Crystal structure of the receptor binding domain of the spike glycoprotein of human betacoronavirus HKU1.

Cell lines

HRT18 (human rectal tumor cell line) and 293 T (human embryonic kidney 293 cell line transformed with SV40 large T antigen) were obtained from ATCC (Manassas, VA, USA), and maintained in DMEM with 10% fetal bovine serum and 2% penicillin, streptomycin and Fugizone (Life Technologies Inc). Primary HTBE cells were obtained from LifeLine Cell Technology (Frederick, MD, USA) and cultured in BronchiaLife Complete Medium (BronchiaLife Basal Medium with BronchiaLife B/T LifeFactors, LifeLine Cell Techonology, Frederick, MD, USA). Till 80-90% confluence, cells were lifted with brief trypsin digestion and plated on 12-well Corning Transwell plates (collagen-coated permeable, 0.4 μm, St Louis, MO, USA) at a density of 2 × 10 5 cells per well. After 48 hrs incubation, cells were switched to differentiation medium (1:1 ratio of BronchiaLife medium and DMEM high-glucose medium (Invitrogen) with the addition of 1.1 mM CaCl 2 and 25 nM retinoic acid) and maintained for 3 weeks in differentiation media at an air-liquid interface. All cell lines were tested for mycoplasma contamination using Mycoplasma Detection Kt (Macgene, Beijing, China) and they were mycoplasma-free.

Constructs and mutagenesis

The full-length, codon-optimized genotype A HKU1 spike gene (Accession #: AY597011 ) preceded by a Kozak sequence was synthesized by GenScript (Piscataway, NJ) and cloned into pcDNA3.1(+) (Invitrogen) between HindIII and XbaI sites for eukaryotic expression. The pcDNA-HKU1A S plasmid was served as a template for the deletion constructs. A series of deletion constructs encoding HKU1 spike protein residues 14-294 (NTD), 14-687, 14-755 (S1), 295-755, 310- 677 (1A-S310-677aa), 307-687, 353-687 and 384-687 were amplified by PCR using primers listed in Supplementary Table 1 and inserted into pFastBac-1 with a hemo signal peptide at the N-terminus and a His tag at the C-terminus. Codon-optimized genotype B HKU1 S1 gene (Access#: AGT17758.1 ) preceded by a Kozak sequence ( Supplementary Table 2 ) was also synthesized by GenScript (Piscataway, NJ) and cloned into pcDNA3.1 (pcDNAHKU1B S1). A DNA fragment encoding HKU1B S310-676aa (residues 310–676) was amplified by PCR using pcDNAHKU1B S1 as the template and cloned into pFastBac-1 for insect cell expression. All mutagenesis was carried out using Q5 mutagenesis kit (NEB, MA, USA). After the entire coding sequences were verified by sequencing, the sequences were cloned back into the pFastBac-1. Protein expression and crystallization Briefly, 1A-S310-677aa and other HKU1 S proteins containing an N-terminal hemo signal peptide and a C-terminal 6His tag were expressed in insect cells using the Bac-to-Bac expression system (Invitrogen). The proteins were purified from the cell supernatantusing nickel-nitrilotriacetic acid (Ni-NTA) affinity chromatography and gel-filtration chromatography using a Superdex 200High Performance column (GE Healthcare). The purified HKU1-S1protein was concentrated to 6 mg ml −1 and stored in buffer containing 20 mM Tris (pH8.0) and 100 mM NaCl. Crystallization of 1A-S310-677aa was achieved at 22 °C using the sitting drop vapour diffusion method with 0.8 μl of protein and 0.8 μl reservoir buffer containing 0.2 M MgCl 2 , 30% (w/v) polyethylene glycol (PEG) 4000, and 0.1 M Tris (pH8.5). The crystals were harvested in reservoir buffer with 8.8% (W/V) pentaerythritol ethoxylate (3/4 EO/OH) and 20% sucrose and flash-frozen in liquid nitrogen.

