Utilisation of glycogen, ATP and arginine phosphate in exercise and recovery in terrestrial red crabs, Gecarcoidea natalis

Comparative Biochemistry and Physiology Part A 133(2002) 813–825

1095-6433/02/$ - see front matter� 2002 Elsevier Science Inc. All rights reserved.PII: S1095-6433Ž02.00217-9

Utilisation of glycogen, ATP and arginine phosphate in exercise andrecovery in terrestrial red crabs,Gecarcoidea natalis�

Stephen Morris *, Agnieszka M. Adamczewskaa, b

aMorlab, School of Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG, UKbInterRidge, Ocean Research Institute, University of Tokyo, 1-15-1 Minamidai, Nakano-ku, Tokyo 164-8639, Japan

Received 25 March 2002; received in revised form 17 July 2002; accepted 20 July 2002

Abstract

Intermittent locomotion by terrestrial crustaceans may under specific circumstances increase walking distance and mayallow partial re-oxidization of anaerobic products, and replenishment of ATP and arginine phosphate. The ChristmasIsland red crabG. natalis undertakes a substantial breeding migration each year. The leg muscles ofG. natalis subjectedto bouts of 2.5 min walking and 2.5 min rest were severally anaerobic. Adenylate energy charge and the large argininephosphate stores were greatly reduced. Walking for 4 min with pauses of only 1 min exacerbated the anaerobiosis andutilised 50% of the endogenous muscle glycogen. Post-exercise, the adenylate energy charge recovered before thearginine phosphate charge and a large and persistent hyperglycaemia accompanied the restoration of glycogen. Argininephosphate functioned as a large, longer term, energy reservoir-almost as part of the adenylate pool. Gluconeogenesis isyet to be generally substantiated in decapod crustaceans butG. natalis appears to remove lactate slowly and toreincorporate exogenous glucose into muscle glycogen in the same time frame as lactate removal from the haemolymph.The 4:1 exerciseypause regimen facilitated access to energy stores and increased walking distance, and it allowedL-lactate and H efflux from the muscle during pausing. These responses are similar to those ofG. natalis in the field,q

except during the migration when walking was entirely aerobic. Determinations of adenylate, fuel and arginine phosphatereserves and usage during the migration are required together with more detailed behavioral analysis to resolve thedichotomy in metabolic response.� 2002 Elsevier Science Inc. All rights reserved.

Keywords: Crab; Exercise; ATP; Arginine phosphate; Lactate; Glycogen; Walking; Gluconeogenesis; Exercise; Glucose

1. Introduction

The metabolic responses of air-breathing crabsto exercise can differ markedly in respect of both

� This paper was originally presented at ‘Chobe 2001’; TheSecond International Conference of Comparative Physiologyand Biochemistry in Africa, Chobe National Park, Botswana– August 18–24, 2001. Hosted by the Chobe Safari Lodgeand the Mowana Safari Lodge, Kasane; and organised byNatural Events Congress Organizing([email protected]).*Corresponding author. Tel.:q44-117-928-9181; fax:q44-

117-925-7374.E-mail address: [email protected](S. Morris).

the maximum velocity that they can sustainedaerobically and in their endurance capacity(e.g.Weinstein and Full, 1992). Foraging crabs, forexample, the ghost crabsOcypode exhibit largeand rapid elevations in aerobic metabolismmatched to high endurance at low speed while therelated fiddler crabUca fatigues more quickly andis unable to elevate oxygen uptake in the sameway (Herreid and Full, 1988). Terrestrial crabsgenerally respond to exercise with a mixture ofelevated aerobic and supplemental anaerobicmetabolism (Herreid and Full, 1988; Full andWeinstein, 1992). Thus, while maximum aerobic

814 S. Morris, A.M. Adamczewska / Comparative Biochemistry and Physiology Part A 133 (2002) 813–825

scope may vary between species and therebyrestrict intensity of immediate exercise there isalso a point in the speed-endurance ‘trade-off’where continued locomotion will be limited by theextent to which glycolysis can be accelerated andby the accumulation of anaerobic end products(Full and Weinstein, 1992; Adamczewska andMorris, 2000a; Weinstein 2001 for reviews).The Christmas Island red crab,Gecarcoidea

natalis, undertakes a startling annual breedingmigration during which crabs may walk over 1km day for several successive days(Adam-y1

czewska and Morris, 2001a). There is no evidence,at the end of each day, that anaerobiosis wasimportant in red crabs completing each day ofmigratory walking although it is important insupporting other activities associated with breeding(Adamczewska and Morris, 2001b). However,G.natalis has relatively low aerobic scope and outsideof the migration is unable to sustain even moderateexercise without recourse to supplemental anaero-biosis (Adamczewska and Morris, 1994, 2000b).It is curious therefore, how this crab is able toaccomplish the annual migration without employ-ing anaerobiosis.A broad spectrum of locomotor strategies are

potentially available to migrating red crabs. Themean walking speed of migrating red crabs is closeto their maximum aerobic speed(MAS), but theyfrequently pause for minutes or rarely for hoursbefore engaging in short periods of locomotion atspeeds significantly in excess of the MAS(Adam-czewska and Morris, 2001a). Intermittent loco-motion is increasingly recognised as havingsignificant energetic and ecological consequences(Kramer and McLaughlin, 2001; Irschick and Gar-land, 2001; for reviews) and just as much so forterrestrial crustaceans as for vertebrates(Full andWeinstein, 1992; Weinstein 2001 for reviews).Intermittent walking can increase overall cost ofhigh speed locomotion(e.g. Weinstein and Full,1998) but may increase the distance that can beachieved, depending on the temperature, the dura-tion and intensity of exercise and the length of therecovery pauses(e.g. Full and Weinstein, 1992;Weinstein and Full, 1998; Weinstein, 2001; Glee-son and Hancock, 2001). As exercise intensityincreases a relatively longer recovery period isneeded, depending on temperature(Weinstein,2001). For example, ghost crabs exercised inter-mittently at 248C were able to walk further thancontinuously exercised crabs if the exercise bouts

and pauses were both of 2 min(Full and Wein-stein, 1992) but at 158C this decreased capacity(Weinstein and Full, 1998). Intermittent exercisecan reduce workload in ghost crabs and allow atleast some reduction ofL-lactate during the pauses(Weinstein and Full, 1992).Herreid and Full(1988) correctly lamented that

there were too few data as to the role of, andchanges in, high energy phosphates in terrestrialcrabs (e.g. Beis and Newsholme, 1975). Thiscircumstance has understandably required compar-ative studies to draw heavily on findings andconclusions from investigations of vertebrates(e.g.Weinstein and Full, 1992; Weinstein, 2001). How-ever, some simple principles emerge.Supplemental anaerobiosis during exercise

requires significantly elevated glycolysis and nec-essarily high glycogen reserves to support it(e.g.England and Baldwin, 1983; Henry et al., 1994).The restoration of high-energy phosphates anddepleted O stores generally occurs more rapidly2

