Molecular basis of passive stress relaxation in human soleus fibers: assessment of the role of immunoglobulin-like domain unfolding

Abstract

Titin (also known as connectin) is the main determinant of physiological levels of passive muscle force. This force is generated by the extensible I-band region of the molecule, which is constructed of the PEVK domain and tandem-immunoglobulin segments comprising serially linked immunoglobulin (Ig)-like domains. It is unresolved whether under physiological conditions Ig domains remain folded and act as "spacers" that set the sarcomere length at which the PEVK extends or whether they contribute to titin's extensibility by unfolding. Here we focused on whether Ig unfolding plays a prominent role in stress relaxation (decay of force at constant length after stretch) using mechanical and immunolabeling studies on relaxed human soleus muscle fibers and Monte Carlo simulations. Simulation experiments using Ig-domain unfolding parameters obtained in earlier single-molecule atomic force microscopy experiments recover the phenomenology of stress relaxation and predict large-scale unfolding in titin during an extended period (> approximately 20 min) of relaxation. By contrast, immunolabeling experiments failed to demonstrate large-scale unfolding. Thus, under physiological conditions in relaxed human soleus fibers, Ig domains are more stable than predicted by atomic force microscopy experiments. Ig-domain unfolding did not become more pronounced after gelsolin treatment, suggesting that the thin filament is unlikely to significantly contribute to the mechanical stability of the domains. We conclude that in human soleus fibers, Ig unfolding cannot solely explain stress relaxation.

FIGURE 1

Measured stress relaxation in passive soleus fibers. (A) Protocol: fiber was stretched and then held for 1 h followed by a release. Top, sarcomere length; bottom, passive tension. The protocol was repeated after KCl/KI extraction (see Materials and Methods). The KCl/KI-sensitive tensions of fibers that had been stretched to different SLs are shown in B and the KCl/KI-insensitive tensions in C. (Mean ± SD of six fibers. Each fiber contributed data to both B and C. Data in B and C were obtained by stretching the fiber from the same initial SL of ∼2.1 μm. For simplicity's sake only selected time points (indicated in A, bottom) are shown in B and C; see text for additional details.)

FIGURE 2

Measured titin-based force versus time at a range of SLs. Titin-based tension was assumed to be equal to the KCl/KI-sensitive tension (same data as in Fig. 1 B), and measured tensions (in mN/mm²) were converted to force per titin molecule (see Materials and Methods). (Inset) Data converted to force-SL relations 1, 4, 16, and 60 min after completion of stretch. (Mean results from six fibers are shown. Numbers at bottom right correspond to sarcomere lengths in μm.)

FIGURE 3

(A) Measured passive tension (black), simulated passive tension (gray), and simulated Ig domain unfolding (gray, bottom) during stretch-hold protocol. During 10-s hold, simulation reveals unfolding of four Ig domains and a predicted decay of passive tension that approximates well the measured tension (measured after stretch from sarcomere length 2.1 μm to 4.0 μm). Simulation assumes α₀ 5 × 10⁻⁴ s⁻¹. The broken lines indicate predicted force (top) and predicted unfolding (bottom) for α₀ of 1 × 10⁻⁷ s⁻¹. (B) Simulated passive force of single titin molecule in sarcomere during 1-h hold period. The molecule was stretched to a peak force of 56 pN (chosen because this is the measured peak force when stretched to SL 4.0 μm, see Fig. 2). During the hold, unfolding of Ig domains is predicted to continue until a maximum of 70 domains are unfolded. Passive force is predicted to decay until a steady value of ∼5 pN is reached. (Simulations assume for unfolding X_u 0.28 nm; α₀ 5 × 10⁻⁴ s⁻¹; polling interval 0.01 s; stretch velocity 100 nm/s. For refolding characteristics and additional details, see Materials and Methods. Upper black traces in each figure represent the sarcomere length changes during the protocol.)

