The Open Access Israeli Journal of Aquaculture – BamidgehAs from January 2010 The Israeli Journal of Aquaculture - Bamidgeh (IJA) will be published exclusively as an on-line Open Access (OA) quarterly accessible by all AquacultureHub (http://www.aquaculturehub.org) members and registered individuals and institutions. Please visit our website (http://siamb.org.il) for free registration form, further information and instructions. This transformation from a subscription printed version to an on-line OA journal, aims at supporting the concept that scientific peer-reviewed publications should be made available to all, including those with limited resources. The OA IJA does not enforce author or subscription fees and will endeavor to obtain alternative sources of income to support this policy for as long as possible.Editor-in-ChiefDan Mires Editorial BoardSheenan Harpaz Agricultural Research Organization

Beit Dagan, Israel

Zvi Yaron Dept. of ZoologyTel Aviv UniversityTel Aviv, Israel

Angelo Colorni National Center for Mariculture, IOLREilat, Israel

Rina Chakrabarti Aqua Research LabDept. of ZoologyUniversity of Delhi

Ingrid Lupatsch Swansea UniversitySingleton Park, Swansea, UK

Jaap van Rijn The Hebrew University Faculty of AgricultureIsrael

Spencer Malecha Dept. of Human Nutrition, Food and Animal SciencesUniversity of Hawaii

Daniel Golani The Hebrew University of JerusalemJerusalem, Israel

Emilio Tibaldi Udine UniversityUdine, Italy

Copy Editor Ellen Rosenberg

Published under auspices ofThe Society of Israeli Aquaculture and

Marine Biotechnology (SIAMB), University of Hawaii at Manoa Library and

University of Hawaii Aquaculture Program in association with

AquacultureHub http://www.aquaculturehub.org

ISSN 0792 - 156X

Israeli Journal of Aquaculture - BAMIGDEH.PUBLISHER: Israeli Journal of Aquaculture - BAMIGDEH -Kibbutz Ein Hamifratz, Mobile Post 25210, ISRAELPhone: + 972 52 3965809http://siamb.org.il

The Israeli Journal of Aquaculture – Bamidgeh 55(4), 2003, 243-257. 243

DEFINING ENERGY AND PROTEIN REQUIREMENTS OF GILTHEAD SEABREAM (SPARUS AURATA)TO OPTIMIZE FEEDS AND FEEDING REGIMES

Ingrid Lupatsch *, George Wm. Kissil

National Center for Mariculture, Israel Oceanographic and Limnological Research, P.O.B. 1212, Eilat 88112, Israel

David Sklan

Faculty of Agriculture, Hebrew University of Jerusalem, P.O.B. 12, Rehovot 76100, Israel

(Received 1.10.03, Accepted 30.11.03)

Key words: energy partitioning, feeding standard, growth prediction, maintenance, protein utilization

AbstractEnergy and protein requirements of growing fish can be quantified as the sum of the amounts ofenergy and protein retained as growth plus the amounts simultaneously lost from the body. Therequirement for dietary gross energy and protein can be calculated using the respective effi-ciencies of utilization. Growth of gilthead seabream as a function of body weight and tempera-ture was predicted by the equation: y = 0.024 x BW0.514 x exp0.060T (where y = daily weight gainin g/fish, BW = body weight in g and T = temperature in °C). The gain was determined in fishranging 1-470 g. The energy content of the fish depended on fish weight and rose from 4.7 to11.0 kJ/g body mass as the fish grew whereas the protein content was constant at 176 mg/gregardless of fish weight. The efficiencies of utilization of digestible energy (DE) and digestibleprotein (DP) for maintenance and growth were determined by feeding the fish at increasing feed-ing levels from zero to the maximum voluntary feed intake. The daily requirement of DE for main-tenance was dependent on temperature and determined as (16.6kJ x exp0.055T)/BW in kg0.82.The maintenance requirement for DP was independent of temperature and equaled 0.62g/ BWin kg0.70. The relationship between DE intake and energy gain was linear, constant at kDEg = 0.67and independent of feed intake and temperature. Efficiency of protein utilization for growth var-ied between 0.33 and 0.80 depending on the DP/DE ratio in the diet. The optimal protein uti-lization for protein deposition was estimated at kDPg = 0.47. Using these values allows optimiza-tion of feeding for seabream culture.

