One of the oldest methods for studying the mechanical properties of active muscle is the application of a rapid mechanical perturbation to an isometrically contracting muscle fiber. In a steady state contraction, myosin cross-bridges generate force asynchronously, which makes it difficult to distinguish the structural changes associated with the individual steps of the cross-bridge cycle because most experimental techniques typically produce only an average measurement over the ensemble. The application of a sudden length perturbation (e.g. 0.12 ms) to an isometric contraction can force attached cross-bridges to respond simultaneously 
When an isometrically contracting muscle at force T0
is allowed to shorten abruptly by a small amount and then held constant at the new, shorter length, tension decreases simultaneously to T1
due to the reduced load on elastic elements, which reside in the filaments and cross-bridges (). This response of the cross-bridges is referred to as phase 1. Force then initially recovers rapidly to T2
(phase 2), which is attributed to the active rotation of lever arms of attached cross-bridges, followed by a slower force recovery or even a slight reversal (phase 3) and a final, slower asymptotic recovery to a new isometric force level T4
(phase 4) 
. For very small step-releases of <5 nm/half sarcomere, the phase 2 recovery is complete during the first 2 ms, indicating that continuously attached cross-bridges undergo a power stroke to restore tension toward the isometric value 
. Quick releases ≥12 nm/half sarcomere drop the tension so far that the phase 2 recovery is absent and tension redevelops slowly through phases 3 and 4.
Muscle fibers become stiffer during a stretch, suggesting that force enhancement by stretch is partly related to the recruitment of additional elastic components 
. When an isometrically contracting muscle at tension T0
is suddenly step-stretched and held at the new length, the force increases during the stretch from T0
due to additional strain on the elastic elements, which consist of myosin heads and the filaments themselves (). Part of this strain is born by the filaments, which elongate slightly and, it is thought, second heads of actin-attached myosin, which bind the thin filament 
. Tension decreases from T1 to T2 during the next few milliseconds as the thick filaments return to their original length at the expense of the cross-bridges whose lever arms “back bend” further to accommodate the higher loading force imposed on them 
Myosin heads consist of a motor domain (MD), which has the catalytic and actin binding functions, and a lever arm that amplifies small structural changes within the MD to produce filament sliding. The lever arm consists of a small folded domain, dubbed the converter, a long α-helical peptide that binds the two light chains, an essential light chain, ELC, and a regulatory light chain, RLC. Attached fluorescent or paramagnetic probes can monitor separately the MD and the lever arm.
Fluorescence polarization measurements show that probes bound to the RLC of myosin in muscle fibers tilt in response to quick length changes during active contraction 
. Similar measurements on contracting/active muscle fibers labeled at Cys707 of the myosin MD show no change in probe orientation in response to length changes 
, suggesting that the myosin head contains an internal hinge, probably between the nucleotide-binding domain and the light chain region. Motions of the lever arm around this hinge rather than changes in MD orientation may underlie force generation. The detected orientations of probes attached to light chains for active isometric contraction showed a broad distribution of tilt angles. Based on stiffness information and filament compliance, it was estimated that the lever arms of 20% of the heads tilt in response to a length step during active contraction 
. However, how much of this motion is passively linked to filament sliding and how much to active force generation still remains a question.
X-ray diffraction of frog muscle has been used to probe with sub-ms precision the mechanically induced changes in average cross-bridge conformation that occur in synchrony with the induced force transients. Because of spacing differences, reflections on either the thick and thin filament can be used to identify changes occurring separately in either the lever arm or the myosin MD. Intensity of the M3 meridional X-ray reflection at 14.5 nm axial spacing, IM3
, arises from the arrangement of myosin heads on the thick filament 
and is used to report changes in tilt of cross-bridges during a force transient. A decrease in IM3
indicates tilting of myosin heads away from 90° and more parallel to the filament axis, which spreads the projection of their cross-bridge mass more uniformly along the filament axis 
. Conversely, intensity increases in the 37–38 nm, 5.9 and 5.1 nm layer lines can detect ordering of the MD interaction with the actin subunits.
