SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  • 1
    The effect on sarcomere organization of stretching intact single skeletal muscle fibres by 50% of their optimum length (Lo) during ten consecutive short tetani was investigated. Stretch reduced tetanic force to 36 ± 4% of the pre-stretch condition. Sarcomere organization was analysed using both electron and confocal microscopy. For confocal microscopy the striation pattern was examined by fluorescently staining F-actin with rhodamine–phalloidin.
  • 2
    Electron microscopy revealed that fibres which had been stretched during contraction contained areas of severe sarcomere disorganization, as well as adjacent sarcomeres of normal appearance.
  • 3
    Confocal images of stretched fibres, which had been fixed and stained with rhodamine–phalloidin, showed focal regions of overstretched sarcomeres and regions where sarcomeres of adjacent myofibrils were out of alignment with each other. Analysis of all sarcomeres along the length of fibres showed regions of sarcomere inhomogeneity were distributed throughout the fibre length and cross-section.
  • 4
    Fibres were microinjected with the fluorescent [Ca2+]i indicator fura-2 before being stretched. Conventional wide-field fluorescence imaging microscopy showed that the tetanic [Ca2+]i was reduced after stretching but remained uniformly distributed.
  • 5
    This study confirms the finding that stretch-induced muscle injury has components caused by disorganization of the myofibrillar array and by failure of tetanic Ca2+ release. The structural damage is spatially heterogeneous whereas the changes in Ca2+ release appear to be spatially homogeneous.

Human and animal studies have shown that stretching skeletal muscles during contraction (eccentric contraction) leads to a long-lasting muscle weakness (Davies & White, 1981; McCully & Faulkner, 1985). Similarly, stretching active single muscle fibres brings about a pronounced decrease in tetanic force production which persists for at least 1 h with no recovery (Balnave & Allen, 1995). Part of this force deficit was shown to be the result of a reduced intracellular free calcium concentration ([Ca2+]i, probably due to reduced Ca2+ release from the sarcoplasmic reticulum (SR). However, the maximum Ca2+-activated force was also reduced following stretch suggesting structural abnormalities (Balnave & Allen, 1995).

Morphological studies have revealed that skeletal muscles which have undergone eccentric contractions in situ exhibit myofibrillar disorganization (Armstrong, Ogilvie & Schwane, 1983; Friden, Sjostrom & Ekblom, 1983; Wood, Morgan & Proske, 1993). Commonly reported abnormalities include sarcomeres which appear to be totally disrupted, Z-lines which have a zigzag appearance, and sarcomeres or half-sarcomeres which are overstretched so that there is no overlap between myofilaments (Friden et al. 1983; Wood et al. 1993). The regions of myofibrillar disorganization are often focal, with regions of normal appearance close by, and are present immediately post-stretch (Newham, McPhail, Mills & Edwards, 1983; Wood et al. 1993). A single stretch during contraction is sufficient to generate this pattern of disorganization (Brown & Hill, 1991; Brooks, Zerba & Faulkner, 1995; Talbot & Morgan, 1996). An equivalent study on the morphology of single fibres has not been performed, so it is unclear whether the stretch-induced reduction in maximum Ca2+-activated force is due to myofibrillar injury or to some other mechanism.

In the study of Balnave & Allen (1995), which showed that the release of Ca2+ from the SR was reduced following stretch, [Ca2+]i was calculated from the spatially averaged fluorescent Ca2+ signal obtained from approximately one-third of the muscle fibre. Therefore, these experiments could not distinguish between a uniform reduction in Ca2+ release and a reduction at irregular intervals along the fibre. For instance, damage to T-tubules might prevent inward conduction of the action potential causing reduced activation in the centre of the fibre (Westerblad, Lee, Lamb, Bolsover & Allen, 1990; Duty & Allen, 1994). Alternatively, there might be a small number of damaged regions in the fibre where Ca2+ release was grossly reduced.