* Corresponding author. Tel.: +972-8-6361453, fax: +972-8-6375761, e-mail: Lupatsch@ocean.org.il

244

IntroductionFish culture in the Mediterranean is presentlybased on two major species, the giltheadseabream (Sparus aurata) and the Europeansea bass (Dicentrarchus labrax). Productionof both species is rapidly expanding. In Israel,the principal mariculture species is the gilt-head seabream and most production takesplace in cages in the Gulf of Aqaba in the RedSea. Given the economic importance of feedand feeding in aquaculture, the need to devel-op nutritionally balanced diets, especially con-cerning protein and energy contents, is evi-dent. This is also crucial with regard to theaquatic environment since feed that is uncon-sumed or unavailable to the fish is lost to thesurroundings, resulting in nutrient enrichmentof the water body. Regulatory authorities oftenimpose limits on waste outputs or feed quotasto limit this problem. Therefore, feeding mod-els that supply the exact amounts of energyand protein needed by the fish to realize theirfull growth potential are essential in fish farm-ing. Feeding charts based on nutritionalbioenergetics have been introduced for rain-bow trout (Cho and Kaushik, 1990; Cho,1992). However, information concerninggrowth and digestible energy needs is lackingfor most Mediterranean warm-water species(Kaushik, 1998). A novel approach for deter-mining protein and energy requirements infish is described here as the sum of the dailyenergy and protein requirements for mainte-nance and growth.

The energy requirement for maintenancein fish depends mainly on body size and tem-perature, thus it is proportional to the meta-bolic body weight. On the other hand, thegrowth requirement depends on the amountand composition of the gained weight. Theformal approach to these calculations is:requirement = (M x BW in kgb) + (G x growth),where BW in kgb is the metabolic body weightand M and G are coefficients describing theefficiency of utilization of dietary energy (orprotein) for maintenance or growth.

This review shows how parameters of thefactorial model for quantifying daily energyand protein needs in growing giltheadseabream were derived. Some of the informa-

tion represents an update of earlier trials(Lupatsch et al., 1998, 2001), presented withnew data.

Materials and MethodsExperimental fish. Gilthead seabream werespawned and raised at the National Center forMariculture (NCM). The NCM broodstock wasestablished some 20 years ago from juvenilegilthead seabream caught in the Bardawillagoon on the Sinai coast of theMediterranean Sea. Prior to the start of allexperiments, fish were fed a local commercialseabream diet (Matmor Inc., M.P. Evtach,Israel) consisting of 450 g crude protein, 190g crude lipid and 20.5 MJ gross energy per kgfeed according to feeding tables developed atNCM.

Growth estimates. To test the growthpotential of the seabream throughout theentire growing cycle until close to market size,a data set was established from growth trialsof fish ranging 1-450 g. For each growth trial,fish were graded before stocking so thatgroups were homogenous in size. In these tri-als, the above-mentioned commercial dietwas fed to fish to satiation twice or three timesdaily, depending on the size of the fish, takingcare that no food was left uneaten. Dependingon the fish size, 0.200 or 3 m3 tanks werestocked at densities ranging 0.25-10 kg/m3.The outdoor tanks were supplied with flow-through sea water (41 ppt, flow rate 2 l/min) atambient temperatures ranging 20-28°C. Fishwere weighed every 14 days and the weightgain and daily feed intake were calculated forthe periods between two successive weight-ings. The corresponding body weight for theperiod was the geometric weight of the fishduring the period.

Diet preparation. Experimental diets wereformulated at the NCM, thoroughly mixed in a25 l batch mixer and pelleted using a labora-tory model California Pellet Mill with a steampretreatment unit. After air-drying and whenfed to the fish, the moisture content of thefeeds was about 80-90 g/kg diet.

Composition of weight gain and loss afterstarvation. To determine the composition of

Lupatsch et al.

gilthead seabream of various sizes, fish weresampled throughout the growing cycle. As thecomposition and energy content of growingfish can be influenced by nutritional status aswell as diet composition, estimation of car-cass composition relied on data obtained for44 groups of 10-20 equal-sized fish within aweight range of 1-470 g, fed the commercialdiet. To determine energy and protein lossesafter starvation, half of the fish in each groupwere sacrificed and frozen. The other half wasstocked in 200 l outdoor tanks up to 30 dayswithout being fed. After the starvation period,fish were sacrificed and stored at -20°C untilanalysis. To calculate the energy and proteinlosses after starvation, the body compositionof the initially-sampled fish was consideredrepresentative.

Efficiency of energy and protein utilization.The efficiencies of energy and protein utiliza-tion for maintenance and growth were deter-mined by feeding the seabream diets contain-ing varying amounts of dietary energy, dietaryprotein and digestible protein/digestible ener-gy ratios (DP/DE; Table 1). Five growth trialswere performed with fish of different startingsizes (Table 2). In each trial, fish were fedincreasing amounts of feed from zero to closeto the maximum feed intake. Feed was giventhree times a day at the high feeding levelsand once daily at the low feeding level toensure equal distribution of the food pelletsamong the fish. Digestibility of protein andenergy were determined as described inLupatsch et al. (1997) using fish weighing300-400 g. Chromic oxide (8 g/kg) was addedto the feed as a marker and feces were col-lected by stripping.

