Getting the jump on skeletal muscle disuse atrophy: preservation of contractile performance in aestivating Cyclorana alboguttata(Günther 1867)

Skeletal muscle is a dynamic and plastic tissue that responds rapidly to changes in frequency and duration of use(Musacchia et al., 1988). An increase in muscle activity builds and maintains muscle mass, while inactivity results in tissue degeneration. In mammals, prolonged immobilisation or unloading of skeletal muscle results in muscle `wasting' or muscle disuse atrophy (Booth, 1982). Muscle disuse atrophy is characterised by a reduction in muscle fibre cross-sectional area due to the combined effects of increased catabolism and decreased anabolism of skeletal muscle protein and an imbalance of muscle biochemistry(Kandarian and Stevenson,2002).

In experimental models of hindlimb immobilisation and unloading the effects of muscle disuse atrophy on muscle morphology and muscle mechanics are variable and depend upon muscle type and fibre type(Ariano et al., 1973; Booth and Kelso, 1973; Edgerton et al., 2002; Peter et al., 1972; Trappe et al., 2004; Witzmann et al., 1982). Most muscles are composed of a heterogenous mixture of fibre types, although there are exceptions. For example, the mammalian soleus muscle is almost completely slow twitch (Ariano et al.,1973).

During normal activity, oxidative muscles tend to be used frequently, for extended periods of time and at low intensities. In comparison, fast-twitch glycolytic muscles are used infrequently, for short periods and at high intensity. When oxidative muscles become inactive the scope of the change in the level of activity is much greater than that experienced by glycolytic,fast-twitch muscles (Hudson and Franklin,2002b). In general, slow-twitch (oxidative) fibres are more vulnerable to muscle disuse atrophy than fast-twitch (glycolytic) fibres(Edgerton et al., 2002; Maier et al., 1973; Nordstrom et al., 1995; Tomanek and Lund, 1974).

Prolonged disuse alters muscle fibre type characteristics, with oxidative fibres acquiring the structural, biochemical and mechanical properties of glycolytic fibres (Anderson et al.,1999; Diffee et al.,1991). Generally, tetanic stress is reduced, maximum rate of shortening is elevated, and fatigue resistance is slightly reduced or unaffected (Booth and Seider,1979a; Roy et al.,1991; Roy et al.,1996; Roy et al.,2002). Commonly, the contractile rates of oxidative muscles are increased after immobilisation leading to quicker twitch times that are more similar to glycolytic muscles (Anderson et al., 1999; Canon and Goubel,1995; Witzmann et al.,1982). Additionally, due to the comparatively greater numbers of mitochondria contained within them, oxidative fibres produce comparatively larger quantities of reactive oxygen species as a byproduct of normal cell metabolism. Reactive oxygen species, or free radicals, are responsible for stochastically damaging cell architecture, including proteins and lipids,which leads to cellular apoptosis and hence muscle disuse atrophy(Kondo et al., 1993a; Kondo et al., 1993b; Kondo et al., 1993c; Kondo et al., 1994). During disuse the dramatic downregulation of activity in the oxidative muscle may trigger profound changes in the biochemical function of the muscle mitochondria.

Few studies have explored the effect of dormancy on muscle disuse atrophy of skeletal muscle (for a review, see Carey et al., 2003). Of the available literature, the majority investigate mammalian models, but virtually none have been conducted on amphibians. In general, the animals studied exhibit some muscle disuse atrophy, though to a lesser degree than observed in humans under similar conditions of immobility(Harlow et al., 2001; Tinker et al., 1998). For example, hibernation for 6 months caused significant atrophy in skeletal muscle from the golden-mantled ground squirrel (Spermophilus lateralis) as measured by whole muscle mass(Wickler et al., 1991b) and whole muscle cross-sectional area, with observable concurrent, but non-significant, declines in the protein content of these muscles(Steffen et al., 1991).