The aim of the present investigation was to determine the nature and distribution of any sarcomere disorganization caused by stretching intact single mammalian skeletal muscle fibres during contraction. In addition, we have studied the distribution of [Ca2+]i, both at rest and during tetanic stimulation, to determine whether the abnormalities of Ca2+ handling were uniform or showed some specific kind of distribution. The overall aim is to explain the reduction in measured force in terms of both Ca2+ handling and sarcomere organization.

METHODS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements

Adult, male mice were killed by rapid cervical dislocation. A single muscle fibre was dissected from the flexor brevis muscle and mounted between a force transducer and the arm of a motor designed to impose known length changes on the fibre. Details of these procedures have been described previously (Balnave & Allen, 1995). Fibres were stimulated with a series of ten 100 Hz tetani, 350 ms in duration with a 4 s interval between each tetanus. In this preparation a 100 Hz tetanus produces about 90% of the maximum force obtained by raising the tetanic [Ca2+]i above maximal levels with caffeine (Balnave & Allen, 1995, 1996). The optimum force-generating length (Lo, ∼800 μm) was determined by increasing the length of the muscle fibre from being slack until tetanic force was maximal. The resting length of all fibres (stretched and control) was set at 100 μm longer than Lo so as to place the fibres on the descending limb of the force–length curve. Fibres were stretched by either 25 or 50%Lo at 5 muscle lengths per second, starting 200 ms after the start of each tetanus. Muscle length was returned to its resting level after completion of the tetanic stimulation. For representative force records see Balnave & Allen (1995). Recovery of force was measured after 30 min. In experiments requiring electron or confocal microscopy, fibres were transferred from the experimental chamber to a second chamber designed for the fixation procedure. Fibre length was reset at approximately the same length as in the experimental chamber.

Electron microscopy

One unstimulated fibre and one fibre which had been stretched by 50%Lo during ten contractions were fixed and their fine structure examined using electron microscopy. The fixative used for electron microscopy was bathing solution containing 2% glutaraldehyde and 4% acrolein (v/v). The fibre was fixed in place in the experimental bath. The mixture was exchanged for more fixative as rapidly as possible, but without draining the solution below the level of the fibre. After 1 h, the fibre was cut from the clamps holding it in the experimental apparatus, transferred to a glass vial, and rinsed in several changes of phosphate buffer solution (28 mm NaH2PO4–72 mm Na2HPO4, pH 7.2). It was fixed overnight in 1% OsO4 in the same buffer, and then rinsed with several changes of buffer solution over a 1 h period. It was then dehydrated through a graded series of ethanol solutions before embedding in Spurr's resin in an embedding capsule. The fibre was sectioned at approximately 50 nm thickness, and stained with uranium and lead. Electron micrographs were obtained with a Philips 201c instrument at a magnification of × 6000.

Confocal microscopy

The sarcomere distributions of five fibres which had been stretched by 50%Lo, and two fibres stretched by 25%Lo, during ten contractions were examined using confocal microscopy. The fixative used to prepare the muscle fibres for confocal microscopy was 4% paraformaldehyde in phosphate buffer solution (28 mm NaH2PO4 and 72 mm Na2HPO4). Once fixed a muscle fibre was placed in an Eppendorf tube containing four units of the fluorescent F-actin stain rhodamine–phalloidin, which had been reconstituted in 200 μl of a solution containing 0.1 M phosphate buffer with 0.5% Triton X-100. The fibre was left in the stain for 2 days, before being placed on a glass coverslip in the 0.1 M phosphate buffer solution to be imaged using confocal microscopy.

An inverted Leica 4D laser scanning confocal microscope, with an Ar–Kr laser, was used to construct two-dimensional images of the distribution of F-actin throughout the fibres. The sample was excited by light of wavelength 568 nm and the emitted signal filtered by a 590 nm long-pass filter. A × 40 oil immersion objective lens with a numerical aperture of 1.0 was used to scan 50 μm × 50 μm sections of each fibre at progressively increasing depths of 3 μm and representative images were then stored. Each 50 μm × 50 μm section shared its border with the adjoining section so that the entire length of each fibre was examined.

