carter et al. 1993; Houlihan et al. 1993b), social status (Abbott and Dill 1989) or variability in digestive efficiency (Torrissen and Shearer 1992). Ultimately genetic variation may explain these differences (Bergot et al. 1981; Hawkins 1991; Torrissen and Shearer 1992). Protein turnover could influence growth efficiency through the link between protein synthesis and energy expenditure so that individuals with reduced protein turnover would be expected to have higher growth rates for given levels of food consumption (Hawkins 1991; Houlihan 1991). The aim of this study was to investigate the
306 source(s) of individual differences in protein growth efficiency in Atlantic salmon, Salmo salar L., by quantifying rates of protein-nitrogen flux. Food consumption rates of individual fish were determined by radiography (Talbot and Higgins 1983) and protein intake and deposition determined in order to calculate protein growth efficiency. The "digestive capacity" of individual fish was estimated from apparent absorption efficiencies for nitrogen and carbon (Austreng 1978), trypsin activity in the pancreas (Pringle et al. 1992) and the weight of the pyloric caeca (Bergot et al. 1981; Ulla and Gjedrem 1985). Social/feeding rank was indicated by the variation in the size of meal consumed on different days by each individual fish (McCarthy et al. 1992; Winberg et al. 1993). Rates of protein synthesis (Garlick et al. 1980; Houlihan et al. 1986) were investigated in selected fish at the end of the three month growth period; it being hypothesised that the proportion of synthesised proteins retained as protein growth should be correlated with protein growth efficiency (Tomas et al. 1991; Carter et al. 1993). Recently there have been a number of attempts at correlating RNA concentration with protein growth rate on fish on the assumption that food consumption rates can alter RNA concentration and that the latter largely determine rates of protein synthesis (Mathers et al. 1992; Foster et al. 1993; Houlihan et al. 1993a). In the present study this hypothesis was tested by analyzing the relationships between RNA concentration and food intake, protein synthesis and protein growth. Furthermore, the relationships between growth efficiency and RNA concentration and RNA activity were investigated.
Material and methods Husbrandy andfeeding Atlantic salmon were supplied by and maintained at Otter Ferry Salmon Ltd, Tighnabruaich, Argyll. At the start of the experiment fish were removed from production tanks, anaesthetized in benzocaine, weighed, individually marked with alcian blue using a Panjet (Hart and Pitcher 1969),
divided between tanks and left to acclimate for a week. The salmon were held in circular tanks containing approximately 4500 1 of water and supplied with filtered, aereated sea water (salinity = 35 ppt: oxygen saturation = 91.5 + 0.5%) from Loch Fyne at a rate of 40 1/min. The fish were fed for five days and then left for a further two days before being reweighed. Twenty-five fish were returned to each tank and feeding started on the next day. At this time ten fish were selected as an initial group, killed, weighed, dissected and the pyloric caeca weighed. The pyloric caeca were returned to the carcass, the carcasses were then stored at - 70°C until further analysis. The mean ( SEM) wet weight of the fish at the start of the experiment was 95.8 + 1.4 g. The experiment lasted for 76 days, from July 18 to October 3. Towards the end of the experiment 19 fish (283.4 + 8.0 g) were taken from stock, deprived of food for 14 days and killed on October 3. The mean weekly water temperatures ranged between 12.1 and 14.1°C, with an overall mean (+ SD) of 12.9 0.6°C. Automatic belt feeders (FIAP, Allersburg) operated continuously from 09:00 to 15:00 each day. Two groups were fed an experimental diet, details of which are given by Carter et al. (1994), at initial rations of either 2.5 %BW/d (percent body weight per day) or 1.25 %BW/d and denoted the high or low ration group, respectively. The fish were weighed during the estimation of individual consumption rates (see below) and the tank rations adjusted accordingly. Elemental carbon (46.9% dry weight) and nitrogen (8.1% dry weight) were measured using a Carlo Erba Elemental Analyzer 1106 (Carter et al. 1992b). Crude protein (47.4% dry weight) was calculated from nitrogen assuming a conversion of 5.85 (Gnaiger and Bitterlich 1984), fat (16.2% dry weight) was measured by Soxhlet extraction, energy content (20.9 kJ/dry g) and ash (16.6% dry weight) were determined with a Gallenkamp Ballistic bomb calorimeter calibrated with benzoic acid. Chromic oxide (2%) was included in the diet fed over the last five days and the apparent absorption efficiency of nitrogen (AEN) and carbon (AEc) calculated according to Fnge and Grove (1979). On the final day of the experiment the fish were
307 first X-rayed (see below) and returned to the tanks to recover. After a further four to six hours the fish were anaesthetized, removed from the tank and stripped of faeces (Austreng 1978). Rates of protein synthesis were determined for 10 fish from the high ration group, 10 from the low ration group and 5 starved fish. These fish were selected to give as wide a range of consumption and growth rates as possible. The remaining fish were killed by a blow to the head and transection of the spinal cord. The pyloric caeca were dissected out and weighed; 500 mg samples were removed for the analysis of trypsin [EC 3.4.21.4] activity (Pringle et al. 1992) and frozen in liquid nitrogen. The pyloric caecal somatic index (PCSI) was calculated as PCSI (o)
= 100 x (weight of pyloric caeca/wet weight of fish)
multiplying the kr by the AEN for individual fish. The effect of feeding rank on protein growth efficiency was assessed using the intra-individual variation in food intake. Intra-individual variation in food consumption was calculated using the coefficient of variation for mean weight specific consumption (CVc) CV (o)
= 100 x (SD/Cm)
where Cm and SD are the mean consumption rates (mg/g/d) and their standard deviations, respectively, for each fish. The inter-individual variation in food consumption was calculated as the proportion of a group meal eaten by each fish on each occasion and averaged to obtain an estimate for the mean share of meal (MSM: %o) eaten by each invididual fish (McCarthy et al. 1992).
