In this work, we assessed the torque and MMG reduction at the end of electrically stimulated contraction of the tibialis anterior in young and old subjects. Our main findings suggest that the age-dependent changes in muscle mechanics during the relaxation process with age are well described by the specific parameters obtained through the analysis of the dynamics of these two signals. These alterations could partially explain the specific features of alternating movements of walking in the elderly.

The comparative analysis of the time course of muscle length changes, indirectly measured using the MMG, and the force in human muscles has been already reported in the literature (Celichowski et al. 1998; Yoshitake et al. 2005, 2008; Orizio et al. 2008; Cogliati et al. 2020; Cè et al. 2017, 2013a, 2013c; Longo et al. 2017, 2016; Shinohara and Søgaard 2006; Jaskólska et al. 2003; Esposito et al. 2011, 2016). Only few papers, however, have investigated the phenomena during constant frequency tetanic stimulation to isolate the mechanical muscle response from the features of the neural control involved in voluntary contraction. In particular, several authors investigated the torque and MMG changes at the onset (Cè et al. 2017, 2013c, 2013a; Cogliati et al. 2020; Esposito et al. 2011, 2016) or at the end of evoked activity (Longo et al. 2016; Esposito et al. 2011, 2016; Cè et al. 2013c, 2014a, 2013b). The MMG detection techniques used the accelerometers (Longo et al. 2016; Esposito et al. 2011, 2016; Cè et al. 2013b, 2017) or the laser distance sensor (Cogliati et al. 2020; Orizio et al. 2013; Yoshitake et al. 2008). Here, the experimental design based on tetanic stimulation allowed, for the first time in humans, to describe the time relationship between the tension decrease and the muscle re-lengthening throughout the entire relaxation process. These results are in substantial agreement with the data reported in cats (Orizio et al. 2003).

We identified the end of electrical activity by analogy and mirroring the criteria generally accepted for the calculation of the time spent for electromechanical coupling during evoked contraction, in which the time interval between the first applied stimulus (onset of EMG signal) and the detectable tension increase is measured. As a consequence, the delay (i.e., the time interval spent before starting the relaxation process) was measured from the end of the electrical activity time instant and the beginning of torque and MMG decay (see Methods for its identification procedure). The comparison of our results with relaxation data from the literature is difficult given the scarce number of papers on this topic. Moreover, the delay of torque and MMG reported in some papers (Booth et al. 1997; Hespel et al. 2002; Cè et al. 2014a, 2013b; Esposito et al. 2011) were obtained using different techniques of signal detection (e.g., MMG was transduced by an accelerometer) and/or included the whole duration of the electrical activity evoked by the last stimulus of the tetanic train. On the contrary, in our procedure, this interval was not considered. For this reason, the absolute values of DT and DMMG reported are not directly comparable with those determined by the cited authors. Nonetheless, the literature data and ours agree that the DMMG is always greater than DT. These differences, in the signals’ decay onset, as well as in the absolute (RRT and RRMMG) and relative (NRRT and NRRMMG) decay rates consistently indicate that force declines before muscle re-lengthening. This last complex process is influenced by several determinants such as the tendon shortening after the fall of tension and other possible factors (elastic energy storage restitution, changes in fluid distribution, variation in intramuscular pressure, etc.) as discussed by Orizio et al. (2003) and partly described by the model suggested by Uchiyama and Hashimoto (2011).

As mentioned above, DT and DMMG measured in this study are not directly comparable with those determined in several studies on the mechanical process of relaxation (Cè et al. 2013b, 2014a; Esposito et al. 2016; Longo et al. 2016, 2017) when the young population is considered. On the contrary, TRT (reflecting the time spent from 80 to 20% force decay) is in line with those reported for the investigated subjects below 30 years old when the different muscle mass and the interval along the tension reduction period are taken into account (Cè et al. 2014a; Booth et al. 1997; Hespel et al. 2002). No data are reported in the literature about the muscle surface dynamics (MMG) during the relaxation phase after tetanic stimulation.

When the aged population is considered, no data about all the above considered mechanical parameters after tetanic stimulation can be found in the literature.

