CLIC and membrane wound repair pathways enable pandemic norovirus entry and infection

Abstract

Globally, most cases of gastroenteritis are caused by pandemic GII.4 human norovirus (HuNoV) strains with no approved therapies or vaccines available. The cellular pathways that these strains exploit for cell entry and internalization are unknown. Here, using nontransformed human jejunal enteroids (HIEs) that recapitulate the physiology of the gastrointestinal tract, we show that infectious GII.4 virions and virus-like particles are endocytosed using a unique combination of endosomal acidification-dependent clathrin-independent carriers (CLIC), acid sphingomyelinase (ASM)-mediated lysosomal exocytosis, and membrane wound repair pathways. We found that besides the known interaction of the viral capsid Protruding (P) domain with host glycans, the Shell (S) domain interacts with both galectin-3 (gal-3) and apoptosis-linked gene 2-interacting protein X (ALIX), to orchestrate GII.4 cell entry. Recognition of the viral and cellular determinants regulating HuNoV entry provides insight into the infection process of a non-enveloped virus highlighting unique pathways and targets for developing effective therapeutics.

Fig. 1. GII.4 capsid protein elicits acidification and endocytosis in HIEs.

a Viral replication at 24 h (black dots) compared to bound virus at 1 h (blue dots) and inhibition of replication in the presence of VLPs compared to untreated at 24 h. Replication was quantified using n = 2 independent HIE replicates for the 1 h and n = 3 independent HIE replicates for 24 h with 2 technical replicates/sample. b LysoTracker staining of acidic compartments in the presence of an anti-GII.4 polyclonal antibody (pAb), GII.4 virus, and pAb mixed with GII.4 virus at 37 ^oC. Right panel: Mean fluorescence intensity was quantified from different regions of interest (ROIs) for pAb (blue bar, ROIs = 32), GII.4 stool (red bar, ROIs = 31), GII.4 stool +pAb (green bar, ROIs = 32). c LysoTracker staining of acidic compartments induced by GII.3 VLP (green, ROIs = 100), GCDCA (purple, ROIs = 100), GII.4 virus (cyan, ROIs = 101), and GII.4 VLP (red, ROIs = 97) compared to media (black, ROIs = 140). d GII.4 replication in the presence/absence of V-ATPase inhibitor bafilomycin A1 at 1 h (bound virus, gray dots) and at 24 h (black dots). Viral GEs were quantified using n = 2 independent HIE replicates for the 1 h and n = 3 independent HIE replicates for 24 h with 2 technical replicates/sample. e FM1-43FX (green) uptake showing GII.4 VLP-induced endocytosis. VLP-induced endocytosis compared to media (n = 4 HIE replicates). Right panel: Mean fluorescence intensity quantified from ROI = 10. f Time lapse microscopy showing GII.4 VLP (green) endocytosis and FM1-43x uptake (red). All the experiments were repeated independently three times with similar results. In a–e, error bars represent mean ± SD with significance (P values) calculated using one-way ANOVA, Dunnett’s multiple comparisons test. Source data are provided as a Source Data file.

Fig. 2. GII.4 entry is dynamin-independent but depends on cholesterol and actin for infection.

a Schematic of major endocytosis pathways, key regulators and their inhibitors. b GII.4 replication (at 37 ^oC) was assessed in the presence of dynamin inhibitors, dynasore and mitmab. Viral RNA replication was quantified at 1 (black) and 24 h (green) by RT-qPCR. c Validation of dynamin inhibitor activity by Dil-complexed low density lipoprotein (Dil-LDL) uptake (red). Right panel: Quantitation of Dil-LDL uptake in untreated (80 ROI), dynasore-treated (38 ROI) and mitmab-treated (55 ROI) HIEs. d GII.4 replication in the presence of cholesterol sequestrants, MßCD and filipin, at 1 h (black) and 24 h (red) at 37 ^oC. e GII.4 replication in the presence of actin depolymerizing agent cytochalasin D (Cyto D) at 1 h (black) and 24 h (green). f GII.4 replication in the presence of receptor tyrosine kinase (RTK) inhibitor genistein at 1 h (black) and 24 h (blue) at 37 ^oC. g GII.4 replication in the presence of nocodazole, a microtubule disrupting agent, at 1 h (black) and 24 h (pink) at 37 ^oC. h Tubulin staining (red) in the presence of genistein and nocodazole. All the experiments were repeated independently three times with similar results. In b, d–g viral GEs were quantified using n = 2 independent HIE replicates for the 1 h and n = 3 independent HIE replicates for 24 h with two technical replicates/sample. In b, viral GEs were quantified using n = 3 HIE replicates. In b–g, error bars represent mean ± SD with significance relative to untreated control calculated using one-way ANOVA, Dunnett’s multiple comparisons test. Source data are provided as a Source Data file.

