Fine structure and metabolism of multiply innervated fast ... - synergy

Cell

Cell Tissue Res (1981) 219:93-109

and

Tissue

Research @

Springer-Verlag

1981

Fine structure and metabolism of multiply innervated fast muscle fibres in teleost fish Ian A. Johnston

and Thomas

W. Moon *

Department of Physiology, University ofSt. Andrews, St. Andrews, Fife, Scotland, Great Britain

Summary. Both the fast and slow muscle fibres of advanced teleost fish are multiply innervated. The fraction of slow-fibre volume occupied by mitochondria is 31.3%,25.5% and 24.6%, respectively, for the myotomal muscles of brook trout (Salvelinusfontinalis), crucian carp ( Carassius carassius), and plaice (Pleuronectes platessa), respectively. The corresponding figures for the fast muscles of these species are 9.3 %, 4.6% and 2.0 %, respectively. Cytochrome-oxidase and citrate-synthetase activities in the fast muscles of 9 species of teleost range from 0.20-0.93 ~moles substrate utilised, 9 wet weight muscle-1 min-1 (at 15° C) or around 4-17% of that of the corresponding slow fibres. Ultrastructural analyses reveal a marked heterogeneity within the fastfibre population. For example, the fraction of fibres with < 1% or > 10% mitochondria is 0,4,42% and 36,12 and 0%, respectively, for trout, carp and plaice. In general, small fibres ( < 500 ~m2) have the highest and large fibres ( > 1,500 ~m2) the lowest mitochondrial densities. The complexity ofmitochondrial cristae is reduced in fast compared to slow fibres. Hexokinase activities range from 0.4-2.5 in slow and from 0.08-0.7 ~moles, g wet weight- I min-1 in fast muscles, indicating a wide variation in their capacity for aerobic glucose utilisation. Phosphofructokinase activities are 1.2 to 3.6 times higher in fast than slow muscles indicating a greater glycolytic potential. Lactate dehydrogenase activities are not correlated with either the predicted anaerobic scopes for activity or the anoxic tolerances of the species studied. The results indicate a considerable variation in the aerobic capacities and principal fuels supporting activity among the fast muscles of different species. Brook trout and crucian carp are known to recruit fast fibres at low swimming speeds. For these species the aerobic potential of the fast muscle is probably sufficient to meet the energy requirements of slow swimming. Send offprint requests to: Ian A. Johnston, Department

of Physiology, University

Andrews, Fife, Scotland, KY169TS, Great Britain * Permanent address: Department of Biology, University

of Ottawa,

of St. Andrews, St.

Ottawa, ' Ontario,

Canada

K1N9B4

0302-766X/81/0219/0093/$03.40

94

.A.

Johnston

and

T.W.

