| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Biological Sciences, University of Calgary, Calgary, Alberta, Canada; and 2Department of Biology, University of California, Riverside, California
Submitted 31 March 2005 ; accepted in final form 3 June 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
54 and 45% the mass of normal muscles, respectively, but the plantaris muscles were not different in mass and soleus muscles were actually 30% larger. In spite of the increased mass, contractile characteristics of the soleus were unchanged in any notable way between mini and normal mice. However, medial gastrocnemius muscles in mini mice were changed markedly toward a slower phenotype, having slower twitches; demonstrated a more curved force-velocity relationship; produced about half the mass-specific isotonic power, 20–50% of the mass-specific cyclic work and power (only 10–25% the absolute power if the loss in mass is considered); and fatigued at about half the rate of normal muscles. These changes would promote increased, aerobically supported running activity but may compromise activities that require high power, such as sprinting. experimental evolution; fatigue; muscle mechanics; power; selective breeding; work
70% more revolutions/day after 10 generations of selection (34), 100% more after 14 generations (17), and 170% more revolutions and about double the average running speed after 23 generations (5, 11, 13). These mice have been the focus of anatomical, behavioral, physiological, and psychological investigations on the correlated effects of such selection (e.g., Refs. 7, 11, 23–25, 28, 29, 35–37).
More recently it has been noted that some individuals in both the selected and control lines express a small-muscle phenotype in which the plantar flexor muscle group (soleus, gastrocnemius, and plantaris) is 44–50% lighter than normal for the body mass (5, 12, 17). Evidence suggests that this small-muscle phenotype, coined "mini muscle" (17), is coded by an allele that was present in the base population at a frequency of 7% and appears to be inherited as a single, autosomal recessive (12). Although the mini-muscle phenotype has fluctuated in frequency between 0 and 10% in the one random-bred control line in which it is observed, it has increased in frequency to 50% by generation 22 in the two selected lines that ever expressed the phenotype, and statistical analyses indicate that it has been favored by the selection protocol (12). The dramatic increase in its expression in these two lines of activity-selected mice leads to obvious questions about its functional significance.
Selective breeding may have favored mini muscles because they possess contractile abilities conducive to powering high levels of endurance running and/or because the underlying allele has pleiotropic effects, such as reduced body mass and increased ventricular mass, which may improve endurance running even if the muscles themselves are not adaptive (12). Individuals with mini muscles sometimes do (12) but sometimes do not (17) appear to have smaller body masses than their normal counterparts, and running did not differ significantly between normal and mini mice in the selected lines at generations 22–24 (5, 12). On the other hand, mini muscles show increased expression of HSP72 independent of access to running wheels (5), and they have increased mass-specific aerobic capacities that almost fully compensate for the reduced muscle mass as well as metabolic enzyme profiles reminiscent of muscles in endurance-trained mammals (17). These observations have left unclear whether the possession of mini muscles or the allele responsible actually confers a selective advantage in the context of the selective breeding protocol.