Structure determination and refinement

To determine the crystal structure of HKU1 1A-S310-677aa, different techniques, including molecular replacement, selenomethionine derivative and heavy atom soaking techniques, were used to solve the phase problem of 1A- S310-677aa, but attempts were unsuccessful. However, 1A-S310-677aa was cysteine-rich and contained 24 cysteines. Its Bijvoet ratio is approximately 1.77% at 5.96 keV, making the 1A-S310-677aa crystal a suitable candidate for native SAD phasing. The diffraction data collection and structure determination using the native SAD method have been described elsewhere 34 . Briefly, all diffraction data were collected at the macromolecular crystallography super-bending magnet beamline X06DA (PXIII) at the Swiss Light Source at an X-ray energy of 6 keV (2.0663 Å wavelength) using the Pilatus 2M-F detector. Sixteen datasets were collected from two isomorphous crystals at a size of 90 × 50 × 20 μm 3 . Each dataset consisted of 1,800 images collected with 60% transmission (flux of ∼0.9 × 10 10 photons per second), and all diffraction data were indexed and integrated using XDS software 36 . Scaling was performed with XSCALE. The data quality was assessed using PHENIX.XTRIAGE 37 . The substructures were determined using SHELXD in 1,000 trials with a cutoff of 3.2 Å. Phasing was performed using SHELXE with combined chain tracing and density modification. From experimentally phased maps, initial models were built with BUCCANEER and PHENIX.AUTOBUILD 37 , and refined by manual model building with COOT 38 . The final dataset had an anomalous signal extending to ∼3.1 Å and the overall resolution was better than 2.3 Å (CC1/2 of highest resolution shell: 89.4%). Final models were obtained using datasets collected at higher energy (1.00 Å) with a resolution at 1.9 Å and by molecular replacement using the initial model generated from the native SAD datasets. ELISA To coat the plate, 0.5 μg of purified proteins were added into each well of Immulon 2HB plates (Thermo, Rochester, NY, USA) and incubated for overnight at 4 °C. After washing three times with washing buffer (PBST, PBS+0.05% Tween 20), the wells were blocked with blocking buffer (3% BSA in PBST) for 1 hr at room temperature (RT). After washing three times with PBST, fourfold serially diluted antibodies in blocking buffer were added into each well and incubated at RT for 1 h. The mAb bound to the truncated S proteins was detected using an horseradish peroxidase-conjugated goat anti-mouse Ig (Cat# 115-035-166, Jackson ImmunoResearch, West Grove, PA, USA) at a dilution of 1:5,000. After 1 h incubation and washing three times with PBST, 100 μl of o -phenylenediamine dihydrochloride (Beijing Wantai Biological Pharmacy Enterprise Co, Beijing, China) was added to each well and incubated for 15 min. The reaction was then stopped by addition of 2 M sulfuric acid, and the optical density at 492 nm was measured using a MultiSkan MK3 plate reader (Thermo, Rochester, NY, USA). Blockade of HKU1 virus entry by soluble HKU1 S proteins Prior to virus infection, apical surface of differentiated HTBE cells were washed twice with DMEM+1% BSA and incubated with various amounts of soluble, truncated HKU1 S proteins for 1 h at 37 °C. After removal of the soluble S protein, cells were then inoculated with 150 μl of passage three of the amplified HKU1 (#21) virus diluted with an equal volume of each HKU1 S protein onto the apical surface for 4 h. After washing three times with DMEM+1%BSA to remove unbound viruses, cells were incubated at 34 °C. HKU1 viruses were harvested by washing once with 150 μl of DMEM+1%BSA at 24 and 48 h post inoculation, and cells were fixed at 48 h post-inoculation and assayed for infection by immunofluorescence using a rabbit 1814 polyclonal antibody to HKU1 S protein and FITC-conjugated goat anti-rabbit IgG (Cat# 111-095-144, Jackson ImmunoResearch, West Grove, PA, USA). Real time PCR Viral RNAs were extracted from 140 μl of virus-containing apical wash using a QIAamp viral mini kit from Qiagen, according to manufacturer’s protocol, and real-time PCR was performed using RNA Ultrasense One-Step qRT-PCR from Invitrogen targeting the N gene of HKU1. About 10 μl of viral RNA extract was mixed with 10 μl of master mix containing 2.3 μl of H 2 O, 1 μl of enzyme, 4 μl of 5 × buffer, 0.2 μl of 10 μM probe (5′-TTGAAGGCTCAGGAAGGTCTGCTTCTAA-3′), 1 μl of 10 μM primer HKU1qPCR-F (5′-CTGGTACGATTTTGCCTCAA-3′), 1 μl of 10 μM primer HKU1qPCR-R (5′-ATTATTGGGTCCACGTGATTG-3′), and 0.5 μl of MgSO 4 . Following an incubation at 42 °C for 5 min and a denaturation step of 5 min at 95 °C, 40 cycles of amplification were performed for 3 s at 95 °C and 30 s at 60 °C on a Roche Light Cycler 480 machine. A serially diluted synthetic DNA fragment containing the sequence from 28,849 to 28,949 of the HKU1 genome was used as the quantitative standard to calculate the viral genome copy number in the samples for real-time PCR.

Data availability

The coordinates and structure factors have been deposited in the Protein Data Bank under the accession codes 5GNB and 5KWB. The GenBank accession codes AGT17758.1 , AY597011 and PDB accession codes 3JCL, 4L72, 4QZV, 5GYQ and 2AJF were used in this study. All other data are available from the corresponding authors upon reasonable request.

Supplementary Material Supplementary Information Supplementary figures and supplementary tables. Supplementary Data 1 Glycan screening array to identify the sugar binding to 1A-S310-677aa. Peer review file