Moon

Key words: Quantitative cytology -Fish muscle -Muscle metabolism -Muscle ultrastructure -Teleosts In all fishes fast fibres comprise between 70-100 % of the trunk musculature (see Greer-Walker andPuI11975). The preponderance of fast fibre types is a reflection of the power requirements of aquatic locomotion [ ""' power cxvelocity3] (Webb 1975). Thus small increments in performance at high speed require the recruitment of increasing numbers of fast motor units. The dogfish (Scyilorhinus cannicula), in common with other elasmobranchs, has fast fibres with single motor endplates (Bone 1964). In this species the fast motor system is reserved for butst locomotory activity (Bone 1966). Fast-muscle in sharks is pure white in colour, poorly vascularised and mitochondria usually occupy less than 1% of fibre volume (Kryvi 1977; Totland et al. 1980). It is likely that these fibres are entirely dependent on anaerobic glycogenolysis for energy (Bone 1978a). Exhaustion in dogfish occurs after only 1-2min of vigorous swimming and coincides with the depletion of glycogen in fast muscles (Bone 1966). In contrast to the elasmobranchs, fast muscle fibres in most teleosts are multiply innervated as are the slow fibres of fish, amphibia, birds and some extra-occular muscles in mammals (Barets 1961; Bone 1964; Hesse 1970). Typically each fibre receives a dense network of innervation with numerous endplates, often derived from more than one motor axon (Barets 1961;'Bone 1964,1970). For example, fast fibres in the scorpeaniform fish Myoxocephalus scorpius have up to 22 endplates each ofwhich is probably derived from a separate axon (Hudson 1969). This type of fast muscle innervation differs from that of all other vertebrates including chondrosteans, elasmobranchs and certain primitive teleost groups (e.g., gonorythichiformes, clupeiformes, anguilliformes) all of which have single motor endplates (Bone 1964, 1970). Electromyographical studies have shown that fast fibres with multiple innervation are recruited at both low-sustainable and burst-swimming speeds (see Bone 1978a; Johnston 1981a, b). For example, the threshold speed for recruitment of fast fibres is only 0.5-2.1 bodylength S-l in carp (Johnston et al. 1977; Bone et al. 1978),0.8-1.9 bodylength S-l in saithe (Johnston and Moon 1980a) and 0.5-1.3 bodylength S-l in rainbow trout (Hudson 1973; Bone et al. 1978). Little is known about the metabolism and energy supply of fast muscles in teleosts at low swimming speeds. However, it seems likely that the metabolism of these fibres may differ from the anaerobic pattern characteristic of elasmobranchs. The present study investigates both the fine structure and energy metabolism of muscles from teleost species with a range of different swimming behaviours and activity levels. Materials

and methods

Fish Marine fish were obtained from the Firth of Forth, Scotland during April 1980. They were used for experiments the same day as capture. Freshwater fish were supplied by local fish farms and maintained at 8-10° C for up to a week prior to sacrifice. Details of the species of fish used and their size ranges are given in Table 1. All fish were killed by a sharp blow to the head and transsection of the spinal cord.

Structure and metabolism of fish muscle Table I. Lengths and weights of species used in th,ese investigations Common name

Scientific name

Habitat

Length (cm)

Weight (g)

Brook trout Crucian carp Tench

Salvelinus fontinalis (Mitchil/) Carassius carassius L. Tinca tinca L. Merlangius merlangus L. Gadus morhua L. Platichthys flesus L. Microstomus kitt L. Pleuronectes platessa L. Limanda limanda L.

Freshwater rivers Freshwater lakes Freshwater lakes Marine Marine Marine/Esturine Marine Marine Marine

18.2:t 0.3 17.9:t0.9 16.9:t 0.5 32.9:t 1.3 44.4:t 2.7 29.5:t 0.8 27.2:t 1.3 24.8:t 2.1 24.2:t 0.7

59.2:: 95.8: : 54.7.: 316.7.: 881.4.: 307.5:: 223.0.: 165.9.: 145.7.:

Whiting Cod Flounder Lemon sole Plaice Dab

Ultrastructural

3.1 10.7 7.5 54.3 147.0 45.0 25.5 17.2 14.1

studies

Muscle samples were initially fixed in situ for 1 h by injecting 3% glutaraldehyde, 0.15 M phosphate buffer pH 7.4 into the posterior third of the dorsal myotomal muscle. During this time fish were kept on ice. In the species studied different types of fibre are anatomically separated (see Bone 1978a). Small bundles of fast and slow fibres can be easily dissected, free from contamination, using a binocular microscope. Samples of fast muscle were dissected at random from all depths within the body. Fibre bundles were held at their resting length by pinning to cork strips. Fixation was continued for a further 2-24h at 4° C in 3% glutaraldehyde, 0.15 M phosphate buffer pH 7.4. Tissue samples were post-fixed in 1% osmium tetroxide in 0.1 M phosphate pH 7.4, dehydrated in a series of alcohols up to 100% and embedded in araldite resin. Ultrathin sections were cut on a Reichart OM U2 ultramicrotome and double stained with uranyl acetate and lead citrate. Sections were examined with a Phillips 301 electron microscope. Orientation of muscle fibres in embedded material was ascertained from examination of 1 ~m sections stained with either toluidine blue or p-phenylene diamine in 1:1 isopropanol:methanol (Hollander

and Vaaland 1968).