than does the removal ofL-lactate and clearing theassociated acid load; and thus intermittent exerciseresulting inL-lactate production will immediatelyrequire longer recovery. Weinstein and Full(1992)pointed out that the dynamics ofL-lactate seemdifferent in crustaceans compared to the mamma-lian model, primarily in the low rate of removal(see also Henry et al., 1994) and in the progressiveincrease even in ghost crabs provided with opti-mum pause duration. There remains controversyas to the site(s) and mechanism(s) of L-lactatemetabolism in crustaceans; whether lactate mightbe re-incorporated into glycogen or more simplyoxidised (e.g. Ellington, 1983; van Aardt, 1988;Hill et al., 1991a,b; Lallier and Walsh, 1992;Oliveira and da Silva, 1997, 2000; Hervant et al.,1999b), or indeed if some is not excreted in aquaticspecies(e.g. Head and Baldwin, 1986; Hervant, etal., 1999a). The possibility that crustaceans mayremoveL-lactate directly by endogenous glyconeo-genesis within the muscles(Hervant et al., 1999b),similarly to vertebrate skeletal muscle(Schulte etal., 1992; Gladden, 2000; Donovan and Paglias-sotti, 2000) could be of considerable potentialimportance to migratory land crabs.The ATP, AMP and ADP concentrations in

decapod crustaceans are not unusually high butwhere arginine phosphate concentrations have beendetermined in decapod muscle they are oftenhigher than in insects and higher than creatinephosphate in vertebrates(Beis and Newsholme,

815S. Morris, A.M. Adamczewska / Comparative Biochemistry and Physiology Part A 133 (2002) 813–825

1975; England and Baldwin, 1983; Hill et al.,1991b; Speed et al., 2001). Arginine phosphate isvery definitely used by decapod crustaceans toprovide energy during intense muscle activity(England and Baldwin, 1983) and during hypoxia(e.g. Hill et al., 1991b; Speed et al., 2001).Arginine phosphate may be important during inter-mittent locomotion(Weinstein and Full, 1992),and under some conditions pauses may permitghost crabs to recover some arginine phosphatebut not ATP(Weinstein and Full, 1998).Some respiratory indices have been investigated

in G. natalis to assess potential strategic benefitsof intermittent locomotion that may assist thebreeding migration of these crabs(Adamczewskaand Morris, 1998). In this previous study, the redcrabs were required to walk either continuously at110% MAS or intermittently with equal durationpauses and exercise bouts at 230% MAS. Inter-mittent locomotion depleted O stores, elevated2

PCO , promoted haemolymph acidosis, and ele-2

vated free glucose andL-lactate to a greater extentthan did continuous exercise. Interestingly thedepletion of venous O increased transport and2

lactate removal preceded more promptly afterintermittent locomotion than after continuouswalking (Adamczewska and Morris, 1998). Thus,pauses allow some slow lactate removal but mightafford a real opportunity to replenish high-energyadenylates and more importantly arginine phos-phate. The present paper reports investigation ofthe role and changes in adenylates, arginine phos-phate, glycogen, glucose and lactate concentrationsin G. natalis subjected to two different exerciseregimens, and especially during recovery fromexercise.

2. Materials and methods

2.1. Animal collection

Terrestrial red crabs,Gecarcoidea natalis(224.4"7.4 g; ns75) were collected during thewet season from the rainforest of Christmas Islandsubsequent to the annual migration under Permitfrom Parks Australia and air-freighted to the lab-oratory under AQIS permit. All crabs used had acomplete and mature complement of limbs andwere clearly intermoult adult females. The crabswere housed in communal terraria at 258C and)85% relative humidity under a 12 h:12 h lightydark regime for at least 8 weeks prior to experi-

mentation. The crabs were fed weekly on a diet ofdry leaves, green vegetables and dry cat biscuit,with freshwater to drink. The day previous toexperimentation the crabs were isolated in individ-ual containers without food but with water.

2.2. Exercise protocol

An independent design was used whereby eachcrab participated in the study in only one of theseven treatments used. Seven different treatmentswere employed;(i) crabs at rest,(ii) crabs thatwere required to walk and rest for intervals of 2.5min each with a total time walking of 10 min(2.5:2.5 regime), (iii ) crabs that were subjected tomore strenuous exercise of 4 min walking and 1min resting with a total time walking of 16 min(4:1 regime), and (iv–vii) crabs recovered fromthe 4:1 exerciseypause regime for 5 min, 30 min,5 h or 24 h. The exercise regimens were construct-ed so that the speed of locomotion integrated overthe entire duration of the experimental exerciseperiod (including pauses-see results) did notexceed the mean speed of just over 1 m miny1

achieved during their migration(Adamczewskaand Morris, 2001a). Eight crabs were employed ineach treatment, with a total sample size of 63 redcrabs. The crabs were exercised in a runway andvoluntarily walked from the illuminated endtowards the shaded end of the runway. Towardsthe end of the exercise regimens the crabs some-times attempted to rest before the allotted walkingperiod had elapsed and were encouraged to com-plete the session by gentle tactile stimulation. Atthe completion of the exerciseyrecovery regimenthe crabs were sampled for haemolymph and forleg muscle. The haemolymph and leg muscle wereassayed forL-lactate and glucose; and the legmuscle additionally for glycogen, ATP, ADP, AMP,arginine phosphate and arginine.

2.3. Sampling and assay protocols

Haemolymph was withdrawn within 30 s ofcompleting the exercise recovery time and analiquot of the sample was mixed(ratio 1:1) with0.6 mol l HClO to denature proteins andy1

4

neutralised with 2.5 mol l K CO . The denaturedy12 3

sample was centrifuged at 10000=g for 10 minand the supernatant was used forL-lactate analysis(Boehringer Mannheim test kit No.138 084).Whole haemolymph samples were frozen and


stored for later analysis for glucose(Sigma Diag-nostics test-kit No. 510).At the appropriate time, each animal was

encouraged to autotomise the penultimate perio-pod, which was wrapped in aluminium foil, andclamp frozen in liquid nitrogen. The time fromremoval of the leg to freezing in liquid nitrogenaveraged less than 45 s. The frozen flesh wasremoved to a ceramic mortar containing liquidnitrogen and ground to a fine powder. A knownmass was then transferred to a glass homogeniser.The tissue was homogenised with two parts 0.9mol l HClO and kept on ice during the processy1

4

of homogenisation. The homogenate was thentransferred into Eppendorf microcentrifuge tubesand one part of 0.9 mol l HClO was added toy1

4

the homogeniser and any remaining tissue homo-genised and poured into Eppendorf tubes. Thehomogenate was centrifuged for 5 min at 6000=gat 4 8C. The sediment was then extracted with 0.2mol l HClO perchloric acid using a third of they1

4

volume of the perchloric acid used for the firstextraction. To neutralise the effects of the acid onepart K CO , 3.75 mol l for five parts of HClOy1

