Phylodynamic reconstruction reveals norovirus GII.4 epidemic expansions and their molecular determinants

Abstract

Noroviruses are the most common cause of viral gastroenteritis. An increase in the number of globally reported norovirus outbreaks was seen the past decade, especially for outbreaks caused by successive genogroup II genotype 4 (GII.4) variants. Whether this observed increase was due to an upswing in the number of infections, or to a surveillance artifact caused by heightened awareness and concomitant improved reporting, remained unclear. Therefore, we set out to study the population structure and changes thereof of GII.4 strains detected through systematic outbreak surveillance since the early 1990s. We collected 1383 partial polymerase and 194 full capsid GII.4 sequences. A Bayesian MCMC coalescent analysis revealed an increase in the number of GII.4 infections during the last decade. The GII.4 strains included in our analyses evolved at a rate of 4.3-9.0x10(-3) mutations per site per year, and share a most recent common ancestor in the early 1980s. Determinants of adaptation in the capsid protein were studied using different maximum likelihood approaches to identify sites subject to diversifying or directional selection and sites that co-evolved. While a number of the computationally determined adaptively evolving sites were on the surface of the capsid and possible subject to immune selection, we also detected sites that were subject to constrained or compensatory evolution due to secondary RNA structures, relevant in virus-replication. We highlight codons that may prove useful in identifying emerging novel variants, and, using these, indicate that the novel 2008 variant is more likely to cause a future epidemic than the 2007 variant. While norovirus infections are generally mild and self-limiting, more severe outcomes of infection frequently occur in elderly and immunocompromized people, and no treatment is available. The observed pattern of continually emerging novel variants of GII.4, causing elevated numbers of infections, is therefore a cause for concern.

Figure 1. Phylodynamics of the GII.4 noroviruses.

The left panel (A, B, C) describes analysis of the polymerase dataset, the right panel (D, E, F) the analysis of the capsid dataset. A) Bayesian Skyline Plot (BSP) of GII.4 NoV partial polymerase sequences, representing the relative genetic diversity, a measure for the number of effective infections, of circulating GII.4 NoVs through time. The black line represents the median posterior value, the grey lines the 95% Highest Probability Density (HPD) intervals. The Y-axis depicts the value of N_eτ on a logarithmic scale. B and E) Surveillance data of NoV GII.4 strains, detected in outbreaks, included for comparison. Only European surveillance data was obtained from the FBVE database and from Dutch surveillance , . The different GII.4 variants detected in the outbreaks are represented by different colors, showing the displacement of subsequent variants through time. The same colors were used in panels D and F, the color legend is shown under panel F. C) Maximum Clade Credibility (MCC) Tree of NoV GII.4 partial polymerase dataset. The MCC tree represents an estimate of the posterior distribution of tree topologies and branch lengths. Different variants are represented in different colors, the tree is scaled in units of time with tips constrained to strain detection dates. D) BSP of the GII.4 NoV capsid dataset. F) MCC tree of the GII.4 NoV capsid dataset.

Figure 2. Bayesian Skyline Plots of different GII.4 variants analyzed separately.

Only the mean population sizes through time were plotted for clarity.

Figure 3. Codon substitution rate classes of GII.4 capsids.

A) Inferred weights for each evolutionary rate class, showing the relative proportion of sites belonging to the given classes. The diagonal line divides the regimes of positive selection (dN/dS>1, above) and negative selection (dN/dS<1, below). B) Approximate inferred posterior distribution of synonymous (alpha) and non-synonymous (beta) substitution rates, showing the variance of each rate estimate. Color intensity is proportional to the square root of the density, and solid oval plots delineate approximate 95% confidence sets.

Figure 4. Sites identified by Bayesian graphical models to co-evolve.

Amino acid sites 297, 372, 368, 394, 407, 534 and 540 of the GII.4 capsid protein were depicted as trees showing different amino acids in different colors. Each connection is associated with posterior probabilities (P) for single (above or left) or two (below or right) dependencies.

Figure 5. Sites identified by molecular adaptation analyses located on top of the GII.4 capsid dimer.

Two parts of the dimer were given different shades of blue, the sites identified as under positive selection, co-evolving with other sites, or sites previously identified to be involved in host-interactions, were colored yellow. Ligands (B trisaccaride) bound in the binding pocket are shown in red. Protein structure 2OBT was used for generating this representation .

Figure 6. Predicted secondary structures in the 5′end of the ORF2 GII.4 RNA, with all possible nucleotides modeled at site 26 (boxed).

Four nucleotides upstream of the first ATG were included, as these are conserved in the 5′end of ORF1 and ORF2 and likely to have a function in translation.