Morphometric

methods

Total cross-sectional areas of fibres were determined by tracing outlines from electron micrographs ( x 4,940), using a summagraphics digitiser in conjunction with a mini-computer (Walesby and Johnston 1980). Quantitative analyses of electron micrographs (magnification 7,000 to 15,000 x) were carried out using a point-grid method (WeibeI1969). Good agreement was found between the stereological methods of Weibel and direct estimates of cell component fractional volumes using the digitiser and minicomputer. The fraction of total fibre volume occupied by mitochondria ( %) (MF) was determined for 50 fast- and 50 slow-muscle fibres from brook trout, crucian carp and plaice. Enzyme assays Superficial red fibers were rapidly dissected from both sides of the entire length of the body. Whitemuscle samples of about 2 9 were dissected from the dorsal third of the trunk musculature. Muscle was minced with scissors and homogenized at 00 C with an Ilado-X10 homogeniser (I.C.A. GmbH, Dottinger, W. Germany) for three periods of 25 s with cooling in 5-8 vols of 50 mM Tris-HCI 5 mM EDTA, 2mM Mg C12, 1mM dithiothreitol pH7.5. Homogenates were centrifuged at 600 g for 20 min and filtered through glass wool. Enzyme activities were determined in the supernatant and expressed in terms of ~oles substrate utilised per g dry weight muscle, min- 1. Measurements of enzyme activity were performed at 15° C with appropriate controls (usually substrate deletion). Concentrations of substrates and co-ions, and pH were established to give conditions for measurements of maximal enzyme activities on the basis of preliminary experiments with trout, carp and plaice. Assay procedures for the individual enzymes were as follows: Citrate synthetase (CS) Citrate synthetase was assayed in a medium of 100mM Tris-HCI, 0.5mM oxaloacetic acid, 0.3mM A""tvl r"A () 1 m M 5-51-rlit.hinhi,;-2-nit.robenzoic in 40 mM ohosohate. oH 8.0. The reaction was

96

A.

Johnston

and

T.W.

Moon

started by addition of oxaloacetic acid and the increase in extinction at 412 nm wavelength monitored. Enzyme activity was calculated using EmM = 13.6. Cytochrome oxidase (CO ) Cytochrome oxidase activity was assayed by following the oxidation of reduced cytochrome C in 50 mM phosphate buffer pH7.6 at 550nm. Enzyme activity was calculated using EmM (red-ox) = 19.1. Hexokinase ( H K) Hexokinase activity was assayed using anA TP-regenerating system in a medium containing 85 mM TrisHCI pH 7.5, 8mM MgCI2 0.8mM EDTA, lmM glucose, 2.5mM ATP, 0.4mM NADP, 10mM phosphoryl creatine, 100 I1g creatine phosphokinase, and 100 I1g glucose-6-phosphate dehydrogenase. Control assays contained the ~bove medium with glucose omitted. Lactate dehydrogenase (LDH) Lactate-dehydrogenase activity was assayed spectrophotometrically buffer pH 7.5, 1 mM sodium pyruvate and 0.27 mM NADH. Phosphofructokinase

in a medium of 50 mM phosphate

( P F K)

Phosphofructokinase activity was assayed spectrophotometrically in a medium of 50 mM Tris-HCl pH7.4, 4.5mM Fructose-6-phosphate, 3mMATP, 0.2mMAMP25mM KC!, 6mM MgCl2, 0.15mM NADH and excess alsolase, triose phosphate isomerase and glycerolpho~phate dehydrogenase. Statistical analyses Results were compared using a Student's t-test. Abbreviations; M F Fraction of total muscle fibre volume occupied by mitochondria ( %); CS Citrate synthetase; CO Cytochrome oxidase; H K Hexokinase; LDH Lactate dehydrogenase; p F K Phosphofructokinase

