Skeletal muscle injury results from trauma (e.g., crush, laceration), but it more commonly results from the performance of contractions themselves, in particular, eccentric contractions. The incidence rate of contraction-induced muscle injury is hard to pinpoint, but it is safe to say that we all experience this injury frequently throughout our lives. The functional implications of this injury are significant and include a decreased joint range of motion, altered fatigability, and decreased muscle shortening velocity, but it is the prolonged strength loss that is the most widely recognized. Strength losses in excess of 40–50% are not uncommon for this type of injury, and complete recovery can take in excess of 1 mo (3,5).
Because the injury-induced strength loss has a long-lasting impact on performance in the workplace, on the athletic field, and in the home, it is important to understand the cause or causes of the strength loss so we can devise means of preventing or at least attenuating the strength loss. The objective of this review is to update the reader on the cellular mechanisms contributing to the strength loss after a bout of eccentric contractions. The explanation for the strength loss that we provide differs from that traditionally given in exercise physiology courses. In these courses, it is usually stated that the strength loss results from damage to fibers, presumably because of the high forces produced during eccentric contractions. The damage, although not explicitly stated by most instructors, is assumed to be a physical disruption of force-bearing structures within the muscle (e.g., myosin crossbridge, z-line, myotendinous junction).
We developed three broad categories that can cover all mechanisms potentially contributing to the strength loss. The categories include mechanisms that result in (1) a physical disruption or alteration of force-generating and/or -transmitting structures, (2) a failure to activate intact force-generating structures, and (3) a frank loss of force-generating and/or -transmitting structures. We discuss the three categories and their relative contribution to the strength loss. We delimit the discussion to that of maximal voluntary contraction strength or maximal electrically stimulated force in the case of animals.
PHYSICAL DISRUPTION OR ALTERATION OF FORCE-GENERATING AND/OR -TRANSMITTING STRUCTURES
This category of mechanisms provides the most intuitive explanation for the strength loss after contraction-induced injury. One expects a loss of strength after seeing muscle histopathology at the light and electron microscopic levels. The premise is that the classic histological lesions associated with this type of injury (e.g., z-line streaming, A-band disruptions) must result in a decreased force production as measured at the tendon ends. Unfortunately, this has never been demonstrated to our knowledge. No force loss has ever been attributed quantitatively or qualitatively to a histological lesion. We previously reviewed in detail the dissociation of histopathology from the strength loss (13) and will not consider it further.
In the single-fiber preparation, the excitation-contraction (E-C) coupling process can be bypassed by skinning or can be overcome pharmacologically, and a deficit in the maximal Ca2+-activated force can be attributed with certainty to disruption and/or alteration of the force-bearing elements. Reductions in single-fiber maximal Ca2+-activated force have been observed immediately after injury, and they are our best evidence that physical disruptions of force-bearing structures contribute to the strength loss. We have reported maximal Ca2+-activated force of fibers isolated from muscles injured in vitro to be reduced by 34% while strength of the muscles from which the fibers were isolated was down by 69% (11). Thus, physical disruption of force-bearing structures could at best account for half of the strength loss of the muscle.
Using an in vitro single-fiber model, Balnave and Allen (1) found maximal Ca2+-activated force was not significantly affected after 10 eccentric contractions in which the fibers were stretched to > 125% of optimal length (Lo) (Figure 1). However, the 28% reduction in maximal isometric force for these fibers must be considered modest. After a more severe injury protocol that reduced maximal isometric force by ˜ 70% (i.e., ≤ 30 eccentric contractions with stretches to > 150% of Lo), maximal Ca2+-activated force was found to be reduced by ˜ 55%. Thus, almost all (i.e., ˜ 80%) of the strength loss in this protocol could be accounted for by physical disruption or alteration of force-bearing structures within the fiber.
