Temporal repeatability of metabolic rate and the effect of organ mass and enzyme activity on metabolism in European eel (Anguilla anguilla)

Comparative Biochemistry and Physiology, Part A 165 (2013) 22–29

Contents lists available at SciVerse ScienceDirect

Comparative Biochemistry and Physiology, Part A

j ourna l homepage: www.e lsev ie r .com/ locate /cbpa

Temporal repeatability of metabolic rate and the effect of organ massand enzyme activity on metabolism in European eel (Anguilla anguilla)

Martin Maagaard Boldsen ⁎, Tommy Norin, Hans MalteZoophysiology, Department of Bioscience, Aarhus University, DK-8000 Aarhus C, Denmark

Abbreviations: BSA, bovine serum albumin; CS, citratoxidase; DTNB, dithionitrobenzoic acid; DTT, dithiothtetraacetic acid; HEPES, hydroxylethyl piperazine ethanhydroxide; MMR, maximummetabolic rate; Mo2, oxygenintegrated transponder; rRMR, residual routine metabolimetabolic rate; rSum organ, residual of summed organrate; SGR, specific growth rate; SMR, standard metaboacid; Tris, hydroxyl methyl aminomethane hydrochloride⁎ Corresponding author.

E-mail address: [email protected] (M.M. Bo

1095-6433/$ – see front matter © 2013 Elsevier Inc. Allhttp://dx.doi.org/10.1016/j.cbpa.2013.01.027

a b s t r a c t

Article history:Received 21 November 2012Received in revised form 30 January 2013Accepted 31 January 2013Available online 4 February 2013

Keywords:FishIntraspecific variationRoutine metabolic rateStandard metabolic rate

Intraspecific variation in metabolic rate of fish can be pronounced and have been linked to variousfitness-related behavioural and physiological traits, but the underlying causes for this variation have receivedfar less attention than the consequences of it. In the present study we investigated whether European eels(Anguilla anguilla) displayed temporal repeatability of body-mass-corrected (residual) metabolic rate over atwo-month period and if variations in organ mass and enzyme activity between individual fish could be thecause for the observed variation in metabolic rate. Both standard metabolic rate (SMR; Pearson's r=0.743)and routine metabolic rate (RMR; r=0.496) were repeatable over the two-month period. Repeatability ofRMR is an interesting finding as it indicates that the level of spontaneous activity in respirometer-confinedfish is not random. Cumulative organ mass (liver, heart, spleen and intestine; mean 1.6% total body mass) wasfound to explain 38% of the variation in SMR (r=0.613) with the liver (one of the metabolically most activeorgans) being the driver for the correlation between organ mass and metabolic rate. No relationships werefound for either liver citrate synthase or cytochrome oxidase activity and metabolic rate in the Europeaneels. Reasons for, and contributions to, the observed variation in metabolic rate are discussed.

© 2013 Elsevier Inc. All rights reserved.

1. Introduction

The cost of living has received much attention over the years asenergy utilisation plays a major role in shaping the behaviour, ecologyand physiology of animals. The minimum cost of living in ectothermsis termed standard metabolic rate (SMR). It is defined as the minimumaerobic maintenance metabolism of a resting individual in a post-absorptive state with no environmental and physiological stress (Brettand Gloves, 1979; Priede, 1985). The upper limit for aerobicmetabolismis the maximum metabolic rate (MMR) and the difference betweenMMR and SMR is the aerobic scope. Elevations in metabolic rate aboveSMR, occurring within this scope and caused by spontaneous anddaily routine activity, is termed routine metabolic rate (RMR). Thevariability in metabolic rate differs between species (interspecific)as well as within species (intraspecific) (Schmidt-Nielsen, 1984;

e synthase; CytOx, cytochromereitol; EGTA, ethylene glycolesulfonic acid; KOH, potassiumconsumption rate; PIT, passivec rate; rSMR, residual standardmass; RMR, routine metaboliclic rate; TNB, thionitrobenzoic.

ldsen).

rights reserved.

Clarke and Johnston, 1999; Glazier, 2005). Interestingly, intraspecificvariation in SMR still exists even when ontogenetic processes aretaken into consideration (i.e., development and total body mass ad-justments) (Burton et al., 2011). Many have tried to unravel thefunctional significance of variation in metabolic rate and how it af-fects the individual animal. These are important issues because nat-ural selection acts on individuals and not on populations. To assessthese questions, one has to show that a physiological trait (such asSMR) in an individual is repeatable. Once this is achieved it is possi-ble to address potential explanations for the observed variation inSMR.

Nespolo and Franco (2007) reviewed the repeatability of whole-animal metabolic rates in several animal groups. They found thatwhole-animal metabolic rate is a repeatable trait and concludedthat no further studies are needed. However, in their meta-analysisNespolo and Franco did not take into account that metabolic rate is afunction of body mass which should be accounted for (Konarzewskiet al., 2005). In the study by Nespolo and Franco (2007) there wasalso a bias towards endotherms, and only two out of the forty-four stud-ies addressed were on fish. A few additional studies on repeatability ofmetabolic rates in fishes are available, but these are concerned mostlywith juveniles, salmonids or both (Cutts et al., 1998; McCarthy, 2000;O'Connor et al., 2000; Cutts et al., 2001; Fu et al., 2007; Seppänen et al.,2010; Norin and Malte, 2011). In other studies, concerning repeatabilityofmetabolic rates in teleosts, there are some kinds of treatments involved(Reidy et al., 2000; Virani and Rees, 2000; Maciak and Konarzewski,2010). To our knowledge, no study to date has assessed the repeatability

23M.M. Boldsen et al. / Comparative Biochemistry and Physiology, Part A 165 (2013) 22–29

of individual residual (body-mass-corrected) metabolic rate in an adultteleost species without any particular treatment involved.

It has been suggested, that in periods that require sustained powerfor migration, or in periods where food is readily available, a highSMR can be beneficial. The reasoning is that high SMR may producea larger aerobic scope, as high SMR may be associated with a dispro-portionately large MMR, leading to greater potential for growth(O'Connor et al., 2000). This has been shown to be true in some stud-ies with salmonids (Metcalfe et al., 1995; Cutts et al., 1998; Millidineet al., 2009). On the other hand, high SMR may be a drawback duringenvironmental hypoxia because of a decrease in aerobic scope togetherwith a high maintenance metabolism. The same argument can alsobe used for other stressors such as increasing temperature or fooddeprivation.

Whole-animal metabolic rate is the sum of the metabolic rate of allthe individual organs and tissues such as the liver, kidney, heart,brain, muscles and intestine. These organs contribute to total metabolicrate by the product of mass and enzyme intensity (Hulbert and Else,2000). As is the case with metabolic rate, the organ mass scalesallometrically with total body mass and the scaling exponent of mostorgans is b1 (i.e., as total body mass increases, the mass-specific organmass decreases). Different organs have different scaling exponentsand it has been suggested that this may partly explain the variation inSMR between different sized animals (Schmidt-Nielsen, 1984). Thus,part of the variation in SMR is due to variation in the relative massof organs and the remainder is due to differences in mass-specificorgan metabolic rate (Krebs, 1950; Rolfe and Brown, 1997; Hulbertand Else, 2000). A large number of studies have assessed the correla-tion between metabolic rate and organ mass in mammals and birds(Daan et al., 1990; Konarzewski and Diamond, 1995; Meerlo et al.,1997; Burness et al., 1998; Speakman and Johnson, 2000; Książeket al., 2004; Speakman et al., 2004; Song and Wang, 2006; Russelland Chappell, 2007). In addition, some studies have assessed therelationship between enzyme activity and total body mass in dif-ferent vertebrate groups (Emmett and Hochachka, 1981; Someroand Childress, 1990; Tripathi and Verma, 2004), indirectly linkingmetabolic rate with enzyme activity. Simon and Robin (1971)showed a close exponential relationship between basal metabolicrate and myocardial cytochrome oxidase (CytOx) activity amongstdifferent vertebrates with varying oxygen consumption rates ( _Mo2)and body mass. Likewise, the oxygen consumption rate of variousrabbit tissues is related to tissue activity of CytOx (Simon andRobin, 1971).