Fig. 3. GII.4 infection in HIEs is sensitive to effectors of CLIC pathway.

a Schematic of macropinocytosis and the Clathrin Independent Carrier (CLIC) pathways, their effectors and inhibitors. b GII.4 replication in the presence of EIPA (Na⁺/H⁺ exchanger inhibitor), ML141 (Cdc42 inhibitor), Wiskostatin (N-WASP inhibitor), NSC23766 (RAC1 inhibitor), CT04 (RhoI inhibitor), Blebbistatin (myosin inhibitor), and LY29402 (PI3K inhibitor) at 1 h (gray) and 24 h (orange). c GII.4 replication in the presence of Arf1 inhibitor Golgicide A (GCA at 1 h (black) and 24 h (blue). d GII.4-induced tubular carriers at 1 h (37 ^oC) detected using guinea pig anti-Sydney VLP polyclonal antibody (Gp Syd-pAb) for viral capsid (green) and phalloidin for actin (red). Images were taken using a ZEISS Laser Scanning Microscope LSM 980 with Airyscan 2. e Electron microscopy to identify CLIC structures in GII.4 VLP-treated HIEs. (1–7 structures/cell) compared to, mock-treated cells (no structures in n = 25 cells). f Confocal microscopy to detect GII.4 VP1 capsid (green) colocalization with gal-3 (red) on the cell surface at 10 min and 1 h (37 ^oC) after VLP treatment using anti-gal-3 and Gp Syd-pAb (n = 3 HIE replicates). g Effect of blocking GII.4 virus-galectin-3 (gal-3) surface interaction using anti-gal-3 antibody on GII.4 replication at 1 h (black) and 24 h (red). h Dot blot analysis investigating GII.4 VLP interaction with purified gal-3. i Probing CLIC carriers utilized in HIEs for endocytosis with Alexa Fluor™ 594 conjugated cholera toxin B (CTxB) (red) and GII.4 VLPs (green) with similar cargoes marked by white arrows (n = 2 HIE replicates). Inset: Co-occurrence of CTxB and GII.4 VLPs in similar cargos, scale = 5 µm. All the experiments were repeated independently three times with similar results. In b, c, and g viral GEs were quantified using n = 2 independent HIE replicates for the 1 h and n = 3 independent HIE replicates for 24 h with two technical replicates/sample. The error bars represent mean ± SD with P values calculated using one-way ANOVA, Dunnett’s multiple comparisons test with comparisons at 24 h relative to untreated control. Source data are provided as a Source Data file.

Fig. 4. GII.4 interacts with ALIX and TSG101, with ALIX being critical for virus entry.

a GII.4 VLP, S and P domain binding to immobilized ALIX by ELISA. Error bars indicate mean ± SD with three replicates. b GII.4 VLP, S and P domain binding to immobilized TSG101 by ELISA. Error bars indicate mean ± SD with three replicates. c Bio-Layer Interferometry (BLI) to determine the binding affinity of GII.4 S and P domains to biotinylated ALIX and TSG101 (bALIX and bTSG101). d Effect of blocking virus interaction with ALIX and TSG101 on viral replication using specific pAbs at 1 h (gray) and 24 h (pink). Error bars indicate mean ± SD calculated using 2 HIE replicates for 1 h and 4 HIE replicates for 24 h (with two technical replicates/sample). e Confocal microscopy to probe GII.4-ALIX colocalization on the cell surface by detecting VP1 (green) and ALIX (red) using Gp Syd-pAb and rabbit anti-ALIX pAb. f ELISA to evaluate binding of GII.4 S- and mutant ∆S-domains to immobilized ALIX. Error bars indicate mean ± SD with 3 replicates. g Endocytosis induced by wild type GII.4 VLP and GII.4 VLP lacking the ALIX–binding motif (Δ VLP) using FM1-43FX in HIEs at 37 ^oC (n = 3 HIE replicates). h Viral replication in the presence of GII.4 WT VLPs (pink bars) and Δ VLPs (blue bars) was compared to untreated (black bar) at 24 h (blue dots). 1 h (gray dots) represents bound virus. Error bars indicate mean ± SD calculated using 2 HIE replicates for 1 h and 3 HIE replicates for 24 h (each condition with two technical replicates). i GII.4 replication in WT J2 and J2^ALIX-KD HIE monolayers indicated by percent fold change in viral RNA using the 2 − ΔΔCT method. Error bars indicate mean ± SD calculated using 3 HIE replicates for each condition with two technical replicates. All the experiments were repeated independently three times with similar results. P values for d, h, and i, relative to untreated control were calculated using one-way ANOVA, Dunnett’s multiple comparisons test. Source data are provided as a Source Data file.