2 3 4

was added with rapid stirring. The tubes were thencentrifuged for 5 min at 6000=g at 4 8C and theresulting supernatant removed and used for thefollowing assays.Muscle glycogen was measured as glucose sub-

sequent to treatment with glycoamylase, and glu-cose before and after this treatment was measuredaccording to the techniques described by Bergmey-er (1985a). Muscle lactate levels were determinedusing a spectrophotometric method with the testkit (Boehringer Mannheim, lactic acid kit� 139084). Arginine and arginine phosphate were deter-mined using the methods in Bergmeyer(1985c).The arginine was determined by the change inabsorbance at 339 nm in the octopine dehyrogen-ase catalysed reaction;

arginineqpyruvateqNADHqH l octopineq

qNAD qH O.q2

To hydrolyse arginine phosphate to arginine andphosphate, 100ml of 1 mol l HCl was added toy1

100 ml of tissue homogenate and incubated intightly capped tubes for 90 s in boiling water. Thehydrolysate was then cooled and neutralised with100 ml of 1 mol l NaOH. The arginine assayy1

was repeated and the previous arginine concentra-tion subtracted to obtain arginine phosphate. The

ATP content of the leg muscle was determinedusing the Sigma ATP test kit(catalogue� 366UV). Each sample was measured in duplicate. TheADP and AMP were determined using an NADHlinked assay as described by Bergmeyer(1985b)and used previously(Adamczewska and Morris,1994; Morris and Oliver, 1999). The adenylateenergy charge was calculated as:

1w x w xADP q ATP )

2w x w x w xAMP q ADP q ATP

and the arginine charge according to Schulte et al.(1992) as:

w xPArgw x w xPArgq Arg

2.4. Statistical analyses

Homogeneity of variance within the data setswas ensured prior to ANOVA using Cochran’s C.Data were compared using a 1 or 2 factor analysisof variance(ANOVA). Post hoc testing was per-formed using either a contrast(A matrix) analysisor Tukeys HSD multiple means comparison anal-ysis where appropriate. Significance levels weretaken at the probability level ofPs0.05 and alldata are presented as mean"S.E.M. unless other-wise stated.

3. Results

The total mean distance travelled by theG.natalis subject to the 2.5 min exercise-2.5 min restregime(2.5:2.5 group) was 10.54"0.45 m; where-as the mean distance covered by all the crabssubject to 4 min exercise with 1 min rest periods(4:1 group) was significantly greater at15.58"0.13 m. However, these distances wereachieved by similar walking speeds of 1.05"0.05and 0.97"0.04 m min . Thus, when the resty1

periods are factored in, the integrated distance overtotal time was covered at 0.60"0.03 m min andy1

0.82"0.03 m min . ANOVA of these rates pro-y1

vided aPs0.052 and thus while it appeared thatthe 4:1 group covered the distance 25% faster thiswas not statistically significant. Thus, the maindiscriminating factors between the groups were theduration of the rest periods and the distance cov-ered, since the 2.5:2.5 group clearly did not walkfaster as a result of longer pauses.


Fig. 1. The concentration of(a) L-lactate,(b) glucose and(c)glycogen(as glycolsyl units) in the leg muscle asmmolØgy1

(solid bars) and in the haemolymph asmmolØml (hatchedy1

bars) of Gecarcoidea natalis at rest, after exercise and duringrecovery. The crabs were intermittently exercised in one of tworegimens; either walking for 2.5 with pauses of 2.5 min orwalking for 4 min with pauses of 1 min. The data for the2.5:2.5 min group are shown by the lighter bars. The recoverydata are all for crabs exercised in the 4:1 min regimen.ns8for each group. * indicates a difference from the resting crabs,� indicates a difference between the 2.5:2.5 and 4:1 treatments.The ‘g’ in panel C indicates differences within the glycogendata.

The L-lactate concentration of restingG. nataliswas low (leg muscle 1.96mmolØg and haemo-y1

lymph 0.42mmolØml ) but increased significant-y1

ly in both the 2.5:2.5 and the 4:1 exercised crabs(Fig. 1). The increase in haemolymphL-lactateconcentration was significantly greater in the 4:1group than in the 2.5:2.5 group(14.72"1.21compared to 6.77"1.26 mmolØml ). Duringy1

recovery from the 4:1 exerciseypause regime theconcentration ofL-lactate continued to rise; reach-ing 28.45"1.89 mmol ml in the haemolymphy1

and 35.43"1.76mmol ml in the muscle after 5y1

min of recovery and remained elevated for at leastthe succeeding 25 min(Fig. 1). A completerecovery of resting L-lactate concentrationappeared to require more than 5 h. The restingconcentrations of glucose were also low at0.05"0.02 mmol g in the muscle andy1

0.37"0.08 mmolØml in the haemolymph(Fig.y1

1b). The increased glycolytic flux needed to sup-port supplemental anaerobiosis andL-lactate pro-duction was reflected in marked increases in freeglucose of the muscle in both the 2.5:2.5 and 4:1exercised groups(Fig. 1b). The tissue concentra-tion of free glucose continued to increase for atleast 30 min post 4:1 exercise, with the result thathaemolymph glucose eventually increased 30 minpost-exercise to 1.69"0.18 mmolØml . Impor-y1

tantly, after 5 h post-exercise tissue concentrationsof free glucose had begun to recover while glucosein the haemolymph continued to rise, eventuallyreaching 2.97"0.42 mmolØml . After 24 h glu-y1

cose concentration in both haemolymph and mus-cle had returned to resting values. Leg muscle inresting G. natalis contained glycogen equivalentto almost 100mmolØg glucose(Fig. 1c). They1

2.5:2.5 exercise regime did not promote significantmobilisation of glycogen. However, after the 4:1exercise treatment the glycogen content was sig-nificantly reduced in the leg muscle by 50% to49.69"13.01mmolØg and thereafter showed ay1

slow recovery and apparently normal glycogenconcentrations after 5 h(Fig. 1c).Despite the apparent fluctuation in ATP concen-

tration between 4.3 and 5.4mmol g the concen-y1

tration was at no time significantly different fromthat in leg muscle of restingG. natalis (Fig. 2).However, the concentration of ADP was increasedby both the 2.5:2.5 and the 4:1 exercise treatmentsand this persisted for at least 5 min into the post-exercise recovery period. Importantly, the 4:1 exer-cise but not the 2.5:2.5 treatment caused a

significant elevation in AMP but this was removedwithin 5 min of recovery(Fig. 2). As a conse-quence of the increases in the AMP and ADP boththe 2.5:2.5 and especially the 4:1 treatment causeda significant decrease in the energy charge to 0.86and 0.85, respectively(Fig. 3). Within 5 min ofrecovery, the restoration of AMP and ADP con-centrations had returned the energy charge to bestatistically similar to that of resting crabs.Arginine phosphate forms a ready and immedi-

ately accessible source of metabolic energy and inexercisedG. natalis reflects the overall metabolicstate(Fig. 4). In the leg muscle of resting crabs


Fig. 2. The concentration of ATP(solid bar), ADP (hatchedbar) and AMP(open bar) in the leg muscle asmmolØg (notey1

axes breaks) of Gecarcoidea natalis at rest, after exercise andduring recovery. The crabs were intermittently exercised in oneof two regimens; either walking for 2.5 with pauses of 2.5 minor walking for 4 min with pauses of 1 min. The data for the2.5:2.5 min group are shown by the lighter bars. The recoverydata are all for crabs exercised in the 4:1 min regimen.ns8for each group. The upward directed arrows indicate significantincreases over resting values.

