Force re-development after shortening reveals a role for titin in stretch–shortening performance enhancement in skinned muscle fibres

Stretch–shortening cycles (SSCs) play a crucial role in everyday movements and are a fundamental type of muscle action during terrestrial locomotion. SSCs involve a combination of muscle lengthening (eccentric contractions) immediately followed by muscle shortening (concentric contractions) (Cavagna et al., 1968). It is widely recognized that the force, work and power output of muscles is significantly enhanced during the shortening phase of SSCs compared with pure shortening contractions (Cavagna et al., 1968). This performance enhancement is known as the SSC effect and has been observed consistently across various scales – from single muscle fibres (Fukutani and Herzog, 2020; Tomalka et al., 2020, 2021), to whole muscle preparations (Hessel et al., 2021; Lee et al., 2001), to human movements (Aeles and Vanwanseele, 2019; Hahn and Riedel, 2018; Navarro-Cruz et al., 2019; Seiberl et al., 2015). Despite this knowledge, the detailed mechanisms responsible for the SSC effect are not yet fully understood (Seiberl et al., 2021). However, recent evidence suggests that the interplay between cross-bridge (XB)-based and non-XB-based structures at the sarcomere level contributes to the SSC effect (Hessel et al., 2021; Tomalka et al., 2020, 2021).

On the muscle fibre level, XB-forces refer to forces generated by the power stroke of the myosin heads attached to the actin filaments, while non-XB-forces refer to forces from viscoelastic proteins such as titin. We have recently shown that the SSC effect depends on contraction velocity and that non-XB forces increase with velocity during the lengthening phase of a SSC (Tomalka et al., 2021). These findings suggest that non-XB structures, such as titin, play a crucial role in eccentric loading because of their adjustable viscoelastic properties (Linke, 2018; Nishikawa, 2020; Tomalka, 2023). However, it is still under investigation whether non-XB structures also play a role in shortening after eccentric loading.

While there is a growing body of research on the SSC effect during its dynamic phases, this study focuses on the force re-development in the transition to a steady state after the SSC to gain a deeper understanding of the SSC effect. In a study by Seiberl et al. (2015) on electrically stimulated human M. adductor pollicis, it was found that the stretch-induced residual force enhancement [rFE; enhanced forces after stretch compared with the muscles' corresponding isometric steady-state force during fixed-end contractions; see Abbott and Aubert (1952) and reviews by Hahn et al. (2023), Herzog et al. (2016) and Nishikawa (2016) on this issue] was not immediately abolished during shortening and contributed to the increased force and work during and after SSCs. Thereby, it was shown that the magnitudes of SSC velocities affect force-enhancing and force-depressing mechanisms: while rFE seems to be largely unaffected by stretch velocity, the amount of force depression decreases with increasing shortening velocities (Hahn et al., 2023; Herzog and Leonard, 2000).

Seiberl et al. (2015) also investigated the rate of force re-development following SSCs and pure shortening contractions and found that force re-development was significantly greater after SSCs than after pure shortening contractions. Although it has not been explicitly examined, an indication of an increased force re-development after SSCs compared with pure shortening contractions can also be found in a study on rabbit muscle fibres (Fukutani and Herzog, 2020). Seiberl et al. (2015) suggested that a tuneable viscoelastic non-XB structure, such as titin, could explain the faster re-development of force following SSCs. However, the potential contributions of XB and non-XB structures to the force re-development are still not well understood (Corr and Herzog, 2005). This is because it remains unclear whether the stretch-induced mechanisms of rFE are offset during the shortening phase in SSCs, either by shortening-induced force depression (FD) (Edman et al., 1993; Maréchal and Plaghki, 1979; see reviews by Chen et al., 2019; Hahn et al., 2023 on this issue) or due to a decrease in rFE during active muscle shortening (Fukutani and Herzog, 2020).

Joumaa et al. (2021a) conducted a study to investigate the energy cost of force production after SSCs and pure shortening contractions in skinned rabbit fibre preparations. Contrary to their hypothesis, they found no difference in ATP consumption per unit of force between the two types of contraction although they observed increased forces following SSCs compared with pure shortening contractions. The authors suggested that the absence of the expected increase in metabolic efficiency after SSCs probably coincided with a simultaneous increase in the proportion of attached XBs, which was concluded from a proportional increase in stiffness with force in the steady state following SSCs and pure shortening contractions. These experimental findings highlight the possible contribution of an increased proportion of attached XBs in addition to increased titin-based forces to the increased force production during and following SSCs compared with pure shortening contractions (Joumaa et al., 2021a).

Therefore, the main objective of the present study was to examine the characteristics of the force response immediately following SSCs and pure shortening contractions to further understand the SSC effect. Subsequently, the study aimed to provide insights into whether an increase in titin-based stiffness or other non-XB stiffness triggered during the stretch phase of SSCs can counteract shortening-induced force depression.

We designed experiments within the physiological range of the force–length relationship, using different SSC velocities (30%, 60% and 85% of the maximum shortening velocity, v₀) and a constant pure shortening contraction velocity (85% v₀) with identical amplitude. By applying the XB-inhibitor blebbistatin, we aimed to identify the distinct contributions of XB and non-XB structures to the force generation in single skinned muscle fibres (Fig. 1; see also Fig. S1). We expected that XB inhibition would lead to substantial force loss and reduced force re-development in the blebbistatin treatment compared with the control. However, we hypothesized that SSCs would result in greater and faster force re-development than pure shortening contractions in both control and blebbistatin treatments. Moreover, given the demonstrated significance of non-XB structures in SSC force generation (Fukutani and Herzog, 2020; Fukutani et al., 2017; Hessel et al., 2021; Tomalka et al., 2020), along with the observation that non-XB energy stored during the eccentric phase contributes to the SSC effect in a velocity-dependent manner (Tomalka et al., 2021), we further hypothesized that the rate of force re-development following SSCs is associated with the contraction velocity.