In fish, the available information on the variation in organ mass andenzyme activity as a predictor of metabolic rate is limited with onestudy investigating the interspecific relationship between key metabolicenzyme activity and metabolic rate in marine fish (Childress andSomero, 1979). In addition, only a couple of studies have dealt with theeffect of organ mass and enzyme activity on intraspecific variation inmetabolic rate (Odell et al., 2003; Norin and Malte, 2012) warranting amore thorough investigation into the causes for inter-individual varia-tion in metabolic rate.

The aim of the present study was to investigate whether metabolicrate is temporally repeatable in individual European eels (Anguillaanguilla Linnaeus 1758) and whether organ mass and/or activity oftwo important metabolic enzymes can explain variation in individ-ual SMR. Cytochrome oxidase was selected for the study because ofits function as the terminal oxidase in the electron transport chain.It is therefore suggested that CytOx is a quantitative biochemicalmarker for functional oxidative phosphorylation. Citrate synthase(CS) was selected for the study because it functions as the pace-makingenzyme in the citric acid cycle that begins with the condensation ofoxaloacetate and the acetyl group of acetyl-CoA to yield citrate andCoA (Wiegand and Remington, 1986). Thus, citrate synthase mayform a bottleneck in the remainder part of the citric acid cycle andoxidative phosphorylation.

2. Material and methods

2.1. Experimental animals

2.1.1. Repeatability experimentEuropean eels (A. anguilla) were obtained from Lyksvad eel farm

(Vamdrup, Denmark) in December 2009. The eels were kept in a 170 Ltank (density: 47 kg m−3) with well aerated tap water (PO2 >18 kPa).Approximately half of the water was replaced once a week. The dailyphotoperiod throughout the experiment was 12 h:12 h light:dark.

2.1.2. Organ and enzyme experimentsEels were obtained from Lyksvad eel farm (Vamdrup, Denmark) in

early April 2010. The fish were maintained in a 280 L (density:12 kg m−3) darkened glass aquarium with well-aerated tap water(PO2 >18 kPa). A five-watt UV-filter was applied to reduce bacterialgrowth. Approximately one third of the water was replaced everyother day. Photoperiod followed natural daily rhythm (latitude56°09′ N, longitude 10°12′ E). Experiments were performed in midOctober 2010.

In both the repeatability and the organ and enzyme experiments,the eels were fed a diet consisting of commercial 3 mm DAN-EX fishpellets (Biomar A/S, Brande, Denmark) at a moderate rate of approxi-mately 0.5% body mass day−1. The holding temperature was 20.5±3 °C. The sex ratio in both populations was unknown.

2.2. Tagging and measurements

2.2.1. Repeatability experimentPrior to the experiment the eelswere anaesthetised in aeratedwater

containing benzocaine (ethyl p-aminobenzoate). The anaestheticwas prepared by dissolving 1.5 g of benzocaine in 5 mL acetoneadded to 5 L tap water to give a final concentration of 0.3 gbenzocaine L water−1. When the animal's movement and ventilationceased, the following morphometric parameters were measured: bodymass (Mb, g), body length (BL, cm), eye horizontal diameter (ED, mm)and eye vertical diameter (VD, mm). From this, condition factor (K)(Bolger and Connelly, 1989) was calculated as

K ¼ Mb � BL−b� �

� 100;

where b=2.76 is the exponent describing the mass vs. length relation-ship in the present study (data not shown). Eye index (IE) was calculatedas

IE ¼ EDþ EVð Þ=4ð Þ2 � πh i

= 10� BL½ �h i

� 100;

which is an indirect measure of sexual maturity where eels with IE>6.5can be classified as sexually mature (Pankhurst, 1982).

Succeeding the morphometric measurements, a small incision wasmade through the mid-ventral body wall into the posterior peritonealcavity, and a 13.5×2.1 mm (weight: 0.1 g) FDX-B Passive IntegratedTransponder (PIT) tag (LoligoSystems, Tjele, Denmark) was pushedinto the peritoneal cavity. According to Baras and Jeandrain (1998)eels with unclosed incisions have a high survival, maximum tag re-tention rate and heal their incision within a month. Surgery lastedless than 5 min per eel. With a handheld scanner the ID (a 15-digitcode) of the individual fish could be recorded. This allowed individualtracking of the fish during the two trial periods in April and June2010. None of the PIT-tags were lost during the experimental period.Three fish were omitted from the final experiment; one due to abnormalbehaviour, one escaped from the tank and one died during respirometrymeasurement because of pump malfunction.

24 M.M. Boldsen et al. / Comparative Biochemistry and Physiology, Part A 165 (2013) 22–29

2.3. Measurement of oxygen consumption

The instantaneous metabolic rate of post-absorptive eels (fasted for20–24 h) was measured as whole-animal oxygen uptake rate ( _Mo2,μmol min−1) using computerised intermittent-closed respirometry(Steffensen, 1989) by means of a LoliResp system (LoligoSystems,Tjele, Denmark). Two acrylic respirometers (vol. 2.4 L) were sub-merged in a 90 L tank. Both of the respirometers were connected to aMINI-DO oxygen electrode (LoligoSystems, Tjele, Denmark) through aset of gas-proof tubing and an Eheim 1046 in-line recirculation pump(EHEIM GmbH & Co. KG, Deizisau, Germany) with a flow rate of5 L min−1. This ensured adequate mixing of water inside the respi-rometer and continuously led water past the oxygen electrode. Theoxygen electrodes were calibrated in oxygen free water (0% calibration)consisting of sodium sulfite (Na2SO3) added to a 0.01 M borax(Na2B4O7) solution as well as in fully aerated water (100% calibration)every day. Both calibrations were done at the experimental temperature.

