Development of a reverse-genetics system for murine norovirus 3: long-term persistence occurs in the caecum and colon

Abstract

Human noroviruses (HuNoV) are a major cause of viral gastroenteritis worldwide, yet, due to the inability to propagate HuNoV in cell culture, murine norovirus (MNV) is typically used as a surrogate to study norovirus biology. MNV-3 represents an attractive strain to study norovirus infections in vivo because it establishes persistence in wild-type mice, yet causes symptoms resembling gastroenteritis in immune-compromised STAT1(-/-) mice. The lack of reverse-genetics approaches to recover genetically defined MNV-3 has limited further studies on the identification of viral sequences that contribute to persistence. Here we report the establishment of a combined DNA-based reverse-genetics and mouse-model system to study persistent MNV-3 infections in wild-type (C57BL/6) mice. Viral RNA and infectious virus were detected in faeces for at least 56 days after inoculation. Strikingly, the highest concentrations of viral RNA during persistence were detected in the caecum and colon, suggesting that viral persistence is maintained in these tissues. Possible adaptive changes arising during persistence in vivo appeared to accumulate in the minor capsid protein (VP2) and the viral polymerase (NS7), in contrast with adaptive mutations selected during cell-culture passages in RAW264.7 cells that appeared in the major capsid protein (VP1) and non-structural protein NS4. This system provides an attractive model that can be readily used to identify viral sequences that contribute to persistence in an immunocompetent host and to more acute infection in an immunocompromised host, providing new insights into the biology of norovirus infections.

Fig. 1.

Recovery of MNV-3 by reverse genetics. (a) Schematic representation of the MNV-3 genomic and subgenomic RNAs. MNV-3 contains four ORFs. Translation of ORF1 results in the synthesis of a viral polyprotein that is then processed proteolytically into different mature non-structural viral proteins as indicated. Translation of the subgenomic RNA results in the synthesis of major capsid protein VP1 (ORF2), minor capsid protein VP2 (ORF3) and virulence factor VF1 (ORF4). The sequence of pT7 : MNV-3 differs from the previously reported sequence in 53 nt positions along the genome, represented as short arrows with non-filled arrowheads (synonymous) and as long arrows with filled arrowheads (non-synonymous). (b) TCID₅₀ virus titres obtained after recovery of pT7 : MNV-1, pT7 : MNV-3 and pT7 : F/S (pT7 : MNV with a frameshift mutation) in BSR-T7 cells 24 and 48 h post-transfection, as explained in the text. (c) Virus titres of MNV-3 samples resulting from serial passage of MNV-3 recovered in RAW264.7 cells. MNV-3 recovered in (b) was used to infect RAW264.7 cells at a low m.o.i. (0.001). MNV-3 produced (passage 1) was used for serial passage at a low m.o.i. of 0.1 (passages 2–5) or 0.3 (passage 6) in RAW264.7 cells.

Fig. 2.

MNV-3 recovered by reverse genetics establishes persistent infections in mice. C57BL/6 mice, 4–5 weeks of age, were inoculated with 100 µl sample containing 1.0×10⁵ TCID₅₀ MNV-3 p2 or 1.5×10⁶ TCID₅₀ MNV-3 p6. Controls included uninfected and mock-infected animals. (a) No significant differences were observed in the weight gain of inoculated animals (○, p2; ▽, p6) compared with controls (▴, uninfected; ▪, mock-infected). (b, c) Viral RNA was extracted from stool pellets collected (○, p2; ▽, p6; ▪, mock-infected) and was detected by quantitative (b) and semiquantitative (c) RT-PCR in all infected animals, but not in mock-infected animals. Quantitative and semiquantitative PCRs were performed as described in the text. Horizontal bars in (a) and (b) represent mean values. −ve, Negative control; Ld, DNA ladder.

Fig. 3.

Low-dose inoculation of MNV-3 results in establishment of viral persistence. C57BL/6 mice 4–5 weeks of age were inoculated with 100 µl sample containing 10⁴, 10³, 10² or 10 TCID₅₀ MNV-3, 10⁵ TCID₅₀ UV-inactivated MNV-3, or filtered cell-culture lysate (Mock). (a) Animals infected with 10 or 10⁴ TCID₅₀ MNV-3 showed detectable viral RNA in faeces by day 28 post-infection. Animals infected with 10² or 10³ TCID₅₀ were also positive (not included in the figure). Animals mock-inoculated with UV-inactivated MNV-3 or cell-culture lysate presented no detectable viral RNA. (b) Viral RNA levels in faeces collected at 28 days post-infection from animals mock-infected (▪) or infected with 10 (◊), 10⁴ (□) or 10⁵ (○) TCID₅₀ MNV-3 (from Fig. 2). (c) Titration by TCID₅₀ in RAW264.7 cells of infectious virus isolated from faecal samples of animals shown in (b). Horizontal bars in (b) and (c) represent mean values.

Fig. 4.

MNV-3 shows fast kinetics of replication in mice. Groups of six mice were infected with a low dose of MNV-3 [10 (•) or 10² (□) TCID₅₀] and secretion of virus was monitored at days 1, 2, 3, 7 and 28 post-infection. MNV-3 shows fast replication kinetics, reaching maximum levels of RNA molecules in stools by day 2 post-infection. Establishment of infection is dose-dependent, with three of six animals inoculated with 10² TCID₅₀, and only one of six animals inoculated with 10 TCID₅₀, being positive for virus shedding at 1 day post-inoculation. Horizontal bars represent mean values.

Fig. 5.

MNV-3 persists in the caecum and the colon. High levels of MNV RNA were detected in caecum and colon samples of animals 28 days post-inoculation. Lower amounts of viral RNA were detected in MLN and small intestine of some animals. MNV RNA levels in liver and spleen were below the detection limit [<4×10² viral RNA molecules (µg total RNA)⁻¹]. The organs and tissues collected included liver, spleen, MLN, different 1 inch sections of the small intestine comprising dudodenum/jejunum (1–2 inches from the stomach end), proximal ileum (3–4 inches from the stomach end) and distal ileum (the last inch before the caecum), caecum and colon. Horizontal bars represent mean values; the dotted line represents the limit of detection.

Fig. 6.

Sequence analysis of viral RNA from stools at days 28 (animals 113, 122, 149, 152, 154) and 56 (animal 16, which presented the highest viral RNA concentration at that time point). Empty and filled symbols represent synonymous and non-synonymous substitutions, respectively. ◊/⧫ represent mutations not totally imposed in the consensus sequence (30–80 % relative to wild-type sequence based on chromatogram quantification); • represents a totally imposed change.