Animals that experience regular bouts of dormancy, such as hibernators and aestivators, exhibit a decline in metabolic and locomotor activity that is correlated with seasonal fluctuations in temperature and resource availability(Guppy and Withers, 1999). During dormancy the animal is primarily reliant upon stored lipids, with protein meeting additional energy demands(Harlow et al., 2001; Pakay et al., 2002; Pedler et al., 1996; Seymour, 1973; Tinker et al., 1998; van Beurden, 1980). Muscle comprises the largest protein store in the body, therefore catabolism of muscle tissue as an energy substrate during hibernation and aestivation may contribute to muscle disuse atrophy(Steffen et al., 1991; Wickler et al., 1987; Wickler et al., 1991b; Yacoe, 1983). Additionally, it has been demonstrated that protein synthesis is downregulated during dormancy,which may further compound the atrophic effects of dormancy on skeletal muscle(Frerichs et al., 1998; Fuery et al., 1998; Pakay et al., 2002). As such,hibernators and aestivators provide a fascinating model for investigating muscle atrophy associated with disuse (see review by Hudson and Franklin,2002b).

Amphibian aestivation usually occurs in response to unfavourable environmental conditions (Kobelt and Linsenmair, 1995), specifically a dry and warm (>10°C)environment, and is characterised by the construction of a thermally stable burrow (van Beurden, 1980),the formation of a cocoon (Withers,1995; Withers,1998) and metabolic rate depression(Flanigan et al., 1991; Pinder et al., 1992). The green-striped burrowing frog Cyclorana alboguttata (Günther,1867) (Meyer et al., 1997) is an amphibian aestivator that inhabits the arid and semi-arid zones of Eastern Australia. This species undergoes aestivation for many months (6–9 months on average) between seasonal rain events(Withers and Richards, 1995). After 3 months of aestivation locomotor performance, wet muscle mass of several hindlimb muscles and contractile properties of the gastrocnemius of C. alboguttata remained unchanged from that of control animals,indicating an absence of muscle disuse atrophy(Hudson and Franklin, 2002a). Long bone mechanical properties, capillary structure and gut function were also preserved during aestivation in C. alboguttata(Cramp and Franklin, 2005; Hudson et al., 2004; Hudson and Franklin,2003).

It is clear that although C. alboguttata does not exhibit typical atrophic responses, we are still unsure as to what exemplifies a typical aestivator response to prolonged muscle disuse. The aim of the present study was to examine the effect of prolonged aestivation on the morphology and contractile performance of whole muscles from C. alboguttata. Unlike previous studies, this project investigated the effect of aestivation over a dramatically longer, more ecologically relevant, time-frame of 9 months. At the time of publication, no other study had examined the effect of aestivation or hibernation on any aspect of muscle physiology or biochemistry over such an extended period. Though Hudson and Franklin's study(Hudson and Franklin, 2002b)demonstrated that C. alboguttata did not exhibit muscle disuse atrophy after 3 months of aestivation, we felt that the results of that study were not indicative of the true capabilities of this animal as a model species.

We hypothesised that aestivation of 9 months duration would result in changes to the fibre-type composition of the sartorius (relatively fast-twitch, glycolytic) and iliofibularis (relatively slow-twitch, oxidative)muscles. Specifically, the iliofibularis muscle would exhibit a compositional change from the atrophy-susceptible oxidative fibre type to the more atrophy-resistant glycolytic fibre type. If evident, we proposed these changes in fibre type would affect muscle contractile performance of aestivating frogs; the slow-twitch iliofibularis muscle would exhibit the contractile properties of a fast-twitch muscle, and the fast-twitch sartorius would become faster still. We measured the effect of aestivation on whole muscle cross-sectional area, muscle fibre type, fibre number and fibre cross-sectional area, and related the observed ultrastructural changes (if any) to muscle mechanical performance as measured by isometric twitch and tetanus kinetics, work-loop power output and work-loop fatigue resistance.