Sample preparation. Fish sampled foranalysis were sacrificed by immersing themfor a short time in ~4°C ice water and freezingthem immediately afterwards. While stillfrozen, fish were cut into smaller pieces andground twice using a meat grinder with a 3mm die. Samples for estimation of dry matterwere taken from the ground fish before theremaining homogenate was freeze-dried. Thefreeze-dried samples were again mixed in ablender before further analysis.

Analytical procedures. Identical analyses

were applied for diets and bodyhomogenates. Dry matter was calculated byweight loss after drying 24 h at 105°C. Crudeprotein was measured using the Kjeldahltechnique and multiplying N by 6.25. Crudelipid was measured after chloroform-methanolextraction (Folch et al., 1957). Samples werehomogenized with a high-speed homogenizerfor 5 min and lipid was determined gravimetri-cally after separation and vacuum drying. Ashwas calculated from the weight loss afterincineration of the samples for 24 h at 550°Cin a muffle furnace. Gross energy content wasmeasured by combustion in a Parr bombcalorimeter using benzoic acid as the stan-dard.

Statistics. Allometric equations wereobtained by applying linear regression analy-sis to the logarithmic transformation of thedata in the form of ln y = ln a + b ln x. Theantilog of this expression produces the equa-tion y = axb. Each calculation represents thecombined fish in a single tank. Descriptivestatistics are means±SE unless otherwisenoted. Analyses were carried out with SPSS6.1 for Windows (SPSS Inc., 1989-1992).

ResultsGrowth and feed intake. The daily weight gainof gilthead seabream at different body weightand temperatures is depicted in Fig. 1 (n =108). The relationship between weight gainand body weight was not linear and the resultsbest fit the following allometric function whenthe body weight (BW) of the fish is 1-450 gand the temperature (T) is 20-28°C:

weight gain (g/fish/day) = 0.024±0.006 xBW (in g)0.514±0.020 x exp0.060±0.009T, r2 = 0.93(equation 1)

Rearranging this equation, the weight Wtafter t days can be calculated from the initialweight W0 as follows:

Wt = [W00.486 + 0.01166 x exp0.060T xdays]2.058 (equation 2)

The feed intake can be described in a sim-ilar manner:

feed intake (g/fish/day) = 0.029±0.007 xBW (g)0.598±0.025 x exp0.057±0.011T, r2 = 0.94(equation 3)

Composition of weight gain. The proxi-

245Energy and protein requirements of gilthead seabream

246

mate body composition of the giltheadseabream is shown in Fig. 2. The protein andash concentrations did not change with theincrease in fish size and averaged 176±8.4and 41.5±4.2 mg/g, respectively. In contrast,energy and lipid concentrations rose andmoisture decreased as the fish weightincreased and best fit the following equations(n = 44):

energy (kJ/g) = 4.66±0.23 xBW(g)0.139±0.010, r2 = 0.89 (equation 4)

lipid (mg/g) = 43.3±3.8 x BW (g)0.243±0.020

r2 = 0.90 (equation 5)moisture (mg/g) = 777±11.4 x BW (g)-

0.041±0.003, r2 = 0.82 (equation 6)Metabolic body weights. The daily energy

and protein losses during starvation were cal-culated from comparative slaughter analysis

Lupatsch et al.

Diet A Diet B Diet C Diet D

Ingredients (g per kg)

Fishmeal 900 780 670 550

Cornstarch - 190 260 330

Fish oil 90 20 50 85

Vitamin premix1 10 10 10 10

Mineral premix2 - - 10 25

Proximate analysis (per kg)

Dry matter (g) 920 911 914 914

Crude protein (g) 575 512 436 367

Crude lipid (g) 183 111 121 140

Ash (g) 146 119 113 109

Gross energy (MJ) 21.09 19.44 19.56 19.69

Digestible protein (g)3 485 450 384 323

Digestible energy (MJ)3 17.36 15.68 15.43 15.20

DP/DE ratio (g/MJ) 27.93 28.70 24.90 21.20

1 per kg diet: A 1600 IU; D3 1900 IU; E 150 mg; thiamine 30 mg; riboflavin 45 mg; niacin 15mg; Ca-pantothenate 30 mg; pyridoxine 5 mg; folic acid 11 mg; B12 0.12 mg; K 11 mg; biotin0.25 mg; inositol 150 mg; ascorbic acid 500 mg and choline chloride 3 g.

2 per kg diet: MgO 2.5 g; KI 1.8 mg; CoCO3 0.66 mg; MnO 73.5 mg; ZnO 75 mg; CuCO3 57.5mg; FeCO3 255 mg; NaCl 1.6 g; KCl 3.6 g; Na2SeO3 0.4 mg.

3 after Lupatsch et al., 1997

Table 1. Composition and proximate analysis (as fed) of experimental diets used to estimateenergy and protein demands of gilthead seabream for maintenance and growth.

247Energy and protein requirements of gilthead seabream

Table 2. Structure of gilthead seabream growth trials for evaluating the effects of increasingenergy and protein intakes and various dietary protein/dietary energy ratios.

