Techniques and tactics used in determining the structure of the trimeric ebolavirus glycoprotein

Abstract

The trimeric membrane-anchored ebolavirus envelope glycoprotein (GP) is responsible for viral attachment, fusion and entry. Knowledge of its structure is important both for understanding ebolavirus entry and for the development of medical interventions. Crystal structures of viral glycoproteins, especially those in their metastable prefusion oligomeric states, can be difficult to achieve given the challenges in production, purification, crystallization and diffraction that are inherent in the heavily glycosylated flexible nature of these types of proteins. The crystal structure of ebolavirus GP in its trimeric prefusion conformation in complex with a human antibody derived from a survivor of the 1995 Kikwit outbreak has now been determined [Lee et al. (2008), Nature (London), 454, 177-182]. Here, the techniques, tactics and strategies used to overcome a series of technical roadblocks in crystallization and phasing are described. Glycoproteins were produced in human embryonic kidney 293T cells, which allowed rapid screening of constructs and expression of protein in milligram quantities. Complexes of GP with an antibody fragment (Fab) promoted crystallization and a series of deglycosylation strategies, including sugar mutants, enzymatic deglycosylation, insect-cell expression and glycan anabolic pathway inhibitors, were attempted to improve the weakly diffracting glycoprotein crystals. The signal-to-noise ratio of the search model for molecular replacement was improved by determining the structure of the uncomplexed Fab. Phase combination with Fab model phases and a selenium anomalous signal, followed by NCS-averaged density modification, resulted in a clear interpretable electron-density map. Model building was assisted by the use of B-value-sharpened electron-density maps and the proper sequence register was confirmed by building alternate sequences using N-linked glycan sites as anchors and secondary-structural predictions.

Figure 1

ZEBOV GP construct design. (a) ClustalW alignment of the glycoprotein primary sequence from the Zaire, Sudan, Ivory Coast and Reston ebolavirus species. Consensus secondary-structural prediction, using the Network Protein Sequence Analysis server (Combet et al., 2000 ▶), is shown for Zaire ebolavirus. Helices and β-strands are shown as coils and arrows, respectively. N- and O-linked glycans predicted by the NetNGlyc (Gupta et al., 2004 ▶) and NetOGlyc servers (Julenius et al., 2005 ▶) are shown by Y and lollipop symbols, respectively. (b) Disordered domain prediction for ZEBOV GP using the DISOPRED2 server (Ward et al., 2004 ▶). (c) Schematic representation of the ZEBOV GP and GP variants generated for expression screening. For clarity, we show only selected examples of mucin-like domain deletions. Disulfide bridges (-S-S-), signal peptide (SP), internal fusion loop (IFL), heptad-repeat region 1 (HR1), heptad-repeat region 2 (HR2), membrane-proximal external region (MPER), transmembrane anchor (TM) and cytoplasmic tail are labeled accordingly. N- and O-linked glycans are shown as red-colored Ys and lollipops, respectively.

Figure 2

Expression, purification and crystallization of ZEBOV GPΔmuc_312–463Δtm. (a) Small-scale expression screening of selected GP truncation variants using a transient transfection HEK293T system. Conditioned media containing ZEBOV GP variants were harvested 4 d post-transfection and 10 µl supernatant was separated by nonreducing SDS–PAGE and probed with anti-HA (linear) or KZ52 (conformational) primary monoclonal antibodies. (b) ZEBOV GPΔmuc_312–463Δtm–KZ52 Superdex 200 10/300 GL chromatogram. (c) The top two selected free-interface diffusion crystallization hits. Crystal form A [OptiMix1 No. 6; 10% ethylene glycol, 10%(w/v) PEG 10 000 and 0.6 M sodium acetate] and crystal form B [OptiMix2 No. 74; 10%(w/v) PEG 6000, 0.1 M PIPES pH 6.5 and 0.6 M NaI] are shown. (d) Various crystal morphologies of ZEBOV GPΔmuc_312–463Δtm–KZ52 were obtained by hanging-drop vapour diffusion in various precipitants and additives and at various pH values: (i) rod crystals, pH 4.2, (ii) pyramidal crystals, pH 4.8, (iii) rhomboid crystals, pH 6.5, (vi) rod crystals, pH 5.0, with cetyl-trimethylammonium bromide (CTAB) additive, (v) rod crystals with cesium chloride and (vi) trapezoidal crystals with ethylene glycol. (e) A single crystal (∼0.2 × 0.2 × 0.2 mm) grown in 8.5%(w/v) PEG 10 000, 0.4 M sodium acetate, 0.1 M Tris–HCl pH 8.5 and 10%(v/v) PEG 200 was washed three times in mother liquor and then dissolved in nonreducing SDS–PAGE sample buffer. The washes (lanes 1–3), dissolved crystal (lane 4), Fab KZ52 (lane 5), PNGaseF-treated GPΔmuc_312–463Δtm (lane 6) and PNGaseF-treated GPΔmuc_312–463Δtm–KZ52 (lane 7) were analyzed by silver-stained SDS–PAGE. Note that Fab KZ52 migrates at a lower molecular weight (∼40 kDa) than the expected 50 kDa for typical antibody fragments and also exists as two isoforms (doublet bands) prior to MonoS ion-exchange purification.

