Role of myoplasmic phosphate in contractile function of skeletal muscle: studies on creatine kinase-deficient mice


Corresponding author H. Westerblad: Department of Physiology and Pharmacology, von Eulers väg 4, Karolinska Institutet, 171 77 Stockholm, Sweden. Email:


  • Increased myoplasmic inorganic phosphate (Pi) has been suggested to have an important role in skeletal muscle fatigue, especially in the early phase. In the present study we used intact fast-twitch muscle cells from mice completely deficient in creatine kinase (CK-/-) to test this suggestion. These CK-/- muscle cells provide a good model since they display a higher Pi concentration in the unfatigued state and fatigue without significant increase of Pi.

  • Tetanic contractions (350 ms duration) were produced in intact single muscle fibres. The free myoplasmic [Ca2+] ([Ca2+]i) was measured with the fluorescent indicator indo-1. The force-[Ca2+]i relationship was constructed from tetani at different frequencies.

  • Compared with wild-type fibres, CK-/- fibres displayed lower force in 100 Hz tetani and at saturating [Ca2+]i (i.e. 100 Hz stimulation during caffeine exposure), higher tetanic [Ca2+]i during the first 100 ms of tetanic stimulation, reduced myofibrillar Ca2+ sensitivity when measurements were performed 100–200 ms into tetani, and slowed force relaxation that was due to altered cross-bridge kinetics rather than delayed Ca2+ removal from the myoplasm.

  • In wild-type fibres, a series of 10 tetani resulted in reduced tetanic force, slowed force relaxation, and increased amplitude of [Ca2+]i tails after tetani. None of these changes were observed in CK-/- fibres.

  • Complementary experiments on isolated fast-twitch extensor digitorum longus muscles showed a reduction of tetanic force and relaxation speed in CK-/- muscles similar to those observed in single fibres.

  • In conclusion, increased Pi concentration can explain changes observed in the early phase of skeletal muscle fatigue. Increased Pi appears to be involved in both fatigue-induced changes of cross-bridge function and SR Ca2+ handling.

Phosphocreatine (PCr) breaks down to creatine (Cr) and inorganic phosphate (Pi) during high-intensity activity in skeletal muscle. In vitro studies on skeletal muscle preparations have shown that Pi affects both the contractile machinery and sarcoplasmic reticulum (SR) Ca2+ handling. Therefore, increased Pi has been suggested to be a key factor in skeletal muscle fatigue (Fitts, 1994; Allen et al. 1995). However, this suggestion has not been experimentally confirmed in intact muscle due to the lack of models where an increase in Pi is obtained without other metabolic changes that might also affect muscle function. Recently, mice completely deficient in creatine kinase (CK-/- mice) have been bred (Steeghs et al. 1997). CK catalyses phosphate exchange between the high-energy phosphates ATP and PCr via the reaction: PCr + ADP + H+⇋ Cr + ATP. Compared with wild-type littermates, fast-twitch muscles of CK-/- mice have a higher Pi concentration at rest and they fatigue without PCr breakdown and associated Pi accumulation (Steeghs et al. 1997; Dahlstedt et al. 2000). Thus, fast-twitch CK-/- muscle provides a model for studying the functional effects of increased Pi in intact muscle.

Fatigue induced by repeated short tetani in intact single muscle fibres has been shown to occur in three phases (Allen et al. 1995). First there is a rapid decline of force to about 85 % of the basal, which is accompanied by increased tetanic [Ca2+]i (phase 1). Then follows a period of more stable tetanic force, the duration of which depends on the tetanic interval and fatigue resistance of the fibre (phase 2). Finally, both tetanic force and [Ca2+]i fall relatively fast until fatiguing stimulation is stopped (phase 3). Several changes in muscle function that develop during phase 1 may be ascribed to increased Pi: (i) reduced cross-bridge force production and myofibrillar Ca2+ sensitivity (e.g. Millar & Homsher, 1990), (ii) increased tetanic [Ca2+]i (e.g. Fruen et al. 1994), (iii) slowed force relaxation due to altered cross-bridge kinetics (Mulligan et al. 1999), and (iv) slowed SR Ca2+ uptake (e.g. Dawson et al. 1980).

In the present study we compare force production and intracellular Ca2+ handling under resting conditions in single fast-twitch muscle fibres of CK-/- and wild-type mice. We also compare changes occurring early during fatiguing stimulation (i.e. during phase 1) in the two groups. Finally, complementary experiments were performed on isolated fast-twitch extensor digitorum longus (EDL) muscles. The results support an important role of increased Pi in changes of muscle function observed in early fatigue (phase 1) in fast-twitch muscle.



