Acta Physiul Scand 1990, 139, 475484
Glycogen and lactate metabolism during low-intensity exercise in man K. N O R D H E I M and N. K. V B L L E S T A D Department of Physiology, National Institute of Occupational Health, Oslo, Norway NORDHEIM, K. & VBLLESTAD, N. K. 1990. Glycogen and lactate metabolism during lowintensity exercise in man. Acta Physiol Scand 139, 475484. Received 22 December 1989, accepted 21 March 1990. ISSN 0001-6772. Department of Physiology, National Institute of Occupational Health, Oslo, Norway. The influence of high lactate concentration on glycogen metabolism in active type I and inactive type I1 fibres was investigated. High muscle lactate concentration (26.7f 1.4 mmol kg-' wet wt) was achieved by three bouts (2 min duration) of bicycle exercise at 112% VO, max. Exercise was continued at 40% VO,max for I h. Serial venous blood samples and biopsies from the vastus lateralis muscle were taken. Over the first 20 min of this low-intensity exercise muscle lactate concentration decreased by 22.9k0.7 mmol kg-' wet wt, while glycogen remained unchanged in type I fibres and increased by 20 mmol kg-' wet wt in type I1 fibres. During3the next 40 min of lowintensity exercise lactate decreased by 1.6+ 1.2 mmol kg-' wet wt, while glycogen concentration decreased by 21 k 7 mmol kg-' wet wt in type I fibres but remained stable in type I1 fibres. In a second series of experiments, in which lactate was allowed to disappear before the light exercise was started, no changes in glycogen concentration were seen in type I1 fibres during the 1 h of 40% VO,max exercise, while a continuous reduction in glycogen of 28 f8 mmol kg-' wet wt was found in type I fibres. The results indicate that in the presence of high lactate levels muscle glycogen was resynthesized in inactive type I1 muscle fibres, while lactate was oxidized in preference to glycogen in type I fibres.
Key wurds : fibre types, gluconeogenesis, glycogenolysis, human muscle, lactate.
After exercise of high intensity most of the lactate produced is eliminated within 1 h (Hermansen & Vaage 1977, Medbe & Sejersted 1985). Part of the elimination occurs in the previously exercising muscles. A study by Hermansen & Vaage (1977) indicates that lactate is converted to glycogen in the muscle cells. Only a minor part of the disappearing lactate could be ascribed to oxidation. In contrast, Brooks & Gaesser (1980) reported that less than 20% of the lactate was used for gluconeogenesis in the muscle. Their observations indicated that oxidation was the main pathway for removal of lactate after exercise.
Correspondence : Dr Nina K. Vellestad, Department of Physiology, National Institute of Occupational Health, PO Box 8149 Dep, N-0033 Oslo 1, Norway.
T h e problem of the fate of intramuscular lactate might be more easily investigated during exercise of moderate intensity, rather than rest, for the following reasons. When submaximal exercise is carried out immediately after the intense lactate-producing exercise, an increased rate of lactate elimination is reported (Jervell 1928, Rammal & Strom 1949, Hermansen & Stensvold 1972). During exercise at 40-60% Vo, max, almost solely type I fibres are active (Vellestad & Blom 1985). Hence, the increased lactate elimination could be due to lactate being the substrate for ATP resynthesis in the active fibres, and a reduced rate of glycogen depletion in these fibres would be expected. On the other hand, gluconeogenesis might occur in type I1 fibres, since they are inactive. In most of the earlier studies of glycogen synthesis, biochemical methods have been used
475
K. Nordheim and N . K. Vdlestad
476
to determine the average glycogen concentration
of the different fibres in a muscle sample. Studies of rat muscle have revealed substantial differences in the rate of gluconeogenesis in the two major fibre types (McLane & Holloszy 1977). By using a quantitative analysis of periodic acid-Schiff (PAS) stain in individual fibres, examination of changes in the glycogen concentration of each fibre type is possible. T h e present study was designed to investigate whether high lactate levels at the start of exercise at 40% Yoz max could reduce the rate of glycogen breakdown in type I fibres and at the same time lead to resynthesis of glycogen in type I1 fibres. A preliminary report of the present study has been given elsewhere (Nordheim & Vdlestad 1987).
M A T E R I A L S AND METHODS A total of 11 males and one female participated in the two parts of the study. Pertinent characteristics of the two groups of subjects are given in Table 1. The subjects were familiarized with the procedures and gave their informed consent to participate. Prior to the experiments the subjects went through a routine medical examination and their maximal oxygen uptake (Vo, max) was determined. The experiment was approved by the Ethics Committee of the institute. Protocol
The subjects arrived at the laboratory in the morning after an overnight fast. They were told not to take part in heavy physical training the day before the experiment. The exercise was carried out on a bicycle ergometer with a pedalling frequency of 1.5 and 1.17 Hz during high- and low-intensity exercise bouts respectively. Part I. In this part of the study the lactate disappearance and changes in glycogen concentration were examined during exercise at 40% YO, max following immediately after intensive exercise. T o ensure a high lactate concentration in the muscle and
in the extracellular space, three bouts of high-intensity exercise with 3 min rest between were carried out. The exercise intensity of the bouts was selected from pretesting as the highest intensity which the subjects could sustain for three bouts of 2 min with 3 min rest between. The intensity of these bouts was 112f3 yo Yoz max, and the last bout was exhaustive. A bout duration of 2 rnin was chosen since this duration is necessary to utilize the entire anaerobic capacity (Medbcr et al. 1988, Medbo & Tabata 1989). Within 40 s after the three bouts of intensive exercise, the subjects continued to exercise at 40% Vo, max for 1 h. Muscle biopsies were taken at rest before any exercise, after the three bouts of high-intensity exercise and after 5, 10, 20, 30 and 60 min of the exercise at 40% Vo, max. Blood samples were obtained at rest and during the last 30 s of the third intense exercise bout. Additional blood samples were obtained every fifth minute during light exercise. Part 11. The objective of part I1 was to study changes in the glycogen content of type I and type I1 fibres during exercise at 40Yo Vo, max when lactate produced during the preceding high-intensity exercise was allowed to disappear before the light exercise was started. Low lactate levels were attained by introducing a 60-min rest period between the intense exercise and the 40% VO, max exercise. Since glycogen concentration will increase during this rest period (Hermansen & Vaage 1977), we ensured about the same glycogen depletion as in part I before the start of the low-intensity exercise by carrying out a total of five bouts of intense exercise. The intensity of these five bouts was 111 f 6 yo YO, max. The first three bouts (2 rnin duration) were carried out as in part I with 3 rnin rest between. Thereafter a 30-min rest was allowed before the last two bouts were carried out with 3 rnin rest between. After a subsequent 60-min rest the same 40% Yo, max exercise as in part I was started. Muscle biopsy samples were taken at rest before any exercise and after the first three bouts of highintensity exercise. Additional biopsies were obtained before the start of the 40% Vo, max exercise and again after 5, 10, 20, 30 and 60 min. Blood samples were obtained at rest and during the last 30 s of the third and the fifth intense exercise bouts. Blood
+
Table 1. Pertinent characteristics of the two groups of subjects (mean SE)
Fibre type distribution Part
n
I
7
11
5
(yo)
Age (years)
Weight (kg)
Vo, max (I min-')
I
IIA
IIAB
IIB
30k2 26f2
76f3 78+1
3.9t0.3 4.0k0.1
52+4 45f4
34+5 33+5
12+2 11+1
2fl 11f3
Glycogen and lactate metabolism samples were also taken after 5 min of the rest periods after the third and fifth intense exercise bouts. During the 40% Vo, max exercise blood samples were obtained every fifth minute.