The objectives of the present study were to compare the voluntary wheel-running behavior of mice with normal and mini muscles in the selected lines at generations 36–38 and then to compare the contractile characteristics of these mini and normal muscles to understand how they may be different and whether such differences would promote high wheel running from a mechanical and energetic perspective. Such differences might include increased resistance to fatigue, perhaps accompanied by faster toward slower fiber-type transitions and increased oxidative capacity as has been observed in mouse muscle during endurance exercise training (1, 3). These changes would likely be accompanied by slowing of the contraction kinetics and reduced mass-specific power output, alterations that would neither promote nor hinder endurance running directly but might constitute an unavoidable "trade-off" for high-power activities such as sprinting. Although comparative studies of species differences have provided correlational evidence to suggest many muscle characteristics that are adaptive for various types of behavior, we know of no examples in which an experimental evolutionary approach (6, 7, 11, 12, 16, 17, 28, 38) has actually been used to identify such characteristics nor of a demonstration of the predicted physiological trade-off between endurance capacity and high power in locomotor muscles of a population under clearly defined selection.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Voluntary wheel running was measured over a 6-day period, exactly as in the routine testing protocol of the selection experiment (34). Mice were 48 ± 0.49 (SE) days old (range 39–55 days) at the beginning of the wheel tests. Mice were housed individually in cages with access to 1.12-m-circumference activity wheels. A computer monitored the total number of wheel revolutions per day, the number of 1-min intervals that the mice were active on the wheels each day (i.e., intervals that showed at least one revolution), and the maximum revolutions per minute observed in any 1-min interval during each day. An approximate average revolutions per minute during daily activity was attained by dividing the total number of wheel revolutions per day by the number of minutes that the mouse was active on the wheel. Values reported are averages of days 5 and 6, as is used to choose breeders in the selected lines (on the basis of total revolutions). Mice were then air shipped from University of California, Riverside to Calgary, where they were housed as littermates in groups of four in filter-top cages with food and water ad libitum, 12:12-h light-dark cycle, at room temperature until being killed for measurements of muscle performance.
Muscles were isolated and mounted in experimental chambers, and initial preparations for measurements were made as described previously (e.g., Refs. 40, 42). Briefly, mice were weighed and then killed by cervical dislocation, and the left leg was removed, skinned, and placed in a dish containing physiological saline (composition in mM: 137 NaCl, 3 KCl, 2 CaCl2, 1 MgSO4, 1 NaH2PO4, 24 NaHCO3, and 10 glucose, pH 7.4) bubbled with 94:6 O2-CO2. The dissection stage was cooled to 15°C, and the saline was changed frequently during dissections to inhibit proteases and limit muscle deterioration. The soleus muscle and medial gastrocnemius muscle (henceforth termed the gastrocnemius unless specified otherwise) were isolated for measures of mechanical performance. The Achilles tendon was split longitudinally so that a section of tendon remained attached to the distal end of each muscle. The proximal tendon of the soleus was cut from the tibia, whereas a small fragment of bone was left attached to the short, proximal tendon of the gastrocnemius. Segments of 5-0 silk suture were tied to the tendons and bone at the end of each muscle. The muscles were then transferred to chambers filled with circulating physiological saline, bubbled with 94:6 O2,CO2, and maintained at 27°C. It was not necessary to maintain normal body temperatures because only relative comparisons between normal and mini muscles were required for purposes of this study; a cooler temperature was selected to sustain viability of the preparations.
Two apparatuses were used to accommodate the smaller soleus and larger gastrocnemius. The tendon on one end of the soleus was tied to the tip of the arm of a model 350 servomotor (Cambridge Technology), and the tendon on the other end was tied to a stainless steel pin attached to a BG-50G force transducer (Kulite Semiconductor Products). The tendon on one end of the gastrocnemius was tied to the arm of a model 305B-LR servomotor (Aurora Scientific), and the tendon on the other end was tied to a stainless steel pin attached to an ELG-V-500G load cell (Entran Sensors and Electronics). The servomotors were used to control and measure the length of the muscles and the force transducers to measure force. Platinum stimulating electrodes were placed alongside the muscles and connected to custom-made, low-impedance current sources gated by Grass SD9 stimulators that were in turn gated by a command signal from a computer. The servomotor and stimulator were controlled by custom software written in LabView 6.1 through PCI-MI0-16-E4 data-acquisition and control cards (National Instruments). Signals from the servomotor, force transducer, and stimulator were collected at 5 kHz on the computer by using custom software written in LabView.
The stimulus voltage was adjusted to 150% of that required to elicit maximum twitch force (1-ms stimulus pulse duration). The muscle length was systematically varied until the length giving maximal isometric twitch force was found. Muscle fiber length was then measured by using a calibrated ocular micrometer, and this length was used for the remainder of the experiments.
