Control of myosin-I force sensing by alternative splicing

Abstract

Myosin-Is are molecular motors that link cellular membranes to the actin cytoskeleton, where they play roles in mechano-signal transduction and membrane trafficking. Some myosin-Is are proposed to act as force sensors, dynamically modulating their motile properties in response to changes in tension. In this study, we examined force sensing by the widely expressed myosin-I isoform, myo1b, which is alternatively spliced in its light chain binding domain (LCBD), yielding proteins with lever arms of different lengths. We found the actin-detachment kinetics of the splice isoforms to be extraordinarily tension-sensitive, with the magnitude of tension sensitivity to be related to LCBD splicing. Thus, in addition to regulating step-size, motility rates, and myosin activation, the LCBD is a key regulator of force sensing. We also found that myo1b is substantially more tension-sensitive than other myosins with similar length lever arms, indicating that different myosins have different tension-sensitive transitions.

Fig. 1.

Expressed myo1b protein constructs. Diagrams show the relationship of the myo1b motor domain (large rectangles) to the IQ motifs (smaller numbered rectangles). The IQ motifs that are alternatively spliced are colored, and the inset shows the amino acid sequences of the splice locations within IQ motifs 4–6. The small black rectangle corresponds to the 15 amino acid sequence that is biotinylated via biotin ligase.

Fig. 2.

Step and substep averages and distributions of myo1b splice isoforms. (A) Histograms of step and substep sizes of unitary interactions acquired in the presence of 1.0 μM ATP. The size of the first substeps of the individual interactions were calculated by averaging 100 points (50 ms of data), 50 ms after attachment. Total step sizes from each interaction were determined by averaging 100 points, 50 ms before detachment. The difference between the two measurements is reported as the second substep. Distributions were fit to a Gaussian function where μ and σ report the means and standard deviations of the Gaussian distributions, respectively. Standard errors of the Gaussian fits are reported in Table S1. Interactions were acquired from multiple coverslips as follows: myo1b^a (4), myo1b^b (6), myo1b^c (6), myo1b^IQ (5). (B) Ensemble averages showing the average step and substep sizes. Total step (black arrows), first substep (green arrows), and second substep (blue arrows) sizes obtained from the ensemble averages are reported. The uncertainty is the standard deviation of the 100 averaged data points (50 ms) from the ensemble trace.

Fig. 3.

Summary of step sizes and distance parameters of myo1b splice isoforms. Step sizes (circles) and distance parameters (squares) plotted as a function of the number of IQ motifs in the LCBD. Total (closed symbols) and second substep (open symbols) sizes are obtained from the fits to the Gaussian distributions as described in Fig. 2. The same protein preparations were used to acquire step sizes and distance parameters. The red line is a linear fit to the d_det points, giving a slope of 2.9 nm/IQ and a y-intercept of 0.054 nm (r = 0.985). Step and substep sizes are reported as the mean and standard errors of the mean, and d_det values are best fits showing 90% confidence limits.

Fig. 4.

Force dependence of myo1b splice isoform kinetics. (Top Row) Representative time traces showing force during unitary actomyo1b attachments acquired with the isometric clamp in the presence of 50 μM ATP. (Center Row) Attachment durations as a function of the average force during the attachment. Interactions were acquired from multiple coverslips as follows: myo1b^a (5), myo1b^b (6), myo1b^c (6), myo1b^IQ (4). (Bottom Row) Force dependence of the actin-detachment rates (k_det) as calculated using the parameters obtained by fitting the attachment durations by maximum likelihood estimation. The best fit values are k_g0 = 0.51( ± 0.11) s^-1, k_i = 0.011(+0.0025/-0.0017) s^-1, and d_det values are given in the figure. Points are the inverse averages of 10 consecutive attachment durations. The shaded areas show the 90% confidence limit based on the uncertainties.

Fig. 5.

Force dependence of attachment lifetimes of the myo1b splice isoforms. Attachment lifetimes (1/k_det) of the four myo1b protein constructs were calculated using the values obtained from the MLE fit of the data in Fig. 4 to Eq. 1. The inset numbers indicate explicit values of attachment lifetimes at 1 pN opposing force (gray line).