Acta Physiol Scand 2000, 168, 473480

Human skeletal muscle mitochondrial capacity

U . F . R A S M U S S E N and H . N . R A S M U S S E N
Department of Biochemistry, August Krogh Institute, University of Copenhagen, Denmark

ABSTRACT
Under aerobic work, the oxygen consumption and major ATP production occur in the mitochondria
and it is therefore a relevant question whether the in vivo rates can be accounted for by mitochondrial
capacities measured in vitro. Mitochondria were isolated from human quadriceps muscle biopsies in
yields of 45%. The tissue content of total creatine, mitochondrial protein and different cytochromes
was estimated. A number of activities were measured in functional assays of the mitochondria:
pyruvate, ketoglutarate, glutamate and succinate dehydrogenases, palmitoyl-carnitine respiration,
cytochrome oxidase, the respiratory chain and the ATP synthesis. The activities involved in
carbohydrate oxidation could account for in vivo oxygen uptakes of 1516 mmol O2 min1 kg1 or
slightly above the value measured at maximal work rates in the knee-extensor model of Saltin and
co-workers, i.e. without limitation from the cardiac output. This probably indicates that the maximal
oxygen consumption of the muscle is limited by the mitochondrial capacities. The in vitro activities of
fatty acid oxidation corresponded to only 39% of those of carbohydrate oxidation. The maximal rate
of free energy production from aerobic metabolism of glycogen was calculated from the mitoc-
hondrial activities and estimates of the DG for ATP hydrolysis and the efciency of the actinmyosin
reaction. The resultant value was 20 W kg1 or 70% of the maximal in vivo work rates of which
1020% probably are sustained by the anaerobic ATP production. The lack of aerobic in vitro ATP
synthesis might reect termination of some critical interplay between cytoplasm and mitochondria.

Keywords ATP synthesis, efciency of ATP utilization, isolated mitochondria, metabolic control,
oxygen uptake, quadriceps muscle, respiratory rates.

Received 18 November 1999, accepted 9 December 1999

Studies of isolated mitochondria have led to solution of
many important problems in bioenergetics. In the
aerobically working muscle, practically all oxygen
consumption and most ATP formation occur in the
mitochondria. It is therefore a relevant question how the
mitochondrial capacities, studied in isolation, relate to
the respiratory rates and work rates of the muscles in vivo.
Under normal conditions of work, several big
muscle groups are in operation and the oxygen uptake
is known to be limited by the cardiac output. The
respiratory rate based on whole body weight is below
2 mmol O2 min1 kg1 (Saltin & Rowell 1980) or
45 mmol O2 min1 kg muscle1 if all oxygen consump-
tion were associated with muscle work. In the
knee-extensor model of Saltin and co-workers, work is
conned to 23 kg quadriceps muscle and respiratory
rates of 1016 mmol O2 min1 kg muscle1 are
normally obtained (Andersen & Saltin 1985, Andersen
et al. 1985, Blomstrand et al. 1997). Under these
conditions, the rates of respiration are not limited by
the circulation, but possibly by the mitochondrial
capacities.
The mitochondrial part of metabolism includes the
substrate accessibility and transport into the mito-
chondria, the pyruvate dehydrogenase reaction or the
b-oxidation system, the reactions of the Krebs cycle,
the transfer of reducing equivalents through the respir-
atory chain and nally the ATP synthesis and the
transfer reactions for ADP, ATP and phosphate. Most
of these reactions may be studied with isolated mito-
chondria and a small-scale preparation of skeletal
muscle mitochondria seemed very favourable in this
respect. In a joint study with the Copenhagen Muscle
Research Centre, the approach was to compare rates
obtained from in vitro measurements with in vivo
measured oxygen uptake and work rate of the knee-
extensors. Strong correlations were obtained between
in vivo and in vitro measurements (unpublished).
Mitochondria have been intensively studied by
metabolic control analysis, based on theories originally

Correspondence: Ulla F. Rasmussen, Department of Biochemistry, The August Krogh Institute, Universitetsparken 13, DK-2100 Copenhagen,
Denmark.

2000 Scandinavian Physiological Society 473

Human skeletal muscle mitochondrial capacity
U F Rasmussen and H N Rasmussen
formulated by Kacser & Burns (1973, 1995). By means
of the relative ux control coefcients, it is possible to
characterize quantitatively the metabolic control of an
integrated system. The mitochondrial metabolism is
especially suited for these kinds of studies because a
large number of specic inhibitors are available. Most
often the substrate has been succinate, but systems
involving physiologically more relevant substrate
combinations have also been studied (Moreno-Sanchez
et al. 1991). It is generally agreed upon that environ-
mental metabolic control is shared by a number of
reactions. The distribution of control depends, however,
on the conditions (for a review see Brown 1992).