Trial 1 Trial 2 Trial 3 Trial 4 Trial 5

Fish/tank 19 13 13 18 21

Initial weight (g) 40.1 72.4 94.4 86.1 21.4

Diets (see Table 1) A A A B, C, D B, C, D

Feeding levels low, medium, high low, medium, high, + starvation maximum + starvation

Replicates 2 2 2 - 2

Duration (days) 41 41 41 65 39

Water temperature (°C) 20.5 20.5 20.5 21.5 28

Fish weight (g)

Feed intake

Weight gain

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

0 50 100 150 200 250 300 350 400 450 500

Wei

ght g

ain

and

feed

inta

ke (

g/fis

h/da

y)

Fig. 1. Daily weight gain (g) and feed intake (g) in relation to increasing body weight in giltheadseabream. Corresponding equations 1 and 3 are presented in the text.

248

for each weight group (Fig. 3). The relation-ships between the losses and fish weightwere not linear and results fitted ln - ln func-tions traditionally used by animal nutritioniststo express metabolic body weight (MBW). Theantilog of these functions describes the allo-metric relationship common in biological mea-surements (n = 44). The daily energy loss perfish can be described by the equation:

energy loss in kJ/fish/day = 41.5±2.1/BW(kg)0.82± 0.026, r2 = 0.95 (equation 7)

Incorporating water temperature improvesthe correlation slightly:

energy loss in kJ/fish/day = 11.6±2.0 xexp0.055± 0.008T/BW(kg)0.82±0.024, r2 = 0.97(equation 8)

The daily protein loss per fish can bedescribed by the equation:

protein loss in g/fish/day =

0.40±0.035/BW(kg)0.70±0.04), r2 = 0.89 (equa-tion 9)

Incorporating the temperature:protein loss in g/fish/day = 0.17±0.055 x

exp0.036±0.013T/BW(kg)0.70± 0.036, r2 = 0.90(equation 10)

The expressions kg0.82 and kg0.70 can thusbe described as the metabolic body weightsfor energy and protein, respectively.

Efficiency of utilization of energy and pro-tein. To examine the relationship betweendietary energy and energy gain for differentsized fish for all trials combined, energyintake and energy gain were expressed permetabolic weight of kg0.82. The relationshipbetween dietary energy intake (x, kJ/kg0.82)and energy retained (y, kJ/kg0.82), as seen inFig. 4, perfectly fits a linear function. Trials Ithrough IV, performed at a water temperature

Lupatsch et al.

Fig. 2. Proximate body composition of gilthead seabream at various sizes. Each data point representsanalysis of a group of fish. Corresponding equations 4 to 6 are presented in the text.

Fish weight (g)

0 50 100 150 200 250 300 350 400 450 500

0

100

200

300

400

500

600

700

800

900

1000

0

2

4

6

8

10

12

14

16

18

20

Ash

Com

posi

tion

of li

ve w

eigh

t (m

g/g)

Lipid ProteinEnergyMoisture

Energy content of live w

eight (kJ/g)

of 21°C, differ from Trial V, which was per-formed at 28°C. Therefore two linear equa-tions are described:

Trials I to IV (21°C): y = -32.2±1.9 +0.68±0.02 x r2 = 0.98 (equation 11)

Trial V (28°C) y = -55.8±2.8 + 0.66±0.02 xr2 = 0.99 (equation 12)

The maintenance energy requirementequals DEmaint = 47.35 kJ/kg0.82/day at a tem-perature of 21°C and 84.54 kJ/kg0.82/day at28°C. However, the slopes of both equations,which refer to the efficiency of energy forgrowth, are nearly the same and average kDEg= 0.67.

On the other hand, the relationshipbetween dietary protein intake (x, g/kg0.70)and protein retained (y, g/kg0.70) per day wasbest represented by an exponential curve(Fig. 5), described by the following equation:

y = a x [1 - exp-b (x - c)], where a =

3.16±0.58, b = 0.159±0.036, and c =0.62±0.039, r2 = 0.97 (equation 13)

No difference was detected in theresponses of the fish at different tempera-tures. The maintenance protein requirementwas defined as the point of zero protein gain,reached at an intake of DPmaint = 0.62g/kg0.70/day. Since the relationship betweenprotein intake and protein gain was not linear,a constant protein efficiency value for growthcould not be determined. Therefore, the effi-ciency of utilization of protein (kDPg) for growthbeyond maintenance was calculated for eachlevel of dietary protein intake as follows:

kDPg = protein gain/(dietary protein fed -DPmaint) where DPmaint = requirement ofdietary protein for maintenance, i.e., 0.62 gDP/kg0.70/day (equation 14)

The efficiency of protein utilization calcu-lated for the different levels of dietary protein

249Energy and protein requirements of gilthead seabream

Fig. 3. Energy (kJ/fish/day) and protein (g/fish/day) losses in gilthead seabream after starvation.Corresponding equations 7 and 9 are presented in the text.

