Fig 2: SCARB2 expression and co-localization with TMEM106A, and mapping of SCARB2/TMEM106A association. (A) Assessment of SCARB2 mRNA level in the presence of over-expressed TMEM106A protein. Total RNAs were isolated from 293A-SCARB2-TMEM106A and 293A-SCARB2 control cells and subjected to RT-qPCR assay. The mRNA level of SCARB2 was normalized to that of GAPDH. The relative mRNA level of control cells was set as 100%. Statistical differences between Ctrl and TMEM106A were assessed using Student’s t test. The data are mean ± SD of three independent experiments. n.s., no statistical significance. (B) Assessment of SCARB2 protein level in the presence of over-expressed TMEM106A protein. Total, cytoplasmic and membrane protein extracts were prepared from 293A-SCARB2-TMEM106A and 293A-SCARB2 control cells. Western blot analysis was carried out by using anti-SCARB2 and anti-TMEMA106A antibody, respectively, to compare the protein levels of SCARB2 as well as TMEM106A. The membrane was then stained to confirm the equal loading of two samples. (C) Immunofluorescence microscopic picture showing the membrane co-localization of SCARB2 and TMEM106A. 293A-SCARB2-TMEM106A and 293A-SCARB2 cells were collected and co-incubated with anti-SCARB2 and anti-TMEM106A antibodies, followed by incubation with corresponding FITC-conjugated secondary antibodies. The cells were then transferred to coverslips and fixed with ice-cold 4% paraformaldehyde. Finally, the cell membrane was permeabilized, and nuclear DNA was stained with DAPI (blue). Bar=10 μm. (D) Crystal structure of SCARB2 ectodomain (PDB:4TW2) (left panel) featuring the binding sites of the antibodies recognizing different domains of SCARB2 (right panel). (E) Conformation analysis of membrane SCARB2 via surface staining analyzed by flow cytometry. 293A-SCARB2-TMEM106A and 293A-SCARB2 cells were collected, incubated with the FITC-conjugated JL-1 or JL-2 antibody, or common antibodies against different regions of SCARB2 (a.a. 27-432 and a.a. 339-437) for 1 h at 4°C, followed by the Alexa Fluor 555-conjugated anti-mouse IgG and Alexa Fluor633-conjugated anti-goat IgG staining. Finally, the cells were analyzed by using flow cytometry. Control 293A-SCARB2 and 293A-SCARB2-TMEM106A cells were incubated with isotype (solid and dashed in black) as florescence negative control. Blue and red lines represent cells incubated with the indicated SCARB2 antibodies.

Fig 3: Proposed mechanism of the suppression of EV-A71 infection by TMEM106A. Both SCARB2 and PSGL-1 act as cell surface receptors enabling EV-A71 infection. SCARB2 facilitates EV-A71 infection by playing critical roles in virus attachment, entry and uncoating. Innate immune response to viral infection, through engagement of receptors, activates signaling pathways that induce the expression and secretion of type I IFN. In turn, IFN molecules bind their cognate receptors, signal through the JAK-STAT pathway, and induce hundreds of ISGs. One of them, TMEM106A, encodes a transmembrane protein, which exerts antiviral activity by blocking the recognition site of EV-A71 through its association with SCARB2, but not PSGL-1.

Fig 4: Mechanistic study of the inhibition of EV-A71 by TMEM106A. (A) The effect of TMEM106A expression on the attachment (left) and endocytosis (right) of EV-A71 was examined by incubating control cells (293A-SCRB2) and cells expressing TMEM106A (293A-SCRB2-TMEM106A) with EV-A71-MZ (MOI=100) for 1 h at 4°C (binding), with or without an additional 30-min incubation at 37°C (endocytosis). The EV-A71 virions were detected by immunofluorescence assay using a primary antibody against EV-A71 VP2 and an Alexa-Fluor 555-conjugated secondary antibody. The cells were counterstained with DAPI to show the nucleus. Bar=10 μm. DIC, differential interference contrast. (B) To examine the effect of TMEM106A expression on the post-entry stages of EV-A71 replication, EV-A71-EGFP RNAs were transcribed from the linearized pWSK-EV71-EGFP, and then transfected into 293A-SCARB2-TMEM106A or control 293A-SCARB2 cells. The expression of EGFP and the cellular viral RNA levels at different time points post transfection were examined by fluorescence microscope. (C) The GFP-positive cells in 50 fields were counted and the mean number was calculated. Statistical differences between Ctrl and TMEM106A were assessed using Student’s t test. * (p < 0.05) and *** (p < 0.001). (D) Eighteen hours after transfection with the viral RNA, the viral yield in 293A-SCARB2-TMEM106A and 293A-SCARB2 cells was measured by plaque assay. (E) The value from the 293A-SCARB2 control cells was set at the 100% of virus production to estimate remaining percentage of viral production in presence of TMEM106A. Statistical differences between control and TMEM106A were assessed using Student’s t test. The data are mean ± SD of three independent experiments. n.s., no statistical significance.

Fig 5: The C-terminal domain of TMEM106A is involved in EV-A71 attachment. (A) Domain architecture of TMEM106A protein and truncations. (B) Antiviral activity of different deletion mutants of TMEM106A. Plasmids expressing myc-tagged wild-type (a.a. 1-262) or truncated TMEM106A were individually transfected into 293A-SCARB2 cells together with a reporter plasmid pCAG-DsRed at a ratio of 3:1, followed by EV-A71-GFP infection at an MOI of 0.1. An empty vector without TMEM106A fragment was used as control. Level of full-length TMEM106A or deletion mutants were analyzed by western blotting using an anti-myc mAb (lower panel). The GFP signal produced by EV-A71-GFP was detected by FACS. The data were normalized according to the GFP produced in the control cells transfected with the empty vector, in which where none of the TMEM106A forms were expressed, set as 100% (upper panel). The results are represented as mean ± SD obtained in three independent experiments. *p < 0.05.