477
mmol kg-l wet wt. Protein was analysed by the biuret methods described by Zamenhof (1957). All analyses were carried out in duplicate. Data analysis and statistics
Anulyticul methods T h e muscle samples were taken from the lateral portion of the quadriceps muscle with the biopsy technique described by Bergstrom (1962). Incisions through the skin and muscle fascia were made after application of a local anaesthetic (10 mg Xylocain). T o allow rapid muscle biopsy sampling during exercise, all incisions were made in advance while the subjects rested in the supine position. All biopsies obtained immediately after intense exercise or during the 40% Vo, max exercise were taken while the subjects were sitting on the bicycle. Biopsies were taken from adjacent but different sites alternately, using the Latin square procedure to avoid systematic errors in sampling (Snedecor & Cochran 1971). T w o muscle samples were taken each time. One sample for biochemical analysis was frozen within 10 s in Freon 22 precooled to about -160 "C, and another was frozen, mounted and stored at - 80 "C for later histochemical analysis. Quantitative determination of glycogen in individual fibres was based on histochemical staining by a periodic acidSchiff (PAS) procedure and fibre classification by the myosin ATPase method as described by Vellestad et al. (1984). T o reduce the possible errors introduced by differences in section thickness, two sections were analysed from each muscle sample, and the mean data used for further calculations. Lately it has been difficult to obtain a stable light intensity using the Hg light source, and hence we used a halogen lamp (Ha) instead. The optical density (OD (PAS)) determined using the Ha light related linearly to the OD(PAS) determined by the H g light (OD,, = 0.033 0.5330Dfi,, r = 0.975, n = 45). We also calibrated the OD(PAS) measured with the Ha light to the biochemically determined glycogen concentration (see Vsllestad et al. 1984 for details), and a good linear relationship was found within a range of 8.7176.5 mmol glycosyl units kg-l wet wt (OD(PAS) = 0.02+0.0026GC, r = 0.91, n = 20). GC is the biochemically determined glycogen Concentration. The regression analysis was calculated as the geometric mean analysis (Sokal & Rohlf 1981). Blood samples were drawn into heparinized syringes through a Teflon catheter inserted into an antecubital vein. Two aliquots of blood were immediately transferred to 0 . 4 ~perchloric acid stored on ice. Glycogen, lactate, creatine phosphate (Crp) and adenosine triphosphate (ATP) were analysed enzymatically according to Lowry & Passonneau (1972). Glycogen was determined as glucosyl units kg-' wet wt, and the concentration given as
+
The resting levels and the degree of depletion of glycogen during the high-intensity exercise varied to a large extent between subjects. To analyse changes in glycogen concentration during the 40% VO, max exercise, these data were given as relative changes compared with the values obtained at start of the lowintensity exercise. All values are given as mean+SE. T h e significance of differences between mean values was determined by the Mann-Whitney rank-sum test. Comparisons of temporal changes of glycogen concentration in different muscle fibre types were performed by MANOVA, whereas least-square linear regression analysis was used to examine temporal changes in biochemically determined metabolites (Sokal & Rohlf 1981). A significance level of 5% was adopted.
RESULTS Part I
Muscle lactate concentration was 0.5 0.2 mmol kg-' wet wt at rest and increased to 26.7f 1.6 mmol kg-' wet wt after the three bouts of intensive exercise (112% of VO,max) (P < 0.001) (Fig. la). Over the first 20 rnin of the subsequent low-intensity exercise (40 % of Vo, max), muscle lactate concentration fell to 3.7k0.5 mmol kg-' wet wt (P < 0.001). At the end of the 60 rnin of low-intensity exercise a further decrease to 2.1 0.4 mmol kg-' wet wt was observed (P < 0.05). The average rate of lactate disappearance in muscle was 1.15 mmol kg-' wet wt min-' over the first 20 rnin and 0.04 mmol kg-' wet wt min-' over the last 40 rnin of low-intensity exercise. Blood lactate concentration was 0.8 'r 0.2 mmol 1-' at rest, 12.8 & 1.4 mmol I-' before the start of low-intensity exercise (t = 0 min) and peaked at 14.4f 1.1 mmol 1-' 4 rnin into the subsequent 40% Vo, max exercise (P < 0.001) (Fig. la). Thereafter a gradual decline was observed. After 21 min blood lactate concentration (5.2 & 0.4 mmol 1-') was significantly higher than muscle lactate concentration (P < 0.025), creating a favourable gradient for lactate uptake by the muscle cells. Blood lactate had declined futher to 1.4k0.2 mmol 1-' after 60 min.
478
K. Nordheim and N K. Vdlestad (a)
30
0 (b) 0.16
60
D
@
c
2 .-
0.12
0.4) when lactate concentration decreased rapidly. Over the last 40 rnin a glycogen depletion of 21 + 4 mmol kg-' wet wt was observed (P
0.5). I n the next 15 min the glycogen concentration increased by 25 f9 (P < 0.002) and 19 f9 mmol kg-' wet wt (P < 0.02) in type I I A and IIAB fibres respectively. These values correspond to average rates of resynthesis of 1.7 and 1.3 mmol kg-' wet wt min-' respectively. Over the last 40 rnin of the low-intensity exercise, when lactate was low and declined
Glycogen and 1a.cfate metabolism 75
50 --
x
a 0
+I +I +I +I LnNtN
-
+I ?-4
*
t-mmf
-50
-25
0
25
50
m
75
Glyzogen conc. for type HA (rnmol kg-')
Fig. 2. Changes in glycogen concentration (calculated from PAS stain intensity) for type IIAB and IIB fibres vs type IIA fibres during the first 20 min of low-
intensity exercise following high-intensity exercise. The straight line is the line of identity. Fibre type IIAB (A), IIB ( 0 ) . slowly, no significant changes in glycogen concentration of type IIA and IIAB fibres were observed ( P > 0.5). A significant fraction of type IIB fibres was found in only three of the seven subjects. Hence, no statistical analysis of the temporal change of this fibre type was carried out. However, when the changes in the glycogen concentration in type IIAB and IIB fibres are plotted against the corresponding changes in type IIA fibres the data points are distributed around the line of identity (Fig. 2). These correlations indicate about the same the rate of glycogen resynthesis for the three subgroups of type I1 fibres. T o check the validity of the histochemical results we also measured glycogen concentration biochemically. This analysis on muscle samples with all types of fibres showed a glycogen depletion of 48 i7 mmol kg-' wet wt during the high-intensity exercise (Table 2). During the first 20 min of the following low-intensity exercise, glycogen concentration increased by 11 f6 mmol kg-' wet wt ( P < 0.05), whereas a shift towards glycogen breakdown rather than synthesis was seen over the subsequent 40 min. Considering that type I and I1 fibres constituted 52% and 48% of the vastus lateralis muscle mass (Table 1) respectively, the two measurements of glycogen changes were almost identical. During the three bouts of intensive exercise,
0
0
9N.9
z
m - 0 0
m
21
+I +I +I +I LnG??W 00
*
2-2 0
s
?N.9
N 0
m o o 0 +I +I +I +I -LnmLn m - 4 s
-
d-
+I a,
d-
0
Ln
0
* 2
??19
2
2 N O O
+I +I +I +I
+I cQir4s 2
-
O N 0 0
0 m
"C
0
m i 0 0 +I +I +I +I
Ln
m
-tl
-ewG?00
m
Z-IZ
Ln
0
N C
'9 h! 9 m o o 0
m
0
Y
!3 h
h h
m
479
480
K. Nordheim and N . K. Vdlestad
CrP concentration fell by 71 yo ( P < 0.001) (Table 2). Following 5 min of 40% VO, max exercise, the CrP concentration had increased to 70% of the resting value and remained at this level for the rest of the low-intensity exercise ( P > 0.4). T h e ATP concentration declined by 25 yo during the high-intensity exercise ( P < 0.025), and this decrease could partly be ascribed to the rise in water content, as shown by a decreased protein concentration (Table 2). No significant changes in ATP concentration were observed during the 40% VO, max exercise (P > 0.1).
(a’
301
T h e changes in protein concentration indicate a 12% increase in water content from rest to after the high-intensity exercise (Table 2). No significant changes were observed during the low-intensity exercise ( P > 0.5), indicating that in this period changes in the measured metabolite concentration reflect alterations in amount.
Part 11 The temporal association between high lactate concentration on the one hand and absence of glycogen depletion in active fibres and glycogen
r 30
0
Time (min)
Fig. 3. Changes in metabolites during low-intensity exercise (LE, 40% VO,max) 60 min after the lactate last high-intensity exercise (HE, 111 % VO, max). (a) Muscle ( 0 )and blood (0) concentration. (b) Glycogen concentration in type I IIA (A) and IIAB (m) fibres. T h e right axis was calculated as described in Materials and Methods. Pre-exercise rest values (R) are shown at the left. Data are given as mean SE for five subjects. Deviating numbers of subjects are shown in the graph.
