Fig 1.
Genome-wide CRISPR screen identifies HRTV entry factors.
A. Schematic of genome-wide CRISPR/Cas9 KO screen in 293T cells. Figure created with BioRender.com. B. Volcano plot analysis of the CRISPR/Cas9 KO screen results. Each dot represents a unique gene identified from sequencing results. The labeled genes are involved in the glycosphingolipid biosynthesis pathway. C. Bubble plot showing the pathway analysis of the statistically enriched genes in the screen. Bubble size and color reflect the statistical significance and the number of genes enriched in the relevant pathway.
Fig 2.
Glucosylceramide biosynthesis pathway is essential for HRTV infection.
A. Diagram of glycosphingolipid biosynthesis pathway. Genes highlighted in red were identified from CRISPR/Cas9 screen. B-C. Effect of NB-DNJ or UGT8-IN on HRTV replication. 293T cells were pre-treated with NB-DNJ or UGT8-IN for 24 hours before infection with HRTV (MOI = 1). Viral RNAs of the S-segment and M-segment were measured by qPCR and normalized to GAPDH mRNA. Data shown are means ± SEM from representative experiments (n = 3 technical replicates). P values were determined by ordinary one-way ANOVA with Dunnett’s multiple comparison tests. ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05. D-E. Effect of NB-DNJ and UGT8-IN on HRTV or rVSV-DBV infection. To generate a replication-competent rVSV-DBV, the VSV glycoprotein in its genome was replaced with DBV Gn/Gc. After incubation with indicated drugs for 24 hours, half-log serially diluted HRTV or rVSV-DBV was added to 4 replicate wells of a 96-well plate containing 293T cells for 72 hours. TCID50 was calculated by plotting cell viability versus dilution factor using GraphPad Prism and fitted by a dose-response (variable slope) algorithm. F. Heat map of the levels of intracellular GalCer and GlcCer in 293T cells with or without NB-DNJ treatment.
Fig 3.
HRTV requires GlcCer for efficient infection.
A-B. Effect of SPTLC2-, UGCG- or UGT8-deficiency on HRTV replication. Control and KO 293T cells were infected with HRTV (MOI = 1) for the indicated times for qPCR analysis. Knockout efficiency of SPTLC2-, UGCG- or UGT8 gene was measured by qPCR. Data shown are means ± SEM from representative experiments (n = 3 technical replicates). P values were determined by multiple unpaired t-tests (A) or two-way ANOVA with Dunnett’s multiple comparison tests (B). ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05. C. Heat map of the levels of intracellular GalCer and GlcCer in SPTLC2-, UGCG-, or UGT8-deficient 293T cell lines. D-E. TCID50 of HRTV and rVSV-DBV in control, SPTLC2-, UGCG-, or UGT8-deficient 293T cells. Half-log serially diluted HRTV or rVSV-DBV was added to 4 replicate wells of a 96-well plate for 72 hours before luminescence measurement.
Fig 4.
Glycosphingolipid downstream of GlcCer is not required for HRTV infection.
A-B. Effect of B4GALT5-, B4GALT6- or B4GALT5/6 double-deficiency on HRTV replication. Control and KO 293T cells were infected with HRTV (MOI = 1) for 24 hours, followed by qPCR analysis. Data shown are means ± SEM from representative experiments (n = 3 technical replicates). P values were determined by ordinary one-way ANOVA with Dunnett’s multiple comparison tests. ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05. C. Heat map of the levels of intracellular GalCer and GlcCer in control or B4GALT5/6 double knockout 293T cell lines.
Fig 5.
Supplementing with GlcCer enhances HRTV infection.