Results Enzyme activity profiles Enzyme activities of slow and fast muscles are shown in Tables 2 and 3 respectively. Based on CS and CO activities the aerobic capacities of the slow muscles studied would appear to be broadly similar. The ratio of the activities of these two enzymes in fast: slow muscles therefore allow a crude assessment of the relative aerobic Table 2. Activities

of some enzymes of energy metabolism in fish slow muscle fibres

Species

No. of

Brook trout

8

Crucialicarp Tench

~ 6 6 5 6 5 8 6

Whiting Flounder Cod Lemon Sole Plaice Dab

Hexokinase

Citrate synthetase

fish

0.3:t 0.03 2.0:t 0.2 0.6:t 0.1 2.2:t 0.1 1.5:t0.4 2.5:t 0.5 0.9:t 0.2 0.4:t 0.06 1.7:t0.3

4.9 to.1 9.2 t 1.3 8.3 to.9 4.8 to.8 7.8 to.9 6.6 t 1.7 5.2 to.6 7.1 to.5 8.7 to.7

Cytochrome Phosphooxidase fructokinase 2.3:t

0.2

S.6:t 0.1 3.2:t

0.4

6.1 :t 0.2 4.1:t0.3 7.1 :t 0.4 S.S:t 0.4 3.9:t

0.2

2.1::1: 0.2

11.4:!:2.4 1.9 :!:0.5 3.6 :!:0.7 2.0 :!:0.4 12.2:!:3.2 4.2 :!:0.5 6.3 :!:0.6 17.1:!:2.2 17.5t 1.1

Lactate dehydrogenase 200:t17 440:t 33 509:t 60 234:t 8 192:t 26 253:t 20 246:t 14 174:t 35 124::!: 34

Enzyme activities are expressed as I.lmoles substrate utilised, 9 wet weight I1iUSCle-l min- 1 at 15° C

~tructure and metaboJi~m of fish muscle

97

Table 3. Activities

of some enzymes of energy metabolism in fast-muscle fibres

Species

No. of

Brook .trout Crucian carp Tench Whiting Flounder Cod Lemon Sole Plaice nab

8 5 6 6 5 6 5 8 6

Hexokinase ti~h

0.6 :!:0.1 0.7 :!:0.1 0.2 :!: 0.04 0.1 :!: 0.01 0.2 :!: 0.08 0.08:!: 0.01 0.08:!: 0.01 0.06:!: 0.07 0.3 :!: "0.06

Citrate synthetase

Cytochrome

Phospho-

Lactate

oxidase

fructokinase

dehydrogenase

O.7:t 0.04 O.9:t 0.1 O.6:t 0.1 O.5:t 0.02 O.3:t 0.06 O.4:t 0.09 O.3:t 0.08 O.3:t 0.06 O.4:t 0.03

0.4:t 0.8:t 0.4:t 0.6:t 0.3:t 0.3:t 0.3:t 0.2:t 0.2:t

14.0:!: 0.8 4.2:!:1.0 9.8:!: 1.6 7.1:!: 1.2 21.7:!:4.3 9.0:!:1.0 13.1 :!: 3.6 29.0:!: 1.2 24.8:!: 3.0

345:!::49 237:!::22 401 :!::47 279:!:: 18 251:!:: 29 200:!:: 16 351:!::18 242:!:: 39 172:!::25

0.08 0.06 0.04 0.03 0.09 0.1 0.04 0.05 0.07

Enzyme activities are expressed as ~moles substrate utilised, 9 wet weight muscle

min-i at 15°C

Table 4. Relative activities of enzyme activities in fast fibres compared to slow fibres ( %) Species