Fish weight (kg)

Ene

rgy

loss

(kJ/

fish/

day)

-22

-20

-18

-16

-14

-12

-10

-8

-6

-4

-2

0

-0.45

-0.40

-0.35

-0.30

-0.25

-0.20

-0.15

-0.10

-0.05

0.00

0.000 0.100 0.200 0.300 0.400 0.500

Energy

Protein

Protein loss (g/fish/day)

250

intake ranged 0.33-0.80 kDPg (Fig. 6). Atdietary protein intakes near the maintenancerequirement, the protein efficiency was high-est (kDPg = 0.80) and as the dietary proteinintake increased the efficiency dropped to aslow as kDPg = 0.33. At the inflection point ofthe response curve, the optimal protein effi-ciency value of kDPg = 0.47 was estimated.

DiscussionWeight gain and feed intake. In contrast to ter-restrial animals, fish seem to grow continu-ously. Growth does not cease and reach anasymptote, which may never be attained inaquaculture. Growth rates in aquaculture aretypically described by a specific growth rate(SGR) or absolute growth in g per day.Temperature affects growth, as in all poikilo-therms, which increases as the temperature

increases to an optimum. Above this opti-mum, growth decreases until the upper lethaltemperature is reached. Although SGR andabsolute weight gain depend on feed intakeand water temperature, they mainly dependon the size of the fish. As a result, growthamong groups of fish of different weights can-not be directly compared.

In this study on seabream, the datadescribing the dependence of weight gain (y)on fish weight and temperature best fit anexponential regression (equation 1), howevergrowth could only be described for the tem-perature range used in this study, 20-28°C.According to our growth prediction, seabreamgrow from 1 to 379 g in one year at an aver-age annual water temperature of 23°C, whichconforms to the conditions in the Red Sea.

Composition of gain. Because a large pro-

Lupatsch et al.

Fig. 4. Daily energy retention per unit metabolic body weight of kg0.82 in gilthead seabream fed dietsdiffering in digestible energy and digestible protein contents at two temperatures. Corresponding equations11 and 12 are presented in the text.

DE fed (kJ/kg0.82/day)

-100

-50

0

50

100

150

0 50 100 150 200 250 300

21°C

28°C

Ene

rgy

gain

(kJ

/kg0

.82 /

day)

portion of the energy and protein consumedby the fish is retained as growth, carcass com-position is a major indicator of the energy andprotein requirements of the fish. When mea-suring whole body composition of fish atincreasing sizes, each unit weight gain isassumed to equal the body compositionchange at that size. Dry matter and fat contentare generally the most variable factors in fishand can increase dramatically especially dur-ing the growth period of small fish (>100 g). Inour study, the protein content per g live weightranged 157-185 mg/g fish, averaging 176mg/g fish. On the other hand, lipid increasedas the fish size increased and, consequently,the caloric content ranged from 5 to close to11 kJ/g live weight. Lipid levels rose from 43to 186 mg/g for market-sized seabream ofabout 400 g. With this high lipid content onemight add gilthead seabream to the categoryof ‘fat’ fish, compared to other cultured fish

such as red drum (Thoman et al., 1999), tur-bot (Regost et al., 2001) and rainbow trout(Dias et al., 1999) which contain 69, 38 and146 mg lipid per g live weight, respectively.Therefore, in estimating requirements for tis-sue deposition and growth, wide variationsbetween species, especially in terms of ener-gy, are expected based on the differing tissuecomposition. Seabream require more dietaryenergy per unit weight gain than leaner fishsuch as red drum or turbot. Fish containingmore moisture (less dry matter) require lessenergy for newly deposited growth.

Metabolism and body weight. Metabolicrate is a measure of the metabolic activityrelated to weight and decreases with increas-ing size at a constant temperature. For poik-ilothermic fish, temperature has an importanteffect on metabolism, although its importancemay be species-specific. In this study, metab-olism was measured at the ambient tempera-

251Energy and protein requirements of gilthead seabream

Fig. 5. Daily protein retention per unit metabolic body weight kg0.70 in gilthead seabream fed diets dif-fering in digestible energy and digestible protein content at two temperatures. The corresponding equation13 is given in the text.

-0.5

0.0

0.5

1.0

1.5

2.0

0 1 2 3 4 5

21°C

28°C

DE fed (kJ/kg0.70/day)

Pro

tein

gai

n (g

/kg0

.70 /

day)

252

tures of 20-28°C. Within this range, the effectof temperature was small compared to theeffect of size. Nevertheless, the correlationcoefficients (equations 8, 10) improved whentemperature was included in the equation. Asthe trials were performed in sea water ranging

20-28°C , the metabolic rate can be describedonly for this range.

