Norovirus VPg Binds RNA through a Conserved N-Terminal K/R Basic Patch

Abstract

The viral protein genome-linked (VPg) of noroviruses is a multi-functional protein that participates in essential roles during the viral replication cycle. Predictive analyses indicate that murine norovirus (MNV) VPg contains a disordered N-terminal region with RNA binding potential. VPg proteins were expressed with an N-terminal spidroin fusion protein in insect cells and the interaction with RNA investigated by electrophoretic mobility shift assays (EMSA) against a series of RNA probes (pentaprobes) representing all possible five nucleotide combinations. MNV VPg and human norovirus (HuNV) VPg proteins were directly bound to RNA in a non-specific manner. To identify amino acids involved in binding to RNA, all basic (K/R) residues in the first 12 amino acids of MNV VPg were mutated to alanine. Removal of the K/R amino acids eliminated RNA binding and is consistent with a K/R basic patch RNA binding motif within the disordered N-terminal region of norovirus VPgs. Finally, we show that mutation of the K/R basic patch required for RNA binding eliminates the ability of MNV VPg to induce a G0/G1 cell cycle arrest.

Keywords: RNA binding; VPg; calicivirus; norovirus; spidroin.

Figure 1

Schematic of MNV VPg. The structural helices and helical core are indicated. Numbers represent the relative positions of amino acids. Amino acid Y26 is essential for nucleotidylylation and F123 is essential for translation. The unstructured N-terminal and C-terminal ends of the protein are indicated. The first 13 amino acids have been shown to be involved in NTP binding by the VPg of HuNV.

Figure 2

Predictions of nucleic acid binding and disorder of MNV VPg. (a) DRNAPred graph depicting predicted RNA (black) and DNA (red) binding probabilities of MNV VPg. (b) Predictions for propensity of disorder of the first 35 N-terminal amino acids of MNV VPg using PrDOS (blue), PONDR VLXT (red), and IUPred2A (red) servers. Regions with predictions above 0.5 were considered disordered [33] and are indicated by the dotted line. (c) Alignment of the first 35 N-terminal amino acids of MNV VPg and the RNA binding region of PVA VPg. An asterisk (*) indicates fully conserved amino acids and a colon (:) indicates amino acids of a similar nature.

Figure 3

Binding of NT*MNV VPg to RNA pentaprobes. (a) Schematic of the NT*MNV VPg, NT*CAT and FOX-1 proteins used in RNA binding experiments. The construct contained a His₆ and Strep II affinity tag, the NT* fusion partner, and the protein to be expressed. An enterokinase cleavage site (EK) was included for cleavage and removal of the fusion partner. (b) Coomassie blue stained SDS-PAGE gel of purified protein (3 μg) used for RNA binding experiments. 1; NT*MNV VPg, 2; NT*CAT, 3; FOX-1. (c) RNA EMSA of purified NT*MNV VPg incubated with 100 nM UTP ATTO 680 labelled pentaprobe RNA. The concentration of NT*MNV VPg was increased from 0.5 μM to 8 μM in 2-fold increments and is indicated by a gradient triangle from low concentration to high concentration. NT*CAT (8 μM) and FOX-1 (8 μM) were included as negative and positive controls, respectively.

Figure 4

NT*MNV VPg preferentially binds RNA. NT*MNV VPg was incubated with 100 nM UTP ATTO 680 labelled pentaprobe RNA and a 100-fold molar excess of either single-stranded or double-stranded unlabelled DNA probe for 30 min. The concentration of NT*MNV VPg was increased from 0.5 μM to 8 μM in 2-fold increments, as shown by a gradient triangle. NT*CAT (8 μM) and FOX-1 (8 μM) were included as negative and positive controls, respectively.