Adult green-striped burrowing frogs Cyclorana alboguttata(Günther, 1867) of both sexes were collected from roadsides after heavy rain at Lake Broadwater, SW Queensland, Australia, between February 2005 and January 2006. The frogs were transported to the School of Integrative Biology at University of Queensland and were sexed. Body mass was measured using an electronic balance (BP310S, Sartorius, Edgewood, NJ, USA) to the nearest 0.01 g and snout–vent length (SVL) measured using digital Vernier callipers (Whitworth, Brisbane, Australia) to the nearest 0.01 mm. Each frog was randomly assigned to either a control (active) or treatment (9-month aestivation) group. There was no significant difference in initial frog size between control and treatment groups, as measured by SVL (mm)(t-test; P=0.896). Control frogs (N=6) were housed individually in 5 l plastic containers with a paper-towel substrate and water to a depth of 10 mm. The water and paper-towel were replaced once per week and frogs were fed ad libitum on crickets and wood cockroaches dusted with ReptiVite™ (ZooMed, San Luis Obispo, CA, USA) diet supplement. Treatment frogs (N=5) were similarly housed and fed for two weeks post-capture before being individually placed in a plastic container filled with wet clay (obtained from the field collection site) to a depth of 200 mm. Most individuals burrowed into the wet clay immediately, while any frog that did not burrow into the clay after 3 consecutive days was removed from the treatment group and replaced. The clay was permitted to dry naturally and the frogs began aestivating for 9 months. All frogs were kept in the facility at a constant 24°C under combined natural and fluorescent lighting conditions. At the conclusion of the 9-month treatment period, the treatment frogs were removed from the dry clay housing and immediately euthenised by cranial and spinal pithing. Control frogs were also pithed at this time. Body mass and SVL were re-measured post-euthanasia.

The skin of both hind limbs was removed to expose the muscle tissue. Slices of approximately 5 mm thickness were dissected from the belly of the right-side iliofibularis and left-side sartorius muscles with a razor blade. The iliofibularis and sartorius were chosen to represent relatively slow and fast skeletal muscles, respectively, based upon previous studies of anuran muscle fibre type (Putnam and Bennett,1983) and preliminary findings by the author. These fresh muscle slices were placed into plastic moulds, mounted in Tissue-Tek™ OCT compound (ProSciTech, Townsville, Australia) and plunged into isopentane(2-methylbutane), cooled to –150°C in liquid nitrogen, for approximately 30 s. The frozen blocks were removed and wrapped in aluminium foil to prevent desiccation, and stored at –80°C in an air-tight container for histochemical analysis. The sartorius muscle from the right leg and the iliofibularis from the left leg of control and treatment frogs were removed, leaving a section of bone attached to the tendon at either end of each muscle. These muscles were stored in refrigerated McKenzies frog Ringer solution (111 mmol l^–1 NaCl, 2.5 mmol l^–1KCl, 1.8 mmol l^–1 CaCl₂. 2H₂O, 1 mmol l^–1 MgCl, 5 mmol l^–1 Hepes, 10 mmol l^–1 glucose, adjusted to pH 7.4 with NaOH) until required for isometric analysis.

Whole muscle and individual fibre morphometrics were determined by histochemical analysis. Frozen muscle blocks were mounted onto stainless steel chucks with OCT compound, and placed in a cryostat chamber (HM505, Microm,Walldorf, Germany) at –20°C. Frozen cross sections (10 μm thickness) were cut and melted directly onto glass slides and air-dried for approximately 3 min. The sections were returned to the cryostat chamber for storage in an air-tight container prior to staining. Best results were achieved for muscle that was sectioned and stained on the same day. Slides were removed from the cryostat chamber and stained immediately for succinic dehydrogenase (SDH) activity to differentiate between oxidative and glycolytic muscle fibres: 50 mmol l^–1 phosphate buffer, 80 mmol l^–1 sodium succinate, 1 mg ml^–1 Nitro blue tetrazolium, adjusted to pH 7.6 with NaOH(Rowlerson and Spurway, 1988). Sections were incubated in the staining medium in air-tight, light-proof containers for at least 90 min. Each slide was washed three times in distilled water, air dried and mounted with Crystal/Mount™ (Biomeda Corp.,Burlingame, CA, USA). Each section was photographed at a 250×magnification with a digital camera (DFC280, Leica, Wetzlar, Germany) mounted to a compound light microscope (BH2, Olympus, LeHigh Valley, PA, USA). Digital images were analysed with SigmaScan™ (SPSS Inc., Chicago, IL, USA) to determine muscle cross-sectional area, fibre area, fibre number and fibre density of each whole muscle section.