Each intermittent cycle lasted 6 min and 45 s and measurements of_Mo2 continued overnight for ~18 h (~160 _Mo2 measurements per fish).A cycle beganwith a flush period using a submerged Eheim 1048 pump(EHEIM) with a flow rate of 10 L min−1, one pump connected to eachrespirometer. The flush period lasted 2 min and ensured that >99% ofthe water in the respirometers was exchanged with the fully aeratedsurrounding water (Steffensen, 1989). This procedure minimises thebuild-up of CO2 and other excretory products inside the respirometers.The flush period was followed by a forty-five second wait period andthen finally a four-minute measuring period. During the measuringperiod the oxygen consumption was calculated from the slope ofthe regression line of PO2 vs. time, using the standard formula:

_Mo2 ¼ k� α � Vresp;

where k is the slope of the regression line (kPa min−1),α is the oxygensolubility in the water (μM kPa−1) and Vresp is the volume of the res-pirometer minus the volume of the fish [L, (assuming a density of1 kg L−1)]. Changes in water PO2 during the measuring period wasmonitored at 1 Hz by the AutoResp™ software (LoligoSystem). Thedecline in water PO2 during the measuring period depended on the_Mo2 of the fish but rarely exceeded more than 2–3 kPa. The temper-ature in the experimental tank was kept at 20.5±1.0 °C by means ofa thermostatic heater. After _Mo2 measurements were concluded,background (bacterial) respiration in the system was determinedfor 1 h. Although bacterial respiration was minor, it was alwaysaccounted for by subtracting bacterial _Mo2 from the total _Mo2(fish+bacterial) during experiments. After each day of experimentall submerged equipment were disassembled, disinfected andcleaned before the next experiment. Fish used in the repeatabilityexperiment were measured for _Mo2 in mid April 2010 (Trial 1) andagain in early June 2010 (Trial 2). In the organ and enzyme activityexperiment measurements of _Mo2 were performed in October 2010.

2.4. Dissection of organs

In the organ and enzyme experiment,when _Mo2 measurementswereterminated, the two eels in the respirometers were killed by cerebralconcussion and subsequently decapitated. Next, the liver, heart, spleenand intestine plus pyloric caeca (combined) were removed. Individualorgans that had been dissected outwere transferred to a vial containingcotton soaked in a physiological solution (300 mOsm) to avoid desicca-tion while handling other organs. Adipose tissue, situated around theintestine and pyloric caeca, was removedmechanically. After dissection,the organs wereweighed to the nearest 0.1 mg. A small part of the liverwas cut free, weighed and flash frozen in liquid nitrogen before beingstored in a freezer at −80 °C. The remaining part of the liver, as wellas the other organs, was dried to constant mass in an oven at 70 °C for48 h and reweighed. Liver samples were taken from the freezer and

stored on ice-cold buffer (5 mM HEPES, 1 mM EGTA, 1 mM DTT and0.10% Triton X-100 adjusted to pH 8.2 at 20 °C with KOH) beforebeing homogenised in a Retsch TissueLyser (Qiagen, CA, USA) for 45 safter which they were refrozen for later protein and enzyme assays.Pilot experiments had shown that homogenate dilutedwith homogeni-sation buffer to concentrations of 10 mg mL−1 and 70 mg mL−1 wasappropriate for protein and enzyme assays, respectively, and thatrefreezing had no effect on later protein analysis.

2.5. Protein assay

To determine total protein content of liver samples, we used acalorimetric Bio-Rad RC DC Protein Assay kit (Bio-Rad Laboratories,Inc., CA, USA) based on the Lowry et al. (1951) assay. Bovine serumalbumin (BSA) was used as a protein standard and absorbance wasread at 750 nm.

2.6. Enzyme activity

The activity of cytochrome oxidase (CytOx; EC 1.9.3.1) wasassessed spectrophotometrically at 550 nm from the oxidation ofcytochrome c according to Christensen et al. (1994). Liver homogenate(40 μL) with a concentration of 70 mg mL−1 was added to a cuvettecontaining 1.2 mL, 13 mM K-phosphate (pH 7.4 at 20 °C) and32.5 μM cytochrome c (bovine heart, purity 61%) that was reducedwith Na-dithionite. Absorbance, as a function of time, was recorded at1 Hz for 200 s. Activitywasmeasured as the fractional change in reducedcytochrome c over time. Since the reaction show first-order kinetics withthe first-order rate constant (km, min−1) being related to the tissueconcentration (g mL−1) in the reaction solution, activity of CytOxis calculated as km divided by the tissue concentration and expressedin mL min−1 g tissue protein−1.

The activity of citrate synthase (CS; EC 2.3.3.1)was assessed spectro-photometrically at 412 nm as described by Birkedal and Gesser (2006).Liver homogenate (20 μL) with a concentration of 70 mg mL−1 wasadded to a cuvette containing 1.2 mL buffer consisting of 100 mMTris, 0.3 mM acetyl CoA (purity 85%), 0.5 mM oxaloacetic acid and0.1 mM DTNB. Absorbance as a function of time was recorded at 1 Hzfor 150 s. The zeroth-order reaction was coupled to the production ofthionitrobenzoic acid (TNB) and the rate constant (slope) obtainedwas used to express the activity of CS. Activity of CS was expressedin μmol min−1 gtissue protein−1 using an extinction coefficient of13.6×103 M−1 cm−1.

The activity of each enzyme was determined twice for each replicate,and the average rate constant of these replicateswas used for further dataanalysis. The activity of both enzymeswas assessed at 21 °C, the acclima-tion temperature of the eels. This was achieved with a thermostattedwater bath connected to the recording cell of the spectrophotometer.All enzyme experiments were conducted in two successive days inNovember 2010 and protein analyses at one day in November 2010.

2.7. Data analysis

Oxygen consumption data were analysed using a Mathematica(Wolfram Research, Inc., Champaign, IL, USA) script and all statisticalanalyses were performed in SigmaPlot® 11.0 (Systat Software Inc.,San Jose, CA, USA). SMR for each individual eel was determined bytaking the mean of the ten lowest _Mo2 measurements obtained duringthe ~18 h experimental period, after which outliers (±2 s.d. from themean) were excluded and the mean of the remaining _Mo2 measure-ments (minimum six data points; cumulative time of at least 24 min)was used to calculate SMR. The routinemetabolic rate (RMR)was calcu-lated as the average of all the _Mo2 measurements of individual eels.Individual residual (body-mass-corrected)metabolic rate was obtainedfrom the _Mo2 vs. bodymass equations ( _Mo2=a×Mb) from each exper-imental trial. Residual metabolic rate (rSMR and rRMR) was calculated

-80 -40 120 160-80

-60

-40

-20

0

20

40

60

80

Residual SMR1 (% from predicted)

Res

idua

l SM

R2 (

% f

rom

pre

dict

ed)

0 40 80

Fig. 1. Repeatability of residual (body-mass-corrected) standard metabolic rate (%from predicted) from April 2010 to June 2010. (Pearson's r=0.743, Pb0.001, N=24).

25M.M. Boldsen et al. / Comparative Biochemistry and Physiology, Part A 165 (2013) 22–29

as the difference between the observed and predicted (based on bodymass) metabolic rate by these equations. Thus, fish with higher _Mo2than expected for their body mass had positive residuals, while thosewith _Mo2 lower than expected had negative residuals. Where correla-tions between residuals from the two trials were made, residuals areexpressed in percent from that predicted since the absolute residualvalues may change over time, due to the effects of scaling, as the fishincrease in size.

In the organ experiment, residual organmass (body-mass-correctedorgan mass) was also calculated by fitting a power function for eachorgan. This was done for both wet weight and dry weight. Since themass of the spleen did not correlate significantly with total bodymass, absolute values (rather than residuals) were used for furtheranalyses. Data was tested for homogeneity of variance and normalityprior to main parametric tests. Temporal repeatability of metabolicrate, relationships between metabolic rate and organ mass, as well asinter-organ relationships, were evaluated from the Pearson product–moment correlation coefficient (r). Inter-organ correlations wereBonferroni corrected to adjust for multiple comparisons. The overall(unadjusted) significance level was 0.05.Where applicable, subsequentvalues are presented as means±s.e.m.

