High aerobic capacities in the skeletal musclesof pinnipeds: adaptations to diving hypoxia

SHANE B. KANATOUS,1 LEONARD V. DIMICHELE,2DANIEL F. COWAN,3 AND RANDALL W. DAVIS2

1Department of Medicine, University of California at San Diego, La Jolla, California, 92092-0623A;2Department of Wildlife and Fisheries Science, Texas A & M University, College Station,Texas 77845; and 3Department of Pathology, University of Texas Medical Branch,Galveston, Texas 77555-0588

Kanatous, Shane B., Leonard V. DiMichele, Daniel F.Cowan, and Randall W. Davis. High aerobic capacities inthe skeletal muscles of pinnipeds: adaptations to divinghypoxia. J. Appl. Physiol. 86(4): 1247–1256, 1999.—Theobjective was to assess the aerobic capacity of skeletalmuscles in pinnipeds. Samples of swimming and nonswim-ming muscles were collected from Steller sea lions (Eumeto-pias jubatus, n 5 27), Northern fur seals (Callorhinus ursi-nus, n 5 5), and harbor seals (Phoca vitulina, n 5 37) by usinga needle biopsy technique. Samples were either immediatelyfixed in 2% glutaraldehyde or frozen in liquid nitrogen. Thevolume density of mitochondria, myoglobin concentration,citrate synthase activity, and b-hydroxyacyl-CoA dehydroge-nase was determined for all samples. The swimming musclesof seals had an average total mitochondrial volume densityper volume of fiber of 9.7%. The swimming muscles of sealions and fur seals had average mitochondrial volume densi-ties of 6.2 and 8.8%, respectively. These values were 1.7- to2.0-fold greater than in the nonswimming muscles. Myoglo-bin concentration, citrate synthase activity, and b-hydroxy-acyl-CoA dehydrogenase were 1.1- to 2.3-fold greater in theswimming vs. nonswimming muscles. The swimming musclesof pinnipeds appear to be adapted for aerobic lipid metabo-lism under the hypoxic conditions that occur during diving.

mitochondria; citrate synthase; b-hydroxyacyl-coenzyme Adehydrogenase; myoglobin

ON THE BASIS of laboratory simulated-dive studies (8), itwas first believed that marine mammals relied on anenhanced anaerobic capacity to maintain ATP synthe-sis under the hypoxic conditions of diving. It was latershown that marine mammals did not possess unusuallyhigh anaerobic enzyme activities compared with terres-trial mammals (3). With the development of dive record-ers, information became available on the free-rangingdiving behavior of marine mammals, and it was foundthat the majority of dives remained within the aerobicdive limit of the animals. The aerobic dive limit isdefined as the longest breath-hold dive that is possiblewithout an increase in lactic acid concentration in theblood during or after the dive. During these dives,marine mammals appear to maintain an overall lowmetabolic rate and rely principally on oxygen stored in

the blood and muscles in order to maintain aerobicmetabolism and physiological homeostasis (25).

In terrestrial mammals, the functional capacity ofthe respiratory pathway to deliver oxygen is matched tothe oxygen requirements of the tissues (15, 18, 29). Forexample, the large aerobic scopes of athletic animals,such as dogs and racehorses, result from both highmitochondrial volume densities in skeletal muscles andcardiorespiratory adaptations that enhance convectiveand diffusive oxygen transport (18, 24). Unlike terres-trial mammals that increase ventilation and cardiacoutput during exercise, marine mammals stop breath-ing, reduce cardiac output, and limit peripheral bloodflow during diving. This dive response (i.e., apnea,bradycardia, and peripheral vasoconstriction) reducesconvective oxygen transport to muscles resulting intissue hypoxia (8) and appears to limit aerobic scope (6).However, studies on the free-ranging diving behavior ofmarine mammals indicate that they still maintainaerobic metabolism during the majority of their dives(6, 13, 25). How, then, do the skeletal muscles ofpinnipeds maintain aerobic metabolism under suchadverse conditions?

Analogies have been drawn between high-altitude-adapted animals and marine mammals in terms ofpossible adaptations in skeletal muscles to maintainaerobic metabolism under hypoxic conditions (12). Un-like high-altitude animals that function under hypoxicconditions but display the typical exercise response ofincreasing ventilation and cardiac output, marine mam-mals exercise under a different form of hypoxic stress.They must function for the duration of a dive with onlya finite amount of oxygen. A number of studies havepostulated a possible downregulation of oxidative me-tabolism, to levels approximately one-half those foundin terrestrial mammals of comparable size, to maintainoxidative metabolism under the hypoxic conditionsassociated with diving (12, 13). The objective of thisstudy was to assess the aerobic capacity of pinnipedskeletal muscle by measuring the volume density ofmitochondria, the activity of key oxidative enzymes,and the concentration of myoglobin in swimming andnonswimming muscles. Our results indicate that pin-niped skeletal muscles have an enhanced oxidativecapacity to maintain aerobic lipid metabolism underthe hypoxic conditions associated with diving and thatthese adaptations are more pronounced in swimmingthan in nonswimming muscles.

The costs of publication of this article were defrayed in part by thepayment of page charges. The article must therefore be herebymarked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734solely to indicate this fact.

8750-7587/99 $5.00 Copyright r 1999 the American Physiological Society 1247http://www.jap.org

MATERIALS AND METHODS

Animals. Muscle samples from healthy adult female Stellersea lions (Eumetopias jubatus; n 5 27, mean body mass260 kg) and healthy adult harbor seals (Phoca vitulina; n 537, mean body mass 61 kg) were collected over a 3-yr periodfrom central and southeast Alaska as part of ongoing studiesconducted by the Alaska Department of Fish and Game andthe National Marine Fisheries Service. Samples from malejuvenile Northern fur seals (Callorhinus ursinus; n 5 5,estimated body mass 29 kg) were obtained immediately afterdeath during a native subsistence hunt in the Pribilof Is-lands, AK.

