Acta physiol. scand. 1975. 95. 153-165 From The Department of Clinical Physiology, Karolinska Sjukhuset, Stockholm, Sweden and August Krogh Institute, Copenhagen University, Denmark
Metabolic Characteristics of Fibre Types in Human Skeletal Muscle BY
B. ESSBN,E. JANSSON, J. HENRIKSSON, A. W. TAYLOR and B. SALTIN Received 24 March 1975
Abstract ESS~N, B., E. JANSSON, J. HENRIKSSON, A. W. TAYLOR and B. SALTIN.Metabolic characteristics 5f fibre types in hrrman skeletal muscle. Acta physiol. scand. 1975. 95. 153-165. Muscle biopsy samples were obtained from healthy subjects in order to evaluate quantitative differences in single fibres of substrate (glycogen and triglyceride) and ion concentrations (Na+ and K+) as well as enzyme activity levels (succinate-dehydrogenase, SDH; phosphofructokinase, PFK; 3-hydroxyacyl-CoAdehydrogenase, HAD; myosin ATPase) between human skeletal muscle fibre types. After freeze drying of the muscle specimen fragments of single fibres were dissected out and stained for myofibrillar-ATPase with preincubations at pH’s of 10.3,4.6,and 4.35.Type I (“red”) and I1 A, B, and C (“white”) fibres could then be identified. Glycogen content was the same in different fibres, whereas triglyceride content was highest in Type I fibres (2-3 X Type 11). No significant differences were observed for Na+ and K+ between fibre types. The activity for the enzymes studied were quite different in the fibre types (SDH and HAD, Type I FJ 1.5 x Type 11; P F K Type I 0.5 x Type 11; Myosin ATPase Type I FJ 0.4x Type 11). The subgroups of Type I1 fibres were distinguished by differences in both SDH and PFK activities (SDH, Type I1 C > A > B; PFK, Type 11 B > A FJ C). It is concluded that contractile and metabolic characteristics of human skeletal fibres are very similar to many other species. One difference, however, appears to be that no Type I1 fibres have an oxidative potential higher than Type I fibres.
Since the first description of difference in colour between fibres in mammalian skeletal muscles many studies have further differentiated and characterized fibres in skeletal muscle. To some extent investigation has now reached the molecular level. However, these studies have not always focused on the characterization of the specific features of different fibre types. This is especially true for studies on human skeletal rnuscIe where it is very difficult to obtain pure samples containing only one fibre type. Recently Peter et al. (1972) described selected biochemical characteristics of the fibres found in skeletal muscles of guinea-pig and rat. They were able to differentiate three types of fibres, which they named fast twitch oxidative (red, Type 11), fast twitch glycolytic (white, Type 11) and slow twitch oxidative (intermediate, Type I) fibres. In most species some muscles or muscle bundles contain only one type of fibres thereby making it relatively easy to characterize biochemically each fibre type separately. In muscles where the fibre types are mixed as in man (Johnson et al. 1973) it is difficult to obtain quantitative 153
154
B. ESSBN, E. JANSSON, J. HENRIKSSON, A.
w.
TAYLOR AND B. SALTIN
measurements of one fibre type. Such information is needed, however, to enable quantitative biochemical characterization of human muscle fibre types, and to better understand their metabolism and degree of adaptability. The difficulties described above can be partially overcome by using human muscle samples which contain known percentages of Type I and Type I1 fibres extrapolating the biochemical analyses of both fibre type to 100% (Taylor, EssCn and Saltin 1974). However, a better approach to this problem is to dissect each fibre type and to base qualitative biochemical measurements on either single or pooled fibres. Using this technique the two main objectives of this study are to describe in detail, methods that can be used to quantify substrate and ion concentrations as well as enzyme activity levels in fibres of skeletal muscles in man, and to describe some biochemical characteristics of these fibre types.
Material and Methods Material. Approximately 150 muscle samples were taken and analysed from 35 healthy males and 5 healthy females. The subjects differed in their background of physical activity and were 20-39 years old.