The limited data available on maximal Ca2+-activated force reductions in injured fibers would indicate that physical disruptions of force-bearing elements within the fibers can account for a widely varying fraction of the strength loss (i.e., from none to almost all of the strength loss). No maximal Ca2+-activated force data are available for fibers isolated from muscles injured in vivo. Such data are needed to ensure that fiber length is constrained within anatomical limits during the eccentric contractions. Our work with rodent ankle plantar flexor and dorsiflexor muscles indicates that in vivo fiber length does not exceed ˜ 130% Lo. Fibers stretched past this length may be predisposed to a strength loss resulting from physical disruptions of force-bearing elements. Also, we note that measures of fiber maximal Ca2+-activated force overestimate the contribution of disrupted force-bearing elements within fibers to the strength loss of the muscle because the single-fiber preparation removes the potential for lateral force transmission to neighboring fibers and thus eliminates a means for a bypassing lesions within a fiber (4). However, the measure may underestimate the contribution of disrupted force-bearing elements to the strength loss of the muscle because it cannot account for losses resulting from disrupted force-transmitting elements external to the fiber.
FAILURE TO ACTIVATE INTACT FORCE-GENERATING STRUCTURES
This category includes mechanisms that result in a failure of E-C coupling. For this review, we broadly define E-C coupling as the sequence of events that starts with the release of acetylcholine at the neuromuscular junction and ends with the release of Ca2+ from the sarcoplasmic reticulum (SR). Figure 2 illustrates the key anatomical structures and physiological events in the E-C coupling pathway.
A number of studies dating back to the late 1970s have suggested that E-C coupling failure occurs after eccentric contraction–biased exercise in humans (2). These studies reported the relative strength loss measured during low-frequency electrical stimulation to be greater than that measured at higher frequencies, suggesting that SR Ca2+ release was depressed at the lower stimulation frequencies.
Despite these observations, the contribution of E-C coupling failure to the strength loss after eccentric contractions was not investigated until the 1990s. The first study was that of Warren and coworkers (14). In this study, mouse soleus muscles were injured using in vitro eccentric contractions and then exposed to caffeine to probe for E-C coupling failure. Caffeine acts to increase free cytosolic Ca2+ concentration ([Ca2+]i) by promoting release of Ca2+ from the SR. Any observed reduction in the caffeine-elicited force would have been interpreted as being due to physical disruption of force-bearing elements and/or intrinsic dysfunction of the SR. However, the caffeine-elicited force of the eccentric contraction-injured muscles was not different from that of control muscle even though injured muscle maximal isometric tetanic force (Po) was down by 43%. These data were interpreted to indicate that (1) E-C coupling failure did occur in this in vitro model of muscle injury and (2) E-C coupling failure accounted for most of the strength loss with minimal evidence for contributions from disruption of force-bearing elements or intrinsic SR dysfunction.
These findings were confirmed and expanded on in a study published 2 yr later by Balnave and Allen (1). Using the single-fiber preparation and injury protocols described here, these researchers were the first to find a reduction in tetanic [Ca2+]i after an injurious bout of eccentric contractions. Following their moderate injury protocol, maximal isometric tetanic [Ca2+]i was found to be reduced by a modest 8%. However, this reduction in [Ca2+]i was thought to completely account for the 28% reduction in Po. Following their severe injury protocol, [Ca2+]i during a maximal isometric tetanus was found to be reduced by 40%, but this reduction could mathematically account for only a small percentage (i.e., ˜ 20%) of the strength loss.
The findings of the two studies just discussed are limited in that (1) the E-C coupling failure may represent some artifact of the in vitro preparation (e.g., overstretching of the fibers) and (2) the E-C coupling failure may be a transient event that accounts for strength loss only in the first few hours after injury. To overcome these limitations, we used our in vivo model for inducing injury to the anterior crural muscles of mice. In this injury model, extensor digitorum longus (EDL) muscle Po is down by ˜ 50% immediately after injury and does not change for 3–5 d; strength begins to recover thereafter but is not complete until ˜ 1 mo after the injury (5,7,9). Using this in vivo injury model, we found that maximal tetanic [Ca2+]i was reduced by 25–45% immediately and 3 d after injury. These findings indicated that the previous observations of E-C coupling failure were not artifacts of the in vitro injury model and the E-C coupling failure was not a transient phenomenon.
