Membrane Condensation and Curvature Induced by SARS-CoV-2 Envelope Protein

The envelope (E) protein of SARS-CoV-2 participates in virion encapsulation and budding at the membrane of the endoplasmic reticulum Golgi intermediate compartment (ERGIC). The positively curved membrane topology required to fit an 80 nm viral particle is energetically unfavorable; therefore, viral proteins must facilitate ERGIC membrane curvature alteration. To study the possible role of the E protein in this mechanism, we examined the structural modification of the host lipid membrane by the SARS-CoV-2 E protein using synchrotron-based X-ray methods. Our reflectometry results on solid-supported planar bilayers show that E protein markedly condenses the surrounding lipid bilayer. For vesicles, this condensation effect differs between the two leaflets such that the membrane becomes asymmetric and increases its curvature. The formation of such a curved and condensed membrane is consistent with the requirements to stably encapsulate a viral core and supports a role for E protein in budding during SARS-CoV-2 virion assembly.


Estimation of SLD and area contribution of the E protein to the bilayer Area occupied by the helix of the transmembrane domain in a lipid bilayer:
approximately 60 Å 2 for a lipid and alpha helix d~10 Å (see figure below), πr 2 ≈ 80 Å 2 , with a protein content of 0.5 mol % in a bilayer → 1.3 area% are occupied by the helix

E protein
The existence of α-helical structure in E protein in the model membranes was determined by circular dichroism spectroscopy.The E protein maintains a considerable percentage of alpha helix structure in all the membrane mixtures tested.Here below the content (in %) of each secondary structure of E protein calculated using the CONTIN analysis (via Dichroweb 2 ).
NRMSD stands for normalized root mean square deviation and represent the goodness of fitting.Ideal would be <0.05,acceptable is <0.1.The data from POPC in presence of E protein was performed to evaluate if the E protein in a lipid bilayer has a 1 st order structural transition between the temperature of the SAXS measurement and the SLB measurement, crossing the physiological temperature.Note that POPC membrane alone does not have a 1 st order phase transition above 0 o C and its melt transition is below 0°C

Figure S1 .
Figure S1.Side view of the E protein structure (left) and SLD profile of the bilayer of a POPC liposome (22 o C) (right).The transmembrane region of E protein (residues 17 to 37, purple circles) is aligned to the hydrophobic core of the bilayer.The position of the α-carbon in these residues is indicated relative to the SLD profile (purple lines).Black circles represent the amino acid backbone and apolar residues.The red plus sign, blue cross and green circles represent polar residues with positive and negative charges, and neutral polar residues, respectively.

Figure S3 .
Figure S3.CONTIN analysis of the percentage of secondary structure of E protein in different bilayer systems.

Figure S4 .
Figure S4.Circular dichroism spectra of E protein in different lipid bilayers and temperatures showing the classic α-helical pattern.

Figure
Figure S5 shows the deviation of the supported lipid bilayers at different footprints and from different preparations, in XRR data and in the analysed SLD.

Figure S5 .
Figure S5.Deviation of the XRR of the POPC and DPPC SLBs without and with co-deposited E protein.The shaded region around the curve represent the standard deviation from multiple measurements, on different footprint and also from various preparations.Arrows show the major change on the reflectivity curve.Deviation of the SLD of the POPC-based bilayers is shown on the upper-right panel.

Figure
FigureS6shows the full set of the XRR data, their fit, and the obtained SLD profiles of the SLBs incubated with E protein containing aqueous phase.

Figure
Figure S6.XRR data and their fits from the fluid POPC SLB (upper), and the gel DPPC SLB (lower), both incubated with E protein containing subphase, the obtained SLD profiles, and the SLD deviation Δρb from the initial pure state (0m).Data are offset by a factor of 10 in XRR, by 5 in SLD and by 2 in Δρb.Dashed lines are the XRR fit, SLD and Δρb of the initial state.Incubation time is entered along the curves.Small insert in the XRR plot is the sum of the squared deviation of measured log10(RQ 4 ) from the initial state, as a function of the incubation time.

Figure S7 .
Figure S7.The complete set of cryo-TEM images of DPPC 0.5 mol% E. Intravesicular vesicles are indicated in yellow, vesicles which seem to shed outwards are indicated in blue and vesicles which seem to shed into the membrane compartment in magenta., the left image is the right without highlighted vesicles.Panel A is the same image of Figure 4A in the main text.

Figure
Figure S8.Cryo-TEM of POPC vesicles in absence of E protein.

Figure S9 .
Figure S9.DSC heating curves (second scan) in the range of 10-60°C.The lines indicate the temperatures of the different xray experiments The temperatures indicate the temperatures of the experiments and the body temperature as orientation.No significant change of the transition temperature or peak height is observed in the DP-systems.The data from POPC in presence of E protein was performed to evaluate if the E protein in a lipid bilayer has a 1 st order structural transition between the temperature of the SAXS measurement and the SLB measurement, crossing the physiological temperature.Note that POPC membrane alone does not have a 1 st order phase transition above 0 o C and its melt transition is below 0°C3 .

Table S1 .
Area per lipid AL, volume Vtail of the hydrophobic chain region of one lipid, and the negative surface charge density obtained from the analysis of the SAXS data. *

3 .
Structural parameters of the POPC bilayer and the co-deposited POPC / 0.5% E bilayer on silicon wafer, obtained from the fitting of the X-ray reflectivity data.Structural parameters of the DPPC bilayer and the co-deposited DPPC / 0.5% E bilayer on silicon wafer at 22 o C and 53 o C, obtained from the fitting of X-ray reflectivity data.