Condensate-mediated shape transformations of cellular membranes by capillary forces

Abstract

Phase-separated biomolecular condensates with liquid-like properties play a key role in the organization and compartmentalization of the intracellular environment. Condensate-mediated capillary forces acting on membranes drive physiologically important reshaping of membrane-bound organelles, such as vacuoles and autophagosomes. Here, we explore condensate-mediated membrane shape transformations. We employ {\textit{in planta}} live-cell imaging, an \textit{in vitro} reconstitution system with tunable interfacial tension, and computer simulations of an elastic membrane model to describe three morphologies of membrane structures localized at condensate interfaces: tubes, sheets, and cups. We find that the forces associated with high interfacial tension drive the formation of stable sheets, while tubes and cups prevail at lower interfacial tension. We calculate the free energies of each membrane shape and identify the energy barriers that govern the transitions between the shapes. With this approach, we find that shape transformations depend on the history of the interfacial membrane and exhibit a tube-to-cup hysteresis. These findings indicate that temporal control of condensate surface properties can mediate the morphogenesis of cup-like structures in cells, such as the formation of "bulbs" within plant vacuoles. Our results further generalize how the interplay of condensates and membranes contributes to intracellular organization.

Figures (5)

• Figure 1: Condensates within plant vacuoles shape internal membrane structures. (A) Each individual A. thaliana plant can be used to sample multiple developmental stages of their embryos. Each fruit (seed capsule) contains 30-60 seeds (light gray) with a single embryo. (B) Embryo cells (white dashed lines to guide the eye, see feeney_protein_2018) at different developmental stages are characterized by single, large vacuoles without condensates (early bent cotyledon stage), single large vacuoles with accessible condensate interfaces (mid bent cotyledon stage), and several small, condensate-filled vacuoles (late bent cotyledon stage). Confocal live-cell imaging of embryonic cotyledon (leaf) cells expressing the tonoplast protein marker TPK-GFP (membrane, green) and condensates (magenta, Neutral Red). (C) Single plant cell at mid bent cotyledon stage with vacuole (largest cellular organelle, green) and condensates (magenta). Left, schematic of plant cell with condensates and interfacial membrane. Right, corresponding confocal live-cell image. Yellow box, condensate with interfacial sheet. (D) Fluorescence intensity line profile along the yellow arrow in (C). (E) Interfacial sheets. Left, confocal section. Centre, 3D reconstruction of the interfacial sheet shown in (C), yellow box. Scanning depth 5 $\mathrm{\mu m}$. Right, sheet schematic. (F) Left, bending interfacial sheet. Centre, interfacial cup. Right, cup schematic. All scale bars, $5~\mathrm{\mu m}$.
• Figure 2: Membrane structures at liquid-liquid interfaces inside GUV. (A) Left, schematics of GUVs (green) cut in half for illustration purposes. The GUV interior is filled with homogeneous dextran/PEG solutions (light magenta). It undergoes phase separation upon hyperosmotic quenching, forming an upper compartment of PEG-rich phase and a lower compartment of dextran-rich phase (magenta). The liquid-liquid interface is located at the equatorial neck region of the GUV. Right, interfacial membrane shapes at GUV neck. Hyperosmotic quenching generates excess membrane that accumulates at the interface in the form of tubes (top), sheets (center), and cups (bottom). (B) Confocal images of interfacial tubes, sheets, and cups. Top, maximum intensity projection. Bottom, slightly tilted 3D projection of the same interface as above. Membrane fluorescence (DilC18) appears black in the upper projection panel and white in the lower 3D panel. (C) Sections of the interface in the xz-plane as in (A). Top left, sheet. Top right, cup. Bottom, line profiles along the arrow segments indicated above show a doubling of sheet and cup fluorescence intensity relative to the outer, unilamellar GUV membrane. (D) Sections of GUVs (green, Atto 647N-DOPE) in the xz-plane as in (A), obtained with STED microscopy. Left, sheet. Right, cup. The outer solution was stained with a water-soluble dye (yellow, Atto 488). (E) Left, averaged STED line profiles