3. Results

3.1. Repeatability experiment

All eels had eye indices≥6.6 with a mean of 9.8±0.5 and could beclassified as sexuallymature (Pankhurst, 1982). This biometricmeasure-ment is fairly similar to the findings of Pelstra et al. (2008) for farmedeels. Mean condition factor (K) was 0.457±0.010 which is higher thanthe findings of Pelstra et al. (2008). This is due to differences in the eval-uation of condition (Bolger and Connelly, 1989). Bodymass of the 24 eelsranged from 0.184 to 0.507 kg (mean 0.300±0.018 kg) in trial 1 and0.171 to 0.504 kg (mean 0.295±0.018 kg) in trial 2. Mean body massdid not change significantly during the two month period (pairedt-test, t=1.221, P=0.235, N=24). Whole-animal SMR ranged from2.04 to 11.95 μmol min−1 (mean 5.16±0.57 μmol min−1) in trial 1and 1.84 to 8.83 μmol min−1 (mean 4.61±0.39 μmol min−1) in trial2. Corresponding values for RMR were 5.04 to 17.05 μmol min−1

(mean 9.27±0.71 μmol min−1) and 3.56 to 12.91 μmol min−1 (mean8.06±0.52 μmol min−1) in trials 1 and 2, respectively. Mean SMR andRMR did not change significantly between trials (paired t-test: SMR,t=0.826, P=0.417; RMR, t=1.423, P=0.168). The relationship be-tween body mass and whole-animal metabolic rate (SMR and RMR)was fitted with a power function at each trial (Table 1). Residuals

Table 1The relationship between body mass (Mb, kg) and oxygen consumption rate ( _Mo2 ,μmol O2 min−1) for SMR and RMR was fitted according to the power function_Mo2=a×Mb

b. Trial 1 corresponds to the period 7–20 April 2010 and trial 2 to the peri-od 31 May–12 June 2010 (N=24 in both trials).

SMR RMR

Trial 1 Trial 2 Trial 1 Trial 2

a 27.75±1.37 16.18±1.32 38.69±5.40 21.90±4.06b 1.44±0.25 1.05±0.22 1.20±0.13 0.82±0.16r2 0.61 0.52 0.81 0.54F 34.16 23.89 92.32 25.85P b0.001 b0.001 b0.0001 b0.0001Residuals:

Min (%) −62.6 −56.8 −26.9 −53.3Max (%) 118.6 60.8 37.0 35.3CV (%) 31.0 28.2 15.7 23.5

Minimum andmaximum residuals express the two largest deviations observed (negativeand positive, respectively) from predicted.CV, coefficient of variation calculated from the standard deviation of residuals according toGarland (1984).Constants a and b are presented ±s.e.m.

from these relationships showed that SMR of individual eels was re-peatable over the two-month experimental period from April to June(Pearson's r=0.743, Pb0.001; Fig. 1). Most eels had a numericallylow rSMR in both periods, meaning that their cost of living was closeto that predicted by the power functions (i.e., origin in Fig. 1). Consis-tency of residual RMR was also tested for and was also found to be re-peatable (r=0.496, P=0.014; Fig. 2). The specific growth rate (SGR)for individual eels ranged from −0.187% day−1 to 0.272% day−1

meaning that some fish lost weight while others gained weight.Although mean SGR was −0.016±0.027% day−1, and did not differsignificantly from zero (one sample t-test; t=−0.584, P=0.565,N=24), individual SGR correlated positively with rSMR at the endof the experiment (r=0.414, P=0.044; Fig. 3). This implies thateels with a higher than expected SMR grew more compared to con-specifics with a lower than expected SMR.

3.2. Organ correlations and enzyme assay

Body mass of the 19 eels ranged from 0.131 to 0.254 kg(mean 0.180±0.008 kg). Whole-animal SMR ranged from 2.12 to5.74 μmol min−1 (mean 3.49±0.25 μmol min−1). The relationshipbetween whole-animal SMR and body mass was fitted with a powerfunction (Table 2) from which residual SMR was calculated.

Of the organs examined, the liver and intestines, respectively,highly dominated in wet as well as dry weights, while the heart andspleen were the smallest organs (data not shown). The liver and intes-tine combined made up 88.1±0.6% and 90.4±0.5% of the total mass ofall examined organs, wet weight and dryweight respectively. However,

-40 -20-60

-40

-20

0

20

40

60

Residual RMR1 (% from predicted)

Res

idua

l RM

R2 (

% f

rom

pre

dict

ed)

0 20 40 60

Fig. 2. Repeatability of residual (body-mass-corrected) routine metabolic rate (% frompredicted) from April 2010 to June 2010. (Pearson's r=0.496, P=0.014, N=24).

-4 -2-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0 2 4

Residual SMR2 (µmol O

2 min-1)

SGR

(%

day

-1)

Fig. 3. Specific growth rate (SGR, % day−1) as a function of residual (body-mass-corrected)standard metabolic rate (μmol O2 min−1), at the end of the experiment (June 2010).(Pearson's r=0.414, P=0.044, N=24).

Table 3Pearson product–moment correlation coefficients (r) for residual SMR (rSMR) on re-sidual organ mass (N=19) for dry weight and wet weight. Values for spleen are abso-lute mass since this organ did not correlate with total body mass. Significance levels inparentheses. Significant correlations are typed in bold.

Organ rSMR

Dry weight Wet weight

rLiver 0.614 (b0.001) 0.624 (b0.005)rIntestine 0.285 (0.238) 0.379 (0.109)rHeart 0.362 (0.128) 0.434 (0.064)Spleen −0.140 (0.567) −0.015 (0.953)rSum organ 0.571 (b0.011) 0.613 (b0.005)

26 M.M. Boldsen et al. / Comparative Biochemistry and Physiology, Part A 165 (2013) 22–29

the dissected organs (wet weight) only made up 1.64±0.05% of thetotal body mass of the eels. The relationship between total body massand organ mass was fitted with a power function (Table 2) from whichresidual organmasswere calculated. Correlations between rSMR and re-sidual organmass (absolute mass for spleen) are summarised in Table 3.For both dry weight and wet weight, residual mass of the liver and theresidual summed mass of all examined organs correlated positivelywith rSMR. This means that individual eels that had a higher SMR thanexpected for their body mass also had a larger liver. Testing the correla-tion between rSMR and residual organmass usingmultiple linear regres-sion, with rSMR as the dependent variable and residual liver, heart,intestine and spleen mass as independent variables, confirmed thatrSMR can be explained by a linear combination of all the organs in ques-tion for both dry weight (F4,18=3.158, P=0.048) and wet weight(F4,18=3.807, P=0.027), although it was still only the liver that reachedsignificance as a single predictor of rSMR (dry weight, P=0.005; wetweight, P=0.004). From correlations amongst organs, individuals withlarger livers than expected also were found to have larger intestines,whether on a wet or dry weight basis (Table 4). Furthermore, theliver is positively correlated with the summed mass of all the organsmeasured, as is the intestine. Performing Bonferroni corrections, toaddress the problem of multiple comparisons, only had an effect on