Measures of mechanical performance of muscles. Several measurements were made to assess contractile characteristics relevant to the capacity for power running, including the ability to produce force, rate of contraction and relaxation, ability to shorten under isotonic load, ability to do work as might occur during locomotion, and fatigue resistance.
Isometric twitches were recorded from which measures of the rates of activation and relaxation were made. These included the time for force to increase from 10 to 90% of maximal during contraction (T10–90), the time for force to fall from 90 to 10% of maximal during relaxation (T90–10), and the duration of the twitch when force is at or above half-maximal amplitude (T50–50). Maximum, isometric tetanic force was elicited with a train of stimulus pulses lasting 500 ms for the soleus and 100 ms for the gastrocnemius. These stimulus durations were long enough to ensure that force reached or closely approached a plateau. A stimulus frequency of 125 Hz was used for the soleus and 175 Hz for the gastrocnemius. Force was expressed relative to the cross-sectional area of the muscle, which was calculated from the muscle length, mass, and assuming a density of 1,050 kg/m3.
Force-velocity characteristics were measured in fully activated muscle and thus reflect the inherent rate functions of the cross bridges and their impact on the ability of the muscle to shorten under load and hence to generate force and power. The muscle was first stimulated tetanically until force reached a plateau. Muscle length was then shortened rapidly by 0.1–0.5% depending on the shortening velocity under study, an amount sufficient to cause force to drop close to the level it would attain during subsequent isovelocity shortening. The muscle was then shortened at a constant velocity, and the associated stable force was measured. This process was repeated over a series of shortening velocities so that forces ranging from isometric to near zero were obtained. The force-velocity data were then fitted to the Hill equation (15) from which the maximum shortening velocity (Vmax) and the curvature of the force-velocity relationship were obtained. Vmax varies as a function of the detachment rate of the cross bridges and was obtained by extrapolating the Hill equation to zero force. Although extrapolation of a Hill curve tends to slightly overestimate the intercepts (26), visual comparison of our data sets to the mathematically derived Vmax confirmed that the result was legitimate and faithfully represented the characteristics of the muscle. The curvature of the force-velocity relationship was described by the ratio a/Po where a is a constant from the Hill equation and Po is isometric force. This ratio reflects the inherent ability of the cross bridges to produce power, where faster, more powerful muscles have straighter relationships and higher a/Po ratios.
The product of shortening velocity and the associated force yields isotonic power, which rises with increasing shortening velocity, reaches a maximum, and then falls with further increases in velocity. Maximum isotonic power was obtained by rearranging the Hill equation to solve for power, taking the first derivative with respect to shortening velocity, equating this to zero, solving for force (which is the force at which power is maximal), and then substituting this force back into the original Hill equation to calculate the associated shortening velocity from which maximal power can then be calculated. This is the power the muscle produces when shortening at a constant, optimal velocity and when the muscle is fully and continuously activated. See Josephson (19) for further details and discussion of analysis of force-velocity characteristics and power output of muscle.
Although maximum isotonic power as calculated above provides useful information about muscle, animals do not use their muscles in this fashion during locomotion (19). Thus cyclic work and power were measured by the work-loop technique, which provides a more realistic estimate of the ability of muscles to produce repetitive movements such as running (e.g., Refs. 21, 41). Briefly, the servomotor was used to impose sinusoidal changes in muscle length that mimic the cyclic lengthening and shortening that muscles experience during locomotion. The amplitude of the length change cycle (strain) was fixed at ±5% of the muscle resting length, and the frequency of the oscillation was set to 2 and 4 Hz for the soleus and 4 and 8 Hz for the gastrocnemius; these frequencies and amplitudes were selected to encompass a range over which the muscles were expected to produce near-maximal and also submaximal work and power output and which is realistically experienced during running by mice and rats (e.g., Refs. 18, 39). Although absolute power and the cycle frequency at which it is maximized would be less at the experimental temperature of 27°C than at body temperature, we only required a relative comparison between power from normal and mini muscles. Thus selection of frequencies and strains that maximized power or precisely mimicked conditions in a running mouse were not essential.