The sarcomere distributions of the five fibres stretched by 50%Lo during ten contractions were compared with those of seven control fibres. Four control fibres were not stimulated, although one of these fibres was passively stretched by 50%Lo. The remaining three control fibres performed ten isometric contractions.

Calcium imaging

The Cai2+ was imaged along the length of six muscle fibres which had been stretched by 50%Lo during ten contractions. The methods and equipment used for imaging Cai2+ in single muscle fibres have been described previously (Westerblad et al. 1990; Duty & Allen, 1994). Briefly, fibres were microinjected with the fluorescent Ca2+ indicator fura-2. After allowing 45 min for the dye concentration to equilibrate along the cell, the fibre was illuminated with ultraviolet light of wavelength 340 or 380 nm using an automated Nikon filter switcher. An image of the emitted fluorescent light of wavelengths longer than 430 nm was then obtained. The ratio of the image produced by 340 nm illumination and the image produced by 380 nm illumination could then be converted to [Ca2+]i using the calibration procedure described by Westerblad & Allen (1991).

To obtain a ratio image of a fibre during contraction, images were taken during two consecutive tetani 14 s apart. The fibre was illuminated at 340 nm during the first tetanus and at 380 nm during the second tetanus. Each image was obtained by averaging over 80 ms, beginning 200 ms after the start of each tetanus. Ratio images produced in this way were taken at rest and during 100 Hz tetani before and 10, 30 and 60 min after the fibres were stretched. Although only about one-third of each fibre could be examined in each image, the pattern of the change in [Ca2+]i was found to be similar in both the middle and at the ends of the fibre.

Statistics

Unless otherwise stated data are quoted as means ±s.e.m. Student's paired t test was used to verify statistical significance with P < 0.05 taken as significant.

RESULTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements

Muscle fibres stretched by 50%Lo during ten contractions showed significant reductions in tetanic force. In the twelve fibres stretched by 50%Lo, force generated by 100 Hz stimulation (here termed 100 Hz force) was reduced to 36 ± 4% of the pre-stretch force after 30 min of recovery. In contrast in three fibres stimulated with ten isometric contractions and one fibre stretched by 50%Lo in the absence of contraction the tetanic force was 99.8 ± 2.3% of the pre-stretch force after 30 min of recovery. These results are similar to our earlier results using the same protocol (Balnave & Allen, 1995).

Electron microscopy

Electron micrographs were taken of an unstimulated control fibre and a fibre which had been stretched by 50%Lo during ten contractions (Fig. 1). The control fibre in Fig. 1A contains sarcomeres of normal appearance organized in a regular array and aligned with the sarcomeres of neighbouring myofibrils. There is no evidence of sarcomere disorganization. In contrast, the stretched fibre in Fig. 1B exhibits many myofibrillar abnormalities. Most notable are Z-lines which have a wavy or zigzag appearance, originally termed Z-line streaming (Friden et al. 1983). In some areas the Z-lines are totally disrupted. Consequently, many sarcomeres are out of alignment with their neighbours and appear either overstretched or reduced in length. In some regions the reduced overlap between myofilaments is limited to the half-sarcomere. Adjacent to these disorganized areas are regions of normal appearance. This pattern of injury has previously been described in human and whole muscle experiments during and immediately after the performance of eccentric muscle contractions (Newham et al. 1983; Brown & Hill, 1991; Wood et al. 1993; Brooks et al. 1995; Talbot & Morgan, 1996).

image

Figure 1. Electron micrograph of a longitudinal section through a control fibre and a fibre stretched by 50%Lo during contraction

The unstimulated control fibre (A) exhibits a normal striation pattern, while the fibre which has been stretched during contraction (B) contains sarcomeres which are disorganized in addition to sarcomeres of normal appearance. The force generated by 100 Hz stimulation (100 Hz force) following stretch was reduced to 41% of the pre-stretch value. Scale bars represent 2 μm.

