A Holistic View of Dietary Carbohydrate Utilization in Lobster: Digestion, Postprandial Nutrient Flux, and Metabolism

A Holistic View of Dietary Carbohydrate Utilization in Lobster: Digestion, Postprandial Nutrient Flux, and Metabolism Leandro Rodrı ´guez-Viera 1,6 * . , Erick Perera 2. , Antonio Casuso 1 , Rolando Perdomo-Morales 3 , Odilia Gutierrez 4 , Idania Scull 4 , Olimpia Carrillo 5 , Juan A. Martos-Sitcha 2 , Tsai Garcı ´a-Galano 1 , Juan Miguel Mancera 6 1Center for Marine Research, University of Havana, Havana, Cuba, 2Instituto de Ciencias Marinas de Andalucı ´a, ICMAN-CSIC, Puerto Real, Cadiz, Spain, 3Biochemistry Department, Center for Pharmaceuticals Research and Development, Havana, Cuba, 4Institute for Animal Science, Mayabeque, Cuba, 5Faculty of Biology, University of Havana, Havana, Cuba, 6Department of Biology, Faculty of Marine and Environmental Sciences, Campus de Excelencia Internacional del Mar (CEI-MAR), University of Cadiz, Puerto Real, Cadiz, Spain Abstract Crustaceans exhibit a remarkable variation in their feeding habits and food type, but most knowledge on carbohydrate digestion and utilization in this group has come from research on few species. The aim of this study was to make an integrative analysis of dietary carbohydrate utilization in the spiny lobster Panulirus argus. We used complementary methodologies such as different assessments of digestibility, activity measurements of digestive and metabolic enzymes, and post-feeding flux of nutrients and metabolites. Several carbohydrates were well digested by the lobster, but maize starch was less digestible than all other starches studied, and its inclusion in diet affected protein digestibility. Most intense hydrolysis of carbohydrates in the gastric chamber of lobster occurred between 2–6 h after ingestion and afterwards free glucose increased in hemolymph. The inclusion of wheat in diet produced a slow clearance of glucose from the gastric fluid and a gradual increase in hemolymph glucose. More intense hydrolysis of protein in the gastric chamber occurred 6–12 h after ingestion and then amino acids tended to increase in hemolymph. Triglyceride concentration in hemolymph rose earlier in wheat-fed lobsters than in lobsters fed other carbohydrates, but it decreased the most 24 h later. Analyses of metabolite levels and activities of different metabolic enzymes revealed that intermolt lobsters had a low capacity to store and use glycogen, although it was slightly higher in wheat-fed lobsters. Lobsters fed maize and rice diets increased amino acid catabolism, while wheat-fed lobsters exhibited higher utilization of fatty acids. Multivariate analysis confirmed that the type of carbohydrate ingested had a profound effect on overall metabolism. Although we found no evidence of a proteinsparing effect of dietary carbohydrate, differences in the kinetics of their digestion and absorption impacted lobster metabolism determining the fate of other nutrients. Citation: Rodrı ´guez-Viera L, Perera E, Casuso A, Perdomo-Morales R, Gutierrez O, et al. (2014) A Holistic View of Dietary Carbohydrate Utilization in Lobster: Digestion, Postprandial Nutrient Flux, and Metabolism. PLoS ONE 9(9): e108875. doi:10.1371/journal.pone.0108875 Editor: Erik V. Thuesen, The Evergreen State College, United States of America Received April 29, 2014; Accepted August 26, 2014; Published September 30, 2014 Copyright: ß2014 Rodrı ´guez-Viera et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper. Funding: This work was partially supported by International Foundation for Science (http://www.ifs.se/) through project no. A/4306-2 granted to EP, and Escuela Internacional de Doctorado en Agroalimentacio ´n (http://www.ceia3.es/eida3) fellowship granted to LRV. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * Email: [email protected] .These authors contributed equally to this work. Introduction Decapod crustaceans live in virtually all marine and freshwater habitats on Earth (also a few species are largely terrestrial), and exhibit a remarkable variation in their feeding behaviors, from filter feeding, scavenging, grazing to hunting, and in the composition of their diet. However, most knowledge about crustacean digestion has come from studies on a few economically relevant decapods due to the importance of optimized formulated feeds for aquaculture success [1–3]. Vegetal-derived flours and starches (native or processed) have been historically used in aquafeeds due to their reasonable price and good results for most species. Starches are, in general, well digested by decapod crustaceans [1], including spiny lobsters [2–6]. However, starch digestibility in shrimp varies from 60% to 96% [1] and in spiny lobsters from 59% (maize) to 91% (wheat) [3]. Those variations depend not only on several features of the starch itself [7–10], but also on the level of inclusion in diet [11], the throughput rate of the digesta [11], and the activity of digestive carbohydrases [12–22]. While digestibility values are indicative of overall digestion, the time-course emergence of digestion products may vary among crustacean taxa, and within the same species for different carbohydrates. High carbohydrate hydrolysis and glucose absorption are widely recognized in crustaceans [1,23–26], but differences occur among different crustacean taxa concerning the location of carbohydrate digestion. The highest amylase activity in PLOS ONE | www.plosone.org 1 September 2014 | Volume 9 | Issue 9 | e108875 shrimp is found within the digestive gland [17], while amylase activity is largely found in the gastric juice of spiny lobsters (P. argus [21], J. edwardsii [12]), and other lobster species such as Homarus gammarus [18]. The effects of the time-course appearance of carbohydrate digestion products on the energy metabolism remain unknown. The utilization of glycogen and free glucose has been studied in crustaceans under different stressful conditions [6,26–29], but the role of carbohydrates during regular feeding is less understood. This issue is particularly interesting in spiny lobsters given the occurrence of high and prolonged hyperglycemia after a meal, which in some species may last 30 h [2]. In shrimp, carbohydrates can spare to a certain extent the dietary proteins in spite of prolonged hyperglycemia [30] by changing the metabolic substrate from protein to a mixture of protein, lipids and carbohydrates [23]. However, there is no direct evidence of a significant use of carbohydrates for energy in spiny lobsters. Spiny lobsters have a marked protein-based energy metabolism [31,32], and have little glycogen stored in their digestive gland [5,6,33]. A slight protein-sparing effect could be demonstrated in P. argus only for lipids under certain dietary conditions [31,32]. This information for lobster, however, was gathered from analysis of the oxygen consumption: ammonia excretion atomic ratio (O:N), without direct evidence at the level of intermediary metabolism. Several studies have evaluated different metabolic, mostly glycolytic, enzymes in economically relevant crustaceans such as penaeid shrimp in relation to carbohydrate level in diet [24,34,35], starvation [29,36], molt cycle [37], and osmotic stress [24,35], but there are no previous reports on the activities of these enzymes in spiny lobster in relation to feeding. In addition, there is poor information on the post-prandial flux of nutrients (i.e., amino acids, lipids) other than glucose [2,4] after feeding in spiny lobsters and other crustaceans, which may also affect the use of dietary carbohydrate. Even for penaeid shrimps, the nutritionally most studied crustaceans during the past four decades, to our knowledge, there is only one report on the postprandial changes of amino acids in hemolymph [38]. Although several studies are available on spiny lobster nutrition [4,5,39–47], the role of dietary carbohydrate on energy