Genetic diversity and population structure of Plasmodium vivax in Central China

Abstract

Background: In Central China the declining incidence of Plasmodium vivax has been interrupted by epidemic expansions and imported cases. The impact of these changes on the local parasite population, and concurrent risks of future resurgence, was assessed.

Methods: Plasmodium vivax isolates collected from Anhui and Jiangsu provinces, Central China between 2007 and 2010 were genotyped using capillary electrophoresis at seven polymorphic short tandem repeat markers. Spatial and temporal analyses of within-host and population diversity, population structure, and relatedness were conducted on these isolates.

Results: Polyclonal infections were infrequent in the 94 isolates from Anhui (4%) and 25 from Jiangsu (12%), with a trend for increasing frequency from 2008 to 2010 (2 to 19%) when combined. Population diversity was high in both provinces and across the years tested (H(E) = 0.8 - 0.85). Differentiation between Anhui and Jiangsu was modest (F'(ST) = 0.1). Several clusters of isolates with identical multi-locus haplotypes were observed across both Anhui and Jiangsu. Linkage disequilibrium was strong in both populations and in each year tested (I(A)(S) = 0.2 - 0.4), but declined two- to four-fold when identical haplotypes were accounted for, indicative of occasional epidemic transmission dynamics. None of five imported isolates shared identical haplotypes to any of the central Chinese isolates.

Conclusions: The population genetic structure of P. vivax in Central China highlights unstable transmission, with limited barriers to gene flow between the central provinces. Despite low endemicity, population diversity remained high, but the reservoirs sustaining this diversity remain unclear. The challenge of imported cases and risks of resurgence emphasize the need for continued surveillance to detect early warning signals. Although parasite genotyping has potential to inform the management of outbreaks, further studies are required to identify suitable marker panels for resolving local from imported P. vivax isolates.

Figure 1

Spatial distribution of P. vivax endemicity in 2010 in China. This map was generated by Zhi Huang, Malaria Atlas Project, University of Oxford. The colour scale reflects the age-standardized P. vivax parasite rate (PvPR), which describes the estimated proportion of the general population that are infected with P. vivax at any one time, averaged over the 12 months of 2010 within the spatial limits of stable transmission [17]. The main figure presents the distribution across all China, with labelling of the provinces from which samples are represented in this study. The plots in the right-hand boxes indicate the prefectures from which the majority of malaria cases in Anhui Province (top) and Jiangsu Province (bottom) originated.

Figure 2

Incidence of indigenous P. vivax cases in Anhui and Jiangsu Province. Data source: Annual case report of parasitic diseases from Ministry of Health and Health Department of Jiangsu Province. Sampling time points are indicated by filled in circles.

Figure 3

Unrooted neighbour-joining trees illustrating genetic relatedness between isolates. Tree generated with data from all markers with the exception of MS10. Only samples with no missing data at these loci are presented, including 103 central Chinese, 1 southern Chinese, and 4 imported isolates. Colour-coding reflects the province (first administrative level) for the Chinese isolates, and country for the imported isolates. Further geographic information on the origin of the Chinese isolates is provided in the sample labels which refer to the prefecture (second administrative level). Multi-locus haplotypes observed 3 or more times are labelled.

Figure 4

Temporal haplotype dynamics. Frequencies of multi-locus haplotypes reconstructed from the genotypes at all loci with the exception of MS10 are presented by quarter. For simple visual representation, only haplotypes observed 3 or more times in the full data set are presented in colour, whilst all other (low frequency) haplotypes are presented in gray. Note - exact dates of collection (only year of collection) were not available for nineteen of the samples from Anhui with full data at the loci investigated. These isolates (including three from 2009 with Hap1, one 2009 with Hap4, one 2008 with Hap33, three 2009 with Hap35, and two 2009 with Hap40) were therefore not presented in the plot.