Isometric analysis was used to determine the twitch and tetanus kinetics of the isolated hindlimb muscles. The bone at either end of the muscle preparation was clamped via crocodile clips to a strain gauge (UF1,Pioden Controls Ltd, Canterbury, UK) at one end, and a motor arm (V201, Ling Dynamics Systems, Royston, UK) attached to an LVDT (Linear Variable Displacement Transformer, DFG 5.0, Solartron Metrology, Bognor Regis, UK) at the other. The LVDT was used to monitor the length changes delivered to the muscle preparation. The whole of the muscle, tendon and bone preparation was then maintained within the organ bath at 25±0.5°C in circulating,oxygenated McKenzies frog Ringer solution. The preparation was held at constant length and stimulated via parallel platinum electrodes to generate a series of twitches. Stimulus amplitude, pulse width and muscle length were adjusted to determine the stimulation parameters and muscle length corresponding to maximal isometric twitch force. Time to peak twitch and time from peak twitch to half relaxation were measured via a Powerlab 4SP(AD Instruments, Colorado Springs, CO, USA) and viewed with Chart 5.0 software(AD Instruments). An isometric tetanic force response was elicited by subjecting the muscle to a 300 ms train of stimulation. Stimulation frequency(130–150 Hz for iliofibularis; 140–170 Hz for sartorius) was altered to determine maximal tetanic force. Time to half peak tetanic force and time from last stimulus to half tetanic force relaxation were measured. A rest period of 5 min was allowed between each tetanic response.

The work-loop technique was used to determine the power output of muscles during cyclical length changes (Josephson,1993). Unlike fixed-length isometric studies and fixed-load isotonic studies of muscle performance, the work-loop technique allows measures of muscle power output under length and activation changes that are generally more indicative of in vivo contractile performance(Caiozzo, 2002; James et al., 1996; Josephson, 1993). Each muscle preparation was subjected to a set of four sinusoidal length changes symmetrical around the length found to generate maximal twitch force. The muscle was stimulated using the stimulation amplitude and frequency found to yield maximal isometric force. Electrical stimulation and length changes were controlled via a data acquisition board and a custom-designed program developed with TestPoint software (Measurement Computing, Norton, MA, USA). Muscle force was plotted against muscle length for each cycle to generate a work loop, the area of which equated to the net work produced by the muscle during the cycle of length change(Josephson, 1993). The net work produced was multiplied by the frequency of length-change cycles to calculate net power output. The total strain of length-change cycles was maintained at 0.10 throughout all experiments, which equated to ± 5% of resting muscle length. The cycle frequency of length change was altered up and down within the range 1–9 Hz for iliofibularis and 2–10 Hz for sartorius to generate power output-cycle frequency curves. During these length change cycles the muscle was usually subjected to phasic stimulation (i.e. active work-loop cycle) but sometimes these length changes were performed without stimulation (passive work-loop cycle) to monitor the net work done on the muscle during the length change cycle. Every 5 min the muscle was subjected to a further set of four work-loop cycles with stimulation duration and stimulation phase parameters being altered until maximum net work was achieved at each cycle frequency. A set of control sinusoidal length change and stimulation parameters were imposed on the muscle every 3–4 sets of work loops to monitor variation in the muscle's ability to produce power/force. Any variation in power was found to be due to a matching change in ability to produce force. Therefore, the power produced by each preparation, prior to the fatigue run, was corrected to the control run that yielded the highest power output, assuming that alterations in power generating ability were linear over time. All muscles still produced over 80%of maximal control run power prior to the fatigue run. On completion of the power output-cycle frequency curve each muscle was subjected to a fatigue run consisting of 30 work-loop cycles at the lowest cycle frequency that had earlier been found to produce 80% of maximal power output.

At the end of the isometric and work-loop experiments, the bones and tendons were removed and each muscle was blotted on absorbent paper to remove excess Ringer solution. Wet muscle mass was determined to the nearest 0.0001 g using an electronic balance (BP211D, Sartorius). Mean muscle cross-sectional area was calculated from muscle length and mass assuming a density of 1060 kg m^–3 (Mendez and Keys,1960). Maximum isometric muscle stress was then calculated as maximum tetanic force divided by mean cross-sectional area (kN m^–2). Normalised muscle power output was calculated as power output divided by wet muscle mass (W kg^–1). Each muscle was dried for 48 h in an oven at 37°C, moved to room temperature for 1 h and then reweighed to calculate dry muscle mass (to the nearest 0.0001 g) and water content (% wet mass).

