Vesicle formation by self-assembly of membrane-bound matrix proteins into a fluidlike budding domain

Abstract

The shape of enveloped viruses depends critically on an internal protein matrix, yet it remains unclear how the matrix proteins control the geometry of the envelope membrane. We found that matrix proteins purified from Newcastle disease virus adsorb on a phospholipid bilayer and condense into fluidlike domains that cause membrane deformation and budding of spherical vesicles, as seen by fluorescent and electron microscopy. Measurements of the electrical admittance of the membrane resolved the gradual growth and rapid closure of a bud followed by its separation to form a free vesicle. The vesicle size distribution, confined by intrinsic curvature of budding domains, but broadened by their merger, matched the virus size distribution. Thus, matrix proteins implement domain-driven mechanism of budding, which suffices to control the shape of these proteolipid vesicles.

Figure 1.

Interaction of M protein with lipid membrane, monitored by patch clamp admittance measurements. (A) Changes of the admittance (ΔIm and ΔRe) and ionic permeability (G_dc) of the patch connected to the membrane reservoir upon application of 2 μM of M protein. Level 1 shows the background level of ΔIm corresponding to the initial area of the patch. ΔIm deviations back and forth to level 1 indicate reversible changes of the patch area, and each single alteration (e.g., around level 2) indicates a budding event. (B) Expanded selection from black box in A. Transient increase in ΔRe (arrow) indicates formation of a thin membrane neck. (C) Cumulative distribution of the values of ΔIm jumps and the corresponding diameters of the spherical membrane particle. The initial part of the distribution (up to ∼1.3 fF) is expanded to show a Gaussian-like profile. (D) Left histogram shows the distribution of small ΔIm jumps from C; right histogram shows the distribution of ΔIm jumps obtained at elevated (5 μM) concentration of M protein.

Figure 2.

Visualization of the budding activity of M protein on a membrane patch. (A) Frame sequence (time in seconds) illustrating budding from a patch pipette (approximately drawn in the first image) containing 2 μM of M protein observed on a GUV. A small part of the large GUV, attached to a platinum electrode used for electroformation, was sucked into the pipette. Recording began after establishing a stable contact between the GUV and the pipette. Bar, 5 μm. (B) Expanded images, corresponding to the area marked by the purple rectangle in A, illustrate brightening of the membrane patch upon M protein adsorption. (C) The scheme outlines a correspondence between the changes in ΔIm and the budding. Levels in and fin show the ΔIm increase caused by formation of a bud. Red arrow indicates bud closure; blue arrow indicates fission of the neck. (inset) The fission shown in detail. Bars: (A) 5 μm; (B) 1 μm; (C, horizontal) 40 ms; (C, vertical) 20 pS.

Figure 3.

Formation of intralumenal vesicles by M protein applied to GUVs of different lipid compositions. (A) Frame sequence shows formation of intralumenal vesicles after M protein application (at 0 s) to GUV (PC–cholesterol mixture). (B) M protein adsorption on LUVs of different lipid compositions (0.005 protein/lipid ratio) measured by gradient flotation technique. The same protein concentration for all bands was loaded and the control fraction (M, no lipids) was taken at the same level as the liposome fraction. (C) Effect of M protein and BSA application (4 μM in the delivery pipette) on the morphology of GUV of different lipid compositions. Images were taken before (top) and ∼2 min after (bottom) protein application. Representative images of three independent experiments are shown. Bars, 2 μm.

Figure 4.

Interaction of M protein with LUVs. (A) M protein adsorption on PC–PE–cholesterol LUVs at different protein/lipid ratios measured by gradient flotation. The same protein concentration was loaded for all bands. The control fraction (M, no lipids) was taken at the same level as the liposome fraction. Positive control shows the M protein band. (B) Sequential additions of 0.3 μM of M protein or BSA to LUV caused dequenching of Rh-DOPE or BODIPY-G_m1 fluorescence (red and black circles). No changes were detected for nonquenched dyes (red and black diamonds) or when BSA was added (dark yellow circles). (E) The same additions of M protein induce the release of LUV-entrapped ANTS/DPX (blue squares) or 70-kD FITC-conjugated dextrans (green squares), seen as changes of normalized fluorescence intensity. The addition of the same amount of the protein mixed with α-chymotrypsin 1:5 causes minor release of ANTS/DPX and dextrans (blue and green triangles). Bars show SD.

Figure 5.

Formation of membrane domains after M protein application to GUVs. (A) Changes of membrane fluorescence and deformations of GUVs (PC–PE–cholesterol) induced by M protein (added at t = 0). Arrowheads show joining of bright domains. (B) Bright spots merger on GUV flattened on the glass surface. (C) Negative staining of M proteins condensing on a lipid monolayer (arrowheads). Bars: (A and B) 5 μm; (C) 50 nm.