244 (2005) 251–262 www.elsevier.com/locate/aqua-online

Evaluation of practical diets for the Caribbean spiny (Latreille, 1804): effects of protein sources on substrate metabolism and digestive proteases

Erick Pereraa,*, Iliana Fragab, Olimpia Carrilloc, Eugenio Dı´az-Iglesiasd, Rau´l Cruza, Marysabel Ba´eza, Germa´n S. Galicha aCenter for Marine Research, Havana University, 16 St. No. 114 between 1st. and 3rd. St. Miramar, Playa Zip Code, 11300, Havana, Cuba bCenter for Fishery Research, Ministry of Fishery Industry. 5th. Ave. and 248 St. Barlovento, Santa Fe, Playa, Havana, Cuba cFaculty of Biology, University of Havana, 25 St. between J and I, Plaza, Havana, Cuba dAquaculture Department, CICESE, Km 107 Tijuana-Ensenada, Baja California, Mexico Received 2 April 2004; received in revised form 19 November 2004; accepted 23 November 2004

Abstract

The formulation of artificial diets is a fundamental issue in the development of lobster aquaculture. The impact of inclusion of clam, squid, chiton, and high-quality fish meal in a local fish meal-based diet on substrate metabolism and digestive proteases in juveniles Panulirus argus was evaluated by using two dietary protein levels, 25% and 35%, with 20% protein the basal level. The oxygen consumption/ammonia excretion ratio showed that fed clam and chiton diets used protein for oxidation at the two dietary protein levels, while lobsters fed high-quality fish meal and squid diets used protein–lipid at 25% protein. Higher protein levels led to an increase in the contribution of protein to energy metabolism. Digestive protease activity increased with squid meal in diets. Our results suggest that the inclusion of squid and high-quality fish meal in local fish meal diets increases the nutritional value of the diet for P. argus juveniles and that squid enhances digestive proteases activities in the hepatopancreas. However, grow-out trials are needed to fully demonstrate the growth-enhancing effect of these protein sources in formulated diets for juveniles P. argus and to decide whether the growth rate increase is sufficient to warrant using these protein sources. D 2004 Published by Elsevier B.V.

Keywords: ; Panulirus argus; Protein sources; O/N ratio; Metabolism; Digestive proteases

1. Introduction

Spiny lobsters are among the worlds most valuable * Corresponding author. Tel.: +537 2030617; fax: +537 and support some of the largest commercial 2042380. fisheries in the world. The spiny lobster Panulirus E-mail address: [email protected] (E. Perera). argus is the most important fishery resource in the

0044-8486/$ - see front matter D 2004 Published by Elsevier B.V. doi:10.1016/j.aquaculture.2004.11.022 252 E. Perera et al. / Aquaculture 244 (2005) 251–262

Great Caribbean and the of highest commer- the inconsistency in quality of those foods leads to cial value in Cuba, accounting for US$90–100 million reduced growth. At higher scales, the restricted access per year. There is an intense interest in the aquaculture to the appropriated natural items and their storage of spiny lobsters all over the world due to their high would become a problem. In addition, if caged value in international markets and because most aquaculture system is used, as in the case of most of stocks are exploited at or above sustainable levels. the Asiatic countries mentioned above, cages can Despite numerous advances in culturing phyllo- release entirely untreated wastewater to the environ- soma larvae (Kittaka, 1997; Ritar et al., 2002; Smith ment. Thus, formulated diet development is one of the et al., 2003), hatchery production of spiny lobsters is key issues in successful aquaculture of lobsters in not yet technically feasible. However, the develop- term of both profitability and sustainability. ment of programs to catch large numbers of pueruli Natural foods have proved to promote higher has motivated the growout of spiny lobsters in some growth than pelleted diets in spiny lobsters (Crear Asiatic countries including Vietnam, Philippines, et al., 2000, 2002; Smith et al., 2003, Thomas et al., Indonesia, India, Thailand, Burma, China, Taiwan, 2003). It has been suggested that the differences in Malaysia, Tahiti, and Singapore (Jeffs and David, spiny lobster growth between fresh foods and 2002) and New Zealand (Jeff and Hooker, 1999). formulated diets are probably due to differences in Commercially, lobsters are successfully fed fresh efficiency of the utilization of protein (Crear et al., fish, , and mollusks, although sometimes 2000; Tolomei et al., 2003). The most commonly used

Table 1 Formulations (%) and proximate composition of the experimental diets Ingredient Experimental diets 12345678 Basal diet (%) Gelatin 5.78 Whole wheat 32.87 Vitamin and mineral premix 5 Soya oil 2 Shark oil 2 Calcium carbonate 2 Dicalcium phosphate 1 Vitamin C 0.10 Vitamin E 0.25 Agglutinanta 2 Local fish meal 17 Stuff b 22 7 23 10 23 9 23 9 Total 92 77 93 80 93 79 93 79 Crude protein in basal diet (%) 19.99 Clam meal 8 23 Squid meal 7 20 Chiton meal 721 High-quality fish meal (from Chile) 721 Total 100 100 100 100 100 100 100 100