The muscle was stimulated phasically during the length change cycle, where phase is the timing of the onset of stimulation expressed relative to the length change cycle. The stimulus phase and duration were altered systematically until the net work done by the muscle was maximal. Net work was measured as the integral of force with respect to muscle length during a complete length cycle; it is the difference between the work done by the muscle when it shortens and the work required to lengthen the muscle and is the net mechanical energy that the muscle contributes during the cycle. Cyclic power was obtained by multiplying the net work done per cycle by the cycle frequency. Work and power were quantified relative to muscle mass. Cyclic power is conceptually different from maximum isotonic power calculated from force-velocity data because it accounts for both the work done by the muscle while it shortens and the energy and time required to lengthen the muscle. Also it is measured while the muscle shortens and lengthens over unsteady and realistic length trajectories and velocities, and it accounts for the dynamic and constraining natures of muscle activation and relaxation during each cycle (e.g., Refs. 8, 20).
The rate that muscles fatigue can be used as an indication of their suitability for powering sustained activities such as endurance (wheel) running. Normal and mini muscles were thus subjected to a fatiguing protocol to compare the rates at which they fatigued. In these protocols the soleus underwent 500-ms-duration, isometric tetanic contractions every 15 s, whereas the gastrocnemius underwent 100-ms-duration, isometric tetanic contractions every 15 s. These were continued until force declined to 70% of the prefatigue level. Because the ability to produce work and power is more severely hampered during fatigue than the ability to produce isometric force, and because the ability to produce work and power is most relevant to movement (33, 42), cyclic work was also recorded during the fatigue protocol at the times when isometric force had decreased by 10, 20, and 30%. Work during fatigue was recorded at a 2-Hz cycle frequency in soleus muscle and at 4 Hz in gastrocnemius muscle by using the same stimulus parameters that resulted in maximum work during the previous measures. At the conclusion of the fatigue protocol, the muscles were allowed to recover until it was evident that the majority of recovery was complete (typically 30–60 min), and preparations that did not show force recovery of at least 85% were discarded. Force and work were then standardized to the initial, unfatigued value, and plotted as a function of time during fatiguing stimulation. The slopes of the least-squares linear regressions relating relative force or work to time were used as measures of the rates of fatigue.
After the conclusion of the experiment the muscle was removed from the chamber, trimmed of external tendon and any fat or other loose connective tissue, blotted on filter paper, sealed in a 0.5-ml centrifuge tube to prevent desiccation, and weighed on an analytical balance. In some animals the lateral gastrocnemius and plantaris muscles were also isolated and weighed to obtain a complete record of how the masses of the components of the entire plantar flexor muscle group have changed in mini mice.
Statistical analysis. Wheel-running data were obtained from 40 normal and 39 mini mice, as classified postmortem after direct observation of the muscle phenotype; a marked reduction in mass of the gastrocnemius muscle was used to identify the mini phenotype. A subset of these same mice was used for muscle contractile measures. Contractile experiments were performed on six soleus and gastrocnemius muscles from both normal and mini mice in line 6, and six soleus and gastrocnemius muscles from mini mice in line 3. Measurements from line 6 normal, line 6 mini, and line 3 mini muscles were compared by ANOVA and analysis of covariance. Covariates in the analyses of covariance included body mass, age, and wheel resistance, depending on the trait. Body masses before and after the wheel running experiments were compared by paired t-tests within groups. Data that did not satisfy assumptions of normality or variance equality were transformed before statistical analysis as indicated in RESULTS. Tests are two-tailed, and P values should be interpreted accordingly. All data are presented as means and SE.