Muscle sampling. The muscle samples (10-50 mg) were usually obtained by a needle biopsy-technique (Bergstrom 1962) and in some few cases by a surgical procedure. Before the biopsy was performed the subjects were informed about the nature of the procedure and, as all of them had participated in similar types of studies before, they were well aware of both immediate and late side effects. The samples were taken at rest from the lateral portion of the thigh as well as from the deltoid, gastrocnemius and soleus muscles. Cfassificationof fibres. Based on histochemical stain for myofibrillar ATPase a t p H 9.4 (Padykula and Herman 1955) after preincubation at p H 10.3 and 4.3 as well as NADH-dehydrogenase, these muscles, as most other skeletal muscles in man1 contain two major fibre types (Gollnick et al. 1972). The fibres staining dark for myofibrillar ATPase at alkaline preincubation have been named Type I1 and the others Type I fibres (Engel 1964). The reverse staining pattern is obtained with the acid preincubalion at p H 4.3, (Guth and Samaha1969). Preparation of fibres. Freeze dried material was used in order to obtain single muscle fibres. Freeze dried material absorbs 0,, N, and moisture when taken out of vacuum but reaches equilibrium in a few seconds (Lowry and Passonneau 1973). Absorption of approximately 5 % of the fibre’s weight can be assumed to occur at 20°C with less than 35% humidity. Our dissection was therefore performed under such conditions. That no further change in fibre weight took place in our room was ensured by weighing the same fibres repeatedly during one day. The freeze dried material was placed under a dissection microscope (40-8OX) and fragments of single muscle fibres were dissected out. The dissecting tools were constructed from wooden sticks with minute metal needles glued to one end. The single fibres were then placed on a microscope slide, the ends cut off and placed in drops of water o n another slide. After evaporation of the water the fibre ends were identified by staining for myofibrillar ATPase at p H 9.4 after preincubation at p H 10.3 for 15 min at 38°C in 0.05 M glycine buffer, 0.03 M CaCl, -0.05 M NaC1. The fibre ends were then stained either positively or negatively. Also the other fibre end was cut off from each fibre and stained after preincubation at p H 4.35 for 5 min a t room temperature in 0.06 M NaAc-0.1 M KC1. This gave the opposite staining pattern also for single fibres. After the fibres were classified the remaining portion of the fibre was weighed o n a quartz-fibre fishpole balance that had been calibrated with quinine hydrobromide. Th e weights of the fibres varied from 0.5 to 5 pg. The above procedure is schematically illustrated in Fig. 1. In addition to the above staining procedure an acid preincubation at pH 4.6 was in some instances also included (Plate 3). As shown by Brooke and Kaiser (1970) this procedure allows a further separation of Type I1 fibres in subgroups A, B and C.
It is well documented that muscles in larynx, chewing muscles, and muscles around the eyes differ in their fibre characteristics.
155
HUMAN MUSCLE FIBRE TYPES
STAl NI NG
Bi
c1
Fig. 1. The picture illustrates schematically the staining procedure of individual muscle fibre types. The fibres are divided in three parts (A) and the two small ends are each placed in a drop of water (B, C) that is allowed to evaporate. The result after staining for myofibrillar ATPase after different p H (Bl, C,) is shown.
To ensure as proper classification in subgroups as possible stains for NADH- and a-glycero-phosphate dehydrogenases were also used. These procedures called for extraordinary long fragments of muscle fibres and could then only be performed on a small number of samples. Biochemical methods. As in our earlier work the biochemical methods are based on NAD-NADP dependent enzymatic fluorimetric (Farrand) assays. Glycogen. A single fibre (0.5-5 pg) was placed in a capillary tube and 8 p1 of 1 M HCI was added. Th e tube was then sealed and placed in an oven at 100°C for 2 h. From the hydrolyzed solution 2 pl was added to 1Sop1 of reagent solution and analyzed for glucose residues (EssBn and Henriksson 1974). The coefficient of variation determined on 12 divided fibres was 6.2%. Triglycerides. Pooled Type I or Type I1 fibres from the same muscle sample (30-1OOpg) were homogenized in 0.5 ml of MeOH and 1 ml of CHCI,. 1.5 ml of saline was added. After 24 hours the CHCI, phase was removed and evaporated under a stream of N,. 7 5 y l of 0.05 M E,NOH in ethanol was added and the sealed tubes were heated at 60°C for 30 min in a water bath. 75 p1 of 0.1 M HCI were then immediately added and the content mixed. Fatty acids were removed by shaking with 2 ml of hexane. The hexane layer was taken off and a n aliquot of the hydrolysate was taken for glycerol determination according to Chernick (1969). Appropriate tripalmitin standards and blanks were treated in the same manner as the samples. A determination of the coefficient of variation on 6 pooled samples of the same type gave values as high as 35 %.As the error of the method of the analytical procedure only was 5 %, the large differences in values most likely reflects the heterogenous storage of lipids in the muscle fibres. Succinare dehydrogenase (SDH; EC 1.3.99.1). From 5 to 10 p g muscle samples (3-15 fibres of similar size from one muscle sample) were used. The pooled fibres were incubated in 75 pl 0.3 M potassium buffer (pH 7.7) with 0.05 % ' BSA for 5 min at room temperature. The sample was made 0.5 m M with regard to PMS (light sensitive), and 150 mM to Na-succinate and was incubated in a water shake bath in the dark for 30 min at 38°C. Fumarate and malate were used a s standards. The reaction was stopped with 25 pl 1 M NaOH and 50 pl brombenzol, mixed and centrifuged for 5 min. A quantitative determination of the fumarate and malate produced was then made o n the supernatant. 10 p1 were added to 1 ml of a 0.1 M hydrazin buffer (pH 9.0) containing 2.0 mM EDTA and 0.36 m M NAD. Blank fluorescence was read and the solution and 0.25 ygjml of fumarase and 5 pg/ml of malate dehydrogenase added, and the reaction allowed to proceed for 2 h. The activity was expressed in mmol
156
B.