No. Hexokinase of fISh

Citrate synthetase

Cytochrome PhosphoLactate oxidase Fructokinase dehydrogenase

Brook trout Crucian carp Tench

8 5 6 6 5 6 5 8 6

184

14

17

36

10

14

29

7

13

6

11

10

15

4

6

3

7

4

9

6

6

15

4 5

10

Whiting Flounder Cod Lemon Sole Plaice f)ah

18

5

123 221 272 355 178 214 208 171 142

173 54 79 119 131 79 143 139 140

Note ratios have been calculated from data expressed to 2 significant figures

capacities of different fast muscles (Table 4). CS and CO activities in fast muscles vary from around 10-15% of that of slow muscles in trout, crucian carp and whiting to only 4-6% in flounder, cod, lemon sole and plaice. Maximal activities of non-equilibrium enzymes are thought to provide a semiquantitative estimate of "maximum" metabolic flux through a given pathway (Newsholme et al. 1978). For example, HK andPFK activities provide a measure of the maximum potential of a tissue for aerobic glucose utilisation and anaerobic glycolysis respectively (Newsholme et al. 1978). HK activities vary some 8-fold among the slow muscles studied (Table 2). The lower activities of hexokinase in trout than in carp slow fibres correlate with the presence of extensive lipid deposits in trout muscle (Figs. 8,9) and may reflect a greater importance for lipid fuels in this species. In the trout HK activities are 2-fold higher in fast than in slow fibres, a reversal of the pattern of the other species (Table 4). Among the species with the highest activities of CO and CS in their fast muscles only carp and trout have significant HK activities ( > 0.3 J.lmoles g wet weight-1 min-1 ). It is therefore unlikely that blood glucose constitutes a major fuel for the fast muscles of these other species under aerobic conditions (Tables 3 and 4). There is wide variation in PFK activities ofboth the fast and slow muscles. The activity ofPFK is affected by a

98

IA. .1ohnston and T .W .Moon

Fig. I. Transverse section through myotomal slow muscle fibres of the brook trout. Note the large deposits of lipid (L) between the muscle fibres; intracellular lipid droplets (L) and high mitochondrial density (M7) Fig. 2. Junction between two slow fibres from brook trout showing the subsarcolemmal mitochondrial zona (MS), and numerous lipid droplets (L)

Structure and metabolism of fish muscle

99

Fig. 3. Junction between two slow fibres from Crucian carp. Note the irregular packing of myofibrils (MY) and high mitochondrial density (MI). N Nucleus; C Capillary Fig. 4. Transverse section through part of a slow fibre from the myotomal muscle of Crucian carp showing extensive deposits of glycogen granules (G) (present as rossettes) and numerous mitochondria ( MT\ with dense cristae structure (CS)

I.A. Johnston and T.W. Mool

100

wide range of allosteric effectors including AMP , Pi and citrate. Thus some of the variation in PFK activities may reflect differing degrees of activation of the enzyme among the species since high-speed homogenates were employed. However, the optimal conditions for fast and slow muscles were found to be similar for carp, trout and plaice, in which conditions for measurement of maximal activities were determined experimentally. It is likely, therefore, that the ratios ofPFK activities between fast and slow muscles proviGe an indication of the relative capacities of these fibre types for anaerobic glycolysis. PFKactivities vary from 1.2 (brook trout) to 3.6 (whiting) times higher in fast than slow muscles (Table 4). LDH activities range from 124-5091!moles g wet weight-1 in slow fibres to 1724011!moles g wet weight-:; 1 in fast muscles (Tables 2, 3). This enzyme functions to maintain redox balance during anaerobic glycolysis by providing a continuous supply of oxidized NAD. No obvious correlation is evident between muscle LDH activity and either anoxic tolerance or known scope for anaerobic activity for the species. Unlike PFK and HK, LDH catalyses a reaction close to equilibrium and cannot be used as a quantitative index of capacity for anaerobic glycolysis (see Newsholme and Start 1973). Ultrastructural investigations Typical structures of slow and fast fibres are shown in Figs. 1-4 and 8-10 respectively. Measurements of activities of CS and CO only provide an average measure of the aerobic capacities of different muscles. Morphometric analyses of mitochondrial content of individual fibres reveal a considerable spread of aerobic capacities within fibre populations, particularly for fast muscles (Figs. / , 15-17). The mean percentage of slow-fibre volume occupied by mitochondria (MF) is similar among species and in the range 25-31% (Table 5). In contrast, the MF of fast fibres varies from 8% of slow muscle values in plaice to 30% in trout (Table 5). A good correlation was obtained for the activities of CS and CO in trout, carp and plaice fast muscles with the mean MF in these fibres (Tables 3, 4 and 5). In general, small fast fibres of the carp have a higher MF than large ones (~1500I.Lm2) (Figs. 13, 14, 18). Fast fibres with areas >2500I.Lm2 invariably contained very few mitochondria (Fig. 18). In plaice white muscle 42% of fibres have a MF ofless than 1% compared to only 4% in carp (Table 5). The proportion Table

