3D motion of vesicles along microtubules helps them to circumvent obstacles in cells

Abstract

Vesicle transport is regulated at multiple levels, including regulation by scaffolding proteins and the cytoskeleton. This tight regulation is essential, since slowing or stoppage of transport can cause accumulation of obstacles and has been linked to diseases. Understanding the mechanisms by which transport is regulated as well as how motor proteins overcome obstacles can give important clues as to how these mechanisms break down in disease states. Here, we describe that the cytoskeleton architecture impacts transport in a vesicle-size-dependent manner, leading to pausing of vesicles larger than the separation of the microtubules. We further develop methods capable of following 3D transport processes in living cells. Using these methods, we show that vesicles move using two different modes along the microtubule. Off-axis motion, which leads to repositioning of the vesicle in 3D along the microtubule, correlates with the presence of steric obstacles and may help in circumventing them.

Keywords: 3D tracking; Motor protein; Super-resolution microscopy; Vesicle trafficking.

Fig. 1.

Vesicles pause at intersections when their size is comparable to microtubule spacing. (A) Single frames from a live-cell movie of LAMP2–mCherry-labeled endolysosomes. The red arrow indicates the tracked endolysosome, which is also shown in a super-resolution image of the same field of view (red square). (B) A two-color super-resolution image of microtubules (green) and endolysosomes (magenta) after fixation. The trajectory of the vesicle determined by single particle tracking is shown in white. Passive transport is shown with a yellow dotted circle. (C) 3D rendering of the endolysosome and (D) intersecting microtubules. The dashed rectangles show the upper (cyan) and lower microtubule (magenta). (E) Distribution of localizations along the z-axis in the super-resolution image of the endolysosome. The diameter of the vesicles was estimated from a pre-established cut-off (black arrow). (F) Distribution of localizations along the z-axis in the super-resolution image of microtubules in the region of the intersection. The blue and green plots correspond to the lower and upper microtubules, respectively. (G) Experimental data showing the pausing (red) and passing (green) probability as a function of the difference between endolysosome size and microtubule separation (n=51 events, n=11 cells). Modeling results are overlaid on top (black curve). (H) Upper plot, pausing probability as a function of vesicle size (n=66 vesicles, n=11 cells). Lower plot, cumulative distribution of microtubule separations (n=51 microtubules, n=11 cells). The color scale bar in A,C and D represents the z-position (between −400 nm, magenta, and 300 nm, red). Scale bars: 1 µm (A), 500 nm (B), 200 nm (C,D). Error bars (s.d.) in G and H are calculated by bootstrapping.

Fig. 2.

Endolysosomes containing rigid microspheres exhibit similar encounter and bypass behavior to that shown by native vesicles. (A) One frame from a live-cell movie of LysoTracker-labeled endolysosomes (magenta) and internalized 450 nm microspheres (green), and an overlay. 94% of motile microspheres were moving inside endolysosomes (n=6 cells, n=50 microspheres). Scale bar: 2 µm. (B) Encounter (left, n=71) and bypass (right, n=45) behavior of endolysosomes containing 450 nm microspheres (green) and native LAMP2–mCherry-labeled endolysosomes (dashed lines). Error bars (s.d.) are calculated by bootstrapping.

Fig. 3.

Internalized microspheres enable correlative 3D tracking and super-resolution microscopy. (A) One frame from a live-cell movie of LysoTracker-labeled endolysosomes (magenta) and internalized 260 nm microspheres (green), and an overlay. 81% of motile microspheres were moving inside vesicles (n=6 cells, n=43 microspheres). The arrows show two microspheres that are in different axial planes. (B) Localized 3D position of fluorescent microspheres embedded in a 3D gel matrix. The green circles and pink crosses indicate the localized position in the live-cell and STORM channels, respectively. (C) The raw z-shift (red) and the xy-shift (blue) after chromatic aberration correction between the localized positions of the microspheres in the two filter sets. The average z-shift was –14±5 nm below and 0±14 nm above the focal plane (n=698 microspheres). The average residual xy-shift after the chromatic aberration correction was 12±2 nm above and 13±1 nm below the focal plane (n=698 microspheres). The reported values represent mean±s.d. The black dashed line corresponds to the position of the coverglass. (D) Overlay of the images of the microspheres in the live-cell channel (green) and the super-resolution channel (magenta) after image registration. Microspheres that adsorbed to the coverslip were used as fiduciary markers, which appear to be colocalized in both channels (magenta squares). The zoomed-in images show xy- and z-projections of the localized center positions of two different reference microspheres on the coverslip after image registration. (E) Boxplot showing the residual registration error in x, y and z calculated as the root mean square distance between the reference microspheres after alignment in different experiments (12±7 nm, 7±2 nm and 16±8 nm in x, y and z, respectively, n=15 experiments). The reported values represent mean±s.d. The box represents the 25–75 percentiles. The solid line, the small square and the whiskers are the median, mean and standard deviation, respectively. Scale bars: 1 µm (A), 2 µm (large microsphere images in D), 100 nm (zoomed images in D).

Fig. 4.