Table 2The relationship between SMR (μmol O2 min−1) and bodymass (Mb, kg) according to thepower function SMR=a×Mb

b as well as the relationship between organ mass (Morgan, g)and Mb for liver, intestine, heart, spleen and summed organ mass (N=19) according tothe power function Morgan=a×Mb

b. Note that spleen did not correlate significantly withbody mass.

a b r2 F P

SMR 38.27±13.92 1.40±0.22 0.69 38.39 b0.0001Liver

ww 7.725±2.325 0.843±0.178 0.564 21.996 0.0002dw 1.759±0.622 0.655±0.208 0.357 9.456 0.0069

Heartww 0.999±0.394 1.352±0.238 0.627 28.602 b0.0001dw 0.278±0.135 1.399±0.2935 0.529 19.111 0.0004

Spleenww 0.271±0.160 0.459±0.345 0.101 1.911 0.1847dw 0.051±0.029 0.341±0.333 0.062 1.121 0.3044

Intestineww 55.969±20.506 2.509±0.231 0.870 113.523 b0.0001dw 11.116±4.150 2.322±0.234 0.849 95.360 b0.0001

Organ sumww 26.121±6.826 1.271±0.158 0.787 62.706 b0.0001dw 5.511±1.654 1.078±0.180 0.667 34.039 b0.0001

ww, wet weight; dw, dry weight.Constants a and b are presented ±s.e.m.

body-mass-corrected liver vs. intestine (wet and dry weights) corre-lations and these correlations lost statistical power.

Due to failure in one protein determination the dataset for proteinand enzyme analyseswas reduced fromnineteen to eighteen replicates.Total protein content of the liver ranged from 0.099 to 0.176 gprotein g tissue−1 (mean 0.129±0.006 g protein g tissue−1). Theactivity of liver CytOx ranged from 153 to 2405 mL min−1 g protein−1

(mean 889±147 mL min−1 g protein−1) and the activity of CSranged from 21.5 to 51.6 μmol min−1 g protein−1 (mean 35.0±2.0 μmol min−1 g protein−1). In contrast to the relationship betweenrSMR and residual liver and summed organmass, we found no correla-tion between CytOx activity and rSMR (r=0.019, P=0.942) or CSactivity and rSMR (r=−0.064, P=0.802). Since organs contribute tototal metabolic rate by the product of mass and enzyme intensitywe also tested for a relationship between total liver enzyme activityand rSMR. We found no correlation between either total liver CytOxactivity and rSMR (r=0.183, P=0.452) or total liver CS activity andrSMR (r=0.306, P=0.203).

4. Discussion

4.1. Repeatability

In the present study, we found that rSMR (r=0.743) and rRMR(r=0.496) were repeatable within a two-month period in Europeaneel. These results are in agreement with the growing body of evidencethat repeatability of metabolic rate is a universal phenomenon inmammals, birds and fish. This applies for both whole-animal metabolicrate (Nespolo and Franco, 2007) as well as for body-mass-correctedmetabolic rate. The correlation coefficients (r) in the present study arehigh, especially for SMR, when comparing with the available literatureon fish, and thus confirms the high repeatability of this physiologicaltrait for eels. Norin and Malte (2011) found correlation coefficientsranging from 0.52 to 0.58 for SMR and 0.32 to 0.38 for MMR over afiveweek period in their study on juvenile brown trout, with repeatabil-ity decreasing over longer time periods. Another study by McCarthy

Table 4Correlations between organ masses (N=19) expressed as Pearson product–momentcorrelation coefficient (r). Significance level in parentheses. Significant correlations aretyped in bold. Interactions indicated with (‡) lost statistical power when performingBonferroni test. Values above diagonal refer to dry weights and values below diagonalrefer to wet weights.

rLiver rIntestine rHeart Spleen rSum organ

rLiver – 0.534(0.019)‡

0.169(0.490)

−0.080(0.744)

0.936(b0.0001)

rIntestine 0.524(0.021)‡

– 0.356(0.134)

0.024(0.922)

0.772(b0.0001)

rHeart 0.129(0.600)

0.361(0.129)

– −0.033(0.892)

0.340(0.154)

Spleen −0.069(0.780)

0.013(0.960)

0.019(0.940)

– 0.046(0.851)

rSumorgan

0.898(b0.0001)

0.811(b0.0001)

0.318(0.185)

0.103(0.674)

–

27M.M. Boldsen et al. / Comparative Biochemistry and Physiology, Part A 165 (2013) 22–29

(2000) found correlation coefficients of 0.68 in Atlantic salmon, andMaciak and Konarzewski (2010) found values of 0.68 and 0.73 forspined loach (Cobitis taenia) in normoxia and hypoxia, respectively.Some repeatability studies are conducted overweeks to severalmonthswith three or more periods of measurements, and it is now believedthat there is a continuous decline in repeatability of _Mo2 over longertime periods (Chappell et al., 1995, 1996; Broggi et al., 2009; Norinand Malte, 2011).

An interesting finding is the repeatability of RMR. This indicatesthat the spontaneous activity within the respirometers is not simplyrandom bouts of movement from time to time, but rather, that the in-dividual fish exhibit specific behavioural patterns when evaluated atdifferent times, even within the confinements of a respirometer.

It has been shown that there is a connection between SMR and bothdominance and aggression in Atlantic salmon (Metcalfe et al., 1995). Ithas also been found that dominance correlates with SGR in this species(Metcalfe et al., 1992) indirectly indicating a positive correlation be-tween SGR and SMR in Atlantic salmon. Similarly, rSMR is positivelycorrelated with SGR in southern catfish (Silurus meridionalis) (Fuet al., 2007). Under these circumstances, one could imagine that havinga positive rSMR (i.e., a high cost of living) would be beneficial becausesuch individuals can out-compete relatives due to high performance.These findings concur with our results on European eels, since those in-dividuals that have relatively high SMR (positive rSMR) also have highSGR. Interestingly, eels with slightly negative SGR also had a lowerthan expected SMR (negative rSMR), raising the question whetherSMR is reduced because of low food intake, or if the food intake is lowbecause of a low SMR. Not all studies to date have found positive rela-tionships between rSMR and growth. In the study by Norin and Malte(2011), brown trout with positive rSMR grew less (or the same) thanindividuals with negative rSMR. This means that there is no advantagein having a relatively high SMR, emphasising the potential disadvantageof possessing such a trait in situations like hypoxia or food deprivation.

In the present study, eels were fed a moderate diet throughout theexperiment, and it is possible that this could have weakened therepeatability of SMR, as has been observed in Atlantic salmon andbrown trout (O'Connor et al., 2000; Norin and Malte, 2011). Evenso, there was still a correlation between SGR and rSMR (r=0.414).

