Fig. 5: Contribution of the expression level of known receptors to RBP cellular tropism.
From: Receptor-binding proteins from animal viruses are broadly compatible with human cell entry factors
a, Multiple regression analysis of pseudotype infectivity as a function of the RNA-seq expression levels of known viral receptors. For each virus, colours indicate the standardized regression coefficient. Significant coefficients are indicated with asterisks (ANOVA; SARS-CoV-2–ACE2, MERS-CoV–DPP4, measles–NECTIN4, Nipah–EFNB2: P < 0.0001; VEEV–LDLRAD3: P = 0.0056, HCV–CLDN1: P = 0.0042; Lassa–DAG1: P = 0.0120; LCMV–DAG1: P = 0.0103; Lujo–NRP2: P = 0.0001; Sin Nombre–PCDH1: P = 0.0320; Mengla–NPC1: P = 0.0193; Cedar–EFNB1: P = 0.0032; Cedar–EFNB2: P = 0.0004; VSV–LDLR: P = 0.0166; VSV–MSR1: P = 0.0330). Lectins and proteins that directly or indirectly attach to lipids were not included because we focused on proteins interacting with RBPs. For the rabies virus pseudotype, the reads per kilobase per million (rpkm) values of all genes encoding cholinergic receptor nicotinic (nAChRs) Alpha or Beta subunits were summed. Scatter plots for several significant cases are shown in Extended Data Fig. 4. b, Effect of receptor overexpression on viral entry. Cell lines were transfected with a receptor-encoding plasmid or an empty vector and then inoculated with pseudotypes. The most appropriate cell line was chosen for each pseudotype–receptor pair based on the criteria detailed in the Methods section. These were SK-OV-3 for HCV–SCARB1, MCF7 for LCMV–DAG1 and Lujo–NRP2, ACHN for Cedar–EFNA2 and Cedar–EFNB1, UACC-62 for SFTSV–MYH9, and EKVX for rabies–NCAM1. The percentage of infection at 24 h post inoculation is shown. Two technical replicates (n = 2) were carried out for each assay. Data points shown in blue and red correspond to receptors with non-significant and significant effects on pseudotype infectivity according to the multiple linear regression shown above, respectively. Experimental results were compared with the empty controls using two-sided t-tests. **P < 0.01; NS, not significant (Lujo: P = 0.0051, Cedar–EFNB1: P = 0.0044). Representative images are shown on the right. Scale bar, 400 µm.