Proximate Crude protein (%) 25.14 34.94 25.24 34.99 25.05 35.17 24.9 34.7 Crude fat (%) 6.24 6.9 6.24 6.37 6.8 7 6.2 6.23 Carbohydrate (%) 23.75 24.44 24.1 24.73 24.03 24.59 23.99 24.5 Gross energy (kJ/g) 11.6 14.0 11.6 13.8 11.8 14.5 11.5 13.7 P/E ratio(mg protein/kJ) 21.7 25.0 21.8 25.4 21.2 24.3 21.7 25.3 a Carboxymethylcellulose. b Talc Protein sources: clam (Arka zebra), squid (Loligo sp.), chiton (Acanthopleura granullata). E. Perera et al. / Aquaculture 244 (2005) 251–262 253 metabolic indicator to assess protein utilization in The formulated diets tested are showed in Table 1. ammonotelic is the oxygen consumption/ The experimental diets were prepared by thoroughly ammonia excretion atomic ratio (O/N; Mayzaud and mixing the dry ingredients with oil and then adding Conover, 1988). The catabolism of protein produces water until the consistency was of stiff dough. This O/N ratios of 3–16, protein and some lipid of 17–50, was passed through a grinder and the resulting strings equivalent amounts of protein and lipid of 50–60 and were air dried at 60 8C and broken into a convenient higher values correspond to the oxidation of lipid and pellet size. Diets were stored at 20 8C until used. carbohydrates (Mayzaud and Conover, 1988). The efficiency of protein utilization depends, in 2.2. Oxygen consumption, ammonia excretion, and part, of the digestive process itself. Proteins ingested O/N ratio are hydrolyzed in the digestive tract into their constitutive amino acids by proteases that play the In this experiment, 80 (10 per diet) intermolt central role in protein digestion. This is of particular juvenile P. argus (120.3F7.73 g wet weight, meanF significance in crustaceans since they lack the acidic standard error) were used. Lobsters were deprived of phase of digestion. The activities of digestive proteases in crustaceans have proven to be affected Table 2 by the protein composition of diet (Le Moulllac et Summary of results of two-way ANOVAs for oxygen consumption, al., 1996). total ammonia excretion, and O/N ratio in juvenile Panulirus argus Understanding the physiological basis of fed experimental diets observed growth in terms of anabolic and catabolic Diet proteins Source of df F P processes will then enable informed decisions to be variation made on the modification of diet. The aim of this Oxygen consumption study was to use O/N ratio and digestive proteases Clam Level 1 3.52 n.s. N0.05 Hour 5 9.56*** V0.001 activity to evaluate the inclusion of different sources Squid Level 1 3.51 n.s. N0.05 of protein: clam, squid, chiton, and high- Hour 5 2.97* V0.05 quality fish meals in local fish meal-based diets for Chiton Level 1 0.92 n.s. N0.05 juvenile P. argus. Hour 5 0.11 n.s. N0.05 Fish Level 1 23.29*** V0.001 Hour 5 5.33*** V0.001

2. Materials and methods Ammonia excretion Clam Level 1 2.94 n.s. N0.05 2.1. Animals, experimental conditions, and diets Hour 5 2.93 n.s. N0.05 Squid Level 1 14.68*** V0.001 Hour 5 0.32 n.s. N0.05 For this study, 125 randomly selected late juvenile P. Chiton Level 1 5.36* V0.05 argus were transported to the Laboratory of Physiol- Hour 5 0.66 n.s. N0.05 ogy, Center for Marine Research, University of Fish Level 1 1.34 n.s. N0.05 Havana, Cuba, from Golfo de Batabano´, in the south- Hour 5 2.77* V0.05 western part of the Cuban archipelago. In the labo- O/N ratio ratory, lobsters were placed individually in a 90-l flow- Clam Level 1 0.002 n.s. N0.05 through tank connected to a 3-ton recirculating aquatic Hour 5 1.75 n.s. N0.05 system with biological–mechanical filtration and aer- Squid Level 1 8.93** V0.01 ated seawater. Experiments were carried out after Hour 5 2.62* V0.05 lobsters were conditioned for 1 week to new conditions Chiton Level 1 0.32 n.s. N0.05 Hour 5 0.84 n.s. N0.05 and formulated diets. During the experiments, the V F Fish Level 1 16.31*** 0.001 conditions were (mean standard error) temperature of Hour 5 4.50*** V0.001 F F x 24.2 0.20 8C, salinity of 36.1 0.09 ,ammonia * Indicates statistical differences for PV0.05. concentration of 0.07F0.009 mg/l, oxygen concen- ** Indicates statistical differences for PV0.01. tration of 5.57F0.010 mg/l, and pH of 7.9F0.02. *** Indicates statistical differences for PV0.001. 254 E. Perera et al. / Aquaculture 244 (2005) 251–262 food for 1 day prior to experiment. For oxygen of the lobster inside and, therefore, the oxygen consumption and ammonia excretion, a continuous concentration within the chambers did never fall below flow respirometer was used (Martı´nez-Otero and Dı´az- 4 mg/l throughout the experiment. Corrections for Iglesias, 1975). Lobsters were introduced in 1.5 l metabolic activity of were made using similar chambers at 07:00 in which they were acclimated for 2 measurements from an empty chamber. Measurements h before the experiment was conducted. The chambers were made at time 0 (09:00, fasting) and 1, 2, 3, 4, 5 h were submerged in a water bath to standardize temper- after food was given and ingested, by an oxymeter ature among them. Oxygen consumption and ammonia Yellow Sprint Instruments, Model 50B (0.01 mg/l excretion rates were estimated from the equations: accuracy) for oxygen concentration and by an ammo-  niometer Orion, Model 290A (0.01 mg/l accuracy) for 4 ½O2 input ½O2 output F ammonia concentration. VO ¼ 2 W Oxygen consumption and ammonia excretion data ˙  were weight-standardized, standardized to 28 8C by the 4 Winberg procedure (Winberg, 1956), and expressed in ½U input ½U output F U ¼ milligrams of oxygen or ammonia per kilogram per W hour at 28 8C (mg/kg/h, 28 8C). The O/N atomic ratio where [O2]input is the oxygen concentration in the was calculated for each lobster from O/N=(oxygen chamber input, [O2]output is the oxygen concentration in consumption62.5)/(ammonia excretion48.47) the chamber output, [U]input is the total ammonia (Mayzaud and Conover, 1988). concentration in the chamber input, [U]output is the total Lobsters were always fed at a rate below that ammonia concentration in the chamber output, W is the required to reach satiation (2% of lobster wet weight wet weight of lobsters, and F represents the water flow according to previous experiments, unpublished) what rate through the chamber. The former comes from a is recommended to minimize differences in food reservoir tank and was regulated according to the size consumption (Smith et al., 2003).