CK-/- mice and their wild-type littermates were generated as described previously (Steeghs et al. 1997). Animals were housed at room temperature with 12 h: 12 h light-dark cycle. Food and water were provided ad libitum. Adult female mice were used and these were killed by rapid neck disarticulation and thereafter muscles were isolated. All procedures were approved by the local ethics committee.


Experiments were performed at room temperature (24 °C). Isolated fibres or muscles were superfused with a Tyrode solution of the following composition (mm): NaCl, 121; KCl, 5.0; CaCl2, 1.8; MgCl2, 0.5; NaH2PO4, 0.4; NaHCO3, 24.0; EDTA, 0.1; glucose, 5.5; 0.2 % fetal calf serum was added to the solution to improve survival of single fibres. The solution was bubbled with 5 % CO2-95 % O2 to give a pH of 7.4. To assess force production at full activation of the contractile machinery, caffeine (5 mm) was added to the solution for 1 min prior to contraction and then rapidly removed.

Single fibre dissection, mounting and stimulation

Intact, single fibres were dissected from the flexor digitorum brevis (FDB) muscle of the hindlimb (Lännergren & Westerblad, 1987). The isolated fibre was mounted between an Akers 801 force transducer (SensoNor, Horten, Norway) and an adjustable holder at the length giving maximum tetanic force. Twitch and tetanic stimulation was achieved by trains of supramaximum current pulses (duration 0.5 ms) delivered via platinum plate electrodes lying parallel to the fibres. The duration of tetani was 350 ms. Under resting conditions, tetani of different frequencies were given at 1 min intervals. The effect of repeated contractions was assessed by giving ten 70 Hz tetani at 2.5 s intervals. Force is expressed in kPa, i.e. kN (m2 cross-sectional area)−1.

[Ca2+]i and force measurements

[Ca2+]i was measured with the fluorescent Ca2+ indicator indo-1 (Molecular Probes Europe B.V., Leiden, The Netherlands); the pentapotassium salt of indo-1 was microinjected into fibres, to avoid problems with loading of organelles. The fluorescence of indo-1 was measured with a system consisting of a xenon lamp, a monochromator and two photomultiplier tubes (PTI, Photo Med GmbH, Wedel, Germany). The excitation light was set to 360 ± 5 nm and the light emitted at 405 ± 5 and 495 ± 5 nm was measured. The ratio of the light emitted at 405 nm to that at 495 nm was translated to [Ca2+]i as described elsewhere (Andrade et al. 1998).

After injection of indo-1, fibres were allowed to rest for at least 60 min. Thereafter some tetanic contractions were produced at 1 min intervals to ensure that force and [Ca2+]i were stable. During contractions, fluorescence and force signals were sampled on-line and stored for subsequent data analysis. The steady-state force-[Ca2+]i relationship in unfatigued fibres was established by measuring [Ca2+]i and force in tetani produced at 10-100 Hz. To get the force at saturating [Ca2+]i, a 100 Hz tetanus in the presence of 5 mm caffeine was produced. [Ca2+]i-force curves were constructed in each fibre by using non-linear regression to fit data points to the following Hill equation:

display math

where P is the relative force, Pmax is the force at full Ca2+ activation, Ca50 is the [Ca2+]i giving 50 % of Pmax, and N is a Hill coefficient that relates to the steepness of the function. Force-[Ca2+]i curves were constructed from measurements of the mean force and [Ca2+]i over the last 100 ms of stimulation (i.e. 250-350 ms into tetani). Force-[Ca2+]i curves were also constructed from measurements performed 100-200 ms into tetani at 10-50 Hz (i.e. during the steep part of the curve) using the same Pmax as above (see Fig. 2).

Figure 2.

The myofibrillar Ca2+ sensitivity is lower in CK-/- fibres when measurements are performed early in the tetanus

Representative [Ca2+]i and force records from 40 Hz tetani produced in a wild-type fibre (A) and a CK-/- fibre (B). Period of stimulation is indicated by bar below force records. Observe the different tetanic [Ca2+]i patterns, with a gradual increase in the wild-type fibre but not in the CK-/- fibre. This has an impact on the force-[Ca2+]i curves from the same fibres, which are shown in C (wild-type fibre) and D (CK-/- fibre). ○ (connected by dashed line) and • (connected by continuous line) refer to measurements performed 100-200 ms and 250-350 ms into tetani, respectively. Note that while the two force-[Ca2+]i curves were similar in the wild-type fibre, there was a clear shift to the right with early measurements in the CK-/- fibre.