Biopsy procedure. The sea lions were darted with Telezol(2.0 mg/kg, tiletamine HCl and zolazepam HCl; Fort DodgeLaboratories, Fort Dodge, IA), intubated, and maintainedwith isoflourane anesthesia by using large-animal anesthesiaequipment and monitoring techniques (11). The harbor sealswere captured with entangling nets and sedated with amixture of Ketalar (2.0 mg/kg ketamine HCl, Parke-Davis,Morris Plains, NJ) and Valium (0.1 mg/kg diazepam, RocheProducts, Manati, PR). Two muscle samples of ,50 mg eachwere collected from a single biopsy site with a 6-mm biopsycannula (Depuy, Warsaw, IN) from both the swimming andnonswimming muscles of both species. For the harbor seals,biopsies of swimming and nonswimming muscles were takenfrom the longissimus dorsi and pectoralis muscles, respec-tively (Fig. 1). Samples from the swimming and nonswim-ming muscles of sea lions and fur seals were taken from thepectoralis muscle and hindlimb complex, respectively. Con-trol samples for electron microscopy were collected from thesoleus muscle, a predominantly slow-twitch oxidative muscle,of laboratory rats (Sprague Dawley) that were killed bycervical dislocation after 2–3 min of carbon dioxide anesthe-sia. Control samples for citrate synthase (CS) and myoglobinassays were collected from the vastus medialis muscle offree-ranging cotton rats (Sigmodon hispidus) that were killedby 2- to 4-min exposure to methane gas. The cotton rats weretrapped as part of a study to assess the distribution andhealth status of small mammals in the wetlands of Texas.

Tissue preparation. Muscle samples were either placed into2% glutaraldehyde and 2% paraformaldehyde fixative in 0.1M cacodylate buffer or were frozen in liquid nitrogen immedi-ately after they were collected. Samples remained in thefixative for 2–14 days before being transferred and stored in0.1 M cacodylate buffer, pH 7.4, for 21–90 days. Frozensamples were stored in liquid nitrogen until they werereturned to Texas A & M University and then stored at 270°Cuntil analysis for CS activity and myoglobin concentration.

Electron microscopy. Fixed muscle samples were rinsed in0.1 M cacodylate buffer and postfixed for 2 h in a 1% solutionof osmium tetroxide. They were stained en bloc with 2%uranyl acetate overnight at 5°C. After dehydration withincreasing concentrations of ethanol (50–100%), they werepassed through propylene oxide and increasing concentra-tions of epoxy (50–100%). The samples were finally embeddedin fresh epoxy and allowed to polymerize overnight at 60°C.Semithick sections (1 µm) were cut with a Leica Ultratomeand stained with toluidine blue to check for fiber orientation.Ultrathin (50–70 nm) transverse sections from four randomlychosen blocks per muscle were cut and contrasted with leadcitrate. Micrographs were taken with a Phillips 201 transmis-sion electron microscope. The number of micrographs permuscle that were analyzed was between 25 and 40, yieldingrelative SE of total mitochondrial volume density [Vv(mt,f)]estimates of ,10% in all muscles. Determination of thevolume densities per volume of muscle fiber of mitochondria,

myofibrils [Vv(fi,f)], and lipid droplets [Vv(li,f)] were per-formed by using standard point-counting procedures (18, 29).

Enzyme and myoglobin assays. Frozen muscle sampleswere thawed, blotted, weighed, and immediately homog-enized at 0°C in 1 ml of buffer containing (in mM): 1 EDTA,2 MgCl2, and 75 Tris ·HCl, pH 7.6, at 25°C (33). The homoge-nates were spun at 2,900 g for 30 min at 4°C. A 500-ml aliquotfrom the supernatant was prepared for myoglobin assays,and the rest was used for the CS and b-hydroxyacyl-CoAdehydrogenase (HAD) assays. CS was assayed on a BeckmanDU series 64 spectrophotometer according to the method ofReed et al. (33). Assay temperature was maintained at 37°Cby using a constant-temperature water bath and a water-jacketed cuvette holder. The assay conditions for CS(EC 4.1.3.7) were (in mM): 50 imidazole, 0.25 5,58-dithiobis(2-nitrobenzoic acid), 0.4 acetyl CoA, and 0.5 oxaloacetate, at pH7.5; DA412, e412 5 13.6, where DA indicates absorbance

Fig. 1. Skeletal muscle anatomy of seals and sea lions. Biopsysamples of swimming muscles were obtained from longissimus dorsiin harbor seals and pectoralis in fur seals and sea lions. Samples ofnonswimming muscles were collected from hindlimb complex in furseals and seal lions and from pectoralis in harbor seals. [Redrawnfrom A. B. Howell (Ref. 20a) by M. Cooley.]

1248 HIGH AEROBIC CAPACITIES IN PINNIPED SKELETAL MUSCLE

wavelength and e is exinction coefficient. Assay conditions forHAD (EC 1.1.1.35) were (in mM): 50 imidazole, 1 EDTA,0.1 acetoacetyl CoA, and 0.15 NADH, pH 7.0, at 37°C; DA340,e340 5 6.22 (33). Enzyme activities, in micromoles per min pergram wet mass muscle, were calculated from the rate ofchange in absorbance at the maximum linear slope. Myoglo-bin was assayed according to the method of Reynafarje (34),with the following modifications. A portion (500 µl) of thehomogenate was further diluted with 1 ml of phosphate buffer(0.04 M, pH 6.6). The resulting mixture was centrifuged for 50min at 28,000 g at 4°C. The supernatant was bubbled withcarbon monoxide for 3 min. Spectrophotometric absorbance(Abs) was measured at 538 and 568 nm, and the concentra-tion of myoglobin, in milligrams per gram wet mass of muscle,was calculated as

(Abs538 2 Abs568) 3 5.865[(1.5/0.5) 3 (mass of sample)]

Statistical analysis. Results are expressed as means 6 SE.Statistical comparisons were made between species by usingan ANOVA (Bonferroni-adjusted P value # 0.01), and com-parisons between swimming and nonswimming muscleswithin a species were made by paired sample t-test (P # 0.05).Correlations between oxidative enzymes and the volumedensity of mitochondria were determined, and significancewas assessed at P # 0.05.

RESULTS

Skeletal muscle morphology. The general morphologyof the skeletal muscle fibers was similar among thethree species of pinnipeds and was typical for mammals(Fig. 2). Electron micrographs of transverse sectionsshowed irregularly shaped myofibrils separated bysarcoplasmic reticulum, mitochondria, and lipid drop-lets. The mean volume density of myofibrils per volumeof muscle fiber was not significantly different betweenthe swimming (76.2 6 1.8 to 81.4 6 2.0%) and nonswim-ming muscles (79.5 6 1.7 to 82.1 6 2.4%) in all threespecies (Table 1).

