Plasma membrane nanodeformations promote actin polymerisation through CIP4/CDC42 recruitment and regulate type II IFN signaling

Abstract number
144
Presentation Form
Poster
Corresponding Email
[email protected]
Session
Poster Session
Authors
Benjamin Ledoux (3), Natacha Zanin (3), Jinsung Yang (1), Vincent Mercier (4), Christine Dupont (2), David Alsteens (2), Pierre Morsomme (2), Henri-François Renard (3)
Affiliations
1. Gyeongsang National University
2. UCLouvain
3. UNamur
4. University of Geneva
Abstract text

Nanotopography is a physical cue known to induce changes in cellular behaviour including cell proliferation, migration, and stem cell differentiation. However, the molecular mechanism involved in nanotopography recognition and its effect on cellular physiology is currently poorly understood.

In their environment, cells constantly experience a variety of external mechanical stimuli and stresses inducing membrane curvature and cell deformation. Among those, cell deformation due to the presence of nanotopographical features is now largely accepted as a biophysical stimulus influencing cellular functions. However, the mechanotransduction cascades involved in nanotopography recognition and their precise molecular effects on cellular physiology are still poorly understood. To bridge the gap, BAR (Bin/Amphiphysin/Rvs) domain proteins are attractive candidates as their intrinsic properties position them right at the interface between membranes and signaling. Indeed, BAR domain proteins form crescent-shaped dimers with a unique intrinsic curvature radius, allowing them to induce and/or bind membrane deformations of different geometries. Then, after binding membranes through their BAR domain, they can promote local molecular responses due to the presence of additional functional domains such as Src homology 3 (SH3) or Rho guanosine triphosphatase (GTPase)–activating protein (GAP) domains. Yet, their ability to translate plasma membrane curvature into molecular responses is only superficially explored. In this study, we investigated the role of BAR domain proteins as mechanosensors of plasma membrane geometry (Ledoux et al., 2023).

We developed a colloidal lithography technique to synthesize fluorescent nanostructured substrates that were used to induce spherical and isotropic nanoscale plasma membrane deformations ranging from 100 to 500 nm in living cells. Cells seeded on these surfaces were then imaged using a variety of microscopy techniques (confocal, airyscan, STED, FIB-SEM and correlative AFM-confocal).

We show that a distinct subset of BAR domain proteins was recruited to each plasma membrane deformation size, where they formed different patterns (Figure 1). Interestingly, the affinity of BAR proteins for a particular curvature radius does not correlate with their usual subfamily classification. Our data further demonstrate that membrane curvature promotes the formation of dynamic actin structures mediated by the Rho GTPase CDC42 (and not Rac1) and the F-BAR protein CIP4 (Figure 2). We propose a model in which activated CDC42 and membrane curvature locally recruit CIP4, where both proteins then cooperate to promote actin polymerisation. Furthermore, using correlative fluidic force microscopy/fast-scanning confocal microscopy, we demonstrate that these actin structures can form at both basal and apical plasma membranes in a very dynamic manner (Figure 3). Lastly, we explore the role of these curvature-induced actin nanodomains in receptor signaling. We show that interferon-γ receptor is localized in these actin nanodomains and that downstream interferon-γ-induced JAK/STAT1 signaling is severely impaired by the actin nanodomains forming around 500-nm nanostructures (Ledoux et al., 2023).

Together, our data demonstrate that plasma membrane deformation induces the formation of CDC42/CIP4/PI(4,5)P2-dependent actin structures and that large actin rings induced by substrate nanotopography can modulate receptor signaling. These findings may be key for understanding the effects of nanotopography on cellular behaviour in biological contexts as diverse as development, cancer progression or for the design of biocompatible materials.

Figure 1: BAR domain proteins bind to plasma membrane deformations. Representative images of controls (PalMyr-GFP and GFP) and selected BAR domain proteins depicting different patterns observed around membrane nanodeformations (Rings for PACSIN3, discontinuous rings for FAM92b and dots for FCHo2).


Figure 2: CDC42 controls actin polymerization and CIP4 recruitment around 100-nm plasma membrane deformations. HeLa cells grown on 100-nm nanostructures and transfected with fluorescent constructs [(A) to (F)] or treated with a specific CDC42 inhibitor (ML141) [(E) and (F)], as indicated. Data shown are quantifications of actin [(B) and (E)], endogenous CIP4 [(C) and (F)], or GFP (F) fluorescence around 100-nm deformations and the corresponding representative Airyscan images (A). (A to D) Effect of transient expression of GFP, GFP-CDC42-WT, GFP-CDC42-T17N, GFP-CDC42-Q61L, or GFP-Rac1-WT on the enrichment of actin, CIP4, and GFP around deformations. Number of deformations: GFP, n = 8146; CDC42-WT, n = 3829; CDC42-T17N, n = 2993; CDC42-Q61L, n = 3058; Rac1-WT, n = 3942. Three independent experiments. (E and F) Effect of dose-dependent inhibition of CDC42 by ML141 (0, 20, 50, or 100 μM) on the enrichment of actin and CIP4 around deformations. Number of deformations: 0 μM, n = 3007; 20 μM, n = 2541; 50 μM, n = 2788; 100 μM, n = 2252. Three independent experiments. Data are means ± SEM; ns, not significant; ****P < 0.0001; ***P < 0.001 (one-way ANOVA with Dunnett’s multiple comparison test). Regions marked by dashed squares, expanded below with individual channels displayed (A, bottom). White arrowheads, colocalization (A). Scale bars, 10 μm (A).


Figure 3: Local CIP4- and CDC42-dependent actin polymerization is dynamic and promoted by plasma membrane curvature. (F and G) Monitoring of CDC42-dependent actin polymerization upon deformation of the plasma membrane by FluidFM. (F) Quantification of normalized LifeAct-mCherry fluorescence intensity around FluidFM tip over time upon coexpression of GFP (control, black), GFP-CDC42-T17N (red), GFP-CDC42-WT (blue), or GFP-CDC42-Q61L (green). Ten-minute time-lapses with 9-s intervals between frames. Number of cells: GFP, n = 20; GFP-CDC42-T17N, n = 18; GFP-CDC42-WT, n = 15; GFP-CDC42-Q61L, n = 17. Two independent experiments. (G) Representative image of FluidFM/confocal experiments upon coexpression of GFP-CDC42-Q61L and LifeAct-mCherry. Region marked by a dashed square surrounding the bead, expanded below at the indicated time points; t = 0 min, initial deformation of the cell membrane. TL, transmitted light (FluidFM cantilever). Scale bar, 10 μm (G).

References

Ledoux, B., Zanin, N., Yang, J., Mercier, V., Coster, C., Dupont-Gillain, C., Alsteens, D., Morsomme, P., & Renard, H. F. (2023). Plasma membrane nanodeformations promote actin polymerization through CIP4/CDC42 recruitment and regulate type II IFN signaling. Science advances9(50), eade1660. https://doi.org/10.1126/sciadv.ade1660