ESSBN, E.
JANSSON, J. HENRIKSSON, A.
w.
TAYLOR AND B. SALTIN
fumarate or malate per kg dry weight per min. The coefficient of variation determined on 8 pooled and paired samples was 6.7 %. Phosphofrucfokinase (PFK; EC 2.7.1.1 1). A single fibre (0.5-2 yg) was added to 50 ,ul 0.04 M Imidazol pH 7.6 and incubated for 2 min at room temperature. The reaction was started with 1 ml reagent solution containing 0.04 M Imidazol buffer p H 7.6, 2 mM MgCI,, 1 m M ATP, 5 mM ADP, 5 mM P,, 0.2 m M 5-AMP, 1 m M G-6-P, 3 mM NH:, 0.01 mM NADH, 1 yg/ml phosphoglucoisomerase, 15 pg/ml aldolase, 3 yg/ml triosphosphatisomerase and 7.5 yg/mlglycerophosphatdehydrogenase. Fructose-1-6-diphosphatewas used as standard. The test tube with the fibre was shaken every second minute. The activity was expressed mmol fructose-1-6-diphosphate produced per kg dry weight per min. The coefficient of variation for PFK determined on pairs of fibres was 6.5 Yo. 3-hydroxyucyl-CoA-dehydrugenaseactivity (HAD; EC I . I .1.35). Determinations were made on a frozen muscle sample homogenized in 0.3 M potassium phosphate buffer (pH 7.0). The homogenate was added to a reagent solution containing 0.05 M Imidazol buffer pH 7.0, 0.05% BSA, 0.1 m M EDTA, 0.05 mM NADH and 0.02 mM acetoacetyl-CoA. N A D H was used as a standard with the blank being handled in the same way as the sample but without acetoacetyl-CoA. The activity was expressed as the maximal initial rate, mmol 3-hydroxyacetoacetyl-CoA produced per kg wet weight per min. Myosin ATPase. Muscle samples weighing 5-15 mg were homogenized in 5 ml of 0.15 M KCI-0.01 M, Histidine-0.005 M EDTA (pH 7.0). After centrifugation at 1500 g for 20 min the supernatant was decanted. The precipitate was washed in 0.15 M KCl-O.01 M Histidine (pH 7.0) and then dissolved in 6.5 ml of 0.6 M KI-6 mM Na,S,O, (pH 7.2). The sample was then centrifuged at 100 000 x g for 30 min to remove actin and muscle residue. The supernatant was dialysed with distilled water overnight. The myosin was removed from the dialysis tube by centrifugation and resuspended in 0.5 M KCI pH 7.5. All the preparative procedures were performed at 4°C. Ca+f activated ATPase was assayed according to Sreter et al. (1973) in a final volume of 1 ml containing 0.05 M TRIS, 0.5 M KCI, 1 mM EDTA and 5 mM ATP. Incubation was stopped after 10 min at 25°C by adding 0.5 ml of 10 % perchloric acid. The liberated Pi was measured according to Rockstein and Herron (1951). Protein was measured as described by Lowry (1951). The activity was expressed as pmol Pi per mg per min. The fibre composition of the sample was determined on a section which was histochemically stained for myofibrillar ATPase after preincubation both at pH 10.3 and pH 4.3. Pooled fibres (150-200 pg) were treated in the same way as the homogenates except that the volume in which they were homogenized was 2 ml and the incubation time was 15 min. K+ and Nu+ concentrations were determined on single fibres (0.5-3 yg) with an integrating flame photometer (Niham AB, Stockholm, Sweden). This apparatus was sufficiently sensitive to detect Na+ and K+ concentrations of mol, thus allowing a fibre size as small as 0.5 y g to be used. The accuracy of the method was within *4%. 0 . 5 ~ of 1 H,O was used to attach the fibre to the platinumirridium loop. Appropriate standards containing both Na+ and K+ were used and 0.5 pl of the solution was placed on the loop. The loop must be dry before it is placed in the flame, which was ensured by the use of radiation heat for at least 1/2 minute. Prior test revealed that neither of the ions nor proteins interfered with the determinations.