The whole-animal SMR reported here is at the lower end of valuesobserved in a number of earlier studies dealing with European eels(see Table 1 in Iversen et al. (2010)), as well as in comparison withother teleost species. Several possibilities could account for the observed,some already suggested by Iversen et al. (2010). Although factors such asenvironment, dietary history, season, size and genetics all influencemetabolism, we suggest that experimental setup andmethod for deter-mination of SMR are probably the main factors for explaining the lowSMR found in this study. Intermittent-closed respirometry allows for agreat resolution in the measurements of _Mo2 and makes it possible toidentify even tiny elevations in _Mo2 due to spontaneous activity. Like-wise, we used a conservative approach for estimating SMR comparedto other studies. With this said, studies using the same respirometrictechnique report values for SMR that are not dissimilar to valuesreported in the present study. For example, Iversen et al. (2010) reportan SMR of 7.9 μmol O2min−1 at 20.5 °C for eelswith amean bodymassof 0.496 kg and McKenzie et al. (2000) found values between 4.6 and9.2 μmol O2min−1 at 23 °C for eels (mean body mass ranging from0.312 to 0.338 kg) fed different diets. When taking into account the dif-ferences in body mass and temperature between studies, these valuesfor SMR corresponds well with those from the present study. The factthat eels do have such a low SMR is also evident from its record fornon-hibernating starvation endurance (1594 days) (McCue, 2010).This may also explain why we still find a relationship between rSMRand SGR even though the fish was fed only moderately.

In the present study, it is clear that the scaling exponents (b) for SMR(Tables 1 and 2) are higher than the traditionally expected 0.75 scalingexponent. Although the ‘3/4 scaling rule’was long accepted as more or

less universal (Schmidt-Nielsen, 1984; Glazier, 2005), recent evidencesuggest that this may not always be the case and that scaling of meta-bolic rate with body mass varies widely between species and, amongstothers, depends on lifestyle and temperature (White et al., 2006; Killenet al., 2010). For example, Degani and Lee-Gallagher (1985) found expo-nents of 0.52 to 0.57 in the American eel (Anguilla rostrata), and Deganiet al. (1989) found scaling exponents of 0.67 to 1.29 in the Europeaneel. According to Clarke and Johnston (1999), scaling exponents rangefrom 0.40 to 1.29 over a broad range of teleost species and the mean ex-ponent are higher than the ‘traditional’ 0.75 scaling exponent. The reasonfor the observed scaling exponents can be described by the metabolic-level boundary hypothesis (Glazier, 2005; Killen et al., 2010). Accordingto this hypothesis, scaling exponents should be inversely related withmetabolic intensity. This is in accordance with the interspecific findingsby Clarke and Johnston (1999) in teleost fish, as well as in the presentstudy. Fish with relatively low metabolic rate (as observed in theAnguilliformes) will have scaling exponents close to 1 and those fishthat have highmetabolic rates (e.g., Gadiformes)will have scaling expo-nents closer to 2/3. Likewise, the hypothesis suggests that a high scalingexponent is to be expectedwhenmetabolism is reduced because of lowfood availability (Glazier, 2005).

4.2. Organ correlations

The few studies that have explored the relationship between met-abolic rate and organ mass in fish have not been conclusive. Odellet al. (2003) found positive correlations between maximummetabol-ic rate and the mass of the heart, gill and swimming motor in Trinida-dian guppies (Poecilia reticulata) whereas Norin and Malte (2012)found no relationship between metabolic rate and the mass of fiveorgans in brown trout (Salmo trutta). A different line of studies onfish have investigated the ontogenetic changes in organ mass andmetabolic rate as a consequence for the mass-specific decrease inwhole-animal metabolic rate (Oikawa and Itazawa, 1984; Oikawaet al., 1992; Oikawa and Itazawa, 1993, 2003). In these studies, therelative decrease in organ mass or organ metabolic rate with increas-ing body mass was measured in one group of fish and indirectlylinked to total oxygen consumption measured in another group offish of the same species. Thus, these studies can only confirm that adecrease in mass-specific rate of metabolism with increasing totalbody mass can be explained by a combination of a decrease in therate of tissue respiration and an increase in the relative size of tissueswith low metabolism.

In the present study, the only individual organ that had a signifi-cant effect on rSMR, regardless of being based on wet or dry weight,was the liver. Since the liver made up most of the organ mass inves-tigated it will contribute to a major part of the variation in thesummed organ mass. This makes it more likely to find a correlationbetween body-mass-corrected summed organ mass (rSum organ)and rSMR, as was the case in the present study. Since the organsonly constituted ca. 1.6% of the total body mass, this may partly ex-plain why we did not find any effect of the other organs on rSMR.The liver likely explains much of the variation in rSMR because ofits high metabolic activity and mass compared to the other organs(i.e., the liver contributes disproportionately more to the total energyexpenditure).

The fact that residual organ mass (in particular the liver) can ex-plain part of the individual variation in body mass corrected SMR ineels is in agreement with the comprehensive literature on inter-and intraspecific differences between organ mass and basal metabolicrate (BMR) in mammals and birds (Daan et al., 1990; Konarzewskiand Diamond, 1995; Meerlo et al., 1997; Burness et al., 1998;Speakman and Johnson, 2000; Speakman et al., 2004; Song andWang, 2006; Russell and Chappell, 2007). Thus, it is indicated thatthere may be no particular differences in ectotherms and endothermsin this respect.

28 M.M. Boldsen et al. / Comparative Biochemistry and Physiology, Part A 165 (2013) 22–29

There is a tendency that individuals with a liver that is larger thanexpected from their total body mass, correspondingly will have alarger intestine. This correlation may be due to the close embryonicorigin of these two organs since the liver is formed from endodermalgut epithelium. The liver is connected with the intestine through thebile duct where the liver secretes bile to the gall bladder. In addition,the two organs are connected through the hepatic portal systemthrough which the liver takes up nutrients from the intestine. Workby Millidine et al. (2009) showed that juvenile Atlantic salmon witha high SMR had higher peak specific dynamic action and a greatertotal energy expenditure when digesting a meal. They also observedthat the duration over which the fish's metabolism was elevatedafter consuming a meal was shorter. Millidine et al. (2009) thereforehypothesised that the greater energy cost the fish experienced afterprocessing a meal could be related to their assimilation efficiency.This might also explain the findings in the present study where indi-vidual eels with larger liver and intestine than conspecifics wouldlikely have increased food assimilation efficiency, resulting in a highergrowth rate but in turn increasing the basic cost of living (SMR).

4.3. Enzymes

The fact that there was no connection between enzyme activityand rSMR is likely because of the low replicate number, resulting ingreater variance. Norin and Malte (2012) found a positive relation-ship between both liver citrate synthase activity and cytochrome ox-idase activity and rSMR in brown trout, suggesting that the activity ofthese two important metabolic enzymes can in fact explain some ofthe variation in metabolic rate. In addition, enzyme activity in otherorgans with high metabolic rate, such as the brain and heart, couldalso contribute to total oxygen consumption, masking the contribu-tion of enzyme activity in the liver.

Schär et al. (1985) showed through histochemical studies thatrainbow trout (Oncorhynchus mykiss) express metabolic zonation inthe liver as previously observed in mammals. However, the functionalheterotopy in trout was found to be less pronounced than in mam-mals. Likewise, Mommsen (1991) showed metabolic and enzymaticheterogeneities in the liver of the teleost Opsanus beta. Since liversamples in the present study were taken from random parts of theliver, this may contribute to the variation in the enzyme assay, partlyexplaining our results.

