Heparan sulfate is an important mediator of Ebola virus infection in polarized epithelial cells

Abstract

Background: Currently, no FDA-approved vaccines or treatments are available for Ebola virus disease (EVD), and therapy remains largely supportive. Ebola virus (EBOV) has broad tissue tropism and can infect a variety of cells including epithelial cells. Epithelial cells differ from most other cell types by their polarized phenotype and barrier function. In polarized cells, the apical and basolateral membrane domains are demarcated by tight junctions, and specialized sorting machinery, which results in a difference in composition between the two membrane domains. These specialized sorting functions can have important consequences for viral infections. Differential localization of a viral receptor can restrict virus entry to a particular membrane while polarized sorting can lead to a vectorial virus release. The present study investigated the impact of cell polarity on EBOV infection.

Methods: Characteristics of EBOV infection in polarized cells were evaluated in the polarized Caco-2 model grown on semipermeable transwells. Transepithelial resistance (TEER), which is a function of tight junctions, was used to assess epithelial cell polarization. EBOV infection was assessed with immunofluorescence microscopy and qPCR. Statistical significance was calculated using one-way ANOVA and significance was set at p < 0.05.

Results: Our data indicate that EBOV preferentially infects cells from the basolateral route, and this preference may be influenced by the resistance across the Caco-2 monolayer. Infection occurs without changes in cellular permeability. Further, our data show that basolateral infection bias may be dependent on polarized distribution of heparan sulfate, a known viral attachment factor. Treatment with iota-carrageenan, or heparin lyase, which interrupts viral interaction with cellular heparan sulfate, significantly reduced cell susceptibility to basolateral infection, likely by inhibiting virus attachment.

Conclusions: Our results show cell polarity has an impact on EBOV infection. EBOV preferentially infects polarized cells through the basolateral route. Access to heparan sulfate is an important factor during basolateral infection and blocking interaction of cellular heparan sulfate with virus leads to significant inhibition of basolateral infection in the polarized Caco-2 cell model.

Keywords: Caco-2; Ebola virus; Heparan sulfate; Polarized cells.

Fig. 1

Establishment of a polarized Caco-2 cell monolayer. a Caco-2 monolayers were seeded at a density of 4 × 10⁴ and allowed to grow for 10 days after seeding. TEER readings were taken every other day and normalized to resistance of unseeded well taken at the same time point. Values plotted are mean ± SD calculated from three independent experiments. b Caco-2 cells were grown for 6 days after seeding on semipermeable membranes and then fixed with 10% PBS buffered formalin (E-cadherin) or ice cold methanol (ZO-1) and examined by immunofluorescence microscopy

Fig. 2

Basolateral infection of EBOV is more efficient in Caco-2 cells a Caco-2 cells infected with EBOV at 3 pfu/cell were assessed for EBOV RNA expression at 6, 24, and 48 hpi, using SYBR-green qPCR assay and normalized to GAPDH expression. Results are expressed in mean ± SD calculated from three independent experiments. Data was analyzed using one-way ANOVA ***p < 0.001. b Caco-2 cells infected with EBOV at 3 pfu/cell were assessed for EBOV-NP protein expression at 48 hpi by Western Blot analysis. GAPDH was used as a loading reference

Fig. 3

Extent of cell polarity affects cell susceptibility to EBOV infection. Caco-2 cells were infected on 4, 6 or 8 days post-seeding at 3 pfu/cell. Cells were assessed for EBOV RNA expression at 6, hpi, using SYBR-green qPCR assay and normalized to GAPDH expression. Results are expressed as mean ± SD fold change calculated from three independent experiments. Data was analyzed using one-way ANOVA ***p < 0.001

Fig. 4

Semiconfluent monolayers are more susceptible to apical EBOV infection. Caco-2 cells were grown to either semiconfluence (day 4 pi) or confluence (day 6 pi) and infected with EBOV at 3 pfu/cell. The monolayers were fixed with 10% buffered formalin and examined for expression of E-cadherin and EBOV-GP by immunofluorescence microscopy

Fig. 5

The integrity of tight junctions is not disturbed by EBOV. Caco-2 cells were grown on semipermeable Transwell supports and infected with EBOV either apically or basolaterally at 3 pfu/ml. TEER was measured daily, and results in Ω are mean values of triplicates. Data was analyzed using one-way ANOVA n.s. > 0.05

Fig. 6

Mechanical damage to Caco-2 monolayer increases susceptibility to apical infection. Caco-2 monolayers were scratched with a pipette tip across the apical surface to expose the underlying basal cells along the injury path (yellow dashed lines). The apical surfaces of injured cultures were immediately infected with EBOV at 3 pfu/cell. At 24 hpi, the cultures were fixed and immunostained with antibody to EBOV-GP, and E-cadherin, and the cultures were examined by fluorescence microscopy. Original magnification 40X

Fig. 7

Treatment with ι-carrageenan or heparin lyase selectively inhibits basolateral EBOV infection. a EBOV was treated with increasing concentrations of ι-carrageenan for 30 min before infection at 4 °C. The treated virus was then added at a concentration of 3 pfu/cell to Caco-2 cells and incubated at 37° for 1 h. b Caco-2 cells were treated with a Heparin lyase I and III blend for 1 h before infection. The cells were then washed with sterile PBS and infected with EBOV either apically or basolaterally at a concentration of 3 pfu/cell and incubated at 37° for 1 h. Following incubation, EBOV RNA expression was measured at 24 hpi, using SYBR-green qPCR assay and normalized to GAPDH expression. Fold change results are expressed in mean ± SD calculated from three independent experiments. Data was analyzed using one-way ANOVA n.s. > 0.05 **p < 0.01, ***p < 0.001

Fig. 8

Treatment with ι-carrageenan or heparin lyase selectively inhibits basolateral EBOV binding. a EBOV was pretreated with media or 20 ng/μl of ι-carrageenan and added to Caco-2 cells at a concentration of 3 pfu/μl at 4 °C for particle binding. Cells were washed 3X with ice cold PBS to remove excess virus and lysed with TRIzol. b Caco-2 cells were treated with a Heparin lyase I and III blend for 1 h before infection. The cells were then washed with sterile PBS chilled to 4 °C and infected with EBOV either apically or basolaterally at the same concentration and incubated at 4 °C for particle binding. Cells were washed 3X with ice cold PBS to remove excess virus and lysed with TRIzol. EBOV RNA expression was measured at 24 hpi, using SYBR-green qPCR assay and normalized to GAPDH expression. Fold change results are expressed in mean ± SD calculated from three independent experiments. Data was analyzed using one-way ANOVA n.s. > 0.05, *p < 0.05, **p < 0.01