Stiffness of cargo-motor linkage tunes Myosin VI motility and response to load

Abstract

We examine the effect of cargo-motor linkage stiffness on the mechanobiological properties of the molecular motor Myosin VI. We use the programmability of DNA nanostructures to modulate cargo-motor linkage stiffness and combine it with high precision optical trapping measurements to measure the effect of linkage stiffness on motile properties of Myosin VI. Our results reveal that a stiff cargo-motor linkage leads to shorter step sizes and load-induced anchoring of Myosin VI, while a flexible linkage results in longer steps with frequent detachments from the actin filament under load. Our findings suggest a novel regulatory mechanism for tuning the dual cellular roles of anchor and transporter ascribed to Myosin VI.

Graphical Abstract

Bustamante et al. describe the cell as a “clockwork” made up of multiple interconnected mechanical elements rather than a “reaction vessel”, the analogy which was prevalent among biochemists more than half a century ago.¹ This emerging paradigm in cell biology emphasizes the importance of mechanical communication in cellular functions. Cellular processes such as cell adhesion, organelle transport, and migration rely on mechanical communication in macromolecular clusters. Effective force transmission in such ensembles depends on the mechanical properties of the linkage between biomolecules. In this study, we focus on the impact of linkage stiffness on the mechanical communication between groups of molecular motors driving cargo transport along cytoskeletal filaments.

Cytoskeletal motors link to their cellular cargoes through diverse mechanisms ranging from adaptor protein intermediates to lipid molecules that directly bind the motor.^2,3 The diversity in cargo-motor attachments translates into distinct linkage stiffness regimes in cellular membrane tråc. Both experimental^4–6 as well as theoretical studies^7–10 have shown that linkage stiffness alters motile properties, such as velocity and run length of cytoskeletal motors. However, the mechanistic basis of these changes, in terms of stepping behavior and force generation is not well understood and forms the focus of this study.

To investigate the impact of linkage stiffness on stepping behavior and force generation of individual motors in ensembles, we need experimental systems wherein the motor number and linkage stiffness can be controlled in concert with high precision force spectroscopy techniques. We have previously demonstrated¹¹ that DNA origami scaffolds offer programmability of motor number and linkage stiffness to impact motile properties. Here, we use optical trapping to measure the intersection of linkage stiffness and force sensitivity of Myosin VI. We have adapted the system to couple DNA origami to micron-sized spherical beads that are subsequently used as probes in an optical trapping setup. Our assay provides resolution of piconewton forces on the motor coupled with measurements of motor movement with nanometer precision, while varying cargo-motor linkage stiffness.

The protocol for coupling DNA origami to beads and oligo-labeled Myosin VI is shown in Figure 1A. Briefly, origami is coupled to avidin-coated beads via a biotinylated DNA strand whereas the oligo-labeled Myosin VI is linked to the bead-origami system via complementary base pairing.¹² To modulate cargo-motor linkage stiffness, the length of the oligo coupled to Myosin VI is varied to achieve a stiff or flexible linkage of the motor to the cargo (Figure 1A, inset). To quantify the change in cargo-motor linkage stiffness, the motor linkage stiffness is measured through a clamping of the bead variance that occurs when a motor engages with the actin filament and exerts force (Figure 1B). The optical trap and the motor linkage can be modeled as springs attached in parallel to the bead (Figure 1C). Thus, the stiffness measured during a motor binding event is a sum of the stiffness of the trap and that of the motor linkage. The stiffness of the linkage can be estimated from the difference between stiffness measured with the motor bound to the actin filament and the trap stiffness measured with the unbound bead. Moreover, the motor linkage is a combination of the oligo spring attaching the motor to the origami and the inherent motor spring connected in series to each other (Figure 1C, inset). Since Myosin VI is common for both a stiff and a flexible linkage, the change in the linkage stiffness is essentially a function of the oligo spring stiffness. Quantification of the linkage stiffness distributions measured for a stiff versus flexible linkage is shown in Figure 1D. We observe a significant decrease (P < 0.001) in cargo-motor linkage stiffness when Myosin VI is coupled to DNA origami through a flexible oligo link (Mean = 0.007 ± 0.003 pN/nm) when compared to a stiff oligo linkage (Mean = 0.012 ± 0.004 pN/nm). Since the applied load could have an impact on the stiffness of the linkage, we measured changes in linkage stiffness as a function of load (Figure 1E and Figure 1F). Indeed, we observed a load-dependent stiffening of the entire linkage at higher loads for both stiff and flexible linkage geometries. Given that the majority of events (~75 %) occur at low loads (< 0.5 pN), the use of single/double stranded DNA linkages to modulate stiffness is appropriate for motor ensembles under low load conditions.

