Confinement Geometry Tunes Fascin-Actin Bundle Structures and Consequently the Shape of a Lipid Bilayer Vesicle

Abstract

Depending on the physical and biochemical properties of actin-binding proteins, actin networks form different types of membrane protrusions at the cell periphery. Actin crosslinkers, which facilitate the interaction of actin filaments with one another, are pivotal in determining the mechanical properties and protrusive behavior of actin networks. Short crosslinkers such as fascin bundle F-actin to form rigid spiky filopodial protrusions. By encapsulation of fascin and actin in giant unilamellar vesicles (GUVs), we show that fascin-actin bundles cause various GUV shape changes by forming bundle networks or straight single bundles depending on GUV size and fascin concentration. We also show that the presence of a long crosslinker, α-actinin, impacts fascin-induced GUV shape changes and significantly impairs the formation of filopodia-like protrusions. Actin bundle-induced GUV shape changes are confirmed by light-induced disassembly of actin bundles leading to the reversal of GUV shape. Our study contributes to advancing the design of shape-changing minimal cells for better characterization of the interaction between lipid bilayer membranes and actin cytoskeleton.

Keywords: actin; encapsulation; fascin; giant unilamellar lipid vesicles; membrane deformation.

FIGURE 1

Actin (top right), lipid (top middle), and merged (top left) confocal images of actin bundles which assembled to deform GUVs and formed long filopodia-like membrane protrusions. Application of 488 nm light at high exposure times (3 s) resulted in the disassembly of actin bundles and reversal of GUV shape without affecting the integrity of the lipid bilayer. Images were captured at 1 frame/3.6 s. Fascin/actin, 0.5 (M/M). Scale bar, 10 μm.

FIGURE 2

(A,B) Actin (left), lipid (middle), and merged (right) confocal images of actin bundles in small GUVs (diameter < 16 μm) at fascin concentrations of 2.5 μM (A) and 0.5 μM (B). Actin, 5 μM. Scale bars, 10 μm. (C) Actin (left), lipid (middle), and merged (right) confocal images of a protruding actin bundle with aggregated actin at the point of initiation of membrane protrusion (arrow) at fascin concentration of 0.5 μM. (D) Actin fluorescence intensity profile across the dashed lines in (C). (E) Actin (left), lipid (middle), and merged (right) confocal images of actin bundles in small GUVs (diameter < 16 μm) at a fascin concentration of 0.25 μM. Actin, 5 μM. Scale bar, 10 μm.

FIGURE 3

(A) Probability of the formation of a single actin bundle (Bundle) or networks (Ntw) in small GUVs (7–16 μm diameter) and large GUVs (>16 μm diameter) at actin concentration of 5 μM with different fascin concentrations indicated. Number of GUVs >30 per experiment per category; ≥2 independent experiments per fascin concentration. (B) The persistence length of actin bundles at two fascin concentrations. Actin, 5 μM; >18 bundles in 6 GUVs across two independent experiments were analyzed per fascin concentration. Error bar represents standard error of the mean. (C) The percentage of small GUVs (7–16 μm diameter) with shape changes (deform and/or protrude) by fascin-actin bundles at various fascin concentrations indicated. Number of GUVs > 30 per experiment per category; ≥2 independent experiments per fascin concentration. (D) The percentage of small GUVs (7–16 μm diameter) with filopodia-like membrane protrusions by fascin-actin bundles at different fascin concentrations indicated. Actin concentration is 5 μM. Number of GUVs > 30 per experiment per category; ≥2 independent experiments per fascin concentration. (E) Schematic representation of the formation of single actin bundles at high fascin concentrations (left) and actin bundle structures around the membrane at low fascin concentrations (right) in small GUVs (7–16 μm diameter). Small geometry of confinement increases the interaction of fascin-bundled actin filaments at high fascin concentrations to merge into a stiff protruding bundle at the GUV mid-plane. These bundles cause GUVs to deform into spindle-like vesicles with filopodia-like protrusions.

FIGURE 4

(A–C) Representative confocal merged images (actin:green, lipid:red) of GUVs encapsulating actin bundles in the presence of fascin and α-actinin. Actin bundles in the lumen of large GUVs do not deform GUVs (A). Localized actin bundles (B) and meshwork (C) at or around the periphery deform GUVs (arrows). Fascin/actin, 0.1 (M/M). α-actinin/actin, 0.1 (M/M). Scale bars, 10 μm. (D) Skeletonized image of encapsulated actin bundles in (C). Actin meshwork in the deformed GUV is indicated by arrow. Scale bar, 10 μm. (E) The percentage of small GUVs (7–16 μm diameter) with shape changes by fascin-actin bundles without and with α-actinin at various concentrations. Fascin/actin, 0.1 (M/M). Actin, 5 μM. Error bars indicate standard error of the mean. Number of GUVs > 30 per experiment per category; three independent experiments per category. (F) The percentage of GUVs with filopodia-like membrane protrusions formed by fascin-actin bundles without and with α-actinin at various concentrations in small (7–16 μm diameter) and large (>16 μm diameter) GUVs. The black line indicates the probability of filopodia-like protrusion in small GUVs encapsulating a single actin bundle. Fascin/actin, 0.1 (M/M). Actin, 5 μM. Number of GUVs > 30 per experiment per category; three independent experiments per category.