During starvation at the average tempera-ture of 23°C, the energy loss was 41.5kJ/BW(kg)0.82/day and the protein loss was0.40g/BW(kg)0.70/day (equations 7, 9). Similar

Lupatsch et al.

Fig. 6. Relationship between protein efficiency for growth (kDPg) and protein gain in gilthead seabreamfed varying levels of digestible protein.

DP intake (g/kg0.70/day)

Pro

tein

effi

cien

cy (

k DP

g)

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

-0.5

0.0

0.5

1.0

1.5

0 1 2 3 4 5 6

Pro

tein

gai

n (g

/kg0

.70 /

day)

to the energy metabolism, the best fit betweenprotein loss and body weight in giltheadseabream was reached using a metabolicbody weight, which was calculated with anexponent of 0.70 (equation 9). Data correlat-ing protein loss with fish weight are sparseand most authors assumed a common expo-nent for the relationships between energy lossand protein loss to body weight. In a studywith Epinephelus aeneus and Dicentrarchuslabrax (Lupatsch et al., 2003), energy andprotein losses were described using metabol-ic body weights with the exponents of 0.80and 0.70, respectively. The similarity of thesecoefficients in all species indicates that ener-gy and protein metabolism cannot bedescribed by the same metabolic bodyweight. Moreover, the exponents are closeenough in value to suggest they might becommon for a number of fish species.

Some criticism has been put forward usingstarvation as a means of determining mainte-nance energy requirements for a long periodof time since the rate of loss of body tissues ishigher during the early weeks of fasting thanin later weeks. This, however, would onlyinfluence coefficient a in the expression a xkgb and, further, loss after starvation is onlyan approximation of maintenance energyrequirements.

Efficiency of energy and protein utilization.By using the metabolic body weight of kg0.82,feeding trials with fish of different sizes can becombined to examine the relationship betweendigestible energy intake and energy retention(Fig. 4). This relationship was linear regard-less of feed intake and the DP/DE ratio andthe efficiency of digestible energy utilization forgrowth (kDEg) averaged 0.67 (equations 11,12). The maintenance requirement for energyusing linear regression was higher at highertemperatures, as expected. At 21°C, the main-tenance requirement was 47.4 kJ/kg0.82 and at28°C, 84.6 kJ/kg0.82. In an earlier study withseabream grown at an average temperature of23.5°C (Lupatsch et al., 1998), the mainte-nance requirement was determined as 55.8kJ/kg0.83, which corresponds favorably withthe current study. Assuming the coefficientincorporating the effect of temperature is the

same under maintenance as well as non-fedconditions, the energy maintenance require-ment depending on temperature can bedefined for seabream as DEmaint = (16.6 kJ xexp0.055T)/BW(kg)0.82/day.

In contrast to the linearity of the relation-ship between digestible energy intake andenergy retained, the relationship between thedigestible protein intake and protein gain wasbetter described by an exponential curve (Fig.5, equation 13). The efficiency of protein uti-lization calculated for increasing levels ofdigestible protein intake ranged widelybetween kDPg = 0.33 and 0.80 (Fig. 6). At adigestible protein intake close to maintenancerequirements, the protein efficiency was high-est because it is limiting. As the digestible pro-tein intake increased, the efficiency dropped.At the inflection point of the response curve ofprotein gain versus digestible protein intake,the optimal protein efficiency value was esti-mated at kDCPg = 0.47. Higher protein efficien-cies seem to be reached only accompaniedby a lower overall gain. Note, however, thatthis value is appropriate only for dietary pro-tein with a balanced amino acid profile suchas fishmeal. In practical diets that are limitedin one or more amino acids, the protein effi-ciency will be lower.

In contrast to the temperature dependencyof energy metabolism, increasing the temper-ature from 21°C to 28°C did not seem to affectthe demand for dietary protein and the main-tenance requirement for seabream was deter-mined as DPmaint = 0.62 g/BW(kg)0.70/day.

Practical applications. The results of thisstudy allow us to calculate the daily recom-mended energy and protein intakes for grow-ing gilthead seabream. By defining thedemands of the fish for maintenance andgrowth, a comprehensive budget can bederived that quantifies the energy and proteinneeded by the fish to achieve its growthpotential at any temperature and during anystage in the growth cycle.

Table 3 shows that the proportion of thetotal digestible energy required for mainte-nance increases as the body weight andgrowth rate decrease, influencing the FCR.Higher temperatures have a positive effect on

253Energy and protein requirements of gilthead seabream

254

feed efficiency and, at 26°C, the growthpotential is much greater. Even though ener-gy requirements for maintenance increasewith temperature, this increase is minor com-pared to the higher weight gain.

The absolute protein requirement per dayper fish depends on the size and weight gainof the fish regardless of the dietary digestibleenergy density. Therefore, as demonstrated inTable 4, the desired protein content of a feedvaries according to the digestible energylevel. The DP/DE ratio decreases as the fishgrow and as the growth potential drops due tothe changing energy to protein ratio of theweight gain and the increasing proportion ofenergy used for maintenance. Fish that areable to consume high amounts of feed due toa larger stomach capacity could be fed lowerenergy diets with low protein levels since,based on the calculations in Table 4, thesame amount of protein per day would beconsumed.