Figure 3

Control of N-linked glycosylation. (a) Ni-affinity purified GPΔmuc_312–463Δtm produced in T. ni insect cells (High Five cells). A fraction of the sample remains as a glycoprotein trimer (∼150 kDa) even in the presence of SDS. (b) Nonreducing Western blots of kifunensine-treated HEK293T-produced GPΔmuc_312–463Δtm. Lane A, HEK293T-expressed fully glycosylated GPΔmuc_312–463Δtm; lane B, kifunensine-treated (1 µg ml⁻¹ final concentration) GPΔmuc_312–463Δtm. The effect of the inhibitor on glycosylation was monitored by nonreducing immunoblot probed with a KZ52 primary mAb. (c) Nonreducing Western blots of selected single-site, double-site and triple-site sugar mutants. All samples were deglycosylated using PNGaseF and probed with a KZ52 primary antibody. The numbers on the gel refer to the sites of threonine residues in the N-X-T/S glycosylation motif that were mutated to valine. (d) Coomassie-stained SDS–PAGE analysis of the effect of urea-assisted PNGaseF deglycosylation. Lane A, HEK293T-expressed fully glycosylated GPΔmuc_312–463Δtm; lane B, PNGaseF-deglycosylated T42V/T230V GPΔmuc_312–463Δtm; lane C, PNGaseF deglycosylation of T42V/T230V GPΔmuc_312–463Δtm in the presence of 2.0 M urea.

Figure 4

Diffraction of ZEBOV T42V/T230V GPΔmuc_312–463Δtm–SeMet KZ52 crystals. (a) Translated T42V/T230V GPΔmuc_312–463Δtm–SeMet Fab KZ52 crystals obtained by hanging-drop vapor diffusion in 8.5%(w/v) PEG 10 000, 0.4 M sodium acetate, 0.1 M Tris–HCl pH 8.5 and 10%(v/v) PEG 200. (b) Diffraction image of GPΔmuc_312–463Δtm–SeMet KZ52 collected on beamline 5.0.2, Advanced Light Source (Berkeley, California, USA) using an ADSC Quantum 315 CCD detector. Reflections are clearly visible to at least 3.5 Å resolution (inset box).

Figure 5

Molecular replacement using an improved search model: Fab KZ52. (a) Comparison of the crystal structures of the uncomplexed (blue; PDB code 3inu) and ZEBOV T42V/T230V GPΔmuc_312–463Δtm-bound Fab KZ52 (green; PDB code 3csy). The ZEBOV T42V/T230V GPΔmuc_312–463Δtm-mediated induced fit of side-chain and main-chain residues in the CDR L1 and CDR H3 regions are shown in the inset boxes. (b) Ab initio SAXS reconstruction of the T42V/T230V GPΔmuc_312–463Δtm–KZ52 molecular envelope docked with three Fabs (shown as blue, red and green ribbons). The knob-like structure in the middle of the SAXS molecular envelope and the space between the molecular-replacement arrangements of Fab KZ52, as shown by the dashed outline, are likely to correspond to the GP. (c) Molecular-replacement solution showing the arrangement of three Fab KZ52 molecules on the crystallographic threefold axis. The location of the threefold symmetry axis in both models is shown by the black triangle.

Figure 6

Initial electron-density maps. (a) Initial experimental histogram-matched density-modified electron-density map calculated in SOLOMON (Abrahams & Leslie, 1996 ▶). (b) Histogram-matched and NCS-averaged density-modified (1500 cycles) electron-density map calculated in DM (Cowtan, 1994 ▶). (c) Final simulated-annealed σ_A-weighted 2mF _o − DF _c electron-density map. The final refined model of ZEBOV GPΔmuc_312–463Δtm–SeMet Fab KZ52 (PDB code 3csy) is superimposed onto each of the electron-density maps. All electron-density maps are contoured at 1σ.

Figure 7

Flowchart of the cross-phasing model-building procedure.

Figure 8

B-value-sharpened electron-density maps. B-value sharpening reveals additional side-chain electron-density features. A series of B-value-sharpened electron-density maps (B _sharp = −50, −75, −100, −150 and −200 Ų) were generated in FFT. For comparison, the initial density-modified (DM) and final σ_A-weighted 2mF _o − DF _c electron-density maps are shown. All electron-density maps are superimposed with the final refined ZEBOV T42V/T230V GPΔmuc_312–463Δtm–SeMet Fab KZ52 coordinates (PDB code 3csy). The B _sharp = −75 Ų and B _sharp = −100 Ų electron-density maps were used to assist with model building, as these maps have improved side-chain densities and minimal background noise.

Figure 9

Structure of ZEBOV T42V/T230V GPΔmuc_312–463Δtm-SeMet Fab KZ52. The molecular surface of the ZEBOV T42V/T230V GPΔmuc_312–463Δtm-SeMet Fab KZ52 trimer is viewed from its side. Three lobes of GP1 (shades of blue) form a chalice and three GP2 subunits (white) wrap around GP1, forming a cradle. The three molecules of Fab KZ52 are depicted as yellow ribbons. This figure was adapted from the cover illustration in Lee et al. (2008 ▶).

Figure 10

Flowchart of suggested tactics and strategies for tackling a challenging new viral or human glycoprotein.