Table 3 Oxygen consumption rate and total ammonia excretion rate of juvenile Panulirus argus fed experimental diets Diets Protein Unfed Fed (%) H1 H2 H3 H4 H5 H6 1 1 Oxygen consumption (mg O2 kg h wet weight basis) Clam 25 40F3.1 70F5.4 67F3.8 74F4.2 77F4.7 72F5.2 35 49F3.8 78F6.4 59F5.6 71F5.2 75F8.4 56F5.3 Squid 25 39F6.2 71F10.0 64F10.0 78F8.0 114F30.8 77F18.0 35 65F9.0 86F14.2 88F11.4 99F14.7 107F10.5 101F15.9 Chiton 25 55F8.7 92F13.3 96F14.8 91F16.6 87F10.0 67F11.4 35 63F8.1 80F9.7 83F10.1 81F11.9 75F9.9 61F6.0 Fish 25 65F10.3 128F6.3 119F19.9 125F6.6 97F14.8 57F14.2 35 49F5.5 48F11.0 73F6.6 84F10.9 75F11.6 66F13.6

+ 1 1 Ammonia excretion (mg NH3–NH4 kg h wet weight basis) Clam 25 4F1.0 7F1.3 9F1.9 10F1.7 8F1.8 6F1.1 35 7F1.4 15F2.4 7F1.6 9F1.2 8F1.5 6F1.7 Squid 25 10F2.7 8F2.5 8F2.3 4F1.1 7F1.4 5F1.6 35 7F1.7 10F2.6 15F2.0 13F2.5 12F1.8 15F1.7 Chiton 25 12F3.3 18F3.7 16F4.7 10F3.2 11F3.0 7F2.1 35 7F2.0 8F1.2 7F1.4 7F1.5 5F1.0 5F1.0 Fish 25 12F1.6 13F2.6 10F3.1 6F3.0 4F0.9 5F2.0 35 8F1.3 7F0.7 7F1.1 10F2.1 8F0.9 5F1.2 Values are meansFstandard error. H1 represents resting metabolism and H2–H6 represent feeding metabolism. E. Perera et al. / Aquaculture 244 (2005) 251–262 255

2.3. Total protease activity in hepatopancreas were centrifuged at 10,600g for 3 min. Then, 83 Al of supernatant was transferred to a microplate, and For examining protease activity in the hepatopan- 166 Al of 0.5 M NaOH and 50 Al Folin–Ciocalteau creas, 45 (5 per diet) intermolt lobsters (197.6F12.4 g reagent (1:3) were added. Optical density of liberated wet weight, meanFstandard error) were fed exper- tyrosine was measured using a microplate reader at imental formulated diets (2% fresh body weight) and 620 nm. Protein concentration measurements were fresh chiton (9% fresh body weight) twice a day over performed by the Biuret method. 30 days. Then, the animals were dissected and hepatopancreas samples were taken and frozen at 2.4. Experimental designs and statistical analysis 20 8C for later analysis. The samples were thomogenized in distilled water at 4 8C during 30 s Data obtained from lobsters that did not ingest the (5 ml dH2O/g tissue) and centrifuged at 10,600g for food were not used for statistical analysis. All data 30 min at 4 8C to eliminate lipids and tissue debris. were checked for normality and homogeneity of Supernatants were used for determination of enzy- variance using Kolmogorov–Smirnov and Levene’s matic activity by the Anson (1938) procedure test, respectively. modified for a microplate. For oxygen consumption, ammonia excretion, and Briefly, 2% bovine hemoglobin (pH 8.0) denatured O/N ratios, each diet was analyzed separately. with urea was used as the substrate. Two hundred and Randomized complete block designs were used, fifty microliters hemoglobin and 50 Al of hepatopan- where the treatments were time after ingestion (6 h), creas tissue were incubated for 10 min at 37 8C. The the blocks were the protein levels in each diet (25% reaction was stopped with 500 Al 5% TCA, and tubes and 35%), and the replicates were the individual

(A) (B) 90 2 2 140 VO2 (25%)= -2.56T +25.13T+20.22 (R =0.83) 2 2 VO2 (35%)= -2.02T +21.34T+46.77 (R =0.95) 80 120 70 /Kg/h) /Kg/h) 100 2 2 60 80

2 2 (mg O 50 (mg O 60

2 VO (25%)= -2.71T +24.31T+22.52 (R = 0.86) 2 2 2 2 VO2 (35%)= -2.53T +18.99T+36.71 (R = 0.41)

VO 40

VO 40 25% 30 25% 20 H1 H2 H3 H4 H5 H6 35% H1 H2 H3 H4 H5 H6 35% Time (hours) Time (hours) (C) (D) 2 2 VO (25%)= - 5.71T +41.12T+23.87 (R =0.91) 2 2 120 2 160 VO2 (25%)= -10.61T +70.62T+12.09 (R =0.91) 2 2 2 2 VO2 (35%)= -3.41T +23.06T+44.68 (R =0.97) VO (35%)= -3.13T +26.91T+19.04 (R =0.76) 110 140 2 100 120 /Kg/h) /Kg/h) 2 2 90 100 80 80 70 (mg O (mg O 2 2 60 60 VO VO 50 40 25% 25% 40 20 H1 H2 H3 H4 H5 H6 35% H1 H2 H3 H4 H5 H6 35% Time (hours) Time (hours)