Results Table I summarizes the observations made on fibres from human muscle which stain positively or negatively for myofibrillar ATPase after alkaline preincubation. It appears as if the difference in staining pattern corresponds to a real difference in the activity for myosin ATPase and the extrapolated values from the linear equation y =0.16 +O.O032x (y =myosin ATPase; x fibre composition) comes out to be 0.16 and 0.48 mmol P, per mg myosin per min for Type I and Type I1 fibres, respectively (Taylor, Essen and Saltin 1974). These findings are further substantiated by determinations on pooled samples of fibres of the major types resulting in 2.2-2.5 times higher ATPase activity levels in the Type I1 than in the Type I fibres.
HUMAN MUSCLE FIBRE TYPES
157
TABLEI. Mean values+ S.D. and range for certain biochemical characteristics of skeletal muscle in man (n = number of subjects). Myofibrillar ATPase stain preincubated at pH 10.3
ATPasea pmol x min-' x mg-' myosin n=22 Glycogenb mmol x kg-'dw n = 7; 617 fibres Triglycerides' mmol x kg-ldw n = 14; 28 samples Phosphofructokinaseb mmol x min-' x kg-' dw n = 12; 101 fibres Succinate dehydrogenaseC mmol x min-I x kg-'dw n = 16; 52 samples K+b mmol x kg-'dw n = 2; 400 fibres
Na+b mmol
kg-'dw
n = 2; 400 fibres
0.48
0.16
355+ 140 (98- 881)
359f92 (1 1.5-749) 74+ 46 (21- 164)
207+ 86 (49- 328)
49.4+ 7.3 (17.3- 99.6)
25.8+ 6.9 (8.0- 61.3)
19.35 7.6
29.6+ 7.2
(6.7- 39.5)
(22.7- 50.6)
632k 18 (598- 702) 102f
617k21 (592- 690)
7
98+
(89- 122)
6
(86- 115) ~
a Extrapolated values from linear regression (see text).
Individual fibres. Pooled fibres.
The mean content of glycogen was 355 and 359 mmol per kg in the Type I and I1 fibres, respectively. The rather wide variation observed around these mean values is partly a function of different resting glycogen levels in the muscle studied. However, in most cases 85-90% of the fibres in one muscle fell within 50 mmol per kg from the mean value. The intensity of the PAS stain varies somewhat but at near or above normal glycogen content in the muscle, the histochemical stain for glycogen content does not appear to be sensitive enough (Plate 1 and Fig. 2). The mean triglyceride concentration is almost 3 times higher in Type I as compared to 11 fibres (207 us. 74 mmol per kg d.w.). A large scatter around the mean value is also found for this substrate. This can probably be attributed to differential distribution of fat in the muscle samples. Thus a t-test on paired determinations, i.e. pooled I1 and I fibres from the same muscle sample, showed a highly significant difference (Fig. 2). The histochemical picture also indicates such a difference (Plate 1). The only glycolytic enzyme studied, PFK, also demonstrated a significant difference
B. E SS~N ,E. JANSSON, J. HENRIKSSON, A.
w.
TAYLOR AND B. SALTIN
.
'Glytosen; mmoles glucolrcx k g 1 'Triglyceride; mmoles p l y c s r o l x k g '
80
.
.
/
/
c
Y
n 0
-
60
k
40
C 0
.3
c
/.
n Y 10
20
y?'
.=PFK; m m o l e r x rnin-'rkg-l
o = S D H ; mmolesx min'lxkg-l
100
200
300
400
Slow Twitch fibres
500
20
40
60
80
Slow Twitch fibres
Fig. 2. The diagrams represent triglyceride and glycogen concentrations and the activity of SDH and PFK in type I (slow twitch) and type 11 (fast twitch) fibres from different muscle samples and individuals. One point shows the concentration of substrate or activity in type I (x-axis) and type I1 (y-axis) in the same sample.
