Mechanisms of force depression caused by different types of physical exercise studied by direct electrical stimulation of human quadriceps muscle

The study involved nine groups of male volunteers and their characteristics are displayed in Table 1. All subjects were physically active and participated in recreational activities 2–3 times per week. They were asked to refrain from any exercise for 1 week prior to the experiment. Each group performed only one of the exercise protocols described below to avoid problems with results being affected by previous exercise; for instance, a single bout of eccentric contractions induces a protection against deleterious effects of subsequent eccentric contractions (repeated bout effect) and this protection lasts for several weeks (e.g. Nosaka et al. 2005). The study was approved by the Kaunas Regional Ethics Committee and is consistent with the principles outlined in the Declaration of Helsinki. Each subject read and signed a written, informed consent form prior to participation.

Electrically evoked peak isometric force of the right leg quadriceps muscle was measured with a Biodex isokinetic dynamometer. The subjects sat upright with the knee joint positioned at an angle of 60°. Shank, trunk, and shoulders were stabilized by belts. Direct electrical muscle stimulation was applied using two carbonized rubber electrodes covered with a thin layer of electrode gel (ECG–EEG Gel; Medigel, Modi’in, Israel). One of the electrodes (6 cm × 11 cm) was placed transversely across the width of the proximal portion of the quadriceps muscle next to the inguinal ligament; the other electrode (6 cm × 20 cm) covered the distal portion of the muscle above the patella. An electrical stimulator (MG 440; Medicor, Budapest, Hungary) delivered square-wave pulses of 1 ms duration. Each subject was familiarized to the experimental procedures and electrical stimulation on a separate occasion before the actual testing. The amplitude of the square-wave current pulses required to obtain maximum force was determined by gradually increasing the voltage until no increment in force response was elicited by a 10 % voltage increase. We measured peak force during 1 s trains of current pulses given at 20 or 100 Hz and separated by a 5 s resting period. Each frequency was only tested once at each time point. Tests were performed before and directly, 5, 10, 30 and 60 min after exercise. The test directly after exercise was performed ~30 s after the end of isometric MVCs, and ~2 min after the end of all-out cycling and drop jumps where subjects had to move to the dynamometer chair. Peak MVC force was measured before and 5, 10, 30 and 60 min after exercise.

We used direct electrical muscle stimulation to be able to assess exercise-induced changes in muscle function independent of any changes neuronal muscle activation. Several studies have shown major differences in the response to exercise performed with voluntary contractions vs. similar exercise with electrically evoked contractions (e.g. Hansen et al. 2009; Jubeau et al. 2008). In the present study, all types of exercise were performed with voluntary contractions and only the testing before and after exercise involved contractions induced by direct electrical stimulation. Thus, it appears unlikely that the observed changes in force production after exercise were significantly affected by the brief electrical stimulation.

Data are presented as mean ± SEM. One way analysis of variance (ANOVA) for repeated measures was used to determine statistical differences in force production during recovery vs. before exercise. If significant effects were found, post hoc testing was performed with Bonferroni correction for multiple comparisons. The level of significance was set at 0.05.

Initial experiments were performed with continuous MVCs. Directly after the fatiguing contractions, 20 and 100 Hz forces were decreased by ~25 % with 30 s MVCs and by ~45 % with 60 s MVCs (Fig. 1a). Forces then showed an initial full recovery, which was followed by a secondary decrease at 20 Hz whereas the 100 Hz force remained high. Thus, there was a prolonged decrease of 20 Hz force and the 20/100 Hz force ratio of ~10 and ~20 % after the 30 and 60 s MVCs, respectively.

The force decrease was more severe when the MVC duration was increased to 120 s, but the general pattern during recovery was similar to that with the shorter MVCs; that is, an early improvement followed by a prolonged decrease, which now also involved 100 Hz force (Fig. 1b). At 60 min of recovery, 20 and 100 Hz force and the 20/100 Hz force ratio were decreased by about 60, 20 and 50 %, respectively.

The force produced during a prolonged MVC gradually declines due to fatigue development, which involves decreased activation from the central nervous system and/or impaired sarcolemmal excitability (Bigland-Ritchie et al. 1983; Kent-Braun 1999), both of which will limit the energy metabolic stress during the contraction. Having this in mind, we also tested the effect of repeated short MVCs (12 × 5 s contractions at 60 s interval), which would induce less activation failure and hence put a larger stress on energy metabolism. Despite a total duration of contraction of only 60 s, the force–time integral of these repeated contractions was similar to that of the 120 s MVCs and consequently the average force was about twice as large with the repeated MVCs (Table 2). The magnitude of the prolonged force depression after the repeated MVCs was similar to that observed with 120 s MVCs, but the initial transient improvement did not occur (Fig. 1c).