In each of the three species, the volume densities oftotal mitochondria (interfibrillar and subsarcolemmal)per volume of muscle fiber were significantly greater(1.7- to 2.0-fold) in swimming than in nonswimmingmuscles (Table 1). Most of the mitochondria (mean,95 6 0.67%; range, 93–96.8%) were interfibrillar, andthe remainder were subsarcolemmal. Harbor seals andfur seals had significantly higher Vv(mt,f) in theirswimming and nonswimming muscles than did Stellersea lions (ANOVA, P 0.01; Table 1).

Roughly spherical lipid droplets ,0.25 µm in diam-eter were distributed in the interfibrillar space. TheVv(li,f) for both the swimming and nonswimmingmuscles was generally ,1% (Table 1). The swimmingmuscles of fur seals had a significantly greater Vv(li,f)than did nonswimming muscles, whereas there were nosignificant differences between the swimming and non-swimming muscles of harbor seals and Steller sea lions.The Vv(li,f) in the swimming muscle of the fur seals wassignificantly greater than in the muscles of the harborseals and Steller sea lions.

Oxidative enzymes. CS activities were 1.1–1.5 timesgreater in the swimming muscles than in the nonswim-ming muscles and were significantly correlated (P ,0.001) to the Vv(mt,f) in all three species (Fig. 3A).

Differences between swimming and nonswimmingmuscles were significant in the Steller sea lions and theNorthern fur seals (paired sample t-test, P value #0.05). The activity of CS was significantly greater in theswimming muscles of the harbor seals and Northernfur seals compared with the swimming muscle of theSteller sea lions and the vastus medialis of the cottonrat (ANOVA, P # 0.01; Table 2). The activities of HADwere 1.0–1.3 times greater in the swimming musclesthan in the nonswimming muscles and were alsosignificantly correlated (P , 0.001) to the volumedensity of mitochondria in all three species (Fig. 3B).HAD values were significantly different in the harborseals and Steller sea lions (ANOVA, P # 0.05). Theywere approximately twofold greater in the swimmingmuscles of the harbor seals and Northern fur sealscompared with the Steller sea lions, and they weresignificantly greater in the muscles of all pinnipedscompared with the vastus medialis of the cotton rat(ANOVA, P # 0.01; Table 2).

The CS/HAD ratio was calculated as an index of theamount of total aerobic metabolism supported by theoxidation of fatty acids. This ratio ranged from 0.59 to0.97 and was similar between all three species ofpinnipeds for both swimming and nonswimmingmuscles (Table 2). These ratios were two- to threefoldlower than in the rat.

Myoglobin. The concentration of myoglobin was sig-nificantly greater in the swimming muscles than in thenonswimming muscles in all three species of pinnipeds(paired samples t-test; P # 0.05). The values in theswimming muscles were significantly greater in theharbor seals than in the Northern fur seals and Stellersea lions (ANOVA, P # 0.01). The concentration ofmyoglobin in the muscles of the pinnipeds was 10- to22-fold greater than those measured in the vastusmedialis of the cotton rat (Table 3).

DISCUSSION

The Vv(fi,f) is inversely related to the oxidativecapacity of the skeletal muscles: the higher the oxida-tive capacity of the muscle, the lower the Vv(fi,f) (17,18). Our results for pinnipeds ranged from 76.2 to82.1%. They are similar to the value of ,80% reportedin the locomotory muscles of athletic terrestrial ani-mals (17). The Vv(mt,f) in pinniped swimming muscleswas 1.7- to 2.2-fold greater than the predicted valuesfor terrestrial mammals of comparable size but wassimilar to that in athletic dogs and ponies (Fig. 4).However, when one compares the volume density ofinterfibrillar mitochondria [Vv(mi,f)] in pinniped swim-ming muscles with that in the locomotory muscles ofathletic terrestrial animals of similar size, values in theswimming muscles were greater (Table 4). In pinni-peds’ muscles, ,7% of their total mitochondrial volumewas located near the sarcolemma. In contrast, ,28% ofthe total mitochondria in athletic terrestrial mammalsare subsarcolemmal (24). Subsarcolemmal mitochon-dria are thought to play an important role in reducingthe diffusion distance for blood-borne substrates andoxygen in terrestrial mammals (5). The increased

1249HIGH AEROBIC CAPACITIES IN PINNIPED SKELETAL MUSCLE

Vv(mi,f) and low volume density of subsarcolemmalmitochondria [Vv(ms,f)] in the skeletal muscles ofpinnipeds may reflect their reduced reliance on blood-borne oxygen and fuel substrates during diving due to

the reduced peripheral perfusion brought about bytheir natural dive response. It is also interesting to notethat a preferential increase in the Vv(mi,f) is also foundin muscles with increased activity under hypoxic condi-

Fig. 2. Representative electron micrographs of swimming muscles ofdifferent species. A: harbor seal; B: Steller sea lion; C: Northern furseal. M, mitochondria; L, lipid droplet; F, myofibril.

1250 HIGH AEROBIC CAPACITIES IN PINNIPED SKELETAL MUSCLE

tions (4, 16, 30), in contrast to training in normoxia inwhich a greater increase occurs in the Vv(ms,f) (30).

It has been found that Vv(mt,f) scales in directproportion to the capillary density in skeletal muscle(29). On the basis of this relationship, the swimmingmuscles should have capillary densities (800–1,000capillaries/mm2) comparable to those measured in ath-letic terrestrial mammals of similar size (5, 23). Capil-lary densities in the swimming muscles in pinnipedshave only been measured in a limited number ofsamples from harbor seals, grey seals, and Antarctic furseals (33). Reed et al. (33) found values ranging from352 to 639 capillaries/mm2 in harbor seals and Antarc-tic fur seals, respectively. These values are lower thanwould be predicted from the regression of Conley (5) ifthese animals had similar Vv(mt,f) to those of theharbor seals and Northern fur seals measured in thisstudy. Capillary density may not scale in direct propor-tion to the total density of mitochondria, becausepinnipeds reduce cardiac output and peripheral bloodflow during diving.