The force relaxation speed was assessed both on real and calcium-derived force records by measuring the time from the end of tetanic stimulation until force had decreased to 70 % of the maximum. In each fibre the calcium-derived force was constructed by converting tetanic [Ca2+]i into force by means of the force-[Ca2+]i curve (Westerblad & Allen, 1993a). The time to relax to 20 % of the maximum force was also measured. The two sets of measurements gave qualitatively similar results both for real and calcium-derived force, and for simplicity we will only report the time to relax to 70 % of the maximum force.

Indo-1 is a relatively slow Ca2+ indicator that cannot accurately follow rapid changes of [Ca2+]i. Therefore, kinetic corrections were performed as described previously (Westerblad & Allen, 1996a). We used an indo-1 Ca2+ dissociation rate (Koff) of 52 s−1, which was obtained in Xenopus frog muscle fibres studied under conditions similar to the present (Westerblad & Allen, 1996a). With an intracellular indo-1 Ca2+ dissociation constant (KD) of 283 nm (Andrade et al. 1998), this gives an indo-1 Ca2+ association rate (Kon) of 184 μm−1 s-1. Two differences were observed when [Ca2+]i records from tetanic contractions with and without kinetic correction were compared (see Fig. 1): a [Ca2+]i transient at the onset of contraction was revealed with correction; uncorrected [Ca2+]i records displayed a small delay (3-6 ms) during the first part of relaxation. Therefore, kinetically corrected indo-1 signals were used when measurements were performed during the first 100 ms of tetanic contraction and during the first 100 ms of relaxation. Also, kinetic correction was used to obtain the peak [Ca2+]i with a single stimulation pulse (twitch). Measurements of [Ca2+]i at rest, between 100 and 350 ms of tetanic contraction, and during the slow decay of [Ca2+]i after the end of tetani were not significantly affected by kinetic correction.

Figure 1.

Tetanic force is markedly smaller in CK-/- fibres than in wild-type fibres despite similar or even higher [Ca2+]i

Typical records of [Ca2+]i and force from 100 Hz tetani produced in a wild-type fibre (A) and a CK-/- fibre (B).[Ca2+]i records are shown both with (dashed line) and without (continuous line) kinetic correction of the indo-1 signal. Observe that kinetic correction reveals a [Ca2+]i transient at the onset of contraction, and that the decrease of [Ca2+]i at the end of stimulation is slightly delayed without correction. Horizontal bars below force records show period of stimulation.

Whole muscle experiments

Intact EDL muscles were dissected from one leg in each animal. Small stainless-steel loops were tied to the tendons of the muscle with thin nylon thread. The muscle was then mounted between a force transducer and an adjustable hook, which allowed the muscle to be stretched to the length giving maximum tetanic force. Tetanic stimulation was produced by applying supramaximal current pulses (0.5 ms duration) via plate electrodes lying on each side of the muscle. After obtaining the length yielding maximal tetanic force, the muscle was allowed to rest for 30 min. The contractile function in the unfatigued state was established by producing a 100 Hz, 300 ms tetanus. The effect of repeated tetanic stimulation was studied by giving a series of 10 tetani at 2 s intervals. Tetanic force and relaxation time were measured as described above for single fibres. The cross-sectional area was not measured in EDL muscles and forces are expressed relative to muscle dry weight. Muscle length was not measured but no consistent differences between wild-type and CK-/- muscles were noted during mounting. Therefore, forces expressed relative to muscle dry weight should provide a reasonable measure of force per cross-sectional area.

There are technical reasons for using different muscles in the single fibre and whole muscle experiments. Muscle fibres of the FDB muscle are much shorter than EDL fibres, which is advantageous during single fibre dissection. However, the FDB muscles are multipennate and have three distinctly different distal tendons, which together with the short fibre length make them less suitable for whole muscle experiments. Mouse FDB and EDL muscles contain almost exclusively fast-twitch, type 2 fibres (Raymackers et al. 2000) and twitch kinetics of the present single FDB fibres and EDL muscles are very similar (Dahlstedt et al. 2000). Thus results from single FDB fibres and whole EDL muscles would give a fair account of the function of fast-twitch muscle in CK-/- and wild-type mice.


Values are presented as means ±s.e.m. Student's unpaired t tests were used to determine significant differences between CK-/- and wild-type muscle; paired t tests were used when analyses were performed within each group. The significance level was set at 0.05 throughout.


