Structure of a vaccine candidate: Much effort is being targeted at developing vaccines that will provide protection against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). A trimeric spike protein that decorates the virus is a primary target of the host immune system and the focus of vaccine development. Bangaru et al. present the structure of a leading vaccine candidate: a full-length spike protein with some modifications aimed at enhancing stability that is formulated in polysorbate 80 detergent. The study confirms that the full-length immunogen is in a stable prefusion conformation and provides a basis for understanding immune responses to the vaccine. Science , this issue p. 1089
22 October 2020 in Proceedings of the National Academy of Sciences
Significance: The HIV-1 Env “glycan shield” masks the surface of the protein from immune recognition, yet intrinsic heterogeneity defies a typical structure–function description. Using an integrated approach of cryo-EM, computational modeling, and mass spectrometry, we visualized the glycan shield structure in a new light. Our approach facilitated development of cryo-EM analysis methods and allowed for validation of models against experiment. Comparison of Env expressed in different cell lines revealed how subtle differences in composition impact glycan shield structure and affect the accessibility of epitopes on the surface, providing insights for vaccine design. Finally, time-resolved cryo-EM experiments uncovered how highly connected glycan clusters help stabilize the prefusion trimer, suggesting the glycan shield may function beyond immune evasion.
Enveloped viruses hijack not only the host translation processes, but also its glycosylation machinery, and to a variable extent cover viral surface proteins with tolerogenic host-like structures. SARS-CoV-2 surface protein S presents as a trimer on the viral surface and is covered by a dense shield of N-linked glycans, and a few O-glycosites have been reported. The location of O-glycans is controlled by a large family of initiating enzymes with variable expression in cells and tissues and hence is difficult to predict. Here, we used our well-established O-glycoproteomic workflows to map the precise positions of O-linked glycosylation sites on three different entities of protein S—insect cell or human cell-produced ectodomains, or insect cell derived receptor binding domain (RBD). In total 25 O-glycosites were identified, with similar patterns in the two ectodomains of different cell origin, and a distinct pattern of the monomeric RBD. Strikingly, 16 out of 25 O-glycosites were located within three amino acids from known N-glycosites. However, O-glycosylation was primarily found on peptides that were unoccupied by N-glycans, and otherwise had low overall occupancy. This suggests possible complementary functions of O-glycans in immune shielding and negligible effects of O-glycosylation on subunit vaccine design for SARS-CoV-2.