To estimate how long the E-C coupling failure contributes to the strength loss after in vivo injury, we measured the force produced by EDL muscles during exposure to caffeine or 4-chloro-m-cresol (i.e., pharmacological agents acting to promote SR Ca2+ release) at 0, 1, 3, 5, and 14 d after the injury (7). The forces elicited by these agents were down moderately (i.e., 11–21%) compared with the Po reduction (i.e., ˜ 51%) at 0–5 d after the injury; by 14 d after the injury, the caffeine-elicited force had returned to normal levels. We concluded that during the first 5 d after the injury, E-C coupling failure could account for 57–75% of the strength deficit, with the remainder of the strength deficit being accounted for by the physical disruption and/or alteration of the force-bearing structures.
The question of interest now moves onto where the failure occurs in the E-C coupling pathway. This is not an easy question because the E-C coupling pathway is not fully understood. Specifically, the means by which the depolarization-induced conformational change of the t-tubular voltage sensor (i.e., dihydropyridine receptor) causes the SR Ca2+ release channel (i.e., ryanodine receptor) to open has not been established. In next three subsections, we discuss the sites in the E-C coupling pathway that have been studied after injury.
If intrinsic SR function was depressed in the injured muscle, tetanic [Ca2+]i would also predictably be depressed. There are, however, little data to indicate that intrinsic SR dysfunction contributes to the E-C coupling failure, at least in the first 24 hr after injury. We measured Ca2+ uptake and release rates in vitro using a SR vesicle–containing fraction isolated from mouse EDL muscle homogenates (7). Immediately after in vivo injury, the rate of SR Ca2+ release was minimally reduced (i.e., 0–6%), as was the rate of SR Ca2+ uptake (i.e., 0–11%). Supporting these data are our observations that the SR Ca2+ release channel–activating agent, 4-chloro-m-cresol, is just as effective in elevating [Ca2+]i in fibers immediately after injury as it is in uninjured fibers (7). Furthermore, [3H]ryanodine equilibrium binding properties of the SR Ca2+ release channel are not altered immediately after injury (6).
SR Ca2+ release and uptake rates measured in vitro decrease gradually by 21–30% over the first 3–5 d after in vivo injury (7). Likewise, 4-chloro-m-cresol–elicited [Ca2+]i in fibers 3 d after injury is depressed by 20% (7), whereas the maximal binding capacity for [3H]ryanodine is reduced by 17% (6). These findings collectively indicate that intrinsic SR function decreases over the first 3–5 d after in vivo injury, but this progressive worsening of SR function is not associated with a further impairment in muscle strength.
The plasmalemma would seem to be a logical site for the E-C coupling failure given the ample evidence for plasmalemmal damage after eccentric contractions (13). This damage could cause a loss of normal ion distribution across the plasmalemma and thus adversely affect action potential conduction by the plasmalemma. Prolonged plasmalemmal damage would predictably result in a reduced fiber resting membrane potential. This has been probed in two studies (10,14). McBride and coworkers (10) found fibers in muscles injured in vivo to be depolarized by 10–15 mV for 1–2 d after injury. On the other hand, we found no evidence of altered resting membrane potential immediately after in vitro injury despite observing a greater strength loss than that reported by McBride and coworkers (14).
The explanation for the discrepant results is not readily apparent, but even if a resting membrane depolarization of 15 mV occurs after injury, the effect on action potential conduction is unknown. Because of this uncertainty, we assessed the capacity of injured muscle fibers to conduct action potentials along their plasmalemma (12). This was done using fine wire electromyographic (EMG) electrodes chronically implanted on mouse tibialis anterior (TA) muscles. EMG activity was recorded during electrical stimulation of the muscle using electrodes chronically implanted on the motor nerve. Under these conditions, a decrease in the EMG signal amplitude after injury would have been interpreted as an impaired capacity of the plasmalemma to conduct action potentials. Immediately after an in vivo eccentric contraction protocol that depressed maximal isometric strength by 47%, EMG amplitude was depressed by 9%. However, the reduction in EMG amplitude was not different from that of control muscles (i.e., muscles that performed an equivalent number of concentric contractions), and these control muscles showed no reduction in maximal isometric strength. Furthermore, EMG amplitude in the injured muscles was recovered by 24 hr after the injury and did not change during the remainder of the 2-wk study. Isometric strength, on the other hand, was still not recovered at 2 wk after the injury. In conclusion, the available data indicate that impaired plasmalemmal action potential conduction plays no major role in the E-C coupling failure induced by eccentric contractions.