Estimation of individual consumption rates
Protein growth rate
The daily consumption rates of individual fish were determined by radiography (Jobling et al. 1989; Carter et al. 1994) on four occasions (days 25, 42, 63, 76). Separate batches of the diet were pelleted to contain 2% Ballotini glass beads (size 9, 290420 gzm, Jencons Scientific Ltd). After the labelled food had been supplied for 6 h the fish were lightly anaesthetized with benzocaine whilst still in the tank before 8 or 9 fish were transferred to a stronger solution of benzocaine (60 ppm) to induce deeper anaesthesia. Radiographs of fish were then taken (Todd Research Machines 80/20 portable X-ray unit; 60 mAs exposure; Kodak Industrex CX Film), the fish identified, weighed and returned to the tank. This procedure took less than five minutes for each batch of fish. Weight specific consumption rates (mg/g/d) were calculated and these data used to calculate a mean weight specific consumption rate (Cm) for each fish for the entire experiment (Jobling et al. 1989; Carter et al. 1992b, 1994). The diet contained 47.4o crude protein and the protein consumption rate (kr: %/d) was calculated as a percent of the initial protein content of fish (see below) consumed per day. The apparent "absorption" rate of dietary protein-nitrogen (ka: %/d) was calculated by
Fractional rates of protein growth (kg) were calculated as kg (%/d) = 100 x (logeP 2 - logeP1)/t where P 1 and P 2 were the initial and final protein contents (g), respectively, after t days (Wootton 1990). The initial protein content was calculated using the initial wet weight of each fish and the average protein content of the initial group. Whole fish used for the determination of rates of protein synthesis were homogenised in 0.5 M PCA (300 ml/100 g wet weight of fish). The homogenate was then spun at 6000 rpm for 5 min at 4°C and the supernatant discarded, the process was repeated four more times (after which no supernatant could be removed). The total precipitate was weighed and triplicate samples of approximately 50 mg used for measurements of protein and RNA and the calculation of whole-animal protein and RNA content and the measurement of protein-bound phenylalaninespecific radioactivities (see below). Protein was measured by the method of Lowry et al. (1951) as modified by Schacterle and Pollock (1973). The final protein contents of the remaining fish (fish not used for protein synthesis measurement) were estimated from the mean final protein content of fish
308 used to measure protein synthesis and from the appropriate ration group. RNA concentrations (lsg/g wet tissue) were calculated from dual wavelength measurements (Ashford and Pain 1986; McMillan and Houlihan 1992) and expressed as the RNA: protein ratio or capacity for protein synthesis (Cs: mg RNA/g protein) (Sugden and Fuller 1991).
