Abstract
The surface spike (S) glycoprotein mediates cell entry of SARS-CoV-2 into the host through fusion at the plasma membrane or endocytosis. Omicron lineages/sublineages have acquired extensive mutations in S to gain transmissibility advantages and altered antigenicity. The fusogenicity, antigenicity, and evasion of Omicron subvariants have been extensively investigated at unprecedented speed to align with the mutation rate of S. Cells that overexpress receptors/cofactors are mostly used as hosts to amplify infection sensitivity to tested variants. However, systematic cell entry comparisons of most prior dominant Omicron subvariants using human lung epithelium cells are yet to be well-studied. Here, with human bronchial epithelium BEAS-2B cells as the host, we compared single-round virus-to-cell entry and cell-to-cell fusion of Omicron BA.1, BA.5, BQ.1.1, CH.1.1, XBB.1.5, and XBB.1.16 based upon split NanoLuc fusion readout assays and the S-pseudotyped lentivirus system. Virus-to-cell entry of tested S variants exhibited cell-type dependence. The parental Omicron BA.1 required more time to develop full entry to HEK293T-ACE2-TMPRSS2 than BEAS-2B cells. Compared to unchanged P681, S-cleavage constructs of P681H/R did not have any noticeable advantages in cell entry. Omicron BA.1 and its descendants entered BEAS-2B cells more efficiently than D614G, and it was slightly less or comparable to that of Delta. Serine protease-pretreated Omicron subvariants enhanced virus-to-cell entry in a dose-dependent manner, suggesting fusion at the plasma membrane persists as a productive cell entry route. Spike-mediated cell-to-cell fusion and total S1/S2 processing of Omicron descendants were similar. Our results indicate no obvious entry or fusion advantages of recent Omicron descendants over preceding variants since Delta, thus supporting immune evasion conferred by antigenicity shifts due to altered S sequences as probably the primary viral fitness driver.