Activation and stretch-induced passive force enhancement—are you pulling my chain? Focus on “Regulation of muscle force in the absence of actin-myosin-based cross-bridge interaction”

in novel experiments on single myofibrils prepared from rabbit psoas muscle, Dr. Walter Herzog and his graduate student Tim Leonard report an intriguing set of findings (16). Myofibrils were held at short sarcomere length, activated with a maximal level of calcium, and then stretched along the descending limb of the force-sarcomere length relation, to a final sarcomere length of ∼6.0 μm. Because stretch reduces overlap between thin and thick filaments, the widely accepted sliding-filament cross-bridge model of muscle contraction (8, 9) predicts that active force decreases during stretch and ceases to exist at the sarcomere length where overlap between thin and thick filaments reaches zero (∼4.0 μm). Instead, Leonard and Herzog report that force continues to rise, without any impediment posed by the absence of filament overlap. Because the myofibrils also develop passive force (produced in absence of calcium activation), the force measured during activation was corrected by subtracting passive force, but this still left an unexplained, stretch-induced large force (>200% of maximal force at optimal overlap). This finding of activation and stretch-induced force enhancement at sarcomere lengths far beyond filament overlap is highly unexpected.

Leonard and Herzog propose that the unexplained force beyond overlap is due to titin (16). Titin spans the half-sarcomere from Z disk to M band and has a large region located in the I band of the sarcomere that functions as a molecular spring that elongates in stretched sarcomeres, developing passive force in the process (5). Titin's spring region consists of two main sequence elements, the tandem Ig segments containing serially linked Ig domains and the PEVK sequence, rich in proline (P), glutamate (E), valine (V), and lysine residues (K) (5). Various mechanisms exist for tuning titin-based force, and any of these could be at play in the present study. A prominent mechanism is the expression of length variants of titin's spring elements, giving rise to titin isoforms that vary in stiffness, a mechanism that is used extensively during muscle development (4, 18, 19, 24). Adult cardiac muscle coexpresses compliant and stiff isoforms in the same sarcomere (21) with changes in the coexpression ratio in disease states (17, 25). Alternative splicing is a slow process that requires days-weeks to be completed and, needless to say, is not expected to take place during a brief stretch protocol.

There are also more rapid mechanisms to tune titin's stiffness, one of which is based on phosphorylation of titin's spring elements. Protein kinase A (PKA) and protein kinase G (PKG) have both been shown to reduce titin's stiffness by phosphorylating the N2B element (14, 27). However, the expression of the N2B element is restricted to cardiac titin isoforms, and thus PKA/PKG phosphorylation also fails to explain the present findings. A recently discovered signaling pathway consists of protein kinase C (PKC) phosphorylation of PEVK sequences; these sequences are expressed in all titin isoforms, and their phosphorylation increases passive force (6, 7). Thus if myofibrils were to contain structurally bound PKC that during stretch of activated myofibrils increases titin's phosphorylation state, increased force would ensue. However, considering that the PKC-induced force increase is on the order of ∼20% in cardiac muscle (and possibly less in skeletal muscle, because of skeletal muscle titin's long spring region), PKC phosphorylation is likely to be insufficient to account for the unexplained force found by Leonard and Herzog (a ∼200% increase at a sarcomere length of 6.0 μm). A second rapid mechanism for increasing titin-based force consists of stiffening of the PEVK element, due to calcium binding to glutamate-rich PEVK exons (3). This results in an increase in passive force on the order of 20–30% (11). Thus, although process might be occurring in the Leonard and Herzog study, it can account for only a relatively small fraction of the measured ∼200% force increase. In summary, it seems that none of the tuning mechanisms that are known to increase titin's force can account for the large unexplained force measured in activated myofibrils that are stretched.

Leonard and Herzog postulate a mechanism that is based on binding between titin's spring region and actin, with enhanced binding in activated myofibrils that are stretched. Such titin-actin binding would increase the degree of stretch of the spring segments that remain free of actin, giving rise to an increase in force. Earlier single-molecule manipulation studies have indeed shown that titin strongly interacts with actin (13), and, furthermore, that titin-actin interaction increases in the presence of high levels of calcium (12). However, these studies did not identify the site along the titin molecule that interacts with actin. Subsequent studies with recombinant titin fragments have failed to detect binding between the tandem Ig segments and actin (15, 26), although it is likely that some Ig domains unfold in sarcomeres longer than ∼4.0 μm (22), and, potentially, these unfolded domains are a site of interaction in extremely stretched myofibrils. Experimental evidence does exist for interactions between the PEVK region and actin (1, 15, 26), and thus these interactions could also be involved in the findings of Leonard and Herzog. An important aspect of their findings is that the level of unexplained force scales with the level of active force that exists when the stretch is initiated (16), and Leonard and Herzog suggest that active-force-induced actin filament strain (23) enhances titin-actin binding (there is precedent for strain-dependent protein-protein interaction; see Ref. 2). Furthermore, to explain that the high force levels are maintained beyond overlap, this enhanced binding must be maintained even though actin strain will have been reduced to zero. Thus, the titin-actin interaction site needs to be established, whether the interaction is enhanced by high levels of actin filament strain, and, once the interaction has been established, whether it can be maintained at low strain levels.

One must note that titin is not the only possible mechanism for this force enhancement. A possible mechanism for the unexplained force is that filament overlap remained at the extreme sarcomere lengths, for example because of thin and/or thick filament elongation or filament misalignment. The reversible extension of thin and thick filaments is <1% for a force that is similar to the maximal active force generated at optimal overlap (10, 23), and filament elongation is therefore unlikely to be sufficient to allow overlap to occur at the longest sarcomere lengths (∼6.0 μm) that were reached in the work. Misalignment of thick filaments in the A-band region or movement of the A band away from the middle of the sarcomere is a more likely explanation, because this is encountered in highly stretched muscle [Granzier HL, unpublished observations, and (22)]. This could result in remaining overlap and cross-bridge-based force at long sarcomere length. An observation that runs counter to this idea is that when Leonard and Herzog withdrew calcium from myofibrils stretched to extreme length, force did not fully relax but remained much higher than the passive force level (16). Thus, for residual filament overlap to explain the high force levels at the extreme sarcomere lengths, cross-bridge cycling under those conditions has to be independent of calcium. While unlikely, it is clear that additional work directed at eliminating cross-bridge-based forces as a contributor is an important control. A possible mechanical means consists of a sinusoidal stiffness analysis in myofibrils stretched to extreme lengths and testing whether dynamic stiffness follows the signature behavior of cross-bridge cycling or that of titin (20).

The titin field has so far made the implicit assumption that titin's properties are largely the same in passive and active muscle (except for the augmentation caused by calcium binding to E-rich PEVK domains, discussed above) and that they are independent of the active force level that exists when a stretch is initiated. The present work challenges these concepts and warrants important follow-up studies on activated myofibrils that are stretched to lengths that far exceed the “zero overlap length,” including 1) further structural and functional studies to test whether overlap and cross-bridge cycling are indeed absent and 2) measuring the fractional extension of titin's spring elements and testing whether some segments extend less and others extend more than in passively stretched myofibrils. In recent years, research on titin has revealed many unique and important properties of this giant protein, and the novel and stimulating findings on single myofibrils by Leonard and Herzog provide a great impetus to study the role of titin in active muscle.

GRANTS

This work was supported in part by funding from National Heart, Lung, and Blood Institute Grant HL-62881.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author.