Article Review: Cold sensitivity of Spike protein ectodomain
In January 2021, Robert J. Edwards et al. published an article in Nature Structural and Molecular Biology looking at the cold sensitivity of the SARS-CoV-2 recombinant Spike protein. Here we review their findings and consider the impact on future research.
SARS-COV-2 viral genetic material is able to enter human host cells through the use of spike ectodomain proteins, which docks to angiotensin converting enzyme 2 (ACE2) on host cells.
This spike protein is a trimeric class I fusion protein that exists in a metastable prefusion conformation. To fuse with the host cell membrane, the receptor-binding domain (RBD) of S1 subunit binds the ACE2 receptor, destabilising the prefusion trimer, resulting in the shedding of the S1 subunit, which generates the S2 subunit a stable postfusion conformation. Binding of the S1 subunit is dependent upon a hinge-like conformational change that either hides or exposes the determinants of receptor binding. The two conformations are receptor accessible ‘up’ state, which is thought to be less stable than the receptor-inaccessible ‘down’ state. This key function of the densely glycosylated spike protein (S-protein) has made it the primary target in the development of the majority of vaccines.
Stable constructs based on S ectodomain (2P constructs) have been designed to mimic spike protein behaviour with the ability to bind ACE2 and present epitopes for neutralising antibodies.
The 2P S construct designed by Edwards et al, comprises residues 1-1208 of SARS-CoV-2 and contains two proline substitutions at residues 986 and 987, a C-terminal foldon trimerization motif and a mutation that abrogates the furin-cleavage site.
Plasmids were transiently transfected into FreeStyle-293F cells and following expression, supernatant purified at room temperature via Streptavidin-tag purification and size exclusion chromatography (SEC).
Purified constructs were assessed for quality by sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE), SEC, differential scanning fluorimetry (DSF) and negative-stain electron microscopy (NSEM). These analyses showed batch to batch variability, caused by a fragile S ectodomain, despite the presence of two proline substitutes incorporated in the design, specifically to increase stability.
Using structural, biophysical and antigenic techniques, the team further investigated the properties of 2P S when stored under different temperatures.
Freshly prepared 2P S samples were assessed on the same day post purification and showed ~75% well-formed spikes, with characteristic kite-shaped morphology (NSEM). This fraction decreased after freeze-thaw and storage at room temperature. However, 1-week storage at room temperature led to 83% well-formed spikes. In comparison, storage at 4 °C for 5-7 days led to 95% degradation of spike constructs. However, structural integrity of the spike constructs could be regained by 3-hours incubation at 37°C. Longer incubation times led to slight aggregation, as seen via SDS-PAGE.
Reports by Zhou et al. suggested that prefusion-spike mechanics is pH dependent and Edwards team investigated if temperature-dependent pH changes were the cause of the 2P S denaturation. SEC purification was used to transfer 2P S from Tris (pH 8.0 at room temperature, but pH 8.6 at 4 °C) into MOPS buffer, which would be pH7.42 at 4 °C. After 4 °C storage, the spike fraction was reduced to 4% in MOPS, in comparison to 5% previously seen in Tris, suggesting that temperature change is the primary cause of 2P S denaturation. Equivalent results were seen when replicated with an acidic MES buffer.
To ensure that cold sensitivity was not an artifact of NSEM sample preparation, the team used DSF to measure transitions in the folding state of the protein. Distinct profile shifts in the inflection point (Ti), the temperature at which a transition occurs, were seen for 2P S samples stored at 4 °C, indicating lower protein stability in comparison to samples stored at 22 or 37 °C. Equally, Differential Scanning Calorimetry (DSC) confirmed storage at 4 °C destabilises 2P S compared to samples stored at 22 or 37 °C. Returning the destabilised 2P S to 37 °C for 3h did restore its stability, but a low-Tm suggested that recovery is only partial.
Ligand binding studies were performed on 2P S using surface plasmon resonance (SPR) and enzyme-linked immunosorbent assay (ELISA). Different binding profiles were seen for antibodies where the epitope was conformation specific. For example, antibody CR3022 showed higher binding at 4 °C, when the ‘up’ state is optimised, in comparison to freshly prepared protein.
These results indicate that 4 °C induced denaturation of the spike protein was associated with increased RBD-exposure of the S ectodomain.
To determine if a ‘down’ state stabilised S ectodomain would be resistant to cold-induced denaturation, Edwards et al. compared 2P S constructs with a rS2d-HexaPro variant, which showed more resistance to cold-induced denaturation in thermostability and binding studies and had 100% ‘down’ conformation present. Further evidence was presented by the analysis of two glycan-deleted mutations with altered RBD-up propensity, which showed a clear reduction of intact spike protein after 4 °C storage. It is believed that resistance to cold-induced denaturation is caused by the presence of inter-promoter disulphide linkages.
Careful consideration of this temperature sensitivity of the furin-cleavage-deficient SARS-CoV-2 S ectodomain is required to ensure consistent results in serology or binding studies using the 2P S or similar constructs. This includes assays designing Covid-19 vaccines to ensure accurate pre-clinical measures vaccine of effectiveness. The data also highlights the importance of correct storage techniques of purified protein.
Zhou, T. et al. Cryo-EM structures of SARS-CoV-2 spike without and with ACE2 reveal a pH-dependent switch to mediate endosomal positioning of receptor-binding domains. Cell Host Microbe (2020); https://doi.org/10.1016/j.chom.2020.11.004.