Our data indicate that in both torque and MMG, the beginning of decay and their reduction velocity (absolute and relative) occur later and are slower in old than young, respectively. To discuss this influence of age on the relaxation process, some basic consideration on its cellular mechanism can be useful. However, we have to keep in mind that the data retrieved from single fibers or isolated myofibrils cannot be directly used to explain experimental data obtained from the whole muscle tendon unit in humans. The cellular data can only highlight some factors that cannot be disregarded, but their action can be “filtered” by the much more complex situation of the “in vivo” experimental setup.

Muscle contraction is a consequence of the electromechanical coupling. Basically, it is determined by the increase of the cytosolic [Ca++] that removes the thin filament inhibitory state and allows the acto-myosin interaction or cross-bridge cycle. Vice versa, during muscle relaxation, the muscle active tension decreases as the decline of cytosolic [Ca++] restores the inhibitory state of the thin filament, so that myosins which end the cross-bridges cycle cannot start a new one. The sarcoendoplasmic reticulum calcium transport ATPase (SERCA) pump plays a key role given its function of removing calcium from the cytoplasm and moving it in the reticulum (Periasamy and Kalyanasundaram 2007). It is important to recall that when [Ca++] decreases below the threshold for thin filament activation, the number of attached active cross bridges initially slowly decays and then undergoes a fast collapse with a sudden increased rate of cross-bridge detachment (Cleworth and Edman 1972; Tesi et al. 2002). Thus, at the end of an isometric contraction, the tension decline takes place in two phases: in the first phase, the sarcomeres are kept in an isometric condition (slow phase), while in the following one the sarcomeres either shorten or elongate with a dramatic force fall (fast or chaotic phase) (for a review see Poggesi et al. 2005; Hill et al. 2021). Even if the extent at which the described phenomena can contribute to the time course of relaxation in intact mammalian muscle at physiological temperature (Hill et al. 2021) is not known, the body of knowledge on the muscular relaxation process must drive our interpretation on the differences in torque and MMG parameters in the young and old populations found in this study.

Following the interpretation outlined above, the rate of myoplasm Ca++ removal likely determines the time at which the threshold for tension decay is reached and probably is monitored by the DT and DMMG parameters. The difference in these parameters in the young and old may be due to differences in their SERCA characteristics. Indeed, Lamboley et al. (2014) reported that two different SERCA isoforms are expressed in fast and slow fibers. They appear to have different rates of calcium re-uptake. The slow fibers, with SERCA2 isoform, have a slower Ca++ re-uptake than the fast fibers with the SERCA1 pump. Given the prevalence of slow fibers in aged tibialis anterior (Orizio et al. 2016; von Haehling et al. 2010; Miljkovic et al. 2015), this can explain the longer delay of mechanical decay in old vs young. Moreover, Xu and Van Remmen (2021) reported a lower functionality of SERCA pump in aged subjects.

The steeper portion of torque and MMG decay, described by RR and TR, is likely and strongly influenced by the myosin cross-bridge kinetics. As shown by Belus et al. (2003) and Stehle et al. (2002), the chaotic and fast phase of tension decline is markedly dependent on myosin isoforms. Therefore, the above-mentioned prevalence of slow fibers in the tibialis anterior of elderly subjects can well explain the lower RR and NRR.

When the whole muscle tendon unit during relaxation after activity is considered, our data showed that the aged tibialis anterior re-elongates at a lesser extent for the same tension reduction compared to the young tibialis anterior (for muscle compliance in the 20–0% tension reduction). This is in agreement with the statistically significant greater value of MMG0T. The explanation can be found in the changes of muscle and tendon stiffness with aging. While tendon stiffness decreases (Magnusson et al. 2008), muscle stiffness increases with aging. The factors influencing the increase in muscle stiffness, passive resistance, in aged mammals are well described in Gajdosik et al. (2005) and can be summarized as: a. a substitution process of the contractile tissue with fat and connective tissue; b. a larger collagen amount in muscles dominated by slow twitch muscle fibers as those of old subjects. By analogy, the compliance of the muscle to re-elongation (tracked by MMG measure) due to the passive restitution of the elastic energy stored during tetanic stimulation may be influenced by the changes of the muscle viscoelastic properties with aging.