Subdomains of Endophilin-NBAR Can Synergistically Drive Membrane Remodeling and Facilitate Controlled Membrane Scission

Using coarse-grained molecular dynamics simulations, this study demonstrates that the synergistic action of endophilin's NBAR subdomains enables them to sort to and generate negative Gaussian curvature, thereby facilitating membrane hemifission during endocytic bud formation.

The Gist

Imagine your cell is a bustling city, and it needs to constantly bring in supplies (like nutrients) or recycle old equipment. To do this, the cell's outer wall (the membrane) has to pinch off little bubbles, grab the cargo, and pull them inside. This process is called endocytosis.

The star player in this operation is a protein called Endophilin. Think of Endophilin as a specialized "membrane sculptor." But how does it work? Is it a single tool, or a team of tools working together? And how does it know exactly where to pinch the membrane to create a bubble without tearing the whole wall apart?

This paper uses computer simulations (like a high-tech video game) to answer these questions. Here is the breakdown in simple terms:

1. The Team vs. The Solo Act

Endophilin isn't just one piece; it's made of different parts:

• The H0 Helix: A sharp, wedge-like spike.
• The BAR Domain: A curved, banana-shaped scaffold.
• The Linker & SH3: A floppy tail that grabs other helpers.

The Discovery: The researchers found that the H0 spike and the BAR scaffold work best when they are glued together as a single unit (called NBAR).

• The Analogy: Imagine trying to bend a flat sheet of paper. If you just push on it with a single finger (the H0 spike), it might poke a hole or do nothing. If you just lay a curved ruler on it (the BAR domain), it might slide off. But if you tape the finger to the ruler, you have a perfect tool that can both push into the paper and hold the curve.
• The Result: When the parts work together, they can bend the membrane much more effectively than if they were working alone. The floppy tail (Linker/SH3) didn't seem to add much extra bending power in this specific simulation, but the core team (NBAR) was the real powerhouse.

2. The "Saddle" Shape Challenge

Most previous studies looked at flat membranes or simple tubes. But in real life, when a cell pinches off a bubble, the neck of that bubble looks like a saddle (think of the shape of a Pringles chip or a horse saddle). This shape has a tricky geometry called "negative Gaussian curvature."

The Discovery: Endophilin loves these saddle shapes.

• The Analogy: Imagine a crowd of people (Endophilin proteins) trying to stand on a trampoline. If the trampoline is flat, they spread out randomly. But if you create a dip in the middle (the saddle shape), the people naturally roll toward that dip and cluster there.
• The Result: The simulations showed that Endophilin doesn't just sit anywhere; it actively seeks out these "saddle" necks. Once there, it organizes itself into rings or strings, hugging the curve perfectly.

3. The "Lipid Reservoir" (The Safety Net)

Here is the most fascinating part. When Endophilin clusters on the neck of the budding bubble, it doesn't just hold the shape; it changes the local environment.

The Discovery: The proteins pull the surrounding lipids (the building blocks of the membrane) tight around them, creating a super-tight, organized patch.

• The Analogy: Imagine a group of people holding a heavy tarp. If they stand loosely, the tarp might rip. But if they pull the tarp tight around their hands, creating a reinforced "reservoir" of fabric, the tarp becomes much stronger and harder to tear.
• The Result: This tight packing creates a "lipid reservoir." It acts like a safety net. When the cell tries to pinch the neck to cut the bubble off, this reinforced area prevents the membrane from snapping too early or leaking. It allows the neck to get very thin and withstand more pressure before finally snapping (scission).

4. Why This Matters for "Fast" Endocytosis

Cells sometimes need to grab things incredibly fast (like in ultra-fast endocytosis). Usually, we think proteins need to build a massive, perfect scaffold to do this.

The Discovery: The study showed that Endophilin can do its job even if it hasn't formed a perfect ring. It just needs to be there, clustered at the neck.

• The Analogy: You don't need a perfectly built dam to stop a small flood; sometimes a few sandbags piled up in the right spot are enough to hold back the water until the main gate closes.
• The Result: This explains how cells can perform these tasks so quickly. They don't need to wait for a perfect, rigid structure to form; the dynamic, flexible clustering of Endophilin is enough to stabilize the neck and facilitate the cut.

Summary

This paper tells us that Endophilin is a master of teamwork and geometry.

1. Teamwork: Its two main parts (the spike and the curve) must work together to bend the membrane.
2. Geometry: It naturally hunts for the tricky "saddle" shapes found at the neck of budding bubbles.
3. Safety: Once it arrives, it tightens the local membrane like a reinforced belt, allowing the cell to pinch off a bubble without tearing the whole wall.

By understanding this, scientists can better understand how cells eat, how they recycle, and potentially how to fix these processes when they go wrong in diseases.

Technical Summary

1. Problem Statement

Endophilin is a critical NBAR-domain protein involved in endocytosis, known for its ability to sense and generate membrane curvature. While previous research has identified three hypothesized mechanisms for its remodeling capabilities—insertion (via the N-terminal amphipathic helix H0), scaffolding (via the BAR domain), and crowding (via the unstructured linker and SH3 domain)—several key questions remain unanswered:

• Synergy: How do these subdomains interact synergistically to drive large-scale membrane remodeling under physiological conditions?
• Negative Gaussian Curvature (NGC): Most previous studies focused on flat membranes (zero curvature) or positive curvature (tubes/vesicles). However, endocytic events involve Ω-shaped geometries with negative Gaussian curvature (saddle shapes) at the neck of budding vesicles. It is unclear how endophilin senses, recruits to, and remodels these specific geometries.
• Scission Mechanism: The precise role of endophilin in the final membrane scission (hemifission) step, particularly in fast endocytic pathways where full scaffold assembly may not occur, is not well understood.