Download figure to PowerPoint

Confocal microscopy

Electron micrographs provide high resolution images but it is difficult to scan spatially the fibre length with this technique. In contrast, with confocal microscopy it is possible to examine systematically sarcomere length distribution throughout a fibre. Figure 2A shows an image taken from an unstimulated control fibre. Each bright band represents the rhodamine–phalloidin-stained F-actin, while each dark band represents the H-zone of the sarcomere, i.e. the region of the A-band where there is no myofilament overlap. Note that the fluorescence intensity varies along the bright band. The non-uniform binding of rhodamine–phalloidin to actin filaments and the Z-line has been described in skeletal muscle myofibrils by other investigators (Bukatina, Sonkin, Alievskaya & Yashin, 1984; Szczesna & Lehrer, 1993; Ao & Lehrer, 1995).

image

Figure 2. Confocal images showing the fluorescence distribution of rhodamine–phalloidin in a control fibre and fibres which had been stretched during contraction

A, confocal image of an unstimulated control fibre. Note the pattern of regularly spaced bright bands indicating F-actin fluorescently stained with rhodamine-phalloidin. B, fibre which had been stretched by 25%Lo during ten contractions. The 100 Hz force was reduced to 94% of the pre-stretch value following 30 min recovery. Again note that sarcomere spacing is regular and uniform. C–E, fibres stretched by 50%Lo during ten contractions. The 100 Hz force produced by these fibres was reduced to 50 (C), 30 (D) and 60% (E) of the pre-stretch value 30 min post-stretch. C shows an area with extremely overstretched sarcomeres (*) and a smaller area with more focal sarcomere inhomogeneity (†). D shows an example of the numerous focal regions of sarcomere inhomogeneity located randomly throughout the fibre. E shows the zigzag appearance of the striation pattern. Scale bars represent 5 μm.

Download figure to PowerPoint

In addition to three unstimulated control fibres, three control fibres were stimulated to produce ten contractions and another fibre was stretched by 50%Lo ten times while at rest. As noted above, these procedures did not affect the developed force. Each fibre was carefully scanned along its length and at 3 μm depths. All displayed a similar uniform appearance to the example in Fig. 2A; sarcomere length was consistent, the dark and bright bands ran parallel to each other, and the distinction between dark and bright bands was clear. In some images we observed darker lines running longitudinally and parallel to the axis of the fibre (e.g. Fig. 2B). Adjacent lines were spaced approximately 1 μm apart and so may indicate the border between neighbouring myofibrils.

Two fibres were stained after being stretched by 25%Lo during ten contractions. After 30 min rest tetanic force had recovered to 100 and 94% of the pre-stretch force of each fibre. Figure 2B shows a typical optical section of one of these fibres. No sarcomere inhomogeneities were observed in any section from either fibre.

The confocal microscope was used to examine five fibres which had been stretched by 50%Lo during ten contractions and stained with rhodamine–phalloidin. All five fibres stretched by 50%Lo during contraction exhibited sarcomere length inhomogeneities which were distributed throughout each fibre. Confocal images of irregularities in the sarcomere pattern, which may contribute to the force deficit, are shown in optical sections from three different fibres in Fig. 2C, D and E. Figure 2C shows an optical section of a region in which the sarcomere spacing is clearly not uniform. The most obvious abnormal region where four sarcomeres appear to be overextended is labelled with an asterisk. Additionally, a smaller area of sarcomere irregularity, which is more common, can be observed at the region labelled with a dagger. These damaged regions are focal and do not extend throughout the depth of the fibre. In fact, with the focal plane 6 μm deeper into the fibre the sarcomere pattern in this region was essentially normal. Therefore, the sarcomere abnormalities observed in Fig. 2C are spatially localized in the z as well as the x–y plane.

The overextended sarcomeres shown in Fig. 2C span the complete diameter of the fibre. However, more commonly, areas of sarcomere inhomogeneity are smaller and appear randomly distributed within a confocal image (Fig. 2D). Another feature of the fibres which had been stretched by 50%Lo during contraction was that in some regions sarcomeres appeared out of alignment with their neighbours. This occurred, in particular, at the longitudinally orientated lines which may represent the border between adjacent myofibrils (Fig. 2E). Therefore, in addition to sarcomere length inhomogeneities, this gave the striation pattern a zigzag appearance.