of protein degradation (kd: %/d) were calculated as the difference between the fractional rates of protein synthesis and protein growth (Millward et al. 1975). The 'anabolic stimulation efficiency' (%o) was calculated as ks/kr (%) = 100 x (ks/kr) (Houlihan et al. 1993a). The 'synthesis retention efficiency' (o) was calculated as kg/ks (%) = 100 x (kg/ks)
Protein synthesis In vivo whole-animal protein synthesis rates were determined for 20 feeding fish, approximately 24 hours after their last meal, and for 5 starving fish. Fish received a single injection of 3 H phenylalanine in the caudal vein without anaesthesia (Houlihan et al. 1986, 1988). The injection solution contained 135 mmol L-phenylalanine and L-[2,6 3 H] phenylalanine (Amersham International) (3.7 MBq/ ml) in Cortland Ringer at pH 7.4 (nominal phenylalanine-specific radioactivity of 1600 dpm/nmol). The dose was 1.0 ml/100 g wet weight (Houlihan et al. 1988). Individually identified fish were injected at a known time, returned to tanks containing aerated seawater at 14°C and then killed after either 81 or 120 min (fed fish) and 150 min (starved fish). The fish were killed by a blow to the head and transection of the spinal cord, weighed and a 300 mg sample of white muscle dissected from below the dorsal fin. White muscle samples were frozen in liquid nitrogen and later used to measure free-pool phenylalanine-specific radioactivities. Protein-bound phenylalanine-specific radioactivities were measured in samples of homogenised whole-animals (see above). Protein-bound and freepool phenylalanine-specific radioactivities were measured as described by Houlihan et al. (1986, 1988). Fractional rates of protein synthesis (k5: %/d) were calculated as
ks = 100 x ((Sb/Sa) x (1440/t)) where Sb (dpm/nmol) is the protein-bound phenylalanine-specific radioactivity at time t and Sa (dpm/nmol) the free-pool phenylalanine-specific radioactivity (Garlick et al. 1983). Fractional rates
(Houlihan et al. 1993a). Protein growth efficiency (%) was calculated as kg/kr (%) = 100 x (kg/kr) (Houlihan et al. 1993a). Translational efficiency of the ribosomes or 'RNA activity' (kRNA: g protein synthesized/g RNA/d) was calculated as kRNA (g protein synthesized/g RNA/d) = 10 x (ks/Cs) (Sugden and Fuller 1991). Statistical analysis Means and their standard errors are used throughout. Significance was accepted at probabilities of 5% or less. Linear regression was by the method of least-squares and significance tested by analysis of variance. Paired t-tests were used to compare pairs of fish with similar consumption rates. Statistical analysis was performed by Statistix (Analytical Software). Results Feeding rank The MSM and CV c of each fish were equated with feeding rank and assumed to be measurable characteristics of that fish (Winberg et al. 1993). Consequently, the data from the two tanks were pooled. The MSM ranged between 2.4 and 8.9% of the food consumed, the mean CVc was 37.8 + 2.9% (n = 47) and varied between 6.7 and 89.0%. There was a significant correlation between MSM
309 l 0
0 S .
-j
0
1
2
3
4
5
6
kr (%/d) Fig. 1. The relationship between fractional rates of protein consumption (kr) and fractional rates of protein growth (kg) for Atlantic salmon. Relationship described by kg = -0.186 + 0.306kr (n = 57; R2 = 0.673; p < 0.0001). Filled circles indicate fish used to measure protein synthesis.
and CV c (n = 47; r = - 0.320; p < 0.05) which indicated that the fish which consumed a greater share of the food had consistently larger meals. The effect of feeding rank on protein growth efficiency was investigated but there was no relationship between protein growth efficiency and CVc.
Digestive capacity Mean values for the pyloric caecal somatic index (PCSI), trypsin activity and apparent absorption efficiencies for nitrogen and carbon were 3.4 + 0.1% (n = 27), 0.072 ± 0.012/Amol pNA/min/g (n = 18), 80.5 1.3% (n = 33) and 61.9 1.5% (n = 33), respectively. There were no significant relationships between protein growth efficiency and these indices of digestive capacity. Nitrogen and carbon absorption efficiencies were not correlated with consumption rate. Individual fish with high nitrogen absorption efficiencies also had high absorption efficiencies for carbon, but absorption efficiency was not significantly correlated with trypsin activity or the weight of the pyloric caeca.
Protein turnover Time course of protein synthesis White muscle free pool concentrations of phe-
0
1
2
3
4
5
6
kr (%/d) Fig. 2. The relationship between fractional rates of protein consumption (kr) and fractional rates of protein synthesis (ks) for Atlantic salmon. Relationship described by k, = 1.082 + 0.513kr (n = 25; R2 = 0.488; p < 0.001).
nylalanine at 81 and 120 min were not significantly different from each other and had a mean value of 386.1 + 18.3 nmol/g white muscle which was approximately 8 times the normal concentration of phenylalanine in Atlantic salmon white muscle (Carter et al., unpublished data). The phenylalanine-specific radioactivity in the white muscle freepools remained stable over the time course of protein synthesis measurements. The free-pool phenylalanine-specific radioactivities were not significantly different at the different times (81 and 120 min) and had an overall mean value of 1323 25 dpm/nmol compared to 1600 dpm/nmol for the injection solution. The mean protein-bound phenylalanine-specific radioactivity was significantly different at 81 (1.74 0.21 dpm/nmol) and 120 (2.94 + 0.27 dpm/nmol) min and indicated linear incorporation with time. Whole-animal protein turnover Protein accretion increased with increasing protein consumption (Fig. 1). The relationship between fractional rates of protein consumption and growth was described by a simple linear model (y = a + bx) from which the maintenance protein ration (i.e. kr at kg = 0) was calculated to be 0.61 %/d (Fig. 1). Fractional rates of protein synthesis increased with increasing protein consumption (Fig. 2) and the fractional rate of protein synthesis at maintenance was predicted to be 1.39 ± 0.26%/d. The
310 6
a. 7
= I.s 0
4
C
.
