Structure and Inhibition of the SARS Coronavirus Envelope Protein Ion Channel

Pervushin, Konstantin; Tan, Edward; Parthasarathy, Krupakar; Lin, Xin; Jiang, Feng Li; Yu, Dejie; Vararattanavech, Ardcharaporn; Soong, Tuck Wah; Liu, Ding Xiang; Torres, Jaume

The accession numbers for the proteins in this paper are SARS-CoV E, NP_828854; TGEV E, AAZ91440; HCoV-229E E, NP_073554; MHV E, O72007 and IBV E, P05139. Figure S1 Building the ETM a-helical pentameric bundle. The skeleton of the ETM bundle (A) was based on orientational data from site specific infrared dichroism [38] . The ETM monomer built from NMR data was superimposed onto the skeleton (B) to obtain the full atom description of the model (C Figure S2 (A-B) Same as Fig. 1 (A-B) , for ETM in the presence of 10 mM HMA. The 2D NOESY spectrum is not shown due to large interference from HMA, causing spectral overlap. (C-E) Same as Fig. 1 (A-C) for ETM in the presence of 100 mM AMT. Figure S7 Clustal X sequence alignment of envelope proteins in coronaviruses, up to the totally conserved Pro residue (P54 in SARS-CoV E), corresponding to SARSCoV sequences (group 2b, black), group 1 sequences (blue), group 2 sequences (red) and group 3 sequences (green). The accession numbers are indicated next to the common name, on the left. The positions N15 and R38 in the SARS-CoV E sequence are indicated above, and the residues that are exposed to the lumen of the pore in ETM are shown with a yellow background. The locations of the two HMA binding sites in the ETM channel are indicated by a red arrow. Found at: doi:10.1371/journal.ppat.1000511.s007 (1.86 MB TIF) Figure S8 Scheme of 2D 1 H N , 1 H aromatic band-selected NOESY, an experiment suitable for detection of 1 H N , 1 H aromatic resonances in membrane proteins in the presence of strong aliphatic resonances of solubilizing detergents. Longitudinal relaxation acceleration scheme [72] prevents saturation of longitudinal magnetization of aliphatic spins and water building up during the mixing period t mix . This magnetization is used to accelerate relaxation of amide and aromatic protons to steadystate Boltzmann thermal equilibrium during the inter-scan delay d 1 . The radiofrequency pulses on 1 H, 15 N, 13 C are applied at 4.7, 118 and 40 ppm, respectively. Narrow and wide black bars indicate non-selective p/2 and p rf-pulses applied with the phase x unless indicated otherwise. Complex shapes on the line marked 1 H indicate the 1 H N , 1 H aromatic band-selective 1.5 ms excitation E-Burp2 pulses with the phase Q 3 and Q 4 and cB 1 = 2733 Hz and the 1.8 ms refocussing Re-Burp pulse [73] with the phase Q 4 and cB 1 = 3050 Hz. The center of the excitation of all 1 H N , 1 H aromatic bandselective pulses is placed at 8.5 ppm. The durations and strengths of the pulsed magnetic field gradients (PFG) applied along the z-axis are selected as G 1 : 500 ms, 80 G/cm; G 2 : 900 ms, 60 G/cm; G 3 : 900 ms, 70 G/cm. Two datasets, with and without 13 C-composite inversion pulse decoupling pulse are acquired using the phases Q 1 = x; Q 2 = x; Q 3 = x, y); Q rec = (x, 2x). The quadrature detection in t 1 dimension is achieved by the States-TPPI method [74] applied to Q 1 . Subtraction of the datasets results in a 2D NOESY spectrum containing NOEs stemming from 1 H covalently bound to 13