Fig. 1. Oxygen consumption rate (meanFstandard error) of juveniles P. argus after being fed clam (A), squid (B), chiton (C), and fish meal (D) diets. Each protein source was tested at 25% (n) and 35% (E) total dietary protein. In each graph H1 represents resting metabolism and H2–H6 represent feeding metabolism. 256 E. Perera et al. / Aquaculture 244 (2005) 251–262 lobsters (10). Two-way analyses of variance range test was used to determine differences in O/N (ANOVA, PV0.05) was performed to detect differ- ratio and protease activity in hepatopancreas. The ence among treatments and between blocks. Addi- Software package Statistica 6.0 Soft, was used for all tionally, regression approach (Ruohonen and tests performed. Kettonen, 2004) was used to describe the pattern of oxygen consumption and ammonia excretion over time. Data were submitted to a stepwise polynomial 3. Results regression analysis and R2 values were calculated as a measure of relative goodness of fit of a regression The oxygen consumption increased ( PV0.05) after curves. Times for maximum responses were calcu- ingestion of diets in which local fish meal was lated from the b1 and b2 coefficients according to Zar supplemented with clam, squid, and high-quality fish (1984). meal (Table 2). In diets with clam and squid meals, For the experiment on proteases activity in the there were no differences between protein levels hepatopancreas, a randomized complete design was (Table 2). used, where the treatments were the diets and 5 Oxygen consumption rates of fasted and fed replications were made (individual lobsters). Tripli- lobsters are showed in Table 3. The time–response cate measurements were performed for each replicate. curves for oxygen consumption are showed in Fig. 1. One-way analysis of variance (ANOVA, PV0.05) was The relationship between oxygen consumption (VO2) performed to detect differences among treatments. and time (T) in lobsters fed 25% clam was described 2 Student–Newman–Keuls (SNK, PV0.05) multiple by the quadratic regression VO2=2.701T +

(A) (B) 2 2 20 U (25%)= 0.34T -3.61T+14.73 (R =0.92) 20 2 2 2 2 U (25%)= -0.63T +4.72T+0.36 (R = 0.97) 18 U (35%)= -0.64T 2+5.62T+2.01 (R =0.80) 18 2 2 U (35%)= -0.40T +2.02T+7.87 (R =0.28) 16 /Kg/h) +

/Kg/h) 16 4 + 14 4 14 12

12 +NH 3 +NH 3 10 10 8 8 6 6 4 4 25% U (mg NH

U (mg NH 25% 2 2 35% H1 H2 H3 H4 H5 H6 35% H1 H2 H3 H4 H5 H6 Time (hours) Time (hours) (C) (D) 30 2 2 2 2 18 U (25%)= -0.001T -1.93T+15.43 (R =0.85) U (25%)= -0.74T +3.66T+11.02 (R =0.71) 2 2 2 2 16 U (35%)= -0.15T +0.66T+6.97 (R =0.54) 26 U (35%)= -0.22T +1.17T+6.03 (R =0.80) /Kg/h)

/Kg/h) 14 + + 22 4 4 12 18 10 +NH +NH 3 3 14 8 6 10 4 6 2 U (mg NH

U (mg NH 25% 25% 2 0 H1 H2 H3 H4 H5 H6 35% H1 H2 H3 H4 H5 H6 35% Time (hours) Time (hours)

Fig. 2. Ammonia excretion rate (meanFstandard error) of juveniles P. argus after being fed clam (A), squid (B), chiton (C), and fish meal (D) diets. Each protein source was tested at 25% (n) and 35% (E) total dietary protein. In each graph H1 represents resting metabolism and H2–H6 represent feeding metabolism. E. Perera et al. / Aquaculture 244 (2005) 251–262 257

24.308T+22.521, R2=0.86 and for those fed 35% lobsters fed 35% fish meal the quadratic regression 2 2 2 clam by the quadratic regression VO2=2.529T + was VO2=3.123T +26.908T+19.040, R =0.76. 18.993T+36.708, R2=0.41. These quadratic regres- These equations predicted that the maximum oxygen sions predicted the maximum oxygen consumption to consumption was 2.3 and 3.4 h after ingestion of 25% be 3.5 and 2.7 h after ingestion of 25% and 35% clam, and 35% fish meal, respectively. respectively. Ammonia excretion rates of fasted and fed lobsters The quadratic regressions of time on VO2 in are showed in Table 3. After ingestion, ammonia 2 lobsters fed 25% squid (VO2=2.565T +25.126T+ excretion showed a non-significant trend to increase 2 20.222, R =0.84) and those fed 35% squid (VO2= for most of the diets. Significant variations ( PV0.05) 2.017T2+21.342T+46.766, R2=0.95) predicted the in ammonia excretion were observed only for lobsters highest oxygen consumption to be 3.9 and 4.3 h after fed fish meal diets (Table 2). Differences in ammonia ingestion, respectively. excretion between protein levels were detected only The response of VO2 to time for lobsters fed 25% for lobsters fed squid diet and chiton diets (Table 2). chiton was described by the quadratic regression The time–response curves for ammonia excretion 2 2 VO2=5.715T +41.125T+23.870, R =0.91 while for are showed in Fig. 2. The relationship between lobsters fed 35% chiton the quadratic regression was ammonia excretion (U) and time (T) in lobsters fed 2 2 VO2=3.407T +23.059T+44.685, R =0.98. These 25% clam was described by the quadratic regres- equations predicted that the maximum oxygen con- sion U=0.633T2+4.717T+0.364, R2=0.97 and for sumptions were 2.6 and 2.4 h after ingestion of 25% those fed 35% clam by the quadratic regression U= and 35% chiton, respectively. 0.396T2+2.021T+7.868, R2=0.28. These quadratic The effect of time on VO2 in lobsters fed 25% fish regressions predicted the maximum ammonia excre- meal was described by the quadratic regression VO2= tion to be 2.7 and 1.5 h after ingestion of 25% and 10.613T2+70.621T+12.089, R2=0.91 whereas for 35% clam, respectively.