The skeletal muscle fibres from rats used for this study were provided by another animal study that was approved according to the regulations of the German Animal Protection Law [Tierschutzgesetz, §4 (3); Permit Number: 35-9185.81/0491].

Muscle preparation, storage and activation techniques for permeabilized single muscle fibres were carried out following the methods described in Tomalka et al. (2017, 2020). Briefly, the experiments were conducted using glycerinated skinned single fibre segments obtained from the soleus muscles of six freshly euthanized female Wistar rats (age 3 months, mass 428–520 g, cage sedentary, 12 h:12 h light:dark cycle, housing temperature 22°C). After permeabilization in a skinning solution, the bundles were stored at −20°C in a storage solution (see table 1 in Tomalka et al., 2017).

On the day of the experiments, small segments of the skinned fibre bundles were dissected, and single muscle fibres were isolated. The fibre bundles were then treated with a relaxing solution (refer to table 1 in Tomalka et al., 2017) containing Triton X-100 (1% v/v) for 2–3 min at 4°C. This treatment ensured the removal of internal membranes without affecting the contractile apparatus (Linari et al., 2007). The two ends of the single muscle fibre were attached to a hook connected to a force transducer (model 403a, Aurora Scientific, Aurora, ON, Canada) and a high-speed length controller (model 322C-I, Aurora Scientific), respectively. During the kinetic experiments, sarcomere length (SL) was measured using a high-speed camera system (model 901B, Aurora Scientific) combined with a ×20 ELWD dry objective (NA 0.40, Nikon) and an accessory lens (×2.5, Nikon). The average sarcomere length of each single fibre was set to 2.4±0.05 µm (mean±s.d.), and was defined as the individual optimal fibre length (L₀). The width (w) and height (h) of the fibres were measured at three different positions in the central segment of the relaxed fibre using a ×10 dry objective (NA 0.30, Nikon) and a ×10 eyepiece, and then averaged. The fibre cross-sectional area (CSA) was determined assuming an elliptical cross-section of single muscle fibres. The CSA was 4.844±1.246 µm² (mean±s.d.). All experiments were conducted at a temperature of 12±0.1°C.

The protocol is described in detail in Tomalka et al. (2020, 2021). It involved different isokinetic ramp perturbations (SSC and pure shortening contraction experiments, referred to as ‘conditions’) to investigate the dynamic and static force responses during and after these perturbations. The experiments were conducted on skinned muscle fibres from six rats, with a total of 16 fibres.

The protocol included two ‘treatments’ of repeated measurements: the control treatment (without XB inhibition) and the blebbistatin treatment (with XB inhibition). In the control treatment (Fig. 1), we aimed to examine the dynamic and static force responses immediately following isokinetic ramp perturbations of single muscle fibres under physiological conditions. In the blebbistatin treatment, the same skinned muscle fibres were subjected to the same ramp perturbations as in the control treatment. However, in this treatment, 20 µmol l⁻¹ of the XB inhibitor blebbistatin was added to all solutions. The purpose of this treatment was to identify and assess the non-XB contributions to muscle force. Blebbistatin is known to inhibit the interaction between actin and myosin, thereby preventing the formation of strongly bound XBs. However, it still allows for the formation of weakly bound XBs, contributing to stiffness rather than force (Iwamoto, 2018). Blebbistatin does not affect titin mobility (Shalabi et al., 2017). Administering blebbistatin resulted in a suppression of active XB forces by up to 98% in fixed-end contractions. Therefore, it is expected that blebbistatin will similarly suppress active forces during SSCs and pure shortening contractions.

The order of the ramp protocol was randomized to minimize any potential bias or systematic effects. Each ramp experiment consisted of three phases: an initial fixed-end phase, a ramp phase (either SSC or pure shortening contraction), and a final isometric phase (Fig. 1). Skinned muscle fibres were activated via calcium diffusion (pCa 4.5) in the presence of ATP.

Activated single skinned fibres were stretched from 2.0 µm (0.8 L₀) to approximately 2.4 µm (1.0 L₀) and immediately shortened back to approximately 2.0 µm (Fig. 1, bottom). The perturbations were conducted at different stretch–shortening velocities of 30%, 60% and 85% of v₀ (Fig. 1, blue, red and yellow lines, respectively). The activated fibres were held at this length until a steady state was achieved. The maximum shortening velocity of skinned soleus muscle fibres from adult male Wistar rats (0.46±0.13 L₀ s⁻¹; n=6) was determined in separate experiments as outlined in Tomalka et al. (2021).

Activated single skinned fibres were shortened from 2.4 µm (1.0 L₀) to approximately 2.0 µm (Fig. 1, black line) at 85% v₀ and held at this length until a steady state was reached.

The relaxing solution contained (in mmol l⁻¹): 100 TES, 7.7 MgCl₂, 5.44 Na₂ATP, 25 EGTA, 19.11 Na₂CP and 10 glutathione (pCa 9.0). The pre-activating solution contained (in mmol l⁻¹): 100 TES, 6.93 MgCl₂, 5.45 Na₂ATP, 0.1 EGTA, 19.49 Na₂CP, 10 glutathione and 24.9 HDTA. The activating solution contained (in mmol l⁻¹): 100 TES, 6.76 MgCl₂, 5.46 Na₂ATP, 19.49 Na₂CP, 10 glutathione and 25 CaEGTA (pCa 4.5). The skinning solution contained (in mmol l⁻¹): 170 potassium propionate, 2.5 MgCl₂, 2.5 Na₂ATP, 5 EGTA, 10 imidazole and 0.2 PMSF. The storage solution was the same as the skinning solution, except for the presence of 10 mmol l⁻¹ glutathione and 50% glycerol (v/v). Cysteine and cysteine/serine protease inhibitors [trans-epoxysuccinil-l-leucylamido-(4-guanidino) butane, E-64, 10 mmol l⁻¹; leupeptin, 20 µg ml⁻¹] were added to all solutions to preserve lattice proteins and thus sarcomere homogeneity (Linari et al., 2007; Tomalka et al., 2017).

pH (adjusted with KOH) was 7.1 at 12°C. Creatine phosphokinase (450 U ml⁻¹) was added to all solutions, except for skinning and storage solutions. Blebbistatin was obtained from Enzo Life Sciences Inc. (NY, USA); creatine phosphokinase was obtained from Roche (Mannheim, Germany); all other chemicals were obtained from Sigma (St Louis, MO, USA).