Since protein and energy demands con-stantly change, different diets would have tobe formulated for growing gilthead seabream.However, it is unreasonable to expect that alarge number of diets would be used in anyfish culture. In Table 5, a range of diets for var-ious growth periods is suggested (theseremain to be tested under practical condi-tions). It is more practical to feed lower energydiets to small fish with a capacity for high feedintake, as it is difficult to create a 18 MJ ener-gy, 60% protein diet (Table 4). Above 200 g,gilthead seabream production would be betterserved by using a diet with a digestible energycontent of at least 18 MJ/kg - which can beachieved only by incorporating a high lipidlevel - as the amount of 15 MJ feed that wouldhave to be consumed by 300 g fish approach-es the physical limits of gilthead seabream.

ReferencesCho C.Y., 1992. Feeding systems for rainbowtrout and other salmonids with reference tocurrent estimates of energy and proteinrequirements. Aquaculture, 100:107-123.Cho C.Y. and S.J. Kaushik , 1990. Nutritionalenergetics in fish: Energy and protein utilisa-tion in rainbow trout (Salmo gairdneri). World

Rev. Nutr. Dietetics, 61:132-172.Dias J., Corraze G., Arzel J., Alvarez M.J.,Bautista J.M., Lopez-Bote C. and S.J.Kaushik , 1999. Nutritional control of lipiddeposition in rainbow trout and Europeanseabass: effect of dietary protein/energy ratio.Cybium, 23:127-137. Folch J., Lees M. and G.H . Sloane, 1957.Simple method for isolation and purification oftotal lipids from animal tissues. J. Biol. Chem.,226:497-507.Kaushik S.J ., 1998. Nutritional bioenergeticsand estimation of waste production in non-salmonids. Aquat. Living Resour., 11:211-217.Lupatsch I., Kissil G. Wm., Sklan D. and E.Pfeffer, 1997. Apparent digestibility coeffi-cients of feed ingredients and their pre-dictability in compound diets for giltheadseabream (Sparus aurata). Aquacult. Nutr., 3:81-89.Lupatsch I., Kissil G. Wm., Sklan D. and E.Pfeffer, 1998. Energy and protein require-ments for maintenance and growth in giltheadseabream (Sparus aurata.). Aquacult. Nutr.,4:165-173.Lupatsch I., Kissil G. Wm., Sklan D. and E.Pfeffer, 2001. Effects of varying dietary pro-tein and energy supply on growth, body com-position and protein utilization in giltheadseabream (Sparus aurata). Aquacult. Nutr.,7:71-80.Lupatsch I., Kissil G. Wm. and D. Sklan,2003. Comparison of energy and protein effi-ciency among three fish species: giltheadseabream (Sparus aurata), Europeanseabass (Dicentrarchus labrax) and whitegrouper (Epinephelus aeneus): energy expen-diture for protein and lipid deposition.Aquaculture, 225:175-189.Regost C., Arzel J., Cardinal M., Robin J.,Laroche M. and S.J . Kaushik, 2001. Dietarylipid level, hepatic lipogenesis and flesh qual-ity in turbot (Psetta maxima). Aquaculture,193:291-309.Thoman E.S., Davis D.A. and C.R . Arnold,1999. Evaluation of growout diets with varyingprotein and energy levels for red drum(Sciaenops ocellatus). Aquaculture, 176:343-353.

Lupatsch et al.

255Energy and protein requirements of gilthead seabreamT

able

3.

Rec

omm

ende

d su

pply

of

dige

stib

le e

nerg

y fo

r gi

lthea

d se

abre

am g

row

n in

wat

er t

empe

ratu

res

of 2

1°C

or

28°C

.

1ac

cord

ing

to e

quat

ion

1 in

Mat

eria

ls a

nd M

etho

ds

2di

gest

ible

ene

rgy

requ

ired

for

mai

nten

ance

= 1

6.6

x ex

p (0

.055

Tem

pera

ture

) /bod

y w

eigh

t (kg

)0.8

2

3re

tain

ed e

nerg

y =

bod

y en

ergy

x w

eigh

t gai

n (e

quat

ion

4)4

dige

stib

le e

nerg

y re

quire

d fo

r gr

owth

, usi

ng e

ffici

ency

for

grow

th k

DE

g=

0.6

75

tota

l dig

estib

le e

nerg

y re

quire

d fo

r m

aint

enan

ce a

nd g

row

th6

dige

stib

le e

nerg

y co

nten

t of f

eed

= 1

7 M

J/kg

Bod

y w

eigh

t (g/

fish)

1050

100

300

Met

abol

ic b

ody

wei

ght (

kg0.