6 0 0
5 0
z 2
E U W) -2
-1
0
1
2
4' 0 0
3 24 1
3
0
S
0
i
i
1
2
3
kg (%/d) Fig. 3. The relationship between fractional rates of protein growth (kg) and fractional rates of protein synthesis (ks) for starved and feeding Atlantic salmon. The dotted line indicates the mean rate of protein synthesis for starving salmon. The full line is weighted to take account of the maintenance rate of protein synthesis indicated by an open circle.
4
5
6
kr (%/d)
b. 7
2 0 0
0
5
~
-~
slope of the line indicated that after accounting for the maintenance requirements approximately 51 %, on average, of the consumed protein ration was partitioned into synthesis. Individual nitrogen absorption efficiencies were measured for 33 of the feeding fish including 14 fish used to measure protein synthesis. Despite individual values for the nitrogen absorption efficiency ranging between 65.4 and 87.4% significant relationships between nitrogen absorption and protein growth and synthesis were found. Thus, kg = -0.202 + 0.365ka (n = 43; R 2 = 0.674; p < 0.0001) and k = 1.148 + 0.598k a (n = 19; R 2 = 0.467; p < 0.001). The mean rate of synthesis in the starved fish was 0.96 0.19%/d and showed no relationship with the rate of loss of body protein. In growing fish synthesis increased with increasing growth rate. After accounting for the predicted rate of synthesis at maintenance (1.39%/d) there was a linear relationship between fractional rates of protein growth and synthesis described by k = 2.04kg (n = 20; R 2 = 0.900; p < 0.001). Incorporation of the predicted value for synthesis at maintenance suggested the relationship between fractional rates of protein growth and synthesis was described by ks = 1.39 + 2.0 4 kg (Fig. 3). The concentration of RNA in the whole-animal
i
i
i
i
z
4
U S
3
~ 0~~~~~
0
_
0_~
1
Ix
v, U
1
0 0
2 1
.
0
1
2
~~~~~~~~~~~~~~~~~~~~~~~ 3
ks
4
5
6
(%/d)
C 7 S
0
6 . 0 o.
5
0 0 *
_,M
z
4
0
0 :0
3
M
bf U
0
.
I
0
s oe
2 _
-2
I
I
I
I
I
-1
o
1
2
3
kg (%/d) Fig 4. The relationships between the capacity for protein synthesis (Cs) and a, the fractional rates of protein consumption (kr) b, the fractional rates of protein synthesis (k,) and c, the fractional rates of protein growth (kg) of seawater Atlantic salmon.
was positively correlated with both rates of protein consumption and protein growth. The increase in RNA resulted in a positive correlation between protein consumption and the capacity for protein syn-
311 Individual variation in protein growth efficiency
100 80 S
60 f1z
0
U, _V
40
0 0 -
20
S W:
0
* 0 . u
0
20
.
i.
40
.
.
60
.
80
.
.
100
kg/kr (%) Fig. 5. The relationship between protein growth efficiency (kg/kr) and synthesis retention efficiency (kg/ks) for feeding seawater Atlantic salmon. thesis (Cs ) described by C s = 2.89 + 0.40kc (n = 25; R2 = 0.357; p < 0.002) Fig. 4a). There were also positive correlations between protein synthesis and C s described by C s = 2.82 + 0.44k5 (n = 25; R 2 = 0.238; p < 0.02) and between protein growth and Cs described by C s = 3.55 + 0. 6 2kg (n = 25; R 2 = 0.234; p < 0.02). This was expected since fish eating more synthesised more protein and grew at higher rates (Fig. 4b and c). The positive correlation (p < 0.03) between protein consumption and RNA activity (kRNA) was weaker than between protein consumption and Cs . This suggested that increased protein synthesis, driven by increased consumption rate, was due primarily to an increased capacity for synthesis (increased RNA concentration) but also to an increase in the RNA activity. The mean protein growth efficiency (kg/kc) was 28.3 + 2.8% and varied between 13.8 and 61.7% for individual fish. There were no significant correlations between protein growth efficiency and rates of protein intake, synthesis or degradation, the capacity for protein synthesis or RNA activity. The mean synthesis retention efficiency was 32.1 + 3.3% and also varied considerably between individuals with values ranging between 16.2 and 65.6%. There was a positive correlation between protein growth efficiency and synthesis retention efficiency (n = 20; r = + 0.500; p < 0.05) (Fig. 5).