Force and length data were collected at a frequency of 1 kHz using real-time software (600A, Aurora Scientific) and an A/D Interface (604A, Aurora Scientific). Data analysis was performed using a custom-written MATLAB script (version R2022b, The MathWorks, Natick, MA, USA). Unless stated otherwise, forces are normalized to the individual maximum muscle force (F/F₀).

The shortening velocity was normalized to the specific maximal shortening velocity of the fibre (v/v₀). Fibre lengths are expressed relative to the optimum fibre length (L/L₀). Average sarcomere lengths are reported as absolute values in micrometres (µm). Minimum force values (F_min) were determined immediately at the end of the shortening phase in both the SSC and pure shortening contraction conditions (Fig. 1, inset i, open circles). Final force values (steady-state force, F_ss) were determined after the completion of the ramp perturbation (SSCs and pure shortening contractions) at the end of the steady-state isometric phase (Fig. 1, inset ii, open diamonds). For data analysis, previous data from isokinetic ramp experiments (SSCs, pure shortening contractions) from Tomalka et al. (2020, 2021) were utilized and re-analysed.

To quantify differences in the transient period of force re-development following active shortening, data were smoothed by using a third-order median filter, and the maximal rate of force re-development (RF_dev) was calculated in both the SSC and pure shortening contraction conditions (Fig. 1; see also Figs S1 and S2). RF_dev was obtained by calculating the first derivative of force over time, representing the maximum change in force divided by the corresponding change in time (ΔF/Δt). Additionally, the time taken to reach the maximum rate of force re-development (referred to as t_RFdev) was determined. To compare RF_dev among different SSC velocities (30%, 60% and 85% v₀) and conditions (SSC versus pure shortening contractions), all force–time data were normalized. Specifically, F_min was set to ‘0’, while F_ss was set to ‘1’ (Seiberl et al., 2015) (refer to Fig. 3).

We conducted a two-way repeated measures analysis of variance (rmANOVA) to evaluate the effects of contraction condition and treatment on the outcome variables (i.e. F_min, F_ss, RF_dev and t_RFdev). The factors included four levels of contraction condition (SSC 30% v₀, SSC 60% v₀, SSC 85% v₀, pure shortening contraction 85% v₀) and two levels of treatment (control and blebbistatin). The statistical analysis aimed to determine: (1) the main effect of contraction condition on the outcome variables, (2) the main effect of treatment on the outcome variables and (3) the interaction effect between contraction condition and treatment. Post hoc pairwise comparisons with Fisher's least significant difference (LSD) correction were conducted if significant main effects or interactions were identified by the rmANOVA. Specifically, for significant main effects of contraction condition and/or interaction, pairwise comparisons were performed between all levels of SSC contractions (SSC 30% v₀, SSC 60% v₀, SSC 85% v₀) to analyse the effect of contraction velocity, and pairwise comparisons of the conditions SSC 85% v₀ and pure shortening contraction 85% v₀ were used to identify the effect of active stretch before shortening. Similarly, we conducted pairwise comparisons for significant main effects of treatment. If violations of sphericity were observed, the Greenhouse–Geisser correction was used. Statistical analyses were realized using SPSS (version 29, 2022, IBM Corp., Armonk, NY, USA).

The level of significance was set at P<0.05. All data are presented as means±s.d. unless otherwise stated. The effect sizes for the ANOVA were classified as small (η²<0.06), medium (0.06≤η²≤0.14) and large (η²>0.14). The effect sizes for the paired t-tests were classified as small (d<0.5), medium (0.5≤d≤0.8) and large (d>0.8) (Cohen, 1998).

For the statistical analysis of temporal differences in continuous/dynamic data (kinetic time series), Statistical Parametric Mapping (SPM) was employed (Serrien et al., 2019). The MATLAB software package spm1d, developed by T. Pataky (Pataky, 2012), was used for this purpose. The specific tests used are indicated in the figure legends.

A significant main effect for the factor ‘treatment’ (control versus blebbistatin) (P<0.01) was observed for all parameters tested (Table 1). The maximum rate of normalized force re-development (RF_dev) (control 0.94±0.20 versus blebbistatin 0.10±0.03 ΔF/Δt; t_66.2=32.53, P<0.001, d=5.76) and the final normalized force (F_ss) (control 0.91±0.05 versus blebbistatin 0.42±0.19 F/F₀; t_77.1=26.92, P<0.001, d=1.58) were greater in the control treatment than in the blebbistatin treatment. The time to reach RF_dev (t_RFdev) (control 107±91 versus blebbistatin 4236±3008 ms; t₆₃=−10.97, P<0.001, d=−1.94) and the minimum normalized force (F_min) (control 0.04±0.05 versus blebbistatin 0.11±0.13 F/F₀; t_83.9=−4.27, P<0.001, d=−0,75) were greater in the blebbistatin treatment than in the control treatment. Furthermore, a significant main effect for the factor ‘condition’ (P≤0.002) was observed for all tested parameters, except for F_ss (P=0.794). Significant interaction effects between treatment and condition were observed for RF_dev and t_RFdev (P<0.001), but not for F_min (P=0.387) and F_ss (P=0.925).