Figure 1.

(A) Schematic illustrating the protocol for coupling DNA origami scaffolds and oligo-labeled Myosin VI to beads for optical trapping (top) and the rationale for changing linkage stiffness via the length of the ss-DNA oligo coupled to Myosin VI (inset,bottom). (B) Representative position v/s time plots of Myosin VI force generation events in an optical trap when connected via a stiff versus flexible linkage. Also shown is the standard deviation (S.D.) in position to illustrate the clamping of bead variance (double-headed arrow) that occurs when Myosin VI binds to an actin filament. (C) Model of cargo-linkage and optical trap coupled to the bead as elastic springs in parallel. The inset shows the cargo linkage spring as being a combination of the oligo and Myosin VI spring connected in series to each other. (D) Distribution of linkage stiffness for a stiff v/s flexible linkage. Both distributions have been fit to a single-peak Gaussian with R² = 0.87 for stiff and 0.99 for a flexible linkage. n = 97 and 89 variance clamping events pooled from three different protein preparations of Myosin VI were used for linkage stiffness calculation for the stiff and flexible linkage respectively. Z-test was used to estimate the statistical significance of the difference between mean stiffness with P < 0.001. Error values are reported as standard deviation. (E) and (F) Linkage stiffness plotted as a function of load for stiff and flexible linkage. Error bars are standard deviation. The number in parentheses below each data point indicates the number of events used to generate the values of that data point. Significance between the linkage stiffness at the lowest load and highest load points was estimated using a one way ANOVA with a post hoc Tukey’s test (P < 0.05). The overall weighted average load experienced by a stiff linkage (0.32 pN) was found to be higher than that of a flexible linkage (0.26 pN).

After quantifying the change in linkage stiffness using a flexible oligo link, we investigated the effect of linkage stiffness on motility of Myosin VI attached to DNA origami. The unloaded motility of stiff vs. flexible linkage was assessed as shown in Figure 2B. Single-molecule motility of Myosin VI without any cargo attached (free motor) is used as a control to assess the effect of cargo binding and cargo size on motility. We observe that the velocities for Myosin VI bound to DNA origami (92 ± 11 nm/s for stiff linkage and 107 ± 5 nm/s for flexible linkage) are lower as compared to velocities for Myosin VI without DNA origami (297 ± 5 nm/s). In contrast, the run lengths for Myosin VI bound to DNA origami (457 ± 23 nm for stiff linkage and 323 ± 18 nm for flexible linkage) are higher as compared to run lengths of Myosin VI without DNA origami (191 ± 5 nm). The physical basis of the differences observed in motile properties of the free versus origami-bound Myosin VI is unclear. Nonetheless, the velocities with flexible and stiff linkages were not found to be significantly different. In contrast, there was a significant decrease (P < 0.001) in run length observed with a flexible linkage (Mean = 323 ± 18 nm) as compared to a stiff linkage (Mean = 457 ± 23 nm) (Figure 2C). The increase in run length with a stiff linkage could result from an effect of linkage stiffness on the load-dependent motor properties of Myosin VI such as step size and detachment rate. Thus, we next measured the impact of linkage stiffness on load-dependent motor properties of Myosin VI using optical trapping (Figure 3A).

Figure 2.

(A) Schematic for the motility assay when origami is attached to the Myosin VI via a stiff linkage (left), via a flexible linkage (middle), and in case of no origami, i.e. free motor (right). (B) Average velocities (nm/s) for stiff linkage, flexible linkage and free motor case. (C) Average run lengths (nm) for stiff linkage, flexible linkage and free motor case. Student’s t-test was used to estimate the significance of the difference between mean values of velocity and run length. N > 1000 motility events across three different protein preparations were used to get mean values of velocity and run length. Error values are reported as standard error of mean.

Figure 3.