Histogram of sarcomere length

In all five fibres stretched by 50%Lo during contraction sarcomere disturbances were distributed randomly throughout the fibre. A detailed analysis of the sarcomere spacing from one of the five fibres that had been stretched by 50%Lo during contraction and one unstimulated control fibre is shown in Fig. 3. Sarcomere length was calculated as the distance between the centres of consecutive bright bands on a confocal image. In the majority of instances the centre of the bright band, which denotes a Z-line, was marked by a distinct peak in fluorescence intensity.

image

Figure 3. Variation in sarcomere length along an unstimulated control fibre and a fibre which had been stretched by 50%Lo during ten contractions

Schematic diagrams of cross-sections through the control fibre (A) and the stretched fibre (B) (not drawn to scale). Fibres have a diameter of approximately 40 μm. The 100 Hz force of the stretched fibre was reduced to 30% of the pre-stretch value after 30 min recovery. Confocal images of the entire length of each fibre were taken at five equally spaced depths (approximately 10 μm), beginning and ending at the upper and lower surfaces of the fibre, respectively. At each depth, sarcomere spacing was measured at five equally spaced intervals (approximately 10 μm) across the fibre, again beginning and ending at the outer edges of the fibre. Each circle and bar represents the mean ±s.d. of the length of all the sarcomeres measured along the fibre in the zone indicated by the position of the circle relative to the mean fibre sarcomere length (3.26 μm for A, 3.1 μm for B) represented by the dashed line at each level. The standard deviations are much larger in the stretched fibre. C and D show individual records of the variation in sarcomere length along the control fibre (C) and the stretched fibre (D) at the positions marked by the open circles in A and B, respectively. Note the much larger variation in sarcomere length along the stretched fibre.

Download figure to PowerPoint

Figure 3A shows a schematic diagram of a cross-section through the control fibre. Fibres have a diameter of approximately 40 μm. The length of every sarcomere along the fibre was measured at a depth and breadth indicated by the position of the circles. Each circle represents the mean sarcomere length of all the sarcomeres along the fibre in that zone. Thus, Fig. 3A illustrates the extent to which the mean sarcomere length of each zone fluctuated from the mean sarcomere length of the whole fibre (dashed lines) and the degree to which sarcomere length varied in each zone (bars indicate ±1 standard deviation (s.d.)) for the control fibre. The equivalent measurements in the stretched fibre are shown in Fig. 3B. Note that, although the mean sarcomere length of each zone did not deviate greatly from the mean sarcomere length of the whole fibre in either cell, individual sarcomere lengths were significantly more variable (P < 0.001; Levene median test for equal variance) following stretch than in the control fibre. This variability in sarcomere length following stretch was observed in each zone analysed. Individual records of this sarcomere length distribution (taken from the zones indicated by open circles in Fig. 3A and B) are shown respectively for the control and stretched fibres in Fig. 3C and D. The greater variability in sarcomere length following stretch compared with the control fibre is apparent. This variability can be observed along the entire length of the fibre (Fig. 3D). Occasionally there are spikes corresponding to highly overstretched or supercontracted sarcomeres. Note that the overstretched sarcomeres are not necessarily found immediately next to the very short sarcomeres.