(a),
Glycogen and lactate metabolism resynthesis in inactive fibres during low-intensity exercise on the other hand suggests that lactate is the preferred substrate for both glycogen synthesis and ATP resynthesis. T o test the validity of this theory five subjects were studied in a separate series of experiments in which muscle lactate concentration was reduced before the start of low-intensity exercise by a 60-min rest. Muscle lactate concentration rose to 26.7 2.2 mmol kg-' wet wt after three bouts of high-intensity exercise from a pre-exercise resting value of 0.6k0.2 mmol kg-' wet wt at rest (Fig. 3a). This change was identical to that observed in part I (Fig. la). Before the start of the 40% Vo, max exercise muscle lactate concentration had declined to 2.4 f0.6 mmol kg-' wet wt, which was only 9% of the lactate concentration found at the start of low-intensity exercise in part I. Muscle lactate concentration remained low and reached 1.0f0.2 and 0.7 +_ 0.3 mmol kg-' wt wt after 20 and 60 rnin of low-intensity exercise respectively. Blood lactate concentration was 0.9 f 0.1 mmol I-' at rest, 13.9f 1.7 mmol 1-' after three bouts of high-intensity exercise (Fig. 3a) and peaked at 17.8+ 1.0 mmol 1-' after 5 min of the following rest period (data point not shown). At the start of the 4076 VO, max exercise after the 60-min rest period, blood lactate concentration was 3.9 f0.3 mmol l-', which was higher than the concentration in muscle (P < 0.05). A gradual decrease to 1.2 f0.1 mmol 1-' occirred over the 60 min of exercise (P< 0.05). Glycogen concentration determined as optical density (OD(PAS)) decreased by 36 f8, 59 f 11 and 54 f 12 mmol kg-' wet wt in type I, IIA and IIAB fibres respectively during the five intensive exercise bouts and the two long rest periods (Fig. 3 b). In type I fibres, the glycogen concentration decreased gradually from the start of the 40% VO, max exercise by 0.44 f0.15 mmol kg-' wet wt min-' (P< 0.03). This depletion rate was not significantly different from the rate calculated for the last 40 rnin in part I (P> 0.87). No significant changes were detected in type I1 fibres during the 60 rnin of exercise at 40% Vo, max (P> 0.6). Biochemical analysis of muscle glycogen showed a glycogen reduction of 4 0 f 6 mmol kg-' wet wt from rest to the start of the lowintensity exercise (Table 2). A further decline of glycogen concentration by 15 +_ 5 mmol kg-' wet
481
wt during the low-intensity exercise was observed
(P< 0.03). Hence the average rate of glycogen breakdown was about 0.25 mmol kg-' wet wt min-', which is in close agreement with the depletion measured with OD(PAS) considering that type I fibres constituted about 45% of the vastus lateralis muscle mass of these subjects.
DISCUSSION During exercise at 40% Vo, max, we observed a clear temporal association between high lactate levels on the one hand and stable glycogen levels in type I and glycogen resynthesis in type I1 fibres on the other hand. When lactate levels were low, a continuous glycogen breakdown was seen in type I fibres and no resynthesis was observed in type I1 fibres. These relationships strongly indicate that lactate was oxidized in type I fibres and also used to synthesize glycogen in type I1 fibres. I Heany exercase-lactate
accumulation
The glycogen breakdown during the highintensity exercise was accompanied by a large accumulation of lactate. The glycogen reduction was 34mmol kg-'wet wt in type I and 61-63 mmol kg-' wet wt in the subgroups of type I1 fibres. A similar relationship between the major fibre types has also been reported for one bout of exhaustive exercise with a duration of 0 . 5 4 rnin (Tabata et al. 1985). Assuming a total active muscle mass of 13 kg, aerobic combustion of glycogen would have required about 35 10,in type I and 65 1 0, in type I1 fibres. With a maximal oxygen uptake of 3.9 1 min-' a total oxygen consumption of well below 20 10,would be expected during the high-intensity exercise. Hence, in both type I and type I1 fibres a large proportion of the glycogenolysis must have resulted in production and accumulation of lactate. Using repeated exercise bouts it is possible to obtain high lactate levels in both muscle and blood (Hermansen & Stensvold 1972; Hermansen & Vaage 1977). With an elevated blood concentration lactate is also taken up by inactive muscles, and probably by other tissues (Karlsson et al. 1975). In the present study the 12 rnin with high lactate levels allowed a good equilibration between blood and other tissue fluids. The rate
482
K. Nordheirn and N . K. Vdlestad
of lactate transport is reported to be about 12 mmol kg-' min-' (Juel & Wibrand 1989). One may therefore expect that in the present study the blood lactate concentration reflects the concentration of most of the body fluids, which are estimated to constitute about 55% of the body mass (Freund & Gendry 1978). Lom-tntensil,y exercise---high lactate
Over the first 20 min of the 40y0 Vo2 max exercise average muscle lactate declined by 24 mmol kg-I wet wt. Earlier studies have shown that blood lactate disappears faster during light exercise compared with resting recovery (Jervell 1928, Rammal & Strom 1949, Hermansen & Stensvold 1972). T h e average glycogen concentration, as determined biochemically in mixed muscle samples, showed a significant increase of 11 mmol kg-' wet wt over the first 20 min of light exercise following immediately after highintensity exercise (Table 2). This increase is in good agreement with the maintained glycogen concentration in type I fibres and an increase of 19-25 mmol kg-' wet w t in type I1 fibres. In the submaximal exercise, type I fibres were presumably the active fibres (V~llestad& Blom 1985), and hence these fibres had the highest ATP demand. Since glycogen concentration remained virtually unchanged while lactate was high, ATP must have been resynthesized from substrates other than glycogen. Lactate could serve this purpose, and the rapid reduction in lactate is consistent with this postulate. Two alternative substrates, glucose and free fatty acids (FFA), can provide only a small fraction of the ATP demand in the active fibres. During light exercise, without prior heavy exercise, glucose uptake can supply about 40y0 of the energy (Ahlborg et al. 1986, Richter et al. 1988), whereas lipid oxidation may account for about 35% (Vdlestad & Blom 1985). However, in the present study the profound acidosis present in the first 20 min of light exercise would have inhibited both glucose uptake and lipolysis (Issekutz et al. 1965, Paul & Issekutz 1967). Hence, the intracellular availability of glucose and FFA was lower than when light exercise was carried out without preceding intense bouts. In resting type I1 fibres, glycogen was synthesized when lactate was high. T h e rate of resynthesis in the type I1 fibres was 10 times faster than when blood-borne glucose was the
main substrate (Vdlestad et a/. 1989). Accordingly, only a minor fraction of the observed glycogen resynthesis could possibly be ascribed to glucose uptake, leaving lactate as the only candidate. This conclusion is in good agreement with previous studies of human and rat muscle during inactive post-exercise recovery. Hermansen & Vaage (1977) examined glycogen and lactate changes in muscle biopsies and fluxes of blood glucose and lactate over the leg, and they concluded that more than 75% of the lactate produced was converted to glycogFn in the first 30 min of the recovery period. Astrand et a/. (1986) examined this in more detail by also measuring the splanchnic flow and lactate uptake, and it was calculated that over the first hour of recovery a maximum of 50 yo of the disappearing lactate could by accountkd for by uptake in the liver and by oxidation. Hence, more than 50% of the lactate was probably resynthesized to glycogen in the leg muscles. More direct evidence was obtained by other investigators (McLane & Holloszy 1977, Shiota et al. 1984) who perfused isolated rat hindlimbs with '*C-labelled lactate and recovered the isotope in glycogen of the type I1 muscle fibres. Our results indicate that glycogen was resynthesized at the same rate in all subgroups of type I1 fibres. This finding compares well with our observation of similar rates of glycogen resynthesis in type IIA, IIAB and IIB fibres after prolonged exhaustive exercise when glucose was the substrate (Vdlestad et al. 1989). I t thus appears that the subgroups of type I1 fibres within one muscle behave in a similar manner with respect to synthesis of glycogen. Studies of rat muscle show that type I fibres (soleus) do not resynthesize glycogen from lactate because of a very low level of fructose biphosphatase activity (McLane & Holloszy 1977). Whether the same is true for type I fibres within a human mixed muscle remains to be seen. However some information may be obtained by comparing our data for type I1 fibres with the resting postexercise data of Hermansen & Vaage (1977) for muscle samples consisting of both type I and I1 fibres. Based on the rate of glycogen resynthesis observed in the present study and a fibre distribution of 48% type 11, the average gluconeogenetic rate for the mixed muscle would be about 0.6 mmol kg-' wet wt min-' without any resynthesis in type I fibres. Hermansen & Vaage observed an average rate of