All of the harbor seals and Steller sea lions sampledin this study were adult animals, whereas the Northernfur seals were 2-yr-old juveniles. Previous research hasfound that there are no significant differences in hemo-globin concentrations, hematocrit, red blood cell concen-trations, or muscle myoglobin concentrations betweenelephant seals (Mirounga angustirostris) and Galapa-gos fur seals (Arctocephalus galapagoenis) older than 6mo and adult animals (19). However, the pinnipedsexamined in this study use different modes of underwa-ter propulsion and vary greatly in body mass (29–240kg). Harbor seals swim by using alternate lateralsweeps of their hind flippers, with thrust generated bythe muscles of the back (e.g., longissimus dorsi muscle;Ref. 33). In contrast, sea lions and fur seals use theirfore flippers as paddles and generate thrust with thechest (pectoralis muscle) and shoulder muscles (e.g.,

deltoideus muscle; Ref. 33). To correct the volumedensities of mitochondria for differences in metabolicrate due to body mass, the absolute volume of mitochon-dria was calculated for the swimming muscles (2, 18)and divided by the animal’s body mass0.75 (28) by usingthe following equation

5(M)(Ms)[Vv(mt,f)/Dmus]/M 0.756

where M represents body mass in kilograms, Ms repre-sents the percent of total body mass that is swimmingmuscle, and Dmus is the density of muscle. Despitedifferences in the mode of propulsion, there were nosignificant differences in the body-mass-adjusted mito-chondrial volume densities of swimming muscles amongharbor seals, fur seals, or sea lions (16.5 6 0.5, 13.1 60.8, and 15.5 6 1.1 ml of mitochondria/kg0.75, respec-tively). In other words, the mitochondrial volume den-sity in the propulsive muscles appeared to be indepen-dent of swimming mode.

The activities of oxidative enzymes were measuredas indicators of total aerobic capacity. As in terrestrialanimals, the enzyme activities were found to be signifi-cantly correlated to the Vv(mt,f) (Fig. 3). The valuesobtained in pinnipeds were higher than those previ-ously reported for the same species (3, 13, 33), whereasthe values for our control agreed well with previouslypublished values (33). The differences in the enzymeactivities between the present study and previousstudies may have resulted from a different collectionprocedure. We collected samples from live animals orfrom animals immediately after death (i.e., 10–15 minpostmortem) and placed them directly into liquid nitro-gen. The immediate freezing of the samples shouldhave prevented any degradation of the samples associ-ated with other freezing techniques. The activities ofaerobic enzymes such as CS normally scale inversely

Table 1. Summary data for the electron microscopy of swimming and nonswimming musclesof harbor seals, Steller sea lions, and Northern fur seals

SpeciesMass,

kg n Muscle

Volume Density, %

Vv(mt,f ) Vv(mi,f ) Vv(ms,f ) Vv(fi,f ) Vv(li,f )

Harbor seals (Phoca vitulina) 6165 10 Swimming (longissimus) 9.760.5‡ 9.160.3‡ 0.660.3 76.261.8‡ 0.1360.0710 Nonswimming (pectoralis) 5.760.4†‡ 5.360.3† 0.460.2 79.561.7 0.0860.06

Ratio of swimming to non-swimming 1.7 1.7 1.5 0.96 1.6

Steller sea lions (Eumetopiasjubatus) 240610 14 Swimming (pectoralis) 6.260.5 6.060.4 0.260.1 78.961.6 0.1060.05

13 Nonswimming (hindlimb) 3.160.4† 2.960.4† 0.260.1 81.561.1 0.0360.01Ratio of swimming to non-

swimming 2.0 2.1 1.0 0.96 3.3

Northern fur seals (Callorhinusursinus) 29* 5 Swimming (pectoralis) 8.860.5‡ 8.560.5‡ 0.360.2 81.462.0 1.060.20§

5 Nonswimming (hindlimb) 4.460.3†‡ 4.260.3† 0.260.2 82.162.4 0.260.06†Ratio of swimming to non-

swimming 2.0 2.0 1.5 0.99 3.0

Values are means 6 SE; n, no. of animals. Vv(mt,f ), volume density of total mitochondria; Vv(mi,f ), volume density of interfibrillarmitochondria; Vv(ms,f ), volume density of subsarcolemmal mitochondria; Vv (fi,f ), volume density of myofibrils; and Vv(li,f ), volume density ofintracellular lipid droplets; all quantities expressed per fiber volume. Soleus of the rat (Sprague-Dawley; n55) was used as control [Vv(mt,f )4.560.2] and agreed with previously published values (30). *Estimated mass from Baker et al. (1). †Significantly different from swimmingmuscle, P # 0.05. ‡Significantly different from Steller sea lion, P # 0.01. §Significantly different from Steller sea lion and harbor seal, P ,0.01.

1251HIGH AEROBIC CAPACITIES IN PINNIPED SKELETAL MUSCLE

Fig. 3. A: citrate synthase activity is significantly correlated to volume density of mitochondria between species(Pearson correlation, r 5 0.71; P , 0.001). Values for emperor penguins (not included in correlation analysis) arefrom Ponganis et al. (32). B: b-hydroxyacyl-CoA dehydrogenase (HAD) activity is significantly correlated to volumedensity of total mitochondria [Vv(mt,f)] between species (Pearson correlation, r 5 0.68; P , 0.001).

Table 2. Enzyme activities of citrate synthase (CS) and b-hydroxyacyl-CoA dehydrogenase (HAD) in swimmingand nonswimming muscles of harbor seals, Steller sea lions, and Northern fur seals

SpeciesMass,

kg n MuscleCS Activity,

IU/g wet wt muscleHAD,

IU/g wet wt muscleCS/HAD

Ratio

Harbor seals 6165 37 Swimming (longissimus dorsi) 37.061.7§ 51.962.4§ 0.7137 Nonswimming (pectoralis) 34.161.7§ 40.562.5†§ 0.84

Ratio of swimming to nonswimming 1.1 1.3

Steller sea lions 26068 27 Swimming (pectoralis) 21.961.2‡ 22.561.3‡§ 0.9727 Nonswimming (hindlimb) 15.061.0† 18.061.0† 0.83

Ratio of swimming to nonswimming 1.5 1.3

Northern fur seals 29* 5 Swimming (pectoralis) 39.362.9§ 47.567.0§ 0.835 Nonswimming (hindlimb) 28.362.8† 48.068.3§ 0.59

Ratio of swimming to nonswimming 1.4 1.0

Cotton rat (Sigmodon hispidus) 0.107 5 Vastus medialis 24.061.4 11.961.7 2.060.3

Values are expressed as means 6 SE. CS and HAD activities are expressed as µmol product formed·min21 ·g wet wt muscle21. *Estimatedbody mass from Baker et al. (1). †Significantly different from swimming muscles of seals (paired sample t-test, P#0.05). ‡Significantlydifferent from harbor seal and fur seal (ANOVA, P#0.01). §Significantly different from rat (ANOVA, P#0.01).