In an attempt to avoid false positive results stemming from repeated comparisons of data from the same individual mice, we employed the False Discovery Rate procedure described by Curran-Everett (10). On the basis of 1) a desired false discovery rate of 0.05, 2) a total of 50 statistical comparisons in the study, and 3) the statistical P values of all the comparisons from the study, the procedure establishes that comparisons with P < 0.03 should be considered statistically significant, and we emphasize discussion of those results. Furthermore, we report test statistics, degrees of freedom, and P values for all statistical tests; independent conclusions may be drawn from these as desired.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
4%, P = 0.015; Table 1). Thus, at the end of wheel trials, mini mice were 7% lighter on average than normal mice (Table 1).
|
Muscle masses.
Of the line 3 mice, 100% of the 22 individuals examined showed the mini muscle phenotype. Of the line 6 mice, 37% of the 27 individuals examined showed the mini muscle phenotype. After adjustment for variation in body mass, the mass of the combined plantar flexor muscle group (soleus + medial and lateral gastrocnemius + plantaris) did not differ between the mini mice in lines 3 and 6 but was only 54% the mass of these muscles in normal mice (Table 2), similar to the original report (12). However, the differences between mini and normal mice were not uniform across the four muscles within this group; rather, the soleus muscles of mini mice were actually 30% heavier, the medial gastrocnemius muscles were 54% of the mass in normal mice, the lateral gastrocnemius muscles were 45% of the mass in normal mice, and the two strains showed no statistical difference in plantaris mass (Fig. 1, Tables 2 and 3).
|
|
|
For the soleus, T10–90 was longer in normal than mini muscles from mice in line 6 but not line 3, and it was not different between mini muscles from mice in lines 3 and 6 (Fig. 2); when the data for mini muscles from lines 3 and 6 were pooled, there was also a highly significant difference between mini and normal muscles (Fig. 2). However, the absolute difference was only 3 ms (20%), which does not constitute a marked change in overall twitch speed. T50–50 and T90–10 of soleus muscles were significantly faster in line 6 mini mice compared with either the line 6 normal or line 3 mini mice (Fig. 2); these differences were on the order of 20–25% and result in a twitch that is considerably faster from a functional perspective. In gastrocnemius muscle, T10–90 was not different between mini and normal muscle or lines (Fig. 2). T50–50 was marginally nonsignificantly different between the groups, but it was significantly different in normal compared with the combined mini muscle groups and was highly nonsignificantly different between mini muscles from lines 3 and 6 (P = 0.74) (Fig. 2). T90–10 was significantly slower in the gastrocnemius of both mini groups compared with normal muscles (Fig. 2).
|
|
Cyclic work and power produced by the muscles were measured at two different cycling frequencies, analogous to different stride frequencies during running (Fig. 4). The net work done per cycle decreased with increasing cycle frequency in both soleus and gastrocnemius muscle, as is typical (e.g., Ref. 18). The power outputs remained relatively unchanged with changes in cycle frequency, suggesting the muscles were working at or near their maximum power at the cycle frequencies chosen for study. For soleus muscle, the only difference in work or power between mini or normal muscles was at 4 Hz, where line 6 mini mice were marginally greater (Fig. 4). For the gastrocnemius, mini muscles of the two lines did not differ, but mini muscles in general produced only 16–40% the mass-specific work and power of normal muscles, the discrepancy being greater at the higher cycle frequency (Fig. 4).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
At generation 22, mice from the selected line 3 exhibited a 70% occurrence of the mini-muscle phenotype, whereas those of selected line 6 exhibited a 40% occurrence (12). We found that, at generations 36–38, all of the mice in line 3 expressed the mini-muscle phenotype, suggesting that the allele responsible has become fixed in this line, whereas in line 6 only 37% of the mice expressed the phenotype, similar to the level at generation 22. The line difference with respect to cross-generational changes in frequency of the mini-muscle allele could be attributable to random genetic drift and/or the interaction between the selection protocol and the "gene pool" of the two lines. In any case, within line 6, mini mice ran 37% more revolutions per day than did their counterparts with normal-sized muscles, which is consistent with the analysis indicating that the mini-muscle allele has been favored by the selective breeding (see Ref. 12). The significantly enhanced running performance in mini mice appears to be a recent phenomenon, or recently augmented, as such differences were not observed after generations 14, 22, and 23 in an analysis of all eight lines (4 selected and 4 control) (5, 12, 17).