Fig. 3. Oxygen consumption: ammonia excretion atomic ratio (O/N) (meanFstandar error) of juveniles P. argus after fed the clam (A), squid (B), chiton (C), and fish meal (D) diets. Each protein source was tested at 25% (n) and 35% (E) total dietary protein. Arrows indicate time of feeding. The dotted line represents the O/N limit between protein and protein–lipid metabolism (see Materials and methods). Different letters indicate statistical differences according to SNK test ( PV0.05). 258 E. Perera et al. / Aquaculture 244 (2005) 251–262

The quadratic regressions of time on U in lobsters from lobsters fed these two diets lies in the theoretical fed25%squid(U=0.345T2+3.614T+14.727, range for pure protein catabolism for both protein R2=0.92) and those fed 35% squid (U=0.636T2+ levels tested (Fig. 1A and B). The O/N ratio changed 5.617T+2.006, R2=0.80) predicted the highest ammo- following inclusion of squid and high-quality fish nia excretion to be 3.4 h after ingestion for 35% squid. meal in local fish meal-based diet, both over time The responses of U to time for lobsters fed 25% ( PV0.05) and between protein levels ( PV0.05). chiton was described by the quadratic regression Feeding 25% protein diets increased the O/N ratio, U=0.741T2+3.656T+11.015, R2=0.71 while for reaching values for protein and lipid oxidation, lobsters fed 35% chiton by the quadratic regression whereas feeding 35% protein diets the O/N ratio U=0.225T2+1.169T+6.027, R2=0.80. These equa- indicated pure protein oxidation for metabolic energy tions predicted that maximum ammonia excretions (Fig. 1C and D). were 1.5 and 1.6 h after ingestion of 25% and 35% Fig. 4 shows the activity of hepatopancreas general chiton, respectively. proteases. No relationship was observed between the The effects of time on U in lobsters fed 25% fish enzymatic activity and protein level in diet. Inclusion meal was described by the quadratic equation U= of squid meal in the local fish meal diet led to an 0.001T2+1.932T+15.429, R2=0.85 whereas for lob- increase of protease activity in hepatopancreas in sters fed 35% fish meal the quadratic equation comparison to those obtained for lobsters fed 35% fish predicted was U=0.147T2+0.657T+6.969, R2=0.60. meal, 25% and 35% clam, and 35% chiton. The highest ammonia excretion was predicted to be 1.2 h after ingestion of 35% fish meal. The variations of O/N atomic ratio over time are 4. Discussion presented in Fig. 3. The dotted horizontal line in Fig. 1 (A, B, C, and D) represents the theoretical division Proteins are needed for growth and maintenance between protein and protein–lipid catabolism accord- and are the most expensive component in a diet. Thus, ing to Mayzaud and Conover (1988). The O/N ratio the selection of nutritional valuable protein sources is did not change ( PN0.05) after the ingestion of clam a key step in feed development for the on-growing of and chiton diets (Table 2). The O/N ratio resulting spiny lobsters. This study has provided information on the metabolic use of different protein sources in formulated diets for juvenile P. argus. Herein, it is 7 demonstrated that supplementation of a local fish meal diet with high-quality fish meal or squid meal 6 increases the efficiency in the use of protein by 5 P. argus juveniles. a a 4 Oxygen consumption and ammonia excretion are ab ab 3 the most commonly used measures for assessing energy metabolism in aquatic animals. There are 2 b b many factors that have been shown to affect the rate 1 b b of oxygen consumption in spiny lobsters such as body 0 weight, dissolved oxygen level, salinity, temperature, -1 activity, handling, diurnal rhythm, and feeding -2 (Buesa, 1979; Crear and Forteath, 2000a; Dı´az- (Digestive proteases activity/mg prot) Clam Squid Chiton Fish Chiton Iglesias et al., 2004). On the other hand, total Diets ammonia excretion of spiny lobsters is also influenced 25% 35% Fresh by temperature, body weight, emersion, daily rhythm, and feeding (Crear and Forteath, 2002a,b). The Fig.4. Digestive proteases activity (meanFstandard error) in hepatopa´ncreas of juvenile P. argus fed experimental diets. One- oxygen consumption rates obtained in this study are way ANOVA: df=7, F=5.04, PV0.05. Different letters indicate similar to those previously observed in our laboratory statistical differences according to SNK test ( PV0.05). (Dı´az-Iglesias et al., 2002; Perera et al., 2003a,b), and E. Perera et al. / Aquaculture 244 (2005) 251–262 259 to those reported by other authors (Buesa, 1979) for energy (Mayzaud and Conover, 1988). The catabo- this specie. Oxygen consumption rates reported here lism of protein produces O/N ratios of 3–16, protein are also similar to those reported for other spiny and some lipid of 17–50, equivalent amounts of lobsters (Crear and Forteath, 2000). protein and lipid of 50–60 and higher values In general, feeding caused the oxygen consumption correspond to the oxidation of lipid and carbohydrates to increase reaching maximum levels 2.3–4.3 h after (Mayzaud and Conover, 1988). Rosas et al., (1995) ingestion depending on diet. The observed pattern in demonstrated that the omnivorous–herbivorous spe- oxygen consumption corresponds to the apparent heat cies like setiferus use protein and lipid as increment (Beamish and Trippel, 1990), a widely energy sources in contrast to omnivorous–carnivorous documented (Rosas et al., 1996; Dı´az-Iglesias et al., species like duorarum which use protein 2002; Perera et al., 2003a,b) but little studied preferentially. The O/N atomic ratio for P. argus after phenomenon (Beamish and Trippel, 1990). In addi- feeding on gastropods, pelecypoda, and crustaceans tion, ammonia excretion showed a strong non- showed that protein and lipid are used as sources of significant trend to increase after feeding reaching energy (Dı´az-Iglesias et al., 2002), while feeding on maximum levels 1.2–3.4 h after ingestion depending chiton and sea urchins lead to the oxidation of pure on diet. The information provided by this study is protein for metabolic energy (Perera et al., 2003a,b). important to manage lobster in captivity. Under The O/N variations observed herein demonstrated intensive culture conditions stocking levels will be that lobsters are able to use different metabolic governed by the system capacity to provide oxygen substrates according to the type of diet they are fed. for metabolic activity and to remove metabolic In this work, the O/N ratio showed that lobsters fed wastes. Implications of oxygen consumption and clam and chiton diets used protein for oxidation at the ammonia excretion levels of spiny lobsters maintained two dietary protein levels tested. Where local fish in recirculated closed systems are discussed in meal was supplemented with high-quality fish meal or Thomas et al., (2000) and Crear and Forteath, squid meal, the anabolic use of protein increased 2 h (2002a,b), respectively. after ingestion, at which time lipid began to have an The nutritive value of a dietary protein