The RF_dev showed no significant difference between the varying SSC velocities (30%, 60% and 85% v₀) tested (P>0.05, n.s.) (Fig. 2A, upper plot), but significantly larger RF_dev for SSC 85% v₀ compared with pure shortening contraction 85% v₀ conditions (1.03±0.17 versus 0.68±0.09 ΔF/Δt; t₁₅=14.40, P<0.001, d=5.61) (Fig. 2A, upper plot; Table 1). Similarly, t_RFdev was not significantly different between varying SSC velocities (30%, 60% and 85% v₀) tested (P>0.05, n.s.) (Fig. 2A, lower plot). Furthermore, when comparing t_RFdev between SSC 85% v₀ and the pure shortening contraction condition, the maximum rate of force re-development was reached at an earlier time following SSCs (77±57 versus 205±113 ms; t₁₅=−5.05, P<0.001, d=−1.43) (Fig. 2A, lower plot; Table 1).

The results of the statistical analysis of the discrete parameters RF_dev and t_RFdev described above are further supported by the analysis of continuous force–time data following SSCs and pure shortening contractions, employing SPM. For the control treatment, the ANOVA using SPM revealed statistically significant differences within the first 1092 ms after the end of shortening (F_3,60=10.43, P<0.001, η²=0.34) (Fig. 3A). These differences occurred between 0 and 3.3% of the entire time needed for full force re-development (Fig. 3A, inset).

In general, administering blebbistatin drastically prolonged t_RFdev by roughly two orders of magnitude (see Fig. 2, lower plots). After the blebbistatin treatment, significant differences were observed in RF_dev between SSC 30% v₀ and SSC 60% v₀ (0.09±0.02 versus 1.12±0.03 ΔF/Δt, t₁₅=−2.77, P=0.001, d=−1.12) and between SSC 30% v₀ and SSC 85% v₀ (0.09±0.02 versus 1.12±0.02 ΔF/Δt, t₁₅=−3.83, P<0.001, d=−1.26), but not for SSC 60% v₀ versus SSC 85% v₀ (1.12±0.03 versus 1.12±0.02 ΔF/Δt, P=0.758) (Fig. 2B, upper plot). Additionally, when comparing RF_dev between SSCs of 85% v₀ and the pure shortening contraction condition, significant differences were observed (1.12±0.02 versus 0.06±0.01 ΔF/Δt; t₁₅=9.11, P<0.001, d=3.34) (Fig. 2B, upper plot; Table 1). Similarly, t_RFdev was significantly different between SSC 30% v₀ and SSC 60% v₀ (4028±1736 versus 2519±1259 ms, t₁₅=4.49, P<0.001, d=1.00) and between SSC 30% v₀ and SSC 85% v₀ (4028±1736 versus 2483±1441 ms, t₁₅=3.28, P=0.003, d=0.97), but not for SSC 60% v₀ versus SSC 85% v₀ (2519±1259 versus 2483±1441 ms, P=0.922) (Fig. 2B, lower plot). Furthermore, a pairwise comparison of t_RFdev between the SSC 85% v₀ and pure shortening contraction 85% v₀ condition revealed a significant difference (2483±1441 versus 7916±3230 ms; t₁₅=−6.40, P<0.001, d=−2.17) (Fig. 2B, lower plot; Table 1).

The ANOVA with SPM revealed statistically significant differences in force re-development between SSC and pure shortening contraction conditions between 1007 and 17,900 ms after the end of shortening (F_3,60=9.07, P<0.001, η²=0.31) (Fig. 3B). These differences occurred in a time window representing 3.1–54.9% of the entire time needed for full force re-development.

The minimum forces (F_min) observed at the end of the SSC shortening phases showed significant differences between SSC 30% v₀ and SSC 60% v₀ (0.10±0.02 versus 0.06±0.02 F/F₀, t₁₅=14.83, P<0.001, d=2.32) and between SSC 30% v₀ and SSC 85% v₀ (0.10±0.02 versus 0.04±0.02 F/F₀, t₁₅=17.88, P<0.001, d=2.97), but not between SSC 60% v₀ and SSC 85% v₀ (0.06±0.02 versus 0.04±0.02 F/F₀, P=0.144, d=0.92) (Fig. 4A, upper plot). Additionally, pairwise comparison of F_min between the SSC 85% v₀ and pure shortening contraction 85% v₀ condition indicated a significant difference (0.04±0.02 versus −0.03±0.04 F/F₀; t₁₅=10.15, P<0.001, d=2.21) (Fig. 4A, upper plot; Table 1).

In contrast, no statistically significant difference was observed in the forces at the end of the steady-state isometric phase (F_ss) between different conditions (SSCs and pure shortening contractions) (P=0.115) (Fig. 4A, lower plot).

After administering the XB inhibitor blebbistatin, no statistically significant difference in F_min was observed between different conditions (SSCs and pure shortening contractions) (P=0.174) (Fig. 4B, upper plot).

Similarly, no statistically significant difference in F_ss at the end of the steady-state isometric phase was shown among the different conditions tested (P=0.989) (Fig. 4B, lower plot).

The purpose of this study was to examine force re-development after both stretch–shortening and pure shortening conditions, aiming to enhance our understanding of the mechanisms involved in the SSC effect. Our findings confirm our hypothesis that force re-development (described by the parameters RF_dev, t_RFdev and F_min) is greater and faster following SSCs than following pure shortening contractions (Figs 2 and 3, Table 1). In general, administering blebbistatin drastically decreased RF_dev and prolonged the time to reach the maximum rate of force re-development (t_RFdev) by roughly two orders of magnitude.

The F_min values exhibited notable differences among all investigated SSC velocities (30%, 60% and 85% v₀) as illustrated in Fig. 4 and detailed in Table 1. Thereby, F_min decreased with increasing shortening velocity from 30% to 85% v₀. This relationship of decreasing F_min with increasing shortening velocity can be explained by the concentric force–velocity (F–v) relationship as XBs produce less force as shortening velocity increases (Fenn, 1924; Hill, 1938; Huxley, 1957).