Frog length (mm) was not correlated with any of the parameters recorded(Pearson correlation; P>0.05), therefore data sets were not normalised to body size (SVL) prior to analysis. Dispersion measurements are given as standard error. In all cases P=0.05 was accepted for statistical significance.

Frog sizes and fibre size distributions were compared using a rank sum t-test, while muscle morphological data was analysed by two-way ANOVA using SigmaStat™. Where differences were detected they were localised by a Holm–Sidak multiple comparison test.

In most cases control and experimental results were compared using independent sample t-tests with correction for unequal variances where appropriate. Control to treatment comparison of power output-cycle frequency data was performed using ANCOVA, with treatment as the factor and cycle frequency as the covariate. Control to treatment comparison of relative power output data during the fatigue tests was performed using ANCOVA, with treatment as the factor and cumulative power as the covariate.

There was no significant difference in dry muscle mass between either muscle type (F_1,20=0.616, P=0.443) or treatment group (F_1,20=0.939, P=0.346)(Fig. 1A). There was no overall difference in muscle cross-sectional area (CSA) between control and 9-month aestivator muscles (F_1,21=0.762, P=0.394), though there were significant differences between muscle types, with the sartorius muscle having a significantly smaller CSA compared to the iliofibularis(F_1,21=13.102, P=0.002)(Fig. 1B).

There was no significant difference in the total number of fibres present in the muscle cross-sections, between either muscle type(F_1,19=2.863, P=0.110) or treatment group(F_1,19=2.557, P=0.129)(Fig. 2A). Overall, there was no significant difference in fibre density between muscle type(F_1,19=4.485, P=0.06) or treatment group(F_1,19=0.882, P=0.362)(Fig. 2B), though there was a significant interaction between muscle type and treatment group(F_1,19=11.53, P=0.004). A Holm–Sidak multiple comparison test revealed that control sartorius fibre density was significantly greater than 9-month aestivator sartorius fibre density(P=0.007) and control iliofibularis density (P=0.001). Control sartorius density was not more dense than 9-month aestivator iliofibularis density, but only arbitrarily (P=0.050). This suggests that, overall, sartorius muscle fibres were more densely packed within the muscle. There was no significant difference in the relative proportions of muscle fibre types (oxidative versus glycolytic) between muscle type(F_1,7=0.000175, P=0.992) or treatment group(F_1,7=0.000175, P=0.992)(Table 1). Oxidative fibres accounted for greater than 63% of the total number of fibres within each muscle, with the control iliofibularis possessing the greatest proportion of oxidative fibres (76%). Glycolytic fibres accounted for approximately 24% of the total number of fibres within each muscle, the greatest proportion of which was within the 9-month treatment sartorius muscle (37%).

There was a significant difference in the CSA of individual fibres between fibre type (F_1,41=48.54, P<0.001) and treatment group (F_3,41=17.15, P<0.001). Within the iliofibularis muscle, the CSA of oxidative fibres from 9-month aestivators were significantly smaller than the control oxidative fibres(P=0.041), and 9-month aestivator glycolytic fibres were significantly smaller than control glycolytic fibres (P<0.001)(Fig. 3A). There was no significant difference between treatment groups for fibre CSA of the sartorius muscle (P=0.221), though there was a significant difference between fibre types: glycolytic fibres were significantly larger than oxidative fibres in both controls (P=0.002) and 9-month aestivators (P=0.037)(Fig. 3B). There was no significant difference in the relative size distribution of fibres between treatment group or muscle type (P>0.05), except for the iliofibularis muscle. In this case, 9-month aestivator oxidative fibres were skewed left, with significantly more fibres occurring in the smallest size classes (P=0.045) (Fig. 4A).

Twitch activation, twitch relaxation and tetanus relaxation times were significantly longer in the 9-month aestivator iliofibularis muscle than in the control iliofibularis muscle (Table 2; P<0.01); however there were no significant differences in the sartorius muscle. There was no significant difference in tetanus activation time between control and 9-month aestivators for either the iliofibularis or sartorius muscles. Maximum twitch stress and maximum tetanic stress tended to increase with aestivation in both the iliofibularis and sartorius muscles (Table 2; P>0.05).