In this paper, we focus on a number of mito-
chondrial activities and the possible maximal in vivo
capacities estimated from these. The activities were
measured in functional assays in which the rates of
respiration of the isolated mitochondria were recorded
under conditions where a single enzyme controlled the
rate and the oxygen stoichiometry was known. Mito-
chondrial matrix enzymes are associated with each
other and with the respiratory chain (Srere 1987, Srere
et al. 1997). Such associations will not be affected in the
functional assays. The important nding was a close
correspondence between the pyruvate dehydrogenase
capacity, the Krebs cycle capacity and the respiratory
chain capacity. The mitochondrial capacity, further-
more, matched the maximal respiratory rates measured
in the above-mentioned study with the knee-extensor
model. With respect to ATP production, the rates
measured in vitro can account for the energy needed for
all normal activities, but not entirely for the energy
needed for the highest work rates.
MEASUREMENTS
The mitochondria were isolated from single needle
biopsies from the vastus lateralis part of m. quadriceps
femoris of healthy young men. Equipment and meth-
odology have been described by Rasmussen et al. (1997)
and Rasmussen & Rasmussen (1993, 1997). The study
was approved by the Ethical Committees of the
Municipalities of Copenhagen and Frederiksberg and
the volunteers were informed in accordance with the
rules.
Table 1 shows data for the mitochondrial prepar-
ation and the markers. All fractions in the isolation
were sampled and analysed for activities of the marker
enzymes pyruvate kinase (cytoplasm) and citrate
synthase (mitochondria). The calculation of the yield of
mitochondria was based on the citrate synthase data.
Low-temperature spectroscopy of the fractions made
possible, for instance, estimation of the content of
cytochrome aa3 and the loss of cytochrome c from the
mitochondria. Furthermore, the relevant fractions were
474
Acta Physiol Scand 2000, 168, 473480
Table 1 Data for the human quadriceps muscle and for the mito-
chondrial preparation
Mean SD
Range (n = 12)
Biopsy (mg) 68108 91 14
Mitochondrial yield (%) 3956 45 5
Mitochondrial protein (g kg)1) 8.713.2 10.3 1.3
Cytochrome aa3 (lmol kg)1) 3.99.9 6.0 1.6
Citrate synthase (mmol min)1 kg)1) 20.038.0 25.8 4.5
Pyruvate kinase (mmol min)1 kg)1) 176285 223 35
S Creatine (mmol kg)1) 25.333.2 28.2 2.6
The markers are referred to tissue mass (wet weight). S creatine is
creatine + phosphocreatine + creatinine. The enzyme activities were
measured at 25 C.
analysed for total creatine content by an enzymatic
method with spectrophotometric measurement.
The yield of about 400-lg mitochondrial protein
sufced for at least 25 single respiratory experiments
made with different substrates under coupled, uncou-
pled and non-coupled conditions regarding ATP
synthesis.
The sampling uncertainty is of critical importance
when measurements made on biopsy material are
extrapolated to muscle mass. The random sampling
error originates from the uncertainty of the biopsy mass
determination (a few percent) and the heterogeneity of
the tissue with respect to muscle bres. The data in all
tables were obtained with preparations from 12
volunteers, i.e. the data reect the biological variation
within this group in addition to the sampling and
measurement errors. Additional preparations were
made from six of these volunteers and the standard
deviation, calculated by pairing creatine per muscle
mass of these six, was 2.3 mmol kg1 or about 8% of
the content (6 degrees of freedom). This standard
deviation most likely represents the random sampling
error, as creatine per muscle bre mass is constant for
the subject and the random error of the analysis was
negligible. As shown in Table 1, the standard error of
creatine per muscle mass was 9%, i.e. the data do not
indicate signicant biological variation of this param-
eter within this group of young men.
The tissue activity of citrate synthase was close to
the value measured by Blomstrand et al. (1997).
M IT O C HO N D R I AL S PE C I FI C
ACTIVITIES
Table 2 presents specic activities, i.e. activities per
mitochondrial protein, of some central reactions. Most
of the activities were obtained in functional assays in
which was measured the maximal respiratory rate with
the proper physiological substrate and ADP or
2000 Scandinavian Physiological Society
Acta Physiol Scand 2000, 168, 473480 U F Rasmussen and H N Rasmussen
Human skeletal muscle mitochondrial capacity
Table 2 Specic activities of human quadriceps mitochondria
Pyruvate dehydrogenase (PDH) 209 24
a-Ketoglutarate dehydrogenase (KgDH) 223 18
Glutamate dehydrogenase (GDH) 238 66
Succinate dehydrogenase (SDH) 539 49
b-Oxidation (2-C units) 81 11
Respiratory chain (NADH-oxidase) 1186 183
Cytochrome oxidase (COX, O-atoms) 3679 545
ATP synthesis 1148 90
The activities are given in lmol min)1 g)1 of mitochondrial protein.