is governed important contribution to the energetic metabolism for by the extent to which its content of amino acids lobsters fed the 25% protein diets. The higher protein reflect the needs of the animal in question. When level (35%) lead to an increase in protein metabolism, dietary protein is in excess of that needed for growth which is in agreement with the protein requirement and maintenance, or has an embalance of amino acids, reported for late juvenile P. argus of 25% dietary they cannot be used for anabolic processes (Mayzaud protein by Fraga et al. (2002) and Perera (2000), and Conover, 1988). Instead, they are used as an despite the two forementioned studies used diets energy source, with the concomitant production of entirely based on local fish meal. Even for small ammonia. Double post-prandial peaks in ammonia juveniles J. edwardsii, the protein requirement has excretion have been well documented in lobsters. been reported to be below 35% (Ward et al., 2003), Crear and Forteath, (2002a,b) suggested that the first although for small juvenile the and larger peak represents the metabolically produced requirement has been reported to be 53% (Smith et al., ammonia and the second represents ammonia from 2002). A protein requirement below 35% for late feces and urine. This first bmetabolic peakQ was 7–8 h juvenile P. argus could explain the low O/N ratios after feeding in edwardsii and observed in lobsters fed the 35% protein diets, even (Crear and Forteath, 2002a,b). For P. argus, the first when ingesting high-quality fish meal and squid diets. peak has been observed within the first 5 h after A previous experiment showed no differences in protein ingestion (Perera et al., 2003a,b). That is why food consumption among the experimental diets (data the first 5 h after feeding was used to calculate the O/ not shown). Because of the difficulty in measuring N ratio as a indicator of metabolic use of dietary feed consumption accurately in the respirometric protein in this study. chambers, lobsters were fed at a rate below satiation The O/N atomic ratio is a useful tool to assess the (see Materials and methods) to minimize differences nutrients used by animals as a source of metabolic in food consumption (Smith et al., 2003). Therefore, 260 E. Perera et al. / Aquaculture 244 (2005) 251–262 food consumption was assumed to be similar among available. This will require improved techniques to treatments. Besides, diets were immediately ingested collect pueruli. A recent study has begun the task of by lobsters avoiding the leaching of nutritional evaluating and improving collection methods in Cuba components that occurs within the first 30 min of (Cruz et al., unpublished). However, feeding contin- exposure of diets to seawater (Tolomei et al., 2003). ues being the main restrictive factor in the develop- The achieved results are, then, supposed to be due to ment of lobster aquaculture worldwide. changes in the proportional composition of the protein As an opportunistic predator, P. argus can use a in the diet that resulted from the supplementation with wide range of different quality proteins in its diet. the different protein sources tested. Nevertheless, our study demonstrates that the supple- Protein quality can affect chymotrypsin like activity mentation of a local fish meal diet with high-quality in Marsupenaeus japonicus juveniles (Van Worm- fish meal or squid meal increases the nutritional value houdt et al., 1986), Litopenaeus vannamei larvae (Le of the diet for P. argus juveniles, and that adding Moulllac et al., 1994), and L. setiferus postlarvae (Brito squid to the diet enhances digestive proteases activ- et al., 2000). Here, supplementation of local fish meal ities. In the future, growout trials are needed to fully with squid meal produced an increase in total proteases demonstrate the growth-enhancing effect of these activity in hepatopanceas of juvenile P. argus. protein sources in formulated diets for juvenile P. Akiyama et al., (1992) stated that the digestibility argus and to decide whether the growth rate increase of squid meal is similar to that for fish meal, so the is sufficient to warrant using these protein sources. increase in proteolytic activity does not seem to be a compensatory response to low protein digestibility of squid meal. Another hypothesis is the presence of Acknowledgements secretory factors, probably of peptide nature, in squid meal (Co´rdova-Murrueta and Garcı´a-Carren˜o, 2002). The authors express their gratitude to the captain Some experimental results are supporting this hypoth- and crew of the research vessel bFelipe PoeyQ for esis. Gastrin/CCK like peptides have been reported in supporting during lobster collection campaigns and to penaeid by Van Wormhoudt et al., (1989). Gaspar Gonza´lez for assistance with statistical model- More recently, Sedlmeier and Sedlmeier, (1999) ing. Special thanks to Charles Derby for English demonstrated the secretory effect of some gastro- correction as well as to the anonymous referee for intestinal hormones from vertebrates (CCK, gastrin, many helpful comments on the manuscript. Additional bombesin, secretin, and peptide P) on thanks to the editor Robert Wilson for the help and hepatopancreas, showing the presence of receptors unfailing guidance during the preparation of this paper. and suggesting that some analogue hormone could be present in crustaceans. Proteins ingested in diet are hydrolyzed in the References digestive tract into their constitutive amino acids by proteases; thus, higher proteases activity should allow Akiyama, D.M., Dominy, W.G., Lawrence, A.L., 1992. Penaeid a higher bio-availability of amino acids. Successful shrimp nutrition. In: Fast, A.W., Lesters, L.J. (Eds.), Marine results have been obtained in culturing fish larvae by Shrimp Culture: Principles and Practices. Elsevier, Amsterdam, increasing enzymatic activities using different meth- pp. 535–568. ods (Kolkovski, 2001). The results for crustaceans, Anson, M.L., 1938. The estimation of pepsin, trypsin, papain and cathepsin with hemoglobin. J. Gen. Physiol. 22, 79–85. however, are inconclusive. Lee et al., (1984) and Chen Beamish, F., Trippel, E., 1990. Heat increment: a static or dynamic and Lin, (1990) have pointed out that shrimp with dimension in bioenergetics models? Trans. Am. Fish. Soc. 119, higher enzymatic activities do not necessarily have 649–661. faster growth rates. Brito, R., Chimal, M.E., Gaxiola, G., Rosas, C., 2000. Growth, Prospects for profitably farming P. argus are more metabolic rate and digestive enzyme activity in the white shrimp early postlarvae fed different diets. J. Exp. likely than for temperate spiny lobster species (Jeffs Mar. Biol. Ecol. 255, 21–36. and David, 2002). To develop the aquaculture of P. Buesa, R., 1979. Oxygen consumption of two tropical spiny lobster argus, an increased number of wild seed must be Panulirus argus (Latreille, 1804) and E. Perera et al. / Aquaculture 244 (2005) 251–262 261