Fig. 3. The adenylate energy charge(AEC-filled bars) andarginine charge(open bars), calculated from the data summar-ised in Figs. 2 and 4, in the leg muscle ofGecarcoidea natalisat rest, after exercise and during recovery. The crabs were inter-mittently exercised in one of two regimens; either walking for2.5 with pauses of 2.5 min or walking for 4 min with pausesof 1 min. The data for the 2.5:2.5 min group are shown by thelighter bars. The recovery data are all for crabs exercised inthe 4:1 min regimen.ns8 for each group. * indicates a sig-nificant depletion of energy charge compared to that in musclesof resting crabs.

Fig. 4. The concentration of arginine(solid bar) and argininephosphate(open bar) in the leg muscle asmmolØg of Gecar-y1

coidea natalis at rest, after exercise and during recovery. Thecrabs were intermittently exercised in one of two regimens;either walking for 2.5 with pauses of 2.5 min or walking for4 min with pauses of 1 min. The data for the 2.5:2.5 min groupare shown by the lighter bars. The recovery data are all forcrabs exercised in the 4:1 min regimen.ns8 for each group.The upward directed solid arrows indicate significant increasein arginine over resting values and the downward directed openarrows a significant decrease in arginine phosphate comparedto resting values.

there is considerably more arginine phosphate(36.5mmol g ) than arginine(9.2mmol g ); ay1 y1

arginine charge of 0.80. After the 2.5:2.5 exercisetreatment this changed markedly so that largeamounts of arginine phosphate were dephosphor-ylated to arginine, and that arginine exceededarginine phosphate and the arginine chargedeclined to 0.40(Figs. 3 and 4). This depletion ofarginine was even more pronounced after the 4:1exercise treatment(arginine charges0.31) andcontinued for at least 5 min into the recoveryperiod (arginine charges0.27). After 30 min ofrecovery the concentration of arginine phosphatein the muscle had returned to near equal to that ofarginine (arginine charges0.55; Figs. 3 and 4).After 5 h of recovery the concentrations of arginineand arginine phosphate were indistinguishablefrom the initial resting concentrations(argininecharges0.78; Figs. 3 and 4).

4. Discussion

4.1. Intermittent exercise

The locomotion imposed on theGecarcoideanatalis in both exercise groups clearly depended


significantly on anaerobic metabolism. Lactate isthe near exclusive end product of glycolytic fer-mentation in crustaceans(Hill et al., 1991b; Lallierand Walsh, 1992; Hervant et al., 1999b, forreviews). Shortening the duration of the pause inthe intermittent exercise regime and therebyincreasing distance walked by 50% produced aneven greater demand on anaerobiosis. Thus, in the4:1 group approximately 50mmolØg of glucosey1

was mobilised from glycogen within the leg mus-cle without free glucose in the tissue ever risingabove 2.5mmolØg and likely therefore remainedy1

as glucose-6-phosphate to directly fuel muscleglycolysis. It is significant that there was noincrease in haemolymph glucose until well intothe post-exercise period(below). This acceleratedglycolysis during exercise was corroborated by thelactate concentration in the muscle which eventu-ally peaked at over 35mmolØg and fluxed intoy1

the haemolymph to reached 26.4mmolØml . Ay1

similarly large haemolymph lactacidosis occurredin G. natalis walking intermittently at 230% MAS(Adamczewska and Morris, 1998). However, thewalking speeds employed in the present studywere slower than those of migrating crabs, whichmanage their migration without accumulatingL-lactate, depleting glycogen or exhibiting anyhyperglycaemia(Adamczewska and Morris, 1994,2001a,b).During muscle work, anaerobic metabolism

essentially is ‘self-contained’ within the musclecell; and thus factors limiting anaerobic capacityinclude the size of the phosphagen and glycogenstores, and sensitivity to anaerobic end products(Baldwin et al. 1999). The glycogen content ofleg muscle inG. natalis, and other land crabs, canbecome extraordinarily high when adequate foodis available (Henry et al., 1994; Adamczewskaand Morris, 1994, 2000b), and appears to becrucial in supporting the elevated level of anaero-biosis in G. natalis and several other species ofcrab(see Table 2 in Henry et al., 1994). The largedepletion of glycogen and the ensuing glycolysisultimately require the large flux ofL-lactate fromthe muscles into the haemolymph, as must alsohave been the case in other exercising decapods(Henry et al., 1994).The relatively few investigations of adenylate

and phosphagen utilisation by decapod crustaceanshave largely focussed on metabolism in theintensely active tail muscle of crayfish and lobster,with less consideration of leg muscles in terrestrial

crabs (Full and Prestwich, 1986; Weinstein andFull, 1998), or did not consider specific tissues(e.g. Hill et al., 1991a,b). High concentrations ofATP and arginine phosphate, and high activities oflactate dehydrogenase are indicative of musclesthat utilise short-term high intensity bursts ofanaerobic muscle work(Beis and Newsholme,1975; England and Baldwin, 1983; Speed et al.,2001). The ATP concentrations inG. natalis legmuscle(4.76 mmolØg ) were higher than thosey1

in ghost crabs or inCarcinus maenas (;2.5mmolØg , Weinstein and Full, 1998; Hill et al.,y1

1991b), while lower than those of some(7.43mmolØg , England and Baldwin, 1983; 6.48y1

mmolØg , Head and Baldwin, 1986) but not ally1

crayfish (4.09 mmolØg ; Gade, 1984) and thosey1 öf spiny lobster tail muscle(8.66mmolØg , Speedy1

et al. 2001), but comparable to concentrations inthe limbs of spiny lobsters which have loweranaerobic capacity(5.14 mmolØg , Speed et al.,y1

2001). Conversely, the high lactate, and higharginine-phosphate(36.45mmØg ) with low argi-y1

nine (9.20 mmol g ) in G. natalis leg muscley1

were similar to crayfish and lobster tail musclewhich utilise predominantly anaerobic energy pro-vision (England and Baldwin, 1983; Speed et al.,2001). The arginine phosphate of red crab legmuscle is more than 2.5-fold greater than in ghostcrab leg muscle(Full and Weinstein, 1992; Wein-stein and Full, 1998). Indeed, the resting levels ofarginine phosphate provide an arginine charge of0.80 inG. natalis which was considerably greaterthan the 0.73 in spiny lobster(Speed et al., 2001)and only 0.59–0.73 in crayfish tail muscle(Eng-land and Baldwin, 1983; Gade, 1984). The sum¨

of the arginine phosphate and ATP concentrationsin the leg muscle ofG. natalis was greater thanthat of the leg muscle of ghost crabs(Weinsteinand Full, 1998) and in the tail muscle ofHom-marus vulgaris (Beis and Newsholme, 1975) andCherax destructor (England and Baldwin, 1983),and not much less than that in tail muscle of wildcaughtJassus edwardsii (Speed et al., 2001).Burst activity byC. destructor tail muscle ini-

tially depleted arginine phosphate, to maintainATP, and then reduced the adenylate energy charge(AEC) before any anaerobic replenishment of ATPand production of lactate occurred(England andBaldwin, 1983; Baldwin et al., 1999). The exercise


regimens of Weinstein and Full(1998) causedghost crabs to deplete arginine phosphate by 47%apparently to defend the ATP concentration andthis seems a general strategic response of crusta-ceans during O shortfall(Hill et al., 1991b). In2