Glycogen and lactate metabolism 0.55 mmol kg-' wet wt min-'. This calculation suggests that human type I fibres also synthesize glycogen from lactate at a much slower rate than type I1 fibres. T h e increase in average C r P concentration of the muscle fibres over the first 5 rnin of light exercise indicates that C r P repletion occurs in type I1 fibres. During this period, glycogen resynthesis in type I1 fibres was lower than for the subsequent 15 min, suggesting that synthesis of glycogen was inhibited until C r P was repleted. Earlier observations of a continued decline in glycogen concentration in the first minutes after heavy exercise supports this view (Hermansen & Vaage 1977). Low-intensity exercise-low
lactate
Uptake of lactate in muscle fibres is linearly related to the extracellular concentration (McLane & Holloszy 1977). Hence, a decreased importance of lactate as a substrate for gluconeogenesis and oxidation would be expected as the muscle and blood concentration is lowered. I n fact, this is what we observed. After 20 min low-intensity exercise in part I and at the start of low-intensity exercise of part 11, muscle lactate concentration was below 4 mmol kg-' wet wt and blood lactate concentration was below 6 mmol kg-' wet wt. Under these conditions no resynthesis of glycogen was observed in type I1 fibres and glycogen declined continuously in type I fibres. I n a previous study we showed a continuous breakdown of glycogen in type I fibres and only minor changes in type I1 fibres during exercise at 43% of VO, max (Vdlestad & Blom 1985). T h e same pattern was found in part I1 of the present study and after 20 min of low-intensity exercise in part I. T h e average rates of glycogen depletion in type I fibres were not significantly different in the two parts of the present study. Hence, glycogen metabolism during exercise at low lactate levels seems to be determined by the current metabolic status, and is only marginally influenced by high lactate levels in the early phase of low-intensity exercise. At low lactate levels, blood glucose is the main substrate for muscle glycogen synthesis. T h e rate of glycogen resynthesis from glucose after exercise is about 10 mmol kg-' wet wt-' (Vdlestad et 01. 1989) and would hardly be detected within the short observation times used
483
in the present study. T h e finding of unchanged glycogen levels in type I1 fibres when muscle lactate was lower than 4 mmol kg-' wet wt is in good agreement with the data presented by Brooks & Gaesser (1980). Using tracer techniques and low lactate concentrations (below 5 mM), they reported that most of the eliminated lactate was oxidized and only a minor fraction resynthesized to glycogen. These results cannot be compared with conditions where lactate concentration is high, as during the first 20 rnin of low-intensity exercise of part I and in the ?tudies by Hermansen & Vaage (1977) and by Astrand et al. (1986). T h e present results strongly indicate a dual effect of lactate upon glycogen metabolism during low-intensity exercise. T h e absence of glycogen depletion in type I fibres when lactate levels are high indicates that glycogenolysis is not regulated by activity only, but can also be modulated by the metabolic status. T h e close correlation between high lactate levels and glycogen resynthesis in type I1 fibres indicates that lactate is incorporated into glycogen. Hence, high lactate levels can serve to spare glycogen in active fibres and restore glycoten levels in inactive fibres. We thank Professor Ole M. Sejersted for valuable comments and suggestions and Roald A. Bjerklund for help with statistical analysis. Excellent technical
assistance from Ada Ingvaldsen, Bjerg Ingrid Selberg and Jorid Thrane Stuenres is gratefully acknowledged. Financial support was received from The Norwegian Confederation of Sports and from The Norwegian Research Council for Science and Humanities.
REFERENCES AHLBORG, G., WAHREN,J. & FELIG, P. 1986. Splanchnic and peripheral glucose and lactate metabolism during and after prolonged arm exercise. 3 Clin Invest 77, 690-699. ASTRAND, P . - 0 . HULTMAN, E., JUHLIN-DANNFELT, A. & REYNOLDS, G. 1986. Disposal of lactate during and after strenuous exercise in humans. 3 Appl Ph,ysiol 61, 338-343. BERGSTROM, J. 1962. Muscle electrolytes in man. Scand 3 Clin Lab Invest 14 (Suppl. 68), 11-13. BROOKS, G.A. & GAESSER, G.A. 1980. End points of lactate and glucose metabolism after exhausting exercise. 3 Appl Physiul49, 1057-1069. FREUND, H. & GENDRY, P. 1978. Lactate kinetics after short strenuous exercise in man. E u r 3 Appl Physiul 39, 123-135.
484
K . Nordheim and N . K. Vollestad
HERMANSEN, L. & STENSVOLD, I. 1972. Production and removal of lactate during exercise in man. Acta Ph,ysiol Scand 86, 191-201. L. & VAAGE,0. 1977. Lactate disHERMANSEN, appearance and glycogen synthesis in human muscle after maximal exercise. A m 3 Physiol 233, E422ZE429. ISSEKUTZ JR, B.J, MILLER,H.I., PAUL,P. & RODAHL, K. 1965. Effect of lactic acid on free fatty acids and glucose oxidation in dogs. Am 3 Physiol 209, 1137-1 144. JERVELL, 0. 1928. Investigation of the concentration of lactic acid in blood and urine. Acta Med Scand SUPPI. 24, 1-135. JUEI., C. & WIBRAND, F. 1989. Lactate transport in isolated mouse muscles studied with a tracer technique kinetics, stereospecificity, pH dependency and maximal capacity. Acta Physiul Scand 137, 33-39. KARISON,J., BONDE-PETERSEN, F., HENRIKSSON, J. & KNUTTGEN, H.G. 1975. Effects of previous exercise with arms or legs on metabolism and performance in exhaustive exercise. 3Appl Physiol38, 763-765. LOWRY,O.H. & PASSONNEAU, J.V. 1972. A Flexible System of Enzymatic Analysis. Academic Press, New York. MCLANE,J.A. & HOLLOSZY,J.O. 1977. Glycogen synthesis from lactate in the three types of skeletal muscle. 3 Biol Chem 254, 6548-6553. MEDBB, J.I. & SEJERSTED, O.M. 1985. Acid-base and electrolyte balance after exhausting exercise in endurance-trained and sprint-trained subjects. Acta Physiol Scand 125, 97-109. MEDBB, J.1. & TABATA, I. 1989. Relative importance of aerobic and anaerobic energy release during shortlasting, exhausting bicycle exercise. 3 Appl Physiol 67, 1881-1886. MEDB0, J.I., MOHN, A.C., TABATA,I., BAHR,R., VAAGE,0. & SEJERSTED, O.M. 1988. Anaerobic capacity determined by maximal accumulated 0, deficit. 3 Appl Physiol 64, 50-60. -
NORDHEIM, K. & VBLLESTAD, N.K. 1987. Can lactate be used as a fuel in active fibres and simultaneously be converted to glycogen in inactive fibres within the same muscle? Acta Ph,ysiol Scand 129, 45A. JR, B. 1967. Role of extramuscular PAUL,P. & ISSEKUTZ energy sources in the metabolism of the exercising dog. 3 Appl Physiol22, 615-622. RAMMAL,K. & STROM,G. 1949. The rate of lactate utilization in man during work and at rest. Acta Physiol Scand 17, 452456. RICHTER,E.A., KIENS,B., SALTIN, B., CHRISTENSEN, N.J. & SAVARD, G. 1988. Skeletal muscle glucose uptake during dynamic exercise in humans : role of muscle mass. A m 3 Physiol 254, E55j-Ej61. SHIOTA,M., GOLDEN,S. & KATZ,J. 1984. Lactate metabolism in the perfused hindlimb. Biochem f 222, 281-292. SNEDECOR, G.W. & COCHRAN, W.C. 1971. Statistical Methods. The Iowa University Press, Ames, Iowa. SOKAL, R.R. & ROHLF,F.J. 1981. Biometry. Freeman, San Francisco. TABATA, I., MEDBB, J.I., V0LLESTAD, N.K. & SEJERSTED, O.M. 1985. Glycogen breakdown and lactate accumulation during shortlasting exercise in normal humans. Clin Ph,ysiol abstr. no. 85. VBLLESTAD, N.K. & BLOM, P.C.S. 1985. Effect Of varying exercise intensity on glycogen depletion in human muscle fibres. Acta Physiol Scand 125, 3955405. L. 1984. VBLLESTAD, N.K., VAAGE, 0. & HERMANSEN, Muscle glucogen depletion patterns in type I and subgroups of type I1 fibres during prolonged severe exercise in man. Acta Physiol Scand 122, 4335441. VBLLESTAD, N.K., BLOM,P.C.S. & GRBNNERBD, 0. 1989. Resynthesis of glycogen in different muscle fibre types after prolonged exhaustive exercise in man. Acta Physiol Scand 137, 15-21. ZAMENHOF,S. 1957. In S. P. Colowick & N. 0. Kaplan (eds.) Methods in Enzymology, pp. 696-703. Academic Press, New York.