Tetanic [Ca2+]i and force

Experiments were performed on seven CK-/- and six wild-type single muscle fibres. Figure 1 shows representative records of [Ca2+]i and force obtained from 100 Hz tetani produced in a wild-type fibre (Fig. 1a) and in a CK-/- fibre (Fig. 1B). Tetanic force was markedly lower in the CK-/- fibre despite the fact that tetanic [Ca2+]i was, if anything, higher in this fibre. Mean data from 100 Hz tetani showed a significantly lower force in CK-/- fibres (241 ± 14 kPa) than in wild-type fibres (329 ± 23 kPa). Measurements of resting [Ca2+]i, performed immediately before tetanic stimulation, did not show any significant difference between CK-/- fibres (72.6 ± 5.7 nm) and wild-type fibres (70.0 ± 5.8 nm). Mean tetanic [Ca2+]i, measured during the last 100 ms of stimulation, was not significantly different between CK-/- fibres (1.58 ± 0.16 μm) and wild-type fibres (1.75 ± 0.21 μm). However, the shape of the tetanic [Ca2+]i records differed between the groups (see Fig. 1). Thus, [Ca2+]i increased significantly during the tetanic plateau in wild-type fibres with a slope of 1.43 ± 0.53 μm s−1 (measured in each fibre as the difference between mean [Ca2+]i during the last and the first 100 ms of stimulation), whereas it tended to fall in CK-/- fibres (slope -1.09 ± 0.65 μm s−1). Hence the mean tetanic [Ca2+]i during the first 100 ms of stimulation was higher in CK-/- fibres (1.85 ± 0.30 μm) than in wild-type fibres (1.40 ± 0.11 μm), but this difference was not significant (P = 0.21). A similar pattern was observed at all tetanic stimulation frequencies; for instance, analysis of [Ca2+]i records from 50 Hz tetani showed a significantly higher [Ca2+]i during the first 100 ms of stimulation in CK-/- fibres (1.02 ± 0.09 μm) than in wild-type fibres (0.77 ± 0.05 μm), whereas there was no difference during the last 100 ms (0.94 ± 0.06 vs. 0.96 ± 0.07 μm). To further study the initial Ca2+ release, twitches were produced and the peak [Ca2+]i measured. Peak [Ca2+]i was significantly higher in CK-/- fibres (3.22 ± 0.41 μm) than in wild-type fibres (2.11 ± 0.20 μm).

The maximum force-generating capacity was assessed by producing 100 Hz tetani in the presence of 5 mm caffeine. Application of caffeine had similar effects on CK-/- and wild-type fibres and increased tetanic [Ca2+]i by about 200 % and force by about 7 %. Thus, tetanic force at saturating [Ca2+]i was significantly lower in CK-/- fibres (257 ± 15 kPa) than in wild-type fibres (353 ± 27 kPa).

Force-[Ca2+]i relationship

The force-[Ca2+]i relationship was determined by measuring force and [Ca2+]i in tetanic contractions of different frequencies. Figure 2 shows representative [Ca2+]i and force records from 40 Hz tetani, which lie on the steep part of the force-[Ca2+]i relationship, from a wild-type fibre (Fig. 2a) and a CK-/- fibre (Fig. 2B). Again, the shape of tetanic [Ca2+]i is different in the two fibres, with a gradual increase during the tetanus in the wild-type fibre and no obvious change in the CK-/- fibre. The shape of the force records, on the other hand, is more similar, with a gradual increase in both fibres, albeit the initial force increase is somewhat faster in the CK-/- fibre. This difference between the fibres affects the force-[Ca2+]i relationship, which is illustrated in Fig. 2C and D. In the wild-type fibre, the force-[Ca2+]i relationship did not depend on measurements being performed early during the tetanus (i.e. between 100 and 200 ms of stimulation; ○) or towards the end (i.e. between 250 and 350 ms of stimulation; •). However, in the CK-/- fibre, the force-[Ca2+]i relationship was shifted to the right with measurements early during the contraction as compared with measurements towards the end. The same pattern was observed in the other fibres and mean data of the force-[Ca2+]i relationship (Table 1) show similar Ca50 values when measurements were performed late. With early measurements, however, there was a significantly higher Ca50 value in CK-/- fibres than in wild-type fibres. Table 1 also shows that Pmax was significantly lower in CK-/- fibres than in wild-type fibres.

Tetanic relaxation speed

The tetanic force records in Fig. 1 show a slower relaxation in the CK-/- fibre. This was a consistent finding and mean data show about 30 % longer relaxation time in CK-/- fibres as compared with wild-type fibres (Table 1). To investigate the mechanism behind this difference between the groups, calcium-derived force records were produced from [Ca2+]i records and the force-[Ca2+]i relationship. Representative force records from the relaxation phase of a wild-type fibre and a CK-/- fibre are shown in Fig. 3. While relaxation of real force was slower in the CK-/- fibre, there was no marked difference in the relaxation speed of calcium-derived force. This was a general finding and the results are summarised in Table 1. The calcium-derived forces used in Table 1 were constructed using the force-[Ca2+]i relationship obtained towards the end of tetani. Qualitatively similar results were achieved when the force-[Ca2+]i relationship obtained earlier in the tetanus was used to construct calcium-derived force, although calcium-derived relaxation then became somewhat faster, especially in CK-/- fibres (data not shown).