G. natalis leg muscle the large arginine phosphatestores are important in supporting exercise sincein the 2.5:2.5 treatment there was a 50% decreaseand in the 4:1 group a 61% decline in the argininephosphate charge(Figs. 3 and 4). Arginine phos-phate usage was semi-proportional to the intensityand duration of walking, rather than being imme-diately depleted. This may be due simply to thelarge amount of arginine phosphate stored. Therequirements on the limb muscles in these migra-tory red crabs may be significantly different fromthose on, for example, lobster tail muscle or inwalking ghost crabs.G. natalis must employ theirwalking limbs for both short but rapid locomotionavoiding predators and in inter-male combat, dur-ing which anaerobiosis becomes important(Adam-czewska and Morris, 2001a,b), as well as for long,persistent locomotion during the breeding migra-tion. Arginine phosphate may provide an ‘energybuffer’ both for both short sprints and for activitybriefly exceeding the MAS during the migration.The latter presupposes that red crabs while gener-ally migrating have sufficient aerobic capacity tosupport rephosphorylation of arginine as well asthat required to re-oxidise lactate to pyruvate(seeWeinstein and Full, 1998; Weinstein 2001). Con-sequently, the overall walking speed of migratingcrabs integrated over a longer period would needto be well within their maximum aerobic speed.The very large increase inL-lactate concentra-

tion in G. natalis indicated, however, that the shortexercise regimens made demands beyond depletingarginine phosphate. In crayfish tail muscleL-lactateaccumulated only once the AEC declined to 0.74(England and Baldwin, 1983). G. natalis in the2.5:2.5 and 4:1 exercise groups exhibited deple-tions in AEC to 0.87 and 0.75, respectively. Thiswas due in both cases largely to the increasedwADPx but in the 4:1 group also becausewAMPxincreased to almost double resting levels. The AECof 0.75 is considerably lower than in red crabsexhausted by continuous locomotion(AECs0.80;Adamczewska and Morris, 1994) or in air exposedspiny lobster(Morris and Oliver, 1999; Speed etal., 2001) and crayfish(Morris and Callaghan,1998). Similar or lower AEC in decapods haveonly been achieved by extreme muscular activity

such as tail flipping to exhaustion in crayfish(England and Baldwin, 1983; Gade et al., 1986;¨Head and Baldwin, 1986) or shrimp(Thebault etál., 1997).The leg muscles ofG. natalis appear generally

to respond like tail muscle of intensely exercisingcrayfish-that is highly anaerobically. The exerciseregimens appear a) to impose a near exhaustingload and b) not to afford pauses of adequateduration to allow any replenishment of adenylateor phosphagen. However, the 4:1 exercise regimefacilitated a more extensive use of energy reservesthan did continuous exercise during which therewas no significant depletion of glycogen beforethe crabs were exhausted(Adamczewska and Mor-ris, 1994). G. natalis has a low factorial aerobicscope(Adamczewska and Morris, 1994, 2000b)and obviously endogenous arginine phosphate andanaerobiosis are insufficient to maintain AEC dur-ing the regimens employed. Pausing may be usefulin extending walking distance by providing aperiod without locomotor demand during whichanaerobiosis in the leg muscle may proceed, albeitalready at maximum, to partially replenish ATPandyor arginine phosphate-similarly to ghost crabs(Weinstein and Full, 1998). Pauses also affordtime for lactate efflux from the muscle and therebya lowering of the product of the equilibriumreaction with pyruvate(Gade, 1984). G. natalisëxercised intermittently showed a much largerhaemolymph acidosis than continuously walkingcrabs(Adamczewska and Morris, 1998) presuma-bly since the pauses allowed greater efflux ofmuscle H . Without pauses in exercise the demandq

for ATP of the walking muscle would rapidlyoutstrip the re-supply from arginine phosphate andultimately from glycolysis generally, leading toexhaustion, and before the glycogen pool could beaccessed to any significant extent. Indeed, ghostcrabs moving at high speed may fatigue due torapid anaerobic energy provision even before therate of aerobic ATP replenishment has becomemaximised(Full and Weinstein, 1992).

4.2. Recovery from a regime of four minute walk-ing with one minute pauses

Changes within the leg muscle ofG. natalisduring recovery from exercise revealed interestingpriorities with respect to energy stores and theimportance of arginine phosphate. There was rapidrecovery of the AEC to 0.88 within 5 min of


Table 1L-lactate clearance rate in leg muscle and haemolymph ofG. natalis after intermittent exercise consisting of 4 min walking periodsand 1 min pauses over a distance of approximately 15 m(details see materials and methods)

Time intervals post-exercise 5–30 min 30 min–5 h 5–24 h

Heamolymph(mmol ml h )y1 y1 7.51 4.55 0.10Musclemmol g FW h )y1 y1 5.25 6.21 0.19

Clearance rates calculated from mean concentrations(ns9) and integrated over the intervals shown. Haemolymph volume wasdetermined by Cr -EDTA dilution as 31% body mass(unpublished) and muscle clearance based on the total muscle mass in all51

periopods.

cessation of exercise and to 0.90 within 25 minafter. In contrast, arginine phosphate levels wereslower to recover since after 5 min the argininecharge had actually declined further and after 30min recovery was still only 0.55. Indeed, it appearsmaintenance of lowered arginine phosphate con-tributes to the re-establishment of AEC duringearly recovery. Normal resting arginine phosphatewas established within 5 h of recovery. In theirstudies ofC. destructor tail muscle England andBaldwin (1983) were of the opinion that argininephosphate was the primary source for ATP resyn-thesis and was rapidly regenerated to supportfuture muscle work. InC. destructor L-lactatecontinued to accumulate implying that anaerobicglycolysis made a significant contribution to therecovery(England and Baldwin, 1983). The verylarge and persistent post-exercise increase inL-lactate inG. natalis was indication that anaero-biosis was important in recovery of ATP levelswithin the walking leg muscle; a conclusionreached also forC. maenas during recovery fromhypoxia (Hill et al., 1991a,b). However, argininephosphate, and arginine charge, was recoveredmore slowly than AEC, within the same time-frame as recovery of restingL-lactate concentra-tions. Rapid aerobic recovery is characteristic ofghost crabs and energy stores were partly replen-ished during pauses(Weinstein and Full, 1992;Full and Weinstein, 1992). Phosphagen in mam-mals and insects is normally much lower than indecapod crustaceans(e.g. Beis and Newsholme,1975) and can only provide a rapid but briefbuffering of ATP levels (e.g. Bogdanis et al.,1996). In G. natalis, and probably other crabs,arginine phosphate makes a more fundamentalcontribution to longer term demand, behaving inmany ways as an extension of the adenylate pool.This is quite a different model to that of a rapiddepletion and replenishment cycle in which phos-

phagen ‘floats’ on top of the ATP pool affordingsome protection to the AEC.The dynamics of lactate recovery are character-