In order to further investigate the link between protein turnover and protein growth efficiency 6 pairs of fish were selected from the fish on which protein synthesis had been measured. Both fish in each pair were from the same tank and had similar (within 5°70 of each other) protein consumption rates but different growth rates (greater than 5% difference). The pairs were divided into low and high (protein growth) efficiency groups which, as a consequence of the selection procedure, had significantly different rates of protein growth and protein growth efficiency (Table 1). There were no significant differences between the two groups in the remaining parameters (Table 1) although the differences between synthesis retention efficiency was close to significance (p = 0.074). However, rates of synthesis and degradation, synthesis retention efficiency, anabolic stimulation efficiency and RNA activity were higher in the high efficiency fish in 4 or more of the pairs (Table 1). These data suggested (together with the previously demonstrated relationship between efficiencies of synthesis retention and protein growth) that small differences in protein turnover which did not attain the level of statistical significance lead to the difference in protein growth efficiency between the two groups. Degradation rates were almost the same for the two groups but the influx of dietary protein-nitrogen caused a greater anabolic stimulation of synthesis of which more was retained in the high efficiency fish. As a proportion of the absorbed protein-nitrogen the mean anabolic stimulation efficiency was higher (109% compared to 977o) for the high efficiency fish. Thus, a greater proportion of the degradation products were recycled into synthesis in the high efficiency fish whereas they were lost (oxidised) in the low efficiency fish. Discussion Methodology This study utilized a variety of techniques to identify possible correlates of protein growth efficiency.
312 Table 1. Comparison between measures of protein-nitrogen flux and protein metabolism of low and high efficiency groups (6 pairs) Parameter
"Selected" parameters kr (%/d) kg (o/d) kg/kr (%) "Non-selected" parameters k5 (%/d) kd (%/d) kg/ks (%) ks/kr (%) Cs (RNA/protein) kRNA g (protein synthesized/g RNA/d)
Low efficiency (mean SEM)
High efficiency (mean + SEM)
2.72 + 0.37 0.64 + 0.06 21.34 + 1.54
2.70 + 0.38 0.88 + 0.06 34.03 + 3.77
2.47 1.89 24.41 97.06 5.05 5.43
+ 0.37 + 0.33 + 4.13 + 14.67 + 0.83 + 1.05
2.73 1.86 31.90 109.12 4.15 6.76
+ + + + + +
0.22 0.16 1.96 14.25 0.28 0.71
p
Trend
0.615 0.0001 0.016
1/6 6/6 6/6
0.510 0.932 0.074 0.416 0.279 0.301
4/6 4/6 5/6 4/6 1/6 4/6
p, significance of paired t-test; trend, indicates the proportion of pairs in which the value was higher in the high efficiency fish.
The measurement of individual consumption rates using radiography has been critically discussed by McCarthy et al. (1993). There is no ideal method for the measurement of absorption efficiency in fish (reviewed by Talbot 1985). Stripping faeces had the advantage of collecting faeces from individual fish as well as preventing leaching and was selected as the most suitable method. In the present study the relationships between protein consumption, growth and synthesis were described using simple linear models in accordance with the scheme proposed by Houlihan et al. (1993a). Linear models relating protein or nitrogen consumption and growth provide good descriptors of this relationship in mammals (reviewed by Boorman 1980) and fish (Carter et al. 1992b; Houlihan et al. 1993a). The linear model was a more appropriate descriptor of the relationship between protein consumption and growth than a curvi-linear model, since the latter model also indicated a slope of close to one for the Atlantic salmon data. In all previous studies the relationships between protein consumption and synthesis and between protein growth and synthesis were linear (e.g., Houlihan et al. 1988, 1989; Carter et al. 1993). The validity of the flooding dose method to measure in vivo rates of protein synthesis in fish has been discussed previously (e.g., Houlihan et al. 1986, 1988, 1992). Protein synthesis was measured
once at the end of the study and over a few hours. It is possible that measured rates of synthesis might not reflect daily protein synthesis or changes in synthesis over time. Rates of protein synthesis remain relatively stable over 24 h in rat white muscle (Reeds et al. 1986). Since fish white muscle is the largest tissue and accounts for the majority of wholeanimal synthesis it was assumed that whole-animal rates of synthesis demonstrated this same stability (Houlihan et al. 1993b).
