Genomic diversity and phylogeography of norovirus in China

Abstract

Background: Little is known about the phylogeography of norovirus (NoV) in China. In norovirus, a clear understanding for the characteristics of tree topology, migration patterns and its demographic dynamics in viral circulation are needed to identify its prevalence trends, which can help us better prepare for its epidemics as well as develop useful control strategies. The aim of this study was to explore the genetic diversity, temporal distribution, demographic dynamics and migration patterns of NoV that circulated in China.

Results: Our analysis showed that two major genogroups, GI and GII, were identified in China, in which GII.3, GII.4 and GII.17 accounted for the majority with a total proportion around 70%. Our demography inference suggested that during the long-term migration process, NoV evolved into multiple lineages and then experienced a selective sweep, which reduced its genetic diversity. The phylogeography results suggested that the norovirus may have originated form the South China (Hong Kong and Guangdong), followed by multicenter direction outbreaks across the country.

Conclusions: From these analyses, we indicate that domestic poultry trade and frequent communications of people from different regions have all contributed to the spread of the NoV in China. Together with recent advances in phylogeographic inference, our researches also provide powerful illustrations of how coalescent-based methods can extract adequate information in molecular epidemiology.

Keywords: Bayesian phylogenetics; China; Demographic dynamics; Norovirus; Phylogeography.

Fig. 1

Distribution of NoV sequences reported to the GenBank between 1976 and 2015 (n = 3134). a Nested pie chart of the NoV genotype distribution proportions. b Distribution of NoV sequences. All NoV sequences were grouped by year and divided into “All NoV”, “GI”, “GII”, “GII.3”, “GII.4” and “GII.17”. The solid line with symbol represents the percentages of “GII.3”, “GII.4” and “GII.17” for the years shown

Fig. 2

Maximum clade credibility (MCC) phylogenies of NoV GI (a) and GII (b) in China. Branches are colored according to the most probable location state of their descendent nodes. The scale bar in the bottom indicates the years before the most recent sampling time (2015)

Fig. 3

Posterior root state probability calculated from GI (a) and GII (b) MCC phylogenies. The histogram shows the posterior probability distributions of root location state of GI and GII

Fig. 4

Bayesian skyline plots of NoV in China estimated from GI (a) and GII (b) VP1 sequences. The plots illustrate the relative effective population size (genetic diversity) of GI and GII NoVs through time. The solid black line represents the mean posterior value and the blue area corresponds to the 95% highest probability density (HPD) intervals

Fig. 5

Spatiotemporal dispersal of NoV among different localities of China reconstructed using discrete phylogeographic analysis of GI (a) and GII (b). Snapshots of the dispersal pattern of GI and GII were provided for 1700, 1900, 1950, 1980, 2000, 2015 and 1860, 1900, 1950, 1980, 2000, 2015, respectively. The red lines represented MCC phylogeny branches projected on the surface, whereas the uncertainty on the locations of NoV was represented by cyan polygons. The 95% HPD regions were obtained by imputing locations on each branch across the posterior distribution using bivariate kernel density estimates for the respective sample. The maps are based on satellite pictures made available in Google Earth (http://earth.google.com)

Fig. 6

Bayesian factor (BF) test for significant non-zero dispersion routes in NoV GI (a) and GII (b). Only routes supported by a BF greater than 3 are plotted. The line color represent the relative strength by which the routes are supported: dark red lines and bright red lines suggest weak and strong support respectively. The maps are based on satellite pictures made available in Google Earth (http://earth.google.com)