A histogram incorporating the length of every sarcomere measured from the confocal images is shown in Fig. 4. Figure 4A shows the histogram of sarcomere lengths in the control fibre. A total of 5592 sarcomeres were measured. The mean sarcomere length was 3.26 μm, with the majority of sarcomeres (>60%) between 3.2 and 3.3 μm. This equated to a standard deviation of 0.14 μm. Since the fibre was fixed at a length of 100 μm longer than Lo, the optimum sarcomere length is estimated as 2.86 μm. This value compares with the values reported by other investigators who measured optimum sarcomere lengths in mammalian skeletal muscle fibres of ∼2.8 μm (Rack & Westbury, 1969; Stephenson & Williams, 1982; Balnave & Allen, 1996). Similar results were obtained from a fibre which had been stimulated to produce ten isometric contractions (537 sarcomeres measured; s.d., 0.09 μm) and in the two fibres stretched by 25%Lo during contraction (497 and 557 sarcomeres measured; s.d., 0.07 and 0.09 μm, respectively).

image

Figure 4. Histograms of sarcomere lengths measured in an unstimulated control fibre and a fibre which had been stretched by 50%Lo during ten contractions

Histograms of sarcomere length for the 5592 sarcomeres measured in the control fibre (A), and for the 4976 sarcomeres measured in the stretched fibre (B). The length of approximately 90% of the total number of sarcomeres in the zones analysed were measured. The length of every sarcomere could not be measured since distinguishing borders between adjacent sarcomeres was, in some cases, very difficult. Sarcomere number is expressed as a percentage of the total number of sarcomeres measured. Note that the spread of sarcomeres away from the mean is much larger in the stretched fibre.

Download figure to PowerPoint

To construct the histogram of sarcomere lengths in the fibre stretched by 50%Lo during contraction (Fig. 4B) 4976 sarcomeres were measured. The mean sarcomere length in this fibre was 3.10 μm, which corresponds to an optimum sarcomere length of 2.67 μm. However, in contrast to the control fibre, 60% of sarcomeres had lengths spread between 2.8 and 3.3 μm, which equated to a standard deviation of 0.40 μm. Therefore, the sarcomere lengths in the stretched fibre were far more variable compared with the control fibre, but the distribution of variability shows no obvious pattern.

Imaging Ca2+ release

We have previously shown that the tetanic [Ca2+]i is reduced following stretch-induced injury (Balnave & Allen, 1995). However, these studies give no indication of the distribution of this reduction in [Ca2+]i. For instance, T-tubular damage might lead to radial gradients of [Ca2+]i (Westerblad et al. 1990; Duty & Allen, 1994). Therefore, using the fluorescent Ca2+ indicator fura-2, we imaged [Ca2+]i in fibres stretched by 50%Lo during contraction.

Figure 5 shows images of the middle third of a typical fibre taken at rest and during a 100 Hz tetanus before and 10, 30 and 60 min post-stretch. At rest (blue) [Ca2+]i was slightly higher after stretch, as indicated by the lighter shade of blue in the images. However, with the resolution of this imaging system, there was no evidence of an uneven distribution of [Ca2+]i within the resting fibre, nor was the standard deviation of the [Ca2+]i in all pixels changed. This observation was consistent in the six fibres analysed. Therefore, it seems unlikely that stretching a contracting muscle fibre causes the surface membrane to tear since we observed no localized regions with a high resting [Ca2+]i where a damaged section of surface membrane should allow Ca2+ to enter the fibre along its large concentration gradient.

Similarly, in the fibre displayed in Fig. 5, the distribution of [Ca2+]i during a 100 Hz tetanus was uniform in the hour after stretch. The paler yellow colour post-stretch indicates that tetanic [Ca2+]i is reduced. In the six fibres analysed tetanic [Ca2+]i was reduced from 664 ± 68 to 501 ± 30 nM (P < 0.05) after 1 h recovery. However, there were no detectable longitudinal or radial gradients of [Ca2+]i and the standard deviation of the [Ca2+]i in all pixels was smaller following stretch.

image

Figure 5. Pseudocolour ratio images from the middle third of one fibre stretched by 50%Lo during contraction showing the distribution of [Ca2+]i along the fibre at rest and during a 100 Hz tetanus

Images were taken prior to stretch and 10, 30 and 60 min post-stretch. In this fibre 100 Hz force was reduced to 31% following 60 min recovery. The colours on the calibration bar indicate the fluorescence ratio of fura-2. The relationship between fura-2 ratio and [Ca2+]i is approximately: 0.7 ≡ 70 nM, 1.0 ≡ 465 nM, 1.2 ≡ 1200 nM.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements

Stretching intact single mammalian skeletal muscle fibres during contraction has been shown to bring about a reduction of tetanic force which lasts for at least 1 h. In a previous investigation, the results of which have been confirmed in the present study, we showed that this stretching protocol resulted in a reduced titanic [Ca2+]i (Balnave & Allen, 1995). This provided more direct evidence for an earlier suggestion that stretch during contraction can cause reduced Ca2+ release from the SR (Warren, Lowe, Hayes, Karwoski, Prior & Armstrong, 1993). In our earlier study we showed that stretching muscle fibres by 25%Lo produced a force deficit which could be completely accounted for by the reduced SR Ca2+ release. However, when the severity of the stretching protocol was increased, by stretching the muscle fibres by 50%Lo, we observed an additional reduction in the maximum Ca2+-activated force which we attributed to sarcomere disorganization, although no structural evidence for this was presented (Balnave & Allen, 1995).

Electron microscopy

The electron micrographs of the stretched fibre revealed that abnormalities in the sarcomere pattern are quantitatively similar to those described in human, animal and whole muscle experiments by other investigators (Armstrong et al. 1983; Friden et al. 1983; Newham et al. 1983; Wood et al. 1993). Therefore, the single fibre model of stretch-induced muscle injury is analogous to the whole animal condition structurally as well as functionally (Balnave & Allen, 1995). Because it is very difficult to sample systematically along a fibre using electron microscopy, we used confocal microscopy to obtain a description of the sarcomere length disruption throughout a single fibre.

Confocal microscopy

In a previous investigation neither ten isometric contractions nor ten stretches of 50%Lo in resting fibres produced a force deficit (Balnave & Allen, 1995). Stretching muscle fibres by 25%Lo during ten contractions was shown to reduce tetanic Ca2+ release but did not affect the maximum Ca2+-activated force. In the present study no notable sarcomere length inhomogeneity was observed in any of the following conditions; (i) unstimulated, unstretched fibres, (ii) unstimulated fibres stretched ten times by 50%Lo, (iii) fibres stimulated with ten isometric contractions, and (iv) fibres stretched by 25%Lo during ten contractions. However, in all five fibres which had performed stretches of 50%Lo during contraction, multiple areas of sarcomere length inhomogeneity of varying degrees were observed using confocal microscopy. Therefore, it appears likely that the reduction in the maximum Ca2+-activated force, observed after stretching a contracting muscle fibre by 50%Lo (Balnave & Allen, 1995), is the result of stretch-induced sarcomere disorganization.

We have shown that stretching a muscle fibre by 50%Lo during ten contractions causes force to fall to 36 ± 4% and produces severe sarcomere disruption. However, a 25% stretch produced no force deficit or sarcomere inhomogeneity. The 50% stretch is very large and it can be questioned whether this result is relevant to events which occur in intact muscles. Although a 50% stretch is large it is still within the range which can occur in muscles (Brooks et al. 1995) and the reduction in force which we observe is similar to that reported by others in the literature. For instance, Brooks et al. (1995) found that a single stretch of less than 30% produced no reduction in force, while a single stretch of 60% reduced force to 35%. These results from intact in situ muscles are not greatly different from ours in isolated single fibres.

Histogram of sarcomere length

The sarcomere length inhomogeneity can be distinguished clearly by examining the histograms of sarcomere length from the control and stretched fibres. Sarcomere length in the control fibre ranged from 2.3 μm (mainly at the ends of the fibre where sarcomere length was shorter than the mean value; Fig. 3C) to 3.7 μm. In contrast, sarcomere length in the stretched fibre ranged from 1.7 to 5.9 μm. This variability is reflected in the standard deviations of sarcomere length of 0.14 and 0.40 μm for the control and stretched fibres, respectively.