2. Methodology

The authors employed Coarse-Grained (CG) Molecular Dynamics (MD) simulations to model endophilin interactions with lipid membranes at a quasi-molecular resolution.

• Model Construction:
  • Protein: The endophilin-A1 structure was derived from PDB ID 1ZWW. The N-terminal H0 helix was added, and the SH3 domain/linker was modeled using AlphaFold2.
  • Resolution: A CG model was developed where one bead represents one amino acid (at the $\alpha$-carbon position). A heteroelastic network model (hENM) was used for protein potentials, calibrated against atomistic MD data.
  • Membrane: Lipids were modeled as four-bead chains (head, interface, two tails) representing a DOPC:DOPS (80:20) composition.
• Simulation Setup:
  • Software: LAMMPS (for CG) and GROMACS (for initial atomistic equilibration).
  • Conditions: NPT ensemble, 25 fs timestep, 0.15M KCl.
  • Constructs Tested: Various subdomains were simulated individually and in combination: H0 helix alone, BAR domain alone, the composite NBAR unit (H0 + BAR), and Full-Length (FL) endophilin (NBAR + linker + SH3).
• Geometries Simulated:
  1. Flat Sheets: To test initial binding and curvature generation.
  2. Cylindrical Tubes: To test remodeling of pre-curved positive curvature.
  3. Catenoid Tubes: To explicitly model Negative Gaussian Curvature (NGC).
  4. Ω-Shaped Bud: A dynamic simulation where a spherical "cargo" is pulled through a flat membrane to mimic endocytic budding and scission.

3. Key Contributions

• Synergistic Remodeling: Demonstrated that the NBAR unit (H0 + BAR) acts as a synergistic composite unit that is significantly more effective at binding and remodeling membranes than either subdomain alone.
• NGC Sensing and Generation: Provided the first molecular-level evidence that endophilin NBAR domains are specifically recruited to and actively generate Negative Gaussian Curvature, a geometry previously difficult to study experimentally.
• Lipid Reservoirs: Discovered that NBAR assemblies create local "lipid reservoirs" (tightly packed lipid head groups) around the protein scaffold, which stabilizes the membrane against tension.
• Scission Dynamics: Revealed that endophilin networks delay membrane scission (hemifission) and prevent leakage by sustaining higher forces, even without a perfectly oriented, complete scaffold. This challenges the "friction-driven scission" (FDS) model.

4. Key Results

A. Subdomain Synergy on Flat and Tubular Membranes

• Binding: Only the NBAR and Full-Length constructs stably bound to flat membranes at 10% surface coverage. Isolated H0 or BAR domains largely remained in the bulk solution.
• Remodeling: The NBAR unit induced significant membrane curvature and organized into string-like assemblies on flat sheets and perpendicular ring-like scaffolds on tubes.
• Conclusion: The H0 helix and BAR domain must be connected to function effectively; the unstructured linker and SH3 domain did not significantly enhance remodeling at these densities, suggesting crowding is not the primary driver for endophilin (unlike amphiphysin).

B. Recruitment to Negative Gaussian Curvature (Catenoid Tubes)

• Sorting: In catenoid-shaped tubes (saddle geometry), NBAR dimers were preferentially recruited to regions of negative and neutral mean curvature, avoiding positive curvature regions.
• Orientation: While generally aligning perpendicular to the tube axis, a significant population in the negatively curved region aligned parallel to the Z-axis, a behavior unique to the catenoid geometry.
• Curvature Generation: The presence of NBAR dimers amplified local Gaussian curvature, creating "patchwork" regions of high curvature.

C. Lipid Reorganization and Reservoirs

• Radial Distribution Functions (RDF): While global lipid packing decreased slightly due to curvature, lipids within 1 nm of NBAR dimers showed a massive increase in packing density (RDF peak increased from ~32 to ~91).
• Implication: NBAR scaffolds sequester lipids to form tightly packed "reservoirs" that buffer membrane tension.

D. Dynamic Bud Formation and Scission (Ω-Geometry)

• Recruitment: During the active pulling of a cargo through a membrane, NBAR dimers were excluded from the bud tip and concentrated at the neck (the region of negative Gaussian curvature).
• Force Resistance:
  • No Protein: Membrane rupture (hemifission) occurred at ~450 kcal/mol/nm.
  • 10% NBAR: Rupture delayed to ~490 kcal/mol/nm.
  • 25% NBAR: Rupture delayed to ~520 kcal/mol/nm.
• Mechanism: The NBAR network allows the neck to narrow continuously until hemifission occurs between scaffold layers, rather than via pore formation at the scaffold edge (FDS). The lipid reservoirs prevent premature leakage.
• Physiological Relevance: Significant benefits (delayed scission, force resistance) were observed even at 10% coverage and without full scaffold orientation, suggesting endophilin can function effectively in ultrafast endocytosis where time for full assembly is limited.

5. Significance

• Mechanistic Insight: This study resolves the debate on endophilin's mechanism by proving that the synergy between H0 insertion and BAR scaffolding is the primary driver of remodeling, rather than crowding by the disordered linker.
• New Geometry Focus: It establishes a computational framework for studying Negative Gaussian Curvature, a critical but under-explored feature of cellular membrane dynamics.
• Scission Model: The findings propose a new model for membrane scission where endophilin acts as a tension buffer via lipid reservoirs, facilitating controlled hemifission rather than relying solely on friction-driven pore formation.
• Therapeutic/Experimental Implications: The results suggest that endophilin's role in fast endocytic pathways (like FEME) is robust even with partial assembly, offering new hypotheses for experimental validation regarding how cells regulate vesicle formation speed and stability.