istically slow in decapods but the fate ofL-lactateremains unclear in crustaceans(e.g. Gade, 1984;¨Gade et al., 1986; Head and Baldwin, 1986; Hillët al., 1991a; Henry et al., 1994), including G.natalis (Adamczewska and Morris, 1998), but iscertainly variable between species(Oliveira andda Silva, 1997; Hervant et al., 1999b). In G.natalis post the 4:1 exercise regime reduction ofL-lactate within leg muscle and the haemolymphwas similar and thus apparently in equilibrium(Table 1); and provides no evidence for the siteof L-lactate removal. InC. destructor the timecourse for change in concentration of lactate inhaemolymph lactate and in muscle lactate werealso similar(Head and Baldwin, 1986) but in thisaquatic species significant amounts of lactate wereexcreted rather than metabolised. In red crab mus-cle, large amounts of glycogen are progressivelybut relatively rapidly resynthesised, within 5 h ofthe cessation of exercise. In view of the utilisationand reestablishment of glycogen within the muscleit seems important to consider the possible meta-bolic fate of L-lactate. Assuming that the lactatein exercisedG. natalis originated from the glyco-genolysis it seems very probable the muscle gly-cogen was re-established via the oxidation ofsignificant amounts of lactate to pyruvate and thenconverted into glycogen, rather than contributingto metabolism in the TCA cycle. This hypothesesis supported by reciprocal time course for lactateremoval and glycogen recovery.Decapod crustaceans incorporatew Cxlactate14

into glucose and glycogen(Thabrew, et al., 1971;Gade, et al., 1986; van Aardt, 1988; Hill et al.,¨1991a; Henry et al., 1994; Oliveira and da Silva,1997, 2000; Hervant et al., 1999b) but the sites ofgluconeogenesis and glyconeogenesis are still


largely undetermined. InG. natalis the leg musclesare rich in glycogen and the midgut gland storesglycogen and glucose(Adamczewska and Morris,2001a). Re-establishment of muscle glycogen post-exercise could proceed using extra-muscular glu-cose derived from the midgut gland, either fromthe stores or by gluconeogenesis from lactate viapyruvate as postulated in the mid gut gland ofChasmagnathus granulata (Oliveira and da Silva,1997). Alternatively glyconeogenesis from glucosederived from L-lactate endogenously within themuscle cells could occur, as is now widely sug-gested for some vertebrate skeletal muscle(Schul-te, et al., 1992; Moyes et al., 1992; Palmer andFournier, 1997; Donovan and Pagliassotti, 2000)and proposed in at least one crustacean(Hervantet al., 1999b). The latter would require the refluxof haemolymphL-lactate into the muscle cellsduring recovery. Gluconeogenesis in the midgutgland would also remove lactate but subsequentlyrequire haemolymph glucose transport to, and into,muscle cells for conversion to glucose-UDP forglyconeogenesis. Henry et al.(1994) did not detectany activity for gluconeogenic enzymes; phospho-nenolpyruvate carboxykinase(PEPCK), pyruvatecarboxylase(PC), and fructose-1, 6-biphosphatase(FBPase) in either muscle or midgut gland of threedecapods, including the terrestrialG. lateralis.Significant PEPCK activity was found in the mid-gut of the semi-terrestrialC. granulata which wasable to incorporate both C-lactate and C-alanine14 14

into glucose in this organ(Oliveira and da Silva,1997). In contrast, Lallier and Walsh(1991) foundappreciable reverse LDH activity and PEPCKactivity in the light levator muscle inCallinectessapidus but almost no LDH in the midgut gland-leading to the conclusion that gluconeogenic activ-ity in the midgut gland would need to be fuelledby circulating pyruvate. In a subsequent analysisthese authors(Lallier and Walsh, 1992) concludedthat any Cori cycling would account for only 1%of the lactate produced during exercise and thus isunimportant in crustaceans. Lallier and Walsh(1992) go on to suggest thatL-lactate oxidationand gluconeogenesis may occur in the dark andlight muscle fibres. Henry et al.(1994) examinedpost-exercise recovery in species including theterrestrialG. lateralis and found that only 1% orless of the C-lactate injected 1h post-exercise14

appeared in glycogen; and derived a glycogenrestoration rate of 0.26mmol.Øg Øh . This iny1 y1

turn prompted the conclusion, that accumulated

lactate was more likely to be converted to amino-acid (see also Hill et al., 1991a), andyor thatgluconeogenesis proceeded through a pathway oth-er than the Cori cycle, or so slowly that very lowenzyme activity was required. More recently Her-vant et al.(1999b) hypothesised the existence oftwo distinct sites: a gluconeogenic organ utilisingextracellular glucose and a separate glyconeogenicorgan(muscle). Although Hervant et al.(1999b)re-enliven the Cori cycle as a useful pathway incrustaceans, the bulk of glycogen synthesisoccurred by an intracellular pathway without glu-cose synthesis and during recovery from hypoxiaoxidative metabolism of lactate was depressed-favouring glyconeogenesis. Similarly Hill et al.(1991a,b) found significant amounts of C-lactate14

appeared in glycogen of post-hypoxicC. maenaswithout, curiously, appreciable amounts appearingglyconeogenic intermediates. It has been arguedthat in the glycolytic route from PEP to pyruvatethat equilibrium for pyruvate kinase(PK) is so fardisplaced in favour of pyruvate as to be essentiallyirreversible (e.g. Connett, 1979; Bonen et al.,1990). However, in vertebrate skeletal muscle PEPmay be formed by the reversal of pyruvate kinase,since the vast majority of ADP measured in musclecells is in fact bound to myosin thus favourablyincreasing the ATP:ADP ratio(Schulte et al., 1992;Donovan and Pagliassotti, 2000). Pyruvate kinasenecessarily occurs at high levels in crab muscles(Lallier and Walsh, 1991) but as there are so fewdata for crustaceans it is impossible to conclude ifPEP formation by reversed PK activity as a basisfor glygoneogensis occurs, although such an alter-native route is increasingly invoked(e.g. Lallierand Walsh, 1992; Henry et al., 1994; Hervant etal., 1999b)Much more work on glycogen restoration in

decapods is required. In red crabs post-exercisemuscle glycogen was re-established at a nett rateof 14.1mmol g h and in the study of Henryy1 y1

et al. (1994) at approximately 16mmolØg hy1 y1

in Callinectes sapidus (see Fig 6 in Henry et al.,1994) despite less than 0.5% of injectedw Cxlactate appearing in the glycogen. The finding14

for C. sapidus is intriguing in that it implies the50% depletion of muscle glycogen is not regener-ated directly from lactate but from glucose origi-nating from an unlabeled source.The dynamics of the data fromG. natalis

suggest a different conclusion. The concentrationof glucose in the leg muscle of red crabs stabilised