The measurements were made at 25 C under conditions of maximal
rates (e.g. ADP or uncoupler present). The activities of the respiratory
chain and cytochrome oxidase were measured with freeze-permea-
bilized mitochondria. Figures are mean values SD (number of
subjects).
uncoupler present. Conditions were selected where the
rate was limited by the activity of a single enzyme and
the oxygen:substrate stoichiometry was known
(Rasmussen & Rasmussen 2000).
Pyruvate dehydrogenase (PDH) activity was assayed
in the presence of malate in order to recycle the
mitochondrial CoA and calculated as half the measured
activity. The enzyme was most likely in its active form
in the mitochondria. Ketoglutarate dehydrogenase
(KgDH) was assayed by the respiratory rate with
a-ketoglutarate in the presence of malonate to prevent
succinate oxidation. Glutamate dehydrogenase (GDH)
was assayed by initial respiratory rates, i.e. before
products accumulated and eventually were oxidized.
Succinate dehydrogenase (SDH) was assayed in the
presence of rotenone to prevent malate oxidation. This
enzyme required some time to reach maximal activity.
Fatty acid b-oxidation was assayed with palmitoyl-
carnitine and malate to recycle CoA. The rate of
respiration was most likely not limited by the activity of
malate dehydrogenase and 2/3 of the respiratory rate
was caused by b-oxidation. The activity of b-oxidation,
in 2-C units/time, was therefore calculated as one-third
the respiratory rate, in O-atoms/time.
The NADH oxidase and cytochrome oxidase
(COX) activities were assayed in freeze-permeabilized
mitochondria. This freeze-permeabilization was critical
and in particular the NADH oxidase activity might be
underestimated. But the procedure was optimized and
the NADH oxidase activities are considered represen-
tative for the activity of the respiratory chain (cf.
Rasmussen & Rasmussen 2000). The COX assay was
not a real functional assay, because the substrates,
ascorbate and NNN N tetramethyl-p-phenylenedi-
amine are not physiological substrates and the maximal
rate could not be measured. But, compared with the
NADH oxidase activity, the reaction of COX with
oxygen is fast and may occur at sufcient rates even at
lower oxygen tensions.
2000 Scandinavian Physiological Society
The rate of ATP synthesis was calculated from the
state 3 rates and the actually measured P/O ratios. The
(12)
(9)
same rate or almost the same rate was obtained with
(12) three different substrate combinations and it may be
(10) regarded as the maximal rate of ATP synthesis, which
(12) can be obtained with these mitochondria (Rasmussen &
(10) Rasmussen 2000).
(12)
(12)
Table 2 shows that PDH apparently has the lowest
activity. Complete oxidation of pyruvate formed
from carbohydrates and oxidized in the Krebs cycle
has an O-atom to pyruvate stoichiometry of 6.
One-sixth of the respiratory chain activity (198 lmol
O-atoms min1 g protein1) corresponded closely to
the PDH activity.
The KgDH activity was not signicantly different
from the PDH activity and agreed with the tissue
activity measured by Blomstrand et al. (1997). These
authors suggest that the KgDH activity may be taken as
representative for the Krebs cycle activity. It is there-
fore of interest that the activities of PDH and the
respiratory chain match the KgDH activity. The GDH
activity was remarkably high in human mitochondria in
comparison with other skeletal muscle mitochondria.
The activity showed a high standard deviation, perhaps
because of increased biological variation.
The SDH activity was clearly in excess of the PDH
activity. If succinyl-CoA is formed directly in metabo-
lism, e.g. from valine, the extra enzyme activity may be
advantageous for the anaplerotic function of the Krebs
cycle.
I N V IT RO T O I N V I V O
EXTRAPOLATIONS
The activities measured with isolated mitochondria and
saturating substrate concentrations may be extrapolated
to in vivo conditions by multiplication with the average
content of mitochondrial protein and a temperature
factor adjusting the rates from 25 to 38 C.
Some extrapolated rates are shown in Table 3. The
enzyme activities on which they are based are indicated
together with the overall reactions. The rate limited by
PDH is the rate which pyruvate formed glycolytically
and oxidized completely in the Krebs cycle. It is very
close to the maximal rate of in vivo respiration measured
by Blomstrand et al. (1997) and in our joint study of the
knee-extensors. The respiratory chain activity also
matched this maximal in vivo rate.
Complete oxidation of fatty acids via b-oxidation
and the Krebs cycle requires six O-atoms per 2-C unit.