Individual variation in protein growth efficiency The present study has revealed inter-individual variation in consumption, growth, apparent absorption efficiency, protein turnover and protein growth efficiency in Atlantic salmon. Although consumption rate was the major determinant of growth rate, individuals differed in their ability to utilize food for growth, irrespective of ration. A major aim of the present study was to ascertain whether differences in feeding rank, digestive capacity, RNA concentration/activity and/or protein turnover were associated with these differences in protein growth efficiency. Some studies have shown that growth efficiency is decreased due to the "physiological stress" of subordination (Koebele 1985; Jobling and Reinsnes 1986; Abbott and Dill 1989). Alternatively a high
313 social position can result in decreased growth efficiency due to increased activity involved in maintaining dominance (Yamagishi et al. 1974). In the present study there was no indication that fish ranked higher in the feeding hierarchy had higher or lower protein growth efficiencies than fish with lower ranks. It was therefore concluded that feeding rank, principally the variation in consumption rate, did not influence protein growth efficiency in this experiment. Absorption efficiency has been demonstrated to decrease with increasing rations of both natural food and formulated pelleted diets (e.g., Elliott 1976; Henken et al. 1985). However, no relationship was observed in some studies (e.g., Birkett 1969; Itawa 1970; Cui and Wootton 1988). In Atlantic salmon individual differences in protein growth efficiency were not correlated with apparent absorption efficiency, trypsin activity or the weight of the pyloric caeca. Studies on Atlantic cod, Gadus morhua L., have indicated that growth rate is positively correlated with the capacity for protein synthesis (RNA/protein ratio) measured in whole-animals (Houlihan et al. 1989) and in white muscle (Foster et al. 1993). In the present study we were able to demonstrate positive correlations between the capacity for protein synthesis and protein consumption, protein synthesis and protein growth for individual fish. These findings supported the hypothesis that increased food consumption rates resulted in an increase in the RNA concentration which provided a mechanism for the resultant increase in synthesis and growth (Mathers et al. 1992; Houlihan et al. 1993a). There were however, no correlations between protein growth efficiency and the capacity for protein synthesis or RNA activity. It is well established that the amount of protein synthesised exceeds the amount of protein retained as growth (Houlihan 1991). The positive correlation between protein growth efficiency and the retention of synthesised protein suggests that this provides a mechanism through which increased protein growth efficiency is achieved. This finding is in agreement with studies on grass carp, Ctenopharyngodon idella (Val.) (Carter et al. 1993) and rainbow trout, Oncorhynchus mykiss (Walbaum)
(McCarthy 1993). Since protein synthesis accounts for a considerable part of the total energy expenditure in fish (Houlihan et al. 1993a, 1993b) costs of growth will therefore be lower in fish which retain a greater proportion of the synthesised protein as growth. Therefore, this study supported the hypothesis that the efficiency with which synthesised protein is retained, irrespective of the amount synthesised, is the important determinant of protein growth efficiency. The partitioning of dietary protein nitrogen between synthesis and oxidation has direct bearing on protein growth efficiency and pairs of salmon with similar consumption rates, but different protein growth efficiencies were selected to investigate this. The high efficiency salmon tended to retain more of the synthesised protein than the low efficiency salmon. The anabolic stimulation efficiency showed that in the high efficiency salmon the rate of synthesis exceeded the amount of protein-nitrogen absorbed by about 10%. This level of anabolic stimulation was not measured in the low efficiency fish. We propose that the greater use of recycled amino acids for protein synthesis and growth, rather than oxidation and excretion, enabled some salmon to retain protein more efficiently. Rainbow trout strains with different nitrogen retention efficiencies can be identified and rates of ammonia excretion were higher in strains with lower nitrogen retention efficiencies (Ming 1985). This suggests differences occur in the partitioning of protein nitrogen between oxidation and growth between fish of the same species and that it has a genetic component. Furthermore, it has been hypothesized that a major influence on individual differences in performance may be due to individual differences in genotype-dependent protein turnover (Hawkins 1991; Tomas et al. 1991). Acknowledgements Our thanks go to Dr. I.D. McCarthy for his critical reading of the manuscript and to David Patterson and Mark Russel (Otter Ferry Salmon Ltd) for feeding and maintenance of the fish. Financial support was from the Ministry of Agriculture, Fisheries and Food.
314