(Latreille, 1804) (, Palinuridae). Crustaceana 36 (1), quality in Penaeus vannamei larvae (Crustacea: Decapoda). 99–107. Aquat. Living Resour. 7, 203–210. Chen, H.Y., Lin, H.F., 1990. The effects of exogenous enzymes on Le Moulllac, G., Klein, B., Sellos, D., Van Wormhoudt, A., 1996. the growth of early postlarval . In: Hiran, R., Adaptation of trypsin, chemotrypsin and alfa-amilase to casein Hanyu, I. (Eds.), Proc. Second Asian Fish. Forum, Tokyo, level and protein source in Penaeus vannamei (Crustacea: Japan, 17–22 April, 1989, pp. 349–352. Decapoda). J. Exp. Mar. Biol. Ecol. 208, 107–125. Co´rdova-Murrueta, J., Garcı´a-Carren˜o, L., 2002. Nutritive value of Martı´nez-Otero, A., Dı´az-Iglesias, E., 1975. Instalacio´n respiro- squid and hydrolyzed protein supplement in shrimp feed. me´trica para el estudio de la accio´n de diversos agentes Aquaculture 210, 371–384. presentes en el agua de mar. Ciencias, Serie 8, Investig. Mar. Crear, B.J., Forteath, G.N.R., 2002a. The effect of extrinsic and No. 18 (6 pp.). intrinsic factors on oxygen consumption by the southern rock Mayzaud, P., Conover, R., 1988. O:N atomic ratio as a tool to lobster, . J. Exp. Mar. Biol. Ecol. 252, 129–147. describe zooplankton metabolism. Mar. Ecol., Prog. Ser. 45, Crear, B.J., Forteath, G.N.R., 2002b. Feeding has the largest effect 289–302. on ammonia excretion rate of the southern rock lobster, Jasus Perera, E., 2000. Bioenerge´tica de la langosta Panulirus argus edwardsii, and the western rock lobster, Panulirus cygnus. (Latreille, 1804), durante el ayuno y la alimentacio´n. Bachelor Aquac. Eng. 26, 239–250. Thesis. Centro de Investigaciones Marinas. Universidad de la Crear, B.J., Thomas, C., Hart, P., Carter, C., 2000. Growth of Habana. 45 pp. juvenile Southern rock lobster, Jasus edwardsii is influenced by Perera, E., Dı´az-Iglesias, E., Ba´ez-Hidalgo, M., Nodas, F., 2003a. diet and temperature, whilst survival is influenced by diet and Ana´lisis bioenerge´tico de la alimentacio´n natural en juveniles de tank environment. Aquaculture 190, 169–182. la langosta comu´n Panulirus argus (Latreille, 1804): 4- Crear, B., Hart, P., Thomas, C., Barclay, M., 2002. Evaluation of Amphineura. Rev. Investig. Mar. 24 (1), 17–22. commercial shrimp grow-out pellets as diets for juvenile Perera, E., Dı´az-Iglesias, E., Ba´ez-Hidalgo, M., Nodas, F., 2003b. Southern rock lobster, Jasus edwardsii: influence on growth, Ana´lisis bioenerge´tico de la alimentacio´n natural en juveniles de survival, color and biochemical composition. J. Appl. Aquac. 12 la langosta comu´n Panulirus argus (Latreille, 1804): 5.Echi- (3), 43–57. noidea. Rev. Investig. Mar. 24 (1), 23–28. Dı´az-Iglesias, E., Ba´ez-Hidalgo, M., Perera, E., Fraga, I., 2002. Ritar, A., Thomas, C., Beech, A., 2002. Feeding Artemia and Respuesta metabo´lica de la alimentacio´n natural y artificial en shellfish to phyllosoma larvae of southern rock lobster (Jasus juveniles de la langosta espinosa Panulirus argus (Latreille, edwardsii). Aquaculture 212, 179–190. 1804). Hidrobiolo´gica 12 (2), 101–112. Rosas, C., Sanchez, A., Dı´az-Iglesias, E., Soto, L., Gaxiola, G., Dı´az-Iglesias, E., Dı´az-Herrera, F., Re-Araujo, D., Ba´ez-Hidalgo, Brito, R., Ba´ez-Hidalgo, M., Pedroza, R., 1995. Oxygen M., Lo´pez-Zenteno, M., Valdez-Sa´nchez, G., Lo´pez-Murillo, A., consumption and ammonia excretion of Penaeus setiferus, P. 2004. Temperature preference and circadian oxygen consump- schmitti, P. duorarum and P. notialis fed purified test diets: tion of the red spiny lobster. Panulirus interruptus (Randall, effect of protein level on substrate metabolism. Aquat. Living 1842). Cienc. Mar. 30 (1B), 169–178. Resour. 8, 161–169. Fraga, I., Galindo, J., De-Arazoza, M., Blanco, W., Dı´az-Iglesias, Rosas, C., Sanchez, A., Dı´az-Iglesias, E., Soto, L., Gaxiola, G., E., Ba´ez-Hidalgo, M., Perera, E., 2002. Efecto de diferentes Brito, R., 1996. Effect of dietary protein level on apparent heat relaciones proteı´na/energı´a en el crecimiento de juveniles de increment and post-prandial nitrogen excretion of Penaeus langosta espinosa (Panulirus argus). CIVA 2002, pp. 74–83. setiferus, P. schmitti, P. duorarum, and P. notialis postlarvae. (http://www.civa2002.org). J. World Aquac. Soc. 27 (1), 92–102. Jeff, A.J., Hooker, S., 1999. Economics of the aquaculture of a Ruohonen, K., Kettonen, J., 2004. Effective experimental designs spiny lobster, Jasus edwardsii, in New Zealand. The Annual for optimizing fish feeds. Aquac. Nutr. 10 (3), 145–151. International Conference and Exposition of the Word Aqua- Sedlmeier, R., Sedlmeier, D., 1999. Release of digestive enzymes culture Society Sydney, 256 April–2 May. 362 pp. from the crustacean hepatopanceas: effect of vertebrate gastro- Jeffs, A., David, M., 2002. An assessment of the aquaculture intestinal hormones. Comp. Biochem. Physiol. 123B, 187–192. potential of the Caribbean Spiny Lobster, Panulirus argus. Gulf. Smith, D., Williams, K., Irvin, S., Barclay, M., Tabrett, S., 2002. Caribbean Fisheries Inst, pp. 1–25. Development of a pelleted feed for juvenile tropical spiny Kittaka, J., 1997. Culture of larval spiny lobster: a review of work lobster (Panulirus ornatus): response to dietary protein and done in northern Japan. Mar. Freshw. Res. 48, 923–930. lipid. Aquac. Nutr. 9, 231–237. Kolkovski, S., 2001. Digestive enzymes in fish larvae and Smith, G., Ritar, A., Dunstan, G., 2003. An activity test to evaluate juveniles—implications and applications to formulated diets. larval competency in spiny lobsters (Jasus edwardsii) from wild Aquaculture 200, 181–201. and captive broodstock held under different environmental Lee, P.C., Smith, L.L., Lawrence, A.L., 1984. Digestive proteases of conditions. Aquaculture 218, 293–307. Penaeus vannamei Boone: relationship between enzyme activ- Thomas, C., Crear, B., Hart, P., 2000. The effect of temperature on ity, size and diet. Aquaculture 42, 225–239. survival, growth, feeding and metabolic activity of the southern Le Moulllac, G., Van Wormhoudt, A., AQUACOP, 1994. Adapta- rock lobster (Jasus edwardsii). Aquaculture 185, 73–84. tion of digestive enzymes to dietary protein, carbohydrates and Thomas, C., Carter, C., Crear, B., 2003. Feed availability and its fiber levels and the influence of protein and carbohydrates relationship to survival, growth, dominance and agonistic 262 E. Perera et al. / Aquaculture 244 (2005) 251–262