metabolism and their interactions with other nutrients remains largely unknown in comparison to more thoroughly studied crustaceans such as penaeid shrimp. The aim of this study was to evaluate the effects of the extent and the time-course of carbohydrate digestion on the metabolism of the spiny lobster P. argus. We used complementary methodologies such as i) in vitro and in vivo assessments of digestibility, ii) measurements of amylase activity in the digestive tract, iii) postfeeding flux of nutrients and metabolites, and iv) determination of activity of key metabolic enzymes, in a step by step approximation to dietary carbohydrate utilization in lobster. While, as widely accepted, the use of carbohydrates by crustaceans depends on their overall digestibility, we showed that differences in carbohydrate digestion kinetics have a profound impact on lobster metabolism and on the utilization of other nutrients. Results presented may assist in selecting physiologically appropriated carbohydrate sources for P. argus, and constitute baseline data for further optimization of formulated feeds in this species. Materials and Methods Preparation of digestive gland extracts Spiny lobsters were collected in the Gulf of Batabano´, Cuba, by SCUBA diving. The collection area was: 21u39.04319N– 83u09.84369W; 21u41.00159N–83u0924639W; 21u40.10169N– 83u11.02979W, and it was performed under permission of the Fisheries Regulator Department from the Ministry of the Fishing Industry of Cuba. This study did not involve endangered or protected species. Lobsters were anesthetized by immersing them into ice-cold water before digestive gland extraction. Samples were immediately frozen in liquid nitrogen and stored at 280uC. Tissue was homogenized with chilled Milli-Q water (90 mg/500 mL) using a glass piston homogenizer, and the homogenate centrifuged at 10 0006g for 30 min at 4uC. The resultant upper lipid layers were discarded and the remaining supernatants were stored at 2 80uC until used for in vitro digestion assays, after determination of its amylase activity. Amylase activity Amylase activities were determined using an HELFA Amilase Assay Kit (Quimefa Biologic Products Inc. Havana, Cuba) with CNPG 3 (2-Chloro-4-nitrophenyl-a-D-maltotrioside) as the substrate, following the manufacturer’s instructions. One unit of amylase activity was defined as the amount of enzyme that produces the release of 1 mmol nitrophenol per minute. Units of amylase activity were expressed per volume, weight of tissue, or soluble protein as needed. Protein concentration Soluble protein concentrations were quantified according to [48] using bovine serum albumin as the standard. In vitro digestion in Eppendorf tubes Carbohydrate digestibility was assessed in vitro in term of glucose released after incubation with digestive gland extracts. Thirteen different carbohydrates sources (300 mg) (Table 1) were dissolved each in 5 mL of Milli-Q water to achieve a concentration of 6% (w/v) and used as substrate solutions as described before [2]. The amount of digestive gland extracts added for each digestion assay (N = 30, per carbohydrate source) were adjusted in term of amylase activity in order to assure similar extract composition against the substrates. In vitro assays were performed using a method previously described [1] and modified [2] as follows: 250 mL of solutions or suspensions of the carbohydrate substrates (Table 1) and digestive gland extracts (0.2 U of final amylase activity in the mixture) were mixed and diluted up to 1 mL with 100 mM citrate-phosphate buffer (pH 5.0) in 2 mL Eppendorf tubes. Tubes were shaken for 60 min at room temperature (26uC) [the rate of carbohydrate hydrolysis was linear up to two hours of incubation under this assay conditions, data not shown]. Then, 20 mL of samples were taken and stored at 220uC for glucose determination. Glucose released was determined using a HELFA RapiGlucoTest glucose oxidase Kit (Quimefa Biological Products Inc., Havana, Cuba) following the manufacturer’s instructions. The assays were performed in duplicate. Blank assays without addition of enzyme extracts or the substrates were carried out for each carbohydrates source to estimate the amount of free glucose present in the extracts and the substrates. The hydrolysis rate (HR) [2] of each substrate was calculated as: HR (nmol glucose min 21 ) = ([glucose] F 2[glucose] 0 )/t, where [glucose] F is the final glucose concentration after incubation with enzyme extracts, [glucose] 0 is the glucose concentration present collectively in the substrate and enzyme extracts, and t is the incubation time in minutes. As glucose is not the only product of carbohydrate hydrolysis, reducing sugars were also determined by the Somogy-Nelson method [49] using maltose as the standard, to express HRs also as mg maltose equivalent released per minute. Carbohydrate Utilization in Lobster PLOS ONE | www.plosone.org 2 September 2014 | Volume 9 | Issue 9 | e108875 In vivo digestibility by the inert marker chromic oxide Apparent in vivo digestibility was assessed using 1% of the inert marker chromic oxide in formulated diets (Table 2). Three experimental isoenergetic diets were formulated to have 45% protein, 10% lipids, and 35% of three different carbohydrates (rice starch, wheat flour, maize starch) of suspected different digestibilities according to previous in vitro results. All feedstuffs were obtained from commercial suppliers, except fish and squid meals that were made at the laboratory as described before [50]. Pellets were made as described previously [51], but extruded twice for a better homogenization of the marker. Spiny lobsters (90–150 g) were collected as described above and transported alive to the Center for Marine Research of the University of Havana, Cuba. Only apparently healthy lateintermolt (C4) specimens (determined according to Lyle and MacDonald [52]) were used. The feeding trial was conducted for 30 days in an indoor facility equipped with recirculated sea water, sand and biological filtration, constant aeration, and photoperiod cycle of 12 h light: 12 h dark. Water quality was monitored in the morning twice a week for temperature (,26uC), pH (,8.0), salinity (36%), dissolved oxygen (,6.0 mg/L), and ammonia-N (,0.07 mg/L). Each of the three experimental diets was sorted at random to six lobsters (N = 6, per diet), housed individually in 60L tanks. Lobsters were acclimated for one week to the experimental diets by gradually reducing fish flesh until they consumed exclusively the pellets. The amount of diet given was progressively adjusted to 2% of body weight per day (BW day 21 ) according to the appetite of the lobsters by checking the bottom of the tanks for excess feed remaining one hour after feeding. This feeding time is enough for lobsters fed close to satiation [53]. Following one week of acclimation to the experimental diets, each lobster was fitted with a fecal collection devise [54] modified as in [3]. Lobsters were fed one ration a day in the morning. Feces were collected once daily, one hour after feeding to reduce the impact of handle stress on feed intake [54]. Feces was carefully removed from the tubes and stored at 220uC. Successive fecal samples collected from the same lobster were pooled together until the total collection of around 100 g of wet feces per lobster [equivalent to about 2 g of dry matter (DM)]. Diets were rotated each week trough the three group of lobsters to avoid adaptation, with two days of fasting inbetween. Any collection that was contaminated with sea water was discarded. Composition analyses of diets and feces were performed according to [55]. Briefly, dry matter was analyzed by weight change following drying at 105uC to a constant weight, ash was determined by weight change following furnace incineration at 550uC for 5 h, total protein was determined by the Kjeldahl method, total carbohydrates were