Development of a Gaussia luciferase-based human norovirus protease reporter system: cell type-specific profile of Norwalk virus protease precursors and evaluation of inhibitors

Abstract

Norwalk virus (NV) is the prototype strain of human noroviruses (HuNoVs), a group of positive-strand RNA viruses in the Caliciviridae family and the leading cause of epidemic gastroenteritis worldwide. Investigation of HuNoV replication and development of antiviral therapeutics in cell culture remain challenging tasks. Here, we present NoroGLuc, a HuNoV protease reporter system based on a fusion of NV p41 protein with a naturally secreted Gaussia luciferase (GLuc), linked by the p41/p22 cleavage site for NV protease (Pro). trans cleavage of NoroGLuc by NV Pro or Pro precursors results in release and secretion of an active GLuc. Using this system, we observed a cell type-specific activity profile of NV Pro and Pro precursors, suggesting that the activity of NV Pro is modulated by other viral proteins in the precursor forms and strongly influenced by cellular factors. NoroGLuc was also cleaved by Pro and Pro precursors generated from replication of NV stool RNA in transfected cells, resulting in a measurable increase of secreted GLuc. Truncation analysis revealed that the N-terminal membrane association domain of NV p41 is critical for NoroGLuc activity. Although designed for NV, a genogroup GI.1 norovirus, NoroGLuc also efficiently detects Pro activities from GII.3 and GII.4 noroviruses. At noncytotoxic concentrations, protease inhibitors ZnCl2 and Nα-p-tosyl-l-lysine chloromethyl ketone (TLCK) exhibited dose-dependent inhibitory effects on a GII.4 Pro by NoroGLuc assay. These results establish NoroGLuc as a pan-genogroup HuNoV protease reporter system that can be used for the study of HuNoV proteases and precursors, monitoring of viral RNA replication, and evaluation of antiviral agents.

Importance: Human noroviruses are the leading cause of epidemic gastroenteritis worldwide. Currently, there are no vaccines or antiviral drugs available to counter these highly contagious viruses. These viruses are currently noncultivatable in cell culture. Here, we report the development of a novel cell-based reporter system called NoroGLuc that can be used for studying norovirus replication and also for screening/evaluation of antiviral agents. This system is based on the fusion between viral protein p41 and a naturally secreted Gaussia luciferase (GLuc) with a cleavage site that can be recognized by the viral protease. Cleavage of this fusion protein by the viral protease results in the release and secretion of an active GLuc. Using NoroGLuc, we demonstrated a cell type-specific activity profile of the viral protease and its precursors and dose-dependent inhibitory effects of two protease inhibitors. This novel reporter system should be useful in probing norovirus replication and evaluating antiviral agents.

FIG 1

Development of NoroGLuc reporter system. (A) Schematic drawing of NV ORF1 and the NoroGLuc fusion protein showing the identical p41/p22 and p41/GLuc cleavage sites. Note that the GLuc sequence begins with Pro-19 (P₁₉) because the first 18 aa of GLuc, which constitute a secretion signal, were deleted. (B) Western blotting of NoroGLuc expressed in HEK293 cells from pCG or pCMV vector at 24 h posttransfection. NoroGLuc was detected by anti-GLuc antibody. A nonspecific protein band, marked by the asterisk, serves as the equal loading control. (C) Immunofluorescence of NoroGLuc expressed in HEK293 cells from pCMV or pCG vector. Cells were fixed at 24 h posttransfection and immunostained with anti-GLuc antibody. Nuclei were counterstained with DAPI. (D) trans cleavage of NoroGLuc in HEK293 cells by NV ORF1 releases an active GLuc. HEK293 cells were cotransfected with NoroGLuc, expressed from pCMV or pCG vector, together with a control vector or ORF1 expression vector. GLuc assay of culture medium was performed at the indicated number of hours posttransfection.

FIG 2

Cell culture expression of NV Pro and precursors. (A) Schematic drawing of NV Pro constructs used in this study. (B) Western blotting (WB) of NV Pro and precursors expressed from the pCG vector in COS-7 and HEK293 cells at 40 h posttransfection using anti-Pro antibody (top panel). The same membrane was overexposed to show mature Pro (middle panel) and then reprobed with anti-VPg antibody (bottom panel). Nonspecific protein bands, marked by triangles, serve as equal loading controls. Note that in the Western blot assay with anti-Pro antibody, VPgProPol was masked by a 95-kDa nonspecific protein band but was visualized by reprobing with anti-VPg antibody. A novel form of VPg at 26 kDa, approximately 6 kDa larger than the mature VPg, was detected by anti-VPg antibody and is marked as VPg*. Additional new cleavage products at 100 kDa and 45 kDa, likely to be VPg*-containing precursors VPg*Pro and VPg*ProPol, respectively, were detected by both anti-VPg and anti-Pro antibodies. Only relevant areas of the membrane are shown.