Figure 3.

Force relaxation is markedly slower in CK-/- fibres than in wild-type fibres

A, real force records of the relaxation phase of 100 Hz tetani obtained in a wild-type fibre (continuous line) and a CK-/- fibre (dashed line). B, calcium-derived force records from the same fibres produced from tetanic [Ca2+]i records and their respective force-[Ca2+]i curves. The time axis starts at the end of tetanic stimulation. Note that the markedly slower relaxation in the CK-/- fibre is not accompanied by a slowed relaxation of calcium-derived force.

Changes in [Ca2+]i and force due to 10 repeated tetani

A series of ten 70 Hz tetani was produced to study changes of [Ca2+]i and force in early fatigue, i.e. during phase 1. In the tenth tetanus the force was reduced to 89 ± 2 % of control in wild-type fibres, whereas it was not significantly changed in CK-/- fibres (104 ± 2 %). However, absolute forces were still lower in CK-/- fibres (222 ± 10 kPa) than in wild-type fibres (271 ± 18 kPa). In the first tetanus, [Ca2+]i (measured as the mean over the first 100 ms of stimulation) was significantly higher in CK-/- fibres (1.40 ± 0.18 μm) than in wild-type fibres (0.91 ± 0.06 μm). Tetanic [Ca2+]i increased significantly during the series of tetani in wild-type fibres whereas it did not change in CK-/- fibres. Thus in the tenth tetanus, tetanic [Ca2+]i was similar in CK-/- fibres (1.34 ± 0.14 μm) and wild-type fibres (1.24 ± 0.10 μm).

The relaxation speed was also measured at the start and at the end of the series of 10 tetani. In wild-type fibres there was a highly significant increase of the relaxation time (31 ± 4 %) in the tenth tetanus. In CK-/- fibres, on the other hand, there was no significant difference between the relaxation times in the first and tenth tetanus. This means that in the tenth tetanus, relaxation times in wild-type and CK-/- fibres were not significantly different (85.8 ± 5.4 ms vs. 83.6 ± 7.0 ms; note that repeated 70 Hz tetani were used and at this stimulation frequency relaxation times of unfatigued fibres were slightly shorter than with 100 Hz tetani).

Tails of elevated [Ca2+]i after the end of tetanic stimulation may be used to assess the function of the SR Ca2+ pumps (Klein et al. 1991). Ten repeated tetani resulted in increased tails of [Ca2+]i in wild-type fibres (Fig. 4a). Measurements in these fibres, performed as the mean over 100 ms at 0.5, 1 and 2 s after the end of tetanic stimulation, showed significant increases of about 15 nm in the tenth as compared with the first tetanus. On the other hand, 10 repeated tetani had no significant effects on tails of [Ca2+]i in CK-/- fibres (Fig. 4B). In the tenth tetanus, tails of [Ca2+]i were similar in wild-type and CK-/- fibres; for instance, at 0.5 s after the end of tetanic stimulation [Ca2+]i was 160 ± 7 nm in wild-type fibres and 162 ± 15 nm in CK-/- fibres.

Figure 4.

The amplitude of tails of elevated [Ca2+]i is increased after 10 repeated tetani in wild-type but not in CK-/- fibres

Mean [Ca2+]i records from the relaxation phase of wild-type fibres (Wt; A) and CK-/- fibres (B). Records were obtained in the first (thin line) and tenth (thick line) repeated tetanus.

Whole muscle experiments

Experiments were performed on five EDL muscles from CK-/- and wild-type mice, respectively. Representative force records from 100 Hz tetani produced in EDL muscles are shown in Fig. 5. Tetanic force was about 25 % lower in EDL muscles from CK-/- mice (90 ± 6 N (g dry weight)−1) as compared with the wild-type (118 ± 7 N (g dry weight)−1). Furthermore, the relaxation time was significantly longer in CK-/- EDL muscles (56.0 ± 4.0 ms vs. 30.4 ± 3.1 ms in wild-type). Thus in the unfatigued state, the differences in contractile performance between CK-/- and wild-type EDL muscles are very similar to those observed in the single fibres.

Figure 5.

Tetanic force is smaller and relaxation slower in CK-/- EDL muscles

Typical force records from 100 Hz tetani produced in EDL muscles from wild-type (continuous line) and CK-/- (dashed line) mice. The period of stimulation is indicated below the records.