Sarcomere inhomogeneities following contractions with stretch have been recognized for many years (e.g. Newham et al. 1983). Morgan (1990) developed and quantified these ideas and proposed the ‘popping sarcomere’ hypothesis to explain many features of contractions with stretch. Morgan suggested that when a stretch is imposed on a contracting muscle the lengthening of individual sarcomeres is not uniform. Due to biological variation some sarcomeres will be weaker than others. The weakest sarcomere tends to stretch the most and once the sarcomere reaches the point on its force–velocity curve when velocity of stretch increases independently of force it elongates extremely rapidly and uncontrollably (popping). If muscle fibre lengthening continues after the weakest sarcomere has popped then the next weakest sarcomere will elongate, and so on until the stretch is complete. Upon relaxation it was proposed that some of the extended sarcomeres do not return to the interdigitating pattern but remain overextended.

In his popping sarcomere hypothesis Morgan (1990) suggested that the weakest sarcomeres are randomly distributed throughout a muscle fibre. Our results support this idea of a random distribution of overstretched sarcomeres. However, we also observed many sarcomeres of very short length. Morgan's theory predicts that muscle fixed during a single contraction with stretch should produce a small peak in the sarcomere length histogram at a long sarcomere length, representing occasional regions of over-stretched sarcomeres, and a large peak at a sarcomere length slightly shorter than the mean length, representing an evenly distributed shortening of the remaining sarcomeres. This prediction has subsequently been confirmed (Brown & Hill, 1991; Talbot & Morgan, 1996). Our results show that, after recovery from ten contractions with stretch, in addition to overstretched sarcomeres the length of some sarcomeres is dramatically reduced while at least 25% remain within 0.5 μm of the mean sarcomere length. Although the distribution of sarcomeres is not what would be predicted from Morgan's hypothesis this may be because in our experiments sarcomere length was measured after the fibre had relaxed and returned to the control length before fixing. It is possible that passive restoring forces cause some over-stretched sarcomeres to resume their interdigitation during and after relaxation and there may also be other processes leading to redistribution of sarcomere lengths.

Imaging Ca2+ release

Our results confirm earlier studies that have shown that 100 Hz tetanic [Ca2+]i is reduced following stretch-induced injury (Warren et al. 1993; Balnave & Allen, 1995), but show that the distribution of the reduced [Ca2+]i is uniform, at least at the resolution of the present imaging system. This result allows us to exclude the possibility that T-tubular damage leading to a uniform failure of inward spread of activation occurs, such as that detected in some types of muscle fatigue (Westerblad et al. 1990; Duty & Allen, 1994). It is also clear that stretch-induced injury does not lead to a small number of restricted areas of reduced Ca2+ release, as this would be very obvious in the images. Another possibility is that the multiple sites of sarcomere disorganization seen in the electron micrograph and confocal images are each associated with similar regions of reduced Ca2+ release. To try to detect this kind of spatial variability of Ca2+ release we compared the standard deviation of Ca2+ across all pixels. The standard deviation was lower following stretch, suggesting that spatial variability is not increased. However, the resolution of conventional imaging is reduced in thick specimens because of the contribution of out-of-focus light (Sandison & Webb, 1994) and it remains possible that a higher resolution method will detect localized regions of reduced Ca2+ release which have the same distribution as the regions of sarcomere damage.

Conclusion

In conclusion, stretching intact single mammalian skeletal muscle fibres during contraction leads to structural disorganization of the contractile apparatus similar to that observed in whole animal and whole muscle investigations. Confocal microscopy can be used to analyse sarcomere inhomogeneity in these stretched fibres and shows that sarcomere length is extremely variable throughout such fibres. Conventional wide-field fluorescence imaging microscopy has been used to show that tetanic and resting [Ca2+]i are uniformly distributed along these single fibres post-stretch. This finding suggests that reduced Ca2+ release occurs regularly throughout stretched muscle fibres.

Acknowledgements

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements

This work was supported by the National Health and Medical Research Council of Australia. The authors would also like to thank Dr Stewart Head and Ms Ann Parkinson for their advice on the rhodamine–phalloidin staining technique.