Fig 2: Blocking GNPTAB cleavage with a SKI-1/S1P inhibitor blocks EBOV infection. a PF-429242 blocks processing of GNPTAB. Huh7 cells were transfected with a plasmid expressing GNPTAB-myc. Twenty-four hours post transfection, PF-429242, or DMSO vehicle, was added to the indicated concentrations. Cell lysates were harvested 48 h post transfection and subjected to immunoblotting. b PF-429242 does not alter NPC1 expression levels. Huh7 cells were treated with PF-429242 at the indicated concentrations for 24 h before cell lysates were harvested and subjected to immunoblotting. c PF-429242 blocks EBOV-GP-mediated entry. Varying concentrations of PF-429242 were added to Huh7 cells 1 h prior to the addition of luciferase-encoding lentiviral particles pseudotyped with the Zaire-EBOV-GP (EBOVpp), or with VSV-G protein (VSVpp). Three days later luciferase activity was determined. Data represent the mean and s.d. of values relative to the DMSO vehicle control, from 8 biological replicates. A representative of 3 independent experiments is shown. d PF-429242 inhibits infection by EBOV-ZsG reporter virus. Huh7 cells were treated with varying concentrations of PF-429242, before being mock-infected or infected with EBOV-ZsG reporter virus. Three days post-infection, cell viability or ZsG fluorescence was measured, respectively. Data represent the mean and s.d. of values relative to the DMSO vehicle control, from four biological replicates. A representative of six independent experiments is shown

Fig 3: Reconstitution of GNPTAB in knockout cells restores EBOV infection. Parental HAP1 and GNPTAB- knockout cells were transduced with lentiviruses to express ß-glucuronidase (GUS) as a negative control, or myc-tagged GNPTAB (GNPTAB-myc). a, b Immunoblotting of lysates from HAP1, knockout and reconstituted cell lines. The migration of molecular mass markers in kDa is shown on the right. c Reconstitution of GNPTAB restores EBOV-induced cytopathic effect. Cells were infected and viability determined 6 days p.i. as for Fig. 2a. Data represent the mean ± s.d. of eight biological replicates. A representative of two independent experiments is shown. Statistical analysis was performed with a two-tailed Student’s t-test with significance shown as **** P < 0.0001. d Reconstitution of GNPTAB restores EBOV infection. HAP1 and GNPTAB- cells reconstituted to express GUS or GNPTAB-myc were infected with EBOV-ZsG virus and fluorescence was determined over time. Data represent the mean ± s.d. of six biological replicates. A representative of two independent experiments is shown

Fig 4: Schematic of a model for the role of GNPTAB in EBOV entry. (1) The SKI-1/S1P protease cleaves the GNPTAB precursor to yield the individual α and β subunits, which then assemble with the γ subunit to form the GlcNAc-phosphotransferase enzyme. (2) GlcNAc-phosphotransferase catalyzes the transfer of mannose-6-phosphate onto the glycans of proteins destined to be transported to the lysosome, such as pro-cathepsin B and pro-cathepsin L. (3) The mannose-6-phosphate receptor recognizes its substrates and mediates their transport to the lysosome. (4) Once in the lysosome, pro-cathepsins cleave to yield the mature cathepsins. (5) The activated cathepsins cleave the EBOV GP, removing the glycan-cap and mucin-like domains, and allowing the receptor-binding site to interact with NPC1, priming the GP for fusion. (6) Following fusion, the viral nucleocapsid enters into the cell cytoplasm where replication can be initiated. Blocking GlcNAc-phosphotransferase activity, either by inhibition of the SKI-1/S1P protease at (1), or by the absence of functional enzyme in (2), has the effect of preventing EBOV entry

Fig 5: GNPTAB is required for efficient EBOV infection. a GNPTAB− cells survive EBOV infection. Parental HAP1, GNPTAB−, or NPC1− knockout cells were infected with wild-type EBOV at an MOI of 0.3, and viability relative to mock-infected cells was determined 6 days p.i. Data represent the mean ± s.d. of three independent experiments. Statistical analysis was performed with a two-tailed Student’s t-test, with significance shown as ***, P = 0.0005. b Infection with the EBOV-ZsG reporter virus is reduced in GNPTAB− knockout cells. Parental HAP1 (orange circles), GNPTAB− knockout (blue squares) and NPC1− knockout (black triangles) cells were infected with EBOV-ZsG at an MOI of 0.1 and ZsG fluorescence was measured over time. Data represent the mean ± s.d. of 4 biological replicates. A representative of three independent experiments is shown. c Wild-type EBOV growth is impaired in GNPTAB− knockout cells. Parental and knockout HAP1 cells were infected with wild-type EBOV at an MOI of 0.5. Samples of culture supernatant were taken at intervals and the EBOV titer was determined. Data represent the mean ± s.d. of three biological replicates. A representative of two independent experiments is shown