As seen from Table 3 the extrapolated rate of respir-
ation corresponded to only 39% of the rate of complete
carbohydrate oxidation. This observation is in accord-
ance with the observation that fatty acids in vivo support
50% of the energy for work rates at 70% of the
475
Human skeletal muscle mitochondrial capacity
U F Rasmussen and H N Rasmussen
Limiting reaction Oxygen stoichiometry
PDH Carbohydrate (3-C) + 3O2 3CO2 + 3H2O
b-Oxidation Palmitoyl-(2-C) + 3O2 2CO2 + 2H2O
Respiratory chain NADH + H+ + O2 NAD+ + H2O
ATP synthesis nADP + nPi + O2 nATP + H2O
(n = P/O = 2.2)
The activities are referred to mass of the quadriceps muscle (wet weight). They are calculated from
the data of Table 2, the tissue temperature of 38 C (factor = 2.44), the content of mitochondrial
protein of 10.3 g kg)1 and the oxygen stoichiometries specied. The maximal oxygen uptake of the
knee-extensors measured in vivo on 6 of the subjects was = 13.2 2.7 mmol O2 min)1 kg)1 .
maximal in vivo oxygen uptake, VO2max (Coyle 1991).
The in vivo limitation is normally considered to be the
supplementation of fatty acids to the mitochondria, but
the present data show that the mitochondrial systems
of the inner membrane and matrix cannot support a
sufcient rate. Furthermore, it is of interest that the
activity is in close agreement with the tissue activity of
carnitine palmitoyltransferase I in human skeletal
muscle reported by Berthon et al. (1998).
The minimal rates of mitochondrial respiration may
also be of relevance in view of the fact that the meta-
bolic rates of skeletal muscles may vary by a factor of
100 (e.g. Hochachka & Matheson 1992). The minimal
or resting rates may be obtained by extrapolation of the
non-phosphorylating (state 4) rates of respiration to
in vivo conditions. The state 4 rates were 44 8 and
42 8 (n 12) lmol O-atom min1 g protein1 with
pyruvate + malate and palmitoyl-carnitine + malate,
respectively. These rates are largely controlled by the
permeability of the mitochondrial inner membrane to
protons (e.g. Rolfe et al. 1994). They therefore apply
directly to the integrated system and the extrapolated
value is 0.55 mmol O2 min1 kg1. Thus, the in vitro
measurements, made at saturating substrate concen-
trations, may account for a ratio between maximal and
minimal respiratory rates of 30. Alteration of the
substrate saturation or additional mechanisms may also
be involved in the in vivo response (cf. Korzeniewski
1998).
Fully coupled mitochondrial respiration is controlled
by the rate of ATP synthesis (Rasmussen & Rasmussen
2000). The ratio between this rate and the rate of
respiration equals the stoichiometric P/O ratio. From
the generally accepted values, carbohydrate oxidation
may be expected to have a P/O ratio of 2.12.3, the
exact value depending on the route of oxidation of the
cytoplasmic NADH. The extrapolated rate of respira-
tion, calculated from the rate of ATP synthesis, corre-
sponded to only 50% of VO2max (Table 3). Supposing
that the mitochondria are loosely coupled under in vivo
476
Acta Physiol Scand 2000, 168, 473480
Table 3 In vivo respiratory activities
Oxygen uptake calculated from in vitro mitochondrial
(mmol O2 min)1 kg)1) activities
15.8
6.1
14.9
6.6
conditions, i.e. that the P/O ratios are lowered, higher
rates of respiration will accompany the same rate of
ATP synthesis.
Submaximal in vivo P/O ratios are not unlikely.
Recently discovered uncoupling proteins may be
effective. One of these, UCP-2, is widely distributed
whereas another, UCP-3, is predominantly expressed in
skeletal muscles (see Schrauwen et al. (1999) for a
review). Loose coupling has even been considered
advantageous as a defence system against reactive
oxygen compounds in vivo (e.g. Starkov et al. 1997). The
P/O ratio could furthermore be lowered at high work
intensities, for instance because of very high Ca2+
concentrations and formation of reactive oxygen
compounds (Richter 1997).
The mitochondrial respiratory activities, extrapolated
to in vivo conditions, appeared to be able to account for
the VO2max. The crucial question is therefore whether
the work rate under VO2max conditions may be
accounted for by the rate of ATP synthesis, calculated
from the in vitro data. The extrapolated rate of mitoc-
hondrial ATP synthesis was 29 mmol min1 kg1 (cf.
Table 2 and the legend for Table 3). Formation of
pyruvate from glycogen is accompanied by formation
of 1.5 mol ATP/mol pyruvate. Some lactate is formed
during knee-extensor excercise, i.e. pyruvate formation
exceeds pyruvate oxidation. Thus the rate of aerobic
cytoplasmic ATP formation may be calculated from the
PDH activity. A rate of 8 mmol min1 kg1 is obtained
and the calculated total rate of aerobic ATP synthesis is
therefore 37 mmol min1 kg1.