between the fibre types. In all cases where a paired comparison between fibres from the same muscle could be made, the highest activity was found in the Type 11 fibres (Fig. 2). The stain for a-glycerophosphate-dehydrogenase is often used as a marker for glycolytic potential and from Plate 2 it can be observed that there is a definite difference between Type I and I1 fibres and minor variations exist between fibres of the same type. Fibres positively stained for myofibrillar ATPase (alkaline preincubation) have the lowest staining intensity for NADH-dehydrogenase. As this latter stain is an indicator of the oxidative potential of the fibre, there is a good agreement between the histochemical picture and the quantitative determinations of SDH (Plate 2 and Fig. 2). Muscle samples from 7 subjects were used to study the SDH-activity in Type I1 A, B and C fibres as well as in Type I fibres. Of these samples, three were large enough to stain not only for myofibrillar ATPase (3 different preincubations) but also for NADH- and a-glycerophosphatedehydrogenases. Moreover on these three samples the PFK-activity was also determined. It appears that within one muscle the SDH-activity in Type I1 A fibres is lower than the activity in Type I fibres but higher than in Type I1 B (p
Metabolic Characteristics of Fibre Types in Human Skeletal Muscle BY
B. ESSBN,E. JANSSON, J. HENRIKSSON, A. W. TAYLOR and B. SALTIN Received 24 March 1975
Abstract ESS~N, B., E. JANSSON, J. HENRIKSSON, A. W. TAYLOR and B. SALTIN.Metabolic characteristics 5f fibre types in hrrman skeletal muscle. Acta physiol. scand. 1975. 95. 153-165. Muscle biopsy samples were obtained from healthy subjects in order to evaluate quantitative differences in single fibres of substrate (glycogen and triglyceride) and ion concentrations (Na+ and K+) as well as enzyme activity levels (succinate-dehydrogenase, SDH; phosphofructokinase, PFK; 3-hydroxyacyl-CoAdehydrogenase, HAD; myosin ATPase) between human skeletal muscle fibre types. After freeze drying of the muscle specimen fragments of single fibres were dissected out and stained for myofibrillar-ATPase with preincubations at pH’s of 10.3,4.6,and 4.35.Type I (“red”) and I1 A, B, and C (“white”) fibres could then be identified. Glycogen content was the same in different fibres, whereas triglyceride content was highest in Type I fibres (2-3 X Type 11). No significant differences were observed for Na+ and K+ between fibre types. The activity for the enzymes studied were quite different in the fibre types (SDH and HAD, Type I FJ 1.5 x Type 11; P F K Type I 0.5 x Type 11; Myosin ATPase Type I FJ 0.4x Type 11). The subgroups of Type I1 fibres were distinguished by differences in both SDH and PFK activities (SDH, Type I1 C > A > B; PFK, Type 11 B > A FJ C). It is concluded that contractile and metabolic characteristics of human skeletal fibres are very similar to many other species. One difference, however, appears to be that no Type I1 fibres have an oxidative potential higher than Type I fibres.
Since the first description of difference in colour between fibres in mammalian skeletal muscles many studies have further differentiated and characterized fibres in skeletal muscle. To some extent investigation has now reached the molecular level. However, these studies have not always focused on the characterization of the specific features of different fibre types. This is especially true for studies on human skeletal rnuscIe where it is very difficult to obtain pure samples containing only one fibre type. Recently Peter et al. (1972) described selected biochemical characteristics of the fibres found in skeletal muscles of guinea-pig and rat. They were able to differentiate three types of fibres, which they named fast twitch oxidative (red, Type 11), fast twitch glycolytic (white, Type 11) and slow twitch oxidative (intermediate, Type I) fibres. In most species some muscles or muscle bundles contain only one type of fibres thereby making it relatively easy to characterize biochemically each fibre type separately. In muscles where the fibre types are mixed as in man (Johnson et al. 1973) it is difficult to obtain quantitative 153
154
B. ESSBN, E. JANSSON, J. HENRIKSSON, A.
w.
TAYLOR AND B. SALTIN
measurements of one fibre type. Such information is needed, however, to enable quantitative biochemical characterization of human muscle fibre types, and to better understand their metabolism and degree of adaptability. The difficulties described above can be partially overcome by using human muscle samples which contain known percentages of Type I and Type I1 fibres extrapolating the biochemical analyses of both fibre type to 100% (Taylor, EssCn and Saltin 1974). However, a better approach to this problem is to dissect each fibre type and to base qualitative biochemical measurements on either single or pooled fibres. Using this technique the two main objectives of this study are to describe in detail, methods that can be used to quantify substrate and ion concentrations as well as enzyme activity levels in fibres of skeletal muscles in man, and to describe some biochemical characteristics of these fibre types.
Material and Methods Material. Approximately 150 muscle samples were taken and analysed from 35 healthy males and 5 healthy females. The subjects differed in their background of physical activity and were 20-39 years old.
