Structure and inhibition of the SARS coronavirus envelope protein ion channel

Abstract

The envelope (E) protein from coronaviruses is a small polypeptide that contains at least one alpha-helical transmembrane domain. Absence, or inactivation, of E protein results in attenuated viruses, due to alterations in either virion morphology or tropism. Apart from its morphogenetic properties, protein E has been reported to have membrane permeabilizing activity. Further, the drug hexamethylene amiloride (HMA), but not amiloride, inhibited in vitro ion channel activity of some synthetic coronavirus E proteins, and also viral replication. We have previously shown for the coronavirus species responsible for severe acute respiratory syndrome (SARS-CoV) that the transmembrane domain of E protein (ETM) forms pentameric alpha-helical bundles that are likely responsible for the observed channel activity. Herein, using solution NMR in dodecylphosphatidylcholine micelles and energy minimization, we have obtained a model of this channel which features regular alpha-helices that form a pentameric left-handed parallel bundle. The drug HMA was found to bind inside the lumen of the channel, at both the C-terminal and the N-terminal openings, and, in contrast to amiloride, induced additional chemical shifts in ETM. Full length SARS-CoV E displayed channel activity when transiently expressed in human embryonic kidney 293 (HEK-293) cells in a whole-cell patch clamp set-up. This activity was significantly reduced by hexamethylene amiloride (HMA), but not by amiloride. The channel structure presented herein provides a possible rationale for inhibition, and a platform for future structure-based drug design of this potential pharmacological target.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Figure 1. ETM forms a continuous α-helix in DPC micelles.

(A) Superimposed 20 conformers of ETM (monomer) calculated from CYANA. The peptide backbone is shown in cyan and the residue side chains in gold. (B) Secondary structure plot for ETM depicted as bands of varying thickness, indicative of the NOE intensity. Sequential and medium range NOE connectivities are shown below the primary sequence. dNN amide backbones and dαN(i,i+3), dαN(i,i+4) connectivities are mostly continuous throughout the length of the peptide, indicating that the peptides adopt a predominantly α-helical conformation . (C) Representative HN/HN region of 2D NOESY spectra.

Figure 2. Intra- and inter-monomeric NOEs.

Dereference 2D homonuclear 1HN, 1Haromatic band-selected NOESY exhibiting NOEs between 1H covalently bound to the 13C spins of L18, L19, and L21 and other proximal 1H spins. Intramonomeric cross-peaks are shown in blue. Cross-peaks which cannot be explained by intramonomer interactions based on the reconstructed secondary structure are assigned to intermonomer NOEs, and are shown in red.

Figure 3. Orientation of ETM relative to the detergent phase and α-helical geometry.

(A) PRE rates of 1HN nuclei for residues A22, V24, V25, L18, L19 and L21 (black solid line), superimposed to predicted PRE from the ETM pentamer using either the immersion depth method (black dotted line) or the distance from the center method (black dashed line)(Fig. S3, BC). (B) RDCs for 1HN nuclei corresponding to A22, V24, V25, L18, L19 and L21 in 4% and 8% polyacrylamide gels compressed axially (▲) or radially (■), respectively. Best-fit sine waves of 3.6 periodicity (red lines) are superimposed onto both PREs and RDCs plots, and show that the stretch of residues 19–24 adopt a regular α-helical structure. The RDCs of flanking residue L18 could not be fitted to this ∼3.6 periodicity, suggesting a local deviation from ideal α-helical geometry.

Figure 4. Pentameric structure of ETM.

(A) Lumen of the α-helical bundle (blue) formed by the ETM α-helices (omitted for clarity), with most constricted region at V25 (green). (B) Radius of the lumen of the ETM channel as a function of residue number, calculated by HOLE . (C) Side view surface representation of the ETM channel showing the lumen volume. Oxygen and nitrogen atoms are colored red and blue respectively.

Figure 5. Chemical shift perturbation in ETM induced by HMA.

(A) Difference in the 1HN chemical shifts (ETM minus ETMHMA) after addition of HMA at the molar ratio described in Methods. Amides exhibiting deviations of chemical shifts of more than 0.05 ppm are highlighted in green. (B) Top view of the ETM pentameric model showing these amides as small spheres.

Figure 6. Binding of HMA to the ETM pentameric channel.

(A) Side view of the binding of HMA to the ETM pentamer in the vicinity of N15. The side chains of amino acids interacting with HMA are shown using a stick representation. (B) Binding of HMA to the C-terminal binding site of the channel, in the vicinity of T35 and R38. The lowest energy conformation of HMA is shown at the centre of bundle. For clarity, one of the ETM monomers has been removed. (C) and (D), top views of panels (A) and (B), respectively.

Figure 7. Current-voltage relationship of SARS-CoV E protein and inhibition by HMA.

(A) Example traces of current flowing through cells expressing SARS-CoV E protein, vector alone-transfected cells and untransfected HEK-293 cells. The cells were held at 0 mV and stepped to various potentials from −100 to 70 mV (in steps of 10 mV). (B) Whole-cell I–V curve in which peak current amplitudes were plotted against test potentials. Notice the significant large inward and outward currents recorded from SARS-CoV E protein, in contrast to vector alone and untransfected HEK-293 cell controls (*, two-tail unpaired T test, p