Glycogen and lactate metabolism during low-intensity exercise in man K. N O R D H E I M and N. K. V B L L E S T A D Department of Physiology, National Institute of Occupational Health, Oslo, Norway NORDHEIM, K. & VBLLESTAD, N. K. 1990. Glycogen and lactate metabolism during lowintensity exercise in man. Acta Physiol Scand 139, 475484. Received 22 December 1989, accepted 21 March 1990. ISSN 0001-6772. Department of Physiology, National Institute of Occupational Health, Oslo, Norway. The influence of high lactate concentration on glycogen metabolism in active type I and inactive type I1 fibres was investigated. High muscle lactate concentration (26.7f 1.4 mmol kg-' wet wt) was achieved by three bouts (2 min duration) of bicycle exercise at 112% VO, max. Exercise was continued at 40% VO,max for I h. Serial venous blood samples and biopsies from the vastus lateralis muscle were taken. Over the first 20 min of this low-intensity exercise muscle lactate concentration decreased by 22.9k0.7 mmol kg-' wet wt, while glycogen remained unchanged in type I fibres and increased by 20 mmol kg-' wet wt in type I1 fibres. During3the next 40 min of lowintensity exercise lactate decreased by 1.6+ 1.2 mmol kg-' wet wt, while glycogen concentration decreased by 21 k 7 mmol kg-' wet wt in type I fibres but remained stable in type I1 fibres. In a second series of experiments, in which lactate was allowed to disappear before the light exercise was started, no changes in glycogen concentration were seen in type I1 fibres during the 1 h of 40% VO,max exercise, while a continuous reduction in glycogen of 28 f8 mmol kg-' wet wt was found in type I fibres. The results indicate that in the presence of high lactate levels muscle glycogen was resynthesized in inactive type I1 muscle fibres, while lactate was oxidized in preference to glycogen in type I fibres.
Key wurds : fibre types, gluconeogenesis, glycogenolysis, human muscle, lactate.
After exercise of high intensity most of the lactate produced is eliminated within 1 h (Hermansen & Vaage 1977, Medbe & Sejersted 1985). Part of the elimination occurs in the previously exercising muscles. A study by Hermansen & Vaage (1977) indicates that lactate is converted to glycogen in the muscle cells. Only a minor part of the disappearing lactate could be ascribed to oxidation. In contrast, Brooks & Gaesser (1980) reported that less than 20% of the lactate was used for gluconeogenesis in the muscle. Their observations indicated that oxidation was the main pathway for removal of lactate after exercise.
Correspondence : Dr Nina K. Vellestad, Department of Physiology, National Institute of Occupational Health, PO Box 8149 Dep, N-0033 Oslo 1, Norway.
T h e problem of the fate of intramuscular lactate might be more easily investigated during exercise of moderate intensity, rather than rest, for the following reasons. When submaximal exercise is carried out immediately after the intense lactate-producing exercise, an increased rate of lactate elimination is reported (Jervell 1928, Rammal & Strom 1949, Hermansen & Stensvold 1972). During exercise at 40-60% Vo, max, almost solely type I fibres are active (Vellestad & Blom 1985). Hence, the increased lactate elimination could be due to lactate being the substrate for ATP resynthesis in the active fibres, and a reduced rate of glycogen depletion in these fibres would be expected. On the other hand, gluconeogenesis might occur in type I1 fibres, since they are inactive. In most of the earlier studies of glycogen synthesis, biochemical methods have been used
475
K. Nordheim and N . K. Vdlestad
476
to determine the average glycogen concentration
of the different fibres in a muscle sample. Studies of rat muscle have revealed substantial differences in the rate of gluconeogenesis in the two major fibre types (McLane & Holloszy 1977). By using a quantitative analysis of periodic acid-Schiff (PAS) stain in individual fibres, examination of changes in the glycogen concentration of each fibre type is possible. T h e present study was designed to investigate whether high lactate levels at the start of exercise at 40% Yoz max could reduce the rate of glycogen breakdown in type I fibres and at the same time lead to resynthesis of glycogen in type I1 fibres. A preliminary report of the present study has been given elsewhere (Nordheim & Vdlestad 1987).
M A T E R I A L S AND METHODS A total of 11 males and one female participated in the two parts of the study. Pertinent characteristics of the two groups of subjects are given in Table 1. The subjects were familiarized with the procedures and gave their informed consent to participate. Prior to the experiments the subjects went through a routine medical examination and their maximal oxygen uptake (Vo, max) was determined. The experiment was approved by the Ethics Committee of the institute. Protocol
The subjects arrived at the laboratory in the morning after an overnight fast. They were told not to take part in heavy physical training the day before the experiment. The exercise was carried out on a bicycle ergometer with a pedalling frequency of 1.5 and 1.17 Hz during high- and low-intensity exercise bouts respectively. Part I. In this part of the study the lactate disappearance and changes in glycogen concentration were examined during exercise at 40% YO, max following immediately after intensive exercise. T o ensure a high lactate concentration in the muscle and
in the extracellular space, three bouts of high-intensity exercise with 3 min rest between were carried out. The exercise intensity of the bouts was selected from pretesting as the highest intensity which the subjects could sustain for three bouts of 2 min with 3 min rest between. The intensity of these bouts was 112f3 yo Yoz max, and the last bout was exhaustive. A bout duration of 2 rnin was chosen since this duration is necessary to utilize the entire anaerobic capacity (Medbcr et al. 1988, Medbo & Tabata 1989). Within 40 s after the three bouts of intensive exercise, the subjects continued to exercise at 40% Vo, max for 1 h. Muscle biopsies were taken at rest before any exercise, after the three bouts of high-intensity exercise and after 5, 10, 20, 30 and 60 min of the exercise at 40% Vo, max. Blood samples were obtained at rest and during the last 30 s of the third intense exercise bout. Additional blood samples were obtained every fifth minute during light exercise. Part 11. The objective of part I1 was to study changes in the glycogen content of type I and type I1 fibres during exercise at 40Yo Vo, max when lactate produced during the preceding high-intensity exercise was allowed to disappear before the light exercise was started. Low lactate levels were attained by introducing a 60-min rest period between the intense exercise and the 40% VO, max exercise. Since glycogen concentration will increase during this rest period (Hermansen & Vaage 1977), we ensured about the same glycogen depletion as in part I before the start of the low-intensity exercise by carrying out a total of five bouts of intense exercise. The intensity of these five bouts was 111 f 6 yo YO, max. The first three bouts (2 rnin duration) were carried out as in part I with 3 rnin rest between. Thereafter a 30-min rest was allowed before the last two bouts were carried out with 3 rnin rest between. After a subsequent 60-min rest the same 40% Yo, max exercise as in part I was started. Muscle biopsy samples were taken at rest before any exercise and after the first three bouts of highintensity exercise. Additional biopsies were obtained before the start of the 40% Vo, max exercise and again after 5, 10, 20, 30 and 60 min. Blood samples were obtained at rest and during the last 30 s of the third and the fifth intense exercise bouts. Blood
+
Table 1. Pertinent characteristics of the two groups of subjects (mean SE)
Fibre type distribution Part
n
I
7
11
5
(yo)
Age (years)
Weight (kg)
Vo, max (I min-')
I
IIA
IIAB
IIB
30k2 26f2
76f3 78+1
3.9t0.3 4.0k0.1
52+4 45f4
34+5 33+5
12+2 11+1
2fl 11f3
Glycogen and lactate metabolism samples were also taken after 5 min of the rest periods after the third and fifth intense exercise bouts. During the 40% Vo, max exercise blood samples were obtained every fifth minute.