The individual eel with the highest total liver protein contentamongst the sample of eels was almost twice as high as the lowest,and on average protein content made up approximately one tenthof the total liver mass. The latter corresponds to ca. half as much asin brown trout (Norin and Malte, 2012). Walsh et al. (1983) foundtotal protein content per mass unit in the liver of A. rostrata to beca. six times higher than in our study. Since body mass of eels intheir study was only 47–65 g, differences in observed protein contentcould be due to scaling effects. Comparing enzyme activity betweenstudies should be done with caution due to differences in protocoland tissues investigated. However, one study to date has examinedCytOx and CS activities in brown trout liver in the same way as thepresent study (Norin and Malte, 2012). The authors found a meanCytOx activity that was 2.2 times higher and a mean CS activity thatwas 0.3 times the findings of this study. The fact that total proteincontent and CytOx activity is higher in brown trout liver than in eelcould simply be that brown trout have a much larger aerobic meta-bolic capacity than eels. This correlates with the findings that CytOxactivity serves as a rough estimate of mitochondrial density (Simonand Robin, 1971; Christensen et al., 1994). Further experiments arerequired to highlight the contribution of enzyme activity in differenttissues on intraspecific variation in metabolic rate.

In summary, European eels display intraspecific variation in SMR, aswell as RMR, that is consistent over a two-month period. Repeatabilityof RMR shows that spontaneous activity in respirometer-confined fish

is not random and the observed inter-individual variation in SMR canpartially be explained by the mass of the liver.

Acknowledgements

The authors wish to thank Hans Gesser and Angela Fago for theirkind advice on enzyme activity and protein determination.

References

Baras, E., Jeandrain, D., 1998. Evaluation of surgery procedures for tagging eel Anguillaanguilla (L.) with biotelemetry transmitters. Hydrobiologia 371/372, 107–111.

Birkedal, R., Gesser, H., 2006. Intracellular compartmentation of cardiac fibres fromrainbow trout and Atlantic cod: a general design of heart cells. Biochim. Biophys.Acta 1757, 764–772.

Bolger, T., Connelly, P.L., 1989. The selection of suitable indices for the measurementand analysis of fish condition. J. Fish Biol. 34, 171–182.

Brett, J.R., Gloves, T.D.D., 1979. Physiological energetics. Fish Physiol. 8, 279–352.Broggi, J., Hohtola, E., Koivula, K., Orell, M., Nilsson, J.-Å., 2009. Long-term repeatability

of winter basal metabolic rate and mass in a wild passerine. Funct. Ecol. 23,768–778.

Burness, G., Ydenberg, R.C., Hochachka, P.W., 1998. Interindividual variability in bodycomposition and resting oxygen consumption in breeding tree swallows,Tachycineta bicolor. Physiol. Zool. 71, 247–256.

Burton, T., Killen, S.S., Armstrong, J.D., Metcalfe, N.B., 2011. What causes intraspecificvariation in resting metabolic rate and what are its ecological consequences? Proc.R. Soc. B 278, 3465–3473.

Chappell, M.A., Bachman, G.C., Odell, J.P., 1995. Repeatability of maximal aerobic perfor-mance in Belding ground squirrels Spermophilus beldingi. Funct. Ecol. 9, 498–504.

Chappell, M.A., Zuk, M., Johnsen, T.S., 1996. Repeatability of aerobic performance in redjunglefowl: effects of ontogeny and nematode infection. Funct. Ecol. 10, 578–585.

Childress, J.J., Somero, G.N., 1979. Depth-related enzymic activities muscle, brain andheart of deep-living pelagic marine teleosts. Mar. Biol. 52, 273–283.

Christensen, M., Hartmund, T., Gesser, H., 1994. Creatine kinase, energy-rich phosphateand energy metabolism in heart muscle of different vertebrates. J. Comp. Physiol.164 B, 118–123.

Clarke, A., Johnston, N.M., 1999. Scaling metabolic rate with body mass and temperaturein teleost fish. J. Anim. Ecol. 68, 893–905.

Cutts, C.J., Metcalfe, N.B., Taylor, A.C., 1998. Aggression and growth depression in juvenileAtlantic salmon: the consequence of individual variation in standard metabolic rate.J. Fish Biol. 52, 1026–1037.

Cutts, C.J., Adams, C.E., Campbell, A., 2001. Stability of physiological and behaviouraldeterminants of performance in Arctic char (Salvelinus alpinus). Can. J. Fish.Aquat. Sci. 58, 73–78.

Daan, S., Masman, D., Groenewold, A., 1990. Avian basal metabolic rate: their associationwith body composition and energy expenditure in nature. Am. J. Physiol. 259,R333–R340.

Degani, G., Lee-Gallagher, M., 1985. The relationship between growth, food conversionand oxygen consumption in developed and underdeveloped American eels, Anguillarostrata Lesueur. J. Fish Biol. 27, 635–641.

Degani, G., Gallagher, M.L., Meltzer, A., 1989. The influence of body size and tempera-ture on oxygen consumption of the European eel, Anguilla anguilla. J. Fish Biol.34, 19–24.

Emmett, B., Hochachka, P.W., 1981. Scaling of oxidative and glycolytic enzymes inmammals. Respir. Physiol. 45, 261–272.

Fu, S.-J., Cao, Z.-D., Peng, J.-L., 2007. The stability of standard metabolic rate and its re-lationship to growth performance variation in southern catfish, Silarus meridionalisChen. J. Fish. Aquat. Sci. 2, 162–167.

Garland Jr., T., 1984. Physiological correlates of locomotory performance in a lizard: anallometric approach. Am. J. Physiol. 247, R806–R815.

Glazier, D.S., 2005. Beyond the “3/4” power law: variation in the intra- and interspecificscaling of metabolic rate in animals. Biol. Rev. 80, 611–662.

Hulbert, A.J., Else, P.L., 2000. Mechanisms underlying the cost of living in animals.Annu. Rev. Physiol. 62, 207–235.

Iversen, N.K., McKenzie, D.J., Malte, H., Wang, T., 2010. Reflex bradycardia does notinfluence oxygen consumption during hypoxia in European eel (Anguilla anguilla).J. Comp. Physiol. B 180, 495–502.

Killen, S.S., Atkinson, D., Glazier, D.S., 2010. The intraspecific scaling of metabolic ratewith body mass in fishes depends on lifestyle and temperature. Ecol. Lett. 13,184–193.

Konarzewski, M., Diamond, J., 1995. Evolution of basal metabolic rate and organ massesin laboratory mice. Evolution 49, 1239–1248.

Konarzewski, M., Książek, A., Łapo, I.B., 2005. Artificial selection on metabolic rates andrelated traits in rodents. Integr. Comp. Biol. 45, 416–425.

Krebs, H.A., 1950. Body size and tissue respiration. Biochim. Biophys. Acta 4, 249–269.Książek, A., Konarzewski, M., Łopo, I.B., 2004. Anatomic and energetic correlates of di-

vergent selection for basal metabolic rate in laboratory mice. Physiol. Biochem.Zool. 77, 890–899.

Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement withthe folin phenol reagent. J. Biol. Chem. 193, 265–275.

Maciak, S., Konarzewski, M., 2010. Repeatability of standard metabolic rate (SMR) in asmall fish, the spined loach (Cobitis taenia). Comp. Biochem. Physiol. A 157, 136–141.

29M.M. Boldsen et al. / Comparative Biochemistry and Physiology, Part A 165 (2013) 22–29

McCarthy, I.D., 2000. Temporal repeatability of relative standard metabolic rate in juve-nile Atlantic salmon and its relation to life history variation. J. Fish Biol. 57,224–238.

McCue, M.D., 2010. Starvation physiology: reviewing the different strategies animalsuse to survive a common challenge. Comp. Biochem. Physiol. A 156, 1–18.