t-Tubular Membrane and Voltage Sensors
The data presented so far indicate that the failure site in the E-C coupling pathway lies below the plasmalemma but above the SR Ca2+ release channel. However, we are aware of only preliminary efforts to probe the sites that lie between. Immediately after in vivo injury to mouse EDL muscle, we exposed the muscles in vitro to Krebs-Ringer buffer containing a high concentration of potassium (i.e., 200 mmol/L) (7). The high concentration of extracellular potassium acts to depolarize all membranes exposed to the extracellular space (i.e., plasmalemmal and t-tubular membranes) so that a site of action potential conduction blockage could be bypassed. In these experiments, we compared the effect of injury on high potassium–elicited force to its effect on Po. A lesser relative reduction in the potassium-elicited force compared with the reduction in Po would have implicated the t-tubular membrane as the failing site. However, injury reduced the high potassium–elicited force to the same extent (i.e., ˜ 50%) as it did Po. Although we consider these experiments to provide less than conclusive results, these data suggest that the E-C coupling failure does not involve the t-tubular membrane and must lie at or below the level of the voltage sensor.
We probed the t-tubular voltage sensor as the site of E-C coupling failure by determining its equilibrium binding properties for the dihydropyridine [3H]PN200-110 (6). Mouse TA muscles exhibiting a 45% reduction in isometric strength were assayed immediately and 3 d after in vivo injury. Maximal binding capacity for the radioligand was increased by ˜ 20% at both time points, whereas the radioligand dissociation constant was unaffected by injury. Thus, a loss of voltage sensors cannot account for the E-C coupling failure, but the rapid increase in maximal binding capacity may reflect a functional change in the voltage sensor. Figure 3 summarizes our knowledge to date on the site of failure in the E-C coupling pathway.
FRANK LOSS OF FORCE-GENERATING AND/OR -TRANSMITTING STRUCTURES
In theory, a rapid increase in the degradation of force-bearing protein structures could contribute to the immediate loss of strength after injury. Also, in the days after injury, increased protein degradation and/or decreased protein synthesis could contribute to an additional loss of strength. There is little evidence, however, to suggest that either of these two scenarios occurs. During the first 6 hr after in vivo injury, muscle protein degradation rate is not increased (9), and as one would predict, neither total nor contractile (i.e., actin and myosin heavy chain) protein content is altered (5,9) (Figure 4). Thus, changes in these protein contents can explain none of the rapid strength loss after injury.
These whole muscle measures of protein metabolism cannot, however, detect changes in the content of a minor protein that plays a key role in force generation and/or transmission. In fact, a rapid loss of the cytoskeletal protein, desmin, has been observed shortly after the onset of an eccentric contraction protocol (8). However, the percent reduction in strength was 1.5- to 2.6-fold greater than the percentage of desmin-negative fibers, so a causal linkage between the two measures is dubious.
In our in vivo injury model, the total protein degradation rate is increasing by 1 d after the injury and has leveled off by 2 d at a rate 60% above normal; the elevated rate is maintained until at least 5 d after the injury (9) (Figure 4). As a result, there is a progressive loss of contractile protein content beginning at ˜ 1 d after the injury, with content being reduced by 20% at 5 d after the injury (5). Surprisingly, there is no additional loss of strength accompanying the reduction in contractile protein content. Between 3 and 14 d after the injury, there is a seemingly paradoxical recovery of strength even though contractile protein content is still decreasing. The relationship between contractile protein content and strength must, however, be reestablished by 14 d after the injury because the strength recovery between 14 and 28 d after the injury is paralleled by a recovery in contractile protein content.
RELATIVE CONTRIBUTIONS TO THE STRENGTH LOSS
In this final section, we provide estimates of the relative contributions of the three categories to the strength loss after the injury (Figure 5). These estimates are based on data consolidated from three comprehensive studies of in vivo contraction-induced muscle injury in which 824 muscles were studied (5,7,9). During the first 3 d of injury, most (i.e., ≥ 75%) of the strength loss is attributed to a failure of E-C coupling. The remainder of the strength loss, at least for the first 1 or 2 d, is attributed to physical disruption and/or alteration of force-bearing elements within the muscle. By 3 d after the injury, the proportion of the strength deficit unaccounted for by E-C coupling failure is ascribed to a decreased contractile protein content, which most likely results from the removal of disrupted myofibrillar structures. The E-C coupling failure is diminishing by 5 d after the injury and is resolved by 14 d after the injury. During that time period, the proportion of the strength deficit accounted for the contractile protein loss increases from 40–45% to almost 100%.