within 5 min post-exercise and then declined, dueto uptake into glycogen, while haemolymph glu-cose continued to increase for at least the next 5h. It seems unlikely that muscle gluconeogenesiscould generate sufficient glucose to produce anefflux since the equilibrium concentration of fruc-tose 6-phosphate would slow the biphosphatase inthe preceding step of gluconeogenesis. Glyconeo-genesis directly from lactate seems equally unlike-ly since it does not explain the hyperglycaemianor is there any decline in ADP that wouldencourage a reversal of pyruvate kianse. The per-sistent and increasing hyperglycaemia must haveits origin in glucose synthesis away from themuscle—a classic Cori cycle. Thus, the data areconsistent withL-lactate efflux from the musclewhich enters gluconeogenesis in some other organ,perhaps the mid gut gland(Oliveira and da Silva,1997), and glucose transport to the muscle andeventually into the muscle for re-storage as gly-cogen. The mid-gut gland of red crabs storesglycogen, glucose and triglycerides but there is noevidence that these are mobilised to support loco-motion but rather they might be seasonal stores(Adamczewska and Morris, 2000b, 2001b). In a250 g red crab an increase in standing haemolymphglucose concentration of approximately 2.5mmoll would provide reservoirs of near 190mmoly1

glucose, which is sufficient to replenish near 4 gof muscle with glycogen. Given the rate ofL-lactate clearance and assuming it leads to glucoseresynthesis it is clear that at least 10 times thisamount of glucose can be recovered which is morethan enough for the total walking muscles in redcrabs. The cost of this putative lactate oxidation(as NAD recycling) and gluconeogenesis inG.q

natalis compared to excretion of lactate(disregard-ing the loss of glycogen stores) as hypothesisedfor Cherax may be a contributing factor to therelatively slow recovery of phosphagen stores inred crabs compared to crayfish.

4.3. Migration in red crabs

It is difficult to reconcile the consequences ofintermittent locomotion with observations onmigrating red crabs. The present study selectedwalking speeds within the mean walking speedrecorded for migrating red crabs and within theaerobic scope of the animals. Many hours ofmigration did not promote lacticidosis, hypergly-caemia or use any part of the glycogen store

(Adamczewska and Morris, 2001b) whereas thevery short exercise regimens in the present studyshowed all the features of incipiently exhaustingactivity. In ghost crabs intermittent exercise withinappropriate exercise pause durations can mark-edly decrease distance(Full and Weinstein, 1992).However, red crabs outside of the migration seasonencouraged, in the field, to just 5 min of slowwalking in the field also showed large lacticidosisas well as glucose mobilisation(Adamczewskaand Morris, 2000b). The inescapable conclusion isthat during the migrationG. natalis are possessedof a very different metabolism compared to duringthe rest of the year(Adamczewska and Morris,2000a). Preliminary evidence supports the sugges-tion that during the wet seasonG. natalis bothproduce more active hyperglycaemic hormone andare more sensitive to the hormone(Adamczewskaand Morris, 2000a; Morris unpublished). Further-more, crabs captured during the migration show aserotonin dependent hyperglycaemia, which pro-gressively disappears during eight weeks of lab-oratory confinement (Morris, unpublished).Intermittent exercise apparently increased distanceby increasing the extent of glycolysis that couldbe completed before AEC fell to entirely debilitat-ing levels, as seemed to be the case previously(Adamczewska and Morris, 1998). Thus the modelpresented here may describe locomotion in redcrabs for the majority of the year but it seems notfor migrating red crabs. Resolution of the role ofphosphagens and adenylates in migrating red crabsawaits the logistically challenging field determi-nation of these compounds, the fate of radio-tracersand delineating the hormonal control mechanismin energy metabolism. Repeating the protocol usedin the present study, under field conditions, oncrabs immediately prior to, during and aftermigrating should finally resolve this question.

Acknowledgments

This work was supported by funds from MorlabEnvironmental Physiology and from NaturalEvents(www.natural-events.com). Our thanks goto the Government Conservator and staff of ParksAustralia, Christmas Island.

References

Adamczewska, A.M., Morris, S., 1994. Exercise in the terres-trial Christmas Island Red CrabGecarcoidea natalis. IIEnergetics of locomotion. J. Exp. Biol. 188, 257–274.


Adamczewska, A.M., Morris, S., 1998. Strategies for migrationin the terrestrial Christmas Island red crabGecarcoideanatalis: intermittent vs. continuous locomotion. J. Exp. Biol.201, 3221–3231.

Adamczewska, A.M., Morris, S., 2000a. Locomotion, respira-tory physiology, and energetics of amphibious and terrestrialcrabs. Physiol. Biochem. Zool. 73, 706–725.

Adamczewska, A.M., Morris, S., 2000b. Respiratory gas trans-port, metabolic status, and locomotor capacity of the Christ-mas Island Red CrabGecarcoidea natalis assessed in thefield with respect to dichotomous seasonal activity levels.J. Exp. Zool. 286, 552–562.

Adamczewska, A.M., Morris, S., 2001a. Ecology and behav-iour of Gecarcoidea natalis, the Christmas Island red crab,during the annual breeding migration. Biol. Bull. 200,305–320.

Adamczewska, A.M., Morris, S., 2001b. Metabolic status andrespiratory physiology ofGecarcoidea natalis, the ChristmasIsland red crab, during the annual breeding migration. Biol.Bull. 200, 321–335.

Baldwin, J., Gupta, A., Iglesias, X., 1999. Scaling of anaerobicenergy metabolism during tail flipping behaviour in thefreshwater crayfish,Cherax destructor. Mar. Freshwater Res.50, 183–187.

Beis, I., Newsholme, E.A., 1975. The contents of adeninenucleotides, phosphagens and some glycolytic intermediatesin resting muscles from vertebrates and invertebrates. Bioch-em. J. 152, 23–32.

Bogdanis, G.C., Nevill, M.E., Boobis, L.H., Lakomy, H.K.A.,1996. Contribution of phosphocreatine and aerobic metabo-lism to energy supply during repeated sprint exercise. J.Appl. Physiol. 80, 876–884.

Bergmeyer, H.U., 1985a. Methods of enzymatic analysis. VolVI, Metabolites 1: Carbohydrates. 3rd Edn.. VCH Verlags-gesellschaft, Weinheim.

Bergmeyer, H.U., 1985b. Methods of enzymatic analysis. VolVII, Metabolites 2: Tri and dicarboxylic acids, purine,pyrimidines, inorganic compounds. 3rd Edn.. VCH Verlags-gesellschaft, Weinheim.

Bergmeyer, H.U., 1985c. Methods of enzymatic analysis.Metabolites 3: lipids, amino acids and related compoundsIIX. 3rd Ed.. VCH Verlagsgesellschaft, Weinheim.

Bonen, A., McDermott, J.C., Tan, M.H., 1990. Glycogenesisand glycogneogenesis in skeletal muscle: effects of pH andhormones. Am. J. Physiol. 258, E693–E700.

Connett, R., 1979. Glycogneogenesis from lactate in frogstriated muscle. Am. J. Physiol. 237, C231–C236.

Donovan, C.M., Pagliassotti, M.J., 2000. Quantitative assess-ment of pathways for lactate disposal in skeletal musclefibre types. Med. Sci. Sport Exer. 32, 772–777.