5. Mitochondrial

Species

content

of fast and slow muscles

Type of fibre

Mitochondrial Mean:tS.E, %

Brook trout

Slow Fast

Crucian carp

Slow Fast

Plaice

Slow Fast

content (Volume

% Fibres ~1%

%

%) Fibres

%

Fibres

?5%

~10%

O

100 36

31.3:!: 0.9 9.3:!:0.7

o o

76

25.5:!: 1.0 4.6:!:0.1

0 4

32

24.6:!: 0.8 2.0:!:0.3

O 42

12

O

O

100

100 0

50

40

30

" FIBRES 20

0

6

18

12

5

II

24

30

36

42

48

36

42

48

MITOCHONDRIA

" FIBRES

6

" FIBRES

0

6

12

7 Fig. 5. Frequency distribution

18

24

30

" MITOCHONDRIA of fractional volume ( %) occupied by mitochondria

in 50 slow fibres from

Fig. 6. Frequency distribution Crucian carp

of fractional volume ( %) occupied by mitochondria

in 50 slow fibres from .

Fig. 7. Frequency distribution

of fractional volume ( %) occupied by mitochondria

in 50 slow fibres from

brook trout

plaice

102

IA. Johnston and T.W. Moon

Fig.8. Semi-thin (1 ~m) section of fast muscle from brook trout stained with p-phenyldiamine. Note the distribution of mitochondria (M7) and the presence of several lipid droplets in each fibre (L). Peripheral myofibrils are elongated in transverse section (MY) Fig.9. Junction between two fast fibres from brook trout in transverse section showing the subsarcolemmal mitochondrial zone (MS), lipid droplets (L), mitochondria (M7), sarcotubular system (S7) and glycogen granules (G) Fig. 10. Transverse section through the central region ofa brook trout fast fibre. Note the small size and relatively simple cristae structure of mitochondria

Fig. 11. Fast-fibre mitochondria

from Crucian carp. Note the relatively simple structure of internal

membrane (CS) Fig. 12. Slow-fibre mitochondria (CS) to fast fibres (Fig.11)

from Crucian carp showing the greater complexity of cristae compared

Fig. 13. Transverse section through a fast fibre from Crucian carp. The total fibre area is 550 !1m2. Note the rather irregular packing of fibrils (MY) and the presence of numerous small mitochondria (MI'). (There are about 27 mitochondria in the field of view) Fig. 14. Transverse section through a fast fibre from Crucian carp. The total fibre area is 2,150 !1m2. Note the lower mitochondrial densitv (MT\ than in Fil!.13

IA. Johnston and T.W. Moon

104

" FIBRES 20

6

2

15

8 "

10

12

14

12

14

18 20

MITOCHONDRIA

" FIBRES

16

0

2

4

6

17 Fig. 15. Frequency distribution brook trout

8

10

16

18

20

22

" MITOCHONDRIA of fractional volume ( %) occupied by mitochondria

in 50 fast fibres from

Fig. 16.Frequencydistribution of fractional volume ( %) occupiedby mitochondria in 50 fast fibres from Crucian carp Fig. 17. Frequency distribution ,,1,,;,..,.

of fractional volume ( %) occupied by mitochondria

in 50 fast fibres from

105

Stnlcture and metabolism of fish muscle

Fig. 18. A graph showing the fractional volume occupied by mitochondria ( %) fibres of Crucian carp of different sizes. Units of fibre area are I1m2, Note the data does not constitute a random sample (unlike that presented in Fig. 16) due to the difficulty of measuring the area of very large fibres at sufficient resolution to distinguish individual mitn"hnnciria ,

« ~ o z O :I: u O f-

~ ~

20

10 r

o~ .(, 0...