Muscle sampling. The muscle samples (10-50 mg) were usually obtained by a needle biopsy-technique (Bergstrom 1962) and in some few cases by a surgical procedure. Before the biopsy was performed the subjects were informed about the nature of the procedure and, as all of them had participated in similar types of studies before, they were well aware of both immediate and late side effects. The samples were taken at rest from the lateral portion of the thigh as well as from the deltoid, gastrocnemius and soleus muscles. Cfassificationof fibres. Based on histochemical stain for myofibrillar ATPase a t p H 9.4 (Padykula and Herman 1955) after preincubation at p H 10.3 and 4.3 as well as NADH-dehydrogenase, these muscles, as most other skeletal muscles in man1 contain two major fibre types (Gollnick et al. 1972). The fibres staining dark for myofibrillar ATPase at alkaline preincubation have been named Type I1 and the others Type I fibres (Engel 1964). The reverse staining pattern is obtained with the acid preincubalion at p H 4.3, (Guth and Samaha1969). Preparation of fibres. Freeze dried material was used in order to obtain single muscle fibres. Freeze dried material absorbs 0,, N, and moisture when taken out of vacuum but reaches equilibrium in a few seconds (Lowry and Passonneau 1973). Absorption of approximately 5 % of the fibre’s weight can be assumed to occur at 20°C with less than 35% humidity. Our dissection was therefore performed under such conditions. That no further change in fibre weight took place in our room was ensured by weighing the same fibres repeatedly during one day. The freeze dried material was placed under a dissection microscope (40-8OX) and fragments of single muscle fibres were dissected out. The dissecting tools were constructed from wooden sticks with minute metal needles glued to one end. The single fibres were then placed on a microscope slide, the ends cut off and placed in drops of water o n another slide. After evaporation of the water the fibre ends were identified by staining for myofibrillar ATPase at p H 9.4 after preincubation at p H 10.3 for 15 min at 38°C in 0.05 M glycine buffer, 0.03 M CaCl, -0.05 M NaC1. The fibre ends were then stained either positively or negatively. Also the other fibre end was cut off from each fibre and stained after preincubation at p H 4.35 for 5 min a t room temperature in 0.06 M NaAc-0.1 M KC1. This gave the opposite staining pattern also for single fibres. After the fibres were classified the remaining portion of the fibre was weighed o n a quartz-fibre fishpole balance that had been calibrated with quinine hydrobromide. Th e weights of the fibres varied from 0.5 to 5 pg. The above procedure is schematically illustrated in Fig. 1. In addition to the above staining procedure an acid preincubation at pH 4.6 was in some instances also included (Plate 3). As shown by Brooke and Kaiser (1970) this procedure allows a further separation of Type I1 fibres in subgroups A, B and C.
It is well documented that muscles in larynx, chewing muscles, and muscles around the eyes differ in their fibre characteristics.
155
HUMAN MUSCLE FIBRE TYPES
STAl NI NG
Bi
c1
Fig. 1. The picture illustrates schematically the staining procedure of individual muscle fibre types. The fibres are divided in three parts (A) and the two small ends are each placed in a drop of water (B, C) that is allowed to evaporate. The result after staining for myofibrillar ATPase after different p H (Bl, C,) is shown.
To ensure as proper classification in subgroups as possible stains for NADH- and a-glycero-phosphate dehydrogenases were also used. These procedures called for extraordinary long fragments of muscle fibres and could then only be performed on a small number of samples. Biochemical methods. As in our earlier work the biochemical methods are based on NAD-NADP dependent enzymatic fluorimetric (Farrand) assays. Glycogen. A single fibre (0.5-5 pg) was placed in a capillary tube and 8 p1 of 1 M HCI was added. Th e tube was then sealed and placed in an oven at 100°C for 2 h. From the hydrolyzed solution 2 pl was added to 1Sop1 of reagent solution and analyzed for glucose residues (EssBn and Henriksson 1974). The coefficient of variation determined on 12 divided fibres was 6.2%. Triglycerides. Pooled Type I or Type I1 fibres from the same muscle sample (30-1OOpg) were homogenized in 0.5 ml of MeOH and 1 ml of CHCI,. 1.5 ml of saline was added. After 24 hours the CHCI, phase was removed and evaporated under a stream of N,. 7 5 y l of 0.05 M E,NOH in ethanol was added and the sealed tubes were heated at 60°C for 30 min in a water bath. 75 p1 of 0.1 M HCI were then immediately added and the content mixed. Fatty acids were removed by shaking with 2 ml of hexane. The hexane layer was taken off and a n aliquot of the hydrolysate was taken for glycerol determination according to Chernick (1969). Appropriate tripalmitin standards and blanks were treated in the same manner as the samples. A determination of the coefficient of variation on 6 pooled samples of the same type gave values as high as 35 %.As the error of the method of the analytical procedure only was 5 %, the large differences in values most likely reflects the heterogenous storage of lipids in the muscle fibres. Succinare dehydrogenase (SDH; EC 1.3.99.1). From 5 to 10 p g muscle samples (3-15 fibres of similar size from one muscle sample) were used. The pooled fibres were incubated in 75 pl 0.3 M potassium buffer (pH 7.7) with 0.05 % ' BSA for 5 min at room temperature. The sample was made 0.5 m M with regard to PMS (light sensitive), and 150 mM to Na-succinate and was incubated in a water shake bath in the dark for 30 min at 38°C. Fumarate and malate were used a s standards. The reaction was stopped with 25 pl 1 M NaOH and 50 pl brombenzol, mixed and centrifuged for 5 min. A quantitative determination of the fumarate and malate produced was then made o n the supernatant. 10 p1 were added to 1 ml of a 0.1 M hydrazin buffer (pH 9.0) containing 2.0 mM EDTA and 0.36 m M NAD. Blank fluorescence was read and the solution and 0.25 ygjml of fumarase and 5 pg/ml of malate dehydrogenase added, and the reaction allowed to proceed for 2 h. The activity was expressed in mmol
156
B.