measured by the acid hydrolysis method, and chromic oxide was determined by the perchloric acid digestion method [55]. Apparent digestibility (AD) was calculated as in [1]: AD (%) = 1006[1– (c i /c f )6(n f /n i )], where c i and c f are the concentrations (dry matter basis) of chromic oxide in the ingested diet and feces respectively, and n i and n f are the concentrations (dry matter basis) of the nutrient in the ingested diet and feces, respectively. Serial collection of gastric juice and hemolymph Another three groups of five lobsters were adapted to experimental diets as above and fed for one week. Later, they were fasted for 2 days and were provided with a ration of the experimental diets for gastric juice and hemolymph collection. Gastric fluid samples were obtained through the oral cavity using disposable insulin syringes with a plastic cannula over the sharp end of the needle as described before [51]. Gastric juice was not sampled prior to feeding as this was presumed to affect feed intake due to stress. Samples (,100 mL) of gastric juice were taken from the same lobster at 2, 6, 12, 24 and 30 h after ingestion, centrifuged at 10,0006g for 10 min, frozen in liquid nitrogen and then stored at 280uC. Lobsters were handled with care and samples were rapidly taken (less than 1 min) to avoid excessive stress. We previously demonstrated that this serial sampling of gastric juice produces no variation in hemolymph glucose concentration as a result of manipulation [51]. Hemolymph was neither sampled prior to feeding as this is known to affect feed intake in other spiny lobster species [2]. Hemolymph sampling began 2 h after feeding, with additional samples at 6, 12, 24, and 30 h from each lobster. Previous studies in spiny lobsters revealed that the effect of serial sampling of hemolymph on hemolymph glucose concentration is negligible [56]. Hemolymph samples (500 mL) were taken from the sinus of the 4 th walking legs [57] in 1 mL pyrogen free disposable syringes containing 500 mLof precooled anticoagulant solution (400 mM NaCl, 10 mM KCl, 10 mM Hepes, 20 mM EDTA, pH 7.3) [58]. An additional group of five lobsters were left unfed and sampled as above, corroborating that no variation in the variables studied occurs due to manipulation or daily rhythm. Table 1. Carbohydrate sources used for in vitro digestion. Carbohydrates Source Agar Sigma-Aldrich (Fluka), cat. No. 05039 Agarose SERVA Electrophoresis GmbH, cat. No. 11403 Alginate Sigma-Aldrich, cat. No. W201502 Carboxymethyl cellulose Sigma-Aldrich cat. No. C5678 Glycogen Sigma-Aldrich, cat. No. G8751 Maize starch * Indias, Baldinelli G.R., Argentina Potato starch * AppliChem-Panreac, cat No. A2223 Rice starch BDH, Merck Chemicals Ltd., cat. No. 30263 Maize flour Yellow, Fine, Iberia Foods Corp., Brooklyn, NY Rice flour Made at the laboratory by finely milling white rice Wheat flour Commercially available regular foodstuff * Marked carbohydrates were also assayed after gelatinisation at 80uC for 20 min (100 g L 21 deionizer water) and dried at 50uC for 48 h [2]. doi:10.1371/journal.pone.0108875.t001 Carbohydrate Utilization in Lobster PLOS ONE | www.plosone.org 3 September 2014 | Volume 9 | Issue 9 | e108875 Time-course of nutrients and metabolites in gastric juice and hemolymph Increasing soluble protein in gastric juice was measured as indicator of solubilization of dietary protein plus enzyme secretion into the foregut. Gastric juice glucose and free amino acids were measured as indicators of rates of carbohydrate and protein hydrolysis in the foregut. Gastric juice triglycerides were measured as indicator of lipid solubilization/emulsification from feeds. The glycemic prandial response was analyzed as indicator of digestibility and assimilation of dietary carbohydrates in lobsters [2,56], and the same rational was applied for free amino acids and triglycerides in hemolymph. Soluble protein and glucose concentrations were quantified as described above. Free amino acid levels were assessed colorimetrically using the nynhidrin method [59,60] using L-alanine as the standard. Triglyceride and lactate concentrations were measured using the commercial kits TAG (Spinreact, Girona, Spain) and Lactate (Spinreact, Girona, Spain), respectively. Metabolites in digestive gland and muscle At the end of the 30 h time-course sampling of gastric juice and hemolymph, the three groups of lobsters were fed during one month with the same experimental diets without disturbance, left unfed for 48 h, and then fed again with the respective diets. They were sacrificed 24 h later in ice-cold water to obtain digestive gland and muscle samples. Five additional lobsters that were fed during one month with fish flesh were also left unfed for 48 h, sampled as above, and referred as fresh fish treatment. Samples were immediately frozen in liquid nitrogen and then lyophilized for metabolite and metabolic enzyme measurements (see below). For metabolites, powder of digestive gland and muscle were weighed and homogenized in water (,20 mg/mL), centrifuged (30 min at 10,0006g, 4uC), and the supernatant was used to assess tissue metabolites. Prior to centrifugation, an aliquot was removed and frozen at 280uC for triglyceride determination. Protein, glucose, free amino acid, triglyceride, and lactate were measured as above. Glycogen concentration was assessed as described before [61]. Table 2. Formulation (%) and proximate composition of the experimental diets. Ingredients Wheat diet Maize diet Rice diet Fish meal a 31 35 35 Squid meal a 13.2 17 17 Gelatin b 555 Wheat flour c 40.9 - - Maize starch d -30Rice starch e --30 Fish oil f 1.9 1.9 1.9 Lecithin g 222 Cholesterol h 111 Vit & Min premix i 111 Chromic oxide j 111 Phosphate/carbonate k 222 Attractants l 111 Talc c - 3.1 3.10 Total 100 100 100 Proximate composition m Crude protein 46.5 46.6 45.9 Crude lipid 10.5 9.7 9.7 Carbohydrate 36.4 35.8 35.7 Ash 7.4 11.3 7.4 Pellets contained 10–12% of water. a Prepared at the laboratory as detailed before [50]. Jack mackerel meal: 79.1% proteins, 16.8% lipids, 5.5% moisture; Squid meal: 76.6% proteins, 10.8% lipids, 8.7% moisture. b Sigma-Aldrich (G2500). c Commercially available regular feedstuff. d Indias, G.R. Baldinelli, Argentina. e BDH (30263), Merck Chemicals Ltd. f Fisheries Research Center Laboratory, Havana, Cuba. g Calbiochem (429415), Merck Chemicals Ltd. h Sigma-Aldrich (C8667). i Premix from DIBAQ-Aquaculture, Segovia, Spain, containing (per kg of feed): vitamin A 15,000 IU, vitamin D3 3000 IU, vitamin E 180 mg, vitamin K 15 mg, vitamin B1 37.5 mg, vitamin B2 37.5 mg, vitamin B6 24.75 mg, vitamin B12 0.045 mg, vitamin H 1.14 mg, D-pantothenic acid 120 mg, nicotinic acid 225 mg, vitamin C 300 mg, folic acid 11.24 mg, Inositol 112.5 mg, zinc 75 mg, selenium 0.3 mg, magnesium 86.25 mg, copper 2.25 mg, manganese 22.5 mg, iodine 7.5 mg, iron 3 mg, cobalt 0.3 mg. J Cr 2 O 3 (Sigma-Aldrich, 393703). k Dicalcium phosphate/Calcium carbonate (1:2), Santa Cruz Fish Feed Factory, Camagu ¨ey, Cuba. l Taurine (Sigma-Aldrich, T0625) 500 mg/Kg diet, Glycine (Sigma-Aldrich, G8898) 500 mg/Kg diet. m Measured as described in Materials and Method section. Crude lipid calculated from proximate composition of ingredients. doi:10.1371/journal.pone.0108875.t002 Carbohydrate Utilization in Lobster PLOS ONE | www.plosone.org 4 September 2014 | Volume 9 | Issue 9 | e108875 Metabolic enzymes in digestive gland and muscle The activities of metabolic enzymes from different pathways were quantified in two key tissues in crustacean metabolism, digestive gland and muscle. The digestive gland of crustaceans is the main site for the synthesis of digestive enzymes, digestion, absorption, metabolism and storage of nutrients, as well as for their mobilization [62], thus being metabolically very active. Gland respiration is known to increase in 56% during the first 6 h after feeding in some crustaceans (e.g. shrimp [63]). Muscle, on the other hand, sustains locomotor activity and growth. Lyophilized samples of digestive gland and muscle were homogenized in 10 volumes of ice-cold buffer containing 50 mM imidazole hydrochloride (pH 7.5), 1 mM 2-mercaptoethanol 50 mM sodium fluoride, 4 mM EDTA, 250 mM sucrose, and 0.5 mM PMSF. Homogenates were centrifuged for 30 min at 10,0006g and supernatants used for assays. Enzymes examined were: hexokinase (HK, EC 2.7.1.11), glycerol-3-phosphate dehydrogenase (G3PDH, EC 1.1.1.8), pyruvate kinase (PK, EC 2.7.1.40), L-lactate dehydrogenase (LDH, EC 1.1.1.27), fructose 1,6-biphosphatase (FBPase, EC 3.1.3.11), glycogen phosphorylase (GPase, EC 2.4.1.1), glucose6-phosphate dehydrogenase (G6PDH, EC 1.1.1.49), aspartate transaminase (AST, EC 2.6.1.1), alanine transaminase (ALT, EC 2.6.1.2), glutamate dehydrogenase (GDH, EC 1.4.1.2), and 3-hydroxyacyl-CoA dehydrogenase (HOAD, EC 1.1.1.35). Amount of sample in each assay was set in preliminary tests to ensure initial velocities for all enzymes studied. Conditions for enzyme assays [e.g. buffer composition, cofactors, additional enzymes for coupled enzyme assays, and electron donor (NADH) or acceptors (NADP, NAD)] were as described elsewhere [65,64]. Substrates were: 5 mM D-glucose for HK, 0.2 mM dihydroxyacetone phosphate for G3PDH, 5 mM Dglucose PK, 6.25 mM lactic acid for LDH, 0.1 mM fructose1,6-bisphosphate for FBPase, 5 mg/mL glycogen for GPase, 1 mM glucose-6-phosphate for G6PDH, 10 mM L-aspartate for AST, 7.5mM L-alanine for ALT, 1.40 mM a-ketoglutarate for GDH, and 0.1 mM acetoacetyl-CoA for HOAD. Reactions without substrates were also performed as controls. Reaction rates of the enzymes HK, LDH, FBPase, GPase, and G6PDH were determined in duplicate by the increase in absorbance at 340 nm and 37uC as a result of the formation of NADPH. Reaction rates of the enzymes G3PDH, PK, AST, ALT, GDH, HOAD were determined in duplicate by the decrease in absorbance at 340 nm and 37uC as a result of the disappearance of NADH. All assays were performed using a Bio-Tek PowerWave 340 Microplate spectrophotometer using KCjunior Data Analysis Software (Bio-Tek Instruments, Winooski, VT, USA). One unit of enzyme activity (U) was defined as the amount of enzyme needed to transform 1 mmol of substrate or produce 1 mmol of product per minute. Enzyme activity was expressed as specific activity as U/mg protein. Statistical analyses Only results from late-intermolt (C4) lobsters were analyzed as molt stage has been found to affect digestive enzyme activities [22] and feeding activity [66] of P. argus. All data were checked for normality and homogeneity of variance using KolmogorovSmirnov and Levene’s tests, respectively, with P#0.05. Logarithmic transformations of data were made when necessary to fulfill the assumptions of ANOVA. In vitro rates of hydrolysis (N = 30 lobsters, per carbohydrate) and in vivo apparent digestibilities (N = 6 lobsters, per dietary treatment) were analyzed by one-way ANOVA (P#0.05), being the carbohydrate substrates and the experimental diets the sources of variation, respectively. Metabolic enzyme activities and metabolites in digestive gland, hemolymph, and muscle 24 h after ingestion were also analyzed by one-way ANOVA (P#0.05). All time-course data were subjected to twoway ANOVA (P#0.05), with diet and time being the two sources of variation. In all cases, the Tukey’s test (P#0.05) was used to determine differences among means. Those variables that significantly varied among diets 24 h after feeding according to univariate analysis detailed above were assumed to be of most predictive value, thus selected for multivariate analysis. We performed a forward stepwise discriminant analysis for two sets of variables, metabolic enzyme activities and metabolites, to understand the combination of variables that can best explain the response of lobsters to diet (carbohydrate) ingested. For the analysis, it was considered F to enter as 0.01, F to remove at 0.0, and minimum tolerance of 0.01. After significant functions development, the relative importance of the original variables in separating the diet ingested by lobster was gauged by standardised values. The software package Statistica 7.0 (StatSoft Inc., Tulsa, OK, USA) was used for all tests performed and figures were generated by GraphPad Prism 5.00 (GraphPad Software, Inc., San Diego, California, US). Results In vitro hydrolysis rate of carbohydrate The liberation of glucose, except for the rice starch, was a suitable measure of carbohydrate digestion as glucose and reducing sugars released, measured by the Somogy-Nelson method, were significantly correlated [mg glucose = .64+9703 (mg maltose equivalents), F = 63.56, P = 0.004, R 2 = 0.95]. The observed values of glucose released from rice starch were 5 times higher than the ones expected from the correlation between glucose and reducing sugars. In vitro digestion of all the carbohydrates sources tested by digestive gland extracts of the spiny lobster P. argus resulted in the liberation of glucose (Fig. 1). The hydrolysis rates (HRs) of the carbohydrate substrates tested were significantly different (one-way ANOVA, F = 68.34, P#0.001) (Fig. 1). Native rice starch displayed the highest HR (nmol glucose min 21 , mean 6SEM) of all the carbohydrate substrates tested (80.6065.46) (Fig. 1). Other carbohydrates were also digested at a high rate such as gelatinised potato starch (70.564.29) and gelatinised maize starch (67.063.94). Intermediate HRs were obtained for rice flour (54.162.98), wheat flour (42.662.29), potato starch (41.664.79), maize flour (32.562.06), glycogen (30.961.87), and maize starch (23.261.40) (Fig. 1). The lowest HRs were found for carboxymethyl cellulose (12.361.61), alginate (3.360.70), agarose (3.260.98), and agar (7.460.83) (Fig. 1). In vivo apparent digestibility Responses in vivo may diverge from those predicted in vitro, thus we assessed in vivo digestibility in this study to ensure that selected carbohydrates (rice starch, wheat flour and maize starch) for studying post-absorptive process in lobster indeed differ in digestibility as predicted in vitro. There were no statistically significant differences in apparent dry matter digestibility among experimental diets (F = 0.61, P.0.05) (Table 3). Conversely, carbohydrate (F = 7.92, P#0.05) and protein (F = 41.99, P# 0.001) in vivo apparent digestibilities were significantly lower for the maize diet than for the other two diets (Table 3). Carbohydrate Utilization in Lobster PLOS ONE | www.plosone.org 5 September 2014 | Volume 9 | Issue 9 | e108875 Soluble proteins and amylase activity in gastric juice Soluble proteins in the gastric juice significantly varied through time (Two-way ANOVA, F = 28.04, P#0.001), and among diets (Two-way ANOVA, F = 4.40, P#0.05), whereas no significant interaction was found between these factors (Twoway ANOVA, F = 2.13, P.0.05). Two peaks of soluble proteins were found in the gastric juice at 6 and 24 h after ingestion all the experimental diets (Fig. 2a). Six hours after ingestion, there were differences among diets (One-way ANOVA, F = 4.75, P# 0.05) in soluble protein concentration of the gastric juice, being significantly higher for wheat than for maize diet (Tuckey’s test, P#0.05) (Fig. 2a). Twenty-four hours after the last meal there were no differences among diets in soluble protein concentration in the gastric juice (One-way ANOVA, F = 2.06, P.0.05), despite lobsters fed with the wheat diet tended to have more soluble proteins in their gastric juice (Fig. 2a). There were no significant differences in amylase activity per volume of gastric juice trough time (Two-way ANOVA, F = 2.49, P.0.05) and among diets (Two-way ANOVA, F = 2.23, P.0.05), neither interaction among these factors (Two-way ANOVA, F = 0.59, P.0.05) (Fig. 2b). Time-course of glucose in gastric juice and hemolymph after feeding Free glucose in the gastric