Fig. 5. GII.4 entry is sensitive to factors controlling endo-lysosomal homeostasis and induces membrane wounding and subsequent wound repair mechanisms.

a GII.4 replication in the presence of PIKfyve inhibitor YM201636 at 1 h (black) and 24 h (red) at 37 ^oC. b GII.4 replication in the presence of vacuolin-1 (a lysosomal exocytosis inhibitor) at 1 h (gray) and 24 h (purple). c GII.4 replication in the presence of acid sphingomyelinase (ASM, amitriptyline) and neutral sphingomyelinase (NSM, GW4869) inhibitors at 1 h (black) and 24 h (blue). d Cell injury determination using propidium iodide (PI) uptake assay. Right panel: Graph quantitating membrane injury calculated by counting number of PI spots (red) counterstained with DAPI (blue) when HIEs were incubated with VLPs (ROI = 10), VLP + Ca²⁺ (ROI = 12) and media (ROI = 10). Significance was calculated using two-way ANOVA, Tukey’s multiple comparisons test. e Immunofluorescence staining showing GII.4-induced lysosomal exocytosis represented by the presence of LAMP-1 on the apical cell surface in media-treated cells compared to VLP-treated cells. LAMP-1 (red) and VP1 (green) colocalization was detected using mouse anti-LAMP-1 mAb and Gp Syd-pAb. (n = 3 HIE replicates). f VP1, ALIX and Gal-3 colocalization in VLP-treated and media-treated cells (1 h at 37^oC) using confocal microscopy. VP1 (green), ALIX (red) and gal-3 (white) were detected using Gp Syd-pAb, rabbit anti-ALIX pAb and rat-anti-gal3. Right panel: Graph showing colocalization between VP1, gal-3 and ALIX as estimated by Pearson Correlation Coefficient using EzColocalization (ROI = 16, black for VLP- and ROI = 17, blue for media-treated HIEs). Error bars indicate mean ± SD and P values were calculated using two-way ANOVA, Šídák’s multiple comparisons test. All the experiments were repeated independently three times with similar results. Error bars for a–c indicate mean ± SD calculated using 2 HIE replicates for 1 h and 3 HIE replicates for 24 h with two technical replicates/sample. P values were calculated using one-way ANOVA, Dunnett’s multiple comparisons test by comparing replication at 24 h to untreated control. Source data are provided as a Source Data file.

Fig. 6. GII.4 uses a complex entry mechanism involving cellular wound repair mechanisms and CLIC-mediated endocytosis.

(1) Binding of GII.4 with its glycan receptor (HBGAs and possibly with a still unidentified co-receptor) on the cell surface (2) induces plasma membrane wounding (3) triggering signaling responses that direct multiple membrane repair cellular components to the injury site. (4) ASM translocation to the plasma membrane surface results in conversion of sphingomyelin (SM) to ceramide. (5) Ceramide formation along with other membrane repair processes involving gal-3 (glycan damage sensor), ALIX (Ca²⁺ sensor) and membrane recycling processes regulated by Rab11 and Rab14 result in membrane reorganization and receptor clustering leading to (6) tubular carrier formation due to multiple interactions of virus with host factors causing membrane bending and (7) endocytosis regulated by Cdc42 and cholesterol. (8) GII.4 entry into the cell results in V-ATPase regulated endosomal acidification causing (9) conformational changes in the virus capsid and release of the viral genome from the endosomal compartment.