1252 HIGH AEROBIC CAPACITIES IN PINNIPED SKELETAL MUSCLE

with body mass (14) unless a species exhibits someadaptive deviation. The activities of CS in the swim-ming muscles were either similar (Steller sea lions) orsignificantly greater (harbor seals and fur seals) thanthose measured in the rat, whereas HAD activitieswere significantly greater in the muscles of all pin-nipeds compared with our control. Because the pin-nipeds were three to four orders of magnitude greaterin mass than the control animal, these results indicatean adaptive increase in aerobic enzyme capacity in theswimming muscles compared with locomotory musclesfrom terrestrial mammals.

The correlations between the oxidative enzymes andthe volume density of mitochondria (Fig. 3) are higherthan those found in terrestrial mammals (20). Thismight indicate a difference in the mitochondria be-tween terrestrial and marine mammal muscle. This

would contrast with the idea that all mitochondria arethe same regardless of the muscle type or species (35).However, recent studies have found differences in therespiratory rate of mitochondria isolated from differentspecies [i.e., hummingbirds (Selasphorus rufus); Ref.37] or fiber types (21). In addition, data from theswimming muscle of another diving animal, the em-peror penguin (Aptenodytes forsteri), show an evengreater CS activity for the volume density of mitochon-dria than was found in pinnipeds (Fig. 3) (32). Thissuggests that there may be differences between mito-chondria from the muscles of diving species as com-pared with the muscles of terrestrial animals.

In the skeletal muscles of pinnipeds, the CS/HADratios ranged from 0.59 to 0.97 and were substantiallyless than those observed in the vastus medialis of therat (2.0 6 0.3). CS/HAD ratios of 1.25 to 2.6 have beenreported in the gluteal muscle, a major locomotorymuscle in horses and steers (22). The ratios measuredin this study agree with previously published values forthe longissimus muscle of the harbor seals (33) andindicate that essentially all of the aerobic metabolismof pinniped skeletal muscles could be supported by theoxidation of fatty acids. These findings agree withearlier studies that found respiratory quotients ofpinnipeds to be ,0.74, thus indicating that fat was theprimary fuel source during exercise (6).

Available data on fiber types in pinniped swimmingmuscles indicate that they consist of a mixed composi-tion of fiber types: slow oxidative, fast oxidative glyco-lytic, and fast glycolytic fibers (33). Similarly, we founda fiber type population distribution of 20% slow oxida-tive, 27% fast oxidative glycolytic, and 53% fast glyco-lytic in a single Steller sea lion (S. B. Kanatous,unpublished observations). Reed et al. (33) found thatharbor seals have a higher oxidative capacity andgreater proportion of slow oxidative fibers than doAntarctic fur seals. However, our data indicated thatthe swimming muscles of harbor seals and Northernfur seals had a greater aerobic capacity than did Stellersea lions, on the basis of the volume density of mitochon-dria and oxidative enzyme activities.

Fig. 4. Percent Vv(mt,f) of locomotory muscles in relation to bodymass. Regression line (y 5 6.75 2 1.34 [log x], r2 5 0.70) showsrelationship for vastus medialis, a primary locomotory muscle, for awide variety of terrestrial mammals ranging in size from 0.04 kg(dwarf mongoose, Helogale pervula) to 450 kg (steer, Bos taurus) (24,29). Dashed lines, 99% confidence intervals for the regression. s,Pinnipeds in this study; r, terrestrial animal athletes (17). %Volumedensities of mitochondria in swimming muscles of sea lions, fur seals,and seals were 1.7, 1.9, and 2.1 times greater, respectively, thanvalues predicted on the basis of their body size. %Vv(mt,f) in vastusmedialis of dog (Canis familiaris, 10.7%) and pony (Equus caballus,6.8%) were 2.2 and 1.8 times greater, respectively, than predictedvalues and were similar to those of comparably sized pinnipeds.

Table 3. Concentration of myoglobin in swimming and nonswimming musclesof harbor seals, Steller sea lions, and Northern fur seals

Species nMass,

kg MuscleMyoglobin Concentration,

mg/g wet wt muscle

Harbor seals 37 6165 Swimming (longissimus dorsi) 37.461.7§37 Nonswimming (pectoralis) 24.160.8†§

Ratio of swimming to nonswimming 1.6

Steller sea lions 27 260610 Swimming (pectoralis) 28.761.5§27 Nonswimming (hindlimb) 20.061.3†§

Ratio of swimming to nonswimming 1.4

Northern fur seals 5 29* Swimming (pectoralis) 22.462.5§5 Nonswimming (hindlimb) 16.060.9§

Ratio of swimming to nonswimming 1.4

Cotton rat 5 0.107 Vastus medialis 1.760.2

Values are means 6 SE. *Estimated body mass from Baker et al. (1). †Significantly different from swimming muscles in seals (pairedsample t-test, P#0.05). §Significantly different from rat, P # 0.01.

1253HIGH AEROBIC CAPACITIES IN PINNIPED SKELETAL MUSCLE

In terrestrial mammals, intracellular stores of oxy-gen are small, and the transport of oxygen from thecapillaries to the mitochondria is critical to sustainoxidative phosphorylation during exercise. In contrast,intracellular stores of oxygen are substantially largerin marine mammals and provide an important sourceof oxygen for the mitochondria, as convective oxygentransport decreases during the dive (33). In addition,myoglobin plays an important role, not only as anoxygen source but also in the facilitated diffusion ofoxygen throughout the muscle, especially under hyp-oxic conditions (10, 33, 39). The myoglobin concentra-tions measured in this study agree with those previ-ously reported for harbor seals and fur seals; there areno published myoglobin values for Steller sea lions (3,33). Like high-altitude-adapted mammals, the pin-nipeds in this study have enhanced concentrations ofmyoglobin in their skeletal muscles compared withsea-level-adapted terrestrial mammals. The combina-tion of an increased myoglobin concentration and vol-ume density of mitochondria will have the apparenteffect of increasing the intracellular diffusing capacityof the skeletal muscles (26).