References cited Abbott, J.C. and Dill, L.M. 1989. The relative growth of dominant and subordinate juvenile steelhead trout (Salmo gairdneri) fed equal rations. Behaviour 108: 104-113. Ashford, A.J. and Pain, V.M. 1986. Effect of diabetes on the rates of synthesis and degradation of ribosomes in rat muscle and liver in vivo. J. Biol. Chem. 261: 4059-4065. Austreng, E. 1978. Digestibility determination in fish using chromic oxide marking and analysis of contents from different segments of the gastrointestinal tract. Aquaculture 13: 265-272. Bergot, P., Blanc, J.M. and Escaftre, A.M. 1981. Relationship between number of pyloric caeca and growth in rainbow trout (Salmo gairdneri Richardson). Aquaculture 22: 81-96. Birkett, L. 1969. The nitrogen balance of plaice, sole and perch. J. Exp. Biol. 50: 373-386. Boorman, K.N. 1980. Dietary constraints on nitrogen retention. In Protein Deposition in Animals. pp. 147-166. Edited by P.J. Buttery and D.B. Lindsay. Butterworths, London. Carter, C.G., Houlihan, D.F., McCarthy, I.D. and Brafield, A.E. 1992a. Variation in the food intake of grass carp, Ctenopharyngodon idella (Val.), fed singly or in groups. Aquat. Living Resour. 5: 225-228. Carter, C.G., Houlihan, D.F. and McCarthy, I.D. 1992b. Feed utilization efficiencies of Atlantic salmon (Salmo salar L.) parr: effect of a single supplementary enzyme. Comp. Biochem. Physiol. 101A: 369-374. Carter, C.G., Houlihan, D.F., Brechin, J. and McCarthy, I.D. 1993. The relationships between protein intake and protein accretion, synthesis and retention efficiency for individual grass carp, Ctenopharyngodon idella (Val.). Can. J. Zool. 71: 392-400. Carter, C.G., Houlihan, D.F., Buchanan, B. and Mitchell, A.I. 1994. Growth and feed utilization efficiencies of Atlantic salmon, (Salmo salarL.), fed a diet containing supplementary enzymes. Aquacult. Fish. Manage. (In press) Cui, Y. and Wootton, R.J. 1988. Bioenergetics of growth of a cyprinid, Phoxinusphoxinus: the effect of ration, temperature and body size on food consumption, faecal production and nitrogenous excretion. J. Fish Biol. 33: 431-443. Elliott, J.M. 1976. Energy losses in the waste products of brown trout (Salmo trutta L). J. Anim. Ecol. 45: 561-580. Foster, A.R., Houlihan, D.F. and Hall, S.J. 1993. Effects of nutritional regime on correlates of growth rate in juvenile Atlantic cod (Gadus morhua): Comparison of morphological and biochemical measurements. Can. J. Fish. Aquat. Sci. 50: 502-512. Fange, R. and Grove, D.J. 1979. Fish Digestion. In Fish Physiology Vol. 8, pp. 161-260. Edited by W.S. Hoar, D.J. Randall and J.R. Brett. Academic Press, New York. Garlick, P.J., McNurlan, M.A. and Preedy, V.R. 1980. A rapid and convenient technique for measurin the rate of protein 3 synthesis in tissues by the injection of H phenylalanine. Biochem. J. 217: 507-516. Garlick, P.J., Fern, M. and Preedy, V.R. 1983. The effect of
insulin infusion and food intake on muscle protein synthesis in postabsorptive rats. Biochem. J. 210: 669-676. Gnaiger, E. and Bitterlich, G. 1984. Proximate biochemical composition and caloric content calculated from elemental CHN analysis: a stoichiometric concept. Oecologia (Berlin) 62: 289-298. Hart, P.J.B. and Pitcher, T.J. 1969. Field trials of fish marking using a jet inoculator. J. Fish Biol. 1: 383-385. Hawkins, A.J.S. 1991. Protein turnover: a functional appraisal. Funct. Ecol. 5: 222-233. Henken, A.M., Kleingeld, D.W. and Tijssen, P.A.T. 1985. The effect of feeding level on apparent digestibility of dietary dry matter, crude protein and gross energy in the African catfish, Clariasgariepinus (Burchell, 1822). Aquaculture 51: 1-11. Houlihan, D.F. 1991. Protein turnover in ectotherms and its relationship to energetics. Adv. Comp. Env. Physiol. 7: 143. Houlihan, D.F., McMillan, D.N. and Laurent, P. 1986. Growth rates, protein synthesis, and protein degradation rates in rainbow trout: Effects of body size. Physiol. Zool. 59: 482-493. Houlihan, D.F., Hall, S.J., Gray, C. and Noble, B.S. 1988. Growth rates and protein turnover in cod, Gadus morhua. Can. J. Fish. Aquat. Sci. 45: 951-964. Houlihan, D.F., Hall, S.J. and Gray, C. 1989. Effects of ration on protein turnover in cod. Aquaculture 79: 103-110. Houlihan, D.F., Wieser, W., Foster, A.R. and Brechin, J. 1992. In vivo protein synthesis rates in larval nase, Chondrostoma nasus L. Can. J. Zool. 70: 2436-2440. Houlihan, D.F., Mathers, E. and Foster, A. 1993a. Biochemical correlates of growth rate in fish. In Fish Ecophysiology. pp. 45-71. Edited by J.C. Rankin and F.B. Jensen. Chapman and Hall, London. Houlihan, D.F., Carter, C.G. and McCarthy, I.D. 1993b. Protein synthesis in fish. (in preparation) In Fish Molecular Biology and Biochemistry. Edited by P. Hochachka and T. Mommsen. (In press). Iwata, K. 1970. Relationship between food and growth in young crucian carps, Carassius auratus cuvieri, as determined by nitrogen balance. Jap. J. Lim. 31: 129-151. Jenkins, T.M. 1969. Social structure, position and microdistribution of two trout species (Salmo trutta and Salmo gairdneri)resident in mountain streams. An. Behav. Monog. 2: 56-123. Jobling, M. and Reinsnes, T.G. 1986. Physiological and social constraints on growth of Arctic charr, Salvelinus alpinus: and investigation of factors leading to stunting. J. Fish Biol. 28: 379-384. Jobling, M., Baardvik, B.M. and Jorgensen, E.H., 1989. Investigation of food-growth relationships of arctic charr, Salvelinus alpinusL., using radiography. Aquaculture 81: 367-372. Koebele, B.P. 1985. Growth and the size hierarchy effect: an experimental assessment of three proposed mechanisms; activity differences, disproportional food acquisition, physiological stress. Env. Biol. Fish. 12: 181-188. Lowry, H.O., Rosebrough, N.J., Farr, A.L., and Randall, R.J. 1951. Protein measurement with Folin phenol reagent. J.