Additionally, the forces at the end of the shortening phase were significantly lower in the pure shortening contraction condition than in the corresponding SSC of 85% v₀ (P<0.001, Fig. 4A, upper plot). This observation agrees with the outcomes of an in vivo study conducted by Seiberl et al. (2015), which investigated SSC and pure shortening contraction conditions in electrically stimulated human m. adductor pollicis under similar experimental conditions. The authors attributed the significantly lower forces at the end of the shortening phase for the pure shortening contraction conditions compared with the corresponding SSCs to the contraction history-dependent properties of skeletal muscle (Bullimore et al., 2007; Herzog and Leonard, 2000). Active muscle lengthening in the eccentric stretch phase leads to increased forces during the subsequent shortening phase in SSCs. Furthermore, Seiberl et al. (2015) noted that force depression following shortening (Abbott and Aubert, 1952; Joumaa et al., 2012; Maréchal and Plaghki, 1979; McDaniel et al., 2010) was consistently smaller or even abolished when active shortening was preceded by an active stretch on the m. adductor pollicis. Our findings at the muscle fibre level are further supported by in vivo investigations of the human thumb adduction forces by Fortuna et al. (2017). They demonstrated significantly greater forces at the end of shortening for SSCs compared with pure shortening contractions, with a notable decrease in forces observed with increasing SSC velocity.

The findings obtained by in vivo experiments on humans align with results from ex vivo experiments on mammalian muscle fibres (Fukutani et al., 2017; Tomalka et al., 2020). Consequently, it has been suggested that in SSCs, mechanisms related to stretch-induced residual force enhancement (Abbott and Aubert, 1952) persisted for a sufficient duration during and/or following shortening. Seiberl et al. (2015) concluded that force-enhancing effects probably persisted during at least part of the shortening phase in SSCs, contributing to a decrease in rFD and enhancing the work and power in SSCs compared with pure shortening contractions. This finding is supported by a more recent study by Fukutani and Herzog (2020). In skinned fibre, they observed that mechanical work, the force at the onset of shortening and the force at the end of shortening (F_min) were greater in the SSC compared with the pure shortening contraction experiments. These results align with our findings, showing that F_min was smaller in pure shortening contractions than with SSCs at corresponding velocities.

Intriguingly, within the pure shortening contraction condition, F_min even attains negative magnitudes (−0.03±0.04 F₀). This experimental finding may be attributed to potential variability in the inherent maximal contraction properties of the muscle fibres tested. Notably, in this study, the maximum shortening velocity of skinned soleus muscle fibres from adult male Wistar rats was 0.46±0.13 L₀ s⁻¹, which is in agreement with the literature data (Degens et al., 1998; Tomalka et al., 2021). Certain individual muscle fibres are likely to have contracted at their maximum unloaded shortening velocity (v₀), potentially causing inertial effects. This may be attributed to the displacement of fibre mass, leading to pressure on the force transducer (Ford et al., 1977). As a result, the output signal of the force transducer dropped below the baseline, consequently producing slightly negative F_min values.

To understand the dynamics of muscle contractions, particularly the physiological mechanisms of the SSC effect, it is crucial to distinguish between XB and non-XB contributions to total muscle force. Despite the trends observed in Fig. 4B (upper plot), which suggest that the mean forces at the end of the pure shortening contraction condition might be smaller compared with those in the SSC condition, no significant interaction was found between treatment and condition on F_min. The high variance observed in the blebbistatin treatment data probably accounted for the lack of a significant difference between the SSC and pure shortening contraction conditions. However, the observed trend in the forces at the end of the shortening phase for SSC and pure shortening contraction conditions aligns with findings of a previous in vitro study conducted on skinned muscle fibres subjected to ramp perturbations (stretching, shortening and combinations thereof) in states where XB-based force production was inhibited by 2,3-butanedione monoxime (BDM) (Fukutani and Herzog, 2020). The difference in F_min between SSCs and pure shortening contractions is primarily attributed to the higher forces observed just before the shortening phase in SSCs compared with pure shortening contractions, as suggested by Fukutani and Herzog (2020). In the blebbistatin treatment, the higher forces arise from a continuous loading of mainly non-XB elastic structures during active stretching until the stretching has stopped (Leonard and Herzog, 2010; Tomalka et al., 2017). The higher minimum forces observed in the blebbistatin treatment, where XBs are inhibited, compared with the control treatment (including both XB and non-XB elements) can probably be attributed to the normalization of the forces. When normalizing individual fibre forces to the maximum isometric force (F/F₀), high eccentric forces averaging approximately 6 F₀ were observed during the stretching phase in the blebbistatin treatment, while in the control treatment, the eccentric forces averaged approximately 1.3 F₀ only. The larger normalized forces during stretching directly contribute to the increased F_min observed at the end of the shortening phase in the SSCs in the blebbistatin treatment compared with the control treatment. The giant filament titin (Maruyama, 1976) emerges as a plausible candidate for energy storage during eccentric contractions (Linke, 2018, 2023; Tomalka, 2023). Titin is extensible (Linke, 2018; Rivas-Pardo et al., 2020) and viscoelastic (Herzog et al., 2014; Mártonfalvi et al., 2014; Tomalka et al., 2021), and its stiffness increases upon muscle activation (Herzog, 2018; Leonard et al., 2010; Nishikawa, 2020; Powers et al., 2016; Tomalka et al., 2017). Thus, if titin acts as a linear spring, adjusting its equilibrium length can modify its force (Jeong and Nishikawa, 2023; Linke, 2023; Monroy et al., 2017). The versatile functional and structural properties of titin facilitate the storage and release of kinetic energy, leading to higher force values at the end of the shortening phase (F_min) in SSCs compared with pure shortening contractions. This SSC effect (increased muscular capability) is associated with increased work and power (Fukutani and Herzog, 2020; Hessel and Nishikawa, 2017; Tomalka et al., 2020, 2021).