The Vv(li,f) per volume of muscle fiber in the pin-nipeds was generally ,1%. The values obtained in theswimming muscles of harbor seals and Steller sea lions(0.13 6 0.07 and 0.1 6 0.05%, respectively) wereapproximately an order of magnitude greater than inthe vastus medialis of comparably sized dogs andponies (0.03 6 0.02 and 0.01 6 0.01%, respectively)(17). The mean value in fur seals was ,37 times greaterthan for a comparably size dog. Unlike the findings infur seals, there were no differences in the Vv(li,f)between swimming and nonswimming muscles in har-bor seals and Steller sea lions. Typically, locomotorymuscles in terrestrial mammals have a greater store of

energy than do nonlocomotory muscles (24). However,previous studies have found in the muscles of free-ranging mammals considerable variation in the Vv(li,f)that was not related to either body size or behavior (29).

Studies of harbor seals led to the hypothesis thatendogenous stores of triglycerides may play an impor-tant role in fuel homeostasis, especially during diving(6). In terrestrial mammals, intramuscular stores oftriglycerides act as a fuel reservoir when the rate ofdelivery of plasma free fatty acids is lower than the rateof utilization in the muscles, especially during submaxi-mal exercise (36). Although the Vv(li,f) values in theskeletal muscles of Steller sea lions and harbor sealswere an order of magnitude greater than in comparablysized terrestrial mammals, the level of Vv(li,f) wassignificantly less than in the swimming muscles of thejuvenile Northern fur seals. In the Steller sea lions, thislevel may have resulted from samples being takenduring the postpartum fast and lactation. Fasting hasbeen found to decrease the content of intramusculartriglycerides in terrestrial mammals (36). The stress ofparturition and fasting may have diminished the lipiddroplet densities to levels less than would normally beseen in actively feeding animals, such as the juvenileNorthern fur seals. However, the relatively low levels oflipid droplets observed in the harbor seals comparedwith the Northern fur seals were more difficult tounderstand. Samples were collected from both femalesand males, during the spring, early summer and fall,when it was believed the animals were actively feeding.However, the population of harbor seals in PrinceWilliam Sound, AK, has been declining since 1984 (9).This decline appears to be a non-density-dependentresponse, possibly attributed to a decreased carryingcapacity of the environment due to food limitations (9).This may be a cause of the Vv(li,f) in the skeletal

Table 4. Comparison of volume density of mitochondria in the locomotorymuscles of pinnipeds and terrestrial animal athletes

Species nMass,

kg Muscle Vv(mt,f ) Vv(mi,f )

Harbor seals 10 61 Longissimus 9.760.5 9.160.2Dog (Canus familiaris) 3 28 Vastus medialis 10.7† 7.3†

Ratio of seals to dogs 0.91 1.25

Stellar sea lion 14 240 Pectoralis 6.260.5 6.060.4Pony (Equus caballus) 3 175 Vastus medialis 6.8† 5.2†

Ratio of sea lions to ponies 0.91 1.2

Fur seals 5 29* Pectoralis 8.860.5 8.560.5Dog 3 28 Vastus medialis 10.7† 7.3†

Ratio of fur seals to dogs 0.82 1.2

Values are means 6 SE; n, no. of animals. *Estimated body mass from Baker et al. (1). †Values from Hoppeler et al. (17).

Table 5. Lipid stores in swimming muscles of harbor seals, Steller sea lions, and Northern fur seals

SpeciesMass,

kg n Muscle Vv(li,f )Energy fromLipid, kJ/kg

Harbor seals 6165 37 Swimming 0.260.1 64Steller sea lions 260610 28 Swimming 0.160.0 32Northern fur seals 29* 5 Swimming 1.160.3 321

Values are means 6 SE; n, no. of animals. Values for Vv(li,f ) are from Table 1. *Estimated body mass from Baker et al. (1).

1254 HIGH AEROBIC CAPACITIES IN PINNIPED SKELETAL MUSCLE

muscles of the harbor seals compared with levels in thefur seals.

How much energy is stored in the form of triglycer-ides in the skeletal muscles? The mass of triglyceridesper kilogram of swimming muscle was estimated fromthe Vv(li,f) and then converted into kilojoules of energyper kilogram of swimming muscle (Table 5). The energywithin the lipid droplets was estimated with the follow-ing assumptions: the lipid droplets consisted entirely oftriglycerides, 94% of the triglycerides mass was fattyacids, and the density of the lipids was 0.92 (deter-mined from cod liver oil). Based on these calculations,there were 321, 44, and 32 kJ of energy from lipid perkilogram swimming muscle in Northern fur seals,harbor seals, and Steller sea lions, respectively. If weassume a resting muscle metabolic rate similar to thosemeasured in humans [,1.6 ml O2·min21 ·kg21 (33)],these endogenous stores of triglycerides would supplyenough energy for 171, 22, and 17 h in the fur seals,harbor seals, and Steller sea lions at rest, respectively.Considering the low CS/HAD ratios, these resultssupport the hypothesis that endogenous triglyceridestores in the muscles are an important source of energyin marine mammals.

Because the aerobic scope of an organism is propor-tional to the volume density of mitochondria (17, 24), anapparent paradox is the low aerobic scope of pinnipedsrelative to the elevated mitochondrial volume densitiesin their swimming muscles. The maximum aerobicscope measured for pinnipeds swimming in a waterflume is approximately ninefold their resting metabolicrate (6), which is less than the 30-fold aerobic scopereported for dogs and horses with similar mitochon-drial volume densities (17). The lower aerobic scope ofpinnipeds may result from the smaller percentage oftotal muscle mass used for swimming compared withthat used for terrestrial locomotion (2). In addition, thedive response, which decreases convective oxygen trans-port to muscles, may limit the aerobic scope to levelsless than the observed maximum.