82)

0.02

30.

086

0.15

10.

373

Wat

er te

mpe

ratu

re (

°C)

2026

20

2620

2620

26

Pre

dict

ed w

eigh

t gai

n (g

/day

)10.

260.

370.

600.

850.

851.

221.

492.

14

DE

mai

nt(k

J/fis

h/da

y)2

1.14

1.59

4.28

5.95

7.55

10.5

018

.58

25.8

5

RE

(kJ

/fish

/day

)31.

672.

394.

786.

857.

5110

.76

15.3

922

.06

DE

grow

th(k

J/fis

h/da

y)4

2.49

3.57

7.12

10.2

011

.19

16.0

422

.93

32.8

7

DE

m+

g(k

J/fis

h/da

y)5

3.63

5.16

11.3

916

.15

18.7

426

.54

41.5

258

.72

% o

f tot

al D

E u

sed

for

mai

nten

ance

31.5

30.8

37.5

36.8

40.3

39.6

44.8

44.0

Fee

d in

take

(g/

fish/

day)

60.

210.

300.

670.

951.

101.

562.

443.

45

FC

R0.

810.

811.

121.

121.

291.

281.

641.

61

256 Lupatsch et al.

Tab

le 4

. Rec

omm

enda

tions

for

diet

ary

ener

gy a

nd p

rote

in s

uppl

y fo

r gr

owin

g gi

lthea

d se

abre

am a

t 23°

C, w

hen

form

ulat

ing

feed

s w

ithdi

ffere

nt e

nerg

y co

nten

ts.

1 ac

cord

ing

to e

quat

ion

1 2

see

note

s 2,

4,5

in T

able

3.

3di

gest

ible

pro

tein

req

uire

d fo

r m

aint

enan

ce =

0.6

2 g/

kg0.

70/d

ay

4re

tain

ed p

rote

in =

bod

y pr

otei

n (1

76 m

g/g)

x w

eigh

t gai

n 5

dige

stib

le c

rude

pro

tein

req

uire

d fo

r gr

owth

usi

ng e

ffici

ency

for

grow

th k

DP

g=

0.4

7 6

tota

l dig

estib

le p

rote

in r

equi

red

for

mai

nten

ance

and

gro

wth

Bod

y w

eigh

t (g/

fish)

1050

100

300

Wei

ght g

ain

(g/d

ay)1

0.31

0.71

1.02

1.79

Ene

rgy

requ

irem

ent

Met

abol

ic b

ody

wei

ght (

kg0.

82)

0.02

30.

086

0.15

10.

373

DE

mai

nt(k

J/fis

h/da

y)2

1.35

5.04

9.33

22.9

7

DE

grow

th(k

J/fis

h/da

y)2

2.98

8.52

13.4

027

.46

DE

m+

g(k

J/fis

h/da

y)2

4.33

13.5

622

.73

50.4

3

Pro

tein

req

uire

men

t

Met

abol

ic b

ody

wei

ght (

kg0.

70)

0.04

00.

123

0.20

00.

431

DP

mai

nt(g

/fish

/day

)30.

025

0.07

60.

124

0.26

7

RP

(g/

fish/

day)

40.

055

0.12

50.

179

0.31

5

DP

grow

th(g

/fish

/day

)50.

117

0.26

70.

381

0.67

1

DP

m+

g(g/

fish/

day)

60.

141

0.34

30.

505

0.93

8

Fee

d fo

rmul

atio

n

DE

den

sity

of f

eed

(per

kg)

15 M

J18

MJ

15 M

J18

MJ

15 M

J18

MJ

15 M

J18

MJ

Fee

d in

take

(g/

fish/

day)

0.29

0.24

0.90

0.75

1.49

1.24

3.29

2.74

DP

con

tent

in fe

ed (

g/kg

)48

658

638

145

733

940

728

534

2

FC

R0.

940.

771.

271.

061.

461.

221.

841.

53

DP

/DE

rat

io (

g / M

J)32

.532

.525

.425

.422

.622

.619

.019

.0

257Energy and protein requirements of gilthead seabream

Table 5. Proposed diet formulations and practical feeding tables for gilthead seabream dur-ing all stages of the growth period (at 23°C).

* assuming digestibility of protein 85% and of energy 80%

1-5 540 19.8 0.16 0.16 25 1.00

5-10 500 19.8 0.26 0.28 19 1.08

10-50 480 20.0 0.51 0.61 78 1.20

50-100 450 20.2 0.86 1.18 58 1.37

100-200 420 20.7 1.23 1.81 81 1.47

200-300 400 21.2 1.61 2.58 62 1.60

300-400 400 22.0 1.92 3.21 52 1.67

Weight(g)

Feed composition(per kg feed)*

Weight gain

(g/fish/day)

Feed intake(g/fish/day)

Days ofgrowth

FCR

Crude protein

(g)

Gross energy(MJ)