ESSBN, E.
JANSSON, J. HENRIKSSON, A.
w.
TAYLOR AND B. SALTIN
fumarate or malate per kg dry weight per min. The coefficient of variation determined on 8 pooled and paired samples was 6.7 %. Phosphofrucfokinase (PFK; EC 2.7.1.1 1). A single fibre (0.5-2 yg) was added to 50 ,ul 0.04 M Imidazol pH 7.6 and incubated for 2 min at room temperature. The reaction was started with 1 ml reagent solution containing 0.04 M Imidazol buffer p H 7.6, 2 mM MgCI,, 1 m M ATP, 5 mM ADP, 5 mM P,, 0.2 m M 5-AMP, 1 m M G-6-P, 3 mM NH:, 0.01 mM NADH, 1 yg/ml phosphoglucoisomerase, 15 pg/ml aldolase, 3 yg/ml triosphosphatisomerase and 7.5 yg/mlglycerophosphatdehydrogenase. Fructose-1-6-diphosphatewas used as standard. The test tube with the fibre was shaken every second minute. The activity was expressed mmol fructose-1-6-diphosphate produced per kg dry weight per min. The coefficient of variation for PFK determined on pairs of fibres was 6.5 Yo. 3-hydroxyucyl-CoA-dehydrugenaseactivity (HAD; EC I . I .1.35). Determinations were made on a frozen muscle sample homogenized in 0.3 M potassium phosphate buffer (pH 7.0). The homogenate was added to a reagent solution containing 0.05 M Imidazol buffer pH 7.0, 0.05% BSA, 0.1 m M EDTA, 0.05 mM NADH and 0.02 mM acetoacetyl-CoA. N A D H was used as a standard with the blank being handled in the same way as the sample but without acetoacetyl-CoA. The activity was expressed as the maximal initial rate, mmol 3-hydroxyacetoacetyl-CoA produced per kg wet weight per min. Myosin ATPase. Muscle samples weighing 5-15 mg were homogenized in 5 ml of 0.15 M KCI-0.01 M, Histidine-0.005 M EDTA (pH 7.0). After centrifugation at 1500 g for 20 min the supernatant was decanted. The precipitate was washed in 0.15 M KCl-O.01 M Histidine (pH 7.0) and then dissolved in 6.5 ml of 0.6 M KI-6 mM Na,S,O, (pH 7.2). The sample was then centrifuged at 100 000 x g for 30 min to remove actin and muscle residue. The supernatant was dialysed with distilled water overnight. The myosin was removed from the dialysis tube by centrifugation and resuspended in 0.5 M KCI pH 7.5. All the preparative procedures were performed at 4°C. Ca+f activated ATPase was assayed according to Sreter et al. (1973) in a final volume of 1 ml containing 0.05 M TRIS, 0.5 M KCI, 1 mM EDTA and 5 mM ATP. Incubation was stopped after 10 min at 25°C by adding 0.5 ml of 10 % perchloric acid. The liberated Pi was measured according to Rockstein and Herron (1951). Protein was measured as described by Lowry (1951). The activity was expressed as pmol Pi per mg per min. The fibre composition of the sample was determined on a section which was histochemically stained for myofibrillar ATPase after preincubation both at pH 10.3 and pH 4.3. Pooled fibres (150-200 pg) were treated in the same way as the homogenates except that the volume in which they were homogenized was 2 ml and the incubation time was 15 min. K+ and Nu+ concentrations were determined on single fibres (0.5-3 yg) with an integrating flame photometer (Niham AB, Stockholm, Sweden). This apparatus was sufficiently sensitive to detect Na+ and K+ concentrations of mol, thus allowing a fibre size as small as 0.5 y g to be used. The accuracy of the method was within *4%. 0 . 5 ~ of 1 H,O was used to attach the fibre to the platinumirridium loop. Appropriate standards containing both Na+ and K+ were used and 0.5 pl of the solution was placed on the loop. The loop must be dry before it is placed in the flame, which was ensured by the use of radiation heat for at least 1/2 minute. Prior test revealed that neither of the ions nor proteins interfered with the determinations.