Vesicles are transported in two separate modes along microtubules. (A) A 3D super-resolution image of microtubules. The highlighted area defines the microtubule segment where off-axis mode motion was observed. (B) A 3D plot placing the microsphere and endolysosome trajectory in the context of its microtubule path (green, not to scale). The smooth trajectory is shown in black as a 3D plot as well as in an xy- and yz-projection. (C) A 3D super-resolution image of microtubules. The highlighted area defines the microtubule segment where on-axis mode motion was observed. (D) A 3D plot placing the microsphere and endolysosome trajectory in the context of its microtubule path (green, not to scale). The smooth trajectory is shown in black as a 3D plot as well as in a xy- and yz-projection. (E–H) Boxplots comparing several motility parameters such as (E) average speed (on-axis mode: 0.4±0.2 µm/s, n=42 and off-axis mode: 0.6±0.4 µm/s, n=15, P=0.05), (F) power-law exponent of the mean square displacement (on-axis mode: 1.7±0.1, n=19 and off-axis mode: 1.8±0.1, n=12, P=0.34), (G) run length (on-axis mode: 0.4±0.4 µm, n=42 and off-axis mode: 1.0±0.5 µm, n=15, P=5×10⁻⁵) and (H) processivity (on-axis mode: 1.2±0.8 s, n=42 and off-axis mode: 2.0±1.0 s, n=15, P=0.005), for on-axis mode (green)- and off-axis mode (magenta)-type motion. The reported values represent mean±s.d. The box represents the 25–75 percentiles. The solid line, the small square and the whiskers are the median, mean and standard deviation, respectively. *P<0.05 (two-tailed two-sample t-test). The color scale bar represents the z-position (between −200 nm, magenta, and 350 nm, red) (A), time (between 0 s, blue, and 3.4 s, red) (B), z-position (between −400 nm, magenta and 400 nm, red) (C), time (between 0 s blue, and 1.1 s, red) (D). Scale bars: 250 nm.

Fig. 5.

Off-axis mode motion helps circumvent vesicular obstacles. (A) Image sequence showing a vesicle–vesicle interaction at different time points. The green and magenta dotted circles highlight the tracked microsphere and the interacting vesicle, respectively. Magenta shows LysoTracker and green shows the microsphere. (B) 2D trajectories of the microsphere-containing and interacting vesicles. (C) The center-to-center distance of the two vesicles before, during and after the encounter. (D) 3D super-resolution image of microtubules. The dotted red circle shows the place of interaction. The color scale bar represents z-position (between −200 nm, magenta, and 300 nm, red). (E) 3D plot placing the microsphere trajectory in the context of its microtubule path (green, not to scale). The smoothed trajectory (10-point sliding window average) and its corresponding xy- and yz-projections are shown in black. For B,C and E, active and passive transport phases in the raw trajectory are shown in blue and red, respectively. Scale bars: 500 nm.

Fig. 6.

Off-axis mode motion helps circumvent intersecting microtubule obstacles. (A,F) A 3D super-resolution image of microtubules. The highlighted area shows the pause-switch (A) and pause-pass (F) events. The arrows indicate the direction of motion. (B,G) 2D trajectory of the tracked endolysosome. Active and passive transport phases are shown in blue and red, respectively. (C,H) 3D projection of the microtubules at the intersection. The green and magenta boxes in (C) and green and yellow boxes in (H) show the upper and lower microtubule, respectively. (D,I) z-position of the raw (gray line) and 10-point sliding window average (black line) trajectory as a function of time. The green and purple lines in (D) and orange and green lines in (I) show the z-position of the upper and lower microtubule, respectively. The red arrows are an estimate of the radius of the vesicle from the center position at the trajectory. (E,J) Cartoon representation of the pause-switch (E) and pause-pass (J) events. (K) Pie charts showing the percentage of on-axis mode (green) and off-axis mode (magenta) events for endolysosomes that directly passed an intersection without pausing (direct pass, n=28) and those that paused before passing (pause-pass, n=17). (L) Boxplots for the speeds (upper) for direct pass (on-axis mode: 0.7±0.3 µm/s, n=19 and off-axis mode: 0.6±0.3 µm/s, n=9) and pause-pass (on-axis mode: 0.3±0.1 µm/s, n=7 and off-axis mode: 0.4±0.2 µm/s, n=10), P=0.005 between pass off-axis mode and pause-pass on-axis mode and P=0.002 between pass on-axis mode and pause-pass on-axis mode; and the power law exponent (lower) for direct pass (on-axis mode: 1.7±0.2, n=17 and off-axis mode: 1.9±0.1, n=8) and pause-pass (on-axis mode: 1.4±0.4, n=6 and off-axis mode: 1.7±0.3, n=6), P=0.009 between pass off-axis mode and pause-pass on-axis mode. The reported values represent mean±s.d. The box represents the 25–75 percentiles. The solid line, the small square and the whiskers are the median, mean and standard deviation, respectively. The P-values are calculated for a two-tailed two-sample t-test and (*P<0.05). The color scale bar represents the z-position (between −100 nm, magenta, and 350 nm, red, in A; −400 nm, magenta, and 400 nm, red, in F). Scale bars: 250 nm (A,C), 500 nm (F), 200 nm (H).