A structural analysis of M protein in coronavirus assembly and morphology

Abstract

The M protein of coronavirus plays a central role in virus assembly, turning cellular membranes into workshops where virus and host factors come together to make new virus particles. We investigated how M structure and organization is related to virus shape and size using cryo-electron microscopy, tomography and statistical analysis. We present evidence that suggests M can adopt two conformations and that membrane curvature is regulated by one M conformer. Elongated M protein is associated with rigidity, clusters of spikes and a relatively narrow range of membrane curvature. In contrast, compact M protein is associated with flexibility and low spike density. Analysis of several types of virus-like particles and virions revealed that S protein, N protein and genomic RNA each help to regulate virion size and variation, presumably through interactions with M. These findings provide insight into how M protein functions to promote virus assembly.

Fig. 1

Coronavirus particles and vesicles in vitreous ice. Cryo-electron micrographs of small unilamellar vesicles (A), MHV-like particles (B), three coronaviruses (C) and cryo-electron tomography of MHV (D) are shown. The longest and shortest visible diameter of each particle was measured, as shown in panel A. Viral particles were distinguished from empty exosomes by the thickness of the envelope. A spikeless particle (∗) and free viral envelopes (e) are marked. Gold particles which were used as fiducial markers in construction of the tomogram are marked with arrows.

Fig. 2

Identification of M and ribonucleoprotein in viral particles. (A) A model viral envelope explains features of an MHV particle by comparing the (B) averages of ∼20 radial density distributions from MHV virus-like particles, apparently spike-depleted virions, normal virions and empty exosomal vesicles. (C) Individual features are mapped onto the radial density pattern of empty vesicles as follows: M protein (blue) is revealed as the difference between EM VLPs and vesicles, S (red) is the difference between normal and spikeless virions and ribonucleoprotein is the difference between EM VLPs and spikeless virions. (D) Differences in the consistency of core and envelope organization are revealed from the standard deviation of 40 MHV radial density plots.

Fig. 3

Evidence for two forms of M protein in virions. Distinct rod-like M_LONG features can be seen in at the edge of cryo-EM images of MHV EMN VLPs (A) and feline coronavirus (B–F). Regions where M appears indistinct, compact and does not appear to contact the ribonucleoprotein at the particle edge are marked with curved arrows. Regions of membrane that resemble protein-free lipid bilayers (D, inset and G) are marked with a double-line. Enlargements of M_LONG (E), M_COMPACT (F) and a vesicle (G) are shown below. Radial density maps (H) and M tail to body ratios (I) are shown to illustrate the difference in appearance between M_LONG and M_COMPACT from three coronaviruses.

Fig. 4

Edge views of packaged M. Reference-free class averages were made from groups of 300–2000 images clipped from viral particle edges. Full contrast transfer function correction through 17 Å resolution was implemented during class average reconstruction. Edge view class averages show M_LONG (A, top) and M_COMPACT (A, bottom) from MHV EMN VLPs. Examples of class averages showing M_LONG from EM VLPs and virions are shown in panel B. Panel C shows two-dimensional projections of M ectodomain, transmembrane domain and endodomain as ellipsoids with the width of M_LONG from class averages and the expected volume of one, two or three copies of M protein (right).

Fig. 5

Association of M_LONG with convex, spike-decorated membranes. Panels A and B depict free viral envelopes that were found in MHV tomograms. Projections through thick (50 nm) and thin (1 nm) subtomograms are shown for comparison. Thicker regions of the membrane ascribed to M_LONG are marked black, and thinner regions ascribed to M_COMPACT are marked white. Visible spikes are marked with circles to demonstrate membrane topology and intact virions (V) are also indicated.

Fig. 6

Organization of M. Small areas which appear to show axial views of small clusters of organized M proteins are marked in images of VLPs (A), and enlarged below (B, left). Reference-free class averages were made from groups of 1500 (EM) or 3000 (EMN) images clipped from VLP centers. Full contrast transfer function correction through 17 Å resolution was implemented during class average reconstruction. Unit cell dimensions are approximately 4 nm (a) by 4.5 nm (b), with an interior angle of 75°. (C) Average spacing of M dimers as measured by various methods.

Fig. 7

Relationship between M conformation and particle shape. (A) The ratio of the longest to the shortest diameter of coronavirus particles from MHV, SARS-CoV and FCoV micrographs and MHV tomograms is plotted against the percentage of particles which had at least one anomalously thin region of membrane. Each datapoint presents the average for 20 particles of similar shape. (B) Boxes were centered on the membrane at each end (stars) and side (circles) of seventy SARS-CoV particles to examine the relationship between M-form and membrane curvature. (C) The ratio of M_LONG to M_COMPACT is plotted against average particle shape. Each datapoint represents two sides or ends from fourteen particles of similar shape. (A, C) Curves were fitted by logistic regression.

Fig. 8

Conversion between M_LONG and M_COMPACT. Purified MHV was imaged by cryo-EM after incubation at pH 7 (A), or a 5 min pulse at pH 5 followed by re-neutralization to pH 7 (B). Arrows in panel B point to flattened edges of the particle with the appearance of M_COMPACT. The proportion of flat-sided virions before and after acidification is shown in table form (C). Radial density maps (D) and M tail to body ratios (E) of the altered regions from acidified MHV, native EM VLPs and native EMN VLPs are shown alongside MHV M_LONG and M_COMPACT data from Fig. 3h-i.

Fig. 9

Relationship of S protein to morphology. (A) The ratio of the longest to the shortest diameter of coronavirus particles from micrographs is plotted against the percentage of particles which had fewer than half the expected number of spikes visible at the particle edge. Each datapoint depicts the average and standard deviation for a group of 20 similar-sized particles. Curves were fitted by logistic regression to datasets in which the association between spike decoration and size was statistically significant (ANOVA, P = 6.71 × 10⁻⁸ for FCoV and 0.046 for MHV). (B) SARS-CoV particles were classified according to shape and spike distribution. The three shape categories refer to particles with d_MAX/d_MIN less than 1.1 (round), 1.1–1.4 (slightly oval) or greater than 1.4 (elongated). The number of particles of each type that were classified is listed.

Fig. 10

Interpretation of coronavirus structure and assembly. (A) A schematic cutaway of an elongated SARS-CoV particle (size ≈ 85 nm, d_MAX/d_MIN ≈ 1.4) is shown to demonstrate how interactions between M_LONG (black), M_COMPACT (white) and the ribonucleoprotein influence particle morphology. Spikes are shown in red and E protein oligomers in yellow. (B, C) Model for how M_LONG could be related to the spacing of spikes and RNP densities in edge view (B) and axial view (C). Hollow shapes mark the nearest potential attachment points for an additional spike or RNP density. (D) A flowchart presents a model for the role of M conformation in assembly. An additional path suggests and explanation for how particle morphology can be altered during purification.