A-B. Exogenous glycosphingolipid supplement for HRTV replication in SPTLC2 or UGCG KO 293T cells. Cells were cultured in media supplemented with 0–50 μM indicated glycosphingolipid for 24 hours before HRTV infection (MOI = 1). Viral RNA levels were measured at 24 hours post-infection. Control cells without glycosphingolipid supplement was shown for comparison. Data shown are means ± SEM from representative experiments (n = 3 technical replicates). P values were determined by ordinary one-way ANOVA with Dunnett’s multiple comparison tests. ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05. C-D. TCID50 of HRTV and rVSV-DBV in SPTLC2 knockout cell with GalCer or GlcCer supplement. Cells were cultured in media supplemented with 50 μM indicated glycosphingolipid for 24 hours before HRTV or rVSV-DBV infection.
Fig 6.
GlcCer is critical for HRTV-induced membrane fusion.
A. Effect of SPTLC2-, UGCG- or UGT8-deficiency on HRTV binding and internalization. For binding assay, control and KO HeLa cells were incubated with HRTV particles at 4°C for 1 hour. For internalization assay, after 4°C incubation, cells were washed by PBS and then moved to 37°C for additional 2 hours before qPCR assay. Data shown are means ± SEM from representative experiments (n = 3 technical replicates). P values were determined by multiple unpaired t-tests. *, P < 0.05. B-C. Subcellular fractionation of HRTV-infected control or UGCG-KO cells. Cells were treated as described in Materials and Methods, and the PNS was fractionated by Optiprep gradient centrifugation. PNS and each fraction were subjected to immunoblotting analysis with the indicated antibodies (B). Total RNAs were extracted from PNS and each fraction for qPCR assay (C). Data shown are means ± SEM from representative experiments (n = 3 technical replicates). P values were determined by multiple unpaired t-tests. ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05. D. Effect of UGCG-deficiency on HRTV trafficking. Control and UGCG-KO 293T cells were infected with HRTV at MOI = 50 for 4 hours. Cells were processed for thin-section electron microscopy according to standard procedures [47]. Bars, 200nm. E-F. Effect of SPTLC2-, UGCG- or UGT8-deficiency on HRTV-induced membrane fusion. HeLa cells were infected with DiOC18/R18-labeled HRTV for 30 min at 4°C and then incubated at 37°C for 2 hours (E). The numbers of fluorescent DiOC18 puncta were quantified by fluorescence microscopy (F). P values were determined by ordinary one-way ANOVA with Dunnett’s multiple comparison test (n = 15). ****, P < 0.0001; *, P < 0.05. G-H. Effect of exogenous GlcCer and GalCer supplement of SPTLC2-KO cells on HRTV-induced membrane fusion. HeLa cells were cultured in media supplemented with 50 μM NBD-labeled GlcCer or GalCer for 24 hours. Cells were subsequently incubated with R18-labeled HRTV for 30 min at 4°C and then for 2 hours at 37°C (G). The numbers of fluorescent R18 puncta were quantified by fluorescence microscopy (H). P values were determined by ordinary one-way ANOVA with Dunnett’s multiple comparison test (n = 10) ****, P < 0.0001.
Fig 7.
HRTV glycoproteins are responsible for GlcCer-mediated membrane fusion.
A. Effect of SPTLC2-, UGCG- or UGT8-deficiency on rVSV-HRTV, rVSV-DBV, and rVSV-G entry. Control and KO 293T cells were infected with the indicated virus for 24 hours, followed by luciferase assay. Luminescence readings were normalized to the respective cell numbers. B. Rescue of rVSV-HRTV and rVSV-DBV entry in UGCG KO cells. Control and UGCG-KO 293T cells were cultured in media supplemented with 50 μM indicated glycosphingolipid for 24 hours, followed by infection with the indicated virus for additional 24 hours for luciferase assay. C. Effect of NB-DNJ treatment on rVSV-HRTV, rVSV-DBV, and rVSV-G entry. 293T cells were treated with indicated concentrations of NB-DNJ for 24 hours, followed by infection with the indicated virus for 24 hours prior to luciferase assay. All the data shown are means ± SEM from representative experiments (n = 3 technical replicates). P values were determined by ordinary one-way ANOVA with Dunnett’s multiple comparison tests. ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05.