juice varied through time of digestion with maximal values 6 h after ingestion (Two-way ANOVA, F = 10.89, P#0.001), and differed between fed and unfed lobsters (Two-way ANOVA, F = 7.34, P#0.01) (Fig. 3a). No interaction Figure 1. In vitro hydrolysis rates of different carbohydrate substrates by digestive gland extracts of the spiny lobsters Panulirus argus .Each value is the mean 6SEM (N = 30 lobsters per diet) Significant differences are marked by different superscript letters (one-way ANOVA, Tukey test, P#0.05). doi:10.1371/journal.pone.0108875.g001 Table 3. Dry matter, carbohydrate, and crude protein apparent digestibility of formulated diets containing 45% protein, 9% lipids, and 30% of different carbohydrate sources (wheat flour, rice starch, maize starch) fed to Panulirus argus. Diet Dry matter (%) Carbohydrate (%) Crude protein (%) Wheat flour 46.467.23 90.764.06 a 83.064.12 a Rice starch 47.862.78 81.463.96 ab 70.664.48 a Maize starch 38.768.98 60.161.06 b 37.661.68 b Values are means 6SEM (N = 6). Significant differences within the same column are indicated by different letters (one-way ANOVA, Tukey test, P#0.05). doi:10.1371/journal.pone.0108875.t003 Figure 2. Changes in concentrations of soluble protein (a) and a-amylase activity (b) in the gastric juice of Panulirus argus after feeding diets with different carbohydrate sources. Diets were named according to the carbohydrate source they contained (wheat flour, rice starch, maize starch). Each value is the mean 6SEM (N = 5 lobsters per diet). For each dietary treatment, the time at which the first statistically difference was found with respect to 2 h is marked by an asterisk. Differences among diets throughout the 30-h experiments are marked by different superscript letters in the legend (Two-way ANOVA, Tukey test, P#0.05). doi:10.1371/journal.pone.0108875.g002 Carbohydrate Utilization in Lobster PLOS ONE | www.plosone.org 6 September 2014 | Volume 9 | Issue 9 | e108875 between diet and time was found (Two-way ANOVA, F = 1.88, P.0.05). Lobsters that ingested the wheat diet tended to have a higher free glucose concentration in the gastric juice 12 h after meal respect to those fed the other diets suggesting a delay in absorption, although no statistical differences were found (Fig. 3a). On the other hand, there were differences among diets (Two-way ANOVA, F = 5.75, P#0.05) and through time (Two-way ANOVA, F = 28.09, P#0.001) in hemolymph free glucose levels, resulting the interaction between both factors also significant (Two-way ANOVA, F = 3.47, P#0.05) (Fig. 3b). Glucose concentration in hemolymph increased with maximal values attained at 12 h (for rice and maize diets) or 24 h (for wheat diet) after ingestion (Fig 3b). Concentration of glucose in hemolymph 12 h after ingestion was significantly lower (One-way ANOVA, F = 23.91, P#0.05; Tukey’s test, P#0.05) in wheat fed lobsters than in lobsters ingesting the other diets (Fig. 3b). Time-course of free amino acid in gastric juice and hemolymph after feeding Free amino acid concentration in the gastric juice did not varied through time of digestion (Two-way ANOVA, F = 2.50, P.0.05), although a slight non-significant increase was observed from 6 to 12 h after ingestion of wheat and rice diets (Fig. 3c), in correspondence with the decrease found in soluble protein during this period (Fig. 2a). Free amino acid levels differed among experimental diets (Two-way ANOVA, F = 26.10, P#0.001) being smaller in maize fed lobsters respect to the other group (Fig. 3c), while there was no interaction between diet and time (Two-way ANOVA, F = 2.82, P.0.05). On the other hand, free amino acid concentration in hemolymph did not differ significantly through time (Two-way ANOVA, F = 2.33, P.0.05) although this parameter tended to raise slowly 12 h after feeding for all diets (Fig. 3d). Concentration of amino acid in hemolymph differed among diets (Two-way ANOVA, F = 4.43, P#0.05) (Fig. 3d), while there was no interaction between diet and time (Two-way ANOVA, F = 0.77, P.0.05). Time-course of triglyceride in gastric juice and hemolymph after feeding Triglyceride concentration in the gastric juice did not vary significantly through time of digestion (Two-way ANOVA, F = 1.55, P.0.05) but differed among experimental diets (Twoway ANOVA, F = 9.32, P#0.001), and the interaction between both factors was also significant (Two-way ANOVA, F = 2.86, P# 0.05) (Fig. 3e). Concentration of triglyceride in the gastric chamber of lobsters was lowest with the maize diet (Fig. 3e). There were differences among fed and unfed lobsters (Two-way ANOVA, F = 8.71, P#0.01) in the triglyceride levels in hemolymph after feeding, although not among formulated feed treatments (Fig. 3f). Yet, non-significant higher values for triglyceride were observed 12 h after ingestion in the wheat diet (Fig. 3f). For all diets, triglyceride concentration in hemolymph varied significantly through time (Two-way ANOVA, F = 15.68, P# 0.001) (Fig. 3f), while the interaction between diet and time was not significant (Two-way ANOVA, F = 1.75, P.0.05). Triglyceride concentration in the hemolymph rose abruptly 12 h after feeding in riceand maize-fed lobster, while it rose earlier in wheat-fed lobsters (Fig. 3f). Lactate in hemolymph after feeding After ingestion, differences were found in lactate concentration in hemolymph among diets (F = 12.89, P#0.001) and throughout time (F = 32.05, P#0.001) (Fig. 4). The interaction between both factors was also statistically significant (F = 4.42, P#0.001). While in riceand wheat-fed lobsters the concentration of lactate increased in hemolymph 12 h after ingestion, in lobsters that ingested the maize diet it started to increase earlier, attaining 12 h after ingestion the highest values recorded for all diets (Fig. 4). Metabolites and metabolic enzymes in tissues of lobsters after feeding different diets The concentrations of glucose (F = 1.21, p.0.05), glycogen (F = 3.13, p.0.05), lactate (F = 1.52, p.0.05), and amino acid (F = 2.17, p.0.05) in the muscle of lobsters did not vary among dietary treatments 24 h after feeding, and differences were only found in triglyceride contents (F = 7.11, p#0.05), with smaller values for the wheat diet (Tukey’s test, P.0.05) (Table 4). At this time, differences were also found in the concentrations of glucose (F = 15.12, p#0.05), lactate (F = 15.19, p#0.05), amino acid (F = 5.64, p#0.05), and triglycerides (F = 5.18, p#0.05) in the hemolymph (Table 4). However, most differences in hemolymph were observed between fresh fish and formulated feed fed lobsters, without significant differences (Tuckey’s test, P.0.05) among formulated feeds, except for triglycerides (Table 4). In the digestive gland, differences were found among dietary treatments in contents of glucose (F = 17.50, p#0.05), glycogen (F = 8.66, p# 0.05), lactate (F = 30.77, p#0.001), amino acid (F = 6.27, p#0.05), and triglyceride (F = 6.94, p#0.05) (Table 4). However, the only significant differences among groups were a smaller glycogen content in the gland of maize-fed lobsters (Tukey’s test, P.0.05), and a low triglyceride content in wheat fed animals (Tukey’s test, P.0.05) (Table 4). Enzymes that differed in activity among dietary treatments 24 h after feeding were (Table 5): G3PDH (F = 6.367, p#0.05), FBP (F = 3.964, p#0.05), GPasa (F = 3.382, p#0.05), and AST (F = 9.922, p#0.05) in muscle, and PK (F = 5.04, p#0.05), LDH (F = 4.482, p#0.05), FBP (F = 4.635, p#0.05), G6PDH (F = 106.8, p#0.0001), AST (F = 6.355, p#0.05), and HOAD (F = 5.463, p# 0.05) in the digestive gland. No significant differences were found in the activity of the following enzymes: HK (F = 0.271, p.0.05), LDH (F = 0.489, p.0.05), G6PDH (F = 0.583, p.0.05), ALT (F = 0.738, p.0.05), GDH (F = 0.728, p.0.05), HOAD (F = 2.565, p.0.05) in muscle, and G3PDH (F = 0.633, p.0.05), ALT (F = 1.13, p.0.05) and GDH (F = 1.248, p.0.05) in the digestive gland. Discriminant