The DG for ATP hydrolysis is probably about
53 kJ mol1, which is the value obtained from probable
parameters (cf. Nicholls & Ferguson 1992) and a
cytoplasmic ATP/ADP ratio of 50100 (Lodi et al.
1997). Accordingly, the calculated rate of aerobic ATP
synthesis corresponds to a rate of free energy produc-
tion of 33 W kg1. Comparison of this rate and the
work rate requires an estimate of the efciency of the
actinmyosin reaction.
2000 Scandinavian Physiological Society
Acta Physiol Scand 2000, 168, 473480 U F Rasmussen and H N Rasmussen
Human skeletal muscle mitochondrial capacity
EFFICIENCY CONSIDERATIONS
Work performed on aerobic carbohydrate metabolism
may for the purpose be divided into three steps: (1)
glycogenolysis + glycolysis + Krebs cycle reactions, (2)
respiratory chain and ATP synthesis and (3) cyto-
plasmic energy transfer and actinmyosin reaction. The
overall efciency will be the product of the efciencies
of these three steps. It may be evaluated in the knee-
extensor model from the VO2max, the work rate and the
caloric equivalent of oxygen. Assuming carbohydrate
metabolism from muscle glycogen, it can be calculated
to maximally 30% from the data of Andersen & Saltin
(1985).
The efciency of the respiratory chain and ATP
synthesis in oxidation of NADH is often in textbooks
calculated to be 42%, but as pointed out by Nicholls &
Ferguson (1992) this calculation is incorrect and the
value is too low. The efciency rather is 60% which is the
value obtained from the redox conditions and a DG of
)53 kJ mol1 for ATP hydrolysis. Taking this efciency
as representative for step (2), the product of the ef-
ciencies of steps (1) and (3) equals 50%, namely
the overall efciency divided by 0.6. The efciency of the
actinmyosin reaction, i.e. work done/free energy lost,
has been estimated to be maximally 50% with isolated
frog sartorii muscles at 0 C by Curtin et al. (1974). An
efciency of 50% in step (3) seems improbable, as it
would require 100% efciency in step (1). The efciency
of step (3) apparently must be higher, perhaps as high as
60%, which goes with 80% efciency in step (1). An
efciency for step (1) of this magnitude may still seem
unlikely in consideration of the large number of reac-
tions, but the cytoplasmic reactions are regarded to occur
in highly organized micro compartment structures with a
high degree of metabolic channelling (Ovadi & Orosz
1997). The efciency of step (3) may also be inuenced
by direct energy channelling from the mitochondria to
selected areas in the sarcomeric muscle via the creatine
kinase system (e.g. Wegmann et al. 1992, Wallimann et al.
1992, Saks et al. 1994).
The notion of an efciency of 60% of the actin
myosin reaction is in fair agreement with the extrapolated
initial efciency reported for one-legged knee-extensor
work (Gonzalez-Alonso et al. 2000). The initial phase of
this work should primarily imply the actinmyosin
reaction.
T H E IN V ITRO AT P PR O DUC T ION
The work rate under VO2max conditions was 28 W kg1
in the knee-extensor study. At an efciency of 60% of
the actinmyosin reaction, the calculated aerobic ATP
production may account for a muscle work rate of
20 W kg1. This may sufce for all normal activities
2000 Scandinavian Physiological Society
limited by the circulation, but only for about 70% of
the maximal work rate. The anaerobic ATP produc-
tion has not been included in the balance.
It may account for 1020% of the work rate under
similar conditions as judged from calculations on the
results of Hellsten et al. (1999, J. Bangsbo, personal
communication). The balance calculations therefore
appear to indicate that the rate of ATP production,
calculated from the in vitro measurements, is somewhat
suboptimal. Four explanations for this may be
considered.
The in vitro assay underestimates the rate of ATP synthesis
This appears unlikely. The rate of ATP synthesis was
measured under conditions where it was limiting for the
rate of respiration. Very similar rates were obtained
with different substrate combinations and the P/O
ratios and specic rates of phosphorylating respiration
equalled or exceeded all literature data (Rasmussen &
Rasmussen 2000).
The extrapolation to tissue activity is not correct
The extrapolation of in vitro rates involves three critical
quantities, namely the biopsy mass, the mitochondrial
yield and the temperature factor. Extrapolation from a
90-mg biopsy to a 23 kg muscle is of course a long
shot. But the samples were probably representative for
the tissue. They were not dissected prior to preparation
and the sampling error was less than 10% with respect
to bre mass (e.g. Table 1).