We also measured force and relaxation speed after ten 70 Hz tetani in EDL muscles. Compared with the first fatiguing tetanus, force in the tenth tetanus was reduced to 84 ± 1 % in wild-type and 92 ± 2 % in CK-/- EDL muscles. In the tenth tetanus the relaxation time was increased by 67 ± 8 % in wild-type muscles and by 45 ± 5 % in CK-/- muscles. Thus, a series of 10 tetani caused a reduction of tetanic force and a slowing of relaxation in both groups, but the changes were significantly smaller in CK-/- muscles.


Increased Pi is considered to be an important cause of skeletal muscle fatigue. Specifically, changes in early fatigue, i.e. during phase 1, have been ascribed to increased Pi. However, lack of suitable models has prevented direct tests of this proposal in intact muscle. In the present study, we used muscles from CK-/- mice to investigate the effect of increased Pi on the function of unfatigued muscle and compared the results with those from early fatigue. CK-/- muscle provides a good experimental model since under resting conditions, there is a markedly higher Pi in fast-twitch EDL muscles of CK-/- animals than in wild-type (Dahlstedt et al. 2000). Furthermore, CK-/- muscles do not display any marked increase of Pi during fatigue (Steeghs et al. 1997; Dahlstedt et al. 2000) or during ischaemia (in't Zandt et al. 1999). It should be noted that rested CK-/- muscle also displays changes in metabolites other than Pi. For instance, ATP and PCr concentrations are lower and the Cr concentration is higher in CK-/- as compared with wild-type muscle (Steeghs et al. 1997; Dahlstedt et al. 2000). However, the differences in these metabolites are relatively modest and would have little effect on force production (Godt & Nosek, 1989). Furthermore, intracellular pH, which has a large effect on force production at the temperature used in the present experiments (Westerblad et al. 1997), is similar in rested CK-/- and wild-type muscles (Steeghs et al. 1997). Thus, CK-/- muscle provides a good model to study contractile effects of increased Pi, but some modulating role of other metabolic differences between CK-/- and wild-type muscles cannot be ignored.

Reduction of cross-bridge force production

When intact muscle fibres are fatigued by repeated tetanic stimulation, there is a reduction of tetanic force already after a few tetani, which has been ascribed to increased myoplasmic Pi (Westerblad & Allen, 1992). This suggestion is strongly supported by the present results. Tetanic force was markedly lower in rested single fibres and EDL muscles of CK-/- mice, which are fast-twitch and have a higher Pi than wild-type fibres (Dahlstedt et al. 2000). Moreover, the decline in tetanic force during phase 1 is not present in CK-/- single fibres, which do not accumulate Pi. Additional support for a coupling between force production and Pi concentration in intact muscle comes from previous studies where a reduced Pi concentration is associated with increased force production (Phillips et al. 1993; Bruton et al. 1997). In this context it should be noted that studies of Pi effects on force production in mammalian muscle, including the present study, have generally been performed at temperatures well below body temperature. There are results showing that the force-depressing effect of Pi is smaller at high temperature (Dantzig et al. 1992), but at 30 °C increased Pi still gives a marked force depression in skinned, fast-twitch muscle fibres (Puchert et al. 1999).

Skinned fibre experiments demonstrate that increased Pi reduces cross-bridge force production (Pate & Cooke, 1989; Millar & Homsher, 1990; Potma & Stienen, 1996). Data from these skinned fibre experiments can be used to assess whether the reduction of force observed in fast-twitch CK-/- muscle is likely to be due to increased Pi. Translating values of Pi in rested EDL muscles of CK-/- (22.6 μmol (g dry weight)−1) and wild-type (11.1 μmol (g dry weight)−1) (Dahlstedt et al. 2000) into millimolar (mm), assuming two litres of intracellular water per kilogram of dry tissue (Sahlin et al. 1983), gives a Pi concentration of about 11.3 and 5.6 mm, respectively. Using data from skinned muscles, this increase in Pi should decrease force by 10-15 % (Pate & Cooke, 1989; Millar & Homsher, 1990; Dantzig et al. 1992; Potma & Stienen, 1996), which is similar to the decrease normally seen in early fatigue (Allen et al. 1995). However, the tetanic force in single muscle fibres and EDL muscles, as well as force at saturating [Ca2+]i in the single fibres, was about 25 % lower in CK-/- than in the wild-type. Thus, the force reduction in fast-twitch CK-/- muscle appears larger than would be expected solely from the higher Pi concentration. In line with this, the absolute tetanic force after 10 fatiguing tetani was still 18 % lower in CK-/- than wild-type single muscle fibres. The additional force deficit in CK-/- fibres may be due to cytoarchitectural abnormalities within the cells and an increased fraction of the myoplasm being occupied by mitochondria and lipid droplets leading to a reduced concentration of myofibrils (Steeghs et al. 1997, 1998; Tullson et al. 1998).