The results from the blebbistatin experiments in this study confirm recent observations that underscore the importance of non-XB structures such as titin in the performance enhancement of SSCs at the muscle fibre level (Fukutani and Herzog, 2019, 2020; Fukutani et al., 2017; Joumaa et al., 2021a; Tomalka et al., 2020, 2021).

The findings of this study demonstrate that the RF_dev following SSCs with various velocities was on average 50% greater compared with that following pure shortening contractions (Fig. 2, upper plot). Also, RF_dev was attained 63% earlier after SSCs in comparison to pure shortening contractions (t_RFdev; Fig. 2, lower plot), signifying a faster force recovery after SSCs. This finding is in agreement with in vivo observations on the human m. adductor pollicis made by Seiberl et al. (2015). The authors demonstrated that at a velocity of 170 deg s⁻¹ (comparable to our slowest speed of 30% v₀) for 30 deg shortening amplitudes, the mean rate of force re-development was 40% greater and occurred approximately 21% earlier after SSCs compared with the pure shortening contraction condition.

Ames et al. (2022) provided a potential explanation for this observation in their recent ex vivo study on rabbit psoas muscle fibres. These authors investigated XB-cycling kinetics during steady state following active shortening and stretching. Their results demonstrated that rFD is associated with significant alterations in XB kinetics, whereas rFE does not exhibit such changes. The lower RF_dev in pure shortening contractions, a key finding of their study, is probably explained by the idea that force re-development is driven by XB kinetics. Specifically, Ames et al. (2022) observed a decrease in the rate of force re-development, stiffness, rate of XB attachment and detachment and the force produced per XB in the pure shortening contraction condition. These findings are explained by the hypothesis that the reduction in force is associated with the amount of strain imposed on actin filaments during active shortening (Joumaa et al., 2012). This strain distorts actin and alters the orientation of XB attachment sites (Daniel et al., 1998), resulting in partially stress-induced inhibition of XBs (Joumaa et al., 2021b; Maréchal and Plaghki, 1979). This inhibition occurs through a decrease in attachment probability or by preventing attached XBs from generating force. However, the mechanism(s) responsible for reducing the number of attached XBs remain(s) unknown and specific measurements of actin deformation in response to stress have not been provided. Other potential mechanisms proposed to induce force depression include sarcomere length non-uniformity (Julian and Morgan, 1979; Morgan et al., 2000) and an increase in fatigue products (Granzier and Pollack, 1989) (see reviews by Hahn et al., 2023 and Herzog, 2004 on this issue).

In SSCs, where stretching is immediately followed by shortening contractions, it is hypothesized that XB kinetics differ compared with pure shortening contractions without a preceding stretch (Fukutani and Herzog, 2020; Joumaa et al., 2021a; Seiberl et al., 2015; Tomalka et al., 2020, 2021). These alterations in XB kinetics may also positively affect the rate of force re-development after SSCs. This could be attributed to several factors, including (i) an increase in elastic energy within the attached XBs, (ii) the contribution of active XB forces, (iii) the mechanosensing in thick filaments, and (iv) the engagement of the giant filamentous structure titin.

Linari et al. (2003) examined how much energy could be stored by stretching XBs and by redistribution of XB states. They extended the model by Piazzesi and Lombardi (1995) to predict energy changes during stretch. According to their model, energy storage in XBs can only occur in the early phase of the stretch, as attached XBs are pulled into a higher energy state. Consequently, XBs alone cannot explain energy storage during active stretch, nor the dissipation of this energy during shortening in muscle fibres. However, an XB state during stretching, characterized by rapid reattachment of forcibly detached XBs, has been proposed by several authors (Flitney and Hirst, 1978; Griffiths et al., 1980; Lombardi and Piazzesi, 1990).

Furthermore, the rate of XB detachment is lower after stretching than after shortening, indicating a reduced detachment rate for heads that have undergone strain (Lombardi et al., 1995). Consequently, the greater forces observed at the end of shortening following a preceding stretch, along with potential alterations in XB dynamics after SSCs, may contribute to a faster and greater force re-development after SSCs compared with pure shortening contractions. However, changes in XB dynamics can only partially account for the reported findings. Therefore, factors beyond the actomyosin apparatus must have played an additional role in the observed greater RF_dev after SSCs compared with pure shortening contractions.

There is an increasing acknowledgement of a mechanism of mechanosensing involving thick filaments in the regulation of muscle contraction (Brunello et al., 2020; Fusi et al., 2016; Linari et al., 2015). Recent X-ray diffraction studies on actively contracting fibres from striated skeletal muscle (Fusi et al., 2016; Linari et al., 2015; Piazzesi et al., 2018) propose that the myosin filament can exist in one of two states: a relaxed state (OFF) and an activated state (ON). In the OFF state, observed in resting muscle, the majority of myosin motors are unavailable for actin binding or ATP hydrolysis (Linari et al., 2015). However, the small fraction of ON motors allows the muscle to respond promptly to calcium activation when the external load is low (Linari et al., 2015). At high loads, the myosin filaments are switched ON by mechanical stress due to stretch-dependent activation, accompanied by the mobilization of more myosin motors that generate additional force (Fusi et al., 2016; Linari et al., 2015). A recent study by Fusi et al. (2016) suggests a potential role of titin in regulating muscle contractility through thick filament activation mediated by the mechanosensory pathway in the myosin filament (Brunello and Fusi, 2020; Linari et al., 2015). This mechanism may play a role in the SSC effect, as the activation of the thick filament is suggested to be stress dependent.