McKenzie, D.J., Piraccini, G., Piccolella, M., Steffensen, J.F., Bolis, C.L., Taylor, E.W., 2000.Effects of dietary fatty acid composition onmetabolic rate and responses to hypoxia inthe European eel (Anguilla anguilla). Fish Physiol. Biochem. 22, 281–296.

Meerlo, P., Bolle, L., Visser, G.H., Masman, D., Daan, S., 1997. Basal metabolic rate in re-lation to body composition and daily energy expenditure in the field vole Microtusagrestis. Physiol. Zool. 70, 362–369.

Metcalfe, N.B., Wright, A.C., Thorpe, J.E., 1992. Relationship between social status, otolithsize at first feeding and subsequent growth in Atlantic salmon (Salmo salar). J. Anim.Ecol. 61, 585–589.

Metcalfe, N.B., Taylor, A.C., Thorpe, J.E., 1995. Metabolic rate, social status and life-history strategies in Atlantic salmon. Anim. Behav. 49, 431–436.

Millidine, K.J., Armstrong, J.D., Metcalfe, N.B., 2009. Juvenile salmon with high standardmetabolic rates have higher energy costs but can process meals faster. Proc. R. Soc.B 276, 2103–2108.

Mommsen, T.P., 1991. Metabolic and enzymatic heterogeneity in the liver of theureogenic teleost Opsanus beta. J. Exp. Biol. 156, 407–418.

Nespolo, R.F., Franco, M., 2007. Whole-animal metabolic rate is a repeatable trait: ameta-analysis. J. Exp. Biol. 210, 2000–2005.

Norin, T., Malte, H., 2011. Repeatability of standard metabolic rate, active metabolic rateand aerobic scope in young brown trout during a period of moderate food availability.J. Exp. Biol. 214, 1668–1675.

Norin, T., Malte, H., 2012. Intraspecific variation in aerobic metabolic rate of fish: relationswith organ size and enzyme activity in brown trout. Physiol. Biochem. Zool. 85 (6),645–656.

O'Connor, K.I., Taylor, A.C., Metcalfe, N.B., 2000. The stability of standard metabolic rateduring a period of food deprivation in juvenile Atlantic salmon. J. Fish Biol. 57,41–51.

Odell, J.P., Chappell, M.A., Dickson, K.A., 2003. Morphological and enzymatic correlatesof aerobic and burst performance in different populations of Trinidadian guppies(Poecilia reticulata). J. Exp. Biol. 206, 3707–3718.

Oikawa, S., Itazawa, Y., 1984. Allometric relationship between tissue respiration andbody mass in the carp. Comp. Biochem. Physiol. A 77, 415–418.

Oikawa, S., Itazawa, Y., 1993. Tissue respiration and relative growth of parts of body ofa marine teleost, porgy, Pagrus major, during early life stages with special referenceto the metabolism-size relationship. Comp. Biochem. Physiol. A 105, 741–744.

Oikawa, S., Itazawa, Y., 2003. Relationship between summated tissue respiration andbody size in a marine teleost the porgy, Pagrus major. Fish. Sci. 69, 687–694.

Oikawa, S., Takemori, M., Itazawa, Y., 1992. Relative growth of organs and parts of amarine teleost, the porgy, Pagrus major, with special reference to metabolism–

size relationships. Jpn. J. Ichthyol. 39, 243–249.Pankhurst, N.W., 1982. Relation of visual changes to the onset of sexual maturation in

the European eel Anguilla anguilla (L.). J. Fish Biol. 21, 127–140.

Pelstra, A., van Ginneken, V., van den Thillart, G., 2008. Cost of transport and optimalswimming speed in farmed and wild European silver eels (Anguilla anguilla).Comp. Biochem. Physiol. A 151, 37–44.

Priede, I.G., 1985. Metabolic scope in fishes. In: Tytler, P., Carlow, P. (Eds.), Fish Energetics:New Perspectives. Croom Helm Ltd., London, pp. 33–44.

Reidy, S.P., Kerr, S.R., Nelson, J.A., 2000. Aerobic and anaerobic swimming performanceof individual Atlantic cod. J. Exp. Biol. 203, 347–357.

Rolfe, D.F.S., Brown, G.C., 1997. Cellular energy utilization and molecular origin of standardmetabolic rate in mammals. Physiol. Rev. 77, 731–758.

Russell, G.A., Chappell, M.A., 2007. Is BMR repeatable in deer mice? Organ mass corre-lates and the effect of cold acclimation and natal altitude. J. Comp. Physiol. B 177,75–87.

Schär, M., Maly, I.P., Sasse, D., 1985. Histochemical studies on metabolic zonation of theliver in the trout (Salmo gairdneri). Histochemistry 83, 147–151.

Schmidt-Nielsen, K., 1984. Scaling: Why Is Animal Size So Important? Cambridge Uni-versity Press, New York.

Seppänen, E., Piironen, J., Huuskonen, H., 2010. Consistency of standard metabolic ratein relation to life history strategy of juvenile Atlantic salmon Salmo salar. Comp.Biochem. Physiol. A 156, 278–284.

Simon, L.M., Robin, E.D., 1971. Relationship of cytochrome oxidase activity to vertebratetotal and organ oxygen consumption. Int. J. Biochem. 2, 569–573.

Somero, G.N., Childress, J.J., 1990. Scaling of ATP-supplying enzymes, myofibrillar pro-teins and buffering capacity in fish muscle: relationship to locomotory habit. J. Exp.Biol. 149, 319–333.

Song, Z.-G., Wang, D.-H., 2006. Basal metabolic rate and body size in Brandt's voles(Lasiopdomys brandtii): effects of photoperiod, temperature and diet quality. Physiol.Behav. 89, 704–710.

Speakman, J.R., Johnson, M.S., 2000. Relationship between resting metabolic rate andmorphology in lactating mice: what tissues are the major contributors to restingmetabolism? In: Heldmaier, G., Klingenspor, M. (Eds.), Life in the Cold. Springer-Verlag, Berlin, pp. 479–486.

Speakman, J.R., Król, E., Johnson, M.S., 2004. The functional significance of individualvariation in basal metabolic rate. Physiol. Biochem. Zool. 77, 900–915.

Steffensen, J.F., 1989. Some errors in respirometry of aquatic breathers: how to avoidand correct for them. Fish Physiol. Biochem. 6, 49–59.

Tripathi, G., Verma, P., 2004. Scaling effects on metabolism of a teleost. J. Exp. Zool.301A, 718–726.

Virani, N.A., Rees, B.B., 2000. Oxygen consumption, blood lactate and inter-individualvariation in the gulf killifish Fundulus grandis, during hypoxia and recovery.Comp. Biochem. Physiol. A 126, 397–405.

Walsh, P.J., Foster, G.D., Moon, T.W., 1983. The effect of temperature on metabolism ofthe American eel Anguilla rostrata (Lesueur): compensation in the summer andtorpor in the winter. Physiol. Zool. 56, 523–540.

White, C.R., Phillips, N.F., Seymour, R.S., 2006. The scaling and temperature dependenceof vertebrate metabolism. Biol. Lett. 2, 125–127.

Wiegand, G., Remington, S.J., 1986. Citrate synthase: structure, control and mechanism.Annu. Rev. Biophys. Biophys. Chem. 15, 97–117.