The effects of hypoxia on muscle structure and func-tion remain controversial. Studies of humans exposedto chronic hypoxia show significant reductions in musclemass, volume densities of mitochondria, and oxidativecapacity (4, 16, 20). In contrast, studies of birds andmammals active at high altitude, including humanstrained under discontinuous hypoxia, display a vastlydifferent response. Under these conditions, skeletalmuscles show an increase in the volume density ofmitochondria, with larger increases in interfibrillarthan subsarcolemmal mitochondria, enhanced aerobicenzyme activities, and increased concentrations of myo-globin compared with sea level controls (8, 15, 16, 27,30, 38). The results of this study are consistent with thefindings of the second group of studies and suggest thatthe increased aerobic capacity in the skeletal musclesof pinnipeds is not an adaptation to hypoxia alone butan adaptation to activity under hypoxic conditions.

In conclusion, the swimming muscles of pinnipedsappear to be adapted for aerobic lipid metabolism evenunder the hypoxic conditions that occur during diving.

Our results are consistent with the hypothesis thatincreased mitochondrial volume densities, aerobic en-zyme capacities, and myoglobin concentrations areadaptations that facilitate aerobic metabolism underhypoxic conditions. The high aerobic capacity of pin-niped swimming muscles, as indicated by the highoxidative enzyme activities, volume densities of mito-chondria, and myoglobin concentrations, may have lessto do with increasing aerobic scope and more to do withmaintaining oxygen flux through the respiratory path-way under hypoxic conditions. Because the mitochon-drial uptake of oxygen under diving conditions is likelyto be diffusion limited, the increased mitochondrialdensities may serve to increase the oxygen sink anddecrease the diffusional distance from extra- and intra-cellular oxygen stores. These results also indicate thatthe overall aerobic capacity of skeletal muscle is en-hanced to a similar degree in mammals adapted toeither low oxygen availability (high-altitude and ma-rine mammals) or high oxygen demand (athletic terres-trial mammals).

We thank the following for their support: Ukeavik Inupiat Corpo-ration, Northern Arctic Research Laboratory, Arctic Research Foun-dation, J. Wen, V. Han, D. Calkins, K. Frost, L. Llowry, T. Loughlin, T.Spraker, D. Bradley, G. Antonelis, O. Mathieu-Costello, A. E. Bull, E.A. Brandon, T. C. Adams, K. D. Dudzinski, C. Hice, J. L. Respess, andT. D. Sparks. We also thank M. A. Castellini, W. E. Evans, M.Horning, O. Mathieu-Costello, C. Ribic, T. M. Williams, G. A. J.Worthy, and B. Wursig for their reviews and help with this manu-script. Special thanks are extended to Dr. G. A. Brooks and twoanonymous reviewers whose comments vastly improved the manu-script.

This work was supported by funds from the Alaska Department ofFish and Game, National Marine Mammal Lab, National MarineFisheries Service, and Texas A & M University.

Address for reprint requests and other correspondence: S. B.Kanatous, Dept. of Medicine 0623A, Univ. of California at San Diego,La Jolla, CA 92092-0623A (E-mail: skanatou@ucsd.edu).

Received 11 May 1998; accepted in final form 2 December 1998.

REFERENCES

1. Baker, J. D., C. W. Fowler, and G. A. Antonelis. Mass changesin fasting male Northern fur seals. Can. J. Zool. 72: 326–329,1994.

2. Bryden, M. M. Relative growth of the major body components ofthe southern elephant seal, Mirounga leonina. Aust. J. Zool. 17:153–177, 1969.

3. Castellini, M. A., G. N. Somero, and G. L. Kooyman. Glyco-lytic enzyme activities in tissues of marine and terrestrialmammals. Physiol. Zool. 54: 242–252, 1981.

4. Cerretelli, P., and H. Hoppeler. Morphologic and metabolicresponse to chronic hypoxia: the muscle system. In: Handbook ofPhysiology. Environmental Physiology, Bethesda, MD: Am.Physiol. Soc. 1996, sect. 4, vol. II, chapt. 50, p. 1155–1181.

5. Conley, K. E. Cellular energetics during exercise. In: Compara-tive Vertebrate Exercise Physiology, edited by J. H. Jones, NewYork: Academic, 1994, p. 1–39.

6. Davis, R. W., M. A. Castellini, T. M. Williams, and G. L.Kooyman. Fuel homeostasis in the harbor seal during sub-merged swimming. J. Comp. Physiol. [B] 160: 627–635, 1991.

7. Desplanches, D., H. Hoppeler, L. Tuscher, M. H. Mayet, H.Speilvogel, G. Ferretti, B. Kayser, M. Leuenberger, A.Grunenfelder, and R. Favier. Muscle tissue adaptations ofhigh-altitude natives to training in chronic hypoxia or acutenormoxia. J. Appl. Physiol. 81: 1946–1951, 1996.

8. Elsner, R., and B. Gooden. Diving and Asphyxia. New York:Cambridge Univ. Press, 1983.

9. Frost, K. J., L. F. Llowry, R. J. Small, and S. J. Iverson.Monitoring, Habitat Use and Trophic Interactions of Harbor

1255HIGH AEROBIC CAPACITIES IN PINNIPED SKELETAL MUSCLE

Seals in Prince William Sound, Alaska. Fairbanks, AK: AlaskaDepartment of Fish and Game, 1996. (Restoration Study No.95064)

10. Guyton, G. P., K. S. Stanek, R. C. Schneider, P. W.Hochachka, W. E. Hurford, D. G. Zapol, G. C. Liggins, andW. M. Zapol. Myoglobin saturation in free-diving Weddell seals.J. Appl. Physiol. 79: 1148–1155, 1995.

11. Heath, R. B., D. Calkins, D. McAllister, W. Taylor, and T.Spraker. Telezol and isoflurane field anesthesia in free-rangingSteller’s sea lions. J. Zoo Wildlife Med. 27: 35–43, 1996.

12. Hochachka, P. W. Balancing conflicting metabolic demands ofexercise and diving. Federation Proc. 45: 2948–2952, 1986.

13. Hochachka, P. W. Metabolic biochemistry of a mesopelagicmammal. Experientia 48: 570–574, 1992.

14. Hochachka, P. W., B. Emmett, and R. K. Suarez. Limits andconstraints in the scaling of oxidative and glycolytic enzymes inhomeotherms. Can. J. Zool. 66: 1128–1138, 1987.

15. Hochachka, P. W., C. Stanley, J. Merkt, and J. Sumar-Kalinowski. Metabolic meaning of elevated levels of oxidativeenzymes in high altitude adapted animals: an interpretivehypothesis. Respir. Physiol. 52: 303–313, 1982.