Line 3 mini mice lost body mass during the standard 6-day wheel test, whereas those of line 6 (both normal and mini) showed a slight increase in body mass (Table 1). This is particularly interesting in light of observations that the mini mice of both lines ran significantly farther and faster but not longer than normal mice (Table 1). We do not have measures of cage activity during wheel tests (see discussion in Ref. 28), and this may also influence changes in body mass. Similar to the mini vs. normal running patterns observed in selected mice in the present study (Table 1), Koteja et al. (24) noted at generation 10 that selected mice spent about the same time running but ran faster than control mice, yet their daily rates of food consumption were only slightly higher (the normal/mini phenotypes were not known for their animals). They concluded that increased running speed as opposed to increasing the time spent running may confer an ability to increase daily movement distances without incurring a substantially greater metabolic cost. Although this mechanism could explain the lack of body mass loss in mini mice from line 6, it would be at odds with the observation that mini mice from line 3 lost mass even though they ran less than mini mice from line 6.
Reports based on mice from generations 14 and 22 agree with our results that the plantar flexor muscle group is 50% smaller in mini than normal mice (12, 17) (Table 2). Furthermore, here we report for the first time a comparison of the masses of the individual muscles in normal and mini mice and note that they do not all change in the same fashion (Fig. 1, Tables 2 and 3). The lateral gastrocnemius, with the largest change in mass in mini mice, would provide the most unequivocal metric for classification of mice as normal or mini. Clearly, the mini phenotype is not expressed as just a uniform reduction in the mass of the ankle extensors but is a specific and differential effect on the individual components. This suggests that the selection protocol may be resulting in unique and functionally relevant changes in each muscle, changes that are borne out to some extent on analysis of the physiology and contractile mechanics of individual muscles, as discussed next.
Sustained wheel running appears to be an aerobic activity in these mice, and there is a substantial effect of voluntary and forced wheel running on aerobic capacity in muscles of rodents (17, 27, 43); thus selection for this behavior might be accompanied by associated changes in the muscles that promote sustained, aerobic locomotion. Houle-Leroy et al. (17) note that the hindlimb muscles of mini mice have higher mass-specific aerobic capacities than their normal counterparts and that these almost fully compensate for any reduced capacity as a result of the reduced muscle mass. Furthermore, only aerobic capacities are compensated in the muscles studied, suggesting that aerobic but not anaerobic capacity is important to support the high levels of voluntary wheel running (note that the soleus, plantaris, and gastrocnemius were excluded from their analyses). It is thus reasonable to expect that changes in the physiology and mechanical characteristics of hindlimb muscles of mice from the selectively bred lines should be in a direction that promotes sustained, aerobic, and economical endurance running, and that mini muscles may exhibit such changes.
As a global observation, the contractile characteristics of the soleus muscles were not markedly different between normal and mini mice. Interestingly, there were some line effects, such that soleus muscles from mini mice in line 6 were slightly faster in their twitch characteristics (Fig. 2), perhaps explaining their ability to do slightly more work and produce more power at the higher operating frequency than the line 3 mini or line 6 normal muscles (Fig. 4). These increases in twitch speed would not translate directly into changes in any of the force-velocity characteristics of the muscle, and they do not appear to have done so (Fig. 3). Thus the contractile physiology of the soleus muscle, being already relatively slow and aerobic, does not appear to be altered substantially in the "renovation" of the plantar flexor group of mini mice. Whether it has remained generally unchanged in response to selection for increased wheel running would require comparisons with the contractile performance of soleus muscles from control lines of mice.