The yield of mitochondria was determined from the
yield of citrate synthase. The tissue activity was deter-
mined by summing the activities of the relevant frac-
tions as the heterogeneity of the homogenate prevented
representative sampling. All fractions were assayed and
the fractionation was quantitatively evaluated as
recommended by DeDuve (1971). Only 7% of the total
activity was not accounted for in the fractions. We may
thus conclude that the estimation of the yield was not
subject to any substantial error.
The yield is also used in the calculation of mito-
chondrial protein pertissuemass. Our value (10.3 mg g1)
is clearly lower than those which may be calculated
from the data of Wibom & Hultman (1990), namely
1623 mg g1. Different purity of the nal mito-
chondrial suspension might explain the difference. The
present mitochondria appeared very pure as indicated
by the high specic rates of respiration and the tissue
content is in accordance with morphometric data (cf.
Hoppeler 1990). It is obvious that specic in vitro rates
should not be extrapolated to the tissue by use of a
value for the tissue content of mitochondrial protein
that was not estimated in the same experiment.
477
Human skeletal muscle mitochondrial capacity
U F Rasmussen and H N Rasmussen
The temperature factor used to adjust the rates from
25 to 38 C was calculated from the Arrhenius equation
for a 2.0 times increase of rate between 25 and 35 C
(data from Byrne & Trounce 1985). The factor is
subject to some uncertainty.
The mitochondria are altered during isolation
The integrity of the quadriceps mitochondria may be
evaluated from the specic respiratory activities, the
respiratory control ratios and the P/O ratios. In
comparison with literature data, all of these appear to
indicate high integrity (Rasmussen & Rasmussen 2000).
The high reproducibility of the preparation also
supports this notion.
It was previously suggested that mitochondrial
integrity might be evaluated from the loss of cyto-
chrome c (Rasmussen & Rasmussen 1997). Spectro-
scopic measurements indicated that the present
mitochondria had lost about 30% of their in situ cyto-
chrome c. But the test parameters mentioned above
were not correlated to the loss of cytochrome c,
although this could be expected. It was therefore
concluded that in this case, the loss of cytochrome c
was not indicative for lowering of the respiratory rates.
The optimum in vivo function cannot be measured in vitro
The insufciency of the in vitro ATP synthesis might
just be a consequence of the removal of the mito-
chondria from the tissue or rather, termination of a
critical interplay between cytoplasm and mitochondria.
However, mitochondria, localized within permeabilized
muscle bres, do respire with the same rates as isolated
mitochondria (Kunz et al. 1993, Saks et al. 1998). This
probably indicates that it is not the mere removal of the
mitochondria that may cause the effect.
ATP synthase is probably a highly regulated enzyme
in vivo, for reviews see Harris & Das (1991) and Brown
(1992). The best-examined tissue is heart, but the
observations probably also apply to skeletal muscle,
which has an even greater need for switching on and
off. ATP synthesis may be regulated by Ca2+, which is
also known to regulate four mitochondrial dehydro-
genases (review by Hansford & Zorov 1998). Cyto-
plasmic Ca2+ concentrations are oscillating in vivo
(Berridge 1990 and Duchen 1999) and the mito-
chondrial accumulation of Ca2+ by one of the pathways
depends on the frequency of the oscillations (Gunter
et al. 1998). The ATP synthase is in vivo also controlled
by two inhibitor proteins, one of which is Ca2+
dependent (cf. Harris & Das. 1991, Brown 1992).
Finally, the rate of the ATP synthase reaction depends
on the occupancy of the three active sites (Boyer 1998)
and may therefore be subject to regulation through
478
Acta Physiol Scand 2000, 168, 473480
local substrate and product concentrations. This might
be inuenced by channelling via the creatine kinase
system (e.g. Wegmann et al. 1992, Wallimann et al.
1992, Saks et al. 1994).
These examples illustrate that the mitochondria,
within the cell, are exposed to conditions that are
different from those in vitro. Seen in this perspective, it
might not be surprising if less than 100% of the ATP
synthesis required at extreme work is measured under
in vitro conditions.
C O N C LU S I O N S
The isolated human quadriceps mitochondria showed
matching activities of the respiratory chain, the Krebs
cycle (KgDH) and PDH. SDH was in 23 times excess.
The in vitro capacities extrapolated to the tissue could
account for the maximal rates of in vivo oxygen uptake,
measured without limitations from the circulation.
Under these conditions the mitochondrial capacities
therefore might be limiting.
A ratio of 30 between fully active and resting
metabolism is calculated from the in vitro measurements
under substrate saturating conditions.
The capacity of the reactions in fatty acid
b-oxidation (from palmitoyl-carnitine) corresponded to
only 39% of the oxygen uptake of carbohydrates.