In the next series of experiments, subjects performed all-out cycling, which involves energetically highly demanding concentric contractions. Subjects performed either 30 s of all-out Wingate cycling (Bar-Or 1987) or bouts of 5 s all-out cycling. The power output showed a marked decline during each 30 s bout, whereas it was well maintained during the 5 s bouts (Fig. 2). The pattern of force changes after one bout of 30 s Wingate cycling was similar to that observed with the prolonged MVCs: an initial force improvement, which was followed by a prolonged secondary reduction especially of 20 Hz force (Fig. 3a). A more severe force depression was observed after three Wingate cycling bouts performed at 4 min interval (Fig. 3b). There was a slight tendency of an initial force recovery after the three Wingate cycling bouts and thereafter all measured parameters remained severely depressed. In a final set of cycling experiments, subjects performed 12 × 5 s all-out cycling bouts at 1 min interval. This resulted in a prolonged force depression of a magnitude similar to that observed with 3 × 30 s Wingate cycling, but it was not accompanied by any sign of transient early recovery (Fig. 3c).

In a final series experiments, subjects performed drop jumps at 30 s interval. This kind of exercise involves repeated eccentric contractions, which means that muscles are exposed to large mechanical stress whereas the demand on energy metabolism is limited. The force at 20 and 100 Hz stimulation and the 20/100 Hz force ratio were all markedly decreased directly after the 10 drop jumps and they remained depressed throughout the 60 min recovery period (Fig. 4a). More severe decreases were observed after the 100 drop jumps with reductions ranging from ~35 % for 100 Hz force to ~65 % for 20 Hz force (Fig. 4b).

Maximum voluntary contraction forces were followed between 5 and 60 min after each of the above presented exercises (Fig. 5). The force reductions after exercise were generally smaller than those obtained with electrical muscle stimulation, which implies that the impaired muscular force production can be compensated for by increased neuronal activation.

A transient initial force recovery was observed with single MVCs (30–120 s duration) and with a single Wingate cycling bout (30 s duration). A common property of these exercises is a shorter duration from the start to the end of exercise (≤2 min) than with repeated MVCs (12 × 5 s with 60 s rest between contractions, totally 12 min) and the all-out cycling protocols (3 × 30 s at 4 min interval, totally 9.5 min; 12 × 5 s at 1 min interval, totally 12 min). Thus, these results indicate that the prolonged force depression (2) developed already during the more long-lasting exercises and hence these showed no fast recovery phase. Two likely mechanisms underlying the decreased force production after the single MVCs and the single Wingate cycling can be proposed. First, prolonged continuous contractions are accompanied by decreased central activation of motor units and possibly impaired action potential propagation within the muscle fibers (Bigland-Ritchie et al. 1983; Kent-Braun 1999). This type of activation failure can be rapidly reversed. For instance, reducing the stimulation frequency during ongoing, electrically induced contractions may actually increase force production, presumably by limiting K⁺- and Na⁺-fluxes over the sarcolemma and hence counteracting problems related to high-frequency action potential propagation (Jones et al. 1979; Westerblad et al. 1990). Second, a major metabolic factor causing decreased force production in acute fatigue is myoplasmic accumulation of inorganic phosphate ions due to breakdown of phosphocreatine (Allen et al. 2008; Dahlstedt et al. 2000). Phosphocreatine recovers to the control level, and hence the increase in inorganic phosphate ions vanishes, within a few minutes after prolonged contractions (Henriksson et al. 1986). To conclude, two mechanisms might explain the rapid initial recovery: reversal of deficient activation and the return of myoplasmic inorganic phosphate ions down to control levels. Deficient activation appears more likely to occur during the prolonged MVCs, where the requirement of a continuous generation of action potentials imposes a high risk of problems related to large K⁺- and Na⁺-fluxes (Bigland-Ritchie et al. 1983; Jones et al. 1979; Kent-Braun 1999; Westerblad et al. 1990). Conversely, Wingate cycling involves repeated brief contractions followed by rest periods, which means a lower risk of action potential failure, whereas the demand on rapid energy metabolism is high and large increases inorganic phosphate are likely to occur.

