Epstein-Barr virus transcytosis through polarized oral epithelial cells

Abstract

Although Epstein-Barr virus (EBV) is an orally transmitted virus, viral transmission through the oropharyngeal mucosal epithelium is not well understood. In this study, we investigated how EBV traverses polarized human oral epithelial cells without causing productive infection. We found that EBV may be transcytosed through oral epithelial cells bidirectionally, from both the apical to the basolateral membranes and the basolateral to the apical membranes. Apical to basolateral EBV transcytosis was substantially reduced by amiloride, an inhibitor of macropinocytosis. Electron microscopy showed that virions were surrounded by apical surface protrusions and that virus was present in subapical vesicles. Inactivation of signaling molecules critical for macropinocytosis, including phosphatidylinositol 3-kinases, myosin light-chain kinase, Ras-related C3 botulinum toxin substrate 1, p21-activated kinase 1, ADP-ribosylation factor 6, and cell division control protein 42 homolog, led to significant reduction in EBV apical to basolateral transcytosis. In contrast, basolateral to apical EBV transcytosis was substantially reduced by nystatin, an inhibitor of caveolin-mediated virus entry. Caveolae were detected in the basolateral membranes of polarized human oral epithelial cells, and virions were detected in caveosome-like endosomes. Methyl β-cyclodextrin, an inhibitor of caveola formation, reduced EBV basolateral entry. EBV virions transcytosed in either direction were able to infect B lymphocytes. Together, these data show that EBV transmigrates across oral epithelial cells by (i) apical to basolateral transcytosis, potentially contributing to initial EBV penetration that leads to systemic infection, and (ii) basolateral to apical transcytosis, which may enable EBV secretion into saliva in EBV-infected individuals.

Fig 1

Model of EBV bidirectional transcytosis. (A) To study apical to basolateral transcytosis, cells were grown on the upper surfaces of Transwell filter inserts, with apical surfaces facing upward, and virus was added to the apical membranes. (B) To study basolateral to apical transcytosis, cells were grown on the lower surfaces of Transwell filter inserts, with apical membranes facing the lower chambers. Addition of virus to the upper chambers of the Transwell inserts allows binding of virions to the apical (A) or basal (B) surfaces of polarized cells. Detection of virus in the culture media of lower chambers indicates EBV transcytosis in the apical to basolateral (A) and basolateral to apical (B) directions. (Middle images) Polarized cells were examined by electron microscopy. Red arrows indicate cells and filter inserts. (Right images) Polarized cells were immunostained for the tight junction protein ZO-1 and analyzed by confocal microscopy. Nuclei are stained in blue with TO-PRO3. Original magnification, ×600.

Fig 2

Bidirectional EBV transcytosis through polarized tonsil epithelial cells. (A) EBV B95-8 virus (10⁷ virions) was added to the upper chambers of Transwell inserts with either apically or basolaterally oriented polarized TNSL#2 tonsil epithelial cells. After 1, 2, 3, and 4 h, culture media from the lower chambers were examined for EBV by qPCR. Similar results were obtained in 3 independent experiments. (B) Polarized cells were propagated from tonsil keratinocytes of 8 independent donors. EBV transcytosis assays were performed in the apical to basolateral and basolateral to apical directions. After 4 h, culture media from the lower chambers were examined for EBV. (C) To measure transcytosis of UV-inactivated EBV, UV-exposed and untreated B95-8 virus was used for bidirectional transcytosis assays using TNSL#2 cells. (D) EBV transcytosis via polarized TNSL#2 tonsil epithelial cells was performed using B95-8 and Akata strains of virus in the apical to basolateral and basolateral to apical directions. (E) EBV virions were incubated with pools of either EBV-positive or -negative sera from healthy individuals. These virions were then used for apical to basolateral or basolateral to apical transcytosis in TNSL#3 cells. Transcytosed virions were quantified by qPCR after 4 h. Data are presented as percent inhibition of transcytosis of virions preincubated with EBV-positive sera relative to transcytosis of virions preincubated with EBV-negative control sera. The results were similar in 2 independent experiments. Error bars show standard errors of the means.