Despite the lack of notable differences in the physiology of the soleus muscles between normal and mini mice, they were significantly larger in mini mice (Tables 2 and 3, Fig. 1). This may reflect an increased contribution to powering sustained running with the increased demands of the higher running speeds and greater distances run by mini mice (Table 1) or with the loss of mass in their agonist, the gastrocnemius muscle. Hypertrophy and an increased contribution of slow muscles such as the soleus, being aerobic and more economical at generating force and power than faster muscles (e.g., Refs. 4, 9), would benefit animals that perform high levels of endurance running. Likewise, a decreased mass and contribution from faster muscle such as the gastrocnemius would confer similar advantages.
In marked contrast to the soleus, the contractile characteristics of the medial gastrocnemius muscles in mini mice became slower in nearly every attribute measured, and substantially so. Slowing of the twitch (Fig. 2) suggests, in part, a slowing of the rate of calcium sequestration in the muscle cells (e.g., Ref. 31). An increase in curvature of the force-velocity relationship and subsequent reductions in the isotonic power output (Fig. 3) suggest a change in the averaged cross-bridge rate constants and perhaps an increase in the thermodynamic efficiency of the muscle (44), as might be expected with a transition toward a slower phenotype and an increased fiber-type heterogeneity (22) in the mini muscles. The lack of an associated reduction of Vmax in mini gastrocnemius muscles was unexpected but would suggest that the muscles still possess a relatively high proportion of fast fiber types and may reflect in part the difficulties of estimating and interpreting Vmax from force-velocity data obtained from muscles with a heterogeneous fiber distribution (22).
Functionally, these changes resulted in the mass-specific mechanical work and power output of the gastrocnemius muscles of mini mice being reduced by one-half to one-sixth, whether measured as isotonic power (Fig. 3) or during cyclic activity (Fig. 4). This is consistent with observations that work and power are reduced in slow relative to fast muscles (e.g., Refs. 18, 39). More strikingly, the reduced mass-specific power in concert with the approximate halving of the mass of the gastrocnemius would result in the absolute power-producing potential of mini gastrocnemius muscles being reduced to only 10–25% of that of normal muscles, depending on the operating conditions. The small increase in mass of the soleus would plainly not be adequate to fully compensate for this loss in power. However, clearly the loss of potential power in the ankle extensors of mini mice did not impede their routine wheel-running abilities (Table 1). Muscle recruitment patterns at different locomotor speeds in rats and cats (reviewed briefly in Ref. 18) suggest that routine wheel running likely utilizes only a fraction of the full power-producing potential of the gastrocnemius. Yet, when high power output is required from these muscles, as when sprinting, a reduced performance in mini mice would be expected. Potentially they can "afford" to sacrifice power because traits such as maximal sprint speed are not a target of selection in this experimental protocol. Likely the primary advantage to having slower and smaller gastrocnemius muscles in mini mice is based in the economy of running or fatigue resistance.
Rates of fatigue of soleus muscles from mini mice were, like most aspects of their mechanical performance, not different from normal mice (Fig. 5). Rates of fatigue of medial gastrocnemius muscles from mini mice were reduced (Fig. 5) and reflective of a slower phenotype with increased aerobic capacity. As has been noted previously (33, 42), the rate of fatigue of work was greater than the rate of fatigue of isometric force (Fig. 5), the difference being attributed primarily to the marked reduction in the ability of fatigued muscle to relax. This suggests that the mini gastrocnemius muscles have acquired a substantial resistance to the fatigue of relaxation (and thus work), allowing them to power dynamic contractions involving muscle shortening and extension (e.g., running) for longer periods. Any reductions in the rate of fatigue of relaxation would also afford a substantial energetic savings to running mice by reducing the energy required to extend their muscles (42).