analysis of metabolic enzymes and metabolites Three significant discriminant functions could be developed from the activity of metabolic enzymes included in the analysis. The first two functions collectively accounted for 96% of the total variance (Table 6). Specific activities of G6PDH in the digestive gland, and FBPase and GPase in muscle were the variables with the highest relevance in the first discriminant function (Table 7), in which wheat fed lobsters showed central values as opposed to those of the lobsters ingesting the other formulated diets or fresh fish (Fig. 5a, axis x). On the other hand, activities of PK, FBPase and HOAD in the digestive gland, and G3PDH, FBPase, AST and GPase in muscle had the highest weight in the second discriminant function (Table 7), which discriminated among all dietary treatments but wheat (Fig. 5a, axis y). No function built from metabolic enzyme activities could strongly discriminate between fresh fish or maize fed lobsters (Fig. 5a). Only two statistically significant functions were able to predict the diet ingested by lobsters from metabolite data, but the single first function explained most of the variability (93%) (Table 6). Carbohydrate Utilization in Lobster PLOS ONE | www.plosone.org 7 September 2014 | Volume 9 | Issue 9 | e108875 The variables that best identified the diet ingested by lobsters were glycogen, glucose, lactate, and triglycerides in the digestive gland, and glucose and amino acid in hemolymph (Table 7). Wheatand rice-fed lobsters were poorly discriminated by both functions (Fig. 5b). Discussion Carbohydrate digestibility in the spiny lobster Methods to assess the in vitro digestibility of feedstuffs were developed as alternatives to expensive and time-consuming in vivo methods [67]. They are particularly suited when there is no previous information for a particular species, such as P. argus,as they allowed for the evaluation of relatively high amounts of feedstuffs. Most starches contain 72–82% of amylopectin and 18–33% of amylose, but proportions vary according to the botanical origin [68]. In general, starch susceptibility to hydrolysis depends on the amylose content [9,69]. High amylose content decreases starch digestibility due to a denser packing of the helicoidal structure [70] and the formation of amylose-lipid complex [7–9]. In addition, starch hydrolysis is also affected by the granule size (i.e., more Figure 3. Changes in concentrations of glucose in gastric juice (a) and hemolymph (b), amino acid in gastric juice (c) and hemolymph (d), and triglyceride in gastric juice (e) and hemolymph (f) of Panulirus argus after feeding diets with different carbohydrate sources. Diets were named according to the carbohydrate source they contained (wheat flour, rice starch, maize starch). Each value is the mean 6SEM (N = 5 lobsters per diet). For each dietary treatment, the time at which the first statistically difference was found with respect to 2 h is marked by an asterisk. Differences among diets throughout the 30-h experiments are marked by different superscript letters in the legend (Two-way ANOVA, Tukey test, P#0.05). doi:10.1371/journal.pone.0108875.g003 Carbohydrate Utilization in Lobster PLOS ONE | www.plosone.org 8 September 2014 | Volume 9 | Issue 9 | e108875 available surface area for enzyme action in small granules) [10,71,72], and type of crystalline polymorphic form [73]. Rice, wheat, and maize starches display the A-type with amylopectins of relatively short branch chains, while potato starch has amylopectins of long branched chains (B-type). In general, A-type starches are more susceptible to a-amylase, especially those of rice and wheat ([73], and reference therein). Rice starch is known to be highly digestible due to its small granule size [74,75] and A-type [73]. Accordingly, native rice starch presented the highest in vitro hydrolysis rate in this study (Fig. 1). For this reason, it is not surprising that the rice flour was also hydrolyzed at a high rate, even higher than other purified starches tested (Fig. 1). Rice flour was the only carbohydrate source tested that produced more free glucose than the expected from the correlation between glucose and maltose released, indicating that products of digestion of rice are better substrates for glucosidase enzymes than the products of hydrolysis of the other carbohydrates studied. Following rice starch, major hydrolysis rates were obtained for gelatinised starches of potato and maize (Fig. 1). During gelatinisation (i.e. heating of starch in presence of water), temperature increases the molecular mobility of pre-hydrated and swelling amorphous regions of starch granule and unravels the double helices of amylose, converting the crystalline material into amorphous/gel material [76] thus increasing digestibility [77]. As in other crustaceans [1,18,78], gelatinisation increased starch digestibility in spiny lobsters such as J. edwardsii [2,3] and P. argus (this study). Nutritional studies in fish [79] and crustaceans (Litopenaeus vannamei [1], Jasus edwardsii [2], Homarus gammarus [18]) provided evidence of the high digestibility of native wheat starch. It is known that high digestibility of native wheat starch is due to the high amylopectin content (,80%) [30,80] of its A-type granules [73]. Accordingly, we observed high in vitro digestibility for wheat flour (Fig. 1, Table 3). Both the rice and the wheat diets presented high in vivo digestibilities without statistical differences (Table 3). On the other hand, maize starch was less digestible in vitro and in vivo than all other starches studied (Fig. 1, Table 3). Maize starch was not well digested in other crustaceans [81–83], including the spiny lobster J. edwardsii [2,3]. Starch from maize has relatively small granules, but a high content of amylose and a polyhedral form, which are two factors that affect hydrolysis negatively [68]. Carbohydrates such as agar, agarose, alginate, and carboxymethyl cellulose (CMC) are often used as binders in formulated feeds [84]. All these carbohydrate sources were deficiently digested Figure 4. Changes in concentration of lactate in hemolymph of Panulirus argus after feeding diets with different carbohydrate sources. Diets were named according to the carbohydrate source they contained (wheat flour, rice starch, maize starch). Each value is the mean 6SEM (N = 5 lobsters per diet). For each dietary treatment, the time at which the first statistically difference was found with respect to 2 h is marked by an asterisk. Differences among diets throughout the 30-h experiments are marked by different superscript letters in the legend (Two-way ANOVA, Tukey test, P#0.05). doi:10.1371/journal.pone.0108875.g004 Table 4. Metabolite levels in digestive gland, muscle and hemolymph of the spiny lobster Panulirus argus 24 h after feeding diets with different carbohydrate sources, or fed with fresh fish under the same experimental conditions. Metabolites Digestive gland Fresh fish Wheat Rice Maize Glucose (mg g 21 ) 1.1360.053 a 0.4160.076 b 0.4960.097 b 0.2460.106 b Glycogen (mg g 21 ) 0.0460.004 a 0.0360.007 a 0.0460.005 a 0.0160.001 b Lactate (mg g 21 ) 0.5260.048 a 0.1660.019 b 0.1760.028 b 0.1360.040 b Amino acid (mg g 21 ) 38.1663.342 a 20.2361.212 b 37.5666.314 ab 16.6862.813 b Triglyceride (mg g 21 ) 91.71618.870 a 16.2667.703 b 34.44617.580 ab 106.26619.210 a Muscle Glucose (mg g 21 ) 4.2760.580 3.6160.455 4.8260.303 4.2160.511 Glycogen (mg g 21 ) 0.2360.017 0.1460.033 0.1360.012 0.1960.032 Lactate (mg g 21 ) 5.9061.075 3.4960.833 4.0060.369 3.8460.745 Amino acid (mg g 21 ) 37.7463.864 26.1963.683 28.7062.294 27.6665.242 Triglyceride (mg g 21 ) 1.3060.194 a 0.5560.133 b 0.6960.114 ab 1.4160.130 a Hemolymph Glucose (mmol L 21 ) 0.5360.043 a 1.1260.195 b 1.40160.092 b 1.0460.110 b Lactate (mg dL 21 ) 2.8960.237 a 1.4160.140 b 1.4860.222 b 1.4560.151 b Amino acid (mmol dL 21 ) 3.6760.715 a 1.1560.311 b 2.0160.376 ab 1.4460.383 b Triglyceride (mg dL 21 ) 34.42614.110 a 7.95860.946 b 13.2762.150 a 14.1361.325 a Diets were named according to the carbohydrate source they contained (wheat flour, rice starch, maize starch). All data are expressed on a dry matter basis. Each value is the mean 6SEM (N = 5 lobsters per diet). Different letters in the same row indicate significant differences among groups (one-way ANOVA, Tukey test, P#0.05). doi:10.1371/journal.pone.0108875.t004 Carbohydrate Utilization in Lobster PLOS ONE | www.plosone.org 9 September 2014 | Volume 9 | Issue 9 | e108875 37. Lesicki A (1977) Activity and kinetic properties of pyruvate kinase from some tissues of the crayfish, Orconectes limosus Raf. (Crustacea: Decapoda) in different stages of the moult cycle. Comp Biochem Physiol 56B: 15–18. 