Fig 3

Infectivity of transcytosed EBV in B lymphocytes. (A) EBV B95-8 bidirectional transcytosis assays were performed in polarized TNSL#3 tonsil epithelial cells. Transcytosed virions were collected and used for infection of EBV-negative B lymphocytes. One set of cells were infected with virus that was used for transcytosis of epithelial cells (direct infection). After 7 days, B cells were treated with PMA for the next 3 days. Cells were collected, and total RNA was isolated and used for qRT-PCR assays of BZLF1 (upper graph) and gp350/220 (lower graph) gene expression. (B and C) Freshly isolated EBV-negative B lymphocytes were infected with EBV B95-8 virions transcytosed through polarized TNSL#3 tonsil epithelial cells. One set of B lymphocytes was not infected and served as a control. (B) After 7 days, half of the cultures were used for qRT-PCR assays of LMP1 and EBNA1 gene expression. *, not detected. (C) The other half of the B lymphocyte cultures were used for cell proliferation assays. OD, optical density. Error bars show standard errors of the means. Similar results were obtained in 2 independent experiments.

Fig 4

EBV apical to basolateral transcytosis is facilitated by macropinocytosis, whereas basolateral to apical transcytosis is initiated by caveolin-mediated endocytosis. (A) Polarized TNSL#3 tonsil epithelial cells were pretreated with nystatin, chlorpromazine, or amiloride for 1 h. Cells were washed and TER was measured. These cells were then used for EBV B95-8 bidirectional transcytosis and after 4 h, TER was again measured. (B) After transcytosis assays, cells were subjected to paracellular permeability assays. Media, culture media without IgG that served as negative control; IgG, horseradish peroxidase-labeled IgG that was added to the upper chamber. (C) One set of cells after transcytosis was used for cell viability assays. (D) Drug-pretreated polarized cells were used for bidirectional transcytosis assays with 10⁷ virions per insert. Transcytosed EBV from the lower chamber of control and drug-treated cells was quantitated by DNA qPCR. Data are presented as percent viral transcytosis in drug-treated cells relative to untreated controls. Controls were considered 100%. Average percent transcytosis is presented from 5 independent experiments. *, P < 0.001 compared with the untreated control cells. (E) Polarized TNSL#3 cells were treated with increasing concentrations of amiloride, and apical to basolateral EBV transcytosis was examined by qPCR assay. (F) TNSL#3 cells were treated with increasing concentrations of nystatin, and basolateral to apical transcytosis of EBV was examined. (G) Polarized TNSL#3 cells were pretreated with nystatin, chlorpromazine, or amiloride at 37°C for 1 h. Cells were then washed and exposed to EBV B95-8 from their apical and basolateral membranes at 4°C for 2 h. One set of control cells was trypsinized to remove membrane-bound virions. Cells were washed and lysed, and viral binding was examined by qPCR. Results are presented as copy numbers of viral DNA in control and drug-treated cells. Results shown are from one representative experiment out of two experiments. Chlorpro, chlorpromazine; RLU, relative luminescence units. Error bars show standard errors of the means.

Fig 5

Visualization of EBV macropinocytosis from apical membranes of polarized tonsil epithelial cells. Polarized TNSL#3 tonsil epithelial cells exposed to EBV B95-8 at their apical membranes were fixed after 30 min and analyzed by electron microscopy. (A) EBV virions surrounded by apical membrane protrusions. (B) Virions are in the invaginated membranes and subapical macropinosome-like vesicles. (C) Virions in the subapical macropinosomes. Insets and green arrowheads show membrane-bound and penetrating virions. Blue arrowheads show virions in the macropinosomes. AP, apical; CYT, cytoplasm; NUC, nuclei.