(A) Schematic illustrating the optical trapping geometries with stiff and flexible cargo-motor linkages for Myosin VI. (B) and (C) Representative force generation events with stiff and flexible linkage for Myosin VI respectively. As illustrated, stiff linkage is characterized by force generation events which last for a longer duration of time as compared to flexible linkage events. (D) Distribution of stall forces for Myosin VI with stiff (black) and flexible (red) cargo-motor linkage. R² = 0.99 and 0.87 for the Gaussian fit of stiff and flexible linkage respectively (P < 0.01). (E) Stall duration distributions for stiff versus flexible linkage fit to a single exponential decay with mean values indicated in the plot. R² = 0.98 and 0.99 for the single exponential fit of stiff and flexible linkage respectively. Flexible linkage exhibits significantly shorter (P < 0.001) stall duration as compared to a stiff linkage. (F) Correlation of stall force against stall duration as a measure of the ability of Myosin VI to sustain load. For a stiff linkage, there is a linear improvement in the stall duration with increasing stall force which does not happen for a flexible linkage of Myosin VI (smaller slope). Slope of linear fit was 3.59 s/pN for stiff and 0.41 s/pN for flexible linkage. n = 192 and 136 force generation events pooled from three different protein preparations of Myosin VI were used for stall force and stall duration calculation for the stiff and flexible linkage respectively. Error bars are standard deviation. Z-test was used for statistical significance in differences.

To investigate the effect of linkage stiffness on Myosin VI behavior under load, we first quantified the force at which the motor stalls (stall force) and the time for which the motor is able to sustain the stall load (stall duration). A stall event is defined as the slowing down of motor velocity to <10 nm/s during a force generation event in the optical trap. Stall duration was the time duration from the beginning of a stall event to the point at which the motor detached and the bead fell back to the center of the trap. The force-generation profile of Myosin VI in Figure 3B and Figure 3C shows that Myosin VI connected via a stiff linkage exhibits longer stalling events, whereas Myosin VI connected via a flexible linkage displays multiple shorter stall events in a similar time window. The stalling profile of Myosin VI was quantified in terms of stall force and stall duration. The stall force distributions shown in Figure 3D indicate that there is an increase (P < 0.01) in the stall force for Myosin VI connected with a stiff linkage (Mean = 0.37 ± 0.19 pN) compared to that of a flexible linkage (Mean = 0.31 ± 0.15 pN). Also, the stall duration (see Figure 3E) was significantly higher (P < 0.001) for Myosin VI connected with a stiff linkage (Mean = 4.6 ± 0.5 s) compared to that of a flexible linkage (Mean = 1.6±0.2 s). Furthermore, the stall duration showed a higher rate of increase with increasing stall force (see Figure 3F) for Myosin VI when it is attached via a stiff linkage (Slope= 3.59 s/pN) as compared to Myosin VI connected via a flexible linkage (Slope= 0.41 s/pN). This faster rise in stall duration with increasing load for Myosin VI attached through a stiff linkage parallels the load-induced anchoring of Myosin VI seen in earlier optical trap measurements.^13,14

To understand the effect of linkage flexibility on stepping behavior of Myosin VI, we analyzed the optical trapping data to obtain the step size and time interval between the subsequent steps (dwell time) of Myosin VI connected via stiff and flexible linkage to the beads (Figures 4A and 4B). Step sizes distributions (Figure 4C) reveal a significant increase (P < 0.001) in step size for Myosin VI connected via a flexible linkage (Mean = 28 ± 13 nm) as compared to a stiff linkage (Mean = 19 ± 10 nm). It was also observed that dwell time between steps changed considerably with linkage stiffness (Figure 4D). The mean dwell time for Myosin connected via a stiff linkage (Mean = 0.9 ± 0.5 s) was found to be significantly higher (P < 0.001) than that for a flexible linkage (Mean = 0.3 ± 0.1 s). Since both step size and dwell time are shown to be load-dependent for other processive motors,^15,16 we next measured step size and dwell time as a function of applied load for both stiff and flexible linkage (Figure 4E–H). We observed no significant trend in step size and dwell time of Myosin VI with increasing load for both stiff as well as flexible linkages. It is important to note here that our measurements are performed at physiological ATP concentrations (2 mM ATP) and dwell time has previously been shown to be dependent on ATP concentration.¹⁴ Therefore, our observations here are consistent with previous literature of Myosin VI stepping where neither the step size nor the dwell time of Myosin VI were observed to have a load dependence at physiological ATP concentrations and loads less than 2 pN.¹⁴

Figure 4.