Nine-months aestivation did not affect the peak power output produced by either the iliofibularis or sartorius muscles(Table 2). Power output-cycle frequency relationships did not differ between control and treatment iliofibularis muscles (Fig. 5A; P=0.29). Aestivation caused a significant shift in the sartorius muscle power output-cycle frequency curve(Fig. 5B; P=0.004),with treatment muscles producing lower power output at higher cycle frequencies. Control sartorius muscles produced higher maximal power output at a higher cycle frequency than the iliofibularis muscle, with the 9-month aestivator sartorius muscle producing a power output-cycle frequency curve similar to that of the iliofibularis. Aestivation tended to reduce fatigue resistance in both the iliofibularis and the sartorius muscles, as evidenced by a more rapid initial decline in 9-month aestivator muscle power output during bouts of repeated work loop cycles(Fig. 6A,B; P=0.53 and P=0.09, respectively).

The results of this study clearly demonstrate that hindlimb muscles of Cyclorana alboguttata do not exhibit a typical atrophic response to prolonged immobilisation during aestivation. Prolonged aestivation of 9-months duration had little or no effect on either the morphology or contractile properties of the iliofibularis and sartorius muscles.

The sartorius and iliofibularis muscles of the 9-month aestivators maintained dry muscle mass and whole muscle cross-sectional area at control levels (Fig. 1A,B). Both sartorius and iliofibularis act to abduct the femur and flex the knee(Duellman and Trueb, 1994),therefore these are not the main extensor muscles used to power jumping. As such, they might represent skeletal muscles less likely to be protected from atrophy during a challenge such as aestivation.

The absence of change at the whole muscle level is in stark contrast to the effects of immobilisation or dormancy previously reported in other animals(Adams et al., 2003; Baldwin and Haddad, 2001; Booth, 1977; Booth and Seider, 1979b). Similar significant reductions in muscle mass were observed in guinea pig muscle after immobilisation for 28 days(Maier et al., 1976). Hibernation caused significant atrophy in the muscles of golden-mantled ground squirrels Spermophilus lateralis(Wickler et al., 1991a), big brown bats Eptesiscus fuscus(Yacoe, 1983) and hamsters(Wickler et al., 1987). However, aestivation for 3 and 6 months in C. alboguttata has previously been shown to have no effect on the wet muscle masses of locomotory muscles (Hudson and Franklin,2002a; Hudson et al.,2006).

There were no significant changes to total fibre number between control and 9-month aestivator groups for either muscle type in C. alboguttata(Fig. 2A), which rules out loss of fibres (hypoplasia) during aestivation. Fibre density declined significantly after aestivation in the sartorius muscle only(Fig. 2B). Combined with the maintenance of sartorius dry muscle mass and fibre number, it is possible that this reduction in fibre density reflects an increase in sartorius muscle length during aestivation. Aestivating C. alboguttata adopt a characteristic water-conserving posture(Withers and Richards, 1995),and it is possible that this behaviour might confer a postural advantage to selected hindlimb muscles, such as the sartorius, by maintaining the muscle in a lengthened position. It has been demonstrated in previous studies that fixing muscle in the lengthened position during immobilisation can delay the onset of muscle disuse atrophy (Booth,1977; Goldspink,1977). Our results are unlike previous studies in which the cross-sectional areas of muscle fibres and numbers of fibres have significantly decreased after immobilisation and dormancy. Immobilisation caused significant reductions in the diameter of guinea pig soleus muscle fibres after 21 days of immobilisation(Tomanek and Lund, 1974), the cross-sectional area of cat tibialis muscle(Nordstrom et al., 1995) and fibre number in rat soleus muscle (Booth and Kelso, 1973). Hibernation caused a significant reduction in the cross-sectional areas of fibres from the soleus and extensor digitorum longus muscles of hibernating golden-mantled ground squirrels Spermophilus lateralis (Steffen et al.,1991) though not in hibernating black bears Ursus americanus (Tinker et al.,1998).