38. Cruz-Ricque LE, Guillaume J, Van Wormhoudt A (1989) Effect of squid extracts on time course appearance of glucose and free amino acids in haemolymph in Penaeus japonicus after feeding: preliminary results. Aquaculture 76: 57–65. 39. Ward LR, Carter CG, Crear BJ (2003a) Apparent digestibility of potential 5 ingredients as protein sources in formulated feeds for the southern rock lobster Jasus edwardsii. In: The Nutrition of Juvenile and Adult Lobsters to Optimize Survival, Growth and Condition(ed. KC Williams), 40–49. Final Report of FRDC 2000/212. Fisheries Research & Development Corporation, Canberra, Australia: Fisheries Research and Development Corporation. 40. Glencross B, Smith M, Curnow J, Smith D, Williams K (2001). The dietary protein and lipid requirements of post-puerulus western rock lobster, Panulirus cygnus. Aquaculture 199: 119–129. 41. Crear BJ, Forteath GNR (2002). Feeding has the largest effect on ammonia excretion rate of the southern rock lobster, Jasus edwardsii, and the western rock lobster, Panulirus cygnus. Aquacult Eng 26: 239–250. 42. Smith DM, Willians KS, Irvin S, Barclay M, Tabrett S (2003) Development of a pelleted feed for juvenile tropical spiny lobster (Panulirus ornatus): response to dietary protein and lipid. Aquacult Nut 9: 231–237. 43. Smith DM, Williams KC, Irvin SJ (2005) Response of the tropical spiny lobster Panulirus ornatus to protein content of pelleted feed and to a diet of mussel flesh. Aquacult Nutr 11: 209–217. 44. Ward LR, Carter CG, Crear BJ, Smith DM (2003b) Optimal dietary protein level for juvenile southern rock lobster, Jasus edwardsii, at two lipid levels. Aquaculture 217: 483–500. 45. Johnston DJ, Calvert KA, Crear BJ, Carter CG (2003). Dietary carbohydrate/ lipid ratios and nutritional condition in juvenil southern rock lobster, Jasus edwardsii. Aquaculture 220: 667–682. 46. Williams KC (2007) Nutritional requirements and feeds development for postlarval spiny lobster: A review. Aquaculture 263: 1–14. 47. Simon CJ, Jeffs A (2008) Feeding and gut evacuation of cultured juvenile spiny lobsters, Jasus edwardsii. Aquaculture 280, 211–219. 48. Bradford M (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254. 49. Robyt JF, Whelan WJ (1968) The ß-amylases. In: Radley, J. A. (Ed.), Starch and its Derivates. Academic Press, London, 477–497. 50. Perera E, Moyano FJ, Rodriguez-Viera L, Cervantes A, Martı ´nez-Rodrı ´- guezhttp://www.sciencedirect.com/science/article/pii/S0044848610006903 – af0015 G, et al. (2010) In vitro digestion of protein sources by crude enzyme extracts of the spiny lobster Panulirus argus (Latreille, 1804) hepatopancreas with different trypsin isoenzyme patterns. Aquaculture 310: 178–185. 51. Perera E, Rodrı ´guez-Viera L, Rodrı ´guez-Casariego J, Fraga I, Carrillo O, et al. (2012a) Dietary protein quality differentially regulates trypsin enzymes at the secretion and transcription levels in the lobster (Panulirus argus) by distinct signaling pathways. J Exp Biol 215: 853–862. 52. Lyle WG, MacDonald CD (1983) Molt stage determination in the Hawaiian spiny lobster Panulirus marginatus. J Crust Biol 3: 208–216. 53. Simon CJ (2009d) Feed intake and its relation to foregut capacity in juvenile spiny lobsters, Jasus edwardsii. New Zeal J Mar Fresh 43: 195–203. 54. Irvin SJ, Tabrett SJ (2005) A novel method of collecting fecal samples from spiny lobsters. Aquaculture 243: 269–272. 55. AOAC (1990) Official methods of analysis. 15 Ed. Association of Official Analytical Chemist. Washington D.C. 1108 56. Radford CA, Marsden ID, Davison W, Taylor HH (2005) Haemolymph glucose concentrations of juvenile rock lobsters, Jasus edwardsii, feeding on different carbohydrate diets. Comp Biochem Physiol 140A: 241–249. 57. Perdomo-Morales R, Montero-Alejo V, Perera E, Pardo-Ruiz Z, AlonsoJimenez E (2007) Phenoloxidase activity in the hemolymph of the spiny lobster Panulirus argus. Fish Shellfish Immunol 23: 1187–1195. 58. Hernandez-Lopez J, Gollas-Galvan T, Gomez-Jimenez S, Portillo-Clark G, Vargas-Albores F (2003) In the spiny lobster (Panulirus interruptus)the prophenoloxidase is located in plasma not in haemocytes. Fish Shellfish Immun 14: 105–114. 59. Yemm EW, Cocking EC, Ricketts RE (1955) The determination of aminoacids with ninhydrin. Analyst 80: 209–214. 60. Rosen H (1957) A Modified Ninhydrin Calorimetric Analysis for Amino Acids. Arch Biochem Biophys 67: 10–15. 61. Keppler D, Decker K (1974). Glycogen. Determination with amyloglucosidase. In: Bergmeyer HU, Ed. Methods of Enzymatic Analysis. New York: Academic Press. p 127–1131. 62. Dall W, Moriarty DJW (1983) Functional aspects of nutrition and digestion. In: L. H Mantel, editor. The biology of crustacea, vol 5. Internal anatomy and physiological regulation. New York: Academic Press. 215–261. 63. Rosas C, Bolongaro-Crevenna A, Sa´nchez A, Gaxiola G, Soto L, et al. (1995) Role of digestive gland in the energetic metabolism of Penaeus setiferus. Biol Bull 189: 168–174. 64. Laiz-Carrio´n R, Martı ´n del Rı ´o MP, Mı ´guez JM, Mancera JM, Soengas J L (2003) Influence of cortisol on osmoregulation and energy metabolism in gilthead sea bream Sparus aurata. J Exp Zool 298A: 105–118. 65. Sangiao-Alvarellos S, Laiz-Carrio´n R, Guzma´m JM, Martı ´ndelRı ´o MP, Miguez JM, et al. (2003) Acclimation of S. aurata to various salinities alters energy metabolism of osmoregulatory and nonosmoregulatory organs. Am J Physiol Regul Integr Comp Physiol 285: R897–R907. 66. Lipcius RN, Herrnkind WF (1982) Molt cycle alterations in behavior, feeding and diel rhythms of a decapod crustacean, the spiny lobster Panulirus argus. Mar Biol 68: 241–252. 67. Tester RF, Qi X, Karkalas J (2006) Hydrolysis of native starches with amylases. Anim Feed Sci Technol 130: 39–54. 68. Bule´on A, Colonna P, Planchot V, Ball S (1998) Starch granules: structure and biosynthesis. Int J Biol Macromol 23: 85–112. 69. Berry C (1986) Resistant starch: formation and measurement of starch that survives exhaustive digestion with amylolytic enzymes during the determination of dietary fibre. J Cereal Sci 4: 301–314. 70. Englyst HN, Cummings JH (1985) Digestion of the polysaccharides of some cereal foods in the human small intestine. Am J Clin Nutr 42: 778–787. 71. Franco CML, Preto SJR, Ciacco CF, Geraldo B (1992) Factors that affect the enzymatic degradation of natural starch granules – Effect of the size of the granules. Starch 44: 422–426. 72. Kong BW, Kim JI, Kim MJ, Kim JC (2003) Porcine pancreatic alpha-amylase hydrolysis of native starch granules as a function of granule surface area. Biotechnol Prog 19: 1162–1166. 73. Jane JL, Wong KS, McPherson AE (1997) Branch-structure difference in starches of Aand B-type X-ray patterns revealed by their Naegeli dextrins. Carbohydr Res 300: 219–227. 74. Manelius R, Bertoft E (1996) The effect of Ca 2+ ions on the a-amylolysis of granular starches from oats and waxy-maize. J Cereal Sci 24: 139–150. 75. Bednar GE, Patil AR, Murray SM, Grieshop CM, Merchen NR, et al. (2001) Starch and fiber fractions in selected food and feed ingredients affect their small intestinal digestibility and fermentability and their large bowel fermentability in vitro in a canine model. J Nutr 131: 276–286. 