Phylogeography of rubella virus in Asia: Vaccination and demography shape synchronous outbreaks

Abstract

Rubella virus causes mild disease in children but for women in the early stages of pregnancy, it can cause spontaneous abortion, congenital rubella syndrome (CRS) and associated birth defects. Despite the availability of an effective vaccine, rubella virus continues to circulate endemically in several regions of the world. This is particularly true in East and Southeast (E/SE) Asia, where control efforts vary widely among countries that are well connected through travel and immigration. It is therefore important to understand how the regional persistence of rubella is affected both by dynamics occurring across countries and susceptibility within countries. Here, we use genetic and epidemiological data from countries in E/SE Asia to explore the phylogeography of rubella virus in this region. Our results underline that metapopulation dynamics are key for rubella persistence and highlight the source-sink population structure of the region. We identify countries that contribute to the regional metapopulation network and link epidemic dynamics to susceptibility profiles within each country. Our results indicate that human movement plays an important role in driving epidemic dynamics in E/SE Asia.

Keywords: Epidemic dynamics; Human movement; Metapopulation; Phylogeography; Rubella virus.

Fig. 1

Phylogenetic trees. Phylogenetic trees for a) genotype 1E and b) genotype 2B. Branch colors represent the most probable location state of the lineage at that time. Posterior support values indicated for each of the identified clades. X-axis represented in years. Branch widths sized according to posterior probability, where thicker branches indicate greater posterior support. Clades shaded in grey represent continuous transmission of a lineage outside of Asia. Interior nodes are colored based on their inferred geographic location and sized by the posterior support for this placement, where larger circles indicate greater posterior support.

Fig. 2

Circulation patterns of rubella virus in Asia. Circulation maps of a) genotype 1E and b) genotype 2B. “Other” represents sequences collected from outside of Asia. Connections between countries are represented by lines colored by their Bayes factor (BF). Dark red (BF > 100) indicates decisive support for the connection, red indicates very strong support (100 > BF > 30), orange indicates strong support (30 > BF > 10). Connections with weak support (10 > BF > 3) are not mapped. Total estimated number of viral importations and exportations per country for c) genotype 1E and d) genotype 2B. Locations identified as being outside the metapopulation circulation network (i.e. no significant connections to the countries of E/SE Asia) are shaded in grey in panels c and d (For interpretation of the references to coloured in this figure legend, the reader is referred to the web version of this article).

Fig. 3

Viral effective population sizes (N_e). Estimates of viral N_e in E/SE Asia for each genotype plotted with total reported cases across the six focal countries. Sparse sample collection precluded estimating dynamics for genotype 2B prior to 2005.