Solid-State NMR of Virus Membrane Proteins

ConspectusEnveloped viruses encode ion-conducting pores that permeabilize the host cell membranes and mediate the budding of new viruses. These viroporins are some of the essential membrane proteins of viruses, and have high sequence conservation, making them important targets of antiviral drugs. High-resolution structures of viroporins are challenging to determine by X-ray crystallography and cryoelectron microscopy, because these proteins are small, hydrophobic, and prone to induce membrane curvature. Solid-state NMR (ssNMR) spectroscopy is an ideal method for elucidating the structure, dynamics, and mechanism of action of viroporins in phospholipid membranes. This Account describes our investigations of influenza M2 proteins and the SARS-CoV-2 E protein using solid-state NMR.M2 proteins form acid-activated tetrameric proton channels that initiate influenza uncoating in the cell. 15N and 13C exchange NMR revealed that M2 shuttles protons into the virion using a crucial histidine, whose imidazole nitrogens pick up and release protons on the microsecond time scale at acidic pH. This proton exchange is synchronized with and facilitated by imidazole reorientation, which is observed in NMR spectra. Quantitative 15N NMR spectra yielded the populations of neutral and cationic histidines as a function of pH, giving four proton dissociation constants (pKa's). The pKa's of influenza AM2 indicate that the +3 charged channel has the highest time-averaged single-channel conductance; thus the third protonation event defines channel activation. In comparison, influenza BM2 exhibits lower pKa's due to a second, peripheral histidine, which accelerates proton dissociation from the central proton-selective histidine. Amantadine binding to AM2 suppressed proton exchange and imidazole reorientation, indicating that this antiviral drug acts by inhibiting proton shuttling. Solid-state NMR 13C–2H distance measurements revealed that amantadine binds the N-terminal pore of the channel near a crucial Ser31, whose mutation to asparagine causes amantadine resistance in circulating influenza A viruses. A second binding site, on the lipid-facing surface of the protein, only occurs when amantadine is in large excess in lipid bilayers. M2 not only functions as a proton channel but also conducts membrane scission during influenza budding in a cholesterol-dependent manner. Solid-state NMR distance experiments revealed that two cholesterol molecules bind asymmetrically to the surface of the tetrameric channel, thus recruiting the protein to the cholesterol-rich budding region of the cell membrane to cause membrane scission.To accelerate full structure determination of viroporins, we developed a suite of 19F solid-state NMR techniques that measure interatomic distances to 1–2 nm. Using this approach, we determined the atomic structures of influenza BM2, SARS-CoV-2 E, and EmrE, a multidrug-resistance bacterial transporter. pH-induced structural changes of these proteins gave detailed insights into the activation mechanisms of BM2 and E and the proton-coupled substrate transport mechanism of EmrE. The SARS-CoV-2 E protein forms pentameric helical bundles whose structures are distinct between the closed state at neutral pH and the open state at acidic pH. These 19F-enabled distance NMR experiments are also instrumental for identifying the binding mode and binding site of hexamethylene amiloride in E, paving the way for developing new antiviral drugs that target these pathogenic virus ion channels.