76. Tester RF, Karkalas J, Qi X (2004) Starch-composition, fine structure and architecture. J Cereal Sci. 39: 151–65. 77. Campanha R, Franco C (2011) Gelatinization properties of native starches and their Naegeli dextrins. J Therm Anal Calorim 106: 799–804. 78. Davis DA, Arnold CR (1993) Evaluation of five carbohydrate sources for Penaeus vannamei. Aquaculture 114: 285–292. 79. Stone DAJ (2003) Dietary carbohydrate utilization by fish. Rev Fish Sci 11: 337–369. 80. Cruz-Suarez LE, Ricque-Marie D, Pinal-Mansilla JD, Wesche-Ebelling P (1994) Effect of different carbohydrate sources on the growth of Penaeus vannamei: economic impact. Aquaculture 123: 349–360. 81. Capuzzo JM, Lancaster BA (1979) The effects of dietary carbohydrate levels on protein utilization in the American lobster (Homarus americanus). Proc World Maric Soc 10: 689–700. 82. Bordner CE, D’Abramo LR, Conklin DE (1983) Assimilation of nutrients by culture hybrid lobsters (Homarus sp.) fed experimental diets. J World Maric Soc 14: 11–24. 83. Koshio S, Castell JD, O’Dor RK (1992) The effect of different dietary energy levels in crab-protein-based diets on digestibility, oxygen consumption, and ammonia excretion of bilaterally eyestalk-ablated and intact juvenile lobsters, Homarus americanus. Aquaculture 108: 285–297. 84. Volpe MG, Monetta M, Di Stasio M, Paolucci M (2008) Rheological behavior of polysaccharide based pellets for crayfish feeding tested on growth in the crayfish Cherax albidus. Aquaculture 274: 339–346. 85. Omondi JG, Stark JR (1996) In vitro carbohydrate digestibility tests in the Indian white shrimp, Penaeus indicus. Aquaculture 139: 315–328. 86. Pavasovic M, Richardson NA, Andersonb AJ, Mann D, Mather PB (2004) Effect of pH, temperature and diet on digestive enzyme profiles in the mud crab, Scylla serrata. Aquaculture 242: 641–654. 87. Perera E, Dı ´az-Iglesias E, Fraga I, Carrillo O, Galich G (2007) Effect of body weight, temperature and feeding on the metabolic rate in the spiny lobster Panulirus argus (Latreille, 1804). Aquaculture 265: 261–270. 88. Irvin SJ, Williams KC (2007) Apparent digestibility of selected marine and terrestrial feed ingredients for tropical spiny lobster Panulirus ornatus. Aquaculture 269: 456–463. 89. Ward LR, Carter CG (2009) An evaluation of the nutritional value of alternative lipid sources to juvenile southern rock lobster, Jasus edwardsii. Aquaculture 296: 292–298. 90. Lee RF, Puppione DL (1978) Serum lipoproteins in the spiny lobster, Panulirus interruptus. Comp Biochem Physiol 59B: 239–43. 91. Grillo LAM, Majerowicz D, Gondim KC (2007) Lipid metabolism in Rhodnius prolixus (Hemiptera: Reduviidae): Role of a midgut triacylglycerol-lipase. Insect Biochem Molec 37: 579–588. 92. Panserat S, Perrin A, Kaushik S (2002) High Dietary Lipids Induce Liver Glucose-6-Phosphatase Expression in Rainbow Trout (Oncorhynchus mykiss). J Nutr 132: 137–141. 93. Parks EJ (2001) Effect of dietary carbohydrate on triglyceride metabolism in humans. J Nutr 131: 2772S–2774S. 94. Song Z, Yang L, Shu G, Lu H, Sun G (2013) Effects of the n-6/n-3 polyunsaturated fatty acids ratio on postprandial metabolism in hypertriacylglycerolemia patients. Lipids Health Dis 12: 181. 95. Fukuda N, Ito T, Ohwada K, Fujii J (2013) Upregulation of fatty acid synthesis and the suppression of hepatic triglyceride lipase as a direct cause of hereditary Carbohydrate Utilization in Lobster PLOS ONE | www.plosone.org 16 September 2014 | Volume 9 | Issue 9 | e108875 postprandial hypertriglyceridemia in rabbits. J Clin Biochem Nutr 53(2): 114– 121. 96. Boulton AP, Huggins AK (1970) Glycolitic activity in crustaceans. Comp Biochem Physiol 33: 491–498. 97. Valentini G, Chiarelli LR, Fortin R, Dolzan M, Galizzi A, et al. (2002) Structure and function of humanerythrocyte pyruvate kinase. Molecular basis of nonspherocytic hemolytic anemia. J Biol Chem 277: 23807–23814. 98. Mun˜oz ME, Ponce E (2003) Pyruvate kinase: current status and functional properties. Comp Biochem Physiol 135B: 197–218. 99. Hochachka PM, Freed JM, Somero GN, Prosser L (1971) Control sites in glycolysis of crustacen muscle. Int J Biochem 2: 25–130. 100. Wells RMG, Lu J, Hickey AJR, Jeffs AG (2001) Ontogenetic changes in enzyme activities associated with energy production in the spiny lobster, Jasus edwardsii. Comp Biochem Physiol 130B: 339–347. 101. Speed SR, Baldwin J, Wong RJ, Wells RMG (2001) Metabolic characteristic of muscles in the spiny lobster, Jasus edwardsii, and responses to emersion during simulated live transport. Comp Biochem Physiol 128A: 435–444. 102. Ji H, Bradley TM, Tremblay GC (1996) Lactate-dependent gluconeogenesis and atractyloside-sensitive flux through pyruvate carboxylase are reduced during smoltification of Atlantic salmon (Salmo salar). J Exp Zool 276: 375– 386. 103. Hers HG, Hue L (1983) Gluconeogenesis and related aspects of glycolysis. A Rev Biochem 52: 617–653. 104. Suarez RK, Mommsen TP (1987) Gluconeogenesis in teleost fish. Can J Zool 65: 1869–1882. 105. Milligan CL, Mcdonald DG (1988) In vivo lactate kinetics at rest and during recovery from exhaustive exercise in coho salmon (Oncorhynchus kisutch)and starry flounder (Platichthys stellatus). J Exp Biol 135: 119–131. 106. Thabrew MI, Poat PC, Andmunday KA (1971) Carbohydrate metabolism in Carcinus maenas gill tissue. Comp Biochem Physiol 40B: 531–541. 107. Walsh PJ, Henry RP (1990) Activities of metabolic enzymes in the deep-water crabs Chaceon fenneri and C.quinquedens and the shallow-water crab Callinectes sapidus. Mar Biol 106: 343–346. 108. Lallier FH, Walsw PJ (1991) Metabolic potential in tissues of the blue crab, Callinectes sapidus. Bull Mar Sci 48: 665–669. 109. Henry RP, Booth CE, Lallier FH, Walsh PJ (1994) Post-exercise lactate production and metabolism in three species of aquatic and terrestrial decapod crustaceans. J Exp Biol 186: 215–234. 110. Ellington WR (1983) The recovery from aerobic metabolism in invertebrates. J Exp Zool 228: 431–444. 111. Taylor EW, Whiteley NM (1989 Oxygen transport and acid–base balance in the haemolymph of the lobster, Homarus gammarus, during aerial exposure and resubmersion. J Exp Biol 144: 417–436. 112. Bridges CR (2001) Modulation of haemocyanin oxygen affinity: properties and physiological implications in a changing world. J Exp Biol 204: 1021–1032. 113. Mayzaud P, Conover RJ (1988) O:N atomic ratio as a tool to describe zooplankton metabolism. Mar Ecol Prog Ser 45: 289–302. 114. Regnault M (1993) Effect of severe hypoxia on some aspects of nitrogen metabolism in the crab Cancer pagurus. Mar Behav Physiol 22: 131–140. 115. Voet D, Voet JG (2011) Biochemistry. 4 th Edition, John Wiley & Sons, 1300 p. 116. Perera E, Rodrı ´guez-Casariego J, Rodrı ´guez-Viera L, Calero J, PerdomoMorales R, et al. (2012b) Differences among trypsin isoforms account for variation in protein digestion in the spiny lobster Panulirus argus: kinetic approximation to physiological efficiency. Biol Bull 222: 158–170. 117. Venou B, Alexis MN, Fountoulaki E (2003) Effects of extrusion of wheat and corn on gilthead sea bream (Sparus aurata) growth, nutrient utilization efficiency, rates of gastric evacuation and digestive enzyme activities. Aquaculture 225: 207–223. 118. Xiao X, Han D, Zhu X, Yang Y, Xie S, et al. (2014) Effect of dietary cornstarch levels on growth performance, enzyme activity and hepatopancreas histology of juvenile red swamp crayfish, Procambarus clarkii (Girard). Aquaculture 426–427: 112–119. 119. Foruichi M, Morita K, Yone Y (1983) Effects of carboxymethyl cellulose supplements on the absortion of dietary nutrients, and on the levels of blood sugar and plasma amino nitrogen. Bulletin of the Japanese Society of Scientific Fisheries 49: 1367–1370. Carbohydrate Utilization in Lobster PLOS ONE | www.plosone.org 17 September 2014 | Volume 9 | Issue 9 | e108875