of sheets and cups as in (D), showing that the outer phase (yellow) is observed between the two membrane bilayers of sheets and cups (green). Each averaged profile was constructed with up to 10 individual line profiles, laterally cut through the adjacent bilayers, and approx. equally spaced along the structure. Right, membrane spacing of sheets ($204\pm23~\mathrm{nm}$) and cups ($105\pm25~\mathrm{nm}$), mean $\pm$ SD. (F) Interaction potential of charged membranes using Derjaguin–Landau–Verwey–Overbeek (DLVO) theory with $50~\mathrm{\mu M}$ ion concentration and $-3.2~\mathrm{mC/m^2}$ surface charge. Data points of sheets and cups from (E). Repulsive forces dominate at membrane spacing of $\leq200~\mathrm{nm}$, see SI. (G) Interfacial tubes and sheet precursors are metastable for $>24~\mathrm{h}$ before they transform within $<1~\mathrm{min}$ into several sheets that merge over time (see Movie S1). Maximum intensity projection (3 slices, $3~\mathrm{\mu m}$ scanning depth). Scale bars are $10~\mathrm{\mu m}$.
• Figure 3: Free energy landscape of in silico vesicles. (A) Top, triangulated sheet-like vesicle (green) wetting a flat condensate surface (magenta) characterized by interfacial tension $\sigma$. Bottom, schematic of the cross-section. (B) Minimized free energy landscape of vesicles with a reduced volume $v= 0.6$ upon changing membrane asymmetry $\Delta a$. Black and left ordinate, $\sigma=0$; magenta and right ordinate, $\sigma=1$. Shape transitions between tube, sheet, and cup are associated with two energy barriers, $H_1$ and $H_2$, that change with $\sigma$. Mean $\pm$ SD for $n = 5$ independent MC minimizations. Snapshots of membranes illustrate the expected shapes at respective metastable points, irrespective of the presence or absence of an interface. (C) Energy barriers $H_1$ (solid lines) and $H_2$ (dashed lines) depend on interfacial tension $\sigma$, and their values can be obtained continuously for all $\sigma$ from our computed energy landscapes (see Fig. S8 for specific transition values of $\sigma$). Colors correspond to $v=0.3-0.6$. (D) Morphological diagram of in silico interfacial vesicles in terms of $\sigma$ and $v$ obtained based on the data in (C). $L_T$ line demarcates the area of unstable tubes (red); $L_S$ demarcates the area of unstable sheets (blue). Middle area (green) allows for all morphologies. Stars correspond to tube-to-sheet (red) and sheet-to-cup (blue) transition conditions used in Fig. \ref{['fig:figure_5']}A,B.
• Figure 4: Morphological behavior of interfacial membranes. (A) In vitro shape frequencies of interfacial membranes (see Fig. \ref{['fig:figure_2']}) following 24 h equilibration. (B) In vitro and (C) in silico free energy differences of cups and sheets relative to the tube state. (B) is obtained from applying \ref{['eq:energy_diff']} to data presented in (A). Shaded regions correspond to free energy differences obtained from theoretical calculations with varying contact angles ($\theta=60^\circ-120^\circ$, see Fig. S11).
• Figure 5: Shape transformation pathways of interfacial vesicles (A) Isometric view of tube-to-sheet intermediate obtained by increasing $\sigma$ from 0 to 8.7. Before increasing $\sigma$, the initial tube-like, triangulated vesicle with $v=0.4$ was equilibrated for $10^7$ MC simulation steps. (B) Side view of a subsequent sheet-to-cup transition by decreasing $\sigma$ from 8.7 to 0.8. (C) Schematic of tension-induced tube-to-sheet and sheet-to-cup transitions. Increasing $\sigma$ breaks the symmetry of the initially symmetric tube and forms a sheet-like domain locally (top right). Expanding the sheet-like section adsorbs the complete tube and spontaneously transforms into a sheet via an irreversible pathway at constant $\sigma$. The sheet-to-cup transition initiates upon decreasing $\sigma$ and is likely reversible, see Fig. S8. However, the whole tube-to-cup transition is hysterical: for the same $\sigma$, either tubes or cups can exist, depending on the history of the vesicle. (D) Transition times obtained from MC simulations, shown in (A,B). Mean $\pm$ SD for $n=50$ independent MC simulations. (E) In vitro ensemble transition times, using the transition model (see Methods). At low $\Sigma$, the tube-to-sheet transition is slow (high $\tau_1$), while the sheet-to-cup transition is fast (low $\tau_2$). With increasing $\Sigma$, $\tau_1$ decreases (faster tube-sheet transition) while $\tau_2$ increases (slower sheet-to-cup transition).