The most dramatic morphological change observed in this study was the significant reduction in iliofibularis muscle fibre cross-sectional area(Fig. 3A). Based on the results of previous studies (e.g. Booth and Kelso,1973; Maier et al.,1976; Tomanek and Lund,1974) we had predicted that the predominantly oxidative iliofibularis muscle would be more susceptible to muscle disuse atrophy than the glycolytic sartorius muscle. During normal activity, oxidative muscles tend to be used frequently, for extended periods of time and at low intensities. In comparison, fast-twitch glycolytic muscles are used infrequently, for short periods and at high intensity. When oxidative muscles become inactive the scope of the change in the level of activity is much greater than that experienced by glycolytic, fast-twitch muscles(Hudson and Franklin, 2002b). Additionally, due to the comparatively greater numbers of mitochondria contained within them, oxidative fibres produce comparatively larger quantities of reactive oxygen species as a byproduct of normal cell metabolism. Reactive oxygen species, or free radicals, are responsible for stochastically damaging cell architecture, including proteins and lipids,which leads to cellular apoptosis and hence muscle disuse atrophy(Kondo et al., 1993a; Kondo et al., 1993b; Kondo et al., 1993c; Kondo et al., 1994). The dramatic downregulation of activity in the oxidative muscle may trigger profound changes in the biochemical function of the muscle mitochondria. As such, it is possible that the susceptibility of oxidative muscles to muscle disuse atrophy is due to changes in mitochondrial density or architecture rather than compositional fibre changes. In other words, the ultimate consequences of muscle disuse atrophy may be more important at the whole muscle level rather than at the individual fibre level.

The reduction in cross-sectional area of the iliofibularis muscle fibres,but not sartorius muscle fibres, supports our initial hypothesis; however, the results of the histochemical analysis show that the iliofibularis muscle of C. alboguttata is not a homogenously oxidative muscle(Table 1). If oxidative fibres are more susceptible to muscle disuse atrophy than glycolytic fibres, we would anticipate that the oxidative fibres would have been preferentially affected over the glycolytic fibres. In actuality, the significant reduction in fibre cross-sectional area in the iliofibularis muscle of 9-month aestivators occurred in both oxidative and glycolytic fibre types(Fig. 3A).

The enzyme histochemistry component of this study does not support the hypothesis that prolonged aestivation induces the transition of fibre types from characteristically (histologically) slow, oxidative to fast, glycolytic. The relative proportions of the two fibre types, measured as a percentage of total number of fibres present in the muscle cross-section, was consistent between controls and 9-month aestivators(Table 1). Furthermore, the size-distribution of fibres, as measured by cross-sectional area, remained consistent between controls and 9-month aestivators(Fig. 4A–D). There was a trend for larger iliofibularis glycolytic fibres (>10 000μm²) to occur more frequently after 9 months of aestivation;however, it was not significant (Fig. 4B).

Previous research has demonstrated that prolonged aestivation has no effect on skeletal motor nerve terminals or membrane potentials in C. alboguttata (Hudson et al.,2005); however, the effect on muscle mechanics was more complex. Twitch activation, twitch relaxation and tetanus relaxation times were significantly longer in 9-month aestivator iliofibularis muscle, but there were no significant changes in sartorius muscle(Table 2). The findings for the faster twitch sartorius muscle concur with previous work on 3-month aestivator C. alboguttata gastrocnemius muscle(Hudson and Franklin, 2002a),suggesting that the isometric kinetics of relatively fast muscles are unaffected by aestivation. The apparent slowing of isometric kinetics of the relatively slower twitch iliofibularis muscle was contrary to the findings of most previous studies (Maier et al.,1973; Nordstrom et al.,1995) and the reverse of our hypothesis. However, the slowing of isometric kinetics in our study was consistent with other findings (Rourke,2004), which showed a non-significant tendency for the expression of myosin heavy chain protein isoforms to shift towards slower isoforms in hibernating ground squirrels. The observed fast-to-slow transition in squirrels may be due to shivering thermogenesis during periodic arousal from hibernation; however,this is clearly not the case for an amphibian aestivator.

There were no significant differences in isometric stress between controls and 9-month aestivators for either sartorius or iliofibularis muscles(Table 2). This is similar to the results found previously in 3-month aestivator C. alboguttatagastrocnemius muscle (Hudson and Franklin,2002a). 12–16 weeks of aquatic hibernation in Rana temporaria has also been previously shown not to affect sartorius muscle isometric stress (West et al.,2006), therefore skeletal muscle stress appears to be conserved in the natural models of anuran dormancy studied so far.