315 Biol. Chem. 193: 265-275. Mathers, E.M., Houlihan, D.F. and Cunningham, M.J. 1992. Nucleic acid concentrations and enzyme activities as correlates of growth rate of the saithe Pollachius virens: growth rate estimates of open-sea fish. Mar. Biol. 112: 363-369. McCarthy, I.D. 1993. Feeding behaviour and protein turnover in fish. Ph.D Thesis, University of Aberdeen. McCarthy, I.D., Carter, C.G. and Houlihan, D.F. 1992. The effect of feeding hierarchy on individual variability in daily feeding of rainbow trout, Oncorhynchus mykiss (Walbaum). J. Fish Biol. 41: 257-263. McCarthy, I.D., Houlihan, D.F., Carter, C.G. and Moutou, K. 1993. Variation in individual food consumption rates of fish and its implications to the study of fish nutrition and physiology. Proc. Nutr. Soc. 52: 427-436. McMillan, D.N. and Houlihan, D.F. 1992. Protein synthesis in trout liver is stimulated by both feeding and fasting. Fish Physiol. Biochem. 10: 23-34. Metcalfe, N.B., Huntingford, F.A., Graham, W.D. and Thorpe, J.E. 1989. Early social status and the development of life-history strategies in Atlantic salmon. Proc. Roy. Soc. Lond. B236: 7-19. Millward, D.J., Garlick, P.J., Stewart, J.C., Nnanyelugo, D.O. and Waterlow, J.C. 1975. Skeletal-muscle growth and protein turnover. Biochem. J. 150: 235-243. Ming, F.W. 1985. Ammonia excretion rate as an index for comparing efficiency of dietary protein utilization among rainbow trout (Salmo gairdneri)of different strains. Aquaculture 46: 27-35. Pringle, G.M., Houlihan, D.F., Callanan, K.R., Mitchell, A.I., Raynard, R.S. and Houghton, G.H. 1992. Digestive enzyme levels and histopathology of pancreas disease in farmed Atlantic salmon (Salmo salar). Comp. Biochem. Physiol. 102A: 759-768. Schacterle, G.R. and Pollock, R.L. 1973. A simplified method for the quantitative assay of small amounts of protein in bio-
logic material. Anal. Biochem. 51: 654-655. Reeds, P.J., Palmer, R.M., Hay, S.M. and McMillan, D.N. 1986. Protein synthesis in skeletal muscle at different times during a 24 hour period. Biosci. Rep. 6: 209-213. Sugden, P.H. and Fuller, S.J. 1991. Regulation of protein turnover in skeletal and cardiac muscle. Biochem. J. 273: 21-37. Talbot, C. 1985. Laboratory methods in fish feeding and nutritional studies. In Fish Energetics - New Perspectives. pp. 211-220. Edited by P. Tyler and P. Calow. Croom Helm, London. Talbot, C. and Higgins, P.J. 1983. A radiographic method for feeding studies in fish using metallic iron powder as a marker. J. Fish Biol. 23: 211-220. Tomas, F.M., Pym, R.A. and Johnson, R.J. 1991. Muscle protein turnover in chickens selected for increased growth rates, food consumption or efficiency of food utilization: effects of genotype and relationship to plasma IGF-I and growth hormone. Brit. Poul. Sci. 32: 363-376. Torrissen, K.R. and Shearer, K.D. 1992. Protein digestion, growth and food conversion efficiency in Atlantic salmon and Arctic charr with different trypsin-like isozyme patterns. J. Fish Biol. 41: 409-415. Ulla, 0. and Gjedrem, T. 1985. Number and length of pyloric caeca and their relationship to fat and protein digestibility in rainbow trout. Aquaculture 47: 105-111. Winberg, S., Carter, C.G., McCarthy, I.D., He, Z.-Y., Nilsson, G.E. and Houlihan, D.F. 1993. Feeding rank and brain serotonergic activity in rainbow trout, Oncorhynchus mykiss (Walbaum). J. Exp. Biol. 179: 197-211. Wootton, R.J. 1990. The Ecology of Teleost Fishes. Chapman and Hall, London. Yamagishi, H., Maruyama, T. and Mashiko, K. 1974. Social relation in a small experimental population of Odontobutisobscurus (Temminck et Schelgel) as related to individual growth and food intake. Oecologia (Berlin) 17: 187-202.