477
mmol kg-l wet wt. Protein was analysed by the biuret methods described by Zamenhof (1957). All analyses were carried out in duplicate. Data analysis and statistics
Anulyticul methods T h e muscle samples were taken from the lateral portion of the quadriceps muscle with the biopsy technique described by Bergstrom (1962). Incisions through the skin and muscle fascia were made after application of a local anaesthetic (10 mg Xylocain). T o allow rapid muscle biopsy sampling during exercise, all incisions were made in advance while the subjects rested in the supine position. All biopsies obtained immediately after intense exercise or during the 40% Vo, max exercise were taken while the subjects were sitting on the bicycle. Biopsies were taken from adjacent but different sites alternately, using the Latin square procedure to avoid systematic errors in sampling (Snedecor & Cochran 1971). T w o muscle samples were taken each time. One sample for biochemical analysis was frozen within 10 s in Freon 22 precooled to about -160 "C, and another was frozen, mounted and stored at - 80 "C for later histochemical analysis. Quantitative determination of glycogen in individual fibres was based on histochemical staining by a periodic acidSchiff (PAS) procedure and fibre classification by the myosin ATPase method as described by Vellestad et al. (1984). T o reduce the possible errors introduced by differences in section thickness, two sections were analysed from each muscle sample, and the mean data used for further calculations. Lately it has been difficult to obtain a stable light intensity using the Hg light source, and hence we used a halogen lamp (Ha) instead. The optical density (OD (PAS)) determined using the Ha light related linearly to the OD(PAS) determined by the H g light (OD,, = 0.033 0.5330Dfi,, r = 0.975, n = 45). We also calibrated the OD(PAS) measured with the Ha light to the biochemically determined glycogen concentration (see Vsllestad et al. 1984 for details), and a good linear relationship was found within a range of 8.7176.5 mmol glycosyl units kg-l wet wt (OD(PAS) = 0.02+0.0026GC, r = 0.91, n = 20). GC is the biochemically determined glycogen Concentration. The regression analysis was calculated as the geometric mean analysis (Sokal & Rohlf 1981). Blood samples were drawn into heparinized syringes through a Teflon catheter inserted into an antecubital vein. Two aliquots of blood were immediately transferred to 0 . 4 ~perchloric acid stored on ice. Glycogen, lactate, creatine phosphate (Crp) and adenosine triphosphate (ATP) were analysed enzymatically according to Lowry & Passonneau (1972). Glycogen was determined as glucosyl units kg-' wet wt, and the concentration given as
+
The resting levels and the degree of depletion of glycogen during the high-intensity exercise varied to a large extent between subjects. To analyse changes in glycogen concentration during the 40% VO, max exercise, these data were given as relative changes compared with the values obtained at start of the lowintensity exercise. All values are given as mean+SE. T h e significance of differences between mean values was determined by the Mann-Whitney rank-sum test. Comparisons of temporal changes of glycogen concentration in different muscle fibre types were performed by MANOVA, whereas least-square linear regression analysis was used to examine temporal changes in biochemically determined metabolites (Sokal & Rohlf 1981). A significance level of 5% was adopted.
RESULTS Part I
Muscle lactate concentration was 0.5 0.2 mmol kg-' wet wt at rest and increased to 26.7f 1.6 mmol kg-' wet wt after the three bouts of intensive exercise (112% of VO,max) (P < 0.001) (Fig. la). Over the first 20 rnin of the subsequent low-intensity exercise (40 % of Vo, max), muscle lactate concentration fell to 3.7k0.5 mmol kg-' wet wt (P < 0.001). At the end of the 60 rnin of low-intensity exercise a further decrease to 2.1 0.4 mmol kg-' wet wt was observed (P < 0.05). The average rate of lactate disappearance in muscle was 1.15 mmol kg-' wet wt min-' over the first 20 rnin and 0.04 mmol kg-' wet wt min-' over the last 40 rnin of low-intensity exercise. Blood lactate concentration was 0.8 'r 0.2 mmol 1-' at rest, 12.8 & 1.4 mmol I-' before the start of low-intensity exercise (t = 0 min) and peaked at 14.4f 1.1 mmol 1-' 4 rnin into the subsequent 40% Vo, max exercise (P < 0.001) (Fig. la). Thereafter a gradual decline was observed. After 21 min blood lactate concentration (5.2 & 0.4 mmol 1-') was significantly higher than muscle lactate concentration (P < 0.025), creating a favourable gradient for lactate uptake by the muscle cells. Blood lactate had declined futher to 1.4k0.2 mmol 1-' after 60 min.
478
K. Nordheim and N K. Vdlestad (a)
30
0 (b) 0.16
60
D
@
c
2 .-
0.12
0.4) when lactate concentration decreased rapidly. Over the last 40 rnin a glycogen depletion of 21 + 4 mmol kg-' wet wt was observed (P
0.5). I n the next 15 min the glycogen concentration increased by 25 f9 (P < 0.002) and 19 f9 mmol kg-' wet wt (P < 0.02) in type I I A and IIAB fibres respectively. These values correspond to average rates of resynthesis of 1.7 and 1.3 mmol kg-' wet wt min-' respectively. Over the last 40 rnin of the low-intensity exercise, when lactate was low and declined
Glycogen and 1a.cfate metabolism 75
50 --
x
a 0
+I +I +I +I LnNtN
-
+I ?-4
*
t-mmf
-50
-25
0
25
50
m
75
Glyzogen conc. for type HA (rnmol kg-')
Fig. 2. Changes in glycogen concentration (calculated from PAS stain intensity) for type IIAB and IIB fibres vs type IIA fibres during the first 20 min of low-
intensity exercise following high-intensity exercise. The straight line is the line of identity. Fibre type IIAB (A), IIB ( 0 ) . slowly, no significant changes in glycogen concentration of type IIA and IIAB fibres were observed ( P > 0.5). A significant fraction of type IIB fibres was found in only three of the seven subjects. Hence, no statistical analysis of the temporal change of this fibre type was carried out. However, when the changes in the glycogen concentration in type IIAB and IIB fibres are plotted against the corresponding changes in type IIA fibres the data points are distributed around the line of identity (Fig. 2). These correlations indicate about the same the rate of glycogen resynthesis for the three subgroups of type I1 fibres. T o check the validity of the histochemical results we also measured glycogen concentration biochemically. This analysis on muscle samples with all types of fibres showed a glycogen depletion of 48 i7 mmol kg-' wet wt during the high-intensity exercise (Table 2). During the first 20 min of the following low-intensity exercise, glycogen concentration increased by 11 f6 mmol kg-' wet wt ( P < 0.05), whereas a shift towards glycogen breakdown rather than synthesis was seen over the subsequent 40 min. Considering that type I and I1 fibres constituted 52% and 48% of the vastus lateralis muscle mass (Table 1) respectively, the two measurements of glycogen changes were almost identical. During the three bouts of intensive exercise,
0
0
9N.9
z
m - 0 0
m
21
+I +I +I +I LnG??W 00
*
2-2 0
s
?N.9
N 0
m o o 0 +I +I +I +I -LnmLn m - 4 s
-
d-
+I a,
d-
0
Ln
0
* 2
??19
2
2 N O O
+I +I +I +I
+I cQir4s 2
-
O N 0 0
0 m
"C
0
m i 0 0 +I +I +I +I
Ln
m
-tl
-ewG?00
m
Z-IZ
Ln
0
N C
'9 h! 9 m o o 0
m
0
Y
!3 h
h h
m
479
480
K. Nordheim and N . K. Vdlestad
CrP concentration fell by 71 yo ( P < 0.001) (Table 2). Following 5 min of 40% VO, max exercise, the CrP concentration had increased to 70% of the resting value and remained at this level for the rest of the low-intensity exercise ( P > 0.4). T h e ATP concentration declined by 25 yo during the high-intensity exercise ( P < 0.025), and this decrease could partly be ascribed to the rise in water content, as shown by a decreased protein concentration (Table 2). No significant changes in ATP concentration were observed during the 40% VO, max exercise (P > 0.1).
(a’
301
T h e changes in protein concentration indicate a 12% increase in water content from rest to after the high-intensity exercise (Table 2). No significant changes were observed during the low-intensity exercise ( P > 0.5), indicating that in this period changes in the measured metabolite concentration reflect alterations in amount.
Part 11 The temporal association between high lactate concentration on the one hand and absence of glycogen depletion in active fibres and glycogen
r 30
0
Time (min)
Fig. 3. Changes in metabolites during low-intensity exercise (LE, 40% VO,max) 60 min after the lactate last high-intensity exercise (HE, 111 % VO, max). (a) Muscle ( 0 )and blood (0) concentration. (b) Glycogen concentration in type I IIA (A) and IIAB (m) fibres. T h e right axis was calculated as described in Materials and Methods. Pre-exercise rest values (R) are shown at the left. Data are given as mean SE for five subjects. Deviating numbers of subjects are shown in the graph.