Increased tetanic [Ca2+]i

There are several reasons why increased myoplasmic Pi might lead to increased tetanic [Ca2+]i. For instance, elevated Pi increases Ca2+-induced SR Ca2+ release and the open probability of isolated SR Ca2+ release channels (ryanodine receptors) (Fruen et al. 1994; Balog et al. 2000). Moreover, Pi may inhibit the SR Ca2+ uptake (Dawson et al. 1980; Stienen et al. 1993) which, at least in the short term, might lead to increased tetanic [Ca2+]i (Westerblad & Allen, 1994). Finally, Pi may decrease Ca2+ binding to troponin C via a reduction in strong cross-bridge attachment (Millar & Homsher, 1990) and therefore reduce the myoplasmic Ca2+ buffering, but this mechanism appears to be of little importance in skeletal muscle (Gordon et al. 2000). Thus, it appears likely that the increase of tetanic [Ca2+]i seen in early fatigue (i.e. during phase 1) is due to increased Pi. This suggestion is supported by the fact that CK-/- muscle fibres, which do not accumulate Pi, did not display any increase of tetanic [Ca2+]i during phase 1. Furthermore, the results from unfatigued fibres show a higher [Ca2+]i in twitches and at the onset of 50 and 70 Hz tetani in CK-/- fibres. However, during the first 100 ms of 100 Hz tetani, [Ca2+]i only showed a tendency (not significant) to be higher in CK-/- fibres. As the tetanus continued, [Ca2+]i tended to decline in CK-/- fibres whereas it increased in wild-type fibres. This difference is exaggerated when tetani are given at short intervals, resulting in a significant transient reduction of tetanic [Ca2+]i in CK-/- fibres (Dahlstedt et al. 2000). This decline of tetanic [Ca2+]i in CK-/- fibres is probably due to a direct inhibition of the SR Ca2+ release channels by reduced ATP and/or increased Mg2+ (Blazev & Lamb, 1999), which occurs during tetanic stimulation in the absence of PCr energy buffering. In accordance with our results, this inhibition would be more marked at high stimulation frequency, where the rate of energy consumption is higher. Thus, at the onset of tetanic contraction of unfatigued CK-/- fibres [Ca2+]i is increased due to the elevated Pi, but [Ca2+]i declines as the tetanus progresses due to the lack of PCr energy buffering.

Increased myoplasmic Pi may also reduce tetanic [Ca2+]i by entering the SR and precipitating with Ca2+, thus reducing the Ca2+ available for release (Fryer et al. 1995; Westerblad & Allen, 1996b). Recent results suggest that the transport of Pi into the SR is facilitated by reduced ATP (Posterino & Fryer, 1998; Ahern & Laver, 1998). Since fast-twitch CK-/- muscle displays both an increased Pi and reduced ATP (Dahlstedt et al. 2000), some Ca2+-Pi precipitation in the SR might decrease the Ca2+ available for release and hence reduce tetanic [Ca2+]i in CK-/- fibres. However, this mechanism, if operating, did not prevent CK-/- fibres from showing a higher [Ca2+]i at the onset of contractions. In this context it must be noted that the situation is markedly different in CK-/- fibres as compared with fibres stimulated to fatigue or injected with Pi. In the latter cases, Pi would rapidly enter the SR and precipitate with Ca2+ and hence reduce the Ca2+ available for release. In CK-/- fibres, on the other hand, Pi is chronically elevated and long-term homeostatic mechanisms would keep the [Ca2+]i constant by adjusting the Ca2+ flux over the sarcolemma. Consistent with this, a decrease of resting [Ca2+]i was observed with injection of Pi, which resulted in reduced tetanic [Ca2+]i (Westerblad & Allen, 1996b), whereas the present results did not show a lower resting [Ca2+]i in CK-/- fibres.

Reduced myofibrillar Ca2+ sensitivity

The force-[Ca2+]i relationship was assessed by measuring mean force and [Ca2+]i over 100 ms in tetani of different frequencies. During the first 100 ms of tetanic stimulation in wild-type fibres, force and [Ca2+]i reached a steady state where [Ca2+]i increased slowly and this was accompanied by an increased force production. This means that the force-[Ca2+]i relationship was similar whether measured 100-200 ms or 250-350 ms into the tetanus, which fits with previous results showing that the method of measuring has little impact on the force-[Ca2+]i relationship (Westerblad & Allen, 1993a; Andrade et al. 1998). In CK-/- fibres, on the other hand, [Ca2+]i tended to decline during the tetanus and this was not accompanied by a corresponding reduction of force. Hence in these fibres, Ca50 was about 20 % higher when measured 100-200 ms as compared with 250-350 ms into the tetanus. The results from wild-type fibres, therefore, indicate that during tetanic stimulation, force can follow slow increases of [Ca2+]i, i.e. additional force-producing cross-bridges are readily being formed when the Ca2+ activation of the thin filament increases. However, the results from CK-/- fibres indicate that the opposite is not true. Due to the co-operativity between strongly attached cross-bridges and thin filament activation (e.g. Swartz et al. 1996; Gordon et al. 2000), strong cross-bridges may promote new cross-bridge attachments despite lowered Ca2+ activation. In accordance with this, during relaxation [Ca2+]i falls much faster than force. Thus, we consider the higher Ca50 obtained early in tetani in CK-/- fibres more likely to be correct.