°

.

\

00

\

.

.°

..

1000

2000

FIBRE

3000

AREA

of fast fibres having MF of more than 10% varies from 0 in plaice to 36% in trout. However, in no case is there a significant overlap between the MF of fast and slow fibres (Figs. 5,--7,15-17; Table 5). The cristae of mitochondria from fast muscles of trout, carp and plaice have a relatively simple structure compared to those of slow muscles (Figs. 11, 12). It therefore seemslikely that the ratio of aerobic capacities of fast/slow fibres is somewhat less than a simple measure of relative mitochondrial densities would indicate. Nevertheless, the results clearly demonstrate that a fraction of the fast fibre population has a significant aerobic capacity. For example, in the case of an active fish, such as brook trout, around 10% of fast-muscle fibre~ have a MF which is more than twice that of the mean (Fig. 15).

Discussion The MF in slow fibres of fish is around 25-31% in the species studied (Table 5). Even higher values have been obtained for the Atlantic mackeral (Scomber scomber) (35.5 %) (Bone 1978b) and the European anchovy (Engraulisencrasicolus) (45.5 %) (Johnston 1981, unpublished results). These are comparable to those of the ventricles of the mouse (38 %) and finch (34 %) (Bossen et al. 1978) and considerably higher than that of slow twitch fibres from mammalian limb muscles. For example, soleus muscles ofrat and guinea pig have a MF of 4.9% (Eisenberg et al. 1974) and 5.7-7.4% (Stonnington and Engel 1973) respectively. Some scrombroid fishes (e.g. Katsuwonus) maintain elevated temperatures in brain and muscle by means of a counter-current vascular heat exchange (Carey and Teal 1969; Stevens and Neill 1978). Slow muscles of the skipjack tuna, Katsuwonus pelamis, have a MF which is only around half that of Scomber scomber which does not operate at temperatures much above ambient (Bone 1978b). It seems possible that the elevated body temperatures of homeotherms allow mitochondria to operate somewhat more efficiently than those of ectotherms. Nevertheless, it is clear that slow fibres of fish have among the highest aerobic capacities of any vertebrate skeletalMF muscles. The average of the teleost fast muscles studied, 2-9% ' (Table 5) are somewhat higher than that reported for specieswith single motor endplates: sharks -

10fi

TA

Jnhn.tnn

and T .W .Moon

Scyliorhinus 1% (Totland et al. 1980); Etmopterus spinax 1 %, Galeus melastomus 0.5% (Kryvi 1977) and the sturgeon Acipenser ste//atus 0.7% (Kryvi et al. 1980). Such low mitochondrial densities are likely to be associated entirely with resting metabolism, e.g., maintenance of ion gradients, protein turnover, etc. There is now abundant evidence that, in contrast to elasmobranchs, some teleosts recruit fast motor units at low sustainable swimming speeds. For example, Bone and his co-workers ( 1978) found that white fibres in common carp ( Cyprinus carpio) were active at all speeds above 0.5 body length S-I. The threshold speed for recruitment of fast fibres in brook trout, of similar size to those in the present study, is around 1.8 bodylength S-I (Johnston and Moon 1980b). It has been generally assumed that contractions of teleost fast fibres are largely supported by anaerobic glycogenolysis as is the case for elasmobranchs (see Biliilski 1974 and Bone 1978a, for reviews). However, anaerobic pathways constitute an inefficient means of producing A TP for sustained activity. The present study provides evidence that there is not a complete dichotomy in the metabolic characteristics of fast and slow muscles in teleosts (Tables 2-5). For example, it can be seen from Table 5 that 36 % of trout fast muscle fibres contain more than 10% mitochondria. A brook trout of 59 g contains around 2.5 g of slow fibres ( 4.2% body weight) and 32.9 g of fast fibres (55.8% body weight). Thus, fibres in which the MF is more than 10% constitute an active mass of muscle at least comparable to that of the slow fibres: Although the aerobic capacity of fast fibres is far less than that of slow fibres on the basis of the abundance and structure of mitochondria, this clearly represents a significant ,