There were no aestivation-related changes to power output-cycle frequency curves in the slow-twitch iliofibularis of C. alboguttata, but the fast-twitch sartorius became slower after 9 months of aestivation(Fig. 5B), which is contrary to the results obtained from previous studies(Diffee et al., 1991) and the reverse of our hypothesis. This result may have been indicative of fibre-type remodelling of the sartorius, though there was no evidence to support this(Table 1). In contrast,sartorius muscle isovelocity power output remained stable after 12–16 weeks of aquatic hibernation in Rana temporaria, suggesting that the muscle mechanical properties of over-wintering frogs remain stable. 9 months of aestivation had no significant effect on fatigue resistance in either the iliofibularis or sartorius muscles (Fig. 6A,B), which was similar to the findings in the gastrocnemius muscle from 3-month aestivator C. alboguttata(Hudson and Franklin,2002a).

Within the present study, the effect of aestivation on isometric kinetics of skeletal muscle partially contradicts the findings from work-loop assessment of mechanical properties. However, previous studies have demonstrated that isometric kinetics and force–velocity properties alone are poor predictors of work-loop shape(Caiozzo, 2002; James et al., 1996). Work-loop shape is also affected by force enhancement, shortening deactivation and passive properties of muscle (Josephson,1993). Therefore, the work-loop technique can give a more functionally relevant estimate of muscular performance capabilities by assessing performance under dynamic conditions(Josephson, 1993).

The apparent slowing of both muscle types, as evidenced by the isometric contractile properties data, indicated that prolonged aestivation did not induce a transition from slow to fast twitch in the iliofibularis and sartorius muscles. As such, prolonged aestivation may have an effect on muscle physiology or biochemistry that indirectly changes the isometric properties of the muscle tissue without affecting fatigue resistance. Changes to muscle mitochondrial structure, function or density may be one possible explanation,though it has been demonstrated that mitochondrial density is not correlated with contraction speed and fatigability in individual fibres(Lipska et al., 1998). A decline in the energetic stores of the individual muscles, rather than fibre type changes, may be responsible for this outcome; however, levels of muscle metabolites, such as phosphorylated creatine and ATP, are known to remain stable in the muscles of cold-submerged hibernating R. temporaria(West et al., 2006), though it should be noted that the quiescent period of this anuran species is substantially shorter than that experienced by C. alboguttata. Furthermore, it has been postulated that one of the mechanisms acting to protect animals from muscle disuse atrophy during dormancy is low body temperature, in conjunction with lowered metabolic rate(Hudson and Franklin, 2002b). In this way, C. alboguttata is truly remarkable as it is the only animal model ever investigated that does not exhibit any muscle disuse atrophy during a temperature-independent dormancy.

The present study demonstrated that 9 months of aestivation had little effect on muscle morphology, histochemistry and dynamic mechanical performance in C. alboguttata. There was no skeletal muscle atrophy of the type normally associated with either experimental limb disuse or hibernation. Whole muscle mass, muscle cross-sectional area, fibre number and proportions of fibre types remained unchanged after prolonged disuse, and relatively minor morphological changes to fibre cross-sectional area and density were observed.

Overall, prolonged aestivation had little effect on the isometric properties of the skeletal muscle of C. alboguttata, with moderate,though significant, changes occurring in the slow-twitch iliofibularis muscle. Though the results of the isometric analysis suggested that the muscles were becoming slower after 9 months of aestivation, the results of the work-loop analysis clearly demonstrate that the contractile properties of aestivating C. alboguttata are maintained at control levels. Despite the fact that significant changes are occurring at the muscle structural level, overall muscle performance, and hence whole animal performance, is preserved. Further investigation is required to elucidate the mechanisms that prevent skeletal muscle disuse atrophy during aestivation.

This project was funded by an ARC Discovery Project Grant to C.E.F.; R.S.J. was supported by a Royal Society International Short Visit Grant. B.L.S. was supported by a University of Queensland Mid-Year Scholarship. The authors would like to thank Rebecca Cramp for technical assistance. The experimental procedures were in full compliance with AEC and EPA regulations and performed under AEC # SIB/431/05/UQ and SPP# WISP03572406.