The rate of aerobic free energy production, calcu-
lated from the in vitro data, could account for about
70% of the maximal work rate or 8085% if the
anaerobic ATP production is also included in the
balance. It is suggested that the insufciency might be
owing to termination of some critical interplay between
cytoplasm and mitochondria.
We thank the members from the Copenhagen Muscle Research
Centre for efcient and inspiring collaboration. The knee-extensor
study is in collaboration with P. Krustrup, B. Quistorff, B. Saltin &
J. Bangsbo. The expert technical assistance of Mrs I-L. Fhns and
H. Lauritzen is gratefully acknowledged.
R E F E RE N C E S
Andersen, P., Adams, R.P., Sjgaard, G., Thorboe, A. &
Saltin, B. 1985. Dynamic knee extension as model for study
of isolated exercising muscle in humans. J Appl Physiol 59,
16471653.
Andersen, P. & Saltin, B. 1985. Maximal perfusion of skeletal
muscle in man. J Physiol 366, 233249.
Berridge, M.J. 1990. Calcium oscillations. J Biol Chem 265,
95839586.
Berthon, P.M., Howlett, R.A., Heigenhauser, G.J.F. & Spriet,
L.L. 1998. Human skeletal muscle carnitine
palmitoyltransferase I activity determined in isolated intact
mitochondria. Am J Physiol 85, 148153.
Blomstrand, E., Radegran, G. & Saltin, B. 1997. Maximum
rate of oxygen uptake by human skeletal muscle in relation
2000 Scandinavian Physiological Society
Acta Physiol Scand 2000, 168, 473480 U F Rasmussen and H N Rasmussen
Human skeletal muscle mitochondrial capacity
to maximal activities of enzymes in the Krebs cycle. J Physiol
501, 455460.
Boyer, P.D. 1998. ATP synthase past and future.
Biochim Biophys Acta 1365, 39.
Brown, G.C. 1992. Control of respiration and ATP synthesis
in mammalian mitochondria and cells. Biochem J 284,
113.
Byrne, E. & Trounce, I. 1985. Oxygen electrode studies with
human skeletal muscle mitochondria in vitro. J Neurol Sci 69,
319333.
Coyle, E.F. 1991. Carbohydrate feedings: effects on
metabolism, performance and recovery. In: F. Brouns (ed.)
Adv Nutrition Top Sport Med Sport Sci, Vol. 32, pp. 114.
Karger, Basel.
Curtin, N.A., Gilbert, C., Kretzschmar, K.M. & Wilkie, D.R.
1974. The effect of the performance of work on total
energy output and metabolism during muscular
contraction. J Physiol 238, 455472.
DeDuve, C. 1971. Tissue fractionation: past and present. J Cell
Biol 50, 20D55D.
Duchen, M.R. 1999. Contributions of mitochondria to animal
physiology: From homeostatic sensor to calcium signalling
and cell death. J Physiol 516, 117.
Gonzalez-Alonso, J., Quistorff, B., Krustrup, P., Bangsbo, J.
& Saltin, B. 2000. Heat production in human skeletal
muscle at the onset of intense dynamic exercise. J Physiol
in press.
Gunter, T.E., Buntinas, L., Sparagna, G.C. & Gunter, K.K.
1998. The Ca2+ transport mechanisms of mitochondria and
Ca2+ uptake from physiological-type Ca2+ transients.
Biochim Biophys Acta 1366, 515.
Hansford, R.G. & Zorov, D. 1998. Role of mitochondrial
calcium transport in the control of substrate oxidation.
Mol Cell Biochem 184, 359369.
Harris, D.A. & Das, A.M. 1991. Control of mitochondrial
ATP synthesis in the heart. Biochem J 280, 561573.
Hellsten, Y., Richter, E.A., Kiens, B. & Bangsbo, J. 1999.
AMP deamination and purine exchange in human skeletal
muscle during and after intense exercise. J Physiol 520, 909
919.
Hochachka, P.W. & Matheson, G.O. 1992. Regulating ATP
turnover rates over broad dynamic work ranges in skeletal
muscles. J Appl Physiol 73, 16971703.
Hoppeler, H. 1990. The different relationship of VO2-max to
muscle mitochondria in humans and quadrupedal animals.
Respiration Physiol 80, 137146.
Kacser, H. & Burns, J.A. 1973. Rate control of biological
processes. Symp Soc Exp Biol 27, 65104.
Kacser, H. & Burns, J.A. 1995. The control of ux. Biochem Soc
Trans 23, 341366.
Korzeniewski, B. 1998. Is it possible to predict any properties
of oxidative phosphorylation in a theoretical way? Mol Cell
Biol 184, 345358.
Kunz, W.S., Kuznetsov, A.V., Schulze, W. et al. 1993.
Functional characterization of mitochondrial oxidative
phosphorylation in saponin-skinned human muscle bers.