When an active muscle fiber is rapidly shortened, IM3
decreases with about the same time course as the phase 2 rapid force recovery, but no detectable change in intensity occurs during the ~0.1 ms length step itself 
. Thus, the myosin head movements underlying the intensity decrease appear associated only with the slightly delayed active force generation early in phase 2, not with the length step and its elastically coupled, prompt force changes 
. In contrast, when a quick stretch is applied to active muscle fibers, part of the intensity decrease of the 14.5 nm reflection occurs during the stretch, which reports the distortion of myosin heads due to the instantaneous elasticity of muscle and a small and fast reversal of the working stroke 
. On the other hand, an X-ray study on muscle fibers utilizing a rapid temperature-jump to induce a tension rise found that the largest effect was found on the actin based layerlines rather than IM3
suggesting that a disordered to ordered transition of the MD on actin, rather than a simple lever arm tilt, was associated with the tension increase 
. Evidence for this disorder-to-order transition of the MD early in force generation had also been provided by earlier EPR studies 
X-ray diffraction has the advantage that the diffraction intensities can be recorded in real time from intact muscle fibers, but with the limitation that the measurements are an average of the ensemble of cross-bridges within the filament lattice. While shifts in this average can be detected, it is not straightforward to translate such shifts into changes of individual cross-bridges or the range and distribution of changes across the ensemble 
. X-ray techniques are sensitive to the mass of the entire head, and do not resolve which fragments of the head bend or tilt. For example, IM3
measures the ordered mass along the 14.5 nm axial periodicity, but this mass could be due to myosin heads ordered with respect to the actin subunits and thus generating force, or they could be disordered with respect to the actin subunits, as would be typical of weak myosin-thin filament interactions. Like X-ray diffraction, electron microscopy (EM) also measures the cross-bridge ensemble but with the distinct difference that individual cross-bridges can be visualized at the same time.
Insect flight muscle (IFM) possesses a highly ordered paracrystalline arrangement of myosin and actin filaments that provides detailed X-ray diffraction patterns 
and electron micrographs 
. Its filament arrangement is ideal for EM because the placement of an actin filament exactly midway between myosin filament pairs permits visualization of all the cross-bridges that attach to it within a section 25–30 nm thick. IFM also permits comparison of two activation pathways, one mechanically triggered by stretching partly activated muscle at pCa <6.0, referred to as stretch activation, the other an isometric contraction induced by saturating calcium at pCa <4.5, that is mechanically equivalent to the same state in vertebrate striated muscle.
IFM exhibits force transients, similar to those of vertebrate skeletal muscles 
. Step changes in sarcomere length of 100 µs duration can be imposed on fibers in Ca2+
-activating solution at the plateau (T0
) of isometric tension. The quick recovery rate increases in the isometric contraction transients, going from the largest stretch to the largest release, indicating that the cross-bridge kinetics of Lethocerus
IFM have a strain dependence similar to that in skeletal fibers from vertebrate muscle.
The fully Ca2+
-activated isometric contractions in IFM have been developed as an experimental model, called High Static Tension (HST), to indicate maximum active force with no stretch activation 
. Our recent EM study of isometric HST (iso-HST) utilized improved data collection, dual axis electron tomography and focused classification of the cross-bridge distribution within the 38.7 nm structural repeats 
. This work resolved the actin subunits on the thin filament thereby enabling the fit of an actin filament atomic model to the density independent of the presence of strong-binding myosin heads, in turn enabling the separate identification of both strong- and weak-binding attachments. Strongly bound cross-bridges were found only in the region exactly midway between successive troponin (Tn) complexes. This region is defined as the target zone 
. Two types of weak-binding attachments were found in and near the target zone, one set attached to actin, the other set contacting tropomyosin (TM). Yet another set of apparently weak attachments were observed contacting Tn. Quasiatomic models of strongly bound attachments show a 77° sweep of lever arms spanning a 12–13 nm power stroke. A plausible sequence of weak-binding attachments toward the strong-binding configuration suggested that the weak-to-strong transition involved primarily azimuthal movements of myosin which may explain temperature jump experiments on isometrically contracting muscle 
as well as providing a mechanism for myosin heads to cycle in place during active contraction 
Here we report on further investigations into the HST state of IFM using a 2 ms duration stretch (str-HST) or release (rls-HST) and their effect on both the structure and distribution of cross-bridges as they adjust to the mechanical perturbation. Changes relative to iso-HST include some new structures and a significant change in the distribution and frequency of previously observed structures. An increase in the number of two-headed attachments occurred after a quick stretch. Following a quick release the number of mask motifs in which myosin heads from successive 14.5 nm levels (crowns) contact a single target zone, as well as myosin heads contacting Tn (troponin bridges) increased. After a mechanical transient, there was also a dramatic change in the number and distribution of weak attachments in or near the target zone. Changes to the cross-bridge lever arm were comparatively small but consistent with the axial direction of the imposed length step.