Figure 5

Viral RNA probes. (a) Schematic of the positive-sense viral RNA probes (5′pos and 3′pos). The genome positions of nucleotides for the positive-sense RNA probes are indicated. (b) Schematic of the negative-sense viral RNA probes 3′neg and 5′neg. The negative-sense RNA probes are the complements of the positive-sense RNA probes. RNA secondary structures (a,b) were predicted using mfold [44,45]. (c) NT*MNV VPg was incubated with labelled 5′pos, 3′pos or 3′neg RNA probes and analysed by EMSA. The concentration of NT*MNV VPg was increased from 0.5 μM to 8 μM in 2-fold increments. NT*CAT (8 μM) and FOX-1 (8 μM) were included as negative and positive controls, respectively.

Figure 6

HuNV VPg binds RNA. (a) Coomassie blue stained SDS-PAGE gel of purified protein (3 μg) used for RNA binding experiments. 1; NT*HuNV VPg, 2; NT*CAT, 3; FOX-1. (b) RNA EMSA of NT*HuNV VPg with pentaprobe RNA. Proteins were incubated with 100 nM UTP ATTO 680 labelled pentaprobe RNA for 30 min, separated on a 5% TBE gel, and directly imaged in the 700 nm channel. The concentration of NT*HuNV VPg was increased from 0.5 μM to 8 μM in 2-fold increments. NT*CAT (8 μM) was included as a negative control and FOX- 1 (8 μM) as a positive control.

Figure 7

Alignment of VPg sequences. An asterisk (*) indicates identical residues and a colon (:) indicates residues that are similar. Conserved amino acids of the N-terminal region are shown in red. The tyrosine residue (Y) corresponding to Y26 of MNV VPg is shown in black in bold. The fully conserved amino acids surrounding the Y26 are shown in blue and similar amino acids in green. The represented genogroups are GI Norwalk virus VPg (AAC64602), GII Sydney 2012 VPg (JX459908), GIII Jena virus VPg (CAA90480) and GIV Lake Macquarie VPg (AFJ21375).

Figure 8

N-terminal lysine and arginine amino acids are important for binding to RNA. (a) Schematic of the first 20 amino acids of MNV VPg, wild-type amino acids are shown in blue. Amino acids mutated to alanine are shown in red for the NT*MNV VPg and NT*MNV VPg K2-12 constructs, respectively. (b) Coomassie blue stained SDS-PAGE gel of purified protein (3 μg) used for RNA binding experiments. 1; NT*MNV VPg TM; NT*CAT, 3; NT* MNV VPg K2-12 and 4; FOX-1. (c,d) RNA EMSA of NT*MNV VPg K2-12 and NT*MNV VPg TM with pentaprobe RNA. Proteins were incubated with 100 nM UTP ATTO 680 labelled pentaprobe RNA for 30 min, separated on a 5% TBE gel and directly imaged in the 700 nm channel. The concentration of NT*MNV VPg TM was increased from 0.5 μM to 8 μM in 2-fold increments. NT*CAT (8 μM) was included as a negative control and FOX-1 (8 μM) as a positive control.

Figure 9

Manipulation of the cell cycle by MNV VPg with N-terminal mutations. RAW-Blue cells were transfected with 4–5 μg of transcript RNA corresponding to MNV VPg, MNV VPg TM, or MNV VPg K2-12. Mock transfected (MT) cells were seeded at the time of transfection as a negative control. At 12 h post transfection, cells were harvested for western blot and flow cytometry analysis of the cell cycle. (a) Representative histograms from one of three experiments. The positions of the G0/G1 and G2/M phase populations of cells are labelled on the t = 0 histogram. The S phase lies between the G0/G1 and G2/M populations and is indicated by hatched lines. (c,e) Analysis of total cell lysate by western blot to confirm expression of MNV VPg, MNV VPg TM, and MNV VPg K2-12, with actin as a loading control. (b,d) The histograms were analysed using MODfit LT 3.0 and the percentage of cells in each phase of the cell cycle are shown. Red indicates percentage of cells in G0/G1 phase, green indicates the S phase, and blue indicates the G2/M phase. The results present the mean and SD from three independent experiments. Statistical significance was determined using a one-way ANOVA with Dunnett’s post-test. ns; not significant, * p ≤ 0.05, ** p ≤ 0.01 and *** p ≤ 0.001.