Results Table I summarizes the observations made on fibres from human muscle which stain positively or negatively for myofibrillar ATPase after alkaline preincubation. It appears as if the difference in staining pattern corresponds to a real difference in the activity for myosin ATPase and the extrapolated values from the linear equation y =0.16 +O.O032x (y =myosin ATPase; x fibre composition) comes out to be 0.16 and 0.48 mmol P, per mg myosin per min for Type I and Type I1 fibres, respectively (Taylor, Essen and Saltin 1974). These findings are further substantiated by determinations on pooled samples of fibres of the major types resulting in 2.2-2.5 times higher ATPase activity levels in the Type I1 than in the Type I fibres.
HUMAN MUSCLE FIBRE TYPES
157
TABLEI. Mean values+ S.D. and range for certain biochemical characteristics of skeletal muscle in man (n = number of subjects). Myofibrillar ATPase stain preincubated at pH 10.3
ATPasea pmol x min-' x mg-' myosin n=22 Glycogenb mmol x kg-'dw n = 7; 617 fibres Triglycerides' mmol x kg-ldw n = 14; 28 samples Phosphofructokinaseb mmol x min-' x kg-' dw n = 12; 101 fibres Succinate dehydrogenaseC mmol x min-I x kg-'dw n = 16; 52 samples K+b mmol x kg-'dw n = 2; 400 fibres
Na+b mmol
kg-'dw
n = 2; 400 fibres
0.48
0.16
355+ 140 (98- 881)
359f92 (1 1.5-749) 74+ 46 (21- 164)
207+ 86 (49- 328)
49.4+ 7.3 (17.3- 99.6)
25.8+ 6.9 (8.0- 61.3)
19.35 7.6
29.6+ 7.2
(6.7- 39.5)
(22.7- 50.6)
632k 18 (598- 702) 102f
617k21 (592- 690)
7
98+
(89- 122)
6
(86- 115) ~
a Extrapolated values from linear regression (see text).
Individual fibres. Pooled fibres.
The mean content of glycogen was 355 and 359 mmol per kg in the Type I and I1 fibres, respectively. The rather wide variation observed around these mean values is partly a function of different resting glycogen levels in the muscle studied. However, in most cases 85-90% of the fibres in one muscle fell within 50 mmol per kg from the mean value. The intensity of the PAS stain varies somewhat but at near or above normal glycogen content in the muscle, the histochemical stain for glycogen content does not appear to be sensitive enough (Plate 1 and Fig. 2). The mean triglyceride concentration is almost 3 times higher in Type I as compared to 11 fibres (207 us. 74 mmol per kg d.w.). A large scatter around the mean value is also found for this substrate. This can probably be attributed to differential distribution of fat in the muscle samples. Thus a t-test on paired determinations, i.e. pooled I1 and I fibres from the same muscle sample, showed a highly significant difference (Fig. 2). The histochemical picture also indicates such a difference (Plate 1). The only glycolytic enzyme studied, PFK, also demonstrated a significant difference
B. E SS~N ,E. JANSSON, J. HENRIKSSON, A.
w.
TAYLOR AND B. SALTIN
.
'Glytosen; mmoles glucolrcx k g 1 'Triglyceride; mmoles p l y c s r o l x k g '
80
.
.
/
/
c
Y
n 0
-
60
k
40
C 0
.3
c
/.
n Y 10
20
y?'
.=PFK; m m o l e r x rnin-'rkg-l
o = S D H ; mmolesx min'lxkg-l
100
200
300
400
Slow Twitch fibres
500
20
40
60
80
Slow Twitch fibres
Fig. 2. The diagrams represent triglyceride and glycogen concentrations and the activity of SDH and PFK in type I (slow twitch) and type 11 (fast twitch) fibres from different muscle samples and individuals. One point shows the concentration of substrate or activity in type I (x-axis) and type I1 (y-axis) in the same sample.
between the fibre types. In all cases where a paired comparison between fibres from the same muscle could be made, the highest activity was found in the Type 11 fibres (Fig. 2). The stain for a-glycerophosphate-dehydrogenase is often used as a marker for glycolytic potential and from Plate 2 it can be observed that there is a definite difference between Type I and I1 fibres and minor variations exist between fibres of the same type. Fibres positively stained for myofibrillar ATPase (alkaline preincubation) have the lowest staining intensity for NADH-dehydrogenase. As this latter stain is an indicator of the oxidative potential of the fibre, there is a good agreement between the histochemical picture and the quantitative determinations of SDH (Plate 2 and Fig. 2). Muscle samples from 7 subjects were used to study the SDH-activity in Type I1 A, B and C fibres as well as in Type I fibres. Of these samples, three were large enough to stain not only for myofibrillar ATPase (3 different preincubations) but also for NADH- and a-glycerophosphatedehydrogenases. Moreover on these three samples the PFK-activity was also determined. It appears that within one muscle the SDH-activity in Type I1 A fibres is lower than the activity in Type I fibres but higher than in Type I1 B (p