(A) and (B) Representative bead position data showing stepping profile for Myosin VI attached with a stiff and a flexible linkage. (C) Distribution of step sizes of Myosin VI with stiff (black) and flexible (red) cargo-motor linkage. The distributions are fit to a single-peak Gaussian with R² = 0.93 and 0.88 for stiff and flexible linkage respectively. There is an increase in mean step size of Myosin VI connected with a flexible linkage (P < 0.001).(D) Distribution of dwell times of Myosin VI with stiff (black) and flexible (red) cargo-motor linkage. The dwell time distributions are fitted to a single exponential with R² = 0.99 and 0.96 for stiff and flexible linkage respectively. Myosin VI attached with a stiff linkage shows a significantly higher mean dwell time (P < 0.001) as compared to that of a flexible linkage. n = 124 and 89 stepping events pooled from three different protein preparations of Myosin VI were used for step size and dwell time calculation for the stiff and flexible linkage respectively. (E) and (F) Step size as a function of load for stiff and flexible linkage. Error values are standard deviation. Significance between the linkage stiffness at the lowest load and highest load points was estimated using a one way ANOVA with a post hoc Tukey’s test. (G) and (H) Dwell time between steps as a function of load for stiff and flexible linkage. Error bars are standard deviation. Significance between the linkage stiffness at the lowest load and highest load points was estimated using a one way ANOVA with a post hoc Tukey’s test. Z-test was used for statistical significance in differences. Error bars are standard deviation.

In this study, we have investigated the impact of cargo-motor linkage stiffness on Myosin VI transport. A stiff cargo-Myosin VI linkage results in load-induced anchoring of Myosin VI characterized by higher stall duration, longer dwell time between steps and shorter step sizes. In contrast, a flexible cargo-Myosin VI linkage leads to frequent detachments under load resulting in shorter stall duration and dwell time between steps. Interestingly, however, a flexible linkage leads to longer step sizes for Myosin VI presumably through extension of the effective Myosin VI lever arm.¹⁷ Theoretical modeling of motor transport has previously predicted an inverse relationship between dwell time and load-dependent detachment rate of the motor from the filament.¹⁸ Also, Kunwar et al. have shown through Monte Carlo simulations of motor transport that a decreased linkage stiffness leads to impaired collective motor function due to stretching of the linkage and higher detachment rates.¹⁹ Our results provide experimental validation for such a role of linkage stiffness in regulating the load-dependent behavior of the motor suggested by the aforementioned theoretical models. A flexible linkage provides greater Brownian diffusion of the two heads relative to the cargo scaffold ($\Delta x=\sqrt{K_{b}T}/\sqrt{K_{\mathit{\text{linkage}}}}$; Figure 4C) stochastically enabling a larger step size. In contrast, the flexible linkage yields smaller resistive loads (Figure 1E,F) for the same distance traveled by the myosin dimer. Consequently, the load-induced anchoring phenomenon that has been well-documented for myosin VI^14,20 is not engaged as frequently in the presence of a flexible linkage.

Myosin VI has been ascribed dual functionality as a transporter and a tether function inside the cell.²¹ It acts as a transporter to processively move organelles such as endosomes and secretory vesicles, whereas it functions as a molecular tether to maintain the structure of stereocilia in inner ear hair cells.²² Myosin VI was shown to display load-induced anchoring at higher loads in optical trapping experiments,^13,14 which likely forms the mechanistic basis for its dual functions. However, previous studies did not explore a role for linkage stiffness in the load-induced anchoring behavior of Myosin VI. Our results here highlight the cargo-Myosin VI linkage stiffness as an additional layer of regulation for load-induced anchoring of Myosin VI whereby both transporter and anchor roles can be tuned by the mechanical properties of the motor-cargo attachment.

In conclusion, the methodology described in this study combines DNA origami with optical trapping and opens up avenues of studying the effect of cargo-motor linkage stiffness on motor transport. Enabled by the methodology, our results reveal that the mechanical properties of cargo-motor linkage play a key role in regulating cellular context-dependent function of Myosin VI.