behaviour of the Southern rock lobster (Jasus edwardsii)in Van Wormhoudt, A., Favrel, P., Guillaume, J., 1989. Gastrin/ captivity. Aquaculture 215, 45–65. Cholecystokinin like post prandial variation: quantitative and Tolomei, A., Crear, B., Johnston, D., 2003. Diet immersion time: qualitative changes in the haemolimph of penaeids (Crustacea: effects on growth, survival and feeding behavior of juvenile decapada). J. Comp. Physiol., B 159, 269–273. southern rock lobster Jasus edwardsii. Aquaculture 219, Ward, L., Carter, C., Crear, B., Smith, D., 2003. Optimal dietary 303–316. protein level for juveniles southern rock lobster Jasus edward- Van Wormhoudt, A., Cruz, E., Guillaume, J., Favrel, P., 1986. sii, at two lipid levels. Aquaculture 217, 483–500. Action de linhibiteur trypsique de soja sur la croissance et Winberg, G., 1956. Rate of metabolism and food requirements of fishes. lactivite´ des enzymes digestives chez Penaeus japonicus Russian (Eng. Transl.: Fish Res. Bd. Can. Trans. Ser., No.194). (Crustacea : Decapoda): role eventuel des hormones gastro- Zar, J.H., 1984. Biostatistical Analysis. Second ed. Prentice-Hall, intestinales. Oce´anis 12, 305–319. Englewood, Cliffs, NJ.