Rare recombination events and occurrence of superinfection exclusion during synchronous and asynchronous infection with homologous murine norovirus strains

Objective:
Human noroviruses (HuNoVs) are recognised as one of the major global causes of non-bacterial gastroenteritis with significant morbidity and mortality in developing countries and a high economic impact in developed countries. Spread primarily via the faecal-oral route, HuNoV infection is typically an acute self-limiting gastrointestinal illness. However, chronic HuNoV infection of immunocompromised persons has been identified as a persistent cause of disease and viral populations in such patients have been postulated as possible reservoirs for novel NoV variants.
The Norovirus genus belongs to the Caliciviridae family of small, non-enveloped, positive sense, single-stranded RNA viruses. This genus is subdivided into at least six genogroups, which infect humans and various animal species. Until the recent report of low-level infection of cultured human B cells, no viable cell culture system existed for the study of HuNoVs. The robustness of this new cell culture system still poses a major hurdle, so that the murine norovirus (MuNoV), replicating efficiently in murine dendritic or macrophage cells, remains the model of choice for in vitro study of noroviruses.
The molecular mechanisms driving viral evolution and specifically that of NoVs, are accumulation of point mutations and recombination, which enables the emergence of new combinations of genetic materials to generate potentially dramatic genomic changes in a recombinant NoV, which clusters within two distinct groups of NoV strains when two different genomic regions are phylogenetically analysed.
The mechanism for NoV recombination is proposed to follow the copy-choice mechanism, involving a template shift during simultaneous replication of two strains infecting the same cell. Numerous NoV recombination events have been highlighted by in silico methods and the phenomenon has recently been shown in vitro with two homologous MuNoV strains.
The object of this study was to qualitatively and quantitatively assess virus progenies generated by the use of different parameters of co- and superinfection of RAW264.7 cells with two homologous MuNoV strains (CW1 and Wu20) and thus help to specify important parameters for the occurrence of recombination events. As prerequisite for recombination events, co-and superinfection are of special interest in viral diseases, such as NoV, for which a persistent stage can be developed.
Methods:
Viruses and Cells
Murine NoVs isolates CW1 and Wu20 were plaque purified and propagated in RAW 264.7 cells (ATCC TIB-71).Virus stocks were produced by infection of RAW 264.7 cells at an MOI of 0.05 and clarified by centrifugation. Passages 8 and 7 for CW1 and Wu20, respectively, were used for the experiments.
Co-infection and superinfection experiments
Monolayers of RAW 264.7 cells were infected with Wu20 at a MOI of 1 on ice. After 1 h, the Wu20 inoculums were removed and stored. The cells were washed twice with PBS and infected with CW1 at various MOI (0.1 ; 1 ; 10) and at various delays of co- or superinfection (0 min, 30 min, 1 h, 2 h, 4 h, 8 h, 12 h and 24 h). For co-infections, CW1 and Wu20 were simultaneously inoculated on the cells for 1 h on ice. Twenty-four hours after CW1 co- or superinfections, cells and supernatants were collected.
Molecular analysis
RNA was extracted both from supernatants from the experiment and from propagations of individual plaques, reverse transcribed into cDNA and analysed via two parallel real time PCR reactions allowing discrimination between CW1 and Wu20 at both genomic extremities (regions 1 and 5, located at the ORF1 and ORF3 terminal respectively), as previously described by Mathijs et al., 2010.
For analysis of the supernatants, quantifications were also performed via real time PCR. Accordingly, amplicons corresponding to region 1 were amplified for both CW1 and Wu20, then cloned and in vitro transcribed to provide a standard curve for RNA copies. Following this, values for genomic copies were deduced and results were normalised with GAPDH gene transcripts.
Isolation and screening of progeny viruses
A plaque assay for virus isolation from the co- or superinfection experiments was set up by modifying the protocol described by Hyde et al. (2009). Thus, after 24 h of incubation, 36 plaques were randomly picked for each condition and propagated by inoculation onto RAW 264.7 cells.
After this amplification step, monitoring of recombination events was performed by PCR and Real-Time PCR on extracted, reverse transcribed viral RNA, using two sets of primers to amplify regions 1 and 5. The use of two pairs of TaqMan probes allowed discrimination of the strains WU20 and CW1 in two different regions and identification of recombinant strains.
Results:
Molecular analysis
The Real-Time PCR performed on supernatants collected at 24 h post infection showed a greater number of copies of MuNoV Wu20 cDNA in almost all conditions, except t 0 h, 0.5 h, 1 h, 2 h at the MOI 10, where an increase in the number of copies of the CW1 strain was noted.
In particular, the latter showed a peak at 1 h at the MOI 10 (89%) followed by a rapid reduction in later times (t 8 h: 20%).
Interestingly, for both viruses expected ratios were never attained during the study with the notable exception of the MOI ratio 0.1/1 and the condition t1 MOI 10/1.
Isolation and screening of progeny viruses
Molecular analysis conducted on plaques selected in the condition of coinfection at MOI 1 and 10 highlighted a predominance of the strain MuNoV CW1 (90%) from t 0h to t 2h, followed by a sharp reduction from t 4h leading to complete absence at t 24h.
The Wu20 strain showed a progressive increase from 4h (10%) to 24h (100%).
Overall, the occurrence of recombination events was very rare. Only three putative recombination events were detected at t1 h MOI 1/1 and t 4 h MOI 1/1.
Conclusion:
The profiles of viral ratios over time are highly interesting. Particularly the infection with CW1 at the MOI 10, with a relative percentage of genomic quantifications of about 50% at the two first time points is intriguing. While the percentage is changed completely at t1 MOI 1 to give the expected ratio of 90/10, it then gradually decreases over the next time points to less than 10% at 24 h post infection.
The presence of numerous recombinant viruses as a possible explanation for the t1 MOI 1 peak seems unlikely, as very few putative recombinants were detected during the screening process.
The single-step growth kinetics established by Mathijs et al in 2010, showing great similarities for both strains, indicate that the replicative cycle dynamics of the viruses are probably also not responsible. The decrease, especially marked after 8 h, is suggestive of a superinfection exclusion mechanism, where productive infection with Wu20 induces a resistance of the cells to infection with the homologous CW1. Alternatively, in view of the above-mentioned growth curve at high MOI, the decrease might also be due to the end of the first replication cycle having been reached, with no more viable cells left for infection. Considering the 50% viral ratio estimated by genomic copies for the early time-points, the identification of CW1 as the predominant strain (90%) after plaque purification for the same time appears to be somewhat of a discrepancy and merits further investigation
Although circulating recombinant NoV strains seem to be common, in vitro recombination is a rare event, at least in the protocol described above, and does not seem to be easily influenced by parameter changes such as time of infection and MOI. Parameters where putative recombinations were identified include t1 MOI 1 and t4 MOI 1. The possible recombinants are yet to be confirmed by sequencing reactions.
Further study is necessary to understand mechanisms favouring the predominance of replication of recombinant virus strain in vivo and the challenges of such a replication in vitro. The occurrence of recombination was theoretically limited to one cycle of replication by the protocol (MOI 1 of Wu20). More than one replication cycle might be necessary to enhance the process of recombination by increasing the number of replicating events that could favour recombination. Thus, initial infection at a lower MOI might be an interesting future consideration. Other mechanisms than a time-dependent coinfection might also be worth exploring.
Acknowledgements:
We thank Professor Herbert Virgin and Dr Larissa Thackray (Washington University, St Louis, MO, USA) for providing the MNV isolates and RAW 264.7 cells.