Contrary to our hypotheses, we found no effect of SSC velocity on force re-development (RF_dev and t_RFdev) under our experimental conditions and for the specific velocities tested. For the in vivo m. adductor pollicis, Fortuna et al. (2017) found significantly decreasing shortening-induced force depression (rFD) following SSCs with increasing SSC velocities. However, they also found that slow SSC velocities (15 and 20 deg s⁻¹) resulted in a similar rFD to that observed for pure shortening contractions. Nevertheless, it is important to note that the authors did not specifically investigate the rate of force re-development after SSCs and pure shortening contractions. Instead, the Fortuna et al. (2017) study focused on other metrics such as the maximum force at the end of the stretch, the minimum force at the end of shortening, and the isometric steady-state force before muscle deactivation.

Consequently, the superposition of various effects, including the influence of muscle length, force–velocity properties, and the amplitude of SSCs might have substantial implications for experimental findings in similar studies existing in the literature.

The explanations for the greater rates of force re-development observed after SSCs versus pure shortening contractions given above are supported by the investigation of non-XB contributions to total muscle force (blebbistatin treatment). The RF_dev was significantly greater and faster after SSCs compared with pure shortening contractions (P<0.001; Fig. 2B, Table 1). However, it cannot be taken for granted that blebbistatin completely eliminates XB-based force production, as blebbistatin seems to affect the contractile apparatus in a complex manner (Minozzo and Rassier, 2010; Rahman et al., 2018). Administering blebbistatin suppressed active XB forces by 98% in this approach. So it is likely that the faster force re-development after SSCs is – at least partially – associated with the rapid transition of attached XBs from weakly bound to strongly bound states (Huxley and Simmons, 1971). However, the minimal contribution of active XB-based force, accounting for approximately 2% F₀, can only elucidate a fraction of the results observed in this study.

Recent findings using chemical inhibitors to hinder or reduce active XB cycling indicate that non-XB viscoelastic structures, such as titin, contribute to enhanced force, work and power during and after SSCs (Fukutani and Herzog, 2020; Hessel et al., 2021; Tomalka et al., 2020, 2021). Titin serves as a crucial modulator of muscle contraction, contributing to stability, elasticity, alignment and even active force production – albeit in a supportive manner (Freundt and Linke, 2019; Linke, 2018; Nishikawa, 2020; Tomalka, 2023). Titin interacts with numerous muscle proteins (Linke, 2018) and facilitates force generation when the muscle is actively stretched. Titin-based force generation primarily depends on the interaction of specific titin spring elements with actin when Ca²⁺ is elevated (Bianco et al., 2007; Dutta et al., 2018; Mártonfalvi et al., 2014; Tahir et al., 2020). Specifically, potential binding partners to actin are the N2A (Dutta et al., 2018; Nishikawa et al., 2020; Tahir et al., 2020) and the PEVK region of titin (Bianco et al., 2007). This interaction leads to a reduction in the free titin spring length located in the proximal I-band region, resulting in an increase in titin-based force. Titin stiffness is widely recognized as a crucial factor in regulating muscle performance in skeletal muscle (for details, please refer to Linke, 2018, 2023). Thus, changes in titin stiffness can significantly impact the contractile properties of the muscle.

Our hypothesis, that greater and faster force re-development is expected after fast SSCs compared with slow SSCs, was supported by our findings. Specifically, fast SSCs at 85% v₀ revealed an increase in RF_dev by about 30% and attained this increase approximately 2 times faster compared with slow SSCs at 30% v₀. Considering titin's viscoelastic behaviour during SSCs, as reported in previous studies (Bianco et al., 2007; Chung et al., 2011; Herzog et al., 2014), it is known that work and power output in the shortening phase of SSCs increases with increasing velocity. This is primarily attributed to titin's ability to store and release elastic energy in a velocity-dependent manner (Freundt and Linke, 2019; Tomalka et al., 2021), thereby contributing to the SSC effect. Thus, it is likely that the viscoelastic behaviour of titin impacts force re-development in a velocity-dependent manner.

Although serving as indirect evidence, a recent study by Jeong and Nishikawa (2023) measured the stiffness-to-force ratio during force re-development after pure shortening contractions at different velocities in electrically stimulated mouse soleus muscles. The authors demonstrated that the slope of the relationship between muscle stiffness and force decreased with decreasing shortening velocity. Jeong and Nishikawa (2023) proposed that a tuneable viscoelastic non-XB structure could explain their results and suggested that titin, rather than weakly bound XBs, can account for the different slopes of the stiffness–force relationship among shortening velocities, including ‘no shortening’. They proposed that force depression can be explained by a shift in the equilibrium length of titin during isometric force re-development following active shortening. Specifically, a simulation-based study by Nishikawa et al. (2012) predicted a decrease in titin's equilibrium length with increasing XB force. A similar idea has been used in muscle modelling to predict rFD (Rode et al., 2009; Schappacher-Tilp et al., 2015; Tahir et al., 2020). Moreover, this alternative explanation is supported by previous evidence from a study on skinned soleus muscle fibres aiming to differentiate between XB and non-XB contributions to total muscle force during SSCs with varying velocities (Tomalka et al., 2021). In that study, it was suggested that the decreasing XB contribution with increasing stretch velocity is compensated for by an increasing non-XB contribution, presumably from titin (see fig. 5 of Tomalka et al., 2021).

Thus, following the line of arguments given above, the titin molecule exhibits several exclusive features, serving as both a ‘passive’ spring and as a generator of ‘active’ contractile work through the folding contractions of its immunoglobulin (Ig) domains (Rivas-Pardo et al., 2016). The mechanical properties of the titin spring can be finely adjusted by various mechanisms, including the binding of chaperones (Unger et al., 2017), by calcium ions (Ca²⁺) (Labeit et al., 2003), or actin (Bianco et al., 2007; Dutta et al., 2018; Li et al., 2018). These modifications, combined with the extensive interactions with other proteins in the muscle cell (Linke, 2018), give rise to a mechanosensory function of titin (Gautel, 2011; Linke, 2018), which is probably dynamically activated depending on the degree of change in titin's spring force (Freundt and Linke, 2019). Thus, the sophisticated interplay of these mechanisms probably enables the increased titin-based forces to persist even in the transition phase after SSCs. Consequently, this potential contribution may result in greater and faster rates of force re-development after SSCs compared with pure shortening contractions.