Biochim Biophys Acta 1144, 4653.
Lodi, R., Taylor, D.J., Tabrizi, S.J. et al. 1997. In vivo skeletal
muscle mitochondrial function in Leber's hereditary optic
2000 Scandinavian Physiological Society
neuropathy assessed by 31P magnetic resonance
spectroscopy. Ann Neurol 42, 573579.
Moreno-Sanchez, R., Devars, S., Lopez-Gomez, F., Uribe,
A. & Corona, N. 1991. Distribution of control of
oxidative phosphorylation in mitochondria oxidizing
NAD-linked substrates. Biochim Biophys Acta 1060,
284292.
Nicholls, D.G. & Ferguson, S.J. 1992. Bioenergetics, Vol. 2,
pp. 5859. Academic Press, London.
Ovadi, J. & Orosz, F. 1997. A new concept for control of
glycolysis. In: L. Agius & H.S.A. Sherratt (eds) Channelling in
Intermediary Metabolism, pp. 237268. Portland Press,
London and Miami.
Rasmussen, H.N., Andersen, A.J. & Rasmussen, U.F. 1997.
Optimimization of preparation of mitochondria from 25 to
100 mg skeletal muscle. Anal Biochem 252, 153159.
Rasmussen, H.N. & Rasmussen, U.F. 1993. Respiration
measurements in small scale. Anal Biochem 208,
244248.
Rasmussen, H.N. & Rasmussen, U.F. 1997. Small scale
preparation of skeletal muscle mitochondria, criteria of
integrity, and assays with reference to tissue function.
Mol Cell Biochem 174, 5560.
Rasmussen, U.F. & Rasmussen, H.N. 2000. Metabolic
characterization of human quadriceps mitochondria.
Mol Cell Biochem in press.
Richter, C. 1997. Reactive oxygen and nitrogen species
regulate mitochondrial Ca+2 homeostasis and respiration.
Bioscience reports 17, 5366.
Rolfe, D.F.S., Hulbert, A.J. & Brand, M.D. 1994.
Characteristics of mitochondrial proton leak and control
of oxidative phosphorylation in the major oxygen-
consuming tissues of the rat. Biochim Biophys Acta 1118,
405416.
Saks, V.A., Khuchua, Z.A., Vasilyeva, E.V., Yu Belikova,
O. & Kuznetsov, A.V. 1994. Metabolic compart-
mentation and substrate channelling in muscle cells.
Role of coupled creatine kinases in in vivo regulation of
cellular respiration A synthesis. Mol Cell Biochem 133,
155192.
Saks, V.A., Veksler, V.I., Kuznetsov, A.V. et al. 1998.
Permeabilized cell and skinned ber techniques in studies
of mitochondrial function in vivo. Mol Cell Biochem 184,
81100.
Saltin, B. & Rowell, L.B. 1980. Functional adaptations to
physical activity and inactivity. Federation Proc 39,
15061513.
Schrauwen, P., Walder, K. & Ravussin, E. 1999. Human
uncoupling proteins and obesity. Obes Res 7, 97105.
Srere, P.A. 1987. Complexes of sequential metabolic enzymes.
Ann Rev Biochem 56, 89124.
Srere, P.A., Sherry, A.D., Malloy, C.R. & Sumegi, B. 1997.
Channelling in the Krebs tricarboxylic acid cycle. In: L.
Agius & H.S.A. Sherratt (eds) Channelling in Intermediary
Metabolism, pp. 201219. Portland Press, London and
Miami.
Starkov, A.A., Simonyan, R.A., Dedukhova, V.I., Mansurova,
S.E., Palamarchuk, L.A. & Skulachev, V.P. 1997.
Regulation of the energy coupling in mitochondria by some
479
Human skeletal muscle mitochondrial capacity
U F Rasmussen and H N Rasmussen
steroid and thyroid hormones. Biochim Biophys Acta 1318,
173183.
Wallimann, T., Wyss, M., Brdiczka, D., Nicolay, N. &
Eppenberger, H.M. 1992. Intracellular compartmentation,
structure and function of creatine kinase isoenzymes in
tissues with high and uctuating energy demands: the
`phosphocreatine circuit' for cellular energy homeostasis.
Biochem J 281, 2140.
480
Acta Physiol Scand 2000, 168, 473480
Wegmann, G., Zanolla, E., Eppenberger, H.M. &
Wallimann, T. 1992. In situ compartmentation of creatine
kinase in intact sarcomeric muscle: the acto-myosin
overlap zone as a molecular sieve. J Muscle Res Cell Motil
13, 420435.
Wibom, R. & Hultman, E. 1990. ATP production rate in
mitochondria isolated from microsamples of human
muscle. Am J Physiol 259, E204E209.
2000 Scandinavian Physiological Society