Select membrane proteins modulate MNV-1 infection of macrophages and dendritic cells in a cell type-specific manner

Abstract

Noroviruses cause gastroenteritis in humans and other animals, are shed in the feces, and spread through the fecal-oral route. Host cellular expression of attachment and entry receptors for noroviruses is thought to be a key determinant of cell tropism and the strict species-specificity. However, to date, only carbohydrates have been identified as attachment receptors for noroviruses. Thus, we investigated whether host cellular proteins play a role during the early steps of norovirus infection. We used murine norovirus (MNV) as a representative norovirus, since MNV grows well in tissue culture and is a frequently used model to study basic aspects of norovirus biology. Virus overlay protein binding assay followed by tandem mass spectrometry analysis was performed in two permissive cell lines, RAW264.7 (murine macrophages) and SRDC (murine dendritic cells) to identify four cellular membrane proteins as candidates. Loss-of-function studies revealed that CD36 and CD44 promoted MNV-1 binding to primary dendritic cells, while CD98 heavy chain (CD98) and transferrin receptor 1 (TfRc) facilitated MNV-1 binding to RAW 264.7 cells. Furthermore, the VP1 protruding domain of MNV-1 interacted directly with the extracellular domains of recombinant murine CD36, CD98 and TfRc by ELISA. Additionally, MNV-1 infection of RAW 264.7 cells was enhanced by soluble rCD98 extracellular domain. These studies demonstrate that multiple membrane proteins can promote efficient MNV-1 infection in a cell type-specific manner. Future studies are needed to determine the molecular mechanisms by which each of these proteins affect the MNV-1 infectious cycle.

Keywords: CD36; CD44; CD98 heavy chain; Infection; Norovirus; Transferrin receptor 1.

Figure 1.

Identification of MNV-1-putative interacting proteins by VOPBA followed by tandem mass spectrometry. (a) Representative image of RAW 264.7 cells VOPBA. Membrane proteins were extracted from RAW 264.7 cells, separated by SDS-PAGE, and VOPBA was performed using MNV-1 or mock lysate. Virus bound to cellular proteins was detected by an anti-MNV rabbit serum. (b) Schematic representation of the identified proteins in the two permissive cell lines tested (RAW 264.7 and SRDC cells). Bands observed by VOPBA (indicated by asterisks) were excised from a Coomassie blue stained gel, and proteins were identified by tandem mass spectrometry. Each hit depicted in the figure represents a distinct protein. Five proteins met the inclusion criteria (outlined in results, section 3.1) as candidate MNV-1-interacting proteins.

Figure 2.

MNV-1 binding is reduced in CD36- and CD44-deficient bone marrow-derived dendritic cells (BMDC) but not bone marrow-derived macrophages (BMM). Wild type (WT), CD36 knockout (CD36 KO), or CD44 KO BMDC (a and c) and BMM (b and d) were infected with MNV-1 (MOI of 2). Bound virus was quantified by RT-qPCR. Results from four independent experiments performed in duplicate are shown as percent binding of virus genome equivalents to cells, which were calculated relative to the WT control cells set to 100%. *p<0.05.

Figure 3.

TfRc and CD98 facilitate MNV-1 binding to RAW 264.7 cells, while CD98 is required for efficient virus infection. RAW 264.7 cells were transfected with CD98 siRNA, TfRc siRNA, or non-targeting (NT) siRNA control. At 48 h post transfection, cells were stained with fluorescently labeled anti-CD98 or anti-TfRc (a) or infected with MNV-1 (b and c). (a) Transfected cells were analyzed by flow cytometry to verify protein knockdown. Left panel: representative image showing CD98 expression in RAW 264.7 cells transfected with NT siRNA (open histogram, solid black line) or CD98 siRNA (grey filled histogram). Right panel: representative image showing TfRc expression in RAW 264.7 cells transfected with NT siRNA (open histogram, solid black line) and TfRc siRNA (grey filled histogram). Open histogram, dotted black line: unstained control, (b) CD98- and TfRc-siRNA knockdown significantly reduced MNV-1 binding to RAW 264.7 cells. Transfected cells were infected with MNV-1 (MOI of 2), and bound virus was quantified by RT-qPCR. Results from three independent experiments performed in duplicate are shown as percent binding of virus genome equivalents to cells, which were calculated relative to the NT siRNA-transfected cells set to 100%. (c) CD98- siRNA knockdown significantly reduced MNV-1 infection of RAW 264.7 cells. Transfected cells were infected with MNV-1 (MOI of 0.05), and viral genome titers were assessed by RT- qPCR at 0 and 8 hpi. Results from three independent experiments performed in duplicate are shown as percent infection, which was calculated relative to the NT siRNA-transfected cells set to 100%. *p<0.05, **p<0.01, ***p<0.001.

Figure 4.

MNV-1 interacts with rCD36, rCD98, and rTfRc via its protruding domain (Pd). MNV-1 and MNV-1 Pd binding to the extracellular domain of recombinant proteins was evaluated by ELISA. (a) Recombinant CD36, CD44, CD98, TfRc or an irrelevant protein (BSA) were coated onto 96 well microtiter plates and incubated with MNV-1 or mock lysate. (b) Recombinant CD36, CD98, TfRc or BSA were coated onto 96 well microtiter plates and incubated with MNV-1 Pd or buffer. Results are represented as the absorbance values measured at 415 nm (A415). The dashed line is drawn at the A415 mean value for the BSA-coated wells plus twice the value of the SEM. Values above the line were considered positive. Results are from at least two independent experiments with conditions tested in duplicate.

Figure 5.

Recombinant CD98 enhances MNV-1 infection in a concentration dependent manner. rCD98 was incubated with MNV-1 prior to infection in 2×l06 fold molar excess (a) or 1,000 fold molar excess (b), or with BSA. RAW 264.7 cells were then infected with MNV-1 (MOI of 0.05) and viral genome titers were assessed by RT-qPCR at 0 and 8 hpi. Results are shown as percent infection, which were calculated relative to the BSA-MNV-1 infected control cells set to 100%. Results are from two independent experiments with conditions tested in duplicate per experiment. ****p<0.0001.