Fig 8.
Replication-competent HRTV or DBV pseudotyped viruses require GlcCer for efficient entry.
A. Diagram of replication-competent rVSV-DBV genome. B. Replication-competent rVSV-HRTV entry in SPTLC2-, UGCG- or UGT8-deficiency cells. Control and KO 293T cells were infected with replication-competent rVSV-HRTV for 16 hours, followed by flow cytometry. C. Rescue of replication-competent rVSV-HRTV entry in SPTLC2-, UGCG- or UGT8-deficiency cells. Control and UGCG-KO 293T cells were cultured in media supplemented with 50 μM DMSO or GlcCer for 24 hours, followed by infection with replication-competent rVSV-HRTV for 16 hours prior to flow cytometry assay. D. Replication-competent rVSV-DBV entry in SPTLC2-, UGCG- or UGT8-deficiency cells. E. Rescue of replication-competent rVSV-DBV entry in SPTLC2-, UGCG- or UGT8-deficiency cells. The experiments were performed similarly as described above.
Fig 9.
The HRTV Gc protein is associated with GlcCer in the lipid bilayer.
A. Induced-fit docking pose of GlcCer with HRTV Gc protein. HRTV Gc ectodomain (PDB: 5YOW) is represented as ribbons and colored by domains (red, domain I; yellow, domain II; blue, domain III). The fusion loops at the tip of domain II are displayed in orange. The bound head group of GlcCer (framed) is drawn as sticks and colored according to atom type (carbon, green; nitrogen, blue; oxygen, red). B. Interaction of GlcCer with HRTV Gc protein in the lipid-head-group binding pocket. The hydrogen bonds between GlcCer and Gc protein are represented as dashed lines and colored in magenta. C. GlcCer binding groove. The HRTV Gc lipid-head-group binding pocket is shown in surface representation and colored according to electrostatic potential. D. 2D representation of docking interaction of GlcCer with binding site residues of HRTV Gc. The hydrogen bonds are shown in magenta lines. E. HRTV Gc protein directly binds to DOPC, GalCer, and GlcCer. Serially diluted DOPC, GalCer, and GlcCer were spotted onto the PVDF membrane. Protein-lipid overlay assay was performed with an anti-6xHis antibody. F. SPR assay of HRTV Gc binding kinetics to the indicated liposomes at neutral pH. All experiments were performed as duplicates, and the calculated fit is shown as a black line. Equilibrium constants (KD) are determined by measuring the association rate constant (Ka) and dissociation rate constant (Kd). G. Binding kinetics of HRTV Gc to the indicated liposomes at acidic pH. Sensorgrams measured in replicates by bio-layer interferometry (BLI) and fitting (black) of the interaction of Gc to the indicated liposomes.
Fig 10.
The D841 residue of Gc protein is crucial for its binding to lipid head group.
A. SPR sensorgrams showing the interaction of Gc WT, D841A, D841K, or D841N mutant with indicated liposomes. WT and mutant Gc proteins were immobilized on the chip surface at the same response units, and the liposome (2.5 μM) was then injected over the sensor chip. B. Expression of WT and mutant glycoproteins of DBV and HRTV. DBV or HRTV glycoprotein expression vectors were transfected into 293T cells for 24 hours and Gc expression was analyzed by immunoblots. C-D. Entry efficiency of D841 mutant-containing rVSV-HRTV or rVSV-DBV in SPTLC2- and UGCG-KO cells. SPTLC2-KO and UGCG-KO 293T cells were cultured in media supplemented with 50 μM GalCer or GlcCer for 24 hours before viral infection. Luciferase activity was measured at 18 hours post-infection. Control cells without glycosphingolipids supplement was included as comparison. Data shown are means ± SEM from representative experiments (n = 3 technical replicates). P values were determined by two-way ANOVA with Dunnett’s multiple comparison tests. ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05.