Measurements performed 100-200 ms into tetani suggest about 15 % higher Ca50 in CK-/- fibres than in wild-type fibres. Taking data from skinned fibre experiments, a two-fold higher Pi concentration would result in an increase in Ca50 of 15-20 % (Millar & Homsher, 1990; Martyn & Gordon, 1992). Thus, the reduced myofibrillar Ca2+ sensitivity observed in CK-/- fibres can be explained by the higher Pi in these fibres.

Slowed relaxation

Several results of the present study support the hypothesis that the slowing of relaxation observed during early fatigue (phase 1) is due to increased Pi. First, in the unfatigued state the relaxation time was markedly longer in CK-/- fibres than in wild-type fibres and the former have a higher Pi concentration. Second, 10 repeated tetani increased the relaxation time in wild-type fibres (in which Pi increases during fatigue) but not in CK-/- fibres (in which Pi does not increase). Third, the reduced relaxation speed seen in unfatigued CK-/- could not be explained by a slowed Ca2+ removal from the myoplasm (see Fig. 3), which is similar to the results obtained in fatigued mouse fibres (Westerblad & Allen, 1993b). Thus, it appears that increased Pi may cause a slowing of relaxation, but the mechanism by which this occurs remains uncertain (for discussion see Westerblad et al. 1997; Mulligan et al. 1999).

In the unfatigued state, CK-/- EDL muscles displayed a markedly longer relaxation time than wild-type muscles, which is in agreement with the single fibre results. However, in early fatigue, EDL muscles differed from single fibres in that both tetanic force and relaxation speed was reduced in CK-/- muscles, albeit to a lesser extent than in wild-type muscles. This difference may be explained by some additional inhibitory factor coming into play in the whole muscle experiments and reduced intracellular pH due to lactate accumulation is a likely candidate. At the temperature used in the present study (24 °C), reduced intracellular pH will cause a marked reduction of tetanic force and relaxation speed (Westerblad et al. 1997). Moreover, single mouse fibres fatigue without significant acidification, probably due to effective lactate-H+ extrusion in the absence of fatigue-induced changes of the extracellular milieu (Westerblad & Allen, 1992). In whole muscles, on the other hand, changes of metabolites in extracellular space (e.g. increased lactate concentration) will inhibit lactate-H+ extrusion and both CK-/- and wild-type EDL muscles display a large increase of lactate at fatigue (Dahlstedt et al. 2000). Furthermore, the acidosis in early fatigue might be exaggerated in CK-/- muscles due to the lack of PCr hydrolysis, which consumes H+ ions.

Slowed SR Ca2+ uptake

There is a marked increase in the amplitude of [Ca2+]i tails after 10 fatiguing tetani (Westerblad & Allen, 1991, 1993b), which might be ascribed to a Pi-induced inhibition of the SR Ca2+ uptake. This may be due to a decreased free energy change of ATP hydrolysis (Dawson et al. 1980; Stienen et al. 1993). The present analyses of [Ca2+]i after the end of tetanic stimulation are in agreement with a Pi-induced inhibition of SR Ca2+ uptake in early fatigue (see Fig. 4). Thus, 10 repeated tetani resulted in increased [Ca2+]i tails in wild-type fibres, where PCr breaks down and Pi accumulates, but not in CK-/- fibres, where this does not occur.


The present results show that fast-twitch muscle of CK-/- mice provides a good model for studying the effect of Pi on muscle function. Using this model we show that increased myoplasmic Pi can explain major functional changes observed in early fatigue in single fast-twitch fibres; that is increased Pi may reduce tetanic force and myofibrillar Ca2+ sensitivity, increase tetanic [Ca2+]i, slow force relaxation, and reduce the rate of SR Ca2+ uptake.


The authors thank Professor Be Wieringa (Department of Cell Biology and Histology, University of Nijmegen, The Netherlands) for donation of the CK-/- mice. The study was supported by grants from the Swedish Medical Research Council (Project 10842), the Swedish National Center for Sports Research, and funds at the Karolinska Institutet.