(a),
Glycogen and lactate metabolism resynthesis in inactive fibres during low-intensity exercise on the other hand suggests that lactate is the preferred substrate for both glycogen synthesis and ATP resynthesis. T o test the validity of this theory five subjects were studied in a separate series of experiments in which muscle lactate concentration was reduced before the start of low-intensity exercise by a 60-min rest. Muscle lactate concentration rose to 26.7 2.2 mmol kg-' wet wt after three bouts of high-intensity exercise from a pre-exercise resting value of 0.6k0.2 mmol kg-' wet wt at rest (Fig. 3a). This change was identical to that observed in part I (Fig. la). Before the start of the 40% Vo, max exercise muscle lactate concentration had declined to 2.4 f0.6 mmol kg-' wet wt, which was only 9% of the lactate concentration found at the start of low-intensity exercise in part I. Muscle lactate concentration remained low and reached 1.0f0.2 and 0.7 +_ 0.3 mmol kg-' wt wt after 20 and 60 rnin of low-intensity exercise respectively. Blood lactate concentration was 0.9 f 0.1 mmol I-' at rest, 13.9f 1.7 mmol 1-' after three bouts of high-intensity exercise (Fig. 3a) and peaked at 17.8+ 1.0 mmol 1-' after 5 min of the following rest period (data point not shown). At the start of the 4076 VO, max exercise after the 60-min rest period, blood lactate concentration was 3.9 f0.3 mmol l-', which was higher than the concentration in muscle (P < 0.05). A gradual decrease to 1.2 f0.1 mmol 1-' occirred over the 60 min of exercise (P< 0.05). Glycogen concentration determined as optical density (OD(PAS)) decreased by 36 f8, 59 f 11 and 54 f 12 mmol kg-' wet wt in type I, IIA and IIAB fibres respectively during the five intensive exercise bouts and the two long rest periods (Fig. 3 b). In type I fibres, the glycogen concentration decreased gradually from the start of the 40% VO, max exercise by 0.44 f0.15 mmol kg-' wet wt min-' (P< 0.03). This depletion rate was not significantly different from the rate calculated for the last 40 rnin in part I (P> 0.87). No significant changes were detected in type I1 fibres during the 60 rnin of exercise at 40% Vo, max (P> 0.6). Biochemical analysis of muscle glycogen showed a glycogen reduction of 4 0 f 6 mmol kg-' wet wt from rest to the start of the lowintensity exercise (Table 2). A further decline of glycogen concentration by 15 +_ 5 mmol kg-' wet
481
wt during the low-intensity exercise was observed
(P< 0.03). Hence the average rate of glycogen breakdown was about 0.25 mmol kg-' wet wt min-', which is in close agreement with the depletion measured with OD(PAS) considering that type I fibres constituted about 45% of the vastus lateralis muscle mass of these subjects.
DISCUSSION During exercise at 40% Vo, max, we observed a clear temporal association between high lactate levels on the one hand and stable glycogen levels in type I and glycogen resynthesis in type I1 fibres on the other hand. When lactate levels were low, a continuous glycogen breakdown was seen in type I fibres and no resynthesis was observed in type I1 fibres. These relationships strongly indicate that lactate was oxidized in type I fibres and also used to synthesize glycogen in type I1 fibres. I Heany exercase-lactate
accumulation
The glycogen breakdown during the highintensity exercise was accompanied by a large accumulation of lactate. The glycogen reduction was 34mmol kg-'wet wt in type I and 61-63 mmol kg-' wet wt in the subgroups of type I1 fibres. A similar relationship between the major fibre types has also been reported for one bout of exhaustive exercise with a duration of 0 . 5 4 rnin (Tabata et al. 1985). Assuming a total active muscle mass of 13 kg, aerobic combustion of glycogen would have required about 35 10,in type I and 65 1 0, in type I1 fibres. With a maximal oxygen uptake of 3.9 1 min-' a total oxygen consumption of well below 20 10,would be expected during the high-intensity exercise. Hence, in both type I and type I1 fibres a large proportion of the glycogenolysis must have resulted in production and accumulation of lactate. Using repeated exercise bouts it is possible to obtain high lactate levels in both muscle and blood (Hermansen & Stensvold 1972; Hermansen & Vaage 1977). With an elevated blood concentration lactate is also taken up by inactive muscles, and probably by other tissues (Karlsson et al. 1975). In the present study the 12 rnin with high lactate levels allowed a good equilibration between blood and other tissue fluids. The rate
482
K. Nordheirn and N . K. Vdlestad
of lactate transport is reported to be about 12 mmol kg-' min-' (Juel & Wibrand 1989). One may therefore expect that in the present study the blood lactate concentration reflects the concentration of most of the body fluids, which are estimated to constitute about 55% of the body mass (Freund & Gendry 1978). Lom-tntensil,y exercise---high lactate
Over the first 20 min of the 40y0 Vo2 max exercise average muscle lactate declined by 24 mmol kg-I wet wt. Earlier studies have shown that blood lactate disappears faster during light exercise compared with resting recovery (Jervell 1928, Rammal & Strom 1949, Hermansen & Stensvold 1972). T h e average glycogen concentration, as determined biochemically in mixed muscle samples, showed a significant increase of 11 mmol kg-' wet wt over the first 20 min of light exercise following immediately after highintensity exercise (Table 2). This increase is in good agreement with the maintained glycogen concentration in type I fibres and an increase of 19-25 mmol kg-' wet w t in type I1 fibres. In the submaximal exercise, type I fibres were presumably the active fibres (V~llestad& Blom 1985), and hence these fibres had the highest ATP demand. Since glycogen concentration remained virtually unchanged while lactate was high, ATP must have been resynthesized from substrates other than glycogen. Lactate could serve this purpose, and the rapid reduction in lactate is consistent with this postulate. Two alternative substrates, glucose and free fatty acids (FFA), can provide only a small fraction of the ATP demand in the active fibres. During light exercise, without prior heavy exercise, glucose uptake can supply about 40y0 of the energy (Ahlborg et al. 1986, Richter et al. 1988), whereas lipid oxidation may account for about 35% (Vdlestad & Blom 1985). However, in the present study the profound acidosis present in the first 20 min of light exercise would have inhibited both glucose uptake and lipolysis (Issekutz et al. 1965, Paul & Issekutz 1967). Hence, the intracellular availability of glucose and FFA was lower than when light exercise was carried out without preceding intense bouts. In resting type I1 fibres, glycogen was synthesized when lactate was high. T h e rate of resynthesis in the type I1 fibres was 10 times faster than when blood-borne glucose was the
main substrate (Vdlestad et a/. 1989). Accordingly, only a minor fraction of the observed glycogen resynthesis could possibly be ascribed to glucose uptake, leaving lactate as the only candidate. This conclusion is in good agreement with previous studies of human and rat muscle during inactive post-exercise recovery. Hermansen & Vaage (1977) examined glycogen and lactate changes in muscle biopsies and fluxes of blood glucose and lactate over the leg, and they concluded that more than 75% of the lactate produced was converted to glycogFn in the first 30 min of the recovery period. Astrand et a/. (1986) examined this in more detail by also measuring the splanchnic flow and lactate uptake, and it was calculated that over the first hour of recovery a maximum of 50 yo of the disappearing lactate could by accountkd for by uptake in the liver and by oxidation. Hence, more than 50% of the lactate was probably resynthesized to glycogen in the leg muscles. More direct evidence was obtained by other investigators (McLane & Holloszy 1977, Shiota et al. 1984) who perfused isolated rat hindlimbs with '*C-labelled lactate and recovered the isotope in glycogen of the type I1 muscle fibres. Our results indicate that glycogen was resynthesized at the same rate in all subgroups of type I1 fibres. This finding compares well with our observation of similar rates of glycogen resynthesis in type IIA, IIAB and IIB fibres after prolonged exhaustive exercise when glucose was the substrate (Vdlestad et al. 1989). I t thus appears that the subgroups of type I1 fibres within one muscle behave in a similar manner with respect to synthesis of glycogen. Studies of rat muscle show that type I fibres (soleus) do not resynthesize glycogen from lactate because of a very low level of fructose biphosphatase activity (McLane & Holloszy 1977). Whether the same is true for type I fibres within a human mixed muscle remains to be seen. However some information may be obtained by comparing our data for type I1 fibres with the resting postexercise data of Hermansen & Vaage (1977) for muscle samples consisting of both type I and I1 fibres. Based on the rate of glycogen resynthesis observed in the present study and a fibre distribution of 48% type 11, the average gluconeogenetic rate for the mixed muscle would be about 0.6 mmol kg-' wet wt min-' without any resynthesis in type I fibres. Hermansen & Vaage observed an average rate of