FIG 3

Cell type-specific activities of NV Pro and precursors as measured by NoroGLuc assay. COS-7 (top) and HEK293 (bottom) cells were cotransfected with NoroGLuc and a control vector or the indicated Pro constructs. GLuc assay of culture medium was performed at the indicated number of hours posttransfection.

FIG 4

Effects of NV Pro and precursors on GLuc secretion. At the last time point of the NoroGLuc assay in Fig. 3, after the medium sample was collected to measure secreted GLuc activity, the remaining medium was removed and the cell monolayer was lysed to measure intracellular GLuc activity. GLuc activities in the medium and cell lysate were used to calculate GLuc secretion efficiency.

FIG 5

Detection of NV RNA replication by NoroGLuc assay. (A) Enrichment of NV particles from stool filtrate by immunocapture. NV particles in stool filtrate were immunocaptured by anti-NV VP1 antibody-coated magnetic beads. Western blot of NV VP1 shows comparison of NV particles in 25 μl stool filtrate versus those from 1 ml stool filtrate that were captured by the beads. (B) Replication of NV stool RNA in HEK293 cells following transfection. HEK293 cells were transfected with NV stool RNA. At 48 h posttransfection, cells were fixed and immunostained with guinea pig anti-NV VP1 antibody. Nuclei were counterstained with DAPI. (C) Detection of NV stool RNA replication in HEK293 cells by the NoroGLuc assay. HEK293 cells were transfected with vector or NoroGLuc, cultured for 24 h, and then transfected with either carrier RNA or NV stool RNA. GLuc assay of culture medium was performed at 18 h post-RNA transfection. Statistically significant difference between samples was calculated by Student's t test (n = 3).

FIG 6

The membrane association domain of p41 is critical for NoroGLuc activity. (A) Schematic showing domain map of p41 in the full-length and N-terminally truncated NoroGLuc constructs. The N-terminal membrane association domain (MAD; aa 1 to 61) is deleted in NoroGLuc-62, and both the MAD and a central region (aa 62 to 148) are deleted in NoroGLuc-149. (B) Comparable expression levels of full-length and N-terminally truncated NoroGLuc constructs in COS-7 cells were confirmed by Western blotting using anti-GLuc antibody. (C) NoroGLuc assay of full-length or N-terminally truncated NoroGLuc constructs together with control vector or NV Pro in COS-7 cells.

FIG 7

NoroGLuc is a suitable substrate for the protease of norovirus GII.3 strain U201. (A) Comparison of the p41/GLuc cleavage site with other p41/p22 cleavage sites among human noroviruses. GenBank accession numbers: GI.1 strain Norwalk virus, NC_001959; GII.3 strain U201, AB039782; GII.4 strain Houston virus, EU310927. (B) NoroGLuc assay of NV Pro versus U201 Pro in HEK293 cells. (C) Cleavage of NoroGLuc into p41 and GLuc and expression of NV Pro and U201 Pro were confirmed by Western blotting using anti-NV p41, anti-GLuc, anti-NV Pro, and anti-U201 Pro antibodies, respectively. Nonspecific protein bands, marked by asterisks, serve as equal loading controls.

FIG 8

NoroGLuc is a suitable substrate for the protease of norovirus GII.4 strain HOV. (A) NoroGLuc assay of NV Pro versus HOV Pro in COS-7 cells. (B) Cleavage of NoroGLuc into p41 (top) and uncleaved NoroGLuc (middle) and expression of NV Pro and HOV Pro (bottom) were confirmed by Western blotting using anti-NV p41, anti-GLuc, and anti-NV Pro antibodies, respectively. Note that both NV Pro and HOV Pro are recognized by the anti-NV Pro antibody.

FIG 9

Dose-dependent inhibitory effect of ZnCl₂ on HOV Pro by NoroGLuc assay. (A) NoroGLuc assay of HOV Pro at increasing concentrations of ZnCl₂ in COS-7 cells. (B) GLuc activities from panel A were replotted in the form of percentages of that of the control (0 μM ZnCl₂) to show a time-independent inhibitory effect for each concentration. (C and D) Cell viabilities at the last time point of the NoroGLuc assay were determined by crystal violet staining (C) and Western blotting of actin (D).

FIG 10

Dose-dependent inhibitory effect of TLCK on HOV Pro by NoroGLuc assay. (A) NoroGLuc assay of HOV Pro at increasing concentrations of TLCK in COS-7 cells. (B) GLuc activities from panel A were replotted in the form of percentages of that of the control (0 μM ZnCl₂) to show a time-independent inhibitory effect for each concentration. (C) Cell viabilities at the last time point of the NoroGLuc assay were determined by crystal violet staining.