A prolonged force depression that was more prominent at 20 Hz than at 100 Hz stimulation was observed after all MVC and all-out cycling protocols. This type of PLFFD can theoretically be due to decreased sarcoplasmic reticulum (SR) Ca²⁺ release and/or decreased myofibrillar Ca²⁺ sensitivity (Allen et al. 2008). Recent studies aimed at identifying the mechanism(s) behind this force depression have revealed a key role of increased production of reactive oxygen/nitrogen species (ROS). The results from these studies show a coherent picture where the cellular ROS handling determines whether the dominating cause of the force depression is decreased SR Ca²⁺ release or reduced myofibrillar Ca²⁺ sensitivity (Cheng et al. 2016), Decreased SR Ca²⁺ release dominates in conditions where accumulation of superoxide anions (O ₂ ^·− ) is favored; conversely, reduced myofibrillar Ca²⁺ sensitivity is the dominating cause with facilitated clearance of O ₂ ^·− , which can be achieved by increased endogenous concentration of superoxide dismutase (converts O ₂ ^·− to hydrogen peroxide (H₂O₂)) or application of exogenous antioxidants (Bruton et al. 2008; Cheng et al. 2015; Watanabe et al. 2015). The SR Ca²⁺ release channel, the ryanodine receptor 1 (RyR1), has been shown to be particularly sensitive to ROS-induced modifications resulting in impaired function and muscle weakness (Andersson et al. 2011; Bellinger et al. 2008). In a recent study (Place et al. 2015), we showed a marked PLFFD and a striking RyR1 fragmentation in muscles of recreationally active subjects after 6 × 30 s Wingate cycling bouts. Intriguingly, the same exercise caused a similar PLFFD in elite endurance athletes, but in this case the RyR1 remained intact and the difference can be explained by a higher superoxide dismutase expression in elite athletes (Place et al. 2015). Thus, the mechanism behind PLFFD induced by MVCs and Wingate cycling bouts can, depending on the training status, be either impaired SR Ca²⁺ release or reduced myofibrillar Ca²⁺ sensitivity.

Repeated drop jumps resulted in marked PLFFD, which was more severe after 100 than after 10 drop jumps and which tended to be more stable than after the metabolically more demanding types of exercise (MVC and all-out cycling). Thus, no initial transient recovery was observed even after 10 drop jumps where the total exercise duration was 4.5 min (9 × 30 s), i.e., less than the 5 min required for the initial force increase to reach its peak after short-lasting MVCs and Wingate cycling. Unaccustomed eccentric contractions are followed by signs of general muscle damage, such as, delayed onset muscle soreness, swelling, protein leakage and inflammation (Clarkson and Hubal 2002). Accordingly, we have previously shown that repeated drop jumps cause a marked PLFFD, which was not fully reversed even after 14 days (Dargeviciute et al. 2013; Skurvydas et al. 2011). They also induced severe delayed onset muscle soreness and substantially increased plasma creatine kinase activity, which indicates substantial protein leakage, and the magnitude and duration of these signs of muscle damage were larger after 100 than after 50 drop jumps (Dargeviciute et al. 2013; Skurvydas et al. 2011). Eccentric contractions exert a large mechanical stress on activated muscle fibers and a likely cause of the force depression after these contractions is myofibrillar damage and disorganized sarcomeres (Fridén et al. 1983; Proske and Morgan 2001; Yu et al. 2004). Still, muscle problems induced by eccentric contractions might also be linked to increased ROS production (Kerksick et al. 2010; Kon et al. 2007; Nikolaidis et al. 2008; Pal et al. 2013) and changes in cellular Ca²⁺ handling (Balnave and Allen 1995; Gehlert et al. 2012; Ingalls et al. 1998). If ROS- and/or Ca²⁺-dependent processes are involved in the PLFFD after drop jumps, these processes are unlikely to be identical to those underlying PLFFD after MVC and all-out cycling exercises, because they were triggered by markedly different major challenges (mechanical vs. metabolic stress) and they developed with different time-courses. Moreover, additional complex interactions may occur; for instance, delayed increases in ROS have been observed in skeletal muscle after eccentric contractions and these have been linked to leukocyte infiltration and inflammation (Nikolaidis et al. 2008).