Glycogen and lactate metabolism 0.55 mmol kg-' wet wt min-'. This calculation suggests that human type I fibres also synthesize glycogen from lactate at a much slower rate than type I1 fibres. T h e increase in average C r P concentration of the muscle fibres over the first 5 rnin of light exercise indicates that C r P repletion occurs in type I1 fibres. During this period, glycogen resynthesis in type I1 fibres was lower than for the subsequent 15 min, suggesting that synthesis of glycogen was inhibited until C r P was repleted. Earlier observations of a continued decline in glycogen concentration in the first minutes after heavy exercise supports this view (Hermansen & Vaage 1977). Low-intensity exercise-low
lactate
Uptake of lactate in muscle fibres is linearly related to the extracellular concentration (McLane & Holloszy 1977). Hence, a decreased importance of lactate as a substrate for gluconeogenesis and oxidation would be expected as the muscle and blood concentration is lowered. I n fact, this is what we observed. After 20 min low-intensity exercise in part I and at the start of low-intensity exercise of part 11, muscle lactate concentration was below 4 mmol kg-' wet wt and blood lactate concentration was below 6 mmol kg-' wet wt. Under these conditions no resynthesis of glycogen was observed in type I1 fibres and glycogen declined continuously in type I fibres. I n a previous study we showed a continuous breakdown of glycogen in type I fibres and only minor changes in type I1 fibres during exercise at 43% of VO, max (Vdlestad & Blom 1985). T h e same pattern was found in part I1 of the present study and after 20 min of low-intensity exercise in part I. T h e average rates of glycogen depletion in type I fibres were not significantly different in the two parts of the present study. Hence, glycogen metabolism during exercise at low lactate levels seems to be determined by the current metabolic status, and is only marginally influenced by high lactate levels in the early phase of low-intensity exercise. At low lactate levels, blood glucose is the main substrate for muscle glycogen synthesis. T h e rate of glycogen resynthesis from glucose after exercise is about 10 mmol kg-' wet wt-' (Vdlestad et 01. 1989) and would hardly be detected within the short observation times used
483
in the present study. T h e finding of unchanged glycogen levels in type I1 fibres when muscle lactate was lower than 4 mmol kg-' wet wt is in good agreement with the data presented by Brooks & Gaesser (1980). Using tracer techniques and low lactate concentrations (below 5 mM), they reported that most of the eliminated lactate was oxidized and only a minor fraction resynthesized to glycogen. These results cannot be compared with conditions where lactate concentration is high, as during the first 20 rnin of low-intensity exercise of part I and in the ?tudies by Hermansen & Vaage (1977) and by Astrand et al. (1986). T h e present results strongly indicate a dual effect of lactate upon glycogen metabolism during low-intensity exercise. T h e absence of glycogen depletion in type I fibres when lactate levels are high indicates that glycogenolysis is not regulated by activity only, but can also be modulated by the metabolic status. T h e close correlation between high lactate levels and glycogen resynthesis in type I1 fibres indicates that lactate is incorporated into glycogen. Hence, high lactate levels can serve to spare glycogen in active fibres and restore glycoten levels in inactive fibres. We thank Professor Ole M. Sejersted for valuable comments and suggestions and Roald A. Bjerklund for help with statistical analysis. Excellent technical
assistance from Ada Ingvaldsen, Bjerg Ingrid Selberg and Jorid Thrane Stuenres is gratefully acknowledged. Financial support was received from The Norwegian Confederation of Sports and from The Norwegian Research Council for Science and Humanities.
REFERENCES AHLBORG, G., WAHREN,J. & FELIG, P. 1986. Splanchnic and peripheral glucose and lactate metabolism during and after prolonged arm exercise. 3 Clin Invest 77, 690-699. ASTRAND, P . - 0 . HULTMAN, E., JUHLIN-DANNFELT, A. & REYNOLDS, G. 1986. Disposal of lactate during and after strenuous exercise in humans. 3 Appl Ph,ysiol 61, 338-343. BERGSTROM, J. 1962. Muscle electrolytes in man. Scand 3 Clin Lab Invest 14 (Suppl. 68), 11-13. BROOKS, G.A. & GAESSER, G.A. 1980. End points of lactate and glucose metabolism after exhausting exercise. 3 Appl Physiul49, 1057-1069. FREUND, H. & GENDRY, P. 1978. Lactate kinetics after short strenuous exercise in man. E u r 3 Appl Physiul 39, 123-135.
484
K . Nordheim and N . K. Vollestad
HERMANSEN, L. & STENSVOLD, I. 1972. Production and removal of lactate during exercise in man. Acta Ph,ysiol Scand 86, 191-201. L. & VAAGE,0. 1977. Lactate disHERMANSEN, appearance and glycogen synthesis in human muscle after maximal exercise. A m 3 Physiol 233, E422ZE429. ISSEKUTZ JR, B.J, MILLER,H.I., PAUL,P. & RODAHL, K. 1965. Effect of lactic acid on free fatty acids and glucose oxidation in dogs. Am 3 Physiol 209, 1137-1 144. JERVELL, 0. 1928. Investigation of the concentration of lactic acid in blood and urine. Acta Med Scand SUPPI. 24, 1-135. JUEI., C. & WIBRAND, F. 1989. Lactate transport in isolated mouse muscles studied with a tracer technique kinetics, stereospecificity, pH dependency and maximal capacity. Acta Physiul Scand 137, 33-39. KARISON,J., BONDE-PETERSEN, F., HENRIKSSON, J. & KNUTTGEN, H.G. 1975. Effects of previous exercise with arms or legs on metabolism and performance in exhaustive exercise. 3Appl Physiol38, 763-765. LOWRY,O.H. & PASSONNEAU, J.V. 1972. A Flexible System of Enzymatic Analysis. Academic Press, New York. MCLANE,J.A. & HOLLOSZY,J.O. 1977. Glycogen synthesis from lactate in the three types of skeletal muscle. 3 Biol Chem 254, 6548-6553. MEDBB, J.I. & SEJERSTED, O.M. 1985. Acid-base and electrolyte balance after exhausting exercise in endurance-trained and sprint-trained subjects. Acta Physiol Scand 125, 97-109. MEDBB, J.1. & TABATA, I. 1989. Relative importance of aerobic and anaerobic energy release during shortlasting, exhausting bicycle exercise. 3 Appl Physiol 67, 1881-1886. MEDB0, J.I., MOHN, A.C., TABATA,I., BAHR,R., VAAGE,0. & SEJERSTED, O.M. 1988. Anaerobic capacity determined by maximal accumulated 0, deficit. 3 Appl Physiol 64, 50-60. -
NORDHEIM, K. & VBLLESTAD, N.K. 1987. Can lactate be used as a fuel in active fibres and simultaneously be converted to glycogen in inactive fibres within the same muscle? Acta Ph,ysiol Scand 129, 45A. JR, B. 1967. Role of extramuscular PAUL,P. & ISSEKUTZ energy sources in the metabolism of the exercising dog. 3 Appl Physiol22, 615-622. RAMMAL,K. & STROM,G. 1949. The rate of lactate utilization in man during work and at rest. Acta Physiol Scand 17, 452456. RICHTER,E.A., KIENS,B., SALTIN, B., CHRISTENSEN, N.J. & SAVARD, G. 1988. Skeletal muscle glucose uptake during dynamic exercise in humans : role of muscle mass. A m 3 Physiol 254, E55j-Ej61. SHIOTA,M., GOLDEN,S. & KATZ,J. 1984. Lactate metabolism in the perfused hindlimb. Biochem f 222, 281-292. SNEDECOR, G.W. & COCHRAN, W.C. 1971. Statistical Methods. The Iowa University Press, Ames, Iowa. SOKAL, R.R. & ROHLF,F.J. 1981. Biometry. Freeman, San Francisco. TABATA, I., MEDBB, J.I., V0LLESTAD, N.K. & SEJERSTED, O.M. 1985. Glycogen breakdown and lactate accumulation during shortlasting exercise in normal humans. Clin Ph,ysiol abstr. no. 85. VBLLESTAD, N.K. & BLOM, P.C.S. 1985. Effect Of varying exercise intensity on glycogen depletion in human muscle fibres. Acta Physiol Scand 125, 3955405. L. 1984. VBLLESTAD, N.K., VAAGE, 0. & HERMANSEN, Muscle glucogen depletion patterns in type I and subgroups of type I1 fibres during prolonged severe exercise in man. Acta Physiol Scand 122, 4335441. VBLLESTAD, N.K., BLOM,P.C.S. & GRBNNERBD, 0. 1989. Resynthesis of glycogen in different muscle fibre types after prolonged exhaustive exercise in man. Acta Physiol Scand 137, 15-21. ZAMENHOF,S. 1957. In S. P. Colowick & N. 0. Kaplan (eds.) Methods in Enzymology, pp. 696-703. Academic Press, New York.