It should be obvious now that there is no simple explanation for the strength loss that occurs after eccentric contractions. The strength loss appears to result from a complex interaction of mechanisms, the details of which remain to be determined. Finally, we acknowledge that our understanding of the strength loss comes almost entirely from studies of mouse muscle in which the contractions have been induced through maximal electrical stimulation. It is thus very much possible that the strength loss induced by this model may differ qualitatively from that elicited by submaximal, voluntary eccentric contractions.
1. Balnave, C.D., and Allen, D.G.. Intracellular calcium
and force in single mouse muscle fibres following repeated contractions with stretch. J. Physiol. (Lond.) 488: 25–36, 1995.
2. Edwards, R.H., Hill, D.K. Jones, D.A. and Merton, P.A.. Fatigue of long duration in human skeletal muscle after exercise. J. Physiol. (Lond.) 272: 769–778, 1977.
3. Howell, J.N., Chleboun, G. and Conatser, R.. Muscle stiffness, strength
loss, swelling and soreness following exercise-induced injury in humans. J. Physiol. (Lond.) 464: 183–196, 1993.
4. Huijing, P.A. Muscle as a collagen fiber reinforced composite: a review of force transmission in muscle and whole limb. J. Biomech. 32: 329–345, 1999.
5. Ingalls, C.P., Warren, G.L. and Armstrong, R.B.. Dissociation of force production from MHC and actin contents in muscles injured by eccentric contractions. J. Muscle Res. Cell Motil. 19: 215–224, 1998.
6. Ingalls, C.P., Warren, G.L. Hamilton, S.L. and Armstrong, R.B.. Dihydropyridine and ryanodine receptor binding after eccentric contractions in mouse skeletal muscle. Med. Sci. Sports Exerc. 31: S72, 1999.
7. Ingalls, C.P., Warren, G.L. Williams, J.H. Ward, C.W. and Armstrong, R.B.. E-C coupling failure in mouse EDL muscle after in vivo
eccentric contractions. J. Appl. Physiol. 85: 58–67, 1998.
8. Lieber, R.L., Schmitz, M.C. Mishra, D.K. and Fridén, J.. Contractile and cellular remodeling in rabbit skeletal muscle after cyclic eccentric contractions. J. Appl. Physiol. 77: 1926–1934, 1994.
9. Lowe, D.A., Warren, G.L. Ingalls, C.P. Boorstein, D.B. and Armstrong, R.B.. Muscle function and protein metabolism following initiation of eccentric contraction-induced injury. J. Appl. Physiol. 79: 1260–1270, 1995.
10. McBride, T.A., Stockert, B.W. Gorin, F.A. and Carlsen, R.C.. Stretch-activated ion channels contribute to membrane depolarization after eccentric contractions. J. Appl. Physiol. 88: 91–101, 2000.
11. Warren, G.L., Hayes, D.A. Lowe, D.A. Williams, J.H. and Armstrong, R.B.. Eccentric contraction-induced injury in normal and hindlimb-suspended mouse soleus and EDL muscles. J. Appl. Physiol. 77: 1421–1430, 1994.
12. Warren, G.L., Ingalls, C.P. Shah, S.J. and Armstrong, R.B.. Uncoupling of in vivo
torque production from EMG in mouse muscles injured by eccentric contractions. J. Physiol. (Lond.) 515: 609–619, 1999.
13. Warren, G.L., Lowe, D.A. and Armstrong, R.B.. Measurement tools used in the study of eccentric contraction-induced injury. Sports Med. 27: 43–59, 1999.
14. Warren, G.L., Lowe, D.A. Hayes, D.A. Karwoski, C.J. Prior, B.M. and Armstrong, R.B.. Excitation failure in mouse soleus muscle injured by eccentric contractions. J. Physiol. (Lond.) 468: 487–499, 1993.