England, W.R., Baldwin, J., 1983. Anaerobic energy metabo-lism in the tail musculature of the Australian yabbyCheraxdestructor (Crustacea, Decapoda, Parastacidae): role ofphosphagens and anaerobic glycolysis during escape behav-iour. Physiol. Zool. 56, 614–622.

Ellington, W.R., 1983. The recovery from anaerobic metabo-lism in invertebrates. J. Exp. Zool. 228, 431–444.

Full, R.F., Prestwich, K., 1986. Anaerobic metabolism duringwlaking and bouncing gaits in ghost crabs. Am. Zool. 26,88A.

Full, R.F., Weinstein, R.B., 1992. Integrating the physiology,mechanics and behaviour of rapid running ghost crabs: Slowand steady doesn’t always win the race. Am. Zool. 32,382–395.

Gade, G., 1984. Effects of oxygen deprivation during anoxiaänd muscular work on the energy metabolism of the cray-fish, Orconectes limosus. Comp. Biochem. Physiol. 77A,495–502.

Gade, G., Graham, R.A., Ellington, W.R., 1986. Metabolic¨disposition of lactate in the horseshoe crabLimulus poly-phemus and the stone crabMenippe mercenaria. Mar. Biol.91, 473–479.

Gladden, L.B., 2000. Muscle as a consumer of lactate. Med.Sci. Sport Exer. 32, 764–771.

Gleeson, T.T., Hancock, T.V., 2001. Modeling the metabolicenergetics of brief and intermittent locomotion in lizardsand rodents. Am. Zool. 41, 211–218.

Head, G., Baldwin, J., 1986. Energy metabolism and the fateof lactate and during recovery from exercise in the Austra-lian freshwater crayfishCherax destructor. Aust. J. Mar.Freshwat. Res. 37, 641–646.

Henry, R.P., Booth, C.E., Lallier, F.H., Walsh, P.J., 1994. Post–exercise lactate production and metabolism in three speciesof aquatic and terrestrial decapod crustaceans. J. Exp. Biol.186, 215–234.

Herreid, C.F., Full, R.J., 1988. Energetics and locomotion. In:Burggren, W.W., McMahon, B.R.(Eds.), Biology of theLand Crabs. Cambridge University Press, New York, pp.333–377.

Hervant, F., Mathieu, J., Messana, G., 1999a. Locomotory,ventilatory and metabolic responses of subterraneanStena-sellus virei (Crustacea, Isopoda) to severe hypoxia andsubsequent recovery. C.R. Acad. Sci. Paris. 320, 139–148.

Hervant, F., Garin, D., Mathieu, J., Freminet, A., 1999b.Lactate metabolism and glucose turnover in the subterraneancrustaceanNiphargus virei during post-hypoxic recovery. J.Exp. Biol. 202, 579–592.

Hill, A.D., Strang, R.H.C., Taylor, A.C., 1991a. Radioisotopestudies of the energy metabolism of the shore crabCarcinusmaenas (L.) during environmental anoxia and recovery. J.Exp. Mar. Biol. Ecol. 150, 51–62.

Hill, A.D., Taylor, A.C., Strang, R.H.C., 1991b. Physiologicaland metabolic responses of the shore crabCarcinus maenas(L.) during environmental anoxia and subsequent recovery.J. Exp. Mar. Biol. Ecol. 150, 31–50.

Irschick, D.J., Garland, T., 2001. Integrating function andecology in studies of adaptation: Investigations of locomotorcapacity as a model system. Ann. Rev. Ecol. Syst. 32,367–396.

Kramer, D.L., McLaughlin, R.L., 2001. The behavioural ecol-ogy of intermittent locomotion. Am. Zool. 41, 137–153.

Lallier, F.H., Walsh, P.J., 1991. Metabolic potential in tissuesof the blue-crab,Callinectes sapidus. B. Mar Sci. 48,665–669.

Lallier, F.H., Walsh, P.J., 1992. Metabolism of isolated hepa-topancreas cells from the blue-crab(Callinectes sapidus)under simulated postexercise and hypoxic conditions. Phy-siol. Zool. 65, 712–723.

Morris, S., Callaghan, J., 1998. The emersion response of theAustralian yabby Cherax destructor to environmentalhypoxia and the respiratory and metabolic responses toconsequent air-breathing. J. Comp. Physiol. 168B, 389–398.


Morris, S., Oliver, S., 1999. Circulatory, respiratory and met-abolic response to emersion and low temperature ofJasusedwardsii: Simulation studies of commercial shipping meth-ods. Comp. Biochem. Physiol. 122A, 299–308.

Moyes, C.D., Schulte, P.M., Hochachka, P.W., 1992. Recoverymetabolism iof trout white muscle: role of mitochondria.Am. J. Physiol. 262, R295–R304.

Oliveira, G.T., da Silva, R.S.M., 1997. Gluconeogenesis inhepatopancreas ofChasmagnathus granulata crabs main-tained on high-protein of carbohydrate-rich diets. Comp.Biochem. Physiol. 118A, 1429–1435.

Oliveira, G.T., da Silva, R.S.M., 2000. Hepatopancreas gluco-neogenesis during hyposmotic stress in crabsChasmagna-thus granulata maintained on high-protein orcarbohydrate-rich diets. Comp. Biochem. Physiol. 127B,375–381.

Palmer, T.N., Fournier, P.A., 1997. Replenishment of muscleglyocogen after high intensity exercise: a role for intramus-cular lactate glyconeogenesis? Biochem. Soc. Trans. 25,25–30.

Schulte, P.M., Moyes, C.D., Hochachka, P.W., 1992. Integratingmetabolism pathways in post exercise recovery of whitemuscle. J. Exp. Biol. 166, 181–195.

Speed, S.R., Baldwin, J., Wong, R.J., Wells, R.M.G., 2001.Metabolic characteristics of muscles in the spiny lobster,

Jasus edwardsii, and responses to emersion during simulatedlive transport. Comp. Biochem. Physiol. 128B, 435–444.

Thabrew, M.I., Poat, P.C., Munday, K.A., 1971. Carbohydratemetabolism inCarcinus maenas gill tissue. Comp. Biochem.Physiol. 40B, 531–541.

Thebault, M.T., Raffin, J.P., Pichon, R., 1997. The effect of´

the stimulation regime onPalaemon serratus tail muscleenergy metabolism. Comp. Biochem. Physiol. 116A,240–337.

van Aardt, W.J., 1988. Lactate metabolism and glucose patternsin the River crabPotamonautes warreni Calman, duringanoxia and subsequent recovery. Comp. Biochem. Physiol.91A, 299–304.

Weinstein, R.B., 2001. Terrestrial intermittent exercise: Com-mon issues for human athletics and comparative animallocomotion. Am. Zool. 41, 219–228.

Weinstein, R.B., Full, R.J., 1992. Intermittent exercise altersendurance in an eight-legged ectotherm. Am. J. Physiol.262, R852–R859.

Weinstein, R.B., Full, R.J., 1998. Performance limits of low-temperature, continuous locomotion are exceeded whenlocomotion is intermittent in the ghost crab. Physiol. Zool.71, 274–284.

Documents

Utilisation of glycogen, ATP and arginine phosphate in exercise and recovery in terrestrial red crabs, Gecarcoidea natalis