The significant interaction effect between condition and treatment on RF_dev and t_RFdev during force re-development could be attributed to two non-linear processes: the restoration of a steady-state XB distribution (e.g. Burton et al., 2006) and viscoelastic non-XB dynamics that contribute to force re-development (e.g. Tomalka et al., 2021) after SSCs and pure shortening. Currently, no muscle fibre model adequately accounts for both XB and non-XB dynamics to explain such interactions. Consequently, it remains uncertain whether the observed results point to unknown aspects of muscle contraction.

The forces (F_ss) observed after completing the ramp contractions at the end of the steady-state isometric phase exhibited no significant differences between the tested conditions (SSCs and pure shortening contractions) (P=0.115, Fig. 4A, lower plot). This observation aligns with experimental in vitro results reported by Fukutani and Herzog (2020). The authors of that study performed SSC and pure shortening contraction experiments in single skeletal muscle fibres from rabbit soleus muscles and reported no differences in steady-state forces 15 s after ramp perturbations. They argued that stretch-induced forces are abolished during the shortening phase in the SSC condition. This is either due to shortening-induced force depression (FD) (Edman et al., 1993; Maréchal and Plaghki, 1979) or a decrease in rFE during active muscle shortening (Fukutani and Herzog, 2018a). Consequently, rFE following SSCs may be completely eliminated (Fukutani and Herzog, 2020; Herzog and Leonard, 2000; Lee et al., 2001). The findings of this in vitro study on skinned muscle fibres were further supported by recent in vivo results on human thumb force development during SSCs and pure shortening contractions by Fortuna et al. (2017). They showed that slow SSC velocities (15 and 20 deg s⁻¹) resulted in similar steady-state isometric forces to those observed after pure shortening contractions. However, these authors also showed that rFD was reduced after SSCs compared with pure shortening contractions for faster shortening velocities (30 and 60 deg s⁻¹). Furthermore, these authors found that the magnitude of rFD did not vary across different SSCs conducted at various velocities.

However, there are conflicting findings indicating a significant difference in steady-state mean forces following pure shortening contractions and SSCs. For example, Joumaa et al. (2021a) observed a substantial increase in the isometric steady-state force following SSCs compared with pure shortening contractions in skinned muscle fibres from rabbit psoas. Furthermore, Tomalka et al. (2020) demonstrated a difference in isometric steady-state force F_ss following SSCs at 85% v₀ compared with pure shortening contractions at 85% v₀ in skinned single muscle fibres of the rat soleus muscle. Both studies attributed their findings to an increase in the proportion of attached XBs and titin stiffness. However, the results of this study did not show any differences in F_ss between SSCs and pure shortening contractions. The high variance within the groups is likely to be masking significant differences between SSCs and pure shortening contractions. Different study results in F_ss may be affected by numerous factors, including not only rFE but also force at the onset and end of shortening, mechanical work during shortening or the duration over which shortening occurs. Factors such as contraction modalities as well as methodological variables including stretch amplitude, contraction velocity, activation levels, experimental temperature, animal model studied and titin isoform may significantly affect the observed F_ss, contributing to variability among similar studies in the literature. As the structural and mechanistic complexity increases with the structural level of muscle, it becomes challenging to compare findings from in vitro animal studies with multi-joint muscle actions in vivo (Tomalka, 2023).

Similar to the control treatment, F_ss in the blebbistatin treatment exhibited no significant differences across all tested SSC velocities, or when comparing the SSC and pure shortening contraction conditions (P>0.05) (see also Tomalka et al., 2021) (Fig. 4B, lower plot). This finding aligns with previous work exploring the SSC effect in skinned muscle fibres, where the influence of the XB contribution is either eliminated or reduced by chemical XB inhibitors such as BDM (Fukutani and Herzog, 2020). Along with Fukutani and Herzog (2020), we conclude that XB cycling – in terms of active force production and strong XB binding to actin – is required for titin–actin interactions and, thus, to induce rFE by an increase in titin-based force and stiffness upon stretch.

Therefore, given the assumption that the titin–actin interaction depends on XB-based force, no differences in F_ss are expected between SSC and pure shortening contraction conditions – as demonstrated in the present study. This assumption finds support in previous studies (Fukutani and Herzog, 2018b; Leonard and Herzog, 2010; Powers et al., 2014), which demonstrated that XB inhibition (using BDM) leads to a decrease in the magnitude of rFE compared with strong XB binding. Consequently, impaired XB binding resulting from administering blebbistatin (or another chemical XB inhibitor) might hinder titin–actin binding, thereby diminishing the contribution of non-XB-based forces to force enhancement during stretching.

In this study, we investigated the force re-development of skinned soleus muscle fibres following SSC and pure shortening contraction muscle actions. We found a faster and greater force re-development after SSC compared with pure shortening contraction conditions and attributed this to the activation of a titin spring mechanism. We propose that titin undergoes stretching during eccentric contractions, and releases elastic energy during shortening, enhancing force production. We also suggest that titin plays a role in mechanosensory coupling, meaning that it transmits the stretch signal to the contractile apparatus and modulates the XB kinetics. Our study reinforces existing insights into XB and non-XB contributions to the SSC effect and provides further support for the role of titin in strain- and phosphorylation-dependent mechano-signalling. Thus, the interplay between XB-based and non-XB-based structures, especially the role of titin, emerges as a critical factor in SSCs, also visible in force re-development after SSCs.