16. Hoppeler, H., and D. Desplanches. Muscle structural modifi-cations in hypoxia. Int. J. Sports Med. 13, Suppl. 1: S166–S168,1992.

17. Hoppeler, H., S. R. Kayar, H. Claassen, E. Uhlmann, andR. H. Karas. Adaptive variation in the mammalian respiratorysystem in relation to energetic demand. III. Skeletal muscles setthe demand for oxygen. Respir. Physiol. 69: 27–46, 1987.

18. Hoppeler, H., O. Mathieu, R. Krauer, H. Classen, R. B.Armstrong, and E. R. Weibel. Design of the mammalianrespiratory system. VI. Distribution of mitochondria and capillar-ies in various muscles. Respir. Physiol. 44: 87–111, 1981.

19. Horning, M., and F. Trillmich. Development of hemoglobin,hematocrit and erythrocyte values in Galapagos fur seals. Mar.Mamm. Sci. 13: 100–113, 1997.

20. Howald, H., D. Pette, J. A. Simoneau, A. Uber, H. Hoppeler,and P. Cerretelli. Effects of chronic hypoxia on muscle enzymeactivities. Int. J. Sports Med. 11, Suppl. 1: S10–S14, 1990.

20a.Howell, A. B. Contribution to the comparative anatomy of theeared and earless seals (genera Zalophus and Phoca). Proc. USNatl. Mus. 73: 1–142, 1928.

21. Jackman, M. R., and W. T. Willis. Characteristics of mitochon-dria isolated from type I and type IIb skeletal muscle. Am. J.Physiol. 270 (Cell Physiol. 39): C673–C678, 1996.

22. Karlstrom, K., B. Essen-Gustavsson, and A. Lindholm.Fiber type distribution and enzymatic profile of locomotor andnonlocomotor muscles of horses and steers. Acta Anat. (Basel)151: 97–106, 1994.

23. Kayar, S. R., H. Hoppeler, R. B. Armstrong, M. H. Laughlin,S. L. Lindstedt, J. H. Jones, K. R. Conley, and C. R. Taylor.Estimating transit time for capillary blood in selected muscles ofexercising animals. Pflugers Arch. 421: 578–584, 1992.

24. Kayar, S. R., H. Hoppeler, S. L. Lindstedt, H. Classen, J. H.Jones, B. Essen-Gustavsson, and C. R. Taylor. Total mitochon-

drial volume in relation to aerobic capacity of horses and steers.Pflugers Arch. 413: 343–347, 1989.

25. Kooyman, G. L., M. A. Castellini, R. W. Davis, and R. A.Maue. Aerobic diving limits of immature Weddell seals. J. Comp.Physiol. [B] 151: 171–174, 1983.

26. Lenfant, C., and K. Sullivan. Adaptation to high altitude. N.Engl. J. Med. 284: 1298–1309, 1971.

27. Leon-Velarde, F., J. Sanchez, A. X. Bigard, A. Brunet, C.Lesty, and C. Monge-C. High altitude tissue adaptation inAndean coots: capillarity, fiber type and enzymatic activities ofskeletal muscles. J. Comp. Physiol. [B] 163: 52–58, 1993.

28. Lindstedt, S. L., and R. G. Thomas. Exercise performance ofmammals: an allometric perspective, in comparative vertebrateexercise physiology. In: Unifying Physiological Principles, editedby J. H. Jones. New York: Academic, 1994, p. 191–217.

29. Mathieu, O., R. Krauer, H. Hoppeler, P. Gehr, S. L. Lind-stedt, R. M. Alexander, C. R. Taylor, and E. R. Weibel.Design of the mammalian respiratory system. VII. Scalingmitochondrial volume in skeletal muscle to body mass. Respir.Physiol. 44: 113–128, 1981.

30. Mathieu-Costello, O., P. J. Agey, L. Wu, J. M. Szewczak, andR. E. MacMillen. Increased fiber capillarization in flight muscleof finch at altitude. Respir. Physiol. 111: 189–199, 1998.

31. Mathieu-Costello, O., J. M. Szewczak, R. B. Logemann, andP. J. Agey. Geometry of blood-tissue exchange with bat hindlimband rat soleus muscle. Am. J. Physiol. 262 (Regulatory Integra-tive Comp. Physiol. 31): R955–R965, 1992.

32. Ponganis, P. J., M. L. Costello, L. N. Starke, O. Mathieu-Costello, and G. L. Kooyman. Structural and biochemicalcharacteristics of locomotory muscles of emperor penguins, Ap-tenodytes forsteri. Respir. Physiol. 109: 73–80, 1997.

33. Reed, J. Z., P. J. Butler, and M. A. Fedak. The metaboliccharacteristics of the locomotory muscles of grey seals (Halicho-erus grypus), harbour seal (Phoca vitulina), and Antarctic furseals (Arctocephalus gazella). J. Exp. Biol. 194: 33–46, 1994.

34. Reynafarje, B. Method for the determination of myoglobin. J.Lab. Clin. Med. 61: 138–145, 1963.

35. Schwerzmann, K, H. Hoppeler, S. R. Kayar, and E. R.Weibel. Oxidative capacity of muscle and mitochondria: correla-tion of physiological, biochemical, and morphometric characteris-tics. Proc. Natl. Acad. Sci. USA 86: 1583–1587, 1989.

36. Staron, R. S., R. S. Hikida, T. F. Murray, F. C. Hagerman,and M. T. Hagerman. Lipid depletion and repletion in skeletalmuscle following a marathon. J. Neurol. Sci. 94: 29–40, 1989.

37. Suarez, R. K., J. R. B. Lighton, G. S. Brown, and O.Mathieu-Costello. Mitochondrial respiration in hummingbirdflight muscles. Proc. Natl. Acad. Sci. USA 88: 4870–4873, 1991.

38. Terrados, T., E. Jansson, C. Sylven, and L. Kaijser. Ishypoxia a stimulus for synthesis of oxidative enzymes andmyoglobin? J. Appl. Physiol. 68: 2369–2372, 1990.

39. Wittenberg, B. A. and J. B. Wittenberg. Transport of oxygenin muscle. Annu. Rev. Physiol. 51: 857–878, 1989.

1256 HIGH AEROBIC CAPACITIES IN PINNIPED SKELETAL MUSCLE