Fig 6

Inhibition of EBV transcytosis by drugs and siRNAs. (A) Polarized TNSL#3 tonsil epithelial cells were pretreated with inhibitors of PI3K (LY294002), PKC (rottlerin), MLCK (ML9), and dynamin (dynasore) for 1 h. Cells were then washed and used for EBV B95-8 bidirectional transcytosis assays. After 4 h, viral transcytosis was examined using qPCR. Data are presented as percent inhibition of EBV transcytosis by drug treatment relative to untreated controls. Average percent inhibition is presented from 2 independent experiments. (B) Polarized tonsil cells transfected with specific siRNAs against Rac-1, Pak1, Cdc42, ARF6, or dynamin 1/2. An irrelevant siRNA was used as a control. After 48 h, one set of cells was used for Western blot analysis using antibodies against Rac-1, Pak1, Cdc42, ARF6, or dynamin 2. The mean densities of pixels in the protein bands were measured by ImageJ software, and the results for each gel are shown as a bar graph under each blot. (C) The next set of cells was used for EBV transcytosis, which was evaluated by qPCR assay. Data are presented as percent inhibition of viral transcytosis by drugs relative to untreated controls. Average percent inhibition is presented from 2 independent experiments. *, no inhibition was detected. (A and C) Error bars show standard errors of the means.

Fig 7

EBV p18 colocalizes with caveolin-containing vesicles upon entry of virus from basolateral membranes of polarized cells. Polarized TNSL#3 tonsil epithelial cells were exposed to EBV B95-8 from apical (A) or basolateral (B) membranes, and 30 min later, cells were fixed and coimmunostained for EBV p18 (green) and either clathrin (red) or caveolin-1 (red). Cells were analyzed by confocal microscopy. Yellow in panel B (lower right image) indicates colocalization of EBV p18 with caveolin-associated vesicles. Nuclei are stained in blue. Original magnification, ×600.

Fig 8

EBV basal to apical transcytosis initiated by caveolin-mediated endocytosis. (A) Polarized TNSL#3 cells transfected with specific siRNAs against clathrin heavy chain or caveolin-1. After 48 h, one set of cells was used for detection of clathrin heavy chain or caveolin-1 on Western blots. The mean densities of pixels in the protein bands are shown as a bar graph under each blot. (B) The next set of siRNA-transfected cells was used for EBV B95-8 transcytosis. Viral transcytosis was evaluated by qPCR, and data are presented as percent EBV transcytosis in specific siRNA-transfected cells relative to cells transfected with control siRNAs. Viral transcytosis in controls was considered to be 100%. Average percent transcytosis is presented from 2 independent experiments. Error bars show standard errors of the means.

Fig 9

Detection of caveolae and EBV-containing caveosomes in basolateral membranes of oral epithelial cells. Polarized TNSL#3 tonsil epithelial cells were exposed to EBV B95-8 from their basolateral membranes, and after 30 min, cells were fixed and analyzed by electron microcopy. (A, inset and arrowheads) Multiple EBV virions were detected within the basolateral membranes of polarized cells, and virions were bound to invaginated membranes. (B, inset) Virions were detected in the deeply invaginated membranes. (C, inset). EBV was detected in the endosome. BM, basal membranes; CYT, cytoplasm; NUC, nuclei.

Fig 10

Cholesterol depletion inhibited EBV basolateral to apical transcytosis. (A) Polarized TNSL#1 tonsil epithelial cells were treated with methyl β-cyclodextrin from their apical and/or basolateral membranes, and culture media were collected from apical or basolateral membranes, separately. Released cholesterol was measured using the Amplex red cholesterol assay kit, and results are expressed as relative fluorescence units. (B) TER of polarized cells was measured 2 h after MβCD treatment. (C) EBV B95-8 was added to the upper chambers of apically or basolaterally oriented cells and incubated for 1 h at 37°C. Cells were then collected by trypsinization, and intracellular EBV was measured by qPCR. Data are presented as percent EBV entry into MβCD-treated cells relative to untreated controls, which was considered to be 100%. Data shown are from one representative experiment out of two experiments. Error bars show standard errors of the means.