FIGURE 4

Immunoelectron microscopy of titin. (A) Top, sequence of I-band region of soleus titin (Labeit and Kolmerer, 1995). Indicated are the binding sites of T12, N2A, 514, I84-86, MIR, and Ti102 antibodies. Bottom, sarcomeres simultaneously labeled with four antibodies. Fibers were stretched and then held for 0 min (top) or 64 h (bottom) followed by fixing and labeling. (B–D) Additional examples of electron micrographs of fibers labeled with anti-titin antibodies used in this work. (Sarcomere lengths: A, 4.1 μm; B, 3.6 μm; C, 3.5 μm; and D, 4.2 μm.)

FIGURE 5

(A) Mid-epitope to mid-Z-line distance of T12, N2A, 514, and Ti102 epitopes in fibers stretched and held for 0 min (green symbols) and 64 h (red symbols). Length of proximal and distal tandem-Ig segment: distance from T12 to N2A and from 514 to Ti102, respectively. Length of PEVK: distance from N2A to 514 eptiopes. (B–D) Lengths of tandem-Ig and PEVK segments versus sarcomere length of fibers stretched and held for 0 min and 64 h. Insets show mean and SD of results binned in 0.1 μm SL bins. Asterisks denote statistically significant differences (P < 0.05 ).

FIGURE 6

Measured and predicted lengths of tandem-Ig (A) and PEVK (B) segments. Measured: green symbols, fibers stretched and then held for 0 min; red symbols held for 64 h. Broken lines are predicted extensions assuming that Ig unfolding is absent and solid lines assuming that a total of 70 domains are unfolded. We assumed that of the 70 unfolded domains, 50 are in the proximal and 20 in the distal segment (the difference reflects the proportion of the total number of domains contained in the segments). (Prediction based on serially linked wormlike chains model in which folded and unfolded tandem-Ig subsegments as well as the PEVK are represented. For details, see Materials and Methods.)

FIGURE 7

Behavior of N2A and T12 epitopes in extremely long sarcomeres. (Top) Examples of labeled sarcomeres (only I-band regions are shown). (Bottom) Scattergram of results. At SLs > ∼4.5 μm the proximal tandem-Ig segment greatly extends. Inset in bottom graph shows length of PEVK (distance between N2A and I84-86 epitopes). Note that extension of proximal tandem-Ig segment at long SLs coincides with a near constant PEVK segment length.

FIGURE 8

Effect of thin filament removal with gelsolin on protein composition (A) and passive tension (B). The gel of single fibers in A reveals that gelsolin largely removed actin and nebulin without major effect on titin and myosin heavy chain (MHC); mechanical measurements in B revealed that thin filament extraction slightly elevated passive tension.

FIGURE 9

Effect of thin filament removal on length of proximal tandem Ig (T12-N2A distance). Top shows example of thin-filament extracted and titin-labeled sarcomere. Bottom right inset shows mean and SD of results binned in 0.1 μm SL bins (gray bars, control; open bars, gelsolin treated). Asterisks denote statistically significant differences (P < 0.05).

FIGURE 10

Model in which unfolding of Ig domains and breakage of random bonds in titin's PEVK region give rise to stress relaxation. Results of Monte Carlo simulation (for details see Materials and Methods) in which molecule was stretched to a maximal force of 56 pN and the number of broken bonds and unfolded domains was calculated during a 1-h hold period. Top left shows a schematic of a bond, Bi. The bond shields part of the chain from external force and removes it from contributing to the contour length. Upon bond breakage, the contour length increases, fractional extension is reduced, and force falls. (A) After stretching to high force and then holding the molecule at constant length, the predicted number of broken bonds increases initially fast and then more gradually. The number of unfolded Ig domains (right scale) gradually increases to ∼4 domains at the end of the 1-h hold phase. (B) Simulated and measured force (SL 4.0 μm; same data as in Fig. 3). Note that simulated forces closely follow measured forces. (Shown are the mean ± SD values of 20 simulation runs. For simulation parameters and other details, see Materials and Methods.)