The higher average and maximal routine wheel-running speeds of mini mice (Table 1) indicate that they may be calling more on the gastrocnemius for power than normal mice, an act they may now afford because of the reduced fatigability of this muscle in mini mice, presumably a result of a slower phenotype and/or increased aerobic capacity. The rate of fatigue is dependent on a complex interaction between power output and cycling frequency and is not related in a simple way to the amount of work done by the muscle or its power output (2, 32). Thus the particular limb cycling frequency chosen by mice while running will impact both running speed and the rate of fatigue, and likely then the magnitude of any benefit incurred by expressing the mini-muscle phenotype. In turn, mini mice may be expected to display a different running style than either normal selected or control mice, much like the more intermittent running behavior exhibited by selected vs. control mice (13), which quite likely has a basis in avoidance of fatigue.
In conclusion, the differences in the masses and contractile physiology of the mini compared with normal muscles are of a nature that should support higher levels of endurance wheel running, as are the differences in aerobic capacity (17) and glycogen depots (14) in muscles of these mice. All of these differences are consistent with the observation that the selective breeding protocol has favored an increase in frequency, to fixation in one line, of the putative Mendelian recessive allele that causes the mini phenotype (12). Moreover, in our sample of mice, mini individuals ran significantly further and faster than their linemates with normal muscles, a difference that was not apparent in a larger sample from generation 22 that was not broken down for within-line analysis (12). Beyond differences in muscle performance per se, we do not yet know how the mini phenotype might impact the energetic costs of running, nor do we know whether mini mice may exhibit trade-offs associated with having smaller, less powerful muscles, such as a reduced maximal sprint speed. An analysis of the fiber-type composition of these animals would also provide valuable information to further understand why this particular muscle allele has been favored by selective breeding for high endurance running.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. M. Middleton, S. A. Kelly, and T. Garland Jr Selective breeding as a tool to probe skeletal response to high voluntary locomotor activity in mice Integr. Comp. Biol., June 18, 2008; (2008) icn057v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hartmann, T. Garland Jr, R. M. Hannon, S. A. Kelly, G. Munoz, and D. Pomp Fine Mapping of "Mini-Muscle," a Recessive Mutation Causing Reduced Hindlimb Muscle Mass in Mice J. Hered., June 9, 2008; (2008) esn040v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Hannon, S. A. Kelly, K. M. Middleton, E. M. Kolb, D. Pomp, and T. Garland Jr Phenotypic Effects of the "Mini-Muscle" Allele in a Large HR x C57BL/6J Mouse Backcross J. Hered., February 28, 2008; (2008) esn011v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Chappell, T. Garland Jr, G. F. Robertson, and W. Saltzman Relationships among running performance, aerobic physiology and organ mass in male Mongolian gerbils J. Exp. Biol., December 1, 2007; 210(23): 4179 - 4197. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. L. Rezende, F. R. Gomes, J. L. Malisch, M. A. Chappell, and T. Garland Jr. Maximal oxygen consumption in relation to subordinate traits in lines of house mice selectively bred for high voluntary wheel running J Appl Physiol, August 1, 2006; 101(2): 477 - 485. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Garland Jr and S. A. Kelly Phenotypic plasticity and experimental evolution J. Exp. Biol., June 15, 2006; 209(12): 2344 - 2361. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. L. Rezende, T. Garland Jr, M. A. Chappell, J. L. Malisch, and F. R. Gomes Maximum aerobic performance in lines of Mus selected for high wheel-running activity: effects of selection, oxygen availability and the mini-muscle phenotype J. Exp. Biol., January 1, 2006; 209(1): 115 - 127. [Abstract] [Full Text] [PDF]
|
|
Significantly different compared with line 6 mini. 
Significantly different from mini muscles of lines 3 and 6 combined (F1,15 = 8.83, P = 0.009 for soleus 10–90; F1,15 = 8.47, P = 0.01 for gastrocnemius 50–50).
